Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Frequency of occurrence of HIV-1 dual infection in a Belgian MSM population

Frequency of occurrence of HIV-1 dual infection in a Belgian MSM population a1111111111 a1111111111 Introduction HIV-1 dual infection is a condition that results from infection with at least two HIV-1 variants from different sources. The scarceness of information on this condition is partly due to the OPENACCESS fact that its detection is technically challenging. Using next-generation sequencing we Citation: Hebberecht L, Vancoillie L, Schauvliege defined the extent of HIV-1 dual infection in a cohort of men who have sex with men (MSM). M, Staelens D, Dauwe K, Mortier V, et al. (2018) Frequency of occurrence of HIV-1 dual infection in a Belgian MSM population. PLoS ONE 13(4): Material & methods e0195679. https://doi.org/10.1371/journal. Eighty-six MSM, diagnosed with HIV-1 subtype B infection between 2008 and 2013 were pone.0195679 selected for next-generation sequencing of the HIV-1 envelope V3. Sequencing was per- Editor: Luis Menendez-Arias, Consejo Superior de formed on 2 plasma samples collected with an interval of> 6 months before the initiation of Investigaciones Cientificas, SPAIN antiretroviral therapy. Maximum likelihood phylogenetic trees were inspected for dual infec- Received: October 20, 2017 tion, defined as the presence of two or more monophyletic clusters with 90% bootstrap Accepted: March 27, 2018 support and a mean between-cluster genetic distance of 10%. To confirm dual infection, Published: April 6, 2018 deep V3 sequencing of intermediate samples was performed as well as clonal sequencing Copyright:© 2018 Hebberecht et al. This is an of the HIV-1 protease-reverse transcriptase gene. open access article distributed under the terms of the Creative Commons Attribution License, which Results permits unrestricted use, distribution, and reproduction in any medium, provided the original Five of the 74 patients (6.8%) for whom deep sequencing was successful, showed clear evi- author and source are credited. dence of dual infection. In 4 of them, the second strain was absent in the first sample but Data Availability Statement: All relevant data are occurred in subsequent samples. This was highly suggestive for superinfection. In 3 patients within the paper and its Supporting Information both virus variants were of subtype B, in 2 patients at least one of the variants was a subtype files. B/non-B recombinant virus. Funding: The AIDS Reference Laboratory of Ghent is supported by the Belgian Ministery of Social Affairs through a fund within the Health Insurance Conclusions System. The funder had no role in study design, Dual infection was confirmed in 6.8% of MSM diagnosed with HIV-1 in Belgium. This preva- data collection and analysis, decision to publish, or preparation of the manuscript. lence is probably an underestimation, because stringent criteria were used to classify viral variants as originating from a new infection event. Competing interests: The authors have declared that no competing interests exist. PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 1 / 15 HIV-1 superinfection in MSM Introduction Human immunodeficiency virus type 1 (HIV-1) dual infection occurs when an HIV-1 infected patient is infected by a new viral variant from another host. This results in the presence of two variants, each originating from an independent infection event. When the second infection occurs nearly simultaneously with the first, or at least before seroconversion, the condition is generally called co-infection. When the second infection occurs after seroconversion it is called superinfection [1]. One of the firmest indications that co- or superinfection is possible is the existence of recombinant viruses. Recombination between genetically distinct viruses can only occur after infection of a single cell with both variants. Recombinant viruses can remain restricted to a single patient (unique recombinant forms; URF) or be further transmitted. Recombinant forms detected in at least three unrelated individuals are called circulating recombinant forms (CRF). CRF contribute extensively to the genetic variability of HIV-1 [2, 3] and the constant increase in number of reported CRF (https://www.hiv.lanl.gov/content/ sequence/HIV/CRFs/CRFs.html) suggests high frequencies of dual infection. The first patient with HIV-1 dual infection was documented in 2002 [4]. Since then, several other, mostly isolated cases of dual infection, have been reported [1]. Estimates of the overall frequency of this condition vary between 0 and 20% [5±8]. The number of well documented superinfections, however, remains low. Difficulties to demonstrate this condition may be the most important reason for that. Heteroduplex mobility assays [9, 10], multi-region hybridiza- tion assays [11, 12] and bulk viral sequencing [13, 14] have been used to identify the presence of different viral variants in the same individual. But all these methods lack sensitivity and specificity and require additional clonal sequencing as confirmation [9, 14]. With the intro- duction of next-generation sequencing, enabling the detection of variants representing 1% or less of a virus population, new possibilities for sensitive detection of dual infection arose [14]. The majority of studies concentrated on intersubtype dual infections [8, 12, 15, 16]. Reports of dual infection with viruses of the same subtype (intrasubtype dual infection) are much more limited [17, 18], probably caused by the challenge to differentiate these closely resembling strains. Alternatively, it is possible that the susceptibility for re-infection is higher when the second variant is genetically more distinct from the one already present. An additional difficulty when trying to demonstrate dual infection is that co-presence of the initial and the superinfecting strain may be limited in time. In an extensive study on the dynamics of the virus population after superinfection, Chaillon et al. [19] found that only in 2 of the 7 patients both strains co-circulated during the entire follow-up period. In 4 patients the original strain was rapidly overgrown by the superinfecting strain and in 1 patient a recombi- nant virus took over [19]. To what extent the level of genetic differences between the initial and the superinfecting strain influences the population dynamics remains unclear. Further research on dual infection is needed to elucidate this. Research on co- and superin- fection may be particularly useful for a better understanding of the host-virus interaction and provide valuable information for vaccine development. In Belgium, the overall burden of HIV-1 is limited, but the prevalence in men having sex with men (MSM) is high, reaching up to 6% in some cities [20]. Because of the high HIV prev- alence in this specific population, a high risk of dual infection can be assumed. As in other Western-European countries, the HIV-1 epidemic in MSM is dominated by subtype B [21]. In this study we therefore specifically focused on the detection of dual infection in HIV-1 subtype B infected MSM. Eighty-six MSM, diagnosed with HIV-1 infection between 2008 and 2013, were included. Roche 454 next-generation sequencing of the variable V3 region of the HIV-1 envelope gene was performed to identify dual infection. Clonal sequencing of the HIV-1 prote- ase-reverse transcriptase region was done to confirm the selected cases of dual infection. PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 2 / 15 HIV-1 superinfection in MSM Materials and methods Study population Patients were selected retrospectively from the cohort of HIV-1 infected individuals followed at Ghent University Hospital. About half of the patients in this cohort are MSM diagnosed with HIV-1 subtype B infection. We focused specifically on this population and selected 86 patients diagnosed with a subtype B infection between 2008 and 2013 from whom leftover plasma, collected at two time points with an interval of at least 6 months before initiation of antiretroviral therapy (ART), was available. Subtyping Subtyping of the HIV-1 infection at diagnosis was done using the subtyping tool of Smartgene IDNS (Smartgene, Zug, Switzerland) and the protease and reverse transcriptase sequences gen- erated for the purpose of baseline resistance analysis. Subtyping of the different virus variants present in the patients with dual infection was done using COMET (Context-based Modeling for Expeditious Typing) HIV-1 (Version 1.0) [22] and all clonal protease-reverse transcriptase sequences. Ethical approval The study was approved by the Ethics Committee of Ghent University Hospital (reference number 2014/0173). All patients provided informed consent for the use of their leftover plasma for scientific research. All samples were anonymized before processing. Viral RNA extraction and preparation of the amplicon library Viral RNA was extracted from 140 μL of EDTA plasma using the QIAamp Viral RNA Mini kit (Qiagen, Hilden, Germany). Ten μL of the extracted RNA was reverse transcribed and subse- quently amplified in one step with the Titan one tube RT-PCR kit (Roche, Basel, Switzerland) using two sense primers5’-GAGGATATAATCAGTTTATGG-3’ and5’-GGATATAATCA GYYTATGGGA-3’ and two antisense primers5’-GGTGGGTGCTATTCCTAATGG-3’ and 5’-GGTGGGTGCTAYTCCYAAITG-3’. The resulting amplicon was used as template for the TM second, nested amplification with the FastStart High Fidelity PCR System (Roche, Basel, Switzerland). The nested primers5’-TCAACHCAAYTRCTGTTAAATGG-3’ and5’-ATTT CTGGRTCYCCKCCTG-3’ were extended with Roche multiplex identifiers (MID 1 to MID 34). One unique MID was used per sample. Generated amplicons, containing 345 bp of the V3 region of env (HXB2 position 6,990 to 7,336) were visualized by electrophoresis on a 2% aga- rose gel. Positive reactions were purified with the Agencourt AMPure XP DNA purification TM kit (Analis, Ghent, Belgium) and quantified using the Qubit 2.0 fluorometer and Qubit TM dsDNA HS Assay kit (Life technologies, Carlsbad CA, USA). Thereafter, amplicons were diluted to a concentration of 10 molecules / μL, pooled and further diluted to a concentration of 10 molecules/μL. Next-generation sequencing Equimolar amplicon mixtures were added to DNA capture beads (Roche, Basel, Switzerland) at a ratio of 1.4 copies per bead. The resulting solution was used for emulsion PCR executed with the Lib-A emPCR kit (Roche, Basel, Switzerland). Next-generation sequencing of the PCR products was performed using the Roche 454 GS Junior next-generation sequencer according to the manufacturer's protocol. PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 3 / 15 HIV-1 superinfection in MSM Clonal sequencing of the HIV-1 protease-reverse transcriptase Viral RNA was extracted from 140 μL of EDTA plasma using the QIAamp Viral RNA Mini kit (Qiagen, Hilden, Germany) and reverse transcribed using SuperScript™ III reverse transcriptase (Life Technologies, Carlsbad CA, USA) and the antisense primer5’-GGGATGTGTACTTCTG AACTTAYTYTTGG-3’. The resulting cDNA was diluted to a concentration that resulted in a PCR success rate of maximum 30%. A nested PCR was run with Platinum™ Taq DNA Polymer- ase High Fidelity (Life Technologies, Carlsbad CA, USA), outer primers5’-CTCAATAAAG CTTGCCTTGAGTGC-3’ and5’-GGGATGTGTACTTCTGAACTTAYTYTTGG-3’ and inner primers5’-AAGTAGTGTGTGCCCGTCTGT-3’ and5’-CACCTGCCATCTGTTTTCCATA- 3’. Amplicons were purified with Agencourt AMPure beads (Analis, Ghent, Belgium) and sub- sequently subjected to sequencing of the protease and reverse trancriptase gene using the Big- Dye™ Terminator v3.1 Cycle Sequencing Kit (Life Technologies, Carlsbad CA, USA) following the procedure described before [23]. Sequences were analyzed using the 3500 Genetic Analyzer (Applied Biosystems, Lennik, Belgium). Proofreading was done using Smartgene IDNS (Smart- gene, Zug, Switzerland). Sequences with mixed nucleotides were eliminated. Population HIV-1 protease-reverse transcriptase sequences were obtained through routine laboratory activities for the purpose of drug resistance analysis using an in-house Sanger sequencing protocol. Data analysis Sequences generated by next-generation sequencing were trimmed to 228 nucleotides. Identi- cal reads were clustered before and after manual correction of homopolymer regions using in- house software and aligned in BioEdit [24]. Reads with a coverage of less than 5 or reads shorter than 228 nucleotides were removed. For the sequences generated by clonal or population sequencing, proofreading was done using Smartgene IDNS (Smartgene, Zug, Switzerland). Concatenated sequences of 831 nucleo- tides long were then aligned in BioEdit [24]. Phylogenetic analysis Phylogenetic trees were constructed with PhyML 3.0 using automatic model selection (http:// www.atgc-montpellier.fr/phyml/). The resulting trees were visualized with iTol [25]. Genetic distances were calculated with MEGA 7 based on the Tamura-Nei model [26]. Highlighter plots were generated using the highlighter tool created by the Los Alamos sequence database (https://www.hiv.lanl.gov/content/sequence/HIGHLIGHT/highlighter_top.html). Recombination analysis was performed with Simplot [27]. Time of infection Dating of the infection at diagnosis was done using the BED HIV-1 Incidence EIA and the HIV-1 LAg-Avidity EIA (both from Sedia Biosciences Corporation, Portland, Oregon, USA). The analy- ses were performed on the first sample collected after diagnosis. Only a concordant recent infec- tion result for both assays was accepted as a valid indication of recent infection, all other results were interpreted as long-term infections. This approach has been extensively validated for accu- racy [28] and allows to reliably differentiate infections 130 days from infections > 130 days. Results Characteristics of the study population Next-generation sequencing was attempted on both samples of 86 selected patients. The sequencing was considered successful if a read coverage of more than 500 was obtained. For PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 4 / 15 HIV-1 superinfection in MSM samples failing this criterion, RNA extraction and next-generation sequencing were repeated. Sufficient read coverage for both samples was finally obtained for 74 of the 86 patients (86%), with a median number of sequencing reads per sample of 1588 (IQR: 851±1915). The median time interval between collection of the two samples analyzed was 25 months (IQR: 15±39). Forty two patients of the 74 (57%) were classified as recently infected at the time of collection of the first sample, 32 were diagnosed with a long-term infection. The median age at diagnosis was 36 years (IQR: 28±46), the median viral load (VL) of the first sample collected was 4.45 log copies per milliliter (cp/mL) (IQR: 4.05±4.87 log cp/mL) and the median CD4 T-cell count was 554 cells/μL (IQR: 422±702 cells/μL). Identification of dual infection All next-generation sequences were aligned to construct per patient alignments. These patient alignments were then randomly combined to assemble 6 large alignments, each including 24 or 25 patients. Each patient was included twice in this exercise but, although shuffling was done at random, care was taken to avoid that duplicates were included in the same alignment. A phylogenetic tree was constructed for each of the 6 alignments and visually examined for clustering. A patient was considered HIV-1 dual infected when his sequences were distributed over 2 or more clusters with a bootstrap value of 90% in one of the phylogenetic trees. Ten patients (patients 08, 16, 17, 20, 29, 35, 50, 54, 60 and 66) fulfilled this criterion. For patients 08, 16, 35 and 66, sequences of at least one other patient separated the two phylogenetic clus- ters, indicating high genetic difference between the variants in these patients. An example phy- logenetic tree is shown in Fig 1. The mean genetic distances of the env sequences in the last sample collected from each of the 74 patients fluctuated between 0.92% and 19.94% (mean 3.28%; IQR 1.78% - 4.02%). For the 10 patients with a tree topology suggesting dual infection, the mean distances were 19.94%, 13.10%, 11.02%, 8.28%, 4.12%, 6.35%, 4.52%, 4.40%, 4.12% and 2.59% for patients 66, 08, 35, 16, 50, 54, 17, 29, 20 and 60 respectively. As the overall mean genetic distance may be biased in case of uneven distribution of the variants and is sensitive to differences in the number of se- quences generated per patient, we also calculated the mean pairwise genetic distance between the sequences of the two clusters for those patients with a dichotomized tree topology. The obtained between-cluster mean distances were 56.18%, 30.65%, 26.63%, 24.78%, 12.40%, 9.80%, 8.80%, 8.76%, 8.10% and 7.80% for patients 66, 08, 16, 35, 50, 54, 60, 29, 17 and 20 respectively. Using a cut off of 10%, 5 patients were defined as dually infected. For the 5 other patients, dual infection was likely but the genetic distance was considered too low for definite conclusion. Highlighter plots constructed for the 5 patients with confirmed dual infection are shown in Fig 2. The highlighter plots for the 5 patients with presumed dual infection are pre- sented in S1 Fig. Further analysis was limited to the patients with confirmed dual infection. The characteristics of these patients and information on the tested longitudinal samples is summarized in Table 1. Analysis of intermediate samples Intermediate samples from the 5 patients with confirmed dual infection were also subjected to next-generation V3 sequencing. Three intermediate samples were available for patient 08 and patient 66 and 2 samples for patient 16 and patient 35. No intermediate samples were available for patient 50. The per patient phylogenetic trees (Fig 3A±3D) clearly revealed the presence of 2 genetically different viral variants in at least one of the intermediate samples in all 4 patients (Table 1). PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 5 / 15 HIV-1 superinfection in MSM Fig 1. Example phylogenetic tree visualized with iTol. The tree, rooted on the HXB2 reference, presents the sequences of 25 patients. The sequences of the patients considered dual infected (patients 35 and 66) are marked with respectively a grey and black circle. Mid-branch filled grey circles indicate a bootstrap value 90%. https://doi.org/10.1371/journal.pone.0195679.g001 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 6 / 15 HIV-1 superinfection in MSM PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 7 / 15 HIV-1 superinfection in MSM Fig 2. Highlighter plots constructed with the env V3 sequences of the 5 patients with dual infection. (A) Patient 08, (B) Patient 16, (C) Patient 66, (D) Patient 35, (E) Patient 50. Sequences are aligned to the most abundant sequence in the initial sample of each patient. Mutations compared to the first sequence are color coded. https://doi.org/10.1371/journal.pone.0195679.g002 A significant increase in viral load coinciding with the presumed time point of superinfec- tion was observed in 2 of the superinfected patients. The viral load increased with 2.19 log cp/ mL in patient 08 and with 0.94 log cp/mL in patient 16 (Table 1). Analysis of HIV-1 protease-reverse transcriptase sequences Limiting dilution clonal sequencing of the protease-reverse transcriptase (pol) gene was per- formed for the viruses isolated from the last sample of the 5 patients with dual infection. Respectively 11, 14, 21, 11 and 20 clonal sequences were obtained for patients 08, 16, 35, 50 and 66. They were aligned and analyzed together with sequences obtained as part of routine drug resistance analysis. Sequence alignment, construction of the phylogenetic tree and high- lighter plots was done as described above for the env sequences. Before phylogenetic analysis, the alignment was supplemented with a selection of 162 reference sequences from MSM reflecting the overall sequence variability in the cohort from where the study patients were selected. Inspection of the resulting phylogenetic tree confirmed the dichotomized topology Table 1. Characteristics of the 5 patients with HIV-1 dual infection. Patient Country of Infection stage at Sample Time since first sample CD4 VL Viral Subtype Subtype ID origin diagnosis date (days) count (log c/ variants variant 1 variant 2 (cells/μl) mL) 08 Belgium Long-term 6/06/2008 0 1060 4.54 1 B Unassigned 20/02/2009 259 757 4.19 1 (B with fragments 22/02/2010 626 660 5.13 1 + 2 of A1/D) 24/01/2011 962 627 4.76 1 + 2 19/12/2011 1291 525 5.12 1 + 2 16 Belgium Recent 26/01/2009 0 403 2.71 1 B B 9/06/2009 134 419 2.98 1 + 2 30/10/2009 277 318 5.17 1 + 2 19/01/2010 358 298 4.83 1 + 2 35 Belgium Recent 24/02/2010 0 1020 4.53 1 Majority B B 10/03/2010 14 624 4.58 1 with fragments 16/06/2010 112 968 4.97 1 + 2 of G 9/01/2013 1050 353 4.32 1 + 2 50 France Long-term 13/01/2011 0 482 3.13 1 + 2 B B 26/03/2012 438 498 3.47 1 + 2 66 Belgium Long-term 19/03/2012 0 555 4.45 1 B B 20/02/2013 338 587 4.69 1 + 2 3/07/2013 471 414 4.28 1 + 2 27/11/2013 618 452 4.83 1 + 2 26/03/2014 737 338 4.76 1 + 2 Subtyping based on the protease-reverse transcriptase sequence VL: Viral Load; MSM: Men who have Sex with Men https://doi.org/10.1371/journal.pone.0195679.t001 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 8 / 15 HIV-1 superinfection in MSM Fig 3. Individual phylogenetic trees of the 4 patients with dual infection from whom more than 2 longitudinal samples were analyzed. (A) Patient 66, (B) Patient 35, (C) Patient 16, (D) Patient 08. Bootstrap values 90% are displayed. Black diamond = HXB2. https://doi.org/10.1371/journal.pone.0195679.g003 for patients 08, 50 and 66 and revealed a division over at least 3 clusters for patients 16 and 35 (S2 Fig). Highlighter plots (Fig 4) showed patterns highly suggestive for recombination between the initial and superinfecting strains in patients 16, 66 and 35, an assumption that was PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 9 / 15 HIV-1 superinfection in MSM Fig 4. Highlighter plots constructed with the pol sequences of the 5 patients with dual infection. (A) Patient 08, (B) Patient 16, (C) Patient 50, (D) Patient 66, (E) Patient 35. Sequences are aligned to the sequence of the first sample collected from each patient. Mutations compared to this sequence are color coded. P = population sequencing (numbered by sampling date from old to recent); C = clonal sequencing (numbered by the amount of clones sequenced). https://doi.org/10.1371/journal.pone.0195679.g004 confirmed by Simplot analysis. The presence of recombinants explains the further branching of the tree. In patient 16, 50 and 66 all sequences were identified as subtype B. In patient 35 and 08 respectively the initial and the superinfecting strain were identified as a subtype B recombinant with fragments attributed to subtype G (patient 35) and subtype A and D (patient 08). PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 10 / 15 HIV-1 superinfection in MSM Discussion Phylogenetic analysis of sequences of the V3 region in the HIV-1 envelope gene showed clear indications for dual infection in 5 (6.8%) of 74 randomly selected MSM diagnosed with HIV-1 in Belgium between 2008 and 2013. This figure is most probably an underestimation, because stringent criteria were used to consider viral variants as originating from dual infection. For 5 other patients the topology of the phylogenetic tree was suggestive for dual infection but the genetic difference between the variants was not high enough to confidently exclude that they evolved from the same founder virus. For 64 (86.5%) of the 74 patients phylogenetic analysis of V3 sequences could not reveal any indication for dual infection. The study further concentrated on the 5 patients with high evidence of dual infection. Addi- tional support for the dual infection was sought by deep sequencing of HIV-1 V3 in intermedi- ate samples and by clonal analysis of a larger fragment in the protease-reverse transcriptase gene. In 4 of the 5 patients, a single variant was present in the initial sample with the second variant detected only in samples collected at a later time point. This is a strong argument for superinfection. In one patient both variants were already present in the first sample, hamper- ing discrimination between co- and superinfection. Clonal analysis of the protease-reverse transcriptase gene of the variants present in the last sample of the 5 patients confirmed the presence of at least two genetically different variants. Interestingly, in 3 of the 5 patients the highlighter plots revealed the presence of new recombinant forms containing fragments of both variants. Based on this information we are confident to state that in the tested MSM pop- ulation, at least 4 (5.4%) patients are identified with superinfection and 1 (1.4%) with co- or superinfection. For an additional 5 patients, indications but no convincing evidence for dual infection were found. These numbers are within the estimated overall frequency of dual infec- tion reported by others [5±8]. The fact that after diagnosis some MSM continue to have unprotected sex with known sero- positive partners may be one explanation for the observed high frequency of superinfection. Some have also suggested that the threshold for superinfection is lower in the early stages of infection when the immune response has not yet come to full maturation [29±31]. Research on this is limited but it may be significant that of the 4 superinfected patients identified, 2 were diagnosed early after infection and superinfection was demonstrated respectively 4 and 5 months after diagnosis. Whether dual or superinfection has implications on the clinical course of the disease is not clear. In line with the observation of others [18, 32±34], we noticed an increase in viral load following superinfection in patient 08 and 16. It is therefore important to consider the possibility of superinfection in case of unexpected rises in viral load. Extension of the research on dual infection is partly hampered by the lack of a good detec- tion method. We used next-generation sequencing to screen the genetic variability of the virus population and defined dual infection based on analysis of the phylogenetic tree, calculation of the between-cluster mean genetic distance and visual inspection of the highlighter plots. The outcome of this approach was validated using clonal sequencing of the HIV-1 protease-reverse transcriptase gene. While establishing the method, the lack of well-defined measures of natu- rally occurring intrapatient evolution in V3 posed a difficulty. Initially all patients for whom the V3 sequences were divided over at least two phylogenetic clusters with a bootstrap support of at least 90% were considered for further analysis. Others have used a less stringent bootstrap cut-off of 80% [35±37]. In most studies on dual infection the criterion of phylogenetic linkage is extended with genetic distance calculation and in general a cut-off for pairwise env genetic distance of 5% is used to discriminate dual infection from intrapatient evolution [18, 34, 37]. The overall mean distance however may be influenced by the distribution of the variants and will be lower if one variant is much less represented than the other. We therefore opted for the PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 11 / 15 HIV-1 superinfection in MSM use of between-cluster mean distances. These between-cluster distances were high, especially considering the fact that most were intrasubtype dual infections, but the tree topology and the highlighter plots indeed confirmed the large differences between the variants. It is obvious that the chance to detect dual infection increases when the variants are genetically more distinct. Difficulties to conclude on dual infection with genetically more closely linked vari- ants may explain why intersubtype dual infection is more frequently reported than intrasubtype dual infection. Considering the fact that many geographic areas have a predominant subtype [38] one would expect a higher frequency of intrasubtype recombination. It is possible that the host immune defense is more successful in preventing re-infection with a highly similar virus than in preventing re-infections with a virus of a more aberrant genetic background, but there are no data today to support this hypothesis. Luan et al. detected 18 dual infections in a population of 64 par- ticipants from the Liaoning province in northeastern China and found equal amounts of intrasub- type and intersubtype dual infections [37]. Redd et al. examined 149 patients from Uganda and detected 7 dual infections, 4 intrasubtype and 3 intersubtype [39]. To our knowledge, only one study so far has attempted to analyze the dynamics of the viral population after dual infection. The results of this study revealed different patterns, with individu- als in whom both the original and second variant remained detectable over longer periods of time, individuals in whom one of both variants rapidly overgrew the other variant, and individuals in whom a recombinant virus took over [19]. This implies that also the timing of the sample col- lection is important for reliable detection of dual infection, even if highly sensitive techniques are applied. Our analysis of the protease-reverse transcriptase sequences clearly revealed the presence of recombinants in 3 of the 5 patients with dual infection. This is remarkable, considering the rela- tively small genomic fragment analyzed, and deserves further investigation. A potential hurdle when studying dual infection is the difficulty to discriminate between co-infection and superinfection. Presence of 2 variants in a single sample can indicate either of both [40] and only examination of longitudinal samples will allow to discriminate between co- and superinfection. A limitation of the deep sequencing method that we used for broad scale screening of dual infection is the short sequence read length. This is a limitation not only of the Roche 454 technology but of most of the currently available next-generation sequencing technologies. Ideally, more than one region of the genome must be sequenced to increase the likeliness of detecting dual infection [41]. In conclusion, using next-generation sequencing, we found a high frequency of dual infec- tion in our HIV-1 infected MSM population. Methods to easily identify dually infected indi- viduals will facilitate more in-depth study of this phenomenon and of its consequences for the patient and the virus. This may provide interesting new insights on the interplay between virus and host, information that will be of particular interest for future vaccine research. Supporting information S1 Fig. Highlighter plots of the 5 patients with potential indications of dual infection. (A) Patient 54, (B) Patient 60, (C) Patient 29, (D) Patient 20, (E) Patient 17. Sequences are aligned to the most abundant sequence in the first sample collected. Mutations are color coded. (PDF) S2 Fig. Phylogenetic tree containing the pol sequences from the 5 dual infected patients. This maximum likelihood phylogenetic tree was constructed using pol sequences from 162 MSM and the 5 dual infected patients (08, 16, 35, 50 and 66), rooted on the HXB2 reference. Visualization with iTol. Mid-branch filled grey circles indicate a bootstrap value 90%. (PDF) PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 12 / 15 HIV-1 superinfection in MSM Author Contributions Conceptualization: Laura Hebberecht, Leen Vancoillie, Virginie Mortier, Chris Verhofstede. Data curation: Laura Hebberecht, Kenny Dauwe. Formal analysis: Laura Hebberecht. Funding acquisition: Chris Verhofstede. Investigation: Laura Hebberecht. Methodology: Laura Hebberecht, Leen Vancoillie, Marlies Schauvliege, Delfien Staelens, Kenny Dauwe. Project administration: Laura Hebberecht, Chris Verhofstede. Resources: Marlies Schauvliege, Delfien Staelens, Kenny Dauwe. Software: Leen Vancoillie. Supervision: Chris Verhofstede. Validation: Chris Verhofstede. Visualization: Laura Hebberecht. Writing ± original draft: Laura Hebberecht. Writing ± review & editing: Leen Vancoillie, Kenny Dauwe, Virginie Mortier, Chris Verhofstede. References 1. Waters L, Smit E. HIV-1 superinfection. Curr Opin Infect Dis. 2012; 25(1):42±50. https://doi.org/10. 1097/QCO.0b013e32834ef5af PMID: 22156898 2. Soares de Oliveira AC, Pessoa de Farias R, da Costa AC, Sauer MM, Bassichetto KC, Oliveira SM, et al. Frequency of subtype B and F1 dual infection in HIV-1 positive, Brazilian men who have sex with men. Virol J. 2012; 9:223. https://doi.org/10.1186/1743-422X-9-223 PMID: 23021203 3. Smyth RP, Davenport MP, Mak J. The origin of genetic diversity in HIV-1. Virus Res. 2012; 169(2):415± 29. https://doi.org/10.1016/j.virusres.2012.06.015 PMID: 22728444 4. 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(10):731±6. https://doi.org/10.1056/NEJMoa020263 PMID: 12213944 5. Sheward DJ, Ntale R, Garrett NJ, Woodman ZL, Abdool Karim SS, Williamson C. HIV-1 Superinfection Resembles Primary Infection. J Infect Dis. 2015; 212(6):904±8. https://doi.org/10.1093/infdis/jiv136 PMID: 25754982 6. Vidal N, Diop H, Montavon C, Butel C, Bosch S, Ngole EM, et al. A novel multiregion hybridization assay reveals high frequency of dual inter-subtype infections among HIV-positive individuals in Camer- oon, West Central Africa. Infect Genet Evol. 2013; 14:73±82. https://doi.org/10.1016/j.meegid.2012.11. 017 PMID: 23232100 7. Redd AD, Ssemwanga D, Vandepitte J, Wendel SK, Ndembi N, Bukenya J, et al. Rates of HIV-1 super- infection and primary HIV-1 infection are similar in female sex workers in Uganda. AIDS. 2014; 28 (14):2147±52. https://doi.org/10.1097/QAD.0000000000000365 PMID: 25265078 8. Ssemwanga D, Lyagoba F, Ndembi N, Mayanja BN, Larke N, Wang S, et al. Multiple HIV-1 infections with evidence of recombination in heterosexual partnerships in a low risk Rural Clinical Cohort in Uganda. Virology. 2011; 411(1):113±31. https://doi.org/10.1016/j.virol.2010.12.025 PMID: 21239033 9. Rachinger A, van de Ven TD, Burger JA, Schuitemaker H, van 't Wout AB. Evaluation of pre-screening methods for the identification of HIV-1 superinfection. J Virol Methods. 2010; 165(2):311±7. https://doi. org/10.1016/j.jviromet.2010.02.016 PMID: 20178816 10. Manigart O, Courgnaud V, Sanou O, Valea D, Nagot N, Meda N, et al. HIV-1 superinfections in a cohort of commercial sex workers in Burkina Faso as assessed by an autologous heteroduplex mobility proce- dure. Aids. 2004; 18(12):1645±51. PMID: 15280775 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 13 / 15 HIV-1 superinfection in MSM 11. van der Kuyl AC, Cornelissen M. Identifying HIV-1 dual infections. Retrovirology. 2007; 4:67. https://doi. org/10.1186/1742-4690-4-67 PMID: 17892568 12. Herbinger KH, Gerhardt M, Piyasirisilp S, Mloka D, Arroyo MA, Hoffmann O, et al. Frequency of HIV type 1 dual infection and HIV diversity: analysis of low- and high-risk populations in Mbeya Region, Tan- zania. AIDS Res Hum Retroviruses. 2006; 22(7):599±606. https://doi.org/10.1089/aid.2006.22.599 PMID: 16831083 13. Fang G, Weiser B, Kuiken C, Philpott SM, Rowland-Jones S, Plummer F, et al. Recombination following superinfection by HIV-1. AIDS. 2004; 18(2):153±9. PMID: 15075531 14. Redd AD, Quinn TC, Tobian AAR. Frequency and implications of HIV superinfection. The Lancet Infec- tious Diseases. 2013; 13(7):622±8. https://doi.org/10.1016/S1473-3099(13)70066-5 PMID: 23726798 15. Andreani G, Espada C, Ceballos A, Ambrosioni J, Petroni A, Pugliese D, et al. Detection of HIV-1 dual infections in highly exposed treated patients. Virol J. 2011; 8:392. https://doi.org/10.1186/1743-422X-8- 392 PMID: 21824422 16. Chohan B, Lavreys L, Rainwater SM, Overbaugh J. Evidence for frequent reinfection with human immu- nodeficiency virus type 1 of a different subtype. J Virol. 2005; 79(16):10701±8. https://doi.org/10.1128/ JVI.79.16.10701-10708.2005 PMID: 16051862 17. Kiwelu IE, Novitsky V, Margolin L, Baca J, Manongi R, Sam N, et al. Frequent intra-subtype recombina- tion among HIV-1 circulating in Tanzania. PLoS One. 2013; 8(8):e71131. https://doi.org/10.1371/ journal.pone.0071131 PMID: 23940702 18. Wagner GA, Pacold ME, Kosakovsky Pond SL, Caballero G, Chaillon A, Rudolph AE, et al. Incidence and prevalence of intrasubtype HIV-1 dual infection in at-risk men in the United States. J Infect Dis. 2014; 209(7):1032±8. https://doi.org/10.1093/infdis/jit633 PMID: 24273040 19. Chaillon A, Wagner GA, Hepler NL, Little SJ, Kosakovsky Pond SL, Caballero G, et al. Dynamics of viral evolution and neutralizing antibody response after HIV-1 superinfection. J Virol. 2013; 87 (23):12737±44. https://doi.org/10.1128/JVI.02260-13 PMID: 24049166 20. Vanden Berghe W NC, Buve  A, Beelaert G, Fransen K, Laga M. A venue-based HIV prevalence and behavioural study among men who have sex with men in Antwerp and Ghent, Flanders, Belgium, Octo- ber 2009 to March 2010. Euro Surveill. 2011; 16(28). 21. Chalmet K, Staelens D, Blot S, Dinakis S, Pelgrom J, Plum J, et al. Epidemiological study of phyloge- netic transmission clusters in a local HIV-1 epidemic reveals distinct differences between subtype B and non-B infections. BMC Infect Dis. 2010; 10:262. https://doi.org/10.1186/1471-2334-10-262 PMID: 22. Struck D, Lawyer G, Ternes AM, Schmit JC, Bercoff DP. COMET: adaptive context-based modeling for ultrafast HIV-1 subtype identification. Nucleic Acids Res. 2014; 42(18):e144. https://doi.org/10.1093/ nar/gku739 PMID: 25120265 23. Vancoillie L, Mortier V, Demecheleer E, Schauvliege M, Vandekerckhove L, Vogelaers D, et al. Drug resistance is rarely the cause or consequence of long-term persistent low-level viraemia in HIV-1- infected patients on ART. Antivir Ther. 2015; 20(8):789±94. https://doi.org/10.3851/IMP2966 PMID: 24. Hall T. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series. 1999; 41:95±8. 25. Anisimova M, Gil M, Dufayard JF, Dessimoz C, Gascuel O. Survey of branch support methods demon- strates accuracy, power, and robustness of fast likelihood-based approximation schemes. Syst Biol. 2011; 60(5):685±99. https://doi.org/10.1093/sysbio/syr041 PMID: 21540409 26. Kumar S, Stecher G, Tamura K. MEGA7: Molecular Evolutionary Genetics Analysis Version 7.0 for Big- ger Datasets. Mol Biol Evol. 2016; 33(7):1870±4. https://doi.org/10.1093/molbev/msw054 PMID: 27. Lole KS, Bollinger RC, Paranjape RS, Gadkari D, Kulkarni SS, Novak NG, et al. Full-length human immunodeficiency virus type 1 genomes from subtype C-infected seroconverters in India, with evidence of intersubtype recombination. J Virol. 1999; 73(1):152±60. PMID: 9847317 28. Verhofstede C, Fransen K, Van Den Heuvel A, Van Laethem K, Ruelle J, Vancutsem E, et al. Decision tree for accurate infection timing in individuals newly diagnosed with HIV-1 infection. BMC Infect Dis. 2017; 17(1):738. https://doi.org/10.1186/s12879-017-2850-6 PMID: 29187159 29. Koelsch KK, Smith DM, Little SJ, Ignacio CC, Macaranas TR, Brown AJ, et al. Clade B HIV-1 superin- fection with wild-type virus after primary infection with drug-resistant clade B virus. AIDS. 2003; 17(7): F11±6. https://doi.org/10.1097/01.aids.0000060361.78202.4c PMID: 12700477 30. Cornelissen M, Pasternak AO, Grijsen ML, Zorgdrager F, Bakker M, Blom P, et al. HIV-1 dual infection is associated with faster CD4+ T-cell decline in a cohort of men with primary HIV infection. Clin Infect Dis. 2012; 54(4):539±47. https://doi.org/10.1093/cid/cir849 PMID: 22157174 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 14 / 15 HIV-1 superinfection in MSM 31. Wagner GA, Landais E, Caballero G, Phung P, Kosakovsky Pond SL, Poignard P, et al. Intrasubtype B HIV-1 Superinfection Correlates with Delayed Neutralizing Antibody Response. J Virol. 2017;91(17). 32. Redd AD, Mullis CE, Wendel SK, Sheward D, Martens C, Bruno D, et al. Limited HIV-1 superinfection in seroconverters from the CAPRISA 004 Microbicide Trial. J Clin Microbiol. 2014; 52(3):844±8. https:// doi.org/10.1128/JCM.03143-13 PMID: 24371237 33. Ronen K, Richardson BA, Graham SM, Jaoko W, Mandaliya K, McClelland RS, et al. HIV-1 superinfec- tion is associated with an accelerated viral load increase but has a limited impact on disease progres- sion. AIDS. 2014; 28(15):2281±6. https://doi.org/10.1097/QAD.0000000000000422 PMID: 25102090 34. Pacold ME, Pond SL, Wagner GA, Delport W, Bourque DL, Richman DD, et al. Clinical, virologic, and immunologic correlates of HIV-1 intraclade B dual infection among men who have sex with men. AIDS. 2012; 26(2):157±65. https://doi.org/10.1097/QAD.0b013e32834dcd26 PMID: 22045341 35. Pernas M, Casado C, Fuentes R, Perez-Elias MJ, Lopez-Galindez C. A dual superinfection and recom- bination within HIV-1 subtype B 12 years after primoinfection. Jaids-J Acq Imm Def. 2006; 42(1):12±8. 36. Kiwelu IE, Novitsky V, Margolin L, Baca J, Manongi R, Sam N, et al. HIV-1 subtypes and recombinants in Northern Tanzania: distribution of viral quasispecies. PLoS One. 2012; 7(10):e47605. https://doi.org/ 10.1371/journal.pone.0047605 PMID: 23118882 37. Luan H, Han XX, Yu XO, An MH, Zhang H, Zhao B, et al. Dual Infection Contributes to Rapid Disease Progression in Men Who Have Sex With Men in China. Jaids-J Acq Imm Def. 2017; 75(4):480±7. 38. Pacold M, Smith D, Little S, Cheng PM, Jordan P, Ignacio C, et al. Comparison of methods to detect HIV dual infection. AIDS Res Hum Retroviruses. 2010; 26(12):1291±8. https://doi.org/10.1089/aid. 2010.0042 PMID: 20954840 39. Redd AD, Mullis CE, Serwadda D, Kong X, Martens C, Ricklefs SM, et al. The rates of HIV superinfec- tion and primary HIV incidence in a general population in Rakai, Uganda. J Infect Dis. 2012; 206 (2):267±74. https://doi.org/10.1093/infdis/jis325 PMID: 22675216 40. Novitsky V, Moyo S, Wang R, Gaseitsiwe S, Essex M. Deciphering Multiplicity of HIV-1C Infection: Transmission of Closely Related Multiple Viral Lineages. PLoS One. 2016; 11(11):e0166746. https:// doi.org/10.1371/journal.pone.0166746 PMID: 27893822 41. Piantadosi A, Ngayo MO, Chohan B, Overbaugh J. Examination of a second region of the HIV type 1 genome reveals additional cases of superinfection. AIDS Res Hum Retroviruses. 2008; 24(9):1221. https://doi.org/10.1089/aid.2008.0100 PMID: 18729772 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 15 / 15 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png PLoS ONE Pubmed Central

Frequency of occurrence of HIV-1 dual infection in a Belgian MSM population

Loading next page...
 
/lp/pubmed-central/frequency-of-occurrence-of-hiv-1-dual-infection-in-a-belgian-msm-YcWdk0RpFW

References (83)

Publisher
Pubmed Central
Copyright
© 2018 Hebberecht et al
ISSN
1932-6203
eISSN
1932-6203
DOI
10.1371/journal.pone.0195679
Publisher site
See Article on Publisher Site

Abstract

a1111111111 a1111111111 Introduction HIV-1 dual infection is a condition that results from infection with at least two HIV-1 variants from different sources. The scarceness of information on this condition is partly due to the OPENACCESS fact that its detection is technically challenging. Using next-generation sequencing we Citation: Hebberecht L, Vancoillie L, Schauvliege defined the extent of HIV-1 dual infection in a cohort of men who have sex with men (MSM). M, Staelens D, Dauwe K, Mortier V, et al. (2018) Frequency of occurrence of HIV-1 dual infection in a Belgian MSM population. PLoS ONE 13(4): Material & methods e0195679. https://doi.org/10.1371/journal. Eighty-six MSM, diagnosed with HIV-1 subtype B infection between 2008 and 2013 were pone.0195679 selected for next-generation sequencing of the HIV-1 envelope V3. Sequencing was per- Editor: Luis Menendez-Arias, Consejo Superior de formed on 2 plasma samples collected with an interval of> 6 months before the initiation of Investigaciones Cientificas, SPAIN antiretroviral therapy. Maximum likelihood phylogenetic trees were inspected for dual infec- Received: October 20, 2017 tion, defined as the presence of two or more monophyletic clusters with 90% bootstrap Accepted: March 27, 2018 support and a mean between-cluster genetic distance of 10%. To confirm dual infection, Published: April 6, 2018 deep V3 sequencing of intermediate samples was performed as well as clonal sequencing Copyright:© 2018 Hebberecht et al. This is an of the HIV-1 protease-reverse transcriptase gene. open access article distributed under the terms of the Creative Commons Attribution License, which Results permits unrestricted use, distribution, and reproduction in any medium, provided the original Five of the 74 patients (6.8%) for whom deep sequencing was successful, showed clear evi- author and source are credited. dence of dual infection. In 4 of them, the second strain was absent in the first sample but Data Availability Statement: All relevant data are occurred in subsequent samples. This was highly suggestive for superinfection. In 3 patients within the paper and its Supporting Information both virus variants were of subtype B, in 2 patients at least one of the variants was a subtype files. B/non-B recombinant virus. Funding: The AIDS Reference Laboratory of Ghent is supported by the Belgian Ministery of Social Affairs through a fund within the Health Insurance Conclusions System. The funder had no role in study design, Dual infection was confirmed in 6.8% of MSM diagnosed with HIV-1 in Belgium. This preva- data collection and analysis, decision to publish, or preparation of the manuscript. lence is probably an underestimation, because stringent criteria were used to classify viral variants as originating from a new infection event. Competing interests: The authors have declared that no competing interests exist. PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 1 / 15 HIV-1 superinfection in MSM Introduction Human immunodeficiency virus type 1 (HIV-1) dual infection occurs when an HIV-1 infected patient is infected by a new viral variant from another host. This results in the presence of two variants, each originating from an independent infection event. When the second infection occurs nearly simultaneously with the first, or at least before seroconversion, the condition is generally called co-infection. When the second infection occurs after seroconversion it is called superinfection [1]. One of the firmest indications that co- or superinfection is possible is the existence of recombinant viruses. Recombination between genetically distinct viruses can only occur after infection of a single cell with both variants. Recombinant viruses can remain restricted to a single patient (unique recombinant forms; URF) or be further transmitted. Recombinant forms detected in at least three unrelated individuals are called circulating recombinant forms (CRF). CRF contribute extensively to the genetic variability of HIV-1 [2, 3] and the constant increase in number of reported CRF (https://www.hiv.lanl.gov/content/ sequence/HIV/CRFs/CRFs.html) suggests high frequencies of dual infection. The first patient with HIV-1 dual infection was documented in 2002 [4]. Since then, several other, mostly isolated cases of dual infection, have been reported [1]. Estimates of the overall frequency of this condition vary between 0 and 20% [5±8]. The number of well documented superinfections, however, remains low. Difficulties to demonstrate this condition may be the most important reason for that. Heteroduplex mobility assays [9, 10], multi-region hybridiza- tion assays [11, 12] and bulk viral sequencing [13, 14] have been used to identify the presence of different viral variants in the same individual. But all these methods lack sensitivity and specificity and require additional clonal sequencing as confirmation [9, 14]. With the intro- duction of next-generation sequencing, enabling the detection of variants representing 1% or less of a virus population, new possibilities for sensitive detection of dual infection arose [14]. The majority of studies concentrated on intersubtype dual infections [8, 12, 15, 16]. Reports of dual infection with viruses of the same subtype (intrasubtype dual infection) are much more limited [17, 18], probably caused by the challenge to differentiate these closely resembling strains. Alternatively, it is possible that the susceptibility for re-infection is higher when the second variant is genetically more distinct from the one already present. An additional difficulty when trying to demonstrate dual infection is that co-presence of the initial and the superinfecting strain may be limited in time. In an extensive study on the dynamics of the virus population after superinfection, Chaillon et al. [19] found that only in 2 of the 7 patients both strains co-circulated during the entire follow-up period. In 4 patients the original strain was rapidly overgrown by the superinfecting strain and in 1 patient a recombi- nant virus took over [19]. To what extent the level of genetic differences between the initial and the superinfecting strain influences the population dynamics remains unclear. Further research on dual infection is needed to elucidate this. Research on co- and superin- fection may be particularly useful for a better understanding of the host-virus interaction and provide valuable information for vaccine development. In Belgium, the overall burden of HIV-1 is limited, but the prevalence in men having sex with men (MSM) is high, reaching up to 6% in some cities [20]. Because of the high HIV prev- alence in this specific population, a high risk of dual infection can be assumed. As in other Western-European countries, the HIV-1 epidemic in MSM is dominated by subtype B [21]. In this study we therefore specifically focused on the detection of dual infection in HIV-1 subtype B infected MSM. Eighty-six MSM, diagnosed with HIV-1 infection between 2008 and 2013, were included. Roche 454 next-generation sequencing of the variable V3 region of the HIV-1 envelope gene was performed to identify dual infection. Clonal sequencing of the HIV-1 prote- ase-reverse transcriptase region was done to confirm the selected cases of dual infection. PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 2 / 15 HIV-1 superinfection in MSM Materials and methods Study population Patients were selected retrospectively from the cohort of HIV-1 infected individuals followed at Ghent University Hospital. About half of the patients in this cohort are MSM diagnosed with HIV-1 subtype B infection. We focused specifically on this population and selected 86 patients diagnosed with a subtype B infection between 2008 and 2013 from whom leftover plasma, collected at two time points with an interval of at least 6 months before initiation of antiretroviral therapy (ART), was available. Subtyping Subtyping of the HIV-1 infection at diagnosis was done using the subtyping tool of Smartgene IDNS (Smartgene, Zug, Switzerland) and the protease and reverse transcriptase sequences gen- erated for the purpose of baseline resistance analysis. Subtyping of the different virus variants present in the patients with dual infection was done using COMET (Context-based Modeling for Expeditious Typing) HIV-1 (Version 1.0) [22] and all clonal protease-reverse transcriptase sequences. Ethical approval The study was approved by the Ethics Committee of Ghent University Hospital (reference number 2014/0173). All patients provided informed consent for the use of their leftover plasma for scientific research. All samples were anonymized before processing. Viral RNA extraction and preparation of the amplicon library Viral RNA was extracted from 140 μL of EDTA plasma using the QIAamp Viral RNA Mini kit (Qiagen, Hilden, Germany). Ten μL of the extracted RNA was reverse transcribed and subse- quently amplified in one step with the Titan one tube RT-PCR kit (Roche, Basel, Switzerland) using two sense primers5’-GAGGATATAATCAGTTTATGG-3’ and5’-GGATATAATCA GYYTATGGGA-3’ and two antisense primers5’-GGTGGGTGCTATTCCTAATGG-3’ and 5’-GGTGGGTGCTAYTCCYAAITG-3’. The resulting amplicon was used as template for the TM second, nested amplification with the FastStart High Fidelity PCR System (Roche, Basel, Switzerland). The nested primers5’-TCAACHCAAYTRCTGTTAAATGG-3’ and5’-ATTT CTGGRTCYCCKCCTG-3’ were extended with Roche multiplex identifiers (MID 1 to MID 34). One unique MID was used per sample. Generated amplicons, containing 345 bp of the V3 region of env (HXB2 position 6,990 to 7,336) were visualized by electrophoresis on a 2% aga- rose gel. Positive reactions were purified with the Agencourt AMPure XP DNA purification TM kit (Analis, Ghent, Belgium) and quantified using the Qubit 2.0 fluorometer and Qubit TM dsDNA HS Assay kit (Life technologies, Carlsbad CA, USA). Thereafter, amplicons were diluted to a concentration of 10 molecules / μL, pooled and further diluted to a concentration of 10 molecules/μL. Next-generation sequencing Equimolar amplicon mixtures were added to DNA capture beads (Roche, Basel, Switzerland) at a ratio of 1.4 copies per bead. The resulting solution was used for emulsion PCR executed with the Lib-A emPCR kit (Roche, Basel, Switzerland). Next-generation sequencing of the PCR products was performed using the Roche 454 GS Junior next-generation sequencer according to the manufacturer's protocol. PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 3 / 15 HIV-1 superinfection in MSM Clonal sequencing of the HIV-1 protease-reverse transcriptase Viral RNA was extracted from 140 μL of EDTA plasma using the QIAamp Viral RNA Mini kit (Qiagen, Hilden, Germany) and reverse transcribed using SuperScript™ III reverse transcriptase (Life Technologies, Carlsbad CA, USA) and the antisense primer5’-GGGATGTGTACTTCTG AACTTAYTYTTGG-3’. The resulting cDNA was diluted to a concentration that resulted in a PCR success rate of maximum 30%. A nested PCR was run with Platinum™ Taq DNA Polymer- ase High Fidelity (Life Technologies, Carlsbad CA, USA), outer primers5’-CTCAATAAAG CTTGCCTTGAGTGC-3’ and5’-GGGATGTGTACTTCTGAACTTAYTYTTGG-3’ and inner primers5’-AAGTAGTGTGTGCCCGTCTGT-3’ and5’-CACCTGCCATCTGTTTTCCATA- 3’. Amplicons were purified with Agencourt AMPure beads (Analis, Ghent, Belgium) and sub- sequently subjected to sequencing of the protease and reverse trancriptase gene using the Big- Dye™ Terminator v3.1 Cycle Sequencing Kit (Life Technologies, Carlsbad CA, USA) following the procedure described before [23]. Sequences were analyzed using the 3500 Genetic Analyzer (Applied Biosystems, Lennik, Belgium). Proofreading was done using Smartgene IDNS (Smart- gene, Zug, Switzerland). Sequences with mixed nucleotides were eliminated. Population HIV-1 protease-reverse transcriptase sequences were obtained through routine laboratory activities for the purpose of drug resistance analysis using an in-house Sanger sequencing protocol. Data analysis Sequences generated by next-generation sequencing were trimmed to 228 nucleotides. Identi- cal reads were clustered before and after manual correction of homopolymer regions using in- house software and aligned in BioEdit [24]. Reads with a coverage of less than 5 or reads shorter than 228 nucleotides were removed. For the sequences generated by clonal or population sequencing, proofreading was done using Smartgene IDNS (Smartgene, Zug, Switzerland). Concatenated sequences of 831 nucleo- tides long were then aligned in BioEdit [24]. Phylogenetic analysis Phylogenetic trees were constructed with PhyML 3.0 using automatic model selection (http:// www.atgc-montpellier.fr/phyml/). The resulting trees were visualized with iTol [25]. Genetic distances were calculated with MEGA 7 based on the Tamura-Nei model [26]. Highlighter plots were generated using the highlighter tool created by the Los Alamos sequence database (https://www.hiv.lanl.gov/content/sequence/HIGHLIGHT/highlighter_top.html). Recombination analysis was performed with Simplot [27]. Time of infection Dating of the infection at diagnosis was done using the BED HIV-1 Incidence EIA and the HIV-1 LAg-Avidity EIA (both from Sedia Biosciences Corporation, Portland, Oregon, USA). The analy- ses were performed on the first sample collected after diagnosis. Only a concordant recent infec- tion result for both assays was accepted as a valid indication of recent infection, all other results were interpreted as long-term infections. This approach has been extensively validated for accu- racy [28] and allows to reliably differentiate infections 130 days from infections > 130 days. Results Characteristics of the study population Next-generation sequencing was attempted on both samples of 86 selected patients. The sequencing was considered successful if a read coverage of more than 500 was obtained. For PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 4 / 15 HIV-1 superinfection in MSM samples failing this criterion, RNA extraction and next-generation sequencing were repeated. Sufficient read coverage for both samples was finally obtained for 74 of the 86 patients (86%), with a median number of sequencing reads per sample of 1588 (IQR: 851±1915). The median time interval between collection of the two samples analyzed was 25 months (IQR: 15±39). Forty two patients of the 74 (57%) were classified as recently infected at the time of collection of the first sample, 32 were diagnosed with a long-term infection. The median age at diagnosis was 36 years (IQR: 28±46), the median viral load (VL) of the first sample collected was 4.45 log copies per milliliter (cp/mL) (IQR: 4.05±4.87 log cp/mL) and the median CD4 T-cell count was 554 cells/μL (IQR: 422±702 cells/μL). Identification of dual infection All next-generation sequences were aligned to construct per patient alignments. These patient alignments were then randomly combined to assemble 6 large alignments, each including 24 or 25 patients. Each patient was included twice in this exercise but, although shuffling was done at random, care was taken to avoid that duplicates were included in the same alignment. A phylogenetic tree was constructed for each of the 6 alignments and visually examined for clustering. A patient was considered HIV-1 dual infected when his sequences were distributed over 2 or more clusters with a bootstrap value of 90% in one of the phylogenetic trees. Ten patients (patients 08, 16, 17, 20, 29, 35, 50, 54, 60 and 66) fulfilled this criterion. For patients 08, 16, 35 and 66, sequences of at least one other patient separated the two phylogenetic clus- ters, indicating high genetic difference between the variants in these patients. An example phy- logenetic tree is shown in Fig 1. The mean genetic distances of the env sequences in the last sample collected from each of the 74 patients fluctuated between 0.92% and 19.94% (mean 3.28%; IQR 1.78% - 4.02%). For the 10 patients with a tree topology suggesting dual infection, the mean distances were 19.94%, 13.10%, 11.02%, 8.28%, 4.12%, 6.35%, 4.52%, 4.40%, 4.12% and 2.59% for patients 66, 08, 35, 16, 50, 54, 17, 29, 20 and 60 respectively. As the overall mean genetic distance may be biased in case of uneven distribution of the variants and is sensitive to differences in the number of se- quences generated per patient, we also calculated the mean pairwise genetic distance between the sequences of the two clusters for those patients with a dichotomized tree topology. The obtained between-cluster mean distances were 56.18%, 30.65%, 26.63%, 24.78%, 12.40%, 9.80%, 8.80%, 8.76%, 8.10% and 7.80% for patients 66, 08, 16, 35, 50, 54, 60, 29, 17 and 20 respectively. Using a cut off of 10%, 5 patients were defined as dually infected. For the 5 other patients, dual infection was likely but the genetic distance was considered too low for definite conclusion. Highlighter plots constructed for the 5 patients with confirmed dual infection are shown in Fig 2. The highlighter plots for the 5 patients with presumed dual infection are pre- sented in S1 Fig. Further analysis was limited to the patients with confirmed dual infection. The characteristics of these patients and information on the tested longitudinal samples is summarized in Table 1. Analysis of intermediate samples Intermediate samples from the 5 patients with confirmed dual infection were also subjected to next-generation V3 sequencing. Three intermediate samples were available for patient 08 and patient 66 and 2 samples for patient 16 and patient 35. No intermediate samples were available for patient 50. The per patient phylogenetic trees (Fig 3A±3D) clearly revealed the presence of 2 genetically different viral variants in at least one of the intermediate samples in all 4 patients (Table 1). PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 5 / 15 HIV-1 superinfection in MSM Fig 1. Example phylogenetic tree visualized with iTol. The tree, rooted on the HXB2 reference, presents the sequences of 25 patients. The sequences of the patients considered dual infected (patients 35 and 66) are marked with respectively a grey and black circle. Mid-branch filled grey circles indicate a bootstrap value 90%. https://doi.org/10.1371/journal.pone.0195679.g001 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 6 / 15 HIV-1 superinfection in MSM PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 7 / 15 HIV-1 superinfection in MSM Fig 2. Highlighter plots constructed with the env V3 sequences of the 5 patients with dual infection. (A) Patient 08, (B) Patient 16, (C) Patient 66, (D) Patient 35, (E) Patient 50. Sequences are aligned to the most abundant sequence in the initial sample of each patient. Mutations compared to the first sequence are color coded. https://doi.org/10.1371/journal.pone.0195679.g002 A significant increase in viral load coinciding with the presumed time point of superinfec- tion was observed in 2 of the superinfected patients. The viral load increased with 2.19 log cp/ mL in patient 08 and with 0.94 log cp/mL in patient 16 (Table 1). Analysis of HIV-1 protease-reverse transcriptase sequences Limiting dilution clonal sequencing of the protease-reverse transcriptase (pol) gene was per- formed for the viruses isolated from the last sample of the 5 patients with dual infection. Respectively 11, 14, 21, 11 and 20 clonal sequences were obtained for patients 08, 16, 35, 50 and 66. They were aligned and analyzed together with sequences obtained as part of routine drug resistance analysis. Sequence alignment, construction of the phylogenetic tree and high- lighter plots was done as described above for the env sequences. Before phylogenetic analysis, the alignment was supplemented with a selection of 162 reference sequences from MSM reflecting the overall sequence variability in the cohort from where the study patients were selected. Inspection of the resulting phylogenetic tree confirmed the dichotomized topology Table 1. Characteristics of the 5 patients with HIV-1 dual infection. Patient Country of Infection stage at Sample Time since first sample CD4 VL Viral Subtype Subtype ID origin diagnosis date (days) count (log c/ variants variant 1 variant 2 (cells/μl) mL) 08 Belgium Long-term 6/06/2008 0 1060 4.54 1 B Unassigned 20/02/2009 259 757 4.19 1 (B with fragments 22/02/2010 626 660 5.13 1 + 2 of A1/D) 24/01/2011 962 627 4.76 1 + 2 19/12/2011 1291 525 5.12 1 + 2 16 Belgium Recent 26/01/2009 0 403 2.71 1 B B 9/06/2009 134 419 2.98 1 + 2 30/10/2009 277 318 5.17 1 + 2 19/01/2010 358 298 4.83 1 + 2 35 Belgium Recent 24/02/2010 0 1020 4.53 1 Majority B B 10/03/2010 14 624 4.58 1 with fragments 16/06/2010 112 968 4.97 1 + 2 of G 9/01/2013 1050 353 4.32 1 + 2 50 France Long-term 13/01/2011 0 482 3.13 1 + 2 B B 26/03/2012 438 498 3.47 1 + 2 66 Belgium Long-term 19/03/2012 0 555 4.45 1 B B 20/02/2013 338 587 4.69 1 + 2 3/07/2013 471 414 4.28 1 + 2 27/11/2013 618 452 4.83 1 + 2 26/03/2014 737 338 4.76 1 + 2 Subtyping based on the protease-reverse transcriptase sequence VL: Viral Load; MSM: Men who have Sex with Men https://doi.org/10.1371/journal.pone.0195679.t001 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 8 / 15 HIV-1 superinfection in MSM Fig 3. Individual phylogenetic trees of the 4 patients with dual infection from whom more than 2 longitudinal samples were analyzed. (A) Patient 66, (B) Patient 35, (C) Patient 16, (D) Patient 08. Bootstrap values 90% are displayed. Black diamond = HXB2. https://doi.org/10.1371/journal.pone.0195679.g003 for patients 08, 50 and 66 and revealed a division over at least 3 clusters for patients 16 and 35 (S2 Fig). Highlighter plots (Fig 4) showed patterns highly suggestive for recombination between the initial and superinfecting strains in patients 16, 66 and 35, an assumption that was PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 9 / 15 HIV-1 superinfection in MSM Fig 4. Highlighter plots constructed with the pol sequences of the 5 patients with dual infection. (A) Patient 08, (B) Patient 16, (C) Patient 50, (D) Patient 66, (E) Patient 35. Sequences are aligned to the sequence of the first sample collected from each patient. Mutations compared to this sequence are color coded. P = population sequencing (numbered by sampling date from old to recent); C = clonal sequencing (numbered by the amount of clones sequenced). https://doi.org/10.1371/journal.pone.0195679.g004 confirmed by Simplot analysis. The presence of recombinants explains the further branching of the tree. In patient 16, 50 and 66 all sequences were identified as subtype B. In patient 35 and 08 respectively the initial and the superinfecting strain were identified as a subtype B recombinant with fragments attributed to subtype G (patient 35) and subtype A and D (patient 08). PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 10 / 15 HIV-1 superinfection in MSM Discussion Phylogenetic analysis of sequences of the V3 region in the HIV-1 envelope gene showed clear indications for dual infection in 5 (6.8%) of 74 randomly selected MSM diagnosed with HIV-1 in Belgium between 2008 and 2013. This figure is most probably an underestimation, because stringent criteria were used to consider viral variants as originating from dual infection. For 5 other patients the topology of the phylogenetic tree was suggestive for dual infection but the genetic difference between the variants was not high enough to confidently exclude that they evolved from the same founder virus. For 64 (86.5%) of the 74 patients phylogenetic analysis of V3 sequences could not reveal any indication for dual infection. The study further concentrated on the 5 patients with high evidence of dual infection. Addi- tional support for the dual infection was sought by deep sequencing of HIV-1 V3 in intermedi- ate samples and by clonal analysis of a larger fragment in the protease-reverse transcriptase gene. In 4 of the 5 patients, a single variant was present in the initial sample with the second variant detected only in samples collected at a later time point. This is a strong argument for superinfection. In one patient both variants were already present in the first sample, hamper- ing discrimination between co- and superinfection. Clonal analysis of the protease-reverse transcriptase gene of the variants present in the last sample of the 5 patients confirmed the presence of at least two genetically different variants. Interestingly, in 3 of the 5 patients the highlighter plots revealed the presence of new recombinant forms containing fragments of both variants. Based on this information we are confident to state that in the tested MSM pop- ulation, at least 4 (5.4%) patients are identified with superinfection and 1 (1.4%) with co- or superinfection. For an additional 5 patients, indications but no convincing evidence for dual infection were found. These numbers are within the estimated overall frequency of dual infec- tion reported by others [5±8]. The fact that after diagnosis some MSM continue to have unprotected sex with known sero- positive partners may be one explanation for the observed high frequency of superinfection. Some have also suggested that the threshold for superinfection is lower in the early stages of infection when the immune response has not yet come to full maturation [29±31]. Research on this is limited but it may be significant that of the 4 superinfected patients identified, 2 were diagnosed early after infection and superinfection was demonstrated respectively 4 and 5 months after diagnosis. Whether dual or superinfection has implications on the clinical course of the disease is not clear. In line with the observation of others [18, 32±34], we noticed an increase in viral load following superinfection in patient 08 and 16. It is therefore important to consider the possibility of superinfection in case of unexpected rises in viral load. Extension of the research on dual infection is partly hampered by the lack of a good detec- tion method. We used next-generation sequencing to screen the genetic variability of the virus population and defined dual infection based on analysis of the phylogenetic tree, calculation of the between-cluster mean genetic distance and visual inspection of the highlighter plots. The outcome of this approach was validated using clonal sequencing of the HIV-1 protease-reverse transcriptase gene. While establishing the method, the lack of well-defined measures of natu- rally occurring intrapatient evolution in V3 posed a difficulty. Initially all patients for whom the V3 sequences were divided over at least two phylogenetic clusters with a bootstrap support of at least 90% were considered for further analysis. Others have used a less stringent bootstrap cut-off of 80% [35±37]. In most studies on dual infection the criterion of phylogenetic linkage is extended with genetic distance calculation and in general a cut-off for pairwise env genetic distance of 5% is used to discriminate dual infection from intrapatient evolution [18, 34, 37]. The overall mean distance however may be influenced by the distribution of the variants and will be lower if one variant is much less represented than the other. We therefore opted for the PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 11 / 15 HIV-1 superinfection in MSM use of between-cluster mean distances. These between-cluster distances were high, especially considering the fact that most were intrasubtype dual infections, but the tree topology and the highlighter plots indeed confirmed the large differences between the variants. It is obvious that the chance to detect dual infection increases when the variants are genetically more distinct. Difficulties to conclude on dual infection with genetically more closely linked vari- ants may explain why intersubtype dual infection is more frequently reported than intrasubtype dual infection. Considering the fact that many geographic areas have a predominant subtype [38] one would expect a higher frequency of intrasubtype recombination. It is possible that the host immune defense is more successful in preventing re-infection with a highly similar virus than in preventing re-infections with a virus of a more aberrant genetic background, but there are no data today to support this hypothesis. Luan et al. detected 18 dual infections in a population of 64 par- ticipants from the Liaoning province in northeastern China and found equal amounts of intrasub- type and intersubtype dual infections [37]. Redd et al. examined 149 patients from Uganda and detected 7 dual infections, 4 intrasubtype and 3 intersubtype [39]. To our knowledge, only one study so far has attempted to analyze the dynamics of the viral population after dual infection. The results of this study revealed different patterns, with individu- als in whom both the original and second variant remained detectable over longer periods of time, individuals in whom one of both variants rapidly overgrew the other variant, and individuals in whom a recombinant virus took over [19]. This implies that also the timing of the sample col- lection is important for reliable detection of dual infection, even if highly sensitive techniques are applied. Our analysis of the protease-reverse transcriptase sequences clearly revealed the presence of recombinants in 3 of the 5 patients with dual infection. This is remarkable, considering the rela- tively small genomic fragment analyzed, and deserves further investigation. A potential hurdle when studying dual infection is the difficulty to discriminate between co-infection and superinfection. Presence of 2 variants in a single sample can indicate either of both [40] and only examination of longitudinal samples will allow to discriminate between co- and superinfection. A limitation of the deep sequencing method that we used for broad scale screening of dual infection is the short sequence read length. This is a limitation not only of the Roche 454 technology but of most of the currently available next-generation sequencing technologies. Ideally, more than one region of the genome must be sequenced to increase the likeliness of detecting dual infection [41]. In conclusion, using next-generation sequencing, we found a high frequency of dual infec- tion in our HIV-1 infected MSM population. Methods to easily identify dually infected indi- viduals will facilitate more in-depth study of this phenomenon and of its consequences for the patient and the virus. This may provide interesting new insights on the interplay between virus and host, information that will be of particular interest for future vaccine research. Supporting information S1 Fig. Highlighter plots of the 5 patients with potential indications of dual infection. (A) Patient 54, (B) Patient 60, (C) Patient 29, (D) Patient 20, (E) Patient 17. Sequences are aligned to the most abundant sequence in the first sample collected. Mutations are color coded. (PDF) S2 Fig. Phylogenetic tree containing the pol sequences from the 5 dual infected patients. This maximum likelihood phylogenetic tree was constructed using pol sequences from 162 MSM and the 5 dual infected patients (08, 16, 35, 50 and 66), rooted on the HXB2 reference. Visualization with iTol. Mid-branch filled grey circles indicate a bootstrap value 90%. (PDF) PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 12 / 15 HIV-1 superinfection in MSM Author Contributions Conceptualization: Laura Hebberecht, Leen Vancoillie, Virginie Mortier, Chris Verhofstede. Data curation: Laura Hebberecht, Kenny Dauwe. Formal analysis: Laura Hebberecht. Funding acquisition: Chris Verhofstede. Investigation: Laura Hebberecht. Methodology: Laura Hebberecht, Leen Vancoillie, Marlies Schauvliege, Delfien Staelens, Kenny Dauwe. Project administration: Laura Hebberecht, Chris Verhofstede. Resources: Marlies Schauvliege, Delfien Staelens, Kenny Dauwe. Software: Leen Vancoillie. Supervision: Chris Verhofstede. Validation: Chris Verhofstede. Visualization: Laura Hebberecht. Writing ± original draft: Laura Hebberecht. Writing ± review & editing: Leen Vancoillie, Kenny Dauwe, Virginie Mortier, Chris Verhofstede. References 1. Waters L, Smit E. HIV-1 superinfection. Curr Opin Infect Dis. 2012; 25(1):42±50. https://doi.org/10. 1097/QCO.0b013e32834ef5af PMID: 22156898 2. Soares de Oliveira AC, Pessoa de Farias R, da Costa AC, Sauer MM, Bassichetto KC, Oliveira SM, et al. Frequency of subtype B and F1 dual infection in HIV-1 positive, Brazilian men who have sex with men. Virol J. 2012; 9:223. https://doi.org/10.1186/1743-422X-9-223 PMID: 23021203 3. Smyth RP, Davenport MP, Mak J. The origin of genetic diversity in HIV-1. Virus Res. 2012; 169(2):415± 29. https://doi.org/10.1016/j.virusres.2012.06.015 PMID: 22728444 4. 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(10):731±6. https://doi.org/10.1056/NEJMoa020263 PMID: 12213944 5. Sheward DJ, Ntale R, Garrett NJ, Woodman ZL, Abdool Karim SS, Williamson C. HIV-1 Superinfection Resembles Primary Infection. J Infect Dis. 2015; 212(6):904±8. https://doi.org/10.1093/infdis/jiv136 PMID: 25754982 6. Vidal N, Diop H, Montavon C, Butel C, Bosch S, Ngole EM, et al. A novel multiregion hybridization assay reveals high frequency of dual inter-subtype infections among HIV-positive individuals in Camer- oon, West Central Africa. Infect Genet Evol. 2013; 14:73±82. https://doi.org/10.1016/j.meegid.2012.11. 017 PMID: 23232100 7. Redd AD, Ssemwanga D, Vandepitte J, Wendel SK, Ndembi N, Bukenya J, et al. Rates of HIV-1 super- infection and primary HIV-1 infection are similar in female sex workers in Uganda. AIDS. 2014; 28 (14):2147±52. https://doi.org/10.1097/QAD.0000000000000365 PMID: 25265078 8. Ssemwanga D, Lyagoba F, Ndembi N, Mayanja BN, Larke N, Wang S, et al. Multiple HIV-1 infections with evidence of recombination in heterosexual partnerships in a low risk Rural Clinical Cohort in Uganda. Virology. 2011; 411(1):113±31. https://doi.org/10.1016/j.virol.2010.12.025 PMID: 21239033 9. Rachinger A, van de Ven TD, Burger JA, Schuitemaker H, van 't Wout AB. Evaluation of pre-screening methods for the identification of HIV-1 superinfection. J Virol Methods. 2010; 165(2):311±7. https://doi. org/10.1016/j.jviromet.2010.02.016 PMID: 20178816 10. Manigart O, Courgnaud V, Sanou O, Valea D, Nagot N, Meda N, et al. HIV-1 superinfections in a cohort of commercial sex workers in Burkina Faso as assessed by an autologous heteroduplex mobility proce- dure. Aids. 2004; 18(12):1645±51. PMID: 15280775 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 13 / 15 HIV-1 superinfection in MSM 11. van der Kuyl AC, Cornelissen M. Identifying HIV-1 dual infections. Retrovirology. 2007; 4:67. https://doi. org/10.1186/1742-4690-4-67 PMID: 17892568 12. Herbinger KH, Gerhardt M, Piyasirisilp S, Mloka D, Arroyo MA, Hoffmann O, et al. Frequency of HIV type 1 dual infection and HIV diversity: analysis of low- and high-risk populations in Mbeya Region, Tan- zania. AIDS Res Hum Retroviruses. 2006; 22(7):599±606. https://doi.org/10.1089/aid.2006.22.599 PMID: 16831083 13. Fang G, Weiser B, Kuiken C, Philpott SM, Rowland-Jones S, Plummer F, et al. Recombination following superinfection by HIV-1. AIDS. 2004; 18(2):153±9. PMID: 15075531 14. Redd AD, Quinn TC, Tobian AAR. Frequency and implications of HIV superinfection. The Lancet Infec- tious Diseases. 2013; 13(7):622±8. https://doi.org/10.1016/S1473-3099(13)70066-5 PMID: 23726798 15. Andreani G, Espada C, Ceballos A, Ambrosioni J, Petroni A, Pugliese D, et al. Detection of HIV-1 dual infections in highly exposed treated patients. Virol J. 2011; 8:392. https://doi.org/10.1186/1743-422X-8- 392 PMID: 21824422 16. Chohan B, Lavreys L, Rainwater SM, Overbaugh J. Evidence for frequent reinfection with human immu- nodeficiency virus type 1 of a different subtype. J Virol. 2005; 79(16):10701±8. https://doi.org/10.1128/ JVI.79.16.10701-10708.2005 PMID: 16051862 17. Kiwelu IE, Novitsky V, Margolin L, Baca J, Manongi R, Sam N, et al. Frequent intra-subtype recombina- tion among HIV-1 circulating in Tanzania. PLoS One. 2013; 8(8):e71131. https://doi.org/10.1371/ journal.pone.0071131 PMID: 23940702 18. Wagner GA, Pacold ME, Kosakovsky Pond SL, Caballero G, Chaillon A, Rudolph AE, et al. Incidence and prevalence of intrasubtype HIV-1 dual infection in at-risk men in the United States. J Infect Dis. 2014; 209(7):1032±8. https://doi.org/10.1093/infdis/jit633 PMID: 24273040 19. Chaillon A, Wagner GA, Hepler NL, Little SJ, Kosakovsky Pond SL, Caballero G, et al. Dynamics of viral evolution and neutralizing antibody response after HIV-1 superinfection. J Virol. 2013; 87 (23):12737±44. https://doi.org/10.1128/JVI.02260-13 PMID: 24049166 20. Vanden Berghe W NC, Buve  A, Beelaert G, Fransen K, Laga M. A venue-based HIV prevalence and behavioural study among men who have sex with men in Antwerp and Ghent, Flanders, Belgium, Octo- ber 2009 to March 2010. Euro Surveill. 2011; 16(28). 21. Chalmet K, Staelens D, Blot S, Dinakis S, Pelgrom J, Plum J, et al. Epidemiological study of phyloge- netic transmission clusters in a local HIV-1 epidemic reveals distinct differences between subtype B and non-B infections. BMC Infect Dis. 2010; 10:262. https://doi.org/10.1186/1471-2334-10-262 PMID: 22. Struck D, Lawyer G, Ternes AM, Schmit JC, Bercoff DP. COMET: adaptive context-based modeling for ultrafast HIV-1 subtype identification. Nucleic Acids Res. 2014; 42(18):e144. https://doi.org/10.1093/ nar/gku739 PMID: 25120265 23. Vancoillie L, Mortier V, Demecheleer E, Schauvliege M, Vandekerckhove L, Vogelaers D, et al. Drug resistance is rarely the cause or consequence of long-term persistent low-level viraemia in HIV-1- infected patients on ART. Antivir Ther. 2015; 20(8):789±94. https://doi.org/10.3851/IMP2966 PMID: 24. Hall T. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series. 1999; 41:95±8. 25. Anisimova M, Gil M, Dufayard JF, Dessimoz C, Gascuel O. Survey of branch support methods demon- strates accuracy, power, and robustness of fast likelihood-based approximation schemes. Syst Biol. 2011; 60(5):685±99. https://doi.org/10.1093/sysbio/syr041 PMID: 21540409 26. Kumar S, Stecher G, Tamura K. MEGA7: Molecular Evolutionary Genetics Analysis Version 7.0 for Big- ger Datasets. Mol Biol Evol. 2016; 33(7):1870±4. https://doi.org/10.1093/molbev/msw054 PMID: 27. Lole KS, Bollinger RC, Paranjape RS, Gadkari D, Kulkarni SS, Novak NG, et al. Full-length human immunodeficiency virus type 1 genomes from subtype C-infected seroconverters in India, with evidence of intersubtype recombination. J Virol. 1999; 73(1):152±60. PMID: 9847317 28. Verhofstede C, Fransen K, Van Den Heuvel A, Van Laethem K, Ruelle J, Vancutsem E, et al. Decision tree for accurate infection timing in individuals newly diagnosed with HIV-1 infection. BMC Infect Dis. 2017; 17(1):738. https://doi.org/10.1186/s12879-017-2850-6 PMID: 29187159 29. Koelsch KK, Smith DM, Little SJ, Ignacio CC, Macaranas TR, Brown AJ, et al. Clade B HIV-1 superin- fection with wild-type virus after primary infection with drug-resistant clade B virus. AIDS. 2003; 17(7): F11±6. https://doi.org/10.1097/01.aids.0000060361.78202.4c PMID: 12700477 30. Cornelissen M, Pasternak AO, Grijsen ML, Zorgdrager F, Bakker M, Blom P, et al. HIV-1 dual infection is associated with faster CD4+ T-cell decline in a cohort of men with primary HIV infection. Clin Infect Dis. 2012; 54(4):539±47. https://doi.org/10.1093/cid/cir849 PMID: 22157174 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 14 / 15 HIV-1 superinfection in MSM 31. Wagner GA, Landais E, Caballero G, Phung P, Kosakovsky Pond SL, Poignard P, et al. Intrasubtype B HIV-1 Superinfection Correlates with Delayed Neutralizing Antibody Response. J Virol. 2017;91(17). 32. Redd AD, Mullis CE, Wendel SK, Sheward D, Martens C, Bruno D, et al. Limited HIV-1 superinfection in seroconverters from the CAPRISA 004 Microbicide Trial. J Clin Microbiol. 2014; 52(3):844±8. https:// doi.org/10.1128/JCM.03143-13 PMID: 24371237 33. Ronen K, Richardson BA, Graham SM, Jaoko W, Mandaliya K, McClelland RS, et al. HIV-1 superinfec- tion is associated with an accelerated viral load increase but has a limited impact on disease progres- sion. AIDS. 2014; 28(15):2281±6. https://doi.org/10.1097/QAD.0000000000000422 PMID: 25102090 34. Pacold ME, Pond SL, Wagner GA, Delport W, Bourque DL, Richman DD, et al. Clinical, virologic, and immunologic correlates of HIV-1 intraclade B dual infection among men who have sex with men. AIDS. 2012; 26(2):157±65. https://doi.org/10.1097/QAD.0b013e32834dcd26 PMID: 22045341 35. Pernas M, Casado C, Fuentes R, Perez-Elias MJ, Lopez-Galindez C. A dual superinfection and recom- bination within HIV-1 subtype B 12 years after primoinfection. Jaids-J Acq Imm Def. 2006; 42(1):12±8. 36. Kiwelu IE, Novitsky V, Margolin L, Baca J, Manongi R, Sam N, et al. HIV-1 subtypes and recombinants in Northern Tanzania: distribution of viral quasispecies. PLoS One. 2012; 7(10):e47605. https://doi.org/ 10.1371/journal.pone.0047605 PMID: 23118882 37. Luan H, Han XX, Yu XO, An MH, Zhang H, Zhao B, et al. Dual Infection Contributes to Rapid Disease Progression in Men Who Have Sex With Men in China. Jaids-J Acq Imm Def. 2017; 75(4):480±7. 38. Pacold M, Smith D, Little S, Cheng PM, Jordan P, Ignacio C, et al. Comparison of methods to detect HIV dual infection. AIDS Res Hum Retroviruses. 2010; 26(12):1291±8. https://doi.org/10.1089/aid. 2010.0042 PMID: 20954840 39. Redd AD, Mullis CE, Serwadda D, Kong X, Martens C, Ricklefs SM, et al. The rates of HIV superinfec- tion and primary HIV incidence in a general population in Rakai, Uganda. J Infect Dis. 2012; 206 (2):267±74. https://doi.org/10.1093/infdis/jis325 PMID: 22675216 40. Novitsky V, Moyo S, Wang R, Gaseitsiwe S, Essex M. Deciphering Multiplicity of HIV-1C Infection: Transmission of Closely Related Multiple Viral Lineages. PLoS One. 2016; 11(11):e0166746. https:// doi.org/10.1371/journal.pone.0166746 PMID: 27893822 41. Piantadosi A, Ngayo MO, Chohan B, Overbaugh J. Examination of a second region of the HIV type 1 genome reveals additional cases of superinfection. AIDS Res Hum Retroviruses. 2008; 24(9):1221. https://doi.org/10.1089/aid.2008.0100 PMID: 18729772 PLOS ONE | https://doi.org/10.1371/journal.pone.0195679 April 6, 2018 15 / 15

Journal

PLoS ONEPubmed Central

Published: Apr 6, 2018

There are no references for this article.