WO2000012454A9 - Novel polycationic lipids - Google Patents

Abstract

A cationic lipid for transfection of macromolecules in which the lipid has a polyether or glyceryl backbone which lipids can be contained in a liposome to effectively transfect a variety of cell types and improve the efficiency of transfection. Compositions containing said lipids and methods of using the same are also disclosed.

Description

Novel Polycationic Lipids

Related Applications

This application claims the priority of U.S. Application Serial No. 60/098,073 filed on August 27, 1998. The contents of that prior application is fully incorporated herein by reference.

Field of the Invention

This invention relates to polycationic lipids useful for the delivery (transfection) of nucleic acids (DNA,RNA) and other negatively charged or neutral molecules into living cells, either in vivo or in vitro.

Background of the Invention

Liposome aggregates made -with polycationic lipids are useful structures capable of complexing with negatively charged macromolecules such as DNA or RNA. These complexes can be taken up by living cells and then, once inside the cytosol, through an unknown mechanism, they are presumed to migrate into the cell nucleus. In the nucleus, there are enzymes capable of "reading" and "expressing" the message coded by the nucleic acids so delivered and produce new proteins, which were not being produced by the cell before the transfection of the foreign nucleic acid. When cells so transfected divide and their daughter cells still have the capability to produce the proteins encoded by the initially transfected DNA, the transfection is said to be stable. That is, the new DNA has stably integrated into the cell nucleus changing the cell's genetic make-up. If, on the other hand, the parent cells can produce the protein encoded by the transfected DNA, but their daughter cells are not capable of expressing that tranfected DNA, the transfection is said to be transient. RNA transfection is always transient. Stable transfection of human or animal cells is the basis of so called gene therapy, since cells which are deficient in a crucial protein for the organism's survival could in principle be repaired by stably transfecting the DNA needed to produce the absent protein. Another type of potential use of DNA/RNA transfection for therapy is antisense therapy. In this approach, a short piece of nucleic acid (oligonucleotide) capable of adhering (hybridizing) to defective DNA (or RNA) which is being expressed by the cells to produce an undesired protein, such as an oncoprotein (cancer causing protein), is transfected into the cells in order to stop the expression of the undesired protein by virtue of its adherence to the defective nucleic acid. This method of therapy does not change the genetic make-up of the cell, but blocks the effect of the genetic disorder already present in the cell's genome. Besides these potential applications of polycationic lipids for use in human therapy, there is already a well established market for these types of chemicals in the research products field. They are currently being used by researchers to deliver nucleic acids and proteins into cells in order to study how the expression of different genes affect cell growth and function.

There are two possible ways to deliver DNA into cells for gene therapy : ex vivo and in vivo transfection. In the ex vivo modality, cells from a patient are removed from the body, cultured and transfected in vitro. Then, the cells are returned into the patient where the beneficial DNA message is hopefully expressed. In the in vivo mode, the DNA is delivered directly into the patient, which makes this procedure simpler and less expensive. To date, the only effective way to deliver DNA in vivo is by using a virus which naturally infects cells of a specific organ (targets that organ) within the body, and whose genetic make up has been modified by adding the DNA beneficial to the patient. Once inside the cells of the patient, the virus can incorporate the new DNA in the genome of the cell (stable transfection) and the parent cell and its daughters can express the beneficial protein. The pathological component of the virus has been deleted before the patient is exposed to such a virus and only the targeting component left intact. Virus can do this process sometimes with nearly 100% efficiency. However, there are risks associated with their use, since they can produce immunological reactions which may be fatal to the patient; the DNA incorporation in the cell's genome is random, therefore it might disrupt needed genes or activate oncogenes; they are also difficult to mass produce, and so forth.

Liposomes or lipid aggregates do not have the side effects of viruses, but are not as efficient as viruses are. There is a constant need to develop newer lipids that can approach the efficiency of viruses without their undesirable side effects (E. Marshall, Science 269,1050 (1995)) . There are several lipids for nucleic acids transfection already in the market. The most relevant of these lipids are: (a) DOTMA (N-[l-(2,3-dioleoyloxy)propyl]- N,N,N - trimethylammonium chloride, U.S. Pat. No. 4,897,355 to D. Eppstein et al.); (b) DMRIE (D,L-l,2-0-dimyristyl-3-dimethylaminopropyl-b-hydroxyethylammoniumchloride,U.S.Pat. No 5,264,618 to Feigner, P.L. et al.); (c) DOTAP (l,2-bis(oleoyloxy)-3- 3(trimethylammonia)propane) Boehringer-Mannheim Catalog No.l 202 375); (d) DOGS (5- carboxysperminyglycine dioctadecylamide, U.S. Pat. No 5,171,678 to Behr, J-P. et al. DOGS is sold under the trade name Transfectam™ by the Promega Corp. Madison, WI ); (e) DOSPA (2,3-dioleyloxy-N-[2(sper-τ necarboxyamido)ethyl]-N,Nκiimethyl-l-prop- aminium trifluoroacetate, U.S. Pat. No. 5,334,761 to Gebeyehu, G. et al.); (f) DDAB (Dimethyloctadecylammoniumbromide, U.S. Pat. No. 5,279,833 to Rose, J.K.); and (g) TMTPS (N,N,N,N-Tetramethyltetrapalmylspermine, PCT Int.Pub.No. WO 95/17373. Haces, A. et al.). DOTMA, DOSPA, DDAB and TMTPS are sold by Life Technologies, Inc., Gaithersburg, MD, under the trade names of Lipofectin, Lipofect AMINE, LipofectACE and CellFECTIN, respectively. A recent relevant publication which deals with art related to the present invention has been reported by Ruysschaert et al.((1994)-5iocbem. Biophys. Res. Commun., 203, 1622 -1228). All of these lipids, except DOGS, are formulated with dioleolylphosphatidylethanolamine (DOPE), which is a neutral lipid devoid of transfection activity, in order to make the active liposomes. These lipids possess some desirable characteristics, however they are far from the ideal vehicle to deliver DNA. Their main drawbacks are low efficiency, non-specificity of targeting, considerably toxicity, low water solubility, and serum inhibition of their action.

Although progress has been made in overcoming some of these obstacles, there is considerable room for improvement. The design of these lipids is still a semi-empirical endeavor, since very little is known about the mechanism by which they act.

Therefore, it is an object of this invention to improve the desired characteristics of these lipids by exploring and incorporating new chemical functionalities as well as spatial or topological arrangements which improve the transfection efficiency and lower the toxicity.

It is also an object of this invention to synthesize polycationic lipids which incorporate a small, non lipid-bilayer-disturbing moiety that mimics a natural molecule, which cells can recognize as their natural effector or ligand, thus facilitating the transfection as well as the specificity of targeting of the macromolecule. Summary of the Invention

In accordance with the invention, a series of new polycationic lipids, their use, and their method of preparation is described. Such lipids are useful as transfection reagents for nucleic acids, oligonucleotides, mononucleotides, polypeptides, and proteins. In addition, some of these lipids are also useful as more effective detergents for cleaning and as vehicles in the cosmetic field.

In one aspect of the present invention, there are described novel oxo and sulfinyl backbone substituted polycationic lipids with ammonium, guanidinium and amidinium positively charged moieties as anchoring groups having Formula I, as shown below and in WO 97/42819 (International Application No. PCT/US97/09093), the contents of which is fully incorporated herein by reference.

The present invention also encompasses a series of novel phosphatidyl and glyceryl guanidinium cationic lipids having the formula 2 shown below and in Figure 8.

In a preferred form of the present invention, the backbone of the substituted polycationic lipis is comprised of a polyether in accordance with Formula 3 as shown in Figure 9:

In a still further aspect of the present invention, there are provided cationic lipids with a glyceryl backbone having Formula 4 as shown below and in Figure 10.

Brief Description of the Drawings

The following drawings are illustrative of embodiments of the invention and are not intended to limit the invention as encompassed by the claims forming part of the application.

Figure 1 is a graphic representation of Scheme I for the preparation of polycationic lipids of the present invention;

Figure 2 is a graphic representation of Scheme II for the preparation of polycationic lipids of the present invention;

Figure 3 is a graphic representation of Scheme HI for the preparation of polycationic lipids of the present invention;

Figure 4 is a graphic representation of Scheme IN for the preparation of polycationic lipids of the present invention;

Figure 5 is a graphic representation of Scheme V for the preparation of polycationic lipids of the present invention;

Figure 6 is a graphic representation of Scheme VI for the preparation of polycationic lipids of the present invention;

Figure 7 is a graphic representation of Formula 1, which represents a series of polycationic lipids of the present invention;

Figure 8 is a graphic representation of Formula 2, which represents a second series of polycationic lipids of the present invention;

Figure 9 is a graphic representation of Formula 3, which represents a third series of polycationic lipids of the present invention; and,

Figure 10 is a graphic representation of Formula 4, which represents a fourth series of polycationic lipids of the present invention.

Figure 11 is a graphic representation of compound 13a of the present invention.

Figure 12 is a graphic representation of compound 13d of the present invention.

Figure 13 is a graphic representation of compound 5a of the present invention.

Figure 14 is a graphic representation of compound 13b of the present invention.

Figure 15 is a graphic representation of compound 6d of the present invention.

Figure 16 is a graphic representation of compound 15 of the present invention.

Detailed Description of the Invention

The compounds of the present invention can be used alone or in mixtures with other

liposome forming compounds (co-lipids) to prepare lipid aggregates which are useful to deliver

macromolecules, specifically negatively charged macromolecules to living cells either in culture (in vitro) or in vivo. Colipids are compounds capable of producing stable liposomes,

alone, or in combination with other lipid components. Such colipids are preferably neutral, although they can alternatively be positively or negatively charged. Some colipids are

disclosed, for example, in U.S. Patent No. 4,897,355, the entire contents of which is

incorporated herein by reference. Such examples include phospholipid-related materials, such

as lecithin, phosphatid lethanolamine, lysolecithin, lysophosphatidylethanolamine, phosphatidylserine, phosphatidylinositol, sphinogomyelin, cephalin, cardiolipin, phosphatidic

acid, cerebrosides, dicetylphosphate, dioleoylphosphatidylcholine (DOPC), dipalmitoylphosphatidylcholine (DPPG), dioleoylphosphatidylglycerol (DOPC), dipalmitoylphosphatidylglycerol (DPPG), dioleoylphosphatidylethanolamine (DOPE),

palmitoyloleoylphosphatidylcholine (POPC) , palmitoyloleoylphosphatidylethanolamine

(POPE) and dioleoylphosphatidylethanolamine 4-(N-maleimidomethyl)cyclohexane-l- carboxylate (DOPE-mal). Additional non-phosphorous containing lipids are, e.g.,

stearylamine, dodecylamine, hexadecylamine, acetyl palmitate, glycerol ricinoleate, hexadecyl

stereate, isopropyl myristate, amphoteric acrylic polymers, triethanolamine-lauryl sulfate, alkyl-aryl sulfate polyethyloxylated fatty acid amides, dioctadecyldimethyl ammonium

bromide and the like.

Compounds of Formula I:

The lipids depicted in Formula I have a hydrocarbon backbone substituted with

heteroatoms which are sterically smaller, but equally or more flexible as the methylene group that they replace. This feature makes these new lipids fit more closely to the macromolecule to be delivered to the cells. This closer fit combined with the polycationic nature of the

backbone produces a tighter binding. In addition, these heteroatoms are hydrophilic; thus,

they not only confer an increased amphiphilic character to the lipids but also make the backbone more linear or "stretched" as compared to the all-methylene groups backbone. The latter being hydrophobic tends to wrap around itself in an aqueous environment, therefore

pulling the positively charged moieties away from the negatively charged phosphates on the

DNA/RNA backbone. This results in a weaker binding between the polycationic lipid backbone and the polyanionic DNA backbone, since the opposite charges can not align

properly in this arrangement. The hydrophilic backbone being linear allows for proper

alignment of the opposite charges, also leading to a tighter binding. In addition, the higher

hydrophilicity conferred by the heteroatoms on the polycationic backbone make possible the addition of more hydrophobic tails without loss of water solubility, thus making these

compounds more densely packed than compounds of the prior art. This is an advantage since

the same molar amount of lipid will produce a higher hydrophobic coating of the nucleic acid to be delivered. In fact, one of the preferred embodiments is compound 5a, as shown in

Figure 13, which has four hydrophobic tails and two positive charges (two tails per charge).

Some of the compounds herein described, such as compounds 13a and 13b (shown in

Figures 11 and 14) have fewer hydrophobic tails and a more hydrophobic backbone (no heteroatom substitution, but less hydrophobic overall). However, these compounds have moieties which mimic small natural biological effectors such as the neurotransmitters gamma

amino butyric acid (GAB A), acetylcholine etc. These moieties bind to their corresponding cell receptors targeting the delivery of nucleic acids to those cells rich in these type of receptors

such as muscle and neural cells. The latter type of cells are among the most difficult to

transfect since they are postmitotic cells (non-dividing) . These small chemical moieties do not perturb the ability of the lipids to form liposomes aggregates and at the same time confer more amphiphilic character to said lipids, since they are polar entities. Particularly preferred

embodiments are compound 13a and compound 13d, as shown in Figures 11 and 12.

Another novel feature of the compounds disclosed herein is the fact that, in addition to the traditional quaternary ammonium salts, guanidino and amidino moieties are used as

permanent positively charged centers. These functional groups are strongly basic and have the same charge as their ammonium counterparts, but have the advantage of being sterically smaller, since they are planar. Thus, they can get closer to the negatively charged phosphates

of the DNA/ RNA backbone producing a stronger binding interaction than that of

ammonium salts. Furthermore, these guanidinium and amidinium moieties have the ability to form hydrogen bridges with the nucleic acids' bases (guanidinium salts are used as

chaotropic agents to precipitate DNA) therefore they have an additional binding mode not

available to ammonium salts. Moreover, the guanidino moiety can also be used to target neural cells, since compounds such as Guanethidine, which possess such a functional group,

are internalized by neurons (Wiener, N. In, The Pharmacological Basis of Therapeutics, Gilman, A.G.; Goodman,L.S.; Rail, T.W.; and Murad, F.; Eds. Macmillan Pubs. Co. New York, 1985,

pp. 181-214.) . Thus, by including the lipidic content as well as the amine and guanidino moieties of Guanethidine in our novel liposome reagents we can target this difficult to transfect cell type. A particularly useful and preferred embodiment of these compounds is compound 6d, as shown in Figure 15, which is the most active of the compounds tested.

Additionally, reduced or no toxicity was observed for these lipids at the concentrations tested.

Compounds of Formula II:

Despite all the reasons given above in order to "rationally design" these lipids, it is still

impossible to predict their DNA transfection activity at this time. In fact, cationic lipids for

DNA transfection already in the market such DOTAP, DOTMA, DMRIE and DORI whose chemical structures are almost identical to that of the cationic lipid known as the Rosenthal

Inhibitor (Rosenthal, A.F. and Geyer, R.P. . Biol. Chem. 235(8):2202 (I960)) have significant

transfection activity, unlike the Rosenthal Inhibitor which is reported to be inactive as a DNA transfection reagent (see U.S. Pat No 5,264,618 to Feigner, P.L. et al.). The formulae for some

prior known compounds, as shown in WO 97/42819 (International Publication Date 20 November 1997) at the top of page 9, illustrate this point more clearly.

Compounds of formula II described herein also have a similar structure to that of the

Rosenthal Inhibitor. However, these compounds differ from the Rosenthal Inhibitor in that

a guanidinium or amidinium functionality is used as the positively charged anchoring group, and they also lack a quaternary ammonium group at the C-1 position of the glycerol

backbone. A preferred embodiment of these latter type of transfection reagents is compound

15, as shown in Figure 16.

An interesting feature of this compound is that it has the ability to form liposomes

without the need of co-adjuvants such as DOPE or DOPC. Thus, it can be used to form

liposomes with other cationic lipid compounds.

Specific Examples of the Invention.

Scheme I

Scheme I, as shown in Figure 1, shows the general synthetic route to prepare

polycationic lipids having a heteroatom substituted anchoring backbone. Thus, diglycolyl

chloride (1) was treated with a suitable primary or secondary amine (2a-d) in methylene

chloride in the presence of a base such as triethyl amine under an inert gas such as argon at room temperature to obtain the corresponding diglycolamides (3a-d). These amides were

reduced with lithium aluminum hydride (LAH) or can be reduced with borane in refluxing

anhydrous tetrahydrofurane (THF) to afford the corresponding amines (4a-d). Secondary

amine (4c) was easily converted to the corresponding tertiary amine (4e) upon treatment with acrylonitrile. Compound (4e) can be treated with ammonium chloride at high temperature

to produce the corresponding amidine(6e). Alternatively, the latter amidine derivatives can be obtained by reacting the dinitrile (4e) with anhydrous hydrogen chloride in ethanol, followed by treatment of the imidoester so obtained with ammonium hydroxide. Primary

amine (4d) was converted to the target compound (6d) by treatment with S-methyl

isothiouronium hydroiodide (S-methyl thiourea) in tetrahydrofurane in the presence of

triethylamine. Additionally, this guanidinium derivative can be alkylated with, for example, iodomethane to produce the corresponding quaternary ammonium salt. Tertiary amines

(4a,b,e) were treated with an alkylating agent such as iodomethane, iodoethyl acetate or 2-

bromo ethyl acetate to afford the quaternary ammonium salts (5a,b,f). The latter compounds

were also synthesized by treating the corresponding tertiary amines with the commercially

available 2-bromoethyl ether (lower panel, scheme I). This route has only two steps, but is not as flexible or prolific as the route depicted in the upper panel of Scheme I.

Scheme II

Compound (7) was readily synthesized, as shown in Figure 2, by treating commercially

available 1,4-diamino butane with acrylonitrile. Diamide-dinitrile (8) was then easily obtained

by treatment of compound (7) with an acyl halide such as palmitoyl chloride in methylene chloride in the presence of triethyl amine. The diamide-dinitrile (8) was reduced with lithium

aluminum hydride or can be reduced with borane in THF to the corresponding tertiary and

primary amines functionalities to afford compound (9). Guanidinium compound (10) can be

obtained in a similar fashion as shown in Scheme I for compound (6d) by reacting the primary

amines of compound (9) with S-methyl thiourea in THF and triethylamine. Scheme HI

Referring to Figure 3, tetrapalmyl spermine (11) (Haces, A. et al. PCT Int. Pub. No

WO 95/17373) was treated with ethyl iodoacetate at room temperature to afford the tetraalkylated derivative (13 a). Similarly, compound (11) was treated with 2-bromoethyl

acetate at high temperature to afford (13b). Reaction of (11) with 4-bromo or 4- chloro butyryl

chloride in methylene chloride in the presence of triethyl amine at low temperature gave the corresponding 4-bromobutyramide derivative (12). Intermediate (12) was immediately treated

with iodomethane to produce the N,'N"- dimethylated intermediate (13c) which in turn was

treated with an excess of ammonium hydroxide in tetrahydrofurane at elevated temperature to convert the bromide (or chloride) into the corresponding primary amine (13d). All of these

compounds have moieties which resemble or mimic the neurotransmitters gamma

aminobutyric acid (GABA, compound (13d)) and acetylcholine (compounds (13a,b)). These small groups do not change substantially the liposomal forming ability of the lipid molecule

and at the same time are capable of being recognized by neural or muscles cells. This

preferential recognition by these type of cells makes these lipids target specific DNA/RNA

delivery agents.

Scheme IV

Referring to Figure 4, commercially available dioleoylphosphatidylethanolamine (14) was treated with an excess of S-methyl isothiouronium hydroiodide in tetrahydrofurane and

in the presence of triethylamine to afford the corresponding guanidinium compound (15). EXAMPLES

Example 1: Synthesis of octadecylcyanoethylamine (2d).

Octadecylamine (2g, 7.4 mmol) and acrylonitrile (15ml) were heated for 3h at 70 C

in a thick wall test tube capped with a teflon lined cap. TLC (silica gel; ethyl acetate) shows a new spot at rf = 0.65 . The excess acrylonitrile was removed in vacuo to afford pure

product (2.40g, 100% yield). H-NMR (CDCI3) : 50.88 (t, 3H), 2.25 (br.s.,32H), 2.53 (t, 2H), 2.62 (t,2H), 2.93 (t,2H). FT-IR (cm^"1 ) 2250 (CN).

Example 2: Synthesis of bis (mono and dialkyl)diglycolamides(3a-d), general procedure.

To a solution of dialkylamine ( 2mmol) and triethylamine (2 mmol) in methylene

chloride (250ml) under argon was added diglycolyl chloride (lmmol) and the resulting solution was stirred for 18h at room temperature. TLC (silica gel; MeOH or CH2CL2/THF,3:1) shows the absence of starting material and a new spot. The methylene

chloride solution was washed with sodium bicarbonate (10% in water), dried -SI^SC^) and

the solvent removed, to afford the desired diamide.

Proceeding as described before and using the appropriate mono or dialkylamine the

following compounds were prepared:

3a. Bis (dioctadecyl) diglycolamide (81% yield), H-NMR (CDCL3) δ: 0.88 (t, 12H), 1.25 (s,

124H), 1.5 (br.s, 8H), 3.15(t,4H), 2.9 (t,4H), 3.2 (s, 4H). R (cm^"1): 1651 (C-O);

3b. Bis (didecyl)diglycolamide (73% yield), H-NMR(CDCL3) δ X: 0.85 (t,12H), 1.25 (s, 64H), 1.5 (br.s,8H), 3.18 (t,4H), 3.3 (t,4H), 4.4 (s, 4H). FT-IR^m^"1): 1657(C = 0);

3c. Bis (octadecyl)diglycolamide (100% yield),H- NMR(CDCL3) δ: 0.85 (t, 6H), 1.25 (br.s, 60H), 1.5 (br.s,4H), 3.3 (q,4H), 4.05(s,4H), 6.4 (br.s,2H);

3d. Bis (octadecylcyanoethyl) diglycolamide (98%yield), H-NMR (CDCI3) δ : 0.85 (br.t, 6H), 1.1-1.6 (br.s, 64H), 2.65 (t, 4H), 3.25 (br.t, 4H), 3.55 (br.t, 4H), 4.3 (s, 4H);

Example 3: Synthesis of N.N.N'.N'-Mono and dialkyl 2,2'-oxybis ethylamines (4a-d), general procedure

To a solution of lithium aluminum hydride ( 6 to 64 molar excess) in dry tetrahydrofurane (THF) was added the corresponding diamide in small portions under a

blanket of argon. The resulting mixture was refluxed for two to three days under argon. The progress of the reaction was followed by TLC (silica gel; CH2CI2/TFH, 3:1 for

dialkylamides; 10% triethylamine in CH2CI2 for monoalkylamides and triethylamine for cyanoethylamides) . The reactions were quenched with sodium hydroxide (10% in water) . The

mixture was filtered, the filtrate dried (-S^SO^.), and the solvent removed in vacuo to afford

the desired products.

Proceeding as described above and using the appropriate diamide the following

compounds were obtained:

4a. N,N,N',N'-dioctadecyl-2,2'-oxobis ethylamine (77% yield), H-NMR(CDC1₃) δ: 0.88 (t, 12H), 1.25 and 1.43 (br.s., 128H), 2.42 (t, 4H), 2.62 (t, 4H), 3.48 (t, 4H), 3.7, (t,4H); 4b. N,N,N',N'-didecyl-2,2'-oxobis ethylamine (93% yield ), H-NMR (CDCL₃) δ: 0.88 (t,12H), 1.25 and 1.42 (br.s, 64H), 2.42 (t, 8H), 2.63 (t, 4H), 3.5 (t, 4H);

4c. N,N' ,-octadecyl-2,2'-oxobis ethylamine (71% yield), H-NMR (CDCI3) δ: 0.87 (t, 6H), 1.25 and 1.45 (br.s., 64H), 2.6 (t,4H), 2.78 (t, 4H), 3.65 (t,4H);

4d. N,N,N',N'-octadecylaminopropyl-2,2'-oxobis ethylamine (80% yield), H-NMR (CDCL3) δ:0.88 (t,6H), 1.25 and 1.45 (br.s,70H), 2.4-2.8 (br.m,16H), 3.4-3.7 (br.m, 4H).

Example 4: Synthesis of N.N.N,N' M'.N' - dioctadecylmethyl-2,2'-oxybisethylammonium iodide. (5a).

N,N,N',N'-dioctadecyl-2,2'-oxy bis ethylamine (19mg, 0.017mmol) was dissolved in iodomethane (1ml) inside a capped thick-wall test tube, and the resulting solution heated for

20h at 75°C . TLC (silicagel ; chloroform:acetone:methanol:water; 50:15:5:5:1) shows only

one spot at Rf = 0.8 , which gives a negative ninhydrin test and no starting material. The excess iodomethane was removed in vacuo to afford the desired product (23 mg, 96%). H-

NMR(CDC1₃) δ: 0.88 (t,12H), 1.15 - 1.5 and 1.7 (br.s.,128H), 3.35 (s, 6H), 3.46 (br.m.,8H),

3.88 (br.s.,4H), 4.28 (br.s., 4H). Proceeding in a similar fashion as per compound (5a) ,

compound (5b) was obtained in 100% yield.

Example 5: Synthesis of N,N.N'.N'-cyanoethyloctadecyl-2.2'-oxybis ethylamine (4e .

A suspension of N,N'-octadecyl-2,2'-oxybis ethylamine (100mg,0.16mmol) in

acrylonitrile (4ml) was heated for 18h at 80 C in a capped, thick-wall test tube (the initially liquid two phase system became a clear homogenous solution after 2h). TLC (silicagel;dichloromethane/ THF, 3:1) shows the absence of starting material and a spot

corresponding to desired material at rf = 0.95. The excess acrylonitrile was removed in vacuo

to afford the desired material. H-NMR(CDC1₃) δ: 0.88 (t, 6H), 1.25 (br.s., 64H), 2.45 (2t,8H), 2.68 (t, 4H), 2.87 (t,4H), 3.5 (t, 4H).

Example 6: Synthesis of N.N,N,N',N'.N'-cyanoethyloctadecylmethyl-2,2'-oxybis ethyl ammonium iodide (5e).

A solution of N,N,N',N'-cyanoethyloctadecyl-2,2'-oxybis ethylamine (60 mg,0.08 mmol) in iodomethane (1.5ml) was heated for 3h at 80 C in a capped, thick wall tested tube.

The excess iodomethane was removed in vacuo to afford desired product. H-NMR (CDCI3)

δ: 0.88 (t, 6H), 1.25 (br.S., 60H), 1.8 (br.s., 4H), 3.45 (br.s., 6H), 3.65 (br.s., 4H), 3.95-4.4

(br.m., 16H).

Example 7: Synthesis of N,N.N,N'.N'.N'- acetoxyethyldioctadecyl-2.2'-oxybis ethyl ammonium iodide(5f).

A solution of N,N,N',N'-dioctadecyl-2,2'- oxybis ethylamine (16mg, 14.3 mmol) and

ethyl iodoacetate (0.5ml) in chloroform (1ml) was heated for 18h at 75 C in a capped thick wall test tube. The chloroform was removed in vacuo and the residue dissolved in tetrahydrofurane (10ml). To this solution was added thiourea and the mixture stirred at room

temperature until no more thiourea went into solution (stoichiometric excess after all iodide

is converted to the isothiouronium salt). The mixture was heated for 2h at 70 C and the

excess solvent removed in vacuo. The residue was then redissolved in dichloromethane (10ml) and the solution washed with water (4x 5 ml), dried (Na2Sθ4), filtered and the solvent removed in vacuo to afford 10 mg of desired product.

Example 8: Synthesis of S-methylisothiouronium hydroiodide.

A solution of thiourea (0.8g,10.5mmol) and iodomethane (8.8g,62mmol) in methanol (25ml) was heated in a capped thick-wall test tube for 4.5h at 50 C. The reaction mixture was

rotaevaporated to afford pure desired material in 100% yield. H-NMR (CD₃OD) δ: 2.62(s,

3H), 4.8 (br.s., 4H).

Example 9: Synthesis of N.N.N'-N'-guanidinopropyloctadecyl oxy bis-2, 2 '-ethylamine hydroiodide (6d).

A solution of N,N,N',N'-aminopropyloctadecyl-2,2'- oxybis ethylamine (lOOmg, 0.14

mmol) , S- methyl isothiouronium hydroiodide (300mg, 1.3 mmol) and triethylamine

(300mg, 3 mmol) in tetrahydrofurane (10ml) were heated in an argon flushed, capped thick- wall test tube for 20h at 95°C. The solvent and methyl mercaptan byproduct were removed

in vacuo in a chemical fumes hood and the residue dissolved in methylene chloride (30ml), the organic phase was washed with brine ( 3x, 20 ml), water (2x 10ml), dried ( ^SO^, filtered and the solvent removed to obtain a reddish solid (100mg,80% ). H-NMR(CDCl₃) δ: 0.86

(t,6H), 1.1-1.6 (br.s., 64H), 2.4-2.8 (br.m., 16H), 3.15-3.7 (br.m., 12H). FTIR (cm^"1): 1653

(C=NH).

Example 10: Synthesis of N.N'-cyanoethyl-l,4-diaminobutane (7).

1,4-diamino butane (2g, 22 mmol) was cooled to 0C (ice bath) and to this solid was

added acrylonitrile (4ml) . The mixture was let reach room temperature slowly (ca 30min) and

then let react for additional 18h at room temperature with stirring. The excess acrylonitrile was removed in vacuo to afford the desired product. H-NMR (CDC1 ) δ: 1.3 (br.s.,2H), 1.5 (br.s, 4H), 2.5 (t, 4H), 2.65 (br.m.,4H), 3.9 (t,4H). FTIR (cm^"1) : 2247 (C = N).

Example 11: Synthesis of N,N,N',N'-cyanoethylpamitoyl-l,4-diaminobutane(8).

To a solution of N,N'-cyanoethyl-l,4-diaminobutane (0.714g, 3.68mmol) and triethylamine (0.744g, 7.36mmol) in dichloromethane (150ml) was slowly added palmitoyl

chloride (2.02g, 7.36 mmol) and the resulting mixture let react at room temperature overnight.

The reaction was washed with sodium bicarbonate (10%, 2x 50ml), water (2x 50ml), dried (Na2Sθ4) , filtered and the solvent removed in vacuo to afford the desired product (2.3g, 93 %). H-NMR (CDCL₃) δ: 1.9 (t,6H), 1.25 and 1.6 (br.s., 56H), 2.32 (m,4H), 2.72 (t, 4H), 3.42

(m, 4H), 3.55 (t,4H). FTIR (cm^"1) 1643 (C=0).

Example 12: Synthesis of N.N.N\N'-aminopropylpalmyl-l,4-diaminobutane (9).

To a solution of lithium aluminum hydride (600mg, 15.9 mmol) in tetrahydrofurane

(50ml) was added N,N,N',N'-cyanoethylpalmitoyl-l,4-diaminobutane (300mg, 0.45 mmol) and the reaction mixture was refluxed for 72 h. Then, a procedure essentially the same as per

example 3 (supra) was followed to afford desired diamine (200mg,70%). H-NMR(CDC1 ) δ:

0.88 (t,6H), 1.25 and 1.5 (br.s, 64H), 2.3-2.7 (br.m, overlap, t, 16H).

Example 13: Synthesis of N.N.N',N'-guanidinopropylpalmyl-1.4-diaminobutane (10).

A procedure identical as per example 6 (supra) was followed. H-NMR (CDC1₃) δ: 0.9

(t, 6H), 1.2-1.35 (br.s., 68H), 1.5-2.0 (br.m., 4H), 2.65-3.5 (m, 24H). FTIR (cm¹) : 1650 (C-N). Example 14: Synthesis of N.N' ' '-4-bromobutyryl-N,N'.N"N' "-tetrapalmylspermine (12). To a cooled (0 °C) solution of N,N' ,N" ,N' "-tetrapalmylspermine (400mg, 0.36 mmol)

and triethylamine (80mg, 0.8 mmol) in dichloromethane (14ml) was added 4-bromobutyryl

chloride (156mg, 0.8 mmol) and the resulting mixture was allowed to react for lh at 0°C. The reaction was quenched and washed with cold sodium bicarbonate solution (10 % in water, 3x

5ml), dried (sodium sulfate), filtered and the solvent removed at room temperature in vacuo

to afford a white foam. H-NMR (CDC1₃) δ: 0.88 (t, 12H), 1.15-1.5 (br.s., 136H), 2.12 (t, 4H), 2.3-2.55 (br.m.,12H), 3.2-3.36 (br.m., 8H), 3.62 (t, 4H).

Example 15: Synthesis of N,N'.N".N'"-tetrapalmyltetraacetoxyethylspermine iodide salt i3a): A solution of tetrapalmylspermine (Haces et al.,PCT Int. Pub. No WO/95/ 17373),

130mg, mmol) in neat ethyl iodoacetate (1.5ml) was heated to 75°C for 18h. The reaction was worked up following essentially the same procedure as per example 13 (supra) to afford the

desired product.

Example 16: Synthesis of N.N"'-4-bromobutyryl-N.N'.N".N"'-tetrapalmyl-N'.N"- dimethylspermine (13 c) .

N,N' ' ' -4-bromobutyryl-N,N' ,N" ,N' ' '-tetrapalmylspermine (350mg,0.25 mmol) was dissolved in iodomethane (3ml) and the resulting solution let react for 2 days at room

temperature. Excess iodomethane was evaporated to afford desired product, which is negative

for ninhydrin test. H-NMR (CDC1₃) δ: 0.88 (t, 12H), 3.41 (br.s., 6H).

Example 17: Synthesis of N.N'"-4-aminobutyryl-N.N'.N".N' "-tetrapalmyl-N'.N"- dimethy lspermine (13d).

To a solution of N,N^,"-4-bromobutyryl-N,N',N",N'"-tetrapalmyl-N',N"- dimethylspermine (100mg,0.05 mmol) in tetrahydrofurane (10ml) was added ammonium hydroxide (20 ml, 28% by weight) and the resulting mixture heated to 70 C for 2 days in a

capped, thick wall reaction tube. The solvent was azeotropically evaporated (ethanol) to

afford a brown solid which is strongly positive for ninhydrin test. H-NMR (CDC1 ) δ: 0.88

(t,12H), 1.1-1.4 (br.s,130H),1.5-2.2 (br.m,16H), 3.0-3.8 (br.m.lOH), 3.4 (br.s, 6H).

Example 18: Synthesis of dioleoylphosphatidyl ethanolguanidine (15).

To a solution of dioleolylphosphatidyl ethanoamine (70mg, 0.094 mmol) and triethylamine (1ml) in tetrahydrofurane (10 ml) was added S-methylisothiouronium hydroiodide (70mg, 0.32 mmol) and the resulting solution was heated for 18h at 70 C. The

solvent was removed in vacuo and the residue redissolved in dichloromethane (25ml). This solution was washed with water (2 x 10 ml), dried ( sodium sulfate), filtered and the solvent

evaporated to afford the desired product (40 mg, 47%). H-NMR (CDC1₃) δ 0.88 (t,6H), 1.2-

1.44 (br.s, 40H), 2.00 (br.d,10H), 2.29 (t, 4H), 3.19 (q,lH), 3.42 (br.s,lH), 3.85-4.20 (br.m,2H),

4.38 (br.m,lH), 5.3 (br.d, 4H). FTIR (cm¹) : 2361, 1741.

Example 19: Liposomes formulation:

Lipids were formulated by mixing appropriate molar amounts of the active lipid

(compounds 5a, 6d and 13d as shown in Tables I and II) with a colipid dioleoylphosphatidyl

ethanolamine (DOPE) in dichloromethane and dispersing this mixture in the final amount of

water using the solvent vaporization method. (David W. Deamer, in Liposome Technology, vol.I, p-29, CRC press Boca Raton, Fl, 1984).

Cell culture and plasmids

Cell lines were from the American Type Culture Collection (Rockville, Maryland) and

were cultured in RPMIl 1649, 10% FCS, pen/strep . Plasmid pCMV a-gal, which contains the

E. Coli a-galactosidase (gene) under the control of the powerful cytomegalovirus promoter (McGregor et al. (1989) Nucleic Acids Res., 17: 2365) was purchased from Clontech, Inc.

Primary cells were from human tracheal isolates and neonatal foreskin.

Example 20: Transfection of HepG2 and HeLa cells.

Cells were plated in 48-well plates (1cm² ) at a concentration of lx 10 cells/ well in 0.5 ml of RPMI-1640, 10% FCS, Pen/step. The next day, lipids aliquats (1,3 and 5μl of 1 mg/ml liposome in water) were diluted in polystyrene tubes containing lOOμl of serum-free,

antibiotic-free RPMI-1640 and to these tubes were added 150 ng of plasmid in lOOμl of the

same medium (suboptimal amount in polypropylene tubes) and incubate for 15 min. The cells were washed twice with Dulbecco's PBS, the lipid:plasmid complexes added to them and then incubated for 7h at 37 C in 5% CO2 atmosphere. Growth medium was added to

the cells for a final volume of 1 ml and a final concentration of 10% FCS, pen/strep, 50μg/ml

gentamicin in RPMI-1640 and were incubated overnight. The results are shown in Tables I

and II and discussed hereinafter.

Example 21: Transfection of Primary Human Tracheobronchial and Epidermal

Keratinocvtes. Cells were grown in serum free medium (SFM) and plated on 35 mm plates (6 wells) such that the confluence after 24h was above 50%. Plasmid reporter (2 μg and 5 μg, respectively) was mixed with variable amounts of liposomes (see tables TV and V) and the

complex formed added to the cells. The cells were transfected during 4h and 6h, respectively.

The DNA/liposome complex was removed by rinsing with SFM and the cells incubated for 48h under normal growth conditions and then assayed for the appropiate marker.

Example 22: Transfection and CAT assay of Turkat Cells (suspension cells).

The cell suspension culture was transferred to a 50 ml conical tube and centrifuged at 400g for 10 min. The cells were washed twice by aspirating off the supernatant and gently

resuspending the cell pellet in 25 ml of sterile PBS and centrifuging again at 400g for lOmin. The pellet was resuspended in a volume of serum-free growth medium such that a final

concentration of 6.25 x 10⁶ cells /ml was obtained (about 10 ml). 35 mm cell culture plates were inoculated with 0.8 ml of the cell suspension. For each well, 10 μg of CAT plasmid

were dissolved in 110 μl of serum-free medium and a separately in another tube were diluted

30 μl of the lipid solution in 70 μl of serum-free medium. The plasmid and lipid solution were mixed and gently swirled and allowed to stand at room temperature for 10 min. The complex DNA/lipid solution was then randomly dropped over the culture well. The wells

were gently swirled and then incubated at 37 ° C under a 7 % C0₂ atmosphere for 5 hours.

After 5h incubation, 4ml of 12.5 % FBS growth medium were added to the wells and the

incubation continued for additional 72 h under the above conditions. The cells were then

transferred to 10 ml Falcon tubes and the wells rinsed with 5ml of sterile PBS. The cell

suspension was washed twice with 5 ml of sterile PBS as previously. The final pellet was resuspended in 400 μl of lysis buffer and transferred to 1.5 ml centrifugation tubes. The tubes were capped and placed horizontally on a rocker and the cells lysated for 30 min.

100 μl aliquots were then assayed for CAT activity following the procedure of

Neumann et al. (1987) Biotechniques 5: 444. The results are shown in Table HI and discussed hereinafter.

Example 23: Assay for transient transfection (adherent cells).

The cells were washed twice with Dulbecco's PBS and stained with freshly prepared

fixative (2% formaldehyde/ 0.2% glutaraldehyde in PBS) for 5 min, washed twice with

Dulbecco PBS) . Then, the cells were stained with 0.5 ml of β-galactosidase histochemical stain (0.1% x-gal, 5mM potassium ferrocyanide, 5mM potassium ferrocyanide, 2mM MgCl₂ in PBS)

for 24h at 37 C in a 5% CO2 atmosphere. Blue cells ( β - gal positive) were counted.

Example 24: Results and Discussion.

The results are summarized in Tables I , II, III, IN and N. Tables I and II show the

relative transfection efficiency of compounds 5a, 6d and 13d versus control compound

TMTPS (Compound 3 in PCT Int. Pub. No. W095/ 17373. Haces, A. et al.) in HepG2 (human hepatocarcinoma) and HeLaS3 (human cervical carcinoma) cells, respectively, and

under suboptimal conditions for activity. In these cell lines, compounds 6d and 13d show a

2-2.4 fold higher efficiency than the TMTPS control, and compound 5a is half as active as the

control in HepG2 cells and showed negligible activity in HeLaS3 cells. Table -H shows an

analogous comparison using Jurkat cells (T-cell leukemia). In this experiment, compounds 5a

and 13d show similar efficiency as TMTPS, but compound 6d shows almost 38% more activity than that of the control. Tables IV and V shows the relative efficiency of

compounds 6d and 13d in the primary human tracheobronchial epithelial and human keratinocytes cells. Primary cells are cells that are freshly isolated from humans or animals and which, unlike the cultured cell lines, reflect the potential behavior of a compound in vivo

more closely. Thus, for genetic therapy to work, it is necessary to be able to transfect these

types of cell lines before any in vivo experiments are tried. These types of cells are also the most difficult to transfect and their transfection efficiencies are usually below 1%. Table IV

shows the relative efficiency of compounds 6d and 13d versus DOTMA (Lipofectin™

Reagent, Life Technologies, Inc., supra) in primary human tracheobrochial cells. Both of these compounds show a relative range of activities of 5.3 to 6.0 times higher than that of the

Lipofectin control. At the same time their cell toxicity was below 5%, unlike the control

which showed toxicity in the 10-20% range. Thus, these lipid reagents are superior to the commercial standards in both respects. In addition, this is a very significant result since

tracheobronchial cells are involved in the genetic disease cystic fibrosis. There are several

genetic therapy clinical trials being conducted at the present time targeting these cells using either viral or liposomal vectors (see 10th Annual North America Cystic Fibrosis Conference, Orlando, FI., Oct 24-27 (1996), Abstracts or Pediatr Pulmonol Suppl, 13: 74-365,

Sept, 1996 ) .

Table V depicts the the percentages of β-gal positive cells (absolute number) which

were obtained in primary human epidermal keratinocytes with compounds 5 a and 6d versus that obtained with DOSPA control (Lipofectamine™ Reagent, Life Technologies, Inc. supra).

Compounds 5a and 6d gave, respectively, 35% and 50% positive cells as compared with 2%

positives for the control. This represents a 15-25 fold better efficiency for these novel liposome reagents when compared with this well known standard. Moreover, primary human

keratinocytes are also a potential target cells for genetic therapy (Fenjves, E.S. et al., Hum

Gene Ther 5: 10,1241-8, Oct. 1994), but its use has been restricted due to the lack of highly efficient transfection vectors.

TABLE I TRANSFECTION OF HEP G2 CELLS

Lipid/DOPE (molar ratio) Optimal Liposome Amount β- Gal Positive Cells (%)

Compound 5a (1: 1.8) 3 /xg 0.8

Compound 6d (1: 1.5) μg 4.3

Compound 13d (1:1.5) 3 μg 2.9

Control TMTPS/DOPE (1:1.5) 5 μ 1.5

Cells were plated in 48 well plates at a density of 1 x 10 per well in 0.5 ml of growth medium. After 24h, the cells were washed with serum free medium and transfected with a suboptimal amount (150 ng) of plasmid pCMV-βgal. using 1,3 and 5μl (1,3 and 5μg) of lipid formulation. The amount giving the highest level of transfection efficiency is shown. The experiment was run in triplicate.

TABLE II TRANSFECTION OF He La S3 CELLS

Lipid/DOPE (molar ratio) Optimal Liposome Amount β Gal Positive Cells (%)

(Mg)

Compound 5a (1:1.8) 5 μg 0.001

Compound 6d (1:1.5) 3 μg 2.4

Compound 13d (1:1.5) lμg 2.1 Control TMTPS/DOPE 5 μg 0.9

(1:1.5)

Cells were plated in 48 well plates at a density of 1 x 10⁵ per well in 0.5 ml of growth medium. After 24h, the cells were washed with serum free medium and transfected with a suboptimal amount (150 ng) of plasmid pCMV- βgal. using 1,3 and 5μl (1,3 and 5μg) of lipid formulation. The amount giving the highest level of transfection efficiency is shown. The experiment was run in triplicate.

TABLE III TRANSFECTION OF JURKAT CELLS

Lipid/DOPE (molar ratio) Optimal Liposome Amount CAT Activity (mU/well) _g)

Compound 5a (1:1.8) 30 μg 196.00

Compound 6d (1:1.5) 30 μg 298.20

Compound 13d(l:1.5) 30 μg 220.00

Control 30 μg 216.44 TMTPS/DOPE(l:1.5)

Wells were inoculated with 6.25 x 10⁶ cells. 10 μg of CAT plasmid were mixed with 30μg (optimal amount known for the control) of the lipids and then added to the cells. After 5h the transfection was quenched with FBS containing medium and the cells incubated for 72h. Cells were lysated in 400 μl of buffer. 100 μl aliquots were assayed for CAT activity.

TABLE IV TRANSFECTION OF PRIMARY HUMAN TRACHEOBRONCHIAL EPITHELIAL CELLS

Lipid/DOPE (molar ratio) Optimal Liposome Amount Luciferase Activity (counts) ϋ)

Compound 6d (1:1.5) 12 μg 7297022

Compound 13d (1:1.5) 12 μg 8343975 Control , Lipofectin™ * 6 μg 1379341

35 mm plates were inoculated with cell isolates and transfected at 90% confluence. 2μg of firefly luciferase plasmid were mixed with 12 μg of the lipids and then added to the cells. After 5h, the transfection was quenched by removal of the DNA/Lipid complex and the cells incubated for 72h. Cells were lysated and aliquats assayed for luciferase activity. Cell toxicity, determined by the trypan blue method, was below 5% for lipids 6d and 13d and between 10-20% for Lipofectin.

* Lipofectamine was also run as a control, but its efficiency was negligible.

TABLE V TRANSFECTION OF PRIMARY HUMAN EPIDERMAL KERATINOCYTES

Lipid/DOPE (molar ratio) Optimal Liposome Amount β-Gal Positive Cells (%) (Mg)

Compound 5a (1:1.5) 40 μg 35 %

Compound 6d (1:1.5) 20 μg 50 %

Control , Lipofectamine™ 25 μg 2 %

Cells were seeded at 2xl0⁵ /well in 35 mm wells and transfected the next day. 5μg of βgal DNA were mixed with the appropiate amount of lipids and added to the cells. After 4h, the medium was replaced and the cells incubated for additional 48h and then assayed. Blue cells were observed under the microscope and counted.

Polyether Backbone Embodiments

In a further embodiment of the present invention, another series of lipids are

provided, as disclosed herein. These lipids have a polyether backbone which generally

makes them more prone to line-up with the negatively charged polyphosphate backbone of

nucleic acids. In addition, the polyether functionality, which is also present in

polyethylene glycol (PEG), a known cell-fusogenic agent, may contribute to the transfection activity of these novel polycationic lipids.

The present lipids have shown to be very effective transfection reagents with

cultured cells, especially with the very useful COS and CHO cell lines. These reagents also work with other cells lines such as NIH3T3. The synthesis of a preferred polycationic lipid having a polyether backbone is described below.

Example 25

Synthesis of triethyleneglycoldipalmitoyldiamide [compound 21]

To a solution of triethylene glycol diamine (Ig, 6.74 mmol) (compound 20) and

triethylamine (1.4g, 13.81 mmol) in dichloromethane ( 400ml) at room temperature under

argon was added (dropwise) palmitoyl chloride (3.79g, 13.81mmol) and the resulting

mixture stirred overnight at room temperature. Since the product is partly soluble in dichloromethane, but soluble in chloroform, the dichloromethane was removed in vacuo

to about one fifth of the initial volume and chloroform (200ml) was added. The organic layer was washed with 10 % aqueous sodium bicarbonate solution (3x, 50 ml), dried (Na2S04), filtered, and the solvent removed in vacuo to afford the desired diamide (3.9, 92%).

Synthesis of N.N=-dipalmityl-triethyleneglycoldiamine (compound 22]

A solution of triethyleneglycoldipalmitoyldiamide (lg, l.όmmol) and lithium

aluminun hydride (0.3 g, 8.0 mmol) in dry tetrahydrofurane (200ml) was refluxed for three

days under argon. The cooled reaction mixture was treated with 5% sodium hydroxide

solution to destroy the excess lithium aluminun hydride. The reaction mixture was then filtered, dried (Na2S04) and the solvent removed to afford pure product (0.8g, 84%).

Synthesis of N,N,N = .N=-dimethyldipalmityl-triethyleneglycoldiamine [compound 23]

N,N=-dipalmityl-triethyleneglycoldiamine (240mg, 0.4mmol) was dissolved in an

excess of iodomethane (3ml) and the solution allowed to react for one day at room temperature in the dark. The unreacted iodomethane was removed under vaccum and the

residue dissolved in ethanol (3 ml). To this solution was added a solution of potassium

hydroxide (3ml,saturated) in ethanol. The red residue solution turned clear and colorless. The new solution was diluted with dichloromethane (50ml) and washed immediately with aqueous sodium bicarbonate solution (3x 20ml). The organic layer was dried (NaS0₄) and

the solvent removed to afford the desired compound (250mg,99%).

Synthesis of N.N.N.N = ,N = .N = -tetramethyldipalmityl-triethyleneglycoldiamine

[compound 241

Compound 23 (250mg, 0.4mmol) was dissolved in an excess of iodomethane (3ml)

and the solution allowed to react overnight at room temperature. The excess reagent was removed under vacuum to produce the the desired tetraamnonium salt (358mg, 99%). This

material was formulated with a colipid dioleoylphosphatidylethanolamine (DOPE) as

discussed above, and as disclosed in International Application No. PCT/US97/ 09093,

International Publication No. WO 97/42819, at a 1:1.5 lipid/DOPE molar ratio.

As indicated above and as shown specifically in Figure 5, commercially available

glycol amine (compound 20), for which n is 2, was reacted with palmitoyl chloride (Rj

= C₁₅H₃₁) in the presence of triethylamine to produce compound 21 in a 92% yield. The resulting compound was reduced with lithium aluminum hydride to produce the

secondary amine (compound 22) in an 84% yield. Sequential iodomethylation, followed by free base liberation and methylation of the intermediate tertiary amine (compound 23)

produced the desired lipid (compound 24) in a 98% overall yield. Compound 24 was

formulated into liposomes in essentially the same manner as described above in connection with Example 19.

TABLE VI TRANSFECTION OF COS 1 CELLS

Lipid/DOPE (molar ratio) Liposome Amount (μg) β-Gal Positive Cells (%)

Compound 24: DOPE, 1:1.5 2 μg less than 5%

Compound 24: DOPE, 1:1.5 6 μg 10 - 15%

Compound 24: DOPE, 1:1.5 8 μg 50 - 80%

Compound 24: DOPE, 1:1.5 10 μg 15 - 20%

Cells were plated in 35 mm well plates at a density of 1 x 10 per well in 1 ml of growth medium. After 24 h (ca 70% confluence), the cells were washed with serum free medium and transfected with 2 μ of plasmid pCMV-β gal. Cells were assayed 48 h after transfection for the β gal gene protein.

TABLE VII TRANSFECTION OF CHO CELLS

Lipid/DOPE (molar ratio) DNA/Liposomes β Gal Positive Cells (%)

Amounts (μg)

Compound 24: DOPE, 1:1.5 b4 less than 10%

Compound 24: DOPE, 1:1.5 2:8 20-25%

Compound 24: DOPE, 1:1.5 3:12 30-70% Compound 24: DOPE, 1:1.5 4:16 30-35%

Cells were transfected with the plasmid pCMV-β gal in 35 mm well plates at 80% confluence in 1 ml of growth medium. After 24 h, the cells were assayed for the expression of the β gal gene protein.

Example 26

The liposomes produced in accordance with Example 25 were used to transfect

COS1 cells and CHO cells following essentially the same procedures described in connection with Examples 20-22 using the amounts of liposomes and slight changes to the procedure as shown in connection with Tables VI and VII respectively.

As shown in Tables VI and VII, as much as 80% of the COS 1 cells and up to 70%

of the CHO cells were transfected employing compound 24 as a transfecting agent.

Glyceryl Backbone Embodiments

In further embodiments of the invention, lipids having a glyceryl backbone are

provided. These glyceryl backbone based lipids have a dicationic bicyclic triethylene diammonium polar head. The high charge density makes the binding of this class of lipids

to DNA molecules stronger than that of the traditional monocationic liposomes for DNA

transfection such as DOTMA (Lipofectin reagent™, Life Technologies, Inc.).

These glyceryl backbone based lipids have shown to be very effective transfection

reagents with cultured cells, particularly with COS cells.

Scheme V of Figure 5 shows the general synthesis of glyceryl backbone

embodiments, which contain the previously unreported triethylenediammonium bicyclic polar head. As shown in the figure, diol (16) was acylated with palmitoyl chloride (R =

C₁₅H₃₁) in the presence of triethylamine to afford the diester (17) in a 59% yield.

Treatment of the bromide (17) with an excess of the very nucleophilic triethylenediamine (trade name Dabco^® ) gave the desired lipid (18) in a 44% yield. Further methylation of

compound (18) with an excess of iodomethane produced diquaternary amine lipid (19). The corresponding ethers of these class of compounds are made by simply using alkyl

iodides or activated esters, such as tosylates or triflates, instead of the acyl halides in the first step. Thereafter, the synthetic route is identical for both types of analogs.

Example 26

Synthesis of 3-bromo-1.2- palmityl propanediol [compound 26]

To a solution of 3-bromo-l,2-propanediol (500 mg, 3.22 mmol) and triethylamine (653 mg, 6.45 mmol) in dichloromethane (20ml) at 0°C under argon was added (dropwise)

palmitoyl chloride (1.77g, 6.45 mmol) and the resulting cloudy mixture stirred for 3h at

room temperature. Then, a 20% excess of reagents was added and the reaction stirred overnight at room temperature. The reaction was diluted with dichloromethane (50ml)

quenched and washed with 10% aqueous sodium bicarbonate (3x 10ml), dried (Na2Sθ4), filtered, and the solvent removed in vacuo to afford impure desired diester. The crude

product was purified by silica gel column chromatography (dichloromethane eluent) to

afford pure product (1.2g, 59%)

Synthesis of 1.2-palmityl-3-triethylenediammonium propanediol [compound 27]

A solution of 3-bromo-l,2-palmitylpropanediol (50mg, 0.07 mmol) and triethylenediamine (Dabco ^®, 150mg, 1.33 mmol) in dry dimethyl formamide (3ml) was

heated for 12h at 70 °C. The reaction mixture was diluted with dichloromethane (15ml) and washed with water (5x, 5ml). The organic layer was dried (Na₂S0₄), filtered and the

solvent removed in vacuo to afford the desired material (26mg, 44%). This material was formulated with a colipid dioleoyl phosphatidyl ethanolamine (DOPE) at a 1:1.5., Lipid:DOPE, respectively.

As indicated above and as shown in Figure 6, compound 26 is produced by acylating the starting diol with palmitoyl chloride in the presence of triethylamine to

produce the diester in a 59% yield. The diester is treated with an excess of the highly

nucleophilic triethylene diamine (Dabco ^®) which gives the desired lipid (compound 27) in a 44% yield. Further methylation with an excess of iodomethane produced a diquaternary amine lipid (compound 28). The corresponding ethers are made by using alkyl iodides

(e.g. methyl iodide or activated esters (e.g. tosylates or triflates) instead of the acyl halides

in the first step.

Example 27

Liposomes containing lipids produced in accordance with the present invention and

particular compounds 6d, 18 and 24 were employed to transfect a variety of cell types as

listed in Table VIII.

TABLE VIII

DNA or RNATRANSFECTION OF OTHER CELL TYPES

WITH LIPOSOMES MADE WITH LIPIDS 6d, 18 and 24

Table VIII shows remarkable transfection efficiencies for liposomes derived from

compounds (6d) and (24). Especially, compound (6d) has proven to be an outstanding reagent for the transfection of transfection-refractory or difficult-to-transfect cells types

such as primary neurones and primary quail myoblasts, respectively. Normally, liposome reagents of the previous art were considered superior if transfection efficiencies

approaching 5-10% were obtained with primary cells. Compound (6d) has produced

transfection efficiencies for the latter type of cells of up to 80%. On the other hand, compound (24) gives very good transfection efficiencies with primary rat smooth muscle

cells and superb efficiencies with difficult-to-transfect insect cells. Moreover, this

compound works, without diminished activity, in the presence of 10% serum. This characteristic is a desirable one for the practice of gene therapy since the cells in the human

body are in permanent contact with pure serum, so any liposome reagent that has little or no activity in the presence of serum has a limited use in this therapeutic arena. Finally, compound (24) is also an outstanding reagent for the transfection of RNA into human

kidney cells.

Thus, as shown in Table VIII, the lipids of the present invention provided

significant assistance in the transfection of a variety of cell types and in several cases resulted in positive transfection rates of as high as 80 or 90%.

Having described the inventions with regard to specific embodiments, it is to be understood that the description is not meant as a limitation since further variations or

modifications may be apparent or may suggest themselves to those skilled in the art. It is

intended that the present application cover all variations and modifications of the

inventions as fall within the scope of the appended claims.

Claims

Claims

What is claimed is:

1 A composition comprising the mixture of a compound having the structure:

I 2Y- I

R₂ _ N⁺ [CH₂CH₂0]_n N⁺ R₂'

R₃ R₃"

R_1} Ri' = independently: linear alkyl -

R₂, R₂' = independently: linear alkyl -C₁₈, linear alkyl - C₄, guanidium or amidinium, aminopropyl, 4-a butyrylamidopropyl, cyanoethyl

R₃, R₃' = independently: linear alkyl C_rC₆, acetoxyethyl, cyanoethyl.

Y = a pharmaceutically acceptable anion

n= 2-50, or higher

and a member of the class consisting of nucleic acids, oligonucleotides, mononucleotides, polypeptides and proteins.

2. A compound having the structure:

R, 2Y- R.'

R₂ — N⁺ — [CH₂CH₂0]_n CH₂CH₂ N⁺ R₂'

I I

R₃ R,'

R_1} R = independently: linear alkyl C,-C₁₈

R₂, R₂' = independently: linear alkyl C_rC₁₈, linear alkyl C1-C4, guanidium or

amidinium, aminopropyl, 4-a butyrylamidopropyl, cyanoethyl

R₃, R₃' = independently: linear alkyl C C , acetoxyethyl, cyanoethyl.

Y = a pharmaceutically acceptable anion

n = 2-10.

A composition comprising the mixture of a compound having the structure

Rι,Rι' = Independently: Ci to C₂₀ acyl group

R₂ = Electron pair, Ci TO C₆ Alkyl group; CH₂CONH₂; Methyl Ethyl or H, CH CN; amino C₂to C alkyl; hydroxy C₂to C₄ alkyl; polyamino alkyl.

X = pharmaceutically acceptable anion(s), and a member of the class consisting of nucleic acids -oligonucleotides, mononucleotides, polypeptides and proteins.

A compound having the structure

Rι,Rι' = Independently: C to C₂₀acyl group

R₂ = Electron pair, C i to C₆ alkyl group; CH₂CONH₂; CH₂CO₂Methyl, Ethyl or H, CH₂CN; amino C₂ to C₄ alkyl; hydroxy C₂ to C₄ alkyl; polyamino alkyl

X = pharmaceutically acceptable anion(s).

5. A liposome comprising the compound of claim 2.

6. A liposome comprising the compound of claim 4.

7. The compound of claim 2 which is N, N, N, N, N', N', N', N' -

tetramethyldipalmityl - triethyleneglycoldiamine.

8. A composition comprising the compound of claim 7 and a member of a class

consisting of nucleic acids, oligonucleotides, mononucleotides, polypeptides, and proteins.

9. The compound of claim 4 which is 1, 2 - palmityl - 3- triethylenediammonium propanediol or alkyhalide derivatives thereof.

10. A composition comprising the compound of claim 9 and a member of a class

consisting of nucleic acids, oligonucleotides, mononucleotides, polypeptides, and

proteins.

11. The composition of claim 1, further comprising a colipid.

12. The composition of claim 3, further comprising a colipid.

13. The composition of claim 8, further comprising a colipid.

14. The composition of claim 10, further comprising a colipid.

15. A method of delivering a negatively charged macromolecule to a cell comprising

delivering to said cell the composition of claim 1.

16. A method of delivering a negatively charged macromolecule to a cell comprising delivering to said cell the composition of claim 3.

17. The method of claim 15, wherein the composition further comprises a colipid.

18. The method of claim 16, wherein the composition further comprises a colipid.