Share this article on:

Envelope V3 amino acid sequence predicts HIV-1 phenotype (co-receptor usage and tropism for macrophages)

Briggs, Daniel R.a; Tuttle, Daniel L.b; Sleasman, John W.ab; Goodenow, Maureen M.ab

Research Letters

Departments of aPediatrics and bPathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida, Gainesville, FL, USA.

Sponsorship: This work was supported by PHS awards HD32259, HL58005, and AI39015. D.R.B. was supported by the Laura McClamma Research Fellowship Award. D.L.T. was an Elizabeth Glaser Pediatric AIDS Foundation Scholar and was supported by NIH grant CA09126.

Received 28 July 2000; accepted: 1 September 2000.

D.R.B and D.L.T. contributed equally in the conception, research, and preparation of this manuscript.

The phenotypic classification of HIV-1 has been based on biological characteristics of viral isolates in culture, which include replication in MT-2 cells (syncytium inducing; SI versus non-syncytium inducing; NSI), replication kinetics in peripheral blood mononuclear cells (rapid–high versus slow–low growth), or host cell range (macrophage tropic versus T cell line tropic) [1,2]. Assessing the HIV-1 phenotype in vitro provides a link to pathogenesis in vivo. The V3 genotype, based on amino acid sequence of envelope (gp120SU), is related to the NSI/SI phenotype and provides a biomarker for disease progression [3,4]. In general, NSI variants are associated with acute infection and display low V3 net charges (≤ + 4), whereas SI variants emerge late in the disease and display high V3 net charges (≥ + 5) [5–7].

An improved system to classify the HIV-1 phenotype is based on chemokine receptor usage, particularly CXCR4 (X4) and CCR5 (R5), and tropism for different cell types in culture. This classification system distinguishes dual tropic viruses that infect both macrophages and T cell lines, as well as viruses that use both R5 and X4 among SI strains [8–12]. Although the use of X4 as a co-receptor and the infection of T cell lines is correlated with advanced disease [7], a relationship between V3 genotype and viral phenotype, as defined by co-receptor usage and tropism, is unclear.

We developed a model for predicting viral phenotype (i.e. co-receptor usage and tropism) based on genotypic variables within the V3 domain of gp120SU. V3 sequences were evaluated for 43 subtype B HIV-1 isolates with known co-receptor usage (Table 1). Cell tropism was available for 17 of the 43 isolates [13–16; unpublished data]. The position and charge of V3 amino acids between Cys263 and Cys296 [4] in relation to the actual phenotype were defined for each virus. The actual phenotype was assigned a numerical value: 1 = R5 and macrophage tropic (M-R5); 2 = X4 alone or in combination with R5 and dual tropic (D-X4 or D-R5X4); or 3 = X4 and T cell line tropic (T-X4). Using multiple linear regression (SigmaStat; SPSS Inc., Chicago, IL, USA), a combination of four genotypic variables in V3 were identified as positive predictors of viral phenotype: (i) number of positively charged amino acid residues (K or R); (ii) number of negatively charged residues (D or E); (iii) net V3 charge [(K + R) − (D + E)]; and (iv) an isoleucine residue at position 292. An equation to predict viral phenotype based on these genetic variables was generated:EQUATION



Table 1

Table 1

Calculated values for predicted phenotypes ranged from 0.7 to 3.4 (Table 1 and Table 2), which were rounded arithmetically. Predicted phenotype values between 0.5 and 1.4 were equated to actual phenotype 1; predicted values between 1.5 and 2.4 equalled actual phenotype 2; and predicted values between 2.5 and 3.4 equalled actual phenotype 3. The net positive charge of V3 among the viruses was a continuum ranging from 2 to 8. The V3 net positive charge between 2 and 4 was invariably related to the M-R5 phenotype (actual phenotype 1). In contrast, V3 net positive charges between 5 and 8 could be associated with D-X4, D-R5X4, or T-X4 (actual phenotypes 2 or 3). The inclusion of absolute numbers of positively and negatively charged amino acids and the presence of Ile292 among the genotypic variables provided increased accuracy to predict phenotypes 2 or 3. When four genetic variables were included, the equation predicted with 100% accuracy the phenotype of all viruses in the initial data set (Table 1).

Table 2

Table 2

Next, the model was applied to predict the phenotype in an independent data set that included V3 sequence, co-receptor usage, and tropism determinations from studies in our laboratory of 24 viral isolates [14; unpublished data] (Table 2). The V3 charge ranged from +2 to +7, whereas actual phenotypes were 1 M-R5, or 2 either D-R5X4 or D-X4. The T-X4 phenotype was not detected among the primary isolates. The model predicted the correct phenotype for 22 of 24 isolates. Only two viruses, 1483-2 and 1907-2 (predicted 1.5, actual 1), were misclassified by the model. Both viruses displayed a rare combination of the M-R5 phenotype with a V3 net charge of +5. Overall, the model to predict phenotype was accurate for over 91% of viruses evaluated.

The predictive value of the model could by enhanced by variables such as the usage of co-receptors other than R5 and X4 or additional genetic determinants within or outside of V3. Nevertheless, the model as currently constituted, is accurate and easily applied to viruses with known V3 genotypes.

Back to Top | Article Outline


The authors would like to thank Jonathon L. Arnold, Brant R. Burkhardt, Paul P. Poole, and Gregory S. Taylor for helpful discussions and for their invaluable assistance in generating the data included in this study.

Daniel R. Briggsa

Daniel L. Tuttleb

John W. Sleasmanab

Maureen M. Goodenowab

Back to Top | Article Outline


1. Cheng-Mayer C, Seto D, Tateno M, Levy JA. Biologic features of HIV-1 that correlate with virulence in the host. Science 1988, 240: 80 –82.
2. Spencer LT, Ogino MT, Dankner WM, Spector SA. Clinical significance of human immunodeficiency virus type 1 phenotypes in infected children. J Infect Dis 1994, 169: 491 –495.
3. Milich L, Margolin B, Swanstrom R. V3 loop of the human immunodeficiency virus type 1 Env protein: interpreting sequence variability. J Virol 1993, 67: 5623 –5634.
4. Hung CS, Vander HN, Ratner L. Analysis of the critical domain in the V3 loop of human immunodeficiency virus type 1 gp120 involved in CCR5 utilization. J Virol 1999, 73: 8216 –8226.
5. Distler O, McQueen PW, Tsang ML. et al. Primary structure of the V3 region of gp120 from sequential human immunodeficiency virus type 1 isolates obtained from patients from the time of seroconversion. J Infect Dis 1995, 172: 1384 –1387.
6. Innocenti-Francillard P, Brengel K, Guillon C, Mallet F, Morand P, Gruters R, Seigneurin JM. Blood monocytes infected in vivo by HIV-1 variants with a syncytium-inducing phenotype. AIDS Res Hum Retroviruses 1994, 10: 683 –690.
7. Connor RI, Sheridan KE, Ceradini D, Choe S, Landau NR. Change in coreceptor use correlates with disease progression in HIV-1-infected individuals. J Exp Med 1997, 185: 621 –628.
8. Lathey JL, Brambilla D, Goodenow MM. et al. Co-receptor usage was more predictive than NSI/SI phenotype for HIV replication in macrophages: is NSI/SI phenotyping sufficient? J Leukoc Biol 2000, 68: 324 –330.
9. Broder CC, Collman RG. Chemokine receptors and HIV. J Leukoc Biol 1997, 62: 20 –29.
10. Berger EA, Doms RW, Fenyo EM. et al. A new classification for HIV-1. Nature 1998, 391: 240. 240.
11. Yi Y, Isaacs SN, Williams DA. et al. Role of CXCR4 in cell–cell fusion and infection of monocyte-derived macrophages by primary human immunodeficiency virus type 1 (HIV-1) strains: two distinct mechanisms of HIV-1 dual tropism. J Virol 1999, 73: 7117 –7125.
12. Tuttle DL, Harrison JK, Anders C, Sleasman JW, Goodenow MM. Expression of CCR5 increases during monocyte differentiation and directly mediates macrophage susceptibility to infection by human immunodeficiency virus type 1. J Virol 1998, 72: 4962 –4969.
13. Speck RF, Wehrly K, Platt EJ. et al. Selective employment of chemokine receptors as human immunodeficiency virus type 1 coreceptors determined by individual amino acids within the envelope V3 loop. J Virol 1997, 71: 7136 –7139.
14. Acquino de Jesus MJ, Anders C, Miller G, Sleasman JW, Goodenow MM, Andiman WA. Genetically and epidemiologically related ‘non-syncytium-inducing’ isolates of HIV-1 display hetergeneous growth patterns in macrophages. J Med Virol 2000, 61: 171 –180.
15. AIDS Research and Reference Reagent Program. Division of AIDS, NIAID, National Institutes of Health.
16. Los Alamos National Laboratory HIV Sequence Database. 2000. (
© 2000 Lippincott Williams & Wilkins, Inc.