Oct 5, 1990 - Baba, M., R. Anoeck, R. Pauwels, and E. De Clercq. 1988. Sulfated ... immunodeficiency virus replication and giant cell formation. Antimicrob.
JOURNAL OF VIROLOGY, Mar. 1991,
p. 1543-1550
Vol. 65, No. 3
0022-538X/91/031543-08$02.00/0 Copyright C) 1991, American Society for Microbiology
Dextran Sulfate Blocks Antibody Binding to the Principal Neutralizing Domain of Human Immunodeficiency Virus Type 1 without Interfering with gpl20-CD4 Interactions LAWRENCE N. CALLAHAN, MICHAEL PHELAN, MARGHERITA MALLINSON, AND MICHAEL A. NORCROSS* Laboratory of Molecular Immunology, Division of Cytokine Biology, Center for Biologics Evaluation and Research,
Food and Drug Administration, Bethesda, Maryland 20892 Received
5
October 1990/Accepted 30 November 1990
The mechanism of the antiviral activity of sulfated polysaccharides on human immunodeficiency virus type (HIV-1) was investigated by determining the effect of dextran sulfate on the binding of CD4 and several anti-gpl20 monoclonal antibodies to both recombinant and cell surface gpl20. Dextran sulfate did not interfere with the binding of sCD4 to rgpl20 on enzyme-linked immunosorbent assay (ELISA) plates or in solution and did not block sCD4 binding to HIV-l-infected cells expressing gpl20 on the cell surface. Dextran sulfate had minimal effects on rgpl20 binding to CD4+ cells at concentrations which effectively prevent HIV replication. In contrast, it potently inhibited the binding of both rgpl20 and cell surface gpl20 to several monoclonal antibodies directed against the principal neutralizing domain of gpl20 (V3). In an ELISA format, dextran sulfate enhanced the binding of monoclonal antibodies against amino-terminal regions of gpl20 and had no effect on antibodies directed to other regions of gpl20, including the carboxy terminus. The inhibitory effects of polyanionic polysaccharides on viral binding, viral replication, and formation of syncytia therefore appear mediated by interactions with positively charged amino acids concentrated in the V3 region. This high local positive charge density, unique to the V3 loop, leads us to propose that this property is critical to the function of the V3 region in mediating envelope binding and subsequent fusion between viral and cell membranes. The specific interaction of dextran sulfate with this domain suggests that structurally related molecules on the cell surface, such as heparan sulfate, may be additional targets for HIV binding and infection. 1
and discuss how this mechanism effects the rationale for therapy.
Sulfated polysaccharides are reported to inhibit the replication of a variety of viruses in vitro (3, 28). A number of sulfated polysaccharides including dextran sulfate have been shown to be potent inhibitors of human immunodeficiency virus (HIV) replication in vitro (6, 7, 13, 19, 27, 31). In vivo, however, orally administered dextran sulfate was largely ineffective against AIDS in a clinical study (1), presumably owing to inefficient absorption from the gastrointestinal tract (11, 18). A number of other sulfated polysaccharides also effectively inhibit HIV and may have better absorptive properties as well as less potent anticoagulant activity than dextran sulfate (4, 6). It is therefore possible that compounds having the same mode of action as dextran sulfate will be effective anti-HIV agents. Several groups have suggested that dextran sulfate inhibits HIV by preventing both binding of the virion to CD4+ cells (19, 22) and subsequent syncytium formation (6, 7, 19). Other groups have claimed that dextran sulfate acts by binding to CD4 (16, 20) at the gpl20-binding site, thereby directly preventing the CD4-gpl20 interaction (16). In this study, however, we were unable to demonstrate that dextran sulfate directly interfered with either the binding of soluble CD4 (sCD4) to recombinant gp 120 (rgpl20), sCD4 to HIV+ cells, or rgpl20 to CD4+ cells. Instead, we found that inhibitory sulfated polysaccharides blocked the interaction between rgpl20 or gp120 on HIV+ cells and neutralizing antibodies that bind to the V3 loop region (amino acids 303 to 338) of gp120. We will present a molecular mechanism developed from these data for the action of dextran sulfate *
MATERIALS AND METHODS Direct solid-phase enzyme-linked immunosorbent assay (ELISA) of rgpl20 binding to sCD4. A 100-,ul portion of a solution containing 1 pug of rgpl20 (HXB2 related) (a gift of Genentech, South San Francisco, Calif.) per ml in 0.1 M NaHCO3 (pH 9) was added to each well of a microfluor plate (Dynatech, Alexandria, Va.) and incubated at 4°C overnight. The plate was then blocked with 100 pul of a 3% solution of nonfat dry milk or 1% bovine serum albumin (BSA) in phosphate-buffered saline (PBS). Various concentrations of either dextran sulfate or dextran were added to duplicate wells, which were then preincubated for 30 min. A 100-ng portion of sCD4 (a gift of Biogen, Cambridge, Mass.) was then added to each well and coincubated with dextran or dextran sulfate for 30 min. The plates were then washed three times with PBS containing 0.5% Tween 20. The amount of bound sCD4 was determined by addition of an anti-CD4 monoclonal antibody, either L-120 or L-83 (100 ng per well; a gift of Becton Dickinson, Emeryville, Calif.), followed by alkaline phosphatase-labeled, affinity-purified, goat anti-mouse immunoglobulin G diluted 1:1,000. All solutions were diluted with PBS containing 0.05% Tween and 3% nonfat dry milk (BLOTTO) or 1% BSA. The plates were then developed with a 50 mM solution of 4-methylumbelliferyl phosphate (Sigma, St. Louis, Mo.) in 100 mM ethanolamine, 5 mM Mg2+, and 5 mM Ca2+. The fluorescence intensity was measured by using a microfluor reader (Dynatech).
Corresponding author. 1543
1544
J. VIROL.
CALLAHAN ET AL.
Antigen capture assay for rgpl20 binding to sCD4. A 100-pul portion of a solution containing 1 ,ug of either the anti-CD4 monoclonal antibodies (L-120 or L-83) or the anti-gpl20 antibodies (110.1 or 110.4, anti-gpl20 LAV; a gift of Genetic Systems, Seattle, Wash.) per ml in 0.1 M NaHCO3 (pH 9) was added to each well of a microfluor plate and incubated at 4°C overnight. The plate was then blocked with 100 [lI of a 1% BSA solution in PBS. In a separate tube, gpl20 (500 ng/ml) and then sCD4 (500 ng/ml) were added to a solution containing a variety of concentrations of either dextran sulfate, dermatan sulfate, or dextran. A 100-pul portion of each solution was then added to each well and allowed to incubate at room temperature for 1 h. The amount of captured antigen was determined by using either 50 ng of a biotinylated form of anti-CD4 antibody (antibody-to-biotin molar ratio, 1:10) per well as previously described (10) if an anti-gpl20 antibody was used to capture, or 50 ng of a biotinylated form of anti-gpl20 antibody per well if anti-CD4 antibody was used to capture. The biotinylated antibodies were detected by using the AP-ABC kit (Vector Labs, Burlingame, Calif.), and the plate was further developed as described above. Antigen capture assay for monoclonal antibodies binding to rgpl2O. A 100-pul portion of a solution containing either 1 ,ug/ml or a 1:2,000 dilution of ascites of anti-gpl20 antibodies 110.1, 110.4 (gifts of Genetic Systems) 6E10, 5B3, 1D10, 6D8, liG5, 10D8, or 1OF6 (a gift of Genentech) in 0.1 M NaHCO3 (pH 9) was added to each well of a microfluor plate. The plate was then blocked as above. In a separate tube, gpl20 (500 ng/ml) was added to a solution containing 100 pug of either dextran sulfate (molecular weight 5,000), dextran (molecular weight 10,000), or dermatan sulfate (molecular weight MW 8,000) per ml. The amount of captured antigen was determined by using 50 ng of a biotinylated form of either anti-gp 120 antibodies 110.1 or 110.4 per well. The plate was further developed as described above. Binding assay for rgpl20 to CD4+ cells. Various amounts of either dextran or dextran sulfate were added to approximately 500,000 Molt-4 cells suspended in 100 pL1 of PBS
containing 10% fetal calf serum. The cells were preincubated for 30 min, and then 200 ng of rgpl2O was added. After a further 30 min of incubation, the cells were washed three times and then 50 plI of a 1:1,000 dilution of ascites fluid of monoclonal anti-gpl20 antibodies, either 110.1 or 110.4, was added to the cells and incubated for 30 min. The cells were again washed three times and incubated with a 1:500 dilution of fluorescein isothiocyanate-labeled goat anti-mouse antibodies (Cappel, Malvern, Pa.). The amount of fluorescence was then quantitated by using flow cytometry (FACS; Becton-Dickinson, Fullerton, Calif.). All incubations were done at 4°C. The extent of binding inhibition was calculated by first converting the intensity from a log scale to a linear scale and then subtracting out the antibody-negative control from both dextran sulfate- and dextran-treated cells. The percent inhibition of binding was equal to 100 times the linear signal from treated cells, divided by the linear signal from untreated cells. Binding assay for sCD4 and anti-gpl20 antibodies to HIV+ cells. Approximately 500,000 H9 cells, previously infected with HIVIIIB, were suspended in RPMI 1640 (GIBCO) containing 10% fetal bovine serum and various concentrations of either dextran or dextran sulfate. The cells were preincubated at 4°C for 30 min, and then 200 ng of sCD4 was added, bringing the total volume to 100 ml. The cells were incubated for 30 min and washed three times, and the amount of bound sCD4 was determined by the addition of
TABLE 1. Effect of dextran sulfate on the CD4-gpl20 interaction % Inhibition' with dextran sulfate at: Interaction format
la. lb. lc. 2. 3.
10 mg/ml 1 mg/ml 100 ,ug/ml 10 pLg/ml
Adsorbed rgpl2O-sCD4 Capture of sCD4-rgpl2O Capture of rgpl2O-sCD4 Cellular CD4-rgpl20b
-71 2 -9 43
Cellular gpl2O-sCD4
-33 -1 9 54 4
-11 -3 0 12 -9
-9 -2 5 8
-2
aPercent inhibition is given as follows: [(dextran sulfate fluorescent signal - background)/(control - background)] x 100. Negative inhibition indicates an enhancement of binding. b Cellular binding inhibition is given as follows: [(linear fluorescent signal for treated cells - background signal)/linear signal for control background] x 100.
100 ,ul of a 1:10 dilution of phycoerythrin-conjugated OKT4 antibody (Ortho Diagnostic Systems, Raritan, N.J.). The fluorescence intensity was measured as described above. For anti-gpl20 antibody binding, HIV+ cells were suspended in solutions containing either dextran or dextran sulfate as above, a 1:1,000 dilution of ascites of the antigpl20 monoclonal antibody 110.4 was added, and the mixture was incubated for 30 min. Cells were washed three times and incubated with fluorescein isothiocyanate-labeled anti-mouse antibody. The fluorescence intensity was determined as described above. Virus inhibition assay. MT-4 cells (2.5 x 104) were cultured with a 1:400 dilution of HIVIIIB stock virus (1 ng of p24 per ml) in the presence of various concentrations of either dextran sulfate or dextran. Cells were incubated for 5 days and labeled with [3H]thymidine (1 ,uCi per well) over the last 24 h of culture. Cells were cultured in RPMI 1640 supplemented with 10% fetal calf serum with penicillin-streptomycin antibiotics. The percent inhibition of cytopathicity is given by the following formula: [(cpm of treated HIVinfected cells - cpm of untreated HIV-infected cells)/(cpm of uninfected cells - cpm of HIV-infected cells)] x 100. RESULTS
Effect of dextran sulfate on the CD4-gpl20 interaction. Previously, Lederman et al. (16) reported that dextran sulfate inhibited the binding of rgpl2O to sCD4. Using similar assay systems, we have been unable to demonstrate any inhibitory effect of dextran sulfate on the rgpl2O-sCD4 interaction at concentrations that are antiviral. Table 1 shows the results of three such assays. In the first assay (assay la), gpl20 was coated on the ELISA plate and then incubated with sCD4 in the presence of dextran sulfate. In this format, dextran sulfate potentiates sCD4 binding at very high concentrations, probably owing to an excluded-volume effect. In the second assay (assay lb), an anti-CD4 antibody was coated on a plate and used to capture CD4-gpl2O in the presence of dextran sulfate. In the third system (assay lc), the capturing antibody was an anti-gpl20 antibody. In these ELISA-based assays, dextran sulfate had little effect on the sCD4-rgpl2O interaction at concentrations as high as 10 mg/ml. The order of addition did not appear to make a difference, as inhibition did not occur regardless of whether sCD4 or rgpl2O was preincubated with dextran sulfate. We have been unable to reconcile these results with those of Lederman et al. (16) except to note that an insect cellderived gpl20 was used in the previous study whereas our experiments used a CHO cell-derived gpl20. To further
DEXTRAN SULFATE BLOCKS HIV NEUTRALIZING DOMAIN
VOL. 65, 1991
4000 -X 3500 C * 3000 1
*
Blank
\
Dextran Sulfate
*
Dermatan Sulfate
1545
LL I-I
o
2500
m
m
2000
0
° 1 500 a)
WL
1 000
I
500 0
6E10 110.1 lDlO 6D8 10D8 10F6 11G5 110.4 Monoclonal Antibodies FIG. 1. Effect of dextran sulfate and dermatan sulfate on the capture of gpl20 by monoclonal anti-gp120 antibodies. rgpl2O was preincubated with 100 ,ug of dextran sulfate or dermatan sulfate per ml or buffer and then captured with monoclonal antibody coated onto ELISA wells. The wells were washed, and then biotinylated 110.1 was used to detect bound gpl20 for all the capture antibodies except 110.1, for which biotinylated 110.4 was substituted.
analyze the effect of dextran sulfate on envelope binding, we examined the binding of rgpl2O to the CD4+ cell Molt-4 (assay 2) and the binding of sCD4 to HIV+ H9 cells (assay 3). In both assays, cells were preincubated with dextran sulfate for 30 min and then coincubated with ligand. In assays measuring rgpl2O binding to Molt-4 cells, partial inhibition was seen at high concentrations of dextran sulfate, whereas minimal effects were observed at lower concentrations that were still active in antiviral assays. High-dose inhibition probably results from the down regulation of cellular CD4, as has been reported to occur when high concentrations of dextran sulfate are incubated with peripheral blood lymphocytes (29). In assays measuring the effect of dextran sulfate on CD4 binding to HIV-infected cells, dextran sulfate had little effect on the binding of soluble CD4 to gpl20 on the cell surface. These results are consistent with those obtained by using the ELISA format and indicate that the reported inhibition of virion binding by dextran sulfate is not simply a result of blocking the CD4-gpl2O interaction. Dextran sulfate blocks the interaction between gpl20 and antibodies directed toward the V3 neutralizing loop of gpl20. During the course of developing a capture assay to measure the CD4-gpl2O interaction, we noticed that when the monoclonal antibody 110.4 was used to capture gpl20, the reaction was inhibited by dextran sulfate. Further experiments showed that the inhibition was due to the interaction between gpl20 and the capturing antibody. Dextran sulfate inhibited the capture of gpl20 by several antibodies (10D8, 10F6, 11G5, and 110.4), all of which are directed against the V3 loop region of gpl20, residues 303 to 338 (7a, 30) (Fig. 1). Antibodies directed against this loop region, including 110.4 (17), have been reported to be capable of neutralizing HIV (9, 14, 21) while not blocking gpl20-CD4 interactions (17,
25). The binding of two antibodies, 6E10 and 110.1 (110.1 reacts with peptide 494 to 517 [30]), was not affected by incubation with dextran sulfate. Interestingly, dextran sulfate increased the binding of antibodies lDlO (peptide 34 to 55) and 6D8 (peptide 21 to 85), both of which are directed to the amino terminus of gpl20 (8). These effects suggest that dextran sulfate binds gpl20 possibly near the V3 site and induces conformational changes that selectively affect other sites. Dermatan sulfate, an HIV-inactive polyanionic compound that has a lower charge density than dextran sulfate and contains negative charges on only one face of the polymer, had little effect on the capture assay regardless of the antibody used (Fig. 1). Taken as a whole, these data indicate that the mechanism of inhibition involves dextran sulfate binding to or disrupting of a site on gpl20 that has been shown previously to be essential for viral infectivity (9, 14, 21, 25). Surprisingly, however, dextran sulfate did not inhibit the direct binding of any of the monoclonal antibodies when gpl20 was directly adsorbed on a polystyrene plate (data not shown). There are two possible explanations for this phenomenon: either part of the site on gpl20 that interacts with dextran sulfate is sensitive to or is blocked by adsorbtion to the polystyrene plate, or the binding of antibodies to solid-phase gpl20 is much stronger than the binding to solution-phase rgpl2O and thereby successfully competes with dextran sulfate binding. The experiments described above demonstrate that the dextran sulfate effect depends on the manner in which rgpl20 is presented to the antibody. To determine whether virion or cell surface gpl20 resembles the adsorbed or solution form of rgpl2O, we assayed the effect of dextran sulfate on the binding of monoclonal antibody 110.4 to H9 cells chronically infected with HIVIIIB. The flow cytometry histograms of chronically infected H9 cells (Fig. 2A) dem-
1546
J. VIROL.
CALLAHAN ET AL.
300
A
C E 200 L L
N U B 100 R
0
1000
1
RED FLUORESCENCE
B 300 C E L L 200 N U
'00
B R 100
'0
0* 0. 0
3
0
~~0
%n 0° ~~~~~ nooqS m .oonoo .10055n O-
1000 100 10 LOG GREEN FLUORESCENCE FIG. 2. Effect of dextran sulfate on sCD4 and monoclonal antibody 110.4 binding to HIV-infected H9 cells. A total of 5 x 105 HIVIIIB-infected H9 cells were incubated with 200 ng of sCD4 (A) or monoclonal antibody 110.4 (1:1,000) (B) in the presence of 100 ,ug of dextran sulfate (O) or dextran (0) per ml, washed, and then stained with OKT4PE (A) or fluorescein isothiocyanate-labeled goat anti-mouse immunoglobulin (B). Background fluorescence with labeled antibodies alone is also shown (A). 1
onstrate that dextran sulfate at 100 ,ug/ml completely blocks the binding of this V3-reactive antibody, whereas dextran has no effect. In contrast, the extent of sCD4 binding is unaffected by dextran sulfate (Fig. 2B), indicating that dextran sulfate does not globally disrupt the structure of gpl20, but binds to specific sites on gpl20. The lack of inhibition of sCD4 binding also demonstrates that dextran sulfate does not down regulate the expression of cell surface gp120 or disrupt the gp4l-gpl2O interaction. This inhibition of anti-V3 antibody binding to cell surface gp120 confirms a recent report (24) showing that polyanionic compounds inhibited the binding of one anti-V3 antibody to HIVinfected cells. However, no data in that study were presented to verify whether gpl20 remained on the cell surface following treatment, because this phenomenon could also be explained by gpl20 shedding. Figure 3 presents the doseresponse curves for inhibition of anti-V3 antibody binding in comparison with antiviral activity. The extent of binding
inhibition of monoclonal antibody 110.4 to HIV+ cells at various concentrations of dextran, dextran sulfate, and dermatan sulfate is shown in Fig. 3A. The antiviral agent dextran sulfate blocked the binding of monoclonal antibody 110.4 to infected cells at concentrations as low as 100 ng/ml, whereas the related compounds dextran and dermatan sulfate, which have no antiviral activity, failed to block this binding. Figure 3B shows a profile of the extent of protection afforded to MT4 cells by various concentrations of dextran sulfate and dextran against challenge by HIVIIIB. Antiviral activity was evident with dextran sulfate at concentrations above 1 ,ug/ml, which is the concentration at which inhibition of anti-V3 antibody binding to gpl20 becomes maximal. The dose-response curves are again consistent with a mechanism of anti-HIV action involving the direct binding of dextran sulfate to the V3 loop region of gpl20. These curves also indicate that for antiviral action to occur, it is necessary to completely bind or block the V3 loop with dextran sulfate;
1 20
1 20
-pI
100 X~ 0
'r 8+ 8 0 -
m _ 40 *0 o
80
0
60
rc
40 20
c
>09 0-
s
O0-
*
0
200
°
'B
1 00
p
_
0
1547
DEXTRAN SULFATE BLOCKS HIV NEUTRALIZING DOMAIN
VOL. 65, 1991
I
0.01
.
C .-
i
-.o
a ............
1000 100 0.1 1 10 Concentration of Polysaccharide gg / m I
G.-o
-2~0 0.001
-e- ---O -&- __e ea
0.01
s
n
a
mmej
0.1
ell-o'-O .
. a
1
G-0--o
nomad
n
2 a n Rona
10
1 00
Concentration of Polysaccharide gg / m I FIG. 3. (A) Concentration dependence of dextran sulfate inhibition of monoclonal antibody 110.4 binding to HIV+ H9 cells. H9-HIVIIIB cells were preincubated with dextran sulfate (O), dermatan sulfate (O), or dextran (0) and then mixed with a 1:1,000 dilution of 110.4 ascities, washed, and stained with fluorescein isothiocyanate-labeled goat anti-mouse immunoglobulin F(ab)2 fragment. Percent inhibition is given as follows: (linear signal from treated cells/total linear signal from untreated cells) x 100. (B) Effect of dextran sulfate concentration on the
cytopathic effects of HIV-11HIB on MT-4 cells. Dextran sulfate (O) or dextran (0) was mixed with MT-4 cells in triplicate and then incubated with a 1:400 dilution of an HIV-1IIIB stock for 5 days. [3H]thymidine incorporation was measured over the last 24 h of culture. Percent inhibition of cytopathicity is given as follows: [(cpm of treated HIV-infected cells cpm of untreated infected cells)/(cpm of uninfected cells cpm of HIV-infected cells)] x 100. -
-
partial action.
occupancy
does not
appear to
lead to antiviral
DISCUSSION These studies indicate that the mechanism by which sulfated polysaccharides inhibit HIV replication appears to be related to an interaction with the V3 loop of gpl20. This interaction is probably mediated by the binding of negatively charged sulfate groups on dextran sulfate to positively charged amino acid side chains in the V3 loop. It is interesting that despite being a highly variable region of gpl20, the V3 loop manages to maintain a highly positive charge density in a wide variety of HIV strains (Table 2). This property suggests that the function of this region may be related to the charge density rather than to the specific amino acid sequence. In fact, this site and the last 25 amino acids of gpl20 are the only regions of gpl20 that consistently have a high positive charge density (Fig. 4). Coincidentally, the carboxyterminal region is recognized by monoclonal antibody 110.1 and dextran sulfate does not affect the binding of this antibody to gpl20 (Fig. 1). Although we have not ruled out effects of dextran sulfate on other functional regions of the envelope glycoprotein, we believe that the primary site of interaction between dextran sulfate and gpl20 lies in the V3 loop. Several factors lead to the conclusion that the interaction with the V3 loop explains the antiviral activity of dextran sulfate. First, similarly to dextran sulfate, polyclonal and monoclonal antibodies (including Mab 110.4 [17]) directed against the V3 loop are neutralizing (9, 14, 21) and prevent cell-cell fusion, but do not substantially block the CD4-gpl2O interaction (17, 25). In this paper we have demonstrated that the binding of gpl20 to CD4 either in recombinant form or on the cell surface is not inhibited at concentrations of dextran sulfate that effectively inhibit viral replication. However, several other groups, as mentioned above, have shown that dextran sulfate blocks the binding of virions to CD4+ cells (7, 19), in contrast to anti-V3 antibodies which do not prevent virion binding (25).
We envision that dextran sulfate binding to gpl20 and other cell surface sites disrupts two (probably related) processes critical to HIV infection: inhibition of HIV binding to the cell surface, followed by interference of the syncytium TABLE 2. Charge density within gp120 V3 loopa b No. of No. of No. of amino acids in loop
positively charged amino acids
negatively charged amino acids
34 34 34 35 33 33 32 34 33 33
9 8 9 8 8 5 8 8 8 7
0 0 0 1 0 2 1 0 1 1
32 32 32 33 32 33 33
7 5 5 6 6 6 5
0 1 0 0 0 0 2
ISY NIHZ
34 33 33
7 9 7
0 1 0
SIV-2C MM142 BK28 SMMH2
32 32 32
5 5 5
1 1 2
Viral strain
North American HXB2 BH10 BH8 RF MN SC WMJ2 LAV SF2 NY5
African Z3 MAL ELI Z6 Z2
JYl Z321 HIV-2 ROD2
a Amino acid sequences from the Los Alamos Human Retroviruses and AIDS data base, 1989. b The V3 loop is from amino acid 303 to amino acid 338. c SIV-2, Simian immunodeficiency virus type 2.
1548
J. VIROL.
CALLAHAN ET AL.
0. 5
-0. 5 0
100
200
300
400
500
600
700
800
900
Residue Number FIG. 4. Charge profile of the HIV-1111B (HxB2) envelope glycoprotein gpl60. Rolling-window analysis averaging charge over a 31-amino-acid moving window, using -1 for glutamic and aspartic acids and + 1 for arginine and lysine. The V3 loop in this sequence is from amino acids 296 to 331. gp120 is from amino acids 1 to 511; gp4l is from amino acids 512 to 856.
fusion step mediated partly by the V3 loop and gp4l fusion peptide. The initial virus-binding event depends on the interaction with the CD4 membrane receptor, but additionally involves overcoming the electrostatic repulsive forces between the cell and virion membranes. Because dextran sulfate has no effect on CD4-gpl20 interactions, we view the early inhibitory effects of dextran sulfate on virion binding as a combination of specific and nonspecific disruption of ionic interactions between charged regions on surface glycoproteins, including gpl20, and membrane phospholipids. Enveloped viruses that bud from cells carry part of the cell membrane and therefore may interact more effectively with the negatively charged cell surface if the outer part of the envelope protein contains regions of high positive charge density (26). We believe that the V3 loop in HIV may provide this function and that the binding of dextran sulfate to the loop not only abolishes positive charge but also adds an additional negative potential, thereby electrostatically preventing the interaction between the virion and the cell. This general mechanism also explains the broad antiviral activity of sulfated polysaccharides against enveloped viruses and predicts that these viruses have domains that help to overcome the negative potentials. One would also predict that these positively charged domains will contain neutralizing epitopes. It is interesting that another enveloped virus, herpes simplex virus, is reported to use cell surface heparan sulfate as an initial binding site (32) and is also inhibited by sulfated polysaccharides. By analogy, heparan sulfate or a related cell surface glycoaminoglycan may provide a similar function to augment HIV membrane binding through the envelope V3 region.
The extreme hypervariability of the V3 loop argues against involvement of the loop in any essential structural, allosteric, or receptor interaction. Therefore, the most likely role for the V3 loop is to assist in membrane fusion after gpl20 binds CD4. Once again, the high positive charge density could be essential to this function, since polycationic peptides and proteins have been shown to facilitate fusion, presumably by bringing two negatively charged membranes together (26). Specifically, this could be accomplished by binding to a negatively charged region on the cell membrane, thereby orienting the fusigenic portion of gp4l and the viral membrane such that fusion occurs. The V3 region may be physically close to gp4l, since mutations in the loop have been reported to lead to gp41-gpl20 dissociation (15). Proteolysis has also been shown to occur within the V3 loop and could be involved in inducing fusion (12). Dextran sulfate binding to the V3 region may therefore block both proteolysis and specific binding by disrupting the local positive charge density. On the other hand, antibodies directed against the V3 domain may not inhibit virion binding, because they do not eliminate the positive charge density but still physically block an essential interaction between the V3 loop and either a cell surface protease or the cell surface membrane. Although we believe that the antiviral effects of dextran sulfate are mediated through the V3 region, several caveats should be discussed. In addition to envelope binding and syncytia inhibition, dextran sulfate may disrupt virus replication at other sites in the virus growth cycle, including those unrelated to viral envelope function. For example, dextran sulfate may affect processes following fusion, such
VOL. 65, 1991
DEXTRAN SULFATE BLOCKS HIV NEUTRALIZING DOMAIN
as uncoating and functional release of viral RNA. It may also bind to and influence other functional regions of the envelope glycoprotein involved in either binding, syncytium formation, or viral replication. Finally, the conformation of rgpl20 used in these experiments may be different from that of gpl20 expressed on viral particles or HIV-infected cells, and, likewise, soluble CD4 may differ from CD4 on the cell surface. Therefore, the effects of inhibitory compounds on the interaction of the virion envelope with cell surface CD4 under native conditions may differ from those seen in these model systems. Dextran sulfate is one of a number of sulfated polysaccharides and small anionic molecules, such as pentosan sulfate (5), sulfated a- and P-cyclodextrin (2), and aurin tricarboxylic acid (23), that are known to inhibit HIV in vitro. In our experiments, all of these compounds and Evan's Blue inhibited the binding of V3 reactive antibodies to gp120 (7b), whereas only Evan's Blue and aurin tricarboxylic acid inhibited the CD4-gpl2O interaction (data not shown). These data confirm that the binding of negatively charged compounds to the V3 loop is responsible for viral inhibition. Although dextran sulfate is capable of blocking the binding of antibodies to the V3 loop of gpl20, it is not effective at disrupting preformed antibody-gpl20 complexes (data not shown). Therefore, since the V3 loop has been shown to be an immunodominant site on gp120, it is possible that antibodies in HIV+ sera block dextran sulfate from binding gp120 and that dextran sulfate interferes with the binding of these neutralizing antibodies. Therefore, it may be difficult to predict the in vivo effectiveness of sulfated polysaccharides on the basis of their in vitro activity. The molecular basis for the anti-retroviral activity, however, suggests that dextran sulfate and other sulfated polysaccharides have a wider range of activity against a variety of viral strains than do antibodies which are more strain specific. It is therefore possible that compounds having the same mode of action but with improved absorption and less potent anticoagulant activity will be effective anti-HIV agents. ACKNOWLEDGMENTS We thank Gary W. Smythers for analysis of the envelope charge content, Judy Beeler for critical review of the manuscript, and Howard Mostowski for cell sorter analysis. We also thank Skip Maino, David Buck, Phil Berman, Shiu-lok Hu, Tim Gregory, and R. Fisher for monoclonal antibodies and recombinant proteins. REFERENCES 1. Abrams, D. I., S. Kuno, R. Wong, K. Jeffords, M. Nash, J. B. Molaghan, R. Gorter, and R. Ueno. 1989. Oral dextran sulfate (UA001) in the treatment of the acquired immunodeficiency syndrome (AIDS) and AIDS-related complex. Ann. Intern. Med. 110:183-188. 2. Anand, R., S. Nayyar, J. Pitha, and C. R. Merril. 1990. Sulphated sugar alpha-cyclodextrin sulphate, a uniquely potent anti-HIV agent, also exhibits marked synergism with AZT, and lymphoproliferative activity. Antiviral Chem. Chemother. 1:4146. 3. Baba, M., R. Anoeck, R. Pauwels, and E. De Clercq. 1988. Sulfated polysaccharides are potent and selective inhibitors of various enveloped viruses, including herpes simplex virus, cytomegalovirus, vesicular stomatitis virus, and human immunodeficiency virus. Antimicrob. Agents Chemother. 32:1742-
1745. 4. Baba, M., E. De Clercq, D. Schols, R. Pauwels, R. Snoeck, C. Van Boeckel, G. Van Dedem, N. KraaiJeveld, P. Hobbelen, H. Ottenhei,m, and F. D. Hollander. 1989. Novel sulfated polysaccharides: dissociation of anti-human immunodeficiency virus activity from antithrombin activity. J. Infect. Dis. 161:208.
1549
5. Baba, M., M. Nakajima, D. Schols, R. Pauwels, J. Balzarini, and E. De Clercq. 1988. Pentosan polysulfate, a sulfated oligosaccharide, is a potent and selective anti-HIV agent in vitro. Antiviral Res. 9:335-343. 6. Baba, M., D. Schols, E. De Clercq, R. Pauwels, M. Nagy, J. Gyorgyi-Edelenyi, M. Low, and G. Sandor. 1990. Novel sulfated polymers as highly potent and selective inhibitors of human immunodeficiency virus replication and giant cell formation. Antimicrob. Agents Chemother. 34:134-138. 7. Bagasra, O., and H. W. Lischner. 1988. Activity of dextran sulfate and other polyanionic polysaccharides against human immunodeficiency virus. J. Infect. Dis. 158:1084-1087. 7a.Berman, P. Personal communication. 7b.Callahan, L. Unpublished data. 8. Dowbenko, D., G. Nakamura, C. Fennie, C. Shimasaki, L. Riddle, R. Harris, T. Gregory, and L. Lasky. 1988. Epitope mapping of the human immunodeficiency virus type 1 gpl20 with monoclonal antibodies. J. Virol. 62:4703-4711. 9. Goudsmit, J., L. Smit, and R. Meloen. 1988. Conserved location and structure of a HIV-1 neutralization epitope with restricted antibody binding specificity, p. 345-355. In L. Lasky (ed.), Technological advances in vaccine development. Alan R. Liss, Inc., New York. 10. Harlow, E., and D. Lane. 1988. Antibodies: a laboratory manual, p. 341. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 11. Hartman, N. R., D. G. Johns, and H. Mitsuya. 1990. Pharmacokinetic analysis of dextran sulfate in rats as pertains to its clinical usefulness for therapy of HIV infection. AIDS Res. Hum. Retroviruses 6:805-811. 12. Hattori, T., A. Koito, K. Takatsuki, H. Kido, and N. Katunuma. 1989. Involvement of tryptase-related cellular protease(s) in human immunodeficiency virus type 1 infection. FEBS Lett. 8:48-52. 13. Ito, M., M. Baba, A. Sato, R. Pauwels, E. De Clercq, and S. Shigeta. 1987. Inhibitory effect of dextran sulfate and heparin on the replication of human immunodeficiency virus (HIV) in vitro. Antiviral Res. 7:361-367. 14. Javaherian, K., A. J. Langlois, C. McDanal, K. L. Ross, L. I. Eckler, C. L. Jellis, A. T. Profy, J. R. Rusche, D. P. Bolognesi, S. D. Putney, and T. J. Matthews. 1989. Principal neutralizing domain of the human immunodeficiency virus type 1 envelope protein. Proc. Natl. Acad. Sci. USA 86:6768-6772. 15. Kowalski, M., J. Potz, L. Basiripour, T. Dorfman, W. C. Goh, E. Terwilliger, A. Dayton, C. Rosen, W. Haseltine, and J. Sodroski. 1987. Functional regions of the envelope glycoprotein of human immunodeficiency virus type 1. Science 237:1351-1355. 16. Lederman, S., R. Gulick, and L. Chess. 1989. Dextran sulfate and heparin interact with CD4 molecules to inhibit the binding of coat protein (gpl20) of HIV. J. Immunol. 143:1149-1154. 17. Lindsley, P. S., J. A. Ledbetter, E. Kinney-Thomas, and S. Hu. 1988. Effects of anti-gpl20 monoclonal antibodies on CD4 receptor binding by the env protein of human immunodeficiency virus type 1. J. Virol. 62:3695-3702. 18. Lorentsen, K. J., C. W. Hendrix, J. M. Collins, D. M. Kornhauser, B. G. Petty, R. W. Klecker, C. Flexner, R. H. Eckel, and P. S. Lietman. 1989. Dextran sulfate is poorly absorbed after oral administration. Ann. Intern. Med. 111:561-566. 19. Mitsuya, H., D. J. Looney, S. Kuno, R. Ueno, F. Wong-Staal, and S. Broder. 1988. Dextran sulfate suppression of viruses in the HIV family: inhibition of virion binding to CD4+ cells. Science 240:646-649. 20. Parish, C. R., L. Low, H. S. Warren, and A. L. Cunningham. 1990. A polyanion binding site on the CD4 molecule. Proximity to the HIV-gpl2O binding region. J. Immunol. 145:1190-1195. 21. Rusche, J. R., K. Javaherian, C. McDanal, J. Petro, D. L. Lynn, R. Grimaila, A. Langlois, R. C. Gallo, L. 0. Arthur, P. J. Fischinger, D. P. Bolognesi, S. D. Putney, and T. J. Matthews. 1988. Antibodies that inhibit fusion of human immunodeficiency virus-infected cells bind a 24-amino acid sequence of the viral envelope, gpl20. Proc. Natl. Acad. Sci. USA 85:3198-3202. 22. Schols, D., M. Baba, R. Pauwels, and E. De Clercq. 1989. Flow cytometric method to demonstrate whether anti-HIV-1 agents
1550
23.
24.
25.
26. 27.
J. VIROL.
CALLAHAN ET AL.
inhibit virion binding to T4' cells. J. Acquired Immune Defic. Syndr. 2:10-15. Schols, D., M. Baba, R. Pauwels, J. Desmyter, and E. De Clercq. 1989. Specific interaction of aurintricarboxylic acid with the human immunodeficiency virus/CD4 cell receptor. Proc. Natl. Acad. Sci. USA 86:3322-3326. Schols, D., R. Pauwels, J. Desmyter, and E. De Clercq. 1990. Dextran sulfate and other polyanionic anti-HIV compounds specifically interact with the viral gpl20 glycoprotein expressed by T-cells persistently infected with HIV-1. Virology 175:556561. Skinner, M. A., A. J. Langlois, C. B. McDanal, J. S. McDougal, D. P. Bolognesi, and T. J. Matthews. 1988. Neutralizing antibodies to an immunodominant envelope sequence do not prevent gp120 binding to CD4. J. Virol. 62:4195-4200. Stegmann, T., R. W. Doms, and A. Helenius. 1989. Proteinmediated membrane fusion. Annu. Rev. Biophys. Chem. 18: 187-211. Sugawara, I., W. Itoh, S. Kimura, S. Mori, and K. Shimada.
1989. Further characterization of sulfated homopolysaccharides anti-HIV agents. Experientia 45:996-998. Takemoto, K. K., and S. S. Spicer. 1965. Effects of natural and synthetic sulfated polysaccharides on viruses and cells. Ann. N.Y. Acad. Sci. 120:365-373. Thiele, B., H. R. Braig, I. Ehm, R. Kunze, and B. Ruf. 1989. Influence of sulfated carbohydrates on the accessibility of CD4 and other CD molecules on the cell surface and implications for human immunodeficiency virus infection. Eur. J. Immunol. 19:1161-1164. Thomas, E. K., J. N. Weber, J. McClure, P. R. Clapham, M. C. Singhal, M. K. Shriver, and R. A. Weiss. 1988. Neutralizing monoclonal antibodies to the AIDS virus. AIDS 2:25-29. Ueno, R., and S. Kuno. 1987. Dextran sulphate, a potent anti-HIV agent in vitro having synergism with zidovudine. Lancet i:1379. WuDunn, D., and P. G. Spear. 1989. Initial interaction of herpes simplex virus with cells is binding to heparan sulfate. J. Virol. 63:52-58.
as
28.
29.
30. 31.
32.
