AU8986198A - Genetic and protein therapy products for treating cystic fibrosis and inducing aglutathione-s-transferase function - Google Patents

Description

I GENE AND PROTEIN THERAPY PRODUCTS FOR TREATING CYSTIC FIBROSIS AND INDUCERS OF A GLUTATHIONE-S TRANSFERASE FUNCTION 5 Mucoviscidosis, also sometimes called cystic fibrosis, is well known to be caused by mutations in the gene coding for the protein designated CFTR, the abbreviation for "Cystic Fibrosis Transmembrane Conductance Regulator". The nucleotide sequence coding for this CFTR protein which 10 comprises 1480 amino acid residues was cloned and identified by Riordan J.R. et al., "Identification of the Cystic Fibrosis Gene : Cloning and Characterization of Complementary DNA" (1989) Science 245 : 1066-1072. The sequence coding for the CFTR is also reproduced in Figure 1 of the application PCT WO/9102796 made 15 public on 7 March 1981 and of which Riordan himself is also one of the co-inventors. This PCT application - and others which have followed - actually describe the cloning of the coding sequence and its expression in a certain number of organisms. According to a recurrent hypothesis in the scientific literature, 20 the normal CFTR protein is supposed to possess a so-called "chloride channel" or "chloride ion channel" function, in other words it regulates anion transfers, more particularly of chloride ions (Cl-), from epithelial cells towards the exterior mediated by its phosphorylation by a cyclic AMP (cAMP)-dependent protein kinase 25 or protein kinase C (Riordan, J.R. et al. (1989) Science 245 : 1066 1073; Frizzell, R.A. et al. (1986) Science 233 : 558-560; Welsh, M.J. and Liedtke, C.M. (1986) Nature 322: 467; Li, M et al. (1988) Nature 331 : 358-360; Hwang, T-C et al. (1989) Science 244 : 1351-1353). At 37C, the CFTR protein is thus considered to pass from the 30 endoplasmatic reticulum to the Golgi apparatus and from there into the plasma membrane, as recalled by Ko et al., (1997) Biochemistry 36 : 5053-5064, who themselves refer to earlier authors. Many mutations of the CFTR have been observed in patients with cystic fibrosis. 35 Epidemiological studies conducted on affected populations have shown in approximately 70% of the cases a CF ("cystic fibrosis") mutation or deletion of the three nucleotides coding for the phenylalanine at position 508 of the amino acid sequence of 2 CFTR. The site of this mutation is thought to be localized more particularly in the so-called NBF1 ("nucleotide binding fold 1") region which contains the first ATP binding domain in CFTR and which, according to Ko et al. already cited, extends from the 5 phenylalanine residue 433 (P433) to the serine residue 589 (S589). This mutation is often held responsible for defective folding of the NBF1 domain in the then CF-defective cells of the patients, a defect which would lead to a blockage of the thus mutated CFTR (designated AF508) in the intracellular organelles (Golgi, 10 endoplasmatic reticulum), then to its destruction before it could complete its post-translation modification. And it is indeed the deficiency of the "chloride ion channel" function which the said "sweat" test, which measures the secretion of chloride in a biological fluid derived from the patient concerned, aims to detect in support 15 of a diagnosis of cystic fibrosis. The correlation often made between clinical phenomena observed in many patients and the mutations observed in the gene coding for CFTR forms the basis of one of the approaches to the treatment of cystic fibrosis already suggested, in this case the 20 approach based on gene therapy, for the purpose of transferring a DNA coding for an "intact" CFTR into the cells in a form then allowing its expression even in CF-defective cells, in particular in epithelial cells of the respiratory tracts of the patients. Thus tests involving a DNA complex coding for the CFTR and 25 cationic lipids were administered to mice by the intra-tracheal route in order to obtain their direct entry into the lungs (Yoshimura et al. "Nucleic Acid Research, 20:3233-3240, 1992) or also by aerosol, Stribling et al., Proc. Natl. Acad. Sci. 89:11277-11281, 1992). It has also been observed that the administration of the gene coding for 30 the CFTR complexed with cationic lipids to a mouse model suffering from cystic fibrosis had the effect of correcting the defective functioning of the chloride ion channel (Hyde et al., Nature 362:250-255, 1993). It has also been suggested, for example in the international 35 applications WO 95/13365 and WO 96/13598, that the expression of an entire CFTR gene supplied from the exterior be induced in the CF- cells of the patients to be treated , this gene having previously been incorporated under the control of suitable regulatory elements 3 into vectors, for example genetically modified adenoviruses or retroviruses, preferably defective but still capable of infecting these cells, of transferring the CFTR gene to them, the latter then being able to be expressed there under the control of the above-mentioned 5 regulatory elements. These methods usually prove to be, if not ineffective, at least inadequate. But as the authors of the PCT application WO 95/25796 observe, the viral vectors of this type can only harbour or package foreign or exogenous genes of limited size, maximally 4.5 10 kb. Now the DNA coding for the natural CFTR reaches this size limit and this makes its expression hazardous in the infected cells under the control of the corresponding viral promoters. To this difficulty is added another apparently major one, namely the low solubility of the expression product in the 15 cytoplasm of the cells in question. Finally, to this is added the probably immunogenic character of the high molecular weight expression products possibly formed. Consequently, the authors of the application WO 95/25796 have suggested replacing the entire gene by a truncated gene 20 lacking a significant part of the C-terminal region of the CFTR, provided however that the deletion of this C-terminal region is not accompanied by the loss of the biological properties of the chloride ion channel type, characteristic of the natural human CFTR or also functionally analogous properties of the recombinant entire CFTR. 25 According to the authors of the application, a recombinant truncated protein still capable of exercising its functions must usually include: - the MSD sequence ("Membrane Spanning Domain") which extends approximately between the amino acid residues 76 to 360 30 (counting from the N-terminus of the CFTR protein), - the NBD1 domain ("Nucleotide Binding Domain-i") - or NBF1 which extends approximately from residue 360 to residue 708 of the entire CFTR, and finally - the regulator domain R which extends approximately from the 35 amino acid residue 590 to 830 of the CFTR, this domain being supposed to ensure, at rest, the closing of the ion chloride channel and its opening mediated by phosphorylation by the cAMP dependent protein kinase or protein kinase C.

4 In particular, a recombinant protein comprising the 884 N terminal amino acids of the CFTR would still exercise the functions required of a chloride ion channel and of a regulator of the opening and closing of this channel. 5 Even if it is supposed that it is actually possible to obtain a better expression with such CFTR amputated of their C-terminal regions, one can not fail to observe the still considerable size of the expression products likely to be obtained and, consequently, to anticipate the persistence of an immunogenic character making its 10 use more than hazardous in gene therapy trials. It should also be noticed that this proposition of gene therapy, which would make use only of the truncated gene of the CFTR corresponding to the conditions described in the patent WO 95/25796, still links the efficacy of the treatment to the 15 reestablishment of the function of the chloride ion channel only, which is exercised in healthy subjects by a functional CFTR protein. However, this is to omit that if the transport defect of the chloride ions actually constitutes a clinical symptom often observed in patients suffering from cystic fibrosis, other symptoms also 20 observed remain unexplained to this day. The invention follows precisely from the observation that the treatment of cystic fibrosis takes place in large measure by a different therapeutic approach since, as demonstrated by the inventors, the CFTR itself plays a particularly important role in the 25 active transport of glutathione (GSH), such that cystic fibrosis might also be linked in an important manner to mutations or other causes which would disturb the capacity of the CFTR to exercise a glutathione transporter function. Hence the invention relates to DNA fragments whose 30 sequences, including sequences derived from the gene coding for the CFTR, are chosen from those which code for an active transport function of glutathione. The invention concerns quite particularly smaller fragments than the DNA fragments whose use in gene therapy for the treatment of cystic fibrosis has already been cited in 35 the technical or scientific literature. It is in fact owing : - on the one hand, to the discovery of particular properties of a fusion polypeptide containing a part of the CFTR sequence which 5 had been produced in the context of studies relating to the study of the three-dimensional structure of the sequence coding for the NBF 1 domain, and - on the other, to observations relating to a durable clinical 5 improvement of the symptoms linked to cystic fibrosis in a patient also treated with cancer chemotherapy which was accompanied by an overexpression of said MRP protein ("Multidrug Resistance Associated Protein"), that the inventors were led to the discovery which is at the root of 10 the invention itself. The study of the three-dimensional structure of the sequence coding for the NBF-1 domain - in particular in order to study the effect of the AF508 mutation presumed to be responsible for the folding defect of the CFTR in the NBF1 domain - requires the prior 15 isolation of considerable quantities of stable, soluble and pure protein (several tens of mg). Since such quantities are not accessible by purification of the natural protein, the inventors therefore undertook the production of NBF1 (in its initial definition) by cloning in a living expression system (E. coli) in the form of a fusion 20 protein with GST (glutathione-S-transferase). Initially, the objective of this choice was to use the specific recognition capacity of glutathione S-transferase by glutathione to purify the desired protein (GST-NBF1) from all of the bacterial proteins. For this purpose, a sequence derived from the CFTR gene (sequence R 4 50 25 1586) and containing the nucleotide sequence coding for the NBF1 sequence was inserted into a plasmid permitting the transformation of E. coli bacteria and the expression in the latter of the insertion sequence in the bacteria transformed, for example, a plasmid of the pGEX-KT type. The placing in culture of the E. coli strain thus 30 transformed did in fact lead to an overproduction of a fusion protein of NBF1 and glutathione-S-transferase (GST). The yields obtained were 5 mg/ml of culture. The soluble protein obtained was sequenced. It was recognized by anti-NBF1 antibodies and also bound ATP. Cleavage with thrombin allowed the free NBF1 35 domains to be prepared which exhibited the same characteristics. However, it is not possible to separate the GST and the NBF1 domain on an affinity column to which glutathione (GSH) is grafted : although completely separated on electrophoresis gel, the two 6 proteins NBF1 and GST co-eluted from a column of immobilized GSH. The activity of the fusion protein was verified by fluorescence an affinity for a fluorescent derivative of ATP (TNP-ATP) was in 5 fact observed. However, many attempts to purify the NBF1 isolated from GST by cleavage with thrombin have remained fruitless. Moreover, the GST-NBF1 protein could be cleaved selectively and quantitatively but the NBF1 domain alone was then revealed to be 10 unstable. It had a strong tendency to form aggregates. These results led the inventors to reconsider or reevaluate the definition of the NBF1 domain, such as had been described by Riordan. In fact, the definition of a domain must correspond to a stable entity, characterized by a correct folding coding for its 15 function and whose stability is independent of the complete protein (here, CFTR) to which this domain belongs. In order to ensure this folding, the limits chosen for NBF1 must contain all of the secondary structural elements (helices and sheets) necessary for the stabilization of the domain in its native conformation. 20 Now, the CFTR belongs to the family of the ABC transporters (ATP Binding Cassette) which form one of the largest families of membrane proteins (A11.96), found in both eukaryotes and prokaryotes. These proteins are usually active transporters which hydrolyse ATP to supply the chemical potential necessary for the 25 transport. In eukaryotes they transport molecules (ions, vitamins, peptides, sugars, drugs, etc...) across all membranes. Their overall organization presents common features : regions (TM) which participate in the selection of the chemical entity to be transported and binding domains for nucleotides. These genes are often derived 30 by fusion between two "half-genes". The overall topology is then: (TM1)-(NBF1)-(TM2)-(NBF2). And as a result, the CFTR belongs more particularly to the subfamily CFTR/MRP, and shares a sequence similarity of about 50% with the human MRP protein (Multi-drug Resistance associated Protein). It is also very similar to 35 the YCF1 protein (Yeast Cadmium resistance Factor1) which confers on the yeast Saccharomyces cerevisiae resistance to cadmium ions (Tom.96). To a lesser extent it is also similar to STE6 (of the subfamily MDR/TAP) which transports the pheromone "factor a" in 7 the yeast Saccharomyces cerevisiae (Tom.93) and human MDR (Multi-drug Resistance protein) or even bovine F1-ATPase, the only protein of known structure in the family of the ABC proteins (Abr.94). 5 And it is because the structure of the bovine ATPase is known that the inventors chose its NBF domain also as reference structure for the reevaluation of the NBF1 domain of the CFTR. This choice, discussed in two recent publications (Ann.97a, Ann.97b), has resulted in a 3D structure model for NBF1 constructed by homology 10 with the structure of bovine ATPase-F1, a method developed in (Pat.96), making it possible among other things: 1. to ascertain the phenotype linked to the mutation AF508, unlike the earlier models (Hyd.90, Mim.91); 2. to ascertain the key role of the consensus sequence LSGGQ, 15 highly conserved in the ABC proteins; 3. to redefine the limits of the NBF1 domain - hereafter "NBF1 revisited" - which is henceforth considered as extending approximately from amino acid residue 430 to approximately residue 650 (overlapping by about 70 residues with the R domain, 20 as initially defined) of the CFTR sequence. This last point follows directly from the hypothesis of a comparable folding for the NBF regions of bovine F1-ATPase and CFTR. The principal argument is that the family of the ABC transporters form a homogeneous group quite distinct from other 25 ATP-binding proteins, whose members have resulted from divergent evolution. In this context, these proteins must exhibit a comparable folding in their conserved regions (typically, in the NBF "elementary bricks"). It is then permissible to model the structure of one of these proteins on the known structure of another. 30 The structural model thus constructed for the NBF1 domain of CFTR, by homology with the known structure of the nucleotide binding domains of the bovine F1-ATPase (Abrahams J.P., Leslie A.G.W., Lutter R., Walker J.F. (1994) Nature 370:621-628), according to the method described by Annereau et al.(FEBS Letters, (1997) 35 407:303-308), comprises a hydrophobic core constituted of a sheet of six parallel strands, respectively: L453-S458, N505-S5 11 , N 5 38-G 5 42, D567-D572, R600-V603, D614-1618). These sheets alternate with six a helices and are organized on 8 either side of the mean plane of the central sheet, as shown in the article published in FEBS Letters, Vol. 407, pp.:303-308, 1997. Loops situated at the surface of the domain connect the helices to the sheets. 5 And in fact the expression product of the fragment corresponding to the gene coding for the CFTR (fragment which itself forms part of the invention) proved to be stable, which substantiates the results of cloning in Ecoli of the "new" NBF1 domain (or NBF1 domain revisited), in E. coli in a polyhistidine 10 system, i.e. a fusion DNA coding for the "NBF1 revisited" domain (it is the fragment coding for the polypeptide extending from amino acid residues 448 to 650 of the CFTR) and a tail consisting of only 6 histidine residues (it may be considered as the "sole" product). The production levels obtained are acknowledged to be higher than for 15 the earlier fragment, attaining for the polyhistidine system about 200 mg per liter of culture (right from the first assay, before optimization of any kind). This purified "NBF1 revisited" fragment was purified in very large quantities, in the form of a soluble and stable protein. 20 The potential existence of a glutathione binding site in NBF1 of CFTR practically superposable with the glutathione binding siite in GST is demonstrated by comparison of this model with the known structure of GST (glutathione-S-transferase), a protein which possesses a glutathione binding site and which catalyses the 25 nucleophilic reactions of glutathione. The residues of NBF1 potentially implicated in glutathione binding are located in a region forming a crevice with the extremities of strands implicated in a sheet (see FEBS letter article). These residues are situated more precisely in the loops, at the N-termini of the sheets 1 to 4 of the 30 NBF1 structure. These loops contain the residues 1448-Q452 (IERGQ),

S

4 78-K 4 81 (SEGK), M498-1506 (MPGTIKENI), A 5 66-L 5 68 (ADLVL), A596-T599 (ANKT). The new definition of NBF1 and the model of the 3D structure constructed have allowed the inventors to put forward a novel 35 function for CFTR, suggested by the unexpected fact that NBF1 is retained on a glutathione column, even after cleavage of the GST NF1 fusion protein. In fact, the comparison of this model with the known 3D structure of GST, a protein which possesses a glutathione 9 binding site and which catalyses the nucleophilic reactions of glutathione (Lim.94), demonstrates the presence of a potential glutathione binding site. Functional information relating to the two ABC proteins the 5 most similar to CFTR (human MRP and yeast YCF1) seem to confirm this hypothesis. In fact, it is now established that MRP exports drugs through the intermediary of glutathione, either in the form of direct "glutathione-drug" adducts, or by simultaneous or sequential binding of glutathione and the drug (Zarn.95). 10 Simultaneously, YCF1 transports cadmium ions in the form of double adducts of glutathione. Thus, it seems that CFTR can be a "glutathione pump", like MRP and YCF1. Now, the human MDR and MRP proteins are implicated in the phenomena of multi-resistance to antitumor drugs (Dean et al., 15 Current Opinion in Genetics & Development 5, 779-785, 1995). These proteins are capable of actively transporting these drugs, thus participating in the cellular detoxification mechanisms. Their overexpression is induced by the drugs themselves and this constitutes the principal reason for the failure of the treatments of 20 cancers by chemotherapy. Now, it has recently been shown that the overexpression of CFTR can also be triggered by anti-tumor drugs and that it confers the multi-resistance phenotype (Ll.95) Consequently, this results in the hypothesis formulated by the inventors that the transporter function of glutathione might be 25 important and that mutations likely to perturb it would also explain many symptoms of cystic fibrosis. The role of CFTR as "glutathione pump" (in addition to its known role as chloride channel) also permits a more complete understanding of cystic fibrosis (Sto.97). A deficit of the glutathione transport function would also 30 account better for the chronic inflammatory reactions and/or the very high number of polymorphonuclear bodies which is often encountered in the respiratory tracts of patients suffering from cystic fibrosis since it is known that glutathione plays a central role in the control of the inflammatory reaction by protecting cells from 35 the aggression of free radicals by means of polymorphonuclear neutrophils and, in addition, that it is implicated in the protection against oxidants such as free radicals or superoxides.

10 Similarly, the organs in which glutathione is usually secreted and in which it exercises a toxification function (lungs, intestines, colon) are usually the ones severely affected by cystic fibrosis. Quite a number of clinical observations are in agreement with 5 the hypothesis which forms the basis of the invention. The GSH level is in fact lowered in the serum and the bronchial mucus of subjects suffering from cystic fibrosis. The organs where CFTR is best expressed are those where GSH plays an important role and is actively transported : pancreas, lung, liver, kidney. The detoxifying 10 hepatic function appears to be impaired, a function in which GSH might play a central role. This results for example in a particular sensitivity of the patients to a heavy metal like mercury. Phenomena of oxidative stress have also been observed for a long time at the pulmonary level. N-Acetyl cysteine, a medicine used 15 with benefit for the patients, is deacetylated on penetrating into the cell, supplying cysteine (semi-essential amino acid for humans). And in fact the validity of the hypothesis already finds confirmation in the durable improvement which has been observed in terms of cystic fibrosis in a 29 years old patient, a carrier of cystic 20 fibrosis (with clinical signs since birth). The patient treated had the exon 12 of CFTR on one allele and G673X situated in exon 13 on the other allele. This patient presented since his birth in 1968 all the clinical signs associated with the disease : positive sweat tests, repeated sinusitis, repeated bronchitis, polyps of the nasal fossa 25 etc... In 1989 he presented a Pseudomonas aeruginosa infection, common in cystic fibrosis and usually practically definitive in these patients. In addition, in April 1993, a fiibrosarcoma of the left thigh was diagnosed. The patient underwent surgical treatment and radiotherapy, then chemotherapy from July to October 1993, 30 followed by further treatment in January 1994. More particularly, the anti-tumor treatment was the following - epirubicin, 110 mg for two days (D1 and D2) - ifosfamide, 3.3 g for five days (D1 to D5) 35 - filgrastim, 300 g per day for eight days (D8 to D15); supplemented by: - granisetron, for the prevention of nauseas - methylprednisolone; 11 - mesna. Three months later, the patient again received six cycles of epirubicin and ifosfamide. The anti-cancer treatment was efficacious. Also regarding the 5 clinical picture of cystic fibrosis, the following improvements were observed following the chemotherapy : - the respiratory state of the patient was considerably improved; - his Pseudomonas aeruginosa infection had disappeared; 10 - his respiratory parameters attain about 75% of the theoretical values; - he had recovered a respiratory state about equivalent to that which he possessed at the start of his adolescence; - he declared himself to be cured of cystic fibrosis. 15 A sweat test performed on him at the beginning of 1997 was shown to be still positive, which signifies that the chloride ion channel function had not been reestablished in this patient. This observation, in conjunction with the fact that the patient declared himself to be cured of cystic fibrosis, leads to the conclusion that 20 another essential function had been reestablished by chemotherapy. It may be supposed in the light of the foregoing that it is in this instance the glutathione transport function which had considerably increased. 25 As has already been indicated, the invention thus relates more particularly to DNA fragments corresponding to parts of the gene coding for the CFTR, these parts also coding for a polypeptide possessing an active transport function for glutathione. In what follows, the DNA sequences are usually defined by reference to 30 peptide sequences corresponding to them, these peptide sequences being each time usually designated for linguistic convenience by the expression "expression product" of the DNA sequences of the invention. These DNA sequences may naturally be reconstituted by the reader, in particular by basing himself on Figure 1 which is 35 essentially a reproduction of the sequence of the gene coding for the CFTR which was published in the article byRiordan et al. already mentioned or also in the PCT application WO 91/02796.

12 In the definitions of the DNA sequences - or their expression products - which follow, reference will frequently be made to regions "downstream" or "upstream" from the DNA sequence or its expression product to which the invention refers. It will be 5 immediately apparent that a region "upstream" from a given DNA sequence within the gene coding for the CFTR represented in Figure 1 or from the corresponding expression product is situated upstream from the 5' end of this DNA sequence or from the N terminus of the amino acid sequence of the corresponding 10 expression product. Conversely, the regions "downstream" are those which are located beyond the 3' end of this DNA sequence or the C-terminus of the amino acid sequence of the corresponding expression product. Equally obviously, the indication of a letter (in the system of 15 symbols for the identification of the amino acids by which each of them is represented by a unique letter) followed by a subscript "n" refers to the "nth" amino acid which follows from examination of the appended Figure 1, counting from the first amino acid residue of the CFTR. 20 In general, the longest DNA molecule in conformity with the invention contains a DNA sequence whose expression product has a peptide sequence inn common with a part of that of the CFTR, this part lacking essentially at least the N-terminal region of the CFTR which normally extends upstream from the MSD region of this 25 latter and which comprises in particular the 360 N-terminal amino acids of the CFTR or, as a variant, any DNA sequence resulting from the modification of the former by deletion, substitution or addition of nucleotides, provided that the capacity of the corresponding expression product to exercise a glutathione 30 transporter function is conserved, at least in part. By "glutathione transporter compound" is meant according to the invention a compound which belongs to the family already listed of the ABC transporters which exert their transport function through the intermediary of glutathione. It also means any fragment 35 or analogue of such a transporter resulting from one or several mutations which conserves the property of transport through the intermediary of the glutathione.

13 By glutathione is meant glutathione or its adducts with other compounds. They may be natural adducts such as the leukotrienes or also detoxification adducts with heavy metals, powerful oxidants (free radicals, peroxides....) or drugs. 5 In particular, the capacities of the expression products of the DNA sequences in conformity with the invention to exercise a glutathione transporter function can be estimated by their capacity to induce in vitro a production of CFTR in epithelial cells, for example in vitro, by implementing the experimental techniques 10 already used to estimate the same type of induction under the effect of antitumor drugs, such as have been described and used by Wey L.Y. et al. in the article entitled "Overexpression of the cystic fibrosis transmembrane regulator in NIH 3T3 cells lowers membrane potential and intracellular pH and confers a multidrug resistance 15 phenotype"; Biophysical Journal (1995) 69:883-895. Recourse may also be had to the measurement of the levels of mRNA corresponding to CFTR in the epithelial cells previously taken from an experimental mammal suffering from cystic fibrosis, such as a mouse to which had previously been administered the test 20 DNA molecule in conformity with the invention, under the control of regulatory elements permitting the transcription and translation of its DNA sequence in epithelial cells, in particular under the conditions described by Yoshimura et al., in the article identified above. 25 TgH (Unc) type mice, obtained by homologous recombination at the level of the exon 10 of the Cftr gene (insertion of the gene expressing neomycin resistance), represent an interesting model for this study. The heterozygous animals show no modification of the phenotype. The homozygous mutants show anomalies similar to 30 human cystic fibrosis in the young mice (Koller B.H. et al., Science, 248, 1227-1229). These mice are adequate models given that they suffer and die of intestinal obstruction before reaching the age of adults, thus drawing attention to a model study of inflammatory exacerbation at the level of this organ. 35 An expression vector utilizable in this test and administrable by the general route would for example be formed by an Av1CF2 type adenovirus (Genetic Therapy Inc., Gaithersbourg, MD, USA) in 14 which the coding sequence to be studied would replace the coding sequence for the entire CFTR. The detection of expression of the glutathione transporter function is then estimated by the increase in the quantities of 5 mRNA induced in the treated mice in comparison with what is observed in untreated control mice with cystic fibrosis, for example after quantification of the mRNAs compared to a RT-PCR internal reference involving the level of mRNA of p2-microglobulin, according to the method of Lechapt-Zalcman E. et al. described in 10 Resp.J. 1997 "F.MDR-1-PgP 170 expression in human bronchus". Furthermore, since the quantities of MDR proteins are high in the intestinal epithelium, the transfection of the test mammals cited above by the CFTR fragment under study ought to provoke a resorption of the exacerbated inflammatory state at the level of the 15 intestine. As a variant, recourse may also be had for defining the DNA molecules of the invention to the test already invoked above. Included in the framework of the invention is any DNA molecule meeting the conditions indicated above and which is characterized 20 by a conservation of the affinity of the corresponding expression product for glutathione, this affinity being capable of being demonstrated by the employment of the test which consists of cleaving the expression product of the DNA molecule in question by means of thrombin into a C-terminal fragment and an N 25 terminal fragment, of recovering the N-terminal fragment and of placing it in contact with an affinity column to which glutathione has been grafted, the affinity of said expression product for glutathione then resulting from its capacities, similar to those of glutathione-S-transferase, to be bound to this activity column and to 30 be eluted from it. However, this DNA molecule is preferably shorter, characterized in particular in that its expression product also essentially lacks at its C-terminus the amino acid sequence of the region of the CFTR which normally extends upstream from the R 35 region of this latter. It even suffices that the amino acid sequence of the expression product of the above-mentioned DNA sequence itself comprises, on the one hand and at its C-terminus, the sub-sequence of 15 approximately 80 amino acid residues (which were previously considered to belong to the R region and which henceforth form part of the "NBF1 revisited" region) which, in the CFTR, extends downstream from its amino acid residue 570 and, on the other and 5 in the direction of its N-terminus, the NBF1 region of the CFTR, even only of a part of this latter, in order that the capacity of the expression product of the above-mentioned DNA sequence to exercise said glutathione transporter function is retained. In general, the DNA sequence according to the invention thus 10 comprises essentially the region of the sequence coding for all or part of the "NBF1 revisited" region of the CFTR which extends from its C-terminus to beyond the presumed binding site for glutathione (naturally in the reverse direction of reading of the amino acid sequence of this "NBF1 revisited", i.e. from its C-terminus towards 15 its N-terminus. A preferred DNA molecule is that whose expression product itself possesses an amino acid sequence which extends approximately between amino acid residues 430 and 660 of the CFTR. 20 Among the preferred DNA molecules according to the invention, mention should be made of those which are characterized in that the amino acid sequences of their respective expression products themselves contain, on the one hand and at their C-terminus the sequence NLQPDFSSKLMGCDS (N635 to S 649 ) 25 of the CFTR and, on the other and in the direction of their N terminus, one or more of the following peptide sites: - ANKT (A596 to T599), - ADLYL (A566 to L570), - MPGTIKENI (M498 to 1506), 30 - SEGK (S 478 to K481), - IERGQ (1448 to Q452), of the CFTR In particular, the DNA molecule according to the invention 35 may be selected from those possessing a DNA sequence characterized in that the amino acid sequence of its expression product includes, on the one hand and at its C-terminus, the 16 sequence NLQPDFSSKLMGCDS (N635 to S 649 ) and, on the other and at its N-erminuus : - either the peptide site ANKT (A596 to T599), - or the peptide site ADLYL (A566 to L570), 5 - or the peptide site MPGTIKENI (M498 to 1506), - or the peptide site SEGK (S478 to K481), - or the peptide site IERGQ (I448 to Q452), of the CFTR. That being so, the amino acid sequence of the expression 10 product of the above-mentioned DNA sequence may also include, downstream from the site NLQPDFSSKLMGCDS (N635 to S649) a segment of 10 to 20 amino acid residues whose sequence corresponds to that which also follows this same site in the CFTR or also, upstream from the above-mentioned site close to its N 15 terminus, a segment of 10 to 20 amino acid residues which also precedes this same site in the CFTR, or both segments at the same time. These segments may also be replaced by (oligonucleotide) linkers containing restriction sites facilitating their incorporation in a vector designed to transfer them into suitable competent cells. 20 In general, the preferred DNAs of the invention are those whose expression products are soluble in water or in a physiological serum, in particular at a concentration of at least 0.5, and preferably at least 1 mg/ml of water, in particular in the form of a physiological solution. 25 Naturally, the invention also relates to any recombinant DNA molecule in which the DNA sequence such as defined above is recombined with exogenous nucleic acid elements, in particular recombinant molecules coding for a fusion protein exhibiting nonetheless the essential biological properties such as were defined 30 above of the expression product of the DNA sequence characteristic of the molecules according to the invention. Another class of DNA molecules in conformity with the invention are those in which the exogenous nucleic acid elements are constitutive of an expression vector, comprising suitable 35 regulatory elements, said nucleic acid sequence being present as an insertion sequence placed under the control of these regulatory elements, the latter then permitting the expression of said nucleic acid sequence in a competent cell organism.

17 In particular, the expression vectors are capable of transforming competent cellular organisms in order to ensure the expression of the DNA sequence characteristic of the invention, the expression product then being optionally recoverable from the cells 5 in culture, even from the culture medium. It is obvious that the invention is not be limited to vectors modified by the DNA sequence according to the invention which could only be used for the purposes of gene therapy. It also extends its effects to any other form of vector capable of being used for the production of the 10 corresponding expression product in cells in culture, in particular for the purposes of producing the recombinant expression product which might itself be used directly, as envisaged below. Reference should be made to the descriptions of the international applications WO 91/02796, already cited, WO 15 91/10374, WO 92/05273 and WO 93/17040, for examples of expression vectors which have already been described for the expression of the entire CFTR gene, and to the application WO 95/25796 already mentioned in which are described usable vectors for the expression of a CFTR whose C-terminal region has been 20 truncated. These different vectors can naturally also be used for the expression of the DNA sequences in conformity with the present invention. The description of these patent applications must be considered as forming part of the description of the present application as regards the variants of expression vectors which can 25 be used for the production of the expression products of the DNA molecules in conformity with the invention. Of particular interest are the vectors permitting the transformation of human cells, in particular epithelial cells. Particularly preferred expression vectors are vectors derived from 30 viruses of the adenovirus or retrovirus type or even viruses related to those (AAV viruses) whose use has already been envisaged for the achievement of gene therapies applied to the treatment of cystic fibrosis. Mention also should be made of the international applications WO 95/13365 and WO 96/13598 which describe 35 vectors and more generally techniques applicable to the expression of the gene coding for the CFTR for the purposes of application in gene therapy in the form of transcription and expression cassettes comprising the sequence coding for the CFTR, recombined with 18 functional in vivo regulatory sequences in mammalian cells. Again these techniques are applicable to the expression of the DNA sequences in conformity with the invention (instead of DNA sequences coding for the entire CFTR) such that the descriptions of 5 these applications must also be considered as forming part of the description of the present application as regards the description of techniques and vectors usable for this purpose. Everything indicates that the expression induced in vivo of the DNA molecules of the invention, which are all designed to code for 10 a polypeptide having a glutathione transporter function, can rectify an insufficiency in this respect of an endogenous CFTR on account of mutations which it carries and to which it seems necessary to attribute at least a part of the clinical signs of cystic fibrosis and, more generally, stimulate an endogenous expression of CFTR and 15 in doing so even reestablish in part the chloride ion channel function provided that the latter has not been completely annihilated by other mutations affecting this endogenous CFTR. The therapeutic effect expected of a gene therapy making use of DNA sequences complying with the above-mentioned definitions 20 may be the greater, the smaller, more soluble and more stable the expression products of these DNA sequences and the more reduced their intrinsic immunogenicity, which in practice may even be nonexistent. In general, the invention also relates to any pharmaceutical 25 composition combining the DNA molecules in conformity with the invention, optionally already in a "vectorized" form, with pharmaceutical vehicles facilitating their penetration into human cells and their expression in the latter. All types of cells capable of being targeted by molecules used 30 directly or indirectly (for example when it concerns polypeptides generated in situ as in the case of gene therapy) are concerned in anti-cystic fibrosis therapy. It concerns for example epithelial cells, in particular in local treatments, for example in the lung, or any other cells for example hepatic cells which are likely to be 35 implicated in a beneficial effect of the treatment, in particular when it results from a treatment by the general route. This latter assumes a very special interest in the framework of the invention, in the light 19 of the preferred mode of action of which it takes advantage, namely the improvement of the glutathione transporter function. The invention thus relates to all forms of pharmaceutical compositions suitable for the treatment of cystic fibrosis, permitting 5 the targeting of the cells selected and containing, combined with a suitable, pharmaceutically acceptable vehicle, a DNA cassette comprising the DNA molecule according to the invention under the control of regulatory elements able to control the transcription and translation of its coding sequence in mammalian, and preferably, 10 human cells. A first type of preferred compositions are pharmaceutical compositions adapted to administration by the general route for the reasons cited above. Other forms of compositions are adapted to administration by the intermediary of the airways, preferred 15 compositions according to the invention then being those which may be used in the form of aerosols, sprays or the like permitting the absorption of the composition in particular by inhalation, and the conveyance of these compositions to direct contact with the lung cells targeted. 20 Examples of pharmaceutical vehicles, in particular of the cationic lipid type, particularly appropriate to this mode of administration and for similar purposes (in relation to gene therapies making use of the entire CFTR) have already been described for example in the international applications WO 25 93/12240, WO 93/12756 or WO 93/24640. It goes without saying that they are also applicable to the constitution of pharmaceutical compositions making use of the DNA molecule according to the invention. Finally, the invention also concerns the polypeptides 30 themselves which are characterized by amino acid sequences such as have been defined above in relation to the DNA sequences characteristic of the DNA molecules according to the invention. These polypeptides are in particular those which were produced in cells in culture, after transformation of the latter by suitable vectors 35 containing DNA sequences coding for these polypeptides under the control of suitable regulatory elements, and recovery of the expression polypeptides obtained from these cell cultures.

20 Reference has already been made above and as examples to several patents already published relative to culture systems whose use has already been described in relation to the production of recombinant CFTR. The same systems may be used with advantage 5 for the production of the polypeptides defined in the present application. Similarly, the invention relates to pharmaceutical compositions combining the above-mentioned polypeptides with suitable pharmaceutical vehicles, in particular of the same type as 10 those considered above for the pharmaceutical compositions based on DNA molecules. The invention is not limited only to the use of a DNA molecule such as has been defined above, in particular as regards the peptide sequence which it possesses in common with the gene 15 of the CFTR. It also relates to the compositions in which this DNA molecule would be replaced - or supplemented - by a distinct DNA molecule whose expression product, structurally different from that of the DNA molecule such as has been defined above, would also exhibit a glutathione transporter function. In particular, DNAs 20 usable for this purpose would be those containing an NBF1 region of proteins of the human MDR or MRP protein type, which also carries such genetic information. As an example, mention should be made of the DNA sequence coding for the human MRP-1 described by Cole S.P.C. et al., Science 258 : 1650-1654 (1992). Similarly, the 25 human MRP protein or the region useful for this purpose may be substituted for - or also added to - the expression product of the DNA molecule in conformity with the invention such as has been described above. The invention is finally not limited to compositions whose 30 active principle would be constituted only by a DNA molecule in conformity with the invention or more generally by a DNA molecule coding for a polypeptide endowed with a transporter activity for glutathione ions or also for corresponding expression products. It also relates to compositions in which this active 35 principle would also be combined with others, for example with a distinct DNA molecule or, depending on the case, with its expression product. This could be, for example, the entire DNA coding for the CFTR or the truncated DNA of the international 21 application WO 95/25796 or the corresponding expression product in order to restore a deficient chloride ion channel function in the patient treated perhaps in a more definitive manner than the DNA molecule according to the invention alone would allow. In this 5 manner it is possible to ensure under the most favourable conditions the simultaneous restoration of the glutathione transporter function and the chloride ion channel function in the patients concerned. Simultaneously, the invention relates to compositions of the 10 type indicated above, for example those in which the DNA molecule according to the invention, in particular that derived from the gene coding for the CFTR, would be combined with a DNA sequence coding for glutathione-S-transferase, such as that borne by the human GSTM1 gene (Zhong S. et al., Biochem.J. 291:41-50 15 (1993)), in particular in order to potentiate the glutathione transporter function of the composition according to the invention. As regards peptide-based compositions, the invention thus also extends its effects to those which would contain, for the same purposes, glutathione-S-transferase. 20 Finally, the clinical indications of the invention are not limited to the treatment of cystic fibrosis. The pharmaceutical compositions of the invention are also applicable to other diseases frequently observed in CF-heterozygous patients, for example asthma and polyarthritis and, more generally, also to the treatment of diseases 25 attributable to a deficit of the glutathione transporter function. Finally, the DNA molecules according to the invention can be used to induce a glutathione transporter function in animals, for example the mouse, in order to create animal models expressing this function and in order to study the effect of other compounds 30 administered to these animal models, whether to stimulate further this glutathione transporter function of their expression products or, on the other hand, to estimate to what extent these compounds interfere with the glutathione transporter functions of these expression products. 35

Claims (21)

1. 2. DNA molecule according to Claim 1, characterized in that the expression product of the above-mentioned DNA sequence also 20 essentially lacks in the direction of its C-terminus the amino acids of the region of the CFTR which normally extends downstream from the R region of this latter and which comprises in particular the 650 C-terminal amino acids of the CFTR. 25 3. DNA molecule according to Claim 1 or Claim 2, characterized in that the amino acid sequence of the expression product of the above-mentioned DNA sequence itself comprises, on the one hand and in the direction of its C-terminus, the subsequence of approximately 80 amino acid residues or more which, in the 30 CFTR, extends downstream from its amino acid residue 570 and, on the other and in the direction of its N-terminus, a sufficient part of the NBF1 region of the CFTR, even of a part upstream from this latter, in order that the capacity of the expression product of the above-mentioned DNA sequence to exercise said function of 35 glutathione transporter is retained.
2. 4. DNA molecule containing: 23 - either a DNA sequence whose expression product has a peptide sequence in common with a part of that of the CFTR, this part essentially lacking at least the N-terminal region of the CFTR which normally extends upstream from the MSD region of this latter and 5 which comprises in particular the 360 N-terminal amino acids of the CFTR, this expression product exhibiting an affinity for glutathione, - or a DNA sequence resulting from the modification of the former by deletion, substitution or addition of nucleotides, provided that 10 the affinity of the corresponding expression product for glutathione is conserved, at least in part.
3. 5. DNA molecule according to Claim 4, characterized in that the expression product of the above-mentioned DNA sequence also 15 essentially lacks in the direction of its C-terminus the amino acids of the region of the CFTR which normally extends downstream from the R region of this latter and which comprises in particular the 650 C-terminal amino acids of the CFTR. 20 6. DNA molecule according to Claim 4 or Claim 5, characterized in that the amino acid sequence of the expression product of the above-mentioned DNA sequence itself comprises, on the one hand and in the direction of its C-terminus, the subsequence of approximately 80 amino acid residues or more which, in the 25 CFTR, extends downstream from its amino acid residue 570 and, on the other and in the direction of its N-terminus , a sufficient part of the NBF1 region of the CFTR, in order that the affinity of said expression product for glutathione is retained. 30 7. DNA molecule according to any one of the Claims 1 to 6, characterized in that the amino acid sequence of the expression polypeptide of the above-mentioned DNA sequence itself contains, on the one hand and in the direction of its C-terminus, the sequence NLQPDFSSKLMGCDS (N63 5 to S649) of the CFTR and, on the other 35 and in the direction of its N-terminus, one or more of the following peptide sites: - ANKT (A596 to T599), - ADLYL (A 5 66 to L 5 7 0), 25
4. 12. DNA molecule according to any one of the Claims 1 to 11, characterized in that its expression product is soluble, in particular at a concentration of at least 0.5 mg/ml of water, for example a physiological solution, and preferably at least 1 mg/ml of 5 physiological solution.
5. 13. DNA molecule according to any one of the Claims 1 to 12, characterized in that it is a recombinant DNA formed between the above-mentioned DNA sequence and a distinct sequence, the whole 10 coding for a fusion molecule conserving the biological properties of the expression product of the former.
6. 14. DNA molecule according to any one of the Claims 1 to 12, characterized in that it consists of an expression vector comprising 15 suitable regulatory elements, said nucleic acid sequence being present as an insertion sequence placed under the control of these regulatory elements, these latter then permitting the expression of said nucleic acid sequence in a competent cellular organism. 20 15. DNA molecule according to Claim 14, characterized in that the competent cellular organism comprises human cells, in particular epithelial or hepatic cells.
7. 16. DNA molecule according to Claim 14 or Claim 15, 25 characterized in that the expression vector is a vector of the adenovirus or retrovirus type or viruses related to them and functionally similar to them.
8. 17. DNA molecule according to any one of the Claims 1 to 16, 30 for administration in gene therapy.
9. 18. Cell host, characterized in that it contains the DNA molecule according to any one of the Claims 1 to 17 in a form permitting its expression within this cell host. 35
10. 19. Composition containing the DNA molecule according to any one of the Claims 1 to 17, combined with a vehicle permitting its penetration into human cells and its expression in the latter. 26
11. 20. Composition according to Claim 19, characterized in that the DNA molecule is combined with a cationic lipid vehicle promoting its expression. 5
12. 21. Composition according to Claim 19 or Claim 20, characterized in that the above-mentioned DNA molecule is combined with a pharmaceutically acceptable vehicle permitting its administration in vivo by the general route. 10
13. 22. Composition according to Claim 19 or Claim 20, characterized in that the above-mentioned DNA molecule is combined with a pharmaceutically acceptable vehicle permitting its administration in vivo by the general route through the 15 intermediary of airways.
14. 23. Composition according to Claim 19 or Claim 20, characterized in that it comprises, in addition, a DNA molecule coding for glutathione-S-transferase, and does so in a form 20 permitting its expression in vivo.
15. 24. Use of the DNA molecule according to any one of the Claims 1 to 17 or of the composition according to any one of the Claims 19 to 23 for the production of a pharmaceutical composition 25 suitable for the treatment of cystic fibrosis.
16. 25. Polypeptide whose sequence coincides with that of the expression product of the DNA sequence such as it has been defined in any one of the Claims 1 to 14. 30
17. 26. Composition containing the polypeptide of Claim 25 in combination with a pharmaceutically acceptable vehicle.
18. 27. Composition according to Claim 26, characterized in that 35 the pharmaceutically acceptable vehicle is of a type which permits administration by the general route. 27
19. 28. Composition according to Claim 26, characterized in that the pharmaceutically acceptable vehicle is of a type which permits administration through the intermediary of the airways. 5 29. Composition according to Claim 28, characterized in that it can be released in the form of aerosols or other form of product administrable by inhalation.
20. 30. Composition according to any one of the Claims 26 to 29, 10 characterized in that it also contains glutathione-S-transferase.
21. 31. Use of the polypeptide composition according to Claim 25 or according to any one of the Claims 26 to 29 for the production of medicines for the treatment of cystic fibrosis. 15 20 25