Passage of HIV-1 Molecular Clones into Different Cell Lines Confers Differential Sensitivity to Neutralization

VIROLOGY

238, 254–264 (1997) VY978812

ARTICLE NO.

Passage of HIV-1 Molecular Clones into Different Cell Lines Confers Differential Sensitivity to Neutralization Yi-jun Zhang,* Robert Fredriksson,* Jane A. McKeating,† and Eva Maria Fenyo¨*,1 *Microbiology and Tumorbiology Centre, Karolinska Institute, 171 77 Stockholm, Sweden; and †School of Animal and Microbial Sciences, University of Reading, Whiteknights, Reading, RG6 2AJ, United Kingdom Received February 27, 1997; returned to author for revision March 25, 1997; accepted September 2, 1997 In this study, progeny viruses of four HIV-1 molecular clones were tested for sensitivity to neutralization following prolonged passage in peripheral blood mononuclear cells (PBMC) and MT-2, H9, and CEM T-lymphoid cell lines. Two of the viruses were able to establish persistent infection with no cytopathic effect in H9 and CEM cells. Such adaptation conferred increased sensitivity to neutralization by a panel of human sera obtained from HIV-1-infected asymptomatic individuals, by soluble CD4 and by monoclonal antibodies directed to a linear epitope in the V3 region (268-D) and a conformational epitope in the CD4 binding site of the envelope gp120 (1.5e). Increased sensitivity to neutralizatiom was paralleled by increased binding of these mAbs to native envelope glycoproteins and by increased binding capacity to CD4 expressed on the cell surface. Our results show that virus–host cell interactions are important in influencing sensitivity to neutralization of HIV-1. In primary PBMC or in cytopathic interactions in cell lines, like in MT-2 cells, envelope epitopes important for neutralization remain masked. In contrast, noncytopathic but productive virus–host cell interactions may lead to an increased exposure of neutralizing epitopes and more efficient binding capacity to CD4 resulting in an increased sensitivity to neutralization. q 1997 Academic Press

INTRODUCTION

isolates thereby allowing selection of preexisting viral populations during in vitro passage, such that the underlying molecular mechanism could not be determined (Sawyer et al., 1994; Wrin et al., 1995). Host cell modification of replicating virus may be a key event in HIV pathogenesis and elucidating its mechanism may help understand the pathogenic process. We previously isolated a number of molecular clones from a primary HIV-1 isolate passaged in PBMC (Fredriksson et al., 1991; Tan et al., 1993). The parental virus isolate 4803 was able to infect cell lines and induce syncytia in PBMC and was therefore classified as rapid/ high phenotype (A˚sjo¨ et al., 1986; Fenyo¨ et al., 1988). Only one of seven molecular clones yielded progeny virus with properties similar to the parental isolate (Fredriksson et al., 1991; Tan et al., 1993). The remaining clones yielded viruses unable to replicate in the H9 and CEM cell lines and induce syncytia in PBMC cultures and were classified as slow/low. More recently (McKeating et al., 1996), when using the MT-2 cell line, all clones were similar to the parental isolate in that they infected and induced syncytia in MT-2 cells. However, these clonally derived viruses had characteristics similar to primary isolates in that they were resistant to neutralization by sCD4, human HIV-1 seropositive sera, and a panel of mAbs specific for epitopes including V2, V3, and multiple sites in the CD4 binding region (McKeating et al., 1996). These clonal viruses with ‘‘primary characteristics’’ provide a useful

The influence of host cell on the biological properties of human immunodeficiency virus type 1 (HIV-1) isolates when propagated in vitro has received much attention recently (Cohen, 1993; Moore et al., 1993, 1995). The sensitivity to neutralization by human sera and by ligands such as soluble CD4 (sCD4) and monoclonal antibodies (mAbs) specific for the envelope glycoproteins has been intensively investigated. In summary, there is a consensus that primary HIV-1 isolates, that is, isolates subjected to a few short passages in peripheral blood mononuclear cells (PBMC), show varying sensitivity to neutralization by sera of HIV-1-infected individuals (Moore et al., 1995; Kostrikis et al., 1996; Nyambi et al., 1996; Weber et al., 1996). Passage of virus isolates in established T cell lines selects for viruses which are sensitive to neutralization (Sawyer et al., 1994; Wrin et al., 1995). Several authors have suggested mechanisms to explain these observations such that mutations occur within the env gene during in vitro passage or that cell line dependent selection of the original quasispecies present within the primary isolate occurs (Turner et al., 1992; Moore et al., 1993, 1995; Mascola et al., 1994; Sullivan et al., 1995). Recent studies reported experiments using nonclonal primary 1 To whom correspondence and reprint requests should be addressed at MTC, Box 280, Karolinska Institute, 171 77 Stockholm, Sweden.

0042-6822/97 $25.00

254

Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID

VY 8812

/

6a53$$$281

10-28-97 14:04:05

viras

AP: VY

CELL LINE ADAPTATION AND NEUTRALIZATION OF HIV-1

model system for determining the effect of host cell on the biological properties of HIV-1.

MATERIALS AND METHODS Cells and media Peripheral blood mononuclear cells from seronegative blood donors were prepared by Ficoll density gradient separation, activated with 2.5 mg/ml of phytohemagglutinin (PHA-P; Gibco) for 3 days in RPMI 1640 medium (Gibco) containing 10% inactivated fetal calf serum (FCS) and antibiotics (RPMI 10%). For virus passage, PHA-activated PBMC were cultured in RPMI 10% supplemented with 10 units of interleukin 2 (IL-2)/ml (Amersham) and 2 mg/ml polybrene. HeLa cells were cultured in DMEM with 10% FCS and antibiotics. The CD4-positive T cell lines H9, CEM, and MT-2 were cultured in RPMI 10% (Koot et al., 1992). HIV-1 molecular clones and their passage/adaptation in different cells Five micrograms of DNA of each of the four HIV-1 molecular clones (cl13, -31, -32, and -82) (Fredriksson et al., 1991, Tan et al., 1993) was transfected into subconfluent monolayers of HeLa cells in 12-well plates (Costar) using the lipofectamine (Gibco) technique as described previously (McKeating et al., 1996). Forty-eight hours after transfection, HeLa cells were cocultivated with 3 1 106 activated PBMC or 2 1 106 MT-2 cells for 24 h. PBMC and MT-2 cells were then recovered, washed, and further cultured. At day 3 freshly activated PBMC and uninfected MT-2 cells were added. Supernatants from both PBMC and MT-2 cultures were collected when p24 antigen values were over 2 ng/ml and used as first-passage virus stocks. One-milliliter aliquots of these virus stocks were used to infect PBMC and MT-2 cells, respectively, for an additional two passages. Supernatants collected at third passage level are designated ‘‘early passage virus.’’ Twomilliliter aliquots of the first-passage virus stocks produced in PBMC were used to infect 6 1 106 PBMC, followed by cocultivation with 3 1 106 H9 or CEM cells at day 7. Cocultures were split twice a week to maintain a cell concentration of 1 1 106 cells/ml. Early passage virus was collected from these cultures after 3 weeks to match the early passage virus from PBMC and MT-2 cells. Subsequently, all cultures were maintained for an additional 8 weeks to conform with the study of Wrin and colleagues (1995). Uninfected PBMC and MT-2 cells had to be added twice a week to PBMC and MT-2 cultures, respectively. At the end of this period, culture medium was changed 2 days before collection of supernatants, termed ‘‘late passage virus.’’ All virus stocks were subjected to low speed centrifugation and frozen at 0707.

AID

VY 8812

/

6a53$$$281

10-28-97 14:04:05

255

Human sera, monoclonal antibodies, and recombinant gp120 protein Human sera were collected from HIV-1-infected asymptomatic Swedish patients previously reported (McKeating et al., 1996). Neutralizing mAbs studied include 2.1H, 1.5e, and 39.13g specific for conformationdependent gp120 epitopes overlapping the CD4 binding site (Cordell et al., 1991; Ho et al., 1991); 268-D recognizing a linear epitope within the V3 region (HIGPGR) (Gorny et al., 1989); and mAb 2G12 specific for a conformationdependent gp120 epitope (kindly provided by Dr. H. Katinger) (Buchacher et al., 1994; Trkola et al., 1996). Polyclonal rabbit serum against recombinant gp120 protein of the HIV-1MN strain was kindly provided by Dr. Phillip W. Berman (Genentech, Inc.). Affinity-purified sheep polyclonal antibody (D7324) recognizing a C-terminal of gp120 epitope was purchased from Aalto Bioreagents (Dublin, Ireland). Soluble CD4 (sCD4) and purified recombinant gp120 protein (HIV-1SF2 strain) produced in CHO cells were obtained through the MRC AIDS Reagent Project (NIBSC, Potters Bar, Hertfordshire, UK). Infectious dose-50 (ID50) titration and neutralization ID50 titration. Seventy-five-milliliter aliquots of virus supernatants were diluted in medium with fivefold, eight steps dilutions in 96-well round-bottom plates (Costar) as described previously (Zhang et al., 1994). Thereafter, an additional 75 ml of medium was added into each well and incubated at 37C7 for 1 h. Subsequently, 105 PHA stimulated PBMC from two donors or, if cell line was used, 2 1 104 cells from each cell line were added to each well. The plates were washed at days 1 and 3, respectively, by centrifugation and medium change. At day 7, 100 ml of culture supernatant was analyzed in an in-house HIV-1 p24 antigen capture ELISA (Sundqvist et al., 1989). The ID50 was defined as the reciprocal of the virus dilutions resulting in 50% of the positive wells for virus p24 antigen production (Reed-Muench calculation). Neutralization assay. Sera were inactivated at 567 for 30 min and subsequently diluted in medium in twofold dilution steps, starting from 1:10. Thirty ID50 of pretitrated virus supernatants in 75 ml were added into each well and incubated at 377 for 1 h and cells were added. Plates were washed as mentioned above and at day 7, 100 ml of supernatants was tested in an in-house HIV-1 p24 antigen capture ELISA. Neutralizing titers were defined as the reciprocal of the highest dilutions of serum, mAb, or sCD4 giving an absorbance value (mean of duplicate wells) below the cutoff value for HIV-1 p24 antigen capture ELISA assay. In each plate, five wells with virus only were included to check the efficiency of washing, five wells with cells only were included as negative control, and five wells with both virus and cells were included as positive controls. The cutoff absorbance value for neu-

viras

AP: VY

256

ZHANG ET AL.

tralization was 0.2–0.25 above the background which corresponds to more than 90% reduction in viral antigen compared with positive virus control. Neutralization assays were not evaluated if the mean absorbance values of the positive controls were less than five times the cutoff of the HIV-1 p24 antigen capture ELISA. Quantitation of gp120 envelope protein in virus stocks Sheep polyclonal antibody D7324 recognizing a conserved epitope at the COOH terminus of gp120 (Aalto Bioreagents) was coated onto 96-well plates (Nunc, Inter Med) at a concentration of 50 ng/ml at 47 overnight. Plates were washed six times with PBS buffer containing 0.5 M NaCl and 0.1% Tween 20 (Sigma) and blocked with 1% BSA (Sigma) in PBS for 2 h at 377. After being washed, 100 ml supernatant of virus stocks was added to the wells and incubated at 377 for 1.5 h. Recombinant gp120 protein was diluted in PBS and used for the standard curve. Rabbit anti-HIV-1MN gp120 serum diluted at 1:1000 in PBS containing 20% sheep serum (Sigma) was added followed by a sheep anti-rabbit horseradish peroxidaseconjugated antibody (Sigma). Assays were developed by addition of substrate O-phenylenediamine dihydrochloride (OPD) (Sigma) and the absorbance measured at 490 nm. The gp120 protein concentrations were calculated by comparison with the standard curve and expressed in nanograms per milliliter. MAb binding assay The binding of mAbs to the native form of viral gp120 proteins was measured by ELISA as described above. In order to use the native form of the envelope protein, virus stocks (without inactivation by detergent) were diluted according to the pretitrated gp120 concentration and a saturating amount of gp120 (0.7 mg/ml) was added to the plates precoated with the D7324 antibody. To ensure that saturating amounts of gp120 were used, unbound gp120 was determined after absorption. The amount of excess gp120 was similar for all wells (about 0.05 mg/ml) and residual infectious virus was barely detected (ID50 on MT-2 cells õ501). After incubation, and washing of excess unbound gp120, increasing concentrations of mAbs were added to duplicate wells. Bound mAbs were visualized using an anti-human IgG monoclonal antibody conjugated to horseradish peroxidase and OPD substrate.

TABLE 1 Replicative and Syncytium-Inducing Capacity of Early Passage Viruses Passage in PBMC

MT-2 cells

H9 cells

CEM cells

Virus

Repla

Syncb

Repl

Sync

Repl

Sync

Repl

Sync

CL13 CL31 CL32 CL82

/ / / /

0 0 / 0

/ / / /

// // // //

0 0 / /

0 0 / 0

0 0 / 0

0 0 / 0

a

Repl, replication: 0, õ0.2 ng/ml p24 antigen production over a period of 3 weeks; /, three sequential increasing values (0.2–ú2 ng/ ml) of p24 antigen. b Sync, syncytia: 0, no apparent syncytia formation; /, large syncytia apparent in every field; //, extensive syncytia formation in every field, uninfected cells had to be added to keep the culture alive.

cells were incubated with 200 ml of FITC-conjugated (1:20) rabbit anti-human IgG (g-chains), followed by washing, and finally fixed by 1% paraformaldehyde in PBS. Cell populations were analyzed by FACScan and the data were processed by using CellQuest software program (Becton Dickinson, U.S.A.). Nucleotide sequencing Late passage virus from MT-2 and H9 cells was used to infect MT-2 and H9 cells, respectively. DNA samples were prepared from 2 1 106 infected cells, PCR amplified with nested primers (outer primer set: JA9 5*CAC AGT ACA ATG TAC ACA TG 3*; JA12 5*ACA GTA GAA AAA TTC CCC TC3*. inner nested primer set: JA10 5*AAA TGG CAG TCT AGC AGA AG 3*; JA53 5*AAT TTC TGG GTC CCC TCC TG 3*) specific for the V3 region of the envelope gene (Albert and Fenyo¨, 1990). Amplified PCR products were cloned into pBluescript plasmid (Stratagene) and the V3 region of one clone from each cell line was sequenced by using the T7 sequencing kit (Pharmacia). RESULTS Biological properties of viruses recovered after shortterm and long-term passage in PBMC and cell lines

Aliquots of virus stocks (0.4 ml) containing similar amounts of infectious virus (503.6 and 503.5 ID50 for MT-2 and H9-passaged virus, respectively) were incubated with 0.25 1 106 MT-2 cells for 0, 10, 30, and 60 min at 47. The cells were washed two times with cold RPMI 1640 medium and incubated with 200 ml of monoclonal antibody 268-D or 1.5e at 10 mg/ml. After being washed,

PBMC and MT-2 cells could support replication of all four clonally derived viruses. Infection was first detected 6 days after cocultivation with transfected HeLa cells and extracellular p24 antigen was greater than 2 ng/ml by days 12 and 8, respectively. In subsequent passages, supernatants were regularly collected 7 days postinfection. MT-2 cultures showed syncytia formation leading to extensive cell death within several days such that uninfected MT-2 cells had to be added to maintain culture viability (Table 1). This pattern of cytopathogenicity

AID

viras

Detection of virus binding to MT-2 cells

VY 8812

/

6a53$$$281

10-28-97 14:04:05

AP: VY

CELL LINE ADAPTATION AND NEUTRALIZATION OF HIV-1 TABLE 2 Replicative Capacity of Late Passage cl32 in Different Host Cells Passage in Titration in

PBMC

MT-2

H9

CEM

PBMC MT-2 H9

2.4a 2.6 2.3

3.1 3.4 3.1

0.2 3.3 2.7

2.6 2.4 0.4

a

Log 5 ID50 titers.

did not change over time, in that, even after 3 months of passage, all four clonally derived viruses from infected MT-2 cell cultures induced syncytia in MT-2 cells. Using other cell systems, clone 32 could be distinguished from the other clones in that only cl32 was able to induce large syncytia in PBMC cultures and to productively infect CEM and H9 cells (Table 1). Syncytia were observed in cl32-infected H9 and CEM cells 10 days after cocultivation with infected PBMC but disappeared within 3 weeks thereafter. Interestingly, these cultures survived without the addition of uninfected cells. Late passage cl32 harvested from H9 cells was, however, still cytopathic for both H9 and MT-2 cells. Both cultures produced similar amounts of infectious virus (53.5 and 53.6 ID50 , respectively) but supernatants from H9 cultures had a 15.5-fold higher gp120 concentration indicating that H9 cultures produced more infectious particles or excess gp120 than

257

MT-2 cultures. When tested on PBMC, late passage cl32 from H9 and CEM cells had a reduced replicative capacity. In addition, virus harvested from CEM cells had a reduced replicative capacity in H9 cells (Table 2). These data show that progeny virus from closely related clones may behave differently on different cell lines with regard to the capacity to replicate and induce syncytia. Neutralization of early passage viruses Viruses derived from the transfection of HeLa cells were passaged three times in PBMC and MT-2 cells and the extracellular virus collected as early passage virus stocks. These ‘‘early passage’’ viruses were tested for sensitivity to neutralization by sera from 10 HIV-1-infected asymptomatic individuals, the human mAb 2G12, and sCD4 using either the MT-2 cell line or PBMC as target cells. Virus passaged either in PBMC or in MT-2 cells was resistant to neutralization by sCD4 even when amounts up to 100 mg/ml were used (Table 3). All viruses were neutralized by mAb 2G12 at a final concentration of 25 mg/ml. Eight of ten human sera showed low titer neutralizing activity against these viruses. MT-2 or PBMC passaged viruses showed similar sensitivity to neutralization, with slight variation in titres between 1:20 and 1:80. When PBMC were used as targets, serum neutralizing titers were similar to those obtained for MT-2 cells (Table 4). It should be noted that ID50 titers of virus stocks showed similar ranges of variation for infection of PBMC and MT-2 cells. Notably, the ID50 titer range for PBMC

TABLE 3 Sensitivity of Early Passage Viruses to Neutralization in MT-2 Target Cellsa Passage in PBMC Neutralization with HIV-1/ human sera 1 3 4 5 6 7 8 9 10 11 Ligands 2G12 (25 mg/ml) sCD4c (100 mg/ml)

MT-2 cells

CL13

CL31

CL32

CL82

CL13

CL31

CL32

CL82

20b 0 20 0 0 20 0 80 0 0

20 20 20 0 0 20 0 0 0 20

80 80 20 0 0 80 0 20 0 80

20 20 20 0 0 0 0 0 0 20

0 0 0 0 0 0 0 20 0 0

20 80 80 0 0 20 0 0 0 20

20 80 20 0 20 20 0 0 0 20

20 20 0 20 0 0 0 0 0 20

/ 0

/ 0

/ 0

/ 0

/ 0

/ 0

/ 0

/ 0

a

Supernatant harvested at third passage level. Viruses were titrated on MT-2 cells and 30 ID50 was used in the neutralization assay. The results of one of two experiments are shown. b Reciprocal neutralizing titers; 0, õ20. c The laboratory strains HIV-1MN and HIV-1IIIB were readily neutralized with 1–5 mg/ml sCD4.

AID

VY 8812

/

6a53$$$281

10-28-97 14:04:05

viras

AP: VY

258

ZHANG ET AL. TABLE 4 Sensitivity of Early Passage Viruses to Neutralization in PBMC Target Cellsa Passage in PBMC

Neutralization with HIV-1/ human sera 1 3 4 5 6 7 8 9 10 11 Ligands 2G12 (25 mg/ml) sCD4c (100 mg/ml)

MT-2 cells

H9 cells

CL13

CL31

CL32

CL82

CL13

CL31

CL32

CL82

CL32

20b 20 0 0 20 0 0 80 0 20

20 20 80 0 0 0 0 0 0 20

80 80 20 20 80 20 0 20 0 20

20 80 20 0 0 20 0 0 0 20

20 0 20 0 0 20 0 80 0 0

20 80 80 20 0 20 0 0 0 20

80 80 80 20 80 20 0 0 0 20

20 20 20 0 0 0 0 0 0 20

80 80 40 0 40 0 0 20 0 20

/ 0

/ 0

/ 0

/ 0

/ 0

/ 0

/ 0

/ 0

/ 50

a Supernatant harvested at third passage level. Viruses were titrated on PBMC and 30 ID50 was used in the neutralization assay. The results of one of two experiments are shown. b Reciprocal neutralizing titers; 0, õ20. c The laboratory strains HIV-1MN and HIV-1IIIB were readily neutralized by 1–5 mg/ml sCD4.

passaged viruses was 503.2 to 504.4 and 502.6 to 504.4, whereas for MT-2 passaged viruses it was 503.6 to 504.8 and 503.2 to 505.2 when tested in PBMC and MT-2 cells, respectively. These data show that three passages, regardless of cell type, do not consistently alter the replicative capacity of these clonally derived viruses and their sensitivity to neutralization by human sera or sCD4. Furthermore, use of PBMC or the MT-2 cells as targets for the neutralization assay does not influence the outcome of neutralization. Neutralization of late passage viruses In these experiments, viruses were carried for 3 months in PBMC, MT-2, H9, and CEM cells, respectively. Late passage cl32 from H9 cells was neutralized at serum titers up to 100-fold higher and at sCD4 concentrations up to 70-fold lower than early passage cl32 (Table 5). This increased sensitivity to neutralization could be detected regardless of the target cell used, MT-2 or H9, in the neutralization assay. That long-term passage indeed increased sensitivity to neutralization was further confirmed by passage of cl82 in H9 cells and cl32 in CEM cells. However, in the case of CEM passaged cl32, the target cells used in the neutralization assay appeared to play a decisive role in determining the neutralization titers of sera, in that titers were up to 64-fold higher in PBMC than in MT-2 target cells. This difference may be explained by the fact that CEM passaged cl32 was more infectious for PBMC than for MT-2 cells, the difference

AID

VY 8812

/

6a53$$$281

10-28-97 14:04:05

being nearly 25-fold. Clearly, much less virus was used in the PBMC assay than in the MT-2 assay and this could have resulted in higher serum neutralizing titers. In contrast, an increased sensitivity to neutralization could not be observed with MT-2 passaged viruses and their resistance to neutralization by sCD4 was maintained. Similarly, long-term PBMC passage did not change the sensitivity of cl32 to neutralization. We previously reported that cl32 and cl82 could not be neutralized by a panel of mAbs to V2, V3, and the CD4 binding site (CD4bs) of the HIV-1 envelope gp120 (McKeating et al., 1996). We selected a number of these mAbs to evaluate their ability to neutralize late passage virus. Both cl32 and cl82 passaged in H9 cells were sensitive to neutralization by mAbs 268-D, 1.5e, and 39.13g but not by 2.1H (Table 6). The results indicate that long-term passage of clonally derived viruses in the H9 cell line results in variants with increased sensitivity to neutralization by mAbs to distinct nonoverlapping epitopes. Such increased sensitivity to neutralization is cell type dependent and does not occur per se upon passage in all established cell lines. Monoclonal antibody binding to the native form of envelope glycoproteins derived from late passage viruses In order to investigate the possible mechanism responsible for the differential sensitivity of late passage viruses, the binding capacity of human mAbs 268-D and

viras

AP: VY

CELL LINE ADAPTATION AND NEUTRALIZATION OF HIV-1

259

TABLE 5 Sensitivity to Neutralization of Late Passage Virusesa Target cells used in neutralization MT-2 cells Neutralization with HIV-1/ human sera 1 3 4 5 6 7 9 11 Ligands 2G12 (25 mg/ml) sCD4c (mg/ml) Log 5 ID50 titers

H9 cells

PBMC

32P

13M

31M

32M

82M

32C

32H

82H

32H

32C

20b 80 40 0 40 80 0 40

0 0 0 0 0 0 20 0

0 20 0 0 0 0 0 0

0 20 0 0 0 0 0 20

0 20 0 0 0 0 0 0

80 160 160 40 80 320 20 320

1280 2560 1280 640 1280 640 80 640

640 640 1280 160 640 320 80 640

5120 640 2560 1280 2560 1280 80 1280

2560 5120 5120 1280 1280 5120 1280 1280

/ 50 2.1

/ 0 3.3

/ 0 3.1

/ 100 3.4

/ 100 3.3

/ 12 2.4

/

/ 1.5 2.3

/ 1.5 2.7

/

3 3.3

6 4.3

a Late passage viruses were obtained from PBMC (P), MT-2 (M), CEM (C), and H9 (H) cells as described in the text. 30 ID50 of virus was used in neutralization based on ID50 titration in each individual target cell. The results of one of two experiments are shown. b Reciprocal neutralizing titers; 0, õ20; repeated experiment with 32H and 32M in serum neutralization resulted similar titers. c The laboratory strains HIV-1MN and HIV-1IIIB were readily neutralized by 1–5 mg/ml sCD4.

1.5e showing differential neutralizing capacity was measured against native viral envelope proteins derived from MT-2 and H9 cells. For the sake of comparison, constant saturating amounts of gp120 protein were used. An increased binding capacity was observed with mAbs 268D and 1.5e on gp120 derived from H9 cells compared to gp120 derived from MT-2 cells (Fig. 1). The increased ability of these mAbs to recognize H9-derived gp120 appeared to be consistent with their capacity to neutralize the H9-adapted viruses. TABLE 6 Sensitivity of Late Passage Viruses to Neutralization by Monoclonal Antibodiesa Passage in H9 cells

MT-2 cells

Monoclonal antibody

CL32

CL82

CL32

CL82

268-Db 1.5eb 2.1Hb 2G12b 39.13gc

1.5 3 ú50 1.5 160

3 0.3 ú50 3 1280

100 ú50 ú50 3 õ40

ú100 ú50 ú50 12 õ40

Virus binding to CD4 expressed on the surface of MT-2 cells Using the mAb 268-D, binding of late passage virus from H9 cells showed a time-dependent increase, whereas no binding could be detected with late passage virus from MT-2 cells, even after 60 min of incubation (Figs. 2A and 2B, respectively). As expected, cell surface bound virus could not be detected by mAb 1.5e, since the CD4 binding site on gp120 was blocked and not available to the antibody (data not shown). To test whether the binding detected with mAb 268-D reflected the binding of virus-associated gp120, MT-2 cells were lysed at the 60 min time point with 0.5% Triton X-100 and the p24 and gp120 antigen contents measured. Cell lysates incubated with late passage virus from H9 cells contained 2 ng p24/ml, whereas cell lysates treated with virus from MT-2 cells contained 0.24 ng p24/ml. Similarly, there was a 14-fold difference in gp120 content of cell lysates. The results indicate that our assay indeed detected binding of virus particles (and not only free gp120) and that late passage virus from MT-2 cells binds less efficiently to CD4 present on the cell surface than virus passaged in H9 cells. Thus passage in H9 cells increases the binding capacity to CD4 and renders the virus sensitive to neutralization by sCD4 as well as some mAbs.

a

Late passage viruses were titrated on MT-2 cells and 30ID-50 of virus was used in the neutralization assay. b Figures denote antibody concentrations (mg/ml) needed for virus neutralization. c Figures denote reciprocal serum dilutions needed for virus neutralization.

AID

VY 8812

/

6a53$$$281

10-28-97 14:04:05

Nucleotide sequencing of HIV-1 V3 loop derived from passage in different cell types Since monoclonal antibody 268-D recognizing a linear epitope in the V3 region showed differential neutralizing

viras

AP: VY

260

ZHANG ET AL.

FIG. 1. Mabs 268D (A) and 1.5e (B) binding to native gp120 of late passage viruses CL32 and CL82 derived from passage in MT-2 (M) and H9 (H) cells, respectively. Binding activity was expressed as absorbance values obtained at different antibody concentrations. The absorbance values represent the mean of two experiments.

capacity to clones derived from passage in MT-2 and H9 cells, the V3 loop including the epitope (HIGPGR) was sequenced. With the exception of cl82/MT-2 (G/K) which remained resistant to neutralization, no amino acid changes were observed in the V3 region from cl32 and cl82 virus passaged in either H9 or MT-2 cells (Fig. 3). These data suggest either that changes outside the region sequenced were responsible for the observed phenotype or that factors other than the primary sequence were involved in conferring altered sensitivity to neutralization.

In the present study we show that sensitivity of HIV-1 to neutralization is dependent on the host cell in which

the virus stocks are prepared. Using progeny viruses of molecular clones we show that the virus–host cell interaction may result in a balance between virus replication and host cell survival. Cl32 in H9 and CEM cells and cl82 in H9 cells were able to undergo such an adaptation after prolonged time of passage and the viruses showed an altered cell tropism concomitant with an increased sensitivity to neutralization by human sera and sCD4. The results also show that adaptation confered an increased binding capacity to CD4. However, prolonged passage alone without adaptation, like in MT-2 cells or PBMC, did not lead to an increased sensitivity to neutralization. Likewise, short-time in vitro passage of clonally derived viruses in various T cell lines did not change the pattern of virus neutralization.

AID

viras

DISCUSSION

VY 8812

/

6a53$$$281

10-28-97 14:04:05

AP: VY

CELL LINE ADAPTATION AND NEUTRALIZATION OF HIV-1

261

FIG. 2. Histogram overlay of cells incubated with medium alone or with late passage cl32 from H9 cells (A and C) or MT-2 cells (B and D) at time points indicated. The detecting antibodies 268-D (A and B) or 1.5e (C and D) were used at 10 mg/ml.

AID

VY 8812

/

6a53$$8812

10-28-97 14:04:05

viras

AP: VY

262

ZHANG ET AL.

FIG. 3. Deduced amino acid sequences of V3 region from late passage viruses derived from passage in MT-2 (CL32/MT-2, CL82/MT-2) and H9 (CL32/H9, CL82/H9) cells, respectively. Sequences are aligned to the sequence of CL32 not subjected to in vitro passage. Identity is indicated by a dash; departures are indicated by the single letter amino acid codes.

Sawyer and colleagues reported that three passages of primary HIV-1 isolates in H9 cells rendered the viruses sensitive to neutralization by human sera (Sawyer et al., 1994). In contrast, Wrin and colleagues (1995) reported that such a phenotypic change only occurred after prolonged passage and adaptation. Such differences may reflect differential selection and adaptation of different primary isolates in the H9 cell line. It is also interesting to note that in the latter study, changes in neutralization were coincident with reduced cytopathogenicity of the late passage virus in H9 cells. However, our data show that late passage cl32 from H9 and CEM cells was as cytopathogenic in MT-2 cells as early passage virus. Nevertheless, there was a difference in cell tropism, in that late passage cl32 showed a reduced replicative capacity in PBMC and H9 cells, respectively (Table 2). Our data are in line with previous observations that passage in different cell lines may exert a differential influence on the biological properties of viruses (Cheng-Mayer et al., 1991; Peden et al., 1991). Passage of the HIV-1LAI molecular clone in SupT1 cells conferred a more restricted cell tropism, whereas passage of the HIV-1SF2 molecular clone in HUT78 cells resulted in a broad host cell tropism (Cheng-Mayer et al., 1991; Peden et al., 1991). We now further extend these observations and show that passage and adaptation of HIV-1 molecular clones in different host cells may affect not only their replicative capacity but also their sensitivity to neutralization (Table 5). We found that H9-adapted viruses showed an increased sensitivity to neutralization and this was associated with an increased binding capacity of mAbs 1.5e and 268-D directed to both linear and conformational epitopes within gp120 (Table 6). Since our binding assay measures, at least in part, the binding activity to oligomeric envelope gp120, our results are in line with those of others showing a correlation between the ability of antibody to bind the oligomeric envelope glycoprotein at the cell surface and their ability to neutralize viral infectivity (Fig. 1) (Sattentau and Moore, 1995; Sullivan et al., 1995; Fouts et al., 1997). It is also interesting to note that antibody 2.1H, which showed no differential binding, also failed to neutralize the clones (Table 6). Together, these

data indicate that binding to the oligomeric form of the envelope protein might be used as a marker for anti-HIV1 antibodies with potential neutralizing activity (Parren et al., 1996). Furthermore, the increased binding of 1.5e and 268-D antibodies to the late passage viral envelope gp120 suggests that the V3 loop as well as part of the CD4bs was masked in early passage viruses. However, adaptation in T cell lines, such as in H9 cells, allows increased exposure of neutralizing epitopes rendering the virus more sensitive to neutralization. Whether such an alteration could also explain the differential sensitivity to antibody neutralization against HIV-1 primary isolates remains to be investigated (Kostrikis et al., 1996; Nyambi et al., 1996; Weber et al., 1996). Several mechanisms underlying the increased exposure of neutralizing epitopes in association with T cell line adaptation of a virus have been considered. In one study, increased sensitivity to neutralization of JR-CSF virus was associated with a single amino acid change in the V3 region of the envelope. The change resulted in increased accessibility of the epitope within the oligomeric complex (Bou-Habib et al., 1994). Interestingly, exposure of epitopes in the V3 loop could be further enhanced in T cell line adapted viruses through binding of sCD4 and a comformational change of envelope gp120 has been proposed to be responsible for this process (Stamatatos and Cheng-Mayer, 1995; Sullivan et al., 1995). In our experiments, however, the amino acid sequence of the linear epitope within the V3 loop recognized by mAb 268-D remained unchanged (Fig. 2). Moreover, it is unlikely that amino acid substitutions in regions outside the binding site of mAb 268-D change the conformation of envelope gp120, since similar amounts of the mAb 2G12, directed against a conformational epitope in gp120 other than the CD4bs, neutralized both H9 and MT-2 passaged viruses (Table 6). Another possibility might be the altered glycosylation of envelope proteins following passage and adaptation in cell lines. Different cell types or individual cell lines may show different glycosylation patterns (Goochee and Monica, 1990) and may conceivably result in viruses with altered envelope properties. Indeed, glycosylation of HIV-1 envelope pro-

AID

viras

VY 8812

/

6a53$$$281

10-28-97 14:04:05

AP: VY

CELL LINE ADAPTATION AND NEUTRALIZATION OF HIV-1

tein has been shown to influence both the virus infectivity and immunogenicity (Benjouad et al., 1992; Lee et al., 1992). A recent report suggested that differential envelope glycosylation was responsible for the observed differences in neutralization sensitivity of viruses derived from macrophages and PBMC (Willey et al., 1996). Similarly, propagation of HIV-1 in T cell lines or in vitro selection by neutralizing mAbs may induce changes in envelope glycosylation that may affect both the tropism and the neutralization sensitivity (Cheng-Mayer et al., 1991; Back et al., 1994; Schonning et al., 1996). However, the potential glycosylation site in the portion of the envelope (V3 region) that has been sequenced in our experiments does not show changes in the viruses tested (Fig. 2). In addition, long-term passage and adaptation of viruses was required to bring about the change in sensitivity to neutralization (Tables 5 and 6). It is therefore unlikely that an altered glycosylation pattern was responsible for the changes observed in our experiments. Our results indicate that adaptation of HIV-1 to growth in the H9 (or CEM) cell line increases binding affinity of the virus to the CD4 molecules. This is in line with a previous report (Kabat et al., 1994) demonstrating differences in CD4 dependence for infectivity of laboratoryadapted and primary isolates. Furthermore, we suggest that increased binding affinity to cell surface expressed CD4 is in turn resposible for the increased sensitivity to neutralization by antibodies and sCD4. This is in line with the observation of Sullivan and colleagues (1995) in that the sensitivity of chimeric HIV-1 to neutralization by mAbs or sCD4 could be predicted by assays dependent on the binding of the inhibitory molecule to the oligomeric envelope glycoprotein complex. This observation has recently been extended to the primary HIV-1 isolate JR-FL. The target cells used in the neutralization assay may also influence the outcome of neutralization. Cl32 adapted in CEM cells appeared to be more sensitive to neutralization on PBMC than on MT-2 target cells. This may be explained by differences in infectious virus titers in PBMC and MT-2 cells (ID50 titers 504.3 and 52.4, respectively). In order to obtain 30 ID50 a higher virus dilution was used on PBMC target cells; consequently, the actual amount of input virus was much lower than on MT-2 cells. Effect of the target cell on neutralization of feline immunodeficiency virus (FIV) has recently been described (Baldinotti et al., 1994). Since the susceptibility to FIV infection of the two cell lines used was not tested, it is difficult to compare with our experiments. In conclusion, our data demonstrate the importance of the host cell in determining sensitivity to neutralization of HIV-1. Adaptation in certain cell lines such as H9 and CEM can dramatically increase sensitivity to neutralization as a result of increased exposure of neutralizing epitopes and increased binding affinity to cell surface expressed CD4. In different virus–host cell interactions,

AID

VY 8812

/

6a53$$$281

10-28-97 14:04:05

263

like in PBMC and MT-2 cells where no adaptation occurs, neutralizing epitopes in the viral envelope remain masked. It is intriguing that virus–host cell interactions appear to have a decisive role in determining sensitivity to neutralization. When designing neutralization assays for use in vaccine studies, the viral phenotype, as reflected by the replicative capacity and cytopathogenicity of HIV-1 isolates should be taken into consideration in relation to the cell type used. ACKNOWLEDGMENTS The expert technical assistance of Kerstin Andreasson is gratefully acknowledged. This work was supported by grants from the Swedish Medical Research Council; the Swedish Board for Technical and Industrial Development (NUTEK) and the Concerted Action for HIV Variability.

REFERENCES Albert, J., and Fenyo¨, E. M. (1990). Simple, sensitive and specific detection of HIV-1 in clinical specimens by polymerase chain reaction with nested primers. J. Clin. Microbiol. 28, 1560–1564. A˚sjo¨, B., Morfeldt-Ma˚nson, L., Albert, J., Biberfeld, G., Karlsson, A., Lidman, K., and Fenyo¨, E. M. (1986). Replicative capacity of human immunodeficiency virus from patients with varying severity of HIV infection. Lancet ii, 660–662. Back, N. K. T., Smit, L., De Jong, J. J., Keulen, W., Schutten, M., Goudsmit, J., and Tersmette, M. (1994). An N-glycan within the human immunodeficiency virus type 1 gp120 V3 loop affects virus neutralization. Virology 199, 431–438. Baldinotti, F., Matteucci, D., Mazzetti, P., Giannelli, C., Bandecchi, P., Tozzini, F., and Bendinelli, M. (1994). Neutralization of feline immunodeficiency virus is markedly dependent on passage history of the virus and host system. J. Virol. 68, 4572–4579. Benjouad, A., Gluckman, J. C., Rochat, H., Montagnier, L., and Bahraoui, E. (1992). Influence of carbohydrate moieties on the immunogenicity of human immunodeficiency cirus type 1 recombinant gp160. J. Virol. 66, 2473–2483. Bou-Habib, D. C., Roderiquez, G., Oravecz, T., Berman, P. W., Lusso, P., and Norcross, M. A. (1994). Cryptic nature of envelope V3 region epitopes protects primary monocytotropic human immunodeficiency virus type 1 from antibody neutralization. J. Virol. 68, 6006–6013. Buchacher, A., Predl, R., Strutzenberger, K., Steinfellner, W., Trkola, A., Purtscher, M., Gruber, G., Tauer, C., Steinndl, F., Jungbauer, A., et al. (1994). Electrofusion and EBV transformation for PBL immortalisation: generation of human monoclonal antibodies against HIV-1 proteins. AIDS Res. Hum. Retroviruses 10, 359–369. Cheng-Mayer, C., Seto, D., and Levy, J. A. (1991). Altered host range of HIV-1 after passage through various human cell types. Virology 181, 288–294. Cohen, J. (1993). Jitters jeopardize AIDS vaccine trials. Science 262, 980–981. Cordell, J., Moore, J. P., Dean, C. F., Klasse, P. J., Weiss, R. A., and McKeating, J. A. (1991). Rat monoclonal antibodies to non-overlapping epitopes of human immunodeficiency virus type 1 gp120 block CD4 binding in vitro. Virology 185, 72–79. Fenyo¨, E. M., Morfeldt-Ma˚nson, L., Chiodi, F., Lind, B., von Gegerfelt, A., Albert, J., Olausson, E., and A˚sjo¨, B. (1988). Distinct replicative and cytopathic characteristics of human immunodeficiency virus isolates. J. Virol. 62, 4414–4419. Fouts, T. R., Binley, J. M., Trkola, A., Robinson, J. E., and Moore, J. P. (1997). Neutralization of the human immunodeficiency virus type 1 primary isolate JR-FL by human monoclonal antibodies correlates

viras

AP: VY

264

ZHANG ET AL.

with antibody binding to the oligomeric form of the envelope glycoprotein complex. J. Virol. 71, 2779–2785. Fredriksson, R., Sta˚lhandske, P., von Gegerfelt, A., Lind, B., A˚man, P., Rassart, E., and Fenyo¨, E. (1991). Biological characterization of infectious molecular clones derived from a human immunodeficiency virus type1 isolate with rapid/high replicative capacity. Virology 181, 55–61. Goochee, C. F., and Monica, T. (1990). Environmental effects on protein glycosylation. Bio/Technology 8, 421–427. Gorny, M. K., Xu, J. Y., Gianakakos, V., and Zolla-Pazner, S. (1989). Generation of human monoclonal antibodies to human immunodeficiency virus. Proc. Natl. Acad. Sci. USA 86, 1624–1628. Ho, D., McKeating, J., Li, X., Moudgil, T., Daar, E., Sun, N., and Robinson, J. (1991). Conformational epitope on gp120 important in CD4 binding and human immunodeficiency virus type 1 neutralization identified by a human monoclonal antibody. J. Virol. 65, 489–493. Kabat, D., Kozak, S. L., Wehrly, K., and Chesebro, B. (1994). Differences in CD4 dependence for infectivity of laboratory-adapted and primary patient isolates of humam immunodeficiency virus type 1. J. Virol. 68, 2570–2577. Koot, M., Vos, A. H. V., Keet, R. P. M., de Goede, R. E. Y., Dercksen, W., Terpstra, F. G., Coutinho, R. A., Miedema, F., and Tersmette, M. (1992). HIV-1 biological phenotype in long-term infected individuals evaluated with an MT-2 cocultivation assay. AIDS 6, 49–54. Kostrikis, L. G., Cao, Y., Ngai, H., Moore, J. P., and Ho, D. D. (1996). Quantitative analysis of serum neutralization of human immunodeficiency virus type 1 from subtypes A, B, C, D, E, F, and I: Lack of direct correlation between neutralization serotypes and genetic subtypes and evidence for prevalent serum-dependent infectivity enhancement. J. Virol. 70, 445–458. Lee, W.-R., Syu, W.-J., Du, B., Matsuda, M., Tan, S., Wolf, A., Essex, M., and Lee, T.-H. (1992). Nonrandom distribution of gp120 N-linked glycosylation sites important for infectivity of human immunodeficiency virus type 1. Proc. Natl. Acad. Sci. USA 89, 2213–2217. Mascola, J. R., Louwagie, J., McCutchan, F. E., Fischer, C. L., Hegerich, P. A., Wagner, K. F., Fowler, A. K., McNeil, J. G., and Burke, D. S. (1994). Two antigenically distinct subtypes of human immunodeficiency virus type 1: Viral genotype predicts neutralization serotype. J. Infect. Dis. 169, 48–54. McKeating, J. A., Zhang, Y. J., Arnold, C., Fredriksson, R., Fenyo¨, E. M., and Balfe, P. (1996). Chimeric viruses expressing primary envelope glycoproteins of human immunodeficienct virus type 1 show increased sensitivity to neutralization by human sera. Virology 220, 450–460. Moore, J. P., Burkly, L. C., Connor, R. L., Cao, Y., Tizard, R., Ho, D. D., and Fisher, R. A. (1993). Adaptation of two primary human immunodeficiency virus type 1 isolates to growth in transformed T cell lines correlates with alterations in the response of their envelope glycoproteins to soluble CD4. AIDS Res. Hum. Retroviruses 9, 529–539. Moore, J. P., Cao, Y., Quing, L., Sattentau, Q. J., Pyati, J., Koduri, R., Robinson, J., Barbas, C. F., III, Burton, D. R., and Ho, D. D. (1995). Primary isolates of human immunodeficiency virus type 1 are relatively resistant to neutralization by monoclonal antibodies to gp120 and their neutralization is not predicted by studies with monomeric gp120. J. Virol. 69, 101–109. Moore, J. P., and Ho, D. D. (1995). HIV-1 neutralization: The consequences of viral adaptation to growth on transformed T cells. AIDS 9, S117–S136. Nyambi, P. N., Nkengasong, J., Lewi, P., Andries, K., Janssens, W., Fransen, K., Heyndrichx, L., Piot, P., and van der Groen, G. (1996). Multivariate analysis of human immunodeficiency virus type 1 neutralization data. J. Virol. 70, 6235–6243. Parren, P. W. H. I., Fisicaro, P., Labrijn, A. F., Binley, J. M., Yang, W.-P., Ditzel, H. J., Barbas, C. F., III, and Burton, E. R. (1996). In vitro antigen

challenge of human antibody libraries for vaccine evaluation: The human immunodeficiency virus type 1 envelope. J. Virol. 70, 9046– 9050. Peden, K., Emerman, M., and Montagnier, L. (1991). Changes in growth properties on passage in tissue culture of viruses derived from infectious molecular clones of HIV-1LAI , HIV-1MAL , and HIV-1ELI . Virology 185, 661–672. Sattentau, Q. J., and Moore, J. P. (1995). Human immunodeficiency virus type 1 neutralization is determined by epitope exposure on the gp120 oligomer. J. Exp. Med. 182, 185–196. Sawyer, L. S. W., Wrin, M. T., Crawford-Miksza, L., Potts, B., Wu, Y., Weber, P. A., Alfonso, R. D., and Hanson, C. V. (1994). Neutralization sensitivity of human immunodeficiency virus type 1 is determined in part by the cell in which the virus is propagated. J. Virol. 68, 1342–1349. Schonning, K., Jansson, B., Olofsson, S., and Hansen, J.-E. S. (1996). Rapid selection for an N-linked oligosaccharide by monoclonal antibodies derected against the V3 loop of human immunodeficiency virus type 1. J. Gen. Virol. 77, 753–758. Stamatatos, L., and Cheng-Mayer, C. (1995). Structural modulations of the envelope gp120 glycoprotein of human immunodeficiency virus type 1 upon oligomerization and differential V3 loop epitope exposure of isolates displaying distinct tropism upon virion-soluble receptor binding. J. Virol. 69, 6191–6198. Sullivan, N., Sun, Y., Li, J., Hofmann, W., and Sodroski, J. (1995). Replicative function and neutralisation sensitivity of envelope glycoproteins from primary and T-cell line-passaged human immunodeficiency virus type 1 isolates. J. Virol. 69, 4413–4422. Sundqvist, V. A., Albert, J., Ohlsson, E., Hinkula, J., Fenyo¨, E. M., and Wahren, B. (1989). Human immunodeficiency virus type 1 p24 production and antigenic variation in tissue culture of isolates with various growth characteristics. J. Med. Virol. 29, 170–175. Tan, W., Fredriksson, R., Bjo¨rndal, A˚., Balfe, P., and Fenyo¨, E. M. (1993). Cotransfection of HIV-1 molecular clones with restricted cell tropism may yield progeny virus with altered phenotype. AIDS Res. Hum. Retroviruses 9, 321–329. Trkola, A., Purtscher, M., Muster, T., Ballaun, C., Buchacher, A., Sullivan, N., Srinivasan, K., Sodroski, J., Moore, J. P., and Katinger, H. (1996). Human monoclonal antibody 2G12 defines a distinctive neutralization epitope on the gp120 glycoprotein of human immunodeficienct virus type 1. J. Virol. 70, 1100–1108. Turner, S., Tizard, R., DeMarinis, J., Pepinsky, R. B., Zullo, J., Schooley, R., and Fisher, R. (1992). Resistance of primary isolates of human immunodeficiency virus type 1 to neutralization by soluble CD4 is not due to lower affinity with the viral envelope glycoprotein gp120. Proc. Natl. Acad. Sci. USA 89, 1335–1339. Weber, J., Fenyo¨, E. M., Beddows, S., Kaleebu, P., Bjo¨rndal, A˚., and collaborators (1996). Neutralization serotypes of human immunodeficiency virus type 1 field isolates are not predicted by genetic subtype. J. Virol. 70, 7827–7832. Willey, R. L., Shibata, R., Freed, E. O., Cho, M. W., and Martin, M. A. (1996). Differential glycosylation, viron incorporation, and sensitivity to neutralizing antibodies of human immunodeficiency virus type 1 envelope produced from infected primary T-lymphocyte and macrophage cultures. J. Virol. 70, 6431–6436. Wrin, T., Loh, T. P., Charron Vennari, J., Schuitemaker, H., and Nunberg, J. H. (1995). Adaptation to parsistent growth in the H9 cell line renders a primary isolate of human immunodeficiency virus type 1 sensitive to neutralization by vaccine sera. J. Virol. 69, 39–48. ¨ hman, P., Biberfeld, G., and Fenyo¨, Zhang, Y.-J., Putkonen, P., Albert, J., O E. M. (1994). Stable biological and antigenic characteristics of HIV2SBL6669 in nonpathogenic infection of macaques. Virology 200, 583–589.

AID

viras

VY 8812

/

6a53$$$281

10-28-97 14:04:05

AP: VY