HK40081641B - Conditionally active polypeptides - Google Patents

Description

Conditionally active peptides

This application is a divisional application of Chinese Patent Application No. 201680077540.7, filed on August 31, 2016, entitled “Conditionally Active Peptide”, and claims priority to US 62/249,907 and PCT/US2016/019242.

Technical Field

This invention relates to the field of providing improved peptides with desired activity. Specifically, this invention relates to a method for generating conditionally active peptides from parental peptides, wherein the conditionally active peptides are more active under one condition than under another in the presence of a particular compound or ionic substance.

Background Technology

Numerous documents describe methods for evolving the properties of proteins (particularly enzymes or antibodies) to make them active or stable under different conditions. For example, enzymes have been evolved to be stable at higher temperatures. In cases where enzyme activity increases at higher temperatures, this increase is largely attributable to the higher kinetic activity generally described in the Q10 rule, which states that for enzymes, the conversion rate is expected to double for every 10°C increase in temperature.

In addition, there are natural mutations that destabilize proteins under their normal operating conditions, thereby reducing their activity under normal operating conditions. For example, there are known temperature mutants that are active at lower temperatures, but their activity is usually reduced compared to the wild-type protein from which the mutant was derived.

It is desirable to generate peptides with conditional activity, such as peptides that are less or almost inactive under one condition but active under another. It is also desirable to generate peptides that are activated or inactivated in specific environments, or peptides that are activated or inactivated over time. Besides temperature, other conditions that can evolve peptides or provide conditional activity include pH, osmotic pressure, molality, oxidative stress, and electrolyte concentration. In addition to peptide activity, it is generally desirable to improve other properties during evolution, including chemical resistance and proteolytic resistance.

Many strategies for evolving peptides have been described previously. For example, US2005/0100985 discloses a rapid and convenient method for generating a set of mutant polynucleotides from a parent template polynucleotide by replacing each original codon position in a template polynucleotide with codons encoding 20 naturally occurring amino acids. This method is simply called saturation mutagenesis and can be used in combination with other mutagenesis methods, such as those that introduce two or more related polynucleotides into a suitable host cell to generate heterozygous polynucleotides through recombination and reduced rearrangement.

Giver et al., “Directed evolution ofathermostable esterase,” Proc. Natl. Acad. Sci. USA, vol. 95, pp. 12809-12813 (1998), used in vitro evolution to study the relationship between the stability and activity of mesophilic esterases. Six generations of random mutagenesis, recombination, and screening significantly stabilized Bacillus subtilis p-nitrobenzyl esterase (increased Tm > 14 °C) without impairing its catalytic activity at lower temperatures. This study found that mutations that increase thermostability while maintaining low-temperature activity are very rare. Improving one property through the accumulation of amino acid substitutions usually comes at the expense of another property, regardless of whether the two properties are inversely related or completely unrelated.

Developing a parent peptide into one that is inactive or nearly inactive (less than 50%, 30%, or 10% activity, especially 1% activity) under its normal operating conditions, while maintaining equivalent or better activity under anomalous conditions, may require a combination of destabilizing mutations and activity-enhancing mutations that do not counteract the destabilizing effect. It is anticipated that the reduction in peptide activity due to destabilizing mutations will be more significant than predicted by standard rules such as Q10; therefore, the ability to develop peptides that function effectively under anomalous conditions—for example, while being low-activity or inactive under their normal operating conditions—will result in conditionally active peptides.

Therefore, conditionally active peptides exhibit enhanced activity compared to their parent proteins under abnormal conditions, while their activity decreases under normal physiological conditions. When used as therapeutic proteins, conditionally active peptides thus preferentially act on sites where abnormal conditions exist, such as the tumor microenvironment. Due to this preferential action, conditionally active peptides potentially cause less damage to normal tissues/organs under normal physiological conditions, thus producing fewer side effects. This allows for longer treatment durations or higher doses of the conditionally active peptide, resulting in greater therapeutic efficacy.

WO2010/104821 and WO2011/009058 disclose methods for evolving and screening conditionally active proteins.

There is still a need for conditionally active peptides that exhibit high activity and/or selectivity in specific environments and/or under specific conditions.

Summary of the Invention

In one embodiment, the present invention relates to a non-naturally occurring polypeptide or isolated polypeptide having an activity ratio of at least 1.3, said activity ratio being the ratio of the activity in a test at a first pH in the presence of at least one substance having a molecular weight of less than 900 a.m.u. and whose pKa differs from the first pH by up to 0.5, 1, 2, 3 or 4 pH units to the activity in a test at a second pH in the presence of the same at least one substance.

In one embodiment, the present invention relates to a non-naturally occurring polypeptide or isolated polypeptide having an activity ratio of at least 1.3, the activity ratio being the ratio of activity in a test at a first pH in the presence of at least one substance having a molecular weight of less than 900 a.m.u. to activity in a test at a second pH in the presence of the same at least one substance, and wherein the pKa of said substance is between the first pH and the second pH.

In one embodiment, the present invention relates to a non-naturally occurring polypeptide or isolated polypeptide having an activity ratio of at least 1.3, said activity ratio being the ratio of the activity in a test at a first pH in the presence of a substance selected from histidine, histamine, hydrogenated adenosine diphosphate, hydrogenated adenosine triphosphate, citrate, bicarbonate, acetate, lactate, sulfide, hydrogen sulfide, ammonium, dihydrogen phosphate, and any combination thereof to the activity in a test at a second pH in the presence of the same substance.

In any of the foregoing embodiments, the ratio of the activity of the peptide in a test at a first pH to its activity in a test at a second pH can be at least 1.5, or at least 1.7, or at least 2.0, or at least 3.0, or at least 4.0, or at least 6.0, or at least 8.0, or at least 10.0, or at least 20.0, or at least 40.0, or at least 60.0, or at least 100.0. The peptide described in the foregoing embodiments can be tested at a first pH that is acidic and a second pH that is alkaline or neutral. The second pH can be a normal physiological pH, within the normal range of physiological conditions at the site of peptide administration to the individual or within the normal range of physiological conditions in the tissue or organ at the site of peptide action in the individual; the first pH can be an abnormal pH, deviating from the normal range of physiological conditions at the site of peptide administration or within the normal range of physiological conditions in the tissue or organ at the site of peptide action.

The first pH can be in the range of 5.5-7.2, or in the range of 6.2-6.8. The second pH can be in the range of 7.2-7.6. The first pH can be approximately 6.0, and the second pH can be approximately 7.4.

The polypeptide described in any of the foregoing embodiments may be a non-naturally occurring mutant polypeptide evolved from a parent polypeptide. The mutant polypeptide described in any of the foregoing embodiments may be derived from a wild-type parent polypeptide that includes a non-naturally occurring polypeptide. The mutant polypeptide described in any of the foregoing embodiments may contain at least one amino acid substitution compared to the parent polypeptide. The mutant polypeptide described in any of the foregoing embodiments may have a higher proportion of charged amino acid residues than the parent polypeptide.

In any of the foregoing embodiments, the polypeptide or mutant polypeptide can be a protein or a protein fragment. In any of the foregoing embodiments, the polypeptide or mutant polypeptide can be selected from antibodies, single-chain antibodies, and antibody fragments, and the activity is antigen-binding activity. In any of the foregoing embodiments, the polypeptide or mutant polypeptide can be the Fc region of an antibody. The polypeptide or mutant polypeptide can be an enzyme, and the activity can be enzymatic activity. The polypeptide or mutant polypeptide can be selected from receptors, regulatory proteins, soluble proteins, cytokines, and fragments of receptors, regulatory proteins, soluble proteins, or cytokines.

The substance described in any of the foregoing embodiments can be hydrogen sulfide, bicarbonate, or sulfide ions. The pKa of any of the substances described in any of the foregoing embodiments can be greater than 6.2.

In any of the foregoing embodiments, the polypeptide may have two functional domains, and the activity is the activity of one of the two functional domains. Both functional domains may have pH-dependent activity. The polypeptide may be a bispecific antibody.

In another embodiment, the polypeptide described in any of the foregoing embodiments can be used to treat solid tumors, inflamed joints, or brain diseases or conditions.

In another embodiment, the present invention relates to a method for treating solid tumors, inflamed joints, or brain diseases or conditions, the method comprising administering a polypeptide as described in any of the preceding embodiments. The polypeptide may be administered as part of a chimeric antigen receptor comprising the polypeptide for T cells or as an antibody-drug conjugate comprising the polypeptide.

In another embodiment, the present invention relates to a chimeric antigen receptor comprising the polypeptide for T cells. In each of the foregoing embodiments, the antibody may be conjugated to nanoparticles.

In another embodiment, the present invention relates to an antibody-drug conjugate comprising the said polypeptide.

In another embodiment, the present invention provides a pharmaceutical composition comprising a conditionally active biological protein and a pharmaceutically acceptable carrier.

In another embodiment, the present invention provides a method for producing conditionally active peptides from parental peptides, the method comprising the following steps:

(i) The parent polypeptide is evolved by mutating at least one region outside the active site of the parent polypeptide to produce one or more mutant polypeptides;

(ii) Performing a first test on the one or more polypeptides and the parent polypeptide under normal physiological conditions to measure the activity of the active site under normal physiological conditions, and performing a second test under abnormal conditions to measure the activity of the active site under abnormal conditions, wherein the normal physiological conditions and the abnormal conditions are the same conditions with different values; and

(iii) Select from the one or more mutant peptides a conditionally active peptide exhibiting the following two properties: (a) decreased activity compared to the same activity of the parent peptide in a first test, and (b) increased activity compared to the same activity of the parent peptide in a second test.

Attached Figure Description

Figure 1 shows the conditionally active antibody produced in Example 9 and its selectivity at pH 6.0 relative to pH 7.4.

Figure 2 shows the binding activity of conditionally active antibodies to antigens tested in different buffer solutions.

Figure 3 shows the effect of changing the composition of Krebs buffer on the binding activity of conditionally active antibodies.

Figure 4 shows that, as described in Example 12, the binding activity of the three different conditionally active antibodies depends on the presence and concentration of bicarbonate at pH 7.4.

Figure 5 is a diagram showing the structure of a chimeric antigen receptor (CAR).

Figure 6 shows the formation of salt bridges in deoxyhemoglobin, where three amino acid residues form two salt bridges that stabilize the T quaternary structure of deoxyhemoglobin, resulting in a lower affinity for oxygen.

Figure 7 shows the activity of conditionally active antibodies against Ror2 in different buffer solutions.

Figure 8 shows the activity of conditionally active antibodies against Axl in different buffer solutions.

Figure 9 shows the activity of the conditionally active antibody against Axl, as observed using a test solution containing 10 mM thiosulfate ions.

definition

To facilitate understanding of the examples provided in this article, some frequently occurring methods and/or terms will be defined in this article. The following terms are incorporated into the definition by reference to U.S. Patent No. 8,709,755B2: “reagent,” “ambiguous base requirement,” “amino acid,” “amplification,” “chimeric property,” “cognate,” “comparison window,” “conserved amino acid substitution,” “corresponding to,” “degradation-effective,” “defined sequence backbone,” “defined sequence kernel,” “digestion,” “directed ligation,” “DNA shuffling,” “drug” or “drug molecule,” “effective amount,” “epitope,” “enzyme,” “evolution” or “evolutionary process,” “fragment” or “derivative” or “analogue,” “single amino acid substitution,” “gene,” “genetically unstable,” “heterologous,” “homologous” or “partially homologous,” “industrial application,” “identical” or “identity,” “identity region,” “isolated,” “isolated nucleic acid,” “ligand,” “ligand,” “connector” or “spacer,” “microenvironment,” “molecular property to be evolved,” “mutation,” “N,N,G/T,” “normal physiological conditions” or “wild-type working conditions,” “nucleic acid molecule.” Nucleic acid molecule, nucleic acid sequence for encoding… or DNA-encoding sequence or nucleotide sequence encoding…, nucleic acid encoding enzyme (protein) or DNA encoding enzyme (protein) or polynucleotide encoding enzyme (protein), specific type of nucleic acid molecule, assembling working nucleic acid samples into a nucleic acid library, nucleic acid library, construct, oligonucleotide (or synonymous “oligonucleotide”), homologous, operablely linked, parental polynucleotide combination, patient or individual, physiological condition, group, precursor form, pseudo-random, quasi-repeated units, random peptide library, random peptide sequence, receptor, recombinant enzyme, synthetic enzyme, associated polynucleotide, reductive rearrangement, reference sequence, repetition index (RI), restriction site, alternative polynucleotide, sequence identity, similarity, specific binding, specific hybridization, specific polynucleotide, strict hybridization conditions, substantially identical, substantially pure enzyme, substantially pure, therapeutic, variable segment, and variant.

As used herein, the term “approximately” in relation to the quantity being measured refers to a normal variation in the quantity being that would be expected by a person skilled in the art to perform the measurement and to implement a level of care commensurate with the purpose of the measurement and the accuracy of the measuring instrument. Unless otherwise stated, “approximately” means a variation of +/- 10% of the given value.

As used in this article, the term "activity" refers to any function a protein can perform, including catalytic reactions and binding to its ligand. For enzymes, activity can be enzymatic activity. For antibodies, activity can be the binding activity between the antibody and its antigen (i.e., binding activity). For receptors or ligands, activity can be the binding activity between the receptor and its ligand.

As used herein, the term "antibody" refers to both complete immunoglobulin molecules and fragments of immunoglobulin molecules, such as Fab, Fab', (Fab') ₂ , Fv, and SCA fragments, which are capable of binding to antigenic epitopes. These antibody fragments retain some ability to selectively bind to the antigens (e.g., peptide antigens) from which the antibody is derived. These antibody fragments can be prepared using methods well known in the art (see, for example, Harlow and Lane, ibid.), and these antibodies are further described below. Antibodies can be used to prepare quantities of antigens by immunoaffinity chromatography. Various other uses of such antibodies include diagnosing and/or staging diseases (e.g., tumorigenesis) and for therapeutic applications to treat diseases such as, for example, tumorigenesis, autoimmune diseases, AIDS, cardiovascular diseases, infections, etc. Chimeric, human-like, humanized, or fully human antibodies are particularly useful for administration to human patients.

Fab fragments consist of monovalent antigen-binding fragments of antibody molecules and can be generated by digesting intact antibody molecules with an enzyme such as papain to produce fragments consisting of intact light chains and partial heavy chains.

The Fab' fragment of an antibody molecule can be obtained by treating the intact antibody molecule with pepsin, followed by reduction to produce a molecule consisting of a complete light chain and a portion of the heavy chain. Treating each antibody molecule in this manner yields two Fab' fragments.

The (Fab') ₂ fragment of an antibody can be obtained by treating the intact antibody molecule with pepsinase without subsequent reduction. The (Fab') ₂ fragment is a dimer of two Fab' fragments linked together by two disulfide bonds.

Fv fragments are defined as genetically engineered fragments containing both light and heavy chain variable regions, represented as two strands.

Single-chain antibodies (“SCA” or scFv) are genetically engineered single-chain molecules containing variable regions on both the light and heavy chains, linked by suitable flexible peptide linkers, and may include additional amino acid sequences at the amino- and/or carboxyl-termini. For example, single-chain antibodies may include tether segments for linking polynucleotides. Functional single-chain antibodies typically contain a sufficient portion of the light chain variable region and a sufficient region of the heavy chain variable region to retain the property of binding to specific target molecules or epitopes of the full-length antibody.

The term "antibody-dependent cell-mediated cytotoxicity" or "ADCC" refers to a cytotoxic form in which secreted immunoglobulins bind to Fc receptors (FcRs) present on certain cytotoxic cells (e.g., natural killer (NK) cells, neutrophils, and macrophages). This enables these cytotoxic effector cells to specifically bind to target cells carrying antigens and subsequently kill the target cells with cytotoxins. Ligand-specific, high-affinity IgG antibodies directed at the target cell surface stimulate the cytotoxic cells and are required for this killing. The lysis of target cells is extracellular, requiring direct cell-to-cell contact and not involving complement.

The ability of any specific antibody to induce target cell lysis via ADCC can be tested. To assess ADCC activity, an antibody of interest is added to target cells, demonstrating the combination of the target ligand with immune effector cells. The target ligand can be activated by the antigen-antibody complex, leading to cell lysis of the target cells. Cell lysis is typically detected by releasing a tag (e.g., a radioactive substrate, a fluorescent dye, or a native intracellular protein) from the lysed cells. Useful effector cells for this test include peripheral blood mononuclear cells (PBMCs) and natural killer (NK) cells. Specific examples of in vitro ADCC assays are described in Bruggemann et al., 1987, J Exp Med, vol. 166, page 1351; Wilkinson et al., 2001, J Immunol. Methods, vol. 258, page 183; Patel et al., 1995, J Immunol. Methods, vol. 184, page 29. Alternatively, the ADCC activity of the antibody of interest can be evaluated in vivo, for example in animal models, as disclosed in Clynes et al., 1998, PNAS USA, vol. 95, p. 652.

As used herein, the term "antigen" or "Ag" is defined as a molecule capable of triggering an immune response. This immune response may involve antibody production, or activation of specific immune-active cells, or both. Those skilled in the art will understand that any macromolecule (including virtually all proteins or peptides) can be used as an antigen. It will be apparent that antigens can be generated, synthesized, or derived from biological samples. Such biological samples may include, but are not limited to, tissue samples, tumor samples, cells, or biological fluids.

As used in this article, the term "antisense RNA" refers to an RNA molecule capable of forming a double helix with a second RNA molecule through complementarity or partial complementarity. Antisense RNA molecules can be complementary to either the translated or untranslated regions of the second RNA molecule. Antisense RNA does not need to be perfectly complementary to the second RNA molecule. Antisense RNA may or may not be the same length as the second RNA molecule; antisense RNA molecules can be longer or shorter than the second RNA molecule. If the second RNA molecule is mRNA, the binding of antisense RNA will completely or partially prevent the mRNA from being translated into a functional protein product.

The use of the terms “biosimilar” or “follow-up biologic” is consistent with the working definition issued by the U.S. Food and Drug Administration (FDA), which defines a biosimilar as “very similar” to a reference product (despite minor clinical differences in inactive components). In practice, there are no clinically meaningful differences between the reference product and the biosimilar product in terms of safety, purity, and potency (Public Health Service (PHS) § 262). A biosimilar can also be a biosimilar that meets one or more of the guidelines adopted by the European Medicines Agency’s Committee on Medicinal Products for Human Use (CHMP) on 30 May 2012 and published by the European Union as “Guidelines for Biosimilars Containing Monoclonal Antibodies – Non-clinical and Clinical Issues” (document cited as EMA/CHMP/BMWP/403543/2010). For example, an “antibody biosimilar” refers to a follow-up version of an innovator antibody (reference antibody) typically manufactured by a different company. Differences between antibody biosimilars and reference antibodies can include post-translational modifications, such as by attaching other biochemical groups like phosphate esters, various lipids, and carbohydrates to the antibody; by post-translational proteolytic cleavage; by altering the chemical properties of amino acids (e.g., formylation); or by many other mechanisms. Other post-translational modifications may result from manufacturing process operations—for example, glycosylation can occur when the product is exposed to reducing sugars. In some cases, storage conditions may allow certain degradation pathways to occur, such as oxidation, deamidation, or aggregation. Because all these product-related variants can be included in antibody biosimilars...

The terms “cancer” and “cancerous” refer to or describe a physiological condition in mammals typically characterized by unregulated cell growth/proliferation. A “tumor” contains one or more cancer cells. Examples of cancer include, but are not limited to, carcinoma, lymphoma, blastoma, sarcoma, and leukemia or lymphoma. More specific examples of such cancers include squamous cell carcinoma (e.g., epithelial squamous cell carcinoma), lung cancer (including small cell lung cancer, non-small cell lung cancer (“NSCLC”), lung adenocarcinoma, and lung squamous cell carcinoma), peritoneal cancer, hepatocellular carcinoma, gastric cancer or stomach cancer (including gastrointestinal cancer), pancreatic cancer, glioblastoma, cervical cancer, ovarian cancer, liver cancer, bladder cancer, hepatoma, breast cancer, colon cancer, rectal cancer, colorectal cancer, endometrial cancer or uterine cancer, salivary gland cancer, kidney cancer or renal cancer, prostate cancer, vulvar cancer, thyroid cancer, liver cancer, anal cancer, penile cancer, and head and neck cancer.

As used herein, the term "chimeric antigen receptor" or "CAR" refers to an engineered receptor that specifically transplants an antigen onto cytotoxic cells, such as T cells, NK cells, and macrophages. The CAR of this invention may include at least one antigen-specific targeting region (ASTR), an extracellular septum domain (ESD), a transmembrane domain (TM), one or more co-stimulatory domains (CSD), and an intracellular signaling domain (ISD). In some embodiments, the ESD and/or CSD are optional. In one embodiment, the ASTR is bispecific and can recognize two different antigens or epitopes. After the ASTR specifically binds to the target antigen, the ISD activates intracellular signaling in the cytotoxic cell. For example, depending on the antigen-binding properties of the CAR, the ISD can redirect T cell specificity and cytotoxicity to the selected target in a non-MHC-restricted manner. Non-MHC-restricted antigen recognition enables CAR-expressing cytotoxic cells to recognize antigens without relying on antigen processing, thereby bypassing major mechanisms of tumor escape. Moreover, when expressed in T cells, the CAR advantageously does not dimerize with the α and β chains of the endogenous T cell receptor (TCR).

The term "conditionally active polypeptide" refers to a variant or mutant of a parent polypeptide that is more active than the parent polypeptide under at least one condition and less active than the parent polypeptide under a second condition, or the term "conditionally active polypeptide" refers to a variant or mutant of a parent polypeptide where the activity of the variant or mutant polypeptide under the first condition is at least 1.3 times that under the second condition. The conditionally active polypeptide may exhibit activity at one or more selected sites in the body and/or enhanced or reduced activity at another site in the body. For example, in one aspect, an evolved conditionally active biological protein is almost inactive at body temperature but active at lower temperatures. Conditionally active polypeptides include conditionally active proteins, protein fragments, antibodies, antibody fragments, enzymes, enzyme fragments, receptors and receptor fragments, cytokines and fragments thereof, hormones and fragments thereof, ligands and fragments thereof, regulatory proteins and fragments thereof, growth factors and fragments thereof, and proteins including stress proteins, fornix-related proteins, neuronal proteins, digestive tract proteins, growth factors, mitochondrial proteins, cytoplasmic proteins, animal proteins, structural proteins, plant proteins, and fragments of any of these proteins. Each conditionally active polypeptide described herein is preferably a conditionally active biological polypeptide.

As used herein, the term "cytokine" or "cytokines" refers to a large class of biomolecules that affect cells of the immune system. This definition is intended to include, but is not limited to, those biomolecules that act locally through blood circulation or at locations other than secretion sites to control or modulate an individual's immune response. Exemplary cytokines include, but are not limited to, interferon-α (IFN-α), interferon-β (IFN-β), and interferon-γ (IFN-γ), interleukins (e.g., IL-1 through IL-29, particularly IL-2, IL-5, IL-6, IL-7, IL-10, IL-12, IL-15, and IL-18), tumor necrosis factors (e.g., TNF-α and TNF-β), erythropoietin (M-CSF), MIP3a, monocyte chemoattractant protein (MCP)-1, intracellular adhesion molecule (ICAM), macrophage colony-stimulating factor (M-CSF), granulocyte colony-stimulating factor (G-CSF), and granulocyte-macrophage colony-stimulating factor (GM-CSF).

As used herein, the term "electrolyte" is used to define a charged mineral in blood or other body fluids. For example, in one aspect, normal physiological conditions and abnormal conditions can be different values of "electrolyte concentration." Exemplary electrolytes include, but are not limited to, ionized calcium, sodium, potassium, magnesium, chloride, citrate, lactate, bicarbonate, and phosphate.

The term "full-length antibody" refers to an antibody that contains an antigen-binding variable region ( _VH or _VL ) and light chain constant domains (CL) and heavy chain constant domains CHI, CH2, and CH3. The constant domains can be native sequence constant domains (e.g., human native sequence constant domains) or their amino acid sequence variants. Full-length antibodies can be classified into different "classes" based on the amino acid sequence of their heavy chain constant domains. There are five main types of full-length antibodies: IgA, IgD, IgE, IgG, and IgM, some of which can be further divided into "subclasses" (isotypes), such as IgG1, IgG2, IgG3, IgG4, IgA, and IgA2. The heavy chain constant domains corresponding to different antibody classes are called α, δ, ε, γ, and μ, respectively.

As used in this article, the term "growth factor" refers to polypeptide molecules that enable cell differentiation. Examples of growth factors include, but are not limited to, epidermal growth factor (EGF), transforming growth factor-α (TGFα), transforming growth factor-β (TGF-β), human endothelial growth factor (ECGF), granulocyte-macrophage colony-stimulating factor (GM-CSF), bone morphogenetic protein (BMP), nerve growth factor (NGF), vascular endothelial growth factor (NEGF), fibroblast growth factor (FGF), insulin-like growth factor (IGF), cartilage-derived morphogenetic protein (CDMP), and platelet-derived growth factor (PDGF).

As used herein, the term "hormone" refers to substances commonly identified as mediators, which are typically released by cells or glands in one part of an organism to act as messengers to the rest of the organism. Exemplary hormones include endocrine hormones released directly into the bloodstream and exocrine hormones (or pheromones) secreted directly into or from ducts, either flowing into the bloodstream or diffusing from cell to cell via a process known as paracrine signaling. Vertebrate hormones can be classified into three chemical classes: peptide hormones, lipid and phospholipid-derived hormones, and monoamines. Peptide hormones consist of polypeptide chains. Examples of peptide hormones include insulin and growth hormone. Lipid and phospholipid-derived hormones are derived from lipids (such as linoleic acid and arachidonic acid) and phospholipids. The main category is steroid hormones derived from cholesterol and arachidonic acid. Examples of steroid hormones are testosterone and cortisol. Monoamines are derived from aromatic amino acids (such as phenylalanine, tyrosine, and tryptophan) through the action of aromatic amino acid decarboxylases. Examples of monoamines are thyroxine and adrenaline.

As used herein, the term "immunomodulator" refers to an agent whose effect on the immune system results in an immediate or delayed enhancement or reduction of the activity of at least one pathway involved in the immune response. This response can occur naturally or be artificially triggered as part of the innate or adaptive immune system, or both. Examples of immunomodulators include cytokines, stem cell growth factors, lymphotoxins such as tumor necrosis factor (TNF), and hematopoietic factors such as interleukins (e.g., interleukin-1 (IL-1), IL-2, IL-3, IL-6, IL-10, IL-12, IL-18, and IL-21), colony-stimulating factors (e.g., granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF)), interferons (e.g., interferon-α, interferon-β, and interferon-γ), stem cell growth factors known as "S1 factors," erythropoietin, and thrombopoietin. Examples of suitable immunomodulatory agents include IL-2, IL-6, IL-10, IL-12, IL-18, IL-21, interferon, and TNF (e.g., TNF-α).

"Individual" or "object" is a mammal. Mammals include, but are not limited to, domesticated animals (e.g., cows, sheep, cats, dogs, and horses), primates (e.g., humans and non-human primates such as monkeys), rabbits, and rodents (e.g., mice and rats).

As used herein, the term "library" refers to a collection of proteins in a single library. Libraries are preferably generated using DNA recombination techniques. For example, a collection of cDNA or any other protein-coding DNA can be inserted into an expression vector to generate a protein library. Alternatively, a collection of cDNA or protein-coding DNA can be inserted into a phage genome to generate a phage display library of wild-type proteins. Collections of cDNA can be generated from selected cell populations or tissue samples, for example, using the methods disclosed by Sambrook et al. (Molecular Cloning, Cold Spring Harbor Laboratory Press, 1989). Collections of cDNA generated from selected cell types are also commercially available from suppliers such as [Provider Name]. Wild-type protein libraries used herein are not collections of biological samples.

As used herein, the term "ligand" refers to a molecule that is recognized by a specific receptor and binds specifically to the receptor at one or more binding sites. Examples of ligands include, but are not limited to, agonists and antagonists of cell membrane receptors, toxins and venoms, viral epitopes, hormones, hormone receptor peptides, enzymes, enzyme substrates, cofactors, drugs (e.g., opioids, steroids, etc.), lectins, sugars, polynucleotides, nucleic acids, oligosaccharides, proteins, and monoclonal antibodies. Typically, a ligand comprises two structural parts: a first part that participates in the binding of the ligand to its receptor and a second part that does not participate in such binding.

As used herein, “receptor” refers to a molecule that has an affinity for a given ligand. Receptors can be naturally occurring or synthetically produced molecules. Receptors can be used in their unchanged state or as aggregates formed with other substances. Receptors can be directly or covalently or non-covalently linked to a binding unit via a specific binding substance. Examples of receptors include, but are not limited to, antibodies (including monoclonal antibodies capable of antiserum reactions with specific antigenic determinants, such as those located on viruses, cells, or other materials), cell membrane receptors, complexes of carbohydrates and glycoproteins, enzyme and hormone receptors. The binding of a ligand to its receptor indicates that the ligand and its receptor molecule combine through specific molecular recognition to form a complex, which can be detected by various ligand-receptor binding assays known to those skilled in the art.

As used in this article, the terms "microRNA" or "miRNA" refer to unprocessed or processed RNA transcripts derived from miRNA genes. Unprocessed microRNA gene transcripts typically contain RNA transcripts of approximately 70–100 nucleotides in length. Transcribed microRNAs can be digested with RNases (e.g., Dicer, Argonaut, or RNase III) into active 19–25 nucleotide RNA molecules. These active 19–25 nucleotide RNA molecules are also referred to as "processed" microRNA gene transcripts or "mature" microRNAs.

As used herein, the term "multispecific antibody" refers to an antibody that has binding specificity to at least two different epitopes. An exemplary multispecific antibody may bind to both BBB-R and the brain antigen. Multispecific antibodies can be prepared as full-length antibodies or antibody fragments (e.g., F(ab') ₂ bispecific antibodies). Engineered antibodies having two, three, or more (e.g., four) functional antigen-binding sites are also considered (see, for example, US2002/0004587A1).

As used herein, the term "nanoparticle" refers to a microscopic particle whose size is measured in nanometers (nm), with a maximum linear size less than about 1000 nm, or less than about 500 nm, or less than about 200 nm, or less than about 100 nm, or less than about 50 nm. As used herein, linear size refers to the distance between any two points on a nanoparticle, measured as a straight line. The nanoparticles of the present invention may be irregular, elliptical, spindle-shaped, rod-shaped, disc-shaped, flat, cylindrical, red blood cell-like, spherical, or substantially spherical in shape, provided that their shape and size allow for binding interactions. The nanoparticles of the present invention are preferably made of biocompatible materials (polymers or lipids).

When applied to objects, the term "naturally occurring" as used herein refers to the fact that an object can exist in nature. For example, polypeptide or polynucleotide sequences present in organisms (including viruses) are naturally occurring, meaning they can be isolated from natural sources and have not been intentionally modified in a laboratory. A polypeptide cleaved from a larger polypeptide is not a naturally occurring polypeptide because the terminal groups of the cleaved polypeptide will differ in the cleaved form from those in the larger, naturally occurring polypeptide, as these terminal groups will no longer bind to adjacent polypeptides. In general, the term "naturally occurring" refers to objects present in non-pathological (non-pathological) individuals, such as those that are ubiquitous within a species.

As used herein, the terms "parental polypeptide" and "parental protein" refer to polypeptides or proteins that can be evolved using the methods of the present invention to produce conditionally active polypeptides or proteins. Parental polypeptides or proteins can be wild-type proteins, including those that are not naturally occurring. For example, therapeutic polypeptides or proteins, or mutant polypeptides or variant polypeptides or proteins, can be used as parental polypeptides or proteins. Examples of parental polypeptides and proteins include antibodies, antibody fragments, enzymes, enzyme fragments, cytokines and fragments thereof, hormones and fragments thereof, ligands and fragments thereof, receptors and fragments thereof, regulatory proteins and fragments thereof, and growth factors and fragments thereof.

The term "pH-dependent" as used in this article refers to peptides that have different properties or activities at different pH values.

As used herein, the term "peptide" refers to a polymer in which the monomers are amino acids linked together by peptide or disulfide bonds. A "peptide" can be a full-length naturally occurring amino acid chain or a fragment, mutant, or variant thereof, such as a selected region of the amino acid chain of interest in binding interactions. A peptide can also be a synthetic amino acid chain, or a combination of a naturally occurring amino acid chain or a fragment thereof with a synthetic amino acid chain. A fragment refers to an amino acid sequence that is part of a full-length protein, typically of about 8 amino acids to about 500 amino acids, preferably about 8 amino acids to about 300 amino acids, more preferably about 8 amino acids to about 200 amino acids, and even more preferably about 10 amino acids to about 50 or 100 amino acids. Furthermore, amino acids other than naturally occurring amino acids (e.g., β-alanine, phenylglycine, and homoarginine) can be included in the peptide. Non-genetically encoded amino acids that are commonly encountered can also be included in the peptide. The amino acid can be a D-optical isomer or an L-optical isomer. D-isomers are preferred for specific situations further described below. In addition, other peptide mimics are also useful, for example, in linker sequences of peptides (see Spatola, 1983, in Chemistry and Biochemistry of Amino Acids. Peptides and Proteins, Weinstein, ed., Marcel Dekker, New York, p. 267). Generally, the term "protein" is not intended to express any significant difference from the term "peptide," except that it includes structures comprising two or more polypeptide chains held together by covalent or non-covalent bonds.

As used herein, the term “recombinant antibody” refers to an antibody (e.g., chimeric, humanized, or human antibody or its antigen-binding fragment) expressed by a host cell containing nucleic acid encoding the antibody. Examples of “host cells” used to prepare recombinant antibodies include: (1) mammalian cells, such as Chinese hamster ovary (CHO), COS, myeloma cells (including Y0 and NSO cells), young hamster kidney (BHK), HeLa and Vero cells; (2) insect cells, such as sf9, sf21 and Tn5; (3) plant cells, such as plants belonging to the genus *Nicotiana* (e.g., tobacco); (4) yeast cells, such as yeast cells belonging to the genus *Saccharomyces* (e.g., *Saccharomyces cerevisiae*) or yeast cells belonging to the genus *Aspergillus* (e.g., *Aspergillus niger*); (5) bacterial cells, such as *Escherichia coli* cells or *Bacillus subtilis* cells, etc.

As used in this article, the term "regulatory protein" refers to any protein that: enhances or reduces the activity of another polypeptide or RNA molecule; increases or decreases the abundance of another polypeptide or RNA molecule; alters the interaction between another polypeptide or RNA molecule and other polypeptides, DNA or RNA molecules, or any other binding substrate; and/or alters the cellular location of another polypeptide or RNA molecule. When regulatory proteins increase or decrease the transcription rate of a gene, they are often referred to as transcription factors that affect the promoter or enhancer regions of the gene. Examples of transcription factors include mammalian transcription factors such as NFkB, NF1, cyclic AMP response element binding protein (CREB), MyoD1, homeobox transcription factors, Sp1, oncogenes and jun, Mep-1, GATA-1, Isl-1, LFB1, NFAT, Pit-1, OCA-B, Oct-1 and Oct-2, yeast A/α, cErb-A, myc, mad and max, p53, mdml, and other factors listed in Latchman, 1998, Eukaryotic Transcription Factors, 3rd Ed., Academic Press: New York. Fusion protein derivatives of these or other transcription factors may also be used, wherein the DNA-binding motif of the fusion protein (providing at least binding specificity) is fused to a small molecule regulator binding site.

As used in this article, the term "small interfering RNA" or "siRNA" refers to RNA or RNA-like molecules that can interact with and cause disruption of mRNA molecules with sequence homology to the siRNA (Elbashir et al., Genes Dev, vol. 15, pp. 188-200, 2001). It is believed that siRNA can be incorporated into a ribonucleoprotein complex called the RNA-induced silencing complex (RISC). RISC uses the siRNA sequence to identify mRNA molecules that are at least partially complementary to the incorporated siRNA strand, and then cleaves these target mRNA molecules or inhibits their translation. Typical siRNAs are double-stranded nucleic acid molecules, each strand having approximately 19 to approximately 28 nucleotides (i.e., approximately 19, 20, 21, 22, 23, 24, 25, 26, 27, or 28 nucleotides). siRNAs can also be single-stranded RNAs, although less effective than double-stranded siRNAs. Single-stranded siRNAs have a length of approximately 19 to approximately 49 nucleotides. Single-stranded siRNAs possess a 5' phosphate group or are phosphorylated in situ or in vivo at the 5' position. Single-stranded siRNAs can be chemically synthesized, synthesized via in vitro transcription, or expressed endogenously from expression vectors or expression cassettes. The 5' phosphate group can be added by a kinase or can be the result of RNA nuclease cleavage.

The term "small molecule" generally refers to molecules or ions with a molecular weight of less than 900 a.m.u., more preferably less than 500 a.m.u., more preferably less than 200 a.m.u., or even more preferably less than 100 a.m.u. In the testing and environment of this invention, small molecules can generally exist as a mixture of molecules and deprotonated ions of molecules, depending primarily on the pH of the testing or environment.

As used herein, the term "therapeutic protein" refers to any protein and/or polypeptide administered to mammals, for example by researchers or clinicians, to elicit a biological or medical response in a target tissue, system, animal, or human. Therapeutic proteins can elicit more than one biological or medical response. Examples of therapeutic proteins include antibodies, enzymes, hormones, cytokines, regulatory proteins, and fragments thereof.

As used herein, the term "therapeuticly effective amount" means any amount that, compared to a corresponding individual who did not receive that amount, results in, but is not limited to, the cure, prevention, or improvement of a disease, symptom, or side effect, or a reduction in the rate of progression of a disease or symptom. Within its scope, the term also includes amounts that effectively enhance normal physiological function and amounts that effectively induce physiological function in the patient that enhances or contributes to the therapeutic effect of a second agent.

The term "tumor microenvironment" as used in this article refers to the microenvironment within a solid tumor that supports the growth and metastasis of tumor cells, as well as the microenvironment surrounding the solid tumor. The tumor microenvironment includes surrounding blood vessels, immune cells, fibroblasts, other cells, soluble factors, signaling molecules, extracellular matrix, and mechanotransmitters that promote tumorigenic transformation, support tumor growth and invasion, protect the tumor from host immune attack, promote therapy resistance, and provide a microenvironment (niches) for dormant metastases. Tumors and their surrounding microenvironment are closely related and constantly interact. Tumors can influence their microenvironment by releasing extracellular signals, promoting tumor angiogenesis, and inducing peripheral immune tolerance, while immune cells within the microenvironment can influence the growth and evolution of cancer cells. See Swarts et al., “Tumor Microenvironment Complexity: Emerging Roles in Cancer Therapy,” Cancer Res, vol., 72, pp. 2473-2480, 2012; Weber et al., “Thetumor microenvironment,” Surgical Oncology, vol. 21, pp. 172-1 77, 2012; Blagosklonny, "Antiangiogenic therapy and tumor progression," Cancer Cell, vol.5, pp.13-17, 2004; Siemann, "Tumor microenvironment," Wiley, 2010; and Bagley, "The tumor microenvironment," Springer, 2010.

As used in this article, the term "wild-type" refers to a polynucleotide that does not contain any mutations. "Wild-type protein," "wild-type protein," "wild-type biological protein," or simply "wild-type biological protein" can refer to a protein that can be isolated from nature and exhibits a certain level of activity under natural conditions, including amino acid sequences found under natural conditions. The terms "parental molecule" and "target protein" also include wild-type proteins.

Detailed Implementation

A. pH-dependent conditionally active peptides

In one aspect, the present invention relates to conditionally active polypeptides having pH-dependent activity in the presence of a substance having a pKa within 0.5, 1, 2, or 4 units of the pH at which the activity is desired. In another aspect, the present invention relates to conditionally active polypeptides having pH-dependent activity in the presence of a substance having a pKa of about 4 to about 10, or about 4.5 to about 9.5, or about 5 to about 9, or about 5.5 to about 8, or about 6.0 to about 7.0. In yet another aspect, the present invention relates to conditionally active polypeptides having pH-dependent activity in the presence of a substance selected from histidine, histamine, hydrogenated adenosine diphosphate, hydrogenated adenosine triphosphate, citrate, bicarbonate, acetate, lactate, hydrosulfide, hydrogen sulfide, ammonium, dihydrogen phosphate, and any combination thereof.

Substances present in the test medium that significantly affect the activity of conditionally active peptides tend to be those with at least two ionization states: an uncharged or lightly charged state and a charged or heavily charged state. Consequently, the pKa of a substance affecting the activity of a conditionally active peptide can determine the extent of that substance's influence on the specific activity of the peptide and/or its effect at a specific pH.

pH-dependent conditionally active peptides exhibit higher activity at a first pH than at different second pH values, both activities being tested in the presence of one or more of the aforementioned substances. To determine the pH dependence of the conditionally active peptide, the same activity of the peptide was tested at two different pH values in the same test medium.

In the same test medium, the ratio of activity at a first pH to the same activity at a second pH can be termed the selectivity of a pH-dependent conditionally active peptide. The selectivity of a pH-dependent conditionally active peptide is at least about 1.3, or at least about 1.5, or at least about 1.7, or at least about 2.0, or at least about 3.0, or at least about 4.0, or at least about 6.0, or at least about 8.0, or at least about 10.0, or at least about 20.0, or at least about 40.0, or at least about 60.0, or at least about 100.0.

It has been observed that pH-dependent conditionally active peptides contain an increased number (or proportion) of charged amino acid residues compared to the parent peptides from which they are derived. There are three positively charged amino acid residues: lysine, arginine, and histidine; and two negatively charged amino acid residues: aspartic acid and glutamic acid. These charged amino acid residues are over-represented in pH-dependent conditionally active peptides compared to their parent peptides. As a result, pH-dependent conditionally active peptides are more likely to interact with charged substances in the test medium due to the increased number of charged amino acid residues. This, in turn, affects the activity of the conditionally active peptide.

It was also observed that, in the test culture medium, pH-dependent conditionally active peptides typically exhibited different activities in the presence of different substances. Substances possessing at least two ionization states (uncharged or poorly charged and charged or heavily charged) can dissociate to a greater extent at a specific pH (depending on the pKa value), thereby increasing the likelihood of interaction with charged amino acid residues present in the conditionally active peptide. This factor can be used to enhance the selectivity and/or pH-dependent activity of conditionally active peptides.

The charge properties of a conditionally active peptide can be a factor in determining suitable substances that influence its activity. In some embodiments, the conditionally active peptide may have more positively charged amino acid residues—lysine, arginine, and histidine—compared to the parent peptide. Therefore, the conditionally active peptide can be selected to interact at a desired level with a specific substance present in an environment where the desired activity is desired, and/or to interact at a desired level with a specific substance present in an environment where the desired activity is reduced. Similarly, the conditionally active peptide may have more negatively charged amino acid residues—aspartic acid and glutamic acid—compared to the parent peptide.

The position of charged amino acid residues on pH-dependent conditionally active peptides can also affect their activity. For example, the proximity of charged amino acid residues to the binding site of the conditionally active peptide can influence the peptide's activity.

In some embodiments, the interaction between charged environmental substances and conditionally active peptides can block or inhibit the activity of pH-dependent conditionally active peptides. For example, charged amino acids interacting with charged environmental substances can exhibit an allosteric effect on the binding site of conditionally active peptides.

In other implementations, it is possible that the interaction between charged environmental substances and conditionally active peptides can form salt bridges between different parts of the peptide, particularly charged or polarized parts. It is known that the formation of salt bridges stabilizes the peptide structure (Donald, et al., "Salt Bridges: Geometrically Specific, Designable Interactions," Proteins, 79(3):898–915, 2011; Hendsch, et al., "Do salt bridges stabilize proteins? A continuous electrostatic analysis," Protein Science, 3:211-226, 1994). Salt bridges can stabilize or immobilize protein structures, which typically undergo continuous, minute structural changes known as "breathing" (Parak, "Proteins in action: the physics of structural fluctuations and formational changes," Curr Opin Struct Biol., 13(5):552-557, 2003). This structural "breathing" is crucial for protein function and binding to its partner, as structural fluctuations allow conditionally active proteins to effectively recognize and bind to their partner (Karplus, et al., "Molecular dynamics and protein functions," PNAS, vol. 102, pp. 6679-6685, 2015). By forming salt bridges, binding sites on conditionally active peptides, particularly binding pockets, are less likely to come into contact with their partner, possibly because the salt bridge directly blocks the partner from entering the binding site. Even if the salt bridge is far from the binding site, allosteric effects can alter the conformation of the binding site to inhibit binding. Therefore, after the structure of a salt-bridged (fixed) active peptide is established, the peptide may lose its activity in binding to its partner, resulting in reduced activity.

A known example of how salt bridges stabilize peptides and their structures is hemoglobin. Structural and chemical studies have shown that at least two sets of chemical groups are responsible for forming salt bridges: the amino termini of histidine β146 and α122, and the side chain, with pKa values close to pH 7. In deoxyhemoglobin, the terminal carboxyl group of β146 forms a salt bridge with a lysine residue in the α subunit of another αβ dimer. If the imidazole group of the histidine residue is protonated, this interaction locks the side chain of histidine β146 in a position where it can form a salt bridge with a negatively charged aspartic acid 94 in the same chain (Figure 6). At high pH, the side chain of histidine β146 is not protonated and does not form a salt bridge. However, as pH decreases, the side chain of histidine β146 is protonated, and a salt bridge forms between histidine β146 and aspartic acid β94. This stabilizes the quaternary structure of deoxyhemoglobin, leading to a greater tendency for oxygen release in metabolically active tissues (with lower pH). Hemoglobin exhibits pH-dependent oxygen-binding activity at low pH, which decreases due to the formation of the salt bridge. Conversely, at high pH, oxygen-binding activity is enhanced due to the absence of the salt bridge.

Similarly, small molecules such as bicarbonates can reduce the binding activity of conditionally active peptides to their counterparts by forming salt bridges in the peptide. For example, at pH levels below that of bicarbonates with a pKa of 6.4, the bicarbonates are protonated and thus uncharged. Uncharged bicarbonates cannot form salt bridges and therefore have little effect on the binding of conditionally active peptides to their counterparts. Therefore, conditionally active peptides exhibit high binding activity to their counterparts at low pH levels. On the other hand, at high pH levels above that of bicarbonates with a pKa, the bicarbonates are ionized by losing protons, becoming negatively charged. The negatively charged bicarbonates will form salt bridges between the positively charged or polarized portions of the conditionally active peptide to stabilize its structure. This will block or reduce the binding of the conditionally active peptide to its counterpart. Therefore, conditionally active peptides exhibit low activity at high pH levels. Thus, conditionally active peptides are conditionally active in the presence of bicarbonates and exhibit higher binding activity at low pH levels than at high pH levels.

When substances such as bicarbonate are absent in the test medium, conditionally active peptides may lose their conditional activity. This may be due to the lack of salt bridges on the conditionally active peptide to stabilize (immobilize) its structure. Therefore, the mating bodies will have similar accessibility to the binding sites on the conditionally active peptide at any pH, producing similar activity at the first and second pH.

In other embodiments, the interaction between the small molecule or ion and the conditionally active peptide can alter the peptide's structure in a way that enhances its activity. For example, structural changes can increase the binding affinity of the conditionally active peptide by altering the position, steric hindrance, or binding energy of the binding site required for binding affinity. In this case, it may be necessary to select a small molecule that binds to the conditionally active peptide at the desired pH.

It should be understood that while salt bridges (ionic bonds) are the strongest and most common way in which compounds and ions influence the activity of conditionally active peptides, other interactions between these compounds and ions and conditionally active peptides may also contribute to stabilizing (fixing) the structure of conditionally active peptides. These other interactions include hydrogen bonds, hydrophobic interactions, and van der Waals interactions.

In some embodiments, to select a suitable compound or ion, the conditionally active peptide is compared with the parent peptide from which it evolved to determine whether the conditionally active peptide has a higher proportion of negatively charged or positively charged amino acid residues. Compounds with appropriate charges at a second pH can then be selected to influence the activity of the conditionally active peptide. For example, when the conditionally active peptide has a higher proportion of positively charged amino acid residues than the parent peptide, a suitable small molecule should generally be negatively charged at the second pH to interact with the conditionally active peptide. Conversely, when the conditionally active peptide has a higher proportion of negatively charged amino acid residues than the parent peptide, a suitable small molecule should generally be positively charged at the second pH to interact with the conditionally active peptide.

In other embodiments, the activity of the conditionally active peptide is controlled by the interaction of a small molecule or ion with the target peptide, which acts as a binding partner of the conditionally active peptide. In this case, the same principles discussed above apply, except that the goal is to generate an interaction between the small molecule or ion and the target peptide. The target peptide can be, for example, an antigen of a conditionally active antibody or a ligand of a conditionally active receptor.

Suitable small molecules can be any inorganic or organic molecule that transitions from an uncharged or less-charged state at a first pH to a charged or more-charged state at a second pH. Therefore, small molecules should generally have a pKa between the first and second pH. For example, bicarbonate has a pKa of 6.4. Thus, at higher pH levels, such as pH 7.4, the negatively charged bicarbonate will bind to the charged amino acid residues in the conditionally active peptide and reduce its activity. On the other hand, at lower pH levels, such as pH 6.0, less charged bicarbonate will bind to the conditionally active peptide in a lesser amount, thus allowing the conditionally active peptide to have higher activity.

The pKa of hydrosulfide is 7.05. Therefore, at higher pH levels, such as pH 7.4, more negatively charged hydrosulfide ions will bind to the positively charged amino acid residues in the conditionally active peptide and reduce its activity. On the other hand, at lower pH levels, such as pH 6.2–6.8, fewer charged hydrogen sulfide/hydrosulfide ions will not bind to the conditionally active peptide at the same level, thus allowing the conditionally active peptide to have higher activity.

Small molecules having a pKa between the first and second pH values are preferably used in this invention. Preferred substances are selected from hydrosulfide, hydrogen sulfide, histidine, histamine, citrate, bicarbonate, acetate, and lactate. Each of these small molecules has a pKa between 6.2 and 7.0. Additionally, other small molecules, such as N-tris(hydroxymethyl)methylglycine (tricine, pKa 8.05) and N,N-dihydroxyethylglycine (bicine, pKa 8.26), can also be used. Other suitable small molecules can be found in textbooks using the principles of this application, such as CRC Handbook of Chemistry and Physics, 96th edition; Chemical Properties Handbook, McGraw-Hill Education, 1998.

The concentration of small molecules in the test medium or environment is preferably at or close to the physiological concentration of small molecules in an individual. For example, the physiological concentration of bicarbonate (in human serum) is in the range of 15 mM to 30 mM. Therefore, the concentration of bicarbonate in the test medium can be 10 mM to 40 mM, or 15 mM to 30 mM, or 20 mM to 25 mM, or about 20 mM. The physiological concentration of hydrosulfide is also very low. The concentration of hydrosulfide in the test medium can be 3 nM to 500 nM, or 5 nM to 200 nM, or 10 nM to 100 nM, or 10 nM to 50 nM.

In this invention, the concentration of the conditionally active peptide is selected and employed such that the normal physiological concentration of a specific substance in the environment will significantly affect the activity of the conditionally active peptide within the pH range of interest. Therefore, in many therapies, it may be advantageous to render the conditionally active peptide low in activity in blood or human serum at approximately pH 7.2-7.4 to allow for hematogenous delivery of the therapy, while minimizing or preventing activation of the conditionally active peptide. Consequently, for such treatments, it will be advantageous to select small molecules with pKa below pH 7.2-7.4 to ensure sufficient ionization of the small molecule at the bloodstream pH, thereby significantly influencing the activity of the conditionally active peptide. Simultaneously, the pKa of the small molecule should be equal to or higher than the pH at which the conditionally active peptide exhibits desired activity, to ensure activation of the conditionally active peptide by releasing its binding sites through protonation of the small molecule.

Small molecules preferably have low molecular weight and/or relatively small conformations to ensure maximum accessibility to small pockets on target peptides or conditionally active peptides by minimizing steric hindrance. For this reason, the molecular weight of small molecules is typically less than 900 a.m.u., or more preferably less than 500 a.m.u., or more preferably less than 200 a.m.u., or even more preferably less than 100 a.m.u. For example, as shown in Examples 13 and 14 below, hydrogen sulfide, bisulfide, and bicarbonate all have low molecular weight and small structures, providing accessibility to pockets on target peptides or conditionally active peptides.

Small molecules can exist at substantially the same concentration in the test or environment, for example, bicarbonate at approximately 20 μM. In some embodiments, small molecules can exist at different concentrations in different environments, and it may be desirable to simulate this in the test. For example, sulfide has a higher concentration in the tumor microenvironment than in human serum. Therefore, one test can simulate a tumor microenvironment with an acidic pH and a higher concentration of sulfide, while a second test can simulate human serum with a neutral or weakly alkaline pH and a lower concentration of sulfide. The acidic pH can be in the range of 6.0 to 6.8, while the neutral or weakly alkaline pH can be approximately 7.4. The higher concentration of sulfide used for the first test can be 30 μM, while the lower concentration of sulfide used for the second buffer can be 10 μM or less, or 5 μM.

In some implementations, the conditionally active peptide is pH-dependent when two or more different small molecules are present (e.g., a combination of bicarbonate and histidine).

Conditionally active peptides may lose their pH dependence in the absence of small molecules. Therefore, in the absence of small molecules, conditionally active peptides can exhibit similar activity between a first pH and a second pH.

In some embodiments, the first pH is an acidic pH, while the second pH is an alkaline or neutral pH. In other embodiments, the first pH is an alkaline pH, while the second pH is an acidic or neutral pH. For example, the first pH may be in the range of about 5.5-7.2, or about 6.0-7.0, or about 6.2-6.8. The second pH may be in the range of about 7.0-7.8, or about 7.2-7.6.

Conditionally active peptides that are more active at acidic pH and less active at alkaline or neutral pH can target acidic tumor microenvironments with pH values of approximately 5.5–7.2 or approximately 6.2–6.8.

In other embodiments, the first pH at which the pH-dependent peptide is more active can be, for example, an alkaline pH of 7.6–7.9, such as in synovial fluid (see Jebens et al., “On the viscosity and pH of synovial fluid and pH of blood,” Journal of Bone and Joint Surgery, vol. 41B, pp. 388–400, 1959). The second pH can be the blood pH of about 7.2–7.6, at which the conditionally active peptide has lower activity. These conditionally active peptides may be suitable for targeting joint diseases, particularly joint inflammation.

In other embodiments, conditionally active peptides can be designed to target the brain. A pH difference exists across the blood-brain barrier, with the brain side having a pH 0.2 pH units lower than the blood side. Therefore, the first brain pH at which the conditionally active peptide is more active could be approximately 7.0–7.2 (brain pH), while the second pH could be approximately 7.4 (blood pH).

Conditionally active peptides can be enzymes, cytokines, receptors, especially cell receptors, regulatory peptides, soluble peptides, antibodies, or hormones.

Conditionally active peptides can be fragments of parental peptides. For example, conditionally active peptides can be antibody fragments, single-chain antibodies, enzyme fragments, receptor fragments, cytokine fragments, or hormone fragments. Antibody fragments can be the Fc fragments of antibodies.

Fc fragments can be used as parent peptides to produce conditionally active Fc fragments, which exhibit higher complement-binding activity at a first pH than at a second pH. The binding of the Fc fragment to complement can be used to provide antibody-dependent, cell-mediated cytotoxicity. The first pH can be acidic, ranging from 5.5 to 7.2 or 6.2 to 6.8, such as the pH in the tumor microenvironment, while the second pH ranges from 7.2 to 7.6. The first pH differs from the pH typically found in lysosomes, which is around 4.0. Furthermore, lysosomes are the sites where Fc fragments (like any other peptide) are targeted for degradation. Lysosomes do not contain complement and do not provide cell-mediated cytotoxicity.

Conditionally active peptides may have two functional domains, at least one of which, preferably both, has pH-dependent activity. These two functional domains can be evolved simultaneously and selected to identify the two functional domains in the same mutant peptide. Alternatively, the two functional domains can be evolved independently and selected to identify pH-dependent activity separately. If the two functional domains are not in the same mutant peptide, they can be fused into a chimeric peptide having two separately identified functional domains.

In one respect, in the presence of factors such as proteins, conditionally active peptides exhibit enhanced activity at a first pH compared to their parent peptides, and decreased activity at a second pH compared to their parent peptides. Proteins can be present in blood, human serum, or in the body's microenvironment (such as the tumor microenvironment, inflamed areas, etc.). A suitable protein could be albumin, particularly mammalian albumin, such as bovine albumin or human albumin.

In one respect, proteins such as albumin are present in test solutions used to screen and select conditionally active peptides from mutant peptides produced by evolutionary steps. In another respect, test solutions containing proteins such as albumin are also used to test the activity of the selected conditionally active peptides under the same or different conditions.

B. Engineering of conditionally active peptides

Conditionally active peptides can be engineered using one or more protein engineering techniques described herein. Non-limiting examples of protein engineering techniques include conjugating conditionally active peptides to nucleic acids, conjugating conditionally active peptides to nanoparticles, engineering conditionally active peptides in chimeric antigen receptors, and engineering masked conditionally active peptides.

The conditionally active peptides of this invention can be conjugated to nucleic acid molecules, such as DNA or RNA molecules, via linkers. Conditionally active peptides can facilitate the delivery of nucleic acid molecules to a target site in an individual that exhibits a condition at which the peptide is more active than at other sites where the condition is absent. For example, a conditionally active peptide can be a conditionally active antibody that exhibits higher binding activity to its antigen under conditions in the tumor microenvironment than under conditions at other sites, such as in human serum. This effect can be used to deliver nucleic acid molecules to the tumor microenvironment by conjugating nucleic acid molecules to conditionally active peptides and applying the conjugate to an individual.

In some implementations, nucleic acid molecules can be agents used to regulate gene expression at target sites. Abnormal gene expression is associated with many diseases. Therefore, correcting abnormal gene expression could help control or even cure these diseases. For example, abnormal gene expression is characteristic of most cancer cells, with some genes exhibiting elevated expression levels in cancer cells, such as many oncogenes (e.g., epidermal growth factor receptor 2 (HER2) is overexpressed in breast cancer cells). Selectively inhibiting constitutively elevated expression of oncogenes provides an opportunity to suppress cancer cell proliferation.

Nucleic acid molecules that can suppress gene expression include antisense RNA, small interfering RNA (siRNA), microRNA, oligoDNA, and oligonucleotide mimics with an uncharged, non-chiral polyamide backbone linked to nucleobases (Pooga et al., Curr Cancer Drug Targets, 1(3):231-9, 2001; Pandey et al., Expert Opin Biol Ther., 9(8):975-89, 2009).

Antisense RNA is a short RNA molecule that can inhibit mRNA expression in a sequence-specific manner by binding to specific complementary regions of mRNA through base pairing. Antisense RNA can induce RNase H (which cleaves mRNA at the site where it binds), or it can physically block translation or other steps in mRNA processing and protein synthesis.

Small interfering RNAs (siRNAs) are typically short double-stranded RNA segments whose sequences are at least partially complementary to the mRNA sequence whose translation is to be blocked. siRNAs function through post-transcriptional gene silencing mechanisms via chromatin remodeling, inhibition of protein translation, or direct mRNA degradation, which are prevalent in eukaryotic cells (Caplen, “Gene therapy progress and prospects. Downregulating gene expression: the impact of RNA interference,” GeneTher., 11(16):1241-1248, 2004) and Bertrand et al., “Comparison of antisense oligonucleotides and siRNAs in cell culture and in vivo,” Biochem Biophys ResCommun., 296(4):1000-1004, 2002).

In particular, through RISC, siRNA can initiate a powerful cascade reaction that specifically degrades mRNA homologous to siRNA (Fire et al., "Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans," Nature 391:806–811, 1998). When siRNA is introduced into cells, it undergoes processing by Dicer's RNase III enzyme, which cleaves long siRNA into short 21-23 nucleotide double strands with symmetrical 2-3 nucleotide 3' overhangs and 5' phosphate groups and 3' hydroxyl groups (Tuschl et al., "Targeted mRNA degradation by double-stranded RNA in vivo," Genes Dev. 13:3191–3197, 1999; Hamilton and Baulcombe, "A species of small antisense RNA in posttranscriptional gene silencing in plants," Science, 286:950–952, 1999). Therefore, effective siRNAs only require a short, consecutive complementary sequence to pair with mRNA to trigger siRNA-mediated silencing (Jackson and Linsley, "Noise amid the silence: off-target effects of siRNAs?" Trends Genet., 20:521–524, 2004). Since siRNAs do not integrate into the genome, they offer greater safety than plasmids or viral vectors.

MicroRNAs (miRNAs) are a class of naturally occurring small, non-coding RNA molecules, 21-25 nucleotides in length. MiRNAs are partially complementary to the mRNA molecules in which they function. The primary function of miRNAs is to reduce gene expression through translational repression, mRNA cleavage, and deadenylateation. A central online repository for miRNA species, sequence data, annotation, and target prediction, hosted by the Sanger Institute in the UK, is called miRBase. MiRNA genes are transcribed by RNA polymerase II to produce pri-miRNAs with a 5' cap and a poly-A tail. In the nucleus, pri-miRNAs are processed by a microprocessing complex (composed of the RNase III enzyme Drosha and double-stranded RNA Pasha/DGCR8) to produce pre-miRNAs. These pre-miRNAs are exported into the cytoplasm via the nuclear transporter exp5 and the Ran-GTP complex, where Ran GTPase binds to Exp5 to form nuclear heterotrimers containing pre-miRNAs. These pre-miRNAs are then processed by the RNase III enzyme Dicer to produce mature microRNAs.

Another class of nucleic acids that can be delivered via conditionally active peptides are oligonucleotide mimics, which contain an uncharged, achiral polyamide backbone linked to nucleobases. Oligonucleotide mimics are commonly referred to as peptide nucleic acids (PNAs). More specifically, PNAs are DNA analogs in which N-(2-aminoethyl)glycine polyamide replaces the phosphate-ribose ring backbone, and a methylene-carbonyl linker connects both native and non-native nucleobases to N-(2-aminoethyl)glycine. Despite this fundamental change in backbone structure, PNAs are able to bind sequence-specifically to DNA and mRNA according to the Watson-Crick base pairing principle.

PNA binds to complementary DNA/RNA with a higher affinity than natural nucleic acids, partly due to the lack of a negative charge on its backbone, thus reducing charge-charge repulsion, and favorable geometry. The PNA-DNA/mRNA complex is highly stable in biological fluids, leading to the repression of target gene transcription and translation through specific hybridization with either DNA or mRNA. PNA is typically synthesized using well-known solid-phase peptide synthesis protocols. See Kim et al., J. Am. Chem. Soc., 115, 6477–6481, 1993; Hyrup et al., J. Am. Chem. Soc., 116, 7964–7970, 1994; Egholm et al., Nature, 365, 566–568, 1993; Dueholm et al., New J. Chem., 21, 19–31, 1997; Wittu (References: ng et al., J. Am. Chem. Soc., 118, 7049–7054, 1996; Leijon et al., Biochemistry, 33, 9820–9825, 1994; Orum et al., BioTechniques, 19, 472–480, 1995; Tomac et al., J. Am. Chem. Soc., 118, 5544–5552, 1996). Compared to DNA, which is depurinated upon treatment with strong acids and hydrolyzed in alkaline hydroxides, PNA is completely acid-stable and sufficiently stable to weak bases.

Another class of nucleic acids that can be delivered by conditionally active peptides is oligoDNA. OligoDNA is a short, single-stranded segment of DNA that, upon entering the cellular plasma, selectively inhibits the expression of genes with sequences complementary to the oligoDNA. For antisense applications, oligoDNA interacts with target mRNA or pre-mRNA to form a double strand, inhibiting its translation or processing, thereby suppressing protein biosynthesis. For antigen applications, oligoDNA must enter the cell nucleus to form a triple strand with double-stranded genomic DNA, inhibiting gene transcription, resulting in less mRNA and consequently, less protein gene product.

Another class of nucleic acids that can be delivered by conditionally active peptides are spherical nucleic acids (SNAs, see Zhang, J.A.M. Chem.Soc., 134(40):16488–16491, 2012). SNAs consist of densely functionalized and highly oriented nucleic acids covalently linked to the surface of a metallic, semiconductor, or insulating inorganic or polymeric core material. They can also be nucleate-free hollow structures composed almost entirely of nucleic acid molecules. These spherical nucleic acids can bypass the individual's natural defenses against exogenous nucleic acids. SNAs utilize the unique properties generated by their densely packed, highly oriented nucleic acid shells to achieve both protection and efficient delivery of nucleic acids. This shell generates regions of high local salt concentration, which, when combined with steric inhibition, serves to reduce nuclease activity and protect nucleic acids from enzymatic degradation. Furthermore, these spherical nucleic acids recruit clearance proteins from the natural extracellular environment to their surface that promote endocytosis.

Once inside the cytoplasm, globular nucleic acids can inhibit the expression of target genes via antisense or siRNA pathways. Therefore, globular nucleic acids offer several advantages over viral vectors and many other synthetic systems, including low toxicity, low immunogenicity, resistance to enzymatic degradation, and more persistent gene knockdown. Conditionally active peptides, particularly conditionally active antibodies, can deliver globular nucleic acids to target sites such as diseased or inflamed tissues (e.g., tumors and inflamed joints).

Conditionally active peptides can also be conjugated to nanoparticles via linkers to facilitate delivery of the nanoparticles to target sites with conditions that enhance the activity of the conditionally active peptides. Nanoparticles are known carriers of toxins, radioactive agents, or other therapeutic agents encapsulated within them.

Therapeutic agents encapsulated in nanoparticles can be proteins that can dedifferentiate tumor cells and thus potentially revert them back to normal cells (Friedmann-Morvinski and Verma, “Dedifferentiation and reprogramming: origins of cancer stem cells,” EMBO Reports, 15(3):244-253, 2014). Nanoparticles can be linked to conditionally active antibodies to selectively deliver the linked nanoparticles and encapsulated therapeutic agents to the environment where the conditionally active antibody is most active.

Several types of nanoparticles with different configurations can be used in this invention. The nanoparticles can be made from a range of biocompatible materials, including biostable polymers, biodegradable polymers, fullerenes, lipids, or combinations thereof. Biostable polymers are polymers that do not degrade in vivo. Biodegradable polymers are polymers that can be degraded after delivery to a patient. For example, when polymers are exposed to bodily fluids such as blood, they can be gradually absorbed and/or eliminated by enzymes in the body. Methods for producing nanoparticles with various degradation rates are known to those skilled in the art, see, for example, U.S. Patent Nos. 6,451,338, 6,168,804, and 6,258,378.

Exemplary nanoparticles of the present invention include liposomes, polymer vesicles, and polymer particles. A liposome is a chamber completely enclosed by a bilayer typically composed of phospholipids. Liposomes can be prepared according to standard techniques known to those skilled in the art. One technique involves suspending a suitable lipid (e.g., phosphatidylcholine) in an aqueous medium and then sonicating the mixture. Another technique involves rapidly mixing a lipid solution in ethanol-water, for example, by injecting the lipid into a stirred ethanol-water solution using a needle. In some embodiments, additionally or optionally, liposomes may also contain other amphiphilic substances, such as sphingomyelin or lipids containing polyethylene glycol (PEG).

Polymer vesicles comprise diblock or triblock copolymers modified to form a bilayer structure similar to liposomes. Depending on the length and composition of the block copolymers, polymer vesicles can be substantially more robust than liposomes. Furthermore, the ability to control the chemical properties of each block of the block copolymer allows for tuning the composition of the polymer vesicles to suit desired applications. For example, the membrane thickness of the polymer vesicle, i.e., the thickness of the bilayer, can be controlled by varying the chain length of individual blocks in the block copolymer. Adjusting the glass transition temperature of each block in the copolymer will affect the fluidity of the polymer vesicle and thus its membrane permeability. Even the release mechanism of the encapsulated reagent can be improved by altering the properties of the copolymer.

Polymer vesicles can be prepared by a method comprising the following steps: (i) dissolving the block copolymer in an organic solvent, (ii) applying the resulting solution to a container surface, and then (iii) removing the solvent, thereby leaving a film of the copolymer on the container wall. This film is then hydrated to form a polymer vesicle. Alternatively, dissolving the block copolymer in a solvent and then adding a solvent that is a weak solvent for one of the blocks of the copolymer will also produce polymer vesicles.

Various techniques can be used to encapsulate therapeutic agents in polymeric vesicles. For example, the therapeutic agent can be mixed in water and then used to rehydrate the copolymer membrane. Another example is the permeabilizing drive of the therapeutic agent into the core of a pre-formed polymeric vesicle, a process known as force loading. Yet another example is the use of dual emulsion technology, which can generate polymeric vesicles with relatively monodispersity and high loading efficiency. Dual emulsion technology involves using microfluidics to produce a dual emulsion containing water droplets surrounded by a layer of organic solvent. These droplets are then dispersed in a droplet-in-a-drop structure in a continuous aqueous phase. Block copolymers dissolve in the organic solvent and self-assemble into protopolymeric vesicles at the concentric interfaces of the dual emulsion. The final polymeric vesicle is formed after the organic solvent is completely evaporated from the shell of the protopolymeric vesicle. This technique allows for precise control over the size of the polymeric vesicles. Furthermore, the ability to maintain complete separation of the internal and external fluids throughout the process allows for very efficient encapsulation of the therapeutic agent.

In contrast to the shell structure of liposomes and polymer vesicles, polymer particles refer to solid or porous particles. Methods for attaching therapeutic agents to the surface of polymer particles or integrating bioactive agents into the structure of polymer particles are known to those skilled in the art.

Polymers that can be used to prepare the nanoparticles of this invention include, but are not limited to, poly(N-acetylglucosamine) (chitosan), chitosan, poly(3-hydroxyvalerate), poly(lactide-co-glycolic acid), poly(3-hydroxybutyrate), poly(4-hydroxybutyrate), poly(3-hydroxybutyrate-co-3-hydroxyvalerate), polyorthoesters, polyanhydrides, polyglycolic acid, polyglycolic acid, poly(L-lactic acid), poly(L-lactide), poly(D,L-lactic acid), and poly(D,L-lactide). Poly(L-lactide-co-D,L-lactide), poly(caprolactone), poly(L-lactide-co-caprolactone), poly(D,L-lactide-co-caprolactone), poly(glycolic acid-co-trimethylene carbonate), polyesteramide, poly(glycolic acid-co-trimethylene carbonate), co-polyether esters (e.g., PEO/PLA), polyphosphazenes, biomolecules (such as fibrin, fibrin glue, fibrinogen, cellulose, starch, collagen, and hyaluronic acid) Polyurethane, polysilicon, polyester, polyolefin, polyisobutylene and ethylene-α-olefin copolymers, acrylic polymers and copolymers other than polyacrylates, halogenated vinyl polymers and copolymers (such as polyvinyl chloride), polyethylene ethers (such as polyvinyl methyl ether), polyvinylidene halides (such as polyvinylidene chloride), polyacrylonitrile, polyvinyl ketone, polyvinyl aromatic compounds (such as polystyrene), polyethylene esters (such as polyvinyl acetate), acrylonitrile-styrene copolymers, acrylonitrile butadiene styrene (ABS) resins, polyamides (such as nylon 66 and polycaprolactam), including tyrosine-based polycarbonates, polyoxymethylene, polyimide, polyether, polyurethane, rayon, synthetic triacetate, cellulose, cellulose acetate, cellulose butyrate, cellulose acetate butyrate, cellophane, cellulose nitrate, cellulose propionate, cellulose ethers and carboxymethyl cellulose.

In some implementations, in addition to selectivity derived from conditionally active peptides, nanoparticles can also provide tissue selectivity through coating. For example, nanoparticles can be coated with electrostatically adsorbed poly(glutamate) peptides to alter the external composition of the core particle. Negatively charged polyglutamate peptides containing arginine-glycine-aspartic acid (RGD) ligands can increase in vitro gene delivery to endothelial cells compared to particles coated with disordered sequences containing RDG instead of RGD. These peptides consist of three parts: a poly(glutamate) segment providing a negative charge, a polyglycine linker, and a terminal sequence that changes charge and has the potential to alter the biophysical properties and tissue selectivity of the particle. The coating and the particles themselves are biodegradable via their amide and ester bonds, respectively. See Harris et al. (“Tissue-Specific Gene Delivery via Nanoparticle Coating,” Biomaterials, vol. 31, pp. 998-1006, 2010).

The mammalian immune system uses T cells to fight substances or cells with foreign antigens. When encountering solid tumors, T cells often fail to respond effectively. Even when T cells reach the tumor site, they face a range of immunosuppressive factors that allow cancer cells to evade the immune system. CAR-T technology uses genetic engineering to reprogram naturally occurring circulating T cells by inserting a chimeric antigen receptor (CAR) into them to create highly specific CAR-T cells. The CAR guides the engineered CAR-T cells to the target tissue by specifically binding to antigens on the surface of the target tissue. Therefore, CAR-T cells can specifically target tumor cells, making them more effective than naturally occurring circulating T cells. CAR-T cells can also be engineered to target other tissues, such as inflamed joints and brain tissue.

The CAR of this invention comprises at least one antigen-specific targeting region (ASTR), an extracellular septum domain (ESD), a transmembrane domain (TM), one or more co-stimulatory domains (CSD), and an intracellular signaling domain (ISD), see Figure 5 and Jensen et al., “Design and implementation of adoptive therapy with chimeric antigen receptor-modified T cells,” Immunol Rev., vol. 257, pp. 127–144, 2014. After the ASTR specifically binds to the target antigen on the tumor or other target tissue, the ISD activates intracellular signaling in the CAR-T cells. For example, utilizing the antigen-binding properties of antibodies, the ISD can specifically and reactivity redirect CAR-T cells to selected targets (e.g., tumor cells or other target cells) in a non-MHC-restricted manner. Non-MHC-restricted antigen recognition enables CAR-T cells to recognize tumor cells and initiate antigen processing, thereby bypassing the main mechanism by which tumors evade immune system surveillance. In one implementation, ESD and/or CSD are optional. In another implementation, the ASTR is bispecific, which allows it to bind specifically to two different antigens or epitopes.

The conditionally active peptides of this invention can be engineered into an ASTR or a portion thereof to enable the CAR to exhibit higher binding activity to target antigens in specific environments (e.g., tumor microenvironment or synovial fluid) than in blood or other parts of the body where different environments exist. Such a CAR can preferentially deliver T cells to disease sites, thereby significantly reducing the side effects caused by T cell attack on normal tissues. This allows for the use of higher doses of T cells to increase therapeutic efficacy and improve individual tolerance to treatment.

These CARs are particularly valuable for developing new therapies needed in subjects for short or limited time. Examples of beneficial applications include high-dose systemic therapy and high-concentration local therapy. See Maher, “Immunotherapy of Malignant Disease Using Chimeric Antigen Receptor Engrafted T Cells,” ISRN Oncology, vol. 2012, article ID 278093, 2012.

ASTRs can contain conditionally active peptides, such as antibodies, particularly single-chain antibodies or fragments thereof, that specifically bind to antigens on tumors or other target tissues. Some examples of peptides suitable for ASTRs include linked cytokines (which lead to recognition of cells carrying cytokine receptors), affibody, ligand-binding domains from naturally occurring receptors, and soluble protein/peptide ligands of receptors (e.g., receptors on tumor cells). In fact, almost any molecule capable of binding a given antigen with high affinity can be used as an ASTR.

In some embodiments, the CAR of the present invention comprises at least two ASTRs targeting at least two different antigens or two epitopes on the same antigen. In one embodiment, the CAR comprises three or more ASTRs targeting at least three or more different antigens or epitopes. When multiple ASTRs are present in the CAR, the ASTRs may be arranged in tandem and may be separated by linker peptides (Figure 5).

In yet another implementation, the ASTR comprises a biantibody. In the biantibody, a linker peptide that is too short for the two variable regions to fold together forms an scFv, thereby driving scFv dimerization. Shorter linkers (one or two amino acids) result in the formation of trimers, also known as tribosomes or tribody antibodies. Tetrabody antibodies can also be used in ASTRs.

The antigen targeted by the CAR is present on the surface or inside the cells of the targeted tissue, such as tumors, glandular hyperplasia (e.g., prostate), warts, and unwanted adipose tissue. Intracellular antigens can also be targeted by the CAR when surface antigens are more effectively recognized and bound by the CAR's ASTR. In some embodiments, the target antigen is preferably specific to cancer, inflammatory diseases, neurological disorders, diabetes, cardiovascular diseases, or infectious diseases. Examples of target antigens include antigens expressed by various immune cells, cancers, sarcomas, lymphomas, leukemias, germ cell tumors, embryonic cells, and cells associated with various blood disorders, autoimmune diseases, and/or inflammatory diseases.

Cancer-specific antigens that can be targeted by ASTR include one or more of the following: 4-IBB, 5T4, adenocarcinoma antigen, alpha-fetoprotein, BAFF, B-lymphoma cells, C242 antigen, CA-125, carbonic anhydrase 9 (CA-IX), C-MET, CCR4, CD152, CD19, CD20, CD200, CD22, CD221, CD23 (IgE receptor), CD28, CD30 (TNFRSF8), CD33, CD4, CD40, CD44 v6, CD51, CD52, CD56, CD74, CD80, CEA, CTLA-4, DR5, EGFR, EpCAM, CD3, FAP, fibronectin extradomain-B, folate receptor 1, GD2, GD3, gangliosides, glycoprotein 75, and GPNMB. HER2/neu, HGF, human scattering factor receptor kinase, IGF-1 receptor, IGF-I, IgG1, LI-CAM, IL-13, IL-6, insulin-like growth factor I receptor, integrin α5β1, integrin αvβ3, MORAb-009, MS4A1, MUC1, mucin CanAg, N-hydroxyacetylneuraminic acid, NPC-1C, PDGF-Rα, PDL192, phosphatidylserine, prostate cancer cells, RANKL, RON, ROR1, SCH900105, SDC1, SLAMF7, TAG-72, tendinin C, TGFβ2, TGF-β, TRAIL-R1, TRAIL-R2, tumor antigen CTAA16.88, VEGF-A, VEGFR-1, VEGFR2, or vimentin.

Antigens specific to inflammatory diseases that can be targeted by ASTRs include one or more of the following: AOC3 (VAP-1), CAM-300, CCL11 (eosinophil activation chemokine-1), CD125, CD147 (basigin), CD154 (CD40L), CD2, CD20, CD23 (IgE receptor), CD25 (IL-2 receptor chain), CD3, CD4, CD5, IFN-α, IFN-γ, IgE, IgE Fc Region, IL-1, IL-12, IL-23, IL-13, IL-17, IL-17A, IL-22, IL-4, IL-5, IL-6, IL-6 receptor, integrin α4, integrin α4β7, Lamagama, LFA-1 (CD11a), MEDI-528, myostatin, OX-40, rhuMAbβ7, scleroscin, SOST, TGFβ1, TNF-α or VEGF-A.

Antigens specific to neuronal diseases that can be targeted by the ASTR of the present invention include one or more of β-amyloid protein or MABT5102A. Antigens specific to diabetes that can be targeted by the ASTR of the present invention include one or more of L-Iβ or CD3. Antigens specific to cardiovascular diseases that can be targeted by the ASTR of the present invention include one or more of C5, cardiac myosin, CD41 (integrin α-lib), fibrin II, β chain, ITGB2 (CD18), and sphingosine-1-phosphate.

Antigens specific to infectious diseases that can be targeted by the ASTR of this invention include one or more of the following: anthrax toxin, CCR5, CD4, aggregation factor A, cytomegalovirus, cytomegalovirus glycoprotein B, endotoxin, Escherichia coli, hepatitis B surface antigen, hepatitis B virus, HIV-1, Hsp90, influenza A hemagglutinin, lipoteichoic acid, Pseudomonas aeruginosa, rabies virus glycoprotein, respiratory syncytial virus, and TNF-α.

Other examples of target antigens include surface proteins present on cancer cells in a specific or amplified manner, such as the IL-14 receptor, CD19, CD20, and CD40 for B-cell lymphoma; LewisY and CEA antigens for various cancers; the Tag72 antigen for breast and colorectal cancer; EGF-R for lung cancer; folate-binding proteins; and the HER-2 protein, which is typically amplified in human breast and ovarian cancer; or viral proteins, such as the gp120 and gp41 envelope proteins of human immunodeficiency virus (HIV), envelope proteins of hepatitis B and hepatitis C viruses, glycoprotein B and other envelope glycoproteins of human cytomegalovirus, and envelope proteins of tumor viruses such as Kaposi's sarcoma-associated herpesvirus. Other possible target antigens include CD4, where the ligand is the HIV gp120 envelope glycoprotein, and other viral receptors, such as ICAM, the receptor for human rhinovirus, and related receptor molecules for poliovirus.

In another implementation, the CAR can target antigens of cancer therapy cells, such as NK cells, to activate these cells by acting as immune effector cells. One example is a CAR that targets the CD16A antigen to enable NK cells to fight against malignant tumors expressing CD30. A bispecific tetravalent AFM13 antibody is an example of an antibody capable of delivering this effect. Further details of this type of implementation can be found, for example, in Rothe, A., et al., “A phase 1 study of the bispecific anti-CD30/CD16A antibody construct AFM13 in patients with relapsed orrefractory Hodgkin lymphoma,” Blood, 25 June 2015, Vol. 125, no. 26, pp. 4024-4031.

In some implementations, the extracellular septum domain and transmembrane domain can be ubiquitination-resistant, which can enhance CAR-T cell signaling and thereby enhance antitumor activity (Kunii et la., "Enhanced function of redirected human t cells expressing linker for activation of t cells that is resistant to ubiquitination," Human Gene Therapy, vol. 24, pp. 27–37, 2013). Within this region, the extracellular septum domain is outside the CAR-T cells and is therefore exposed to different conditions, and may conditionally develop ubiquitination resistance.

C. Engineered masked conditionally active peptides

The conditionally active peptides of the present invention, particularly conditionally active antibodies, can mask their conditional activity and/or mask the activity of reagents conjugated to them by means of a masking moiety. The masked activity is obtained once the masking moiety is removed from or cleaved from the conditionally active peptide. Suitable masking techniques are described, for example, in Desnoyers et al., “Tumor-Specific Activation of an EGFR-Targeting Probody Enhancements Therapeutic Index,” Sci. Transl. Med. 5, 207ra144, 2013.

In some embodiments, the conditionally active antibody is linked to a masking moiety that masks the conditional activity and/or the activity of any reagent conjugated thereto. For example, when the conditionally active antibody is linked to the masking moiety, this linking or modification causes a structural change that reduces or inhibits the ability of the conditionally active antibody to bind specifically to its antigen. Once the conditionally active antibody reaches the target tissue or microenvironment, the masking moiety is cleaved by an enzyme present in the target tissue or microenvironment, thereby releasing the masked activity. For example, the enzyme could be a protease that is normally active in the tumor microenvironment, which can cleave the masking moiety to release the conditionally active antibody that is active within the tumor tissue.

In some embodiments, the activity is masked to less than about 50%, or less than about 30%, or less than 10%, or less than 5%, or less than 2%, or less than 1%, or less than 0.1%, or less than 0.01% of the original activity. For example, in some embodiments, to ensure sufficient delivery time, the masking effect is designed to persist for at least 2, 4, 6, 8, 12, 28, 24, 36, 48, 60, 72, 84, 96 hours, or 5, 10, 15, 30, 45, 60, 90, 120, 150, 180 days, or 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 months or longer during target replacement for in vivo measurements or in vitro immunoadsorption assays.

In some embodiments, the masking portion is structurally similar to the natural binding partner (antigen) of the conditionally active antibody. The masking portion may be a modified natural binding partner of the conditionally active antibody containing amino acid changes that at least slightly reduce the affinity for binding to the conditionally active antibody and/or the antibody's antigenicity. In some embodiments, the masking portion does not contain the natural binding partner of the conditionally active antibody or is substantially non-homological to the natural binding partner of the conditionally active antibody. In other embodiments, the sequence identity between the masking portion and the natural binding partner of the conditionally active antibody does not exceed 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, or 80%.

The masking portion can be provided in various different forms. In some embodiments, the masking portion can be a known binding partner of the conditionally active antibody, provided that the binding affinity and/or antibody affinity of the masking portion to the conditionally active antibody is lower than the binding affinity and/or antibody affinity of the target protein targeted by the conditionally active antibody after the masking portion is cleaved, in order to reduce interference of the masking portion with the desired binding to the target. Therefore, the masking portion is preferably a masking portion having the following characteristics: before the masking portion is cleaved, the masking portion masks the conditionally active antibody from binding to the target, but once the masking portion has been cleaved from the antibody, it does not substantially or significantly interfere with or compete with the binding of the active molecule to the target. In a specific embodiment, the conditionally active antibody and the masking portion do not contain the amino acid sequence of a naturally occurring binding partner pair, such that at least one of the conditionally active antibody and the masking portion does not have an amino acid sequence that is a member of a naturally occurring binding partner.

Alternatively, the masking portion may not specifically bind to the conditionally active antibody, but rather interfere with the binding of the conditionally active antibody to the target through non-specific interactions (such as steric hindrance). For example, the masking portion may be positioned such that the structure or conformation of the antibody allows the masking portion to mask the conditionally active antibody through, for example, charge-based interactions, thereby interfering with the target's access to the conditionally active antibody.

In some embodiments, the masking portion is covalently linked to the conditionally active antibody. In another embodiment, the conditionally active antibody is prevented from binding to its target by binding the masking portion to the N-terminus of the conditionally active antibody. In yet another embodiment, the conditionally active antibody is linked to the masking portion via a cysteine-cysteine disulfide bond between the masking portion and the conditionally active antibody.

In some embodiments, the conditionally active antibody is also linked to a cleavable moiety (CM). The CM can be cleaved by an enzyme, reduced by a reducing agent, or photolyzed. In one embodiment, the amino acid sequence of the CM may overlap with or be contained within the masking moiety. In another embodiment, the CM is located between the conditionally active antibody and the masking moiety. It should be noted that all or part of the CM can help mask the conditionally active antibody before cleavage. When the CM is cleaved, the binding activity of the conditionally active antibody to its antigen is enhanced.

CM can be a substrate of an enzyme that co-localizes with the target antigen at the therapeutic site of an individual. Alternatively, CM can have a cysteine-cysteine disulfide bond that can be broken by reduction. CM can also be a light-activated, light-labile substrate.

Enzymes that cleave CMs should preferentially be located in the desired target tissue of conditionally active antibodies, where the antibodies are more active under the conditions (abnormal conditions) provided by the target tissue (such as diseased or tumor tissue). For example, there are known proteases present in many cancers (e.g., solid tumors) at elevated levels. See, for example, La Rocca et al., (2004) British J. of Cancer 90(7):1414-1421. Non-limiting examples of these diseases include: all types of cancer (breast cancer, lung cancer, colorectal cancer, prostate cancer, head and neck cancer, pancreatic cancer, etc.), rheumatoid arthritis, Crohn's disease, melanoma, SLE, cardiovascular injury, local ischemia, etc. Therefore, suitable CMs can be selected that contain peptide substrates that can be cleaved by proteases present in tumor tissues, especially proteases present in tumor tissues at elevated levels compared to non-cancerous tissues.

In some embodiments, CM can be a substrate of an enzyme selected from cardamomin, plasmin, TMPRSS-3/4, MMP-9, MT1-MMP, cathepsin, caspase, human neutrophil elastase, β-secretase, uPA, and PSA. The level of the enzyme that cleaves CM in the target tissue at the treatment site (e.g., diseased or tumor tissue; for example, for therapeutic or diagnostic treatment) is relatively high compared to the level in tissues at non-treatment sites (e.g., healthy tissue). Therefore, the conditional activity of the antibody is higher in diseased or tumor tissue, and the enzyme present in diseased or tumor tissue can cleave CM, which further enhances the activity of the conditionally active antibody or the agent conjugated thereto. Unmodified or uncleaved CM can effectively inhibit or mask the activity of the conditionally active antibody, resulting in lower activity of the conditionally active antibody in normal tissues (normal physiological conditions). This dual mechanism of inhibiting the activity of the conditionally active antibody in normal tissues (conditional activity and masking portion) allows the use of high doses of the conditionally active antibody without causing significant adverse reactions.

In some implementations, CM may be a substrate of an enzyme selected from the enzymes listed in Table 1 below.

Table 1. Exemplary enzymes/proteases

Alternatively or additionally, CM may include disulfide bonds of cysteine pairs that can be broken by a reducing agent such as a cellular reducing agent, including glutathione (GSH), thioredoxin, NADPH, flavin, ascorbate, etc., which may be present in large quantities within or around the solid tumor tissue.

In some embodiments, the conditionally active antibody comprises both a CM and a masking moiety. The masking of the activity of the conditionally active antibody is released when the enzyme cleaves the CM. In some embodiments, it may be necessary to insert one or more linkers (e.g., flexible linkers) between the antibody, the masking moiety, and the CM to provide flexibility. For example, the masking moiety and/or the CM may not contain a sufficient number of residues (e.g., Gly, Ser, Asp, Asn, especially Gly and Ser, especially Gly) to provide the desired flexibility. Therefore, it may be beneficial to introduce one or more amino acids to provide a flexible linker. For example, the masked conditionally active antibody may have the following structure (where the following formula represents an amino acid sequence in the N-to-C-terminal or C-to-N-terminal direction):

(MM)-L ₁ -(CM)-(AB)

(MM)-(CM)-L ₁ -(AB)

(MM)-L ₁ -(CM)-L ₂ -(AB)

Ring [ _L1 -(MM) _-L2 -(CM) _-L3 -(AB)]

MM is the masking part, AB is the conditionally active antibody; _L1 , _L2 and _L3 are present independently and optionally, and are the same or different flexible linkers containing at least one flexible amino acid (e.g., Gly); in the presence of a ring, the entire structure is in the form of a ring structure due to the presence of a disulfide bond between a pair of cysteine residues at or near the N-terminus and C-terminus of the structure.

The adapters suitable for use in this invention are generally adapters that provide flexibility to the masking portion in order to inhibit the activity of conditionally active antibodies. Such adapters are generally referred to as flexible adapters. Suitable adapters can be readily selected and can have any suitable and different lengths, such as 1 amino acid (e.g., Gly) to 20 amino acids, 2 amino acids to 15 amino acids, 3 amino acids to 12 amino acids, including 4 amino acids to 10 amino acids, 5 amino acids to 9 amino acids, 6 amino acids to 8 amino acids, or 7 amino acids to 8 amino acids, and can be 1, 2, 3, 4, 5, 6, or 7 amino acids.

Exemplary flexible joints include glycine polymers (G) _n , glycine-serine polymers (including, for example, (GS) _n , (GSGGS) _n , and (GGGS) _n , where n is an integer at least 1), glycine-alanine polymers, alanine-serine polymers, and other flexible joints known in the art. Exemplary flexible joints include, but are not limited to, Gly-Gly-Ser-Gly, Gly-Gly-Ser-Gly, Gly-Ser-Gly-Ser-Gly, Gly-Ser-Gly-Gly, Gly-Gly-Gly-Gly, Gly-Ser-Ser-Ser-Gly, etc.

WO2010081173A2 describes some techniques for masking the activity of conditionally active antibodies.

This invention provides a method for preparing conditionally active peptides from parental peptides, such as wild-type peptides or therapeutic peptides. The method includes the steps of: evolving DNA encoding the parental peptide using one or more evolution techniques to generate mutant DNA; expressing the mutant DNA to obtain the mutant peptide; subjecting the mutant peptide and the parental peptide to tests under a first condition and a second condition; and selecting conditionally active peptides from the mutant peptides that possess both of the following properties: (a) decreased activity compared to the parental peptide in the test under the first condition, and (b) increased activity compared to the parental peptide in the test under the second condition. The tests under the first and second conditions are performed in a test solution containing at least one component selected from inorganic compounds, ions, and organic molecules. In some embodiments, the first condition is a normal physiological condition, and the second condition is an abnormal condition.

Conditionally active peptides are reversibly or irreversibly inactivated under a first condition or normal physiological condition, but remain active under a second condition or abnormal condition at the same or equivalent level as the first condition or normal physiological condition. These conditionally active peptides and methods for producing them have been described in U.S. Patent No. 8,709,755B2. Conditionally active peptides are particularly valuable for developing novel therapies that are active in the host for a short time or limited period. This is especially valuable in cases where prolonged action of a given dose of therapeutic agent would be harmful to the host, but its limited activity is required for the desired treatment. Beneficial applications include high-dose local or systemic treatment, as well as high-concentration local treatment. Inactivation under a first condition or normal physiological condition can be determined by a combination of drug administration and the peptide inactivation rate. This condition-based inactivation is particularly important for enzyme therapeutics, where catalytic activity causes large negative effects over a relatively short period.

This invention also relates to methods for engineering or evolving parental peptides to produce conditionally active peptides, said conditionally active peptides being reversibly or irreversibly activated or inactivated over time, or only when they are in certain microenvironments in vivo, including specific organs such as tumor microenvironments, synovial fluid, bladder, or kidneys. In some embodiments, the conditionally active peptide is an antibody or antibody fragment against one or more target proteins (antigens) described herein.

The conditionally active polypeptide can be a isolated polypeptide with pH-dependent activity, wherein, in the presence of a substance selected from histidine, histamine, adenosine diphosphate, adenosine triphosphate, citrate, bicarbonate, acetate, lactate, sulfide, hydrogen sulfide, ammonium, dihydrogen phosphate, and any combination thereof, the activity at a first pH is at least about 1.3 times that at a second pH. In the absence of small molecules, the same activity is not pH-dependent. In some embodiments, the activity at the first pH is at least about 1.5, or at least about 1.7, or at least about 2.0, or at least about 3.0, or at least about 4.0, or at least about 6.0, or at least about 8.0, or at least about 10.0, or at least about 20.0, or at least about 40.0, or at least about 60.0, or at least about 100.0 times that at the second pH. The first pH can be an abnormal pH in the range of approximately 5.5–7.2 or approximately 6.2–6.8, while the second pH can be a normal physiological pH in the range of approximately 7.2–7.6.

D. Parental polypeptide

Parental peptides can be wild-type peptides, including those that are not naturally occurring, mutant peptides derived from wild-type peptides such as therapeutic peptides, chimeric peptides derived from different wild-type peptides, or even synthetic peptides. Parental peptides can be selected from antibodies, enzymes, cytokines, regulatory proteins, hormones, receptors, ligands, biosimilars, immunomodulators, growth factors, and fragments of these peptides.

U.S. Patent No. 8,709,755B2 describes the suitable wild-type peptides and the ways in which they can be evolved and selected for the production of conditionally active peptides.

In some embodiments, the parental peptide can be selected from libraries of wild-type or mutant peptides, such as phage display libraries. In such embodiments, a large number of candidate peptides are expressed in the phage library, particularly via surface display technology. Candidate peptides from the library are screened to obtain suitable parental peptides. A typical phage library may contain phages that express thousands or even millions of candidate peptides in a bacterial host. In one embodiment, the phage library may contain multiple phages.

To construct a phage library, filamentous phages, such as filamentous E. coli phage M13, are typically genetically modified by inserting an oligonucleotide encoding a candidate peptide into the coding sequence of one of the phage coat proteins. The phage coat protein is then expressed along with the candidate peptide, causing the candidate peptide to be displayed on the surface of the phage particle. The displayed candidate peptides can then be screened to obtain suitable parent peptides.

A common technique for screening suitable parental peptides involves immobilizing phage particles containing the desired candidate peptide onto a support. The support can be a plastic plate coated with a "decoy" that binds to the desired candidate peptide. Unbound phage particles can be washed off the plate. The phage particles bound to the plate (containing the desired candidate peptide) are eluted by washing, and the eluted phage particles are amplified in bacteria. The sequence encoding the candidate peptide in the selected phage particle can then be determined by sequencing. The relationship between the candidate peptide and the decoy can be, for example, a ligand-receptor or antigen-antibody relationship.

Another common technique for screening suitable parental peptides is to use an enzyme assay to test the desired enzyme activity of each phage clone, which is the enzyme activity exhibited by the candidate peptide. Based on the specific enzyme activity, those skilled in the art can design appropriate assays to screen candidate peptides with the desired level of enzyme activity.

In some embodiments, the phage library is provided as an array, such that each phage clone occupies a specific position on the array. This array can be provided on a solid support, such as a membrane, agar plate, or microtiter plate, with each phage clone of the library placed on or adhered to a specific predetermined position on the solid support. In the case of an agar plate, such a plate preferably includes a bacterial growth medium to support bacterial growth. When the array is set on a membrane (e.g., a nitrocellulose or nylon membrane), a bacterial culture is applied to the membrane and the membrane is immersed in a nutrient growth medium. Alternatively, phage clones can also be provided on beads, in which case individual phage clones can be adhered to individual beads. Alternatively, each phage clone can be positioned at the end of an optical fiber, in which case the optical fiber is used for optical transmission of ultraviolet radiation from a light source.

A typical phage library can contain ^10⁶ to ^10¹⁰ phages, each characterized by a coat protein carrying a different candidate polypeptide (e.g., gp3 or gp8 of phage M113). The bacterial host for the phage library can be selected from bacterial genera, including, for example, *Salmonella*, *Staphylococcus*, *Streptococcus*, *Shigella*, *Listeria*, *Campylobacter*, *Klebsiella*, *Yersinia*, *Pseudomonas*, and *Escherichia*.

Oligonucleotides encoding candidate polypeptides can be collections of cDNAs encoding wild-type polypeptides. Methods are known for synthesizing cDNA from biological samples that can express suitable parental polypeptides. Any genetic information expressing physiological activity via transcripts can be harvested as cDNA. When generating cDNA, full-length cDNA must be synthesized. Several methods exist for synthesizing full-length cDNA. For example, suitable methods include labeling the 5' cap site using a cap-binding protein from yeast or HeLa cells (I. Edery et al., "An Efficient Strategy To Isolate Full-length cDNAs Based on amRNA Cap Retention Procedure (CAPture)", Mol. Cell. Biol., vol. 15, pp. 3363-3371, 1995); and removing phosphate from incomplete cDNA lacking a 5' cap using alkaline phosphatase, followed by treatment of the complete cDNA with a decapping enzyme from tobacco mosaic virus, thereby obtaining only full-length cDNA. Methods involving phosphate-containing NAs (K. Maruyama et al., "Oligo-capping: a simple method to replace the cap structure of eukaryotic mRNAs with oligoribonucleotides", Gene, vol. 138, pp. 171-174, 1995; and S. Kato et al., "Construction of a human full-length cDNA bank", Gene, vol. 150, pp. 243-250, 1995).

In embodiments where the parental peptide is an antibody, a library of candidate peptides can be generated using recombinant antibodies derived from a complete antibody profile of an organism. The genetic information representing this antibody profile is compiled into a large collection of complete antibodies, which can be screened to obtain suitable parental antibodies possessing the desired antigen-binding activity and/or one or more other functional characteristics. In some embodiments, B cells from animals immunized with an antigen (e.g., immunized humans, mice, or rabbits) are isolated. mRNA from the isolated B cells is collected and converted to cDNA, which is then sequenced. The most common cDNA fragments encoding the light chain and the most common cDNA fragments encoding the heavy chain are assembled into antibodies. In one embodiment, the 100 most common cDNA fragments encoding the light chain and the 100 most common cDNA fragments encoding the heavy chain are assembled to generate candidate antibodies. In another embodiment, the most common cDNA fragments encoding only the variable region of the heavy chain and the variable region of the light chain are assembled to generate antibody fragments containing only the variable region and not the constant region.

In some implementations, a cDNA fragment encoding the variable region of the IgG heavy chain is assembled with the most common variable region of IgK, or the IgK light chain. The assembled antibody contains the heavy chain variable region from IgG and the light chain variable region from IgK, or IgK.

Then, cDNA encoding the assembled antibody is cloned and expressed, preferably in a plate-based format. The binding activity of the expressed antibody can be tested using a bead-based ELISA assay, and suitable parental antibodies can be selected based on the ELISA assay. Alternatively, the cDNA encoding the assembled antibody can be expressed in a phage display library, and one or more desired parental antibodies can then be screened using any of the techniques disclosed herein.

In embodiments where the parental polypeptide is the antibody, the parental antibody preferably possesses at least one characteristic that makes it easier to evolve the parental antibody into a conditionally active antibody. In some embodiments, the parental antibody may have similar binding activity and/or characteristics under both normal and abnormal conditions. In such embodiments, the parental antibody is selected based on the combination of the most similar binding activity and/or the most similar one or more characteristics under both normal and abnormal conditions. For example, if the normal and abnormal conditions are pH 7.4 and pH 6.0, respectively, then the parental antibody with the most similar binding activity at pH 7.4 and pH 6.0 can be selected relative to antibodies with less similar binding activity at pH 7.4 and pH 6.0.

E. Identification of conditionally active peptides

After selecting the parental peptide, the DNA encoding the parental peptide is evolved using a suitable mutagenesis technique to generate mutant DNA. The mutant DNA can then be expressed to produce a mutant peptide, which is then screened to identify conditionally active peptides. In some embodiments, the evolution may be minimal; for example, only a small number of mutations are introduced into the parental peptide to generate a mutant peptide with the desired conditional activity. For example, introducing fewer than about 20 variations (possibly fewer than about 18 variations) at each site via comprehensive positional evolution (CPE) may be sufficient to generate a suitable conditionally active peptide. For comprehensive positional synthesis (CPS), combinations of fewer than about 6, 5, 4, 3, or 2 moderating mutations in the parental peptide may be sufficient to generate the desired conditionally active peptide.

In some implementations, evolution and expression steps may be unnecessary when the candidate peptide library (e.g., a phage library and/or a recombinant antibody library) is sufficiently large. Such a large library may contain parental peptides with conditional activity characteristics (lower activity in tests under normal physiological conditions and higher activity in tests under abnormal conditions compared to a reference peptide, or lower activity in tests under normal physiological conditions than in tests under abnormal conditions). In these implementations, the parental peptides in the library undergo a selection step to discover conditionally active peptides with the characteristic that their activity in tests under normal physiological conditions is lower than their activity in tests under abnormal conditions. In one implementation, the parental peptides and reference peptides in the library are tested under normal physiological conditions and abnormal conditions, respectively. The conditionally active peptides selected from the library are those that exhibit lower activity under normal physiological conditions compared to the reference peptide and higher activity under abnormal conditions compared to the reference peptide. In this implementation, since the library is already sufficiently large, candidate peptides with conditional activity characteristics are present. Therefore, evolution of the parental peptides is not necessary for the purpose of discovering conditionally active peptides.

In some implementations, the reference peptide may not have conditional activity because it exhibits similar or identical activity under both normal and abnormal physiological conditions. The reference peptide is a peptide of the same type as the candidate peptides in the library, such as a similar type of enzyme, cytokine, regulatory protein, antibody, hormone, or functional peptide. The reference peptide may also be a similar type of tissue plasminogen activator, streptokinase, urokinase, renin, hyaluronidase, calcitonin gene-associated peptide (CGRP), substance P (SP), neuropeptide Y (NPY), vasoactive intestinal peptide (VTP), vasopressin, or angiostatin. For example, when the library contains a large number of candidate antibodies against an antigen, the reference peptide is an antibody against the same antigen that exhibits the same or similar antigen-binding activity under both normal and abnormal physiological conditions.

Therefore, in one embodiment, candidate peptides and reference peptides in the library are tested under normal physiological conditions and under abnormal conditions, respectively. Conditionally active peptides selected from the library exhibit the following two characteristics: (a) decreased activity compared to the reference peptide under normal physiological conditions, and (b) increased activity compared to the reference peptide under abnormal conditions.

F. Methods for generating conditionally active peptides

One or more mutagenesis techniques are used to evolve the DNA encoding the parental peptide to generate mutant DNA; the mutant DNA is expressed to generate the mutant peptide; and the mutant peptide is screened under a first condition that may be normal physiological conditions and a second condition that may be abnormal conditions. Conditionally active peptides are selected from those mutant peptides that exhibit both of the following characteristics: (a) decreased activity compared to the parental peptide in the test under the first condition, and (b) increased activity compared to the parental peptide in the test under the second condition. The decrease in activity of the conditionally active peptide under the first condition or normal physiological conditions may be reversible or irreversible.

In some embodiments, the polypeptide to be evolved may be a fragment of a wild-type polypeptide, a fragment of a therapeutic polypeptide, or an antibody fragment. In other embodiments, the parent polypeptide may be a polypeptide selected from mutant polypeptides generated by mutagenesis methods that select polypeptides with desired properties, such as high binding activity, high expression levels, or humanization. The selected polypeptide may be used as the parent polypeptide to be evolved in the methods disclosed herein.

A method for generating mutant DNA from DNA encoding a parental polypeptide has been described in U.S. Patent No. 8,709,755B2.

Evolving the DNA encoding the parental polypeptide to produce mutant DNA can be done using point mutations (substitution, insertion, and/or deletion) or mutations in large segments of the DNA. In some respects, the evolution steps do not alter the active site of the parental polypeptide, but rather change only one or more regions around the active site and/or one or more regions distant from the active site.

In one aspect, the evolution step involves converting the parental full-length antibody into a single-chain antibody. In this case, although the active site, i.e., the mutable region, particularly the CDR, may not have any mutations relative to the parental antibody, the environment in which the active site exists has been altered by eliminating the constant region. In one embodiment, the parental full-length antibody is an IgG antibody, and the mutant antibody is a single-chain antibody derived from it.

In some respects, a single-chain antibody is a bispecific antibody with two arms, each arm binding to a different epitope. A mutation in one arm may affect the activity of the other arm. Therefore, the evolution step may involve mutating only one arm of the parent polypeptide that is a bispecific antibody. In one embodiment, the arm can be shortened by deletion or lengthened by insertion to evolve the length of one arm. Alternatively, the evolution step may evolve both arms of the bispecific antibody in the same evolution step or in successive evolution steps, optionally with screening after each step.

On the other hand, the parent polypeptide is an antibody or antibody fragment. Evolutionary steps can mutate the Fc region. Mutations in the Fc region can be substitutions, insertions, and/or deletions. The Fc region can be shortened by deleting a fragment from the Fc region, or lengthened by inserting a fragment into the Fc region.

In another aspect, the parent peptide contains multiple complementarity-determining regions that are interrupted by the backbone region. For example, such a parent peptide can be a variable region of an antibody, a light chain, or a heavy chain. In some embodiments, the evolution step may involve mutating only the backbone region or a combination of mutating the complementarity-determining region and the backbone region. The evolution of the backbone region and the complementarity-determining region can be performed in a single step or in multiple consecutive steps, optionally followed by screening after each step.

On the other hand, the parental peptide has several regions outside its active site. These regions can be mutated sequentially in multiple evolutionary steps, optionally followed by screening after one or more evolutionary steps. For example, an evolutionary step could evolve one region of the peptide, followed by screening for conditionally active peptides; then evolve another region of the peptide, followed by screening for conditionally active peptides; then further evolve yet another region of the peptide, followed by yet another step of screening for conditionally active peptides.

In some cases, evolution of one or more regions (e.g., surrounding or remote regions) outside the active site of the parent peptide and/or the mutated conditionally active peptide can alter the activity of the active site. Mutating a surrounding or remote region instead of the active site can, in some cases, result in the mutated peptide's active site having more or less activity than the parent peptide's active site under specific conditions. In other embodiments, desired conditional activity or improved selectivity is achieved by evolving one or more regions of the parent peptide or mutated peptide outside the region containing the active site.

In some respects, conditionally active peptides derived from regions other than the active site region of the parent peptide can produce at least two, at least three, or at least five selectivity.

A suitable method for expressing mutant DNA to produce mutant polypeptides has been described in U.S. Patent No. 8,709,755B2.

A method for screening mutant peptides to select conditionally active peptides has been described in U.S. Patent No. 8,709,755B2.

Test conditions for screening and selecting conditionally active peptides

The screening step can use a first and second condition, or normal physiological and abnormal conditions, selected from temperature, pH, osmolarity, osmolality, oxidative stress, electrolyte concentration, and combinations of two or more such conditions. For example, a normal physiological condition for temperature could be the normal human body temperature of 37.0°C, while an abnormal temperature condition could be a temperature different from 37.0°C, such as a temperature in a tumor microenvironment that is 1-2°C higher than normal physiological temperature. In another example, normal physiological and abnormal conditions could also be a normal physiological pH in the range of 7.2-7.8 or 7.2-7.6 and an abnormal pH in the range of 5.5-7.2, 6-7, or 6.2-6.8, as presented in a tumor microenvironment.

The tests under the first and second conditions, or under normal physiological and abnormal conditions, can be performed in a test medium. The test medium can be a solution containing, for example, buffers and other components. Common buffers that can be used in the test medium include citrate buffer (e.g., sodium citrate), phosphate buffer, bicarbonate buffer (e.g., Krebs buffer), phosphate-buffered saline (PBS), Hank's buffer, Tris buffer, HEPES buffer, etc. Other buffers known to those skilled in the art as suitable for this test can be used. These buffers can be used to mimic the characteristics or components of human or animal bodily fluids (such as plasma or lymph).

The test solution used in the method of the present invention may contain at least one component selected from inorganic compounds, ions, and organic molecules, preferably components commonly found in the body fluids of mammals (e.g., humans or animals). Examples of such components include nutrients and metabolites, as well as any other components that may be present in body fluids. The present invention contemplates that this component may or may not be part of the buffer system. For example, the test solution may be a PBS buffer supplemented with bicarbonate ions, wherein the bicarbonate ions are not part of the PBS buffer. Alternatively, the bicarbonate ions may be part of a bicarbonate buffer.

This component can be present in both test solutions at substantially the same concentration (for both the first and second conditions), even though the two test solutions differ in other aspects, such as pH, temperature, electrolyte concentration, or osmotic pressure. Therefore, this component is used as a constant, rather than as a difference between the first and second conditions or between normal and abnormal conditions.

In some implementations, the component is present in both test solutions at a concentration close to or the same as the normal physiological concentration of the component in mammals, particularly humans.

The inorganic compound or ion may be selected from one or more of the following: boric acid, calcium chloride, calcium nitrate, diammonium phosphate, magnesium sulfate, monoammonium phosphate, monopotassium phosphate, potassium chloride, potassium sulfate, copper sulfate, ferric sulfate, manganese sulfate, zinc sulfate, magnesium sulfate, calcium nitrate, chelates of calcium, copper, iron, manganese and zinc, ammonium molybdate, ammonium sulfate, calcium carbonate, magnesium phosphate, potassium bicarbonate, potassium nitrate, hydrochloric acid, carbon dioxide, sulfuric acid, phosphoric acid, carbonic acid, uric acid, hydrogen chloride, urea, phosphate ions, sulfate ions, chloride ions, magnesium ions, sodium ions, potassium ions, ammonium ions, iron ions, zinc ions and copper ions.

Examples of normal physiological concentrations of some inorganic compounds include: uric acid (2-7.0 mg/dL), calcium ions (8.2-11.6 mg/dL), chloride ions (355-381 mg/dL), iron ions (0.028-0.210 mg/dL), potassium ions (12.1-25.4 mg/dL), sodium ions (300-330 mg/dL), carbonic acid (15-30 mM), citrate ions (approximately 80 μM), histidine ions (0.05-2.6 mM), histamine (0.3-1 μM), HAPT ions (hydrogenated adenosine triphosphate) ions (1-20 μM), and HADP ions (1-20 μM).

In some implementations, the ions present in the test solution used for testing under the first and second conditions or under normal physiological and abnormal conditions are selected from hydroxide ions, halide ions (chloride ions, bromide ions, iodide ions), halide ions, sulfate ions, magnesium ions, calcium ions, hydrogen sulfate ions, carbonate ions, bicarbonate ions, sulfonate ions, halide ions, nitrate ions, nitrite ions, phosphate ions, hydrogen phosphate ions, dihydrogen phosphate ions, persulfate ions, monopersulfate ions, borate ions, ammonium ions, or organic ions such as carboxylate ions, phenolate ions, sulfonate ions (organic sulfates such as methyl sulfate), vanadate ions, tungstate ions, borate ions, organic borate ions, citrate ions, oxalate ions, acetate ions, pentaborate ions, histidine ions, and phenolate ions.

The organic compounds present in the test solution under the first and second conditions or under normal physiological and abnormal conditions may be selected from, for example, amino acids such as histidine, alanine, isoleucine, arginine, leucine, asparagine, lysine, aspartic acid, methionine, cysteine, phenylalanine, glutamic acid, threonine, glutamine, tryptophan, glycine, valine, pyrrolidone, proline, selenocysteine, serine, tyrosine, and mixtures thereof.

Examples of normal physiological concentrations of some amino acids include: alanine 3.97±0.70 mg/dL, arginine 2.34±0.62 mg/dL, glutamate 3.41±1.39 mg/dL, glutamine 5.78±1.55 mg/dL, glycine 1.77±0.26 mg/dL, histidine 1.42±0.18 mg/dL, and isoleucine 1.60±0.31 mg/dL. Leucine 1.91±0.34 mg/dL, Lysine 2.95±0.42 mg/dL, Methionine 0.85±0.46 mg/dL, Phenylalanine 1.38±0.32 mg/dL, Threonine 2.02±6.45 mg/dL, Tryptophan 1.08±0.21 mg/dL, Tyrosine 1.48±0.37 mg/dL, and Valine 2.83±0.34 mg/dL.

The organic compounds present in the test solution used for testing under the first and second conditions, or under normal physiological and abnormal conditions, may be selected from non-protein nitrogenous compounds such as creatine, creatinine, guanidinoacetic acid, uric acid, allantoin, adenosine, urea, ammonia, and choline. Examples of normal physiological concentrations of some of these compounds include: creatine 1.07 ± 0.76 mg/dL, creatinine 0.9–1.65 mg/dL, guanidinoacetic acid 0.26 ± 0.24 mg/dL, uric acid 4.0 ± 2.9 mg/dL, allantoin 0.3–0.6 mg/dL, adenosine 1.09 ± 0.385 mg/dL, urea 27.1 ± 4.5 mg/dL, and choline 0.3–1.5 mg/dL.

The organic compounds present in the test solution used for testing under the first and second conditions, or under normal physiological and abnormal conditions, can be organic acids, such as citric acid, α-ketoglutarate, succinic acid, malic acid, fumaric acid, acetoacetic acid, β-hydroxybutyric acid, lactic acid, pyruvic acid, α-keto acids, acetic acid, and volatile fatty acids. Examples of normal physiological concentrations of some of these organic acids include: 2.5 ± 1.9 mg/dL citric acid, 0.8 mg/dL α-ketoglutarate, 0.5 mg/dL succinic acid, 0.46 ± 0.24 mg/dL malic acid, 0.8–2.8 mg/dL acetoacetic acid, 0.5 ± 0.3 mg/dL β-hydroxybutyric acid, 8–17 mg/dL lactic acid, 1.0 ± 0.77 mg/dL pyruvic acid, 0.6–2.1 mg/dL α-keto acids, and 1.8 mg/dL volatile fatty acids.

The organic compounds present in the test solution used for testing under the first and second conditions, or under normal physiological and abnormal conditions, may be selected from sugars (carbohydrates), such as glucose, pentoses, hexoses, xylose, ribose, mannose, and galactose, as well as disaccharides including lactose, GlcNAcβ1-3Gal, Galα1-4Gal, Manα1-2Man, GalNAcβ1-3Gal, and O-glycosides, N-glycosides, C-glycosides, or S-glycosides. Examples of normal physiological concentrations of some of these sugars include: glucose 83±4 mg/dL, polysaccharides (such as hexoses) 102±73 mg/dL, glucosamine 77±63 mg/dL, hexuronates (as glucuronic acid) 0.4-1.4 mg/dL, and pentoses 2.55±0.37 mg/dL.

The organic compounds present in the test solution used for testing under the first and second conditions, or under normal physiological and abnormal conditions, may be selected from fats or their derivatives, such as cholesterol, lecithin, cephalin, sphingomyelin, and bile acids. Examples of normal physiological concentrations of some of these compounds include: 40-70 mg/dL of free cholesterol, 100-200 mg/dL of lecithin, 0-30 mg/dL of cephalin, 10-30 mg/dL of sphingomyelin, and 0.2-0.3 mg/dL of bile acids (as bile acids).

The organic compounds present in the test solution used for testing under the first and second conditions, or under normal physiological and abnormal conditions, may be selected from proteins, such as fibrinogen, antihemophilic globulin, immunogamma globulin, immunoeuglobulin, allolectins, β-pseuglobulins, glycoproteins, lipoproteins, and albumins. For example, the normal physiological concentration of mammalian serum albumin is 3.35-5.0 g/dL. In one embodiment, the albumin is bovine serum albumin.

The organic compounds present in the test solution used for testing under the first and second conditions, or under normal physiological and abnormal conditions, may be selected from vitamins, such as vitamin A, carotene, vitamin E, ascorbic acid, thiamine, inositol, folic acid, biotin, pantothenic acid, and riboflavin. Examples of normal physiological concentrations of some of these vitamins include: vitamin A 0.019–0.036 mg/dL, vitamin E 0.90–1.59 mg/dL, inositol 0.42–0.76 mg/dL, folic acid 0.00162–0.00195 mg/dL, and biotin 0.00095–0.00166 mg/dL.

The concentration of inorganic compounds, ions, or organic molecules in the test solution (for testing under first and second conditions, or under normal physiological and abnormal conditions) can be within the normal physiological concentration range of inorganic compounds, ions, or organic molecules in human or animal serum. However, concentrations outside the normal physiological range may also be used. For example, the normal range for magnesium ions in human serum is 1.7–2.2 mg/dL, and for calcium, it is 8.5–10.2 mg/dL. The concentration of magnesium ions in the test solution can be from about 0.17 mg/dL to about 11 mg/dL. The concentration of calcium ions in the test solution can be from about 0.85 mg/dL to about 51 mg/dL. Typically, the concentration of inorganic compounds, ions, or organic molecules in the test solution can be as low as 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, or 80% of the normal physiological concentration of inorganic compounds, ions, or organic molecules in human serum, or 1.5 times, 2 times, 3 times, 4 times, 5 times, 7 times, 9 times, 10 times, or even 20 times the normal physiological concentration of inorganic compounds, ions, or organic molecules in human serum. Different components of the test solution can be used at concentration levels different from their respective normal physiological concentrations.

Tests under first and second conditions, or normal physiological and abnormal conditions, are used to measure the activity of mutant peptides. During the test, the mutant peptide and its binding partner are present in the test solution. The relationship between the mutant peptide and its binding partner can be, for example, antibody-antigen, ligand-receptor, enzyme-substrate, or hormone-receptor. For the mutant peptide to express its activity, it should be able to contact and bind to its binding partner. The activity of the mutant peptide against its binding partner is then demonstrated and measured after binding.

In some embodiments, ions used in the test can play a role in forming a bridge between the screened mutant peptide and its binding partner, particularly those containing charged amino acid residues. Thus, ions can bind to both the mutant peptide and its binding partner via hydrogen bonds and/or ionic bonds. This can facilitate the binding between the mutant peptide and its binding partner by enabling the ions to reach sites that might be difficult for larger molecules (the mutant peptide or its binding partner) to reach. In some cases, ions in the test solution can increase the likelihood of the mutant peptide and its binding partner binding to each other. Furthermore, ions can additionally or optionally assist the binding between the mutant peptide and its binding partner by binding to larger molecules (the mutant peptide or its binding partner). Such binding can alter the conformation of the larger molecule and/or maintain the larger molecule in a specific conformation favorable for binding to its binding partner.

It has been observed that ions can facilitate the binding between mutant peptides and their binding partners, likely through the formation of ionic bonds between the mutant peptide and its binding partner. Therefore, screening can be significantly more efficient than the same test without ions, and can identify a greater number of hits (candidate conditionally active peptides). Suitable ions can be selected from magnesium ions, sulfate ions, bisulfate ions, carbonate ions, citrate ions, HAPT ions, HADP ions, carbonate ions, nitrate ions, nitrite ions, phosphate ions, hydrogen phosphate ions, dihydrogen phosphate ions, persulfate ions, monosulfate ions, borate ions, lactate ions, citrate ions, histidine ions, histamine ions, and ammonium ions.

It has been found that ions at pH conditions close to their pKa facilitate binding between mutant peptides and their binding partners. Preferably, such ions are relatively small relative to the size of the mutant peptide.

In one implementation, when the anomalous condition is a pH different from the normal physiological pH under normal physiological conditions, the ions suitable for increasing the number of hits of candidate conditionally active peptides can be selected from ions at the anomalous pH to be tested in the pKa proximity test. For example, the pKa of the ion can differ from the anomalous pH by up to 2 pH units, up to 1 pH unit, up to 0.8 pH units, up to 0.6 pH units, up to 0.5 pH units, up to 0.4 pH units, up to 0.3 pH units, up to 0.2 pH units, or up to 0.1 pH units.

The pKa of the ions that can be used in this invention can vary slightly at different temperatures. Exemplary ions with the following pKa are: ammonium ion with a pKa of approximately 9.24, dihydrogen phosphate ion with a pKa of approximately 7.2, acetic acid ion with a pKa of approximately 4.76, histidine ion with a pKa of approximately 6.04, bicarbonate ion with a pKa of approximately 6.4, citrate ion with a pKa of approximately 6.4, lactate ion with a pKa of approximately 3.86, histamine ion with a pKa of approximately 6.9, HATP ion with a pKa of 6.95, and HADP ion with a pKa of 6.88.

In one embodiment, the conditionally active peptide is tested and selected in the presence of hydrosulfite. The pKa of hydrosulfite is 7.05. In some embodiments, different concentrations of hydrosulfite can be used in tests representing normal and abnormal physiological conditions. Alternatively, test media for normal and abnormal conditions may have approximately the same hydrosulfite concentration, with some numerical differences for specific conditions, such as testing at different pH values. The concentration of hydrosulfite used in the test can be from 1 mM to about 100 mM. Preferably, the concentration of hydrosulfite used in the test medium is 2-500 nM, 3-200 nM, or 5-100 nM. In some aspects, the hydrosulfite concentration can be from 1 mM to 20 mM or from 2 mM to 10 mM. Tests performed in the presence of hydrosulfite are known.

In some embodiments, once the pH of the anomalous condition (i.e., the anomalous pH) is known, the ions suitable for increasing the hits of candidate conditionally active peptides can be selected from ions with pKa that is the same as or close to the anomalous pH. For example, the pKa of the candidate ion can differ from the anomalous pH by up to 4 pH units, up to 3 pH units, up to 2 pH units, up to 1 pH unit, up to 0.8 pH units, up to 0.6 pH units, up to 0.5 pH units, up to 0.4 pH units, up to 0.3 pH units, up to 0.2 pH units, or up to 0.1 pH units.

As described above, ions most effectively facilitate the binding of mutant peptides to their binding partners when the pH and the pKa of the ions are similar or identical. For example, it has been found that bicarbonate ions (pKa approximately 6.4) are not very effective in binding mutant peptides to their binding partners in test solutions with pH values of 7.2–7.6. When the pH of the test solution decreases to 6.7 and further to approximately 6.0, bicarbonate ions become increasingly effective in binding mutant peptides to their binding partners. As a result, more hits can be identified in tests at pH 6.0 compared to tests at pH 7.2–7.6. Similarly, histidine is not very effective in binding mutant peptides to their binding partners at pH 7.4. When the pH of the test solution decreases to 6.7 and further to approximately 6.0, histidine becomes increasingly effective in binding mutant peptides to their binding partners, again enabling the identification of more hits in a pH range of, for example, approximately 6.2–6.4.

This invention has surprisingly discovered that when the pH of the test solution differs between normal physiological conditions (i.e., normal physiological pH) and abnormal conditions (i.e., abnormal pH), ions with the following pKa can significantly facilitate the binding between the mutant peptide being screened and its binding partner: the pKa is approximately midway between normal physiological pH and abnormal pH, extending to approximately the abnormal pH. As a result, the screening test is more efficient in finding more hits or candidate conditionally active peptides that exhibit high activity under abnormal conditions.

In some embodiments, the pKa can even differ from the anomalous pH by at least one pH unit. When the anomalous pH is acidic, the pKa of a suitable ion can be in the range from (abnormal pH - 1) to the midpoint between the anomalous pH and the normal physiological pH. When the anomalous pH is alkaline, the pKa of a suitable ion can be in the range from the midpoint between the anomalous pH and the normal physiological pH to (abnormal pH + 1). Ions can be selected from those described in this application. However, other ions not explicitly described in this application may also be used. It should be understood that once the anomalous pH and normal physiological pH for the screening test are selected, those skilled in the art can use the guidelines of this invention to select any ion with a suitable pKa to improve screening efficiency in identifying more hits that are highly active under anomalous conditions.

For example, when the exemplary screening target an abnormal pH of 8.4 and a normal physiological pH of 7.4, any ion with a pKa in the range of approximately 7.9 (midpoint) to 9.4 (i.e., 8.4+1) can be used for screening. Some ions in this pKa range include those derived from N-tris(hydroxymethyl)methylglycine (pKa 8.05), hydrazine (pKa 8.1), N,N-dihydroxyethylglycine (pKa 8.26), N-(2-hydroxyethyl)piperazine-N'-(4-butyrylic acid) (pKa 8.3), N-tris[hydroxymethyl]methyl-3-aminopropanesulfonic acid (pKa 8.4), and taurine (pKa 9.06). As another example, when the exemplary screening target an abnormal pH of 6 and a normal physiological pH of 7.4, any ion with a pKa in the range of approximately 5 (i.e., 6-1) to 6.7 (midpoint) can be used for screening. Some ions with pKa in this range include those derived from malate (pKa 5.13), pyridine (pKa 5.23), piperazine (pKa 5.33), carboxylate (pKa 6.27), succinate (pKa 5.64), 2-(N-morpholino)ethanesulfonic acid (pKa 6.10), citrate (pKa 6.4), histidine (pKa 6.04), and bis-tris (6.46). Those skilled in the art will be able to identify known compounds, including both inorganic and organic compounds, by referring to numerous chemical handbooks and textbooks. Among compounds with suitable pKa, those with smaller molecular weights may be preferred.

Therefore, this invention unexpectedly discovered that the production of conditionally active peptides ultimately identified depends not only on generating the correct peptide mutants but also on using ions with suitable pKa in the test solution. This invention envisions that, in addition to generating large mutant peptide libraries (e.g., via CPE and CPS), efforts should be made to find suitable ions (with appropriate pKa) for use in the test solution, as these ions can facilitate efficient selection of highly active mutants from large libraries. Furthermore, it is envisioned that without suitable ions, screening efficiency is low, and the likelihood of finding highly active mutants is reduced. Therefore, in the absence of suitable ions, multiple rounds of screening may be required to obtain the same number of highly active mutants.

The ions in the test solution can be formed in situ from the components of the test solution or are directly contained in the test solution. For example, _CO2 from the air can dissolve in the test solution to provide carbonate and bicarbonate ions. As another example, sodium dihydrogen phosphate can be added to the test solution to provide dihydrogen phosphate ions.

The concentration of this component in the test solution (for testing under a first condition or normal physiological condition and for testing under a second condition or abnormal condition) can be the same as or substantially the same as the concentration of the same component commonly found in the naturally occurring bodily fluids of mammals (e.g., humans). In other embodiments, the concentration of the component may be higher, particularly when the component is an ion capable of facilitating the binding between the mutant peptide and its binding partner, as higher concentrations of such ions have been observed to form ionic bonds with the mutant peptide and its binding partner, thus effectively promoting binding and increasing the likelihood of discovering more hits or candidate conditionally active peptides.

In some implementations, the concentration of ions in the test solution is positively correlated with the probability of detecting more hits using the test, especially when concentrations exceeding normal physiological levels are used. For example, the concentration of bicarbonate ions in human serum is approximately 15-30 mM. In one instance, as the concentration of bicarbonate ions in the test solution increased from 3 mM to 10 mM, to 20 mM, to 30 mM, to 50 mM, and to 100 mM, the number of hits in the test also increased with each increase in bicarbonate concentration. Therefore, the range of bicarbonate concentrations that can be used in the test solution is from about 3 mM to about 200 mM, or from about 5 mM to about 150 mM, or from about 5 mM to about 100 mM, or from about 10 mM to about 100 mM, or from about 20 mM to about 100 mM, or from about 25 mM to about 100 mM, or from about 30 mM to about 100 mM, or from about 35 mM to about 100 mM, or from about 40 mM to about 100 mM, or from about 50 mM to about 100 mM.

In another embodiment, the concentration of citrate in the test solution may be from about 30 μM to about 120 μM, or from about 40 μM to about 110 μM, or from about 50 μM to about 110 μM, or from about 60 μM to about 100 μM, or from about μM to about 90 μM, or about μM.

In one implementation, normal physiological conditions are a normal physiological pH of 7.2–7.6, and abnormal conditions are abnormal pH ranges of 5.5–7.2, 6–7, or 6.2–6.8. Test solutions used for testing under normal physiological conditions have a normal physiological pH and 50 mM bicarbonate ions. Test solutions used for testing under abnormal conditions have an abnormal pH and 50 mM bicarbonate ions. Because the pKa of bicarbonate ions is approximately 6.4, bicarbonate ions can facilitate the binding between the mutant peptide and its binding partner at abnormal pH levels of 6.0–6.4 (e.g., pH 6.0 or 6.2).

In another embodiment, normal physiological conditions are a normal physiological pH of 7.2–7.6, and abnormal conditions are abnormal pH of 5.5–7.2, 6–7, or 6.2–6.8. The test solution for testing under normal physiological conditions has a normal physiological pH and 80 μM citrate ions. The test solution for testing under abnormal conditions has an abnormal pH and 80 μM citrate ions. Since the pKa of citrate ions is 6.4, citrate ions can effectively facilitate the binding between mutant peptides and their binding partners in test solutions used under abnormal conditions (pH 6.0–6.4). Therefore, more candidate conditionally active peptides with higher binding activity at pH 6.0–6.4 and lower activity at pH 7.2–7.8 can be identified. Other ions, including acetate, histidine, bicarbonate, HATP, and HADP, function in a similar manner to enable test solutions containing these ions to effectively screen for mutant peptides that exhibit high binding activity at pH values near the ion's pKa and low binding activity at pH values different from the ion's pKa (e.g., normal physiological pH).

In another embodiment, normal physiological conditions are defined as a normal physiological temperature of 37°C, and abnormal conditions are defined as an abnormal temperature of 38-39°C (the temperature in certain tumor microenvironments). The test solution used for testing under normal physiological conditions has a normal physiological temperature and 20 mM bicarbonate ions. The test solution used for testing under abnormal conditions has an abnormal temperature and 20 mM bicarbonate ions.

In another implementation, normal physiological conditions are specific concentrations of electrolytes in normal human serum, while abnormal conditions are different, abnormal concentrations of the same electrolyte, which may be located in different locations in animals or humans, or may be caused by diseases in animals or humans that alter the normal physiological concentrations of electrolytes in human serum.

The binding between the mutant peptide and/or its binding partner can also be affected in many other ways. Typically, this effect is achieved by including one or more other components in the test solution. These other components can be designed to interact with the mutant peptide, the binding partner, or both. Furthermore, these other components can influence binding using combinations of two or more interactions, as well as combinations of two or more types of interactions.

In one embodiment, the binding interaction of interest is between an antibody and an antigen. In this embodiment, one or more other components may be included in the test solution to influence the antibody, antigen, or both. In this way, the desired binding interaction can be enhanced.

In addition to ions that can form ionic bonds with the mutant polypeptide and/or its binding partner to facilitate binding between the mutant polypeptide and its binding partner, the present invention also includes other components that can be used to assist binding of the mutant polypeptide to its binding partner. In one embodiment, molecules capable of forming hydrogen bonds with the mutant polypeptide and/or its binding partner are used. In another embodiment, molecules capable of hydrophobic interactions with the mutant polypeptide and/or its binding partner are used. In yet another embodiment, molecules capable of van der Waals interactions with the mutant polypeptide and/or its binding partner are contemplated.

As used in this article, the term "hydrogen bond" refers to a relatively weak non-covalent interaction between a hydrogen atom covalently bonded to an electronegative atom (such as carbon, nitrogen, oxygen, sulfur, chlorine, or fluorine) (hydrogen bond donor) and an electron donor atom with a non-shared electron pair (such as nitrogen, oxygen, sulfur, chlorine, or fluorine) (hydrogen bond acceptor).

Components capable of forming hydrogen bonds with mutant peptides and/or their binding partners include organic molecules and inorganic molecules with polar bonds. Mutant peptides and/or their binding partners typically contain amino acids capable of forming hydrogen bonds. Suitable amino acids have side chains containing polar groups capable of forming hydrogen bonds. Non-limiting examples of suitable amino acids include glutamine (Gln), glutamic acid (Glu), arginine (Arg), asparagine (Asn), aspartic acid (Asp), lysine (Lys), histidine (His), serine (Ser), threonine (Thr), tyrosine (Tyr), cysteine (Cys), methionine (Met), and tryptophan (Trp).

These amino acids can act as both hydrogen donors and acceptors. For example, the oxygen atom in the -OH group (e.g., found in Ser, Thr, and Tyr), the oxygen atom in the -C=O group (e.g., found in Glu and Asp), the sulfur atom in the -SH group or -SC- (e.g., found in Cys and Met), the nitrogen atom in the _-NH3 ⁺ group (e.g., found in Lys and Arg), and the nitrogen atom in the -NH- group (e.g., found in Trp, His, and Arg) can all act as hydrogen acceptors. Furthermore, groups containing hydrogen atoms in this list (e.g., -OH, -SH, _NH3 ⁺ , and -NH-) can act as hydrogen donors.

In some embodiments, the backbone of the mutant peptide and/or its binding partner may also participate in the formation of one or more hydrogen bonds. For example, the backbone may have a repeating structure of -(C=O)-NH-, such as in peptide bonds. The oxygen and nitrogen atoms in this structure can act as hydrogen acceptors, while the hydrogen atoms can participate in hydrogen bonding.

Inorganic compounds having at least one polar bond containing a hydrogen or oxygen atom that can bond with hydrogen can include, for example, _H₂O , _NH₃ , _H₂O₂ , _hydrazine , carbonates, sulfates, and phosphates. Organic compounds include alcohols; phenols; thiols; aliphatic, amine, and amide compounds; epoxides and carboxylic acids; ketones; aldehydes; ethers, esters; organochlorides, and organofluorines. Compounds capable of forming hydrogen bonds are well-known in the chemical literature, such as those discussed in, for example, "The Nature of the Chemical Bond," by Linus Pauling, Cornell University Press, 1940, pp. 284-334.

In some embodiments, alcohols may include methanol, ethanol, propanol, isopropanol, butanol, pentanol, 1-hexanol, 2-octanol, 1-decanol, cyclohexanol, and higher alcohols; diols such as ethylene glycol, propylene glycol, glycerol, diethylene glycol, and polyalkylene glycols. Suitable phenols include hydroquinone, resorcinol, catechol, phenol, o-cresol, m-cresol, and p-cresol, thymol, α-naphthol and β-naphthol, pyrogallol, guaiacol, and resorcinol. Suitable thiols include methanethiol, ethanethiol, 1-propanethiol, 2-propanethiol, butanethiol, tert-butanethiol, pentathiol, hexanethiol, thiophene, dimercaptosuccinic acid, 2-mercaptoethanol, and 2-mercaptoindole. Suitable amines include methylamine, ethylamine, propylamine, isopropylamine, aniline, dimethylamine and methylethylamine, trimethylamine, aziridine, piperidine, N-methylpiperidine, benzidine, cyclohexylamine, ethylenediamine, hexamethylenediamine, o-toluidine, m-toluidine and p-toluidine, and N-phenylpiperidine. Suitable amides include acetamide, N,N-dimethylacetamide, N,N-dimethylformamide, N,N-dimethylmethoxyacetamide, and N-methyl-N-p-cyanoethylformamide. Epoxides may include ethylene oxide, propylene oxide, tert-butyl hydroperoxide, styrene oxide, epoxide glycidol, cyclohexene oxide, di-tert-butyl peroxide, cumene hydroperoxide or ethylbenzene hydroperoxide, isobutylene oxide, and 1,2-epoxyoctane. Carboxylic acids may include terephthalic acid, isophthalic acid, phthalic acid, salicylic acid, benzoic acid, acetic acid, lauric acid, adipic acid, lactic acid, citric acid, acrylic acid, glycine, hexahydrobenzoic acid, o-toluic acid, m-toluic acid and p-toluic acid, nicotinic acid, isonicotinic acid and p-aminobenzoic acid. Ketones may include acetone, 3-propanone, butanone, pentanone, methyl ethyl ketone, diisobutyl ketone, ethylbutyl ketone, methyl isobutyl ketone, methyl tert-butyl ketone, cyclohexanone, acetone, methyl ethyl ketone, methyl propyl ketone, methyl butyl ketone, methyl pentanone, methyl hexyl ketone, diethyl ketone, ethyl butyl ketone, dipropyl ketone, diisobutyl ketone, diacetone alcohol, phorone, isophorone, cyclohexanone, methyl cyclohexanone and acetophenone. Aldehydes may include formaldehyde, acetaldehyde, propionaldehyde, butyraldehyde, benzaldehyde, cinnamaldehyde, isobutyraldehyde, pentanal, octanal, benzaldehyde, cinnamaldehyde, cyclohexanone, salicylaldehyde, and furfural. Esters include ethyl acetate, methyl acetate, ethyl formate, butyl acetate, ethyl lactate, ethyl butyrate, propyl acetate, ethyl formate, propyl formate, butyl formate, pentyl formate, methyl acetate, ethyl acetate, propyl acetate, butyl acetate, pentyl acetate, methyl isopentyl acetate, methoxybutyl acetate, hexyl acetate, cyclohexyl acetate, benzyl acetate, methyl propionate, ethyl propionate, butyl propionate, methyl butyrate, ethyl butyrate, butyl butyrate, pentyl butyrate, methyl acetoacetate, and ethyl acetoacetate, etc. Ethers that can be used in this invention include dimethyl ether, methyl ethyl ether, diethyl ether, methyl propyl ether, and dimethoxyethane. Ethers can be cyclic, such as ethylene oxide, tetrahydrofuran, and dioxane.

Organochlorinated compounds include chloroform, pentachloroethane, dichloromethane, trichloromethane, carbon tetrachloride, tetrachloromethane, tetrachloroethane, pentachloroethane, trichloroethylene, tetrachloroethylene, and dichloroethylene. Organofluorine compounds can include fluoromethane, difluoromethane, trifluoromethane, trifluoroethane, tetrafluoroethane, pentafluoroethane, difluoropropane, trifluoropropane, tetrafluoropropane, pentafluoropropane, hexafluoropropane, and heptafluoropropane.

Hydrogen bonds can be classified into strong, moderate, or weak hydrogen bonds according to their bond strength (Jeffrey, George A.; An introduction to hydrogen bonding, Oxford University Press, 1997). Strong hydrogen bonds have donor-acceptor distances with energies in the range of 14-40 kcal/mol. Moderate hydrogen bonds have donor-acceptor distances with energies in the range of 4-15 kcal/mol. Weak hydrogen bonds have donor-acceptor distances with energies in the range of <4 kcal/mol. Some examples of hydrogen bonds with energy levels are F-H…:F (38.6 kcal/mol), O-H…:N (6.9 kcal/mol), O-H…:O (5.0 kcal/mol), N-H…:N (3.1 kcal/mol), and N-H…:O (1.9 kcal/mol). For more examples, see Perrin et al. “Strong” hydrogen bonds in chemistry andbiology, Annual Review of Physical Chemistry, vol.48, pp.511-54 4,1997; Guthrie, "Short strong hydrogen bonds:can they explain enzymic catalysis?" Chemistry&BiologyMarch 1996,3:163-170.

In some embodiments, the components used in this invention can form strong hydrogen bonds with mutant peptides and/or their binding partners. These components often contain atoms with strong electronegativity. The atoms with the strongest electronegativity are known to be F>O>Cl>N in that order. Therefore, this invention preferably uses organic compounds containing fluorine, hydroxyl, or carbonyl groups to form hydrogen bonds. In one embodiment, an organofluorine compound can be used in this invention to form strong hydrogen bonds.

In another embodiment, components capable of hydrophobic interactions with the mutant peptide and/or its binding partner are used. These components include organic compounds having hydrophobic groups.

As used in this paper, the term "hydrophobic interaction" refers to a reversible attractive interaction between a hydrophobic compound or a hydrophobic region of a compound and another hydrophobic compound or another compound. This interaction has been described in the literature "Hydrophobic Interactions," A. Ben-Nairn (1980), Plenum Press, New York.

Hydrophobic substances are repelled by water molecules due to their nonpolarity. When relatively nonpolar molecules or groups in an aqueous solution associate with other nonpolar molecules but not with water, it is called a "hydrophobic interaction".

Mutant polypeptides and their binding partners typically contain amino acids capable of hydrophobic interactions. These amino acids are typically characterized by having at least one side chain containing a nonpolar group capable of hydrophobic interactions. Hydrophobic amino acids include, for example, alanine (Ala), isoleucine (Ile), leucine (Leu), phenylalanine (Phe), valine (Val), proline (Pro), glycine (Gly), and to a lesser extent, methionine (Met) and tryptophan (Trp).

Components capable of hydrophobic interactions with mutant proteins and/or their binding partners include organic compounds that are hydrophobic molecules or molecules containing at least one hydrophobic moiety. In some embodiments, these hydrophobic components may be hydrocarbons selected from aromatics, substituted aromatics, polyaromatics, aromatic or non-aromatic heterocycles, cycloalkanes, alkanes, alkenes, and alkynes. Hydrophobic groups may include aryl, alkyl, cycloalkyl, alkenyl, and alkynyl groups. As used herein, the terms "alkyl," "alkenyl," and "alkynyl" refer to unsaturated aliphatic groups having one to thirty carbon atoms, including straight-chain alkenyl/alkynyl, branched alkenyl/alkynyl, cycloalkenyl (alicyclic) groups, alkyl-substituted cycloalkyl, and cycloalkyl-substituted alkenyl/alkynyl groups. Such hydrocarbon moiety may also be substituted at one or more carbon atoms.

It is understood that the strength of hydrophobic interactions is based on the available amount of “hydrophobic” components that can interact with each other. Therefore, hydrophobic interactions can be modulated, for example, by increasing the amount and/or “hydrophobicity” of the hydrophobic portion in the molecule involved in the interaction. For example, the hydrophobic portion (whose original form may include a hydrocarbon chain) can be modified to increase its hydrophobicity (the ability to increase the strength of the hydrophobic interactions involved) by giving it a hydrophobic side chain attached to one of the carbon atoms of its carbon backbone. In a preferred embodiment of the invention, this may include the linking of various polycyclic compounds, including, for example, various steroid compounds and/or their derivatives, such as sterol-type compounds, more specifically cholesterol. Typically, the side chain can be linear, aromatic, aliphatic, cyclic, polycyclic, or any other type of hydrophobic side chain that would be conceived by those skilled in the art.

The type of component capable of van der Waals interactions with mutant peptides and/or their binding partners is typically, but not always, a compound with a polar portion. As used herein, “van der Waals interaction” refers to the attractive force between atoms, parts, molecules, and surfaces, arising from dipole-dipole interactions and/or correlations that occur as a result of quantum dynamics in the wave polarization of neighboring atoms, parts, or molecules.

The van der Waals interactions of this invention are the attractive forces between mutant peptides or binding partners and their components. Van der Waals interactions can originate from three sources. First, some molecules/parts, while electrically neutral, can be permanent electric dipoles. Due to the fixed deformation of the electron charge distribution in the structure of some molecules/parts, one side of the molecule/part is always slightly positively charged, while the opposite side is slightly negatively charged. This tendency of permanent dipoles to align with each other results in a net attractive force. This is the interaction between two permanent dipoles (Keesom force).

Second, the presence of a molecule as a permanent dipole can temporarily deform the electron charge in other nearby polar or nonpolar molecules, thereby causing further polarization. The interaction between the permanent dipole and the neighboring induced dipole produces another attractive force. This interaction between the permanent dipole and the corresponding induced dipole can be called the Debye force. Third, although the involved molecules are not permanent dipoles (e.g., the organic liquid benzene), there is an attractive force between molecules with two transient induced dipoles. This interaction between two transient induced dipoles can be called the London dispersion force.

Mutant peptides and/or binding partners contain numerous amino acids capable of van der Waals interactions. These amino acids can have polar side chains, including glutamine (Gln), asparagine (Asn), histidine (His), serine (Ser), threonine (Thr), tyrosine (Tyr), cysteine (Cys), methionine (Met), and tryptophan (Trp). These amino acids can also have side chains containing nonpolar groups, including alanine (Ala), isoleucine (Ile), leucine (Leu), phenylalanine (Phe), valine (Val), proline (Pro), and glycine (Gly).

Components capable of van der Waals interactions with mutant peptides and/or their binding partners include polar or nonpolar inorganic compounds soluble in the test solution. The test solution is typically an aqueous solution, therefore these polar or nonpolar inorganic compounds are preferably soluble in water. Preferred substances for van der Waals interactions are polar substances, such that they can undergo dipole-dipole interactions. For example, AIF ₃ has a polar Al-F bond and is soluble in water (approximately 0.67 g/100 ml water at 20 °C). HgCl ₂ has a polar Hg-Cl bond and is soluble in water at a concentration of 7.4 g/100 ml water at 20 °C. PrCl ₂ has a polar Pr-Cl bond and is soluble in water at a concentration of approximately 1 g/100 ml water at 20 °C.

Suitable polar compounds capable of van der Waals interactions include alcohols, thiols, ketones, amines, amides, esters, ethers, and aldehydes. Suitable examples of these compounds have been described above in the section on hydrogen bonding. Suitable nonpolar compounds capable of van der Waals interactions include aromatics, substituted aromatics, polyaromatics, aromatic or non-aromatic heterocycles, cycloalkanes, alkanes, alkenes, and alkynes.

The binding of a mutant peptide to its binding partner can be influenced in a variety of ways using hydrogen-bonding components, hydrophobic components, and van der Waals components. In one embodiment, hydrogen bonds, hydrophobic interactions, and/or van der Waals interactions can form a bridge between the mutant peptide and its binding partner. Such a bridge can bring the mutant peptide and the binding partner closer to each other to facilitate binding and/or position the mutant peptide and/or the binding partner relative to each other in a manner that promotes binding.

In another embodiment, hydrogen bonding and/or hydrophobic interactions can increase the likelihood of the mutant peptide binding to its binding partner, for example, by causing the peptide and binding partner to aggregate or associate with each other in a manner that increases the likelihood of binding. Therefore, one or more of these interactions can be used alone or in combination to cause the mutant peptide and binding partner to aggregate more closely together, or to arrange the mutant peptide and binding partner in a manner that promotes binding, for example, by bringing the binding sites closer together or arranging the non-binding portions of the molecule further apart, thereby bringing the binding sites closer together.

In another embodiment, hydrogen bonding and/or hydrophobic interactions can influence the conformation of the mutant peptide and/or its binding partner to provide a conformation more favorable to the binding of the mutant peptide to its binding partner. Specifically, binding or interaction with one or more amino acids of the mutant peptide and/or binding partner may cause one or more conformational changes in the mutant peptide or binding partner that are favorable to the mutant peptide/binding partner binding reaction.

This invention involves two pairs of tests. One pair of tests is to find mutant peptides with the following characteristics: in tests under normal physiological conditions, the mutant peptide exhibits reduced activity compared to the activity of the parent peptide from which the mutant peptide is derived under the same normal physiological conditions. The second pair of tests is to find mutant uniqueness with the following characteristics: in tests under abnormal conditions, the mutant peptide exhibits enhanced activity compared to the activity of the parent peptide from which the mutant peptide is derived under the same abnormal conditions.

The conditions in the two pairs of tests used in this invention can be selected from temperature, pH, osmotic pressure, molality, oxidative stress, electrolyte concentration, and the concentration of any other component of the test solution or culture medium. Therefore, a specific component of the test medium can be used in both pairs of tests at substantially the same concentration. In this case, the presence of this component is typically used to simulate a specific environment in a human or animal, such as serum, tumor microenvironment, muscle environment, neural environment, or any other environment that may be encountered at the application site, any other environment that the applied treatment may experience, or any other environment that may be encountered at the treatment site. An important aspect of selecting one or more components to simulate these environments is that it can improve the results of the selection process using the two pairs of tests. For example, simulating a specific environment allows the specific components of that environment to be obtained during the selection process to assess various effects of the mutant peptide. Components of the specific environment can, for example, alter or bind to the mutant peptide, inhibit the activity of the mutant peptide, inactivate the mutant peptide, etc.

In some embodiments, one or more components of the test solution are preferably small compounds, such as hydrosulfide, hydrogen sulfide, histidine, histamine, citrate, bicarbonate, lactate, and acetate. In one embodiment, the small molecule component is preferably present in the test solution at a concentration of about 100 μM to about 100 mM, or more preferably about 0.5 mM to about 50 mM, or about 1 mM to about 10 mM.

The concentration of a component in the test solution may be the same as or substantially the same as the concentration of the same component typically found in the naturally occurring body fluids of mammals (e.g., humans). This concentration may be referred to as the normal physiological concentration of that component in the body fluid. In other embodiments, the concentration of a particular component in the test solution may be less than or greater than the concentration of the same component typically found in the naturally occurring body fluids of mammals (e.g., humans).

In another embodiment, the component may be present at significantly different concentrations in each pair of tests. In this case, the presence, absence, or concentration of the component becomes the condition being tested, because it is precisely the concentration of the component that creates the difference between the test solution used for testing under normal physiological conditions and the test solution used for testing under abnormal conditions. Therefore, the conditionally active peptides produced by this embodiment of the method according to the invention will be selected to obtain conditionally active peptides whose activity is at least partially dependent on the concentration of said component.

In some implementations, a component may be present in one pair of test solutions but completely absent in another pair. For example, the lactate concentration in the test solution under abnormal conditions may be set to a level that simulates the lactate concentration in the tumor microenvironment. In a pair of test solutions used for normal physiological conditions, lactate may be absent.

In one implementation, normal physiological condition is a first lactate concentration representing normal physiological condition, and abnormal condition is a second lactate concentration representing abnormal condition present at a specific location in the body.

In another example, glucose may be absent in the test solution under anomalous conditions to simulate glucose deficiency that may be found in the tumor microenvironment, while in a pair of test solutions used for normal physiological conditions, glucose can be set to a level that simulates plasma glucose concentration. This feature can be used to preferentially deliver conditionally active peptides to the location or environment where they are inactive or have minimal activity during transport, and are activated when they arrive at an environment where the component concentration in the test solution for anomalous conditions is present.

For example, the tumor microenvironment typically has a lower glucose concentration and a higher lactate concentration compared to human serum. The normal physiological concentration of glucose in serum ranges from about 2.5 mM to about 10 mM. On the other hand, the glucose concentration in the tumor microenvironment is typically very low, ranging from 0.05 mM to 0.5 mM. In one embodiment, the test solution for testing under normal physiological conditions has a glucose concentration ranging from about 2.5 mM to about 10 mM, and the test solution for testing under abnormal conditions has a glucose concentration ranging from about 0.05 mM to about 0.5 mM. The resulting conditionally active peptide exhibits higher activity in a low glucose environment (in the tumor microenvironment) than in a high glucose environment (in normal tissue or blood). This conditionally active peptide will function in the tumor microenvironment but has lower activity during transport in the bloodstream.

Normal physiological concentrations of lactate in serum range from about 1 mM to about 2 mM. On the other hand, in the tumor microenvironment, lactate concentrations are typically in the range of 10 mM to 20 mM. In one embodiment, the test solution for testing under normal physiological conditions has a lactate concentration of about 1 mM to about 2 mM, and the test solution for testing under abnormal conditions has a lactate concentration of about 10 mM to about 20 mM. The resulting conditionally active peptide exhibits higher activity in a high lactate environment (in the tumor microenvironment) than in a low lactate environment (in normal tissue or blood). Therefore, the conditionally active peptide functions in the tumor microenvironment but exhibits lower activity during transport in the bloodstream.

Similarly, painful muscles are known to have a higher (abnormal) lactate concentration than normal muscles. Therefore, when seeking mutant peptides active in a painful muscle environment, a pair of tests under aberrant conditions can be performed in the presence of higher lactate concentrations to mimic the painful muscle environment, while a pair of tests under normal physiological conditions can be performed in the presence of lower lactate concentrations or in the absence of lactate. In this way, mutant peptides with enhanced activity in a painful muscle environment with increased lactate concentrations can be selected. Such conditionally active peptides can be used, for example, as anti-inflammatory agents.

In another embodiment, two or more components can be used in two pairs of test solutions. In this type of test, the characteristics of the two tests described above can be used to select conditionally active peptides. Alternatively, two or more components can be used to increase the selectivity of conditionally active peptides. For example, with regard to the tumor microenvironment, a pair of tests under abnormal conditions can be performed in a test medium with a high lactate concentration and a low glucose concentration, while the corresponding pair of tests under normal physiological conditions can be performed in a test medium with a relatively low lactate concentration and a relatively high glucose concentration.

The present invention envisions that each component selected from inorganic compounds, ions, and organic molecules can be used alone or in combination to select conditionally active peptides that are more active at one concentration of the component than at different concentrations of the same component.

Tests that rely on different concentrations of one or more metabolites as a distinguishing condition between normal environments (normal physiological conditions) and abnormal environments (abnormal conditions) may be particularly suitable for selecting conditionally active peptides that are more active in the tumor microenvironment than in plasma, since the tumor microenvironment typically has a significant number of metabolites at concentrations different from the same metabolites in plasma.

The literature Kinoshita et al., "Absolute Concentrations of Metabolites in Human Brain Tumors Using In Vitro Proton Magnetic Resonance Spectroscopy," NMR INBIOMEDICINE, vol. 10, pp. 2-12, 1997, compared metabolites in normal brain and brain tumors. This group found that the concentration of N-acetylaspartate in normal brain was 5000-6000 μM, but only 300-400 μM in glioblastoma, 1500-2000 μM in astrocytoma, and 600-1500 μM in anaplastic astrocytoma. Furthermore, the concentration of inositol in normal brain is 1500-2000 μM, in glioblastoma it is 2500-4000 μM, in astrocytoma it is 2700-4500 μM, and in anaplastic astrocytoma it is 3800-5800 μM. The concentration of phosphatidylethanolamine in normal brain is 900-1200 μM, but in glioblastoma it is 2000-2800 μM, in astrocytoma it is 1170-1370 μM, and in anaplastic astrocytoma it is 1500-2500 μM. The concentration of glycine in the normal brain is 600-1100 μM, but in glioblastoma it is 4500-5500 μM, in astrocytoma it is 750-1100 μM, and in anaplastic astrocytoma it is 1900-3500 μM. Alanine has a concentration of 700-1150 μM in the normal brain, but in glioblastoma it is 2900-3600 μM, in astrocytoma it is 800-1200 μM, and in anaplastic astrocytoma it is 300-700 μM. These metabolites may also have different concentrations in the blood; for example, N-acetylaspartate has a blood concentration of approximately 85000 μM, inositol approximately 21700 μM, glycine approximately 220-400 μM, and alanine approximately 220-300 μM.

Therefore, these metabolites, including at least N-acetylaspartic acid, inositol, glycine, and alanine, can be used in test solutions at different concentrations to select conditionally active peptides that are active in brain tumors but inactive in blood or normal brain tissue. For example, a test solution with an N-acetylaspartic acid concentration of 85,000 μM can be used for a pair of tests under normal physiological conditions, and a test solution with an N-acetylaspartic acid concentration of 350 μM can be used for a pair of tests under abnormal conditions to select conditionally active peptides that are active in the tumor microenvironment of glioblastoma but inactive or at least have low activity in blood or normal brain tissue.

The literature Mayers et al., "Elevated circulating branched chain amino acids are an early event in pancreatic adenocarcinoma development," *Nature Medicine*, vol. 20, pp. 1193-1198, 2014, investigated the concentrations of various metabolites, including branched-chain amino acids, in the pre-diagnosis plasma of pancreatic patients. They found that the concentrations of several metabolites in the bloodstream of pancreatic cancer patients differed from the concentrations of the same metabolites in the blood of individuals without pancreatic cancer. Mayers et al. also found that the concentrations of branched-chain amino acids in the plasma of pancreatic cancer patients were significantly increased compared to normal individuals. Branched-chain amino acids present at elevated concentrations included isoleucine, amino acids, and valine (Table 1 of Mayers et al.). Figure 1 from Mayers' work shows that the concentrations of other metabolites in the plasma of pancreatic cancer patients differed significantly from those in healthy individuals. These metabolites include at least acetylglycine, glycine, phenylalanine, tyrosine, 2-aminoadipic acid, taurideoxycholic acid/tauridechenodeoxycholate, aconitate, isocitrate, lactic acid, α-glycerophosphate, and uric acid. Therefore, based on the finding that certain metabolites are present at different concentrations in the plasma of pancreatic cancer patients and healthy individuals, it can be predicted that the concentrations of these metabolites in the tumor microenvironment of pancreatic cancer will also differ from their concentrations in the pancreatic microenvironment of healthy individuals.

Therefore, in one embodiment, one or more of these metabolites can be used in a test solution under normal physiological conditions at concentrations close to those found in the plasma of a healthy individual (i.e., normal physiological concentrations of the metabolites). For example, the known normal physiological concentrations of isoleucine in the plasma of a healthy individual are approximately 1.60 ± 0.31 mg/dL, leucine is approximately 1.91 ± 0.34 mg/dL, and valine is approximately 2.83 ± 0.34 mg/dL. The test solution under normal physiological conditions can have normal physiological concentrations of one or more of these branched-chain amino acids within the aforementioned range. The concentration of the same branched-chain amino acid in the test solution under abnormal conditions is approximately 5 times, or approximately 10 times, or approximately 20 times, or approximately 50 times, or approximately 70 times, or approximately 100 times, or approximately 150 times, or approximately 200 times, or approximately 500 times the normal physiological concentration of the corresponding branched-chain amino acid in a healthy individual. Since the high concentrations of these branched-chain amino acids found in plasma by Mayers et al. originated from the tumor microenvironment and were diluted in the bloodstream, their findings suggest that the concentrations of these branched-chain amino acids in the pancreatic tumor microenvironment are expected to be significantly elevated. Similarly, even if the concentrations of specific metabolites in cancer patients are significantly lower than in healthy individuals, tests under abnormal conditions can still reflect the concentrations of other metabolites in the blood of pancreatic cancer patients. In this way, screening can simulate the actual environment, thereby ensuring the selection of mutants with the highest activity in that specific environment.

In some other embodiments, the test solution for normal physiological conditions may contain one or more branched-chain amino acids at concentrations that mimic the concentrations found in the plasma of pancreatic cancer patients, thereby simulating the actual plasma environment of these patients. In such embodiments, the test solution for abnormal conditions may contain the same branched-chain amino acids at concentrations approximately 2, 3, 4, 5, 7, 8, 10, 15, 20, or 50 times higher than the corresponding branched-chain amino acid concentrations in the plasma of pancreatic cancer patients, to reflect that these higher concentrations originate from the tumor microenvironment and that the concentrations in the bloodstream represent a dilution of the actual concentrations in the tumor microenvironment. Similarly, other metabolites may also have different concentrations in the test solutions for normal physiological conditions and those for abnormal conditions to reflect the expected actual differences based on data collected from the bloodstream. In some cases, a deficiency of specific metabolites may be observed in the bloodstream of pancreatic patients. In such cases, concentrations reflecting the levels measured in the bloodstream can be used in tests under normal physiological conditions, and even lower concentrations can be used in tests under abnormal conditions to account for the anticipated consumption of the metabolites in the tumor microenvironment. Conditionally active peptides selected using these test solutions will be more active in the pancreatic cancer microenvironment than in the plasma of pancreatic cancer patients.

In some embodiments, whole plasma from pancreatic cancer patients can be used in this invention. For example, in one embodiment, one or more components of pancreatic cancer patient plasma can be simulated in the test solution for one or both of the tests for normal physiological conditions and abnormal conditions. In one exemplary embodiment, the pH of the test solution for normal physiological conditions is in the range of 7.2-7.6, in which 30 wt% of pancreatic cancer patient plasma is added, and the pH of the test solution for abnormal conditions is in the range of 6.2-6.8, in which 30 wt% of pancreatic cancer patient plasma is added. In this embodiment, the presence of pancreatic cancer patient plasma ensures that (1) the conditionally active peptide is not activated in blood at a pH of 7.2-7.6, even in the presence of a combination of metabolites found in the blood of pancreatic cancer patients, and (2) also ensures that the conditionally active peptide can be activated in the tumor microenvironment at a pH of 5.5-7.2, 6-7, or 6.2-6.8. This will allow for customized treatment for pancreatic cancer patients.

In another exemplary embodiment, the pH of the test solution under normal physiological conditions is in the range of 7.2-7.6, wherein 30 wt% of plasma from a pancreatic cancer patient is added, and the pH of the test solution under abnormal conditions is in the range of 5.5-7.2 or 6.2-6.8, wherein no plasma from a pancreatic cancer patient is added.

The same components selected from inorganic compounds, ions, and organic molecules can be used in each of the aforementioned types of tests. For example, in the case of lactate, lactate can be used at substantially the same concentration in test solution pairs under normal physiological and abnormal conditions. However, normal physiological and abnormal conditions will differ in one or more other aspects (e.g., temperature, pH, concentration of another component, etc.). In different implementations, lactate can be used as one of the differentiating factors between normal and abnormal conditions to reflect the fact that the concentration of lactate in the abnormal tumor microenvironment is higher than that in the normal physiological condition (non-tumor microenvironment).

In some embodiments, two or more components are added at substantially the same concentration to two test solutions for normal physiological conditions and abnormal conditions. For example, both citrate and bovine serum albumin (BSA) are added to the test solutions. In both test solutions, the concentration of citrate may be approximately 80 μM, and the concentration of BSA may be approximately 10-20%. More specifically, the test solutions for a pair of tests under normal physiological conditions may have a pH in the range of 7.2-7.6, with a citrate concentration of approximately 80 μM and a BSA concentration of approximately 10-20%. The test solutions for a pair of tests under abnormal conditions may have a pH in the range of 6.2-6.8, with a citrate concentration of approximately 80 μM and a BSA concentration of approximately 10-20%.

In one embodiment, human serum can be added to both the normal physiological condition and abnormal condition test solutions at substantially the same concentration. Because human serum contains a large number of inorganic compounds, ions, and organic molecules (including polypeptides), the test solutions will have a variety and a large number of components selected from inorganic compounds, ions, and organic molecules, which are present at substantially the same concentration in both test solutions. The test solutions may contain 5-30 vol%, or 7-25 vol%, or 10-20 vol%, or 10-15 vol% serum. In some other embodiments, the test solutions for normal physiological conditions and abnormal conditions do not contain serum. The serum can be human serum, bovine serum albumin, or serum from any other mammal. In some other embodiments, the test solutions do not contain serum.

Test solutions under normal physiological conditions and test solutions under abnormal conditions can have different pH values. By using bicarbonate, the pH of this test solution can be adjusted using the _CO₂ and _O₂ levels in the buffer solution.

In some other embodiments, at least one of two or more components is added at different concentrations to a test solution under normal physiological conditions and a test solution under abnormal conditions. For example, both lactate and bovine serum albumin (BSA) are added to the test solution. The lactate concentration in the test solution under normal physiological conditions and the test solution under abnormal conditions may differ, while the BSA concentration may be the same in both test solutions. In the test solution under abnormal conditions, the lactate concentration may be in the range of 30-50 mg/dL, and in the test solution under normal physiological conditions, the lactate concentration may be in the range of 8-15 mg/dL. On the other hand, the BSA concentration is the same in both test solutions, for example, about 10-20%. In the presence of BSA, conditionally active peptides selected using these test solutions are more active at a high lactate concentration of 30-50 mg/dL than at a low lactate concentration of 8-15 mg/dL.

In some embodiments, test solutions can be designed to select conditionally active peptides that exhibit activity dependent on two or more conditions. In one exemplary embodiment, the conditionally active peptide may exhibit activity dependent on both pH and lactate. The test solution used to select such a conditionally active peptide may be a test solution under normal physiological conditions with a pH of 7.2-7.6 and a lactate concentration in the range of 8-15 mg/dL. A test solution under abnormal conditions may have a pH of 6.2-6.8 and a lactate concentration in the range of 30-50 mg/dL. Optionally, both the test solutions under normal physiological conditions and those under abnormal conditions may further contain ions to facilitate binding between the mutant peptide and its binding partner, thereby increasing the number of hits for the candidate bioactive peptide.

In another exemplary embodiment, the conditionally active peptide may have pH-, glucose, and lactate-dependent activities. The test solution used to select such a conditionally active peptide may be a test solution under normal physiological conditions with a pH of 7.2-7.6, a glucose concentration in the range of 2.5-10 mM, and a lactate concentration in the range of 8-15 mg/dL. The test solution under abnormal conditions may have a pH of 6.2-6.8, a glucose concentration in the range of 0.05-0.5 mM, and a lactate concentration in the range of 30-50 mg/dL. Optionally, both the test solution under normal physiological conditions and the test solution under abnormal conditions may further contain ions to facilitate the binding between the mutant peptide and its binding partner, thereby increasing the number of candidate bioactive peptides that bind to the binding partner under conditions of pH 6.2-6.8. The conditionally active peptides selected using this test solution are more active in an environment with pH 6.2-6.8, glucose concentration of 0.05-0.5 mM, and lactate concentration of 30-50 mg/dL than in an environment with pH 7.2-7.6, glucose concentration of 2.5-10 mM, and lactate concentration of 8-15 mg/dL.

Two or more components selected from inorganic compounds, ions, and organic molecules are used to prepare an anomalous condition test solution that simulates the environment at the location/site (i.e., target site) where the selected active peptide will be delivered. In some embodiments, at least three components present in the environment at the target site may be added to the test solution, or at least four components present in the environment at the target site may be added to the test solution, or at least five components present in the environment at the target site may be added to the test solution, or at least six components present in the environment at the target site may be added to the test solution.

In one embodiment, fluid extracted from a target site (where the conditionally active peptide will be more active) can be used directly as a test solution for testing under abnormal conditions. For example, synovial fluid can be extracted from an individual, preferably from an individual with a joint disease requiring treatment. The extracted synovial fluid, optionally diluted, can be used as a test solution in a pair of tests under abnormal conditions to select the conditionally active peptide. By using the extracted synovial fluid, optionally diluted, as a test solution for testing under abnormal conditions and as a test solution simulating human plasma for testing under normal physiological conditions, the selected conditionally active peptide (e.g., TNF-α) is more active at the joint than at other sites or organs. For example, TNF-α can be used to treat individuals with inflammatory joints (e.g., arthritis). However, TNF-α often has serious side effects that damage other tissues and organs. Conditionally active TNF-α, which is more active in synovial fluid but inactive or less active in the blood, can deliver the activity of TNF-α to the joint while mitigating or potentially eliminating the side effects of TNF-α on the rest of the body.

The development of conditionally active peptides with activity dependent on multiple conditions will improve the selectivity of conditionally active peptides for target sites in an individual. Ideally, conditionally active peptides are inactive or at least significantly less active at other sites where only some of these conditions are present. In one embodiment, conditionally active peptides active under conditions of pH 6.2-6.8, glucose concentration of 0.05-0.5 mM, and lactate concentration of 30-50 mg/dL can be specifically delivered to the tumor microenvironment, as all of these conditions are present in the tumor microenvironment. Other tissues or organs may contain one or two of these conditions, but not all three, and therefore the conditionally active peptide may not be fully activated in other tissues or organs. For example, post-exercise muscle may have a low pH in the range of 6.2-6.8. However, there is no other testing condition. Therefore, the conditionally active peptide is inactive or at least less active in post-exercise muscle.

In some implementations, steps can be taken to confirm that the activity of the conditionally active peptide is indeed dependent on the conditions used to select the conditionally active peptide. For example, the conditionally active peptide is selected to depend on three conditions: pH 6.2–6.8, a glucose concentration of 0.05–0.5 mM, and a lactate concentration of 30–50 mg/dL. The selected conditionally active peptide can then be tested individually under each of the three conditions and in multiple environments having two of the three conditions to confirm that the conditionally active peptide is inactive or has low activity in these tests.

In some implementations, certain components of serum may be intentionally reduced or removed from the test medium. For example, when screening for antibodies, serum components that bind or adsorb antibodies may be reduced or removed from the test medium. Such bound antibodies may produce false positives, including mutant antibodies that lack conditional activity and bind only to components present in serum under various conditions. Therefore, by carefully selecting test components to reduce or remove those that may bind to mutants in the test, this method can reduce the number of nonfunctional mutants that may be unintentionally identified as conditionally active positives because they bind to components rather than the desired binding pair in the test. For example, in some implementations of screening mutant peptides that tend to bind to components in human serum, BSA may be used in the test solution to reduce or eliminate the possibility of false positives caused by the binding of mutant peptides to components in human serum. Other similar alternatives may be made in certain cases to achieve the same objective.

In some implementations, the screening environment mimics the environment near the cell membrane, such as the environment inside, at, or outside the cell membrane, or the environment within a joint. When screening in a cell membrane environment, factors that may affect binding activity include receptor expression, internalization, and antibody-drug complex (ADC) titer.

The test can be any suitable test known to those skilled in the art. Examples include ELISA, enzyme activity assays, in vitro screening of real tissues (organs, etc.), tissue sections, whole animals, cell lines, and the use of 3D systems. For example, WO2013/040445 describes a test based on suitable cells, US 7,993,271 describes a tissue-based test, US2010/0263599 describes a whole-animal-based screening method, and US2011/0143960 describes a 3D system-based screening method.

In some embodiments, the evolution step can produce mutant peptides that simultaneously possess other desired properties besides the aforementioned conditionally active characteristics. Suitable other desired properties that can be evolved may include binding activity, expression, humanization, etc. Therefore, the present invention can be used to produce conditionally active peptides that also possess at least one or more of these improved other desired properties.

In some embodiments, the present invention generates conditionally active peptides. In, for example, a second evolution step, the selected conditionally active peptide may be further mutated using a mutagenesis technique disclosed herein to improve another property of the selected conditionally active peptide, such as binding activity, expression, humanization, etc. Following the second evolution step, the mutant peptides can be screened to obtain conditional activity and improved properties.

In some embodiments, after evolving the parent peptide to generate a mutant peptide, a first conditionally active peptide is selected that exhibits both of the following characteristics: (a) decreased first activity compared to the parent peptide in tests under normal physiological conditions, and (b) increased first activity compared to the parent peptide in tests under abnormal conditions. The first conditionally active peptide can then be further subjected to one or more other evolution, expression, and selection steps to select at least a second conditionally active peptide that (1) exhibits both of the following characteristics: (a) decreased second activity compared to the parent peptide in tests under normal physiological conditions, and (b) increased second activity compared to the parent peptide in tests under abnormal conditions; or (2) a greater ratio of first activity under abnormal conditions to first activity under normal physiological conditions compared to the first conditionally active peptide and/or the parent peptide. It is noted that the second conditionally active peptide may have higher first and second activities under abnormal conditions compared to the parent peptide, and lower first and second activities under normal physiological conditions compared to the parent peptide.

In some embodiments, the present invention aims to produce conditionally active peptides that have a larger ratio of activity under abnormal conditions (or a second condition) to activity under normal physiological conditions (or a first condition) (e.g., greater selectivity between abnormal and normal physiological conditions). The ratio or selectivity of activity under abnormal conditions (or the second condition) to activity under normal physiological conditions (or the first condition) may be at least about 2:1, or at least about 3:1, or at least about 4:1, or at least about 5:1, or at least about 6:1, or at least about 7:1, or at least about 8:1, or at least about 9:1, or at least about 10:1, or at least about 11:1, or at least about 12:1, or at least about 13:1, or at least about 14:1, or at least about 15:1, or at least about 16:1, or at least about 17:1, or at least about 18:1, or at least about 19:1, or at least about 20:1, or at least about 30:1, or at least about 40:1, or at least about 50:1, or at least about 60:1, or at least about 70:1, or at least about 80:1, or at least about 90:1, or at least about 100:1.

In one embodiment, the conditionally active peptide is an antibody whose activity under abnormal conditions is in a ratio of at least about 5:1, or at least about 6:1, or at least about 7:1, or at least about 8:1, or at least about 9:1, or at least about 10:1, or at least about 15:1, or at least about 20:1, or at least about 40:1, or at least about 80:1. In one embodiment, the conditionally active peptide is used to target tumor sites, wherein the conditionally active peptide is active at the tumor site (tumor microenvironment) but has significantly lower or no activity at non-tumor sites (normal physiological conditions).

In one embodiment, the conditionally active polypeptide is an antibody intended to conjugate with another reagent (such as those disclosed elsewhere herein). The conditionally active antibody exhibits a higher ratio of activity under abnormal conditions to activity under normal physiological conditions. For example, the ratio of activity under abnormal conditions to activity under normal physiological conditions of the conditionally active antibody to be conjugated with the other reagent is at least about 10:1, or at least about 11:1, or at least about 12:1, or at least about 13:1, or at least about 14:1, or at least about 15:1, or at least about 16:1, or at least about 17:1, or at least about 18:1, or at least about 19:1, or at least about 20:1. This is particularly important when the conjugated reagent is, for example, toxic or radioactive, as it is desirable that such a conjugated reagent be concentrated at the site of disease or treatment (where abnormal conditions exist).

G. Production of conditionally active peptides

It can produce selectable conditionally active peptides with reversible or irreversible activity for therapeutic, diagnostic, research and related purposes, and/or enable these conditionally active peptides to undergo one or more additional evolution and selection cycles.

Peptides can be used to express cells to produce conditionally active peptides from hosts or organisms. To make the production process more efficient, the DNA encoding the conditionally active peptide can be codon-optimized for cell production, hosts, or organisms. Codon optimization has been previously described, for example, in Narum et al., "Codon optimization of gene fragments encoding Plasmodium falciparum merozoite proteins enhances DNA vaccine protein expression and immunogenicity in mice," Infect. Immun. 2001 December, 69(12). :7250-3, which describes codon optimization in the mouse expression system; Outchkourov et al., "Optimization of the expression of Equistatin in Pichiapastoris, protein expression and purification", ProteinExpr. Purif. 2002 February; 24(1):18-24, which describes codon optimization in the yeast system; Feng et al., "High level expr "Efficient expression and mutagenesis of recombinant human phosphatidylcholine transfer protein using a synthetic gene: evidence for a C-terminal membrane binding domain" Biochemistry 2000 Dec. 19, 39(50): 15399-409, which describes codon optimization in Escherichia coli; Humphreyse tal., "High-level periplasmic expression in Escherichia coli using aeukaryotic signal peptide: importance of codon usage at the 5' end of the coding sequence", Protein Expr. Purif. 2000 Nov. 20(2): 252-64, which describes how codon usage affects protein secretion in Escherichia coli.

The cell-generating host can be a mammalian cell line selected from CHO, HEK293, IM9, DS-1, THP-1, Hep G2, COS, NIH 3T3, C33a, A549, A375, SK-MEL-28, DU145, PC-3, HCT116, MiaPACA-2, ACHN, Jurkat, MM1, Ovcar 3, HT 1080, Pancreatic Cancer-1 (Panc-1), U266, 769P, BT-474, Caco-2, HCC1954, MDA-MB-468, LnCAP, NRK-49F, and SP2/0; and mouse spleen cells and rabbit PBMCs. In one embodiment, the mammalian system is selected from the CHO or HEK293 cell line. In one specific aspect, the mammalian system is the CHO-S cell line. In another embodiment, the mammalian system is the HEK293 cell line.

In some embodiments, the cell-producing host is a yeast cell system, such as *Saccharomyces cerevisiae* or *Pichia pastoris* cells. In some embodiments, the cell-producing host is a prokaryotic cell such as *Escherichia coli* (Owens RJ and Young RJ, *J. Immunol. Meth.*, vol. 168, p. 149, 1994; Johnson S and Bird RE, *Methods Enzymol.*, vol. 203, p. 88, 1991). Conditionally active peptides can also be produced in plants (Firek et al. *Plant MoI. Biol.*, vol. 23, p. 861, 1993).

Conditionally active peptides can also be produced by synthetic methods using chemical methods known in the art. See, for example, Caruthers, "New chemical methods for synthesizing polynucleotides," Nucleic Acids Res. Symp. Ser. 215-223, 1980; Horn, "Synthesis of oligonucleotides on cellulose. Part II: design and synthetic strategy to the synthesis of 22 oligodeoxynucleotides coding for Gastric Inhibitory Polypeptide (GIP)," Nucleic Acids Res. Symp. Ser. 225-232, 1980; Banga, A.K., Therapeutic Peptides and Proteins, Formulation, Processing and Delivery Systems, Technomic Publishing Co., Lancaster, Pa, 1995. For example, various solid-phase techniques can be used for peptide synthesis (see, for example, Roberge, "A strategy for a convergent synthesis of N-linked glycopeptides on a solid support", Science 269:202, 1995; Merrifield, "Concept and early development of solid-phase peptide synthesis", Methods Enzymol. 289:3-13, 1997); and automated synthesis can be achieved, for example, using an ABI 43IA peptide synthesizer (Perkin Elmer) according to the manufacturer's instructions.

Solid-phase chemical peptide synthesis methods have been known in the field since the early 1960s (Merrifield, RB, "Solid-phase synthesis I. The synthesis of atetrapeptide", J. Am. Chem. Soc, 85: 2149-2154, 1963) (see also Stewart, J. M. and Young, JD, Solid Phase Peptide Synthesis, 2nd Ed., Pierce Chemical Co., Rockford, 111, pp. 11-12)), and have recently been used in commercially available laboratory peptide design and synthesis kits (Cambridge Research Institute of Biochemistry). Commercially available laboratory kits typically utilize the teachings of HMGeysen et al., "Use of peptide synthesis to probe viral antigens for epitopes to a resolution of a single amino acid," Proc. Natl. Acad. Sci., USA, 81:3998, 1984, and provide a method for synthesizing peptides on the tips of multiple "bars" or "needles," all of which are attached to a single plate. When using such a system, the plate with the attached bars or needles is inverted and inserted into a second plate containing a solution for attaching or anchoring the appropriate amino acid to the tip of the needle or bar. By repeating this process step—inverting the tips of the bars and needles and inserting them into the appropriate solution—the amino acids are combined into the desired peptide. In addition, a large number of FMOC peptide synthesis systems are available. For example, the Applied Biosystems, Inc. 431A ^™ Automated Peptide Synthesizer can be used for the assembly of peptides or fragments on a solid support. This device makes it easy to obtain the peptides of the present invention by direct synthesis or by synthesizing a series of fragments that can be linked using other known techniques.

Conditionally active peptides can also be glycosylated. Glycosylation can be added post-translation chemically or through cellular biosynthetic mechanisms, the latter involving the use of known glycosylation motifs, which may be native to the sequence or can be added as a peptide or as a nucleic acid coding sequence. Glycosylation can be O-linked or N-linked.

Conditionally active polypeptides include all forms of "mimics" and "peptides". The terms "mimic" and "peptide mimic" refer to synthetic compounds having substantially the same structural and/or functional properties as the polypeptides of the present invention. Mimics may consist entirely of synthetic, non-natural amino acid analogs, or chimeric molecules comprising a portion of natural peptide amino acids and a portion of non-natural amino acid analogs. Mimics may also contain any amount of such substitutions, provided that the conservative substitution of natural amino acids does not substantially alter the structure and/or activity of the mimic. With respect to the polypeptides of the present invention as conservative variants, conventional experiments will determine whether the mimic is within the scope of the invention, i.e., whether its structure and/or function have not been substantially altered.

The peptide mimic compositions of the present invention may comprise any combination of non-natural structural components. In another aspect, the mimic compositions of the present invention comprise one or all of the following three structural groups: a) residue-linking groups other than natural amide bonds (“peptide bonds”); b) non-natural residues that substitute for naturally occurring amino acid residues; or c) residues that induce secondary structure mimicry, i.e., residues that induce or stabilize secondary structures (e.g., β-turns, γ-turns, β-sheets, α-helical conformations, etc.). For example, when all or some residues of the peptides of the present invention are linked by chemical means other than natural peptide bonds, it may be called a mimic. Individual peptide mimic residues may be linked by peptide bonds, other chemical bonds, or coupling methods, such as glutaraldehyde, N-hydroxysuccinimide esters, bifunctional maleimides, N,N'-dicyclohexylcarbodiimide (DCC), or N,N'-diisopropylcarbodiimide (DIC). Linking groups that can be used as alternatives to traditional amide bonds (“peptide bonds”) include, for example, ketomethylene (e.g., replacing -C(=O) _-NH- with -C(=O)-CH2-), aminomethylene ( _CH2 -NH), ethylene, olefin (CH=CH), ether ( _CH2 -O), thioether ( _CH2 -S), tetrazolium ( _CN4- ), thiazole, retroamide, thioamide, or ester (see, for example, Spatola (1983) in Chemistry and Biochemistry of Amino Acids, Peptides and proteins, Vol. 7, pp. 267-357, "Peptide Backbone Modifications", Marcell Dekker, NY).

When the polypeptides of the present invention contain all or some non-natural residues in place of naturally occurring amino acid residues, the polypeptides of the present invention may also be referred to as mimics. Non-natural residues are described in detail in scientific and patent literature; several exemplary non-natural compositions and guidelines for using mimics of natural amino acid residues are described below. Aromatic amino acid mimics can be generated by using, for example, the following substances: D- or L-naphthylalanine; D- or L-phenylglycine; D- or L-2-thienylalanine; D- or L-1, -2, 3, or 4-pyrenoylalanine; D- or L-3-thienylalanine; D- or L-(2-pyridyl)-alanine; D- or L-(3-pyridyl)-alanine; D- or L-(2-pyrazinyl)-alanine; D- or L-(4-isopropyl)-phenylglycine. D-(trifluoromethyl)-phenylglycine; D-(trifluoromethyl)-phenylalanine; D-p-fluoro-phenylalanine; D- or L-p-biphenylphenylalanine; D- or L-p-methoxy-biphenylphenylalanine; D- or L-2-indole(alkyl)alanine; and D- or L-alkylamines, wherein the alkyl group may be substituted or unsubstituted methyl, ethyl, propyl, hexyl, butyl, pentyl, isopropyl, isobutyl, sec-isotyl, isopentyl, or a non-acidic amino acid. Aromatic rings of non-natural amino acids include, for example, thiazolyl, thiophene, pyrazolyl, benzimidazolyl, naphthyl, furanyl, pyrroleyl, and pyridyl aromatic rings.

Analogs of acidic amino acids can be generated by, for example, replacing them with non-carboxylated amino acids while retaining a negative charge, by substituting with (phosphono)alanine, or by substituting with sulfated threonine. The carboxyl side group (e.g., aspartic or glutamic) can also be selectively modified by reacting with a carbodiimide (R'～N—C--N--R') such as 1-cyclohexyl-3-(2-morpholino-(4-ethyl))carbodiimide or 1-ethyl-3-(4-azacation-4,4-dimethylpentyl)carbodiimide. Aspartic or glutamic groups can also be converted to asparagyl and glutamic acid residues by reacting with ammonium ions. Analogs of basic amino acids can be generated by replacing them with, for example (in addition to lysine and arginine), the following amino acids: ornithine, citrulline, or (guanidino)-acetic acid or (guanidino)alkyl-acetic acid, where alkyl is defined as above. Nitrile derivatives (e.g., containing a CN- moiety replacing COOH) can be substituted to yield asparagine or glutamine. The asparagine acyl and glutamine acyl residues can be deamidated to yield the corresponding asparagine or glutamine residues. Arginine residue mimics can be generated by reacting the arginyl group with, for example, one or more conventional reagents (including, for example, benzoylformaldehyde, 2,3-butanedione, 1,2-cyclohexanedione, or ninhydrin), preferably under basic conditions. Tyrosine residue mimics can be generated by reacting the tyrosine acyl group with, for example, an aromatic diazo compound or tetranitromethane. N-acetylimidazolium and tetranitromethane can be used to form O-acetyltyrosine acyl substances and 3-nitro derivatives, respectively. Cysteine residue mimics are generated by reacting the cysteine residue with, for example, an α-haloacetic acid ester (such as 2-chloroacetic acid or chloroacetamide) and the corresponding amine to yield carboxymethyl or carboxyaminomethyl derivatives. Cysteine residue mimics can also be generated by reacting cysteyl residues with substances such as bromotrifluoroacetone, α-bromo-β-(5-imidazolyl)propionic acid, chloroacetyl phosphate, N-alkylmaleimide, 3-nitro-2-pyridyl disulfide, methyl-2-pyridyl disulfide, p-chloromercuryl benzoate, 2-chloromercuryl-4-nitrophenol; or chloro-7-nitrobenzo-oxa-1,3-diazole. Lysine mimics can be generated (and have their amino-terminal residues modified) by reacting lysyl with substances such as succinic anhydride or other carboxylic anhydrides. Lysine and other α-amino residue mimics can also be generated by reacting with imine esters such as pyridinylimine methyl ester, pyridoxal phosphate, pyridoxal, chloroborohydride, trinitrobenzenesulfonic acid, O-methylisourea, 2,4-pentanedione, and by transaminases-catalyzed reactions with glyoxylic acid. Methionine mimics can be generated by reaction with, for example, methionine sulfoxide. Proline mimics include, for example, piperidinic acid, thiazolidinic acid, 3- or 4-hydroxyproline, dehydroproline, 3- or 4-methylproline, or 3,3-dimethylproline. Histidine residue mimics can be generated by reacting histidine with, for example, diethyl orthocarbonate or p-bromobenzoylmethyl bromide. Other mimics include, for example, mimics generated by hydroxylation of proline and lysine; mimics generated by phosphorylation of the hydroxyl groups of serine or threonyl residues; mimics generated by methylation of the α-amino groups of lysine, arginine, and histidine; mimics generated by acetylation of N-terminal amines; mimics generated by methylation of skeletal amide residues or substitution with N-methyl amino acids; or mimics generated by amidation of the C-terminal carboxyl group.

The residues (e.g., amino acids) of the polypeptides of the present invention can also be replaced by amino acids (or peptide-like residues) of opposite chirality. Thus, any amino acid that is naturally present in the L-configuration (which may also be referred to as R or S depending on the structure of the chemical entity) can be replaced by an amino acid of the same chemical structure type but with opposite chirality or a peptide mimic with opposite chirality, referred to as a D-amino acid, which may also be referred to as the R- or S-form.

This invention also provides methods for modifying conditionally active peptides via natural processes (e.g., post-translational processing, such as phosphorylation, acylation, etc.) or via chemical modification techniques. Modifications can occur anywhere in the peptide, including the peptide backbone, amino acid side chains, and amino or carboxyl terminals. It should be understood that the same type of modification can be present at several sites in a given peptide in the same or different amounts. A given peptide can also have many types of modifications. Modifications include acetylation, acylation, PEGylation, ADP-ribosylation, amidation, covalent linkage of flavin, covalent linkage of heme moieties, covalent linkage of nucleotides or nucleotide derivatives, covalent linkage of lipids or lipid derivatives, covalent linkage of phosphatidylinositol, cross-linking cyclization, disulfide bond formation, demethylation, formation of covalent cross-links, formation of cysteine, formation of pyroglutamic acid, formylation, γ-carboxylation, glycosylation, GPI anchoring, hydroxylation, iodination, methylation, myristylation, oxidation, polyethylene glycolation, proteolytic processing, phosphorylation, isopentenylation, racemization, selenization, sulfation, and transfer RNA-mediated amino acid addition to proteins, such as arginylation. See, for example, Creighton, T.E., as—Structure and Molecular Properties, 2nd Ed., W.H. Freeman and Company, New York (1993); Posttranslational Covalent Modification of proteins, B.C. Johnson, Ed., Academic Press, New York, pp. 1-12 (1983).

H. Engineering of conditionally active antibodies

The conditionally active antibodies of this invention can be engineered using one or more antibody engineering techniques described herein. Non-limiting examples of antibody engineering techniques include antibody conjugation, engineered multispecific antibodies, engineered bispecific conditionally active antibodies against surface antigens and target antigens of immune effector cells, and the Fc region of engineered antibodies.

Suitable methods for conjugating conditionally active antibodies are described in WO2015/175375. In one embodiment, the ratio of the activity of the conjugated conditionally active antibody under abnormal conditions to its activity under normal physiological conditions is preferably at least about 10:1, or at least about 12:1, or at least about 14:1, or at least about 16:1, or at least about 18:1, or at least about 20:1, or at least about 22:1, or at least about 24:1, or at least about 26:1.

In some embodiments, conditionally active antibodies may be conjugated to the Fc region of these antibodies. The aforementioned conjugated molecules, compounds, or drugs can be conjugated to the Fc region, as described in U.S. Patent No. 8,362,210. For example, the Fc region can be conjugated to a cytokine or toxin for delivery to a site where the conditionally active antibody exhibits preferential activity. Methods for conjugating peptides to the Fc region of antibodies are known in the art. See, for example, U.S. Patents 5,336,603, 5,622,929, 5,359,046, 5,349,053, 5,447,851, 5,723,125, 5,783,181, 5,908,626, 5,844,095, and 5,112,946; EP 307,434; EP 367,166; EP 394,827; WO91/06570, WO96/04388, WO96/22024, WO97/34631, and WO99/04813; Ash kenazi et al., Proc.Natl.Acad.Sci.USA, vol.88, pp.10535-10539, 1991; Traunecker et al., Nature, vol.331, pp.84-86, 1988; Zheng et al., J. Immunol., vol. 154, pp. 5590-5600, 1995; and ViI et al., Proc. Natl. Acad. Sci. USA, vol. 89, pp. 11337-11341, 1992.

In some embodiments, the conditionally active antibody can be covalently linked to a conjugated reagent via an intermediate linker having at least two reactive groups, one of which reacts with the conditionally active antibody and the other with the conjugated reagent. The linker may be selected to include any compatible organic compound such that the reaction with the conditionally active antibody or with the conjugated reagent does not adversely affect the reactivity and/or selectivity of the conditionally active antibody. Furthermore, the connection of the linker to the conjugated reagent must not destroy the activity of the conjugated reagent. The ratio of the anticancer molecule conjugated to the conditionally active peptide molecule is 3:1, 4:1, 5:1, or 6:1. In one example, the ratio of the anticancer molecule to the conditionally active peptide is approximately 4:1.

Suitable linkers for oxidized conditional antibodies include linkers containing a group selected from primary amines, secondary amines, hydrazides, acylhydrazides, hydroxylamines, phenylhydrazides, aminoureas, and aminothioureas. Suitable linkers for reduced conditional antibodies include linkers having certain reactive groups capable of reacting with the thiol group of the reduced conditional antibody. Such reactive groups include, but are not limited to, reactive haloalkyl groups (including, for example, haloacetyl groups), p-mercuric benzoate groups, and groups capable of Michael addition reactions (including, for example, maleimides and the types of groups described in the literature Mitra and Lawton, J. Amer. Chem. Soc. Vol. 101, pp. 3097-3110, 1979).

A suitable method for engineering multispecific conditionally active antibodies has been described in WO2015/175375.

Conditional active antibodies can be engineered to produce bispecific conditional active antibodies against both immune effector cell surface antigens and target antigens. The bispecific conditional active antibodies of this invention can attract immune effector cells to disease sites where the target antigen is present. A bispecific conditional active antibody is an antibody capable of specifically binding to two different antigens: an immune effector cell surface antigen and a target antigen. A bispecific antibody can be a full-length antibody comprising two arms, one arm binding to the immune effector cell surface antigen and the other arm binding to the target antigen. A bispecific antibody can be an antibody fragment comprising only a heavy chain variable domain ( _VH ) and a light chain variable domain ( _VL ). In one embodiment, the antibody fragment comprises at least two _VH / _VL units: one for binding to the immune effector cell surface antigen and the other arm binding to the target antigen. In another embodiment, the antibody fragment comprises at least two single variable domains ( _VH or _VL ): one for binding to the immune effector cell surface antigen and the other for binding to the target antigen. In some embodiments, a bispecific conditional active antibody comprises two scFvs: one binding to the immune effector cell surface antigen and the other binding to the target antigen.

Attracted immune effector cells are active in binding to target antigens on both immune effector cells and diseased cells or tissues. They are drawn to virus-carrying cells or diseased tissues containing the target antigens. These attracted effector cells then attack the diseased cells or tissues, thus aiding in the treatment of the disease, as they can suppress or even destroy diseased cells or tissues. For example, immune effector cells can destroy tumor cells or infected cells. Immune effector cells include natural killer cells, macrophages, lymphokine-activated killer (LAK) cells, and T cells.

Bispecific conditional antibodies possess two binding activities: one against an immune effector cell surface antigen and the other against a target antigen. In one embodiment, both binding activities are conditional, meaning that the binding activity of the bispecific conditional antibody against both the immune effector cell surface antigen and the target antigen is lower than that of the parent antibody under normal physiological conditions, but higher under abnormal conditions. In another embodiment, only one of the two binding activities is conditional, meaning that either the binding activity of the bispecific conditional antibody against the immune effector cell surface antigen or its binding activity against the target antigen is conditional. In this case, either the binding activity of the bispecific conditional antibody against the immune effector cell surface antigen or its binding activity against the target antigen is lower than the corresponding activity of the parent antibody under normal physiological conditions, but higher under abnormal conditions.

The two arms (e.g., two _VH - _VL units or two scFvs) of a bispecific conditionally active antibody can be linked together using conventional methods. As is well known in the art, the smallest antibody fragment containing a complete antigen-binding site has a dimer of a heavy-chain variable domain and a light-chain variable domain ( _VH and _VL ) linked non-covalently. This structure corresponds to the structure found in natural antibodies, where three complementarity-determining regions (CDRs) of each variable domain interact to define the antigen-binding site on the surface of the _VH - _VL dimer. A total of six CDRs confer antigen-binding specificity to the antibody. The flanking backbones (FRs) of the CDRs have a tertiary structure that is largely conserved in natural immunoglobulins of species such as humans and mice. These FRs serve to maintain the proper orientation of the CDRs. The constant domains are not essential for binding function but contribute to stabilizing the _VH - _VL interaction. Even a single variable domain (or half of the Fv containing only three CDRs specific to the antigen) has the ability to recognize and bind to the antigen, but the binding affinity is generally lower than that of the entire binding site (Painter et al., "Contributions of heavy and light chains of rabbit immunoglobulin G to antibody activity. I. Binding studies on isolated heavy and light chains," Biochemistry, vol. 11 pp. 1327-1337, 1972). Therefore, the domain of the binding site of a bispecific conditionally active antibody can be constructed as a pair of _VH - _VL , _VH - _VH , or _VL - _VL domains of different immunoglobulins.

In some embodiments, bispecific conditionally active antibodies can be constructed into continuous polypeptide chains using recombinant DNA technology, for example, by expressing nucleic acid molecules encoding bispecific conditionally active antibodies to construct continuous polypeptide chains (see, for example, Mack et al., "A small bispecific antibody construct expressed as a functional single-chain molecule with high tumor cell cytotoxicity," Proc. Natl. Acad. Sci. USA, vol. 92, pp. 7021-7025, 2005). The order of the _VH and _VL domains within the polypeptide chain is not critical to this invention, as long as the _VH and _VL domains are arranged such that the antigen-binding site can be appropriately folded to form one binding site against the surface antigen of immune effector cells and one binding site against the target antigen.

Some of the techniques described in this article for engineering multispecific conditionally active antibodies can be used to generate bispecific conditionally active antibodies against both immune effector cell surface antigens and target antigens.

Bispecific antibodies can be constructed as a single polypeptide chain, as described in WO99/54440; Mack, J. Immunol. (1997), 158, 3965-3970; Mack, PNAS, (1995), 92, 7021-7025; Kufer, Cancer Immunol. Immunother., (1997), 45, 193-197; Loftier, Blood, (2000), 95, 6, 2098-2103; Bruhl, J. Immunol., (2001), 166, 2420-2426. A particularly preferred configuration for bispecific antibodies is a polypeptide construct in which the _VH and _VL regions are interconnected by a linker domain. The order of the _VH and _VL regions in a single polypeptide chain is not important. In one embodiment, a single polypeptide chain is constructed as _VH1 -adaptor domain- _VL1 -adaptor domain- _VH2 -adaptor domain- _VL2 . In another embodiment, a single polypeptide chain is constructed as _VL1 -adaptor domain- _VH1 -adaptor domain- _VL2 -adaptor domain- _VH2 . In yet another embodiment, a single polypeptide chain is constructed as _VH1 -adaptor domain- _VH2 -adaptor domain- _VL1 -adaptor domain- _VL2 . In yet another embodiment, a single polypeptide chain is constructed as _VH1 -adaptor domain- _VL2 -adaptor domain- _VL1 -adaptor domain- _VH2 . The single polypeptide chain can fold into two arms, each arm capable of binding to an immune effector cell surface antigen or a target antigen.

The linker domain in a bispecific conditionally active antibody is a peptide fragment of sufficient length to allow intermolecular linkage between these _VH and _VL domains. Designs of linkers suitable for this purpose are described in the prior art, such as EP623679B1, U.S. Patent No. 5,258,498, EP 573551B1, and U.S. Patent No. 5,525,491. The linker domain is preferably a hydrophilic flexible linker of 1-25 amino acids, selected from glycine, serine, and/or glycine/serine. In one embodiment, the linker domain is a 15-amino acid linker sequence (Gly ₄ Ser) ₃ .

Other linker domains include oligomerization domains. Oligomerization domains can facilitate the folding of two or more _VH and _VL domains into two arms, each capable of binding to immune effector cell surface antigens or target antigens. Non-limiting examples of oligomerization domains include leucine zippers (e.g., jun-fos, GCN4, E/EBP; Kostelny, J. Immunol. 148 (1992), 1547-1553; Zeng, Proc. Natl. Acad. Set 94 (1997), 3673-3678; Williams, Genes Dev. 5 (1991), 1553-1563; Suter, "Phage Display of Peptides and Proteins", Chapter 11, (1996), Academic Press), antibody-derived oligomerization domains such as constant domains CH1 and CL (Mueller, FEBS Letters). 422(1998), 259-264) and/or tetramerized domains such as GCN4-LI (Zerangue, Proc. Natl. Acad. Sci. 97(2000), 3591-3595).

In some implementations, the knock-in-hole technique can be used to stabilize the folding of single-chain bispecific conditional antibodies. Ridgway et al. ("'Knobs-into-holes' engineering of antibody CH3 domains for heavy chain heterodyne formation," Protein Eng. 1996 Jul; 9(7): 617-21) described the knock-in-hole technique. This method has been used for the stacking of amino acid side chains between adjacent α-helices, where the side chains of residues in the α-helices are represented as spaced-out knockers alternating with the knockers on the surface of a cylinder, which may contain knockers of adjacent α-helices (O'Shea et al., (1991) Science, 254, 539-544).

Immune effector cell surface antigens should be specific to one or a class of immune effector cells. Many immune effector cell surface antigens are known. Natural killer cells possess surface antigens, including CD56, CD8, CD16, KIR family receptors, NKp46, NKp30, CD244(2B4), CD161, CD2, CD7, CD3, and killer cell immunoglobulin-like receptors (Angeliset et al., “Expansion of CD56-negative, CD16-positive, KIR-expressing natural killer cells after T cell-depleted haploidentical hematopoietic stem cell transplantation,” Acta Haematol. 2011; 126(1):13-20; Dalle et al., “Characterization of Cord Blood Natural Killer Cells: Implications for Transplantation and Neonatal Infections,” Pediatric Research (2005) 57,649– 655; Agarwal et al., "Roles and Mechanism of Natural Killer Cells in Clinical and Experimental Transplantation," Expert Rev Clin Immunol. 2008; 4(1):79-91).

Macrophages possess surface antigens, including CD11b, F4/80, CD68, CSF1R, MAC2, CD11c, LY6G, LY6C, IL-4Rα, CD163, CD14, CD11b, F4/80 (mouse)/EMR1 (human), CD68 and MAC-1/MAC-3, PECAM-1 (CD31), CD62, CD64, CD45, Yml, CD206, CD45RO, 25F9, S100A8/A9, and PM-2K (Murray et al., “Protective and pathogenic functions of macrophage subsets,” Nature Reviews Immunology, 11, 723-737; Taylor et al., "Macrophage receptors and immunerecognition," Annu Rev Immunol 2005; 23:901-44; Pilling, et al., "Identification of Markers that "Distinguish Monocyte-Derived Fibrocytes from Monocytes, Macrophages, and Fibroblasts," PLoS ONE 4(10):e7475.doi:10.1371/journal.pone.0007475, 2009).

Lymphokine-activated killer (LAK) cells possess surface antigens, including T3, T4, T11, T8, T1I, Leu7, and Leu1. (Ferrini et al., “Surface markers of human lymphokine-activated killer cells and their precursors,” Int J Cancer. 1987 Jan 15; 39(1): 18-24; Bagnasco et al., “Glycoproteic nature of surface molecules of effector cells with lymphokine-activated killer cells”) phokine-activated killer(LAK)activity," Int J Cancer.1987Jun 15;39(6):703-7; Kaufmann et al., "Interleukin 2induces human acute lymphocyt ic leukemiacells to manifest lymphokine-activated-killer(LAK)cytotoxicity,” The Journal of Immunology, August 1, 1987, vol.139no.3977-982).

T cells, particularly cytotoxic T cells, possess surface antigens, including CD2, CD3, CD4, CD5, CD6, CD8, CD28, T58, CD27, CD45, CD84, CD25, CD127, and CD196 (CCR6), CD197 (CCR7), CD62L, CD69, TCR, T10, T11, and CD45RO (Ledbetter et al., "Enhanced transmembrane signaling activity of monoclonal antibody heteroconjugates suggests molecular interactions better) n receptors on the T cell surface," Mol Immunol. 1989 Feb; 26(2):137-45; Jondal et al., "SURFACE MARKERS ON HUMAN TAND B LYMPHOCYTES," JOURNAL OF EXPERIMENTAL MEDICINE, VOLUME 136, 1972, 207-215; Mingari et al., "Surface markers of human Tlymphocytes," Ric ClinLab. 1982Jul-Sep; 12(3):439-448).

Bispecific conditionally active antibodies, upon binding to immune effector cells, can transport these cells to cells or tissues containing the target antigen, preferably the target antigen being present on the surface. Once the bispecific conditionally active antibody (and the immune effector cells) are bound to the target antigen, the immune effector cells can attack diseased cells or tissues. Immune effector cells such as natural killer cells, macrophages, LAK cells, and T cells (cytotoxic) can kill and/or destroy diseased cells or tissues, such as destroying tumor tissue.

Diseased cells or tissues can be selected from cancer, inflammatory diseases, neuronal diseases, diabetes, cardiovascular diseases, or infectious diseases. Examples of target antigens include antigens expressed by various immune cells, cancers, sarcomas, lymphomas, leukemias, germ cell tumors, germ cell tumors, and cells associated with various hematologic disorders, autoimmune diseases, and/or inflammatory diseases.

Cancer-specific target antigens that can be targeted by bispecific conditionally active antibodies include 4-IBB, 5T4, adenocarcinoma antigen, alpha-fetoprotein, BAFF, B lymphoma cells, C242 antigen, CA-125, carbonic anhydrase 9 (CA-IX), C-MET, CCR4, CD152, CD19, CD20, CD200, CD22, CD221, CD23 (IgE receptor), CD28, CD30 (TNFRSF8), CD33, CD4, CD40, CD44 v6, CD51, CD52, CD56, CD74, CD80, CEA, CNT0888, CTLA-4, DR5, EGFR, EpCAM, CD3, FAP, fibronectin superdomain B, folate receptor 1, GD2, GD3 gangliosides, glycoprotein 75, and GPN. MB, HER2/neu, HGF, human scatter factor receptor kinase, IGF-1 receptor, IGF-1, IgG1, LI-CAM, IL-13, IL-6, insulin-like growth factor I receptor, integrin α5β1, integrin ανβ3, MORAb-009, MS4A1, MUC1, mucin CanAg, N-hydroxyacetylneuraminic acid, NPC-IC, PDGF-Rα, PDL192, phosphatidylserine, prostate cancer cells, RANKL, RON, ROR1, SCH 900105, SDC1, SLAMF7, TAG-72, tendinin C, TGFβ2, TGF-β, TRAIL-R1, TRAIL-R2, tumor antigen CTAA16.88, VEGF-A, VEGFR-1, VEGFR2, or vimentin.

The types of cancers treated with the genetically engineered cytotoxic cell or pharmaceutical compositions of this invention include carcinomas, germ cell tumors, and sarcomas, as well as certain leukemias or lymphomas, benign and malignant tumors, and malignant tumors such as sarcomas, carcinomas, and melanomas. Cancers may be non-solid tumors (such as hematologic malignancies) or solid tumors. Adult tumors/cancers and pediatric tumors/cancers are also included.

Blood cancers are cancers of the blood or bone marrow. Examples of hematologic (or blood-related) cancers include leukemia, which includes acute leukemia (such as acute lymphoblastic leukemia, acute myeloid leukemia, acute myeloid leukemia and myeloblastic leukemia, promyelocytic leukemia, myelomonocytic leukemia, monocytic leukemia and erythroleukemia), chronic leukemia (such as chronic myeloid (granulocytic) leukemia, chronic myeloid leukemia and chronic lymphocytic leukemia), polycythemia vera, lymphoma, Hodgkin's disease, non-Hodgkin's lymphoma (painless and advanced forms), multiple myeloma, Waldenström macroglobulinemia, heavy chain disease, myelodysplastic syndrome, hairy cell leukemia, and myelodysplastic syndrome.

Solid tumors are abnormal masses of tissue that do not typically contain cysts or fluid-filled areas. Solid tumors can be benign or malignant. Different types of solid tumors are named according to the cell type that forms them (e.g., sarcomas, carcinomas, and lymphomas). Examples of treatable solid tumors include sarcomas and carcinomas, including fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteosarcoma and other sarcomas, synovoma, mesothelioma, Ewing's tumor, leiomyosarcoma, rhabdomyosarcoma, colon cancer, malignant lymphomas, pancreatic cancer, breast cancer, lung cancer, ovarian cancer, prostate cancer, hepatocellular carcinoma, squamous cell carcinoma, basal cell carcinoma, adenocarcinoma, sweat gland carcinoma, medullary thyroid carcinoma, papillary thyroid carcinoma, pheochromocytoma, sebaceous gland carcinoma, papillary carcinoma, papillary adenocarcinoma, medullary carcinoma, bronchial carcinoma, renal cell carcinoma, hepatocellular carcinoma, bile duct carcinoma, choriocarcinoma, and Wilms' disease. Tumors, cervical cancer, testicular tumors, seminoma, bladder cancer, melanoma, and CNS tumors (such as gliomas (e.g., brainstem glioma and mixed glioma), glioblastoma (also known as glioblastoma multiforme), astrocytoma, CNS lymphoma, germ cell tumor, medulloblastoma, schwannoma, ependymoma, pineal tumor, angioblastoma, acoustic neuroma, oligodendroglioma, meningioma, neuroblastoma, retinoblastoma, and brain metastases).

Bispecific conditionally active antibodies can target inflammatory disease-specific antigens including AOC3 (VAP-1), CAM-3001, CCL11 (eosinophil chemokine-1), CD125, CD147 (basal immunoglobulin), CD154 (CD40L), CD2, CD20, CD23 (IgE receptor), CD25 (IL-2 receptor chain), CD3, CD4, CD5, IFN-α, IFN-γ, IgE, and IgE F. c region, IL-1, IL-12, IL-23, IL-13, IL-17, IL-17A, IL-22, IL-4, IL-5, IL-6, IL-6 receptor, integrin α4, integrin α4β7, Lama Glama, LFA-1 (CD1 1a), MEDI-528, myostatin, OX-40, rhuMAbβ7, sclerosing protein, SOST, TGFβ1, TNF-α or VEGF-A.

The bispecific conditionally active antibodies of the present invention can target neurological disease-specific antigens including one or more of β-amyloid protein or MABT5102A. The bispecific conditionally active antibodies of the present invention can target diabetes-specific antigens including one or more of L-Iβ or CD3. The bispecific conditionally active antibodies of the present invention can target cardiovascular disease-specific antigens including one or more of C5, cardiac myosin, CD41 (integrin α-lib), fibrin II, β-chain, ITGB2 (CD18), and sphingosine-1-phosphate.

The bispecific conditionally active antibody of the present invention can target one or more of the following infectious disease-specific target antigens: anthrax toxin, CCR5, CD4, agglutination factor A, cytomegalovirus, cytomegalovirus glycoprotein B, endotoxin, Escherichia coli, hepatitis B surface antigen, hepatitis B virus, HIV-1, Hsp90, influenza A hemagglutinin, lipoteichoic acid, Pseudomonas aeruginosa, rabies virus glycoprotein, respiratory syncytial virus, and TNF-α.

Other examples of target antigens include surface proteins present on cancer cells in a specific or amplified manner, such as the IL-14 receptor, CD19, CD20, and CD40 in B-cell lymphoma; LewisY and CEA antigens in various cancers; the Tag72 antigen in breast and colorectal cancer; EGF-R in lung cancer; folate-binding proteins and HER-2 proteins, which are typically amplified in human breast and ovarian cancer; or viral proteins, such as the gp120 and gp41 envelope proteins of human immunodeficiency virus (HIV), envelope proteins of hepatitis B and hepatitis C viruses, glycoprotein B and other envelope glycoproteins of human cytomegalovirus, and envelope proteins of methylviruses such as Kaposi's sarcoma-associated herpesvirus. Other possible target antigens include CD4, where the ligand is the HIV gp120 envelope glycoprotein, and other viral receptors, such as ICAM, the receptor for human rhinovirus, and related receptor molecules for poliovirus.

Human immunodeficiency virus (HIV) cannot enter human cells unless it first binds to two key molecules on the cell surface: CD4 and a co-receptor. The initially recognized co-receptor was CCR5, but later, during the viral life cycle, another chemokine receptor, CXCR4, became the co-receptor for HIV-1 (D'Souza, Nature Med. 2, 1293 (1996); Premack, Nature Med. 2, 1174; Fauci, Nature 384, 529 (1996)). HIV-1 strains that are primarily transmitted through sexual contact are called M-tropical viruses. These HIV-1 strains (also known as non-syncytial induced (NSI) primary viruses) can replicate in primary CD4 ⁺ T cells and macrophages, using the chemokine receptor CCR5 (and the less common CCR3) as their co-receptor. T-tropic viruses (sometimes called syncytial-induced (SI) primary viruses) can also replicate in primary CD4+ T cells, but can also infect established CD4+ T cell lines in vitro, where they perform the aforementioned functions via the chemokine receptor CXCR4 (fusion viral protein). At least under certain in vitro conditions, many of these T-tropic virus strains can use CCR5 in addition to CXCR4, and some can enter macrophages via CCR5 (D'Souza, Nature Med. 2, 1293 (1996); Premack, Nature Med. 2, 1174; Fauci, Nature 384, 529 (1996)). Since M-tropic HIV-1 strains are involved in approximately 90% of HIV sexual transmission, CCR5 is a major co-receptor for the virus in patients.

The number and identity of co-receptor molecules on target cells, and the ability of HIV-1 strains to enter cells via different co-receptors, appear to be key determinants of disease progression. High expression of CCR3 and CCR5 has also been observed in T cells and B cells of lymph nodes derived from Hodgkin's disease patients. Type 1 diabetes is considered a T cell-mediated autoimmune disease. In relevant animal models, expression of the CCR5 receptor in the pancreas has been associated with the progression of type 1 diabetes (Cameron (2000) J. Immunol. 165, 1102-1110). In one embodiment, a bispecific conditional active antibody binding to CCR5 as a target antigen can be used to inhibit HIV infection in host cells and slow the progression of other diseases.

Several antibodies that specifically bind to (human) CCR5 are known in the art, including MC-1 (Mack (1998) J. Exp. Med. 187, 1215-1224) or MC-5 (Blanpain (2002) Mol Biol Cell. 13: 723-37, Segerer (1999) Kidney Int. 56: 52-64, Kraft (2001) Biol. Chem. 14; 276: 34408-18). Therefore, bispecific conditionally active antibodies preferably contain, for example, _VL and _VH domains (i.e., the second domain derived from Ig) of antibodies specific to CCR5 (especially human CCR5) and _VH and _VL domains of antibodies specific to CD3 antigens on T cells.

In another embodiment, the present invention provides a bispecific conditionally active antibody targeting CD3 and CD19 on T cells. CD19 has proven to be a highly useful medical target. CD19 is expressed throughout the entire B-cell lineage, from precursor B cells to mature B cells, and is expressed uniformly on all lymphoma cells, but is absent in stem cells (Haagen, Clin Exp Immunol 90 (1992), 368-75; Uckun, Proc. Natl. Acad. Sci. USA 85 (1988), 8603-7). Combination therapies using antibodies against CD19 and other immunomodulatory antibodies have been disclosed for the treatment of B-cell malignancies (WO02/04021, US2002006404, US2002028178) and autoimmune diseases (WO02/22212, US2002058029). WO00/67795 discloses the use of antibodies against CD19 for the treatment of indolent and aggressive B-cell lymphomas, as well as the formation of acute and chronic lymphocytic leukemias. WO02/80987 discloses the therapeutic use of immunotoxins based on antibodies against the CD19 antigen for the treatment of diseases such as B-cell non-Hodgkin lymphoma, Hodgkin lymphoma, or B-cell leukemia (e.g., B-cell acute lymphoblastic leukemia (B-ALL)), (e.g., pilocellular lymphoma) B-cell precursor acute lymphoblastic leukemia (pre-B-ALL), and B-cell chronic lymphocytic leukemia (B-CLL)).

In another embodiment, the present invention provides a bispecific conditionally active antibody against CD3 on T cells and CD20 as a target antigen. CD20 is one of the cell surface proteins present on B lymphocytes. The CD20 antigen is present in normal and malignant precursor B lymphocytes and mature B lymphocytes, including those in more than 90% of B-cell non-Hodgkin lymphomas (NHL). This antigen is not present in hematopoietic stem cells, activated B lymphocytes (plasma cells), and normal tissues. Several major murine antibodies have been described: 1F5 (Press et al., 1987, Blood 69/2, 584-591), 2B8/C2B8, 2H7, and 1H4 (Liu et al., 1987, J Immunol 139, 3521-3526; Anderson et al., 1998, US Patent No. 5,736,137; Haisma et al., 1998, Blood 92, 184-190; Shan et al., 1999, J Immunol. 162, 6589-6595).

CD20 has been described in immunotherapy strategies using DNA encoding scFv linked to a carrier protein for the treatment of plasma cell malignancies (Treon et al., 2000, Semin Oncol 27(5), 598), and immunotherapy using a CD20 antibody (IDEC-C2B8) has been shown to be effective in the treatment of non-Hodgkin's B-cell lymphoma.

In some implementations, the bispecific conditionally active antibody is a single polypeptide chain encoded by a polynucleotide molecule. The polynucleotide can be, for example, DNA, cDNA, RNA, or synthetically produced DNA or RNA, or a recombinant chimeric nucleic acid molecule containing any of those polynucleotides, alone or in combination. The polynucleotide can be part of a vector, such as an expression vector, including plasmids, granules, viruses and bacteriophages, or any expression system conventionally used in genetic engineering. The vector can contain other genes, such as marker genes, which allow selection of the vector under appropriate conditions in a suitable host cell.

In one aspect, the polynucleotide is operatively linked to an expression regulatory sequence that allows expression in prokaryotic or eukaryotic cells. Expression vectors derived from viruses (such as retroviruses, vaccinia viruses, adeno-associated viruses, herpesviruses, or bovine papillomaviruses) can be used to deliver the polynucleotide or vector into mammalian cells. Vectors containing the polynucleotides of the present invention can be transformed into host cells using well-known methods that vary depending on the type of cell host. For example, calcium chloride transfection is commonly used for prokaryotic cells, while calcium phosphate treatment or electroporation can be used for other cell hosts.

In another aspect, conditionally active peptides can be engineered to produce bispecific conditionally active peptides. The methods used to engineer conditionally active peptides are similar to those described in WO2015/175375 for engineering bispecific conditionally active antibodies. For example, a bispecific conditionally active peptide may have two active sites, each possessing conditional activity, i.e., lower activity than the parental site under normal physiological conditions and higher activity than the parental site under abnormal conditions. These two conditionally active sites can evolve and be screened independently, and then linked together to form a single bispecific conditionally active peptide using a linker. In one aspect, the linkers that can be used in bispecific conditionally active antibodies are known linkers adapted to generate bispecific conditionally active peptides by linking two conditionally active sites to the conditionally active peptide.

A suitable method for engineering the Fc region of conditionally active antibodies has been described in WO2015/175375.

A suitable method for engineering conditionally active viral particles has been described in WO2015/175375.

In some respects, using the methods described in WO2015/175375, conditionally active peptides can be inserted into viral particles that are oncolytic viruses. Oncolytic viruses are viruses that kill tumor cells upon contact with them. Conditionally active peptides inserted into oncolytic viruses may be more active in the tumor microenvironment but less active in other parts of the individual. For example, conditionally active peptides may be more active at the pH (e.g., pH 6.2–6.8) or other conditions present in the tumor microenvironment, but less active at the pH or other conditions present at another location in the individual (e.g., pH 7.2–7.6), such as normal physiological conditions. Conditionally active peptides inserted into oncolytic viruses can be used to facilitate the delivery of oncolytic viruses to tumors, where the oncolytic viruses can target and kill tumor cells.

Oncolytic viruses of interest include adenoviruses, herpes simplex virus-1, vaccinia virus, parvovirus, reovirus, and Newcastle disease virus. Vaccine viruses are of particular interest.

In one respect, oncolytic viruses are selected from paramyxoviruses, reoviruses, herpesviruses, adenoviruses, and Semlikie forest viruses. In another respect, paramyxoviruses are selected from Newcastle disease virus (NDV), measles virus, and mumps virus. In yet another respect, NDV is derived from strains selected from MTH68/H, PV-701, and 73-T.

On the other hand, oncolytic viruses are selected from herpesviruses, reoviruses, E1B-deficient adenoviruses, vesicular stomatitis viruses, and poxviruses. These oncolytic viruses may not only destroy tumor cells but also release antigens from the destroyed tumor cells, thereby triggering an immune response.

Specific examples of oncolytic viruses include, but are not limited to, adenoviruses (e.g., δ-24, δ-24-RGD, ICOVIR-5, ICOVIR-7, Onyx-015, ColoAdl, H101, AD5/3-D24-GMCSF), reoviruses, herpes simplex viruses (HSV, OncoVEX GM-CSF), Newcastle disease virus, measles virus, retroviruses (e.g., influenza virus), poxviruses (e.g., vaccinia virus, including Copenhagen, Western Reserve, and Wyeth strains), myxoma viruses, rhabdoviruses (vesicular stomatitis virus (VSV)), piconenucleoviruses (e.g., Seneca Valley virus, SW-001), Coxsackieviruses, and parvoviruses.

In one aspect, the oncolytic virus is an adenovirus comprising any one of its 57 human serotypes (HAdV-1 to 57). In one embodiment, the adenovirus is the Ad5 serotype. Alternatively, the adenovirus may be a heterozygous serotype that may or may not contain the Ad5 component. Non-limiting examples of suitable adenoviruses include δ-24, δ-24-RGD, ICOVIR-5, ICOVIR-7, ONYX-015, ColoAd1, H101, and AD5/3-D24-GMCSF. ONYX-015 is a hybrid of viral serotypes Ad2 and Ad5, with deletions in the E1B-55K and E3B regions to enhance cancer selectivity. HI 01 is a modification of Onyx-015. ICOVIR-5 and ICOVIR-7 contain a deletion of the Rb binding site of E1A and an E2F promoter substitution for the E1A promoter. ColoAd1 is a chimeric Addl lp/Ad3 serotype. AD5/3-D24-GMCSF (CGTG-102) is a serotype 5/3 capsid-modified adenovirus encoding GM-CSF (the Ad5 capsid protein nob is replaced by the serotype 3 nob domain).

In a particularly preferred embodiment, the oncolytic virus is δ-24 or δ-24-RGD adenovirus. δ-24 is described in US2003/0138405 A1 and US2006/0147420A1. δ-24 adenovirus is derived from adenovirus type 5 (Ad-5) and contains a 24-base pair deletion within the CR2 portion of the E1A gene. δ-24-RGD further comprises the RGD-4C sequence of an H1 loop inserted into the fibrin knockb protein (which strongly binds to αvβ3 and αvβ5 integrin) (Pasqualini R. et al., Nat Biotechnol., 15:542-546, 1997).

Oncolytic adenoviruses can be further modified to improve their ability to treat cancer. Jiang et al. (Curr. Gene Ther. 2009 Oct 9(5):422-427) have described such modifications to oncolytic adenoviruses, and see also US2006/0147420A1.

Oncolytic viruses, including those containing conditionally active peptides, can be administered locally or systemically. For example, but not limited to, oncolytic viruses can be administered intravascularly (intra-arterial or intravenous), intratumorally, intramuscularly, intradermally, intraperitoneally, subcutaneously, orally, parenterally, intranasally, intratracheally, percutaneously, intraspinally, intraocularly, or intracranially.

Oncolytic viruses can be administered in single or multiple doses. The virus can be administered at doses of at least 1× ^10⁵ plaque-forming units (PFU), at least 5× ^10⁵ PFU, at least 1× ^10⁶ PFU, at least 5× ^10⁶ or at least 5× ^10⁶ PFU, 1× ^10⁷ , at least 1× ^10⁷ PFU, at least 1× ^10⁸ or at least 1× ^10⁸ PFU, at least 1× ^10⁸ PFU, at least 5× ^10⁸ PFU, at least 1× ^10⁹ or at least 1× ^10⁹ PFU, at least 5× ^10⁹ or at least 5× ^10⁹ PFU, at least 1× ^10¹⁰ PFU or at least 1× ^10¹⁰ PFU, at least 5× ^10¹⁰ or at least 5× ^10¹⁰ PFU, at least 1× ^10¹¹ PFU or at least 1× ^10¹¹ PFU, at least 1× ^10¹² PFU or at least 1× ^10¹³ PFU. For example, oncolytic viruses can be administered at doses of approximately ^10⁷ – ^10¹³ PFU, approximately ^10⁸ – ^10¹³ PFU, approximately ^10⁹ – ^10¹² PFU, or approximately ^10⁸ – ^10¹² PFU.

In some respects, cancers that can be treated with oncolytic cells include any solid tumor, such as lung cancer, ovarian cancer, breast cancer, cervical cancer, pancreatic cancer, stomach cancer, colon cancer, skin cancer, laryngeal cancer, bladder cancer, and prostate cancer. In another respect, the cancer is cancer of the central nervous system. Cancer can be a neuroepithelial tumor, such as astrocytic tumors (e.g., astrocytoma, anaplastic astrocytoma, glioblastoma, gliosarcoma, pilocytic astrocytoma, giant cell astrocytoma, pleomorphic xanthoastrocytoma), oligodendroglioma, ependymoma, oligodendroglioma, glioblastoma, astrocytoma, choroid plexus papilloma, choroid plexus carcinoma, gangliocytoma, ganglioglioma, neurocytoma, neuroepithelial tumor, neuroblastoma, pineal region tumors (e.g., pineal cell carcinoma, pineal blastoma, or mixed pineal cell carcinoma/pineal blastoma), medullary epithelioma, medulloblastoma, neuroblastoma or ganglioblastoma, retinoblastoma or ependymoma. Cancer can be a tumor of the central nervous system, such as a sellar region tumor (e.g., pituitary adenoma, pituitary carcinoma, or craniopharyngioma), a hematopoietic tumor (e.g., primary malignant lymphoma, plasmacytoma, or granulocytoma sarcoma), a germ cell tumor (e.g., germ cell tumor, embryonal carcinoma, yolk sac tumor, choriocarcinoma, teratoma, or mixed germ cell tumor), a meningioma, a mesenchymal tumor, a melanoma, or a skull or spinal nerve tumor (e.g., schwannoma or neurofibroma). Cancer can be a low-grade glioma (e.g., ependymoma, astrocytoma, oligodendroglioma, or mixed glioma) or a high-grade (malignant) glioma (e.g., glioblastoma multiforme). Cancer can be a primary or metastatic brain tumor. Conditionally active peptides or products engineered from conditionally active peptides can be used in pharmaceutical compositions. Suitable pharmaceutical compositions are described in U.S. Patent No. 8,709,755B2.

The pharmaceutical composition can be used to treat various types of cancer, including carcinoma, germ cell tumors, and sarcomas, as well as certain leukemias or lymphomas, benign and malignant tumors, and malignancies such as sarcomas, carcinomas, and melanomas. Cancers can be non-solid tumors (such as hematologic malignancies) or solid tumors. Adult tumors/cancers and pediatric tumors/cancers are also included.

Blood cancers are cancers of the blood or bone marrow. Examples of hematologic (or blood-related) cancers include leukemia, which includes acute leukemia (such as acute lymphoblastic leukemia, acute myeloid leukemia, acute myeloid leukemia and myeloblastic leukemia, promyelocytic leukemia, myelomonocytic leukemia, monocytic leukemia and erythroleukemia), chronic leukemia (such as chronic myeloid (granulocytic) leukemia, chronic myeloid leukemia and chronic lymphocytic leukemia), polycythemia vera, lymphoma, Hodgkin's disease, non-Hodgkin's lymphoma (painless and advanced forms), multiple myeloma, Waldenström macroglobulinemia, heavy chain disease, myelodysplastic syndrome, hairy cell leukemia, and myelodysplastic syndrome.

Solid tumors are abnormal masses of tissue that do not typically contain cysts or fluid-filled areas. Solid tumors can be benign or malignant. Different types of solid tumors are named according to the cell type that forms them (e.g., sarcomas, carcinomas, and lymphomas). Examples of treatable solid tumors include sarcomas and carcinomas, including fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteosarcoma and other sarcomas, synovoma, mesothelioma, Ewing's tumor, leiomyosarcoma, rhabdomyosarcoma, colon cancer, malignant lymphomas, pancreatic cancer, breast cancer, lung cancer, ovarian cancer, prostate cancer, hepatocellular carcinoma, squamous cell carcinoma, basal cell carcinoma, adenocarcinoma, sweat gland carcinoma, medullary thyroid carcinoma, papillary thyroid carcinoma, pheochromocytoma, sebaceous gland carcinoma, papillary carcinoma, papillary adenocarcinoma, medullary carcinoma, bronchial carcinoma, and renal cell carcinoma. Hepatocellular carcinoma, cholangiocarcinoma, choriocarcinoma, Wilms' tumor, cervical cancer, testicular tumors, seminoma, bladder cancer, melanoma, and CNS tumors (such as gliomas (e.g., brainstem glioma and mixed glioma), glioblastoma (also known as glioblastoma multiforme), astrocytoma, CNS lymphoma, germ cell tumor, medulloblastoma, schwannoma, ependymoma, pineal tumor, angioblastoma, acoustic neuroma, oligodendroglioma, meningioma, neuroblastoma, retinoblastoma, and brain metastases).

Pharmaceutical compositions comprising conditionally active peptides or products engineered from conditionally active peptides can be formulated according to known methods for preparing pharmaceutical compositions. In these methods, the conditionally active peptide is typically combined with a mixture, solution, or composition containing a pharmaceutically acceptable carrier.

A pharmaceutically acceptable carrier is a substance that the receiving patient can tolerate. Sterile phosphate-buffered saline is one example of a pharmaceutically acceptable carrier. Other suitable pharmaceutically acceptable carriers are well known to those skilled in the art (see, for example, Gennaro (ed.), Remington's Pharmaceutical Sciences (Mack Publishing Company, 19th ed. 1995)). The formulation may also contain one or more excipients, preservatives, solubilizers, buffers, albumin to prevent protein loss from the vial surface, etc.

The form, route of administration, dosage, and regimen of a pharmaceutical composition naturally depend on the condition to be treated, the severity of the disease, the patient's age, weight, and sex, etc. Those skilled in the art can take these factors into account to formulate a suitable pharmaceutical composition. The pharmaceutical compositions of the present invention can be formulated for topical, oral, parenteral, intranasal, intravenous, intramuscular, subcutaneous, or intraocular administration, etc.

Preferably, the pharmaceutical composition contains a carrier that is pharmaceutically acceptable for an injectable formulation. These carriers may be, in particular, isotonic sterile saline solutions (sodium or disodium phosphate, sodium chloride, potassium chloride, calcium chloride or magnesium chloride, or mixtures of these salts), or dry, particularly freeze-dried compositions that, upon addition of, for example, sterile water or physiological saline, allow the formation of an injectable solution.

In some embodiments, a tensioning agent (sometimes referred to as a "stabilizer") is present to adjust or maintain the tension of the liquid in the composition. They are often referred to as "stabilizers" when used with large charged biomolecules such as proteins and antibodies because they can interact with the charged groups on the side chains of amino acids, thereby reducing the likelihood of intermolecular and intramolecular interactions. Tensioning agents can be present in any amount from 0.1% to 25% by weight, preferably from 1% to 5% by weight, of the pharmaceutical composition. Preferred tensioning agents include polyols, preferably ternary or higher sugar alcohols, such as glycerol, erythritol, arabinitol, xylitol, sorbitol, and mannitol.

Other excipients include one or more of the following agents that can be used as: (1) fillers, (2) solubilizers, (3) stabilizers, and (4) agents that prevent denaturation or adhesion to the container wall. Such excipients may include: polyols (listed above); amino acids such as alanine, glycine, glutamine, asparagine, histidine, arginine, lysine, ornithine, leucine, 2-phenylalanine, glutamic acid, threonine, etc.; organic sugars or sugar alcohols such as sucrose, lactose, lactitol, trehalose, stachyose, mannose, sorbitol, xylose, ribose, ribitol, inositol galactose, inositol, galactose, galactitol, glycerol, cyclic polyols (e.g. Inositol, polyethylene glycol; sulfur-containing reducing agents such as urea, glutathione, lipoic acid, sodium thioglycolate, thioglycerol, α-thioglycerol and sodium thiosulfate; low molecular weight proteins such as human serum albumin, bovine serum albumin, gelatin or other immunoglobulins; hydrophilic polymers such as polyvinylpyrrolidone; monosaccharides (e.g. xylose, mannose, fructose, glucose); disaccharides (e.g. lactose, maltose, sucrose); trisaccharides such as raffinose; and polysaccharides such as dextrin or dextran.

Nonionic surfactants or detergents (also known as "wetting agents") can be used to help solubilize therapeutic agents and protect therapeutic proteins from agitation-induced aggregation. This also allows the formulation to be exposed to shear surface stress without causing denaturation of the active therapeutic protein or antibody. Nonionic surfactants can be present in concentrations ranging from about 0.05 mg/ml to about 1.0 mg/ml, preferably from about 0.07 mg/ml to about 0.2 mg/ml.

Suitable nonionic surfactants include polysorbates (20, 40, 60, 65, 80, etc.), poloxamer (184, 188, etc.), polyols, polyoxyethylene sorbitol monoether (etc.), polylaurin 400, polyoxyethylene 40 stearate, polyoxyethylene hydrogenated castor oil 10, 50, and 60, glyceryl monostearate, sucrose fatty acid esters, methylcellulose, and carboxymethylcellulose. Anionic detergents that can be used include sodium dodecyl sulfate, sodium dioctyl sulfosuccinate, and sodium dioctyl sulfonate. Cationic detergents include benzalkonium chloride or benzyl chloride.

The dosage used for administration can be adjusted as a function of various parameters related to the relevant pathology or the desired duration of treatment, particularly as a function of the method of administration. To prepare the pharmaceutical composition, an effective amount of the conditionally active peptide, or a product further engineered from the conditionally active peptide, can be dissolved or dispersed in a pharmaceutically acceptable carrier or aqueous medium.

Suitable injectable pharmaceutical formulations include sterile aqueous solutions or dispersions; carriers such as sesame oil, peanut oil, or propylene glycol aqueous solutions; and sterile powders for the temporary preparation of sterile injectable solutions or dispersions. In all cases, the formulation must be sterile and, to some extent, flowable to facilitate injection. It must be stable under manufacturing and storage conditions and must be protected against contamination by microorganisms such as bacteria and fungi.

Solutions of conditionally active peptides, as free bases or pharmaceutically acceptable salts, can be prepared in water appropriately mixed with surfactants. Dispersions can also be prepared in glycerol, liquid polyethylene glycol and mixtures thereof, and oils. Under normal storage and use conditions, these formulations contain preservatives to prevent microbial growth.

Conditionally active peptides and products engineered from conditionally active peptides can be formulated into compositions in the form of salts. Pharmaceutically acceptable salts include acid addition salts (forming with the free amino group of a protein) that are formed with inorganic acids (such as hydrochloric acid or phosphoric acid) or organic acids (such as acetic acid, oxalic acid, tartaric acid, mandelic acid, etc.). Salts formed with free carboxyl groups can also be derived from inorganic bases (such as sodium hydroxide, potassium hydroxide, ammonium hydroxide, calcium hydroxide, or iron hydroxide) and organic bases (such as isopropylamine, trimethylamine, histidine, procaine, etc.).

The carrier can also be a solvent or dispersion medium containing, for example, water, ethanol, polyols (e.g., glycerol, propylene glycol, and liquid polyethylene glycol), suitable mixtures thereof, and vegetable oils. For example, by using a coating such as lecithin, the desired particle size can be maintained in the dispersed state, and appropriate flowability can be maintained by using surfactants. Microbial action can be prevented by various antimicrobial and antifungal agents, such as parabens, chlorobutanol, phenol, sorbic acid, thimerosal, etc. In many cases, isotonic agents, such as sugars or sodium chloride, are preferred. The absorption of the injectable composition can be prolonged by using agents that delay absorption, such as aluminum monostearate and gelatin.

A sterile injectable solution is prepared by incorporating the required amount of the conditionally active peptide with one or more other components listed above (if necessary) into a suitable solvent, followed by filtration and sterilization. Typically, a dispersion is prepared by incorporating various sterile active ingredients into a sterile carrier containing a base dispersion medium and any other desired components from those listed above. In the case of sterile powders used to prepare sterile injectable solutions, a preferred method of preparation is vacuum drying and freeze-drying techniques, which produce a powder of the active ingredient plus any other desired components from a previously sterile filtered solution.

The preparation of more concentrated or highly concentrated solutions for direct injection is also envisioned, in which the use of dimethyl sulfoxide (DMSO) as a solvent is expected to result in extremely rapid penetration, delivering high concentrations of the active agent to small tumor areas.

After preparation, the solution will be administered in a manner compatible with the dosage formulation and at a therapeutically effective amount. This preparation is readily administered in various dosage forms, such as the injectable solution type described above, but drug-release capsules, etc., may also be used.

For parenteral administration, such as in the form of an aqueous solution, the solution should be appropriately buffered (if necessary) and the liquid diluent should first be isotonic with sufficient saline or glucose. These specific aqueous solutions are particularly suitable for intravenous, intramuscular, subcutaneous, and intraperitoneal administration. In this regard, sterile aqueous media that can be used according to this disclosure are known to those skilled in the art. For example, a dose may be dissolved in 1 ml of isotonic NaCl solution, or added to 1000 ml of subcutaneous infusion solution, or injected at the recommended infusion site (see, for example, "Remington's Pharmaceutical Sciences," 15th Edition, pp. 1035-1038 and 1570-1580). Depending on the individual's condition, some dosage variation will inevitably occur. In any case, the person responsible for administration will determine the appropriate dose for the individual.

Conditionally active peptides and products engineered from conditionally active peptides can be formulated in a therapeutic mixture to deliver about 0.0001-10.0 mg, or about 0.001-5 mg, or about 0.001-1 mg, or about 0.001-0.1 mg, or about 0.1-1.0 mg, or even about 10 mg per dose. Multiple doses can also be administered at selected time intervals.

In addition to compounds formulated for parenteral administration, such as intravenous or intramuscular injection, other pharmaceutically acceptable forms include, for example, tablets or other solids for oral administration; delayed-release capsules; and any other forms currently in use.

In some embodiments, it is envisioned to use liposomes and/or nanoparticles to introduce conditionally active peptides or products derived from further engineered conditionally active peptides into host cells. The formation and use of liposomes and/or nanoparticles are known to those skilled in the art.

Nanocapsules can typically encapsulate compounds in a stable and reproducible manner. To avoid side effects caused by intracellular polymer overload, such ultrafine particles (approximately 0.1 μm in size) are usually designed using polymers that can degrade in vivo. It is envisioned that biodegradable polycyanoacrylate nanoparticles meeting these requirements could be used in this invention.

Liposomes are formed from phospholipids dispersed in an aqueous medium and spontaneously forming multilayered concentric bilayer vesicles (also known as multilayered vesicles (MLVs)). MLVs typically have a diameter of 25 nm to 4 μm. Sonication of MLVs results in the formation of small monolayer vesicles (SUVs) with diameters ranging from 200 nm to 4 μm, whose cores contain an aqueous solution. The physical properties of liposomes depend on pH, ionic strength, and the presence of divalent cations.

Pharmaceutical formulations containing conditionally active peptides or products derived from further engineered conditionally active peptides, as described in this invention, are prepared as lyophilized formulations or aqueous solutions by mixing with one or more optional pharmaceutically acceptable carriers (Remington's Pharmaceutical Sciences 16th edition, Osol, A.Ed. (1980)). Pharmaceutically acceptable carriers include, but are not limited to: buffers, such as phosphates, citrates, and other organic acids; antioxidants, including ascorbic acid and methionine; preservatives (e.g., octadecyl dimethyl benzyl ammonium chloride; hexamethyl diammonium chloride; benzalkonium chloride; benzyl chloride; phenolic alcohol, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (less than about 10 residues) multi-... Peptides; proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers such as polyvinylpyrrolidone; amino acids such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextrin; chelating agents such as EDTA; sugars such as sucrose, mannitol, trehalose, or sorbitol; salt-forming counterions such as sodium; metal complexes (e.g., Zn-protein complexes); and/or nonionic surfactants such as polyethylene glycol (PEG).

Exemplary pharmaceutically acceptable carriers described herein also include interstitial drug dispersants, such as soluble neutral active hyaluronidase glycoprotein (sHASEGP), such as human soluble PH-20 hyaluronidase glycoprotein, such as rHuPH20 (Baxter International, Inc.). Certain exemplary sHASEGPs and methods of use, including rHuPH20, are described in U.S. Patent Publications 2005/0260186 and 2006/0104968. In one aspect, sHASEGP is combined with one or more additional glycosaminoglycans, such as chondroitinase.

Exemplary lyophilized antibody formulations are described in U.S. Patent No. 6,267,958. Aqueous antibody formulations include those described in U.S. Patent No. 6,171,586 and WO2006/044908, the latter of which contain a histidine acetate buffer.

The formulations described herein may also contain more than one active ingredient as needed for a specific indication being treated. Preferably, ingredients having complementary activities that do not adversely affect each other may be combined into a single formulation. For example, in addition to the conditionally active antibodies, antibody fragments, or immunoconjugates of the present invention, it may be necessary to provide EGFR antagonists (e.g., erlotinib), anti-angiogenic agents (e.g., VEGF antagonists that may be anti-VEGF antibodies), or chemotherapeutic agents (e.g., taxanes or platinum preparations). These active ingredients are suitable to be present in combination in amounts effective for the intended purpose.

The active ingredient can be encapsulated in microcapsules, for example, prepared by coagulation techniques or by interfacial polymerization. For example, hydroxymethyl cellulose or gelatin microcapsules and poly(methyl methacrylate) microcapsules can be used, respectively, in colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles, and nanocapsules) or in macroemulsions. These techniques are disclosed in Remington's Pharmaceutical Sciences 16th edition, Osol, A.Ed. (1980).

Sustained-release formulations can also be prepared. Suitable examples of sustained-release formulations include a semi-permeable matrix of a solid hydrophobic polymer containing the antibody or antibody fragment, said matrix being in the form of a molded article, such as a film or microcapsule.

In some embodiments, the conditionally active peptide or a product engineered from the conditionally active peptide can be used to produce an article containing a substance for treating, preventing, and/or diagnosing said disease. The article includes a container and a label or instruction manual on or associated with the container. Suitable containers include, for example, bottles, vials, syringes, IV solution bags, etc. The container can be formed from a variety of materials, such as glass or plastic. The container contains the composition itself or a combination of the composition with another composition that is effective in treating, preventing, and/or diagnosing the condition, and may have a sterile inlet (e.g., the container may be an intravenous solution bag or a vial with a stopper that can be punctured by a hypodermic needle). At least one active agent in the composition is the conditionally active peptide of the present invention or a product further engineered from the conditionally active peptide. The label or instruction manual indicates that the composition is used to treat the selected condition. Furthermore, the article may comprise (a) a first container containing the composition, wherein the composition contains the conditionally active peptide or a product engineered from the conditionally active peptide; and (b) a second container containing the composition, wherein the composition contains additional cytotoxins or other therapeutic agents. The article in this embodiment of the invention may also include instruction manual indicating that the composition can be used to treat a specific condition. Alternatively or additionally, the article may also comprise a second (or third) container containing pharmaceutically acceptable buffer solutions, such as water for injection (BWFI), phosphate-buffered saline, Ringer's solution, and glucose solution. It may also include other substances desired from a commercial and user perspective, including other buffer solutions, diluents, filters, needles, and syringes.

The article may optionally include a container as part of a parenteral, subcutaneous, intramuscular, intravenous, intra-articular, intrabronchial, intra-abdominal, intracystic, intracartilaginous, intracavitary, intrabody, intracerebellar, intravenous, intracolonic, intracervical, intrastomal, intrastomal, intramuscular, intrapelvic, intraperitoneal, intraperitoneal, intrapleural, intraprostatic, intrapulmonary, intrarectal, intrarenal, intraretinal, intraspinal, intrasynovial, intrathoracic, intrauterine, intrabladder, pill, vagina, rectum, oral cavity, sublingual, intranasal, or transdermal delivery device or system.

Conditionally active peptides or products engineered from conditionally active peptides may be included in a medical device wherein the device is adapted to contact or administer the conditionally active peptide or products further engineered from the conditionally active peptide by at least one of the following modes: parenteral, subcutaneous, intramuscular, intravenous, intra-articular, intrabronchial, intraperitoneal, intracystic, intracartilaginous, intracavitary, intracavitary, intrabody, intracerebellar, intravenous, intracolonic, intracervical, intragastric, intrahepatic, intramyocardial, intraosseous, intrapelvic, intraperitoneal, intraperitoneal, intraperitoneal, intrapleural, intraprostatic, intrapulmonary, intrarectal, intrarenal, intraretinal, intraspinal, intrasynovial, intrathoracic, intrauterine, intrabladder, pill, vagina, rectum, oral cavity, sublingual, intranasal, or transdermal absorption.

In other embodiments, the conditionally active peptide or a product engineered from the conditionally active peptide may be contained in the kit in a lyophilized form in a first container, and optionally in a second container containing sterile water, sterile buffered water, or at least one preservative selected from the following compounds: phenol, m-cresol, p-cresol, o-cresol, chlorocresol, benzyl alcohol, phenylmercuric nitrite, phenoxyethanol, formaldehyde, chlorobutanol, magnesium chloride, alkyl p-hydroxybenzoate, benzalkonium chloride, benzyl chloride, sodium dehydroacetate, and thimerosal, or mixtures thereof in an aqueous diluent solution. In one aspect, in the kit, the concentration of the conditionally active peptide or a product engineered from the conditionally active peptide in the first container is restored to a concentration of about 0.1 mg/ml to about 500 mg/ml by the contents of the second container. In another aspect, the second container further includes an isotonic agent. In yet another aspect, the second container also contains a physiologically acceptable buffer. In one aspect, the invention provides a method of treating at least one parent protein-mediated condition, comprising administering the formulation provided in the kit to a patient in need, and reconstitution prior to administration.

The following examples are illustrative and not intended to limit the invention. Various suitable modifications and adaptations to conditions and parameters commonly encountered in the art and readily apparent to those skilled in the art are within the scope of this disclosure.

Example

Examples 1-5: Examples 1-5 for the preparation of conditionally active peptides have been described in U.S. Patent No. 8,709,755B2, which is incorporated herein by reference.

Example 6: Evolution of the light or heavy chain of the antibody

CPE was used to evolve the heavy and light chains of antibodies F1-10F10, respectively. Light chain mutants were screened to identify 26 light chain mutants with conditional activity, in which the mutants were more active than the wild type at pH 6.0 and less active than the wild type at pH 7.4. These 26 light chain mutants had mutations at 8 different positions in the light chain. More than 5 of the 26 light chain mutants had mutations at 3 of these 8 positions. These 3 positions were considered hotspots in the light chain. Heavy chain mutants were screened to identify 28 heavy chain mutants with conditional activity. These 28 heavy chain mutants had mutations at 8 different positions in the heavy chain. More than 5 of the 28 heavy chain mutants had mutations at 3 of these 8 positions. These 3 positions were considered hotspots in the heavy chain. The conditional activity of the light and heavy chain mutants was confirmed by ELISA.

The optimal conditionally active antibody produced by this example exhibits a 17-fold difference in activity at pH 6.0 compared to pH 7.4. Furthermore, many conditionally active antibodies exhibit reversible activity at pH levels between the normal physiological pH 7.4 and the abnormal pH 6.0. Interestingly, when the activity of conditionally active antibodies was tested by ELISA within a pH range of 5.0–7.4, most of the conditionally active antibodies produced by this example showed optimal binding activity at approximately pH 5.5–6.5.

The activity of the conditionally active antibody generated in this embodiment was also confirmed using whole-cell FACS (fluorescence-activated cell sorting) assays, in which CHO cells were used to express the antibody antigen at pH 6.0 and pH 7.4. The conditionally active antibody was added to CHO cells to measure binding activity. The FACS assay confirmed the overall trend in the results of an ELISA assay used to measure the selectivity of the conditionally active antibody at pH 6.0 relative to pH 7.4.

Example 7: Selection of conditionally active antibodies in specific buffer solutions

The mutant antibodies generated according to the present invention via an evolutionary step were tested at normal physiological pH 7.4 and at an abnormal pH 6.0. Both tests were performed using a phosphate-buffered saline (PBS) solution containing bicarbonate, which is found in human serum. The concentration of bicarbonate in the solution was typical of human serum, i.e., physiological concentration. Comparative tests were performed using the same PBS solution without bicarbonate.

In this embodiment, the test for measuring the binding activity of the mutant antibody or the conditionally active antibody is an ELISA test, which is performed according to the following steps:

1. The day before ELISA: Coat the wells with 100 μl of Ab-AECD his tag (histidine tag) antigen (2.08 mg/ml) at a concentration of 1 μg/ml in PBS.

3. Remove the buffer solution from the 96-well plate coated with antibody Ab-A-His antigen and blot dry on a paper towel.

4. Wash the plate three times with buffer N or PBS.

5. Block the plate with 200 μl of the specified buffer at room temperature for 1 hour.

6. Dilute the selected CPE/CPS mutant and wild-type proteins to 75 ng/ml in the designated buffer solution according to the layout. The pH of the buffer solution should be set to 6.0 or 7.4 (hereinafter referred to as the "designated buffer solution").

6. Discard the buffer solution and add 100 μl of 75 ng/ml sample to each well according to the plate layout.

7. Incubate the plate at room temperature for 1 hour.

8. Tap the buffer solution out of the 96-well plate and blot dry on a paper towel.

9. Wash the plate a total of 3 times with 200 μl of the specified buffer according to the layout.

10. Prepare anti-Flag HRP at a dilution of 1:5000 in the specified buffer solution, and add 100 μl of Anti-Flag horseradish peroxide (HRP) to each well according to the layout.

11. Incubate the plate at room temperature for 1 hour.

13. Wash the plate three times with 200 μl of the specified buffer solution.

14. Develop the plate with 50 μl of 3,3',5,5'-tetramethylbenzidine (TMB) for 1.5 minutes.

The test was found to significantly increase the success rate of selecting conditionally active antibodies in PBS buffer containing bicarbonate. Furthermore, conditionally active antibodies selected using PBS buffer containing bicarbonate tended to exhibit a much higher ratio of activity at pH 6.0 to activity at pH 7.4, thus providing significantly higher selectivity.

Further observation revealed that when conditionally active antibodies selected using PBS buffer with bicarbonate were tested in PBS buffer without bicarbonate, the selectivity of the conditionally active antibodies at pH 6.0 was significantly reduced compared to that at pH 7.0. However, the same selectivity of the conditionally active antibodies was restored when physiological amounts of bicarbonate were added to the PBS buffer.

In another test, the selected conditionally active antibody was tested in Krebs buffer supplemented with bicarbonate. A higher ratio of activity at pH 6.0 to activity at pH 7.4 was also observed in this bicarbonate-supplemented Krebs buffer. This appears to be at least partly due to the presence of bicarbonate in the Krebs buffer solution.

When the concentration of bicarbonate in PBS buffer was reduced to below its physiological concentration, enhanced activity of the conditionally active antibody was observed at the normal physiological pH of 7.4. This enhanced activity of the conditionally active antibody at pH 7.4 was correlated with the decrease in bicarbonate concentration in the PBS buffer.

When tested at pH 7.4, wild-type antibodies are not affected by different amounts of bicarbonate in the PBS buffer solution because their activity remains the same at all bicarbonate concentrations in the PBS buffer solution used to test conditionally active antibodies.

Example 8: Selection of conditionally active antibodies in different buffer solutions

The mutant antibody generated according to the present invention via an evolutionary step was subjected to ELISA testing at normal physiological pH (7.4) and at abnormal pH (6.0). The two ELISA tests were performed using different buffers, including a Krebs-based buffer containing bovine serum albumin (BSA) and a PBS-based buffer containing bicarbonate and BSA.

The ELISA test is performed according to the following steps:

1. The day before ELISA: Coat the wells with 100 μl of Ab-AECD his tag (2.08 mg/ml) antigen at a concentration of 1 μg/ml in coating buffer (carbonate-bicarbonate buffer).

2. Remove the buffer solution from the 96-well plate coated with antibody Ab-A-His antigen and blot dry on a paper towel.

3. Wash the plate three times with 200 μl of 20 buffer solution.

4. Block the plate with 200 μl of 20 buffer solution at room temperature for 1 hour.

5. Dilute the mutants and chimeras to 75 ng/ml in 20 buffer solutions according to the layout.

6. Remove the buffer solution and add 100 μl of diluted sample to each well according to the plate layout.

7. Incubate the plate at room temperature for 1 hour.

8. Tap the buffer solution out of the 96-well plate and blot the plate dry with a paper towel.

9. Wash the plate a total of 3 times with 200 μl of 20 buffer solution.

10. Prepare anti-flag IgG HRP at a dilution of 1:5000 in 20 buffer solutions. Add 100 μl of anti-flag IgG HRP solution to each well.

11. Incubate the plate at room temperature for 1 hour.

12. Wash the plate a total of 3 times with 200 μl of 20 buffer solution.

13. Develop the plate with 50 μl of 3,3',5,5'-tetramethylbenzidine for 30 seconds.

Conditionally active antibodies selected for tests using PBS buffer with bicarbonate showed a significantly higher ratio of activity at pH 6.0 to activity at pH 7.4 compared to those selected for tests using PBS buffer without bicarbonate. Furthermore, Krebs buffer with added bicarbonate also provided a higher ratio of activity at pH 6.0 to activity at pH 7.4 when compared to tests using PBS buffer without bicarbonate. It appears that bicarbonate is important for selecting the desired conditionally active antibody.

Example 9: Selection of conditionally active antibodies in different buffer solutions

In this example, conditionally active antibodies against the antigen that exhibited higher activity than wild-type antibodies at pH 6.0 and lower activity than wild-type antibodies at pH 7.4 were screened. The screening steps were performed using the buffers listed in Tables 2 and 3 below. The buffers in Table 2 are based on Krebs buffer, and other added components are shown in column 1 of Table 2.

Some test buffers with added components based on PBS buffer are shown in Table 3 below. Please note that the components in the buffers in Tables 2 and 3 are expressed in grams (grams) added per liter of buffer. However, the concentration of human serum is 10 wt% of the buffer.

Table 3. Test buffers based on PBS buffer

Screening was performed using ELISA tests with these test buffers. The ELISA tests were performed as described in Examples 7-8. The conditionally active antibodies selected from the 10 test buffers are shown in Table 4 below. OD 450 absorbance was negatively correlated with binding activity in the ELISA test.

Table 4. Conditionally active antibodies (mutants) selected using different test buffers

The selectivity of some of the selected conditionally active antibodies was confirmed using buffers 8 and 9, and it was found that they indeed exhibited the desired selectivity at pH 6.0 (relative to pH 7.4), as shown in Figure 1. Note that using different buffers affects the selectivity of the conditionally active antibodies.

Example 10: Activity of conditionally active antibodies in different buffer solutions

The activity of conditionally active antibodies derived from two monoclonal antibodies (mAb 048-01 and mAb 048-02 as parental antibodies) was measured in two different buffers (Figure 2). The two buffers were phosphate-buffered saline (condition IV) and Krebs buffer (condition I). Six conditionally active antibodies were derived from mAb 048-01: CAB Hit 048-01, CAB Hit 048-02, CAB Hit 048-03, CAB Hit 048-04, CAB Hit 048-05, and CAB Hit 048-06. Three conditionally active antibodies were derived from mAb 048-02: CAB Hit 048-07, CAB Hit 048-08, and CAB Hit 048-09.

This study shows that the selectivity of conditionally active antibodies (the ratio of activity in a pH 6.0 test to activity in a pH 7.4 test) is affected by the buffer used in the test. The conditionally active antibody evolved from wild-type mAb 048-02 showed significantly higher selectivity in Krebs buffer than in phosphate buffer (Figure 2).

Example 11: Selectivity of conditionally active antibodies and bicarbonate

In Example 10, a higher selectivity for the conditionally active antibody was observed in Krebs buffer (condition I) than in phosphate buffer (condition IV). This involved identifying the component in the Krebs buffer that contributed most significantly to the higher selectivity observed in Example 10. The selectivity of a conditionally active antibody was tested again in a buffer derived from Krebs buffer, from which one component was subtracted one at a time (Figure 3, left column group). When using the full Krebs buffer, the conditionally active antibody exhibited high selectivity, with an activity ratio of approximately 8 at pH 6.0/7.4. While the selectivity of the conditionally active antibody decreased with each subtraction of components A through F from the Krebs buffer, it was not lost. However, when component G (bicarbonate) was subtracted from the Krebs buffer, the selectivity of the conditionally active antibody was completely lost. See Figure 3. This indicates that the high selectivity of the conditionally active antibody in Krebs buffer is at least partially due to bicarbonate.

The selectivity of the same conditionally active antibody was then measured in phosphate-buffered saline (condition IV) without bicarbonate, and a complete loss of selectivity was observed in phosphate-buffered saline. When bicarbonate was added to phosphate-buffered saline, the selectivity of the conditionally active antibody recovered to the level observed in Krebs buffer. This confirms that bicarbonate is essential for the selectivity of this conditionally active antibody.

Example 12: Bicarbonate inhibits binding at pH 7.4

This example measured the binding activity of three conditionally active antibodies (CAB HitA, CAB Hit B, and CAB Hit C) at pH 7.4 in buffer solutions with different concentrations of bicarbonate (from 0 to physiological concentrations of bicarbonate (approximately 20 mM)) (Figure 4). It was observed that the binding activity of all three conditionally active antibodies at pH 7.4 decreased in a dose-dependent manner as the bicarbonate concentration increased from 0 to physiological concentration (Figure 4). On the other hand, the binding activity of the wild-type antibody was not affected by bicarbonate. This study suggests that the selectivity of conditionally active antibodies in the presence of bicarbonate may be at least partially attributed to the loss of binding activity of conditionally active antibodies at pH 7.4 due to interaction with bicarbonate.

Example 13: Activity of conditionally active antibodies against ROR2 in different buffer solutions

The conditionally active antibodies against ROR2 were tested in different buffers using test solutions containing sodium bicarbonate (as described in Example 9): CAB-P was a standard phosphate saline buffer (PBS buffer) used at pH 6.0 or 7.4; CAB-PSB was a PBS buffer at pH 6.0 or 7.4 supplemented with 15 mM sodium bicarbonate; and CAB-PSS was a PBS buffer at pH 6.0 or 7.4 supplemented with 10 mM sodium sulfide nonahydrate.

The activity of these conditionally active antibodies was measured using the following ELISA protocol:

1. The day before ELISA: Coat the plate with 100 μl of antigen in PBS at a concentration of 1 μg/ml at 4°C.

2. Wash the plate twice with 200 μl of CAB-P, CAB-PSB, or CAB-PSS buffer, depending on the plate layout.

3. Depending on the plate layout, block the plate with 200 μl of CAB-P, CAB-PSB or CAB-PSS buffer at room temperature for 1 hour.

4. Dilute the antibody sample and positive control in CAB-P, CAB-PSB, or CAB-PSS buffer as shown in the plate layout.

5. Tap the blocking buffer out of the 96-well plate and blot dry on a paper towel.

6. Add 100 μl of diluted antibody sample, positive control, or negative control to each well according to the plate layout.

6. Incubate the plate at room temperature for 1 hour.

7. Prepare the second antibody in CAB-P, CAB-PSB, or CAB-PSS buffer according to the plate layout.

8. Tap the buffer solution out of the 96-well plate and blot it dry on a paper towel.

9. Wash the plate a total of 3 times with 200 μl of CAB-P, CAB-PSB or CAB-PSS buffer, depending on the plate layout.

10. Depending on the plate layout, add the second antibody diluted in CAB-P, CAB-PSB, or CAB-PSS buffer to each well.

11. Incubate the plate at room temperature for 1 hour.

12. Tap the buffer solution out of the 96-well plate and blot it dry on a paper towel.

13. Wash the plate a total of 3 times with CAB-P, CAB-PSB or CAB-PSS buffer.

14. Place the 3,3',5,5'-tetramethylbenzidine (TMB) substrate at room temperature.

15. Tap the buffer solution off the plate and blot it dry on a paper towel.

16. Add 50 μl of TMB substrate.

17. Stop the color development with 50 μl of 1N HCl. The color development time is 3 minutes.

18. Use a plate reader to take readings at OD 450nm.

The activities of these conditionally active antibodies against ROR2 are presented in Figure 7. In CAB-PSB buffer, the conditionally active antibodies showed higher activity at pH 6.0 than at pH 7.4, indicating selectivity at pH 6.0 relative to pH 7.4. This selectivity of several conditionally active antibodies was lost or significantly reduced in CAB-P buffer. However, this selectivity at pH 6.0 relative to pH 7.4 was also observed in CAB-PSS buffer. Conversely, wild-type antibodies showed relatively minimal or no selectivity in any buffer.

This example demonstrates that thiosulfate and bicarbonate have similar functions in mediating the conditional binding of the tested conditionally active antibody against ROR2.

Example 14: Activity of conditionally active antibodies against Axl in different buffer solutions

Conditional antibodies against Axl selected using test solutions containing sodium bicarbonate (as described in Example 9) were tested in different buffers as described in Example 13: CAB-P, CAB-PSB, and CAB-PSS. The activity of these conditional antibodies against Axl was measured using the same ELISA protocol as described in Example 13 and is presented in Figure 8. In CAB-PSB buffer, the conditional antibodies showed higher activity at pH 6.0 than at pH 7.4, i.e., selectivity at pH 6.0 versus pH 7.4. This selectivity of these conditional antibodies was lost or significantly reduced in CAB-P buffer. This selectivity at pH 6.0 versus pH 7.4 was also observed in CAB-PSS buffer. In contrast, wild-type antibodies showed essentially no selectivity in any buffer.

This embodiment also demonstrates that thiosulfate and bicarbonate have similar functions in mediating the conditional binding of the tested conditionally active antibody against Axl.

Example 15: Generation of conditionally active antibodies against Axl in a test solution containing hydrosulfide ions

The test solution used in this embodiment was prepared as follows:

CAB-PSB buffer pH 6.0 (containing 1% BSA)

1. Add 1.26 g/L sodium bicarbonate (Sigma S5761) to PBS-Cellgro to achieve a final sodium bicarbonate concentration of 15 mM.

2. Add BSA (MP, CATNo0218054991) to a final concentration of 1% (in 1 liter).

3. Adjust the pH to 6.0 with 1N HCl while stirring.

4. Store at 4°C. Check pH before use. Adjust pH with 1N HCl if necessary.

CAB-PSB buffer pH 7.4 (containing 1% BSA)

1. Add 1.26 g/L sodium bicarbonate (Sigma S5761) to PBS-Cellgro to achieve a final sodium bicarbonate concentration of 15 mM.

2. Add BSA (MP, CATNo0218054991) to a final concentration of 1% (in 1 liter).

3. Adjust the pH to 7.4 with 1N HCl while stirring.

4. Store at 4°C. Check pH before use. Adjust pH with 1N HCl if necessary.

CAB-PSS buffer with 10 mM hydrosulfide, pH 6.0 (containing 1% BSA)

1. Add 2.4 g/L sodium sulfide nonahydrate ( _Na₂S · _9H₂O ) (ACROS, #424425000) to PBS-Cellgro, with a final concentration of 10 mM.

2. Add BSA (MP, CATNo0218054991) to a final concentration of 1% (in 1 liter).

3. Adjust the pH to 6.0 with 1N HCl while stirring.

4. Store at 4°C. Check pH before use. Adjust pH with 1N HCl if necessary.

CAB-PSS buffer with 10 mM hydrosulfide, pH 7.4 (containing 1% BSA)

1. Add 2.4 g/L sodium sulfide nonahydrate ( _Na₂S · _9H₂O ) (ACROS, #424425000) to PBS-Cellgro, with a final concentration of 10 mM.

2. Add BSA (MP, CATNo0218054991) to a final concentration of 1% (in 1 liter).

3. Adjust the pH to 7.4 with 1N HCl while stirring.

4. Store at 4°C. Check pH before use. Adjust pH with 1N HCl if necessary.

Specifically, the 1mM CAB-PSS buffer solution is pH 6.0 (containing 1% BSA).

1. Dilute BioAtla CAB-PSS buffer pH 6.0 (containing 1% BSA) 10M by 1/10 to a final concentration of 1mM.

2. Adjust the pH to 6.0 with 1N HCl while stirring.

3. Store at 4°C. Check pH before use. Adjust pH with 1N HCl if necessary.

Specifically, the 1mM CAB-PSS buffer solution has a pH of 7.4 (containing 1% BSA).

1. Dilute BioAtla CAB-PSS buffer pH 7.4 (containing 1% BSA) 10M by 1/10 to a final concentration of 1mM.

2. Adjust the pH to 7.4 with 1N HCl while stirring.

3. Store at 4°C. Check pH before use. Adjust pH with 1N HCl if necessary.

CAB-P buffer pH 6.0 (containing 1% BSA)

1. Add BSA to PBS-Cellgro to a final concentration of 1% (in 1 liter).

2. Adjust the pH to 6.0 with 1N HCl while stirring.

3. Store at 4°C. Check pH before use. Adjust pH with 1N HCl if necessary.

CAB-P buffer pH 7.4 (containing 1% BSA)

1. Add BSA to PBS-Cellgro to a final concentration of 1% (in 1 liter).

2. Adjust the pH to 7.4 with 1N HCl while stirring.

3. Store at 4°C. Check pH before use. Adjust pH with 1N HCl if necessary.

The method of this invention is performed using a wild-type antibody against Axl to generate a mutant antibody using a similar protocol described in the foregoing examples. The mutant antibody is tested using a test solution containing 10 mM hydrosulfide ions at pH 6.0 or pH 7.4 to select a conditionally active antibody. The selected conditionally active antibody (BAP063.1-CAB1-8) is shown in Figure 9.

When tested in a 10 mM hydrosulfide concentration, the selected conditionally active antibodies exhibited higher binding activity to Axl at pH 6.0 than at pH 7.4. However, in a test solution with only 1 mM hydrosulfide, the activity differences between pH 6.0 and pH 7.4 were significantly reduced for all selected conditionally active antibodies except for one. See Figure 9.

It should be noted that, as used herein and in the appended claims, the singular forms “a” and “the” include plural references unless the context clearly specifies otherwise. Furthermore, the terms “a”, “one or more”, and “at least one” are used interchangeably herein. The terms “comprising,” “including,” “having,” and “consisting of” are also used interchangeably.

Unless otherwise stated, all figures used in the specification and claims to indicate the amount and properties of components, such as molecular weight, percentage, ratio, reaction conditions, etc., should be understood to be modified by the term "about" in all cases, regardless of whether the term "about" is present. Therefore, unless indicated otherwise, the numerical parameters set forth in the specification and claims are approximate values that may vary according to the desired properties sought to be obtained according to the invention. At least, and without attempting to limit the application of the principle of equivalence to the scope of the claims, each numerical parameter should be interpreted at least according to the number of significant figures reported and by applying ordinary rounding techniques. Although the numerical ranges and parameters describing the broad scope of the invention are approximate, in specific embodiments we report the listed values as precisely as possible. However, any numerical value necessarily inherently contains some error caused by the standard deviation found in the respective test measurements of these values.

It should be understood that each component, compound, substituent or parameter disclosed herein should be interpreted as disclosed for use alone or in combination with one or more of each other component, compound, substituent or parameter disclosed herein.

It should also be understood that each amount/value or range of each component, compound, substituent, or parameter disclosed herein is to be interpreted as also disclosed in combination with each amount/value or range of any other component, compound, substituent, or parameter disclosed herein. Therefore, for the purposes of this specification, any combination of amounts/values or ranges of two or more components, compounds, substituents, or parameters disclosed herein is also disclosed in combination with each other.

It should also be understood that each range disclosed herein should be interpreted as a disclosure of each specific value within a range having the same number of significant figures. Therefore, the ranges 1-4 should be interpreted as an explicit disclosure of values 1, 2, 3, and 4. It should also be understood that each lower limit of each range disclosed herein should be interpreted as a combination of each upper limit of each range for the same component, compound, substituent, or parameter and each specific value within each range disclosed herein. Therefore, the invention is interpreted as a disclosure of all ranges obtained by combining each lower limit of each range with each upper limit of each range or each specific value within each range, or by combining each upper limit of each range with each specific value within each range.

Furthermore, the specific amounts/values of components, compounds, substituents, or parameters disclosed in the specification or examples should be interpreted as disclosures of a lower or upper limit of a range, and thus can be combined with any other lower or upper limit or specific amounts/values disclosed elsewhere in this application for a range of the same component, compound, substituent, or parameter to form a range for that component, compound, substituent, or parameter.

All references to this document are incorporated herein by way of citation in their entirety, or provide for the disclosures to which they particularly rely. The applicant does not intend to make any disclosed embodiments publicly available, and any disclosed modifications or alterations that may not, in their literal sense, fall outside the scope of the claims are considered part of this invention under the doctrine of equivalents.

However, it should be understood that although numerous features and advantages of the invention, as well as details of its structure and function, have been set forth in the foregoing description, this disclosure is merely illustrative and may be subject to detailed changes under the principles of the invention, particularly changes to the shape, size, and arrangement of components, within the full scope indicated by the broad general meaning of the terms used to express the appended claims.

Claims (16)

1. A method for generating a conditionally active antibody, single-chain antibody, or antibody fragment from a parental antibody, a single-chain antibody, or an antibody fragment, the method comprising the steps of: (i) The parent antibody, single-chain antibody or antibody fragment is evolved by mutating at least one amino acid to produce one or more mutated antibodies, single-chain antibodies or antibody fragments, wherein the antibody fragment is selected from Fab, Fab', (Fab') ₂ , Fv and SCA fragments capable of binding epitopes of antigens; (ii) Perform a first ELISA test and a second ELISA test on the one or more mutated antibodies, single-chain antibodies or antibody fragments, wherein the first ELISA test is performed at a first pH of 6.0-6.8 in the presence of small molecules or ions to measure the binding activity of the one or more mutated antibodies, single-chain antibodies or antibody fragments, and the second ELISA test is also performed at a second pH of 7.2-7.8 in the presence of the small molecules or ions to measure the binding activity of the one or more mutated antibodies, single-chain antibodies or antibody fragments, wherein the small molecules or ions have a molecular weight of less than 100 a.m.u. and a pKa that differs from the first pH by at most 4 pH units; (iii) Select a mutated antibody, single-chain antibody, or antibody fragment whose binding activity in the first test is at least 3.0 to the binding activity in the second test; (iv) Obtain the conditionally active antibody, single-chain antibody, or antibody fragment by selecting one of the mutated antibodies, single-chain antibodies, or antibody fragments selected in step (iii) whose activity is pH-independent in the absence of the small molecule or ion. The small molecules or ions and their concentrations are selected from 3mM-200mM bicarbonate ions, 10mM-100mM hydrosulfide ions, and 100μm-100mM hydrogen sulfide. 2. The method according to claim 1, wherein the parent antibody, single-chain antibody, or antibody fragment is a heavy chain or light chain of an antibody, and the at least one amino acid is located within the complementarity-determining region of the parent antibody, single-chain antibody, or antibody fragment. 3. The method according to claim 2, wherein the at least one amino acid is located in the Fc region of the parent antibody, single-chain antibody, or antibody fragment. 4. The method of claim 3, wherein the evolution step comprises replacing the Fc region with a variable region of a different antibody to form a bispecific antibody. 5. The method of claim 3, wherein the evolution step comprises replacing the Fc region with a variable region of a different antibody to form a single-chain antibody. 6. The method according to claim 2, wherein the at least one amino acid is located within the backbone region of the parent antibody, single-chain antibody, or antibody fragment. 7. The method according to any one of claims 1-6, wherein the evolution step comprises substitution, insertion, deletion, or a combination thereof. 8. The method of claim 1, wherein step (ii) further comprises performing the following tests on the one or more mutated antibodies, single-chain antibodies, or antibody fragments: a test measuring the binding activity of the one or more mutated antibodies, single-chain antibodies, or antibody fragments at a first pH in the absence of the small molecule or ion, and a test measuring the binding activity of the one or more mutated antibodies, single-chain antibodies, or antibody fragments at a second pH in the absence of the small molecule or ion; step (iii) further comprises selecting mutated antibodies, single-chain antibodies, or antibody fragments having pH-independent binding activity in the absence of the small molecule or ion. 9. The method of claim 1, wherein the ratio of binding activity in the first test to binding activity in the second test is at least 4.0. 10. The method according to claim 1, wherein the small molecule or ion is bicarbonate. 11. The method according to claim 1, wherein the small molecule or ion is hydrogen sulfide. 12. The method according to claim 1, wherein the pKa of the small molecule or ion differs from the first pH by at most 0.5, 1, 2, 3 or 4 pH units. 13. The method of claim 1, wherein the pKa of the small molecule or ion is between the first pH and the second pH. 14. The method according to claim 1, wherein the pKa of the small molecule or ion is greater than 6.2 and less than 7.2. 15. The method of claim 1, wherein the antibody, single-chain antibody, or antibody fragment is a non-naturally occurring mutated antibody, single-chain antibody, or antibody fragment evolved from a parent antibody, single-chain antibody, or antibody fragment. 16. The method of claim 15, wherein the mutated antibody, single-chain antibody, or antibody fragment has a higher proportion of charged amino acid residues compared to the parent antibody, single-chain antibody, or antibody fragment.