Predicts Viral Load Outcome Positive Individuals − Epitope in HLA-B*27 HIV Immune Escape at an Immunodominant

HIV Immune Escape at an Immunodominant Epitope in HLA-B*27−Positive Individuals Predicts Viral Load Outcome

This information is current as Palanee Ammaranond, David J. van Bockel, Kathy of September 28, 2021. Petoumenos, Marylin McMurchie, Robert Finlayson, Melanie G. Middleton, Miles P. Davenport, Vanessa Venturi, Kazuo Suzuki, Linda Gelgor, John M. Kaldor, David A. Cooper and Anthony D. Kelleher

J Immunol 2011; 186:479-488; Prepublished online 29 Downloaded from November 2010; doi: 10.4049/jimmunol.0903227 http://www.jimmunol.org/content/186/1/479 http://www.jimmunol.org/ Supplementary http://www.jimmunol.org/content/suppl/2010/11/29/jimmunol.090322 Material 7.DC1 References This article cites 66 articles, 36 of which you can access for free at: http://www.jimmunol.org/content/186/1/479.full#ref-list-1

Why The JI? Submit online. by guest on September 28, 2021

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

HIV Immune Escape at an Immunodominant Epitope in HLA-B*27–Positive Individuals Predicts Viral Load Outcome

Palanee Ammaranond,*,†,‡,1 David J. van Bockel,*,1 Kathy Petoumenos,x Marylin McMurchie,{ Robert Finlayson,‖ Melanie G. Middleton,x Miles P. Davenport,# Vanessa Venturi,# Kazuo Suzuki,* Linda Gelgor,x John M. Kaldor,x David A. Cooper,*,x and Anthony D. Kelleher*,x

The CTL response in HLA-B*27+ HIV-infected individuals is characterized by an immunodominant response to a conserved epitope in gag p24 (aa 263–272, KRWIILGLNK; KK10). Mutations resulting in substitution of the arginine (R264) at position 2 of this epitope have been identiﬁed as escape mutations. Nineteen HLA-B*27+ long-term nonprogressors were identiﬁed from an

Australian cohort with an average follow-up of 16 y following infection. Viral and host genetic factors impacting on disease Downloaded from progression were determined at multiple time points. Twelve of 19 had wild-type sequences at codon 264 at all time points; 7 of 19 carried CTL escape variants. Median viral load and CD4+ T cell counts were not signiﬁcantly different between these groups at enrollment. Viral load, as judged by levels at their last visit (1,700 and 21,000 RNA copies/ml, respectively; p = 0.01) or by time- weighted area under the curve was higher in the escape group (p = 0.02). Escape mutants at other HLA-B*27–restricted epitopes were uncommon. Moreover, host polymorphisms, such as CCR5Δ32, CCR2-64I, and SDF1-39A, or breadth of TCR repertoire

responding to KK10 did not segregate to wild-type or escape groups. Host and viral factors were examined for a relationship to http://www.jimmunol.org/ viral load. The only factor to affect viral load was the presence of the R264 escape mutations at the immunodominant epitope. CTL escape at R264 in the KK10 epitope is a major determinant of subsequent viral load in these HLA-B*27+ individuals. The Journal of Immunology, 2011, 186: 479–488.

he phenomenon of HIV mutational escape from CTL escape mutations occurring during acute or early infection, which pressure is well documented. Mutations within epitopes ultimately impact upon disease progression (14, 18–22). T and in flanking regions resulting in reduced binding to the In models of macaque SIV infection, there is evidence for a restricting MHC class I allele, altered T cell recognition, and al- relationship between generation of escape mutants and disease tered processing have been described within most proteins of HIV-1 progression or vaccine failure (23–27). However, even here the by guest on September 28, 2021 (1–12). Furthermore, there is evidence for the imprinting of mu- numbers are small and are confounded by the fact that some of the tations within viral genomes as evolution occurs within individ- most rapid progressors were those that never mounted a detectable uals and within populations (13–17). There are accumulating data CTL response (26). Therefore the relationship between disease suggesting that CD8+ T cell responses are closely associated with progression and generation of escape mutations is still unclear. One of the best-described CTL responses to HIV-1 is the immunodominant HLA-B*27–restricted response to a conserved *Immunovirology Laboratory, St Vincent’s Centre for Applied Medical Research; xNational Centre in HIV Epidemiology and Clinical Research, University of New epitope in Gag p24 (aa 263–272, KRWIILGLNK [KK10]) (28). South Wales; {East Sydney Doctors, Darlinghurst; ‖Taylor Square Private Clinic, Robust CTL responses directed toward this epitope are detectable # Sydney, New South Wales 2010; Complex Systems in Biology Group, Centre for from early in primary infection and are maintained throughout Vascular Research, University of New South Wales, Kensington, New South Wales 2052, Australia; and †Department of Transfusion Medicine and ‡Chulalongkorn Uni- disease. T cell responses are usually detectable, despite the frequent versity Centenary Academic Development Project, Faculty of Allied Health Scien- generation of a common L268M variant at position 6, until well- ces, Chulalongkorn University, Bangkok 10330, Thailand characterized escape mutations occur at codon 264 (5, 6, 29, 30). 1 P.A. and D.J.v.B. contributed equally to this work. These mutations result in conversion of the arginine (R) residue at Received for publication October 5, 2009. Accepted for publication October 13, position 2 of the epitope, which is critical for binding the B pocket 2010. of HLA-B*27 to lysine, glycine, threonine, or glutamine (5, 31, 32). This work was supported by the Australian Government Department of Health and Ageing, the National Health and Medical Research Council, and the Australian Re- Each of these substitutions results in an epitope with decreased search Council. binding affinity to the HLA-B*27 molecule, resulting in poorly The views expressed in this publication do not necessarily represent the position of presented epitopes that are weak stimulators of the CTL response. the Australian Government. The National Centre in HIV Epidemiology and Clinical These mutations usually take years to arise. This delay, despite Research is affiliated with the Faculty of Medicine, University of New South Wales. concerted immune pressure, seems to be due to the critical nature of Address correspondence and reprint requests to Dr. David J. van Bockel, St Vincent’s the arginine at 264 for the formation and stability of the multimeric Centre for Applied Medical Research, 405 Liverpool Street, Darlinghurst, New South Wales, 2010, Australia. E-mail address: d.vanbockel@cfi.unsw.edu.au p24 capsid (33–35). The presence of a mutation at codon 264 The online version of this article contains supplemental material. in isolation has a substantial impact on viral fitness. The virus is Abbreviations used in this paper: ART, antiretroviral therapy; AUC, area under the intolerant of this nonsynonymous mutation and it is only rarely curve; DS, Simpson diversity index; KK10, KRWIILGLNK; MFI, median fluorescence detected as an isolated variant in clinical samples. Viral fitness is index; N/A, not able to amplify; ND, not done; NS, not able to be sequenced using the restored by the presence of concomitant compensatory mutations, traditional sequencing; PID, patient identifier; p-MHC, peptide-MHC; WT, wild-type. predominantly involving S173A and L268M, which restore the Copyright Ó 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 capacity for capsid unfolding (6, 35, 36). Within individuals, www.jimmunol.org/cgi/doi/10.4049/jimmunol.0903227 480

Table I. Characteristics of HLA-B*27+ individuals at enrollment and end point visits

At Study Entry At End Point

Sequence at Gag aa Sequence at Gag aa Age Infected with Plasma HIV CD4+ T Cell 263–272: WT Sequence Follow- Plasma HIV CD4+ T Cell 263-272: WT Sequence PID (y) HIV (y) RNA (copies/ml) Count (cells/ml) KRWIILGLNKa up (y) RNA (copies/ml) Count (cells/ml) KRWIILGLNKa LT1 41 10 620 651 -----M---- (16) 15 19,900 464 -----M---- (15) R------(4) -----I---- (5) LT2 38 9 21,000 510 -----M---- (20) 10 21,000 396 -----M---- (14) -G------(4) -K--VI---- (2) LT3 40 12 52,000 651 -----M---- (20) 19 .75,000 247 -G------(18) -G------N (2) LT4 47 9 4,800 532 -K--VI---- (4) 12 9,600 504 -K--VI---- (6) LT5 31 9 73,000 644 ------(20) 9 69,000 630 -----M---- (15) ------(5) LT6 60 9 16,000 726 -Q---M---- (16) 13 1,580,000 250 -Q---M---- (15) LT7 42 10 1,100 754 -----M---- (20) 17 800 682 -----M---- (20) LT8 39 11 1,800 500 ------(20) 19 7,790 700 ------(8) ------H- (8) ------T- (3) OUTCOME LOAD VIRAL PREDICTS ESCAPE IMMUNE HIV -----M---- (1) LT9 40 10 200 882 -----M---- (5) 17 869 430 -----M---- (5) LT10 36 8 200 561 8 200 561 -----M---- (4) LT11 36 10 1,100 1170 10 1,100 1170 ------(1) ------H- (5) LT12 29 10 14,600 840 -----M---- (4) 16 50 1131 -----M---- (5) LT13 47 12 4,000 360 -G------(5) 19 130,200 340 -G------(7) LT14 38 16 1,190 696 16 1,190 696 -K---M---- (7) LT15 61 11 7,000 630 -----M---- (4) 11 4,400 540 -----M---- (4) LT16 39 11 200 921 11 200 921 ------(4) LT17 39 10 1,900 1050 -----M---- (20) 17 2,300 638 -----M---- (18) ---T-M---- (2) LT18 40 8 9,200 608 ------(6) 14 8,920 262 ------(8) LT19 44 12 8,800 814 12 8,800 814 -K---M---- (5) LT20 33 11 146 1190 NS LT21 39 9 50 910 NS Subjects generating escape variants at Gag codon 264 are shown in bold type. For LT10, LT11, LT14, L16, and LT19, only a single time point was available for evaluation. aThe number in parentheses indicates the number of clones with the reported sequence.

PID, patient identiﬁer; NS, not able to be sequenced using standard sequencing.

Downloaded from from Downloaded http://www.jimmunol.org/ by guest on September 28, 2021 28, September on guest by The Journal of Immunology 481 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 1. Compiled individual time course for CD4 counts (A), plasma viral load (B), and plasma virus sequence variation (C) in KK10 HLA-B*27+ patients included in this study. The x-axis represents time from enrollment in the study (years of follow-up). Patient data for CD4+ T cell count and viral load (legend for A and B) are represented by viral sequence analysis; WT (solid line) and immune escape (dashed line). C, Viral immune escape is shown in parallel represented by common nonsynonymous mutations at codons 264 and 268. 482 HIV IMMUNE ESCAPE PREDICTS VIRAL LOAD OUTCOME mutations seem to occur at the time of disease progression or Enumeration of CD8+ T cell responses by ELISPOT and MHC vaccine failure (37). class I tetrameric complexes Because carriage of HLA-B*27 has been associated with long- IFN-g ELISPOT and peptide-MHC (p-MHC) class I tetrameric complexes term nonprogression (38–40), it has been speculated that this ad- were used to enumerate responses to CD8+ T cell-specific epitopes on thawed vantage is specifically related to the negative impact of the escape PBMCs, as previously described (31, 45). Calculation of Ag-specific (tetra- mutation at codon 264 on viral fitness and the requirement for mer+) CD8+ T cell population frequency and median fluorescence index fitness-restoring compensatory mutations. However, possible con- (MFI) was performed using BD FACSDiva software (v 4.0; BD Biosciences, San Jose, CA). founders that could impact on disease progression, such as the carriage of chemokine receptor polymorphisms, have not been Clonotypic analysis of TCR repertoire responding to KK10 systematically addressed. epitope We chose to study the effect of the generation of the KK10 escape PBMCs were stained with HLA-B*2705 tetrameric complexes carrying mutation on surrogate markers of disease progression, while taking either of the two major variants (L or M268) of the KK10 epitope se- into account known viral and host determinants that are associated quenced from plasma. HLA-A*0201 tetrameric complexes carrying the NV9 epitope (NLVPMVATV) from the pp65 matrix protein of CMV or with slower progression. This study was performed in the context HLA-B*2705 complexes carrying the RL9 epitope (RRIYDLIEL) from the of a cohort of long-term nonprogressors (LTNPs) prospectively EBN3AC nuclear Ag were used as comparator immunodominant pop- recruited and followed since 1994 (41). During follow-up, more than ulations. Cells were fixed with formaldehyde and sorted into tetramer+ half of this cohort has progressed in terms of reduced CD4+ T cell populations, and RNAwas extracted as previously described (45). Unbiased amplification of TCR b-chain first-strand cDNA template was achieved counts or being started on antiretroviral therapy (ART). We screened using a nonnested anchored template-switch RT-PCR with a singular 39 the cohort for individuals who carried HLA-B*27 and then exam- TCR b-chain C region primer (59-TTCTGATGGCTCAAACACAGCGAC- Downloaded from ined these individuals for the presence of CTL escape mutants at 39), as described previously (45, 46). Purified product was ligated into a R264 in KK10. We subsequently determined whether the presence TA-cloning vector (Invitrogen, Carlsbad, CA) and transformed into chem- of escape mutations at this single immunodominant epitope con- ically competent Escherichia coli. A minimum of 23 clones were selected and vector product amplified using generic M13 primers and then se- tributed to long-term outcome. The presence of escape mutations at quenced using a BigDye v3.1 sequencing kit and an ABI 3730xl capillary R264 in the immunodominant HLA-B*27–restricted KK10 epitope sequencer (Applied Biosystems). Sequence data were aligned using

impacts upon viral load outcomes in these individuals. Sequencher (Gene Codes), and clonotype identity was confirmed using the http://www.jimmunol.org/ Immunogenetics online sequence analysis algorithm (http://imgt.cines.fr). Materials and Methods Statistical analysis Study subjects The Fisher exact test was used to compare categorical variables, and the Mann–Whitney U test was used to analyze continuous variables derived The Australian Long-Term Non-Progressor cohort was established in 1994, from the HLA-B*27+ individuals falling into wild-type (WT) and escape recruiting and following 94 patients. After gaining written informed con- groups. Plasma viral load and CD4+ T cell covariates were measured as sent, patients with asymptomatic HIV disease who had been infected with medians at censored time point (last available time point or start of treat- HIV for $8 y and had a CD4+ T cell count $500 cells/ml in the absence of + + ment), as well as mean time-weighted area under the curve (AUC) for viral any ART were enrolled in this study. Parameters, such as CD4 and CD8 + load (copies/ml) and for CD4 T cell count. Time-weighted AUC was T lymphocyte counts, viral load, treatment history, and clinical indicators calculated as AUC (using the trapezoidal rule) divided by the patient’s by guest on September 28, 2021 of disease progression as defined by the revised World Health Organization total duration of follow-up (47). These tests were performed using SAS clinical staging of HIV infection (42), were recorded and entered into a software (SAS Institute, Cary, NC). database. PBMCs and plasma samples were cryopreserved at up to yearly Comparison of clonotype distribution was standardized for multiple intervals. For this study, data were censored at the last available visit or at samples using the Simpson diversity index (DS), which ranges in value from the visit at which ART was commenced. 0 (minimal diversity) to 1 (maximal diversity), accounting for the variety HLA genotyping and detection of CCR5, CCR2, and SDF1 of TCR clonotypes and their clonal dominance (48). It is used to quanti- tatively assess differences in diversity between TCR repertoires. A rando- polymorphisms mization procedure was used to estimate the value of these indices using Genomic DNA was extracted from 5 3 106 PBMCs (Qiagen, Valencia, 23 clonotypic TCR sequences per time point to account for differences in CA). HLA class I and II DNA typing was performed by standard PCR the number of clonal sequences obtained per TCR repertoire. The calcu- sequence-specific priming through the Australian Red Cross Blood Ser- lation of diversity measures and the randomization technique were per- vice, Sydney. The identification of the CCR5D32, CCR2-64I, and SDF1- formed using Matlab (The Mathworks, Natick, MA). 39A polymorphisms was performed with minor variations of the previously described methodologies (43, 44). Results Demographic characteristics of HLA-B*27+ individuals Viral extraction, PCR, and sequencing reactions Twenty-two of the 94 patients enrolled in the LTNP cohort were Viral RNAwas extracted and reverse transcribed from plasma, as previously HLA-B*2705+. On chart review, one subject was found to be on described (31), from 21 HLA-B*27+ patients at multiple visits during up to 10 y of follow-up. Full-length gag p24 gene sequences were amplified by combined ART at cohort entry and was excluded from this study. nested PCR and cloned, as previously described (31). A minimum of four clones were sequenced at each time point. Sequences were manually edited and assembled with Sequencher (Gene Codes, Ann Harbor, MI). The same methodology was used to amplify sequences from p17 and nef using the following primer sets: p17, outer primers: P1A (59-AGTG- GCGCCCGAACAGG-39) and P1B (59-GAAGTGACATAGCAGGAACT- 39), inner primers: P2A (59-TCTCGACGCAGGACTCG-39) and P2B (59- GAGGAAGCTGCAGAATGG-39); nef, outer primers: nefof (59-AGAGT- TAGGCAGGGATATTCACC-39) and nefor (59-CAGCTGCTTATATGCA- GGATC-39), inner primers: neff (59-ACCTCAGGTACCTTTAAGACC- AAT-39) and nefr (59-GTCCCCAGCGGAAAGTC-39). A 720-bp fragment of gp41 was amplified using primers gp412 (59-AAGCCCTGTCTTATT- FIGURE 2. Features of KK10-specific CD8+ T cells response at study CTTCTAGGTA-39), gp413 (59-ATACCTAAAATACCTAAAGGATCAA- CAGCTC-39) and gp411 (59-TGCTCTGGAAAACTCATTTG-39). PCR entry as measured by frequency (%) and avidity (MFI) of cells staining products were analyzed on a 1.5% agarose gel before bulk sequencing. with HLA-B*27 tetramer carrying cognate peptide in patients maintaining Sequencing reactions were carried out using a BigDye terminator cycle WT virus (LT1, LT5, LT8, LT17) and those developing an escape mutant sequencing kit (Applied Biosystems, Foster City, CA). (LT2, LT3) during the period of observation. The Journal of Immunology 483

Table II. Distribution of virological and immunological parameters among WT and R264 escape mutation groups in HLA-B*27+ individuals

Parameter WT (n = 12) Escape (n =7) p Value Median total follow-up (y) 16 18 0.18 Median follow-up at censored time point (y) 14.5 13 0.50 Surrogate markers Median viral load to censored time point (copies/ml) 1,700 21,000 0.01 Median CD4+ T cell count to censored time point (cells/ml) 634 396 0.09 Viral load time-weighted AUC (copies/ml) 1,784 21,000 0.02 CD4+ T cell count time-weighted AUC (cells/ml) 745 518 0.02 Host factors CCR5 WT/WT 10 3 0.13 Δ32/WT 2 4 0.13 Δ32/Δ32 0 0 – CCR2 WT/WT 8 7 0.25 64I/WT 4 0 0.25 64I/64I 0 0 – SDF1 WT/WT 6 3 0.99 Downloaded from 39A/WT 5 4 0.65 39A/39A 1 0 0.99 Viral factors Escape at gp41 2 1 .0.99 Escape at nef 2 0 0.51 Escape at p17 0 0 – nef-deleted virus 0 0 – http://www.jimmunol.org/ Parameters reaching statistical signiﬁcance are indicated in bold type. –, The parameter was not found in any individual.

All 21 patients included were men, with a median age of 45 y Enumeration of immunological response to KK10 epitope (range: 29–64 y). Reported risk factors for HIV infection were men variants who have sex with men (47.6%) and were unspecified in the re- Staining of PBMC populations with HLA-B*27 tetrameric com- mainder. At study entry, median viral load was 4,000 copies/ml + plexes bearing the appropriate cognate epitope KRWIILGLNK or (range: 50–73,000 copies/ml), and median CD4 T cell count KRWIIMGLNK revealed substantial proportions of CD8+ T cells by guest on September 28, 2021 was 651 cells/ml (range: 360–1190; Table I). All were infected with specific for this immunodominant epitope at time points when WT + , subtype B virus. Subject LT13 had a CD4 T cell count 500 cells/ or the L268M variant sequence was present. For patient groups ml at study entry; however, because he had not received any ART + maintaining p24 Gag WT or viral immune escape sequence, there and had been diagnosed as HIV 12 y before inclusion in the study, was no clear difference in the frequency of KK10-specific CD8+ he was included in this analysis. Two subjects were excluded from T cells (median: 4.1 versus 3.6%; range, 1.6–8.8%; Fig. 2). the analysis because they consistently had low-level plasma viral load (,200 copies/ml), and the gag p24 region could not be amplified from plasma or PBMC samples, despite repeated attempts with a range of primer pairs. Table III. Distribution of polymorphisms in chemokine coreceptors CCR5 and CCR2 and chemokine SDF1 Characterization of immunodominant HLA-B*27–restricted epitope KK10 PID CCR5 CCR2 SDF-1 Δ The gag p24 region was successfully sequenced from 19 individ- LT1 WT/ 32 WT/WT WT/WT LT2 WT/Δ32 WT/WT WT/3A uals. At a minimum, each of these patients had clonal sequences LT3 WT/WT WT/WT WT/WT derived from plasma collected at enrollment or the censoring time LT4 WT/WT WT/WT WT/3A point (Fig. 1, Table I). Twelve of 19 carried R264 in all sequences at LT5 WT/WT WT/WT WT/WT Δ each sampled time point (median: 14.5 y; range, 8–19 y). Among LT6 WT/ 32 WT/WT WT/WT LT7 WT/WT WT/64I WT/WT these, four carried the WT sequence (KRWIILGLNK) and eight LT8 WT/WT WT/WT WT/WT carried the common variant L268M sequence (KRWIIMGLNK). LT9 WT/WT WT/64I WT/3A Seven carried variant strains with mutations at aa 264, consistent LT10 WT/WT WT/WT WT/3A with previously described CTL escape mutants: KKWIIMGLNK LT11 WT/WT WT/64I 3A/3A (n = 2), KKWIVIGLNK (n = 1) (5, 6, 29, 32), KGWIILGLNK LT12 WT/WT WT/WT WT/3A LT13 WT/WT WT/WT WT/WT (n = 3) (6, 49), or KQWIIMGLNK (n = 1) (31) (Table I). In five of LT14 WT/Δ32 WT/WT WT/3A seven (LT4, LT6, LT13, LT14, and LT19), the virus carried one of LT15 WT/WT WT/WT WT/3A these mutations at the initial and last visits. One of seven (LT3) LT16 WT/WT WT/WT WT/3A Δ had WT virus at aa 264 at entry but developed an escape mutation LT17 WT/ 32 WT/WT WT/WT LT18 WT/WT WT/64I WT/WT (R264G) 2 y later, which then increased to fixation. One of seven LT19 WT/Δ32 WT/WT WT/3A (LT2) had a WT virus at the baseline visit; 9 mo later, a mixture of Δ32 indicates the presence of a 32-bp deletion in the CCR5 gene; 3A indicates viral sequences coding for R264, R264K, and R264G was ob- G/A mutation at position 801 in the untranslated region of SDF-1b; 64I indicates served (Fig. 1). the presence of mutation at position 64. 484 HIV IMMUNE ESCAPE PREDICTS VIRAL LOAD OUTCOME

Table IV. Diversity of TCR repertoire at study entry for HIV (KK10) codon 264. The time courses of plasma viral RNA levels, CD4+ + and control herpesvirus CMV and EBV epitope-specific CD8 T cell T cell counts, and proportions of molecular clones carrying WT or populations variant sequence in individual patients are summarized in Fig. 1. At study entry there was no difference in CD4+ T cell count HIV (KK10) CMV (NV9)/EBV (RL9)a (median cells/ml: WT = 703; escape = 651; p=0.19) or viral load Clonotypes Clonotypes (median copies/ml: WT = 1850; escape = 8800; p = 0.06) between PID Detected (n) DS Detected (n) DS these two groups. These groups were compared at the last follow- WT up time point available or on the day that ART was commenced. LT1a 3 0.25 2 0.51 The median follow-up to this time point from diagnosis was 14.5 LT5 3 0.45 2 0.40 and 13.0 y for the WT and escape groups, respectively. Median LT8 6 0.65 5 0.76 + LT17 2 0.00 – – CD4 T cell counts at the follow-up time point were 634 and Escape 396 cells/ml in WT and escape groups, respectively (p=0.09). LT2 3 0.43 3 0.50 However, viral load in the WT group was significantly lower than LT3 4 0.32 2 0.40 that of the escape group (median copies/ml: 1,700 and 21,000, + Populations were assessed for clonotype variety using diversity calculations (DS) respectively; p = 0.01; Table II). Time-weighted AUC for CD4 and absolute clonotype number. Data for calculated diversity and clonotype fre- T cell count and viral load variables were compared, to allow for quency were obtained from raw, unbiased sequence analysis of the TRCB CDR3 region. FACS-isolated cellular populations were specific for p-MHC class I tetramer correction of different lengths of follow-up between subjects. that presented the cognate HIV (KK10) epitope sequenced from plasma in the ab- Viral load was significantly lower in those who maintained R264 sence of R264 mutation or for control CMV (NV9) or EBV (RL9) herpesvirus in KK10 compared with those who did not (copies/ml: WT = Downloaded from epitope (where indicated). aAn alternative tetramer to NV9 was used. 1,784; escape = 21,000; p=0.02). Using this analysis, there was –,Absence of an Ag-specific CD8+ T cell population detected using MHC class I also a statistical difference in this measure of change in CD4+ tetramers. T cell numbers between these two groups (cells/ml: WT = 745; escape = 518; p=0.02; Table II). Similarly, substantial IFN-g responses were measured by ELI- Finally, we considered the commencement of ART as another

SPOT from PBMCs stimulated with WT or L268M variant of the clinical marker of disease progression. Four of 7 in the escape http://www.jimmunol.org/ epitope if measured at pre-escape time points. Responses were not group but only 1 of 12 in the WT group had commenced therapy seen to peptides representing escape variants KKWIIMGLNK, during the period of observation (p=0.04, Fisher exact test). This KGWIILGLNK, or KQWIIMGLNK at pre- or postescape time series of univariate analyses suggests that development of an es- points. Following generation of these escape variants at KK10, cape mutation at KK10 results in higher levels of viral load and responses to the subdominant p17 epitope IRLRPGGKK were de- perhaps lower levels of CD4+ T cells. Further to this observation, tected in patients LT4 and LT14 subsequent to development of es- we subsequently investigated whether there were other differences cape mutations at codon 264. Consistent with the published data, between these two groups, apart from the generation of the escape these responses were not detected in LT3 and LT9, who had not mutation at codon 264, which might explain the higher viral loads developed an escape mutation at codon 264 at the study entry time observed. Particular attention was paid to host and viral factors by guest on September 28, 2021 point examined (Supplemental Fig. 1) (32). that are associated with altered disease progression. The effects of CTL escape at KK10 on surrogate markers of Characterization of host chemokine receptor polymorphisms disease progression and TCR repertoire We next investigated the effect of carrying an escape mutation at The presence of CCR5D32, CCR2-64I, and SDF1-39A poly- codon 264 of Gag on surrogate markers of disease progression by morphisms was assessed in all 19 LTNPs (Table III). Thirteen dividing the group into those that carried WTor escape mutations at (68.4%) were homozygous for WT CCR5. Six (31.6%) were

Table V. Summary of sequences found at subdominant HLA-B*27–restricted epitopes in p17, gp41, and nef (derived from bulk sequence data)

p17 (aa 19–27) gp41 (aa 786–795) Nef (aa 105–114) ID IRLRPGGKK GRRGWEALKY RRQDILDLWI LT1 N/A ------V------LT2 ------E-----V LT3 ------T------E-----V LT4 ------Q------N/A LT5 ------LT6 ------I--- N/A LT7 ------P------E------LT8 ------LT9 ------V--- -G------V LT10 ------V------V LT11 N/A -H----V--- N/A LT12 ------V LT13 ------E------LT14 ------V LT15 ------L------V LT16 ------N/A ------V LT17 ------I------V LT18 ------H----V--- -E------V LT19 ------V N/A, not able to amplify. The Journal of Immunology 485

Table VI. The effect of potential confounders on viral load in HLA-B*27+ individuals

Median Viral Load (copies/ml)

Host or Viral Factor WT Mutation p Value Host factors CCR5Δ32 4,400 14,350 0.22 CCR2-64I 8,800 985 0.19 SDF1-39A 8,295 1,100 NAa Viral factors Escape mutation in other B27-restricted epitope gp41 7,790 8,920 0.95 nef 6,095 4,894 0.87 p17b ––– nef-deleted virusb ––– aNo comparison was possible because 18 patients were homozygous WT or heterozygous; 1 was homozygous for the SDF1- 39A mutation. bNo comparison was possible because no sequence variants were found. NA, not applicable.

heterozygous for the D32 mutation, and none was homozygous for the KK10 and control herpesvirus epitopes was determined using Downloaded from this polymorphism. Fifteen were homozygous for WT CCR2 DS, which standardizes the analysis of T cell repertoire data be- (78.9%), four were heterozygous (21.0%), and none was homo- tween patients. No substantial differences were found between zygous for the CCR2-64I polymorphism. Nine (47.4%) were ho- WT or escape patients responding to the KK10 or control epitopes mozygous for WT SDF1, nine (47.4%) were heterozygous, and one with regard to the number of clonotypes (range: WT = 2–6; es- (5.3%) was homozygous for the SDF1-39A mutation. None of the cape = 2–4) or repertoire diversity (DS range: WT = 0.00–0.76; Δ subjects who were homozygous or heterozygous for the CCR5 32 escape = 0.32–0.50; Table IV). http://www.jimmunol.org/ and CCR2-64I polymorphisms carried both. Three subjects were MFI of staining with p-MHC class I tetramers was analyzed as heterozygous for CCR5Δ32 and SDF1-39A, whereas only one in- a surrogate marker of Ag-speciﬁc CD8+ T cell avidity for Ag. The dividual was heterozygous for CCR2-64I and SDF1-39A. The calculated MFI of patients maintaining WT virus at codon 264 of carriage of these host genetic factors was not different between KK10 was lower than those who developed escape (median MFI the groups carrying WT or escape viral sequence at codon 264 of KK10-speciﬁc CD8+ T cells: WT = 5,396; escape = 19,138; (Table II). Fig. 2). However, this analysis was limited because only two of the TCR repertoire at study entry for cells responding to cognate escape group had WT sequence at entry to the study. variant epitope (KK10) in four patients who did not develop an escape mutation in KK10 (LT1, LT5, LT8, and LT17) was com- Viral variation at subdominant HLA-B*27–restricted epitopes by guest on September 28, 2021 pared with two patients who eventually developed an escape The HLA-B*2705–restricted subdominant epitope in Gag p17, aa mutation in KK10 (LT2 and LT3). Clonal expansion or diversity to 19–27, IRLRPGGKK, which may become dominant after escape at the p24 (aa 263–272) epitope (5, 32), was bulk sequenced at the last available time point. All patients carried the WT sequence at Table VII. Summary table of HLA class I and II DNA typing this epitope (Table V). determined from genomic DNA In 18 of 19 patients, the HLA-B*27–restricted epitope in gp41 (aa 786–795), GRRGWEALKY, was successfully sequenced at HLA Class I HLA Class II the last assessed time point (Table V). Fifteen carried the WT PID A B DRB1 sequence; 3 of 18 carried the possible escape mutations R787Q WTa (n = 1) or R787H (n = 2). Ten carried mutations at position 792 LT1 A1, 3 B8, 27 DR1, 15 (A792V/T/I/P/L). This is not an anchor residue for binding to LT5 A2, 3 B27, 51 DR1, 4 HLA-B*27. LT7 A2, 25 B27, 62 DR4, 16 As previously documented, none of these individuals carried LT8 A2, 32 B27, 62 DR13, 15 LT9 A3, 29 B27, 35 ND a nef-deleted virus (50). The region of nef containing the HLA- LT10 A2, – B27, – ND B*27–restricted epitope RRQDILDLWI (aa 105–114) was suc- LT11 A2, – B27, – DR1, 7 cessfully sequenced from plasma virus of 16 patients. Two of 16 LT12 A3, 32 B27, 44 DR11, 15 carried mutations at the anchor residue R106G (n = 1) and R106E LT15 A1, 32 B27, 37 DR4, 8 (n = 1), which represent possible escape mutations. Four of 16 had LT16 A2, 11 B27, 51 DR4, 10 LT17 A11, 24 B27, 35 DR3, 4 mutations at position 108 (D108E), and 11 had amino acid sub- LT18 A3, 26 B18, 27 DR1, 4 stitutions of V to I at position 114 (Table V). Escape There was no difference between the WT and escape groups with LT2 A2, 3 B27, – DR11, – regard to the carriage of mutations of the anchor residue, which LT3 A2, 3 B27, – DR1, – LT4 A1, 31 B8, 27 DR3, 4 represent candidate escape mutations, within any of these three LT6 A1, 30 B18, 27 DR1, 103 subdominant HLA-B*2705–restricted epitopes (Table II). Varia- LT13 A1, 2 B8, 27 DR1, 11 tion was not noted in either group within the Gag p17 (aa 19–27) LT14 A2, 26 B27, 44 DR1, – epitope. Two of those in the WT group and one in the escape LT19 A23, 33 B27, 44 DR3, 9 group had a mutation at position 2 in the gp41 epitope (aa 786– HLA alleles are in World Health Organization-assigned format, as required for 795; p . 0.99). Two subjects in the WT group and none in the HLA nomenclature (described in Ref. 51). aWT sequence at codon 264. escape group had mutated sequences at position 2 of the nef –, Allelic homozygosity; ND, not done. epitope (aa 105–114; p = 0.51; Table II). 486 HIV IMMUNE ESCAPE PREDICTS VIRAL LOAD OUTCOME

Distribution of possible confounding host and viral factors WT virus. This was true whether the measure of viral load was associated with viral load between WT and escape groups taken at the last available time point or was measured by time- Because polymorphisms within the genes coding for the SDF1 weighted AUC analysis, which compensates for the different chemokine and CC chemokine receptors have been associated with periods of follow-up among individuals. The magnitude of the delayed disease progression, we looked for confounding associa- difference in viral load of 1.1 log10 copies/ml is relatively large tions between carriage of these mutations and viral load. No dif- compared with that attributed to escape mutations occurring in ference was observed in the median viral load when patients were chronic infection; a recent analysis suggested that the presence of , divided by carriage of CCR2-64I or CCR5Δ32 polymorphisms a single escape mutation is associated with a 10% increase in (p = 0.22 and p = 0.19, respectively; Table VI). It was not possible viral load (63). The size of this effect is likely to reflect the im- to statistically interpret differences in viral load for the SDF1-39A portance of this immunodominant CTL response to viral control in polymorphism, because only 1 of the 19 patients studied was these individuals prior to escape and the extent of viral fitness homozygous for this mutation. Further, there was no impact on postescape when compensatory mutations are present (35, 64). viral load when patients were divided by carriage of viral immune This finding suggests that the presence of escape mutations at an immunodominant epitope is associated with a higher viral load. In escape variants in subdominant HLA-B*27–restricted epitopes in + gp41 (p=0.95) or Nef (p = 0.87; Table VI). Moreover, when contrast, CD4 T lymphocyte numbers were not consistently sta- determining confounding immunological parameters, there was no tistically different between those carrying WT and escape poly- segregation of TCR diversity in those who had diversity of TCR morphisms at codon 264 of KK10. This may be due to treatment repertoire determined with viral load (Table IV). guidelines at the time of commencing ART in the majority of these HLA class I and II genotyping was assessed to determine patients (late 1990s, early 2000s). Those guidelines, based on Downloaded from whether other HLA alleles might influence disease progression the contemporaneous Department of Health and Human Services following immune escape (Table VII) (51). HLA-A and -B alleles guidelines, encouraged early administration of therapy, prior to significant immune depletion and at relatively high CD4+ T cell were determined for all (n = 19) patients, and HLA-DRB1 alleles + were determined for 17 patients. HLA types were examined for counts (65). This would prevent the CD4 T cell count depletion that usually follows increased viral loads. Therefore, the failure to featured haplotypes, homozygosity, and alleles previously asso- + ciated with control (52, 53), in the absence of immunodominant consistently demonstrate a deleterious effect of escape on CD4 http://www.jimmunol.org/ T cell count may reflect treatment of high or increasing viral load influence (54) or with escape (38, 55, 56). Equal distribution of + HLA-B*27 homozygosity observed in WT and escape groups (n = prior to any substantial decrease in CD4 T cell count. This argu- 2 in both), combined with a lack of a haplotype or another HLA- ment is strengthened by the observation that those with escape allele associated with the escape group, suggests that other HLA- mutations were more likely to begin ART than were those who A, -B, and -DR alleles had no considerable impact on viral load, maintained WT sequence at codon 264 of gag. although the sample size precluded any formal analyses. Together, The distribution of other possible determinants of plasma viral these analyses are consistent with there being no significant im- load was not different between those developing escape mutations pact of these cofactors on viral load variation in our cohort. Taken and those maintaining WT sequences at codon 264. This indicated by guest on September 28, 2021 together, the univariate analyses performed suggest that the pre- that generation of escape mutations was not influenced by these sence of an escape mutation at codon 264 in p24 was the only factors. Further, if the effect of these other possible confounding variable measure that was associated with higher viral load out- factors on viral load were considered separately, there was no come (p = 0.01). significant influence on viral load outcome within this group. The data presented in this article are consistent with previously published data, in that the CD8+ T cell responses to KK10 prior to escape were Discussion robust (5, 57, 66, 67). Further, as expected, the generation of KK10 Although escape mutations have been repeatedly demonstrated in escape mutants with low affinity for HLA-B*27 was associated HIV-infected individuals, the association of escape variants and with loss of this immune response, providing circumstantial evi- disease progression is variable, depending upon the escape mutation dence that loss of this immunodominant response results in reduced and the restricting HLA element (14, 18). In particular, the asso- control of the virus replication. ciation of mutations at R264 inducing reduced affinity to HLA- The data are also consistent with the published literature in that B*27 and progression has not been definitively demonstrated (5, 6, in the majority of individuals who carry HLA-B*27, these escape 57, 58). To address this question, we chose a candidate epitope mutations take many years to arise. This is likely due to the need (KK10) that is known to be uniformly immunodominant (28, 30, for the accumulation of compensatory mutations that overcome 59) and that is restricted by an HLA allele associated with reduced the negative effects on viral fitness induced by escape mutations at rates of disease progression (38–40). In this study, we considered codon 264 in KK10 if present alone (35, 64). However, the fact the effect of a range of other viral and host factors that might impact that escape seems to arise while individuals have relatively intact on disease progression, in addition to considering the effect of viral immune systems seems to be inconsistent with the observation sequence variation resulting in known escape mutations within the from Gao et al. (68), in that the major benefit of carriage of HLA- conserved immunodominant KK10 epitope in Gag p24. These B*27 occurs late in the disease process in those with ,200 CD4+ factors included: 1) host genetic factors described as impact- T cells/ml. These late protective effects can likely be attributed to ing upon disease progression: polymorphisms in chemokine cor- the long-term maintenance of WT KK10 sequence in a subset of eceptors (CCR5, CCR2) and the natural ligand of CXCR4 (SDF1); HLA-B*27+ individuals (35). Intervention with ART limits our 2) diversity of TCR repertoire responding to the KK10 epitope; ability to explore this hypothesis in the context of this data set. and 3) sequence variation at the known subdominant HLA-B*27– Interestingly, there was no statistical difference between the restricted epitopes in gag, env, and nef genes (6, 28, 30, 57, 60, 61), WT and the escape groups with regard to the presence of escape as well as the presence of deletions within the nef gene (62). mutations at HLA-B*27–restricted epitopes in Gag p17, Nef, and We found that, even among this group selected for their slow gp41, indicating that there is no rapid accumulation of escape disease progression, those with escape mutations at codon 264 in mutants at subdominant epitopes following the generation of es- KK10 developed higher viral loads relative to those that maintained cape in KK10. Where present, variations in these subdominant The Journal of Immunology 487 epitopes did not show any statistically significant impact on CD4+ 6. Kelleher, A. D., C. Long, E. C. Holmes, R. L. Allen, J. Wilson, C. Conlon, C. Workman, S. Shaunak, K. Olson, P. Goulder, et al. 2001. Clustered mutations T cell counts or viral load. in HIV-1 gag are consistently required for escape from HLA-B27-restricted Unlike a previous study in vaccinated macaques challenged with cytotoxic T lymphocyte responses. J. Exp. Med. 193: 375–386. SIV (20), we were unable to demonstrate, using a standardized 7. Klenerman, P., U. C. Meier, R. E. Phillips, and A. J. McMichael. 1995. The effects of natural altered peptide ligands on the whole blood cytotoxic T lymphocyte quantitative measurement, a relationship between the breadth of response to human immunodeficiency virus. Eur. J. Immunol. 25: 1927–1931. TCR repertoire and the propensity for control of virus during 8. Klenerman, P., S. Rowland-Jones, S. McAdam, J. Edwards, S. Daenke, chronic infection. This may be an artifact of sample size, because D. Lalloo, B. Ko¨ppe, W. Rosenberg, D. Boyd, A. Edwards, et al. 1994. Cytotoxic T-cell activity antagonized by naturally occurring HIV-1 Gag variants. Nature estimations from this data set make experiments to formally test 369: 403–407. this hypothesis unfeasibly large. Although this study attempted to 9. Meier, U. C., P. Klenerman, P. Griffin, W. James, B. Ko¨ppe, B. Larder, address the impact on a range of host factors, as well as viral A. McMichael, and R. Phillips. 1995. Cytotoxic T lymphocyte lysis inhibited by viable HIV mutants. Science 270: 1360–1362. escape on viral control, it does not address the impact of quali- 10. Price, D. A., P. J. Goulder, P. Klenerman, A. K. Sewell, P. J. Easterbrook, tative differences in the T cell response to the KK10 epitope. M. Troop, C. R. Bangham, and R. E. Phillips. 1997. Positive selection of HIV-1 cytotoxic T lymphocyte escape variants during primary infection. Proc. Natl. Clonal turnover and senescence may impact outcome, as identified Acad. Sci. USA 94: 1890–1895. by a previous study (66). 11. Sewell, A. K., G. C. Harcourt, P. J. Goulder, D. A. Price, and R. E. Phillips. 1997. To the best of our knowledge, this is the first study of an HLA- Antagonism of cytotoxic T lymphocyte-mediated lysis by natural HIV-1 altered peptide ligands requires simultaneous presentation of agonist and antagonist B*27 population to show an exclusive association between the peptides. Eur. J. Immunol. 27: 2323–2329. presence of escape mutations and increases in plasma viral load. 12. Yokomaku, Y., H. Miura, H. Tomiyama, A. Kawana-Tachikawa, M. Takiguchi, Although escape mutations in the HLA-B*27 population were A. Kojima, Y. Nagai, A. Iwamoto, Z. Matsuda, and K. Ariyoshi. 2004. Impaired processing and presentation of cytotoxic-T-lymphocyte (CTL) epitopes are

+ Downloaded from reported in other studies, ours is the largest study of HLA-B*27 major escape mechanisms from CTL immune pressure in human immunodefi- individuals to date. Although our sample size is small, it is more ciency virus type 1 infection. J. Virol. 78: 1324–1332. than double the number of HLA-B*27+ subjects included in any 13. Frater, A. J., C. T. Edwards, N. McCarthy, J. Fox, H. Brown, A. Milicic, N. Mackie, T. Pillay, J. W. Drijfhout, S. Dustan, et al. 2006. Passive sexual other published work (5, 6, 29, 49). An additional caveat is the transmission of human immunodeficiency virus type 1 variants and adaptation in possibility that the escape mutations present at study entry had been new hosts. J. Virol. 80: 7226–7234. 14. Frater, A. J., H. Brown, A. Oxenius, H. F. Gu¨nthard, B. Hirschel, N. Robinson, present since transmission (32, 37, 69) or at least arose very early in A. J. Leslie, R. Payne, H. Crawford, A. Prendergast, et al. 2007. Effective T-cell the disease course (6, 30), and the observed loss of viral control is responses select human immunodeficiency virus mutants and slow disease pro- http://www.jimmunol.org/ related to unmeasured parameters. The frequency of these escape gression. J. Virol. 81: 6742–6751. 15. Moore, C. B., M. John, I. R. James, F. T. Christiansen, C. S. Witt, and mutations within databases and the literature suggest that this is an S. A. Mallal. 2002. Evidence of HIV-1 adaptation to HLA-restricted immune unlikely scenario but is one that cannot be completely excluded. responses at a population level. Science 296: 1439–1443. Significant challenges remain for the identification of correlates of 16. Wolinsky, S. M., B. T. Korber, A. U. Neumann, M. Daniels, K. J. Kunstman, A. J. Whetsell, M. R. Furtado, Y. Cao, D. D. Ho, and J. T. Safrit. 1996. Adaptive protection in nonprogressors. In conclusion, we found that the evolution of human immunodeficiency virus-type 1 during the natural course of plasma RNA level was significantly higher following the de- infection. Science 272: 537–542. velopment of escape compared with those that maintain WT virus 17. Yusim, K., C. Kesmir, B. Gaschen, M. M. Addo, M. Altfeld, S. Brunak, A. Chigaev, V. Detours, and B. T. Korber. 2002. Clustering patterns of cytotoxic (p = 0.01). These results suggest that escape mutation in HLA- T-lymphocyte epitopes in human immunodeficiency virus type 1 (HIV-1) proteins B*27+ individuals is a major determinate of disease control, as reveal imprints of immune evasion on HIV-1 global variation. J. Virol. 76: 8757– by guest on September 28, 2021 determined by plasma concentrations of HIV-1 RNA. 8768. 18. Brumme, Z. L., C. J. Brumme, J. Carlson, H. Streeck, M. John, Q. Eichbaum, B. L. Block, B. Baker, C. Kadie, M. Markowitz, et al. 2008. Marked epitope- and Acknowledgments allele-specific differences in rates of mutation in human immunodeficiency type 1 (HIV-1) Gag, Pol, and Nef cytotoxic T-lymphocyte epitopes in acute/early We thank Drs. Ben Anderson (St Leonard’s Medical Centre, St Leonard’s, HIV-1 infection. J. Virol. 82: 9216–9227. New South Wales, Australia), Nicholas C. Doong (Burwood Road Medical 19. Duda, A., L. Lee-Turner, J. Fox, N. Robinson, S. Dustan, S. Kaye, H. Fryer, Centre, Burwood, New South Wales, Australia), Mark Kelly (AIDS Med- M. Carrington, M. McClure, A. R. McLean, et al; SPARTAC Trial Investigators. 2009. HLA-associated clinical progression correlates with epitope reversion ical Centre, Brisbane, Queensland, Australia), Ross Price (Taylor Square rates in early human immunodeficiency virus infection. J. Virol. 83: 1228–1239. Private Clinic), and all patients who contributed to this study. 20. Price, D. A., S. M. West, M. R. Betts, L. E. Ruff, J. M. Brenchley, D. R. Ambrozak, Y. Edghill-Smith, M. J. Kuroda, D. Bogdan, K. Kunstman, et al. 2004. T cell receptor recognition motifs govern immune escape patterns in Disclosures acute SIV infection. Immunity 21: 793–803. The authors have no financial conflicts of interest. 21. Wang, Y. E., B. Li, J. M. Carlson, H. Streeck, A. D. Gladden, R. Goodman, A. Schneidewind, K. A. Power, I. Toth, N. Frahm, et al. 2009. Protective HLA class I alleles that restrict acute-phase CD8+ T-cell responses are associated with viral escape mutations located in highly conserved regions of human immuno- References deficiency virus type 1. J. Virol. 83: 1845–1855. 1. Allen, T. M., M. Altfeld, X. G. Yu, K. M. O’Sullivan, M. Lichterfeld, S. Le Gall, 22. Goonetilleke, N., M. K. Liu, J. F. Salazar-Gonzalez, G. Ferrari, E. Giorgi, M. John, B. R. Mothe, P. K. Lee, E. T. Kalife, et al. 2004. Selection, transmission, V. V. Ganusov, B. F. Keele, G. H. Learn, E. L. Turnbull, M. G. Salazar, et al; and reversion of an antigen-processing cytotoxic T-lymphocyte escape mutation in CHAVI Clinical Core B. 2009. The first T cell response to transmitted/founder human immunodeficiency virus type 1 infection. J. Virol. 78: 7069–7078. virus contributes to the control of acute viremia in HIV-1 infection. J. Exp. Med. 2. Borrow, P., H. Lewicki, X. Wei, M. S. Horwitz, N. Peffer, H. Meyers, 206: 1253–1272. J. A. Nelson, J. E. Gairin, B. H. Hahn, M. B. Oldstone, and G. M. Shaw. 1997. 23. Allen, T. M., L. Mortara, B. R. Mothe´, M. Liebl, P. Jing, B. Calore, Antiviral pressure exerted by HIV-1-specific cytotoxic T lymphocytes (CTLs) M. Piekarczyk, R. Ruddersdorf, D. H. O’Connor, X. Wang, et al. 2002. Tat- during primary infection demonstrated by rapid selection of CTL escape virus. vaccinated macaques do not control simian immunodeficiency virus SIVmac239 Nat. Med. 3: 205–211. replication. J. Virol. 76: 4108–4112. 3. Draenert, R., S. Le Gall, K. J. Pfafferott, A. J. Leslie, P. Chetty, C. Brander, 24. Barouch, D. H., J. Kunstman, J. Glowczwskie, K. J. Kunstman, M. A. Egan, E. C. Holmes, S. C. Chang, M. E. Feeney, M. M. Addo, et al. 2004. Immune F. W. Peyerl, S. Santra, M. J. Kuroda, J. E. Schmitz, K. Beaudry, et al. 2003. Viral selection for altered antigen processing leads to cytotoxic T lymphocyte escape escape from dominant simian immunodeficiency virus epitope-specific cytotoxic in chronic HIV-1 infection. J. Exp. Med. 199: 905–915. T lymphocytes in DNA-vaccinated rhesus monkeys. J. Virol. 77: 7367–7375. 4. Furutsuki, T., N. Hosoya, A. Kawana-Tachikawa, M. Tomizawa, T. Odawara, 25. Chen, Z. W., A. Craiu, L. Shen, M. J. Kuroda, U. C. Iroku, D. I. Watkins, M. Goto, Y. Kitamura, T. Nakamura, A. D. Kelleher, D. A. Cooper, and G. Voss, and N. L. Letvin. 2000. Simian immunodeficiency virus evades A. Iwamoto. 2004. Frequent transmission of cytotoxic-T-lymphocyte escape a dominant epitope-specific cytotoxic T lymphocyte response through a mutation mutants of human immunodeficiency virus type 1 in the highly HLA-A24- resulting in the accelerated dissociation of viral peptide and MHC class I. J. positive Japanese population. J. Virol. 78: 8437–8445. Immunol. 164: 6474–6479. 5. Goulder, P. J., R. E. Phillips, R. A. Colbert, S. McAdam, G. Ogg, M. A. Nowak, 26. Evans, D. T., D. H. O’Connor, P. Jing, J. L. Dzuris, J. Sidney, J. da Silva, P. Giangrande, G. Luzzi, B. Morgan, A. Edwards, et al. 1997. Late escape from T. M. Allen, H. Horton, J. E. Venham, R. A. Rudersdorf, et al. 1999. Virus- an immunodominant cytotoxic T-lymphocyte response associated with pro- specific cytotoxic T-lymphocyte responses select for amino-acid variation in gression to AIDS. Nat. Med. 3: 212–217. simian immunodeficiency virus Env and Nef. Nat. Med. 5: 1270–1276. 488 HIV IMMUNE ESCAPE PREDICTS VIRAL LOAD OUTCOME

27. Friedrich, T. C., A. B. McDermott, M. R. Reynolds, S. Piaskowski, S. Fuenger, encoded HLA-B27-restricted cytotoxic T lymphocyte epitope despite specific I. P. De Souza, R. Rudersdorf, C. Cullen, L. J. Yant, L. Vojnov, et al. 2004. in vitro reactivity. Eur. J. Immunol. 21: 2637–2640. Consequences of cytotoxic T-lymphocyte escape: common escape mutations in 50. Rhodes, D. I., L. Ashton, A. Solomon, A. Carr, D. Cooper, J. Kaldor, N. Deacon; simian immunodeficiency virus are poorly recognized in naive hosts. J. Virol. 78: Australian Long-Term Nonprogressor Study Group. 2000. Characterization of 10064–10073. three nef-defective human immunodeficiency virus type 1 strains associated with 28. Nixon, D. F., A. R. Townsend, J. G. Elvin, C. R. Rizza, J. Gallwey, and long-term nonprogression. J. Virol. 74: 10581–10588. A. J. McMichael. 1988. HIV-1 gag-specific cytotoxic T lymphocytes defined 51. Holdsworth, R., C. K. Hurley, S. G. Marsh, M. Lau, H. J. Noreen, with recombinant vaccinia virus and synthetic peptides. Nature 336: 484–487. J. H. Kempenich, M. Setterholm, and M. Maiers. 2009. The HLA dictionary 29. Nietfield, W., M. Bauer, M. Fevrier, R. Maier, B. Holzwarth, R. Frank, B. Maier, 2008: a summary of HLA-A, -B, -C, -DRB1/3/4/5, and -DQB1 alleles and their Y. Riviere, and A. Meyerhans. 1995. Sequence constraints and recognition by association with serologically defined HLA-A, -B, -C, -DR, and -DQ antigens. CTL of an HLA-B27-restricted HIV-1 gag epitope. J. Immunol. 154: 2189–2197. Tissue Antigens 73: 95–170. 30. Wilson, J. D., G. S. Ogg, R. L. Allen, C. Davis, S. Shaunak, J. Downie, W. Dyer, 52. Kawashima, Y., K. Pfafferott, J. Frater, P. Matthews, R. Payne, M. Addo, C. Workman, S. Sullivan, A. J. McMichael, and S. L. Rowland-Jones. 2000. H. Gatanaga, M. Fujiwara, A. Hachiya, H. Koizumi, et al. 2009. Adaptation of Direct visualization of HIV-1-specific cytotoxic T lymphocytes during primary HIV-1 to human leukocyte antigen class I. Nature 458: 641–645. infection. AIDS 14: 225–233. 53. Kiepiela, P., A. J. Leslie, I. Honeyborne, D. Ramduth, C. Thobakgale, S. Chetty, 31. Ammaranond, P., J. Zaunders, C. Satchell, D. van Bockel, D. A. Cooper, and P. Rathnavalu, C. Moore, K. J. Pfafferott, L. Hilton, et al. 2004. Dominant in- A. D. Kelleher. 2005. A new variant cytotoxic T lymphocyte escape mutation in fluence of HLA-B in mediating the potential co-evolution of HIV and HLA. HLA-B27-positive individuals infected with HIV type 1. AIDS Res. Hum. Ret- Nature 432: 769–775. roviruses 21: 395–397. 54. Leslie, A., P. C. Matthews, J. Listgarten, J. M. Carlson, C. Kadie, T. Ndung’u, 32. Goulder, P. J., C. Brander, Y. Tang, C. Tremblay, R. A. Colbert, M. M. Addo, C. Brander, H. Coovadia, B. D. Walker, D. Heckerman, and P. J. Goulder. 2010. E. S. Rosenberg, T. Nguyen, R. Allen, A. Trocha, et al. 2001. Evolution and trans- Additive contribution of HLA class I alleles in the immune control of HIV-1 mission of stable CTL escape mutations in HIV infection. Nature 412: 334–338. infection. J. Virol. 84: 9879–9888. 33. Gamble, T. R., F. F. Vajdos, S. Yoo, D. K. Worthylake, M. Houseweart, 55. Flores-Villanueva, P. O., H. Hendel, S. Caillat-Zucman, J. Rappaport, A. Burgos- W. I. Sundquist, and C. P. Hill. 1996. Crystal structure of human cyclophilin A Tiburcio, S. Bertin-Maghit, J. A. Ruiz-Morales, M. E. Teran, J. Rodriguez-Tafur, bound to the amino-terminal domain of HIV-1 capsid. Cell 87: 1285–1294. and J. F. Zagury. 2003. Associations of MHC ancestral haplotypes with Downloaded from 34. Momany, C., L. C. Kovari, A. J. Prongay, W. Keller, R. K. Gitti, B. M. Lee, resistance/susceptibility to AIDS disease development. J. Immunol. 170: 1925– A. E. Gorbalenya, L. Tong, J. McClure, L. S. Ehrlich, et al. 1996. Crystal 1929. structure of dimeric HIV-1 capsid protein. Nat. Struct. Biol. 3: 763–770. 56. Itescu, S., U. Mathur-Wagh, M. L. Skovron, L. J. Brancato, M. Marmor, 35. Schneidewind, A., M. A. Brockman, R. Yang, R. I. Adam, B. Li, S. Le Gall, A. Zeleniuch-Jacquotte, and R. Winchester. 1992. HLA-B35 is associated with C. R. Rinaldo, S. L. Craggs, R. L. Allgaier, K. A. Power, et al. 2007. Escape accelerated progression to AIDS. J. Acquir. Immune Defic. Syndr. 5: 37–45. from the dominant HLA-B27-restricted cytotoxic T-lymphocyte response in Gag 57. Feeney, M. E., Y. Tang, K. A. Roosevelt, A. J. Leslie, K. McIntosh, N. Karthas, is associated with a dramatic reduction in human immunodeficiency virus type B. D. Walker, and P. J. Goulder. 2004. Immune escape precedes breakthrough 1 replication. J. Virol. 81: 12382–12393. human immunodeficiency virus type 1 viremia and broadening of the cytotoxic

36. Altfeld, M., and T. M. Allen. 2006. Hitting HIV where it hurts: an alternative T-lymphocyte response in an HLA-B27-positive long-term-nonprogressing http://www.jimmunol.org/ approach to HIV vaccine design. Trends Immunol. 27: 504–510. child. J. Virol. 78: 8927–8930. 37. Betts, M. R., B. Exley, D. A. Price, A. Bansal, Z. T. Camacho, V. Teaberry, 58. Phillips, R. E., S. Rowland-Jones, D. F. Nixon, F. M. Gotch, J. P. Edwards, S. M. West, D. R. Ambrozak, G. Tomaras, M. Roederer, et al. 2005. Charac- A. O. Ogunlesi, J. G. Elvin, J. A. Rothbard, C. R. Bangham, C. R. Rizza, et al. terization of functional and phenotypic changes in anti-Gag vaccine-induced 1991. Human immunodeﬁciency virus genetic variation that can escape cyto- T cell responses and their role in protection after HIV-1 infection. Proc. Natl. toxic T cell recognition. Nature 354: 453–459. Acad. Sci. USA 102: 4512–4517. 59. Altfeld, M., E. T. Kalife, Y. Qi, H. Streeck, M. Lichterfeld, M. N. Johnston, 38. Carrington, M., G. W. Nelson, M. P. Martin, T. Kissner, D. Vlahov, J. J. Goedert, N. Burgett, M. E. Swartz, A. Yang, G. Alter, et al. 2006. HLA Alleles Associated R. Kaslow, S. Buchbinder, K. Hoots, and S. J. O’Brien. 1999. HLA and HIV-1: with Delayed Progression to AIDS Contribute Strongly to the Initial CD8(+) heterozygote advantage and B*35-Cw*04 disadvantage. Science 283: 1748– T Cell Response against HIV-1. PLoS Med. 3: e403. 1752. 60. Goulder, P. J., A. K. Sewell, D. G. Lalloo, D. A. Price, J. A. Whelan, J. Evans, 39. Hendel, H., S. Caillat-Zucman, H. Lebuanec, M. Carrington, S. O’Brien, G. P. Taylor, G. Luzzi, P. Giangrande, R. E. Phillips, and A. J. McMichael. 1997.

J. M. Andrieu, F. Scha¨chter, D. Zagury, J. Rappaport, C. Winkler, et al. 1999. by guest on September 28, 2021 Patterns of immunodominance in HIV-1-specific cytotoxic T lymphocyte New class I and II HLA alleles strongly associated with opposite patterns of responses in two human histocompatibility leukocyte antigens (HLA)-identical progression to AIDS. J. Immunol. 162: 6942–6946. siblings with HLA-A*0201 are influenced by epitope mutation. J. Exp. Med. 40. Kaslow, R. A., M. Carrington, R. Apple, L. Park, A. Munõz, A. J. Saah, J. J. Goedert, C. Winkler, S. J. O’Brien, C. Rinaldo, et al. 1996. Influence of 185: 1423–1433. combinations of human major histocompatibility complex genes on the course of 61. Goulder, P. J., M. A. Altfeld, E. S. Rosenberg, T. Nguyen, Y. Tang, HIV-1 infection. Nat. Med. 2: 405–411. R. L. Eldridge, M. M. Addo, S. He, J. S. Mukherjee, M. N. Phillips, et al. 2001. 41. Ashton, L. J., A. Carr, P. H. Cunningham, M. Roggensack, K. McLean, M. Law, Substantial differences in specificity of HIV-specific cytotoxic T cells in acute M. Robertson, D. A. Cooper, J. M. Kaldor; Australian Long-Term Nonprogressor and chronic HIV infection. J. Exp. Med. 193: 181–194. Study Group. 1998. Predictors of progression in long-term nonprogressors. AIDS 62. Deacon, N. J., A. Tsykin, A. Solomon, K. Smith, M. Ludford-Menting, Res. Hum. Retroviruses 14: 117–121. D. J. Hooker, D. A. McPhee, A. L. Greenway, A. Ellett, C. Chatfield, et al. 1995. 42. World Health Organization. 2007. WHO case definitions of HIV for surveillance Genomic structure of an attenuated quasi species of HIV-1 from a blood trans- and revised clinical staging and immunological classification of HIV-related fusion donor and recipients. Science 270: 988–991. disease in adults and children. WHO Press, Geneva, Switzerland, p. 8–15. 63. Brumme, Z. L., I. Tao, S. Szeto, C. J. Brumme, J. M. Carlson, D. Chan, C. Kadie, 43. Stewart, G. J., L. J. Ashton, R. A. Biti, R. A. Ffrench, B. H. Bennetts, N. Frahm, C. Brander, B. Walker, et al. 2008. Human leukocyte antigen-specific N. R. Newcombe, E. M. Benson, A. Carr, D. A. Cooper, J. M. Kaldor; The polymorphisms in HIV-1 Gag and their association with viral load in chronic Australian Long-Term Non-Progressor Study Group. 1997. Increased frequency untreated infection. AIDS 22: 1277–1286. of CCR-5 delta 32 heterozygotes among long-term non-progressors with HIV-1 64. Miura, T., M. A. Brockman, Z. L. Brumme, C. J. Brumme, F. Pereyra, A. Trocha, infection. AIDS 11: 1833–1838. B. L. Block, A. Schneidewind, T. M. Allen, D. Heckerman, and B. D. Walker. 44. Voevodin, A., E. Samilchuk, and S. Dashti. 1999. Frequencies of SDF-1 che- 2009. HLA-associated alterations in replication capacity of chimeric NL4-3 mokine, CCR-5, and CCR-2 chemokine receptor gene alleles conferring re- viruses carrying gag-protease from elite controllers of human immunodefi- sistance to human immunodeficiency virus type 1 and AIDS in Kuwaitis. J. Med. ciency virus type 1. J. Virol. 83: 140–149. Virol. 58: 54–58. 65. Vujovic O. 2009. Initiation of antiretroviral therapy in the naive patient. In HIV 45. van Bockel, D., D. A. Price, T. E. Asher, V. Venturi, K. Suzuki, K. Warton, Management in Australia: A Guide for Clinical Care. J. Hoy, S. Lewin, J. J. Post, M. P. Davenport, D. A. Cooper, D. C. Douek, and A. D. Kelleher. 2007. Vali- A. Street, eds. Australian Society of HIV Medicine (ASHM) Press, Sydney, dation of RNA-based molecular clonotype analysis for virus-specific CD8+ Australia, p. 77–92. T-cells in formaldehyde-fixed specimens isolated from peripheral blood. J. 66. Almeida, J. R., D. A. Price, L. Papagno, Z. A. Arkoub, D. Sauce, E. Bornstein, Immunol. Methods 326: 127–138. T. E. Asher, A. Samri, A. Schnuriger, I. Theodorou, et al. 2007. Superior control 46. Douek, D. C., M. R. Betts, J. M. Brenchley, B. J. Hill, D. R. Ambrozak, of HIV-1 replication by CD8+ T cells is reflected by their avidity, poly- K. L. Ngai, N. J. Karandikar, J. P. Casazza, and R. A. Koup. 2002. A novel functionality, and clonal turnover. J. Exp. Med. 204: 2473–2485. approach to the analysis of specificity, clonality, and frequency of HIV-specific 67. Lichterfeld, M., D. G. Kavanagh, K. L. Williams, B. Moza, S. K. Mui, T. Miura, T cell responses reveals a potential mechanism for control of viral escape. J. R. Sivamurthy, R. Allgaier, F. Pereyra, A. Trocha, et al. 2007. A viral CTL es- Immunol. 168: 3099–3104. cape mutation leading to immunoglobulin-like transcript 4-mediated functional 47. Byakwaga, H., J. Zhou, K. Petoumenos, M. G. Law, M. A. Boyd, S. Emery, inhibition of myelomonocytic cells. J. Exp. Med. 204: 2813–2824. D. A. Cooper, and P. W. Mallon. 2009. Effect of nucleoside reverse transcriptase 68. Gao, X., A. Bashirova, A. K. Iversen, J. Phair, J. J. Goedert, S. Buchbinder, inhibitors on CD4 T-cell recovery in HIV-1-infected individuals receiving long- K. Hoots, D. Vlahov, M. Altfeld, S. J. O’Brien, and M. Carrington. 2005. AIDS term fully suppressive combination antiretroviral therapy. HIV Med. 10: 143–151. restriction HLA allotypes target distinct intervals of HIV-1 pathogenesis. Nat. 48. Venturi, V., K. Kedzierska, S. J. Turner, P. C. Doherty, and M. P. Davenport. Med. 11: 1290–1292. 2007. Methods for comparing the diversity of samples of the T cell receptor 69. O’Connell, K. A., R. K. Pelz, J. B. Dinoso, E. Dunlop, J. Paik-Tesch, repertoire. J. Immunol. Methods 321: 182–195. T. M. Williams, and J. N. Blankson. 2010. Prolonged Control of an HIV Type 1 49. Meyerhans, A., G. Dadaglio, J. P. Vartanian, P. Langlade-Demoyen, R. Frank, Escape Variant Following Treatment Interruption in an HLA-B*27-Positive B. Asjo¨, F. Plata, and S. Wain-Hobson. 1991. In vivo persistence of a HIV-1- Patient. AIDS Res. Hum. Retroviruses. In press.