WO2023012353A1 - Compounds for use in the therapeutic treatment of batten disease - Google Patents

Abstract

The invention relates to the use of selective estrogen receptor modulators (SERMs) for the therapeutic treatment of Batten Disease. Batten Disease is the common name for a group of recessively inherited fatal diseases of the nervous system also known as neuronal ceroid lipofuscinoses (NCLs), which typically arise in childhood and are the most common rare neurodegenerative disorders in children. SERMs have been found reduce intracellular accumulation of globotriaosylceramide (Gb3) though a mechanism of action that is surprisingly independent of the estrogen receptors. The scope of the invention also includes an in vitro method of assessing the effectiveness of a therapeutic treatment administered to a subject suffering from a neuronal ceroid lipofuscinosis, as well as an in vitro method of diagnosing a neuronal ceroid lipofuscinosis in a subject, which are both based on the finding that an increased level of globotriaosylceramide (Gb3) is a marker for Batten Disease.

Description

Compounds for use in the therapeutic treatment of Batten Disease

The present invention relates to a novel therapeutic use of known compounds for Batten Disease, in particular selective estrogen receptor modulators (SERMs). Selective estrogen receptor modulators (SERMs) are a group of nonsteroidal compounds that have estrogen-like effects (i.e., agonism) on some tissues (such as bone, skin, heart or vaginal epithelium), but antiestrogen effects (i.e., antagonism) on other tissues (such as breast or uterus).

Currently, there are two classes of clinically approved SERMs: triphenylethylene derivatives (e.g., tamoxifen) and benzothiophene derivatives (e.g., raloxifene).

Typical indications for SERMs include treatment or prevention of breast cancer (e.g., tamoxifen, toremifene, raloxifene), treatment or prevention of postmenopausal osteoporosis (e.g., raloxifene, bazedoxifene) and amelioration of symptoms of menopause symptoms (e.g., ospemifene).

The following is a non-exhaustive list of compounds within the group of SERMs: tamoxifen, (E) tamoxifen, toremifene, raloxifene, clomifene, ospemifene, bazedoxifene, nafoxidine, lasofoxifene, zuclomifene, afimoxifen, N-Desmethyl tamoxifene, droloxifene, tamoxifen aziridine, Idoxifene, 2- Methyl-4-hydroxytamoxifen, endoxifen, phenyltoloxamine, tamoxifen N-oxide, tamoxifen epoxide, diethylaminoethoxyhexestrol, etoloxamine, tesmilifene, tamoxifen-d5, myoparkil.

The IUPAC names and structural formulae of the aforementioned SERMs are provided below.

Tamoxifen: {2-[4-( 1 ,2-diphenylbut-1 -en-1 -yl)phenoxy]ethyl}dimethylamine

Raloxifene: [6-hydroxy-2-(4-hydroxyphenyl)-benzothiophen-3-yl]-[4-[2-(1-piperidyl)ethoxy]phenyl]- methanone

Toremifene: 2-(p-[(Z)-4-chloro-1 ,2-diphenyl- 1 -butenyl]phenoxy)-N,N-dimethylethylamine

Clomifene: (2-[4-(2-chloro-1 ,2-diphenylethenyl)phenoxy]-N,N-diethylethanamine;2-hydroxy-1 ,2,3- propanetricarboxylate) -chloro-1 ,2-diphenylbut-1 -enyl]phenoxy]ethanol

Bazedoxifene: 1-[[4-[2-(azepan-1 -yl)ethoxy]phenyl]methyl]-2-(4-hydroxyphenyl)-3-methylindol-5-ol

Lasofoxifene: (5R,6S)-6-phenyl-5-[4-(2-pyrrolidin-1 -yl-ethoxy)-phenyl]-5,6,7,8-tetrahydro- naphthalen-2-ol

Afimoxifene: (4-hydroxy-tamoxifen) (Z)-4-(1 -(4-(2-(dimethylamino)ethoxy)phenyl)-2-phenylbut-1 - enyl)phenol

Droloxifene : 3-[(E)-1 -[4-[2-(Dimethylamino)ethoxy]phenyl]-2-phenylbut-1 -enyl]phenol

Tamoxifen aziridine: 1 -[2-[4-[(Z)-1 ,2-diphenylbut-1 -enyl]phenoxy]ethyl]aziridine

2-Methyl-4-hydroxytamoxifen: 4-[(E)-1 -[4-[2-(dimethylamino)ethoxy]phenyl]-2-phenylbut-1 -enyl]-

3-methylphenol

Phenyltoloxamine: 2-(2-Benzylphenoxy)-N,N-dimethylethanamine

Tamoxifen N-oxide: 2-[4-[(Z)-1 ,2-diphenylbut-1 -enyl]phenoxy]-N,N-dimethylethanamine oxide

Tamoxifen epoxide: 2-[4-[(2S,3R)-3-ethyl-2,3-diphenyloxiran-2-yl]phenoxy]-N,N- dimethylethanamine

Diethylaminoethoxyhexestrol (Coralgil): 2-[4-[4-[4-[2-(diethylamino)ethoxy]phenyl]hexan-3- yl]phenoxy]-N,N-diethylethan amine

Etoloxamine: 2-(2-benzylphenoxy)-N,N-diethylethanamine

Tesmilifene: 2-(4-Benzylphenoxy)-N,N-diethylethanamine

Tamoxifen-d5: N,N-dimethyl-2-[4-(3,3,4,4,4-pentadeuterio-1 ,2-diphenylbut-1 - enyl)phenoxy]ethanamine

The present inventors have surprisingly identified a new activity of compounds belonging to the class of SERMs that ameliorate major clinical hallmarks of Batten Disease and therefore make these compounds suitable for the treatment of Batten Disease. Batten Disease is the common name for a group of recessively inherited fatal diseases of the nervous system also known as neuronal ceroid lipofuscinoses (NCLs), which typically arise in childhood and are the most common rare neurodegenerative disorders in children. The neuronal ceroid Lipofuscinosis (NCLs or CLNs or Batten disease, BD) affect children and adults with a combined incidence rate ranging between 1 :14,000 and 1 :67,000. NCLs are progressive degenerative brain diseases clinically characterized by a decline of mental and other capacities, epilepsy, and vision loss due to retinal degeneration, and histopathologically characterized by intracellular accumulation of an autofluorescent material, called ceroid lipofuscin, in the neuronal cells in the brain and retina. Eventually, these symptoms lead to premature death of affected patients in their teens or twenties (see Table 1 ). There are no existing treatments able to eliminate the progressively worsening effects of NCLs due to CNS degeneration, but only palliative treatments to reduce some symptoms, such as anticonvulsants to reduce seizures.The clinical expression of NCLs varies widely among the different existing forms, but the clinical hallmark is a combination of dementia, visual loss, seizures, and epilepsy. Depending on the form, manifestations may begin between the neonatal period and the young adult age, leading to the original classification of NCLs by age at onset into congenital, infantile, late infantile, juvenile and adult NCL subgroups. At least 14 genetically distinct NCLs, designated CLN1 to CLN14, are presently known and are briefly outlined below.

The majority of NCLs are inherited in an autosomal recessive manner, however, autosomal dominant inheritance has been reported in one adult-onset form designated as a CLN4 disease. CLN7 disease follows an autosomal recessive inheritance. The gene responsible for CLN7 is the major facilitator superfamily domain containing eight (CLN7/MFSD8) gene, that encodes the lysosomal transmembrane protein MFSD8 (518 aa, ~58 kDa, with 12 predicted transmembrane domains). CLN7 is caused by biallelic mutations in the CLN7/MFSD8 gene, causing a complete loss of function of the encoded MFSD8 protein. MFSD8 has recently been characterized as an endolysosomal chloride channel, but its precise function is still unknown.

Genes involved in human NCLs encode soluble lysosomal enzymes (CLN1/PPT1 , CLN2/TPP1 , CLN10/CTSD, CLN13/CTSF), a soluble lysosomal protein (CLN5), a protein in the secretory pathway (CLN11/GRN), two cytoplasmic proteins that also peripherally associate with membranes (CLN4/DNAJC5, CLN14/KCTD7), and transmembrane proteins located in the endoplasmic reticulum (CLN6), endoplasmic reticulum/cis Golgi (CLN8) and in lysosomes (CLN3, CLN7, CLN12). Despite the identification of causative genes of BD, for many of them, their functions are not fully understood, making it difficult to identify intracellular pathways as therapeutic targets.

CLN1 disease, infantile onset

The CLN1 gene, located on chromosome 1 , directs the production of an enzyme called palmitoyl- protein thioesterase 1 (PPT1 ). A deficiency in the PPT1 protein or its poor operation results in the abnormal buildup of lipids and proteins. In the classic infantile form, symptoms are seen before age 1 and progress rapidly. Developmental skills such as standing, walking, and talking are not achieved or are gradually lost. Children often develop seizures by age 2 and eventually become blind. By age 3, children may become completely dependent on their caregivers, and some may need a feeding tube. Most affected children die in early to mid-childhood.

CLN1 disease, juvenile onset

Some children with CLN1 abnormalities develop the disease after infancy — around age 5 or 6 — and have slower disease progression. Affected children may live into their teenage years. Others may not develop symptoms until adolescence and may live into adulthood.

CLN2 disease, late-infantile onset

The CLN2 gene, located on chromosome 11 , produces an enzyme called tripeptidyl peptidase 1 (TPP1 ) that breaks down proteins. The enzyme is insufficiently active in CLN2 disease. Developmental delay begins around the end of age 2. Children develop seizures and begin to gradually lose the ability to walk and speak. Brief, involuntary jerks in a muscle or muscle group (called myoclonic jerks) typically begin around age 4-5. By age 6 most children are completely dependent on their caregivers, and many will require a feeding tube. Most children with CLN2 disease die between the ages of 6-12 years.

CLN2 disease, later-onset

Some children with CLN2 abnormalities develop the disease later in childhood — around age 6 or 7 — and have slower disease progression. In later-onset CLN2 disease, loss of coordination (ataxia) may be the initial symptom. Affected children may live into their teenage years.

CLN3 disease, juvenile onset (ages 4-7)

The disease is caused by a mutation in the CLN3 gene, located on chromosome 16. The gene directs the production of a protein called battenin, which is found in the membranes of the cell. Most children suffering from CLN3 disease have a missing part in the gene, which in turn results in inability for the protein to be produced. Rapidly progressive vision loss begins between ages 4 and 7. Children develop learning and behaviour problems, and slow cognitive decline (dementia) and then start having seizures around age 10. In the teenage years, children affected by CLN3 disease develop slow movement, stiffness, and loss of balance (also referred as “parkinsonism”). They also develop difficulty with speech and language. As they age, children and teenagers become increasingly dependent on their caregivers. Most children with the disease die between the ages of 15 and 30.

CLN4 disease, adult onset

Also known as Kufs disease type B, this is a very rare form which typically begins in early adulthood (normally around age 30) and causes problems with movement and early dementia. The symptoms progress slowly, and CLN4 disease does not cause blindness. It is related to mutations in the DNAJC5 gene on chromosome 20. The age of death varies among affected individuals.

CLN5 disease, variant late-infantile onset

This disease is caused by problems with a lysosomal protein called CLN5, whose function is unknown. The CLN5 gene is located on chromosome 13. Children progress normally for the first few years of life before they start losing skills and develop behaviour problems. Seizures and myoclonic jerks begin usually between ages 6 and 13. Vision deteriorates and is eventually lost. Children have learning disabilities and problems with concentration and memory. Some may need a feeding tube. Most children with CLN5 live into their late childhood or teenage years.

CLN6, variant late-infantile onset

The gene CLN6, located on chromosome 15, directs the production of the protein CLN6, also called linclin. The protein is found in the membranes of the cell (most predominantly in the endoplasmic reticulum). Its function has not been identified. Symptoms vary among children, but typically start after the first few years of life and include developmental delay, changes in behaviour, and seizures. Children eventually lose skills for walking, playing, and speech. They also develop myoclonic jerks, problems sleeping, and vision loss. Most children with CNL6 die during late childhood or in their early teenage years.

CLN6, adult onset

Also known as Kufs disease Type A, this form of CLN6 disease shows signs in early adulthood that include epilepsy, inability to control muscles in the arms and legs (resulting in a lack of balance or coordination, or problems with walking), and slow but progressive cognitive decline.

CLN7, variant late-infantile onset

This disease is caused by mutations in the CLN7 gene located on chromosome 4, which produces the protein MFSD8 — a member of a protein family called the “major facilitator superfamily”. This superfamily is involved with transporting substances across the cell membranes. As with all the other forms of Batten Disease, the defect in the gene results in lack of production of the protein. Developmental delays begin after a few years of what seems to be a normally-developing child. Children usually develop epilepsy between the ages of 3 and 7, along with problems sleeping and myoclonic jerks. Children begin to lose the ability to walk, play, and speak as the disease progresses, with a rapid advancement of symptoms between the ages of 9 and 1 1 . Most children with the disorder live until their late childhood or teenage years.

CLN8 disease with Epilepsy with Progressive Mental Retardation (EPMR) Abnormalities in the CLN8 gene cause epilepsy with progressive decline in mental function. The gene, located on chromosome 8, encodes a protein also called CLN8, which is found in the membranes of the cell — most predominantly in the endoplasmic reticulum. The protein’s function has not been identified. Onset of symptoms begins between ages 5 and 10 and include seizures, cognitive decline, and behavioural changes. Seizures typically become very intermittent after adolescence. Loss of speech occurs in some individuals. Affected individuals can live into adulthood. A very rare form of the disorder is sometimes called Northern Epilepsy syndrome, because it occurs in certain families in an area of Finland.

CLN8 disease, late-variant onset

Affected children begin showing symptoms between ages 2 and 7, which include loss of vision, cognitive problems, unsteadiness, myoclonic jerks, and behavioural changes. Children develop treatment-resistant epilepsy and a marked loss of cognitive skills by age 10. Many children lose the ability to walk or stand unassisted. Life expectancy is uncertain; some children have lived into their second decade of life.

CLN10 disease

This very rare disease is caused by a mutation in the CTSD gene, located on chromosome 11 , which produces a protein known as cathepsin D. Cathepsin D is an enzyme that breaks apart other proteins in the lysosome. The disease typically is seen soon after birth, although it can occur later in childhood or adulthood. Some children have microcephaly — an abnormally small head size with reduced brain size.

CLN11/PGRN disease, adult-onset

The CLN11 gene encodes progranulin (PGRN). Mutations in the PGRN gene were first reported in two siblings suffering from an adult-onset NCL. Although PGRN has been shown to be transported to the lysosome via sortilin, its function(s) in this organelle remains unclear. It has been reported that PGRN-deficiency causing lysosomal dysfunction can be explained based on lipidomic and transcriptomic considerations. In a recent comprehensive review, Paushter and colleagues have provided new insights into the lysosomal function of PGRN and its link to multiple neurodegenerative diseases.

CLN12/ATP13A2

The CLN12 disease is caused by loss of function mutations in the predominantly neuronal P-type ATPase (ATP13A2) gene. The patients afflicted with this syndrome manifest characteristics of not only typical NCL, but also show extrapyramidal involvement. Postmortem pathological examination of the brain tissues from a Kufor-Rakeb syndrome patient with homozygous missense mutations in the ATP13A2 gene showed extensive deposition of lipofuscin in the retina, cerebral cortex, basal ganglia and cerebellum. CLN12/ATP13A2 gene product is targeted to the acidic compartments of the cell including the late endosome and lysosome. The loss of transporter function of the ATP13A2 may explain the dysregulated neurotransmission and eventual dementia characteristic of CLN12 disease.

CLN13/CTSF

The CLN13 gene encodes cathepsin F (CTSF) and mutations in this gene were originally reported in mice, which develop neurological disease with accumulation of autofluorescent material in neurons of the cerebral cortex, hypothalamus, cerebellar Purkinje cells and other regions of the brain. Recently, 3 families with adult-onset NCL causing dementia and motor disturbances without epilepsy have been described. These patients carried rare mutations in the CTSF gene. CTSF is expressed at a high level in cerebrocortical, hippocampal and cerebellar neurons. Although CTSF is recognized as a lysosomal cysteine proteinase, its in vivo function remains obscure. Pathologic analysis of the CTSF-deficient neurons showed accumulation of eosinophilic granules that had characteristics of lysosomal lipofuscin as well as elevated levels of autofluorescent lipofuscin, which are characteristic of all NCLs. These findings may indicate that CTSF either only mildly affects the phenotype or is mildly compensated by other gene product(s).

CLN14

Progressive myoclonic epilepsy (PME) is a clinically defined epileptic syndrome that manifests as myoclonic seizures and progressive neurological dysfunction. Mutations in the potassium channel tetramerization domain-containing protein 7 (KCTD7) have been extensively linked to progressive myoclonic epilepsy. Moreover, homozygous mutations in the KCTD7 gene have been reported to cause a subtype of NCL discovered in two siblings in a Mexican family who presented with infantileonset, progressive myoclonic epilepsy, cognitive impairment, loss of vision, motor regression, and premature death. Pathological analysis showed prominent NCL-type storage material. Further analyses showed that these patients carried a missense mutation in the KCTD7 gene and pathological analysis showed autofluorescent storage material characteristic of the NCLs. Currently, KCTD7 gene represents CLN14.

The most common form of NCL, or Batten Disease, is the juvenile form (JNCL) which - as mentioned - resulting from mutations in the CLN3 gene. Up to 90% of patients with Batten Disease carry the 1.02 kb deletion, which, in homozygous form, always causes a severe phenotype, including blindness, epilepsy, dementia, and early death at approximately 20-30 years of age. The mechanisms possibly involved in the pathological features of JNCL are defects in the lysosomal and autophagic pathways (S.E. Mole et al. 2014) which includes defects in traffic of proteins, cholesterol, sphingolipids, neurotransmitters, accumulation of damaged organelles or its components as the subunit c of the mitochondrial ATP-synthase (SCMAS) ( S.E. Mole et al. 2014). Mutations in CLN7/MFS domain-containing 8(CLN7/MFSD8) are responsible for late-infantile onset NCL (LINCL or CLN7 disease), with disease onset at 1.5-5 years of age. The CLN7 protein is localized primarily in lysosomes. In addition to the features shared with the CLN3 form (SCMAS and the newly discovered accumulation of the glycosphingolipid Gb3), mouse embryonic fibroblasts derived from CLN7 deficient mice show a loss of lysosomal proteins and deficits in mTOR reactivation. Mutant mice show lysosomal dysfunction and a potential impairment of autophagy.

The only treatment approved by the U.S. Food and Drug Administration for Batten Disease is Brineura (cerliponase alfa), an enzyme replacement therapy designed to slow down the loss of walking ability in children affected by the CLN2 form. Brineura replaces TPP1 , the enzyme that is mutated in the CLN2 form. However, there are no treatments available for the CLN3 and CLN7 forms.

Accordingly, there is a strong need to provide an effective drug therapy for the treatment of the group of diseases identified as neuronal ceroid lipofuscinoses (NCLs), commonly known as Batten Disease (BD), in order to address these terrible infantile diseases for which so far there is no cure.

More in particular, there is a need to provide small molecule-based therapeutics for Batten Disease to counteract the difficulties in replacing the affected proteins or genes and reaching tissues such as the CNS.

These and other needs have been met by the present inventors, who observed that the glycosphingolipid globotriaosylceramide (herein Gb3) significantly accumulates within the lysosomes of cellular and murine models of Batten Disease.

The inventors screened a library of 1280 FDA drugs and identified 9 compound hits. These include two compounds belonging to the stilbenoid class of drugs that are selective estrogen receptor modulators (tamoxifen and toremifene), one alkaloid (apomorphine), three phenylpiperazines (itraconazole, ketoconazole, aripiprazole), a derivative of cholesterol (pregnenolone), a diphenylmethane (benztropine), and an acetylcholinesterase inhibitor (donepezil dihydrochloride).

The inventors surprisingly found that tamoxifen and other SERMs are capable of significantly reducing the intracellular accumulation of Gb3 in CLN3 and CLN7 cellular models. The inventors also studied the mechanism by which SERMs reduce intracellular accumulation of Gb3 and surprisingly found that it is independent of the estrogen receptors but involves the activation of the transcription factor EB (TFEB), a master gene of lysosomal function and autophagy. The inventors additionally found that in vivo administration of tamoxifen to the CLN7Aex2 mouse model significantly reduces the accumulation of Gb3 and SCMAS, decreases microglia activation, hindlimb clasping, and motor discoordination. Thus, tamoxifen and other SERMs are promising drug candidates for the treatment of Batten Disease, in particular for the treatment of patients affected by the CLN3 and CLN7 forms of Batten Disease. Accordingly, a first aspect of the present invention is a selective estrogen receptor modulator (SERM) for use in the therapeutic treatment of a neuronal ceroid lipofuscinosis.

In a preferred embodiment, the SERM is effective in reducing neuroinflammation, discoordination and/or visual loss in a patient affected by a neuronal ceroid lipofuscinosis.

In another preferred embodiment, the SERM is effective in reducing accumulation of sphingolipids, in particular Gb3.

In yet another preferred embodiment, the neuronal ceroid lipofuscinosis is selected from the group consisting of CLN1 , CLN2, CLN3, CLN4, CLN5, CLN6, CLN7, CLN8, CLN9, CLN10, CLN11 , CLN12, CLN13 and CLN14. More preferably, the neuronal ceroid lipofuscinosis is selected from CLN3 and CLN7.

In a further preferred embodiment, the selective estrogen receptor modulator (SERM) contains a tertiary amine in its chemical structure.

Preferred SERMs for use according to the invention are selected from the group consisting of tamoxifen, (E) tamoxifen, toremifen, raloxifen, clomifene, ospemifene, bazedoxifene, nafoxidine, lasofoxifene, zuclomifene, afimoxifen, N-Desmethyl tamoxifene, droloxifene, tamoxifen aziridine, Idoxifene, 2-Methyl-4-hydroxytamoxifen, endoxifen, phenyltoloxamine, tamoxifen N-oxide, tamoxifen epoxide, afimoxifene, diethylaminoethoxyhexestrol, etoloxamine, tesmilifene, tamoxifen- d5, myoparkiltamoxifen and any pharmaceutically acceptable salt, ester, ether, isomer, mixture of isomers, complex, derivative or deuterated form thereof.

Particularly preferred SERMs for use according to the invention are selected from the group consisting of tamoxifen, (E) tamoxifen, toremifen, clomifene, ospemifene, zuclomifene, afimoxifen, N-desmethyl tamoxifene, droloxifene, tamoxifen aziridine, Idoxifene, 2-methyl-4-hydroxytamoxifen, endoxifen, tamoxifen N-oxide, tamoxifen epoxide, tamoxifen-d5, and any pharmaceutically acceptable salt, ester, ether, isomer, mixture of isomers, complex, derivative or deuterated form thereof.

SERMs for use according to the invention are administered to a patient suffering from neuronal ceroid lipofuscinosis in the form of a pharmaceutical composition, which comprises, in addition to the pharmaceutical active ingredient (SERM), a pharmaceutically acceptable excipient, vehicle and/or diluent. Of course, the selection of pharmaceutically acceptable excipients, vehicles and/or diluents, as well as the dosage of the active ingredient, depend on a variety of factors, including the type and severity of the disease to be treated, the age of the patient, as well as the intended dosage regimen and route of administration.

Generally, pharmaceutical dosage forms suitable for topical, oral, parenteral, intravitreal, transretinal or ophthalmic administration are preferred. Pharmaceutical dosage forms suitable for oral administration include for example tablets or capsules. Ophthalmic formulations may take the form of a solution, a suspension, an ointment or an emulsion. One of the most popular forms of opthalmic formulation is the topical instillation, i.e. a solution that enables the active pharmaceutical ingredient (API) to be administered directly onto the surface of the eye. Intravitreal and transretinal administrations are usually performed by injection.

The formulation of the pharmaceutical composition of the invention in a suitable dosage form is fully within the abilities of the person skilled in the art.

As mentioned, many patients affected by Batten Disease take anticonvulsants, which are a group of medicines commonly used to treat seizures in epilepsy. Different classes of anticonvulsants work through different mechanisms. Depending on the type of seizure that the Batten Disease patient has, different anticonvulsants can be used. Many Batten Disease patients, over the course of their disease, will be given more than one kind of anticonvulsant.

The main classes of anticonvulsants known in the art are ion channel modulators and GABA potentiating agents, although there are other anticonvulsants that act through multiple mechanisms of action, and others that use mechanisms that are yet not fully understood.

By way of non-limiting examples, anticonvulsants commonly used in juvenile NCL are valproic acid and levetiracetam. Additionally, an open-label study in juvenile NCL patients, published in 1999, showed that the anticonvulsant lamotrigine was well-tolerated and reduced seizure severity and frequency. Another study, from 2000, found that seizures could be controlled with valproic acid or lamotrigine, either as monotherapies or given in combination with clonazepam in juvenile NCL patients. Other medications prescribed to Batten Disease patients include zonisamide, carbamazepine, topiramate, phenytoin, and oxcarbazepine.

Accordingly, the present invention also includes a pharmaceutical composition as described above, which additionally comprises one or more anticonvulsants, preferably selected from the group consisting of valproic acid, levetiracetam, lamotrigine, clonazepam, zonisamide, carbamazepine, topiramate, phenytoin and oxcarbazepine.

Alternatively, the combination therapy with SERM and anticonvulsant(s) may be administered to the Batten Disease patient through a combined preparation. Accordingly, the scope of the invention also encompasses a kit-of-parts or combined preparation, comprising a SERM as defined above and one or more anticonvulsants as defined above, for simultaneous, separate or sequential use in the therapeutic treatment of a neuronal ceroid lipofuscinosis as defined above.

Alternatively, the combination therapy may be performed by the simultaneous, separate or sequential administration of a SERM as defined above and a gene therapy or a gene editing procedure. Such genetic techniques are aimed at replacing or repairing the mutated gene in the Batten Disease patient. Non-limiting examples are AAV or Lentivurus- based gene therapies aimed at replacing the mutated gene (eg. AAV-TPP1 , AAV-CLN3, AAV-CLN6), or CRISPR/Cas or Zn finger -based editing strategies either intended to replace the mutated gene or to correct the specific mutation allowing expression of the gene product, including base editing approaches replacing the mutated nucleotide(s).

A further alternative is to perform the combination therapy by the simultaneous, separate or sequential administration of a SERM as defined above and an enzyme replacement therapy wherein the wild-type version of the protein encoded by the mutated gene is administered to the Batten Disease patient. A non limiting example of an enzyme replacement therapy suitable for use in a combination therapy is the TPP1 enzyme replacement therapy for CLN2, which, as such, is already available on the market.

Another aspect of the present invention is an in vitro method of assessing the effectiveness of a therapeutic treatment administered to a subject suffering from a neuronal ceroid lipofuscinosis, as well as an in vitro method of diagnosing a neuronal ceroid lipofuscinosis in a subject, which are both based on the inventors’ finding that an increased level of globotriaosylceramide (Gb3) is a marker for Batten Disease.

Within this context, it is preferred that the neuronal ceroid lipofuscinosis is not CLN6. In a further preferred embodiment, the neuronal ceroid lipofuscinosis is selected from the group consisting of CLN1 , CLN2, CLN3, CLN4, CLN5, CLN7, CLN8, CLN9, CLN10, CLN1 1 , CLN12, CLN13 and CLN14.

The in vitro methods according to the invention are as defined in the appended claims, which form an integral part of the description.

Experimental evidence in support of the present invention is provided in the experimental section below, which is provided for illustrative purposes only and is not limiting the scope of the invention as defined in the appended claims.

The experimental section refers to the appended drawings, in which:

Figure 1 shows images and graphs relating to accumulation of Gb3 in Batten Disease cellular models. (A) Representative Opera images of WT and CLN3 KO ARRE 19 cell lines stained using fluorescent-conjugated cholera toxin (to detect GM1 ), LipidTox (to detect neutral lipids), fluorescent- conjugated Shiga toxin and is quantification (to detect Gb3), Filipin III (to detect cholesterol). Data are presented as mean ± SD ***: P < 0.0001 , as determined by Student’s t-test (n=3 biological replicas in duplicate). Scale bars: 50 pm. (B-C) Representative confocal images and their quantification of Gb3 accumulation within the lysosome, detected by fluorescent-conjugated Shiga toxin, from WT, CLN3-KO and CLN7-KO ARPE-19 cells treated with PDMP or silenced for Gb3 synthase (siGb3S) (*** vs WT, ⁰⁰⁰ vs DMSO). Data are presented as mean ± SD, °°°/***- p < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm. (D) Lipidomic analysis of ARPE-19CLN3KO and ARPE-19CLN7KO. Data are presented as mean ± SD, *: P < 0.01 , **: P < 0.001 , ***: P < 0.0001 , as determined by ANOVA (n=4 technical replicas).

Figure 2 shows images and graphs relating to neural accumulation of Gb3 in brain areas of the CLN7Aex2 mouse. (A-B) Representative confocal images of Gb3 accumulation, revealed by STX staining, in brain sections from CLN7Aex2 mice at 3 and 7.5 months of age compared with CLN7 WT mice. Scale bars: 80 pm. (C-D) Gb3, NeuN, and GFAP distribution in brain areas of CLN7Aex2 mice at 7.5 months of age. Scale bars: 40 pm, 20 pm. (E) Lipidomic analysis of neurons isolated from CLN7 WT and CLN7Aex2 mouse forebrain. Data are presented as mean ± SD, **: P < 0.001 , ***: P < 0.0001, as determined by ANOVA (n= 3 biological replicas).

Figure 3 shows images and graphs pertaining to the role of Gb3 in Batten Disease and the identification of correctors of Gb3 accumulation. (A) Representative confocal images and quantification of SCMAS staining within the lysosome of ARPE-19 WT and CLN3 KO upon depletion of Gb3S (siGb3S), LacCer (siLCS), and GM3s (siGM3S). (*** vs WT, ⁰⁰⁰ vs CLN3 KO siSCR). Data are presented as mean ± SD, ^oo7“*: p < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm. (B) Identification of FDA-compounds reducing Gb3 accumulation: Plot showing the ability of compound hits to reduce Gb3 within the lysosomal compartment compared with DMSO treated mutant cells. Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). (C) Table showing the EC50s (Half maximal effective concentration) from the dose-response curves of compound hits. EC50s were calculated using the Prism software. (D) Representative confocal images and quantification of STX staining within the lysosome of ARPE-19 WT and CLN7 KO in DMSO or treated 48h with Tamoxifen. (*** vs WT, ⁰⁰⁰ vs CLN3 KO DMSO). Data are presented as mean ± SD, °°°/***- p < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm. (E) Representative confocal images and quantification of STX staining within the lysosome of NPCs WT and derived from CLN7^Pa474 patient IPSCs in DMSO or treated with Tamoxifen for 48h. (*** vs WT, ⁰⁰⁰ vs CLN7^Pa474 DMSO). Data are presented as mean ± SD, °°7***: p < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm.

Figure 4 depicts images and graphs showing that tamoxifen-mediated clearance of Gb3 is ER- independent but TFEB-dependent. (A) Representative confocal images and Quantification of STX within the lysosome in U2-OS and HeLa cells after acute silencing of CLN3 (siCLN3) in DMSO or treated 48h with Tamoxifen. (*** vs siSCR DMSO, ⁰⁰⁰ vs siCLN3 DMSO). Data are presented as mean ± SD, °°7***: p < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm. (B) Representative confocal image and quantification of Gb3 in ARPE-19 CLN3 KO cells silenced with siRNA against scramble sequence and TFEB (siTFEB) for 72h and treated for the last 48h with DMSO or Tamoxifen. Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm. (C) Representative confocal image and quantification of TFEB in U2-OS cells silenced with siRNA against scramble sequence and CLN3 (siCLN3) for 72h and treated for the last 48h with DMSO or Tamoxifen (5 pM and 10 pM). Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm.

Figure 5 depicts images and graphs showing that tamoxifen dephosphorylates TFEB. (A) Immunoblot analysis and quantification of pTFEB S21 1 in HeLa cells stably expressing TFEB-GFP. Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3 biological replicas). GAPDH immunoblotting was performed as a loading control. (B) Immunoblot analysis of TFEB shift in ARPE-19 CLN3 KO cells, p-actin immunoblotting were performed as a loading control. (C) Representative confocal image and quantification of TFEB localization in HeLa TFEB-GFP transfected with RagC for 48h and treated for the last 3h with DMSO, Torinl , Tamoxifen or Ospemifene. Ratios of nuclear to cytosolic TFEB localization in RagC non-expressing (RagC-) and RagC-expressing cells (RagC+) are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm. D) Representative confocal images and quantification of STX in ARPE-19 CLN3 KO cells transfected with empty vector or HA-RagC for 48h and treated for 48h with DMSO or Tamoxifen. STX average spot area presented as mean ± SD, ***: P < 0.0001, as determined by ANOVA (n=6). Scale bars: 20 pm.

Figure 6 depicts images and graphs showing that tamoxifen reduces Gb3 accumulation in the brain of CLN7Aex2 mice. Representative confocal images of STX in the Cortex, Hippocampus, and Cerebellum brain section derived from 7.5 months-old mouse WT or CLN7Aex2 injected with vehicle or Tamoxifen (Tamox). Quantification of confocal images, the plot shows the quantification of the STX average spot area normalized for the number of Hoechst positive cells. (*** vs WT, °°°/°° vs CLN7Aex2 Vehicle). Data are presented as mean ± SD, **/oo: P < 0.001 , ***/ooo: P < 0.0001 , as determined by ANOVA (N>3 biological replicas). Scale bars: 60 pm.

Figure 7 depicts images and graphs showing that tamoxifen reduces SCMAS accumulation in the brain of CLN7Aex2 mice. Representative confocal images of SCMAS in the Cortex, Hippocampus, and Cerebellum brain section derived from mouse WT or CLN7Aex2 injected with vehicle or Tamoxifen. Quantification of confocal images, the plot shows the quantification of the SCMAS average spot area normalized for the number of Hoechst positive cells. (*** vs WT, ⁰⁰⁰ vs CLN7Aex2 Vehicle). Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (N>3 biological). Scale bars: 60 pm.

Figure 8 depicts images and graphs showing that tamoxifen ameliorates CLN7Aex2 phenotype. (A- B) Representative confocal images and quantification of IBA-1 in the Cortex, Hippocampus, and Cerebellum brain section derived from WT or CLN7Aex2 mice injected with the vehicle or Tamoxifen. (***/**/* vs WT, °°TT vs CLN7Aex2 Vehicle)Data are presented as mean ± SD, */o: P < 0.01, **/oo: P < 0.001 , ***: P < 0.0001 , as determined by ANOVA (N>3 biological replicas). Scale bars: 50 pm. (C) Plots show the quantification latency to fall from the rotarod. Data are presented as mean ± SD, *: P < 0.01 , as determined by ANOVA (N>3 biological replicas). (D) Representative Images of hindlimb clasping test in mouse WT or CLN7Aex2 injected with vehicle or Tamoxifen.

Figure 9 depicts images and graphs concerning accumulation of lipids and cell-based STX assay validation. (A) Quantification of WT and CLN3 KO ARPE-19 cell lines stained using Filipin III (to detect cholesterol) fluorescent-conjugated cholera toxin (to detect GM1 ) and LipidTox (to detect neutral lipids). Data are presented as mean ± SD *: P < 0.01, **: P < 0.001 as determined by Student’s t-test (n=3 biological replicas in duplicate). (B) Representative confocal images and quantification of Gb3 accumulation within the lysosome detected by Shiga toxin in WT and GLA KO HeLa cells treated with DMSO (controls) or PDMP or silenced for the Gb3 synthase (siGb3S). (*** vs WT, ⁰⁰⁰ vs DMSO) Data are presented as mean ± SD, ⁰⁰⁰ /***: P < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate) Scale bars: 20 pm. (C) Representative Opera images and their quantification of STX staining in WT ARPE-19 and HeLa cells silenced for 72h with siRNAs against a scrambled sequence, CLN3, CLN6, and CLN7. Data are presented as mean ± SD, ***: P < 0.0001, as determined by ANOVA (n=3 biological replicas in duplicate) Scale bars: 50 pm. (D) Representative confocal images and quantification of Gb3 accumulation, revealed by STX staining, in brain sections from CLN3+/Aex7/8 and CLN3Aex7/8 mice at 7.5 months of age. Data are presented as mean ± SD, **: P < 0.001 , ***: P < 0.0001 , as determined by ANOVA (n=4 biological replicas) Scale bars: 60 pm.

Figure 10 shows graphs concerning the dose-response analysis of compound hits derived from the FDA-screening to identify correctors of Gb3 accumulation. Data are presented as mean ± SD (n=4 technical replicas).

Figure 11 depicts images and graphs showing that tamoxifen induces clearance in Batten Disease models. (A) Representative confocal images and quantification of STX staining within the lysosome of human WT fibroblasts and CLN3 mutants in DMSO or treated with Tamoxifen. (*** vs WT, ⁰⁰⁰ vs DMSO) Data are presented as mean ± SD, ^ooo/***-_ p < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate) Scale bars: 20 pm. (B) Representative confocal images and quantification of STX within the lysosome in ARPE-19 cells silenced with siRNA scramble or against CLN7, in DMSO, or treated with Tamoxifen. (*** vs siSCR, ⁰⁰⁰ vs DMSO) Data are presented as mean ± SD, ^ooo/***-_ p < 0.0001 , as determined by ANOVA (n= 3 biological replicas in duplicate) Scale bars: 20 pm. (C) Representative confocal images and quantification of SCMAS accumulation within the lysosomes in wild type and ARPE-19 CLN3 KO cells in DMSO, or treated with Tamoxifen. (*** vs WT, ⁰⁰⁰ vs DMSO) Data are presented as mean ± SD, °°7***: p < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate) Scale bars: 20 pm. (D) Representative confocal images and quantification of STX within the lysosome. HeLa cells after acute silencing of CLN3 (siCLN3) in DMSO or treated 48h with Tamoxifen. (*** vs siSCR, ⁰⁰⁰ vs DMSO). Data are presented as mean ± SD, °°7***: p < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate) Scale bars: 20 pm.

Figure 12 depicts images and graphs showing that tamoxifen induces TFEB activation. (A) Representative confocal image and quantification of TFEB localization in ARPE-19 CLN3 KO treated for 3h with DMSO, Torinl or Tamoxifen. Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm. (B) Representative confocal image and quantification of STX in ARPE-19 CLN3 KO infected with an inducible vector expressing a nuclear-localized mutant form of TFEB. Data are presented as mean ± SD, ***: P < 0.0001, as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm.

Figure 13 depicts images and graphs relating to the effects of ER modulators on Gb3 clearance and TFEB activation. (A) Representative Opera and confocal images and quantification of TFEB subcellular localization, and STX accumulation within the lysosome of ARPE-19 CLN3 KO cells treated with DMSO, Tamoxifene or Toremifene for 48 h. Data are presented as mean ± SD, ***: P < 0.0001, as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 50 pm top and 20 pm down. (B) Representative Opera images and quantification of STX and TFEB localization in ARPE-19 CLN3 KO in DMSO, Tamoxifen, or Ospemifene after 48h. Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 50 pm top and 20 pm down. (C) Representative Opera images and quantification of Lysotracker-Red staining in ARPE-19 CLN3 KO cells cultivated for 3 h and 48 h in the absence (DMSO) or presence of Tamoxifen (Tamox) or Ospemifene. Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3 biological replicas in duplicate). Scale bars: 20 pm.

Figure 14 is a graph relative to Figures 1 and 9. Quantitative PCR showing mRNA levels of silenced gene Gb3S compared to a scramble sequence (siSCR). Data are presented as mean ± SD, ***: P < 0.0001, as determined by ANOVA (n=3)

Figure 15 is a graph relative to Figure 9. Quantitative PCR showing mRNA levels of silenced genes CLN3, CLN7 and CLN6 compared to a scramble sequence (siSCR). Data are presented as mean ± SD, ***: P < 0.0001 , as determined by (n=3)

Figure 16 is a graph relative to Figure 3. Quantitative PCR showing Gb3 synthase (Gb3S), LacCer synthase (LCS) and GM3 synthase (GM3S) mRNA levels in ARPE-19 CLN3 KO cells transfected with scrambled (SCR, black bars) or gene-specific siRNAs (grey bars). Data are presented as mean ± SD, ***: P < 0.0001 , as determined by (n=3)

Figure 17 depicts graphs relative to Figure 3. (A) Quality data determination of the cell-based high content screening performed (Z-score and SSMD score values) and (B) correlation between plate replicates of the screening. Figure 18 is an image relative to Figure 3. Representative confocal images of Nestin in NPCs WT and derived from a CLN7 patient IPSC. Scale bars: 20 pm.

Figure 19 depicts graphs relative to Figure 4 and Figure 11. A) Quantitative PCR showing mRNA levels of silenced gene CLN3 (siCLN3) compared to a scramble sequence (siSCR). Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3). B) Quantitative PCR showing mRNA levels of Estrogen receptor 1 and 2 in HeLa and U2OS cells compared to MCF-7. Data are presented as mean ± SD, ***: P < 0.0001 , as determined by ANOVA (n=3).

Figure 20 is a graph showing that tamoxifen reduces Gb3 accumulation in multiple models of NCL generated by siRNA-based knock-down of CLN genes. Shiga toxin staining was used to detect Gb3 accumulation in human ARPE-19 cells depleted of CLN genes by siRNAs.

Figure 21 depicts a graph and an image relative to Figure 4. Quantitative PCR and immunoblot of TFEB expression in WT and ARPE 19-CLN3 KO cells transfected with scrambled (siSCR) and TFEB siRNAs (siTFEB). p-actin immunoblotting was performed as loading control. Data are presented as mean ± SD, ***: P < 0.0001 , as determined by one-way ANOVA (n=3).

Figure 22 depicts a graph and images relative to Figure 5. Tamoxifen dephosphorylates TFEB. (A) Opera images and quantification of Phospho-S6 ribosomal protein (pS6) intensity levels in ARPE- 19 CLN3 KO cells incubated in the absence (DMSO) or presence of torinl or tamoxifene (Tamox). Data are presented as mean ± SD, ***: P < 0.0001 , as determined by one-way ANOVA (n=3). Scale bars: 40 pm. (B) Immunoblot analysis of the mTORCI substrates ULK1 , p70S6K (S6) and 4EPB in wild type and in ARPE-19 CLN3 KO cells.

As illustrated in more detail below, the inventors combined cell-based phenotypic screening and repurposing of FDA drugs for the identification of correctors of lysosomal storage in CLN3 and CLN7 cellular models of BD. CLN3 disease (MIM # 204200) represents the most common form of NCL worldwide, whereas CLN7 disease (MIM # 610951 ) is one of the most prevalent BD in southern and Mediterranean Europe. The inventors found a significant endogenous accumulation of the glycosphingolipid globotriaosylceramide (Gb3) (Welford et al, 2018) within the lysosomes of human ARPE-19 cells depleted of CLN3 or CLN7 genes by CRISPR genome editing. Gb3 was also found in human juvenile CLN3 patient fibroblasts, neuronal progenitor cells (NPCs) derived from CLN7 patient iPSCs, and neurons in brain tissues from both Cln3Aex7/8 and Cln7Aex2 mutant mice, suggesting that Gb3 accumulation is part of the pathological storage in these diseases.

By using fluorescent-conjugated bacterial toxins to label Gb3, the inventors developed a cell-based high content imaging (HCI) screening assay for the repurposing of FDA-approved compounds able to reduce the accumulation of Gb3 within the lysosomes of BD cells. They found that tamoxifen and other SERMs significantly reduce the intracellular accumulation of Gb3 in CLN3 and CLN7 cell models through a mechanism that is independent of estrogen receptors but involves activation of the transcription factor EB (TFEB), a master gene of lysosomal function and autophagy (Sardiello et al, 2009). TFEB activation by tamoxifen is triggered by lysosomotropic-mediated inhibition of mTORCI . Furthermore, in vivo administration of tamoxifen significantly rescued Cln7Aex2 mice from brain cortex Gb3 accumulation, reducing SCMAS storage, hindlimb clasping, and motor discoordination. These data indicate that Gb3 is a novel biomarker for CLN3 and CLN7 diseases, and tamoxifen and other SERMs may be suitable drugs for their treatment.

EXPERIMENTAL SECTION

Material and Methods

Cell culture and siRNA transfection

ARPE-19 (retinal pigment epithelium (RPE) cell line), U2-OS, and HeLa cells were purchased at ATCC and cultured in DMEM F12 and DMEM, supplemented with 10% fetal bovine serum, 200pM L - glutamine, 100pM sodium pyruvate, 5% CO2 at 37°C. Human ARPE-19 cells were chosen because they are diploid and non-transformed. ARPE-19 depleted of CLN3 was generated by Dr. J. Monfregola at TIGEM (Naples) and was cultured in DMEM F12 supplemented with 10% fetal bovine serum, 200pM L - glutamine, 100pM sodium pyruvate, 5% CO2 at 37°C.

Human control patient fibroblasts were provided by Professor Brunetti (TIGEM), CLN3 patient fibroblasts were purchased from Coriell Institute and cultured in DMEM supplemented with 15% fetal bovine serum, 200pM L - glutamine, 5% CO2 at 37°C. HeLa TFEB/TFE3 KO cells were generated from Dr. R.J. Youle from the National Institutes of Health, Bethesda. U2OS was purchased at ATCC and cultured in DMEM supplemented with 10% fetal bovine serum, 200pM L - glutamine, 5% CO2 at 37°C.

Cells were silenced with 25nM of siRNA against all each CLN genes, Gb3S, and TFEB for 72 hours using Lipofectamine RNAimax (Thermofisher) according to the protocol from the manufacturer. All control experiments to confirm silencing efficiency (by qPCR or immunoblot) are reported.

Generation ofARPE- 19 CRISPR/Cas9 CLN3 KO and CLN7 KO cell lines

Generation of ARPE-19 CRISPR/Cas9 CLN3 KO and CLN7 KO cell lines. ARPE-19 (ATCC CCRL- 2320) cells carrying a homozygous deletion of a C were generate by using the CRISPR/Cas9 system. The gRNA sequence with low off-target score have been selected using the http://crispor.tefor.net/crispor.py tool. An “ALL in One” vector expressing Cas9, the specific gRNA and GFP was obtained from SIGMA (CAS9GFPP). The CAS9GFPP was nucleofected in ARPE19 cells using the Amaxa Cat No VCA-1003 and transfected GFP-positive cells were FACS sorted into 96 well plates to obtain single-cell derived colonies carrying the INDEL mutations. Upon genomic DNA extraction and DNA Sanger sequencing, clones carrying the c. 1055 del A for CLN3 KO cells and c.103del C for CLN7 KO cells were selected and expanded. Drugs and cellular treatments

The following drugs were used to perform the assays: Tamoxifen (10 pM-SIGMA 3-48h), uM Toremifene (10 pM-SIGMA 48h), Ospemifene (10 pM-SIGMA 48h).

Screening and dose-response: Cells were plated on 384-well plates (2x10⁴ cells per well). After 24h, cells were treated with 10 pM compounds or 0.1% dimethyl sulfoxide (DMSO) in complete medium. The Prestwick Library consists of 1 ,280 FDA-approved drugs, all off-patent, dissolved in DMSO. The drugs from the 96-well source plate were diluted and compacted in 384-well plates to a concentration of 100pM in the DMEM medium (working plate). To study the effect of the drugs, 5pl of the drugs at 100pM in DMEM medium were added to plates containing 45pl of medium (10pM final drug concentration with 0.1 % DMSO). As a positive control of Gb3 reduction, the glucosylceramide synthase inhibitor PDMP was used.

For the dose-response confirmation test, compounds were serially diluted from 10mM stock into complete medium and added to plates starting at 30 pM to 0.1 pM. The final concentration of DMSO did not exceed 0.3% in the dose-response assays.

Cells were incubated together with drugs 48 h at 37 °C and 5% CO2.

Screening quality control analysis.

The robustness of the STX-assay was confirmed using two different quality control scores (Z-score and SSMD-score) (Figure 17-A). To exclude toxicity in the primary screening, compounds reducing the cell viability to 40% compared to DMSO-treated controls were discarded. To ensure reproducibility, the correlation between plate replicates was also analyzed (Figure 17-B). As a cutoff for hit selection, compounds capable of reducing the lysosomal accumulation of Gb3 greater than the mean of Gb3 in DMSO-treated CLN3 cells minus two standard deviations were selected.

Antibodies and western blotting

The following antibodies were used: p-Actin (Santa Cruz SC 47778, 1 :4000), LILK1 (cell signaling cat. 8054 1 :1000), Phospho-ULK1 (Ser757) (Cell signaling cat. 6888 1 :1000), p70 S6 Kinase (Cell signaling cat. 2708 1 :1000), Phospho-p70 S6 Kinase (Thr389) (Cell signaling cat. 9205 1 :1000), GAPDH (6C5) (Santa Cruz sc-32233, 1 :2000), 4EBP (cell signaling cat. 9644 1 :1000), p4EBP (Cell signaling cat. 9456 1 :1000), TFEB (Cell signaling cat. 4240S 1 :1000) and TFEB-pS21 1 (customgenerated in collaboration with Bethyl Laboratories 1 :1 ,000). For immunoblot, the total cell lysates were prepared by solubilization of cell pellets in 10mM T ris HCI pH 8.0 and 0.2% SDS supplemented with protease and phosphatase inhibitors (SIGMA). Protein concentration was determined by the Bradford method. After SDS-PAGE and immunoblotting the protein recognized by the specific antibody were visualized by chemiluminescence methods (Luminata Crescendo Western HRP substrate, Millipore) using peroxidase-conjugated anti-rabbit or anti-mouse secondary antibodies (Millipore). Membranes were developed using a Chemidoc UVP imaging system (Ultra-Violet Products Ltd) and densitometric quantification was performed in unsaturated images using Imaged (NIH).

Immunofluorescence

For immunofluorescence, the following antibodies were used: LAMP1 (Santa Cruz cat. sc-2001 1 , 1 :400), TFEB (cell signalling cat. 4240S 1 :200), Phospho-S6 Ribosomal Protein (Ser235/236) (cell signalling cat. 9865 1 :400), anti-ATP-synthase C (Abeam ab181243 1 :500), HA.1 1 clone 16B12 (Biolegend 901501 1 :500), Nestin (Thermo Fisher MA1 -1 10 1 :200). Cells were fixed in PFA 4% for 20 minutes and permeabilized with blocking buffer saponin whereas for TFEB immunostaining cells were permeabilized in 0.1% (w/v) Triton-X100, 1% (w/v) horse serum, and 1 % (w/v) BSA in PBS. Cells were incubated with the indicated primary antibodies for 2 hours and subsequently incubated with secondary antibodies for 45 minutes (AlexaFluor 488 A21202, AlexaFluor 488 A11008, AlexaFluor 568, A10037, AlexaFluor 568 A10042, all Thermo Fisher 1 :400). For confocal imaging, the samples were examined under a Zeiss LSM 800 confocal microscope. Optical sections were obtained under a x63 or x40 immersion objective at a definition of 1024 x 1024 pixels (average of eight or sixteen scans), adjusting the pinhole diameter to 1 Airy unit for each emission channel to have all the intensity values between 1 and 254 (linear range). For high content images, the OPERA high content imager from PerkinElmer was used.

For image analysis, Columbus 2.6.0.127073 (built at 03:56 on 05/02/19) released by PerkinElmer was used. This online platform is based on Harmony High-Content Imaging and Analysis Software which provides an easy quantification of complex cellular phenotypes.

Lipidomic analysis

For Lipidomics analysis, 200000 cell culture lysates or 10 pg of protein of homogenized imunoisolated neuron samples were spiked with 4.28 pL of internal standard lipid mixture containing 500 pmol of Chol-d6, 100 pmol of Chol-16:0-d7, 100 pmol of DAG 17:0-17:0, 50 pmol of TAG 17:0- 17:0-17:0, 100 pmol of SM 18:1 ;2-12:0, 30 pmol of Cer 18:1 ;2-12:0, 30 pmol of GalCer 18:1 ;2-12:0, 50 pmol of LacCer 18:1 ;2-12:0, 300 pmol of PC 17:0-17:0, 50 pmol of PE 17:0-17:0, 30 pmol of PI 16:0-16:0, 50 pmol of PS 17:0-17:0, 30 pmol of PG 17:0-17:0, 30 pmol of PA 17:0-17:0, 25 pmol of Gb3 18:1 ;2- 17:0, 25 pmol of GM3 18:1 ;2- 18:0-d5, 25 pmol of GM2 18:1 ;2- 18:0-d9, 25 pmol of GM1 18:1 ;2- 18:0-d5m 25 pmol of GD1 a 18:1 ;2- 17:0 and subjected to lipid extraction at 4 °C, as described elsewhere (Sampaio et al, 2011). Briefly, the sample was dissolved in 200 pL of 155 mM ammonium bicarbonate and then extracted with 1 mL of chloroform-methanol (10:1 ) for 2 h. The lower organic phase was collected, and the aqueous phase was re-extracted with 1 mL of chloroform-methanol (2:1 ) for 1 h. The lower organic phase was collected and evaporated in a SpeedVac vacuum concentrator. Lipid extracts were dissolved in 100 pL of infusion mixture consisting of 7.5 mM ammonium acetate dissolved in propanol:chloroform:methanol [4:1 :2 (vol/vol)]. Samples were analyzed by direct infusion in a QExactive mass spectrometer (Thermo Fisher Scientific) equipped witha TriVersa NanoMate ion source (Advion Biosciences). 5 pL of sample were infused with gas pressure and voltage set to 1 .25 psi and 0.95 kV, respectively.

HexCer was detected in the 10:1 extract, by negative ion mode FTMS as a deprotonated ion by scanning m/z= 520-1050 Da, at Rm/z=200=280 000 with lock mass activated at a common background (m/z=529.46262) for 30 seconds. Every scan is the average of 2 micro-scans, automatic gain control (AGC) was set to 1 E6 and maximum ion injection time (IT) was set to 200ms. Hex2Cer and Hex3Cer were detected in the 2:1 extract, by positive ion mode FTMS as protonated ions by scanning m/z= 800-1600 Da, at Rm/z=200=280 000 with lock mass activated at a common background (m/z=1 194.81790) for 30 seconds. Every scan is the average of 2 micro-scans, automatic gain control (AGC) was set to 1 E6 and maximum ion injection time (IT) was set to 50ms. GM3 was detected in the 2:1 extract, by polarity switch to negative ion mode FTMS as a deprotonated ion by scanning m/z= 1100-1650 Da, at Rm/z=200=280 000 with lock mass activated at a common background (m/z=1 175.77680) for 30 seconds. Every scan is the average of 2 microscans, automatic gain control (AGC) was set to 1 E6 and maximum ion injection time (IT) was set to 50ms. All data was acquired in centroid mode.

All data were analyzed with the lipid identification software, LipidXplorer (https://doi.org/10-1186/gb- 2011-12-1-r8). Tolerance for MS and identification was set to 2 ppm. Data post- processing and normalization to internal standards were done manually. For the sake of simplicity, only the pertinent data is displayed (HexCer, Hex2Cer, Hex3Cer and GM3) and normalized to the total lipid identified. Raw data are represented in Appendix Table S1 -S2-S3-S4.

Induced pluripotent stem cells (iPSC) and Neural Progenitor Cells (NPC) generation. iPSC were generated from a CLN7 patient (Pa474), then characterized and differentiated to NPC as previously described (FitzPatrick et al, 2018). Human iPSC-derived NPCs from an age-matched control patient and patient Pa474(c.1393C>T; p.R465W) harboring the indicated CLN7 homozygous mutations, were plated on Matrigel. Matrix in Nunc™ Lab-Tek™ 8-well Chamber Slides and cultured in Neural Expansion Medium (NEM) with DMEM/F12, NEAA, N-2 supplement, B-27 supplement, heparin, bFGF protein, penicillin/streptomycin.

Fluorescent assays

Cholera toxin: Cells were cultured on 96-well plates and incubated in a serum-free medium containing 1 pg/ml AlexaFluor488-labelled cholera toxin subunit B (C22841 Thermo Fisher Scientific) for 30 min at 33 °C. Subsequently, cells were washed three times with PBS and fixed in 4 % (w/v) paraformaldehyde for 10 min at room temperature. Nuclei were stained with Hoechst for 10 min. Filipin III: Cells were fixed with 4% paraformaldehyde 10 min. The paraformaldehyde was rinsed with PBS and quenched with 50 mM glycine in PBS. Cells were then incubated for 2 h at room temperature in PBS containing 50 pg/ml filipin III (from Streptomyces filipinensis SIGMA F4767). Nuclei were stained with DRQ5 1 :5000 (62254 Thermo Fisher Scientific) for 10 min.

Shiga Toxin: Cells were fixed with 4% paraformaldehyde 10 min and permeabilized in 0.1% (w/v) saponin, 0.5% (w/v) BSA and 50 mM NH4CI in PBS (blocking buffer saponin). STX were incubated alone or with LAMP1 antibody in blocking buffer saponin for 2h (1 :50000) and subsequently incubated with secondary antibodies for 45 minutes. Nuclei were stained with Hoechst for 10 min.

Lysotracker: Cells were cultured on 96-well plates and incubated in a serum-free medium containing 1 :10000 AlexaFluor568-labelled Lysotracker Red (L7528 Thermo Fisher Scientific) for 20 min at 33 °C. Subsequently, cells were washed three times with PBS and fixed in 4 % (w/v) paraformaldehyde for 10 min at room temperature. Nuclei were stained with Hoechst for 10 min.

RNA extraction and quantitative PCR

Total RNA was extracted from cells using the RNeasy Plus Mini Kit (Qiagen). Reverse transcription was performed using the QuantiTect Rev Transcription Kit (Qiagen). Real-time quantitative Reverse Transcription PCR (qRT-PCR) was performed using the LightCycler® System 2.0 (Roche Applied Science). HPRT was used for qRT -PCR as a reference gene. The parameters of real-time qRT -PCR amplification were according to Roche recommendations. Primer sequences are available upon request.

Ethical Use of Animals. Mice were bred at the Animal Experimentation Unit of the University of Salamanca. All protocols were performed according to the European Union Directive 86/609/EEC and Recommendation 2007/526/EC, regarding the protection of animals used for experimental and other scientific purposes, enforced in Spanish legislation under the law 6/2013. All protocols were approved by the Bioethics Committee of the University of Salamanca. The Cln3Aex7/8 knock-in mice were bred in a pathogen-free animal facility at the University Medical Center Hamburg-Eppendorf according to institutional guidelines.

Mice genotyping by a polymerase chain reaction.

For Cln7Aex2 genotyping, a PCR with the following primers was performed 5’- TGGTGCATTAATACAGTCCTAGAATCCAGG-3’ (DEQ ID NOU ), 5’- CTAGGGAGGTTCAGATAGTAGAACCC-3’ (SEQ ID NO:2), 5’-

TTCCACCTAGAGAATGGAGCGAGATAG-3’ (SEQ ID NO:3), resulting in a 290 bp band in the case of Cln7Aex2 mice, and 400 bp for wild type (18). In the case of Cln3Aex7/8 knock-in mice, a band with 250 bp is obtained for wild type, and a 500 bp in the case of Cln3Aex7/8 knock-in mice using the following primers: 5’- CAGCATCTCCTCAGGGCTA-3’ (SEQ ID NO:4), 5’- CCAACATAGAAAGTAGGGTGTGC-3’ (SEQ ID NO:5), 5’- GAGCTTTGTTCTGGTGCCTTC-3’ (SEQ ID NO:6), 5’- GCAGTCTCTGCCTCGTTTTCT-3’ (SEQ ID NO:7) (19).

Neuronal cell isolation from the brain cortex.

Adult mouse brain (from 6 months animals) tissue was dissociated with the Adult Brain Isolation Kit (Miltenyi Biotec). Neurons were separated in the dissociated cells, after removal of debris and red blood cells, neurons were separated with the Neuron Isolation Kit (Miltenyi), according to the manufacturer’s protocol. The identity of the isolated fraction was confirmed previously (Lopez-Fabuel et al, 2016) by western blot against the neuronal marker microtubule associated protein 2 (MAP2).

Tamoxifen administration.

Twice per week 40 mg of tamoxifen per gram of body weight (from a stock solution of tamoxifen in 20% (vol/vol) ethanol and 80% (vol/vol) of sunflower oil) were injected intraperitoneally to male mice of 2.5 months old until they reach 7.5 months old. Body weight was evaluated before injections, as well as the general aspect of the mice (eyes, fur, and behavior) to see the influence of tamoxifen in mice. Control animals received injections with the tamoxifen vehicle (20% (vol/vol) ethanol and 80% (vol/vol) of sunflower oil). A rotarod test was done before treatment beginning, and every month during it, to all the studied animals.

Locomotor assessment (Rotarod Test).

The rotarod test (Rotarod apparatus, model 47600, Ugo Basile) was used to analyze motor balance and coordination. Male mice were previously trained during three consecutive days, two days before the test. The rotarod conditions were gradually accelerated from 4 to 45 r.p.m., reaching the final speed at 270 s. The latency to fall was evaluated and averaged for each animal during the three days of the experiment.

Mouse perfusion and immunohistochemistry. Male mice were anaesthetized by intraperitoneal injection of a mixture of xylazine hydrochloride (Rompun; Bayer) and ketamine hydrochloride/chlorbutol (Imalgene; Merial) (1 :4) at 1 ml per kg body weight and then perfused intraaortically with 0.9% NaCI followed by 5 ml per g body weight of Somogyi (Paraformaldehyde, 4% (wt/vol) and picric acid, 0.2% (vol/vol), in 0.1 M PB, pH 7.4. After perfusion, (i) brains were dissected out sagittally in two parts and post-fixed with Somogyi for 2 h at room temperature; (ii) eyes were extracted, stabbed in the lens, post-fixed overnight in 4% (wt/vol) of paraformaldehyde, and then were subjected to 5 washes of 10 minutes in 0.1 M of PB solution, followed by cryoprotection in 15% and 30% of sucrose (wt/vol) in 0.1 M PB sequentially at 4^eC. Brain blocks were rinsed successively for 15 min, 30 min, 1 h and 2 h with 0.1 M PB solution and cryoprotected in 10%, 20% and 30% (wt/vol) sucrose in PBS sequentially, until they sank. After cryoprotection, 40 pm-thick sagittal sections were obtained with a freezing-sliding cryostat (Leica; CM1950 AgProtect). The sections were collected serially in a 12-well plate in 0.1 M PB, rinsed three times for 10 min in 0.1 M PBS and used for subsequent immunohistochemistry and autofluorescence evaluation. The sectioncontaining wells that were not used were kept in freezer mix (polyethylene glycol, 30% by volume and glycerol 30% by volume in 0.1 M PB) at - 20 °C. In the case of autofluorescence, sections were mounted with Fluoromount (Sigma-Aldrich) aqueous mounting medium and lamelles cover-objects (Thermo Fisher Scientific). For immunohistochemistry, sections were incubated sequentially in (i) 5 mg/ml sodium borohydride in PBS for 30 min (to remove aldehyde autofluorescence); (ii) three PBS washes of 10 min each; (iii) 1 :500 anti-IBA-1 (019-19741 , Wako), 1 :200 anti-ATP-synthase C (ab181243 abeam), 1 :500 anti-NeuN (MAB377 Millipore), 1 :500 anti-GFAP (G6171 , Sigma) and 1 :1000 STX in 0.02% Triton X-100 (Sigma-Aldrich) and 5% goat serum (Jackson Immuno-Research) in 0.1 M PB for 72 h at 4 °C; (iv) three PB washes of 10 min each; (v) fluorophore conjugated secondary antibodies, 1 :500 Cy2 goat anti-mouse or 1 :500 Cy3 goat anti-rabbit (Jackson ImmunoResearch) in PB for 2 h at room temperature; and (vi) 0.5 pg/ml Hoechst in PB for 10 min at room temperature. After being rinsed with PB, sections were mounted with Fluoromount.

Imaging and quantification.

For confocal imaging, the sections were examined under a Zeiss LSM 800 confocal microscope. Optical sections were obtained under a x63 or x40 immersion objective at a definition of 1024 x 1024 pixels (average of eight or sixteen scans), adjusting the pinhole diameter to 1 Airy unit for each emission channel to have all the intensity values between 1 and 254 (linear range). For image analysis Columbus 2.6.0.127073 (built at 03:56 on 05/02/19) released by PerkinElmer was used. This online platform is based on Harmony High-Content Imaging and Analysis Software.

Statistical Analysis.

Microsoft Excel, GraphPad Prism, and R software packages were used to analyze the data.

Sample numbers and other information (mean or SD, number of replicates and specific statistical tests) are indicated in the main text or Figure legends.

Data has been analysed with the support of the Bioinformatics Core in TIGEM.

The Shapiro-Wilk test was first used to check normality assumption. The p-value in the data obtained is not significant, so normality can be assumed. Then One-way or Two-way ANOVA have been applied for all charts with more than two groups. Student's t-test was used for statistical analysis when comparing only two groups.

Results

Pathological accumulation of Gb3 in CLN3 and CLN7 diseases

Batten Disease (BD) accumulates autofluorescent material, called ceroid lipofuscin, within lysosomes (Mole et al, 2005). The nature of this material is heterogeneous (Katz & Robison, 2002; Double et al, 2008), and some might derive from oxidation of either modified protein residues or lipids, including triglycerides, free fatty acids, cholesterol, and phospholipids (Double et al, 2008). Additionally, various CLN3 disease models present elevation of lipids belonging to the glycosphingolipid pathway such as ceramide, LacCer, GalCer, and gangliosides (Puranam et al, 1999; Rusyn et al, 2008; Somogyi et al, 2018).

The inventors tested different fluorescently-labeled reporters for the detection of intracellular lipids and sphingosines including; Cholera Toxin B subunit (CTB) to detect GM1 ganglioside (Rusyn et al, 2008); LipidTOX for neutral lipids (Somogyi et al, 2018); Shiga Toxin subunit B (STX) for globotriaosylceramide (Gb3) (Mallard & Johannes, 2003); and Filipin III to label Cholesterol (Namdar et al, 2012). These reporters were assessed in human ARPE-19 cells depleted for CLN3 (ARPE-19- CLN3-KO) cells by CRISPR genome editing. Although increasing levels of various lipid reporters were observed using these cells, the strong elevation of fluorescent-labeled Shiga Toxin subunit B (STX) staining suggest a dramatic accumulation of globotriaosylceramide (Gb3) in ARPE-19-CLN3- KO cells compared to their WT counterparts (Figure 1 A). Also, in comparison with the staining by the other lipid reporters, the STX labelling presents the best signal window for development of a cellbased assay in ARPE-19-CLN3-KO cells (Figure 1 A, Figure 9A). To confirm STX staining is selective for Gb3, HeLa cells lacking alpha-galactosidase A (HeLa-GLA-KO) were used, as they are a cell model of Anderson-Fabry disease that results from mutations in the GLA gene leading to Gb3 accumulation (Namdar et al, 2012) (Figure 9B). Further, STX staining was negative in cells depleted for Gb3 synthase (Gb3S) or treated with the glucosylceramide synthase inhibitor d-threo-1 -phenyl- 2-decanoylamino-3-morpholino-propanol (PDMP) (Abe et al, 2000; Raa et al, 2009) in both HeLa- GLA-KO and ARPE-19-CLN3-KO cells (Figure 1 B; Figure 9B). Co-staining with the lysosomal marker LAMP1 indicates that in addition to general elevation of Gb3 in cellular membranes, most intracellular Gb3 accumulates within the lysosomal compartment (Figure 1 B).

Next, the inventors investigated whether accumulation of Gb3 is a unique phenotypic feature of CLN3 disease or whether is present in other BDs. Thus, the inventors performed STX staining in cells depleted of either CLN6 or CLN7 genes by acute silencing using specific siRNAs in two cell lines, HeLa and ARPE-19 cells (Figure 9C). As expected, STX staining revealed that the silencing of CLN3 induces Gb3 accumulation in both HeLa and ARPE-19 cells (Figure 9C). Interestingly, the depletion of CLN7 also induces a significant elevation of Gb3 (Figure 9C). However, Gb3 does not accumulate in cells depleted of CLN6 (Figure 9C). PCR analysis of the Gb3S, CLN3, CLN6, and CLN7 mRNA levels confirms the efficiency of gene depletion by siRNAs (Figures 14-15). The selectivity of STX towards Gb3 was again confirmed in ARPE-19-CLN7-KO cells (Figure 1 C). Thus, either genetic or pharmacological targeting of Gb3 synthase significantly lowers STX/Gb3 staining in ARPE-19-CLN3-KO, ARPE-19-CLN7-KO, and GLA-KO cells (Figure 1 B-C, Figure 9B). Moreover, the direct measurement of Gb3 content by lipidomics analysis of both ARPE-19-CLN3-KO and ARPE-19-CLN7-KO cells shows a doubling of the total Gb3 present in these subtypes of BD (Figure 1 D).

Next, we investigated whether the accumulation of Gb3 observed upon CLN gene depletion in vitro also occurs in vivo. Thus, we analysed tissue samples derived from the CLN7Aex2 mouse, an animal model that well recapitulates the neurological phenotype observed in the human disease (Brandenstein et al, 2016). By using the STX labelling of Gb3, we stained brain sections from mice at 3 and 7.5 months of age, which represent the early and late stages of disease progression (Figure 2A-B). We observed a significant accumulation of Gb3 in different brain areas such as the cortex, hippocampus, and cerebellum compared with their corresponding healthy siblings (Figure 2A-B). The abnormal storage of Gb3 appears to be an early pathologic hallmark in the CLN7Aex2 disease since it was already present at three months of age (Figure 2A). Also, co-staining with neuronal nuclear antigen NeuN, but not with the astrocyte marker GFAP, indicates that Gb3 mostly accumulates in neurons and not in astrocyte (Figure 2C-D). In line with these observations, the Lipidomics analysis of immuno isolated neurons from fresh CLN7Aex2 mouse forebrain revealed a dramatic neuronal accumulation of both Gb3 and GM3, by 20- and 15-fold, respectively (Figure 2E). As expected, we also observe a similar accumulation of Gb3 in the same regions of the brain from a CLN3 mouse model at 7.5 months (Cotman et al, 2002; Staropoli et al, 2012), confirming that Gb3 storage is a signature of both BD variants (Figure 9D).

The accumulation of Gb3 or GM3 could be a collateral effect arising from the progressive lysosomal dysfunction in the BD models tested, or in contrast, play a direct role in the pathological mechanisms of these diseases. Thus, we silenced Gb3 synthase (siGb3S) to reduce Gb3 generation using specific siRNAs and measured the accumulation of SCMAS, a characteristic component of the pathological storage of CLN3 and CLN7 disease (Palmer, 2015). We observed that acute depletion of Gb3S causes a decrease of SCMAS within the lysosomes of CLN3-depleted cells (Figure 3A). A similar result was obtained by silencing of a more upstream enzyme in the glycosphingolipid synthesis pathway, LacCer synthase (siLCS), whereas silencing of the unrelated GM3 synthase (siGM3S) did not clear the accumulation of SCMAS (Figure 3A). PCR analysis of the Gb3S, LCS, and GM3S mRNA levels confirms the efficiency of the depletion of these genes by siRNAs (Appendix Figure 16). Together, these results indicate that the specific accumulation of Gb3 and not GM3 might play a role in the pathogenesis of two subtypes of Batten Disease, and therefore its targeting might ameliorate the BD phenotype.

Identification of small molecules reducing Gb3 accumulation in a cell model of Batten Disease.

We have found that abnormal accumulation of Gb3 appears to be pathologic in both in vitro and in vivo models of CLN3 and CLN7 diseases. Thus, we used the STX assay to identify FDA-approved compounds able to reduce the lysosomal accumulation of Gb3 by quantifying its co-localization with the lysosomal membrane protein LAMP1 in ARPE-19-CLN3-KO cells (see methods and Figure 17). The screening of 1280 FDA drugs resulted in the identification of 9 compound hits. These include two compounds belonging to the stilbenoid class of drugs that are selective estrogen receptor modulators (tamoxifen and toremifene), one alkaloid (apomorphine), three phenylpiperazines (itraconazole, ketoconazole, aripiprazole), a derivative of cholesterol (pregnenolone), a diphenylmethane (benztropine), and an acetylcholinesterase inhibitor (donepezil dihydrochloride) (Figure 3B). All nine compounds were further confirmed and tested in the same STX assay in a doseresponse format to determine EC50 and cell viability (Figure 3C, Figure 10). Tamoxifen treatment resulted in the most potent reduction of lysosomal STX accumulation with an EC50 of 0,75 pM without compromising vitality (Figure 3C). We only observed a very weak reduction in the number of nuclei at the highest concentration of tamoxifen that may suggest potential cytotoxicity at doses of >30pM (Figure 10).

Tamoxifen is a readily available EMA- and FDA-approved drug used for several decades for treating breast cancer and other hormone-related disorders. Importantly, it is also safe in pediatric conditions (Gayi et al, 2018). Given the well-established and widespread prescription of this drug, the inventors decided to focus on tamoxifen for further studies. 10pM tamoxifen was shown to promote the clearance of lysosomal Gb3, stained with STX, in human CLN3 patient fibroblasts (Figure 11 A), ARPE-19 cells depleted of CLN7 by siRNAs (Figure 11 B), ARPE-19 CLN7-KO cells (Figure 3D), and Nestin positive neuronal precursor cells (NPCs) derived from CLN7 patient iPSCs (Figure 3E, Figure 18). Tamoxifen was also able to reduce SCMAS levels in the same cells (Figure 1 1 C).

Tamoxifen is a selective estrogen receptor modulator (SERM) and the most commonly used drug for the treatment of estrogen receptor (ER) positive breast cancer (Shagufta & Ahmad, 2018). Thus, the inventors investigated whether tamoxifen's ability to reduce the accumulation of Gb3 could be through targeting ERs. Surprisingly, at two concentrations, tamoxifen was able to promote Gb3 clearance in two cell lines silenced for CLN3 that do not express ERs, U2OS and HeLa cells (Kallio et al, 2008; Selyunin et al, 2019) (Figure 4A, Figure 1 1 D, Figure 19 A-B). Thus, a siRNA-based survey has been performed to determine how many NCL subtypes (see Table 1 ) are accumulating Gb3 and, most importantly, whether they respond to tamoxifen treatment. The inventors found that all NCLs with the exception of CLN6 disease accumulates Gb3 and tamoxifen was able to reduce Gb3 accumulation in all tested NCL subtypes (Figure 20). These results indicate that tamoxifen induces Gb3 clearance in BD cellular models by a mechanism that is independent of the modulation of ERs. Together, the inventors have developed a novel phenotypic screening tool for repurposing compounds able to reduce lysosomal Gb3 accumulation in BD cells and identified tamoxifen as a potential corrector of all NCLs with the exception of CLN6 subtype. These data strongly suggest that tamoxifen may be a suitable drug to treat most subtypes of NCLs extending its new therapeutic application to multiple Batten Disease. The inventors recently published in vitro and in vivo results showing data included in this invention (Soldati et al., 2021 ). Table 1

^a Most prevalent phenotype in bold, phenotype variants and other details see MIM links ^b NCL-like phenotypes ^c Online Mendelian Inheritance in Man - the database of Human Genes and Genetic Disorders

Tamoxifen induces Gb3 clearance through activation of the transcription factor TFEB The clearance activity of tamoxifen in two different types of BD through a mechanism that is ER- independent might be explained by the activation of the transcription factor TFEB. This is a master gene of lysosomal function that, upon activation, induces lysosomal clearance of pathological storage in various LSDs, including CLN3 disease (Medina et al, 201 1). The inventors found that tamoxifen was able to significantly induce TFEB nuclear translocation in ARPE-19-CLN3-KO cells (Figure 12 A). To determine whether TFEB activation is a requirement for tamoxifen-mediated clearance of Gb3, the inventors tested tamoxifen in ARPE-19-CLN3-KO cells depleted of TFEB by using siRNAs (Figure 4B). While tamoxifen was effective in reducing Gb3 in ARPE-19-CLN3-KO cells treated with scrambled siRNAs (Figure 4B), it was not active in the TFEB-silenced CLN3-KO cells (Figure 4B). PCR analysis of the TFEB mRNA levels and western blotting confirmed the efficiency of siRNA-mediated depletion of TFEB (Figure 21 A-B). Consistently, viral-mediated transduction of an inducible vector expressing a nuclear-localized mutant form of TFEB was sufficient to clear Gb3 in ARPE-19-CLN3 KO cells (Figure 12 B). Indeed, tamoxifen was able to induce TFEB nuclear translocation in U2-OS cells that do not express ERs (Figure 4C), indicating that, like tamoxifen-mediated induction of Gb3 clearance, TFEB nuclear translocation is also ER- independent. mRNA expression analysis by qPCR shows the efficiency of the CLN3 gene depletion in U2-OS cells (Figure 19).

Lysosomotropic compounds possess weak-base properties that favor their accumulation in lysosomes by ion-trapping mechanisms (Ohkuma & Poole, 1981 ; Pisonero-Vaquero & Medina, 2017). Recent work shows that lysosomotropic anti-cancer drugs promote lysosome-mediated cancer drug resistance by stimulating activation of TFEB and the consequent increase in lysosomal biogenesis, lysosomal exocytosis, and autophagy (Zhitomirsky et al, 2018). Indeed, two compound hits in the screening, namely tamoxifen, and toremifene, both present a tertiary amine that makes them weak-bases able to transiently modify endolysosomal pH by a mechanism that is independent of the ER (Altan et al, 1999; Lu et al, 2017; Selyunin et al, 2019). Consistently, tamoxifen and toremifene were able to reduce Gb3 accumulation and induce TFEB nuclear translocation (Figure 13 A). Conversely, ospemifene, which is structurally related to tamoxifen (Taras et al, 2001 ), possesses similar potency targeting ER-mediated pathways (31 ), but has a hydroxyl group in place of the tertiary amine of tamoxifen in its side chain, was not able to induce TFEB nuclear translocation or Gb3 clearance at 10 pM (Figure 13 B). These observations suggest that the effects of tamoxifen on inducing TFEB activation and reducing intracellular Gb3 storage in BD models are due to its weak base property. Indeed, while 10 uM tamoxifen was able to transiently alkalinize the lysosome after 3h treatment, measured by a reduction in lysotracker staining, ospemifene was not effective at a similar concentration (Figure 13 C). Also, the inventors confirmed previous data showing that the effect of tamoxifen on lysosomal alkalinization is reversible (Figure 13 C) (Actis et al, 2021 ). mTOR kinase is the major kinase involved in the negative regulation of TFEB (Laplante & Sabatini, 2012). Thus, the inventors investigated whether tamoxifen induces TFEB nuclear translocation by inhibiting mTORCI activity. Interestingly, mTOR kinase activity measured by both high content imaging assays and immunoblot of its classical substrates such as p70S6K, 4EPB, and ULK1 , was not affected by tamoxifen treatment (Figure 22 A-B). However, and in agreement with the induction of TFEB nuclear translocation, tamoxifen reduced the phosphorylation of TFEB while ospemifene did not (Figure 5A-B). Recent work has shown that unlike other substrates of mTORCI , such as those tested above, TFEB is strictly dependent on the activation of RagC and RagD GTPases (Napolitano et al, 2020). Indeed, the overexpression of a constitutive active form of RagC (HA-GST- RagCS75L) was able to block tamoxifen-mediated TFEB nuclear translocation as well as the clearance of Gb3 (Figure 5C-D). As a negative control, the inventors treated HelaTFEB-GFP cells with the selective mTOR kinase inhibitor torin 1 that translocates TFEB even in the presence of active RagC (Napolitano et al, 2020) (Figure 5C). As expected, ospemifene was not effective at inducing TFEB translocation and was not affected by RagC expression (Figure 5C). Together, the inventors’ results indicate that the lysosomotropic feature of tamoxifen induces TFEB nuclear translocation through the inhibition of mTORCI via a Rag-dependent mechanism.

Tamoxifen ameliorates pathologic hallmarks of the CLN7 \ex2 mouse model

To test the efficacy of tamoxifen in vivo, the inventors selected the CLN7Aex2 mouse model (Brandenstein et al, 2016), which has a more severe phenotype than the existing CLN3 mouse models (Huber et al, 2020) and recapitulates very well the phenotype of human CLN7 patients. Thus, CLN7Aex2 mice show the accumulation of autofluorescent material and SCMAS in the central nervous system, as well as brain gliosis, clasping, hind limb paralysis, and seizures (Damme et al, 2014; Brandenstein et al, 2016; Huber et al, 2020). The inventors first tested the ability of tamoxifen to reduce pathologic hallmarks of disease by intraperitoneal injections of tamoxifen (40 mg/kg, twice per week) starting from 2.5 months old-mice to 7.5 months of age when the disease phenotype is well established (Brandenstein et al, 2016). The inventors investigated the ability of tamoxifen to reduce Gb3 storage by using Shiga toxin staining assay. The inventors found a significant reduction of Gb3 in the cortex and the cerebellum, but not the hippocampus, of 7.5 months-old tamoxifen- treated CLN7Aex2 mice compared with their age-matched untreated CLN7Aex2 mice (Figure 6). Then, the inventors investigated whether other features of the disease, such as the accumulation of SCMAS and the activation of microglia, could be reversed by the treatment with tamoxifen. Similar to the reduction of Gb3, the inventors found that tamoxifen treatment was able to significantly reduce the accumulation of SCMAS in the cortex and the cerebellum, but not the hippocampus (Figure 7). To test neuroinflammation, the inventors first confirmed that the levels of the small calcium-binding protein IBA1 , a specific marker of both resting and activated populations of microglia, were upregulated in brain sections from the CLN7Aex2 mouse (Figure 8A-B) compared to the age- matched wild type mice. By contrast, tamoxifen-treated CLN7Aex2 mice presented a significant reduction of IBA1 -positive cells in both the cortex and the cerebellum (Figure 8A-B). The inventors found a similar trend in the hippocampus, although it was not statistically significant (Figure 8A-B). Together, these results strongly indicate that tamoxifen treatment can reduce the pathological storage of GB3 and SCMAS as well as reduce signs of neuroinflammation in the brain of CLN7Aex2 mice.

Motor deficits are one of the primary clinical features of BD (Raininko et al, 1990; Kovacs et al, 2006; Mole et al, 2019). Indeed, by eight months of age, Cln7Aex2 mice began to manifest signs of neurological deterioration attested by clasping phenotype, hind-leg paralysis, tremor, and myoclonus epilepsies (Brandenstein et al, 2016). Motor deficits and balance are detectable by measuring the latency to fall from the rotarod and can be used as a read-out of the efficacy of potential therapeutic compounds in BD models (Finn et al, 2011 ). The inventors performed the rotarod test (Finn et al, 2011 ) during the whole period of treatment (6 measurements in total). Interestingly, tamoxifen- treated wild-type mice improved with age. CLN7Aex2 mice exhibit a marked locomotor dysfunction in the late stages of the disease, while tamoxifen-treated CLN7Aex2 mice displayed a higher latency to fall when compared with the untreated CLN7Aex2 mice, although did not improve to the extent of wild-type mice (Figure 8C). The inventors tested motor dysfunction by using the hindlimb clasping test (Lieu et al, 2013). In healthy mice, both hindlimbs remain splayed outward away from the abdomen with splayed toes. Partial retraction of one or both hindlimbs towards the body indicates a moderate phenotype. Severe motor dysfunction correlates with both hindlimbs partially retracted toward the body and touching the abdomen. As expected, the wild-type animals showed a normal extension reflex in the hindlimbs, while 7.5-month-old mutant mice did not. The inventors found that hindlimb clasping improved in the tamoxifen-treated CLN7Aex2 mice (Figure 8D). Together, the results of the motor tests suggest a partial recovery of motor coordination capacity in CLN7Aex2 mice treated with tamoxifen. In conclusion, in vivo administration of tamoxifen improves biochemical markers and motor deficits of CLN7 disease.

Discussion

Using cellular models of CLN3 and CLN7 diseases, and NPCs generated from iPS cells derived from CLN7 patient fibroblasts, the inventors found that Gb3 accumulates within lysosomes as a consequence of BD disease. This Gb3 accumulation is even more striking in neurons of CLN3 and CLN7 mouse models. In vitro, silencing of Gb3 synthase leads to the reduction of Gb3 levels, and also decreases the characteristic disease storage of subunit SCMAS, indicating that the altered levels of Gb3 might be part of the neuropathological features characterizing these diseases. Previous alterations of some glycosphingolipids such as ceramide, LacCer, and GM3 have been described in BD models (Puranam et al, 1999; Rusyn et al, 2008; Schmidtke et al, 2019). Indeed, the inventors found that in addition to endogenous Gb3 accumulation, the lipidomic analysis of CLN3 and CLN7 KO models showed increased levels of LacCer, the common precursor of both Gb3 and GM3 synthesis. Lipidomics analysis of freshly isolated neurons from the forebrain of CLN7 mice, displays an even higher elevation of Gb3 and GM3 and their precursor GlcCer. However, the inventors observed that while the depletion of LacCer synthase or Gb3 synthase reduced SCMAS accumulation, the silencing of the unrelated GM3 synthase did not, indicating that unbalancing the pathway involving the synthesis/degradation of Gb3 may contribute to the pathogenesis of CLN3 and CLN7 diseases. Recent SILAC-based quantitative analysis of the lysosomal proteome of MEFs from CLN7 mice showed significant differences in the expression of proteins involved in lipid trafficking and glycosphingolipid catabolism (Danyukova et al, 2018). Thus, these changes may be related to the pathologic accumulation of Gb3 in CLN7 disease. The same study revealed that the CLN5 protein is also downregulated in CLN7 MEFs (Danyukova et al, 2018). CLN5 mutations present with a similar disease onset, progression, and phenotypes as CLN7 disease suggesting that both genes may act in a common pathway that is disturbed in both diseases. Additionally, CLN5 can interact with CLN3 (Vesa et al, 2002), suggesting that CLN5 disease, together with CLN3 and CLN7, may belong to a subset of BDs accumulating Gb3. Future studies are needed to confirm this hypothesis and to determine the interaction of these three proteins in the regulation of brain Gb3 levels.

The inventors’ observations allowed to develop a cell-based HCI for Gb3 accumulation assay to screen >1200 FDA compounds. Among the compound hits, the inventors focused on the selective estrogen receptor modulator (SERM) tamoxifen that promotes the clearance of lysosomal Gb3 in CLN3 and CLN7 cells. The inventors found that tamoxifen activity is independent of its ER modulation but requires TFEB activation. TFEB can promote clearance of pathological storage both in vitro and in vivo in various models of LSDs (Sardiello et al, 2009; Spampanato et al, 2013; Palmer, 2015; Kauss et al, 2019). Indeed, the inventors confirmed that TFEB expression was sufficient to promote Gb3 clearance in ARPE-19-CLN3-KO cells. Thus, tamoxifen-mediated clearance via TFEB activation may represent a small molecule-based strategy to treat common types of BD. Using ospemifene, an analog of tamoxifen that does not contain the tertiary amine conferring lysosomotropic properties, the inventors determined that the activation of TFEB requires the weak- base nature of tamoxifen. Consistently, another lysosomotropic analog of tamoxifen, toremifene, also induces Gb3 clearance and TFEB activation in vitro. Another lysosomotropic SERM, raloxifene (Selyunin et al, 2019), is effective in neuroprotection and immunomodulatory effects in a mouse model of Parkinson’s disease (Poirier et al, 2016), supporting the potential benefits of repurposing approved stilbenoids to treat LSDs and more common neurodegenerative disorders. Since most of the approved CNS-penetrant drugs are lysosomotropic, future studies are needed to elucidate whether all compounds with this feature can promote clearance of pathological storage through the activation of TFEB or other properties are involved. Also, the logP and pKa properties of five out of the nine hits identified correspond to drugs with potential lysosomotropic properties (log P>2 ; pKa 6- 11 ), supporting further studies of these compounds in BD models.

Mechanistically, the inventors observed that tamoxifen induced TFEB nuclear translocation by specifically impairing mTORCI -mediated phosphorylation of TFEB without affecting mTORCI activity towards canonical substrates such as S6K, 4EBP, and ULK1 . Giving the recent observations that RagC/D GTPase activity can mediate selective phosphorylation of mTORCI substrates (Napolitano et al, 2020), the inventors postulated that lysosomotropic properties of tamoxifen specifically affect RagC/D activity leading to the dephosphorylation of TFEB. A few reports suggest that tamoxifen can alter glycosphingolipid metabolism in cancer cells (Lavie et al, 1997; Morad & Cabot, 2015). Thus, future studies are needed to determine whether the reported effects of tamoxifen on glycosphingolipid regulation may contribute to Gb3 clearance and whether the activation of TFEB is involved.

The inventors tested the efficacy of tamoxifen as a therapeutic agent by treating CLN7Aex2 mice. Tamoxifen treatment using a therapeutic concentration of 40 mg/kg ameliorated various phenotypic hallmarks of CLN7 mouse including; a) the accumulation of Gb3, b) SCMAS, c) microglia activation, and d) improved motor dysfunction measured by rotarod and hindlimb clasping.

In humans, tamoxifen is used orally, and crosses the blood-brain-barrier. It has shown neuroprotective activity in rat and dog models of brain ischemia and stroke, respectively (Kimelberg et al, 2003; Kimelberg, 2008; Boulos et al, 201 1 ), and it has been used in the treatment of a variety of childhood disorders (Maddalozzo et al, 1993; Walter et al, 2000; Derman et al, 2003; Lawrence et al, 2004; Kreher et al, 2005). Adverse effects in these populations have been rare, and tamoxifen seems to have an excellent safety profile overall. The data obtained by the present invents provide support for tamoxifen as well as other SERMs as novel therapeutic agents for the treatment of BD, in particular for the CLN3 and CLN7 forms of the disease.

References

Abe A, Gregory S, Lee L, Killen RD, Brady RO, Kulkarni A & Shayman JA (2000) Reduction of globotriaosylceramide in Fabry disease mice by substrate deprivation. J Clin Invest 105: 1563-1571

Actis C, Muzio G & Autelli R (2021 ) Autophagy Triggers Tamoxifen Resistance in Human Breast Cancer Cells by Preventing Drug-Induced Lysosomal Damage. Cancers (Basel) 13

Altan N, Chen Y, Schindler M & Simon SM (1999) Tamoxifen inhibits acidification in cells independent of the estrogen receptor. PNAS 96: 4432-4437

Boulos AS, Deshaies EM, Dalfino JC, Feustel PJ, Popp AJ & Drazin D (2011 ) Tamoxifen as an effective neuroprotectant in an endovascular canine model of stroke. J Neurosurg 1 14: 1 117-1126

Boustany R-MN (2013) Lysosomal storage diseases-the horizon expands. Nat Rev Neurol 9: 583- 598

Brandenstein L, Schweizer M, Sedlacik J, Fiehler J & Storch S (2016) Lysosomal dysfunction and impaired autophagy in a novel mouse model deficient for the lysosomal membrane protein Cln7. Hum Mol Genet 25: 777-791 Cotman SL, Vrbanac V, Lebel L-A, Lee RL, Johnson KA, Donahue L-R, Teed AM, Antonellis K, Bronson RT, Lerner TJ, et al (2002) Cln3 Aex7/8 knock-in mice with the common JNCL mutation exhibit progressive neurologic disease that begins before birth. Hum Mol Genet 11 : 2709-2721

Damme M, Brandenstein L, Fehr S, Jankowiak W, Bartsch U, Schweizer M, Hermans-Borgmeyer I & Storch S (2014) Gene disruption of Mfsd8 in mice provides the first animal model for CLN7 disease. Neurobiol Dis 65: 12-24

Danyukova T, Ariunbat K, Thelen M, Brocke-Ahmadinejad N, Mole SE & Storch S (2018) Loss of CLN7 results in depletion of soluble lysosomal proteins and impaired mTOR reactivation. Human Molecular Genetics 27: 171 1-1722

Derman O, Kanbur NO & Kutluk T (2003) Tamoxifen treatment for pubertal gynecomastia. Int J Adolesc Med Health 15: 359-363

Double KL, Dedov VN, Fedorow H, Kettle E, Halliday GM, Garner B & Brunk UT (2008) The comparative biology of neuromelanin and lipofuscin in the human brain. Cell Mol Life Sci Q5: 1669- 1682

Finn R, Kovacs AD & Pearce DA (2011 ) Altered sensitivity of cerebellar granule cells to glutamate receptor overactivation in the Cln3(Aex7/8)-knock-in mouse model of juvenile neuronal ceroid lipofuscinosis. Neurochem Int 58: 648-655

FitzPatrick LM, Hawkins KE, Delhove JMKM, Fernandez E, Soldati C, Bullen LF, Nohturfft A, Waddington SN, Medina DL, Bolanos JP, et al (2018) NF-KB Activity Initiates Human ESC-Derived Neural Progenitor Cell Differentiation by Inducing a Metabolic Maturation Program. Stem Cell Reports 10: 1766-1781

Gayi E, Neff LA, Massana Munoz X, Ismail HM, Sierra M, Mercier T, Decosterd LA, Laporte J, Cowling BS, Dorchies OM, et al (2018) Tamoxifen prolongs survival and alleviates symptoms in mice with fatal X-linked myotubular myopathy. Nat Common 9: 4848

Haltia M & Goebel HH (2013) The neuronal ceroid-lipofuscinoses: A historical introduction. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 1832: 1795-1800

Huber RJ, Hughes SM, Liu W, Morgan A, Tuxworth Rl & Russell C (2020) The contribution of multicellular model organisms to neuronal ceroid lipofuscinosis research. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 1866: 165614

Huber RJ & Mathavarajah S (2018) Secretion and function of Cln5 during the early stages of Dictyostelium development. Biochim Biophys Acta Mol Cell Res 1865: 1437-1450 Kallio A, Guo T, Lamminen E, Seppanen J, Kangas L, Vaananen HK & Harkonen P (2008) Estrogen and the selective estrogen receptor modulator (SERM) protection against cell death in estrogen receptor alpha and beta expressing U2OS cells. Molecular and Cellular Endocrinology 289: 38-48

Katz ML & Robison WG (2002) What is lipofuscin? Defining characteristics and differentiation from other autofluorescent lysosomal storage bodies. Arch Gerontol Geriatr 34: 169-184

Kauss V, Dambrova M & Medina DL (2019) Pharmacological approaches to tackle NCLs. Biochim Biophys Acta Mol Basis Dis: 165553

Kimelberg HK (2008) Tamoxifen as a powerful neuroprotectant in experimental stroke and implications for human stroke therapy. Recent Pat CNS Drug DiscovS 104-108

Kimelberg HK, Jin Y, Charniga C & Feustel PJ (2003) Neuroprotective activity of tamoxifen in permanent focal ischemia. J Neurosurg 99: 138-142

Kovacs AD, Weimer JM & Pearce DA (2006) Selectively increased sensitivity of cerebellar granule cells to AMPA receptor-mediated excitotoxicity in a mouse model of Batten disease. Neurobiol Dis 22: 575-585

Kreher NC, Eugster EA & Shankar RR (2005) The use of tamoxifen to improve height potential in short pubertal boys. Pediatrics 1 16: 1513-1515

Laplante M & Sabatini DM (2012) mTOR Signaling in Growth Control and Disease. Cell 149: 274- 293

Lavie Y, Cao H t, Volner A, Lucci A, Han TY, Geffen V, Giuliano AE & Cabot MC (1997) Agents that reverse multidrug resistance, tamoxifen, verapamil, and cyclosporin A, block glycosphingolipid metabolism by inhibiting ceramide glycosylation in human cancer cells. J Biol Chem 272: 1682-1687

Lawrence SE, Faught KA, Vethamuthu J & Lawson ML (2004) Beneficial effects of raloxifene and tamoxifen in the treatment of pubertal gynecomastia. J Pediatr 145 71-76

Lieu CA, Chinta SJ, Rane A & Andersen JK (2013) Age-related behavioral phenotype of an astrocytic monoamine oxidase-B transgenic mouse model of Parkinson’s disease. PLoS ONE S e54200

Lopez-Fabuel I, Douce JL, Logan A, James AM, Bonvento G, Murphy MP, Almeida A & Bolanos JP (2016) Complex I assembly into supercomplexes determines differential mitochondrial ROS production in neurons and astrocytes. PNAS 113: 13063-13068

Lu S, Sung T, Lin N, Abraham RT & Jessen BA (2017) Lysosomal adaptation: How cells respond to lysosomotropic compounds. PLoS ONE 12: e0173771

Maddalozzo J, Tenta LT, Hutchinson LR, Crawford SE & Morse DS (1993) Juvenile fibromatosis: hormonal receptors. Int J Pediatr Otorhinolaryngol 25: 191-199 Mallard F & Johannes L (2003) Shiga Toxin B-Subunit as a Tool to Study Retrograde Transport. In E. coli: Shiga Toxin Methods and Protocols, Philpott D & Ebel F (eds) pp 209-220. Totowa, NJ: Humana Press

Medina DL, Fraldi A, Bouche V, Annunziata F, Mansueto G, Spampanato C, Puri C, Pignata A, Martina JA, Sardiello M, et al (2011 ) Transcriptional activation of lysosomal exocytosis promotes cellular clearance. Dev Ceii 2 : 421-430

Mole SE (2014) Development of new treatments for Batten disease. The Lancet Neurology 13: 749- 751

Mole SE, Anderson G, Band HA, Berkovic SF, Cooper JD, Kleine Holthaus S-M, McKay TR, Medina DL, Rahim AA, Schulz A, et al (2019) Clinical challenges and future therapeutic approaches for neuronal ceroid lipofuscinosis. Lancet Neurol 18: 107-1 16

Mole SE & Cotman SL (2015) Genetics of the Neuronal Ceroid Lipofuscinoses (Batten disease). Biochim Biophys Acta 1852: 2237-2241

Mole SE, Williams RE & Goebel HH (2005) Correlations between genotype, ultrastructural morphology and clinical phenotype in the neuronal ceroid lipofuscinoses. Neurogenetics 6: 107-126

Morad SAF & Cabot MC (2015) Tamoxifen regulation of sphingolipid metabolism-Therapeutic implications. Biochim Biophys Acta 1851 : 1134-1145

Namdar M, Gebhard C, Studiger R, Shi Y, Mocharla P, Schmied C, Brugada P, Luscher TF & Camici GG (2012) Globotriaosylsphingosine accumulation and not alpha-galactosidase-A deficiency causes endothelial dysfunction in Fabry disease. PLoS ONE 7: e36373

Napolitano G, Di Malta C, Esposito A, de Araujo MEG, Pece S, Bertalot G, Matarese M, Benedetti V, Zampelli A, Stasyk T, et al (2020) A substrate-specific mTORCI pathway underlies Birt-Hogg- Dube syndrome. Nature 585: 597-602

Ohkuma S & Poole B (1981 ) Cytoplasmic vacuolation of mouse peritoneal macrophages and the uptake into lysosomes of weakly basic substances. Journal of Cell Biology 90: 656-664

Palmer DN (2015) The relevance of the storage of subunit c of ATP synthase in different forms and models of Batten disease (NCLs). Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 1852: 2287-2291

Pisonero-Vaquero S & Medina DL (2017) Lysosomotropic Drugs: Pharmacological Tools to Study Lysosomal Function. Curr Drug Metab 18: 1147-1158 Poirier A-A, Cote M, Bourque M, Morissette M, Di Paolo T & Soulet D (2016) Neuroprotective and immunomodulatory effects of raloxifene in the myenteric plexus of a mouse model of Parkinson’s disease. Neurobiology of Aging 48: 61-71

Puranam KL, Guo W-X, Qian W-H, Nikbakht K & Boustany R-M (1999) CLN3 Defines a Novel Antiapoptotic Pathway Operative in Neurodegeneration and Mediated by Ceramide. Molecular Genetics and Metabolism 66: 294-308

Raa H, Grimmer S, Schwudke D, Bergan J, Walchli S, Skotland T, Shevchenko A & Sandvig K (2009) Glycosphingolipid Requirements for Endosome-to-Golgi Transport of Shiga Toxin. Traffic 10: 868-882

Raininko R, Santavuori P, Heiskala H, Sainio K & Palo J (1990) CT findings in neuronal ceroid lipofuscinoses. Neuropediatrics 21 : 95-101

Rusyn E, Mousallem T, Persaud-Sawin D-A, Miller S & Boustany R-MN (2008) CLN3p impacts galactosylceramide transport, raft morphology, and lipid content. Pediatr Res 63: 625-631

Sampaio JL, Gerl MJ, Klose C, Ejsing CS, Beug H, Simons K & Shevchenko A (201 1 ) Membrane lipidome of an epithelial cell line. PNAS 108: 1903-1907

Sardiello M, Palmieri M, di Ronza A, Medina DL, Valenza M, Gennarino VA, Di Malta C, Donaudy F, Embrione V, Polishchuk RS, et al (2009) A gene network regulating lysosomal biogenesis and function. Science 325: 473-477

Schmidtke C, Tiede S, Thelen M, Kakela R, Jabs S, Makrypidi G, Sylvester M, Schweizer M, Braren I, Brocke-Ahmadinejad N, et al (2019) Lysosomal proteome analysis reveals that CLN3-defective cells have multiple enzyme deficiencies associated with changes in intracellular trafficking. J Biol Chem 294: 9592-9604

Selyunin AS, Hutchens S, McHardy SF & Mukhopadhyay S (2019) Tamoxifen blocks retrograde trafficking of Shiga toxin 1 and 2 and protects against lethal toxicosis. Life Science Alliance 2

Shagufta & Ahmad I (2018) Tamoxifen a pioneering drug: An update on the therapeutic potential of tamoxifen derivatives. European Journal of Medicinal Chemistry 143: 515-531

Somogyi A, Petcherski A, Beckert B, Huebecker M, Priestman DA, Banning A, Cotman SL, Platt FM, Ruonala MO & Tikkanen R (2018) Altered Expression of Ganglioside Metabolizing Enzymes Results in GM3 Ganglioside Accumulation in Cerebellar Cells of a Mouse Model of Juvenile Neuronal Ceroid Lipofuscinosis. Int J Mol Sci 19

Soldati C, Lopez-Fabuel I, Wanderlingh LG, Garcia-Macia M, Monfregola J, Esposito A, Napolitano G, Guevara-Ferrer M, Scotto Rosato A, Krogsaeter EK, Paquet D, Grimm CM, Montefusco S, Braulke T, Storch S, Mole SE, De Matteis MA, Ballabio A, Sampaio JL, McKay T, Johannes L, Bolanos JP, Medina DL. Repurposing of tamoxifen ameliorates CLN3 and CLN7 disease phenotype. EMBO Mol Med. 2021 Oct 7;13(10):e13742. doi: 10.15252/emmm.202013742. Epub 2021 Aug 19. PMID: 34411438; PMCID: PMC8495452

Spampanato C, Feeney E, Li L, Cardone M, Lim J-A, Annunziata F, Zare H, Polishchuk R, Puertollano R, Parenti G, et al (2013) Transcription factor EB (TFEB) is a new therapeutic target for Pompe disease. EMBO Mol Med 5: 691-706

Staropoli JF, Haliw L, Biswas S, Garrett L, Holter SM, Becker L, Skosyrski S, Silva-Buttkus PD, Calzada-Wack J, Neff F, et al (2012) Large-Scale Phenotyping of an Accurate Genetic Mouse Model of JNCL Identifies Novel Early Pathology Outside the Central Nervous System. PLOS ONE 7: e38310

Taras TL, Wurz GT & DeGregorio MW (2001 ) In vitro and in vivo biologic effects of Ospemifene (FC- 1271 a) in breast cancer. The Journal of Steroid Biochemistry and Molecular Biology 77: 271-279

Vesa J, Chin MH, Oelgeschlager K, Isosomppi J, DellAngelica EC, Jalanko A & Peltonen L (2002) Neuronal ceroid lipofuscinoses are connected at molecular level: interaction of CLN5 protein with CLN2 and CLN3. Mol Biol Cell 13: 2410-2420

Walter AW, Gajjar A, Reardon DA, Thompson SJ, Langston JW, Jones-Wallace D, Kun LE & Heideman RL (2000) Tamoxifen and carboplatin for children with low-grade gliomas: a pilot study at St. Jude Children’s Research Hospital. J Pediatr Hematol Oncol 22: 247-251

Welford RWD, Muhlemann A, Garzotti M, Rickert V, Groenen PMA, Morand O, Ugeyler N & Probst MR (2018) Glucosylceramide synthase inhibition with lucerastat lowers globotriaosylceramide and lysosome staining in cultured fibroblasts from Fabry patients with different mutation types. Hum Mol Genet 27: 3392-3403

Zhitomirsky B, Yunaev A, Kreiserman R, Kaplan A, Stark M & Assaraf YG (2018) Lysosomotropic drugs activate TFEB via lysosomal membrane fluidization and consequent inhibition of mTORCI activity. Cell Death & Disease 9: 1-15

Claims

42 CLAIMS

1 . A selective estrogen receptor modulator (SERM) for use in the therapeutic treatment of a neuronal ceroid lipofuscinosis.

2. A selective estrogen receptor modulator (SERM) for use in the therapeutic treatment of neuroinflammation in a patient affected by a neuronal ceroid lipofuscinosis.

3. A selective estrogen receptor modulator (SERM) for use in the therapeutic treatment of motor coordination impairment in a patient affected by a neuronal ceroid lipofuscinosis.

4. A selective estrogen receptor modulator (SERM) for use in the therapeutic treatment of vision loss in a patient affected by a neuronal ceroid lipofuscinosis.

5. A selective estrogen receptor modulator (SERM) for use in reducing accumulation of sphingolipids, in particular globotriaosylceramide (Gb3), in a patient affected by a neuronal ceroid lipofuscinosis.

6. The selective estrogen receptor modulator (SERM) for use according to any of claims 1 to 5, wherein the neuronal ceroid lipofuscinosis is selected from the group consisting of CLN1 , CLN2, CLN3, CLN4, CLN5, CLN6, CLN7, CLN8, CLN9, CLN10, CLN11 , CLN12, CLN13 and CLN14.

7. The selective estrogen receptor modulator (SERM) for use according to claim 6, wherein the neuronal ceroid lipofuscinosis is CLN3 or CLN7.

8. The selective estrogen receptor modulator (SERM) for use according to any of claims 1 to 7, wherein the SERM contains a tertiary amine in its chemical structure.

9. The selective estrogen receptor modulator for use according to any of claims 1 to 7, which is selected from the group consisting of tamoxifen, (E) tamoxifen, toremifen, raloxifen, clomifene, ospemifene, bazedoxifene, nafoxidine, lasofoxifene, zuclomifene, afimoxifen, N-Desmethyl tamoxifene, droloxifene, tamoxifen aziridine, Idoxifene, 2-Methyl-4-hydroxytamoxifen, endoxifen, phenyltoloxamine, tamoxifen N-oxide, tamoxifen epoxide, afimoxifene, diethylaminoethoxyhexestrol, etoloxamine, tesmilifene, tamoxifen-d5, myoparkil and any pharmaceutically acceptable salt, ester, ether, isomer, mixture of isomers, complex, derivative or deuterated form thereof.

10. The selective estrogen receptor modulator (SERM) for use according to claim 9, which is selected from the group consisting of tamoxifen, (E) tamoxifen, toremifen, clomifene, ospemifene, zuclomifene, afimoxifen, N-desmethyl tamoxifene, droloxifene, tamoxifen aziridine, Idoxifene, 2- methyl-4-hydroxytamoxifen, endoxifen, tamoxifen N-oxide, tamoxifen epoxide, tamoxifen-d5, and any pharmaceutically acceptable salt, ester, ether, isomer, mixture of isomers, complex, derivative or deuterated form thereof. 43

11 . The selective estrogen receptor modulator (SERM) for use according to any one of claims 1 to 10, wherein the use comprises the simultaneous, separate or sequential administration of a gene therapy, an enzyme replacement therapy or a gene editing procedure.

12. A pharmaceutical composition comprising a selective estrogen receptor modulator (SERM) preferably as defined in any of claims 8 to 10 together with pharmaceutically acceptable excipient, vehicle and/or diluent, for the use as defined in any of claims 1 to 7.

13. The pharmaceutical composition according to claim 12, which additionally comprises one or more anticonvulsants.

14. The pharmaceutical composition according to claim 13, wherein the one or more anticonvulsants are selected from the group consisting of valproic acid, levetiracetam, lamotrigine, clonazepam, zonisamide, carbamazepine, topiramate, phenytoin and oxcarbazepine.

15. The pharmaceutical composition for use according to any of claims 12 to 14, which is formulated in a dosage form for topical, oral, parenteral, intravitreal, transretinal or ophthalmic administration.

16. The pharmaceutical composition for use according to claim 15, which is formulated as a tablet, a capsule, or as an ophtalmic solution, suspension, ointment or emulsion.

17. A kit-of-parts comprising a selective estrogen receptor modulator (SERM) preferably as defined in any of claims 8 to 10 and one or more anticonvulsants preferably selected from the group consisting of valproic acid, levetiracetam, lamotrigine, clonazepam, zonisamide, carbamazepine, topiramate, phenytoin and oxcarbazepine, for simultaneous, separate or sequential use in the therapeutic treatment of a neuronal ceroid lipofuscinosis,

18. A kit-of-parts comprising a selective estrogen receptor modulator (SERM) preferably as defined in any of claims 8 to 10 and one or more anticonvulsants preferably selected from the group consisting of valproic acid, levetiracetam, lamotrigine, clonazepam, zonisamide, carbamazepine, topiramate, phenytoin and oxcarbazepine, for simultaneous, separate or sequential use for the therapeutic treatment of neuroinflammation, motor coordination impairment and/or vision loss in a patient affected by a neuronal ceroid lipofuscinosis.

19. A kit-of-parts comprising a selective estrogen receptor modulator (SERM) preferably as defined in any of claims 8 to 10 and one or more anticonvulsants preferably selected from the group consisting of valproic acid, levetiracetam, lamotrigine, clonazepam, zonisamide, carbamazepine, topiramate, phenytoin and oxcarbazepine, for simultaneous, separate or sequential use in reducing accumulation of sphingolipids, in particular globotriaosylceramide (Gb3), in a patient affected by a neuronal ceroid lipofuscinosis. 44

20. The kit-of-parts according to any of claims 17 to 19, wherein the neuronal ceroid lipofuscinosis is selected from the group consisting of CLN1 , CLN2, CLN3, CLN4, CLN5, CLN6, CLN7, CLN8, CLN9, CLN10, CLN11 , CLN12, CLN13 and CLN14.

21 . An in vitro method of assessing the effectiveness of a therapeutic treatment administered to a subject suffering from a neuronal ceroid lipofuscinosis, the method comprising the steps of: a) determining the level of globotriaosylceramide (Gb3) accumulation in a biological sample from the subject after or during administration of the therapeutic treatment, and b) comparing the level determined in step a) with the level of globotriaosylceramide (Gb3) accumulation in a biological sample from the subject before administration of the therapeutic treatment, wherein a decreased level of globotriaosylceramide (Gb3) accumulation after treatment compared to before treatment is indicative of an effectiveness of the therapeutic treatment.

22. An in vitro method of diagnosing a neuronal ceroid lipofuscinosis in a subject, the method comprising the steps of: a) determining the level of globotriaosylceramide (Gb3) accumulation in a biological sample from the subject, and b) comparing the level determined in step a) with a predetermined globotriaosylceramide (Gb3) accumulation threshold which is indicative of a healthy state, wherein a level of globotriaosylceramide (Gb3) accumulation determined in step a) above said predetermined threshold is indicative of the subject being affected by a neuronal ceroid lipofuscinosis.

23. The in vitro method according to claim 21 , wherein said therapeutic treatment is with a selective estrogen receptor modulator (SERM).

24. The in vitro method according to claim 23, wherein the selective estrogen receptor modulator (SERM) is as defined in any of claims 8 to 10.

25. The in vitro method according to any of claims 21 to 24, wherein the neuronal ceroid lipofuscinosis is selected from the group consisting of CLN1 , CLN2, CLN3, CLN4, CLN5, CLN7, CLN8, CLN9, CLN10, CLN1 1 , CLN12, CLN13 and CLN14, preferably the neuronal ceroid lipofuscinosis is CLN3 or CLN7.

26. The in vitro method according to any of claims 21 to 25, wherein the biological sample is a serum sample or a plasma sample.

27. The in vitro method according to claim 26, wherein the level of globotriaosylceramide (Gb3) accumulation is determined by tandem mass spectrometry and lipidomics analysis.

28. The in vitro method according to any of claims 21 to 25, wherein the biological sample is a cell sample, preferably a sample of retinal pigment epithelial cells, fibroblasts or neurons.

29. The in vitro method according to claim 28, wherein the level of globotriaosylceramide (Gb3) accumulation is determined by a cell-based high content imaging (HCI) screening assay.