CA2075067A1 - Auto-inactivating aviary vector - Google Patents

Abstract

2075067 9111190 PCTABS00105 The present invention relates to a self-inactivating viral vector for integration and expression of at least one heterologous gene in avian cells, characterized in that it is formed from the proviral genome of a retrovirus. avian and that it comprises between two avian LTR sequences the said heterologous gene (s) which is (are) under the control of a heterologous internal promoter, deletions or mutations in the 3 'LTR having were carried out so as to inactivate the viral promoters present in the LTRs.

Description

WO 91/11190 PCI/FRgl/00062 207aOfi7 ~UR SELF-INACTIVATING AVIAN

The present invention relates to new vectors retrovirals, of avian origins, integration and expression of at least one heterologous gene in avian cells.
The present invention constitutes a development of the inventions described in European patent applications EP 178 996 and international WO 89/03877 in the name of the present applicant. He agree to refer to it if necessary for the best understanding of the present invention.
By heterologous gene, we mean a gene that does not not found in the ~ enome avian of origin and preferably which does not not found in the genome of a virus. It may be a useful gene encoding for a protein of industrial interest intended to be obtained by culture cell or a gene that may be of particular interest for raising birds, in particular chickens or quails, or a gene encoding an immunogenic protein intended for use as a vaccine, or yet it may be a marker gene in particular for resistance to a antibiotic.
Retroviruses are now widely used for the gene transfer because of their high efficiency with respect to the times their penetration through the cell membrane ee the integration of their genome into the DNA of the host cell. Retroviral vectors in which the inserted foreign sequences are transcribed directly under the control of internal promoters, are particularly interesting. They allow efficient translation of the RNA transcribed in the form of a native protein. However, functional interferences are observed between viral and internal promoters5.
C'-5t why we developed ~ according to the present invention a new generation of avian retroviral vectors. These vectors carry mutations in their terminal "long repeats"
(so-called LTR sequences) which inactivate the transcription of the integrated provirus without affecting the expression of internal promoters.

WO 91/III90 PCr/FR9ltOO~2

2 0 ~ 2 These vectors are called SIN ("self inactivating"), i.e.
self-inactivating.
The article by Dougherty and Temin 1987, PNAS 84, 1197-1201, describes a self-inactivating vector constructed from SNV turkey virus 5 which can only be amplified by the D-17-Rev-A line, which in ou ~ re only produces infectious virions on the cells of turkey and dog, the effectiveness of infection on other avian species being very weak.
More specifically, the present invention relates to a 10 viral vector for integration and expression of at least one heterologous gene in avian cells, consisting of a proviral genome from avian retroviruses AEV and related avian retroviruses, in particular RAV or RSV type and comprising between two avian LTR sequences the said heterologous gene(s) which is/are under the control of 15 promoter (s) internal (s) he ~ erologist (s), characterized in that deletions or mutations in the 3' LTR have been made in such a way as to inactivate the viral promoters present in the LTRs, said 3' LTR originating from the retroviruses RAV-O, RAV-I, RAV-2.
The structural modifications of the viral genome brought about by the vectors according to the invention totally inactivate the viral promoters present in the LTRs of the proviral forms integrated into the cells transfer targets.
By heterologous internal promoter is meant a promoter that is not found naturally in the genome of a 25 avian retrovirus and which was placed between the two 5 LTR sequences.
In one embodiment 5 according to the invention, the promoter or the enhancer sequence, depending on the case, of the 3' LTR of the viraJ genome are deleted in whole or in part. The virus resulting from this genome introduced into Je genome of infected cells a provirus whose LTR S' is copied from the 30 LTR 3' according to the classic process of proviral DNA synthesis from virionic RNA.
More specifically, a deletion is made in the U3 region of the 3' LTR.

WO 91/11190 PCI~FR91/00062

3 207~67 In particular, mutations or deletions can be made in the CAAT and TATA boxes or eliminate said boxes.
Since the U3 region of the 3' LTR serves as a matrix for the synthesis of the corresponding region of the 5' LTR in ~a 5 reverse transcription, any mutation in U3 in the 3' LTiR appears in both LTRs after one cycle of viral replication. This is why the mutation that destroys the 3' LTR promoter gives rise to a inactive provirus in infected cells. However, it has been observed that a transcription from an internal promoter can still take place.
In the case where a 3' LTR from a virus whose LTR are naturally very inactive, it is then sufficient to carry out - point deletions or mutations to inactivate viral promoters present in the LTRs. We can cite as LTR 3 'naturally very little activates that of the RAV-0 virus.

Deletions in the 3' LTR influence not only the inactivation of the vectors but also the efficiency of the transfer in the cells therefore, according to another characteristic of the present invention, the vectors carry in 5′ the LTR of the RSV virus (sarcoma virus 20 of Rous). This LTR has a very high transcriptional activity.
notably higher than that of the RAV-I and RAV-2 ~ Ce5 viruses constructs improve the transcription of the viral genome Jors of the transfcction of helper cells and consequently increase the number of viral particles produced~
In the vectors according to the invention, the5 heterologous genes (selection genes or useful gene) are appropriately under the control heterologous internal promoters. However, in the case of the gene for selection, this one can be animated directly by the LTR 5' and not associated with its own internal promoter, the heterologous gene used being under control of heterologous internal promoters.
In the genomes of the SIN vectors according to the inventlon. one can integrate transcriptional units in which the heterologous gene WO 91/111g0 PCI/FR91/00062 is associated with its own promoter in its original genome, as heterologous internal promoter. In infected cells, only these internal promoters will then be functional.
The correct operation of these vectors has been verified and the results are presented below.
On the one hand, the expression of the inserted genes is initiated at the level of their respective internal promoters. On the other hand, in cells infected the LTRs of the vectors become inactive. Finally, the vectors can emerge from infected cells, even after infection of these latest with a competent helper virus.
The vectors according to the invention therefore provide a great safety of use since in the infected cells on the one hand, inactive LTRs cannot alter the functioning of the sequences genes adjacent to the site of insertion of the provirus, and on the other hand the provirus can no longer generate transmissible viruses.
In a particular embodiment, the vectors have been constructed from genome elements derived from retroviruses RAV-I, RAV-2 and/or AEV for the portion between the two LTRs.
Mention may in particular be made of vector constructions consisting essentially of the genome of the RAV-2 retrovirus in which the CAAT and TATA boxes have been deleted in the 3' LTR of said genome. In one embodiment, the vector comprises as 3' LTR:
- the U3 sequence of the LTR of RAV-2 in which the boxes CAAT and TATA have been removed, and - the R and U5 sequences of the LTR of RAV-I.
Appropriately, the vectors according to the invention comprise two heterologous genes, namely an 8th selection gene and a useful gene.
When the e5t vector constructed from the genome of avian erythroblastosis retrovirus (AEV), heterologous genes are advantageously placed in positions of the v-erbA and v-erbB ~ oncogenes By selection gene is meant a marker gene for resistance, in particular resistance to a drug analogous to a antibiotic. Mention may be made in particular of the hygroR gene for resistance to hygromycin G418 and the neo gene for resistance to neomycin.

.

WO 91/11190 PCI'/FWI/00062 2o7~o67 Of course, the marker genes that have been cited do not sound~
purely by way of illustrative examples. It is possible to use any another resistance gene as a marker gene.
By useful gene is meant a gene coding for a protein 5 of interest other than a selection gene.
In a suitable embodiment, the genes heterologous are placed under the control of internal promoters heterologous, in particular the simian promoter SV40 or metallothionein human MTlla.
In one embodiment of the vectors according to the invention, the selection gene is placed under the control of the promoter simian SV40 and the useful heterologous gene is placed under the control of the human metallothionein lla (MTlla) promoter.
In another embodiment, the selection gene 15 is driven directly by the 5' LTR and not associated with a promoter own.
The direction of transcription of a heterologous gene inserted into the vector must be opposed to that of the LTR in the case where said gene heterologous contains a polyadenylation control sequence. In the 20 contrary case, the heterologous gene could be inserted in the direction of - transcription of the LTR, the transcript of said gene then using the signal polyA provided by the 3' LTR. In the case where the heterologous gene inserted contains introns, this gene must be inserted in the direction reverse transcription, in order to avoid any interference between the signals 25 splicing control of this gene and those of the viral genome.
In one embodiment, the selection gene is inserted in the direction of retroviral transcription and the useful gene is inserted into the direction opposite to that of retrovrral transcription.
As an example of a useful heterologous gene, the 30 gene encoding histone H5 and it has been shown that the expression of H5 in normal or transformed quail or chicken fibroblasts did not significantly affected cell growth properties transformed, but inhibited the proliferation of normal fibroblasts.

WO 91/IIIgO PCI/FR91/00062 6~

According to another example, the heterologous gene is the gene of E.
coli encoding beta-galactosidase.
In particular vectors carrying a 3' LTR derived from the RAV-0 viruses, naturally devoid of enhancer sequences, have been 5 constructs in which the CAAT box is deleted:
These vectors carry the following sequences of 51 in 3~:
- a 5' LTR of the RSV virus, - a leader sequence of the RAV^1 virus, - a resistance ~ ene, such as NeoR, 10 - a heterologous ~ ene of interest, such as the E. Coli gene encoding a - beta-galactosidase provided with a nuclear adre55age signal, under the con ~ role the promoter of the SV40 virus, - the 3' non-coding sequence and the 3' LTR of the RAV-0 virus deleted from the CAAT box.
15These vectors have been te5té5 and 50nt functional and offer a maximum safety of inactivation since the LTR RAV-0 in 3' underwent a deletion in the CAAT promoter sequence.
Given the fact that avian retroviruses are viruses to RNA, the terminolo ~ ie virus can sometimes designate the proviral genome 20 in the form of DNA as well as in some cases the foreign gene may be in the form of RNA, especially when inserted into the genome of a virion. The techniques allowing to pass from a DNA form to 12 RNA form and vice versa, 50nt known and are the transcription and the reverse transcription. The recombinant virus can be used as such under 25 form of RNA virions or proviral genome in the form of DNA comprising the in ~ ertion of a foreign gene also in the form of DNA and corresponding to the reverse transcription of the RNA of ~irUs~ can be integrated into a DNA vector such as a plasmid, a cosmid, or a phage for example.
For convenience, the constructions have been 30 carried out first on plasmids comprising a composite sequence of DNA corresponding to the reverse transcript of an RNA retroviral vector of a virus as previously described.

WO 91/111g0 PCI'/FIW1/00062 20750~7 The present invention also relates to a method of preparation of a vector according to the invention characterized in that ~ one carries out deletions or mutations in the avian 3' LTR so as to inactivating the viral promoters present in said LTRs and in that one inserts said heterologous gene(s) under the control of a promoter internal heterolo ~ ue or promoter of the LTR S '.
In one embodiment of the method, one inserts into the vector pDAI at the Xbal site a transcriptional cassette of a gene heterologous.
For example, the transcriptional cassette of the ~ ene hetero-logue consists of the vector pBO ~ as shown in Figure 6 in which one, inserts said heterogeneous gene ~ ue at the restriction sites of the polylinker between the transcription initiation sequence under shape of the MTlla promoter and transcription termination sequence and polyadenylation derived from the histone HS gene, said cassette being framed by two XbaI sites allowing its isolation and then its insertion into the unique Xbal site of the pDAI vector.
The present invention further relates to a method of ~, modification of cells characterized in that a transfection is carried out or infection depending on the case of said cells with vectors according to the invention.
The present invention therefore relates to transfected cells ;, or infected with the vectors according to the invention whether they are in the form of plasmid or virus respectively. It is evident that the transfection Primary 2S will be generally carried out with plasmid-type vectors, , ~ that is to say that the avian retrovirus will be in the form of the DNA of its genome but in the presence of the DNA of a helper virus of 5 cells transfected will in this case transform and release the viruses recombinants which will be able to transfect other cells. This kind of process also makes it possible to constitute stocks of virus vectors subsequently usable in place of the plasmid vectors.

~, : 35 ..

.

WO 91/11190 PCr/FR91/00062 ?,Qr~6~
Retroviruses such as AEV, RAV-I, RAV-2 are defective for their replication. To replicate, these viruses require the presence in the same host cell of a functional helper virus as is known from the skilled person.
The present invention also relates to a method of modification of cells characterized in that a co-infection is effected or co-transfection depending on the case with a helper virus which comprises the genes ensuring the replication of said vector.
The modified cells according to the invention can be various but are preferably avian cells, of hematopoietic type, germline or fibroblast type.
Preferred cells for the implementation of the vectors of the invention are chicken or quail embryo fibroblasts. Nevertheless the vectors according to the invention make it possible to modify turkey cells, duck and other birds.
Finally, the invention relates to the process for the preparation of a determined protein in which the cells modified by a vector according to the invention in which a foreign gene encodes said protein and in that the protein is recovered from the cell culture.
Other advantages and characteristics of the present invention will become apparent from the detailed description of an example of realization given for purely illustrative purposes and as a model for integration and expression of a heterologous gene, using the gene of histone H5.
The following description is made with reference to the figures I to 8.

~`

..

~, ;~..

.

WO 91/llt90 PCT/FR91/00062 20750~7 Fi ~ ure 1 diagram illustrating the transfer of the deletlon of the 3'LTR a la S'LTR (1) A ~ector retrotrlrai delot~ tans the promotour of 3'LTR t wearing a g~no otrang-r (~) a~oo its own promoter (P), ost transfects in back helper cells ~2) Apre~
S m tegratior, the proiral DNA is transcribed from the promoter of 5'LTR, and trlral transcripts containing the deletion in the realon U3 are encapsid~J in dos part~culos vlralo6 snfectleuse~ (3) Dan~ cells mfected by this~ partlculcs v rale ~ devoid of helper, the viral RNA is transcribed by transcrlptlon m~rerse in a double-brln DNA molecule shows a rogion U3 dol~t~o dan~ los sweet LTR A m sl, tans ces cells, the transcriptlon trlralo is a ~ olie and soule the transcription from the int-rne pro~ot-us may take place on trlangle mt ~ qu- the t ~ lotlon; ca-- bachuro- rogion U3, case po mt~ o rogion US, caJo clairo rogion R

Figs~! LTR structure normalos (in ~aut) and deleted (in as~ La iocaii~atlon your rogulatlon slgnaux transcrlptlonneile est ~ndlqu~o LJ ZOn-J hatch-o ot po m till-- in the roglon U3 ~ 20 represent reJpeCtlvem nt the reg~ons tu enhancor and du ; p~omotour The rogion U3 teleteo ~provonant to provir DNA~1 RA~-2) at oto la~o, by los lD~ers Clal, to the R ot US reg~ons of the LTR te RAV-1 L- trlanglo po~nt~llo ~ntiquo l'~t-ntuo do la et~on~ (~) Struoturo doo voctouro rotrov~raux. Dan~ pT2~3', ter~vo of AEV, the gono to ro~stanoo a la ~ooryo~no (3001 ost transcrlt from the LTR v~ralo, while quans pDAl, ~1 ost ~ranscr~ta from promotsur to SV40 m t-rno (SV) pDAl, doslve essontlellomont to RAV-~, contlont un- reglon U3 doleteo dan~ 13 ; ~ 'LTR pDA ~ Y and pDA ~ S have been derlv ~ s to pDAl on msérant respect the nph genes (reslstanco a l'hygromyc me) e~ ~i ~os ~loohos ~ntlqu-nt la tlroct~on transor~ptlonnolle dos ~f'erent genes In pDA~Y, 13 r~g~on coding for hph, llée ~u 2a~'6~ 1 o slte of polyadenylatlon of the gene of thyme ~ dme kina ~ e (Tk), has been placed SOUQ the SV40 promoter control. In pDA~5, the promoter of mctal1Othlonelno humalno MT-IIA (MT) was Ut:il3e to transcrlre the gene ~5 containing its own slgnal of polyadenylatlon (Pol HS). Ca~es hatched. clalre and po ~ nt; llee.
respectively LTR of AEV, RAV-2 and RAV-1; SD and SA:
r-~p-ctlve~ nt donn-ur t acoept-ur d'~p~Jage;~U3; region U3 deleted; ~o: ~hol; ~: a~al; C:Clal. The eohelle (in pairs of kilobases:kb) is listed below.
F~ e 3.4 analysis by Southorn blot mg of lnrected OT6 cells by DAH5. 15 ~ g of DNA, l ~ olé of cell ~ lnfecte-~ (band ~ 5 and ~!
and not 1 ~ fected (ba ~ of ~ 4 and 6), were digested by ~ bal (bands

4 and ~) or ~ hol ~ bands t; and 1). After electrophoresis and transmission on n~trocellulo~e, the transfer range ("blot") has been hybridized a ~ cc an opeciflque H5 probe. Leo quantity of fragment of reotrictlon oorre~ bridged to 10.1 and 0.1 oopio~ of gene H5 by genome (roopectl~rent bante~ 1, 2 ot 3) ¢nt ~t~ tralteeJ en : 20 parallel. The~bda DNA dig~r~ by EooRI ot ~intIII a ~t~ uses coume marker dc size. A ch~m~tlque representation of the . ~ o ~ rte restriction DA ~ 5 is shown in ~ a ~ of the figure.

Fi~.~ç~Eo~al of protectlon by the S1 nuoleaoe of RNA
transorlt cytoplasm in CT6 cells. 20 µg of RNA
. cytopl ~ 8rique, i5010 of oellules 30n infeotoes (band l) and of cellulos mrocteos by DA~5 (bandos 2 and 53, have eto bybrl~es a~ec the probe do H5 DanJ the bandOJ 3 ot 4. los oollul-~ have ~te trait~os rosp-ctlvement with 50 mM and 100 ~M of Z~Cl~, ~vant the ~ solement do RNA. Dan- la ~ and- 5, d-- o~llul-s ~n~eot~es have been maintained for 50 generatlons without ~eleot~on by G4le avan ~ the extrac: ion. A fragment of 1~25 p~ isolated from pMTX5 ee representing the probe of H5 ~ bando 6) has protection ~ 15 nuoleotldes after d~gostlon Sl The letters in the left margin represent the sizes, in nucleotides, of the tragmenes used as molecular weight markers 207~0~7 Fi ~ ure 5: Detection by Western blotting of H5 in QT6 cells infected. The nuclear proteins, isolated from QT6 cells infected with DAH5 (lanes 1, 2 and 3) or uninfected QT6 cells (lanes 4, 5 and 6), were subjected to polyacrylamide-SDS gel electrophoresis and transferred to nitrocellulose. H5 was detected after incubation with anti-H5 antibodies and protein A1. The cells were treated with 100 mM ZnCl2 (lanes 2 and 5). Bands 7, 8 and 9 represent the equivalent of 0.5 μg, 1 μg and 2 μg of purified HS, and bands 10, 11 and 12 show the amount of H5 present tans 5.106, 10.106 and 20.106 chicken erythrocytes.
FiRure 6: represents p~OL. This plasmid was constructed from H5 transcription cassette containing the MTIIA promoter, the sequence coding of H5 and the polyA sequence. The coding sequence of H5 was eliminated by EcoRI + Smal digestion and replaced by an oli ~ onucleotide of synthesis presents figure 6. it is possible to consider replacing the promoter MTIIA by another promoter.
Fi ~ ure 7: Structure of OVA retrovirus vectors. The black triangles indicate the positions of internal transcription promoters, while the white triangles indicate the position of U3 deletions. The presence or the absence of an enhancer element, a CAAT box and a TATA box - in each downstream LTR is indicated on the right.
Fi ~ ure 8: represents the construction of OVA type vectors.
EXAMPLE 1.
Transfer and expression of histone H5 were performed dar ~ normal avian fibroblasts and transformed by a vector retroviral SIN. The vector, called pDAH5, was obtained by mutation in Ia LTR 3 'by eliminating the boxes; your CAAT and TATA. The coding sequence for H5 inserted into pDAHS was controlled by the metallo-promoter thioneine MT-IIA human.
Infection of QT6 quail cells by the DAH5 virus leads to the integration of one copy of a foreign H5 gene per genome and to the formation of properly initiated mRNA transcripts and H5 protein However, expression of the exoene H5 sequence was constitutive and non-inducible, by Zn ions. The level of H5 in QT6 cells was equivalent to that of a mature chicken erythrocyte.

5~12 Expression of H5 in transformed QT6 cells infected with DAH5 or transformed chicken embryo fibroblasts by AEV-ES4 had only slight effects on growth velocity e~
did not inhibit cell replication. Conversely, the expression of H5 in 5 of the normal quail and chicken fibroblasts was spectacular: the cells acquired the appearance of senescent fibroblasts, they grew very slowly and the nuclei seemed compact, they were often devoid of cytoplasm.
Analysis of post-translational modification of H5 in 10 QT6 cells showed that phosphorylation may be involved in Ies differential effects of protein in normal cells e ~
transformed.

I - MATERIALS AND METHODS
1. Construction of a retroviral vector for the transfer and expression of the coding sequence for HS
The principle of SIN vectors is illustrated in Figure 1.
The construction of the vector pTXN3' has been published recently (3 and WO 89/03877). The pDAI genome is derived from the virus genome ! 20 associated with Rous sarcoma-2 (RAV-2) after modification of the 3' LTR.
An avian LTR mutated by deletion of the U3 region isolated from the LTR of the avian leukosis virus RAV-2 between the nucleotides -149 and -15 (facing the transcriptional cap site: nucleotide +1). That is, the sequence downstream of the Sph1 site in U3 was 25 removed. This deletion eliminated the CAAT and TATA boxes but left intact the enhancer region located upstream of position -149 (6,16, 22J.
The deleted RAV-2 U3 region was then linked to the 5 R and U5 regions isolated from RAV-I LTR. That is to say, the dice sequence of U3 has was replaced by the Taql-Sstl fragment derived from the RAV-I LTR. This 30 fragment essentially contains the R and U5 sequences and 14 nucleotides residuals of U3. The Clal linker was inserted at the junction of the fragments derivatives of RAV-I and RAV-2. The hybrid LTR therefore has a deletion of 134 nucleotides which eliminated the CAAT and TATA boxes. The deleted LTR
thus reconstituted was used to replace the normal 3' LTR in the pDAI vector. The pDA1 vector was constructed from RAV-2 and it carries a transcriptional cassette containing the neo gene linked to WO 91/11190 PCI/Fl~91/00062 13 2~7aO67 -early promoter of SV40 inserted in the same direction of transcription of the viral transcription (fi~ure 3). Specifically, the internal Xho I fragment RAV-2's huge ~ has been eliminated and replaced with a cassette containing ~
in the order of the ga8 sequences thus carrying Xbal, the enhancer and the simian virus SV40 early promoter and gene coding sequence bacterial neomycin phosphotransferase (neo) (2S).
As a control virus, the enome pDAHY derived of pDAI by inserting a cassette containing the sequence coding for 0 I ~ hy ~ romycin ~ phosphotransferase (hph ~ between the early promoter of SV40 and if ~ nal polyadenylation of ~ SV Tk. This tape of transcript was inserted in the opposite direction relative to the neo gene.
More specifically, the pDAHY vector was derived from pDAI by inserting at Xbal site a cassette containing the sequence encoding hy ~ romycin ~
bacterial phosphotransferase (bph) ( I l ). This ~ene was put 50US on control of the SV40 enhancer and early promoter. The signal of polyadenylation of the herpes virus (HSV) thymidine kinase (Tk) gene was added at the 3' end.
As an additional control, the pTXN3' virus, a vector which expresses the viral LTR neo gene, was used in parallel.
The pDAHS vector genome was constructed from pDAI
` by inserting a transcription cassette containing the sequence coding for `1 HS linked to the 8ene promoter of human metallothionein MT-IIA. In 1 this construction, the HS sequence provides its own signal of poly-- 25 adenylation and it was inserted in the antisense direction with respect to the . retroviral transcription. More specifically, the ~ enome pDAH5 was constructed from pDAI by inserting a cassette containing the sequence encoding HS with its own polyadenylation signal linked to the MT-IIA promoter (13~. The transcription unit of H5 as a cassette of transcription of the 8th hph was inserted in the opposite direction with respect to transcription controlled by the LTR.
All of these retroviral genomes were transfected into the born from helper G cells that only released vectors retrovirals without helper viruses (helper free) ~24). The ~iter of the virus DAH5 without an assistant (helper free) was 100 rfu/ml (unit forming resistance/ml) as determined by its ability to induce resistance to G4 18 in QT6 recipient quail cells.

WO 91/11190 PCT/FR91/000~2

6'~14 2. Colluloc The ~ flbroblasts of chicken em ~ ryon t OE F) have e ~ e prepare and cultivate as previously described (9). The em ~ ryon sibroblasts of callle ~ OE F) were prepared and culti~ec dan~ leo m~moo oonditlonc. Loo chicken cell ~EV-ES4, let fibroblaoteo of quail CT6 ~l7) and leo cell~
quail helper G (~4) were cultured in the culture medium CEF Los AEY-ES4 cells were derived from OE F lntected by the vlrUs AE-ES4 (9); oeo cells have been retained in culture while ~using~ molc before being used The select~on for neomyc cell columns m e-recl ~ aunt ~ has been ~ alte in presence of G~18 (Gibco) at a concentration of 200 ~g/mi ;

3. Tr~ f-otio~ ti~lfootio~. Los helper G cells have been transfected by the polybrene-dimethyl sulfoxide method (14). Lcc suspected ~lraleo have ote rooueillieo laying p rlodeJ de 12 houroo ~ part~r doo oolluloo du productour rouc-confluentec (3 ml per box of 100 mm), pul~ centrlfugee~
for 15 mm ~ 2500 9 and filtered ~ cur dec Filtrec li ~ nt ,albi~ment the protelnes (~porei. Lo~ ceilulos. ensomocees d ralson do 5 105 ooiluloo per boito do 60 mm. have been lnfooteeJ
with l ml of Jolutlon ~rlrale for 12 hours~, pUl~
selected ~ tan~ G4l~ The remaining colonies were evaluated ;0 to 15 ~ours ~near the m sootlon 4. I~d~otio~ p~r l- z~od~ pro-ot-ur do ~T~ .d~ loe O-11Q1-ei~f-ot~--. LJ oellul-- have ~t~ lnou~-o~ laying 24 h in lo mlllou de culturo CEF ~upplomonee a~teC so ~M or 100 ~M of _3Cl2 (Yer~k, Lo ZnCl~ was proparo JOUS formo of Jolutlon of rosor-e 10 ~Y in a ta~pcn s~lsn ~30tonlquo and stocko ~ ~4'C

5. Cl~otiquo do oroios~noo doe oolluloa i~fooeOoa. The~
cell ~ have been seeded ~ at a rate of 5,105 cells per box.

WO 91/11190 PCr/FR91/00062 207~0~7 60 mm and the mll ~ had in ~ ulee was changed every ~ our. Every délal, u ~ e or trols box ~ by type cellul ~ ire have been trypslnlsées and the ~ ceilulos have been counted ~.

6. ~ULa1~JO of ~DN ot of ~N . A ~DN of high molecular weight~
was extracted as previously described (10). ApreQ digestion of 15 ~ g by ~ enzym ~ of ~ te ~ trlctlon approprlee ~, the ~ } }ragment3 have e~e ~paree par eleotrphore~e ~ur do~ gel~ d'agaro~ea 0.8%
The ~ DN has in ~ ulte eee tran ~ fere ~ filter ~ in nitrocellulo ~ e and hybrlde with a randomly initiated marked wave. Filters have summer fm aiemont lavo~ dan~ do~ extreme conditions expoae~ has a pell~cule ~odak-~-Omat. The ~wave ut~ ee was an EcoRI-~mnl fragment derived from pHTH5 The total cellular RNA was purlfied after lysis of ~ cells in a buffer containing 1 m ~ of Dithlotreltol and 0.2 ~ of Nom det P40 e ~ entiellorent com ~ e decrlt procedommont ~ 10). From~
exporienco~ dc cartographio S1 o~t eto offoctuoo~ in accordance Fa~aloro et al. (~). A fragment nt d- ro~trlctlon Bam~IP~tl contonating the ~equonce coding for ~5 a and ~i~ole a partlr of pMTH5 and ~ arc ~ the end 5 'by ~ r 32Pl ~ TP ct the polynucleoelde : ~ina~e. The probe has etc coprecipitéo a ~ eo 10 ~ gd' total MN e ~
rerl6e in ~ u ~ pon6 ~ 0n in the hybrldatlon Sl buffer containing 40 nM of Pipes lpip~raz me NN' bl~(ac~de 2-ethane ~ulfo m que. pH
6.4) r 400 mff of NaCl, lm ~ of EDT ~] The hybrldatlon ~ ete effectllee hangs ~ nt 6-9 b ~ 56-C, then the hybrids6 were digested with 900 U
do nuoleaso Sl par mi ao tampon do dlgostlon (~0 m~ of acetate ae ~od~um p~ 4.5, 250 m~ do NaC1, 2 mM do Z~S04) for 30 mln ~
31'~. The DNA reslJtant ~ the nucleus ~ a-- S1 has te ~ f malely not ~ e in un gel do polyaorylamldo ~ 6~ oe expo ~ ea uno pelliculo Yodak-~-Omat. The identity of the protected fragments has been determined by comlgratlon with appropriate molecular weight markers

7. Tr-~of-rt ~l-otrophor~tiqu- d-- prot~ o our p~pior do nitrooolluloJo. Los hlstonos totalos one ote oxeralte~ ~vec H~S04 Qes nuclei purlfies ~19). Lo~ ex~ral~s gather on; e~e WO 91/11190 ~ t ~ 6 PCT / FR9l / 00062 dlalyses and freeze-dried ~ es. The hl-tones were analyzed by electrophoresis on dc polyacrylamide-SD~ gel (dodecyl Qulfat~ from sodlum) (15). The eloctrtranstert of poiyacrylam~de gei on n ~ trocoiluosis paplor was carried out for 1 hour ~ 24 v and the filters were f malely lncub ~ avoc antlcorps antl-H5 and protein A labeled with 125I. 1-10.105 spm (Amer~ham Corp.) The productlon of antlcorp ~ in your lap ms lmmu m 3es by ~ 5 has been rapportoo prooodom nt (18) L-~ anticorpJ antl-~S ont ~t~
pure ~ fies lmmuno ~ p ~ clfiquoment by affi ~ lte chromatography with H5 designed at Sepharosc 4~. Lcur Immunoreactlvite ~ ete monitored by solid phase radlol immunoassay (19).

8- I ~ om nofluor ~ o-noa. The cells ~ have been cultured on ~
lame~ L~b Tek ~i}es labora~or~es), puls fixed in tans l~ethanol aDSO1U SU1V1 d acctono at am~lante temperature. After washing with PES-PMS~ [10 mM of pho-phato do sodlum ta2pon, p~ 7.4, 0.15 M
of NaCl, 0.1 m ~ of ph ~ yl methyl sulphonyl fluoride], the cells were lncuD ~ es for 30 m ~ na 4-C with ~ ec of ~ antl-H5 antibodies purifle~ by aff m lt~. DJ antloorp~ anti-lapln marquoJ a la fluor-soolne ont on~ult- ote a3OutoJ ot loJ o-llul-J oolorooJ ont summer exam m ee~ ~u mlcroscope Olympu~ equips you with a ~pl-illum matlon.
The ~ phot~micrograph~ were taken with an Olympus oamcra and Kodak color film. ~e~ temp~ d'expo~t~on oe deve-ioppe nt oon ~ tants have been utlll ~ eJ for the lmpro ~ Jlon.

9. ~u~ d-pho~phoryl~tio~ d-ln prot~ino ~5. The collulea ont oto ancuDeeJ pentant 4 h tanJ un mlllou es~en~le ~n~mal x-mpt do PhOJPhat~, PU~ Ont and~ mart~uees pentant 1 h tan~ the same m~liou by ~ mCi de H3P04 m~rqu~ au 32p ~Amer~h~m Corp 1 Le~ hlJtonos rlches en ly~nc have been xtra~teJ with the ~cate perchiorlche ~ 5~ e~ submitted ~ an electrophore~ e like decrlt higher ~ high. The pho5phorylat~0n of H5 was detected after r.ran~tcrt tu qe1 sur paplcr de nltrocellulo~ee~ autorad~ograp~1e de la plaae de transrere ("blot.'l. The mome place de eransrert ~'blot") was used for the detection of protelne H5 with !~n antlbody ant~-~5 and ur. con~ugue protects me A-peroxidase from -2075~b7 ramfort ~mcr~bam) ~5 ~ been lnally ~lsuali~cc by lncuba~lon in 10 m~ of Trls, p~ 7.1, 0.0~5% of H202, 0.04% of 4-amlnoantlpyrlne ~sigma! and 0.2~ of phenol.

ll - RESULTS

ulal~-- d- l'~ t d- l'lJU~ ir~ d~o le a-lluloe cnt6 i~fo~t~-rp~r OU~ls. CT6 cells, a line of chemically transformed quail fibers (17), were :. 10 lnsected~ by a reserve dc vlru6 DA~5 depour~ud ~ssistant ("helper free"~
pll6 selected by G418 and amplified in cultures polyclonal. These cells have no viruses and the ~
supernatants of cultures superinfected by a ~ lrus assistant (helper) normai etalent lncapa ~ los to transmit the resistor ~ G41 ~.
Since no vlru~ could be found. we conclude that in the cells infected by D~S, the pro~irus mtegre e~t inactive from the point of ~ ue transoription. To determine the : presenoo, the gold pn~satlon and the number of coples of the DNA vector ntrodult, a ~DN of h~ut polds l~oulairo pro~-nant of cells CT6 mfoctoes or not rnroctees has ~ te prepared lS ~ g of DNA have been digested ~ by restriction enzymes and were 0Xam ~ 30S by the : metnode of the Soutnern blottlng with a- probe oont-nant the reglon coding for ~5 ~flg 3! L'-nzyme de rostrlotion ~hol ooupe aeux fols ~ the inter~ our iu vector and the prosonoo d a fragment of 4.6 : 25 kiloka ~ s (kb) or therefore characteristic of the genes ~ntacts ~5 and noo. The expected fraament do 4.6 kb has been deleted tanJ leJ oellule~ infootoos ~bando ~), ma~s 11 n'~talt pa~ pr~ent in DNA do ~ oelluleJ non lnfoct ~ e5 ~ b ~ nde 6).

' ~18 preo ~ ment the presence of an MTII-H5 fragment lntact, the AON of Infected and non-infected GT6 cells ~ was cut by ~ bal.
As expected. a 2.1 kb fragment was detected in the 6 cells lnfected ~ lband 5) and has not been detected in cells no ~
lnfected (band 4) This data shows that the ln~ere provlrus do not have any major rearrangements6 and present the expected structure. Le~ rragment~ de haut po~ds molecuia~re : ob~er~es in the ccllule~ mfectce~ and not m focteos correspond to the endogenous H5 gene. A p rtlr ~ntensltes ~es bands at 2.1 and 9.5 kb in the cell ~ i ~ fected, we have estlme that an exogenous ~5 gene has been integrated per genome ceilulalre To conflrmer this ostamat ~ on, we compared l lntcnslte of the oande of 2.1 kb has ~it thundered by a ~uantlt~
; ^onne of sequences H5. 15 ~ g DNA ganom ~ that avlalre repre ~ hear o.106 genomes before alres. Bandos 1. 2 and 3 contain respectively 105. 6.106 and 6.107 cop ~ es of plasmlde p ~ Tho dl ~ ere by EcoRI. If we compare the hybrldatlon intonslte of the 0.3 kb fraymcnt from bando 2 to 2.1 kb fragment of bando 5, we con~rmon~ that a COple tU proVlrU~ 3A~S has been msered by genome To determine the rate of specific transcripts of gene Y5 in CT6 cells mrected by DAH5 and not mrected~ ~a quantitative analysis with the Sl nuolcaje was used frlg 4) The ~ ~ cnde of ~ DN ct. ~ lt a 5 ~ m ~ I-Pstl fragment of 1026 bp ~ sole of the plasn~ of pMT~5. This fragment contains the promoter te MT-;;A ee lk nucl ~ otmde ~ of ~ equence coding for H5. Lo~ tran~crlts ~n1t1es to the promoter do MT-II ~ do ~ have otro protogcs aur uno lengthuour de 316 pb (l3~ Co~mo planned, a ~ragmont protected do pros do 315 nuclèat ~ te6 has been detected with RNA te cells OT6 ~ ntec ~ eec by ~AH5 (bands ~ at 5~ A protected rracment of the mo~me tallle has been detected in aes collulos OT6 lnrocteoJ by DAH5 ma m~onu-s en culture for 50 aeneratlon6 without selecting on p~r G41~ (band 5~
No protected fragment born ~ e detected with cell RNA c 3TG
non msec~-e~ (band 1~. The promoter of NT-I;A huma~ne little~e~re ,ndu~ pas des me~ ml~ur~ ~el~ que 7n~ 3) Po~ es~e:

20750b`7 the increased expression of the HS gene ~ partlr of the internal promoter, OT6 infected were exposed for 24 ha SO ~M (band 3) or 100 ~ M ZnCl ~ (bante 4) avan ~ isolation of RNA. The quantity d mRNA of ~5 did not increase significantly compared to the A ~ N of the cells not ~ ntected (band 2!. We can conclude that the transcription of the sequence coded for HS was effectively and corroctoment im tiated in the cells ~ OT6 and that the addition of Zn~ did not increase the degree of transcriptlon. Of plus the ~equence coding for ~5 a king~ tran~reree tan~ the OT6 cells eealt expressed stably laying at ~ olns 50 ~eneratlons of culture~ freed~ from the ~electlon by G418 2. EspreJ~ion do 1~ prot~in~ ~5 dnn~ d~s fibrobla~tes OT6 infe~t~ pnr 1- ir~JD ~ 5. The JyntheJe of hl~tone H5 ae~e ~e~urea dan~ de~ cell~ CT6 lnfected by DA~S and not ln~ectecs The total hlstones were extracted from isolated purified nuclei cells~ re~ poctivo~. Los prote mo~ have been analyzed by electrophoreJo Jur gol of polyaorylamido, tranJfereoJ on nitrocellulo~e and the pro~ence of ~5 was detected after ~ncubat~on the transfer range ("blot") with an antibody antl-~5 and the prote me Al~SI~fig. S). Antibody cost na ton no slonal when he was to~te against 10J hia~one~ to~aies te cell~ OT6 no m foctee~ tralteos aroc 0 ~band 6), 50 µM ~band Si ou 100 ~M do ~nC12 (band 4) A slgDal clearly readable with ~ec the me~e ~ob~l~to of ~S pu~1f~ee (bandJ 7, ~, 9) or of ~S
of hi ~ toneJ total chicken erythrocytes (bands 10.! 1. '2). 3 eto detocto avoc los hlstonos totaloJ do 5,106 oollulos ~6 l~octoes per ~5 tra~too~ avoc 100 U~ do ZnCl2 (bante 1). 50 µM
of ZnCl~ (bando 2~ or no ZnC12 (bando 3).
: quant ~ fler the productlon of H5 in back cells OT6 infoctee ~, your known quantltes of H5 purllee. 0.5 µg Iband 7). 1~g lcande ~ and ~ ua ~bando 91, or d'hlstono~ totalos extra~tes de - lQ6 (band l0). l0 l06 (band ll~ or ~0.l06 ~band 12!
erythrocytes of adult chickens were subjected to a eiectro, D~ore~o on the gel mome. A partlr to coq roouitats, we can conciure that ouant ~ te of H5 pre ~ onte in 5 10 - wo gl/lllgo ~ Q ~ 20 PCT/FR9l/00062 cell- OT ~ lntectee ~ p ~ r D ~ H5 is eqUl ~ alente A la au ~ ntlte dc H5 produced by 5.10 erythrocytes of adult chickens and that ie did not au~men~e the ~5 productlon ~egre. The io~ailiatlsn nuciealrs of H5 in le3 3T6 cells lntec~oes? r DA~5 was monitored by munofluorescence.

3. Propri~t~rd~a oollul~a 0~6 ~td~of ibrobla~t~i trnnafor~e pnr ~IEr-ES~ ~spri ~nt 1~prot~ino ~5. From cultures of OT6 inserted by DA~5 have developed large ro~ers ~ n ~ es containing ~ cell ~ phenotypically not different ~ i ^
0~6 cells not lntected or lnfeceeec cells by the vlru--.emol~ e ia even f~corm chicken .lbroblasts passed on by the retrovirus to the AEV-ES4, were infected by 3AffS ~es celP:les cn~ develops deo foyer~ ne pou~ant ~re dlstlr ~ gue ~ of those of cells lnfected by TgN3 'and fl3 n have could not bereamplltleos~not shown~. ~our ox~m~ner de racon plus approrondle sl the protolne H~ pouvalt a~olr a cer~aln effe~ on the ~ tran ~ cells formed, we have comçare the clnetique of roljsance of uninfested OT6 stools to that of infested OT6 by DA~5 or p~r ~N3'. Collulalros ~rols typos have been ~ nseeds lndl ~ duellemon ~ ~ the momo donslte in boi ~ e ~ this ju mm and -ul: lves dan ~ the same conditions. Ensult-, ~dlfferen~r del~ls, i~ cells have been tried and counted It can be seen that the QT6 cells infected with DAH5 have presented a growth piu~ slower than lç~ coilule~ ~T6 no ~nfeotoes or Infeotoos pa~ TDN3' Los temp3 do gonorat~on may have been es~me _0 h for e3 ceiiuls~ OT~ ~emo1ns and at '5 h for ioC coilules CS~
?rOdUlSant A protelne H5 We can therefore conciurs that i express:-r. ie i~ protein H seems to have little effect ~ u:
13 _role~anoe des ^cellule~ 0~5 et le5 fibroblaste3 tr n3~0rms p3r A~V-ES~

4. Espre~io~ and effects of 1~ protein ~5 daD~ of fibro~lasto# nor us 3rd ~ibroblasts ~'embr70n chicken WO 91/11190 PCT / FR9l / 00062 207aO~'7 (OEF) and quail embryo fibroblasts (OEF) were lnfected either by the vlrus DA}I5 SOlt not the vlrus temolns, T ~ 3' or DA~Y 31X ~bear apros selectlon in a mlileu containing G~i6 large clones donated to fibroblasts have been observed in ies Cultures of Ca~lle and Chicken Cells Infected with Cnaque VlsUS temo m In all infected crops by DA ~ 5 only a few approved petlts of dispersed cells have been oDJerveJ. These cells grew very slowly and it could not be effectively replicated, in therefore, they could not be amplified In addition, 1~
part of these cells are sick and presently a morphoioale apiat~e eeo~l~e ~dont quo a cell~ de flbro~ia~tes _enescents The same observations were made in SlX
md~pendantos experiments using fibroblasts secondaries of chicken or ca~lle Since the ceiiuies ~nteceeos by the vlrus tomoln DA~Y sombla~ont aUJs~ sa~nes that aes normal fibroDlastes, the ofot obJcr~o with the vlrus D~5 is the most likely, the expression of the sequence coding for ~5 As these collulo~ ao to~oloppa were too bad, we couldn't effoctuor d'analyJe l~oulair- ot on oon~ once the oxpro~s~on of the protein ~ 5 was analyJee by ~ ounofluoresconce;
a coloring according to Giemsa Wri~ht of the same field a and~
ettectuoe for a~st~ngu~r ies ot cyeopiasm LOJ f~robla~to~ do poulot normal ~nfoctes par DA~5 one exhibited heterogeneous fluorescence when wiped with anti-HS All cells presented with anti-HS
nuclei sluoroJoen~s ee the lntonslto to the ~tluor-soonc~ est on Donno oorr~latlon av-o the h ochromatlne struoture The nuclei bautom~nt condonJos pre~onta~-nt la fluor-saonce l~ pius :lve, tandl; nuclei with condensed mo ms chromatin presentalene uno falble tluorosconc-Some nautomone nuclei cond-nJ-J s-mbialene eere oxpuises Tnverse cells normal non-ln ~ s tibroblastos do not suffer from any significant luorescence containing nuclei with a typical dark chromate .

wo g~ go Q~ 22 PCT/FR9l/00062 5- ~5 e~t phoephor~l~e dnn~lo~oellulo~ CT6. It requires results obtained that the expresslOn of 1 ~ protelne ~ 5 has only one ea ~ ble erret on the growth of transrormees cells. Am sl, as next step. we looked for mods ~ one poet-traductlonnelles possible~ of this protelne entailing its lnact1~atlon in these cells. ~ as do ~ born as pho phorylatlon could be lmpllquee in the m actlvatlon of ~5 at the stages precoceJ of erythropol ~ so normal ~27), we examined i ~ extent of the phosphorylatlon of H5 in lo cell ~ OT6. From OT6 cells infected with DA~5 and uninfected were incubated in the presence of ~3P04 mark at 32p Le~ h~stone~ rlches en lys me were followed by eXtraltes, subjected to electrophoresis on a gel te polvacryiar ~ de-SDS, eransrees on r.ltrocellulose and l lmportance of phosphorylatlon a ~ t~ dlroctomont anaiyQee by autorad~ograph of the transfer range ("blot"~. A
pho~phorylated protelne, absent in OT6 te~olns cells (band 1~ a ote dotoctoo in back cellulo~ OT6 in~toct~os by DAH5 (band 2\ For id-ntlfi-r cotto protein, the same place of transfer ("blot") was treated with ~ oc an antl-H5 antibody, a combines ~rotate me A-peroxidase and 4-a~ino-antipyr~ne as cubstrat. The prote me H5 has been dotectoo in the CT6 collules m rectee~ by DAH5 (bando 4i ot dan~ lo~ hlston-s totalos ^xtral~çs d~eryt~rocy~es de poulçe!k~nde 5), while none : one has not ob ~ or ~ oe in the ~ OT6 temo ~ ns cells (band 3). The protect me phosphorylated detected in the bante ~ basically ia meme moblllte as the H5 protelne identified in leJ bandoJ 4 and. 5 We can therefore conclude that the protein H5 05t phosphorylec in cellulo~ OT6 l~OCt~O~ by D~S

6. Discussion These example temon ~ re the transfer and 'expresslon eff ~ cacos of hlJtone H5 in normal fibrobiastos e ~
~ransfo,mes The SIN vector decrlt in this study has ett _~nstru~ta parta: ~u ~!lruS of leukosis a~rla~re RA'-~ La WO 91/11190 23'; PCT/FR91/00062 2075~!h'7 delee ~ on de ~ ~ equenoes of the promoter dan ~ the 3'LTR contult ~
the integration of a pro ~ irus mactif tan ~ the ~ infected cells.
Vector transferred copy de oequenoo ~5 etrangero in the genome do a~a~ue oelluie lnfeotee It was found that this transfer was very stable being gives that the inserted sequence H5 is still present in cult cells v~oJ laying 50 gonoratmons without pres~lon selectlve.
In cells infected with the SIN vector of DAHS, the ~ sequences ~ 5 were deleted from the promoter of MT-IIA Internal. ~ as long as no vector virus could be recovered from mt cells, ected even after 5 ~ ur mf, ectlon with a auxiliary rus, no transcription has been m tlee a partlr of the 5'LTR. The SIN tecrlt lCl voctour is quite reliable and efficient.
lS
We did not observe the effects slgnificatives doJ ion~ Zn~ on the effioaoito to tranJcription of H5 tans let oellulos infoctooJ. ~ome without stimulation. the bottom of transcrlpt~on te H5 int~groo was actmvo. The actlvlto con tltutlve of the MT-IIA promoter and its absence of repon ~ e ~ ux lons Zn ~;
in the ~ sector of DA ~ 5 could otro m horontos a co promoteu:
in an avlairo collule. or can~ we~ otro du~ to the eftfots st m~lators croe~ by the promotour of SV40 prox~mal oY the LTR

The quantlto do H5 produmte by collulo OT6 m roctoo not DAH5 'e~t ~omblablo a oolle trôuveo in doJ orythrooytot av~alres maturoJ C'e~t why~, 10J offoto do H5 danJ lo~ oellulo~
in~ect~o~ can be compared to oonditlonJ phyJ~oloq~que~
natl~os.
The exprorJlon do H5 in back fibroblastos tran~form~, -'est-~-say quail fibroblasts 0~6 or doJ fi~robla3te~
do~oryon do pouiet transtormos par AEV-ES4, n'a J lDhlbe ieur orolssanco ln vltro. OT6 lnfoct~os not DA~5 no proJen~ent that a small increase in their temp ~ do goratlon, from 20 ~ 25 - hours, but their ability to csoitre commo a li ~ born established ~ a!. and WO 91/11190 ~ 3 ~ 6 24 PCT / FRgl / 00062 not been diminished. This result clearly shows that the constant production of ~5 has little effect on the phenotypo te ' transformative f~robiasts.

Normal quail or chicken fi~roblasts express the exogenous HS protelne present a potential you csolssance tre ~
l~m~te and could not be m replicated m developed. Moreover. those cell~ looks like tos cellulo~ ~onoooonto~ avoc uno flattened morphology, a vacuoliated cytoplasm of the nuclei o condensed. The condensation of chromatin in these cells etalt dlrectement llee has the expre ~ lon of ~ 5 given that the more condensed new ones have pre-contaminated the fluoro-cence i rained VlVe when -ls have been treated with the antlcorp ~ ~ nt ~ -H5 In addition, some nuclei with highly condensed chromatin were expels cells. These aonneos therefore show ~ue HS
shows very different erfots in normal tibroblasts and transformers.

Lo rolo blolog~quo ao ~5 a oto rogulloremont lio a l'ldoe it is about an aonoral reproJ~ our. Co po mt do vuo ost conflsme by the fact quo, when the nuclei have been roactlves by the chicken erythrocytoJ fusion with HeL cell back, the evonemene-2 procooo~ findings include a decondonsatlon toe ia cnronat mo, uno initiatlon do ynth--- d'RNA ot uno di~parlelon ae HS (~) A recent realiJeo study by mlcro-injectlon of H5 dan~
proliferating L6 rat mqobla ~ tos. cellular lignoo qul ConcJorVe le poton~lel do se differonclor on myotubulos, a mon~re that; a ml ~ ratlon and the localisaelon to H5 da ~ - lo- noynux ~ alon-.
llees a la dlm~nut~on do la taillo dos cells, la vacuDl~satlon of the cytoplasm, the ta~somont doJ nuclei and the inhi~lt~on of;~
tran ~ crlptlon and do the repllcatlon 14).

;e fa ~ t that H5 is highly pho ~ phoryl in the ~ collule ~
^T5 transformos ~nfoctee~ reports the procoJsus o~serve in 1 normal erythropolosis. The pro~ence te ~5 was d~tectse ~ a nlveau redult in dos çrythroblaJteJ on ~tat de dlv~slon prococe in these cells. 1~ pro~elne newly oynthet~ee es~

25 2075o67 trln~porte d~n~ the ~oy~u where it is modified by a process mplquant the unidirectional pho ~ phorylatlon continues during laqueile neut phosphorylated groups can be introduced.
At the termlnal stages of the maturation of erythrocytos, ia ; phosphorylated protein is dephosphorylated e~ this last event~
e ~ t correlates well with the arrest of DNA synthesis and of MN ee the condensatlon of chromatlne (S. 26). We have not could not analyze the extent of H5 phophorylation in normal fibroblasts mfected with DA~S given that these cell ~ n paQ could be a ~ plified.

Lc moca m ~ me exact lntoractlon of ~ 5 avoc i'ADN in io ochromatlne does not ~ t ~ known. However, one can envlsager a qUl permot mechanism ~ the cell of traitor do grandeJ quantltes iS of HS without deterloratlon of the m ~ tabollsmo ccllulalre, The pho ~ phoryiatlon of H; proault uno prot~lno qul a mo~ns d'afflnlte for DNA. by neutrali~atlon de~charge~, C'e~t why, the~
modified amino acids must otro looalioo~ dano domains what are dec~fJ for this intor~ction. that is to say the region C-terminal (re~idus 100-182) which where the region that contributes the more has the condensation of chromatin by neutrallsatlon partlel;e doJ phoJphates of DNA. The unmodified ~S showing plus a'afr m~t~ for lo~ structuroJ d'ordro ~uperleur de cnroma~ mo ot from ACN. ~ol~oralt otrolterNnt a 1'ADN, spoclatoment to the r~glon~ rlcho~ in A-T. on ~ o condon5ant ot/ou en chan ~ e ~ nt sa confor ~ at ~ on t20) ot, ~ e oette manioro. by blocking the binding of the maoh~nerle of transcrlptlon and ropllcat~on.

Il sora ~ntoros~ant d'otudior lo~ vo~oo motabollque~ qu~ son~
d~rectomont rosponsablos do phosphorylation do ~S in back -ellulos eransftormoos and lour rolat~on avoa l'oxpro~slon d' oncoaenes Since lo~ tibroblaJtos OT6 and AEV-ES4 ~have ~ransforre~ by do~ lnltlaux events differ~. It is likely that both processes aerate common H5-kinases L'analv~o des motiflcatlons te HS dan~ dlvorJos collules nfected by DAHS can ^rre utll! for det~ mr the role of oette WO 91/11190 PCr / FR91 ~ 00062 6~26 protein in cell replication and differentiation and in the knowledge of the mechanisms of onco~enic transformation.

Vectors carrying RAV-0 virus sequences and an RSV 5' LTR
Two vectors carrying a 3' LTR derived from the RAV-0 virus, naturally devoid of enhancer sequences, were constructed.
The OVA-ZZ vector carries the following sequences from 5' to 3':
- An LTR of the RSV virus, - A leader sequence of the RAV-1 virus, - The NeoR resistance gene, - - The ~ ene of E. Coli coding for a beta-galactosidase provided with a slgnaJ
nuclear targeting, under the control of the SV40 virus promoter (gene SVnlslacZ, supplied by JF Nicolas of the lnstitut P~asteur in Paris. see Bonnerot et al. 1987, Proc. Natl. Acad. Science. USA 84. 6795-6799), - The 3' non-coding sequence and the LTR of the RAV-0 virus.
The OVA-Zdel vector is identical to the previous one except in its 3' part where the 3' LTR of RAV-0 deleted by 50 nucleotides comprising its CAAT box.
OVA-Zdel virus can be produced at a titer of 1000 particles/ml conferring beta-galactosidase activity on infected cells.
On the other hand the ~ ene NeoR is not expressed in the infected cells, this which confirms the self-inactivating character (SIN) of this vector.

~OrA-2Z
LTR RSV NtoR SVnl~l~cZ l.TR RAY~) ~ } OYA-Zdol L ~r~--~
LTR RSV NeoR SYnl~lecZ LTR RAV(0) deleted '':'' , ~, .
' WO 91/11190 PCI/FRg1/00062 2o75o67 The OVA-ZZ and OVA-Zdel vectors were transfected into the Isolde helper line. Supernatants from pools of resistant helper cells with G418 were titrated on QT6 cells on the one hand by selection with G418 (NeoR title) and on the other hand by in situ revelation of the beta-5 galactosidase ~ lacZ titer). The ratio of the two titles reveals that only OVA-Zdel is capable of significant self-inactivation: OVA-ZZ gives 100 G418rfu/ml and 100 lacZfu/ml. The G418 resistance activity of cells infected with this vector probably results from the transcription initiated at the 5' LTR, which therefore remains fairly active. So we have builds OVA-Zdel, a vector similar to OVA^ZZ but whose LTR 3~, still from RAV-0 has been deleted from its CAAT box. The titration was produced from 20 clones of transfected helper cells. One of the clones produced 1000 lacZfu/ml and none (~20/ml) G418ru. This result demonstrates the efficacy and safety of the OVA-Zdel vector.
Subsequent experiments have provided a better understanding the factors that control the efficiency of gene transfer and self-inactivation, by comparing OVA-ZZ and OVA-Zdel with a construct carrying an RSV 3' LTR (OVA-D):
- The transient expression of nlslacZ ~ ene in QT6 cells transfected cells is reduced 5 to 10-fold by deletion of the RAV-0 LTR.
- The level of messenger RNA initiated at the LTR S' in the cells G418-resistant helper is not affected by 3' sequences.
- The encapsidation of the RNA carrying the RAV-0 sequences is reduced by a factor of 2 or 3.
2S These results explain why the lacZ titer obtained with OVA-Zdel remains 100 times lower than that obtained with OVA-D.
Avian leukemia-derived vectors were constructed with a internal transcription promoter and various non-coding sequences in 3'. Deletions are introduced into the downstream U3 LTR element ~3') to 30 achieve self-inactivation of LTR-mediated transcription after a . replication cycle. However, the 3' non-coding sequences seem determine not only the self-inactivation ability of vectors, !NaiS also the efficiency of gene transfer. Further analysis reveals the influence of these sequences both on the expression of the internal gene 35 than on RNA encapsidation. Elements built with a 5' LTR

.. . ~~'. ' . .

:~ .

WO 91/11190 ~ PCr/FR91/00062 of RSV make it possible to achieve efficient gene transfer while strongly inactivating LTR-promoted transcription at 5~.
The ability of self-inactivating vectors to produce a high-fire viral preparation and fully inactivate the promoter LTR relies mostly on the mutation that is created- However, the consequences of the muta ~ ion of non-coding viral sequences in 3 'on the transcription, processing, translation, encapsidation and transcription RNA reverses are very unpredictable (R34, R43, R47).
We describe here the construction of several vectors derived from 0 avian leukosis with various 3' non-coding sequences. The influence of 3'-ceasequences on virus expression is further analyzed and reveals its various consequences on virus expression and replication.
MATERIALS AND PROCESSES
Construction of OVA plasmids (fi ~ ures 7, 8) DAI is constructed by assembling in pBR322 the fra ~ ments following: an XhoI fragment from pRAV-2 (R45) carrying an LTR
Fully active RAV-2 and a leader sequence, an XhoI-XbaI from pTXN3' carrying part of the GAG sequence of the avian erythroblastosis virus (R29), an Accl-Bgltl fragment of pSV2Neo (R50) in the forward orientation, and a deleted 3' LTR. The deleted LTR is created by assembling the 3' part of the RAV-2 genome (from XhoI to a Clal linker inserted in the site SphI which is located 149 nucleotides upstream of the capsule site). This deletion removes the TATA and CAAT boxes.
OVA is built by assembling the polylinker of pBluescript (Stratagene SanDiego), the following fragments: the Ndel-EcoRI fraRment of pUT507(RS2) (carrying 0.5 kh of Schmldt-Ruppin strain RSV sequences D, covering part of the v-src gene and the upstream part of U3), EcoRI-XhoI
of pRAV-I (R35) (from U3 to GAG) an XhoI-Xbal of pTXN5' (R29b WO91/1119~ PCT/FR91/00062 .: 29 207aO~7 (NeoR gene) and the SalI-BamHI fragment of pMuMLVSVnlslacZ
(R24) (the SV40 promoter) followed by a mutated version of the lacZ gene of E. coli encoding a targeted protein to the cell nucleus). Given the insertion small fragments of pBluescript polylinker, this structure is flanked by two restriction sites spI.
To build OVA-D, we transfer the Ndel-XhoI fragment of OVA between the Saml and XhoI from pBluescript (to provide a sequence not encoding in 3' complete RSV type), then we insert the OVA Spel fragment in the upstream Spel site in : right orientation. We construct OVA-ZZ by transferring rant the HindIII-BamHI fragment of pGJ18 ~ R19) carrying IS two copies of RAV~0) LTR in pBluescrlpt (giving BluZZ~, pu~s by inserting the OVA Spel fragment into BluZZ's ~indIII site in the correct orientation, after partial filling of the protruding ends by Klenow polymerase. We get OVAZdel by . 20 deleting BluZZ by digestion with Ndel and religation at low concentration of DNA ~ giving BluZdel) and by inserting the OVA Spel fragment at the HindIII site, Cellular culture All reagents are purchased from Gibco. Isold ~R44 helper cells are cultured) in a William5 medium supplemented with 54 of serum calf fetus, 2~ chicken rum, NaCl ~60 mM), penicillin ~1000 u/ml), streptomycin (0.05 ~, phl ~ omycin (50 ~ g / ml) and hygromycin (50 ~ g / ml). QT6 (R36) and fibroblasts are cultured chicken embryos prepared from .

WO91~11190 PCT/FR9ltO0062 ~30 from 10-day-old S/O SPAPAS embryos (CEP, R45) in a HamPl0 medium supplemented with 5% of calf serum, 1% chicken serum, 2% broth phosphate tryptose, penicillin and streptomycin. We change the medium every three days.
Transfection, virus production and titration Purified plasmid DNA is transfected by production of CsCl (R39) bands in cells

10 Isold either by calcium phosphate precipitation (R40) or polybrene (R41). ~n sele~ion the cells (250 ug/ml) with G418 (Gibco) 24 hours after tranfection, for at least 10 days. We harvest viral supernatant from cells 15 subconfluent 16 hours after changing the medium, and centrifuged (2000 g, S minutes ~ to remove cellular debris. For titration, we add ~l to 500 ~l of helper cell supernatant to QT6 or CEF cells (5 x 10 cells per 25 cm2).
Cells infected with G418 are selected at after 24 hours, or they are stained to determine nlslacZ activity after 48 hours ~RS3) RNA analysis RNA is extracted from helper cells 25 or centrifuged supernatant (45 minutes, '5000 rpm in a rotor of ~ ackman SW41) by digestion at the prot~inase K and ph~nol-CHCl3 extraction as it is said (R39). We hybridize the blotting products Northern slitter and blotter with a sample of RNA synthesized in vitro labeled with 32P using standard techniques. We proceed to establish ment of the Sl card as it is said (R39).

-~
.
'--WO91/11190 PCT/FR9l/00062 207aO67 RESULTS
A large deletion in the promoter elements 3' LTR reduces titer virus Figure 7 shows the structure of DAl, vector carrying the g ~ the ~ eoR promoted by the promoter SV40 and inserted in direct orientation(compared to the transcription of the virus). We delete the sequence U3 of the 3' LTR from nucleotide -149 to -15 10 (to capsule site; see materials and methods).
This deletion includes the CMT box and the box TATA but neither the enhancer sequence nor the signal of polyadenylation. Several other vectors are derived from DAl by insertion of an additional gene 15 between the 5' LTR and the SV40 promoter, in the orientation reverse tation.
These constructs are transfected in the helper cells, which are then selected titrated with G418, and the recombinant viruses are titrated 20 from the cell supernatant. The Viral Headlines are low, ranging from 3U G418 rfu/ml to 200 G418 rfu/ml when testing cell populations clonal. Re-infected cells are superinfected resistant with a competent helper virus for the 25 replication to rescue uninactivated viruses.
As no recombinant virus was rescued in a set of about 100 clonea cellulair-a 1nfoct~s, we conclude that the deletion effectively inacts the 5' LTR promoter after a cycle of replication 30 true cation.
Sl RNA analysis can detect LTR- and SV40-promoted RNAs in cells WO91/11190 ~ PCT/FR91/00062 ~32 transfected helper resistant to G418. (An expe-control test was successfully carried out with pTXN3', previously described vector (R29J). Thereby, it is likely that the stability and/or treatment The 3' end of the RNA is affected by the 3' LTR mutation. Adding a poly-signal SV40 adenylation downstream of the 3' LTR fails to restore blir viral titer ~data not shown). By Consequently, an approach similar to that which has 10 has been used successfully to generate self-generating viruses.
inactivating agents derived from spleen necrosis virus rat results in poor success with derived viruses avian leukosis.
Modifi ~ ation of non-coding sequences in 3! only To allow a direct comparison of various 3' non-coding sequences and study precisely their influence on the efficiency of the trans-gene transfer, a series of vectors are being developed which differ only in their 3' end. We : 20 achieves by constructing OVA, a plasmid that can provide as a single restriction fragment ; cassette carrying (Ae 5' to 3') a fully active LTR
of RSV type, a RAV(1) leader sequence, the sequence encoding NeoR, and the nlslacZ gene controlled by the 25 SV40 transcription promoter (such as the SV40-nlslacZ is placed in the front orientation, it e3~
conceivable that reinitiation of translation could occur at low frequency ~on the transcripts promoted by LTR in 5', and reflect a low 30 part of the translation of nlslacZ).
The OVA cassette is inserted upstream of various L~R

WO91/11190 PCT/FR~1/00062 33 207~7 that provide the signal needed for processing 3' end of LTR-promoted transcripts in 5' and SV40.
As a positive control, LTR is used.
RSV fully active as LTR in 3' and we obtain OVA-D. OVA-ZZ has the RAV(0) LTR no mutated as a 3' sequence. This LTR is naturally resides in an endogenous provirus which can be cultured at a relatively high titer (R32) and has been shown 10 possess promoter, terminator, but not to enhance ~Rl4, Rl9). We therefore expect build product OVA-ZZ a very limited self-inactivation but a transfer effective gene. To increase the ability to self-inactivation, OVA-Zdel is also constructed in deleting 50 nucleotides comprising the CAAT box of RAV(0).
Viral titer is influenced by sequences non-coding in 3' Virus assays are performed based on : in situ revelation of beta-galac- activity tosidase or G418 selection from cells stably transfected helper ~Table I).
The titer expressed as lac+fu/ml ~'itre lacZ) 25 results from the combined transfer efficiencies of gene and internal gene expression. The compa-reason with the title expressed in the form G418r~u/ml ~NeoR titer~ gives self-inactivation estimate~
OVA-D seems to give a high titer in both assays, confirming that avian leukosis viruses .. such as murine leukemia viruses, can harbor the internal promoter. To our surprise, the efficiency WO 91/11190 PCI`/FWt/OOn62 ~34 of gene transfer is dramatically reduced . with OVA-2Z (100 lac+fu/ml) and the vector does not present no signs of self-inactivation (100G418rfu/ml~. -The deletion of the 3' CAAT box in OV-Zdel i results in an improvement in the efficiency of the trans-gene fer and the increase in the ratio between the title lacZ and NeoR. To improve both the ef-efficiency of gene transfer and self-inactivation, 20 clones of transfected helper cells are isolated 10 with OVA-Zdel and both the expression nlslacZ and NeoR. 10 of these clones do not give no virus (< 20 lac+ or G418rfuiml). The results obtained with the four clones which give the title highest lacZ are reported in Table II. the 15 ratio between the two viral titers (i.e.
the effectiveness of self-inactivation) varies widely from clone to clone and can exceed 60. Therefore, OVA-Zdel can be considered as a vector of effective self-inactivating retrovirus.
Self-inactivation is still confirmed in a failed virus rescue experiment to give a recombinant virus after superinfection by RAV(l) ~<20 lac+fu/ml).
The ~ equences in 3 'influence the eYpression 25 internal gene transient To acquire information 9supplé-on the factors that determine the production tion of virus in the helper cells, we transfect Isold or QT6 cells with OVA plasmids 30 and assayed 48 hours later transient expression tory of the SV40-nlslacZ gene by ln situ staining (Table III). Under these conditions, OVA-D and OVA-20 7'i0 67 ZZ give similar results. U3 deletion in OVA-Zdel alters transient expression of SV40-nlslacZ. These results support the idea than the modest viral titer obtained with OVA-Zdel is linked to a defect in the expression of SV-nlslacZ
; and that the deleted part of the RAV(0) U3 element carries an important signal for gene expression viral. On the other hand, the low viral titer obtained with OVA-ZZ is not correlated with the express-transient nlslacZ, indicating that a post-transcriptional step of viral production is also upset when using the sequence ~AV(O) 3l. Transcripts promoted by SV40 are less abundant than promoted transcripts by L~R in 5' in the cells transfected in a way stable but the constant level of both RNA in these cells is not clearly dependent 3' non-coding sequences.
Although this observation is an argument against a possible engagement of the 3' sequences in ` the viral transcription, it can be biased by the fact that only cells producing enough ment of viral RNA to resist G418 selection are considered in the latter case.
V~ri~t;on clo~le d~ns the espre~sion of the gene SV~0-nlslacZ in the cells infoct~o-On coloro dos clones d- collulo5 infected d ~ itant to G418 ~ clones G418R) to d ~ finish the acti-vity of ~ -galactosidase. Although this analysis is carried out on a limited number of clones, it aboutlt ~ several interesting observations:

: . ' . :, ' O91/11190 ~ PCT/FR91/00062 q~~36 - Some G418R clones only contain no cells lac+ cell. This could result of a mutation occurring in the nlslacZ gene.
However, since G418R/lac helper cells can produce recombinant lac+ virus (data not represented) and cor, - ~ e a low level of expression enough NeoR gene to provide resistance to G418 (see above and R29), it is more likely that G418R/lac- cells actually express 10 nlslacZ, but below the necessary threshold level to stain positively in the in situ assay.
A position effect may explain the reason for the phenotype is almost constant in a single clone G418R. It is therefore likely that a 15 in situ lacZ staining leads to an underestimation the effective efficiency of gene transfer.
- The relative proportion of G418R/lac- clones and G418R/lac+ depends on the vector that is used for infect cells. As expected from 20 transient dosages, this ratio is similar for OVA-D and OVA-ZZ and is higher than that which obtained with OVA-Zdel.
RNA analysis reveals that the non-coding sequences in 3' influence the effectiveness of the envelopment of 25 the AR~.
A low viral titer may result either reduced production of viral particles both the vector DNA by the9 cell8 helper, or a low level of viral gene expression in 30 infected cells. To identify what is the limiting step in the present situation, we 37 2075~fi7 analyzes retrovirus RNAs in transfected helper cells stable by OVA-D, OVA-Z7 and OVA-Zdel (clone A) and in particles viruses they produce. Slit blot analysis shows that the viral titer correlates with the amount of retrovirus RNA
5 enveloped in the virus particles but not with the constant level of vector RNA present in the cells. Quantitative analysis reveals that the OVA-ZZ and OVA-Zdel transcripts initiated at the 5' LTR are sub-represented in virus particles. The envelope of RNA therefore seems be a limiting step in virus production for OVA-7' e~
10 OVA-Zdel.
It is shown here that 3' non-coding sequences influence the ability to inactivate retrovirus vectors but also ~
the efficiency of retroviral-mediated gene transfer into cells avians. As tested U3 deletions affect titer 15 virus, one is inclined to suppose that the LTR sequences play a role in this phenomenon.
However OVA-Zdel with a 5' LTR of RSV allows us to produce 100 lacZfu/ml while fully inactivating the LTR promoter.

20 Table 1: Virus titer Helper cells are transfected and subjected to selection by G418. The cell surna ~ eant is then titrated on QT6 cells and ~; CEF for both NeoR and lacZ expression. We take the average of the tlter viral from at least 3 different transfection experiments.
Table ll: Clonal variation in viral titer obtained with OVA-Zdel 20 clones of helper cells transfected in a manner stable with OVA-Zdel for NeoR gene expression as. Iac'. A, B, C
and D are the four clones which give the highest lacZ titer.

W091/11190 PcT/FR9l/ooo62 20~ rl 38 Table III: Transient E ~ pressure of SVnlslacZ
in the OVA vectors.
Transient expression: we transfect QT6 cells by phosphate precipitation of calcium or with polybrene ~see materials and pro-ceded~. The cells are fixed after 48 hours and stained to determine the activity of ~-galactosidase. When more than 500 cells are positive for the transitive expression of nlslacZ
on a single box, we extrapolate the total number of lac+ cells from an exhaustive count performed on a limited area. The last track indicates average ratios for OVA-D.

Table I
¦Construct ¦DA(l) IOVA-D¦ OVA ZZ ¦OVA Zdel ~aL~L 10,000 100~ 250 NeoR fu/ml ¦ ~ 10010,000 100 100 Table II
Construct~ :~ OVA-ZZ OVA-Zdel Ca Phosphate _6000 6000 _ Polybrene182 154 = 26 Table III
~ ~ b ~ 1200 1400 G4 1 8rfu/ml _ 2000 250 -- ~ 20 200 Ratio 7.3 6.4 60 _ 2.2 39 207~067 BIBLIOGRAPHY

1. Affolter, M., J. Cote, J. Renaud and A. Ruiz-Carillo. 1987. Regulation of histone and BA-globin gene expression during differentiation of chicken erythroid cells. soft. Cell. Biol. 7, 3663-3672.
2. Appels, R., L. ~olund and RN Rin~ertz. 1974. Biochemical analysis of reactivated chick erythrocyte nuclei isolated from chick HeLa heterokaryons. J.Mol. Biol. 85:182-190.
3. Benchaibi, M., F. Mallet, P; Thoraval, P. Savatier, JH Xiao, G. Verdier, J. Samarut and JV Nigon. 1989. Avian retroviral vectors derived from avian defective leukemia virus: role of the translational context of the inserted gene on efficiency of the vectors. Virolo~y 169~ 26.
4. Bergman, MG, E. Wawra and M. Win~e. 1988. Chicken histone H5 inhibits transcription and replication when introduced into proliferating cells by microinjection. J. Cell Sci. 91:201-209.
5. Briand, G., D. Kmiecik, P. Sautiere, D. Wouters, O. Borie-Loy, G. Biserte, A. Mazen and M. Champagne. 1980. Chicken erythrocyte histone H5.
FEBS Lett. 112:147-6. Cullen, BR, K. Raymond and G. Ju. 1985. Functional analysis of the transcription control region located within the avian retroviral long terminal repeat. Mol. Cell. Biol. 5:438-447.
7. Emerman, M. and HM Temin. 1984. Genes with promoters in retroviruses vectors can be independently suppressed by an epigenetic mechanism.
Cell 39:4S9-467.
8. Favaloro, J., R. Treisman and R. Kamen. 1980. Transcription maps of polyoma virus specific RNA:analysis by two dimensional nuclease Sl gel mapping. enzyme methods. 6S:718-749.
9. Gandrillon, O., P. Jurdic, M. Benchaibi, JH Xiao, J. Ghysdael and J.
Samurut. 1987. Expression of the v-erb A oncogene in chicken embryo fibroblasts stimulate their proliferation in vitro and enhance tumor growth in vivo. Cell 49:687-697.

WO 91/11190 PCI'/FR91/00062 6~ -10. Garcia, M., R. Welliner, A. Vessaz and H. Diggelman. 1986. ~ new site of integration for mouse mammary tumor virus proviral DNA common ~o ~alb/c (C3H) mammary and Kidney adenocarcinomas. 1:127-134.
He. Gritz, L., and J. Davies. 1983. Plasmid-encoded hygromycin 3 resistance: the sequence of hydromycin B phosphotransferase gene and its expression in Escherichia coli and Saccharomyces cerevisiae. Embarrassed 25: 179-188.
12. Hwan~, LS and E. Gilboa. 1984. Expression of genes introduced into cells by retroviral infection is more efficient than tha~ of genes introduced into cells by DNA transfection. J. Virol. 50:417-424.
13. Karin, M. and Rl Richards. 1982. Human metallothionein genes-primary structure of the metallothionein-ll gene and a related processed gene. Nature (London) 299:797-802.
14. Kawai, S. and M. Nishizawa. 1984. New procedure for DNA transfection with polycation and dimethyl sulfoxide. Mol. Cell. ~iol. 4:1172-1174.
IS. Laemli, VK 1970. cleavage of structural protein during the assembly of the head of bacteriophage T4. Nature 2S6:680-685.
16. Laimins, LA, P. Tsichlis and G. Khoury. 1984. Multiple enhancer domains in the 3' terminus of the prague strain of Rous sarcoma virus.
Nucleic Acids Res. 12:6427-6442.
17. Moscovici, D., MG Moscovici, H. Jimenez, MMC Lai, MJ Hayman and PK Vogt. 1977. Continuous tissue culture cell lines derived from ; chemically induced tumors of japanese quail. Cell 11:95-103.
18. Mura, CV and BD Stollar. 1981. Fluorescence-activated sorting of isolated nuclei. Exp. Cell. Res. 135:31-37.
19. Mura, CV and BD Stollar. 1981. Serological detection of homologies of H10 with H5 and H1 histones. J. E~iol. Chem. 256s9767-9769.
20. Mura, CV and BD Stollar. 1984. Interaction of H1 and H5 histones with polynucleotides of E- and Z- conformation 3iochem. 23:6141-6152.

WO 91/11190 PCI/FR9l/00062 41 207~S0~7 -21. Mura. CV, G. Setterfield and JM Neelin. 1982. Immunofluorescent localization of Iysine-rich histones in isolated nuclei from adult and embryonic chicken erythrocytes. Can. J. Biochem. 60:215-223.
22. Norton, P. and JM Coffin. 1987. Characterization of Rous Sarcoma virus sequences essential for viral gene expression. J. Virol. 61:1171-1179 23. Richards, Rl, A. Heguy and M. Karin. 1984. Structural and functional analysis of the human metallothionein-lA gene: differential induction by metal ions and glucocorticoids. Cell 37:263-272.
24. Savatier, P., C. Agnis, P. Thoraval, D. Poncet, M. Belakebi, F. Mallet, C. Legras, FL Cosset, JL Thomas, Y. Chebloune, C. Faure, G. Verdier, J. Samarut and V. Nigon. 1989. Generation of a helper cell line for packaging avian leukosis virus-based vectors. J. Virol. 63:513-522.
25. Southern, PJ and P. Ber~. 1982. Transformation of mammalian cells to antibiotic resistance with a bacterial gene under control of the ; 15 SV40 early region promoter. J.Mol. Appl. Broom. 1:327-341.
26. Sung, MT and EF Freedlender. 1978. Sites of in vivo phosphorylation of histone H 5. Biochem. 12: 1864-1869.
27. Sung, MT, J. Harford, J. Bundman and G. Vidalakas. 1977. Metabolism of histones in avian erythroid cells. Biochem. 16:279-285.
28. Varmus, H. and R. Swanstrom. 1982. Replication of retroviruses. In (R.
Weiss, N. Teich, H. Varmus and J. Coffin, Eds.), RNA Tumor Viruses Molecular biology of Tumor Viruses" pp. 369-512. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY
29. Wu, RS, HT Panusz, CL Hatch and WM Gonner. 1986. histones and their modifications. CRC Crit. Rev. I~iochem. 20:201-210.

WO 91/11190 PCI`/FR91/00062 rl 4 2 BlaLlOGRAPHlCAL REFERENCES FOR EXAMPLE 2 -R3: Galileo. DS. GE Gray, GC Cwens, J. Maprs and JA Sanes. 1990. Neurons ana glia anse ~rom a common progenitor in chicken optic tectum: Demonstration ~nlh two retrovirus3s and cell-type specific ar~odies. proc. Natl. academic n.d. USA. 87:458-462.
- R4: Danix. JL. 1986. Circuiarizalion ot reboviral genomic RN~ and the cor~ol o~ RNA
translation, packaging anrJ reverse lrarWdptiort. Biochemistry 68: g41~949.
- A7: Darlix. JL. 1986. Cor~o~ ot Rous w~rna vinus transbffon and packa9in9 by the 5 and 3 untranslated sequences. J.Mol. Biol. 189: 421~4.
R 10: Flanagan. JR. AM Krieg, EE Max and AS Kahn. 1989. Negative control region al tne 5 end ot rnunne leukernia vin~s bng terminal repeats. Mol. C~U. Biol. 9: 739-746 A 11: Murphy, I. and SP Gott. 1989. Constnuction and anal~rsis ol deletion mutations in me U5 reglon ot Moloney rnunne leukernia virus: Effects on RNA parking ana reverse ~ranscnptlon.
R 13: Ludw, P A., JM Bishop. HE Varrnus and MR Capecchi. 1983. Location and lunctlon ot retroviral and SV40 se4uences that enhance bioehernicat transtormation atter mlcrolinjection ot DNA. Cell 33: 705-716.
R14: Herrnan, SA, and JM Coffin. 1986. Dillerer~trar~ption trom the bng terminal repeats ot integrated avian teukosis vinus DNA. J. Virol. 60: 497-505.
- R16: Emernan. M. and HM Temin. 1986. Ouanlltaive anaiyds ot gene suppression in integrated retrovirus vectors. Mol. This U. Biol. 6: 792-800.
- A17: Swain. A. anrJ JM cOnin. 1989. Po~n at cor~ct sites In genome RNA is not requlred tor retrovirus replication of genome encp~tion. J. Virol. 63: 3301-3306.
A18: No. PA ~nct JM CoHln. 1987. Characterization of Rous sarcorna vin~s sequences essential to viral gene expression. J. Virol. 61: 1171-1179.
R19: Cullen, BR, AM Skatt<a and G. Ju. 1983. Endogenous avian retroviruses contain 2 5 oe~iclent promoter and leader sequences. proc. NaU. Acati. SC'. USA 80: 2946-2950 R23: 0Ougheny, JP and H. M Ternin 1987. A prornoterieU retrovlral vQctor indlcat~ls ~hat tnere are seoiuences In U3 requlred lor 3 RNA proCeuung. proc. Nation. Acad. Scl. USA 84 1 197-1201.
R24: Bonnerot. CD Roc;tncoun, P. Brlant. G. Grimber and JF Nicolas. 1987. Year galaaosidase hybrid proteln targe~ed to nucld as a mari'er tor developmental stuaies Proc Natl. Acad. Science. USA 84: 6795-6799.
R29: ~enchal~i. M., F. Mauet, P Thoraval, P. Savaber. JH Xiao, ~;. Verdier, J. Samarut ana V
M. Niqon 1989. Avian retro~nral veaors derived ~rom avian de~eative leukemia virus: Role ot the ranslatlonal context ot the Insened gene on etHdency o~ the vectors. Virobgy 169: 15-26 R30: Cone. RC. Weber-aenarous A., Baorta. D and RC MulUgan .19a7 Regulated exPresslon ot a comPlele human 13-glo~in gene ancoded by a transmissible retrovirus vectors Mol Cell. Biol. 7:B87-897 207~067 - R31: Emerman. M. ana HM Temln. (1984). Genes With promoters in relrovinus vectors can De ndepenaently suPpressed cy an eplgene~lc mecnanism. Cell 39: 459-467.
A32: Sorge. Y.A. W. Ricci and S. H Hughes. 1983. Cis aing RNA pacl(aging lo~s ln the 115- nucleonde direct repeal ot Rous sarcoma vinus. J. Virol. 48: 667-675.
5 R33: RNA lumor vinuses.19B5 R. Weiss, N. Teich, H. Vamnus and J. Coffin eas. Cold Spring HarOor s~oralory. Cold Spring Harbor. NY

- R34: Yu. SF, T. Von RDden, P. Karrlott, C. Ga~r, M. S~r~t, U. RDlher, WF Anderson, EF Wagner formerly E. Giiboa.1986. Sen-ina~iivating retrovirai vectots designed tortrans~ar o~
whole genes into mammalian cells. Prr c. Natl. Acad. Scl. USA 83: 3194-3193.

- R35: Payne. GS. SA Cou,r~neidge, LB Cnttenden, AM Fadlv, JM Bishop and HM
Varnus. Analysis ot a~vian /eukosis DNA and flNA in b~sal turnors: Viral gene expression is no~
rer~uirea for the main~enance of the tumor state. C~D23, 311 19~1 R~6: Mosoovid, C.. MC Mosowid, H. Jirner~z, MMC Lai, MJ Hayman and PK Vog~.
1976. Continuous tissue cu(ture cell lines derived from trom chemically inauced tumors ot japanes l 5 quail. Cell 11: 95-103.
R~9: Maniatis, T., EF F~itsch and JM Samarook.1982. Mobwhr clon~ng: a laboratory rnanual. Cold spnng harOor labotatory, Colci Sprhg Harbor. NY
R40: Dusty MiUer, A., DR Trauber and C. 8u~irnore: 1986. Facsors imohed in produaion ot helper vlnus-tree relrovir ts vectors. Sum. This U. Mol. Broom. 12: 175-1B3.
A41: Chaney. WG. Dr Howard. JW Poltard, S. SaUustio and P. Stan~ey.1986. High-2~ ~requency IranSteCtiOn ot CHO cells using polyOrene. Sum. This U. Mol. Genet.12: 237-244.
A43: Yeah. JK, JC Moores, DJ JoUy, JA wOm and JG Respess.1987. Embarrassed expression trom IranscJi~ionaUy disa01ed retroviral vectors. proc. Natl. Acad. Science. USA 84:
5197-5201.
- R44: Cossel, FL.. C. Legras, Y. Chebbune. P. Savatier, P. Thoraval, JL Thomas J.
Samanut. VM Nigon and G. Verdier.1990. A new avian ktukosis vinus basea Dac~taging cell 2 5 line Ihat uses two separate transcornplemerlting heiper genomes. J. Virol. 64: 1070-1078.
R45: Bizub D., RA Katz and AM SSkaika 1984 Nucieollde sfJquence ot non coding reglons m Rous sssoaat~d vinus-2: compan50n dl~lineats conufv~i reglons Irnponant In replicaoon and oncogenesis.
A46: Hughes. SH. in Viral vectors.1988. Y. Giuzman and SH Hughes and others. Cold Spring Hardor LaDoratory. Cold Spring Hardor. NY
R47: IwaKaza K.ana HM Temln.1990. The U3 region is not necessary tor 3 end tormanon ot spleen necrosls vlrus RNA. J. Virol. 64: 6329-6334.
R48: Denome RM formerly CN Colle. 1988. Panem ot polyadenyiation site seleaion in gene constnucts contalning muniple ~oolyaaenylation signal. Mol. Call. Biol. 8:4849-4a39.
A49: Katz RA. BR Culian. R. Malavarca and A. h~. Skaika.1986. Role ol the a~an retroYlnJs mRNA leaaemn expresslon: Evidence to novel translational control. Mol. Cell. Biol. 6:372-379.

3&~

R50: Sou~hem PJ formerly P. Berg. 1982- Transtonmation ot mammalian cenns to antibiotic reslstance wlth a oactenal gene unaer control o~ the SV40 eany promoter reglon.
J.Mol. Appl. Broom. l: 327 341.
A51: Sealy. L.. ML Privalsky. G. Moscovlci. C: Moscovia anct JM Bishop. 1983. Sile- -speatic mutagenesls ot avian erythroblastosls virus: erbB is requlred tor oncogeniaty. Virology 130: 155-178.
RS2: Gormen CM, GT Merlino, MC Willingham, I. Past~an and B.~.
Howard. 19e2. The Fous s = virus lcng te~m~n21 repeat is a strcng promDter when introduced in a variety of eukariotic cells by DNA-mediated t~ansfection. proc. ~latl. Acad. Bcc. USA 7~: 6777-6781.
.RS3: Sanes JR, JLR Rubenstein and JF Nioolas. 1986. Use o~ a recombinant retrovirus to study post-implantation cell lineage in mDuse embryos. EM30 J.5: 3133-3142.

.

.~

Claims (18)

1. Self-inactivating viral vector for integration and expression of at least one heterologous gene in avian cells, consisting of from the proviral genome from avian retroviruses, mainly AEV and/or related retroviruses in particular of the RAV type, and comprising between two avian LTR sequences the said heterologous gene(s) which find(s) under the control of a heterologous internal promoter, characterized in that deletions or mutations in the 3' LTR have been made so as to inactivate the viral promoters present in the LTRs, the 3' LTR from retroviruses RAV-0, RAV-1, RAV-2. 2. Vector according to claim I, characterized in that the promoter or enhancer sequence of the 3' LTR has been deleted complete or partial. 3. Vector according to claim 2 or 3, characterized in that the LTR sequence has undergone deletions in the U3 region. 4. Vector according to one of the preceding claims, characterizes in Fe that the CAAT and TATA boxes have been deleted. 5. Vector according to one of the preceding claims.
characterized in that it is constructed essentially from the genome of the RAV-2 retrovirus whose 3' LTR comprises the U3 sequence of the LTR of RAV-2 in which the CAAT and TATA boxes have been deleted and the RAV-1 LTR R and U5 sequences. 6. Vector according to one of claims 1 to 3, characterized in that the 3' LTR is derived from a RAV-0 virus whose LTRs are naturally very little active and has undergone deletions or point mutations. 7. Vector according to claim 6, characterized in that 12 RAV-0 3' LTR has been deleted in the CAAT promoter sequence. 8. Vector according to one of claims 1 to 7, characterized in that the 5' LTR consists of the 5' LTR of the RSV virus. 9. Vector according to one of the preceding claims, characterized in that it comprises two heterologous genes, namely a gene useful place under the control of a heterologous internal promoter and a gene of selection placed under the control of a heterologous internal promoter or of the 5' LTR promoter. 10. Vector according to the preceding claim, characterized in that said heterologous genes replace the v-erbA oncogenes and v-erbB. 11. Foreign gene integration and expression vector characterized in that it is of the plasmid type and comprises a sequence composite of reverse transcribed DNA of an RNA viral vector according to one of claims 1 to 10. 12. Method for preparing a vector according to one of previous claims. characterized in that one or more deletion(s) or mutation(s) in the avian 3' LTR so as to inactivate the viral promoters present in said LTRs, and said genes are inserted heterologous under the control of an internal heterologous promoter or the 5' LTR promoter. 13. Process for modifying cells characterized in that that an infection or a transfection of said cells is carried out with retroviral vectors according to one of claims 1 to 10. 14. Process according to claim 13, characterized in that performs a co-infection or co-transfection with a helper virus which contains the genes ensuring the replication of said retroviral vector. 15. Cell modified by implementing a method according to one of claims 13 and 14. 16. Cell according to claim 15, characterized in that it are avian cells. 17. Cell according to claim 16 characterized in that it These are chicken or quail embryonic fibroblasts. 18. Process for the preparation of at least one specific protein by cell culture according to one of Claims 15 to 17, the gene being a foreign gene encoding said protein the protein being recovered after culture.