Introduction
Heterosexual transmission is a major route for HIV-1 spread. It has been estimated that 80% of HIV-1 infections in Africa are acquired through heterosexual contacts, but heterosexual transmission is also increasing in those industrialized countries where injecting drug users (IDU), usually engaging in heterosexual relationships, are the main risk group for AIDS [1]. In order to identify effective strategies for the control of the HIV-1 epidemic, it is important to learn about the conditions that favour heterosexual transmission.
Although HIV-1 has been isolated from semen and cervico-vaginal fluid of infected subjects [2,3], a clear correlation between the frequency of sexual exposure and the transmission of the infection has not been found. In fact, some subjects may be exposed for several years and still remain seronegative, whereas others become infected after a few instances of sexual intercourse. Several studies analysed the association of sexual transmission with clinical and behavioural cofactors, such as the type of sexual exposure, the clinical and immunological conditions of the infected partner, concomitant antiretroviral treatments and the coexistence of other sexually transmitted diseases [4-9], and came to the conclusion that each of these factors may affect HIV-1 heterosexual transmission. In addition, we recently demonstrated that the biological properties of the virus [10] as well as host defence mechanisms, detected as reactivity against selected V3 loop peptides [11], may play a role in sexual transmission.
In this study, we examined the presence of neutralizing antibodies, viral load and biological phenotype of HIV-1 isolates derived from HIV-1-seropositive transmitting and non-transmitting male index cases. Sera were tested for their capacity to neutralize autologous and heterologous primary HIV-1 isolates. Virus isolates were studied for replicative capacity in peripheral blood mononuclear cells (PBMC) and MT-2 cells. The results show that the neutralizing antibody response in the host, the viral load and the replicative capacity of the virus may all influence the outcome of male-to-female sexual transmission.
Materials and methods
Study population
Fourteen couples were selected from a larger group enrolled in a continuous cohort study on heterosexual transmission at the Clinic of Infectious Diseases, University of Bari, Bari, Italy, and included in this study. The criteria of eligibility included heterosexual orientation, a documented HIV-1 infection (index cases), and a long-term heterosexual partnership with an initially seronegative individual (partner) who had not been exposed to any other known risk factor for HIV-1 infection and who had unprotected sex for at least 1 year before blood sampling.
The 14 couples selected for this study had no clinical signs of sexually transmitted diseases during the observation period. Index cases were male IDU and were divided into two groups (seven transmitting and seven non-transmitting index cases) according to the serological status of their partners. Partners were tested for HIV-1-specific antibodies every 6 months. Blood samples from the HIV-1-positive index cases were collected every 12 months, and PBMC and sera were stored frozen. For the purpose of the present study, virus isolates and serum samples from the transmitting index cases were obtained simultaneously with the diagnosis of the partner's seroconversion whereas, in the group of non-transmitting index cases, virus isolations were performed from samples collected 9-12 months prior to confirmed seronegativity of the partner.
Quantification of HIV-1 plasma RNA
Plasma HIV-1 RNA was measured by branched DNA (bDNA) signal amplification-based hybridization (Quantiplex HIV-1 RNA assay Kit, Chiron, Emeryville, California, USA), according to the manufacturer's instructions. Briefly, viral particles were recovered from plasma sample specimens (preserved at -70°C) by centrifugation in a refrigerated microcentrifuge (70 000 g for 1 h), lysed and incubated with target probes complementary to the pol gene. After centrifugation at 70 000 g for 15 min, supernatants were added in duplicate to contiguous microwells coated with capture probes. The plate was left overnight at 53°C to allow hybridization of the complex viral-RNA probes to the capture probe. After washing, bDNA amplifier molecules were hybridized to the bound RNA and then hybridized to multiple alkaline phosphatase-labelled probes. Finally, a chemiluminescent substrate was added to the microwells and light emission was measured in a luminometer. The light emission was directly proportional to the amount of RNA in the sample. The light emission of sample specimens was compared with a four-point standard curve obtained by testing in parallel four different dilutions of a DNA plasmid ranging from 104 to 1.6 × 106 RNA equivalents per ml (number of copies/ml). The limit of sensitivity of this technique is 104 molecules/ml of plasma.
Virus isolation
Viruses were isolated, as previously described [12], from PBMC of HIV-1 seropositive index cases by cocultivation with healthy blood-donor PBMC stimulated with 2.5 μg/ml phytohaemagglutinin (PHA) for 48-72 h. The cultures were carried in RPMI containing 10% foetal calf serum (Gibco, Paisley, Scotland) and antibiotics. Twice weekly, half of the medium was removed and replaced with fresh medium and once a week fresh PHA-stimulated blood-donor PBMC were added to the cultures. For preparation of virus stocks, virus isolates were passaged once in PBMC; culture supernatants, harvested 7-10 days after the infection, were stored at -70°C, at a concentration of at least 2 ng/ml p24 antigen.
MT-2 cell assay
The syncytium-inducing capacity of viral isolates included in the study was evaluated in a cocultivation assay using infected PBMC and the MT-2 cell line. For this purpose, 5 × 106 healthy PHA-stimulated PBMC were infected with 1 ml viral stock, and 7 days later 1 × 106 PBMC were cocultivated with 3 × 106 MT-2 cells. The cultures were kept for 27 days, and twice-weekly supernatants were tested for the presence of p24 antigen and reverse transcriptase. Cultures were checked daily for the appearance of syncytia. A virus was considered MT-2 positive if cultures showed typical syncytia and supernatants were positive for p24 antigen and/or reverse transcriptase on at least three consecutive occasions.
Virus titration and neutralization assays
Fifty per cent infectious dose (ID50) titrations were performed in PBMC as previously described [13]. Briefly, virus aliquots were diluted in interleukin (IL)-2 medium to give six fivefold dilution steps starting with a 1:5 dilution; 75 μl of each dilution was added to five parallel wells of a round-bottom 96-well culture plate (Nunc, Roskilde, Denmark) and 1 × 105 PHA-stimulated blood donor PBMC in 150 μl IL-2 were added to each well. At days 1 and 3, the plates were centrifuged and the supernatants were replaced with fresh IL-2 medium. On day 7, a 100-μl sample from each well was analysed in an in-house HIV-1 p24 antigen enzyme-linked immunosorbent assay.
The ID50 was defined as the reciprocal of the virus dilution resulting in 50% positive wells (Reed-Muench calculation). ID50 ranged from 95 to 3200 for the different primary isolates. Neutralization assays were run simultaneously with each ID50 titration, using three virus dilutions (1: 5, 1: 25, 1: 125) for each serum sample. Sera, inactivated 30 min at 56°C, were diluted directly in the plate, in a total volume of 75 μl IL-2 medium. Five steps of fourfold serum dilutions, starting with 1:10, were used and virus was added in an equal volume. After 1-h incubation at 37°C, 1 × 105 PHA-stimulated PBMC in 75 μl IL-2 medium were added and the plates further incubated. The plates were washed on days 1 and 3, and supernatants were tested for the presence of HIV-1 antigen at day 7 [14]. All sera and virus isolates obtained from one patient were assayed simultaneously. The neutralizing titre of a serum was defined as the reciprocal of the highest serum dilution at which no virus was detected in the HIV-1 antigen assay.
Results
Clinical correlates of heterosexual transmission
The two groups of index cases were similar with regard to the clinical stage of infection and use of antiviral drugs (Table 1). The two groups differed, however, with regard to p24 antigen in serum and CD4+ T-lymphocyte counts, such that transmitting index cases had more frequently detectable p24 antigen in serum and lower CD4 lymphocyte counts (< 250 cells × 106/l) than non-transmitting index cases (Fisher exact test, P = 0.01 and 0.13, respectively).
HIV-1 RNA plasma load
Plasma HIV-1 RNA levels were determined in 12 of the patients enrolled (Table 1). All of the transmitting index cases tested showed high HIV-1 RNA copy numbers (between 46200 and 342 220 copies/ml of plasma). Among the non-transmitting index cases, three individuals were HIV-1 RNA negative in plasma, being below the sensitivity limit of the method, whereas in three additional cases HIV-1 plasma RNA could be detected but at a lower level (between 16 500 and 41 870 copies/ml of plasma) than in transmitting index cases, and this difference was statistically significant (Mann-Whitney U test: P = 0.0039).
Biological characterization of virus isolates from index cases
Characterization of HIV-1 isolates was performed by means of two different approaches: in PBMC by measuring ID50 titres; and in MT-2 cells by testing for replication and cytopathogenicity. Comparison of ID50 titres in PBMC showed differences between the two groups, in that viruses isolated from non-transmitting index cases had ID50 titres 6.5-fold lower than those from transmitting index cases (mean ID50 titres, 121.8 and 791.7, respectively). Differences were also found with regard to MT-2 tropism. In fact, four isolates from transmitting index cases, but only one isolate from non-transmitting index cases, were able to replicate and induce syncytia in MT-2 cells. Interestingly, MT-2-positive viruses replicated up to sixfold higher titres in PBMC than MT-2 negative viruses (mean ID50 titre, 824.1 and 134.9, respectively). None of these differences were statistically significant.
When ID50 titres in PBMC and MT-2 tropism were considered in relation to transmission, we noted similar ID50 titres with isolates from transmitters, regardless of the virus ability to grow in MT-2 cells (mean ID50 for MT-2-positive and MT-2-negative viruses, 728.6 and 884.4, respectively). MT-2-negative viruses from non-transmitters, however, differed from all the remaining isolates in that they had 10-fold lower ID50 (mean ID50, 81.6). In fact, three of these isolates did not yield enough infectious virus for evaluation in the neutralization test (ID50 < 5). Thus it appears that both qualitative and quantitative aspects of viral phenotype may influence the outcome of transmission.
Neutralizing activity in patient sera
Sera from 14 index cases were tested for neutralizing activity against seven heterologous primary isolates, i.e., against viruses derived from two non-transmitting and from five transmitting index cases (Table 2). Sera from 11 out of 14 (78%) subjects were able to neutralize at least one isolate. Interestingly, sera from transmitting and non-transmitting index cases could be distinguished by the neutralizing capacity. In fact, sera from non-transmitters showed neutralizing activity in 20 out of 47 reactions performed (42%), whereas sera from transmitters showed neutralizing activity in only 12 out of 44 reactions (27%). When the patients' sera were stratified into three groups according to the number of heterologous isolates neutralized, a broad-spectrum neutralizing activity was found in four out of seven non-transmitting index cases (Table 3). Moreover, all non-transmitting index cases showed a neutralizing antibody response to at least one heterologous isolate. In contrast, sera from three transmitting index cases could neutralize none of the seven heterologous isolates. The Jonckhecre-Terpstra exact test, a Kruskal-Wallis-like exact test, although not statistically significant (P = 0.0775), indicates that individuals with a broadly cross-reactive neutralizing antibody response are less likely to transmit the infection to their sexual partner.
Sera from seven transmitting and four non-transmitting index cases were tested for the presence of neutralizing antibodies against an autologous virus isolate obtained simultaneously with the serum sample (Table 2). Overall, only four out of 11 sera tested showed neutralizing activity to autologous virus, namely, three out of seven transmitting and one out of four non-transmitting index cases. Isolates derived from three non-transmitting index cases did not yield enough infectious virus (< 5 ID50) to be evaluated in the neutralization assay.
Discussion
In the present study we have shown that HIV-1-infected men with a low viral load, measured as plasma viral RNA or p24 antigen serum, are less likely to transmit HIV-1 infection to their female sexual partner. Recently, it has been suggested that 100 000 viral RNA molecules/ml plasma is the threshold above which the risk of HIV-1 transmission from mother-to-child is increased [15]. Remarkably, none of our non-transmitting index cases had a viral load higher than 100 000 viral RNA molecules/ml plasma, indicating that the threshold suggested for mother-to-child transmission may also apply for sexual transmission.
In addition to viral load, the index cases were studied for serum neutralizing activity to primary HIV-1 isolates and for the biological phenotype of their virus isolates. All index cases with a female sexual partner who remained seronegative over more than 1-year period had serum neutralizing activity. In the majority of cases, this neutralizing activity had broad specificity and the majority of sera were able to neutralize at least four heterologous primary isolates. In contrast, index cases whose sexual partners became infected frequently lacked neutralizing activity or had neutralizing activity with limited specificity. This result is reminiscent of mother-to-child transmission, in that sera from mothers with uninfected children neutralized a broader range of primary isolates [16].
Another difference between transmitting and non-transmitting index cases was observed in the replicative capacity of viral isolates. Extending our previous observations [10], the present data demonstrate that slow replication and low titres in PBMC characterize viruses obtained from index cases with HIV-1-seronegative sexual partners over a long time. Most of these viruses lack the capacity to replicate and induce syncytia in MT-2 cells. In contrast, viruses from index cases whose sexual partners seroconverted replicated efficiently in PBMC and, in the majority of cases, the viruses also replicated and induced syncytia in MT-2 cells. Again, the results are similar to those previously obtained in studies of mother-to-child transmission; mothers with an infected child more often had a rapid/high virus than mothers with an uninfected child [17]. The outcome of HIV-1 transmission, whether from mother-to-child or sexually from male to female, thus seems to correlate with both the replicative capacity of the virus and the presence of neutralizing antibodies. Conceivably, neutralizing antibodies as well as the replicative properties of the virus may exert their effect through the control of viral load. Our data also suggest that viruses with high replicative capacity would not only replicate more efficiently but that they also may be less susceptible to neutralization. The impact of neutralizing antibodies on viral load cannot be adequately evaluated in the present study, due to the limited number of patients. The finding that transmitting index cases carry a significantly higher viral load, however, suggests that the interplay between the host immune response and viral replication may influence the transmission of HIV-1.
It has been shown that antiretroviral treatment reduces the risk of heterosexual [9] and vertical transmission [18]. Our findings that subjects with high viral load more easily transmit the infection to the sexual partner support the concept that measures capable of reducing viral load, like antiretroviral drugs, may represent an important tool to reduce HIV-1 spread.
In addition to differences in viral replicative capacity, the viral phenotype may influence sensitivity to neutralization. Comparison of MT-2 tropism of the viral isolates and sensitivity to neutralization showed that MT-2-negative isolates were more readily neutralized than MT-2-positive isolates (Table 1). Neutralization was achieved in 54% of the reactions carried with MT-2-negative viruses, but in only 21% of those carried out with MT-2-positive viruses. This finding indicates that susceptibility to neutralization varies with the biological phenotype of the virus.
We also considered viral phenotype in relation to the host's ability to mount a serum neutralizing activity against autologous virus. One of five individuals who yielded MT-2-positive virus, and three out of six individuals with MT-2-negative virus were able to neutralize the autologous virus. This suggests that hosts with MT-2-positive viruses are less able to mount a neutralizing antibody response than those who harbour MT-2-negative viruses. This was to be expected, as the presence of MT-2 positive virus has been shown to correlate with an increased severity of HIV-1 infection and this, in turn, is associated with a decreased ability to mount a neutralizing antibody response [19].
In addition, the virological and immunological microenvironment of the male genital tract should be considered as important in transmission. Even though neutralizing antibodies and viruses with low replicative capacity can be demonstrated in the blood, other variables, like genital tract inflammations, irregular viral shedding and local antibody concentrations, may influence the outcome of HIV-1 transmission. Sexual transmission of HIV-1 is a complex phenomenon with several cofactors involved; to clarify exactly the role of each of these factors, larger patient groups will have to be tested.
Acknowledgements
We thank the patients in the study, A. La Gioia and A. Spinelli for invaluable technical assistance, and G. Scarlatti and L. Monno for critical reading of the manuscript and helpful discussions.
References
1. Quinn TC: Epidemiologic trends in HIV infection. Curr Opin Infect Dis 1991, 5:189-200.
2. Krieger JN, Coombs RW, Collier NC, et al.: Recovery of human immunodeficiency virus type 1 from semen: minimal impact of stage of infection and current antiviral therapy. J Infect Dis 1991, 163:386-388.
3. Hanin J, Mandelbrot L, Henrion R, Pradinaud R, Coulaud JP, Montagner L: Virus excretion in the cervicovaginal secretions of pregnant and non pregnant women. J Acquir Immune Defic Syndr 1993, 6:72-75.
4. Carael M, Van De Perre PH, Lepage PH, et al.: Human immunodeficiency virus transmission among heterosexual couples in Central Africa. AIDS 1988, 2:201-205.
5. European Study Group: Risk factors for male to female transmission of HIV. BMJ 1989, 298:411-415.
6. Padian NS, Shiboski SC, Jewell NP: The effect of number of exposures on the risk of heterosexual HIV transmission. J Infect Dis 1989, 161:883-887.
7. Plummer FA, Simonsen JN, Cameron DW, et al.: Cofactors in male-to-female transmission of human immunodeficiency virus type-1. J Infect Dis 1991, 163:233-239.
8. Basset MT, Latif AS, Katzenstein DA, Emmanuel JC: Sexual behaviour and risk factors for HIV infection in a group of male factory workers who donated blood in Harare, Zimbawe. J Acquir Immune Defic Syndr 1992, 5:556-559.
9. Musicco M, Lazzarin A, Nicolosi A, et al.: Antiretroviral treatment of men infected with human immunodeficiency virus type-1 reduces the incidence of heterosexual transmission. Arch Intern Med 1994, 154:1971-1976.
10. Fenyö EM, Fiore JR, Karlsson A, Albert J, Scarlatti G: Biological phenotypes of HIV-1 in pathogenesis and transmission. Antib Chemother 1994, 46:18-24.
11. Fiore JR, Jansson M, Scarlatti G, et al.: Correlation between seroreactivity to HIV-1 V3 loop peptides and male-to-female heterosexual transmission. AIDS 1993, 7:29-31.
12. Fiore JR, Calabro' ML, Angarano G, et al.: HIV-1 variability and progression to AIDS: a longitudinal study. J Med Virol 1990, 32:252-256.
13. Albert J, Abrahamsson B, Nagy K, et al.: Rapid development of isolate-specific neutralizing antibodies after primary HIV-1 infection and consequent emergence of virus variants which resist neutralization by autologous sera. AIDS 1990, 4:107-112.
14. Albert J, Björling E, von Gegerfelt A, Thorstensson R, Zhang J, Fenyö EM: Antigen detection is a reliable method for evaluating HIV/SIV neutralization assays. AIDS Res Hum Retrovirol 1993, 6:501-504.
15. Fang G, Burger H, Grimson R, et al.: Maternal plasma human immunodeficiency virus type-1 RNA level: a determinant and projected threshold for mother-to-child transmission. Proc Natl Acad Sci USA 1995, 92:12100-12104.
16. Scarlatti G, Albert J, Rossi P, et al.: Mother-to-child transmission of human immunodeficiency virus type-1: correlation with neutralizing antibodies against primary HIV-1 isolates. J Infect Dis 1993, 168:207-210.
17. Scarlatti G, Hodara V, Rossi P, et al.: Transmission of human immunodeficiency virus type-1 (HIV-1) from mother-to-child correlates with viral phenotype. Virology 1993, 197:624-629.
18. Connor EM, Sperling RS, Gelber R, et al.: Reduction of maternal-infant transmission of human immunodeficiency virus type-1 with zidovudine treatment. N Engl J Med 1994, 331:1173-1180.
19. Von Gegerfelt A, Albert J, Morfeldt-Manson L, Broliden K, Fenyö EM: Isolate-specific neutralizing antibodies in patients with progressive HIV-1 related disease. Virology 1991, 185:162-181.
© Lippincott-Raven Publishers.