use of persistent infections with vaccinia virus recombinants to introduce alterations in foreign...

Virus Research

E L S E V I E R Virus Research 46 (1996) 45 56 , ,

Use of persistent infections with vaccinia virus recombinants to introduce alterations in foreign proteins: an application to

HIV-1 e n v protein

Mariano Esteban*, Juan Ramdn Rodriguez, Victoria Jimenez, Dolores Rodriguez

Department of Cell and Molecular Biology, Centro Nacional de Biotecnologla, CSIC, Campus Universidad Autdnoma, 28049 Madrid, Spain

Received 16 November 1995; revised 10 July 1996; accepted l0 July 1996

Abstract

With the aim of generating a virus-cell system to introduce alterations in proteins of interest--which may be of use in studies of their biological functions--we established a persistent infection on a B-lymphoma cell line (A20.2J) with vaccinia virus (VV) recombinants. As a model, we used a vaccinia virus recombinant expressing the human immunodeficiency virus HIV-1 env gene. In this unique virus-cell system, we found that it is possible to introduce several structural and functional alterations in the env protein with passage numbers. From passage 10-20, two new env products emerged: an uncleaved gpl60 and a glycoprotein fragment of 110 kDa. The uncleaved gpl60 exhibit interesting properties as an immunogen. This protein forms stable oligomers, is not released from the cells, cannot fuse CD4 + presentirLg HeLa cells and activates a stronger cellular immune response than the parental cleaved env. In contrast, the 110 kDa product is a poor immunogen, since it lacks the gp41 domain, cannot form oligomers, accumulates intracellLularly and cannot fuse CD4 + cells. In the persistently infected cells we have also found alterations in another heterologous protein--fl-galactosidase--a gene inserted in the same locus of VV as the env

gene. This alteration resulted in a truncation of the (fl-galactosidase protein from 125 kDa to about 70 kDa. A similar size truncation of env and of fl-galactosidase was observed in many of the isolated VV recombinants. Copyright © 1996 Elsevier Science B.V.

Keywords: Vaccinia virus; Persistent infections; HIV env

1. Introduction

* Corresponding author. Tel.: + 34 1 5854503; fax: + 34 1 5854506; e-mail: [email protected]

It has been previously described that persistent

infections with vaccinia virus ( W ) can be estab-

lished in Fr iend erythroleukemia (FEL) cells

0168-1702/96/$15.00 Copyright © 1996 Elsevier Science B.V. All rights reserved Pll S01 68-1702(96)01 380-9

46 M. Esteban et al. / Virus Research ,16 (1996) 45 56

(Pogo and Friend, 1982; Paez et al., 1985) and in a human leukemic cell line (K562) (Pogo et al., 1991). We have shown that during the virus persistence in FEL cells numerous changes are introduced in the virus genome, i.e point mutations, and deletions, which result in alterations in cer- tain viral proteins (Paez et al., 1985, 1987). Sig- nificantly, the mutant viruses obtained after several passages had markedly reduced patho- genicity and induce in animals a protective immu- nity against VV (Paez et al., 1987).

In this investigation we investigated the possible effect of persistent infections on recombinant proteins expressed by VV. With this purpose, we established a persistent infection in a B-lymphoma cell line (A20.2J) with a recombinant VV expressing the HIV-1 env protein, and we examined the alterations introduced in the env protein during persistence. The env mutants exhibit unique properties with regard to oligomerization, cleavage, cell surface expression and immune response.

2. Materials and methods

2.1. Cells and viruses

ings in MEM and cells were transferred to 25 cc flasks in complete medium. Cells were passaged once a week and, at various cell passages aliquots were removed for virus titrations, Hoescht staining for viral factories, virus plaque isolations, and protein analyses. Cell viability was monitored by trypan blue exclusion. Virus titrations were carried out in African green monkey kidney cells (CV-1 or BSC-40), as previously described (Dallo and Esteban, 1987). For virus isolation at different cell passages, BSC-40 cells were infected with a cell extract derived from persistently infected cells and individual plaques were grown in BSC- 40 cells and plaque purified three times.

2.3. Protein analyses

A20 cell extracts obtained at different cell passages, or cell extracts derived from CV-1, BSC-40, HeLa or chick embryo fibroblast (CEF) cells infected with plaque purified viruses isolated at various passages, were run on 10% SDS-PAGE gels, proteins transferred to nitrocellulose paper and blots reacted with specific antibodies, as previously described (Dallo and Esteban, 1987; Ro- driguez et al., 1987).

A20.2J cells, an established B lymphoma cell line from mouse Balb/c (Kim et al., 1979), was obtained from Bryan Thomas, National Institute for Medical Research, London, UK. Cells were cultured in a humidified atmosphere of 5% CO2 in MEM containing 2 mM glutamine and 10% fetal calf serum (complete medium). Cells were diluted twice a week from 1 x 106 to 0.4 X 106 with fresh medium. VV recombinants, VV-env-1 and M7- env-1, expressing the entire env gene of HIV-1 (strain III B) have been described previously (Ro- driguez et al., 1989). M7-env-1 is derived from the highly attenuated VV mutant 48-7 (Dallo and Esteban, 1987).

2.2. Establishment o f virus persistence

To establish virus persistence, A20 cells at 107

cells/ml in MEM were infected with 1 PFU/cell of purified VV recombinants. After 1 h of virus adsorption, virus inoculum was removed by wash-

2.4. Antibodies

The antibodies used were: rabbit polyclonal anti-VV (Paez et al., 1987), rabbit polyclonal anti- gpl20 (Rodriguez et al., 1989), monoclonal anti- V3 loop (NEA 9305; DuPont), monoclonal anti-gp41 (M-H) (was a gift from J. Skehel, and recognizes the amino terminus region of gp41), monoclonal antibodies (aa 329-431; aa 42-113; aa 256-273; aa 723-741) were obtained from George Lewis through AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases.

2.5. Cell surface localization o f env by ELISA

BSC-40 cells grown on microtiter plates were infected (5 PFU/cell) with VV-env recombinants in duplicate. At 18 hpi cells were made non-per- meable by fixation with 0.1% glutaraldehyde in PBS, washed and incubated in 5% non-fat dry

M. Esteban et al. / Virus Research 46 (1996) 45 56 47

milk in PBS (blotto) with different dilutions of the corresponding antibodies. Cells were washed, incubated in blotto with a second antibody labeled with peroxidase and washed• Antibody reactivity was developed with 50 ktl of 25 mM sodium citrate, pH 4.5, containing 0.02% 3,3',5,5'- tetramethyl-benzidine dihydrochloride and 0.01% hydrogen peroxide. The reaction was stopped with 50 /~1 of sulfuric acid (1 M) and optical densities were measured at 450 nm. Values given were corrected afte:r subtracting the non-specific antibody bound to uninfected cells.

2.6. Proliferative response to HIV-1 env

Bulk spleen cells (5 x 105) from mice, immunized with VV-env recombinants, were cultured with various dilutions of antigen in 0.2 ml of RPMI 1640 medium supplemented with glutamine (2 mM), 100 /ag/ml each of streptomycin and penicillin and 10% fetal calf serum in 96-well round bottom plates. After 4 days of incubation at 37°C, the cells were pulsed with 1.25 pCi of 3H-thymidine per well for 16-18 h. Cells were harvested onto glass filter paper with a semiauto- matic cell harvester and 3H-thymidine incorporation was measured in 1.0 ml of liquid scintillation and counted.

(Fig. 1B), there is heterogeneity in env with a passage number. We maintained the persistent infection for about 6 months. During this time the

A

206.-o,,

1 1 6 " - I ~

66.--~

1 2 3 4 5 6 7 8

4 5 - ~

29.-~

B , ~ ~ ' , , IMP ~ . ~ "~,"-'- gp 1 6 0

g p 1 2 0 ~ ~,~ ~ '41Bgp l 20

gp70 "-ID"

g p S 0 - - I ~ ~ • "

3. Results

3.1. Establishment o f a persistent infection with vaccinia recombinants expressing the env gene of HIV-1 in a B-lymphoma cell line

To determine whether alterations could be introduced in a foreign protein like the env protein of HIV-1, we established a persistent infection in A20 lymphoma B cells infected with two VV recombinants expressing the env gene (VV-env-1, and M-env-1). The advantage of A20 cells is that this cell line expresses MHC class 1 and class I1 receptors and can be used as antigen presenting cells to CD4 + and CD8 + T cells. As shown by a western blot in Fig. 1A, VV proteins are produced at all different ceil passages• In western blots reacted with a polyclonal antibody against gpl20

'~ 2 '3 4 5 6 7

Fig. 1. Vaccinia virus and HIV env polypeptide patterns during VV-env-I and M7-env-1 persistence in A20 cells. At different cell passages during persistence of A20 cells infected with VV recombinants, cell extracts were processed for SDS PAGE and western blots analyzed after reactivity with anti- VV or anti gpl20 antibodies. Panel A. lmmunoreactivity with anti-VV antibodies. CV-I cells, uninfected (lane 1) and infected (1 pfu/cell) for 24 h with wild type VV (lane 2); A20 cells persistently infected with M7-env-1 at passage 10 (lane 3) and passage 20 (lane 4); A20 cells persistently infected with VV-env-1 at passage 10 (lane 5), passage 15 (lane 6), passagel6 (lane 7) and passage 20 (lane 8). Panel B. lmmunoreactivity with anti-gpl20 antibodies, Purified gpl20 from VV-env- l , infected HeLa cells (lane 1); A20 cells persistently infected with VV-env-1 at passage 10 (lane 2), passage 15 (lane 3), passage 16 (lane 4); A20 cells persistently infected with M7-env-I at passage 20 (lane 5). HeLa cells infected (1 pfu/cell) for 24 h with parental VV-env-I (lane 6) or with M7-env-1 (lane 7).

48 M. Esteban et al. / Virus Research 46 (1996) 45-56

Table 1 Cell viability and virus yields in A20 cells persistently infected with vaccinia virus

Cultures Passage number Number of cells/ml % alive Virus yields (PFU/ml)

1 W R 6 1.1 x 106 20 3.5 x 105 WR-env 6 0.5 × 106 50 2.5 × 103 M7-env 6 1.4 x 106 20 1.5 x 106

2 WR 6 1.2x 106 20 1 x 106 WR-env 6 2.1 X 10 6 20 5 x 104 M7-env 6 4.2 × 105 50 6 x 106

At passage 6, aliquots of cell cultures infected with different viruses (wild type, WR; VV recombinant expressing HIV-1 env; mutant7 expressing HIV-1 env) were collected and cells counted. Cell viability was determined after counting the cells that exclude trypan blue. Virus yields were determined after titration in BSC-40 cells. Results from two independently established cell cultures are shown.

virus titers vary with passage number from about 103 t o 106 PFU/ml. For example, at passage 6 the number of viable cells is between 20-50% in two independently established sets of cultures (Table 1). From the yields of virus in the cultures we can estimate that all cells in a culture should be infected, because the ratio of virus yields to cells is over 1 pfu. We confirmed by Hoescht staining of viral factories that nearly all cells are infected. This was further confirmed after the persistently infected cells were used as targets in a CTL assay with splenocytes derived from mice immunized with purified recombinant viruses or with infected cells (Table 2). This assay shows that persistently- infected cells were as good targets as newly infected A20 cells. At passage 17, the number of cells infected in the population were about 20%, as determined by the formation of virus plaques appearing from single cells plated on top of BSC- 40 cells, that were subsequently fixed by addition of agar. The virus persistence could not be established with other cell lines, like H9 and U937. We conclude that a persistent infection can be established in A20 cells and that survival of cells is probably mediated by host factors that limit the extent of VV infection.

3.2. Alterations in the size o f H I V - I env during pers&tent infections with VV-env recombinants

To characterize the virus population present in cell cultures during persistence, we isolated 20 plaque purified viruses at passages 10 and 20, and

infected different cell types (HeLa and CEF) with these isolates. The size of env was analyzed by western blots. For comparative studies, we also examined the size of fl-galactosidase, since the VV recombinants contain the lac Z gene in the same locus as the env gene. At passage 10 and with representative isolates (Fig. 2), we observed that some of the VV recombinants expressed an uncleaved gpl60 (panel A, lanes 4, 5 and 9), while other VV recombinants express fl-galactosidase of shorter size, 70 instead of 125 kDa (panel B, lanes 2, 3, 7 and 8). At passage 20, a higher proportion of VV recombinants express env with a size of about 110 kDa (panel C, lanes 5 -1 l ) and fl- galactosidase with a size of about 70 kDa (panel D, lanes 3-5, 7-9, and 11). Some of the VV recombinants obtained at late cell passages were highly fusogenic in infected, monkey BSC-40 cells (data not shown). To define the genotype of these variants, we carried out Southern blot analysis with specific probes using HindI l l and E c o R I restriction enzyme digestion of DNA isolated from cells infected with some of the mutants. We could not detect the occurrence of deletions, because the restriction patterns were the same. To detect if point mutations were present in the DNA, we sequenced the gpl20-gp41 cleavage site and found no differences (data not shown). The above findings demonstrate that alterations in the size of env are produced during a persistent infection of A20 cells with VV-env recombinants. Point mutations inserted at one or more sites of env are likely to be responsible for changes in protein size.


Table 2 CTL assay on A20 cells infected with vaccinia virus

Effector cells Target cells

Splenocytes from mice immunized with A20 A20-infected with WR A20-persistently infected with M7

WR-env 6 57 62 M7-env 1 37 20 A20-infected with WR 0 29 33 A20-infected with M7 0 22 17

Effector cells were splenocytes derived from Balb/c mice inoculated i.p with either purified VV recombinant viruses at 107 PFU/mouse or with 106 ~pleen cells infected with VV. Target cells were A20 cells infected with VV (2 PFU/cell) and A20-persistently infected with VV at passage 6. The effector to target ratio was 30:1 in the standard chromium release assay. The percent values are represented.

3.3. Differential structural properties o f env with

passage number o f ,420 cells persistently infected

with VV-env recombinants

Since naturally occurring gpl60 exists as a heavily glycosylated protein in the form of oligomers (Weiss et al., 1990; Schawaller et al., 1989), we next examined the extent of oligomerization of env, by infecting cells with different VV env recombinants isolated at passage 10 and 20. Thus, BSC-40 cell,.~ were infected with VV-env

recombinants, cell extracts were prepared in non reducing buffer or in a buffer containing 1 mM DTT, fractionated by SDS-PAGE and env

oligomers were analyzed by immunoblot after reactivity with anti-gpl20 antibodies. As shown in Fig. 3 (panel A), virus isolates at passage 10 that produced uncleaved gpl60 form oligomers (lanes 2 4 and 10) similar in size to the oligomers observed by the VV recombinant expressing the wild type env (lane 1). However, oligomers were not formed by VV recombinants isolated at passage 20 that produced env with a size of 110 kDa (Fig. 3, lanes 5-9). The oligomers derived from uncleaved gpl60 were more stable than those produced by wild type env, as revealed after treatment of cell extracts with 1 mM DTT (Fig. 3, panel B). However, it is clear that while oligomers are dissociated in wild type env by the reducing agent, oligomers are still observed after 1 mM DTT treatment, with uncleaved gpl60. The extent of glycosylation was analyzed after tunicamycin treatment. The size of the uncleaved gpl60 was

reduced to 80 kDa-- the predicted size for the unglycosylated gpl60 (Stein and Engleman, 1990), while the size of the 110 kDa product was reduced to 60 kDa, indicating that this glycoprotein resulted from truncation in the coding sequence (data not shown). This was also confirmed by immunoblot analyses with antibodies reacted against gpl20 and gp41 (Fig. 4). We found that VV recombinants producing the 110 kDa glycoprotein (M-env-7) lack the C-terminal region of env, including the entire gp41 region.

Because gpl20 is released from cells after cleavage (Willey et al., 1988), we then analyzed the env

products present in the medium after infection of cells with VV recombinant viruses. There was no release of gpl20 or any other env product in the medium from cells infected with any one of the isolated VV recombinants (Fig. 5, lanes 2 and 4). Failure to release env should lead to an accumula- tion of env, possibly at the cell membrane. To investigate the localization of env, we tested by ELISA with a panel of antibodies the extent of antigen presentation with non permeabilized cells. The comparative results from cells infected with VV recombinants that produce uncleaved gpl60 (M-env-8), the 110 kDa truncated env (M-env-7) and wild type env (VV-env-1) are shown in Fig. 6. It is clear that some differences exist in the extent of antibody reactivity between the different env

forms. For example, antibodies to the V3 loop recognized poorly uncleaved env. This is not due to differences in abundance of env at the cell surface, since similar optical density for uncleaved

50 M. Esteban et al. / Virus' Research 46 (1996) 45 56

and cleaved e n v was obtained after reactivity with polyclonal antibodies against gpl60. There were also differences in antibody binding with mAbs against gp41 and to the conserved domains C1 (aa: 42-113) and C2 (aa: 256-273). The ELISA

data was confirmed by immunofluorescence analysis. The VV recombinants producing uncleaved gpl60 and the 110 kDa truncated protein were unable to induce gpl20-CD4 mediated cell fusion in HeLa cells expressing human CD4 and infected

A 1 2

g p 1 6 0 -.,1~ g p 1 2 0 ''m'='

HeLa CEF

3 4 5 6 7 8 9 10

B

Hg-HW VV-env-1 M7-env-1 p10"4 p10~5 p10-12 W-env-1 MT-mv-1 p10-5 plG-1Z

M7-env-1 M7-env-1

CEF HeLa 1 2 3 4 5 6 7 8 9 10

B -G a I ' - I~ j ~ ~ ~ ~,-.e-e .....

p10-12 plO-$ plO-4 M7-env-t VV-elv-1 p10-12 p10-5 plO-4 M7-env-~ W-env*l

M7-env-1 M7-env~l

1 2 3 4 5 6 7 8 9 1 0 11

gp 16 0 " ~ !

gpl 20-~ ~ ..........

C

B - G a I - I ~

1 2 3 4 5 6 7 8 9 1 0 11

D

Fig. 2.


in the presence o f rifampicin (not shown). F r o m the findings o f Figs;. 4 - 6 , we conclude that specific muta t ions are introduced in e n v during virus persistence, that these muta t ions prevent the cleaveage o f gp l60 at the gpl20-gp41 boundary , they enhance the stability o f oligomers o f e n v and can also enhance or decrease presentat ion o f different domains o f e n v at the cell surface.

3.4. I m m u n e r e s p o n s e to V V r e c o m b i n a n t s

e x p r e s s i n g m u t a n t f i ) r m s o f e n v

We next examined the ability o f VV recombinants to trigger humora l and cellular immune responses in Balb/c mice. Groups o f five mice were primed by the i.p route with 5 × 10 7 P F U of recombinant virus, 20 days later animals were boosted with 108 P F U of the same virus and serum samples were collected after 20 days. Anti- e n v reactivity in pooled serum was analyzed by E L I S A in plates coated with purified gpl60. The results o f Fig. 7 (panel A) show an increase in an t ibody levels in zLnimals immunized with a VV recombinant producing uncleaved gp l60 in com- !barison with animals immunized with a VV recombinant producing wild type env . There was poor an t ibody response in animals immunized with a VV recombinant producing the 110 k D a truncated e n v protein. The cellular immune response was examined in splenocytes. As shown in Fig. 7 (panel B), the proliferative response was higher in animals immunized with the VV recombinant virus expres~sing uncleared gp l60 relative to the VV recombinant expressing wild type e n v

or the 110 kDa truncated env protein.

4. Discussion

VV is a highly cytocidal virus and cells in culture are destroyed as a result o f infection. However, it has been previously shown that persistent infections can be established in mouse and human leukemic cell lines (Pogo and Friend, 1982; Paez et al., 1985; Pogo et al., 1991). In this investigation we showed that VV persistent infections can also be established in mouse B lymphoid cells. It is significant that lymphocytes are a target for poxvirus replication in vivo. In fact, in a generalized poxvirus infection blood cells are known to carry the virus to the target organs where cont inuous virus replication occurs (Fenner et al., 1989). The ability o f the virus to replicate in lymphoid cells might represent a strategy of the virus to avoid the host immune response. The readiness o f the virus to establish a persistent infection in these cells in the absence o f immunological pressure could be interpreted as the result o f the adapta t ion o f the virus to grow under these conditions.

H o w is the persistent infection established?. The simplest interpretation is that only a fraction o f the cells are infected, and that the infected cells die and liberate the virus, which poor ly infects surviving cells. However, as indicated in Table 1, there is enough infectious virus in the cultures to infect all cells. We estimated that in some passages all or nearly all o f the cells are infected, while in other passages only about 20% o f the cells are infected. Our interpretat ion o f the mech- anism of virus persistence is that the host provides

l~ig. 2. HIV env and fl-gal protein patterns after 10 and 20 cell passages during virus persistence. Stocks of plaque purified viruses obtained at passage 10 and 20 from A20 cells persistently infected with M7-env-l , were used to infect for 24 h monolayer cultures of HeLa cells and CEF cells. Cell extracts were prepared and the polypeptide patterns were analyzed by western blots. Panels A and B, represent env and fl-gal protein patterns at passage 10. Panel A. lmmunoreactivity with anti-gpl20 antibodies. Cell extracts from H9 cells chronically infected with HIV (lane 1). Cell extracts from HeLa cells infected for 24 h with parental VV-env-1 (lane 2) or M7-env-I (lane3); HeLa cells infected with three different M7-env-1 virus isolates (p10-4, lane 4), (p10-5, lane 5), (p10-12, lane 6). CEF infected for 24 h with parental VV-env-I (lane 7) or M7-env-1 (lane 8). CEF infected with M7-env-1 virus isolates (p10-5, lane 9) and (p10-12, lane 10). Panel B. lmmunoreactivity with anti fl-gal antibodies. CEF infected with M7-env-1 virus isolates (p10-12, lane 1), (p10-5, lane 2), (p10-4, lane 3) or with the parental viruses M7-env-1 (lane 4) and VV-env-1 (lane 5). HeLa cells infected with the M7-env-1 virus isolates (p10-12, lan e 6), (p10-5, lane 7), (p10-4, lane 8) or infected with parental M7-env-1 virus (lane 9) and VV-env-1 (lane 10). Panels C and D, represent env and B-gal protein patterns from several plaque-purified viruses obtained after 20 passages from A20 cells persistently infected with M7-env-1 and reacted with anti-gpl20 (panel C) or anti-fl-gal (panel D) antibodies. Panel C. Cell extracts from HeLa cells infected with parental viruses, M7-env-I (lane 1) and VV env-I (lane 2). Lanes 3-10, represent different virus isolates. Panel D. Same as C, but the blot was reacted with anti-fl-gal antibodies.

52 M. Esteban et al. / Virus Research 46 (1996) 4 5 - 5 6

a restriction to the virus. This restriction could be

at the level of virus entry or through induct ion of

cytokines and act ivat ion of apoptosis. We con-

sider the latter possibility most likely, since A20

cells are infected with VV and lymphoid cells

infected with VV die by apoptosis (manuscr ipt in

preparat ion). Since VV encodes an array of genes which counteract cytokine responses (Smith,

1993), it is likely that persistence in A20 cells is

established because of a balance between VV in-

terference products and the act ion of cytokines.

This is under investigation.

In the process of HIV-1 infection numerous

genetic changes are in t roduced in the e n v gene.

These muta t ions are due to the error rate of the

reverse transcriptase enzyme and variants with

4 1 IP,.lln

-100

- I l l

m -&ll

'2"/

1 2 3 4 5 6 7 8 9 1 0 A

gp160 gp120

B

gp160 ..-- gp120 --

B

u t i gl)'41

1 2 3

Fig. 3. Extent of oligomerization of env from VV-env recombinants isolated during virus persistence of A20 cells. BSC-40 cells were infected (1 pfu/cell) for 24 h with different recombinant viruses, cell extracts were prepared in non-reduping buffer or in buffer containing I mM DTT and then fractionated by SDS-PAGE. The extent of oligomer formation of env was analyzed by immunoblot after reactivity with anti-gpl20 antibodies. Panel A. Gel run under non reducing conditions. Extracts from parental M7-env-1 infected cells (lane 1); extracts from cells infected with different M7-env- I isolates at passage 10 (lanes 2-4 and 10); extracts from cells infected with different virus isolates of M7-env-1 at passage 20 (lanes 5-9). Panel B. Same as panel A, but the gel was run in a buffer containing 1 mM DTT.

Fig. 4. Truncation of env occurs during virus persistence of A20 cells with vaccinia-env recombinants. Representative virus isolates at passage 10 (M-env-8) and passage 20 (M-env-7) were used to infect BSC-40 cells for 24 h and cell extracts were ahalyzed by SDS-PAGE. Western blots were reacted with anti-gpl60 antibodies (panel A) or against a mAb specific for gp41 (panel B).

muta t ions in the hypervariable regions of e n v

accumulate with time of infection. The i m m u n o -

logical pressure established in vivo contr ibutes to

the selection of a viable popu la t ion of HIV that

persist in an infected individual (Ho et al., 1995;

Wei et al., 1995). The con t r ibu t ion of the defective

M. Esteban et al. / Virus Research 46 (1996) 4 5 - 5 6 53

e n v molecules, which are produced in vivo but do not become encapsidated into infectious particles, and their role in the', activation of immunological response and in the control of HIV infection has not been established (Levy, 1993; Meyers et al., 1993; Temin, 1993; Ho et al., 1995; Wei et al., 1995; Chainer et al., 1994).

The objective of using a recombinant virus expressing HIV-1 e n v to establish a persistent infection was to explore the possibility of generating mutant forms of env , which might or might not be produced in the natural infection but yet retain important biochemical functions relevant to their ability to actbcate significant immunological responses. The results obtained demonstrate that during a persistent infection of a B-lymphoma cell line with vaccinia-HIV e n v recombinants, various mutant forms of e nv are produced with or without truncations at the C terminal site of the protein. These protein alterations can enhance or decrease

r- 7 ~o - 9 > >

z :~ z Z tu Z LU

> UU

-!- =E > =£ 1£

gp160- i op12O- 11111

0.2 ' "

0.1

1/400 1/800 1/1600 serum dilutions

0 0 8 . , ,

o o ll i 0 , , , ,L

1/400 1/800 1/1600 serum dilutions Eo0 _ o oo61, L

, i O ~ o , , ,

v 1/400 1/800 1/1600 1/400 1/800 1/1600 s e r u m d i l u t i o n s

0,12-- =E anti aa: 7 3 2 - 7 4 1

1/400 1/800 1/1600

serum ddut~o~

012 -

= I ~ E anti 9p160 o

o,o

1/400 1 /800 1 / 1 6 0 0

serum dlMhOn%

s e r u m d d u t l o n s

01 11,: i

1/400 1 /800 1 /1600 serum dilutions

[ ] VV-env-1

[ ] M-env-8

[ ] M-env-7

[ ] WR

Fig. 6. Differential localization of env at the cell membrane from cells infected with VV-env recombinants. BSC-40 cells were infected with VV-env recombinants (5 pfu/cell) and 18 h later cells were fixed in non-permeabilized conditions. The extent of cell surface expression of env was analyzed by ELISA using a panel of antibodies. Bars, denote the virus origin. W R is the VV wild type (Western Reserve). The s tandard deviation of the mean was less the 5% in all cases.

1 ?- 3 4 5

Fig. 5. VV-env recombinants isolated during virus persistence of A20 cells do not release env products from the cells. Representative virus isolates were used to infect BSC-40 cells fro 24 h and env products in the medium were recovered by concentration in Amicon filters. The products were analyzed by SDS-PAGE and blots were reacted with anti-gpl20 serum. Extracts from H9 cells persistently infected with HIV (lane 1); supernatants from cells infected with virus isolates at passage 20 (M-env-7, lane 2) and 10 (M-env-8, lane 4) or from cells infected with the parental virus VV-env-1 (lane 3) and M7-env- 1 (lane 5).

the ability of the e n v protein to form oligomers, to be localized on the cell surface and to activate immunological responses.

In the novel persistent VV-cell system described here, the appearance of the genetic alterations in env appear to be similar to those found during short and prolonged passage of simian immunodeficiency virus (SIV). In the SIV system extensive deletions occur in the cytoplasmic tail of e n v during long term passages, whereas minimally passaged viruses express the full-length glycoprotein (Chakrabart i et al., 1989; Hirsch et al.,

54 M. Esteban et al. / Virus Research 46 (1996) 45 56

1,4~ A

8 0,8 g

"~ 0,6

~ 0,4

o 0,20

1/250 1/500 1/1000 1/2000 1/4000 serum dilution

• WR

[ ] VV-env-1 [ ] M7-env- 1

M-env-8 [7 M-env-7

B

x -'o

c -

~ L M-ENV-7 .--(~ __ M-ENV-8 ~ 2 VV-LUC .~ •

0 Oil ]

gp 160 l.tg/ml

Fig. 7. (A) Humoral immune response in mice immunized with VV-env recombinants. Groups of mice (5 per group) were primed (i.p) and booster as described in the text. Anti HIV reactivity in pooled serum was analyzed by ELBA in plates coated with purified recombinant gpl60 (American Biotechnologies). (B) Proliferative response of spleen cells from mice primed and boosted with VV-env recombinants. Mice were primed ip with (5 x l07 pftl) of VV-env-1, M-env-7, M-env-8 and VV-LUC, and 20 days later animals were boosted with 108 pfu of the corresponding virus. Spleens were removed eleven months after priming and spleen cells (5 x 105) were incubated with various concentrations of purified gpl60 (obtained from American Biotechnologies). 3H-thymidine incorporation was measured as described under Materials and Methods. VV-LUC is a recombinant expressing luciferase, and was used as a control.

1989; Ritter et al., 1993). Extensive C-terminal deletion of gpl60 has also been observed in HIV variants arising during long-term cultures of

chronically infected cells (Zaides et al., 1994). Our results revealed that truncations in the C-terminal region of e n v can occur independently of HIV

M. Esteban et al. / Virus Research 46 (1996) 4 5 - 5 6 55

infection, that VV recombinants contain the entire e n v gene after 20 petssages and that truncation of e n v was not due to a DNA deletion. This was confirmed by Southern blot hybridization analysis with specific probes. Moreover, nucleotide sequence analyses of 150 nucleotides spanning the gpl20-gp41 cleavage site of e n v , fail to detect base substitutions within this region in virus isolates that produced uncleaved gpl60 or from virus isolates that produced a 110 kDa product (data not shown). This indicates that genetic lesions are introduced in other domains of e n v . In fact, this is supported by the increased binding of some mAbs (aa 42-113; aa 256--273) that recognize conserved regions (C1 and C2) of gpl20 (Moore et al., 1994), by a decrease binding of a mAb to the V3 loop region (Fig. 6) and by the inability of uncleaved gpl60 to fuse with CD4 + presenting HeLa cells (data not shown). Increased exposure at the cell surface of domains of e n v which are normally poorly presented or silent, might trigger a better immunological response than wild type e n v . The results obtained measuring humoral and proliferative response with uncleaved gpl60 sup- port this view (Fig. 7).

An interesting feature about the persistent virus-cell system described in this investigation is the dominant phenotype of the VV recombinants that are selected, which contain truncations in e n v

and l a c Z proteins. These truncations occur largely when the initial virus inoculum used for the establishment of the persistent infection was derived from a VV recombinant based on a mutant 48-7 virus isolated at passage 48 during a persistent infection in FEL cells (Paez et al., 1985). In fact, when the persistent infection was established with a VV recombinant derived from wild type VV, fewer virus isolates contained alterations on e n v or #-gal after 20 passages. This is probably related to the superior adaptation of mutant 48-7 to growth during virus persistence in B cells.

Generation of VV recombinants expressing e n v

molecules with structural alterations, might lead to molecules with interesting properties as im- munogens. These novel molecules, with the ability to enhance or decrease the immunological response of the host, may in turn be good antigenic

candidates, able to modulate the extent of activation of Thl versus Th2 cell populations. Recent studies suggest the importance of the balance between Thl and Th2 cells in the control of HIV infection (Salk et al., 1993; Cohen, 1993; Clerici and Shearer, 1993).

Although in this investigation we have charac- terized e n v molecules with between 10 and 20 cell passages, it is likely that with longer passages additional alterations might be introduced in the e n v protein. In fact, we have described the introduction of numerous genetic alterations in the VV genome of wild type virus during 2 years of virus persistence in FEL cells (Paez et al., 1987). A potential advantage of the persistent virus-cell system described here is that it should allow the correlation in vivo of a defective function in a foreign protein with a specific genetic alteration. This system could be used in the selection of multiple variants of e n v and of other VV heterologous proteins for biochemical and immunological studies.

Acknowledgements

This investigation was supported in part by Research Grants A1 34552 from the National Institutes of Health, USA, and BIO92-0401 from the Comision Interministerial de Ciencia y Tec- nologia of Spain. We are particularly grateful to Cecilio Lopez Galindez for DNA sequence analysis of the gp120-41 domain of e n v . D.R and J.R.R were supported by research contracts from CSIC and MEC of Spain.

References

Chainer, 1.R., Heurichwork, S., Hadida, F., Pelletier, E., Ok- seuhendler, E., Autran, B. and Wain-Hobson. (1994) HIV and T cell expression in splenic white pulps is accompanied by infiltration of HIV specific cytotoxic T lymphocytes. Cell. 78, 373-387.

Chakrabarti, L., Emerman, M., Tiollais, P. and Sonigo, P. (1989). The cytoplasmic domain of simian immunodeficiency virus type 1 envelope glycoprotein. J. Virol. 63, 4395 -4403.

56 M. Esteban et al. / Virus Research 46 (1996) 45 56

Clerici, M. and Shearer, G.M. (1993). A Thl-Th2 switch in a critical step in the etiology of HIV infection, lmmunol. Today. 14, I07-111.

Cohen, J.T. (1993). T cell shift: key to AIDS therapy?. Science. 262, 175 176.

Dallo, S. and Esteban, M. (1987). Isolation and characteriza- tion of attenuated mutants of vaccinia virus. Virology. 159, 408 -422.

Fenner, F., Wittek, R. and Dumbell, K.R. (1989). The Ortho- poxviruses. Academic Press, San Diego.

Hirsch, U.M., Edmonson, P., Muphey-Corb, M., Arbeille, B., Johnson, P.R. and Mullins, J.I. (1989). SIV adaptation to human cells. Nature 341, 573 574.

Ho, D.D., Neumann, A.U., Perelson, A.S., Chen, W., Leonard, J.M. and Markowitz, M. (1995). Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 373, 123 126.

Kim, K.J., Kanellopoulos-Iangevin, C., Merwin, R.M., Sachs, D.H. and Asofsky, R. (1979). Establishment and character- ization of Balb/c lymphoma lines with B cell properties. J. lmmunol. 122, 549-554.

Levy, J.A. (1993). Pathogenesis of human immunodeficiency virus infection. Microbiol. Rev. 57, 183-289.

Meyers, G., Wain-Hobson, S., Korber, B., Smith, R.F. and Paulakis, G.N. (1993). Human retroviruses and AIDS. A compilation and analyses of nucleic acid and amino acid sequences. Theoretical biology and byophysics group T-10. Los Alamos National Laboratory, Los Alamos, NM, USA.

Moore, J.P., Willey, R.L., Lewis, G.K., Robinson, J. and Sodroski, J. (1994) Immunological evidence for interactions between the first, second and fifth conserved domains of the gpl20 surface glycoprotein of human immunodeficiency virus type 1. J. Virol. 68, 6836-6847.

Paez, E., Dallo, S. and Esteban, M. (1985). Generation of a dominant 8 MDa deletion at the left terminus of vaccinia virus DNA. Proc. Natl. Acad. Sci. USA. 82, 3365-3369.

Paez, E., Dallo, S. and Esteban, M. (1987). Virus attenuation and identification of structural proteins of vaccinia virus that are selectively modified during virus persistence. J. Virol. 61, 2642-2647.

Pogo, B.G. and Friend, D. (1982). Persistent infection of Friend erythroleukemia cells with vaccinia virus. Proc. Natl. Acad. Sci. USA. 79, 4805-4809.

Pogo, B.G., Lai, A.C., Joesten, M.E., Royston, M.E. and Holloway, D. (1991). Changes in cell gene expression in human leukemic cells persistently infected with vaccinia virus. Virus Res. 19, 131-138.

Ritter, G.D., Mulligan, M., Lydy, S.L. and Compans, R.W. (1993). Cell fusion activity of the simian immunodeficiency virus envelope protein is modifed by the intracytoplasmic domain. Virology. 197, 255 264.

Rodriguez, D., Rodrlguez, J-R., Rodriguez, J.F., Trauber, D. and Esteban, M. (1989). Highly attenuated vaccinia virus mutants for the generation of safe recombinant viruses. Proc. Natl. Acad. Sci. USA. 86, 1287 1291.

Rodriguez, J.F., Paez, E. and Esteban, M. (1987). A 14000-M r envelope protein of vaccinia virus is involved in cell fusion and forms covalently linked trimers. J. Virol. 61,395-404.

Salk, J., Bretscher, P.A. and Salk, P.A. (1993). A strategy for prophylactic vaccination against HIV. Science. 260, 1270 1272.

Schawaller, M., Smith, G.E., Skehel, J.J. and Wiley, D.C. (1989). Studies with crosslinking reagents on the oligomeric structure of the env glycoprotein of HIV. Virology. 172, 367 369.

Smith, G.L. (1993). Vaccinia virus glycoproteins and immune evasion. J. Gen. Virol. 74, 1725 1740.

Stein, B.S. and Engleman, E.G. (1990). Intracellular process- ing of the gpl60 HIV-I envelope precursor. J. Biol. Chem. 265, 2640-2649.

Temin, H.O. (1993). Retrovirus variation and reverse tran- scription: abnormal strand transfers results in retrovirus genetic variation. Proc. Natl. Acad. Sci. USA. 90, 6900- 6903.

Wei, Y., Ghosh, S.K., Taylor, M.E., Johnson, U.A., Emini, E.A., Deutsh, E.P., Lifson, J.D., Bonhoeffer, S., Nowak, M.A., Hahn, B.H., Saag, M.S. and Shaw, G.M. (1995). Viral dynamics in human immunodeficiency virus type 1 infection. Nature 373, 117 122.

Weiss, C.D., Levy, J.A. and White, J.M. (1990). Oligomeric organization of gpl20 on infectious human immunodeficiency virus type 1-particles. J. Virol. 64, 5674 5677.

Willey, R.L., Bonifacino, J.S., Potts, B.J., Martin, M.A. and Klansner, R.D. (1988). Biosynthesis, cleavage and degrada- tion of the human immunodeficiency virus 1 envelope glycoprotein gpl60. Proc. Natl. Acad. Sci. USA. 85, 9580- 9584.

Zaides, Y., Yagello, M., Veselovskaya, T., Schmitt, D., Rykova, L., Fenouillet, E. and Gluckman, J.C. (1994). Extensive C-terminal deletion in human immunodefieiency virus type 1 env glycoprotein arising after long-term culture of chronically infected cells. J. Gen. Virol. 75, 2963-2975.

use of persistent infections with vaccinia virus recombinants to introduce alterations in foreign...

Documents