JOURNAL OF VIROLOGY, June 2005, p. 7777–7784 0022-538X/05/$08.00⫹0 doi:10.1128/JVI.79.12.7777–7784.2005 Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Vol. 79, No. 12

Resistance of Human Immunodeficiency Virus Type 1 to the High-Mannose Binding Agents Cyanovirin N and Concanavalin A Myriam Witvrouw,1† Valery Fikkert,1† Anke Hantson,1 Christophe Pannecouque,1 Barry R. O’Keefe,2 James McMahon,2 Leonidas Stamatatos,3 Erik de Clercq,1 and Anders Bolmstedt4,5* Rega Institute for Medical Research, Katholieke Universiteit Leuven, Minderbroedersstraat 10, Leuven, Belgium1; Molecular Targets Development Program, Center for Cancer Research, National Cancer Institute, Frederick, Maryland2; Seattle Biomedical Research Institute and Department of Pathobiology, University of Washington, Seattle, Washington 981093; Centre for Microbiological Preparedness, Swedish Institute for Infectious Disease Control, Solna, Sweden4; and Goteborg University, Department of Clinical Virology, Goteborg, Sweden5 Received 27 April 2004/Accepted 3 February 2005

Due to the biological significance of the carbohydrate component of the human immunodeficiency virus type 1 (HIV-1) glycoproteins in viral pathogenesis, the glycosylation step constitutes an attractive target for anti-HIV therapy. Cyanovirin N (CV-N), which specifically targets the high-mannose (HM) glycans on gp120, has been identified as a potent HIV-1 entry inhibitor. Concanavalin A (ConA) represents another mannosebinding lectin, although it has a lower specificity for HM glycans than that of CV-N. For the present study, we selected CV-N- and ConA-resistant HIV-1 strains in the presence of CV-N and ConA, respectively. Both resistant strains exhibited a variety of mutations eliminating N-linked glycans within gp120. Strains resistant to CV-N or ConA displayed high levels of cross-resistance towards one another. The N-glycan at position 302 was eliminated in both of the lectin-resistant strains. However, the elimination of this glycan alone by site-directed mutagenesis was not sufficient to render HIV-1 resistant to CV-N or ConA, suggesting that HIV-1 needs to mutate several N-glycans to become resistant to these lectins. Both strains also demonstrated clear cross-resistance towards the carbohydrate-dependent monoclonal antibody 2G12. In contrast, the selected strains did not show a reduced susceptibility towards the nonlectin entry inhibitors AMD3100 and enfuvirtide or towards reverse transcriptase or protease inhibitors. Recombination of the mutated gp160 genes of the strains resistant to CV-N or ConA into a wild-type background fully reproduced the (cross-)resistance profiles of the originally selected strains, pointing to the impact of the N-glycan mutations on the phenotypic resistance profiles of both selected strains. cellular lipid bilayer (18). Recently, the fusion inhibitor enfuvirtide (T20) was approved by the Food and Drug Administration as the first HIV entry inhibitor to be used in humans (49). HIV-1 represents one of the most heavily glycosylated viruses known. The carbohydrate component of the major glycoprotein of HIV-1 gp120 comprises approximately half of the molecular weight of the complete molecule (26). This carbohydrate component plays an important role in the biological activities of the glycoprotein, i.e., folding, intracellular transport, receptor binding, and immune cell recognition (8, 33, 35, 46, 48). Due to the biological significance of carbohydrates in viral pathogenesis, the glycosylation step constitutes an attractive target for antiviral therapy (20, 22). Several compounds inhibiting the glycosylation process have therefore been evaluated for anti-HIV activity, and some glucosidase inhibitors have shown promising results (37, 38). Furthermore, the antiHIV activities of a variety of compounds binding directly to the carbohydrates on the virion surface have been analyzed. In this context, several plant lectins, including Urtica dioica agglutinin (3), Galanthus nivalis agglutinin (4), Astrocarpus heteophyllus agglutinin, Myrianthus holstii lectin (19), and concanavalin A

The current treatment of human immunodeficiency virus type 1 (HIV-1) infection focuses primarily on the inhibition of the viral enzymes reverse transcriptase (RT) and protease (PR). However, in many patients residual replication in the presence of the selective pressure of antiviral drugs allows the emergence of drug-resistant strains, finally resulting in therapeutic failure. Therefore, the development of new drugs, preferentially acting on new targets in the HIV replication cycle, remains an important issue. A potentially powerful target in addition to RT and PR is the first event in the viral replicative cycle, HIV entry. HIV entry involves the interaction of the viral protein gp120 with the CD4 receptor on the surface of the target cell and the subsequent interaction of gp120 with the coreceptor CCR5 (for R5 strains) or CXCR4 (for X4 strains). This interaction results in a conformational change of the viral glycoprotein gp41, wherein the interaction of heptad region 1 (HR1) and HR2 is followed by the fusion of the virus with the * Corresponding author. Mailing address: Department of Clinical Virology, University of Go ¨teborg, Guldhedsgatan 10B, S-413 46 Go ¨teborg, Sweden. Phone: 46 705 95 15 29. Fax: 46 31 82 70 32. E-mail: [email protected]. † Both authors contributed equally to this work. 7777

7778

WITVROUW ET AL.

(ConA) (28), have shown significant anti-HIV activities in vitro. The microbicide cyanovirin N (CV-N), an 11-kDa protein originally isolated from the cyanobacterium Nostoc ellipsosporum, was found to target high-mannose glycans in an oligosaccharide-dependent manner (10, 12, 40, 47). CV-N acts as a virus entry inhibitor and potently inhibits the infectivity of an extremely broad spectrum of HIV strains, including X4, R5, and X4/R5 strains and primary HIV-1 isolates as well as HIV-2 and simian immunodeficiency virus strains (12, 20). More “classical” mannose-binding lectins such as ConA, whose binding to N-glycans can be blocked by free monomeric mannose units (4), recognize the trimannose core of many N-linked glycans (41). In contrast, CV-N exhibits a high specificity for terminal Man␣(1-2)Man␣ moieties on high-mannose Man8 or Man9 glycans but does not bind glycans with Man7 or lower (6, 7, 11, 47). The high affinity of CV-N for high-mannose Nglycans suggests that the binding of CV-N to certain N-glycans on the virion surface may be the prime determinant of the anti-HIV activity of CV-N. However, the position of these N-glycans on the peptidyl main chain of the HIV glycoprotein(s) has not yet been identified. From studies of recombinant gp120 expressed in CHO cells, it has been shown that virtually all of the 23 N-glycosylation consensus sequences, N-X-S/T (X ⫽ any amino acid except P), are occupied by N-linked glycans consisting of high-mannose (HM), hybridtype, and complex-type (CT) structures. Approximately 11 of these glycans are suggested to have an HM or possibly HT structure and therefore represent potential binding targets for CV-N (34). For this study, we selected a CV-N-resistant HIV-1 strain in the presence of the HM-binding agent CV-N. Subsequent sequencing of this strain revealed the positions of the N-glycans within HIV-1 gp120 that served as targets for CV-N. Following a similar resistance selection procedure in the presence of the mannose-binding lectin ConA, we found that HIV-1 strains resistant to CV-N or ConA display high levels of cross-resistance towards one another. Both selected strains exhibited a variety of mutations eliminating N-linked glycans in gp120. The broadly HIV-neutralizing monoclonal antibody 2G12 recognizes an HM-glycan-dependent epitope on gp120 which is primarily composed of carbohydrate, probably with no direct involvement of the gp120 polypeptide surface. The epitope resides on a face orthogonal to the CD4 binding face on gp120, on a surface proximal to but distinct from that implicated in coreceptor binding (17, 29, 43, 44). We monitored the strains that were resistant to CV-N or ConA for their susceptibility to the inhibitory effect of 2G12 and other HIV entry inhibitors. Finally, the mutated gp160 genes of both selected strains were recombined in the genetic background of a wild-type proviral clone with a deletion of gp160. The subsequent recombined strains were analyzed phenotypically in order to assess the impact of the described mutations on the resistance phenotype of both selected strains. MATERIALS AND METHODS Compounds. Purified recombinant cyanovirin N, a cyanobacterial protein, was produced in Escherichia coli as reported previously (40). Concanavalin A was purchased from Sigma-Aldrich (Sweden). The HIV-neutralizing antibody 2G12 was a kind gift from C. Bewley, National Institutes of Health, Bethesda, Md.

J. VIROL. AMD3100 was provided by G. Bridger and G. Henson, AnorMED (Langley, British Columbia, Canada) and was synthesized as described previously (13). 3⬘-Azido-3⬘-deoxythymidine (AZT) was synthesized according to the method described by Horwitz et al. (30). Ritonavir (ABT538) was obtained from J. M. Leonard, Abbott Laboratories (Abbott Park, Ill.). Virus and cells. The HIV-1(NL4.3) strain was derived from the molecular clone pNL4.3 (1) (obtained from the National Institutes of Health, Bethesda, Md.) and was used as the wild-type strain in all resistance selection experiments. MT-4 cells (39) were cultured in RPMI 1640 supplemented with 10% heatinactivated fetal calf serum, 2 mM L-glutamine, 0.1% sodium bicarbonate, and 20 ␮g/ml gentamicin. Selection of resistant HIV-1 strains. Selection was initiated at a low multiplicity of infection (MOI; 0.01) with a drug concentration of one to five times its 50% effective concentration (EC50), as determined in an MT-4/MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay described elsewhere (42). Every 3 to 4 days, the culture was monitored for the appearance of an HIVinduced cytopathic effect (CPE). When CPE was observed, the cell-free culture supernatant was used to reinfect fresh, uninfected cells in the presence of an equal or higher concentration of the compound. When no virus breakthrough was observed, the infected cell culture was subcultivated in the presence of the same concentration of compound. The concentration of the compound was then incrementally increased. PCR amplification and sequencing of the envelope genes. (i) PCR amplification of gp160-encoding sequences. A 3,590-nucleotide fragment (corresponding to positions 5448 to 9037) was amplified by a PCR using the Expand High Fidelity PCR system. The PCR was performed on a Biometra Trioblock with the primers AV310 (5⬘-AGC AGG ACA TAA YAA GGT AGG-3⬘, corresponding to positions 5448 to 5468) and AV319 (5⬘-GCT SCC TTR TAA GTC ATT GGT CT-3⬘, corresponding to positions 9015 to 9037). The primer positions correspond to the HIV-1(LAV-1) recombinant clone pNL4.3 (GenBank accession no. M19921). The cycling conditions were as follows: a first denaturation step of 2 min at 95°C, followed by 40 cycles consisting of 15 s at 95°C, 30 s at 55°C, and 3 min 40 s at 68°C. A final extension was performed at 72°C for 10 min. (ii) Sequencing of gp120-encoding regions. PCR products were purified by use of a PCR purification kit (QIAGEN). For sequencing reactions, an ABI PRISM Dye Terminator cycle sequencing core kit (Perkin-Elmer, Brussels, Belgium) was used. The primers used to sequence the entire gp120 gene were as follows: AV304 (5⬘-ACA TGT GGA AAA ATG ACA TGG T-3⬘, corresponding to positions 6501 to 6522), AV305 (5⬘-GAG TGG GGT TAA TTT TAC ACA TGG-3⬘, corresponding to positions 6572 to 6595), AV306 (5⬘-TGT CAG CAC AGT ACA ATG TAC ACA-3⬘, corresponding to positions 6937 to 6960), AV307 (5⬘-TCT TCT TCT GCT AGA CTG CCA T-3⬘, corresponding to positions 6999 to 7020), AV308 (5⬘-TCC TCA GGA GGG GAC CCA GAA ATT-3⬘, corresponding to positions 7304 to 7327), AV309 (5⬘-CAR TAG AAA AAT TCY CCT CYA CA-3⬘, corresponding to positions 7346 to 7368), and AV313 (5⬘-TCC YTC ATA TYT CCT CCT CCA GGT C-3⬘, corresponding to positions 7620 to 7644). The samples were loaded onto an ABI PRISM 310 genetic analyzer (Perkin-Elmer). The sequences were analyzed with the program Geneworks 2.5.1 (Intelligenetics Inc., Oxford, United Kingdom). (iii) Sequencing of gp41-encoding region. The primers used to sequence the gp41 gene were as follows: AV322 (5⬘-AAG CAA TGT ATG CCC CTC C-3⬘, corresponding to positions 7509 to 7527), AV323 (5⬘-CTG CTC CYA AGA ACC CAA-3⬘, corresponding to positions 7771 to 7790), AV324 (5⬘-GGC AAA GAG AAG AGT GGT-3⬘, corresponding to positions 7714 to 7731), AV326 (5⬘-TTG GGG YTG CTC TGG AAA AC-3⬘, corresponding to positions 7999 to 8018), AV327 (5⬘-TTT TAT ATA CCA CAG CCA-3⬘, corresponding to positions 8246 to 8263), AV328 (5⬘-ATA ATG ATA GTA GGA GG-3⬘, corresponding to positions 8270 to 8286), AV329 (5⬘-GTC CCA GAA GTT CCA CA-3⬘, corresponding to positions 8557 to 8573), AV330 (5⬘-GGA RCC TGT GCC TCT TCA-3⬘, corresponding to positions 8496 to 8513), and AV331 (5⬘-TCT CAT TCT TTC CCT TA-3⬘, corresponding to positions 8833 to 8849). Drug susceptibility assay. The inhibitory effects of antiviral compounds on the HIV-induced CPE in human lymphocyte MT-4 cell cultures were determined by an MT-4/MTT assay as described elsewhere (42). This assay is based on the reduction of the yellow compound MTT by the mitochondrial dehydrogenase of metabolically active cells to a blue formazan derivative, which can be measured spectrophotometrically. The 50% cell culture infective doses (CCID50) of the HIV strains were determined by titration of the virus stocks in MT-4 cells. For drug susceptibility assays, MT-4 cells were infected with 100 to 300 CCID50 of the HIV strains in the presence of fivefold serial dilutions of antiviral drugs. The concentration of each compound achieving 50% protection against the CPE of HIV, defined as the EC50, was determined.

VOL. 79, 2005

HIV RESISTANCE TO MANNOSE BINDING LECTINS

7779

TABLE 1. Genotypic analysis of the gp120 genes of HIV-1(NL4.3) strains selected in the presence of CV-N or Con A Strain

Amino acid mutated in gp120

NL4.3/CV-Nres

N302K Deletion of aa 364–376 (TWFNSTWSTEGSN) N418S N204K S261F N302K T311I

NL4.3/ConAres

env recombination assay. MT-4 cells were subcultured at a density of 500,000 cells/ml 1 day prior to transfection. The cells were pelleted and resuspended in phosphate-buffered saline at a concentration of 6.25 ⫻ 106 cells/ml. For each transfection reaction, 5 ⫻ 106 cells (0.8 ml) were used. Transfections were performed by electroporation using an EASYJECT system (Eurogentec, Seraing, Belgium) and electroporation cuvettes (Eurogentec, Seraing, Belgium). For gp160 recombination experiments, MT-4 cells were cotransfected with 10 ␮g of a linearized NL4.3 clone with a deletion of gp160 and with 2 ␮g of the purified and concentrated AV310-AV319 PCR product (PCR purification kit; QIAGEN, Westburg, Leusden, The Netherlands). The electroporation conditions used for all transfections were 250 V and 1,500 ␮F. The transfected cell suspension was incubated in 5 ml of culture medium at 37°C in a humidified atmosphere with 5% CO2. The recombinant virus was harvested by centrifugation once a full CPE was microscopically observed in the culture (at about 8 days posttransfection) (25). Aliquots of 1 ml were stored at ⫺80°C for subsequent infectivity (CCID50) and drug susceptibility (EC50) determinations in the MT-4/MTT assay. Construction of HIV-1(NL4.3)gp120N302K mutant strain. Two mutant PCR fragments were constructed with the primer set N302Kupper (5⬘-TCTACTAA TTTTACAATGTGCT-3⬘, corresponding to positions 7236 to 7257 of NL4.3 and containing the mutation N302K) and AV310 (5⬘-AGC AGG ACA TAA T/CAA GGT AGG-3⬘, corresponding to positions 5447 to 5467 of NL4.3), the primer set N302Klower (5⬘-AGCACATTGTAAAATTAGTAGA-3⬘, corresponding to positions 7236 to 7257 of NL4.3 and containing the mutation N302K) and AV319 (5⬘-GCT G/C CC TTA/G TAA GTC ATT GGT CT-3⬘, corresponding to positions 9001 to 9024 of NL4.3), and DNA from MT-4 cells that were infected with HIV-1(NL4.3). The cycling conditions were as follows: a first denaturation step of 2 min at 95°C, followed by 40 cycles consisting of 45 s at 95°C, 45 s at 55°C, and 2 min at 68°C. A final extension step was performed at 72°C for 10 min. These PCR products were purified and concentrated with a PCR purification kit (QIAGEN, Westburg, Leusden, The Netherlands). In another PCR, the purified PCR products were annealed to each other by use of the primers AV311 (5⬘-CTA CTT TAT ATT TAT ATA ATT CAC TTC TCC-3⬘, corresponding to positions 7630 to 7659 of NL4.3) and AV312 (5⬘-AGA A/GGA C/TAG ATG GAA CAA GCC CCA G-3⬘, corresponding to positions 5549 to 5573 of NL4.3). The cycling conditions were as follows: a first denaturation step of 10 min at 95°C, followed by 40 cycles consisting of 30 s at 95°C, 30 s at 55°C, and 4 min at 68°C. A final extension step was performed at 72°C for 10 min. Topo cloning (Topo TA cloning kit for sequencing; Invitrogen, Merelbeke, Belgium) and sequencing of this PCR product were used to check for the mutation. Two micrograms of the purified and concentrated AV311-AV312 PCR product was used for recombination in MT-4 cells (as described earlier), using 10 ␮g of the linearized NL4.3 clone with a deletion of gp160. The recombinant mutant virus was harvested by centrifugation once a full CPE was microscopically observed in the culture (at about 6 days posttransfection). Aliquots of 1 ml were stored at ⫺80°C for subsequent infectivity (CCID50) and drug susceptibility (EC50) determinations in the MT-4/MTT assay. Fluorescence assay for CV-N blocking of HIV infection. TZM-b1 cells (3,000 per well) were plated in a 96-well flat-bottomed TC plate (Corning). TZM-b1 cells have an integrated copy of luciferase under the control of the HIV-1 promoter. The following day, the cell medium was aspirated and replaced with fresh medium containing CV-N (0.16 to 500 nM) for 90 min at 37°C. Triplicate wells were used for each CV-N concentration. The cell medium was then aspirated, and virus (100 50% tissue culture infective doses) was added (50 ␮l per well) for 3 days. The cells were then lysed with 100 ␮l of 1⫻ cell lysis buffer (Promega). Sixty microliters of cell lysate was used to determine the cell-associated luciferase levels by use of a Fluoroskan Ascent fluorometer. The percent neutralization for each CV-N concentration was determined.

Region flanking mutation

HC (N3K) IS NS (⌬T-N) NT SS (N3S) IT TF (N3K) GT NQ (S3F) YE HC (N3K) IS NN (T3I) TL

Position(s) of eliminated N-glycans

N302 N362, N367, N376 N418 N204 N259 N302 N309

RESULTS In vitro selection of HIV-1(NL4.3) strains in the presence of mannose-binding lectin CV-N or ConA. HIV-1 strains resistant to CV-N or ConA were selected by serial passages of HIV1(NL4.3) in the presence of increasing concentrations of CV-N or ConA, respectively. After 19 and 21 passages, respectively, the strains selected in the presence of CV-N or ConA were able to grow in the presence of compound concentrations of 0.5 and 13 ␮g/ml, respectively. These final concentrations were 125- and 5-fold higher than the concentrations required to inhibit the replication of wild-type HIV-1(NL4.3) by 50% (EC50) (0.004 and 2.6 ␮g/ml, respectively). Higher concentrations proved to be cytotoxic. The strains resistant to CV-N and ConA were designated NL4.3/CV-Nres and NL4.3/ConAres, respectively. Genotypic analysis of the env genes of HIV-1(NL4.3) strains selected in the presence of CV-N or ConA. Several mutations in the gp160-encoding region were detected in the NL4.3/CVNres and NL4.3/ConAres strains compared to the DNA sequence of the wild-type HIV-1(NL4.3) strain (Table 1). Part of the HIV-1(NL4.3) virus population that was passaged 19 times in the presence of CV-N carried an N418S mutation, whereas an N302K mutation and a deletion of TWFNSTWSTEGSN were present in the complete virus population. Part of the NL4.3/CV-Nres population contained a leucine residue at position 216 of gp41, whereas proline was present in the wild-type HIV-1(NL4.3) strain. After 21 passages in the presence of ConA, the mutations N204K, S261F, N302K, and T311I were present as mixtures with the wild-type amino acids in gp120 of NL4.3/ConAres (Table 1). No mutations were found in the gp41 gene of NL4.3/ConAres. Evaluation of phenotypic (cross-)resistance of HIV1(NL4.3) strains selected in the presence of CV-N or ConA. The antiviral activities of CV-N and ConA against strains HIV1(NL4.3), NL4.3/CV-Nres, and NL4.3/ConAres were determined in parallel with those of an HIV-neutralizing antibody (2G12), the viral entry inhibitors AMD3100 and enfuvirtide (T20), the reverse transcriptase inhibitor zidovudine (AZT), and the protease inhibitor ritonavir (Table 2). NL4.3/CV-Nres showed a 143-fold reduced susceptibility to CV-N. This strain also displayed cross-resistance to ConA (⬎8.8-fold) and 2G12 (⬎43.0-fold). The sensitivity of the NL4.3/ConAres strain to ConA was decreased ⬎9.3-fold. We could not evaluate higher concentrations of ConA because of its cytotoxicity. CV-N and 2G12 had decreases of 156- and ⬎46.1-fold, respectively, in their inhibitory effects against the replication of NL4.3/

50% effective concentration (EC50) or concentration required to inhibit the CPE of different HIV strains by 50% in MT-4 cells. Fold increase in the EC50 of the compound for in vitro-selected HIV-1(NL4.3) or the recombined selected strain compared to the EC50 of the compound for the parental HIV-1(NL4.3) strain or the recombined HIV-1(NL4.3) strain, respectively. c HIV-1(NL4.3) wild-type strain recombined with the gp160 gene of pNL4.3. d HIV-1(NL4.3) strain selected in vitro in the presence of CV-N or ConA. e HIV-1(NL4.3) wild-type strain recombined with the gp160 gene of the respective selected HIV-1(NL4.3) strain. b

a

0.009 ⫾ 0.0002 (0.5) 2.2 (1.1) 0.0004 ⫾ 0.0002 (0.6) 0.02 ⫾ 0.002 (0.6) 0.02 ⫾ 0.02 (1.7) 3.8 ⫾ 0.7 (2.6) 0.001 ⫾ 0.0004 (2.8) 0.04 ⫾ 0.007 (1.4) 0.02 ⫾ 0.02 (1.4) 2.7 ⫾ 0.8 (1.8) 0.001 ⫾ 0.0003 (2.6) 0.03 ⫾ 0.001 (1.2) 0.03 ⫾ 0.01 (1.3) 3.6 ⫾ 0.9 (1.7) 0.001 ⫾ 0.0004 (1.8) 0.04 ⫾ 0.02 (1.4) 0.02 ⫾ 0.01 2.1 ⫾ 1.1 0.0006 ⫾ 0.0003 0.03 ⫾ 0.007 Reference compounds AMD3100 Enfuvirtide Zidovudine Ritonavir

0.01 ⫾ 0.008 1.5 ⫾ 0.7 0.0005 ⫾ 0.0002 0.03 ⫾ 0.002

⬎40 (⬎40.0)

0.03 ⫾ 0.01 (1.5) 3.8 ⫾ 0.3 (1.8) 0.001 ⫾ 0.0004 (2.2) 0.04 ⫾ 0.01 (1.4)

0.006 ⫾ 0.002 (1.4) ⬎2.2 ⫾ 0.4 (⬎0.9) ⬎40 (⬎43.0) ⬎21.5 ⫾ 0.8 (⬎9.7) ⬎40 (⬎40.0) 0.004 ⫾ 0.002 2.6 ⫾ 1.4 0.9 ⫾ 0.2

0.003 ⫾ 0.001 2.2 ⫾ 0.06 1.0 ⫾ 0.1

0.2 ⫾ 0.01 (77)

0.6 ⫾ 0.3 (143) ⬎22.7 ⫾ 1.3 (⬎8.8) ⬎40 (⬎43.0)

0.6 ⫾ 0.4 (156) ⬎23.9 ⫾ 1.0 (⬎9.3) ⬎40 (⬎46.1)

J. VIROL.

Carbohydrate binding inhibitors CV-N ConA 2G12

R160/ConArese NL4.3/ConAresd R160/CV-Nrese HIV-1(NL4.3)

R160/NL4.3c

NL4.3/CV-Nresd

Strains resistant to ConA Strains resistant to CV-N WT strains Compound

EC50a(␮g/ml) (fold increase)b

TABLE 2. Susceptibility of selected and gp160-recombined HIV-1 strains to the inhibitory effect of various antiviral compounds

HIV-1(N302K)

WITVROUW ET AL.

Strains with mutated gp120

7780

ConAres (Table 2). Other compounds such as AMD3100, enfuvirtide, zidovudine, and ritonavir retained full activity against the different selected strains (Table 2). Evaluation of the susceptibility of HIV-1 strains with recombined gp160 genes selected in the presence of CV-N or ConA. We established a gp160 recombination assay wherein different gp160 genes could be recombined into a proviral HIV1(NL4.3) clone with a deletion of the gp160 gene (25). gp160 recombination was performed with the following strains: HIV1(NL4.3) and the HIV-1(NL4.3) strains selected in the presence of CV-N (NL4.3/CV-Nres) and ConA (NL4.3/ConAres). The strains with recombined gp160 are referred to as R160/ NL4.3, R160/CV-N, and R160/ConA, respectively. The susceptibilities of the strains with recombined gp160 to antiviral agents were determined by the standard MT-4/MTT assay (Table 2). The resistance of both parental selected strains in comparison to HIV-1(NL4.3) was reproduced after gp160 recombination, since the R160/CV-Nres and R160/ConAres strains displayed comparable losses in susceptibility to that of the wild-type recombined strain R160/NL4.3 for the inhibitory effects of CV-N, ConA, and 2G12. As expected, the compounds AMD3100, enfuvirtide, zidovudine, and ritonavir retained their full anti-HIV activity against both strains with recombined gp160 (Table 2). Evaluation of the susceptibility of a mutated HIV-1(N320K) strain expressing recombinant gp120. Wild-type HIV1(NL4.3) and a mutated HIV-1(N302K) strain expressing recombinant gp120 were compared for their susceptibility to antiviral drugs and 2G12 as described above. We found no resistance to CV-N and ConA for the HIV-1 strain lacking the N302 glycan (Fig. 1 and Table 2). This result was confirmed with an HIV-1(SF162)-based system. With this system, a mutated SF162 clone, lacking the N-glycan corresponding to the N302 glycan, and its parental wild-type strain were compared for their sensitivity to CV-N by the use of TZM-b1 cells (data not shown). DISCUSSION The glycosylation of gp120 represents an interesting aspect of HIV inhibition due to the role of carbohydrates in (co)receptor binding and evasion of the host immune response. The HIV-1 envelope glycoproteins display an extremely large amount of carbohydrate. More than half of the molecular weight of gp120 is due to N-linked glycans (26). Most broadly neutralizing antibodies are directed against functionally conserved gp120 regions involved in either CD4 or coreceptor binding and are known as CD4 binding site or CD4-induced antibodies, respectively. An inspection of the gp120 crystal structure reveals that although the receptor-binding regions lack glycosylation, sugar moieties lie proximal to both receptor-binding sites on gp120 and thus in proximity to both the CD4 binding site and CD4-induced antibody epitopes (53). Carbohydrates that flank receptor-binding regions on gp120 protect HIV-1 from antibody-mediated neutralization. Since the carbohydrates on viral glycoproteins are encoded by the host cell glycosylation machinery, the glycosylated viral surface appears as “self” to the host immune system. Indeed, the large extent of N-linked carbohydrates covering this so-called “silent face” of gp120 dramatically reduces its immunogenicity. The

VOL. 79, 2005

HIV RESISTANCE TO MANNOSE BINDING LECTINS

7781

FIG. 1. Carbohydrate positions in gp120 proteins of HIV strains resistant to CV-N or ConA in comparison with those in gp120 of the HIV-1(NL4.3) strain. (A) Wild-type NL4.3; (B) NL4.3/CV-Nres; (C) NL4.3/ConAres.

“silent face” of gp120 seldom elicits antibodies and thus appears to avoid the generation of potential neutralizing humoral responses (31, 32, 53). The massive carbohydrate exposure of the “silent face” of gp120 indicates that without protection as a result of glycosylation, this part of the protein may be highly vulnerable to HIV-neutralizing antibodies (8, 9, 14–16, 27, 36, 45, 46). The lectin cyanovirin (CV-N) efficiently inhibits the replication of HIV as a result of binding to large high-mannose (HM) glycans on the virion surface (7, 12, 47). In order to identify the positions of HM glycans on gp120 involved in the anti-HIV activity of CV-N, we investigated the development of HIV escape mutants by in vitro selection of CV-N-resistant HIV1(NL4.3) strains in the presence of increasing concentrations of CV-N. Since CV-N shows a very strict specificity for terminal Man␣1-2Man␣ moieties on Man8 or Man9 (7, 10, 47), we analyzed in parallel the development of resistance towards the HIV-inactivating mannose-binding lectin concanavalin A (ConA) (28). ConA is less restricted than CV-N regarding its binding specificity for HM glycans (2, 4). The HIV-1(NL4.3) strains that were selected in vitro in the presence of CV-N or ConA are referred to as NL4.3/CV-Nres and NL4.3/ConAres, respectively. Interestingly, all mutations found in the gp120 genes of both NL4.3/CV-Nres and NL4.3/ ConAres specifically eliminated N-glycosylation sites believed to harbor HM or hybrid-type glycans (34). Furthermore, these mutations appear to map to the “silent face” of gp120 in the NL4.3/CV-Nres and NL4.3/ConAres strains, based on the crystal structure of the gp120 core. Mutations associated with resistance to CV-N abolished the

glycosylation sites at positions N302 and N418. This indicates that the N-glycans at these sites are critical for the anti-HIV activity of CV-N and most likely correspond to large HM glycans (Man8 or Man9) exposing the terminal CV-N binding oligosaccharide structure Man␣(1-2)Man␣ (6, 10, 47). The 13amino-acid deletion observed in the V4 region of gp120 of NL4.3/CV-Nres eliminated the three N-linked glycans at positions N362, N367, and N376, of which only the glycan at position N362 is predicted to have an HM structure, while the others are predicted to be CT glycans. Deletions of N-glycancontaining peptide sequences within this part of the V4 region have previously been reported to be associated with HIV-1 resistance towards other nonlectin bicyclam entry inhibitors, suggesting that this part of gp120 constitutes an efficient target for entry inhibitors (21, 24, 51). However, the NL4.3/CV-Nres strain harboring this deletion did not display cross-resistance to the low-molecular-weight bicyclam AMD3100, showing that the CV-N-induced deletion in V4 was not sufficient to also render the virus resistant to AMD3100. The strain selected in the presence of ConA displayed four point mutations eliminating individual N-glycosylation sites in gp120 (Table 1; Fig. 1). Strains resistant to CV-N or ConA showed significant crossresistance towards one another, but only one mutation, N302K, was seen for both the NL4.3/CV-Nres and NL4.3/ ConAres strains. However, the elimination of the N302 glycan on its own was not enough to obtain resistance to CV-N or ConA, indicating that resistance to these lectins requires the elimination of other N-glycans from gp120 as well. It is still interesting, however, that the elimination of only a limited number of the 24 total N-glycans on gp120 was sufficient to

7782

WITVROUW ET AL.

render HIV-1 resistant to these lectins. The exact mechanism of lectin-mediated neutralization is not fully understood, but there are at least two potential ways, one direct and one indirect, to explain why the elimination of only a few glycans is sufficient for HIV-1 to become lectin resistant. The direct way is that the mutated N-glycosylation sites harbor the “key target glycans” for lectin-mediated neutralization and that these are eliminated by the mutations. Lectins binding to these glycans either block HIV binding to cellular receptors or interfere with postbinding conformational changes of gp120 (5). The indirect mechanism suggests that the elimination of one or more Nglycosylation sites renders N-glycans at other sites more exposed to carbohydrate-processing enzymes in the endoplasmic reticulum and the Golgi. Local changes in glycosylation site utilization can have global effects on the glycosylation setup and also on protein folding (52). Thus, lectin-binding highmannose glycans on the wild type might be more processed in the lectin-selected strains, resulting in a loss of their affinity for the high-mannose-binding lectins. To confirm that the HIV strains selected in the presence of high CV-N or ConA concentrations were indeed drug resistant, we determined the antiviral activities of CV-N and ConA against NL4.3/CV-Nres and NL4.3/ConAres. In parallel, the sensitivities of these strains to the HIV-neutralizing monoclonal antibody 2G12, targeting gp120 HM glycans, and to previously identified HIV inhibitors, i.e., the CXCR4 antagonist AMD3100, the fusion inhibitor enfuvirtide (T20), the reverse transcriptase inhibitor zidovudine (AZT), and the protease inhibitor ritonavir, were determined (Table 2). Both NL4.3/ CV-Nres and NL4.3/ConAres displayed pronounced (cross-)resistance to CV-N, ConA, and 2G12. The other inhibitors evaluated (AMD3100, enfuvirtide, zidovudine, and ritonavir) retained their full activity against both strains (Table 2). For an evaluation of the impact of the described mutations in gp160 on the observed phenotype, recombination of the mutated gp160 genes into a wild-type background seemed appropriate. Therefore, we used a gp160 recombination assay, wherein a gp160 gene can be recombined into a proviral HIV1(NL4.3) clone with a deletion of the gp160 gene, as previously described (25). gp160 recombination was performed with the viral strains resistant to CV-N and ConA. The loss of antiviral susceptibility of the different strains with recombined gp160 to the compounds CV-N, ConA, and 2G12 with respect to the wild-type recombined strain mirrored the decreased sensitivity of the corresponding parental selected strains for these compounds (Table 2). These data indicate that the observed mutations in gp160 are fully responsible for the observed phenotypic resistance of both selected strains. The 2G12 epitope comprises carbohydrate structures on Nlinked glycans located within or in close connection to the “silent face” of gp120. The glycans at positions N302 and N309 have previously been reported to represent a critical part of the 2G12 epitope (31, 32, 44, 53), explaining the observed crossresistance of NL4.3/CV-Nres and NL4.3/ConAres towards this monoclonal antibody. Both CV-N and 2G12 recognize terminal Man1-2Man structures, and the capability of these agents to bind to the structure formed by the two closely located N302 and N309 glycans in the “silent face” of gp120 may offer an explanation for their uniquely broad neutralizing capacities (Fig. 2). However, CV-N probably binds to a larger number of

J. VIROL.

FIG. 2. Surface model of the core structure of HIV-1 gp120 (gray) bound to a ribbon diagram of CD4 (green). Deletion mutations (blue) in the V4 loop and point mutations (red) as a consequence of HIV resistance development to CV-N result in the loss of glycosylation sites in the gp120 envelope glycoprotein. The red arrow indicates the point mutation at residue 302 near the V3 loop that abolishes an important glycosylation site from the gp120 molecule.

N-glycans on gp120 than does 2G12, since CV-N is capable of blocking the binding of 2G12 to gp120, but not vice versa (23, 44). Taken together, we have identified N-glycans on gp120 that are involved in the anti-HIV activity of CV-N and ConA by the selection of CV-N- and ConA-resistant HIV-1 strains. The herein identified N-glycans map to or close to the highly glycosylated “silent face” of gp120 (Fig. 2). Mannose-specific lectins display novel resistance profiles on HIV compared to what is known for existing compounds against HIV infection and are therefore of interest in the development of new antiviral therapies (5). The elimination of epitope-shielding N-glycans in the vulnerable silent face region of gp120 by a lectin-resistant virus might cause difficulties for the virus since it runs the risk of becoming more sensitive to neutralizing antibodies (16, 36, 45). As shown here, CV-N differs from 2G12 in that 2G12 only requires the single N302 glycan for resistance while CV-N requires additional mutations for resistance, suggesting that CV-N might constitute a more broadly effective microbicide than 2G12. Recently, CV-N has been shown to have antiviral effects in in vivo as well as in vitro models, and it has been suggested as a candidate for testing in humans (50). For future therapeutic or prophylactic purposes, it may be worth considering the development of nontoxic HM-binding virucides, such as CV-N, and/or HM-binding 2G12-like antibodies with the capacity to recognize N-glycans on the vulnerable “silent face” of gp120. ACKNOWLEDGMENTS These investigations were supported in part by the Stimulating Flemish Participation in EU Research Programs (Verkennende Internationale Samenwerking, VIS A6895, Vlaamse Gemeenschap). Anke Hantson was funded by a grant from the Flemish Institute supporting Scientific-Technological Research in Industry (IWT). We are grateful to Istvan Botos for graphical assistance with modeling the mutations on gp120 and to Barbara Van Remoortel and

VOL. 79, 2005 Liesbet De Dier for their excellent technical assistance. We also thank Carole Bewley for the donation of the HIV-neutralizing antibody 2G12. REFERENCES 1. Adachi, A., H. E. Gendelman, S. Koenig, T. Folks, R. Willey, A. Rabson, and M. A. Martin. 1986. Production of acquired immunodeficiency syndromeassociated retrovirus in human and nonhuman cells transfected with an infectious molecular clone. J. Virol. 59:284–291. 2. Baenziger, J. U., and D. Fiete. 1979. Structural determinants of concanavalin A specificity for oligosaccharides. J. Biol. Chem. 254:2400–2407. 3. Balzarini, J., J. Neyts, D. Schols, M. Hosoya, E. Van Damme, W. Peumans, and E. De Clercq. 1992. The mannose-specific plant lectins from Cymbidium hybrid and Epipactis helleborine and the (N-acetylglucosamine)n-specific plant lectin from Urtica dioica are potent and selective inhibitors of human immunodeficiency virus and cytomegalovirus replication in vitro. Antiviral Res. 18:191–207. 4. Balzarini, J., D. Schols, J. Neyts, E. Van Damme, W. Peumans, and E. De Clercq. 1991. Alpha-(1–3)- and alpha-(1–6)-D-mannose-specific plant lectins are markedly inhibitory to human immunodeficiency virus and cytomegalovirus infections in vitro. Antimicrob. Agents Chemother. 35:410–416. 5. Balzarini, J., K. Van Laethem, S. Hatse, K. Vermeire, E. De Clercq, W. Peumans, E. Van Damme, A. M. Vandamme, A. Bolmstedt, and D. Schols. 2004. Profile of resistance of human immunodeficiency virus to mannosespecific plant lectins. J. Virol. 78:10617–10627. 6. Bewley, C. A. 2001. Solution structure of a cyanovirin-N:Man alpha 1-2Man alpha complex: structural basis for high-affinity carbohydrate-mediated binding to gp120. Structure (Cambridge) 9:931–940. 7. Bewley, C. A., and S. Otero-Quintero. 2001. The potent anti-HIV protein cyanovirin-N contains two novel carbohydrate binding sites that selectively bind to Man(8) D1D3 and Man(9) with nanomolar affinity: implications for binding to the HIV envelope protein gp120. J. Am. Chem. Soc. 123:3892– 3902. 8. Bolmstedt, A., S. Olofsson, E. Sjogren-Jansson, S. Jeansson, I. Sjoblom, L. Akerblom, J. E. Hansen, and S. L. Hu. 1992. Carbohydrate determinant NeuAc-Gal beta (1–4) of N-linked glycans modulates the antigenic activity of human immunodeficiency virus type 1 glycoprotein gp120. J. Gen. Virol. 73:3099–3105. 9. Bolmstedt, A., S. Sjolander, J. E. Hansen, L. Akerblom, A. Hemming, S. L. Hu, B. Morein, and S. Olofsson. 1996. Influence of N-linked glycans in V4-V5 region of human immunodeficiency virus type 1 glycoprotein gp160 on induction of a virus-neutralizing humoral response. J. Acquir. Immune Defic. Syndr. Hum. Retrovirol. 12:213–220. 10. Bolmstedt, A. J., B. R. O’Keefe, S. R. Shenoy, J. B. McMahon, and M. R. Boyd. 2001. Cyanovirin-N defines a new class of antiviral agent targeting N-linked, high-mannose glycans in an oligosaccharide-specific manner. Mol. Pharmacol. 59:949–954. 11. Botos, I., B. R. O’Keefe, S. R. Shenoy, L. K. Cartner, D. M. Ratner, P. H. Seeberger, M. R. Boyd, and A. Wlodawer. 2002. Structures of the complexes of a potent anti-HIV protein cyanovirin-N and high mannose oligosaccharides. J. Biol. Chem. 277:34336–34342. 12. Boyd, M. R., K. R. Gustafson, J. B. McMahon, R. H. Shoemaker, B. R. O’Keefe, T. Mori, R. J. Gulakowski, L. Wu, M. I. Rivera, C. M. Laurencot, M. J. Currens, J. H. Cardellina II, R. W. Buckheit, Jr., P. L. Nara, L. K. Pannell, R. C. Sowder II, and L. E. Henderson. 1997. Discovery of cyanovirin-N, a novel human immunodeficiency virus-inactivating protein that binds viral surface envelope glycoprotein gp120: potential applications to microbicide development. Antimicrob. Agents Chemother. 41:1521–1530. 13. Bridger, G. J., R. T. Skerlj, D. Thornton, S. Padmanabhan, S. A. Martellucci, G. W. Henson, M. J. Abrams, N. Yamamoto, K. De Vreese, R. Pauwels, et al. 1995. Synthesis and structure-activity relationships of phenylenebis(methylene)-linked bis-tetraazamacrocycles that inhibit HIV replication. Effects of macrocyclic ring size and substituents on the aromatic linker. J. Med. Chem. 38:366–378. 14. Burton, D. R. 1997. A vaccine for HIV type 1: the antibody perspective. Proc. Natl. Acad. Sci. USA 94:10018–10023. 15. Burton, D. R., and P. W. Parren. 2000. Vaccines and the induction of functional antibodies: time to look beyond the molecules of natural infection? Nat. Med. 6:123–125. 16. Burton, D. R., R. A. Williamson, and P. W. Parren. 2000. Antibody and virus: binding and neutralization. Virology 270:1–3. 17. Calarese, D. A., C. N. Scanlan, M. B. Zwick, S. Deechongkit, Y. Mimura, R. Kunert, P. Zhu, M. R. Wormald, R. L. Stanfield, K. H. Roux, J. W. Kelly, P. M. Rudd, R. A. Dwek, H. Katinger, D. R. Burton, and I. A. Wilson. 2003. Antibody domain exchange is an immunological solution to carbohydrate cluster recognition. Science 300:2065–2071. 18. Chan, D. C., and P. S. Kim. 1998. HIV entry and its inhibition. Cell 93:681– 684. 19. Charan, R. D., M. H. Munro, B. R. O’Keefe, R. Sowder, T. C. McKee, M. J. Currens, L. K. Pannell, and M. R. Boyd. 2000. Isolation and characterization of Myrianthus holstii lectin, a potent HIV-1 inhibitory protein from the plant Myrianthus holstii(1). J. Nat. Products 63:1170–1174.

HIV RESISTANCE TO MANNOSE BINDING LECTINS

7783

20. De Clercq, E. 2002. Highlights in the development of new antiviral agents. Mini Rev. Med. Chem. 2:163–175. 21. de Vreese, K., V. Kofler-Mongold, C. Leutgeb, V. Weber, K. Vermeire, S. Schacht, J. Anne, E. de Clercq, R. Datema, and G. Werner. 1996. The molecular target of bicyclams, potent inhibitors of human immunodeficiency virus replication. J. Virol. 70:689–696. 22. Dwek, R. A., T. D. Butters, F. M. Platt, and N. Zitzmann. 2002. Targeting glycosylation as a therapeutic approach. Nat. Rev. Drug Discov. 1:65–75. 23. Esser, M. T., T. Mori, I. Mondor, Q. J. Sattentau, B. Dey, E. A. Berger, M. R. Boyd, and J. D. Lifson. 1999. Cyanovirin-N binds to gp120 to interfere with CD4-dependent human immunodeficiency virus type 1 virion binding, fusion, and infectivity but does not affect the CD4 binding site on gp120 or soluble CD4- induced conformational changes in gp120. J. Virol. 73:4360– 4371. 24. Este, J. A., D. Schols, K. De Vreese, K. Van Laethem, A. M. Vandamme, J. Desmyter, and E. De Clercq. 1997. Development of resistance of human immunodeficiency virus type 1 to dextran sulfate associated with the emergence of specific mutations in the envelope gp120 glycoprotein. Mol. Pharmacol. 52:98–104. 25. Fikkert, V., P. Cherepanov, K. Van Laethem, A. Hantson, B. Van Remoortel, C. Pannecouque, E. De Clercq, Z. Debyser, A. M. Vandamme, and M. Witvrouw. 2002. env chimeric virus technology for evaluating human immunodeficiency virus susceptibility to entry inhibitors. Antimicrob. Agents Chemother. 46:3954–3962. 26. Geyer, H., C. Holschbach, G. Hunsmann, and J. Schneider. 1988. Carbohydrates of human immunodeficiency virus. Structures of oligosaccharides linked to the envelope glycoprotein 120. J. Biol. Chem. 263:11760–11767. 27. Hansen, J. E. 1992. Carbohydrates of human immunodeficiency virus. APMIS 27(Suppl.):96–108. 28. Hansen, J. E., C. M. Nielsen, C. Nielsen, P. Heegaard, L. R. Mathiesen, and J. O. Nielsen. 1989. Correlation between carbohydrate structures on the envelope glycoprotein gp120 of HIV-1 and HIV-2 and syncytium inhibition with lectins. AIDS 3:635–641. 29. Hong, P. W., K. B. Flummerfelt, A. De Parseval, K. Gurney, J. H. Elder, and B. Lee. 2002. Human immunodeficiency virus envelope (gp120) binding to DC-SIGN and primary dendritic cells is carbohydrate dependent but does not involve 2G12 or cyanovirin binding sites: implications for structural analyses of gp120-DC-SIGN binding. J. Virol. 76:12855–12865. 30. Horwitz, J. P., J. Chua, and M. Noel. 1964. The monomesylates of 1-(2⬘deoxy-␤-D-lyxofuranosyl)thymidine. J. Org. Chem. 29:2076–2078. 31. Kwong, P. D., R. Wyatt, E. Desjardins, J. Robinson, J. S. Culp, B. D. Hellmig, R. W. Sweet, J. Sodroski, and W. A. Hendrickson. 1999. Probability analysis of variational crystallization and its application to gp120, the exterior envelope glycoprotein of type 1 human immunodeficiency virus (HIV-1). J. Biol. Chem. 274:4115–4123. 32. Kwong, P. D., R. Wyatt, J. Robinson, R. W. Sweet, J. Sodroski, and W. A. Hendrickson. 1998. Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 393:648–659. 33. LaBranche, C. C., T. L. Hoffman, J. Romano, B. S. Haggarty, T. G. Edwards, T. J. Matthews, R. W. Doms, and J. A. Hoxie. 1999. Determinants of CD4 independence for a human immunodeficiency virus type 1 variant map outside regions required for coreceptor specificity. J. Virol. 73:10310–10319. 34. Leonard, C. K., M. W. Spellman, L. Riddle, R. J. Harris, J. N. Thomas, and T. J. Gregory. 1990. Assignment of intrachain disulfide bonds and characterization of potential glycosylation sites of the type 1 recombinant human immunodeficiency virus envelope glycoprotein (gp120) expressed in Chinese hamster ovary cells. J. Biol. Chem. 265:10373–10382. 35. Li, Y., L. Luo, N. Rasool, and C. Y. Kang. 1993. Glycosylation is necessary for the correct folding of human immunodeficiency virus gp120 in CD4 binding. J. Virol. 67:584–588. 36. Malenbaum, S. E., D. Yang, L. Cavacini, M. Posner, J. Robinson, and C. Cheng-Mayer. 2000. The N-terminal V3 loop glycan modulates the interaction of clade A and B human immunodeficiency virus type 1 envelopes with CD4 and chemokine receptors. J. Virol. 74:11008–11016. 37. Mehta, A., T. M. Block, and R. A. Dwek. 1998. The role of N-linked glycosylation in the secretion of hepatitis B virus. Adv. Exp. Med. Biol. 435:195– 205. 38. Mehta, A., N. Zitzmann, P. M. Rudd, T. M. Block, and R. A. Dwek. 1998. Alpha-glucosidase inhibitors as potential broad based anti-viral agents. FEBS Lett. 430:17–22. 39. Miyoshi, I., H. Taguchi, I. Kobonishi, S. Yoshimoto, Y. Ohtsuki, and T. Akagi. 1982. Type C virus-producing cell lines derived from adult T cell leukemia. Gann. Monogr. 28:219–228. 40. Mori, T., K. R. Gustafson, L. K. Pannell, R. H. Shoemaker, L. Wu, J. B. McMahon, and M. R. Boyd. 1998. Recombinant production of cyanovirin-N, a potent human immunodeficiency virus-inactivating protein derived from a cultured cyanobacterium. Protein Expr. Purif. 12:151–158. 41. Naismith, J. H., and R. A. Field. 1996. Structural basis of trimannoside recognition by concanavalin A. J. Biol. Chem. 271:972–976. 42. Pauwels, R., J. Balzarini, M. Baba, R. Snoeck, D. Schols, P. Herdewijn, J. Desmyter, and E. De Clercq. 1988. Rapid and automated tetrazolium-based

7784

43.

44.

45. 46.

47.

48.

WITVROUW ET AL.

colorimetric assay for the detection of anti-HIV compounds. J. Virol. Methods 20:309–321. Sanders, R. W., M. Venturi, L. Schiffner, R. Kalyanaraman, H. Katinger, K. O. Lloyd, P. D. Kwong, and J. P. Moore. 2002. The mannose-dependent epitope for neutralizing antibody 2G12 on human immunodeficiency virus type 1 glycoprotein gp120. J. Virol. 76:7293–7305. Scanlan, C. N., R. Pantophlet, M. R. Wormald, E. Ollmann Saphire, R. Stanfield, I. A. Wilson, H. Katinger, R. A. Dwek, P. M. Rudd, and D. R. Burton. 2002. The broadly neutralizing anti-human immunodeficiency virus type 1 antibody 2G12 recognizes a cluster of alpha132 mannose residues on the outer face of gp120. J. Virol. 76:7306–7321. Schonning, K. 2003. Antibody neutralization of human immunodeficiency virus type 1 (HIV-1). APMIS 111(Suppl.):1–42. Schonning, K., A. Bolmstedt, J. Novotny, O. S. Lund, S. Olofsson, and J. E. Hansen. 1998. Induction of antibodies against epitopes inaccessible on the HIV type 1 envelope oligomer by immunization with recombinant monomeric glycoprotein 120. AIDS Res. Hum. Retrovir. 14:1451–1456. Shenoy, S. R., B. R. O’Keefe, A. J. Bolmstedt, L. K. Cartner, and M. R. Boyd. 2001. Selective interactions of the human immunodeficiency virus-inactivating protein cyanovirin-N with high-mannose oligosaccharides on gp120 and other glycoproteins. J. Pharmacol. Exp. Ther. 297:704–710. Sjolander, S., A. Bolmstedt, L. Akerblom, P. Horal, S. Olofsson, B. Morein, and A. Sjolander. 1996. N-linked glycans in the CD4-binding domain of human immunodeficiency virus type 1 envelope glycoprotein gp160 are es-

J. VIROL.

49.

50.

51.

52.

53.

sential for the in vivo priming of T cells recognizing an epitope located in their vicinity. Virology 215:124–133. Tashima, K. T., and C. C. Carpenter. 2003. Fusion inhibition—a major but costly step forward in the treatment of HIV-1. N. Engl. J. Med. 348:2249– 2250. Tsai, C. C., P. Emau, Y. Jiang, M. B. Agy, R. J. Shattock, A. Schmidt, W. R. Morton, K. R. Gustafson, and M. R. Boyd. 2004. Cyanovirin-N inhibits AIDS virus infections in vaginal transmission models. AIDS Res. Hum. Retrovir. 20:11–18. Witvrouw, M., V. Fikkert, W. Pluymers, B. Matthews, K. Mardel, D. Schols, J. Raff, Z. Debyser, E. De Clercq, G. Holan, and C. Pannecouque. 2000. Polyanionic (i.e., polysulfonate) dendrimers can inhibit the replication of human immunodeficiency virus by interfering with both virus adsorption and later steps (reverse transcriptase/integrase) in the virus replicative cycle. Mol. Pharmacol. 58:1100–1108. Wu, Z., S. C. Kayman, W. Honnen, K. Revesz, H. Chen, S. Vijh-Warrier, S. A. Tilley, J. McKeating, C. Shotton, and A. Pinter. 1995. Characterization of neutralization epitopes in the V2 region of human immunodeficiency virus type 1 gp120: role of glycosylation in the correct folding of the V1/V2 domain. J. Virol. 69:2271–2278. Wyatt, R., P. D. Kwong, E. Desjardins, R. W. Sweet, J. Robinson, W. A. Hendrickson, and J. G. Sodroski. 1998. The antigenic structure of the HIV gp120 envelope glycoprotein. Nature 393:705–711.