Persistence of multidrug-resistant HIV-1 in primary infection leading to superinfection

Brenner, Bluma; Routy, Jean-Pierre; Quan, Yudong; Moisi, Daniela; Oliveira, Maureen; Turner, Dan; Wainberg, Mark A; co-investigators of the Quebec Primary Infection Study

Basic Science

Objective: The authors previous studies documented persistence of multidrug resistance (MDR) acquired in five primary HIV-1infection (PHI) cases for 1–2 years in the absence of antiretroviral treatment. This study characterizes the evolution of transmitted wild-type (WT) (n = 15), resistant (n = 10), and MDR (n = 6) infections. Long-term persistence of MDR infections (2–7 years), leading to one observed MDR superinfection is documented.

Methods: Genotypic changes in circulating viral quasi-species were evaluated over 1.5–7 years in patients (n = 31) enrolled in the PHI study. Sequencing of reverse transcriptase and protease regions identified nucleotide substitutions in the viral quasi-species and mutations at sites implicated in resistance to antiretroviral drugs. Phylogenetic and clonal analysis were performed to confirm one observed superinfection.

Results: Patients acquiring WT, drug-resistant and MDR infections showed little quasi-species evolution (> 99.6% homology) for more than 1.5 years, regardless of route of transmission. Transmitted resistance mutations (other than 184V) persisted for 2–7 years. MDR persistence in two PHI cases contrasted with the corresponding rapid reversion of MDR infections to WT in their partners following treatment interruption. One MDR transmission eliciting low-level viremia resulted in clearance of the original MDR infection followed by re-infection with a second heterologous MDR strain from a different partner. Phylogenetic and clonal analysis of source and index partner confirmed the superinfection. Both MDR species showed approximately 13-fold reductions in replication capacity relative to the homologous WT strain isolated from the source partner.

Conclusions: Genotypic analysis in PHI may identify superinfection and MDR infections that represent important determinants of virological and treatment outcome.

Author Information

From the McGill University AIDS Centre, Jewish General Hospital and McGill University Health Centre, Montreal, Quebec, Canada. *See Appendix.

Correspondence to Mark A. Wainberg, McGill AIDS Centre, Jewish General Hospital, 3755 Cote Ste Catherine Road, Montreal, Quebec, Canada H3T 1E2.

E-mail:; tel: +1 514 340 8260; fax: +1 514 340 7537.

Received: 2 March 2004; revised: 22 April 2003; accepted: 26 April 2004.

Article Outline
Back to Top | Article Outline


Highly active antiretroviral therapy (HAART), including drugs that inhibit the reverse transcriptase (RT) and protease (PR) enzymes of human immunodeficiency virus type 1 (HIV-1), has resulted in declining morbidity and mortality [1]. The failure to completely suppress viral replication allows for the development of genotypic changes in HIV-1 that confer resistance to each of the three major classes of antiretroviral drugs [2–4]. Cumulative data indicate that single drug-resistant variants can be transmitted to 10–20% of newly infected persons, with transmission of dual and triple-class multidrug resistance (MDR) observed in 3–5% of cases [5–8].

There is concern that the transmission of MDR viruses in primary HIV-1 infection (PHI) may limit future therapeutic options. Treatment failure has been observed in several individuals harboring MDR infections [8–10]. Our findings show an impaired fitness of transmitted MDR variants compared with wild-type (WT) infections acquired in PHI [11]. MDR infections transmitted in five PHI patients persisted in the absence of treatment [11]. This persistence differs from the rapid outgrowth of WT viruses in established infections upon treatment interruption, due to the selective growth advantage and fitness of WT variants [11–13]. Taken together, these findings suggest that archival WT viruses may not exist in MDR infections transmitted during PHI.

Several reports in the past year have documented four cases of inter-subtype superinfection (A/E and B) in recently infected intravenous drug users (IDU) [14,15]. Other studies have failed to confirm superinfection following IDU exposure, suggesting that superinfection is a relatively rare event [16,17]. Two subsequent reports have demonstrated superinfection in subtype B infections. In one case, a WT superinfection arose following a primary MDR infection [18,19].

It is, therefore, important to assess the virological consequences of transmission of drug-resistant variants in primary infection, as well as the time to disappearance in those patients not initially treated. The current study is a follow-up over more than 2 years of WT (n = 15) and resistant infections (n = 16) arising from IDU or male-to-male [men who have sex with men (MSM)] routes of transmission.

We describe several novel findings. Genotypic analysis indicates that a single dominant HIV-1 species persists for more than 2 years in circulating plasma and peripheral blood mononuclear cells (PBMC), regardless of route of transmission. Resistant and MDR infections can persist for 2–7 years following PHI. We also document the first described case of superinfection with a second MDR strain in a patient originally infected with a MDR strain from an identified source partner.

Back to Top | Article Outline

Patients and methods

Patients from the Quebec PHI cohort, acquiring WT (n = 15), drug-resistant (n = 10) or MDR (n = 6) infections were followed for two or more years following infection [5,11]. The time of original infection was estimated to be 2 weeks prior to symptom onset and/or presence of positive p24 antigenemia with negative enzyme-linked immunoabsorbent assay (EIA) [5]. Clinical reports identified route of transmission and treatment intervention(s). In two of the primary MDR infections, source partners of infection were identified and followed over time. Plasma samples from all patients were obtained with informed written consent.

Genotypic analysis of plasma viral RNA was performed as described [5,11]. In some samples, in which viral load decreased below levels of detection (< 50 copies/ml), plasma was concentrated by ultracentrifugation prior to viral RNA extraction. The TRUGENE HIV-1 Genotyping Assay was used in conjunction with the Open Gene automated DNA sequencing system (Bayer Diagnostics Inc., Toronto, Canada) to sequence the RT and PR regions of reverse-transcribed HIV-1 cDNA [5,11].

Nucleotide substitutions in the circulating plasma viral quasi-species were monitored over time. Substitutions were stratified according to changes at non-resistance sites or resistance sites, defined by the IAS–USA resistance panel [3,4]. Sequence evolution over time was calculated as total nucleotide changes relative to the original sequence observed at the first visit following PHI, expressed as a percentage of the total number of possible nucleotide transitions (300 PR nucleotides and 630 RT nucleotides between codons 40–250).

In the case suggestive of MDR superinfection, standard molecular cloning procedures were performed using cDNA amplified from PBMC isolated from the PHI patient at 2 and 16 months following the estimated date of infection. Briefly, the entire PR–RT region was amplified by nested PCR and inserted into the NcoI–SalI sites of a modified pTWIN2 vector, then grown, amplified and cloned in DH5α competent cells. Plasmid DNA amplified from five representative clones at both time points were sequenced. A phylogenetic tree was performed using BioEdit sequence alignment software (North Carolina State University). In parallel, genotypic analysis on plasma viral RNA and PBMC-derived cDNA was performed on the original source partner of infection at the time of initial infection and following treatment interruption (TI).

The replicative capacity of two MDR strains derived at 2 and 16 months following PHI and a homologous WT variant derived from the source partner were compared, based on RT enzymatic assays performed 7 days post-infection [11]. The relative fitness of both MDR variants and the WT species were also compared using dual-infection competition assays as previously described [11]. Both viral strains were normalized with respect to p24 antigen levels and titer, and mixed at the varying percentages, namely 0, 10, 25, 50, 75, 90 and 100% in regard to total inoculum. Enzymatic assays were performed weekly prior to new rounds of infection and the virus cultures were genotyped on the third week of infection.

Back to Top | Article Outline


Evolution of WT and drug-resistant quasi-species acquired in PHI

The evolution of plasma viral quasi-species was analyzed at two or more time points (18–24 months) following primary infection. The original WT quasi-species established in PHI was stable for 18 to 24 months post-infection regardless of MSM (n = 6) or IDU (n = 9) routes of transmission. There were fewer than four nucleotide changes at different time points in each patient indicating > 99.6% sequence homology (Table 1). In contrast, there were 25 to 36 signature nucleotide differences among different patients with 96–97% sequence homology at codon sites not implicated in drug resistance. PR polymorphisms, including 36I, 63P, 71V and 77I, also persisted over time.

Polymorphisms and mutations at resistance sites were also stable in the plasma species isolated from ten PHI cases (PHI-19 to PHI-28) harboring variants resistant to at least one drug (Table 1). There was no reversion of acquired mutations observed at the time of presentation. Two patients (PHI-19 and PHI-22) stably expressed T215 D/S mutations that probably arose from a single nucleotide reversion of T215F/Y codons (Table 1). None of the mutations in these ten patients were sufficient to confer phenotypic drug resistance per se.

Back to Top | Article Outline
Evolution of MDR variants acquired in PHI

As observed in WT infections, the dominant plasma quasi-species transmitted in MDR infections persisted over time (Table 1). Five PHI patients acquired MDR infections (PHI-2 to PHI-6) that conferred the expected phenotypic resistance to PR and RT inhibitors (unpublished results) (Table 1). Apart from the loss of M184V and the observed partial 215 transitions (T215C/S), MDR genotypes persisted for 15 to 81 months post-infection (Table 1).

PHI case 3 harbored a stable infection with K103N and PR mutations that persisted for over 3 years (39 months) (Table 1). His partner, however, lost the homologous MDR variant (99.2% sequence identity), due to WT viral outgrowth during a 6-month treatment interruption that started 37 months after initial presentation of PHI case 3 (unpublished results).

PHI case 5 showed a stable V77I/K103N/T215Y infection that persisted for over 6 years. During 5.6 years following his initial infection, this patient received zidovudine/lamivudine-containing regimens that included one of several protease inhibitors (saquinavir, indinavir or nelfinavir). His viremia remained below 1000 copies/ml throughout his treatment and no additional mutations other than M184V was acquired (see Methods for description of procedures used to perform genotyping in patients with viral load < 1000 copies/ml). The K103N mutation persisted in the absence of efavirenz and nevirapine usage throughout his treatment regimen. A 3-month treatment interruption at 67 months post-PHI presentation resulted in a T215Y ← T215C reversion that coincided with a 1.9 log increase in plasma viremia. Increased viremia due to enhanced viral fitness has been recently reported for the 215Y ← C transition [10].

PHI patient 6 harbored a M41L/T215Y/K101E mutational pattern conferring five- to six-fold phenotypic resistances to zidovudine, nevirapine, delaviridine, and efavirenz (data not shown). Introduction of a zidovudine/lamivudine/abacavir regimen led to a triple-class-resistant quasi-species mixture after 3 months of therapy, which was 20 months after presentation (Table 1). Of note, the MDR quasi-species included PR mutations although protease inhibitors (PI) were not included in his regimen. The transient appearance of a triple-class-resistant variant may have represented a minor and less fit species than was present in the original infection. Alternatively, the patient may have been superinfected with a second MDR strain originating from the same partner.

Back to Top | Article Outline
Superinfection in one PHI case with a second divergent MDR strain

Unlike PHI cases 2 to 6, the MDR infection in PHI case 1 showed considerable divergence after 10 months of infection, suggestive of re-infection by a divergent strain acquired from a second partner (Table 2). This patient acquired his first triple-class MDR infection from his partner who harbored the homologous MDR strain (Table 2). Although his partner had high viremia at the time of infection (102 600 copies/ml), the MDR infection in PHI case 1 was associated with low viremia (1305 copies/ml) and rapid CD4+ cell rebound (1200 × 106 cells/l) after PHI. Remarkably, viremia went to undetectable levels (< 50 copies/ml) at 4 and 8 months in the absence of treatment, contrasting with a mean of 79 432 copies/ml in our PHI cohort of WT infections. His low viremia was not attributable to host factors associated with favorable progression, namely the presence of polymorphisms in the HIV co-receptor, CCR5 Δ32 and relevant B27 or B57 MHC alleles (data not shown).

At 10 months following infection, the viremia of patient PHI-1 showed a 2.44 log increase to 13 888 copies/ml in association with a new MDR infection acquired 3 weeks following exposure by a different partner (Table 2). Unlike the original infection, this new strain had no nucleoside reverse transcriptase inhibitor or non-nucleoside reverse transcriptase inhibitor mutations other than M184V, and there were also differences in the profile of PI resistance mutations. Moreover, there were 41 nucleotide differences at non-resistance sites between the two viral variants (Table 2).

Molecular cloning techniques were performed on viruses isolated from PBMCs in PHI case 1 at the time of first presentation, and at 16 months following the estimated date of infection. All fourteen sequenced clones at visit 1 (P1A-T1) showed triple-class MDR variants, with silent mutations and polymorphisms that were identical to those in the original source partner. Seven clones isolated after 16 months (P1B-T6) represented the PI-resistant variant with homology to the second quasi-species. One of the seven clones harbored M184V in addition to the PI mutations.

A phylogenetic tree and detailed analysis of genotypic diversity at both resistance and non-resistance sites shows the extensive divergence of the two variants in PHI case 1, with viruses at 2, 4 and 8 months representative of one quasi-species (P1A- T1-T3) and viruses present 10 to 16 months post-infection (P1B- T4-T6) representative of a second quasi-species (Fig. 1a and b).

Moreover, virus isolated from the source partner at the time of presentation was clonal, with no MDR reversion in more than 2 years, other than loss of M184V in laboratory culture. In the absence of drug pressure, it might have been expected that rapid outgrowth of WT or the more fit PI-resistant variant could have occurred. Indeed, the source partner underwent a treatment interruption at 4 years after the presumed transmission of virus to PHI case 1 and this led to the in vivo outgrowth of a WT virus with homology to the original quasi-species observed in PHI case 1 (Table 2, Fig. 1b).

The replicative capacities of the initial MDR species transmitted in PHI and the superinfecting MDR variant were 7.5 and 7.8% of that observed for the WT homologous strain isolated from the source partner (maximal RT enzymatic activities of 19 981 ± 1495, 20 939 ± 8241 and 266 146 ± 13 551 cpm, respectively). In dual competitive fitness assays, the WT variant showed a selective advantage at ratios ranging between 10 and 90% of the inocula for both MDR species. Surprisingly, although both MDR species showed similar RT enzymatic activities, there appeared to be a replicative advantage of the initial MDR species relative to the superinfecting variant in two separate dual competition experiments.

Taken together, our findings indicate the likelihood of superinfection rather than co-infection with two variants in this case. The selective clearance of the first MDR species may have been related to its impaired fitness. The superinfection with another unfit resistant species may be related to host or viral factors, including inocula size and cytotoxic T lymphocyte response, rather than viral fitness per se [20].

Back to Top | Article Outline


Numerous PHI studies conducted between 1995 and 2003 have shown transmission of dual or triple class MDR in 4% of newly infected patients [5–8]. Further knowledge of the virological consequences of primary MDR infections is important. Our findings indicate that viruses transmitted in HIV infection are monophyletic, having few if any nucleotide changes in RT or PR over the course of more than 2 years. In viruses harboring resistance mutations, there was no major reversion of resistance codons other than M184V and partial T215Y/F transitions to T215C/D/N/S. This differs from the well-documented reversion to WT in established HIV infections consequent to treatment interruption [12,13].

We have documented the first reported case of a person harboring an MDR primary infection who was apparently infected with a second MDR variant. In spite of a rapid decline in plasma viremia suggestive of effective immune response, this patient was susceptible to a second infection. Five other subtype B superinfections have been described, as well as three intersubtype A/E and B superinfections [14,15,18–21]. Including our study, six of the seven superinfections described have occurred in the first year following initial infection.

Many have attributed superinfection to co-infection during primary infection. Two longitudinal studies involving IDU populations (n = 37 in both studies) indicated that superinfection is a rare phenomenon that was not observed during 1–12 years of follow-up spanning 215 and 1072 total years of exposure [18,19]. However, it is not known whether any patients were recruited within the first year of HIV-1 exposure in these studies. In contrast, we have been able to identify the source partner of infection and can argue against co-infection.

Findings of HIV-1 superinfection are a matter of concern insofar as such results challenge the assumption that immune responses can protect against re-infection. In our study, the impaired viral fitness of the initial MDR infection may be a factor in permitting superinfection. The initial MDR strain showed a 13-fold impaired replicative capacity from a WT variant strain from the isolated source partner following a treatment interruption. Fitness considerations may also have been important in a WT superinfection of an initial MDR infection and cases of subtype B superinfection following A/E infections that elicited low-level viremia [14,15].

In newly infected individuals, multi-mutated viruses conferring MDR may represent a new determinant of virological outcome. Persistence of MDR in the absence of treatment raises serious issues regarding HIV-1 management. For recently infected MDR patients, drug resistance analysis and viral fitness may provide useful information in regard to ultimate therapeutic strategies.

Back to Top | Article Outline


We thank Maureen Oliveira and Mervi Detorio for their technical support.

Sponsorship:This work was supported by the Canadian Institutes for Health Research and the Réseau SIDA of the Fonds de la Recherche en Santé du Québec.

Back to Top | Article Outline


1. Palella FJ, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med 1998; 338:853–860.
2. Wainberg MA, Friedland G. Public health implications of antiretroviral therapy and HIV drug resistance. JAMA 1999; 279:1977–1983.
3. Hirsch MS, Brun-Vézinet F, Clotet B, Conway B, Kuritzkes DR, D'Aquila RT, et al. Antiretroviral drug resistance testing in adults with human immunodeficiency virus type 1: 2003 recommendations of an International AIDS Society-USA panel. Clin Infect Dis 2003; 37:113–128.
4. D'Aquila RT, Schapiro JM, Brun-Vezinet F, Brun-Vezinet F, Clotet B, Conway B, et al. Drug resistance mutations in HIV-1. Top HIV Med 2003; 11:92–96.
5. Salomon H, Wainberg MA, Brenner BG, Quan Y, Rouleau D, Cote P, et al. Prevalence of HIV-1 viruses resistant to antiretroviral drugs in 81 individuals newly infected by sexual contact or intravenous drug use. AIDS 2000; 14:F17–23.
6. Yerly S, Kaiser L, Race E, Bru JP, Clavel F, Perrin L, et al. Transmission of antiretroviral-drug-resistant HIV-1 variants. Lancet 1999; 354:729–733.
7. Boden D, Hurley A, Zhang L, Cao Y, Guo Y, Jones E, et al. HIV-1 drug resistance in newly infected individuals. JAMA 1999; 282:1135–1141.
8. Little SJ, Holte S, Routy JP, Daar ES, Markowitz M, Collier AC, et al. Antiretroviral-drug resistance among patients recently infected with HIV. N Engl J Med 2002; 347:385–394.
9. Hecht GM, Grant RM, Petropoulos CJ, Dillon B, Chesney MA, Tian H, et al. Sexual transmission of an HIV-1 variant resistant to multiple reverse-transcriptase and protease inhibitors. N Engl J Med 1998; 339:307–311.
10. Gandhi RT, Wurcel A, Rosenberg ES, Johnston MN, Hellmann N, Bates M, et al. Progressive reversion of hunman immunodeficiency virus type 1 resistance mutations in vovo after transmission of a multiply drug-resistant virus. Clin Infect Dis 2003; 37:1693–1698.
11. Brenner BG, Routy JP, Petrella M, Moisi D, Oliveira M, Detorio M, et al. Persistance and fitness of multidrug-resistant of human immunodeficiency virus type 1 acquired in primary HIV infection. J Virol 2002; 76:1753–1761.
12. Verhofstede C, Wanzeele FV, Van der Gucht B, De Cabooter N, Plum J. Intrerruption of reverse transcriptase inhibitors or a switch from reverse transcriptase transcriptase to protease inhibitors resulted in a fast reappearance of virus strains with a reverse transcriptase inhibitor-sensitive genotype. AIDS 1999; 13:2541–2546.
13. Devereux HL, Youle M, Johnson MA, Loveday C. Rapid decline in detectability of HIV-1 drug resistance mutations after stopping therapy. AIDS 1999; 13:F123–F127.
14. Jost S, Bernard MC, Kaiser L, Yerly S, Hirschel B, Samri A, et al. A patient with HIV-1 superinfection. N Engl J Med 2002; 347:731–736.
15. Ramos A, Hu DJ, Nguyen L, Phan KO, Vanichseni S, Promadej N, et al. Intersubtype human immunodeficiency virus type 1 superinfection following seroconversion to primary infection in two injecting intravenous drug users. J Virol 2002; 76: 7444–7452.
16. Gonzales MJ, Delwart E, Rhee SY Tsui R, Zolopa AR, Taylor J, et al. Lack of detectable human immunodeficiency virus type 1 superinfection during 1072 person-years of observation. J Infect Dis 2003; 188:397–405.
17. Tsui R, Herring BL, Barbour JD, Grant RM, Bacchetti P, Kral A, et al. Human immunodeficiency virus type 1 superinfection was not detected following 215 years of injection drug user exposure. J Virol 2004; 78:94–103.
18. 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 infection. Nature 2002; 420:434–439.
19. Koelsch KK, Smith DM, Little SJ, Ignacio CC, Macaranas TR, Brown AJ, et al. Clade B HIV-1 superinfection with wild-type virus after primary infection with drug-resistant clade B virus. AIDS 2003; 17:F11–16.
20. Allen T, Altfeld M. HIV-1 superinfection. J Allergy Clin Immunol 2003, 112: 829–835.
21. Smith D, Wong J, Hightower, Kolesch K, Ignacio C, Daar E, et al. Incidence of HIV superinfection following primary infection. XI Conference on Retroviruses and Opportunistic Infections, February 2004. San Francisco, Abstract 21.
Back to Top | Article Outline

Co-investigators of the Quebec Primary Infection Study include J-G Baril, M. Bélanger, P Côté, S. Dufresne, F. Leplante, J. Lebel, Clinique du Quartier Latin; M. Boissonnault, H. Lavoie, B. Lessard, C. Olivier, R., B. Trottier, S. Vézina, Clinique l'Actuel; Clinique Goldberg LeBlanc and Rosengren; N. Gilmore, M. Klein, R. Lalonde, J. MacLeod, G. Smith, McGill University Health Centre; P. Cholette, Clinique 1851; N. Laponte, J. Samson, Hôpital Sainte Justine; C. Frenette, C. Valois, Hôpital Charles- Lemoyne; M. Bélanger, CLSC des Faubourgs, Montréal. Cited Here...


HIV transmission; viral load; drug resistance; M184v mutation

© 2004 Lippincott Williams & Wilkins, Inc.