US20090227769A1

US20090227769A1 - CFTR with a partially deleted R domain and uses thereof

Info

Publication number: US20090227769A1
Application number: US11/999,797
Authority: US
Inventors: Michael J. Welsh; Lynda S. Ostedgaard; Joseph Zabner
Original assignee: University of Iowa Research Foundation UIRF
Current assignee: University of Iowa Research Foundation UIRF
Priority date: 2002-02-19
Filing date: 2007-12-07
Publication date: 2009-09-10
Also published as: WO2003070900A2; US7339046B2; AU2003211119A8; AU2003211119A1; US20030235885A1; WO2003070900A3

Abstract

The present invention offers new therapies for treating Cystic Fibrosis (CF), that are based on novel DNA molecules and proteins encoded by the DNA molecules. The present invention features DNA molecules encoding CFTR having a partially deleted R domain. The partial deletions in the R domain are between residues 708 and 835 of the wild-type CFTR.

Description

CROSS REFERENCE TO RELATED APPLICATIONS

This is a divisional application of U.S. Ser. No. 10/367,507 filed Feb. 14, 2004, which is a nonprovisional application claiming priority under 35 U.S.C. § 119(e) to Provisional Application No. 60/358,074, filed on Feb. 19, 2002.

GRANT REFERENCE

This invention was made in part with government support from the National Heart, Lung and Blood Institute (NHLBI). Therefore the United States Government has certain rights in the invention.

FIELD OF THE INVENTION

This invention relates to DNA molecules encoding partially deleted CFTR and the CFTR proteins encoded thereby which are useful for treating cystic fibrosis (CF) airway disease.

BACKGROUND OF THE INVENTION

Various attempts have been made develop gene therapy for cystic fibrosis (CF) airway disease.
Airway disease is the major cause of morbidity and mortality in cystic fibrosis (CF), an autosomal recessive disease caused by mutations in the gene encoding the cystic fibrosis transmembrane conductance regulator (CFTR) Cl⁻ channel. Welsh et al., The Metabolic and Molecular Basis of Inherited Disease, eds. Scriver, C. R., Beaudet, A. L., Sly, W. S., Valle, D., Childs, B. & Vogelstein, B. (McGraw-Hill, New York). Gene transfer offers the potential for a new and effective treatment for CF airway disease. For reviews see Davies, Geddes & Alton, 2001, J. Gene Med. 3:409-417; Flotte, 1999, Curr. Opin Mol. Ther. 1:510-516; and Welsh, 1999, J. Clin. Invest. 104:1165-1166. Previous studies have shown the feasibility of transferring the CFTR cDNA to CF airway epithelial cells in vitro and in vivo. However, with most vectors two main problems limit gene transfer: gene transfer from the apical surface of differentiated airway epithelia is inefficient, and DNA molecule expression is transient. See Davies, Geddes & Alton, 2001, J. Gene Med. 3:409-417; Flotte, 1999, Curr. Opin Mol. Ther. 1:510-516; and Welsh, 1999, J. Clin. Invest. 104:1165-1166.
For developing CF gene therapy, adeno-associated virus (AAV) vectors have several potential advantages.
One limitation of AAV vectors is the small size of a DNA molecule that can be inserted. Studies testing the insert size suggest that 4100-4900 bp is the optimal genome size for packaging. See Dong, Fran & Frizzell, 1996, Hum Gene Ther. 7:2101-2112. In comparison, the coding sequence of full length CFTR is 4450 bp. Riordan et al., 1989, Science 245:1066-1073. Addition of the two inverted terminal repeats of AAV (300 bp), and minimal 3′ and 5′ untranslated regions (˜100 bp) yields an insert (4850 bp) that leaves little room for promoter-enhancer elements, most of which are >600 bp. Some studies have attempted to circumvent this limitation by using AAV sequences as a promoter. See Zhang et al., 1998, Proc. Natl. Acad. Sci. 95:10158-10163; and Flotte et al., 1993, J. Biol. Chem. 268:3781-3790. However, their utility in differentiated airway epithelia and in vivo is uncertain.
A potential solution to this problem is to shorten the DNA molecule by selectively deleting coding sequence. This strategy has been proposed with a mini-dystrophin gene for Duschennes muscular dystrophy (Phelps et al., 1995, Hum. Mol. Genet. 4:1251-1258) and for CFTR (Zhang et al., 1998, Proc. Natl. Acad. Sci. 95:10158-10163; and Flotte et al., 1993, J. Biol. Chem. 268:3781-3790).
The CFTR R (regulatory) domain (for reviews on the R domain see Ostedgaard, Baldursson & Welsh, 2001, J. Biol. Chem. 276:7689-7692; Sheppard & Welsh, 1999, Physiol. Rev. 79:S23-S45; Gadsby & Nairn, 1999, Pysiol. Rev. 79:S77-S107; and Ma, 2000, News Physiol. Sci. 15:154-158) has been speculated to be an important domain. Earlier studies in heterologous cells indicated that the CFTR R domain is predominantly random coil and that parts of the R domain can be deleted without abolishing channel function. Phosphorylation of the R domain by the cAMP-dependent protein kinase (PKA) controls CFTR Cl⁻ channel activity. Although this domain contains several conserved serines that are phosphorylated by PKA, no one phosphoserine is required and several different phosphoserines contribute to regulation. While the boundaries of the R domain are not precisely defined, they extend approximately from residues 634-708 at the N-terminus to approximately 835 at the C-terminus. See Ostedgaard, Baldursson & Welsh, 2001, J. Biol. Chem. 276:7689-7692; Ostedgaard, et al., 2000, Proc. Natl. Acad. Sci. U.S.A. 97:5657-5662; and Csandy et al., 2000, J. Gen. Physiol. 116:477-500. Previous work has shown that residues 708-831 regulate activity, but in solution they are predominantly random coil. Ostedgaard, et al., 2000, Proc. Natl. Acad. Sci. U.S.A. 97:5657-5662. These studies suggest that selective deletions might not severely disrupt structure and that retention of consensus phosphorylation sites might be sufficient for PKA-dependent regulation. Importantly, several earlier studies deleted portions of the R domain without abolishing channel function. Zhang et al., 1998, Proc. Natl. Acad. Sci. U.S.A. 95:10158-10163; Rich et al., 1991, Science 253:205-207; Rich et al., 1993, Receptors Channels 1:221-232; Ma et al., 1997, J. Biol. Chem. 272:28133-28141; Vankeerberghen et al., 1999, Biochemistry 38:14988-14998; and Xie et al., 2000, Biophys. J. 78:1293-1305.
While these earlier studies suggested that a DNA molecule with R domain deletions might be of value in gene therapy applications, some alterations induced channel activity in the absence of phosphorylation, reduced the response to PKA-dependent phosphorylation, and/or reduced net channel activity. Zhang et al., 1998, Proc. Natl. Acad. Sci. U.S.A. 95:10158-10163; Ostedgaard, Baldursson & Welsh, 2001, J. Biol. Chem., 276:7689-7692; Rich et al., 1991, Science 253:205-207; Rich et al., 1993, Receptors Channels 1:221-232; Ma et al., 1997, J. Biol. Chem. 272:28133-28141; Vankeerberghen et al., 1999, Biochemistry 38:14988-14998; and Xie et al., 2000, Biophys. J. 78:1293-1305. Moreover, previous studies have only examined CFTR expressed in heterologous cell lines and studied activity using the patch-clamp technique, planar lipid bilayers, or anion efflux. There is no information, prior to this invention, about their function in airway or other epithelia. Expression in epithelia is key in assessing their value for gene transfer because deletions could alter protein-protein interactions, targeting to the apical membrane, constitutive and stimulated activity, phosphorylation-dependent regulation, and perhaps toxicity.
The present invention solves these problems by deleting regions within the CFTR R (regulatory) domain (for reviews on the R domain see Ostedgaard, Baldursson & Welsh, 2001, J. Biol. Chem. 276:7689-7692; Sheppard & Welsh, 1999, Physiol. Rev. 79:S23-S45; Gadsby & Nairn, 1999, Pysiol. Rev. 79:S77-S107; and Ma, 2000, News Physiol. Sci. 15:154-158) to provide a partially deleted CFTR capable of forming Cl⁻ channels in airway epithelia in vitro and in vivo.

BRIEF SUMMARY OF THE INVENTION

The present invention offers new therapies for treating Cystic Fibrosis (CF), that are based on novel DNA molecules and proteins encoded by the DNA molecules. The present invention features DNA molecules encoding CFTR proteins having a partially deleted R domain. The partial deletions in the R domain are between residues 708 and 835 of the wild-type CFTR.
In a preferred embodiment, the DNA molecules of the present invention encode a CFTR comprising a partially deleted R domain which is capable of normal targeting to the apical membrane, wild-type biosynthesis, and generating transepithelial Cl⁻ current in CF epithelia (see Examples below). In addition, the CFTR protein comprising a partially deleted R domain corrects the Cl⁻ transport defect in a CF subject when expression in their nasal mucosa (see Examples below). In one aspect, the CFTR comprising a partially deleted R domain provides low constitutive Cl⁻ current in CFTR channels and provides a functional chloride ion channel in CF airway epithelia cells.
In a particularly preferred embodiment of the present invention, the CFTR comprising a partially deleted R domain has a deletion selected from the group consisting of Δ708-759, Δ708-723/749-783/832-8-35 and Δ760-835.
The CFTR comprising a partially deleted R domain may also comprise deletions in other regions as long as it maintains the ability to provide a functional chloride ion channel in CF airway epithelia cells. Additional deletions may be useful in producing a DNA molecule encoding a CFTR protein which is better accommodated by a vector and to ensure efficient packaging.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention may be better understood with reference to the attached figures in which

FIG. 1A shows a graphic representation of exemplary embodiments of CFTR proteins of the present invention; 1B shows bumetanide-sensitive short-circuit current in well-differentiated CF epithelia expressing the exemplary proteins shown in panel A; 1C shows basal current;

FIG. 2 shows an example of short-circuit current in well-differentiated airway epithelia expressing wild type CFTR and GFP;

FIG. 3 shows the current from inside-out patches of membrane containing multiple CFTR channels in the presence of 1 mM ATP and PKA. 3 A is Δ708-759, and 3B is Δ708-723/749-783/832-835;

FIG. 4A depicts gels showing CFTR at indicated time after pulse with .sup.³⁵S-methionine and showing the disappearance of band B (immature) and band C (mature); 4B is a graph plotting the number of counts in band B (solid lines) and band C (dashed lines) which were determined by phosphorimaging;

FIG. 5 is an immunostaining of differentiated airway epithelia expressing exemplary embodiments of the CFTR proteins of the present invention; and

FIG. 6 shows the voltage across nasal epithelium (Vt) in CF mice expressing indicated exemplary embodiments of the CFTR proteins of the present invention in the nasal mucosa.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

The present invention is based on the surprising finding that a defective DNA molecule, namely a DNA molecule encoding a CFTR protein comprising a partially deleted R domain, expresses a protein capable of providing a functional chloride ion channel in CF epithelia cells. Based on this finding, the invention features methods for making the DNA molecules expressing CFTR protein comprising a partially deleted R domain. The DNA molecules and CFTR protein encoded thereby can be used, for example, therapeutically in CF gene and protein replacement therapies.
As used herein the following words and phrases have the meaning set forth below:
“DNA molecule” shall mean a sequence of genetic material that carries the information representing a protein.
Unless otherwise indicated, “protein” shall mean a protein, polypeptide or peptide.
“CFTR or Cystic Fibrosis Transmembrane Conductance Regulator protein” refers to a 1480 amino acid protein containing two membrane-spanning domains (MSDs), two nucleotide binding domains (NBDs) and a unique R domain, that functions as a chloride channel regulated by phosphorylation and by nucleoside triphosphates.
The phrase “cystic fibrosis transmembrane conductance regulator (CFTR) activity or function”—is meant to refer to functions normally performed by wild-type CFTR. Such functions can include mediation of ion, (e.g. chloride ion) transport across cellular membranes.
A “Cystic Fibrosis (CF) cell” is a cell that lacks cystic fibrosis transmembrane conductance regulator function. Examples include CFTR mutants of which over 1000 different varieties have been identified to date (see for example, <HTTP://genet.sickkids.on.ca>).
“R (regulator) domain” refers to a domain that keeps a chloride channel closed at rest and which opens the channel when phosphorylated (e.g. by cAMP-dependent protein kinase (PKA) or protein kinase C (PKC)). The R domain of CFTR is encoded by a portion of exon 13, and generally comprises 128 amino acid residues that span from about amino acid residues 708 to 835 of full length CFTR or a lesser portion within this stretch. Ostedgaard et al., 2000, Proc. Natl. Acad. Sci. USA 97:5657-5662.
“Partially deleted R domain” refers to deletion of part, but not all, of the R domain.
“CF gene therapy” refers to the transfer of genetic material (e.g., DNA or RNA) encoding CFTR functional activity into a host to treat or prevent Cystic Fibrosis (CF).
“CF protein replacement therapy” refers to transfer of a protein having CFTR functional activity into a host to treat or prevent CF.
The nucleotide and amino acid sequence for full-length CFTR and modifications encoding CF mutant are known in the art (See, e.g., European Patent No. 0446017). Based on this information, one of skill in the art can obtain DNA molecules encoding CFTR comprising a partially deleted R domain using techniques that are well-known. For example, DNA molecules encoding CFTR can be isolated from appropriate cells or plasmids using standard techniques (e.g. restriction enzyme cleavage). Genetic material encoding full-length CFTR can then be modified (e.g. via deletion mutagenesis using Quik Change™ Mutagenesis, Stratagene, La Jolla, Calif.) to obtain a DNA molecule encoding a CFTR comprising a partially deleted R domain. Alternatively, a DNA molecule encoding a CFTR protein comprising a partially deleted R domain can be generated synthetically using standard modes of polynucleotide synthesis. A candidate gene can be tested to determine whether it in fact encodes functional CFTR activity, for example, using the techniques detailed below in the Examples.
An “expression cassette” comprising the gene encoding a CFTR comprising a partially deleted R domain operably linked or under the control of transcriptional and translational regulatory elements (e.g. a promoter, ribosome binding site, operator or enhancer) can be made and used for expression of CFTR protein comprising a partially deleted R domain in vitro or in vivo. The choice of regulatory elements employed may vary, depending, for example, on the host cell to be transfected and the desired level of expression. Several promoters for use in mammalian cells are known in the art and include, inter alia, the phosphoglycerate (PGK) promoter, the simian virus 40 (SV40) early promoter, the Rous sarcoma virus (RSV) promoter, the adenovirus major later promoter (MLP) and the human cytomegalovirus (CMV) immediate early 1 promoter. However, any promoter that facilitates suitable expression levels can be used in the present invention. Inducible promoters (e.g., those obtained from the heat shock gene, metallothionene gene, beta interferon gene, or steroid hormone responsive genes) may be useful for regulating transcription based on external stimuli.
A preferred DNA molecule encodes a CFTR protein comprising a deletion in the R domain wherein the deletion is selected from the group consisting of Δ708-835 (SEQ ID NO:1), .Δ708-759 (SEQ ID NO:2), Δ708-723/749-783//832-835 (SEQ ID NO:3), Δ708-723/749-783/819-1-835 (SEQ ID NO:4), Δ708-759/819-835 (SEQ ID NO:5), Δ760-835 (SEQ ID NO:6), Δ708-783 (SEQ ID NO:7), and Δ708-783/823-835 (SEQ ID NO:8). A preferred CFTR protein comprises a deletion in the R domain wherein the deletion is selected from the group consisting of Δ708-835 (SEQ ID NO:9), Δ708-759 (SEQ ID NO:10), Δ708-723/749-783//832-835 (SEQ ID NO:11), Δ708-723/749-783/81-9-835 (SEQ ID NO:12), Δ708-759/819-835 (SEQ ID NO:13), Δ760-835 (SEQ ID NO: 14), A708-783 (SEQ ID NO:15), and Δ708-783/823-835 (SEQ ID NO:16). More preferably, the DNA molecule encodes a CFTR protein comprising a deletion in the R domain selected from the group consisting of Δ708-759 (SEQ ID NO:2), Δ708-723/749-783H/832-835 (SEQ ID NO:3), and Δ760-835 (SEQ ID NO:6) and the CFTR protein comprises a deletion in the R domain selected from the group consisting of Δ708-759 (SEQ ID NO: 10), Δ708-723/749-783//832-835 (SEQ ID NO: 11), and Δ760-835 (SEQ ID NO: 14). In a particularly preferred embodiment, the DNA molecule encodes a CFTR protein comprising a deletion in the R domain of a of Δ708-759 (SEQ ID NO:2) and the CFTR protein comprises a deletion in the R domain Δ708-759 (SEQ ID NO: 10). The CFTR protein of the present invention which comprises a deletion in the R domain is capable of providing a functional chloride ion channel in CF airway epithelia cells.
The DNA molecule of the present invention, and the protein encoded thereby may further comprise deletions of other regions of CFTR provided that the resultant CFTR protein is capable of providing a functional chloride ion channel in CF airway epithelia cells.
In another aspect of the invention, there is provided a DNA molecule encoding a CFTR protein comprising a partially deleted R domain wherein the encoded CFTR has low constitutive Cl⁻ current. As used herein, “low constitutive Cl⁻ current” means an amount of Cl⁻ current as determined in patch-clamp studies (described by Baldursson et. al., 2001, J. Biol. Chem. 276:1904-1910) which is less than 2 μA·cm⁻². In one embodiment, the CFTR having low constitutive Cl⁻ current is selected from the group consisting of SEQ ID NO:10, SEQ ID NO: 12, and SEQ ID NO: 13. The corresponding DNA molecule is selected from the group consisting of SEQ ID NO:2, SEQ ID NO:4 and SEQ ID NO:5.
The CFTR proteins of the present invention which comprise a partially deleted R domain can be made by introducing the DNA molecules of the present invention into cells in culture using standard techniques (e.g. via calcium phosphate or calcium chloride co-precipitation, or via infection with a recombinant virus, such as a recombinant adenovirus, comprising the DNA molecule, DEAE dextran mediated transfection, lipofection, or electroporation). Recombinant cells can then be cultured in vitro in a manner that allows expression of the CFTR proteins of the present invention. Preferred host cells for generating the CFTR proteins of the present invention include, inter alia, mammalian cells, such as HeLa cells, COS cells, C127 cells; yeast cells, insect cells and bacterial cells.
The CFTR proteins of the present invention which comprise partially deleted R domains can be purified from host cell membranes using known methods, such as ion exchange chromatography, gel filtration chromatography, electrophoresis and affinity chromatography. (Tilly et. al., 1992, The Journal of Biological Chemistry 267:9470-73). A preferred method of purification involves first solubilizing the protein in the presence of a nondenaturing detergent.
The CFTR proteins of the present invention comprising partially deleted R domains produced as described herein can be used, for example, in protein replacement therapies and the DNA molecule in gene therapies for Cystic Fibrosis as described in detail below.
Protein therapy may be accomplished by any method that effectively introduces the CFTR protein of the present invention into the membrane of CF defective cells to imbue on those cells CFTR activity. An effective amount of a CFTR protein of the present invention comprising a partially deleted R domain (i.e. an amount sufficient to reduce or eliminate the symptoms associated with CF and/or to provide a functional chloride ion channel in CF airway epithelia cells) can be administered alone or in association with an agent that facilitates passage (e.g. via fusion or endocytosis) through cell membranes to CF patients (i.e. patients having CF defective cells). The “effective amount” can be determined by one of skill in the art based on such factors as the type and severity of symptoms being treated, the weight and/or age of the subject, the previous medical history of the subject, and the selected route for administration of the agent.
Preferably for use in protein therapy, the CFTR proteins comprising partially deleted R domains are associated with lipids, such as detergents or other amphipathic molecule micelles, membrane vesicles, liposomes, virosomes, or microsomes. Lipid compositions that are naturally fusogenic or can be engineered to become fusogenic (e.g. by incorporating a fusion protein into the lipid) are especially preferred. Fusion proteins can be obtained from viruses such as parainfluenza viruses 1-3, respiratory syncytial virus (RSV), influenza A, Sendai virus, and togavirus fusion protein. Nonviral fusion proteins include normal cellular proteins that mediate cell-cell fusion. Other nonviral fusion proteins include the sperm protein PH-30 which is an integral membrane protein located on the surface of sperm cells that is believed to mediate fusion between the sperm and the egg. See Blobel et al., 1992, Nature 356:248-251. Still other nonviral fusion proteins include chimeric PH-30 proteins such as PH-30 and the binding component of hemaglutinin from influenza virus and PH-30 and a disintegrin (e.g. bitistatin, barbourin, kistrin, and echistatin). In addition, lipid membranes can be fused using traditional chemical fusogens such as polyethylene glycol (PEG).
A CF patient can be treated by administration of an effective amount of a CFTR protein comprising a partially deleted R domain, optionally in a pharmaceutically acceptable carrier or diluent. An effective amount of a CFTR protein comprising a partially deleted R domain is an amount sufficient alleviate the symptoms of CF and/or an amount to provide a functional chloride ion channel in CF airway epithelia cells. A CFTR protein comprising a partially deleted R domain can be administered subcutaneously, intravenously, intraperitoneally, intramuscularly, parenterally, orally, submucosally, by inhalation, or other appropriate route of administration in an effective dosage range. A preferred route of administration is by inhalation (e.g. of an aerosolized pharmaceutical composition). If necessitated by a particular mode of administration, CFTR proteins comprising partially deleted R domains can be encapsulated within a material that protects it from enzymatic degradation. In addition, prior to administration, it may be useful to administer agents to clear mucus (e.g. using a DNAse) and/or bacterial infection.
Alternatively, a preparation of the gene encoding a CFTR protein comprising a partially deleted R domain can be incorporated into a suitable vector for delivering the gene into a CF patient's defective cells. As many of the symptoms of CF manifest themselves in the respiratory tract, the preparation can be delivered directly to the airways of CF patients.
The first generation of CF gene therapy may be transient and may require repeated delivery to the airways. Eventually, however, gene therapy may offer a cure for CF when the identity of the precursor or stem cell to air epithelial cells becomes known. If genetic material encoding CFTR proteins comprising partially deleted R domains were incorporated into airway stem cells, all subsequent generations of such cells would make authentic CFTR protein comprising a partially deleted R domain from the integrated sequences and would correct the physiological defect almost irrespective of the biochemical basis of the action of CFTR.
For use in treating CF, appropriate vectors must: 1) effectively infect lung epithelia or other tissue manifesting the disease and deliver the therapeutic nucleic acid encoding CFTR function; 2) be appropriately maintained in host cells; and 3) be safe. The following describes a number of approaches and vectors that may prove useful for performing CF gene therapy. The following listing, however, is not intended to be exhaustive and many other vectors should prove useful for performing gene therapy with the novel genes disclosed herein.
Retroviruses—Although defective retroviruses are one of the best characterized system (Miller, A. D., 1990, Blood 76:271), the major issue in relation to CF is the requirement for dividing cells to achieve DNA integration and gene expression. Were conditions found to induce airway cell division, the in vivo application of retroviruses, especially if repeated over many years, would necessitate assessment of the safety aspects of insertional mutagenesis in this context.
Adeno-Associated Virus—(AAV) is a naturally occurring defective virus that requires other viruses such as adenoviruses or herpes viruses as helper viruses (Muzyczka, N., 1992, Current Topics in Microbiology and Immunology 158:97). It is also one of the few viruses that may integrate its DNA into non-dividing cells. Vectors containing as little as 300 base pairs of AAV can be packaged and can integrate, but space for exogenous DNA is limited to about 4.5 kb. AAV vectors therefore may prove useful for expressing genes encoding the CFTR proteins of the present invention comprising partially deleted R domains, although genes encoding fall length CFTR approach AAV's upper limit. For reviews see Flotte, 1999, Curr. Opin. Mol. Ther. 1:510-516; Carter & Samulski, 2000, Int. J. Mol. Med. 6:17-27; and Athanasopoulos & Dickson, 2000, Int. J. Mol. Med. 6:363-375. AAV has already been successfully used to produce Factor IX in humans with hemophilia B. In AAV vectors, viral genes are deleted, thereby minimizing cell-mediated immune responses. AAV vectors can transduce non-dividing cells, such as airway epithelia. And DNA molecule expression can be prolonged. Although, most previous studies have used type 2 AAV vectors, its receptor is on the basolateral membrane and thus inaccessible to vector applied apically. See, Summerford & Samulski, 1998, J. Virol. 72:1438-1445. Recent studies have discovered that type 5 AAV can efficiently transduce well-differentiated human airway epithelia, and that its receptor lies on the apical membrane. See Zabner et al., 2000, J. Virol. 74:3852-3858; Walters et al. 276:20610-20616 Type 6 AAV is also a promising vector for airway epithelia. See Halbert, Allen & Miller, 2001, J. Virol. 75:6615-6624.
Naked DNA—Naked plasmid can be introduced into muscle cells by injection into the tissue. Expression can extend over many months but the number of positive cells is low (Wolff, J. et al., 1989, Science 247:1465).
DNA-Lipid Complexes—Lipid carriers can be associated with naked DNA (e.g. plasmid DNA) to facilitate passage through cellular membranes. Cationic, anionic, or neutral lipids can be used for this purpose. However, cationic lipids are preferred because they associate better with DNA, which generally has a net negative charge. Cationic lipids have been shown to mediate intracellular delivery of plasmid DNA (Felgner, P. and Ringold, G. M., 1989, Nature 337:387). Injection of cationic lipid plasmid DNA complexes into the circulation of mice has been shown to result in expression of the DNA in lung (Brigham, K. et al., 1989, Am. J. Med. Sci. 298:278). Instillation of cationic lipid plasmid DNA into lung has also been found to be expressed in epithelial cells but the efficiency of expression has been reported as being relatively low and transient (Hazinski, T. A. et al., 1991, Am. J Respir., Cell Mol. Biol. 4:206).
Receptor Mediated Entry—In an effort to improve the efficiency of plasmid DNA uptake, attempts have been made to utilize receptor-mediated endocytosis as an entry mechanisms and to protect DNA in complexes with polylysine (Wu, G. and Wu, C. H., 1988, J. Biol. Chem. 263:14621). One potential problem with this approach is that the incoming plasmid DNA enters the pathway leading from endosome to lysosome, where much incoming material is degraded. One solution to this problem is the use of transferrin DNA-polylysine complexes linked to adenovirus capsids (Curiel, D. T. et al., 1991, Proc. Natl. Acad. Sci. USA 88:8850). The latter enter efficiently but have the added advantage of naturally disrupting the endosome thereby avoiding shuttling to the lysosome.
Adenovirus—Defective adenoviruses may also be useful for CF gene therapy (Berkner, K. L., 1988, BioTechniques 6:616). Adenovirus can be manipulated such that it encodes and expresses the desired gene product, and at the same time is inactivated in terms of its ability to replicate in a normal lytic viral life cycle. In addition, adenovirus has a natural tropism for airway epithelia. The viruses are able to infect quiescent cells as are found in the airways, offering a major advantage over retroviruses. Adenovirus expression is achieved without integration of the viral DNA into the host cell chromosome, thereby alleviating concerns about insertional mutagenesis. Furthermore, adenoviruses have been used as live enteric vaccines for many years with an excellent safety profile (Schwartz, A. R. et al., 1974, Am. Rev. Respir. Dis. 109:233-238). Finally, adenovirus mediated gene transfer has been demonstrated in a number of instances including transfer of alpha-1-antitrypsin and CFTR to the lungs of cotton rats (Rosenfeld, M. A. et al., 1991, Science 252:431-434; Rosenfeld et al., 1992, Cell 68:143-155). Furthermore, extensive studies to attempt to establish adenovirus as a causative agent in human cancer were uniformly negative (Green, M. et al., 1979, Proc. Natl. Acad. Sci. USA 76:6606).
A first generation adenovirus encoding full length CFTR has been prepared and includes viral DNA derived from the common relatively benign adenovirus 2 serotype. A similar vector can be prepared to express CFTR proteins comprising partially deleted R domains. The E1a and E1b regions of the viral genome, which are involved in early stages of viral replication have been deleted. Their removal impairs viral gene expression and viral replication. The protein products of these genes also have immortalizing and transforming function in some non-permissive cells.
The following properties would be desirable in the design of a viral vector to transfer the gene for a CFTR protein comprising a partially deleted R domain to the airway cells of a CF patient. The vector should allow sufficient expression of the CFTR protein, while producing minimal viral gene expression. There should be minimal viral DNA replication and ideally no virus replication. Finally, recombination to produce new viral sequences and complementation to allow growth of the defective virus in the patient should be minimized.
The present invention is further illustrated by the following examples which in no way should be construed as being further limiting.

EXAMPLES

Example 1

Construction of CFTR Variants

DNA molecules encoding exemplary embodiments of CFTR proteins comprising partial deletions in the R domain were made in pTM1-CFTR4 by PCR deletion mutagenesis (Quik Change Mutagenesis™, Stratagene, La Jolla, Calif.) and confirmed by sequencing. Constructs were ligated into an adenovirus serotype 5 vector in which the CMV promoter drives cDNA expression. The exemplary CFTR proteins were named by the residues that were deleted; for example in Δ708-835, residues between and including aa 708 and 835 are deleted. An identical adenovirus expressing green fluorescent protein (GFP) was used as a negative control. FIG. 1A shows the eight variants constructed which include, Δ708-835 (SEQ ID NO:9), Δ708-759 (SEQ ID NO: 10), Δ708-723/749-783/832-835 (SEQ ID NO:11), Δ708-723/749-783/819-835 (SEQ ID NO:12), Δ708-759/819-835 (SEQ ID NO:13), Δ760-835 (SEQ ID NO: 14), Δ708-783 (SEQ ID NO:15), and Δ708-783/823-835 (SEQ ID NO: 16). FIG. 1A indicates the deletions by crosshatching. Serines that are phosphorylated in vivo are indicated in FIG. 1A with residue number at the top. First and last residue of deleted regions are indicated above each construct. The number of nucleotides deleted in each variant is shown on the right of FIG. 1A.

Example 2

Protein Biochemistry

To confirm protein size and phosphorylation, HeLa cells were infected with 200 MOI of recombinant adenovirus in Eagles minimal essential media (EMEM) for 45 min. Cells were lysed 18-24 hr later, CFTR immunoprecipitated, and phosphorylated with γ-³²P-ATP and the catalytic subunit of PKA as described previously. Baldursson et al., 2001, J. Biol. Chem. 276:1904-1910. For pulse chase studies, HeLa cells were infected as above, and after 18-24 hr cells were methionine starved, labeled with ³⁵S-methionine, and pulse-chase studies carried out as described previously (Ostedgaard, Zeiher & Welsh, 1999, J. Cell Sci. 112:2091-2098. Proteins were separated on 8% SDS-PAGE, stained, destained, dried and exposed to phosphorscreens. After phosphorimaging, counts in bands B (immature) and C (mature) were quantitated. FIGS. 4A&B show that two representative CFTR proteins of the present invention comprising partially deleted R domains, namely Δ708-759 and Δ708-723/749-783/832-835, demonstrate similar disappearance of band B and appearance of band C as wild type. FIG. 4A shows the appearance in a gel and FIG. 4B is the quantitation of the bands from 3-4 experiments. Band B is shown as counts relative to counts at time=0; band C is shown as counts relative to counts at time=0.5 hr. (n=3-4 for all points.)

Example 3

Well-Differentiated CF Airway Epithelia

Cultures of human airway epithelia were obtained from CF bronchus (μF508/μF508 or μF508/other genotypes) and cultured at the air-liquid interface as previously described (Karp et al., 2002, Epithelial Cell Culture Protocols, ed. Wice (Human, Totowa, N.J.) 188:115-137, incorporated herein by reference. Epithelia were used at least 14 days after seeding when they were well-differentiated with a surface consisting of ciliated cells, goblet cells and other non-ciliated cells. They also retained the functional properties of airway epithelia including transepithelial electrolyte transport and resistance. FIG. 2 shows the short circuit current in well-differentiated airway epithelia in the presence of wild-type CFTR and GFP, demonstrating that wild-type CFTR can provide a functional chloride ion channel in CF airway epithelia. Bars at top of FIG. 2 indicate additions to solutions (detailed below in Example 5). Zero current level is shown by dashed line.
Epithelia were infected with 200 MOI adenovirus vector using 5 mM EGTA applied to the apical surface to transiently disrupt the tight junctions as previously described (Walters et al., 1999, J. Biol. Chem. 274:10219-10226.

Example 4

Immunocytochemistry

Three days following gene transfer, epithelia were fixed with 4% paraformaldehyde, permeabilized with 0.2% Triton X-100, blocked with 5% normal goat serum in SuperBlock (Pierce, Rockford, Ill.), and stained with anti-CFTR (24-1, R&D Systems, Minneapolis, Minn.) and anti-ezrin primary antibodies. Appropriate Alexa Fluor-conjugated secondary antibodies were then applied and epithelia were examined by confocal laser scanning microscopy. FIG. 5 shows X-Z confocal image reconstructions.

Example 5

Ussing Chamber Studies

Three days following gene transfer, short-circuit current was measured in symmetrical solutions containing: 135 mM NaCl, 1.2 mM MgCl₂, 1.2 mM CaCl₂, 2.4 mM K₂PO₄, 0.6 mM KH₂PO₄, 5 mM dextrose and 5 mM Hepes, pH 7.4, as previously described (Zabner et al., 1998, Mol. Cell. 2:397-403. After measuring baseline current, mucosal amiloride (10⁻⁴M), mucosal 4,4′-diisothiocyanoto-stilbene-2,2′-disulfonic acid (DIDS, 10⁻⁴M); the cAMP agonists mucosal forskolin (10⁻⁵M) plus 3-isobutyl-2-methylxanthine (IBMX, 10⁻⁴M), and submucosal bumetanide (10⁻⁴M) were sequentially added (see FIG. 2). For a limited number of studies, epithelia were treated with forskolin (10⁻⁵M) and IBMX (10⁻⁴M) for 24 hr prior to study in Ussing chambers to minimize basal CFTR current.

Example 6

Patch-Clamp Studies

The methods, solutions, and procedures for excised, inside-out patch-clamp recording were identical to those previously described (Carson, Travis & Welsh, 1995, J. Biol. Chem. 270:1711-1717). Patches containing multiple CFTR channels were studied at room temperature (˜24° C.) in the presence of 1 mM ATP±75 nM PKA added to the bath solution. Membrane voltage was clamped at −40 mV; data were filtered at 100 Hz and digitized at 250 Hz.

Example 7

Nasal Voltage Study in CF Mice

For in vivo analysis, we used 6-8 wk old ΔF508 homozygote CF mice (Zeiher et al., 1995, J. Clin. Invest. 962051-2064. Mice were lightly anesthetized in a halothane chamber. Adenovirus vectors (5.times.10⁹particles) were administered intranasally as Ad:CaPi coprecipitates (Fasbender et al., 1998, J. Clin. Invest. 102:184-193) in two 5 μl instillations delivered 5 min apart. Four days later animals were anesthetized with ketamine and xylazine and the transepithelial electric potential difference across the nasal epithelium (Vt) was measured as previously described (Zeiher et al., 1995, J. Clin. Invest. 96:2051-2064). During measurement of Vt, the nasal mucosa was perfused at a rate of 50 μl/min with a Ringer's containing (in mM) 135 NaCl, 2.4 KH₂PO₄, 0.6 K₂HPO₄, 1.2 CaCl₂, 1.2 MgCl₂, and 10 HEPES (pH 7.4 with NaOH). Three solutions were used: a) Ringer's containing 100 μM amiloride; b) Ringer's containing 135 mM Na-gluconate substituted for NaCl plus amiloride; and c) Na-gluconate Ringer's containing 10 μM isoproterenol and amiloride. Measurements were made after perfusion for 5 min.

Example 8

Results

A. Generation of CFTR with R Domain Deletions
Portions of the R domain were selectively deleted based on known PKA motifs and earlier structure and function studies (Ostedgaard, Baldursson, & Welsh, 2001, J. Biol. Chem. 276:7689-7692; Sheppard & Welsh, 1999, Physiol. Rev. 79:S23-S45; and Gadsby & Nairn, 1999, Physiol Rev. 79:S77-S107). Because previous work showed that residues 708-835 are the largest deletion that yields a functional channel in mammalian cells (Rich et al., 1993, Receptors Channels 1:221-232), deletions were made in this region. In addition, constructs were produced that retained different numbers of the phosphoserines. FIG. 1A shows the deletion constructs. The cDNA for each variant was inserted into a recombinant adenovirus vector. Infection of HeLa cells produced approximately equivalent amounts of protein of the predicted size; it was recognized by CFTR antibodies and was phosphorylated in vitro by the catalytic subunit of PKA.
B. Function of R Domain Variants in Well-Differentiated CF Airway Epithelia
To determine whether the R domain variants can complement the CF Cl⁻ transport defect, the variants were expressed in well-differentiated CF airway epithelia and the short-circuit current response to several interventions was measured. FIG. 2 shows the interventions and an example of the currents. The following were sequentially added: a) amiloride to inhibit apical Na⁺ channels, hyperpolarize the apical membrane, and thereby generate a driving force for Cl⁻ secretory currents; b) DEDS to inhibit DIDS-sensitive apical Cl⁻ channels; c) cAMP agonists to activate CFTR; and d) bumetanide to inhibit basolateral Cl⁻ co-transport. Under these conditions, bumetanide-sensitive current provides the most accurate assessment of CFTR-dependent transepithelial Cl⁻ transport.
All the CFTR variants produced transepithelial Cl⁻ currents (FIG. 1B). The data in FIG. 1B represent the difference in current generated by adding bumetanide corrected for current in GFP expressing epithelia and normalized to current generated by wild type CFTR. Bumetanide-sensitive current for epithelia expressing wild type CFTR was 20.3±1.6 μA·cm⁻². The asterisks in FIG. 1B indicate the value different from wild type (p<0.05, one way ANOVA) (n=18 for wild type and 6-15 for each variant).
Because the constructs in CF epithelia obtained from multiple different lungs were tested, in each culture the responses of the variants were compared to epithelia expressing GFP (as a negative control) and then normalized current to the response of wild-type CFTR. As shown in FIG. 1B, the Δ708-835 variant generated the least Cl⁻ current, consistent with patch-clamp studies showing that this channel has a low open state probability ((Winter & Welsh, 1997, Nature 389:294-296; and Rich et al., 1993, J. Biol. Chem. 268:20259-20267). Two variants generated current similar to wild-type CFTR: Δ708-759 and Δ708-723/749-783/832-835 (FIG. 1B). The other variants produced intermediate levels of Cl⁻ current (FIG. 1B).
Amiloride-inhibited current has been reported to be increased in CF epithelia (Boucher, 1994, Am J. Respir. Crit. Care Med. 150:271-281; Schweibert et al., 1999, Physiol. Rev. 79:S145-S166). However, the responsible mechanism remains uncertain and a direct effect of CFTR on the Na⁺ currents has not been uniformly observed (Schweibert et al., 1999, Physiol. Rev. 79:S145-S166; and Nagel et al., 2001, EMBO Rep. 2:249-254). Prior studies showed limited and variable effects on Na⁺ current. However, in the present invention, amiloride-inhibited current is influenced not only by the activity of epithelial Na⁺ channels, but also by the basal Cl⁻ current which is increased when amiloride hyperpolarizes the apical membrane. Moreover, in the present invention, there was no control for the percentage of cells infected in different experiments. Although gene transfer to 5-10% of cells is sufficient to correct the CF Cl⁻ transport defect (Davies, Geddes, & Alton, 2001, J. Gene Med. 3:409-417; Flotte, 1999, Curr. Opin. Mol. Ther. 1:510-516; and Welsh, 1999, J. Clin. Invest. 104:1165-1166), alteration of Na⁺ current may depend on the percentage of infected cells over a wide range (Johnson et al., 1995, J. Clin. Invest 95:1377-1382).
Patch-clamp studies in heterologous cells demonstrated that some of the CFTR proteins of the present invention comprising partially deleted R domains opened even without PKA phosphorylation; i.e., they were constitutively active (Ostedgaard, Baldursson, & Welsh, 2001, J. Biol. Chem. 276:7689-7692). To assess constitutive activity, epithelia were first treated with cAMP agonists for 24 hr prior to mounting them in Ussing chambers; this treatment minimizes basal CFTR Cl⁻ channel activity. Then the current remaining after treatment with amiloride and DIDS, but before addition of cAMP agonists was measures, as shown in FIG. 1C, and FIG. 2). In FIG. 1C, the basal current was measured in the presence of amiloride and DIDS and corrected for current in epithelia expressing GFP. All epithelia were pre-treated with cAMP agonists for 24 hr. Asterisks in FIG. 1C indicate values different from wild type (p<0.05, one way ANOVA) (n=3-6 for each construct).
Interestingly, Δ708-835 produced a large basal current, consistent with previous patch-clamp studies showing that it generates significant constitutive but little total Cl⁻ current. Wild type and the other CFTR proteins of the present invention comprising partially deleted R domains surprisingly showed low basal/constitutive current. In one embodiment of the present invention, such low basal/constitutive current is preferred.
C. Constitutive Activity of CFTR with R Domain Deletions
To test further for constitutive activity, we examined the two variants generating the largest Cl⁻ currents in airway epithelia by expressing them in HeLa cells and measuring activity in excised, inside-out patches. Consistent with the transepithelial studies shown in FIG. 1C, FIG. 3 shows that Δ708-723/749-783/832-835, but not Δ708-759 generated constitutive current. Specifically, FIG. 3A shows that Δ708-759 showed no current before phosphorylation with PKA and FIG. 3B shows that Δ708-723/749-783/832-835 activity was stimulated with ATP alone. The ratio of current with ATP alone to the maximal current with PKA and ATP was 0.22±0.01, n=4.
D. Biosynthesis and Localization of the R Domain Variants
The glycosylation state of CFTR traces its progress through the biosynthetic pathway (Cheng et al., 1990, Cell 63:827-834). In the endoplasmic reticulum, CFTR appears as a partially glycosylated intermediate, band B (immature). In the Golgi complex, the protein becomes fully glycosylated, appearing as band C (mature); this is the form that traffics to the plasma membrane. A pulse-chase analysis to assess biosynthesis of the R domain variants was used. FIG. 4 shows results for wild-type CFTR and three of the CFTR proteins comprising partially deleted R domains of the present invention. The rates at which band B disappeared and band C appeared were similar for each of the CFTR proteins of the present invention and wild type.
CFTR resides in the apical membrane of non-CF epithelia where it provides a pathway for Cl⁻ flow (Welsh et al., 2000, The Metabolic and Molecular Basis of Inherited Disease, eds. Scirver, Beaudet, Sly, Valle, Childs & Vogelstein (McGraw-Hill, New York) pp 5121-5189); an apical location is critical for its function in transepithelial Cl⁻ transport. The exemplary embodiments of CFTR proteins comprising partially deleted R domains of the present invention were expressed in well-differentiated CF airway epithelia, immunostained CFTR, and the pattern of fluorescence using confocal microscopy was examined. All the constructs showed the same apical localization as wild type CFTR; FIG. 5 shows examples for Δ708-759 and Δ708-723/749-783/832-835. In FIG. 5, data are X-Z confocal images. Arrows indicate the position of the apical membrane and the top of the filter support. Anti-CFTR immunostaining is green and anti-ezrin staining is red. Ezrin stains the apical region of the epithelial cells.
E. In Vivo Function of R Domain Variants in the Nasal Epithelia of CF Mice
As an additional test of their combined biosynthesis, localization, and functional activity, the variants were tested in an art recognized animal model in vivo. Zeiher et al., 1995, J. Clin. Invest. 96:2051-2064. Nasal epithelia of CF mice were infected with adenovirus vectors expressing wild type and the two CFTR proteins of the present invention that generated the largest Cl⁻ current in human airway epithelia. Epithelia were treated with amiloride to inhibit Na⁺ channels and then Vt in response to perfusion was measured with solutions containing a low Cl⁻ concentration and isoproterenol to elevate cellular cAMP levels. As shown in FIG. 6, Expression of Δ708-759 and Δ708-723/749-783/832-835 corrected the nasal voltage defect to a similar extent as wild-type CFTR and to levels similar to those previously observed in non-CF mice (Zeiher et al., 1995, J. Clin. Invest. 96:2051-2064). In FIG. 6, values of Vt obtained from untreated CF and wild type mice are indicated by dashed lines. The three interventions are indicated at the bottom of FIG. 6 (n=13 for wild type and 14 for the deleted variants).
The data show that CFTR constructs with multiple R domain deletions retain normal biosynthesis, apical targeting and Cl⁻ channel function when expressed in differentiated CF airway epithelia. These results have implications for developing CF gene therapy and for understanding CFTR structure and function.
The data also establish the feasibility to generate a smaller CFTR DNA molecule to accommodate the limited packaging capacity of AAV in vitro and in vivo. For optimal use in an AAV vector for CF gene therapy, the DNA molecule would have two characteristics. The DNA molecule would be short to facilitate packaging, and the protein product would correct the CF defect to the same extent as wild type CFTR.
Of these three constructs, Δ708-759 most closely resembled wild type, in that it produced no constitutive Cl⁻ current. In contrast, Δ708-723/749-783/832-835 and Δ708-759/819-835 had greater, but still low, basal Cl⁻ currents than wild type and showed constitutive Cl⁻ current when examined in patch-clamp studies.

Example 9

Structure of CFTR

The CFTR proteins of the present invention which comprise partially deleted R domains provide insight into CFTR structure. All the CFTR proteins comprising partially deleted R domains of the present invention were capable of targeting exclusively to the apical membrane, indicating that important apical targeting motifs are not likely located within this region of the R domain. CFTR proteins of the present invention also showed normal biosynthesis, suggesting that sequences in the deleted regions are not required for normal processing. Apical targeting and biosynthesis were also normal with and without constitutive activity, suggesting that Cl⁻ channel activity may not influence these processes. Other studies have shown that R domain deletions and missense mutations in this region of the R domain generate band C (mature) protein (see Vankeerberghen et al., 1999, Biochemistry 38:14988-14998; and Vankeerberghen et al., 1998, Hum. Mol. Genet. 7:1761-1769). Other studies have also shown that elimination of a single arginine-framed motif (residues 764-766) did not impair processing (Chang et al., 1999, Mol. Cell 4:137-142).
The CFTR proteins comprising partially deleted R domains of the present invention reveal several aspects of R domain function. a) Length. In general, the more the R domain deleted, the less the Cl⁻ current. However, length alone does explain the results as evidenced by the finding that Δ708-783/823-835 (267 bp deleted) had as much current as Δ708-723/749-783/819-835 (204 bp deleted). b) Specific phosphoserines. Although all of the CFTR proteins of the present invention retained Ser660 and Ser700, the number of additional phosphoserines failed to predict the amount of current. For example, Δ760-835 (with one additional phosphoserine) had at least as much current as Δ708-783 (two additional phosphoserines) and Δ708-723/749-783/819-835 (three additional phosphoserines). These results are consistent with previous work suggesting that not all the phosphoserines are necessary for activity and no one phosphoserine is dominant. c) Charge. No correlation between Cl⁻ current and net charge present within the region between aa 708 and 835 was found. d) Ser737. Mutation of Ser737 suggested it has an inhibitory function on CFTR studied in Xenopus oocytes (Wilkinson et al., 1997, Am J. Physiol. 273:L127-L133). In airway epithelia, the CFTR proteins comprising partially deleted R domains of the present invention did not reveal inhibition. e) Residues 817-838. This stretch of negatively charged amino acids has been suggested as a stimulatory region (Xie et al., 2000, Biophys. J. 78:1293-1305). Deletion of this region decreased current in Δ708-723/749-783/819-835 compared to Δ708-723/749-783/832-835. However, deletion of this region in Δ708-783/823-835 did not reduce current as compared to Δ708-783. f) Residues 760-783. It was previously suggested that these residues prevented constitutive activity (Baldursson et al., 2001, J. Biol. Chem. 276:1904-1910). The present invention provides support for this hypothesis. g) Structure. The ability to alter the sequence of the R domain in so many different ways and yet retain Cl⁻ channel function and phosphorylation-dependent activity supports the hypothesis that there are few or no required structural motifs in this portion of the R domain. That conclusion is consistent with the recent finding that this region of the R domain is predominantly random coil (Ostedgaard et al., 2000, Proc. Natl. Acad. Sci. 97:5657-5662).

Claims

1. A CFTR protein having a deletion in the R domain selected from the group consisting of SEQ ID NO:10.

2. A CFTR protein having a deletion in the R domain, wherein said CFTR protein provides low constitutive chloride ion current in CFTR channels, and wherein said CFTR provides a functional chloride ion channel in CF airway epithelia cells, and comprising a CFTR protein with a deletion of amino acids 708-759.

3. The CFTR protein of claim 2, wherein said CFTR protein is selected from the group consisting of SEQ ID NO:10.

4. A CFTR protein comprising approximately 95% homology to the protein of SEQ ID NO:10 wherein said CFTR protein has bases 708-759 deleted from wild-type CFTR sequence.