Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Viral Evolution and Cytotoxic T Cell Restricted Selection in Acute Infant HIV-1 Infection


Antiretroviral therapy-naive HIV-1 infected infants experience poor viral containment and rapid disease progression compared to adults. Viral factors (e.g. transmitted cytotoxic T- lymphocyte (CTL) escape mutations) or infant factors (e.g. reduced CTL functional capacity) may explain this observation. We assessed CTL functionality by analysing selection in CTL-targeted HIV-1 epitopes following perinatal infection. HIV-1 gag, pol and nef sequences were generated from a historical repository of longitudinal specimens from 19 vertically infected infants. Evolutionary rate and selection were estimated for each gene and in CTL-restricted and non-restricted epitopes. Evolutionary rate was higher in nef and gag vs. pol, and lower in infants with non-severe immunosuppression vs. severe immunosuppression across gag and nef. Selection pressure was stronger in infants with non-severe immunosuppression vs. severe immunosuppression across gag. The analysis also showed that infants with non-severe immunosuppression had stronger selection in CTL-restricted vs. non-restricted epitopes in gag and nef. Evidence of stronger CTL selection was absent in infants with severe immunosuppression. These data indicate that infant CTLs can exert selection pressure on gag and nef epitopes in early infection and that stronger selection across CTL epitopes is associated with favourable clinical outcomes. These results have implications for the development of paediatric HIV-1 vaccines.


Clear differences in the natural history of acute infection exist between adults and infants both clinically and with respect to viral population dynamics and the nature of the cytotoxic T cell (CTL) response. Where adults typically progress to AIDS over 8–10 years if untreated, infants experience extremely rapid progression, with 50% of African infants dying before they are 2 years old1,2. Despite the uncontrolled viral replication and high mortality rates associated with vertical HIV-1 transmission, a proportion of infants experience delayed clinical and immunological progression in the absence of antiretroviral therapy that can persist into childhood and even adolescence3. The mediators of infant immunity that contribute to slow progression in paediatric HIV-1 infection remain largely unknown.

Acute HIV-1 infection in adults is characterised by a steep decline from peak viral load during week 12–20 post-infection to a stabilised viral load set point that has been positively correlated with the rate of disease progression4,5. This decline and set point maintenance is largely mediated by HIV-1 specific CTL responses. Vertically infected infants experience a higher peak viraemia compared to adults6, a slower decline in viral load over the first two years of infection7 and lack a well-defined viral load set point. However, albeit on a lower magnitude compared to adults, vertically infected infants can mount HIV-1 specific CTL responses from birth8,9,10. This indicates that infant HIV-1 specific responses can be functional and select for viral escape variants11,12.

The rate of evolution in HIV-1 is one of the highest in nature, and a striking degree of viral genetic diversity is observed within HIV-1-infected individuals13. The error prone nature of the viral reverse transcriptase, a high rate of recombination, and a high replicative capacity all contribute to the genetic diversity. This plasticity enables viral adaptation to the host immune response14 and the functional efficacy of the CTL response can be observed by the selection of viral escape variants and differences in CTL-restricted epitopes15,16,17. Detailed longitudinal characterisation of the HIV-1 specific immune response has shown that CTLs exert a significant selective pressure on HIV-1 in early adult infection15,18. However, it is not known to what extent CTL-mediated selection pressure drives HIV-1 evolution in early life of infants, and how these functional differences may determine viral control and ultimate clinical outcome. Selection pressure such as that imposed by the host CTL response will influence the evolutionary rate of HIV-1 and the ratio of nonsynonymous and synonymous substitutions (dN/dS) provides a robust measure of both the mode and strength of selection. The intra-host evolutionary dynamics of HIV-1 and its relationship to clinical outcome has been studied in both adults19,20,21,22 and infants23,24,25,26, with the majority of studies focusing on the envelope gene encoding the primary targets of the humoral immune response27. The major targets of the CTL response, however, lie principally in the gag, pol and nef genes. Although the relationship between the evolutionary dynamics in these genes and clinical outcomes has been explored to some extent in adults28,29, little is known about this relationship during paediatric HIV-1 infection.

In a cohort of perinatally infected Kenyan infants, we estimated joint and individual evolutionary rates and selection across gag, pol and nef. We also investigated the influence of the infant CTL response on evolutionary parameters. Finally we assessed the association between clinical outcome and selection in CTL epitopes. Our results indicate a clear association between stronger immune selection on CTL-restricted epitopes and less severe immunosuppression during early infant HIV-1 infection.


Participant characteristics

HIV-1 gag, pol or nef sequences at three or more longitudinal time points were recovered from 19 of 23 eligible infants (Table 1). From 83 available specimens sequentially collected during the first 24 months of infant HIV-1 infection, sequence recovery was 98% for gag, 64% for pol and 87% for nef. Of the 19 infants, 13 (68%) were male, 12 (63%) became infected peripartum and seven (37%) became infected in utero. The median peak viral load was 7.2 log10 viral RNA copies/mL of plasma (interquartile range [IQR]: 6.6–7.6); the median absolute CD4 T cell count and CD4% at 6 months of age were 1415 cells/mL (IQR: 760–2195 cells/mL) and 22% (IQR: 16–26%), respectively (S1 Table). As expected, phylogenetic analysis showed clustering of sequences sampled longitudinally for the same infant (Fig. S1). The baseline characteristics of the 19 infants were representative of infants in the parent cohort (S1 Table). Subtype analysis showed that 11 infants were infected with HIV-1 sub-subtype A1 (58%) and one infant with HIV-1 subtype D (5%). This result was concordant in all the analysed genetic regions. Seven infants (37%) showed discordant subtype classification between HIV-1 genes that was unique from previously described circulating recombinant forms (CRFs), indicating a large proportion of unique recombinant forms (URFs, S2 Fig.).

Table 1 Infant characteristics.

HIV-1 evolutionary rates were highest in nef followed by gag and pol

The average evolutionary rates for gag, pol, and nef were 5.9 (95% confidence intervals [CI]: 5.0–7.0), 1.5 (CI: 1.0–2.0) and 12.6 (CI: 10.0–15.0) nucleotide substitutions/1000 sites/year, respectively. The average evolutionary rates were significantly higher in gag and nef compared to pol (P = 0.002 and P < 0.0001, respectively), and in nef compared to gag (P = 0.03, Fig. 1A). The patient-specific evolutionary rates across gag and nef did not correlate (ρ = 0.2, P = 0.5; Fig. 1B) and across pol it was consistently low between individuals, indicating considerable evolutionary constraint. Overall, dN/dS ratios indicated moderate purifying selection across gag and nef (0.7 [CI: 0.5–0.8] and 0.8 [0.6–0.9], respectively), and strong purifying selection across pol (0.3 [CI: 0.2–0.4]). Individual estimations indicated that four (IDs: 170, 211, 291 and 334), zero and one (ID: 211) infants showed overall positive selection across gag, pol and nef, respectively.

Figure 1: Evolutionary rates across HIV-1 gag, pol and nef genes from infants sampled under 2 years of age.

(A) Summary of nucleotide substitution rates across gag, pol and nef with median (line) 1st and 3rd quartiles (box) and the 10th and 90th percentile (whiskers) shown. The Mann Whitney U test was used to assess differences between groups. (B) Rank-based Spearman correlations between the nucleotide substitution rate across gag and nef.

Evolutionary rate was lower in infants with non-severe immunosuppression than in infants with severe immunosuppression across gag and nef

In order to assess associations between clinical outcomes and evolutionary parameters, infants were stratified into groups based on severe immunosuppression (nadir CD4% ≤15), peak viral load, mode of transmission, and HIV-1 subtype (Table 2). The evolutionary rate was significantly lower in infants with non-severe immunosuppression compared to those with severe immunosuppression across gag and nef (P = 0.008 and 0.008, respectively; Table 2). In addition, selection was significantly stronger in infants with non-severe compared to those with severe immunosuppression across gag (1.4 [CI: 1.1–1.8] vs. 0.4 [CI: 0.3–0.5]; P = 0.006). Three of four infants with overall positive selection across gag (IDs: 170, 211 and 334), and the single infant with global positive selection across nef (ID: 211) were all infants with non-severe immunosuppression. No significant difference was observed in median peak viral load between non-severely immunosuppressed and severely immunosupressed infants (6.6 vs 7.3 log10 viral RNA copies/mL of plasma, P = 0.2 [Mann-Whitney U test]).

Table 2 Global evolutionary rates and dN/dS rate ratios in infants stratified by clinical parameters.

The evolutionary rate in nef was higher in infants with high peak viral loads (above median peak viral load) compared to those with low peak viral loads (below median, P = 0.007). The evolutionary rate was also higher in nef in infants infected with HIV-1 non-A subtypes compared to those infected with HIV-1 subtype A (P = 0.018). Interestingly, five of seven non-subtype A infected infants were infected with URFs, and all these five infants had nef sequences that clustered with subtype A-like sequences. No significant differences were observed between evolutionary rate estimates and selection across pol between infant groups (Table 2), or between evolutionary parameters in any genetic region in infants stratified by mode of transmission.

CTL targeted regions were under stronger selection than non-CTL targeted regions in gag and nef

To investigate the influence of infant CTL responses in early viral evolution, we used infant HLA types to partition viral sequences into epitopes putatively restricted by the infant CTL response (CTL+) and non-restricted parts (CTL−). Differences in estimates of evolutionary rate and selection strength was assessed on both the group level and the individual level. Overall, no significant differences in evolutionary rate between CTL+ and CTL− regions was seen across gag and pol. However, on the individual level, three infants (IDs: 261, 313 and 334) had significantly higher evolutionary rate in CTL+ vs. CTL− regions. (P < 0.05 for all comparisons). Overall, the nef gene showed contrasting results with significantly higher evolutionary rate in CTL+ regions (18.4 nucleotide substitutions/1000 sites/year [CI: 14.0–24.0]) compared to CTL− regions (9.1 [CI: 7.0–12.0], P < 0.001). Individual comparisons indicated faster rates in CTL+ regions in five infants (IDs: 170, 211, 231, 281, 411 and 454) and slower rates in one infant (ID: 424) compared to CTL− regions (P ≤ 0.05 for all comparisons).

Estimates of selection in partitioned genetic regions indicated that dN/dS ratios in gag were significantly higher in CTL+ (1.0 [CI: 0.7–1.4]) vs. CTL− regions (0.5 [CI: 0.4–0.7], P = 0.008). Individual estimates indicated that the same three individuals with faster evolutionary rates in CTL+ vs. CTL− regions, also had significantly stronger selection in CTL+ vs. CTL− regions (P < 0.05 for all comparisons). Similar results were seen in nef with significantly higher dN/dS ratios in CTL+ (1.2 [CI: 0.8–1.6]) vs. CTL− regions (0.5 [CI: 0.4–0.7]; P = 0.001) on the group level. Individual estimates indicated that six infants had significantly stronger selection in CTL+ vs. CTL− regions (P < 0.05 for all comparisons), four of which (IDs: 170, 211, 231, and 411) were also the infants with higher evolutionary rates in CTL+ vs. CTL− regions. No significant differences in selection pressure between CTL+/− regions were observed in pol at either the group level or the individual level.

CTL targeted regions across gag and nef were under stronger selection in infants with non-severe immunosuppression in contrast to infants with severe immunosuppression

To further explore the relationship between CTL-mediated selection in gag and nef and the clinical outcome of infection, we estimated differences in dN/dS ratios in CTL+/− regions in infant groups stratified by the severity of immunosuppression. The pol region was not considered due to the consistently low variation and evolutionary signal. The dN/dS ratios were higher in CTL+ vs. CTL− regions in infants with non-severe immunosuppression in both gag (2.3 [CI: 1.6–3.3] vs. 1.0 [CI: 0.7–1.5], P = 0.006) and nef (2.6 [CI: 1.7–3.8] vs. 0.5 [CI: 0.2–0.8], P < 0.001; Table 3). No significant differences were found between dN/dS ratios across CTL+ and CTL− regions in infants with severe immunosuppression in either gag or nef.

Table 3 dN/dS rate ratios across gag and nef in HLA and non-HLA targeted epitopes in infants stratified by immunosuppression.


The positive correlation between HIV-1 evolutionary rate and disease progression rate has been demonstrated several times in adults20,22,30. Here, we show that this is true in infant HIV-1 infection as well. We also demonstrate a clear association between stronger immune selection on CTL-restricted epitopes and less severe immunosuppression during early infant HIV-1 infection. However, in contrast to adults, evidence of CTL-mediated selection is only seen in a minority of vertically-infected infants who maintain relatively high CD4 percentages. Our results support the notion that infant CTL responses may have the functional capacity to exert selective pressure on cognate epitopes, and that possession of functional CTL responses is associated with favourable clinical outcomes. The strengths of our study approach include the use of historical samples obtained before the era of infant ART, a clinically well characterised cohort, a relatively large sample size which enabled stratification of individuals according to clinical parameters, and the analysis of longitudinal changes in viral sequence in early infant HIV-1 infection.

The few studies that have addressed the association between CTL-mediated evolution and differential clinical outcomes in infant HIV-1 infection have reported divergent findings. Recently, Adland et al.31 indicated that HLA played a limited role in protection from HIV-1 disease progression in children compared to adults, with modest associations between the presence of a protective HLA allele and disease severity in children or with in vitro viral replicative capacity of patient derived chimeric viruses. In addition, no evidence of de novo CTL escape was seen in viral sequences from severely immunosuppressed infants (CD4% <20 within the first year of life) bearing protective HLA B alleles. Our data indicate that CTL-mediated selection indeed can occur as a consequence of the infant CTL response and that this is associated with non-severe immunosuppression. This suggests a possible protective role for HLA in mediating HIV-1 disease outcomes in early infection following vertical transmission, in some but not all infants, potentially explaining the inconsistent findings in previous reports. Our results are in accordance with a study of five infants with no evidence of severe immune suppression (based on the CDC paediatric HIV-1 classification), which reported global positive selection in nef and CD8 T cell driven mutations in HLA A*24-restricted nef epitopes over the first year of life11. Similarly, amino acid changes in CTL-associated nef epitopes was reported for children with non-progression but not for children with disease progression32. As discussed by Adland et al.33, further evidence of functional immune driven selection in neonatal HIV-1 infection comes from observations that replicative capacity of infant Gag-Protease chimeric viruses was significantly reduced compared with matched maternal chimeric viruses34. Our data suggest that within the first year of life, a proportion of infants may select viral variants that harbour CTL-driven mutations. Further studies are needed to determine associations between CTL-driven mutations, fitness costs and viral load dynamics in early infant infection.

Despite the apparent association between CTL-driven selection in HIV-1 genes and clinical outcome in this study, uncontrolled viremia was observed throughout the first two years of life in the majority of infants. Some of the unique features of paediatric immunity may account for this discrepancy, particularly with regards to the increased CD4 T cell counts that are seen in infancy both in the periphery and in mucosa when compared to adulthood. Although naïve (CCR5) CD4 T cells predominate in the periphery, evidence from the rhesus macaque model indicate that the pool of activated and proliferating CD4+ targets is greatly increased in neonatal tissues compared to adults (10-fold in the gut), and that these cells are preferentially targeted and depleted by simian immunodeficiency virus35,36. The abundance37 and profound loss of CCR5+ CD4 coupled with the high rate of thymic turnover during infancy, may supply HIV-1 with a virtually unlimited supply of susceptible target cells. As a consequence, functional infant CTL responses may not reflect the dynamics of peripheral blood viraemia in early life because these are masked by rapid viral turnover in the context of overabundance of susceptible target cells. Notwithstanding, we observed a large effect size of the median difference between peak viral loads in non-severely immunosuppressed vs severely immunosuppressed groups that did not reach significance, likely due to a power effect. This suggests a relationship between infant pathogenesis, peak viral load and evolutionary dynamics that merit further studies.

Due to the use of viral population sequences, our estimates of evolutionary rate across gag, pol and nef probably provide a lower bound on this parameter17. Our estimates ranged from 10−2 nucleotide substitutions/site/year in nef, to 10−3 nucleotide substitutions/site/year in gag and pol which is in accordance with rates estimated during adult infection27,38, although direct comparisons with previous reports may be difficult to interpret due to methodological differences. Our analysis was adapted from Hightower et al. who reported an average rate of 0.72 nucleotide substitutions/1000 sites/year across pol during adult chronic HIV-1 subtype B infection17. Our estimate of 1.5 nucleotide substitutions/1000 sites/year in pol is approximately twice as high and could reflect differences in evolutionary rate between early and chronic infection19 or differences between adult and infant infection (e.g. higher viral loads in infants that may influence evolutionary rates). However, the low evolutionary rate across pol relative to gag and nef suggests considerable evolutionary constraints in this region, as would be expected from genes encoding enzymes with critical functions in the viral replication cycle.

In accordance with previous estimates27, nef was shown to evolve at the highest rate, a reflection of the molecular plasticity of its encoded protein that is involved in diverse roles during viral replication and pathogenesis. In addition, the influence of CTL selection was most apparent in nef as a) six infants showed significantly stronger selection in CTL+ vs. CTL− regions compared to three across gag, and b) the overall association between CTL selection and favourable clinical outcome resulted in higher LRT scores in nef compared to gag. This observation probably reflects the early targeting of nef by CTL responses as seen in both adult HIV-1 infection and in SIV infection. Comparably, CTL targeting of gag epitopes has been shown to occur at a later stage of infection. Alternatively, these observed differences may reflect strong protective effects of CTL selection in nef during acute HIV-1 infection.

We also found a lower evolutionary rate in nef among HIV-1 subtype A compared with non-subtype A infected infants. This result together with the higher viral evolutionary rate seen in infants with severe immunosuppression are in line with previous studies on adults that show a positive correlation between rate of evolution and disease progression20,22,30 and a slower disease progression in HIV-1 subtype A-infected adults compared with other HIV-1 subtypes and CRFs39,40. These results warrants further studies on HIV-1 strain-specific effects on disease progression among both infants and adults.

In summary, we find that infant CTLs exert significantly stronger selection across gag and nef epitopes in early infection. In addition, we find that stronger selection across CTL epitopes is associated with favourable clinical outcomes. These results highlight the potential functionality of infant CTLs despite a lack of viral control with implications for the development of protective paediatric HIV-1 vaccines.


Cohort characteristics, sampling and ethics

Samples were collected, as part of the CTLs and the Prevention of HIV-1 Transmission Study, in Nairobi, Kenya between 1999 and 200241. As described in detail elsewhere41, the parent cohort consisted of 465 HIV-1-infected women recruited during pregnancy and followed for up to two years postpartum with their infants. Mothers were provided with short-course zidovudine (ZDV) during the final trimester of pregnancy to reduce the risk of HIV-1 transmission according to contemporaneous guidelines42.In the absence of a national ART programme for children, all infants were treatment-naïve during the study period. Maternal and infant CD4 T cell counts and RNA viral loads were serially assessed at birth, and months 1, 3, 6, 12, 15, 18, 21, and 24. Infant viral loads were measured from cryopreserved plasma specimens, using the Gen-Probe assay43 (limit of detection of 50 copies/ml). Infant HIV-1 infection was defined as the first detection of HIV-1-gag DNA by PCR from dried blood spots44 or HIV-1 RNA from plasma43. Infection and classification into in utero, peripartum/early breast milk, or late breast milk transmission were defined as previously described41. Ninety transmission events were detected, with timing of infection available for 87 infants. Infant HLA types to two digit resolution were determined as previously described45. For evolutionary analyses, infants were selected if they were infected by one month of age and if they had three of more longitudinal plasma specimens available for study. Seventy two infants were infected by one month of age and 23 infants had samples from three or more longitudinal time points totalling 89 plasma specimens.

Ethics statement

This study was conducted according to the principles expressed in the Declaration of Helsinki and the study protocol was approved by the Kenyatta National Hospital Ethics Review Committee and the Institutional Review Board of the University of Washington. HIV-1 seropositive pregnant women were enrolled after provision of written informed consent as detailed previously41.

HIV-1 amplification and sequencing

As detailed in the Supplementary methods, viral RNA was isolated from cryopreserved plasma samples and purified using the QIAamp Viral RNA extraction kit (Qiagen, Limburg, Netherlands) followed by reverse transcription and nested PCR (Table S2) for gag, pol and nef. Amplified DNA was purified and sequenced on a 48 capillary ABI-3730 DNA analyser using Applied Biosystems BigDye Terminator v3.1 (Waltham, MA, USA) chemistry. Contigs were assembled and open reading frames edited using Geneious v6.1.646 (Biomatters Ltd.). Full length gag and nef sequences corresponding to HXB2 nucleotide positions 790–2289 and 8797–9414, respectively, were obtained (except for gag sequences from infant 313, HXB2 positions 880–2289) as well as near full length pol sequences (2.0–2.8 kb spanning the protease through to integrase coding sequences, HXB2 positions 2358–5096). Ambiguous bases were identified using the Find Heterozygotes tool packaged in Geneious v6.1.6 and defined as peaks with quality scores >10 with a secondary peak height above 25% of the maximal peak. GenBank accession numbers for the sequences are indicated in Table S4.

HIV-1 subtyping

The earliest infant sample following infection was used for subtyping and gag, pol and nef, sequences were analysed separately. Pairwise alignments of infant samples were made using the version and settings of MUSCLE47 packaged in Geneious v6.1.6. Profile alignments of infant sequences and the Los Alamos National Library HIV sequence database subtype reference alignment (2010)48 were made in Clustal X249. All alignments were manually inspected and edited in MEGA v5.250. Maximum-likelihood phylogenetic reconstruction (GTR + γ + I) with branch support estimated by the Approximate Likelihood Ratio Test (aLRT)-Shimodaira-Hasegawa-like (SH) procedure51 in PhyML 3.0 was carried out. Branches with aLRT-SH values above 0.85 were considered as statistically well supported. Sequences that either clustered with a known circulating recombinant form (CRF), clustered within different subtypes in distinct genetic regions, showed intermediate clustering across distinct genetic regions or presented long branches within a cluster were analysed as recombinant subtypes. Characterisation, using Simplot and Bootscan analysis in Simplot v3.5.152, was carried out as detailed in the Supplementary methods.

Phylogenetic estimation of intra-patient rates of evolution in gag, pol and nef

We estimated evolutionary rates separately for each individual infant using a maximum likelihood (ML) approach implemented in the HyPhy package53 with suitable modifications needed to analyse gag, pol and nef, respectively17,54. The Tamura-Nei (TN93) model of nucleotide evolution and a strict molecular clock were used to estimate evolutionary rates based on the procedure described by Hightower et al.17, having determined that rate estimates in all genes were unaffected by the complexity of the evolutionary model (Fig. S4). Briefly, sequences from each individual were aligned, and equipped with a “linear” tree, where the sequence at the oldest time point as the root, the second oldest sequence as the direct descendant of the root, and so on. Sequences from all individuals were analysed both individually (one individual at a time), and jointly (all individuals together) by profile likelihood in the ML phylogenetic framework. The likelihood ratio test (LRT) was used to test for deviations from molecular clock in a given individual, assuming the χ2N−2 asymptotic distribution of the test statistic (N = number of time points). Estimates of codon substitution rates, and the dN/dS ratio were obtained using the Muse-Gaut 94 x REV (MG94) codon-based model of evolution55. Infant HLA types at two-digit resolution and the 2013 LANLDB annotation of CTL-restricted epitopes56 were used to partition gag, pol and nef sequences into regions restricted by infant CTL and those that were not. The software pipeline implementing the analysis can be downloaded from

Data analysis

Comparisons of global ML estimates of the evolutionary rates between HIV-1 genes were carried out using the Mann-Whitney U test. Two-sided p-values were calculated for analysis and values of p < 0.05 were considered significant. Correlations were tested using the Spearman’s rank correlation coefficient. Statistical analysis and plots were made in GraphPad Prism version 6.0 (GraphPad Software inc.). Severe immunosuppression was defined as a nadir CD4% ≤15 at any time point up to 15 months of age due to the association with infant mortality within 2 years of life in this cohort57. The Holm-Ŝidák test was used to correct for multiple comparisons.

Additional Information

How to cite this article: Garcia-Knight, M. A. et al. Viral Evolution and Cytotoxic T Cell Restricted Selection in Acute Infant HIV-1 Infection. Sci. Rep. 6, 29536; doi: 10.1038/srep29536 (2016).


  1. 1

    Mbori-Ngacha, D. et al. Morbidity and mortality in breastfed and formula-fed infants of HIV-1-infected women: A randomized clinical trial. Jama 286, 2413–2420 (2001).

    CAS  Article  Google Scholar 

  2. 2

    Spira, R. et al. Natural history of human immunodeficiency virus type 1 infection in children: a five-year prospective study in Rwanda. Mother-to-Child HIV-1 Transmission Study Group. Pediatrics 104, e56 (1999).

    CAS  Article  Google Scholar 

  3. 3

    Warszawski, J. et al. Long-term nonprogression of HIV infection in children: evaluation of the ANRS prospective French Pediatric Cohort. Clin. Infect. Dis. 45, 785–794 (2007).

    Article  Google Scholar 

  4. 4

    McMichael, A. J., Borrow, P., Tomaras, G. D., Goonetilleke, N. & Haynes, B. F. The immune response during acute HIV-1 infection: clues for vaccine development. Nat. Rev. Immunol. 10, 11–23 (2010).

    CAS  Article  Google Scholar 

  5. 5

    Mellors, J. W. et al. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 272, 1167–1170 (1996).

    CAS  ADS  Article  Google Scholar 

  6. 6

    Richardson, B. A. et al. Comparison of human immunodeficiency virus type 1 viral loads in Kenyan women, men, and infants during primary and early infection. J. Virol. 77, 7120–7123 (2003).

    CAS  Article  Google Scholar 

  7. 7

    Melvin, A. J. et al. HIV-1 dynamics in children. J. Acquir. Immune Defic. Syndr. Hum. Retrovirol. 20, 468–473 (1999).

    CAS  Article  Google Scholar 

  8. 8

    Thobakgale, C. F. et al. Human immunodeficiency virus-specific CD8+ T-cell activity is detectable from birth in the majority of in utero-infected infants. J. Virol. 81, 12775–12784 (2007).

    CAS  Article  Google Scholar 

  9. 9

    Lohman, B. L. et al. Longitudinal assessment of human immunodeficiency virus type 1 (HIV-1)-specific gamma interferon responses during the first year of life in HIV-1-infected infants. J. Virol. 79, 8121–8130 (2005).

    CAS  Article  Google Scholar 

  10. 10

    Luzuriaga, K. et al. HIV-1-specific cytotoxic T lymphocyte responses in the first year of life. J. Immunol. 154, 433–443 (1995).

    CAS  PubMed  PubMed Central  Google Scholar 

  11. 11

    Sanchez-Merino, V., Nie, S. & Luzuriaga, K. HIV-1-specific CD8+ T cell responses and viral evolution in women and infants. J. Immunol. 175, 6976–6986 (2005).

    CAS  Article  Google Scholar 

  12. 12

    Pillay, T. et al. Unique acquisition of cytotoxic T-lymphocyte escape mutants in infant human immunodeficiency virus type 1 infection. J. Virol. 79, 12100–12105 (2005).

    CAS  Article  Google Scholar 

  13. 13

    Jenkins, G. M., Rambaut, A., Pybus, O. G. & Holmes, E. C. Rates of molecular evolution in RNA viruses: a quantitative phylogenetic analysis. J. Mol. Evol. 54, 156–165 (2002).

    CAS  ADS  Article  Google Scholar 

  14. 14

    Goulder, P. J. R. & Walker, B. D. HIV and HLA class I: an evolving relationship. Immunity 37, 426–440 (2012).

    CAS  Article  Google Scholar 

  15. 15

    Goonetilleke, N. et al. The first T cell response to transmitted/founder virus contributes to the control of acute viremia in HIV-1 infection. J. Exp. Med. 206, 1253–1272 (2009).

    CAS  Article  Google Scholar 

  16. 16

    Liu, M. K. P. et al. Vertical T cell immunodominance and epitope entropy determine HIV-1 escape. J. Clin. Invest. 123, 380–393 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  17. 17

    Hightower, G. K. et al. HIV-1 clade B pol evolution following primary infection. PLoS One 8, e68188 (2013).

    CAS  ADS  Article  Google Scholar 

  18. 18

    Borrow, P. et al. Antiviral pressure exerted by HIV-1-specific cytotoxic T lymphocytes (CTLs) during primary infection demonstrated by rapid selection of CTL escape virus. Nat. Med. 3, 205–211 (1997).

    CAS  Article  Google Scholar 

  19. 19

    Shankarappa, R. et al. Consistent viral evolutionary changes associated with the progression of human immunodeficiency virus type 1 infection. J. Virol. 73, 10489–10502 (1999).

    CAS  PubMed  PubMed Central  Google Scholar 

  20. 20

    Lemey, P. et al. Synonymous substitution rates predict HIV disease progression as a result of underlying replication dynamics. PLoS Comput. Biol. 3, e29 (2007).

    ADS  MathSciNet  Article  Google Scholar 

  21. 21

    Esbjörnsson, J. et al. Inhibition of HIV-1 disease progression by contemporaneous HIV-2 infection. N. Engl. J. Med. 367, 224–232 (2012).

    Article  Google Scholar 

  22. 22

    Mild, M. et al. High intrapatient HIV-1 evolutionary rate is associated with CCR5-to-CXCR4 coreceptor switch. Infect. Genet. Evol. 19, 369–377 (2013).

    CAS  Article  Google Scholar 

  23. 23

    Zhang, H. et al. Restricted genetic diversity of HIV-1 subtype C envelope glycoprotein from perinatally infected Zambian infants. PLoS One 5, e9294 (2010).

    ADS  Article  Google Scholar 

  24. 24

    Carvajal-Rodríguez, A. et al. Disease progression and evolution of the HIV-1 env gene in 24 infected infants. Infect. Genet. Evol. 8, 110–120 (2008).

    Article  Google Scholar 

  25. 25

    Ganeshan, S., Dickover, R. E., Korber, B. T., Bryson, Y. J. & Wolinsky, S. M. Human immunodeficiency virus type 1 genetic evolution in children with different rates of development of disease. J. Virol. 71, 663–677 (1997).

    CAS  PubMed  PubMed Central  Google Scholar 

  26. 26

    Raghwani, J., Bhatt, S. & Pybus, O. G. Faster Adaptation in Smaller Populations: Counterintuitive Evolution of HIV during Childhood Infection. PLOS Comput. Biol. 12, e1004694 (2016).

    ADS  Article  Google Scholar 

  27. 27

    Alizon, S. & Fraser, C. Within-host and between-host evolutionary rates across the HIV-1 genome. Retrovirology 10, 49 (2013).

    CAS  Article  Google Scholar 

  28. 28

    Piantadosi, A. et al. HIV-1 evolution in gag and env is highly correlated but exhibits different relationships with viral load and the immune response. AIDS 23, 579–587 (2009).

    CAS  Article  Google Scholar 

  29. 29

    Noviello, C. M. et al. Maintenance of Nef-mediated modulation of major histocompatibility complex class I and CD4 after sexual transmission of human immunodeficiency virus type 1. J. Virol. 81, 4776–4786 (2007).

    CAS  Article  Google Scholar 

  30. 30

    Markham, R. B. et al. Patterns of HIV-1 evolution in individuals with differing rates of CD4 T cell decline. Proc. Natl. Acad. Sci. U. S. A. 95, 12568–12573 (1998).

    CAS  ADS  Article  Google Scholar 

  31. 31

    Adland, E. et al. Discordant Impact of HLA on Viral Replicative Capacity and Disease Progression in Pediatric and Adult HIV Infection. PLoS Pathog. 11, e1004954 (2015).

    Article  Google Scholar 

  32. 32

    Casartelli, N. et al. Structural defects and variations in the HIV-1 nef gene from rapid, slow and non-progressor children. AIDS 17, 1291–1301 (2003).

    CAS  Article  Google Scholar 

  33. 33

    Goulder, P. J. R. & Watkins, D. I. Impact of MHC class I diversity on immune control of immunodeficiency virus replication. Nat. Rev. Immunol. 8, 619–630 (2008).

    CAS  Article  Google Scholar 

  34. 34

    Carlson, J. M. et al. HIV transmission. Selection bias at the heterosexual HIV-1 transmission bottleneck. Science 345, 1254031 (2014).

    Article  Google Scholar 

  35. 35

    Wang, X. et al. Simian immunodeficiency virus selectively infects proliferating CD4+ T cells in neonatal rhesus macaques. Blood 116, 4168–4174 (2010).

    CAS  Article  Google Scholar 

  36. 36

    Wang, X. et al. Profound loss of intestinal Tregs in acutely SIV-infected neonatal macaques. J. Leukoc. Biol. 97, 1–10 (2014).

    CAS  Google Scholar 

  37. 37

    Bunders, M. J. et al. Memory CD4 + CCR5 + T cells are abundantly present in the gut of newborn infants to facilitate mother-to child transmission of HIV-1 Memory CD4 + CCR5 + T cells are abundantly present in the gut of newborn infants to facilitate mother-to child transmissio. 120, 4383–4391 (2012).

  38. 38

    Novitsky, V., Wang, R., Rossenkhan, R., Moyo, S. & Essex, M. Intra-host evolutionary rates in HIV-1C env and gag during primary infection. Infect. Genet. Evol. 1, (2013).

  39. 39

    Palm, A. A. et al. Faster progression to AIDS and AIDS-related death among seroincident individuals infected with recombinant HIV-1 A3/CRF02_AG compared with sub-subtype A3. J. Infect. Dis. 209, 721–728 (2014).

    CAS  Article  Google Scholar 

  40. 40

    Kiwanuka, N. et al. Effect of human immunodeficiency virus Type 1 (HIV-1) subtype on disease progression in persons from Rakai, Uganda, with incident HIV-1 infection. J. Infect. Dis. 197, 707–713 (2008).

    Article  Google Scholar 

  41. 41

    John-Stewart, G. C. et al. HV-1-specific cytotoxic T lymphocytes and breast milk HIV-1 transmission. J. Infect. Dis. 199, 889–898 (2009).

    CAS  Article  Google Scholar 

  42. 42

    Shaffer, N. et al. Short-course zidovudine for perinatal HIV-1 transmission in Bangkok, Thailand: a randomised controlled trial. Bangkok Collaborative Perinatal HIV Transmission Study Group. Lancet 353, 773–780 (1999).

    CAS  Article  Google Scholar 

  43. 43

    Emery, S. et al. Evaluation of performance of the Gen-Probe human immunodeficiency virus type 1 viral load assay using primary subtype A, C, and D isolates from Kenya. J. Clin. Microbiol. 38, 2688–2695 (2000).

    CAS  PubMed  PubMed Central  Google Scholar 

  44. 44

    Panteleeff, D. D. et al. Rapid method for screening dried blood samples on filter paper for human immunodeficiency virus type 1 DNA. J. Clin. Microbiol. 37, 350–353 (1999).

    CAS  PubMed  PubMed Central  Google Scholar 

  45. 45

    Bunce, M. et al. Phototyping: comprehensive DNA typing for HLA-A, B, C, DRB1, DRB3, DRB4, DRB5 & DQB1 by PCR with 144 primer mixes utilizing sequence-specific primers (PCR-SSP). Tissue Antigens 46, 355–367 (1995).

    CAS  Article  Google Scholar 

  46. 46

    Kearse, M. et al. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28, 1647–1649 (2012).

    Article  Google Scholar 

  47. 47

    Edgar, R. C. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32, 1792–1797 (2004).

    CAS  Article  Google Scholar 

  48. 48

    Los Alamos National Laboratory, HIV sequence database., (2010) (Date of access: 10/06/2010).

  49. 49

    Larkin, M. A. et al. Clustal W and Clustal X version 2.0. Bioinformatics 23, 2947–2948 (2007).

    CAS  Article  Google Scholar 

  50. 50

    Kumar, S., Stecher, G., Peterson, D. & Tamura, K. MEGA-CC: computing core of molecular evolutionary genetics analysis program for automated and iterative data analysis. Bioinformatics 28, 2685–2686 (2012).

    CAS  Article  Google Scholar 

  51. 51

    Anisimova, M. & Gascuel, O. Approximate likelihood-ratio test for branches: A fast, accurate, and powerful alternative. Syst. Biol. 55, 539–552 (2006).

    Article  Google Scholar 

  52. 52

    Lole, K. S. et al. Full-length human immunodeficiency virus type 1 genomes from subtype C-infected seroconverters in India, with evidence of intersubtype recombination. J. Virol. 73, 152–160 (1999).

    CAS  PubMed  PubMed Central  Google Scholar 

  53. 53

    Pond, S. L. K., Frost, S. D. W. & Muse, S. V. HyPhy: hypothesis testing using phylogenies. Bioinformatics 21, 676–679 (2005).

    CAS  Article  Google Scholar 

  54. 54

    Pond, S. L. K. Pol Evolution. (Date of access: 23/01/2016) (2013).

  55. 55

    Muse, S. V. & Gaut, B. S. A likelihood approach for comparing synonymous and nonsynonymous nucleotide substitution rates, with application to the chloroplast genome. Mol. Biol. Evol. 11, 715–724 (1994).

    CAS  PubMed  PubMed Central  Google Scholar 

  56. 56

    Los Alamos National Laboratory, HIV molecular immunology database SVG Epitope Maps. (Date of access: 25/11/2013) (2013).

  57. 57

    Obimbo, E. M. et al. Predictors of Early Mortality in a Cohort of Human Immunodeficiency Virus Type 1-Infected African Children. Pediatr. Infect. Dis. J. 23, 536–543 (2004).

    Article  Google Scholar 

Download references


We would like to thank the women and infants who participated in this study. We would also like to thank Esther Wambui Kiragu for assistance with sequencing. This research was funded by the Rosetrees Trust (M403 to SRJ), the British HIV Association (BH14 to SRJ), the Wellcome Trust (089567/Z/09/Z to MGK.), the Swedish Research Council (350-2012-6628 to JE), the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) (R01-HD023412, K24-HD054314 to GJ-S) and also in part by the University of Washington CFAR (P30 AI027757) and UW Global Center for Integrated Health of Women, Adolescents and Children (Global WACh). This manuscript was submitted for publication with permission from the Director of the Kenyan Medical Research Institute (KEMRI).

Author information




M.A.G.-K., J.S., T.I.d.S., S.L.R.-J. and J.E. conceived the study and designed the experiments; M.A.G.-K., B.K. and B.C. performed the experiments; G.J.-S. and D.M.-N. designed the parent cohort and oversaw cohort recruitment and clinical procedures; J.S. and B.L.P. led specimen collection, processing and field-based cellular assays; M.A.G.-K., J.E. and S.L.K.P. analysed the data; all authors helped write and/or review the manuscript.

Corresponding author

Correspondence to Miguel A. Garcia-Knight.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Supplementary information

Rights and permissions

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Garcia-Knight, M., Slyker, J., Payne, B. et al. Viral Evolution and Cytotoxic T Cell Restricted Selection in Acute Infant HIV-1 Infection. Sci Rep 6, 29536 (2016).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing