In the latter part of 2002 and early 2003, there were four published reports of superinfection of HIV-1–infected individuals with a second strain of HIV-1 [1–4]. This information has been received with both surprise and dismay. While the surprise expressed by some individuals is, in itself, surprising, the dismay appears to be justified because of its negative implications relative to development of an HIV-1 vaccine. If the superinfections occurred in infected individuals with high titers of antiviral antibodies, CD4 and CD8 T-cell responses, and little or no measurable dysfunction of their immune systems, it would imply that the prospects for an efficacious vaccine that could prevent HIV infection or limit dissemination and disease progression were not good. While the former goal is the ideal, albeit unrealistic, one, the latter might be possible.
In the four recent reports, a total of five individuals infected initially with one HIV-1 strain and subsequently with a second strain were identified. Two of the superinfections occurred in injecting drug users and the other three occurred during sex between men, indicating that the apparent failure of ongoing immune responses to prevent the second infection was not restricted to parenteral or mucosal transmission. Furthermore, in two of the five individuals, superinfection occurred with strains from the same subtype (B) whereas the other three involved subtype B and A/E recombinant strains. While a priori one might predict that superinfection was more likely when it involved strains of different subtypes, the apparent occurrence with strains from the same subtype, which are more closely related genetically, is unsettling. If ongoing immune responses generated with antigens from one HIV-1 subtype cannot prevent infection with a virus from the same subtype, then the possibility for vaccine-mediated protection against infection by other subtypes, which can differ by as much as 30%, is virtually non-existent, especially if univalent vaccines are used.
Since the early 1990s, many cases of HIV-1 infections with two or more strains have been reported [5–16]. However, unless there was definitive evidence that infection with one strain occurred several months after the first, it was generally assumed that exposure to and infection by the second strain occurred simultaneously or during a ‘window period’ before the HIV-1–specific immune response had matured enough to provide protection against infection. Moreover, polymerase chain reaction (PCR) amplification and sequence analysis generally revealed variants with genetic diversity indicative only of intrastrain evolution, even in multiply exposed individuals, and this finding was used to argue against superinfection. The fallacy in this reasoning was that in multiple infections, one strain might exist as a provirus in substantially more cells than a second strain and, to detect a minor population, strain-specific, not universal, PCR primers must be used and often multiple independent PCR analyses must be made [17,18]. Since it is rare that the sequences of strains to which an individual is exposed are known, the use of strain-specific primers is not always possible. Therefore, it is probable that the number of co- and superinfections has been underestimated.
Statistically, it is highly unlikely that all or most of these dual infections occurred during the same transmission event or during the few weeks before an immune response was established. Therefore, one must conclude that an ongoing immune response, in the majority of cases, was unable to prevent infection with a second strain of HIV-1, especially with a virus belonging to a different subtype. At the recent Tenth Conference on Retroviruses and Opportunistic Infections, Perrin et al. , in a study of 156 Swiss intravenous drug users infected with HIV-1, identified five (5/156, 3.2%) individuals who were coinfected with a subtype B strain and a circulating recombinant form, CRF-11. Of these five coinfected persons, two (2/156, 1.3%) patients infected with subtype B strains for 3 and 5 years experienced a sudden increase in HIV-1 plasma RNA levels and HIV-1 CRF-11 was identified. Although both strains persisted more than 2 years, the second virus strain, CRF-11, ultimately supplanted the subtype B viruses. This study indicates that dual infection is not uncommon and that, among dually infected individuals, superinfection is not a rare event.
The strongest evidence for relatively frequent occurrences of HIV-1 co- and superinfection is the large number of recombinant viruses identified in many countries, especially in those where several HIV-1 subtypes are cocirculating [20–25]. For recombination to occur, two distinct strains must infect and replicate in the same cell at the same time. This criterion is satisfied by HIV-1 since most HIV-1–infected cells appear to harbor two or more proviruses [26,27]. In recent studies characterizing HIV-1 subtypes in various cohorts, from 21 to 37% of all HIV-1 isolates evaluated were recombinants between strains representing different clades [22–25]. These percentages reflect unique recombination events and do not include the CRF strains that are prevalent and being transmitted in some populations and regions. Among a cohort of 57 injecting drug users in China, Yang et al.  identified 8.8% ‘second-generation’ recombinants, formed between two CRF HIV-1 strains (both distinct intersubtype B/C recombinants). Even recombinant viruses from HIV-1 groups M and O have been reported, despite overall homology of only 65% [29,30]. One of these individuals was infected with two group O viruses and a M/O recombinant .
This extraordinary amount of recombination can cause stepwise increases in the rate of genetic diversity and, in all likelihood, changes in biological properties, such as acquisition of drug resistance or increased pathogenicity. Moutouh et al.  described the emergence in vitro of HIV-1 recombinants that were resistant to two drugs during growth of single drug-resistant strains in the presence of both drugs. In addition, Wooley et al.  demonstrated in vivo recombination between two mutant SIVmac239 clones to generate a wild-type virus. Generation of recombinant HIV-1 strains with enhanced biological properties as well as rapid progression to AIDS in patients coinfected with two genetically divergent subtype B strains have been described [15,16]. In one report, multiple recombinants were detected and one ultimately predominated . Because of the greater diversity between strains in the M, O, and N groups, particular emphasis should be placed on monitoring those persons infected with intergroup recombinants and evaluating not only the virus–host interactions but also the recombinants themselves for novel properties, such as changes in receptor usage and cell tropism.
That an HIV-infected person can become infected with a second, unrelated strain, even when an apparently strong anti-HIV-1 response is present, is not surprising to those who are familiar with studies of lentivirus animal models. Superinfection has been reported for feline immunodeficiency virus infection of cats, SIV infection of macaques, and HIV-1 infection of chimpanzees [33–38]. Innumerable preclinical vaccine trials in macaques have shown that immunization with various attenuated SIV strains, in general, does not prevent infection, but it does lead to more rapid clearance of virus and delayed progression of disease . However, recombination between an attenuated vaccine strain and the challenge strain, with generation of a more virulent virus, has been reported . Although clinicians and other researchers often ignore the results of animal studies, much can be learned from relevant models. In fact, some studies of HIV-1–infected chimpanzees have been predictive of later studies in HIV-1–infected patients. Superinfection of a chimpanzee with a second subtype B strain was reported as early as 1987  and with a second strain of a different subtype, with subsequent recombination, in 1997 . Of note, the one confirmed case of terminal AIDS in a chimpanzee occurred in an animal that had been superinfected with a second HIV-1 subtype B strain . During the course of infection, recombinant viruses that were more pathogenic in this species were generated and predominated at the time of death [41,42]. In our published and unpublished studies, we attempted to superinfect chimpanzees previously infected with various HIV-1 strains (for 8 to 64 months) with a second HIV-1 from the same or a different subtype on 22 different occasions; we documented infection with the second HIV-1 strain 18 times, for a success rate of more than 80%. Of these, nine intraclade attempts were made and all nine (100%) were successful. Superinfections were established by both intravenous and mucosal routes and, regardless of route, time of the second exposure, or subtype, HIV-1 strains that replicated more efficiently and to higher titers always established a secondary infection. One chimpanzee initially infected with HIV-1LAV−1b was superinfected twice: first rectally with 90CR402_A/E and later intravenously with another subtype B strain, JC499 (unpublished data). The majority of HIV-1–infected chimpanzees maintains a fully functional immune system, in contrast to some HIV-1–infected humans, who are immunosuppressed. Therefore, being immunocompetent at the time of exposure to and infection with a second strain of HIV-1 supports the contention that superinfection occurs readily despite HIV-1–specific responses.
What are the implications of HIV-1 superinfection? The most obvious answer relates to its predictive value for potential vaccine efficacy, which a priori would appear to be low, especially if responses elicited by a vaccine are similar to those induced by active infection. Although vaccine-induced immunity may differ from natural infection, it is generally believed that attenuated, live replicating vaccines are the most efficacious for viruses because they potentially can elicit both humoral and cellular immune responses to multiple epitopes. If prior infection with HIV-1 elicits a seemingly broad immune response, as in the patient described by Altfeld et al. , yet superinfection occurs, then the possibility of an efficacious HIV-1 vaccine seems remote. However, whether elicited by infection or by a vaccine, the ability of any ongoing immune response to prevent infection by another HIV-1 strain is likely to depend on multiple parameters that will differ for each person . These parameters are related not only to the virus and virus-specific immune responses but also to the general immunological status of the exposed individual and the physical milieu into which the virus or virus-infected cells are deposited (if exposure is to a mucosal surface). Whether HIV-1 successfully establishes infection will depend on the route of exposure, the infectious dose of the inoculum, and the inherent replicative properties of the virus. From studies of HIV-1 transmission to women, we know that ease of infection via the vaginal mucosa is dependent on the concentration of the inoculum and on whether other infectious agents or vaginal lesions are present [44,45]. In essence, every transmission event is unique relative to the complexity and biological properties of the HIV-1 inoculum and the immune and physical status of the recipient. Since efficiency of transmission and progression to disease are related directly to the amount of virus in the inoculum and viral burden, respectively, the daunting task of an HIV-1 vaccine is to lower the infectious dose and/or to maintain low to undetectable viral loads so that transmission will be interrupted and time to disease will be lengthened. Based on the available information regarding dual infections, superinfections, and recombinant HIV-1 strains in humans, and on the results obtained in animal models, it is likely that some measure of vaccine-mediated protection will occur in a minority of individuals. Furthermore, it is likely there is no universal correlate(s) of protection because of the large number of variables influencing each transmission event.
The degree of efficacy of a vaccine against HIV-1 is likely to vary in different populations and cannot be predicted at this time, other than that it probably will be low. However, if more innovative vaccines and formulations can be developed, then this conclusion might be disproved. For example, the use of antigens or epitopes from several subtypes might broaden the immune response such that the induced antibodies could recognize or bind a larger percentage of HIV-1 strains, which potentially could be eliminated by neutralization or antibody-dependent cellular cytotoxicity. More recently, the use of engineered vaccine antigens based on consensus or ancestral sequences has been proposed . It is believed that this approach will reduce the overall genetic diversity between contemporary strains and the vaccine. Even the best vaccination strategy is likely to provide more protection against disease progression than against infection itself. This occurrence, in itself, would be beneficial since any decrease in disease progression should be accompanied by lower viral burdens and, as a secondary effect, rates of transmission would decrease.
From the above discussion, it is obvious that substantial numbers of superinfections will continue to occur, that recombinant viruses will be generated, and that some of these recombination events will result in more-fit viruses. Some of these viruses will acquire new properties that enable them to be transmitted more easily and to be more pathogenic, with increased rates of disease progression. Although superinfection might result in increased antiviral immunity that recognizes a broader array of viral epitopes, it also might induce a decline in specific CD8 T-cell responses, as demonstrated by Altfeld et al. . Therefore, a better understanding of HIV–HIV and HIV–host interactions as well as prospective studies to obtain more accurate estimates of the frequency of superinfection are needed. We are left to conclude that all HIV-infected persons should adhere to safe-sex practices, not only to lessen the risk to themselves but also to their partners and infants. Regardless of HIV status, behavioral modification to institute these practices should be encouraged and all high-risk individuals should be counseled.
The author thanks Dr Grace Aldrovandi for helpful comments on the manuscript.
1. Ramos A, Hu DJ, Nguyen L, Phan K-O, Vanichseni S, Promadej N, et al
. Intersubtype human immunodeficiency virus type 1 superinfection following seroconversion to primary infection in two injection drug users. J Virol
2. Jost S, Bernard M-C, Kaiser L, Yerly S, Hirschel B, Samri A, et al
. A patient with HIV-1 superinfection. N Engl J Med
3. Altfeld M, Allen TM, Yu XG, Johnston MN, Agrawal D, Korber BT, et al
. HIV-1 superinfection despite broad CD8+ T-cell responses containing replication of the primary virus. Nature
4. Koelsch KK, Smith DM, Little SJ, Ignacio CC, Macaranas TR, Brown AJL, et al
. Clade B HIV-1 superinfection with wild-type virus after primary infection with drug-resistant clade B virus. AIDS
5. Sala M, Zambruno G, Vartanian J-P, Marconi A, Bertazzoni U, Wain-Hobson S. Spatial discontinuities in human immunodeficiency virus type 1 quasispecies derived from epidermal Langerhans cells of a patient with AIDS and evidence for double infection. J Virol
6. Artenstein AW, VanCott TC, Mascola JR, Carr JK, Hegerich PA, Gaywee J, et al
. Dual infection with human immunodeficiency virus type 1 of distinct envelope subtypes in humans. J Infect Dis
7. Zhu T, Wang N, Carr A, Wolinsky S, Ho DD. Evidence for coinfection by multiple strains of human immunodeficiency virus type 1 subtype B in an acute seroconvertor. J Virol
8. Diaz RS, Sabino EC, Mayer A, Mosley JW, Busch MP, and the Transfusion Safety Study Group. Dual human immunodeficiency virus type 1 infection and recombination in a dually exposed transfusion recipient. J Virol
9. Butto S, Argentini C, Mazzella AM, Iannotti MP, Leone P, Leone P, et al
. Dual infection with different strains of the same HIV-1 subtype. AIDS
10. Takehisa J, Zekeng L, Miura T, Ido E, Yamashita M, Mboudjeka I, et al
. Triple HIV-1 infection with group O and group M of different clades in a single Cameroonian AIDS patient. J Acquir Immune Defic Syndr Hum Retrovirol
11. Janini LM, Tanuri A, Schechter M, Peralta JM, Vicente ACP, dela Torre N, et al
. Horizontal and vertical transmission of human immunodeficiency virus type 1 dual infections caused by viruses of subtypes B and C. J Infect Dis
12. Belda FJ, Barlow KL, Murphy G, Parry JV, Clewley JP. A dual subtype B/E HIV type 1 infection with a novel V3 loop crown motif among infections acquired in Thailand and imported into England. AIDS Res Hum Retroviruses
13. Ramos A, Tanuri A, Schechter M, Rayfield MA, Hu DJ, Cabral MC, et al
. Dual and recombinant infections: an integral part of the HIV-1 epidemic in Brazil. Emerg Infect Dis
14. Mellquist JL, Becker-Pergola G, Gu J, Guay L, Himes L, Kataaha P, et al
. Dual transmission of subtype A and D HIV type 1 viruses from a Ugandan woman to her infant. AIDS Res Hum Retroviruses
15. Wang B, Lal RB, Dwyer DE, Miranda-Saksena M, Boadle R, Cunningham A, et al
. Molecular and biological interactions between two HIV-1 strains from a coinfected patient reveal the first evidence in favor of viral synergism. Virology
16. Liu S-L, Mittler JE, Nickle DC, Mulvania TM, Shriner D, Rodrigo AG, et al
. Selection for human immunodeficiency virus type 1 recombinants in a patient with rapid progression to AIDS. J Virol
17. Rodrigo AG, Goracke PC, Rowhanian K, Mullins JI. Quantitation of target molecules from polymerase chain reaction-based limiting dilution assays. AIDS Res Hum Retroviruses
18. Wei Q, Javadian A, Lausen N, Fultz PN. Distribution and quantification of human immunodeficiency virus type 1, strain JC499, proviral DNA in tissues from an infected chimpanzee. Virology
19. Perrin L, Yerly S, Jost S, Monnat M, Telenti A, Chave JP, et al
. Co/super-infection in recently HIV-1 infected IVDUs.
In Tenth Conference on Retroviruses and Opportunistic Infections
. Boston, February 2003 [abstract 153].
20. Robertson DL, Sharp PM, McCutchan FE, Hahn BH. Recombination in HIV-1. Nature
21. Peeters M. Recombinant HIV sequences: their role in the global epidemic.
In: HIV Sequence Compendium 2000
. Edited by Kuiken C, Foley B, Hahn B, Mellors JW, Hahn B, Wolinsky S, et al
. Los Alamos, NM: Los Alamos National Laboratory; 2000: I39–I54.
22. Guimaraes ML, Dos Santos Moreira A, Loureiro R, Galvao-Castro B, Morgado MG, and the Brazilian Network for HIV Isolation and Characterization. High frequency of recombinant genomes in HIV type 1 samples from Brazilian southeastern and southern regions. AIDS Res Hum Retroviruses
23. Harris ME, Serwadda D, Sewankambo N, Kim B, Kigozi G, Kiwanuka N, et al
. Among 46 near full length HIV type 1 genome sequences from Rakai District, Uganda, subtype D and AD recombinants predominate. AIDS Res Hum Retroviruses
24. Renjifo B, Gilbert P, Chaplin B, Vannberg F, Mwakagile D, Msamanga G, et al
. Emerging recombinant human immunodeficiency viruses: uneven representation of the envelope V3 region. AIDS
25. Cornelissen M, Kampinga G, Zorgdrager F, Goudsmit J and the UNAIDS Network for HIV Isolation and Characterization. Human immunodeficiency virus type 1 subtypes defined by env show high frequency of recombinant gag genes. J Virol
26. Gratton S, Cheynier R, Dumaurier M-J, Oksenhendler E, Wain-Hobson S. Highly restricted spread of HIV-1 and multiply infected cells within splenic germinal centers. Proc Natl Acad Sci USA
27. Jung A, Maier R, Vartanian J-P, Goudsmit J and the UNAIDS Network for HIV Isolation and Characterization. Multiply infected spleen cells in HIV patients. Nature
28. Yang R, Kusagawa S, Zhang C, Xia X, Ben K, Takebe Y. Identification and characterization of a new class of human immunodeficiency virus type 1 recombinants comprised of two circulating recombinant forms, CRF07_BC and CRF08_BC, in China. J Virol
29. Takehisa J, Zekeng L, Ido E, Yamaguchi-Kabata Y, Mboudjeka I, Harada Y, et al
. Human immunodeficiency virus type 1 intergroup (M/O) recombination in Cameroon. J Virol
30. Peeters M, Liegeois F, Torimiro N, Bourgeois A, Mpoudi E, Vergne L, et al
. Characterization of a highly replicative intergroup M/O human immunodeficiency virus type 1 recombinant isolated from a Cameroonian patient. J Virol
31. Moutouh L, Corbeil J, Richman DD. Recombination leads to the rapid emergence of HIV-1 dually resistant mutants under selective drug pressure. Proc Natl Acad Sci USA
32. Wooley DP, Smith RA, Czajak S, Desrosiers RC. Direct demonstration of retroviral recombination in a rhesus monkey. J Virol
33. Okada S, Pu R, Young E, Stoffs WV, Yamamoto JK. Superinfection of cats with feline immunodeficiency virus subtypes A and B. AIDS Res Hum Retroviruses
34. Petry H, Dittmer U, Stahl-Hennig C, Coulibaly C, Makoschey B, Fuchs D, et al
. Reactivation of human immunodefiency virus type 2 in macaques after simian immunodeficiency virus SIVmac superinfection. J Virol
35. Fultz PN, Srinivasan A, Greene CR, Butler D, Swenson RB, McClure HM. Superinfection of a chimpanzee with a second strain of human immunodeficiency virus. J Virol
36. Fultz PN, Wei Q, Stallworth J, Barre-Sinoussi F, Deslandres A, Girard M. Superinfection of chimpanzees with HIV-1 strains representing the same or different subtypes.
In: Retroviruses of Human AIDS and Related Animal Diseases
[Dixieme Colloque des Cent Gardes
]. Edited by Girard M, Dodet B. Paris: Elsevier; 1995:173–177.
37. Shibata R, Siemon C, Cho MW, Arthur LO, Nigida SM, Matthews T, et al
. Resistance of previously infected chimpanzees to successive challenges with a heterologous intraclade B strain of human immunodeficiency virus type 1. J Virol
38. Fultz PN, Yue L, Wei Q, Girard M. Human immunodeficiency virus type 1 intersubtype (B/E) recombination in a superinfected chimpanzee. J Virol
39. Letvin NL. Strategies for an HIV vaccine. J Clin Invest
40. Gundlach BR, Lewis MG, Sopper S, Schnell T, Sodroski J, Stahl-Henning C, et al
. Evidence for recombination of live, attenuated immunodeficiency virus vaccine with challenge virus to a more virulent strain. J Virol
41. Novembre FJ, Saucier M, Anderson DC, Klumpp SA, O'Neil SP, Brown CR, et al
. Development of AIDS in a chimpanzee infected with human immunodeficiency virus type 1. J Virol
42. Wei Q, Fultz PN. Extensive diversification of human immunodeficiency virus type 1 subtype B strains during dual infection of a chimpanzee that progressed to AIDS. J Virol
43. Nathanson N, Mathieson BJ. Biological considerations in the development of a human immunodeficiency virus vaccine. J Infect Dis
44. Quinn TC, Wawer MJ, Sewankambo N, Serwada D, Li C, Wabwire-Mangen F, et al
. Viral load and heterosexual transmission of human immunodeficiency virus type 1. N Engl J Med
45. Fiore JR, Zhang Y-J, Bjorndal A, Angarano G, Pastore G, Fenyo EM. Biological correlates of HIV-1 heterosexual transmission. AIDS
46. Gashen B, Taylor J, Yusim K, Foley B, Gao F, Lang D, et al
. Diversity considerations in HIV-1 vaccine selection. Science