5 Figure 48 illustrates the failure of CAMP or torskolin in stimulating membrane arrrents in a PLJ done 6 cell.
Figure 4C is a graph depicting the instantaneous current-voltage relations of forskoiin-induced currents in NaCI, k~w CI and Na-tree baths.
Figure 5 depicts a stabilization scheme for a CFTR construct.
Figure 6 is~ a restriction map of a plasmid-based vector used in the practice of the present invention.
DETAILED DESCRIPTION OF THE
PREFERRED EMBODIMENTS OF THE INVENTION
The abs~rnce of functional CFTR or CFTR function which is not at physiologically-acxeptable levels and which arises from a defect in the CFTR
gene is treated through gene transfer of a normal CF'TR gene into CFTR defective cells. By physiologically-acceptable level of CFTR function' is meant a level of CFTR
function at which a cell population or patient exhibits the normal physiological effects presence of the normal amounts of CFTR. Examples of insuffidencies in CFTR function include but are not limited to abnormal CI channel regulation in epithelial cells, such as that exhibited in cystic fibrosis.
A recombinant viral vector of the present invention comprises DNA of at least a portion of retrovral genome which portion is capable of infecting the target cells and a normal CFTR gene operatively linked to thereto. By 5nfection' is generally meant the process by which a virus transfers genetic material to its host or target cell. Preferably the retrovirus used in the construction of a vector of the invention is also rendered roptication-defecflive to remove the effects of viral replication on the target cells. In such cases, the replication-defective viral genome can be packaged by a helper virus in accordance v~rtth conventional techniques. Generally any retrovirus meeting the above criteria of irrfectiausness and capabilities of CFTR gene transfer can be employed in the practice of the present invention may also be desirable.
Suitable retroviruses for the practice of this invention indude, for example, PLJ, pZip, pWe and SUBSTITUTE SHEET
s WO 92/05273 i?CT/US91 /06660 _ 6 pEM weA kncnvn to those skilled in the art. Suitable packaging virus lines for repiicatlon-defective retroviruses indude, for example, ~Crip, ~Cre and ~2 and ~Am.
!t will be appredated that when viral vector schemes are employed for CFTR
transfer, the use of attenuated or aviruient viruses may also be desirable.
WhFre applicable in the practice of the invention, amplfication of the CFTR gene can also be utilized to enh~ence the levels of normal CFTR expression.
The genetic material to be recombined with the retroviral vector or transferred through other methods of the invention is preferably provided through conventional Boning methods, i.e. cDNA, through overlapping oligonucteotide sequences or any other suitable method yielding the desired sequence. When used in diagnostic or screening assays, the genetic material is usually provided by cloning of patient DNA
or, alternatively, through the use of patient genomic DNA. As stated previously, by normal CFTR gene, is meant any nucleic acid sequence which codes for functional CFTR.
The cells targeted for transduction or gene transfer in accordance with the present invention include any cells to which the delivery of the CFTR gene is desired.
Generally speaking, the cells are those with the CFTR gene defect, such as CF
cells.
In the case of CF, the cells targeted are preferably epithelial cells, including pancreatic, sweat gland, liver, intestinal, kidney and even more preferably epithelial airway cells, such as lung cells.
Cells or cell populations can be treated in accordance with the present invention in vnvo or in irr vitro. For example, in in vivo treatments, CFTR
vectors of the present invention can be administered to the patient, preferably in a biologically compatible solution or pharmaceutically acceptable delivery vehicle, by ingestion, injection, inhalation or any number of other methods. The dosages administered will vary from patient to patient and will be determined by the level of enhancement of CFTR function balanced against any risk or deleterious side effects.
Monitoring levels of transduclion, CFTR expression and/or the presence or levels of normal CFTR
will assist in selecting and adjusting the dosages administered. in vitro transduction is also contemplated within the present invention. Cell populations with defective CFTR genes can be remo~red from the patient or otherwise provided, transduced with a normal CFTR gene in accordance with the principles of the invention, then (re)introduced into the patient CFTR-defective cell lines, such as transformed CF lines, can also be transduced in accordance with the present invention. Such cell lines are useful, for example, in SUBSTITUTE $~
WO 9r. i?SZ73 ~ ~ ' PCT/L?S91 /06660 complementation assays for waluating CF mutations to diagnose CF and screen for carriers. For example, patient CFTR cONA can be transferred into CF aalts and the Celts screened for complementation, i.e. CFTR function, to confirm or rule out of CF or CFTR gene defects.
tin the first set of SpeciOc Examples which follow, rstrovirus-mediated gene _ transfer was used 'to complement the cystic fibrosis (CF) defect in CI
regulation in epithelial Celts of a CF patient. Amphotropic retroviruses were used to transduce a functional cystic fibrosis transmembrane conductanaa regulator (Cl-TR) cDNA
into CFPAC-t, a pancxe.stic adenocaranoma aaU Gne derived from a patient with CF.
This cell line stably expresses abnom~aiitiss in electrolyte transport that are dlaracteristic of the CF defect, i.e. they iadK CAMP-stimulated CI transport CFPAC-1 aalls were exposed to corttroi virus (Pt.,n and CFTR-expressing vivre (PLJ-CFTH); viral transduaad Bones were isolated and subjected to molecular and physiologic analysis.
Agarose gel blot analysis re~realed unrearranged proviral sequsnaas in 10 of 10 PLJ
clones and 0 of 10 PLJ-CFTR doves. I~NA analysis detected a viral-derived CFTFi transcript in all of the PLJ-CFTFi Bones. t~NA analysis detected a viral-derived CFTR transcript in all of the PL,I-CFTR dlonss that contained unrearranged proviral sequences.
Mion (~~I) efflux was used to examine PL! and PLJ-CFTR Bones for CAMP
and Ca stimulated anion transport Agents that increase intracellular CAMP
stimulated ~~I efflux in PLJ-CFTR Bones but not PLJ doves. Whole-cell patch-clamp performed on two respondin~l doves showed that the anion efflux responses were due to CAMP
stimulation of d~loride conductance. Calcium ionophore incxeased ~~1 efflux and chloride arrrents in ail PLJ and PLJ-CFTR doves. These findings indicate that expression of the normal CFTR gene confers CAMP-dependent CI channel regulation on CF epithelial aails.
The second set of Speafic Examples describes gene transfer to airway epithelial aalts as well as pancreatic calls, administration of CFTR gene therapy and aitemabve gene 'transfer delivery systems, induding fipofection and DNA-protein complex- mediated gene transfer.
. 3p BPECIF1C EXAMPLES ~ I -Recomblnartt Reims Earty attertipts to reconstitute a tuU length CFTR cONA from overlapping Bones were unsuxessfut. The exact crease of these ditficu~ies remains to be defined, but our data indicated that prokaryotic transcription from internal CFTR cONA
sequences may result in the expression of a protein that is toxic to bacteria. The introduction of three sua~ stir i ._._._~.. . .~ _.v...._....~._-_._._._. .__ WO 9Z/tN5273 iPCT/US91 /06660 silent mutations (T bo C at 930, A to G at 933, and T to C at 936) into a restriction , tngmeM of CFTR beat spans eicon 6b completely ablated this toxic efted~
potentially by interfering with the cryptic prokaryotic promoter, and enabled the reconstruction of -4.6 kb of contigucrs CFTR cDNA sequence. The nucleotide sequence of this reconstructed cDNA was re-determined and found tder~ical to that published previously ' with the exception crf the three silent mutations noted above. See Riordan et al., supra.
Sequence Listing set forth before the Claims illustrates the nucleotide sequence of cDNA encoding !~F transmembrane conductance regulator along with the deduced ~rr~ino add sequence. The DNA sequencing was performed by the didaoxy dzain termination methocf with uS-labeled nucleotides or by the Dupont Genesis 2000 t ~ ~l)t"°~!~~!~i. ~ i''..T' ; fv ~ automatic DNA seqpencer. Numbers on the right of the columns in ~~g~e~3 indicate ,;
v~ r base positions. TtN first base position corresponds to the first nucleotide in the 5' extension done PA3-5, which is one nucleotide longer than T82-7. The 3' end and ~2a~;%;~~°?. = .-.:;~ _...
the noncoding sequence are also shown in ttie~Figure-~(nudeotides 4561 to 6129 plus the poty(A)+ taiQ. Zhe arrows shown indicate the position of y~anscriptionYinit~ Lion site by primer extension analysis. As shown abbreviated in theCFignre, nudeottde is followed by a poly(A) tact The positions of axon junctions are between the following base positions: 186-186; 296-297; 405-446; 621-622; 711-712; 1001-1002;
124&1249; X1341-1342; 1523-1524; 1716-1717; 1811-1812; 1898-1899; 2621-2622;
2789-2790; 3040-1041; 3120-3121; 3499-3500; 3599-3600; 3849-3850; 4005-4006;
4095-4096; 4268-4,269; and 4374-4375.
algorithm of Eisent>ert, et al., J. Mol. BroL 179.125 (1984) and nucleic add designations in the ~fFrgr~e. -Amindadds comprising putative ATP-binding ~:~.~~,r ~ ~.r~-~rr~J~) x 25 folds are underlined under the animo add designations in thee: Possible sites of phosphorylation by protein kinsses A or C are indicated by o and ', respectively, tnd + designates lalycosyiation sites. The open triangle indicates the position at which -3 by are deleted irn CF. Abbreviations for the amino add residues used in~ligu~5-an: A, Ale; C, Cry; D, Asp; E, Glu; F, Phe; G, Giy; H, His; Ile; K, Lys; L, Leu; M, Met; .
N, Asn; P, Pro; O, Gin; R, IArg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Try.
The modified CFTR cDNA was Boned into the retroviral vector PLJ previously desuibed by ttonman, A.J. et ai., PNAS (USA) 84:2150-2154 (1987). The proviral - .
component of thi:o recombinarrt vector, called PlJ-CFTR, is depicted in Figure 1 A.
Important structun~l components of the vector illustrated in the Figure include the long-terminal repeat so<auences (LTR), CFTR cDNA, sequences from the origin of SV40, the Potential membrane-spanning segments wars ascertained with flee use of the are ~'r~~ed under the - SUBSTtTUT'~ $~
' CA 02289222 1999-11-24 WO 9I/05273 iPCT/US91 /06660 gene that confers; resistance to 6418 (Neo), and the origin of replication for pBR322 (pBR). Transaiption from the 5'LTR produces the 8.5 kb genomic transcript that is responsible for viral passage and CFTR expression. Truucription from the SV40 sequences leads to the fom~ation of the-second transcriptions! unit that expresses a Neo-selectable marker. Sites of transcriptions! initiation ue indicated in Figure 1 A with urows at the 5' ILTR and ~temal SV40 sequences. Eiecogn~ion sites for restriction endonudeases K~pn I and Hind 111 ue also indicated. Probes sped5c for the CFTR
gene (axon 13) and the Neo gene that were used in DNA and FtNA blot analysis ue indicated below the vector.
Truisfection of PLJ and PLJ-CFT'R vectors into the vivre packaging cell line qCrip led to the transient production of replication-defective virus. Limiting dilutions of virus stocks were used to infect CFPAG1 tails which were subsequently cultured in the presence of 6418 in order to select for transduced Bones. Transiently produced PLJ-CFTR vivre a~todcs had a lower titer (50-100 fold) than those produced with PLJ
vector. Ten individual doves of cells were isolated from infections pertormed with each type of virus (named PL.! Bones t through 10 and PLJ-CFTR Bones 1 through 10) and subjected to mol~ecutar and physiologic analysis.
Tnnsduced Cla~es Express Retrorfral CFTR Sequences Retrovirally transduced Bones of CFPAG1 cells were analyzed for proviral sequences ss described for otter cells types by Wilson, J.M. et al., PNAS
(USA) 85:
4421-4425 (1988) and fNlson, J.M. et al., Science 248:1413-1416 (1990). CFPAG1 cells were infectsrd with PLJ or PLJ-CFTR virus and selected in the presence of 6418 in order to isolate individual Bones. High molecular weight DNA was isolated from oath done and analyzed by the method of gel blot hybridization as shown in Figure 1 B. in the top panel of Figure 18, DNA was digested with Kpn I and the filter was hybridized to a Neo sped5c probe, whereas in the bottom panel, DNA was digested with Hind ITI and the fitter was hybridized to the excn 13 CFTR speafic probe.
The 4.3 kb band in elf hones uises from the endogenca CFTR gene. Samples include:
CFPAG1 DNA (1I0 fig); lane '1 copy' - CFPAG1 DNA (10 fig) and lanes 'PLJ-CFTR
1 through 10' -1DNA (10 fig) from PLJ-CFTR doves 1 through 1 G. Along the right border of the figure, molecular size standuds in kilobases ue indicated.
Digestion of high molecular weight DNA with restriction enzyme iCpn I, which has unique :ttes in the vector LTRs, released all integrated forms of the PLJ-CFTR
provirus as a connmon 8.5 kb fragment As indicated in the top pane! of Figure 18, gel blot hybridization of Kpn I-restricted DNA revealed unrearranged proviral sequences suBS~nTUTS Bt~T
with the expected abundance of one copy per cell in 10/10 PLJ Bones and 9/10 PLJ- .
CFTR Bones. Hybriidization of the fitter with a Neo-spedfic probe detected a markedly rearranged proving In PLJ-CFTR done 2; tfiis virus apparently deleted a major part of -»e CFTR cDNA (d~aa not shoMm). .
5 The results of get blot hybridization analysis to study the complexity and uniqueness of each putative Pi,.J-CFTR done is shown in the bottom panel of Figure ..
1 B. High molecular weight DNA was isolated and digested with Hind BI, a restriction enzyme with two bnemai sites in PLJ-CF'fR, and analyzed with the axon 13 CFTR
specific probe. ~1s iuustraisd in the bottom panel of Figure 1 B, this analysis 10 demonstrated a single unique integration sfte in 9/10 PLJ-CFTR Bones. Ths CFTR
specific probe failed to detect the provirus in DNA from PLJ-CFTR done 2 because of th apparent deletion described above.
Expression of the retroviral transduced CFTR gene was studied by RNA blot analysis of CFPAC-1 Bones using the CFTR axon 13 probe and is shown in Figure 2.
Clones of retrovin,~s transduced CFPAG1 cells were isolated and analyzed for the presence of CFTR'transcripts. Total cellular RNA was harvested from individual clones and subjected to RNA blot analysis using the axon 13 CFTR probe to hybridize with the filter as shown in the top panel of Figure 2. In the bottom panel of Figure 2, the fitter was stripped .and rehybridized with a probe derived from human Y-actin cDNA in order to control far variation in sample loading. RNA samples (10 fig) were derived from the following calls: lanes T84' - duplicate samples from the colonic tumor cell line T84; lane 'CF'PAG1' - norttransduced CFPAGt cells; lane 'PLJ 6" - CFPAC-1 stone ~6 from the PLJ infection; and tanes'PLJ-CFTFi 1 to 10' - CFPAGt clones ~1 through X10 tronn the Pl~l-CFTR infection. Along the left border of the figure, molecular size standards in kilobases are indicated.
As illustrated in Figure 2. t~ cellular RNA from the previousty described human colon tumor cep line, T84, demonstrated high levels of the endogenous CFTR
transcript. No CFTR transcript was detected by Northern analysis in mode infected CFPAG1 cells or PLJ Bones 1 though 10 CFTR RNA can be detected in CFPAGt by .
RNA-PCR. A virFd directed CFTR transcript of the expected size (i.e., 8.5 kb) was ' .
detected in 9/10 F~LJ-CFTR doves; the CFTR probe failed to detect a transcript in RNA .
from the done thn contains the deleted provinu (PLJ-CFTFi done 2). ' Trsnsduced Clones Show Forskolln Stlmulsfion of Anion Transport Isotopic afnion (~~~ effluxes were measured to screen the PLJ and PLJ-CFTR
Bones for CAMP- and Ca-stimulated anion transport. The efflux assay, described by SUBSTITUTE SHEET
WO 9Z..rSZ'!3 _ ~ ~ ~ ~ PCT/US91 /06660 Vengiarik, C.J. W al., I4n. J. Physlol. 259:C358-0364, (1Q90) provides a qualitative estimate of agonist-stimulated CI conductance pathways in G-secreting epithelia, as judged from the Mhibitory effects of CI channel blodcers and depolarizing membrane potentials on ~'~I efflux. Figure 3A shows the time-course of the ~~1 efflux rate s constant (r) in rwo Bones, PLJ s and PLJ-CFTR s, with and without the addition of forskolin, an agent which stimulates adenytate cydase. 10 ~M of forskoiin was added at the time indicated in Figure 3A. Following a basal efflux period in the absence of agonist (not shown), forskolin increased t~l efflux rate from PLJ-CF'TR done 6 from 0.32 to 0.70 mint; PLJ 8 did not respond. r values obtained before forskolin addition and during the peak of the forskolin response provided an estimate of the relative stimulation of tzsl efflux (i.e. r~/r~). in the responding PLJ-CFTR doves, the peak forskolin effect on anion efflux was observed during the first three flux periods following forskolin addition (1 S-45 sec). The mean +/-SEM was n=9 for all Bones except PLJS
where n=7.
Data deriived from twenty Bones is illustrated in Figure 3B. The r values were taken before and after the addition of forskolin. For PLJ-CFfR clone 2 the same scaling applies below 1Ø The values ue nean +/-SEM; n=9 for all doves except PLJS where n= ~~. As illustrated in Figure 38, seven of ten P~J-CFTR Bones showed significant incre~~ses in t~l efflux in response to forskolin, whereas none (0/1 O) of the control PLJ doves responded to forskolin. The parent cell line, CFPAC-1, also shows no response to Iorskolin or CAMP analogues as described by Schoumacher, R.A.
et al., PNAS (USA) 87:4012-4016 (1990). PLJ-CFTR done 2 showed a major deletion in its CFTR cDNA by gel blot hybridization as shown in Figure 1 B, accounting for the failure of forskoliin to stimulate t~l efflux. In the seven responding PLJ-Cl=TR clones, the relative stimulation of anion efflux by forskolin ranged from 1.8 to 28-fold. This compares well vwith the 3.5-fold stimulation of efflux reported recently for the colonic tumor cell line T~34 by Venglarik, supra. Our results indicate that expression of CFTR
eDNA endows CFPAC-1 cells with CAMP-responsive anion efflux.
The correlation between forskolin responsiveness of the pLJ-CFTR Bones and their CFTR mRNA levels was not striking as i0ustrated by a comparison of Figures 2 and 38. Three csf the best responders in efflux assay showed high mRNA levels (i.e., PLJ-CFTR doves 1, 6 and 10). !n other instances, however, the correlation was not as good. For e:Kample, doves 7 and 8 showed approximately a 2 fold response to torskolin but had relatively low mRNA levels, and doves 3 and 9 showed a low forskotin response, despite the presence of readily detectable CFT'R mRNA.
SUBSTITUTE gHEEZ' Additlon~of the Ca ionophore, ionomydn, incxeased ~~I etfiux in all control and CFTR doves. Values of rionolrbasal averaged 14+/ 2 in PLJ and 14+/ 1 in Pt.J-CFTR
(n=20) in e~~ch group; no significant differences were dstectad between individual doves. The extent of response of PLJ Bones is ionorrfyan is simila~ to that observed previously in wiid~ype CFPAC-1 cells by Schoumacher et al., supra (1990), and is about three times the response of 784 cells observed by Venglarik et al., supra. The ability of Ca ionophores and Ca-mediated agonists to stimulate Cl secretion has been reported for ~urway and sweat gland cells derived firom both normal individuals and CF
patients. Seta Sato, K et al., J. Clln. Invest 73a 763-1771 (1984); Frizzell et al., supra (1986); Wiifurnsen, N.J. et aL,Nn. J. Physlal. 256:C226-C233 (1989). The presence of this response in CF cells indicates that CFTR is not required for Ca-mediated CI
transport stimulation. The lack of significant differences in the extent of Ca stimulation in PLJ and PILJ-CFTR doves suggest that CFTR does not modulate the activity of Ca-msdiated regulatory pathways that govern CI secretion.
Clones Tran~:duced with the CFTR Retrovlrus Show CAMP-Induced Cl Currents Whoh~-cell path~lamp recordings were used to determine whether the CAMP-aduce~d increase in anion efflux in PLJ-CFTR doves of Figure 3 was due to stimulation off CI conductance pathways as described by Cliff, W.H. et al., PNAS (USA) 87:4956-4960 (1990). A typical response of PLJ-CFTR done 1 is shown in Figure which illustrates stimulation of inward currents by 5lrM forskolin. Membrane voltage was held at -~10 mV and pulsed to 0 and -84 mV. The gap in the record represents time (6 min) during which bath solution substitutions were performed to determine ion selectivity of 1~e forskolin-induced amertt. The pulse protocols for determining the I-V
relations were run at the times indicated in the Figure. Chloride currents were measured ae; the inward a,urent produced by voltage pulses to -84 mV. Similar jnasases in inward cunsnt were observed in 11 of 13 cells firom PL)-CFTR
clones 1, 8, and 10 in which additjon of forskolin (5 kM) or CAMP (200 to B00 ~M) increased iinward currents from 220 +/-68 pA to 1690 +/-495 pA in responding cells. The magnitude of this response compares favorably with that observed in 784 cells by Cliff _ et ai., supra. As shown in Figure 48, CAMP (400 ~M) of forskolin (5 ~M) failed to -stimulste membrane currents in cells from the control done, PLJ 6 (n=6). The membrane voltage wss held at 20 mV and pulsed to 0 mV and -84 mV. Similar results ~ -were obtained in 5 PLJ done 8 cells. As observed from the ~~I efflux determinations, lonomyan (2 ~M) inatased inward arrrents in both PLJ (n=4) and PL.f-CFTFi (n=3) Bones.
suBSrrTU~ 8~T
. . ~ CA 02289222 1999-11-24 . ~ ~ . I.
WO 9Z/Q5273 ~ PCf/US91/06660 . As shown in Figure 4C, instantaneous arrrsnt-voltage (LV) relations of the torskolin-stimulated arrant in a NsCI bath, a low d bath, end a Na free bath were obtained from PLJ-CFTR done 1. ForskofiMnduced amerrts wars obtained by digital subsVactior~ of currents before and after stirtuarlation. The values shown In Figure 4C
wars recorded 6 cosec eRer the initiation of voltage pulses. These data were obtained from the PLJ-CF1~ done 6 cell record shown in Figure 4A during the 6 rein.
recording cap.
As tuustrated in Figure 4C, the I-V nlaiion of the stimulated current appeared to be linear, es observed in T84 cells by Clot et al., supra. Currents were determined using equal bath end pipette CI ooncentr~ations reversed near the CI
equilibrium potential of 0 mV'. Redudng bath CI to 8 mM (glutamate replacement) decreased the outward currents and shitted the reversal potential for current flow to +66 mV, a value dose to the CI equilibrium potential (+80 mV) for this outwardly-directed CI
gradient.
Replacement of bath Na by N-methyl-D-glucsmine (NMDG) did not signi~cantfy after the 1-V relation. These findings indicate that the forskolin-stimulated current is CI-select'rve, end that the stimulation of anion efflux in PLJ-CFTR Bones is due to acwation of CI conductance pathways.
EXPERIMENTAL PROCEDURES
The foilornring experimental procedures were employed in the Specific Examples set forth above:
CFPAGt cells were maintained in culture as described previously by Schoumacher et al., supra (1990); cells used for retroviral infewon were at passage 72 Infection populations of CFPAG1 cells were selected in medium containing (1 mg/ml) in order to isolate individual Bones. Transduced CFPAG1 cells were removed from selection soon after they were expanded as doves. This was not assodated with itn apparent loss of proviral sequences or proviral expression.
The - amphotropic packaging aell line pCrip, was maintsined in Dulbecco's modfied Eagle's medium supplemented with 10% calf serum and peniciliin/streptomycin as described by Oanos, O. et iv., PNAS (USAJ 85:5460-fi46~4 (1988).
Construcdfon of CFfR cpNA
The cDNA was constructed by joining the overlapping Bones 10-1, T16-1 and T16-4.5 as desaibed by Riordan et aL, supra. 10-1 end T16-1 were Ggated at the ' unique Nnr I site in axon 4 and fhe resuhant canstr~rc~ spanning exans 1 thorough 73, joined to T16-4.5. This was done by inserting a Sac I-Eco RI partial digestion product of T16~.5, extending from axon 13 to axon 24, into the respective sites of the 5' 13-suBSTnTUr~ sty ". t WO 9Z/052i3 PCT/US91/OG6b0 axon construct.. These manipulations generated a 4.5 kb done containing the .entire .
coding sequence as previcusty described by Riordan et al., supra. tt was observed that most Bones generated firom these construction attempts were grossly rearranged.
Upon sequendng of an apparentfy~intact construct. a 57 by deletion v~
identified in axon 6b oarurring between the two copies of a 13 by direct repeat. On Inspection, this interval was noted to amiain a consensus prokaryotic promoter sequence.
In an attempt to disrupt the ropeat, three single nudeotide alterations were made by in vitro mutagenesis. The i»troduasd d~anges which do not after the Ct-TR translation product and result In a stable construct, Indude substitution of C for T at position 930, G for A at 933, and C for T ai position 936. The modified reconstructed CFTH pfasmid is called CFTH 4.6.
The above described changes were accomplished by synthesis of an oligonudeotide whidt matched the normal sequence except for the presence of G
ai 933 and C at 936. The antisense strand of this segment of the CF'i R cONA was Boned into single-strande~~ M13 phage, and mutagenrzed with the oligonucleotide using standard techniques as desGibed by Smith, M., (1989) Annu. Re. Genet 19:423 (~ 985); Sanbrook, J., et al.: Molecular doping. A laboratory Manual, 2nd ed., Cold Spring Harbor Press, 15.81-75.80 (1989). The resulting done, shown in Fgure 6, was sequenced and found to have an additional unexpected base d~ange at position 930, whidi is also in a silent nudeotide position not uttering the encoded protein.
tt will else be appre~dated that other methods to stabilize the full-length CFTR
cONA can be used in the practice of the invention. Any alteration in the fortuitous E.
call promoter in axon fib, ~~s shown in Fgure 5 whid~ renders it non-funetianal while preserving the correct amirno add coding sequence for CFTH will a~omplish this same goal. For example, mutag~anesis of the CATACT sequena~ underlined in Fgure 5 can be accomplished in severe ways which wilt not after the amino acid sequence (e.g.
CGTAT1~, but will inac~vp~t~ the sequence as a prokaryotic promoter, rendering it stable in the usual doping vxtors.
Retrovfral Vea'ors end Recamblnsnt Refrovfrrrsas 3p pi9astion of the modified C~ ptasmid with Sac I released the modfied CFTR ' cDNA on a 4.6 kb resirie*,ion fragment The Sac I sites were converted to Bct I
sites .
with oligonudeotides and the Tinkered fragment was Boned into the Bum I site of the ratrcviral vec:or PL.t previously desG~ibed by Kcrman art al., supra. This recombinant vector, called PLJ-Ci-~, i~ presented in Fgura 1A. Aetroviral vec;or PLJ and PLJ-C~'i R were transfec:ed into the amphotropic packaging cEtl line TCrip as described.
St?8S"i'TIJfE SST
Tissue culture media was removed from platE~ containing the transfected packaging cells 24 hours later in order to harvest the transiently produced amphotropic virus.
CFPAGt cells, passaged t :5 onto 10 cm2 plates, were exposed to viral supernatants sup~plemsnted with polybnns (4 ~g/mi) for 12 to 16 hours. When the 5 calls nad~ed confluence, they were passaged 1:10 into medium containing 6418 (1 mg/ml). Clones cd cells were isolated, expanded, and ayopreserved.
DNA and RNA Analysis of CFPAC~! Clones High mole~~rlar weight DNA was isolated from CFPAC-1 pelts as described and analyzed by gel blot hybridization as described by Wilson et al., supra (1988). Total 10 cellular RNA was purified and subjected to RNA blot analysis of Wilson et al, supra (1988). Filters ware hybridized with a variety of DNA probes that were labeled to a high spedfic acti<rtty using the random priming method of Feinberg, A.P. et el., Anal.
Biochem. 132:6-13 (19183). These probes indude: 1) Exon 13 of CFTR isolated following PCR unpification of doped cONA using oligonudeotides that flank the 15 borders of this exon, (NT 1900 to 2611 ); 2) Neo-speafic sequences on a 950 base pair Hind 111 to Nco I fragment of pSV2Neo, and 3) human Y actin cDNA.
Anion Efllux Massuramenls Radioisotopic anion efflux was determined as described by Venglarik et al., supra. Briefly, celll monolayers were preloaded with ~~i for 30 min; after two washes, efflux was monitored at 15 sec intervals using a sample-replace procedure. At the end of the experiment, tracer remaining in the cell monolayer was extracted with 0.1 N
HP03. The efflux rate constant (r) for each sampling interval was calculated as follows:
r = [In (R~) _ tn~ (R~)]/(t1 _ ti), where R~ and R2 are the percent of loaded remaining in the monolayer at times (t) 1 and 2. Forskolin or ionomycin were added after the fifth 15-sec sampling interval. The degree of agonist stimulation is expressed as Y/Y~ where ris the maximal value observed in the presence of agonist and r~~ is taken from flux interval immediately prior to agontst addition.
Most of the extraceltular ~~1 washout occurs during the initial 60 sec of sampling as set lforth by Vanglarfk et al., supra; this period was ignored in the rate .
constant calculations. However, a small residual efflux from the extracellular space after 60 sec leads to a slight underestimate of the agonist response because the extraceilular compartment washes out faster than the cellular compartment.
Therefore, when there is no efflux response to forskolin, r determined immediately after forskolin addition is slighth~ less than that measured before forskolin is added. This accounts sued s~EET
. ' CA 02289222 1999-11-24 r WO 9I/05Z73 .PCT/US91 /06660 for the finding thsl; Y~IY~ ~ ~~een 0.9 and 1.0 in the PLJ doves shown in , Figure 38.
Whole-Cell Currun Reaordlngs Maaoscopic csurents were recorded during whole~etl patch-clamp by methods previously described by Cliff et al., supra. Recordings were made at 37°C with the following solution: (mM); bath: 115 NaCI, 40 N-rnethyl-D~9iucamine (NMDG) glutamate, 5 K~lutamate, 2 MgCLj,1 CaCLt,10 HEPES (pH 72); pipette: 17 5 KCI, NMDG-glutamate, 025 EGTA, 0.09 CaC>Z (100 nM tree Ca), 2 MgCLZ, 2 Na2ATP, 020 NazGTP,10 HEPI::S (pH 72). Membrane poterrttals were damped alternately for msec duration at three voltages, two of which were chosen to equal the equilibrium potentials for CI (D mV) and K (_84 mV). This pem~its the CI and K currents to be monitored during agonist responses as described by Cliff et al., supra.
Pulsing was interrupted to determine current-voltage relations by stepping the clamp voltage between +/_100 mV at 20 mV increments as shown in Figure 4C.
SPECIFIC EXAMPLES - II
Retrovlnrs~Medla~ted Trsnsducfion of PancreatJc and Pulmonary Eplfheliel Cells Retrovirus-mediated gene transduction iMo various epithelial cells was .
optimized using a replication defective retrovirus that expresses the ~-galactosidase gene from E. colt. This was used because expression of viral directed a-gafactosidase can be detected In alto using cytochemical reaction that stains the transduced cell blue. The ampinotropic virus producer cell line made from the ~-gafactosidase expressing BAG vector, which has been described previously, was used as a source of virus. This virus produdng call Une is called BAGS. The supernatant over a confluent plate o! BAGS cells was harvested, filtered, and used to infect various .
23 epithelial cells as: described below. _ .
Pancrestlc Epltl~ielfal GIl Une CFPAGt is a cell line derived from an adenocardnoma of a patient with CF
which expresses the cellular defect d~araMeristic of CF (i.e. chloride channels are not activated in the presence of CAMP agonists). CFPAG1 cells were split at various dilutions (1:2, 1:5, 1:10, and 1:20) and 24 hours later exposed to fresh virus ' supemataMs th;~t had been supplemented with polybrene (4 ~glml). Twelve hours Later the vinrs was replaced with firesh medium. When confluent, the cells were analyzed for the expression of viral directed a-galactosidase as described.
Optimal infecKion effidency was obtained with CFPAGt cells that were split 1 to 5 the day before infecrtion" Under optimal conditions, a single exposure to virus led to stable SUBSTITUTE $HE~T
" CA 02289222 1999-11-24 WO 9Z/aSZ73 Ir'Cf/US91 /06660 transduction of 'the ~-gatactosidssa gene into 30-40% of the oslls. Expression of ~-paJactosidase fuss been stable in arttured Celts for over 2 months. Attempts to reinfect CFPAC-1 tails cm subsequent days led to little augmentation of infection efficiency.
Alrwsy Epitthellal fills As discussed pn~viously, a~~way epithelial cells are the most desirable targets for pane transfer because the pulmonary complications of CF are usually its most morbid and fee-limiting. Taussig, supra (1884). Since airway epithelia! cells are easily trtfectsd with recxambinant n~trovin~es, the gees transfer approaches described in the preceding and lioliowing examples will also be useful for gene therapies directed to t airway epithelial Celts such as those of the lung.
An epithelial cell line derived from an airway of a patient with CF was used as a potential target for ntrovirus-mediated gene transfer. These cells had been described previously and have been called T43 c;elts. Freshly harvested BAG5 virus was supplemented with poiybrene and exposed to T43 cells that had been split 1:5.
24 hours previously. Celts were exposed to vine for 12-18 hours and aJiowed to grow to confluence bE~fore being analyzed for viral directed ~-galactosidase expression using the previously described cytochsmicat assay. Under optimal conditions, greater than 25% of CFPAC cells wore stabty transduced with the ~-gslactosidase gene after a single exposure to virus.
Dlnct Delivery o! CFTf~ F.xprasslnp Vectors to the Airway Eplfhellsl Cells One approach to the use of recombinant rstroviruses and the treatment of CF
is to introduce a. functional CFfR gene into epithelial cells in vivo by directly delivering retroviruses into the airway. Several approaches can be taken for the direct delivery of retroviruses. The more invasive approach would be to intubate the patient and savage the airway with concentrated solutions of CFTR expressing retrovinrses.
Stable retroviral expression requires that the provirus integrates into chromosomal DNA. This occurs most effiaentiy tt the reapient cells are dividing. n may be necessary to stimulate regeneration of the epithelial soon artter exposure to virus. This could be accomplished Hrtth mechanical or chemical irtitation of the e~rway.
The less morbid approach would be to deliver the normal CFTR gene to airway eptifieliaJ cells bmwo by a nebutized preparation that can be inhaled. Many different pharmacologic agents are effiaently delivered to a large surface of the airway by nebutized treatments. It is possible that the beneficial effect achieved by this method may be transient. tt may, therefore, be necessary to give repeated doses of the drug.
The gene delivery system used for direct gene introduction may not have to be viral SUBSnTUTE $HE~
t'''191/I?5173 - FCTlUS91 /066 based. Direct inhalation of DNA protein complexes or DNA expression vectors in ~posomes may be a safer and more effecZiw gene delivery system than retroviruses.
Tnnsplarrfaffon of Genetically Mod~ed AJrwsy Eplfheilal Cells This approach to somatic gone therapy of CF is similar in concept to bone marrow directed gene therapy. We would propose to isolate airway epithelial cells from the CF patient, establish a~tturss of the cells, use recombinant retroviruses described In thi:: invention to stabty correct the defect In the cells, and transplant the genetically modified cells into the patient so they can repopulate the airway.
In order to achieve effident repopulation in the airway with-genetically modfied cells, ft may be ~~~Y ~ F~~~ ~e ~a9rrty ~ ~ endogenous epithelial lining through mechanical or chemical irritation.
Atttmatlve Gene T~arrsfer Delivery Systems - Other gone deliveries systems for genetic correction of CF defects also fall within the scope of the present invention. For these experiments ptasmid-based DNA
vectors will be aced. An example of suds a vector is i3A-CFTp BD presented in Figure 6 '. This is a simple 7762 by transfeciion-based vector in which transcription is initiated from actin flanking systems and temninated from heterologous 3' potyadenylation p sequences.
The vector was canstrucied in the following manner. The backbone contained sequences from PC18 (nudeotide 6928 to 4563) the 5' flanking region of the chicken ~ actin gene (nudeotide 6928 to 7754) and 3' flanking sequences of Bovine growth hormone potyadertyiation signal (nudeotide 4827 to 4553). The full length CFT'R
sequences spanning the entire calling region, and containing the three nucleotide changes discussed earlier, were removed from the vector CFTFI on a Sac I to Sal I
fragment, and clone into the vector backbone described above.
n will be »ppreciated by those skilled In the art that this vector could be used in several gene delivery systems.
Upofec#ion The previously described procedure is based on the encapsidation of DNA
tiposames. When cells are incubated with Gposomes, they take up the DNA and express it. We proposed io dilute DNA of an expression vector and lipid (DOTMA) to 7.5 ml in Hepes; buffered saline and mix these constituents to form lipid-DNA
complexes. Liposomes could then be used to transfeC, airway tails in vivo by lavaging an inlubated patient with liposome captaining solution or by administering the GPosomes by inhtdaiion.
SUBSTtTUTE g~
H'O 9Z/05173 PCT/US91 /06660 ' 10 ONA~Proteln Gunple~ces M atterr~atiw approach to targeted gene dsGvery is through the formation of a ONA protein complex This type of gene transfer substrate is constructed in the toUowing manna:r. A potypeptide tigand for a receptor on a respiratory epithelial calf b conjugated to poiyiyaine with ethylidene diamino carbodiimide as described.
This ' protein conjuga~ce is compiexed to DNA of a transfection vector by mixing equal mass quantities of protein conjugate and DNA in 025 molar sodium chloride. The DNAlprotein complex is taken up by respiratory airway cells and the gene is expressed. Thiac could be used to directly deliver the CFTH gene to airway epithelial cells In vlvo using the approaches described for dposomes.
It is app~~reni that many modifications and variations of this invention as set forth as may be made without departing from tfie spirit and scope thereof. The specific embodiments described herein are given by way of example only and the invention is limited only by bhe terms of the appended daims.
suBSrlruT~ s~
SEQUENCE LISTING
M
Q R S P L E K A S V Y S K L F
F S W T R P I L R K G Y R Q R
L E L S D I Y Q I P S V D S A
D N L S E K L E R E W D R E L
A S K K N P K L I N A L R R C
F F W R rF M F Y G I F L Y L G
A S Y D P D N K E E R A I
Y L G I G L C L L F I V R T L
T TA CAC CCA GCC ATT TTT~GGC CTT CAT CAC ATT 6GA ATG CAG 585 M R I A M F S L I Y K K T L K
AGT CTC CTT TCC AAC AAC CTG AAC AAA TTT GAT 6AA GGA ~ 720 ~ S L L S N N L N K F D E G A
CA 02289222 1999-11-24 .
M G L ~ W E L L Q A S A F C G
E R L Y I T S E M I E N I Q S
GT? AAG GCA TAC TGC TGG GAA GAA GCA ATG GAA AAA ATG ATT 6AA 990 Y K A Y C W E E A M E K M I E
N L R Q T E L K L T R K A A Y
V R Y F N S S A F F F S G F F
~TG GTG TT? TTA TCT GTG CTT CCC TAT GCA CTA ATC AAA A AT 1125 V V F L S V L P Y A L I K
R M A~ T R Q F P W A V Q T W
AAG CAA GAA, TAT AAG ACA TTG GAA TAT AAC TTA ACG ACT ACA GAA 1305 K Q E Y K T L E Y N L T T T E
V Y M E N Y T A F W E E G F G
6AA TTA TT1' 6AG AAA 6CA AAA CAA AAC AAT AAC AAT AGA AAA ACT 1395 E L F E K A K Q N H N N R K T
TCT AAT GG1' 6AT CAC AGC CTC TTC TTC AGT AAT TTC TCA CTT CTT 1440 S N G D D S L F F S N F S L L
GGT ACT CC1' GTC CTG AAA GAT ATT AAT TTC AAG ATA 6AA AGA GGA 1485 TTG TCC ACT TCA
L
M
ATTAAGCAC.AGT6GA AGA ATT TTC TCT CAG TCC 1620 TCA TGT TTT TGG
K
H
ATTATGCC1'6GCACC ATT AAA AAT AT~ 6GT 6TT 1665 GAA ATC TCC
I M P G T I K E N I I F G Y S
AGC ATC TGC CTA
GCA AAA ATA CTT
E E D I S K F A E K D N I V L
GGAGAAGG1'GGAATC ACA CTG GGA CAA CGA AGA 1800 AGT GGT GCA ATT
AAA GCT TAT TTA
GACTCTCC1'TTTGGA TAC CTA GTT ACA GAA GAA 1890 GAT TTA AAA ATA
D S P F G Y L D V L T E K E I
CTG GCT ACT ATT
F E S C V C K L M A N K T R I
TTGGTCAC1'TCTAAA ATG GAA TTA AAA GCT AAA 1980 CAT AAG GAC ATA
L Y T S K M E H L K K A D K I
E L Q N L Q P D F S S K L M G
TGT 6AT TC1' TTC 6AC CAA TTT AGT GCA 6AA AGA AGA AAT TCA ATC 2115 C D S F D Q F S A E R R N S I
L T E T L H R F S L E G D A P
* *o V S W T E T K K Q S F K Q T G
* o S I R K F S I V Q K T P L Q M
S L V P D S E Q G E A I L P R
I S V I S T G P T L Q A R R R
Q S V L N L M T H S Y N Q G Q
o * * 0 N I H R K T T A S T R K V S L
A P Q A N L T E L D I Y S R R
GAA GAA
L S Q E T G L E I S E E I N E
TTT ATG
E D L K E C L F D D M E S I P
TAC TAT
A V T T W N T Y L R Y I T V H
ATT ATT
TGG
TGC
TTA
K S L I F V L I W C L V I F L
GAG T TGG
GTG T
TTG
TT
+ +
' T P L Q D K G N S T H S R N N
S Y A Y I I T S T S Y Y V F
' . CA 02289222 1999-11-24 F R G L P L V H T L I T V S K
I L H H K M L H S V L Q A P M
6AT 6CA TTG CTT CTT ~
S K D I A I L D D L L P L T
TTC CAG TTA GCT GCA
TTA
ATT
GTG
~T- GTC CAA TAC ATC TT A ACA TG~A_ 3195 GCA TT CCC TTT
TTA
V V A ~ Q P Y I F Y A ' V P
T
Q T
TCA CAG CA,A CTC AAA CAA CTG 6A,A TCT GAA GGC AGG AGT CCA ATT 3285 S Q Q L K Q L E S E G R S P I
F T H L V T S L K G L W T L R
GCC TTC GGA CGG CAG CCT TAC TTT GAA ACT CTG TTC CAC AA,A GCT 3375 A F G R Q P Y F E T L F H K A
CTG AAT TTA, CAT ACT GCC AAC TGG TTC TTG TAC CTG TCA ACA CTG 3420 L N L H T A N W F L Y L S T L
R W F Q M R ~, I E M I F V I F F
ATT GCT GTI ACC TTC ATT TCC ATT TTA ACA ACA ~ 6AA 6GA GAA 3510 GGA AGA ~ GGT ATT ATC CTG ACT TTA GCC ATG AAT ATC ATG AGT 3555 G R Y G I I L T L A M N I M S
ATG CGA TCT 6TG AGC CGA GTC TTT AAG TTC ATT GAC ATG CCA ACA 3b45 L S K V M I I E N S H Y K K D
*
D I W P S G G Q M T V K D L T
A K Y T E G G N A I L E N I S
GGG ACT TTG TTG
TTA TCA AGA
GCT
G S G K S T L L S A F L R L L
TTG TGG TTT
m TCT AGA
K Y F I F S G T F R K N L D P
CAG GAT TGG
CTC GTG TTT
CTT GGG CTA
ATG GCT CTC
0 L M ~L A R S Y L S K A K I
CTT CCC TTG
L L L D E P S A H L D P V T Y
ATT ACT GCA
CTC CAC GCA
V I L C E H R I E A M L E C Q
Q F L V I E E N K V R Q if 0 S
I Q K L L N E R S L F R Q A I
*
S P S D R V K L F P H R N S S
K C K S K P Q I A A L K E E T
E E E V Q D T R L
AAAACAAGGATGAATTAAGTTTTTtTTTAAAAAAGAAACATTTGG 4725 TAAGGGGAATT'GAGGACACTGATATGGGTCTTGATAAATGGCTTC 4770 CTGGCAATAGI'CAAATTGTGTGAAAGGTACTTCAAATCCTTGAAG 4815 AT'TTACCACT1'GTGTTT?GCAAGCCAGATTTTCCTGAAAACCCTT 4860 CTAGTTGATCAGCTTA?TGTCTAGTGAAACTCGT?AATTTGTAGT 4950 GTTGGAGAAGAACTGAAATCATACT?CTTAGGGTTATGATTAAGT 4995 AATGATAACT(~GAAACTTCAGCGGTTTATATAAGCTTGTATTCCT 5040 TTTTCTCTCC1~CTCCCCATGATGTTTAGAAACACAACTATATTGT 5085 TTGCTAAGCA'~TCCAACTATCTCATTTCCAAGCAAGTATTAGAAT 5130 ACCACAGGAA1:CACAAGACTGCACATCAAAATATGCCCCATTCAA 5175 CATCTAGTGAi;CAGTCAGGAAAGAGAACTTCCAGATCCTGGAAAT 5220 CAGGGTTAGTi4TTGTCCAGGTCTACCAAAAATCTCAATATTTCAG 5265 ATAATCACAA'fACATCCCTTACCTGGGAAAGGGCTGTTATAATCT 5310 TTCACAGGGG~4CAGGATGGTTCCCTTGATGAAGAAGTTGATATGC 5355 CTTTTCCCAAi;,TCCAGAAAGTGACAAGCTCACAGACCTTTGAACT 5400 . AGAGTTTAGC'fGGAAAAGTATGTTAGTGCAAATTGTCACAGGACA 5445 T11GATGTATAG1;TT6AT6GTGGTATGTTTTCAGGCTAGATGTATG 5580 TGTTTCAAACA'TATATTACAATGCTGTATTTTAAAAGAATGATTA 5760 CCTTTTGGTCTGGAGGGAAGCCTTGGGGCTGATC~~AGTTGTTGCC 5940 CATTTGTGTAA
Rev. 2/19/93