WO1997046573A1

WO1997046573A1 - Method of inhibiting antigen processing, chimeric proteins reflecting this method, and methods of expressing those proteins

Info

Publication number: WO1997046573A1
Application number: PCT/US1997/009294
Authority: WO
Inventors: Michael Kurilla; Maria Masucci
Original assignee: The University Of Virginia Patent Foundation
Priority date: 1996-06-07
Filing date: 1997-06-06
Publication date: 1997-12-11
Also published as: AU3291697A

Abstract

Cytopathic T lymphocyte immune responses that are major histocompatibility complex Class-I mediated, in response to the expression in a cell of foreign protein, are inhibited by the presence, in the foreign protein, of a Gly-Ala repeat domain. A minimum of about 30 amino acids in the repeat domain is required to inhibit CTL immune responses. DNA segments encoding the repeat sequence are preferably inserted downstream of the sequence encoding the CTL epitope of the structural protein. Host cells, transfected with the DNA, on expression of the resultant chimeric protein, do not trigger CTL attack.

Description

Title of the Invention;

METHOD OF INHIBITING ANTIGEN PROCESSING, CHIMERIC PROTEINS REFLECTING THIS METHOD, AND METHODS OF EXPRESSING THOSE PROTEINS

Field of the Invention:

This invention pertains to chimeric nuclear proteins, and methods of expression of the same, which proteins do not trigger a conventional major histocompatibility complex reaction where inserted into a cell of a patient. Specifically, proteins, particularly nuclear proteins, provided with a Gly-Ala repeat domain appear to suppress recognition by cytotoxic T lymphocytes (CTL) and inhibit an induced immune response, interfering with antigen presentation.

Background of the Invention:

It has long been recognized that the injection of foreign proteins into mammalian systems, including human systems, presents a desirable therapeutic treatment method, but is frustrated by the immune response induced. Certain therapies, including gene therapy, as well as more conventional vaccine design, are limited by the CTL-mediated immune response to most exogenous proteins that are candidates for therapeutic measures by insertion into (infection of) the hosts cells. The use of a wide variety of potential therapeutic agents would be made possible, if a method could be found to suppress the CTL-mediated immune response induced by presentation of a foreign (that is, not encoded by the hosts original DNA) protein. Whether the protein is a recognized antigen, or merely foreign and therefore immunogenic, the response is generally obtained. The Epstein- Barr virus is a well-studied herpes virus, which transforms cells through expression of episomal DNA in the host cell. The same avenue is the basic modality employed in gene therapy, where foreign genetic material is inserted into the cells of the target patient, to achieve expression of that genetic material, producing the therapeutic protein(s) .

The immune response discussed above includes attack by cytotoxic T lymphocytes (CTL) mediated by the major histocompatibility complex (MHO . In this type of response, cytotoxic T cells are drawn to, and attack, the foreign material by recognition of the MHC proteins present in response to the inserted foreign or exogenous material. Thus, proteins, and nuclear proteins in particular, are susceptible to MHC Class I- mediated CTL immune response, given the cell surface glycoprotein pattern associated therewith. Class II-mediated immune response is also associated with this phenomenon, but not necessarily directly targeted to nuclear proteins.

The EBV-encoded nuclear antigen EBNA1, is a viral protein regularly detected in all malignancies associated with EBV. See, e.g., Masucci et al . , Trends in Microbiology 2, 125-130 (1994) . No MHC-I mediated CTL response has been demonstrated in connection with this nuclear protein. Rickinson et al .. A New Look at Tumor Immunology, 53-80 (Cold Spring Harbor Laboratory Press, 1992) . If the method by which the MHC-I mediated CTL response to this nuclear protein is inhibited could be adopted for therapeutic proteins, a wide class of proteins could be better utilized for the therapies described.

Accordingly, it remains an object of those of skill in the art to develop a method for preventing or inhibiting the presentation of antigens in MHC Class-I CTL response processing.

SUMMARY OF THE INVENTION

The above objects, and other objects demonstrated more clearly below, are achieved by expression of a Gly-Ala repeat domain in the exogenous protein whose antigen processing is sought to be inhibited, by in-frame insertion of the Gly-Ala repeat domain DNA, in the sequence of the protein itself. By "Gly-Ala repeat domain" is intended herein to mean a series of three tripeptides, wherein each tripeptide consists of Gly and Ala residues. Three triplets define a nine amino acid sequence, which is then repeated. Each triplet contains one Gly, one Ala and a third ammo acid, either Ala or Gly, in any order. The sequence of 3 triplets is repeated in the domain. A total of about 35-50 amino acids appear effective suppressing MHC-I mediated CTL response to the proteins, in "infected cells".

Below about 35 amino acids, the domain may be so small as to not sufficiently inhibit toxicity. While the domain may be incorporated m the chimeric protein at any point, it is preferably incorporated downstream of the epitope triggering the

CTL response . The presence of the repeat appears to inhibit presentation/recognition of the epitope, and a downstream location is demonstrated to achieve this effectively. While applicants do not wish to be bound by this theory, it is possible that the Gly-Ala repeat present occupies or "redirects" the mechanism responsible for MHC mediated CTL response to such a degree the epitope-tπggered response does not occur possibly due to failure to express foreign protein-induced MHC glycoprotein fragments on the cell surface. By insertion "in frame", t s intended that the DNA sequence inserted into the sequence encoding the antigen must be inserted so as to encode the repeat itself. That is, the inhibition of the CTL response is a protein-associated phenomenon, not a DNA-associated phenomenon.

Larger Gly-Ala repeats may be used, if desired. Repeats encoded by DNA segments of up to 700 base pairs have been demonstrated to be effective. Optimal repeat size can be determined by empirical methods, but it appears that 100-130 amino acid domains may be effective in MHC Class-I and perhaps Class-II mediated CTL immune response.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic outline of chimaera prepared according to the invention, including a (full-length) EBNA1 chimaera. The position of insertion of the EBNA4 416-424 epitope (E4) and the EBNA1 internal repeat region (IR) are indicated by arrows pointing to the amino-acid number in the B95.8 sequence. The Gly-Ala repeats are indicated in solid marking, while the nuclear localization signal is dark-shaded, and DNA binding and dimerization domains are indicated by cross-hatching.

DETAILED DESCRIPTION OF THE INVENTION:

Applicants' invention resides in the discovery that the "infection" by an antigenic protein, of a cell, which would otherwise induce MHC-mediated CTL immune response, can be expressed while inhibiting that response if the expression includes a Gly-Ala repeat domain within the protein. The antigenic presentation of the protein, incorporating the Gly-Ala repeats, is somehow suppressed, such that CTL response is not triggered. The Gly-Ala repeat domain should be on the order of 30 amino acids-150 amino acids. Optimally, it appears that an insertion domain of 50-120 amino acids provides significant inhibition of CTL response.

Much of the inventors' initial work done in this field, including that discussed below, is described in Levitskava et al, Nature, 375, 685-688 (June, 1995) . As the information contained therein is not essential to an enablement, the disclosure therein is incorporated herein-by-reference. Applicants note that the activities and thoughts reflected in the application constitute the work of the inventors, although other authors are named recognizing their non-inventive contribution to the subject matter disclosed.

Expression of the chimeric nuclear proteins that are the subject of this invention is achieved by conventional recombinant technology, inserting the DNA segment responsible for the Gly-Ala repeat domain into the DNA encoding the antigenic peptide itself. Preferably, the Gly-Ala repeat segment is inserted downstream of that segment of DNA encoding the CTL epitopic portion of the protein, where such is known. Restriction, insertion, ligation, plasmid preparation and cell insertion or "infection" are all conventional technologies familiar to those of ordinary skill in the art, and are not detailed herein, saved for their expression in terms of the examples set forth below. Applicants' discovery was keyed by the recognition that the nuclear antigen EBNAl encoded by the EBV is expressed in latent EBV-infected B lymphocytes that persist for life in healthy virus characters, is the only viral protein regularly detected in all malignancies associated with EBV, and not bee demonstrated to be subject to a MHC-mediated CTL response. Mapping of the antigenic protein demonstrated the presence of a Gly-Ala repeat domain, which interferes with antigenic processing and image Class - I, and perhaps Class - II presentation. By fixing the epitope, and inserting the DNA encoding the Gly-Ala repeat downstream of that epitope, CTL recognition of otherwise CTL-recognized epitopes was inhibited.

This discovery has been demonstrated to be suitable for use in the suppression or inhibition of MHC Class-I mediated CTL response in other nuclear proteins. While the exact mechanism of this "viral escape" from CTL response is not clear, it appears that it may include a mechanism whereby a protein inducing the CTL response is redirected or sequestered, but is not presented in an antigenic fashion. In any event, it should be clear that there is no absolute number of repeat sequences essential to achieve inhibition. In some antigens, and some hosts, a lower number may be appropriate, while in other antigens and hosts, a higher number may be required. Thus, the value set forth above, including the lower level of about 30 and the higher value of about 120, are proximal in nature. Reducing the numbers below the minimums recited may have some, but less, inhibitory characteristic.

This invention is disclosed, below, in the context of the EBNAl nuclear antigen. The invention is not so limited, and the suppression of CTL response to the insertion of any protein, particularly nuclear proteins, in accordance with the invention as disclosed and claimed, is contemplated by the Applicants.

EXAMPLE

EBNAl is a phosphoprotein composed of unique amino and carboxy-terminal domains (amino acids 1-89 and 327-641, respectively, in the prototype B95.8 EBV strain) joined by a repetitive sequence of Arg-Gly-containing motifs, surrounding an internal Gly-Ala repeat. While Gly-Ala repeats of different le-.-gths are present in all EBV isolates and represent the major target of EBNA-specific antibody responses, the specific function of these repeats has not been previously described.

EBV induces CTL responses. Caucasian leukocyte antigen- (HLA)All-positive individuals are often dominated by All- restricted reactivities to peptide epitopes corresponding to residues 399-408 and 416-424 of the EBNA4 protein. Dittmar et al, Proc. Natl. Acad. Sci. USA, 81,4652-4656 (1984) . Accordingly, recombinant vaccinia viruses expressing chimeric genes that contain the EBNA4 416-424 epitope inserted in-frame within the attached EBNAl sequence, or within EBNAl deletion mutants that did not contain the Gly-Ala repeats were prepared. Expression of the chimeric proteins in human fibroblast was confirmed by immunofluorescence and immunoblotting. EBNAl deletion (the native protein with the Gly-Ala repeat domain excised) containing the EBNA4 416-424 epitope inserted at the His 39, Pro 446 or Lys 520 positions, relative to the B95.8 sequence sensitized HLA All-positive fibroblasts to lysis by EBV-specific CTL. The level of killing was in each case comparable to that observed after infection with a vaccinia recombinant expressing EBNA4. In contrast, fibroblasts expressing a chimeric full-sized EBNAl with the EBNA4 416-424 epitope inserted at His 39 (E1N-E4) were not recognized by the CTL. This was true even after infection for up to 24 hours, although the recombinant protein was detected in most infected cells starting from 12 hours. Fibroblasts infected with the EBNA4 epitope inserted at the His 39 positions (El Δgan-E4) were lysed after six hours, and remain sensitive to lysis until 24 hours when the experiment had to be terminated because of extensive cytopathic effects. The differential recognition of cells infected with the chimeric protein including the CTL-recognized epitope, and the CTL- recognized epitope together with the Gly-Ala repeat sequence, indicates that the presence of the Gly-Ala repeat sequence is the cause of the failure of the Enl-E4 chimera desensitized target cells to undergo lysis.

To confirm this possibility, the CTL sensitivity of cells infected with vaccinia recombinants expressing EBNA4, or chimeric EBNA4 protein with the EBNA 1 Gly-Ala repeats region inserted in- frame (that is, for expression of Gly-Ala) between Trp 624 and Pro 625, that is, downstream of the 399-408 and 416-424 epitopes, was compared. The recombinant proteins were expressed with substantially identical kinetics and at similar levels. HLA 11- positive fibroblasts infected with EBNA4 were sensitive to CTL clones specific for the 399-408 or 416-424 epitopes with maximum lysis after infection for six hours. In contrast, cells infected with E4IR (that is, including the repeat) were lysed only weakly, even after infection six-fold higher M.O.I, or for prolonged periods of time. Infection of a spontaneous lymphoblastoid carrying a Chinese EBV isolate with mutations abrogating recognition of the endogenous EBNA4 399-408 and 416-424 epitopes gave essentially similar results. Thus, the EBNAl G.y-Ala repeats seem to inhibit processing, and MHC Class-I presentation, independently of the target cell lineage.

Further experiments demonstrated that over-expression of EBNAl itself did not, alone, influence antigen processing. Over- expression of EBNAl or EBNA3 did not prevent recognition of the infected fibroblast. Accordingly, the presence of the repeats within the target protein is responsible for either preventing processing, or sequestering processing products, such that they are inaccessible to MHC Class-I presentation.

DESCRIPTION OF SPECIFIC METHODS A Schematic outline of the full-length EBNAl chimaera (E1N-E4) , the Gly-Ala-deleted EBNAl chimeras (E1ΔGA) and the Gly- Ala-containing EBNA4 chimaera (E41R) is set forth in Figure 1. The position of insertion of the EBNA4 416-424 epitope (E4) and the EBNAl internal repeat region (IR) are indicated by arrows pointing to the amino-acid number in the B95.8 sequence. The EBNAl Gly-Ala repeats (solid) ; Arg-Gly repeats (light shaded) ; nuclear localization signal (dark shaded) ; DNA binding and dimerization domains (cross-hatched) are indicated. Quantification of expression of the recombinant proteins was detected by (ACIF) . E1N-E4 and E1ΔGAN-E4 reached a plateau of expression between 12 to 18 h whereas the expression of EBNA4 and E4IR reached a maximum at 12 h and decreased thereafter. The negative orientations E1N-E4 and E1ΔGAN-E4 were expressed with slower kinetics and at lower levels compared to the positive orientation E1ΔGAN-E4 whereas the orientation of insertion did not affect the expression of E4IR. It should be noted that the expression of E1N-E4 and E4IR is probably overestimated because of the presence of Gly-Ala-specific antibodies in the human serum.

Vaccinia recombinants carrying the coding sequences of EBNAl and EBNA4 (also known as EBNA-3B) from B95.8 were used. Chimeric

(5EQ NO : I) proteins containing the EBNA4 416-424 epitope (IVTDFSVIK)^ ithin

EBNAl (El) , or an EBNAl deletion mutant that lacks the Gly-Ala repeat region (E1ΔGA) were produced by inserting an oligonucleotide corresponding to the E4 epitome in the Ncol ,

PffMl or Bsu361 site of the EBNAl coding sequence (genomic positions: 108,067, 109,291 and 109,510, respectively) . The oligonucleotide pairs E1N-E4F/E1N-E4R (5' -CATGCCATAGTAACTGACTTT-

AGTGTAATCAAG-3' /5' -CATGCTTGATTACACTTAAGTCAGTTACTAT-3 ' ) ; E1P-

E4F/E1P-E4R (5' - CGATCGTAACTGACTTTAGTGTAATCAGG- 3 ' /5 ' -

CCTTGATTACACTAAAGTCAGTTACGATCGTGC-3' ) and Ξ1B-E4F/E1B-E4R (5'-

TAACGATCGTAACTGACTTTAGTGTAATCAAGG - 3 ' / 5 ' -

TTACCTTGATTACACTAAGTCAGTTACGATCG-3 ' ) were annealed and ligated to appropriately digested pBS-El or pBS-ElΔGA at 100:1 ratio^.

The chimeric El and E1ΔGA open reading frames were excised by

BStYT and Hindi ! digestion and inserted at the Smal site of pSCll. The pBS-E4IR plasmid was constructed by inserting an

Ncol-Apal EBNAl fragment (IR, gp: 108,067-109,261) nto the scI site of EBNA4 (gp: 97,302) . The E4IR open reading frame was excised by EcoRI and SstI digestion and ligated into pSCll. Insert containing plasmids were sequenced to determine correct orientation and reading frame alignment . Recombinant viruses were generated by transfection into CV-I cells infected with WR strain wild- ype vaccinia virus, and recombinant viruses were selected on TK-143 cells. Viral stocks were prepared and titrated in CV-l cells. Western blot analysis using a previously characterized human serum containing high antibody titres to all EBNAs (HR) confirmed that the E1N-E4 chimera runs as a major band of 80K, corresponding to the full-length protein, and a 70K band, probably arising from recombination within the Gly-Ala sequence, whereas the three E1ΔGA chimeras have a rough size of 52K. Attempts to detect the E4IR chimera by immunoblotting gave inconsistent results, probably because the poor transfer of the 1,334 amino-acid-long product. The HR serum was routinely used in complement enhanced immunofluorescence (ACIF) staining, Reedman et al. International Journal of Cancer 11, 499-520 (1973) to detect expression of the recombinant proteins in vaccine virus-infected human fibroblasts. Parallel slides were stained with affinity-purified antiGly-Ala human antibodies and with an EBNAl-specific monoclonal antibody. For quantification experiments the fibroblasts were grown in 8-chamber slides. Separate chambers were infected at a M.O.I, of 10 starting from 24 hours before the termination of the assay. Pairs of slides were stained with the HR serum and with an EBV antibody-negative serum as controls. The percentage of positive cells was estimated visually using an epifluorescence microscope. Digital image recording was done with a cooled CCD camera controlled by the HiPic software. The images were analyzed with the NIH image 1.56 software were fluorescence is proportional to the pixel value on a gray scale from 0 to 255. The mean fluorescence of each sample was calculated from the pixel value of at least 50 nuclei.

Fibroblast lysis experiments were conducted by growing semiconfluent monolayers of fibroblasts from the donor. These were grown in 96-well plates. Infection was done in the assay wells in the presence of 3 μCi^slNaCrO„ per well. CTL clones specific for the EBNA4 416-424 epitope were obtained by stimulation of lymphocytes from an EBV seropositive donor with the autologous B95.8 virus transformed LCL as described in Rickinson et al. supra. The cytotoxic activity was assayed in triplicate in standard four hour ⁵¹Cr-release assays.

Cell lysis assays were conducted using semiconfluent monolayers of fibroblast, infected and labelled as described above. Aliquots of 5 x 10⁶ cells were placed in 5-ml tubes and infected for one hour at 37°C with 1 ml concentrated virus before addition of 2 ml complete medium and further incubation for the indicated times. 100 μCi⁵¹ NaCr0₄ were added to each tube two hours before the initiation of the assay.

The invention is generally applicable to all MHC-mediated CTL response. Of specific interest is MHC Class-I restricted CTL responses. Class II mediated responses may be similarly inhibited, but since the T cells implicated in such response are generally different from those in Class I responses, the inhibitory effects achieved may be of lower importance . By modifying proteins to include a Gly-Ala repeat, by modifying the DNA encoding the same to include a Gly-Ala repeat domain segment encoding at least about 25-30 amino acids effectively inhibits presentation of the CTL-responsive epitope. Preferably, the repeat sequence is inserted downstream of the CTL-responsive epitope.

Ideally, the chimeric proteins, and DNA encoding the same, are effectively used in plasmids and other constructs effective in gene therapy, vaccine design and the like, where CTL response would interfere with attempts to employ exogenous proteins in therapy. One interesting utility is the modification of autoantigens, or potential autoantigens, expressed by an individual. Individuals (mammals) who are either suffering from autoimmune disease, or at risk to develop autoimmune disease, may be treated by providing them with DNA expressing the autoantigen with a Gly-Ala repeat domain as described. This may be used to render the expressed protein non-i munogenic.

This invention has been described both generically, and by reference to specific examples. The examples are not intended as limiting, except where so indicated, and adaptations and variations will occur to those of ordinary skill in the art without the exercise of inventive faculty. In particular, other proteins, epitopes, and insertion schemes will occur to those of skill in the art without deviating from the invention of the Applicants, saved as limited by the recitations of the claims set forth below.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: KURILLA, MICHEAL MASUCCI, MARIA

(ii) TITLE OF INVENTION: METHOD OF INHIBITING ANTIGEN PROCESSING, CHIMERIC PROTEINS REFLECTING THIS METHOD, AND METHODS OF EXPRESSING THOSE PROTEINS

(ϋi) NUMBER OF SEQUENCES: 7

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: OBLON, SPIVAK, MCCLELLAND, MAIER & NEUSTADT,

P.C.

(B) STREET: 1755 S. JEFFERSON DAVIS HIGHWAY, SUITE 400

(C) CITY: ARLINGTON

(D) STATE: VA

(E) COUNTRY: USA

(F) ZIP: 22202

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentln Release #1.0, Version #1.30

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: US 08/660,272

(B) FILING DATE: 07-JUN-1996

(C) CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: OBLON, NORMAN F.

(B) REGISTRATION NUMBER: 24,618

(C) REFERENCE/DOCKET NUMBER: 494-195-0

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: 703-413-3000

(B) TELEFAX: 703-413-2220

(2) INFORMATION FOR SEQ ID NO:l:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 9 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:l: lie val Thr Asp Phe Ser Val lie Lys

1 5 (2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 33 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: other nucleic acid

(A) DESCRIPTION: /desc = "SYNTHETIC PRIMER"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

CATGCCATAG TAACTGACTT TAGTGTAATC AAG

(2) INFORMATION FOR SEQ ID NO: 3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 31 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: other nucleic acid

(A) DESCRIPTION: /desc = "SYNTHETIC PRIMER"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

CATGCTTGAT TACACTTAAG TCAGTTACTA T

(2) INFORMATION FOR SEQ ID NO:4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 29 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: other nucleic acid

(A) DESCRIPTION: /desc - "SYNTHETIC PRIMER"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:

CGATCGTAAC TGACTTTAGT GTAATCAGG

(2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 33 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: other nucleic acid

(A) DESCRIPTION: /desc = "SYNTHETIC PRIMER" (xi) SEQUENCE DESCRIPTION: SEQ ID NO:5: CCTTGATTAC ACTAAAGTCA GTTACGATCG TGC (2) INFORMATION FOR SEQ ID NO:6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 33 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: other nucleic acid

(A) DESCRIPTION: /desc = "SYNTHETIC PRIMER"

( i) SEQUENCE DESCRIPTION: SEQ ID NO: 6:

TAACGATCGT AACTGACTTT AGTGTAATCA AGG

(2) INFORMATION FOR SEQ ID NO:7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 32 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: other nucleic acid

(A) DESCRIPTION: /desc = "SYNTHETIC PRIMER"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:7:

TTACCTTGAT TACACTAAGT CAGTTACGAT CG

Claims

CLAIMS:

1. A method of introducing a foreign protein nto a cell of a host, without triggering a cytotoxic T lymphocyte response to said cell, comprising expressing said protein as a chimeric protein which chimeric protein is comprised of a structural domain which contains an epitope and a domain of about 30-150 amino acids consisting of a repeat domain of Gly-Ala, said repeat domain being downstream of said epitope, as said chimeric protein is expressed, wherein said structural protein, if expressed in said host cell without said Gly-Ala repeat domain, triggers a major histocompatibility complex-mediated cytotoxic T lymphocyte (CTL) immune response.

2. The method of Claim 1, wherein said chimeric protein is introduced to said cell by expression with that cell encoded by foreign DNA introduced into said host cell.

3. The method of Claim 2, wherein said foreign DNA s introduced nto said host cell by means of a recombinant virus comprising said foreign DNA.

4. A chimeric protein, comprising a structural protein which comprises an epitope which induces a CTL immune response in a host when a cell of said host is transfected with said protein, said chimeric protein further comprising a Gly-Ala repeat domain of about 30- about 150 ammo acids.

5. The chimeric protein of Claim 4 , wherein said CTL immune response is induced by an epitope in said protein, and said repeat domain is downstream, as the chimeric protein is expressed, of said epitope.

6. The chimeric protein of Claim 4, wherein said CTL immune response is a MHC Class-I mediated response.

7. DNA which encodes a chimeric protein, comprising DNA which encodes a protein, said DNA further comprising a nucleotide sequence which encodes a domain of about 30-150 amino acids, said 30-150 amino acids consisting of a Gly-Ala repeat domain.

8. The DNA of Claim 7, wherein said protein comprises an epitope which, when said protein is expressed in the cell of a foreign host without said Gly-Ala repeat domain, said epitope induces a CTL immune response, and wherein said DNA encoding said Gly-Ala repeat domain is inserted downstream of DNA encoding said epitope.

9. An expression vehicle, comprising the DNA of Claim 7, operably linked to a promoter sequence.

10. A virus comprising the expression vehicle of Claim 9.