Figures
Abstract
In HIV infection, TLR7-triggered IFN-α production exerts a direct antiviral effect through the inhibition of viral replication, but may also be involved in immune pathogenesis leading to AIDS. TLR7 could also be an important mediator of vaccine efficacy. In this study, we analyzed polymorphisms in the X-linked TLR7 gene in the rhesus macaque model of AIDS. Upon resequencing of the TLR7 gene in 36 rhesus macaques of Indian origin, 12 polymorphic sites were detected. Next, we identified three tightly linked single nucleotide polymorphisms (SNP) as being associated with survival time. Genotyping of 119 untreated, simian immunodeficiency virus (SIV)-infected male rhesus macaques, including an ‘MHC adjusted’ subset, revealed that the three TLR7 SNPs are also significantly associated with set-point viral load. Surprisingly, this effect was not observed in 72 immunized SIV-infected male monkeys. We hypothesize (i) that SNP c.13G>A in the leader peptide is causative for the observed genotype-phenotype association and that (ii) the underlying mechanism is related to RNA secondary structure formation. Therefore, we investigated a fourth SNP (c.-17C>T), located 17 bp upstream of the ATG translation initiation codon, that is also potentially capable of influencing RNA structure. In c.13A carriers, neither set-point viral load nor survival time were related to the c.-17C>T genotype. In c.13G carriers, by contrast, the c.-17C allele was significantly associated with prolonged survival. Again, no such association was detected among immunized SIV-infected macaques. Our results highlight the dual role of TLR7 in immunodeficiency virus infection and vaccination and imply that it may be important to control human AIDS vaccine trials, not only for MHC genotype, but also for TLR7 genotype.
Citation: Siddiqui RA, Krawczak M, Platzer M, Sauermann U (2011) Association of TLR7 Variants with AIDS-Like Disease and AIDS Vaccine Efficacy in Rhesus Macaques. PLoS ONE 6(10): e25474. https://doi.org/10.1371/journal.pone.0025474
Editor: Nick Gay, University of Cambridge, United Kingdom
Received: June 22, 2011; Accepted: September 5, 2011; Published: October 13, 2011
Copyright: © 2011 Siddiqui et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was supported by the German Ministry of Education and Research (BMBF) through National Genome Research Network (NGFN) grant 0313360. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Toll-like receptor 7 (TLR7) localizes to intracellular vesicles in antigen-presenting cells such as plasmacytoid dendritic cells, macrophages, memory B cells and T cells. In these cells, TLR7 functions as a receptor for pathogen recognition in that it binds ligands like uridine-rich single-stranded (ss) RNAs or certain small interfering RNAs [1], [2], [3], [4]. Through the recognition of ssRNA derived from RNA viruses, TLR7 can also act as a danger receptor for viral infection (reviewed in [5], [6], [7]). Moreover, TLR7 activation modulates the innate and adaptive antiviral immune response by triggering the production of type I interferons, cytokines and chemokines [8], [9]. The X-linked TLR7 gene appears to have been subject to positive selection during primate evolution, thereby underpinning its relevance in host-pathogen interaction [10].
TLR7 activation is an important mediator of vaccine efficacy. Targeting of TLR7 by synthetic agonists like polyUs21, or conjugation of TLR7 to a protein vaccine, can prime high-frequency polyfunctional type 1 T helper cell (Th1) and cytotoxic T-lymphocyte (CTL) response, probably through the activation of plasmacytoid dendritic cells (PDC) [11], [12], [13]. The stimulation of Toll-like receptors, in particular TLR7, also appears to be a determinant of the greater magnitude and Th1 polarization of the immune response induced by inactivated whole-virus H5N1 influenza vaccine, as compared to split or subunit vaccine-induced responses [14]. Furthermore, IGA responses important for mucosal immunity potentially also depend upon TLR7 activation [15]. Targeting TLR7 may therefore improve vaccine efficacy.
In HIV infection, TLR7-triggered IFN-α production appears to be a double-edged sword [16]. On the one hand, stimulation of antigen-presenting cells by synthetic TLR7 agonists like imidazoquinoline compounds augment HIV-1-specific T cell responses in vitro [17]. Moreover, administration of IFN-α or triggering of TLR7/8 by imidazoquinoline compound R848 has direct anti-HIV effects by inhibiting viral replication in vitro, and potentially in vivo, through multiple pathways [18], [19], [20], [21], [22], [23]. On the other hand, TLR7-triggered pathways appear to be directly involved in immune pathogenesis leading to AIDS [24], [25]. Chronic immune stimulation via TLR7 in mice caused progressive lymphoid system destruction resembling HIV-mediated immunopathology [26]. Opposing effects of TLR7 have also been observed in the context of respiratory syncytial virus infection (RSV) in mice. Whereas administering TLR7/8 agonists during immunization did hardly influence the phenotype upon challenge with RSV, the immunostimulatory properties of the same agonists increased disease severity when used in mice that had already been infected [27]. Finally, since IFN-α production is particularly enhanced in females upon TLR7 stimulation, the latter may also contribute to sex-specific disease progression [24], [28].
The pathophysiological role of TLR7 can be studied in some detail in humans owing to a functional polymorphism located in the signal peptide. This Q11L amino acid replacement (rs179008) has been shown to confer susceptibility to asthma [29] and to influence the course of hepatitis C infection [30], HIV disease progression and potentially HIV-1 acquisition [31]. Peripheral blood mononuclear cells (PBMCs) from individuals carrying the minor allele (11L) secreted less IFN-α upon stimulation by TLR7 agonist Imiquimod compared to PBMCs from individuals carrying the major allele [31]. Interestingly, the allele associated with diminished ex vivo IFN-α secretion was also associated with an accelerated time to onset of antiviral therapy (CD4 counts>350 µl) and higher set-point viral load in HIV-infected patients [31], indicating that the protective antiviral effects of TLR7 may prevail in the initial and chronic phases of infection.
The important role of TLR7 in viral infection and immunization prompted us to characterize TLR7 gene polymorphisms in simian immunodeficiency virus (SIV)-infected rhesus macaques (Macaca mulatta), which represent the most important animal model of HIV infection. Moreover, the expression patterns of TLR7 and other TLRs in antigen-presenting cells such as dendritic cells, monocytes and B cells are similar in macaques and humans, but differ markedly from the expression profiles in mice [32]. TLRs also respond to the same ligands in macaques as in humans [32]. Upon targeted TLR7 re-sequencing in 36 rhesus macaques of Indian origin, we detected 12 polymorphisms and subsequently investigated their possible association with AIDS-free survival and viremia in SIV-infected animals. We identified two polymorphisms, located immediately upstream and downstream of the TLR7 translation initiation codon, to be associated with survival time and/or set-point viral load in untreated SIV-infected macaques. Interestingly, no such association was evident in immunized-infected macaques.
Results
Variability of the rhesus TLR7 gene
Like in humans and many other species, the X-linked TLR7 gene in rhesus macaque contains three exons, namely two untranslated exons encompassing the 5′ untranslated region (5′UTR) and the ATG translation initiation codon, and exon 3 spanning the sequence coding for the leader peptide, the mature TLR7 polypeptide and a 3′ untranslated region (3′UTR) of about 1700 bp not yet annotated firmly in rhesus macaque. In contrast to the human situation, however, no polymorphisms in the rhesus TLR7 gene were known prior to this study. We therefore had to establish a panel of polymorphisms first before we could assess the impact of TLR7 gene variability upon the progression and outcome of SIV infection in this species.
Initial DNA sequence analysis of all three exons and some intronic parts of the TLR7 coding region, comprising a total of 7853 bp, was performed in a pilot sample 36 animals. Twelve variable positions were identified in the DNA sequence encoding the TLR7 mRNA (Figure S1), four of which were located in the protein-coding region of TLR7 (Figure S2). Two of the coding polymorphisms were non-synonymous whilst two were synonymous. The rhesus TLR7 amino acid sequence differs from human TLR7 by 19 amino acid substitutions (Figure S2). Notably, the amino acid sequences encoding the transmembrane and the Toll/IL-1 receptor (TIR) domain important for transport and associated signalling are identical in humans and rhesus macaques, and no polymorphisms were detected in these regions (Figure S2). Seven of the eight non-coding polymorphisms were located in the 3′UTR of exon 3 whereas one was found in exon 2, 17 bp upstream of the ATG translation initiation codon. Since the rhesus genome is not yet annotated as thoroughly as the human genome, we provisionally labelled the SNPs following the recommendations for the description of DNA sequence variants made by the Human Genome Variation Society [33]. Allele frequencies observed among the 36 macaques of the pilot sample are given in Table 1.
Four single nucleotide polymorphisms (SNPs) were also genotyped in a larger sample. Thus, a total of 237 animals of Indian origin (46 females, 191 males), including the 36 samples used in the re-sequencing experiment, were investigated for SNPs c.-17C>T (5′UTR), c.13G>A (V5M), c.1710T>A (V570V) and c.*1604C>T (3′UTR). The last three polymorphisms were in strong pair-wise linkage disequilibrium (LD) in our sample (all pair-wise allelic r2>0.90) so that the respective allele frequencies were very similar (Table 1). Allele frequencies also differed hardly between the initial re-sequencing panel and the larger validation sample (data not shown).
Association of TLR7 polymorphisms c.13G>A, c.1710T>A and c.*1604C>T with survival time and viral load in untreated SIV-infected rhesus macaques
The initial re-sequencing experiment was carried out on 36 unrelated SIV-infected rhesus macaques that displayed the whole spectrum of disease progression (i.e., AIDS-free survival from 6 to >580 weeks post infection). Three of the discovered SNPs, namely c.13G>A, c.1710T>A, and c.*1604C>T, were suggestive of an association with survival time in the pilot sample. For further validation, a larger sample of 237 SIV-infected macaques was genotyped for the three SNPs either by Sanger sequencing or using a TaqMan-based allelic discrimination assay. In the following, we will only report upon c.13G>A, because the other two SNPs were in strong LD with this polymorphism (see above).
Among the 119 untreated-infected male macaques genotyped, AIDS-free survival was found to be significantly associated with the c.13G>A hemizygous status. Thus, male macaques carrying the minor allele (c.13A) progressed faster to AIDS than macaques carrying the major allele (c.13G; log-rank χ2 = 5.04, 1 d.f, p = 0.025) (Fig. 1A). Furthermore, c.13A carriers had a significantly higher viral load at set-point (F = 6.82, p = 0.011) (Fig. 2A). No difference in peak viremia at week 2 post infection was observed, indicating that initial viral replication is not strongly influenced by c.13G>A genotype (F = 1.33, p = 0.252).
A: untreated-infected macaques (c.13A: n = 35, c.13G: n = 84) (p = 0.025); B: untreated-infected macaques after MHC adjustment (c.13A: n = 26, c.13G: n = 69) (p = 0.077); C: untreated-infected macaques after MHC adjustment carrying haplotype c.-17T,c.13A (blue discontinuous line, n = 14), c.-17C,c.13A (straight blue line, n = 11), c.-17T,c.13G (red discontinuous line, n = 18), c.-17C,c.13G (straight red line, n = 51); (c.13G carriers: c.-17C>T: p = 0.017); D: immunized-infected macaques carrying c.-17C,c.13A , c.-17T,c.13A or c.-17T,c.13G (blue line, n = 38) or c.-17C,c.13G (red line, n = 34) (p>0.3).
Geometric mean and 95% confidence interval are indicated by error bars. Significant differences are marked by asterisks. A: untreated SIV-infected carriers of allele c.13A (n = 23) or c.13G (n = 54), and untreated SIV-infected monkeys after MHC adjustment carrying c.13A (n = 17) or c.13G (n = 46). B: untreated SIV-infected rhesus macaques stratified by their joint c.-17C>T and c.13G>A genotype: c.-17C,c.13A (n = 12), c.-17T,c.13A (n = 10), c.-17T,c.13G (n = 12), c.-17C,c.13G (n = 34). C: plasma RNA copy numbers of immunized SIV-infected macaques carrying c.13A or c.-17T,c.13G (n = 33), c.-17C,c.13G (n = 31), MHC adjusted: c.13A or c.-17T,c.13G (n = 20), c.-17C,c.13G (n = 18), carrying MHC alleles associated with slow disease progression: c.13A or c.-17T,c.13G (n = 11) , c.-17C,c.13G (n = 10).
Since the MHC genotype is known to influence both survival time and viral load in HIV-infected humans as well as in SIV-infected macaques, we controlled our analysis for the presence for MHC genotypes known to strongly influence disease progression (Figure S3). Inspection of the data revealed that MHC genotypes previously found to be associated with either rapid or slow disease progression were not evenly distributed among the male animals studied here [34]. In the group of c.13A carriers, MHC-genotypes associated with rapid disease progression were significantly overrepresented compared to c.13G carriers (17.1% vs. 2.4%; Fisher's exact test p = 0.008) whereas MHC genotypes associated with slow disease progression (Mamu-A1*001, -B*008) were underrepresented (8.3% vs. 20.2%), although the second difference failed to reach statistical significance (Fisher's exact test, p = 0.389). After excluding all of these animals from the analysis (‘MHC adjustment’), survival times were no longer significantly different between the two c.13G>A genotypes (log-rank χ2 = 4.14, 1 d.f., p = 0.077) but still showed a trend towards longer survival being associated with c.13G (Fig. 1B). However, even in the smaller, MHC-adjusted sample comprising 63 monkeys, c.13A carriers still had a significantly higher set-point viral load than c.13G carriers (F = 6.02, p = 0.017) whilst peak viral load did not differ between genotypes (F = 0.61, p = 0.439).
We also investigated 28 untreated SIV-infected females available to seek evidence for a potential sex difference in terms of the observed genotype-phenotype relationship. Whilst only one female was homozygous for -c.13A, c.AG heterozyotes and c.GG homozygotes were almost equally frequent (13 vs. 14). The relevant MHC genotypes were evenly distributed between both groups and no significant difference in AIDS-free survival time was apparent between AG and GG females (log-rank χ2 = 0.155, 1 d.f., p = 0.694). Further analyses were not deemed meaningful in view of the small sample size.
TLR7 polymorphism at position −17 (c.-17C>T) influences set-point viral load and survival time in c.13G carriers
Polymorphisms close to the ATG start codon can affect translation initiation and thus the protein expression. We therefore examined the potential effect of the c.-17C>T polymorphism, located in exon 2, which is not in linkage disequilibrium with the polymorphisms in exon 3. Exon 2 of the TLR7 gene is remarkably conserved between human and rhesus macaque, and no polymorphisms have been detected in the respective human sequence so far. Since the polymorphisms at position −17 and +13 might both alter the local RNA secondary structure, we analyzed the phenotypic effect of the −17 polymorphism separately in both, c.13A and c.13G males. Among the 34 c.13A carriers, the c.-17C>T status was not significantly associated with either survival time, set-point viral load (both p>0.5) or peak viremia (p>0.1). From this analysis, only a single long-term non-progressing monkey carrying Mamu-A1*001 and -B*017 was excluded because all other MHC genotypes were almost identically distributed within the two c.13G>A genotype-defined groups (Figures 1C, 2B). Note, however, that only MHC adjusted monkeys are shown in Figures 1C and 2B in order to allow direct comparison with Figures 1B and 2A. MHC adjustment reduced the number of monkeys analysed but did not change the results of the statistical analysis.
In c.13G carriers, by contrast, the c.-17C>T polymorphism was significantly associated with survival time (log-rank χ2 = 5.66, 1 d.f., p = 0.017), but not with set-point or peak viral load (both p>0.5). Only MHC adjusted monkeys were included in this analysis because of the overrepresentation of slow progressor genotypes among c.-17C,c.13G carriers. Interestingly, survival time (p>0.6) and set-point viral load (p>0.1) did not differ significantly between c.13A and c.-17T,c.13G carriers, but between c.13A and c.-17C,c.13G carriers (survival time: log-rank χ2 = 6.21, 1 d.f., p = 0.013; set-point viral load: F = 8.10, p = 0.006). Thus, the inclusion of c.-17C>T genotype was capable of refining the genotype-phenotype association of c.13G>A. Eventually, only the c.-17C,c.13G haplotype turned to be significantly associated with prolonged survival and low set-point viral, compared to the other haplotypes (see Figures 1C, 2D).
No association of TLR7 polymorphisms with survival time and viral load in immunized SIV-infected macaques
Next, we analyzed the influence of the TLR7 SNPs upon AIDS-free survival and viral load in 72 immunized-infected male monkeys. These animals were analyzed separately because their survival time was significantly prolonged and their set-point viral load significantly reduced compared to untreated-infected macaques (both p<0.001). This difference remained significant even after MHC adjustment (data not shown). The macaques were immunized with different vaccine formulations, but the majority of the respective vaccination trials included adenovirus-based vectors [35], [36], [37], [38]. In all experiments, peak viremia was lower among immunized than among untreated-infected macaques. Reduced peak viremia has been found to be associated with prolonged survival and reduced set-point viral load before, probably reflecting the preservation of central memory CD4+ T cells [35], [39].
Owing to the relatively small number of immunized male monkeys available in this study, we did not initially perform any MHC adjustment although rapid progressors, as predicted by their MHC genotype, were underrepresented among c.13G carriers (2.3%) compared to c.13A carriers (11%). In the immunized-infected macaques, neither AIDS-free survival (log-rank χ2 = 0.33, 1 d.f., p = 0.564), nor set-point viral load (F = 0.23, p = 0.636) nor peak viremia (F = 0.32, p = 0.574) were found to be significantly associated with c.13G>A genotype. Subsequent MHC adjustment led to insignificant results as well (all p>0.3). Immunized macaques carrying TLR7 genotypes associated with more rapid disease progression in untreated -infected animals (i.e. c.13A and c.-17T,c.13G) did also not differ from carriers of c.-17C,c.13G (i.e., the haplotype associated with slow disease progression) neither in terms of survival time nor viral load,, and even after MHC adjustment (Figures 1D, 2C).
A trend towards prolonged survival was observed in immunized-infected female macaques compared to untreated-infected females (log-rank χ2 = 3.31, 1 d.f., p = 0.069). However, owing to the small sample size (n = 18), a detailed statistical analysis of the genotype-phenotype relationship was not considered warranted.
Prediction of RNA secondary structure
Since RNA secondary structure is a major determinant of translation kinetics and efficiency, we analysed the potential effects upon RNA secondary structure of the SNPs associated with AIDS-free survival. Of the three tightly linked SNPs in exon 3, only c.13G>A, but not the two other SNPs, was found to influence RNA secondary structure (CentroidFold program) or to change the minimum free energy by more than −2.1 kcal/mol (RNAfold, Vienna Websuite). For the signal peptide-encoding sequence (position 1–78), both prediction programs calculated an identical RNA secondary structure for the major c.13G allele, with a minimum free energy (MFE) of −13.8 kcal/mol (Fig. 3A). The c.13A allele disrupts this structure. The CentroidFold program predicted no strong secondary structure (MFE = 0 kcal/mol) at all for this sequence (Fig. 3B) whereas the Vienna RNA software predicted a secondary structure with an MFE of −8.9 kcal/mol. Despite some minor differences, both programs therefore suggested a destabilisation of the secondary structure of TLR7 signal peptide RNA sequence by the c.13A allele. The c.-17C>T polymorphism in exon 2 is located in a region critical for ribosomal entry and also influences RNA secondary structure. Both programs predict a stronger secondary structure for c.-17C compared to c.-17T containing RNA (MFE Centroidfold: −12.7 vs. 0 kcal/mol, RNAfold: −20.2 vs. −17.6 kcal/mol). Analyzing the effect of the c.-17C>T polymorphism in c.13A and c.13G carriers individually revealed that the folding effects of the two polymorphisms may be interdependent. Centroidfold predicted an MFE of −39.7 kcal/mol for the RNA sequence spanning exons 1 and 2 and the signal peptide-encoding region in the case of the c.-17C,c.13G haplotype whereas the structures of the other haplotypes had an MFE between −30.0 and −29.1 kcal/mol (Table 2). RNAfold predicted an MFE of −51.5 kcal/mol for the c.-17C,c.13G haplotype and of −49.7 to −47.0 kcal/mol for the other three haplotypes. Restriction of the in silico analysis to smaller RNA strands (25 bases up- and downstream of the c.-17 and c.13 position) yielded similar results (Table 2). Here, the c.-17C,c.13G haplotype also had the highest MFE , and the other haplotypes had a lower but relatively similar MFE. Despite some quantitative discrepancies, both programs therefore suggested (i) a joint effect of positions −17 and 13 on the RNA secondary structure and (ii) an RNA secondary structure that is most stable for the c.-17C,c.13G haplotype.
RNA secondary structure as predicted by CentroidFold [36] of rhesus c.13G (A), rhesus c.13A (B), human (C) signal peptide-encoding sequences, corresponding to amino acid positions 1–26. Each predicted base pair is coloured according to a gradient (from blue to red) corresponding to the base-pairing probability (from 0 to 1). The heat colour can be interpreted as a confidence measure of the predicted base-pairing. Red arrows indicate the polymorphic positions in the respective signal peptide DNA sequence. Plain RNA secondary structures as predicted by RNAfold [37] are shown for rhesus c.13A (D) and human (E) signal peptide-encoding sequences. Both programs predicted identical RNA secondary structures for rhesus c.13G. The minimal free energy (MFE) of each structure is given.
For comparison, we also analyzed the potential RNA secondary structure of the human TLR7 signal peptide-coding sequence. CentroidFold predicted an RNA secondary structure with an MFE of “0.0 kcal/mol” (Fig. 3C), whereas RNAfold predicted a structure with an MFE of −10.2 kcal/mol, indicating that the human TLR7 signal peptide-coding RNA structure is probably more similar to that of the rhesus c.13A allele. Both programs predicted that the Q11L (rs179008) polymorphism, which has been reported to influence viral load and time to AIDS in humans, does not influence the secondary structure of the human TLR7 signal peptide-coding RNA [31].
Finally, we investigated the influence, upon the potential RNA structure, of those SNPs that were not associated with AIDS-free survival in the initial sample of 36 macaques. The two remaining coding polymorphisms and three of the six polymorphisms in the 3′UTR did not change the predicted RNA secondary structure. The polymorphisms at position c.*236, c.*867 and the insertion/deletion c.*908_1911delGGCT changed the calculated MFE by more than 3 kcal/mol. None of polymorphisms in the 3′UTR was located close to a likely functional element [40]. Notably the polymorphisms at position c.*867 and c.*908_1911delGGCT flank the site of another functional polymorphism (TLR7 3′UTR SNP, rs3853839) reported in the human TLR7 gene [41].
Discussion
Prior to this study, no polymorphisms in the TLR7 gene locus of rhesus macaques were known. We therefore re-sequenced the TLR7 gene in 36 rhesus macaques of Indian origin and identified 12 polymorphisms. An initial analysis of 36 monkeys revealed three tightly linked SNPs to be associated with disease progression. Subsequent investigation of a larger sample of SIV-infected monkeys confirmed this initial result. The polymorphisms were significantly associated with viral load at set-point in untreated-SIV-infected male rhesus macaques and showed a trend towards an association with survival time even after the exclusion of monkeys carrying MHC alleles associated with rapid or slow disease progression. One of the polymorphisms of interest is located in the 3′ UTR region of the TLR7 mRNA, one is a silent nucleotide substitution located at amino acid position 570, and the remaining variant encodes a non-synonymous amino acid substitution (5VM) in the leader peptide. Formally, we could not distinguish which of these tightly linked polymorphisms may be responsible for the differential viral load. Furthermore, each of the three types of genetic variation seen in TLR7 is known to affect protein expression. Mutations in the 3′UTR , for instance, can influence mRNA stability [42], silent mutations may influence translation efficiency and protein folding [43], and variations in the signal peptide can effect the transport, localization and synthesis of membrane-bound proteins [44], [45], [46].
The c.13G>A (V5M) SNP in the signal peptide represents a conservative exchange in the hydrophilic N-terminal region of the signal peptide. Using the SignalIP program [47], we found no evidence that the polymorphism would greatly influence processing of the signal peptide, although this cannot be ruled out completely without additional experimental evidence. In contrast, the 11L (rs179008) substitution in the human TLR7 signal peptide is potentially capable of altering signal peptide function (see supplement ref [44]). More convincing results were obtained in an analysis of RNA secondary structure. The rhesus c.13A allele very likely destabilizes an RNA secondary structure predicted for the c.13G allele. Interestingly, a strong RNA secondary structure comparable to the rhesus wild-type sequence was not predicted for the human TLR7 signal peptide RNA sequence, and it is unlikely that the Q11L (rs179008) polymorphism will affect the secondary structure of the signal peptide-encoding RNA sequence. Furthermore, neither the silent mutation at amino acid position 570 nor the substitution in the 3′ UTR altered the potential secondary structure of the RNA around the respective SNPs. Based upon these predictions, the rhesus macaque c.13A allele is a functional candidate for the observed association with higher viral load in untreated-infected macaques [45], [46]. This result encouraged us to also investigate the influence of an adjacent polymorphism at position −17, which may affect the same local RNA structure and resides in a region critical for translation initiation. In silico analyses predicted that the effect of c.-17C>T upon RNA secondary structure was dependent upon the nucleotide present at position 13. Notably, c.-17C>T was associated with differential survival time only in c.13G carriers. Although translation represents a dynamic process that cannot be captured by mere sequence analyses, the coincidence of in silico prediction and experimental data therefore suggests that both SNPs exert a synergistic influence upon RNA structure.
It should be noted that, in whole genome screens involving human HIV-1-infected patients [48], [49], [50], the association between human TLR7 SNPs and disease progression did not attain statistical significance after Bonferroni correction. The validity of the respective associations has therefore been questioned [51]. While variation in the MHC region is undoubtedly the most important host factor for determining disease progression, however, it explains only a fraction of the variability in HIV viremia [48], [49], [50], [52]. The identification of additional host gene polymorphisms contributing weakly to the variability of viremia or time to treatment initiation will require much larger cohorts and, because of the variability of the virus, the exact size of meaningful cohorts of HIV-1-infected patients may not even be known. We therefore proposed cross-species comparisons as a valuable means to validate potential host gene polymorphisms influencing the course of HIV-infection [53]. This report is the third one describing similar effects of co-localising functional genetic polymorphisms in humans and rhesus macaques even although the affected biochemical pathways may differ between the two species [53], [54]. For TLR7, the rhesus substitution at position 13 and −17 are more likely to influence translation efficiency whereas the human Q11L variation probably influences transport or localization. At last, however, both variants would affect TLR7 expression.
Most interestingly, the TLR7 polymorphisms did not influence plasma viral set-point and survival time in immunized macaques. One explanation for this result could be that vaccination-induced immune reactions such as T and B cell response had a much stronger influence upon viral replication than differential TLR7 expression.
To better define the influence of TLR7 polymorphisms on vaccine efficacy, larger numbers of macaques treated with the same AIDS vaccine are required. Furthermore, it is paramount to investigate the exact contribution of the polymorphisms on TLR7 expression and cytokine secretion in vitro and in vivo, in untreated and infected, and in immunized subjects. Our work highlights the dual role of TLR7 in immunodeficiency virus infection and vaccination. Assuming that the c.13A and c.-17T,c.13G variants are associated with enhanced TLR7 expression, the data suggest that efficient triggering of TLR7 may improve AIDS vaccine efficacy while attenuating the action of TLR7 may decelerate disease progression. Finally, our results imply that it may be important to control human vaccine trials, not only for MHC genotype, but also for TLR7 genotype.
Materials and Methods
Animals and laboratory parameters
We retrospectively analyzed genetic and phenotypic data from 237 SIV-infected rhesus macaques (Macaca mulatta) of Indian origin, seronegative for STLV-1 and D-type virus. These animals were housed and treated at the German Primate Research Centre (‘Deutsches Primatenzentrum’, DPZ). All relevant protocols of the DPZ comply with the German Animal Protection Act, which in turn follows European Union guidelines on the use of non-human primates for biomedical research. This includes a 12∶12 light dark schedule, provision of dry food supplemented with fresh fruit twice a day and constant water access. The monkeys were kept under permanent medical care. In cases of suffering monkeys were humanely killed. All experiments were approved by the independent governmental veterinary authority of Lower Saxony. Major characteristics of the monkeys under study and the algorithm used to define AIDS-like disease in these animals have been described before [53]. For the majority of samples viral RNA was isolated from frozen plasma samples following the MagAttract Virus Mini M48 protocol (Qiagen, Hilden, Germany). Purified SIV RNA was quantified using TaqMan-based real-time PCR on an ABI-Prism 7500 sequence detection system (Applied Biosystems) as described [55] using gag forward (5′-ACCCAGTACAACAAATAGGTGGTAACT-3′), gag reverse primer (5′-TCAATTTTACCCAGGCATTTAATGT-3′) and a fluorescent labeled probe (5′-6FAM(6-carboxyfluorescein)-TGTCCACCTGCCATTAAGCCCGAG-TAMRA(6-carboxytetramethylrhodamine-3′). Amplified viral RNA was expressed as SIV-RNA copies per mL plasma. Samples from animals infected before 2004 were analyzed using another technique [56]. Peak and set-point viremia were quantified by the RNA viral copy number at weeks 2 and 20 post infection, respectively. If in addition data at week 24 post infection were available, set-point viremia was equaled to the mean copy number at weeks 20 and 24. Note that, for some animals included in the survival analysis, no viral load data were available. Macaques were designated ‘untreated’ if they had not received any AIDS vaccine.
MHC-genotyping
Monkeys were genotyped for the MHC as described before and classified as either ‘slow progressor’, ‘rapid progressor’ or ‘unspecified’ [34]. Survival curves used as a guide for this ‘MHC adjustment’ of the subsequent statistical analyses are shown in Figure S3.
DNA sequence analysis of TLR7
Genomic DNA was isolated from peripheral blood or from Herpes papio virus-transformed B-cell lines using the Genomic DNA Isolation Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The rhesus TLR7 gene was screened for polymorphisms by genomic re-sequencing of all three exons and parts of the flanking intron sequence in 36 monkeys (chrX:10564627–10566043; chrX:10582811–10589400 in the former version, rhemac2, still available at http://genome.ucsc.edu/cgi-bin/hgGateway). The primers used for nested PCR were designed using Primer3Plus [57] and are listed in Table S1. Between 25 and 50 ng of genomic DNA was used to generate the PCR template. Sequencing was performed using ABI Big Dye chemistry and the ABI3730xl Analyzer (ABI, Foster City, Calif., USA). Overlapping amplicons were sequenced to ensure that no polymorphisms in the primer binding sites were missed. Polymorphisms were searched for with the GAP software (version 4.11 of the Staden package) using the implemented SNP candidate viewer, and by visual inspection. The DNA sequences of the re-sequenced genomic regions and the polymorphisms identified in these regions are deposited in GenBank (accession number JF691587).
Genotyping of TLR7
Three polymorphisms in the TLR7 gene were initially genotyped by means of Sanger sequencing. For amplifying the TLR7 c.13G>A substitution, we used primers TLR7-13-for (5′GGAAAATGCTGCTTCTACCATC) and TLR7-13-rev (5′ATGGACCAGTCTGTGAAAGGA); the silent c.1710T>A substitution was amplified with primers TLR7-1710-for (5′AAGTATGGGCAGACCTTGGA) and TLR7-1710-rev (5′CAAAAACTCCAGAAGGCAAGA). For typing 3′ UTR polymorphism c.*1604C>T, we used primers TLR7-1604-for (5′TCAAAGATTGAAACCTGACCAA) and TLR7-1604-rev (5′CAAACCACCGTACTTAACCTCA). PCR was performed using 150 ng genomic DNA and included an initial denaturation step at 94°C for 5 min, followed by 30 cycles with denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and elongation at 72°C. PCR products were either precipitated with ethanol or gel-purified using standard techniques (Qiagen, Hilden, Germany). DNA sequencing was performed using dye terminator chemistry and ABI 3730/3700 technology. DNA sequences were aligned using the CLUSTALW function of the Bioedit freeware (http://www.mbio.ncsu.edu/bioedit/bioedit.html).
Alternatively genotyping was performed by means of a TaqMan-based allelic discrimination assay (ABI7500 Real Time PCR System). The reaction volume of 25 µl contained 12.5 µl ImmoMix™ (Bioline, Luckenwalde, Germany), 660 nM ROX (Invitrogen, USA) as an internal standard, 90 nM of the respective primers, 20–24 nM labelled probes and 30 ng DNA. All samples were run in triplicates. PCR was performed with an initial denaturation step at 95°C for 10 min, a pre-read step at 60°C for 1 min, followed by denaturation at 92°C for 16 s and annealing/amplification for 1 min at the indicated temperature for 40 cycles and a post read step for 1 min in at 60°C. Primers and probes for detecting the c.13G>A polymorphism were: 13-forward (5′AAAGAGAGGCAGCAAATGGGA), 13-reverse (5′AACCATCTAGCCCCAAGGAGTT), and probes FAM-5′-TTTCCAGTGTGGACACTGAAGA-3′-BBQ (20 nM) and YAK-5′-TTTCCAATGTGGACACTGAAGA-3′-BBQ (24 nM, annealing and amplification at 57°C). Oligonucleotides for detecting the c.1710T>A polymorphism were: 1710-forward (5′CAACAACCGGCTTGATTTACTC), 1710-reverse (5′GCTGGTGGAGGAAGAGATGTCA3′), and probes: FAM-5′-TTGGAAGTTCTGGATATAAGCAGT-3′-BBQ (20 nM) and YAK-5′-AATTGGAAGTACTGGATATAAGCAGT-BBQ (24 nM, annealing/amplification at 60°C). For identification of the polymorphism at position c.*1604, we used 1604-forward (5′GCAGTGCCAAAGGCTTTACTT), 1604-reverse (5′GATTAATTTGGGAATTTGTGACCCT), and the LNA-probes FAM-5′-TGAGTACACACTAAATGAATT-3′-BBQ (24 nm), YAK-5′-TGTGAGTATACACTAAATGAAT-3′-BBQ (24 nmol, annealing/amplification at 59°C). The c.-17C>T polymorphism was typed using −17-forward (5′ATCTCAAGCTGATCTTGGCACC), −17 reverse (5′GGTTTCAGTTGTTTCCACAAAGATT), and LNA-probes FAM-5′TCCAAAATGGAATGTAGAGGTCTG-3′-BBQ, YAK-5′-TCCAGAATGGAATGTAGAGGTCTG-3′-BBQ (20 nM each, annealing/amplification at 60°C). Primers and probes were designed and synthesized by TIPMolbiol, Berlin, Germany. Allele calling was performed on the basis of the amplification and post-amplification plate reads.
Statistical analysis
Survival analysis was performed using the LIFETEST and LIFEREG procedures of the SAS software package (SAS version 9.2, SAS Inc., Cary, NC). Differences between genotype-specific survival curves were assessed for statistical significance using a log-rank test with the appropriate degrees of freedom. Unbalanced analysis of variance (ANOVA) of log-transformed viral load data was performed using SAS procedure GLM. Kaplan-Meier survival curves were prepared with SigmaPlot 8.0 (IBM SPPS Inc. Somers, NY). All other statistical analyses were carried out using the FREQ procedure of SAS.
Analysis of RNA secondary structure
Two online prediction tools were employed for RNA secondary structure analysis, namely the Contrafold option in CentroidFold [58] and RNAfold from the Vienna RNA Websuite [59]. Secondary structure prediction of the region encoding the signal peptide covered nucleotide positions 1 to 78 while positions −136 to 78 were involved in the analysis of the c.-17C>T polymorphism. For the remaining two polymorphisms, 30 base pairs both upstream and downstream of the site of interest were included.
Supporting Information
Figure S1.
Alignment of rhesus and human TLR7 gene DNA sequences. The rhesus genomic DNA and refseq mRNA (acc. no. NM_001130426) sequences were taken from the USCS genome browser (http://genome.ucsc.edu/). Untranslated regions are still hypothetical. Coding sequences are shown in black, untranslated regions are given in purple. Polymorphic positions in the rhesus sequence are marked.
https://doi.org/10.1371/journal.pone.0025474.s001
(DOC)
Figure S2.
Alignment of rhesus and human TLR7 amino acid sequence. Amino acid exchanges between human and rhesus are shown in bold, identity between deduced amino acid residues is depicted by a dot. Signal peptide, transmembrane and TIR domain are marked by red letters. Polymorphic sites in the rhesus TLR7 sequence are annotated (green for synonymous, blue for non-synonymous nucleotide exchanges).
https://doi.org/10.1371/journal.pone.0025474.s002
(DOC)
Figure S3.
Kaplan Meier Survival curves of SIV-infected male rhesus macaques of Indian origin stratified by MHC genotype. ‘MHC adjusted’ refers to the exclusion of animals carrying either Mamu-A1*001, B*008 or an MHC genotype known to be associated with rapid disease progression. A: untreated SIV-infected macaques (p<0.01). B: immunized SIV-infected macaques (p<0.01).
https://doi.org/10.1371/journal.pone.0025474.s003
(TIF)
Table S1.
Primers used for PCR amplification of the rhesus macaque TLR7 gene region.
https://doi.org/10.1371/journal.pone.0025474.s004
(DOC)
Acknowledgments
We thank Ivonne Heinze, Nicole Leuchte and Heidi Meyer for expert technical assistance, and Monika Franz for support in the animal experiments. We especially thank Christiane Stahl-Hennig for her continuous support of our work and for providing access to data from the experimental animals that were generated under her supervision.
Author Contributions
Conceived and designed the experiments: MP RS US. Performed the experiments: RS US. Analyzed the data: MK RS US. Wrote the paper: MK US.
References
- 1. Diebold SS, Kaisho T, Hemmi H, Akira S, Reis e Sousa C (2004) Innate antiviral responses by means of TLR7-mediated recognition of single-stranded RNA. Science 303: 1529–1531.
- 2. Heil F, Hemmi H, Hochrein H, Ampenberger F, Kirschning C, et al. (2004) Species-specific recognition of single-stranded RNA via toll-like receptor 7 and 8. Science 303: 1526–1529.
- 3. Lund JM, Alexopoulou L, Sato A, Karow M, Adams NC, et al. (2004) Recognition of single-stranded RNA viruses by Toll-like receptor 7. Proc Natl Acad Sci U S A 101: 5598–5603.
- 4. Hornung V, Guenthner-Biller M, Bourquin C, Ablasser A, Schlee M, et al. (2005) Sequence-specific potent induction of IFN-alpha by short interfering RNA in plasmacytoid dendritic cells through TLR7. Nat Med 11: 263–270.
- 5. Xagorari A, Chlichlia K (2008) Toll-like receptors and viruses: induction of innate antiviral immune responses. Open Microbiol J 2: 49–59.
- 6. Uematsu S, Akira S (2007) Toll-like receptors and Type I interferons. J Biol Chem 282: 15319–15323.
- 7. Kawai T, Akira S (2010) The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors. Nat Immunol 11: 373–384.
- 8. Zhang SY, Jouanguy E, Sancho-Shimizu V, von Bernuth H, Yang K, et al. (2007) Human Toll-like receptor-dependent induction of interferons in protective immunity to viruses. Immunol Rev 220: 225–236.
- 9. Gibson SJ, Lindh JM, Riter TR, Gleason RM, Rogers LM, et al. (2002) Plasmacytoid dendritic cells produce cytokines and mature in response to the TLR7 agonists, imiquimod and resiquimod. Cell Immunol 218: 74–86.
- 10. Ortiz M, Guex N, Patin E, Martin O, Xenarios I, et al. (2009) Evolutionary trajectories of primate genes involved in HIV pathogenesis. Mol Biol Evol 26: 2865–2875.
- 11. Wille-Reece U, Flynn BJ, Lore K, Koup RA, Kedl RM, et al. (2005) HIV Gag protein conjugated to a Toll-like receptor 7/8 agonist improves the magnitude and quality of Th1 and CD8+ T cell responses in nonhuman primates. Proc Natl Acad Sci U S A 102: 15190–15194.
- 12. Wille-Reece U, Flynn BJ, Lore K, Koup RA, Miles AP, et al. (2006) Toll-like receptor agonists influence the magnitude and quality of memory T cell responses after prime-boost immunization in nonhuman primates. J Exp Med 203: 1249–1258.
- 13. Rajagopal D, Paturel C, Morel Y, Uematsu S, Akira S, et al. (2010) Plasmacytoid dendritic cell-derived type I interferon is crucial for the adjuvant activity of Toll-like receptor 7 agonists. Blood 115: 1949–1957.
- 14. Geeraedts F, Goutagny N, Hornung V, Severa M, de Haan A, et al. (2008) Superior immunogenicity of inactivated whole virus H5N1 influenza vaccine is primarily controlled by Toll-like receptor signalling. PLoS Pathog 4: e1000138.
- 15. Bessa J, Jegerlehner A, Hinton HJ, Pumpens P, Saudan P, et al. (2009) Alveolar macrophages and lung dendritic cells sense RNA and drive mucosal IgA responses. J Immunol 183: 3788–3799.
- 16. Schlaepfer E, Audige A, Joller H, Speck RF (2006) TLR7/8 triggering exerts opposing effects in acute versus latent HIV infection. J Immunol 176: 2888–2895.
- 17. Lore K, Betts MR, Brenchley JM, Kuruppu J, Khojasteh S, et al. (2003) Toll-like receptor ligands modulate dendritic cells to augment cytomegalovirus- and HIV-1-specific T cell responses. J Immunol 171: 4320–4328.
- 18. Schlaepfer E, Speck RF (2008) Anti-HIV activity mediated by natural killer and CD8+ cells after toll-like receptor 7/8 triggering. PLoS One 3: e1999.
- 19. Goujon C, Malim MH (2010) Characterization of the alpha interferon-induced postentry block to HIV-1 infection in primary human macrophages and T cells. J Virol 84: 9254–9266.
- 20. Brassard DL, Grace MJ, Bordens RW (2002) Interferon-alpha as an immunotherapeutic protein. J Leukoc Biol 71: 565–581.
- 21. Tavel JA, Huang CY, Shen J, Metcalf JA, Dewar R, et al. (2010) Interferon-alpha produces significant decreases in HIV load. J Interferon Cytokine Res 30: 461–464.
- 22. Yamada O, Hattori N, Kurimura T, Kita M, Kishida T (1988) Inhibition of growth of HIV by human natural interferon in vitro. AIDS Res Hum Retroviruses 4: 287–294.
- 23. Adalid-Peralta L, Godot V, Colin C, Krzysiek R, Tran T, et al. (2008) Stimulation of the primary anti-HIV antibody response by IFN-alpha in patients with acute HIV-1 infection. J Leukoc Biol 83: 1060–1067.
- 24. Meier A, Chang JJ, Chan ES, Pollard RB, Sidhu HK, et al. (2009) Sex differences in the Toll-like receptor-mediated response of plasmacytoid dendritic cells to HIV-1. Nat Med 15: 955–959.
- 25. Mandl JN, Barry AP, Vanderford TH, Kozyr N, Chavan R, et al. (2008) Divergent TLR7 and TLR9 signaling and type I interferon production distinguish pathogenic and nonpathogenic AIDS virus infections. Nat Med 14: 1077–1087.
- 26. Baenziger S, Heikenwalder M, Johansen P, Schlaepfer E, Hofer U, et al. (2009) Triggering TLR7 in mice induces immune activation and lymphoid system disruption, resembling HIV-mediated pathology. Blood 113: 377–388.
- 27. Johnson TR, Rao S, Seder RA, Chen M, Graham BS (2009) TLR9 agonist, but not TLR7/8, functions as an adjuvant to diminish FI-RSV vaccine-enhanced disease, while either agonist used as therapy during primary RSV infection increases disease severity. Vaccine 27: 3045–3052.
- 28. Berghofer B, Frommer T, Haley G, Fink L, Bein G, et al. (2006) TLR7 ligands induce higher IFN-alpha production in females. J Immunol 177: 2088–2096.
- 29. Moller-Larsen S, Nyegaard M, Haagerup A, Vestbo J, Kruse TA, et al. (2008) Association analysis identifies TLR7 and TLR8 as novel risk genes in asthma and related disorders. Thorax 63: 1064–1069.
- 30. Schott E, Witt H, Neumann K, Bergk A, Halangk J, et al. (2008) Association of TLR7 single nucleotide polymorphisms with chronic HCV-infection and response to interferon-a-based therapy. J Viral Hepat 15: 71–78.
- 31. Oh DY, Baumann K, Hamouda O, Eckert JK, Neumann K, et al. (2009) A frequent functional toll-like receptor 7 polymorphism is associated with accelerated HIV-1 disease progression. Aids 23: 297–307.
- 32. Ketloy C, Engering A, Srichairatanakul U, Limsalakpetch A, Yongvanitchit K, et al. (2008) Expression and function of Toll-like receptors on dendritic cells and other antigen presenting cells from non-human primates. Vet Immunol Immunopathol 125: 18–30.
- 33. den Dunnen JT, Antonarakis SE (2000) Mutation nomenclature extensions and suggestions to describe complex mutations: a discussion. Hum Mutat 15: 7–12.
- 34. Sauermann U, Siddiqui R, Suh YS, Platzer M, Leuchte N, et al. (2008) Mhc class I haplotypes associated with survival time in simian immunodeficiency virus (SIV)-infected rhesus macaques. Genes Immun 9: 69–80.
- 35. Schulte R, Suh YS, Sauermann U, Ochieng W, Sopper S, et al. (2009) Mucosal prior to systemic application of recombinant adenovirus boosting is more immunogenic than systemic application twice but confers similar protection against SIV-challenge in DNA vaccine-primed macaques. Virology 383: 300–309.
- 36. Stolte-Leeb N, Bieler K, Kostler J, Heeney J, Haaft PT, et al. (2008) Better protective effects in rhesus macaques by combining systemic and mucosal application of a dual component vector vaccine after rectal SHIV89.6P challenge compared to systemic vaccination alone. Viral Immunol 21: 235–246.
- 37. Suh YS, Park KS, Sauermann U, Franz M, Norley S, et al. (2006) Reduction of viral loads by multigenic DNA priming and adenovirus boosting in the SIVmac-macaque model. Vaccine 24: 1811–1820.
- 38. Suh YS, Park KS, Sauermann U, Kim KS, Ahn SS, et al. (2008) Prolonged survival of vaccinated macaques after oral SIVmac239 challenge regardless of viremia control in the chronic phase. Vaccine 26: 6690–6698.
- 39. Mattapallil JJ, Douek DC, Buckler-White A, Montefiori D, Letvin NL, et al. (2006) Vaccination preserves CD4 memory T cells during acute simian immunodeficiency virus challenge. J Exp Med 203: 1533–1541.
- 40. Grillo G, Turi A, Licciulli F, Mignone F, Liuni S, et al. (2010) UTRdb and UTRsite (RELEASE 2010): a collection of sequences and regulatory motifs of the untranslated regions of eukaryotic mRNAs. Nucleic Acids Res 38: D75–80.
- 41. Shen N, Fu Q, Deng Y, Qian X, Zhao J, et al. (2010) Sex-specific association of X-linked Toll-like receptor 7 (TLR7) with male systemic lupus erythematosus. Proc Natl Acad Sci U S A 107: 15838–15843.
- 42. Chen JM, Ferec C, Cooper DN (2006) A systematic analysis of disease-associated variants in the 3′ regulatory regions of human protein-coding genes I: general principles and overview. Hum Genet 120: 1–21.
- 43. Kimchi-Sarfaty C, Oh JM, Kim IW, Sauna ZE, Calcagno AM, et al. (2007) A “silent” polymorphism in the MDR1 gene changes substrate specificity. Science 315: 525–528.
- 44. Jarjanazi H, Savas S, Pabalan N, Dennis JW, Ozcelik H (2008) Biological implications of SNPs in signal peptide domains of human proteins. Proteins 70: 394–403.
- 45. Mohren S, Weiskirchen R (2009) Non-synonymous gene polymorphisms in the secretory signal peptide of human TGF-beta1 affect cellular synthesis but not secretion of TGF-beta1. Biochem Biophys Res Commun 379: 1015–1020.
- 46. Kertesz M, Wan Y, Mazor E, Rinn JL, Nutter RC, et al. (2010) Genome-wide measurement of RNA secondary structure in yeast. Nature 467: 103–107.
- 47. Emanuelsson O, Brunak S, von Heijne G, Nielsen H (2007) Locating proteins in the cell using TargetP, SignalP and related tools. Nat Protoc 2: 953–971.
- 48. Fellay J, Ge D, Shianna KV, Colombo S, Ledergerber B, et al. (2009) Common genetic variation and the control of HIV-1 in humans. PLoS Genet 5: e1000791.
- 49. Fellay J, Shianna KV, Ge D, Colombo S, Ledergerber B, et al. (2007) A whole-genome association study of major determinants for host control of HIV-1. Science 317: 944–947.
- 50. Limou S, Le Clerc S, Coulonges C, Carpentier W, Dina C, et al. (2009) Genomewide association study of an AIDS-nonprogression cohort emphasizes the role played by HLA genes (ANRS Genomewide Association Study 02). J Infect Dis 199: 419–426.
- 51. McMichael A (2010) Host Genetics and Viral Diversity: Report from a Global HIV Vaccine Enterprise Working Group. Nature Preceedings.
- 52. Emu B, Sinclair E, Hatano H, Ferre A, Shacklett B, et al. (2008) HLA class I-restricted T-cell responses may contribute to the control of human immunodeficiency virus infection, but such responses are not always necessary for long-term virus control. J Virol 82: 5398–5407.
- 53. Siddiqui RA, Sauermann U, Altmuller J, Fritzer E, Nothnagel M, et al. (2009) X chromosomal variation is associated with slow progression to AIDS in HIV-1-infected women. Am J Hum Genet 85: 228–239.
- 54. Francis PJ, Appukuttan B, Simmons E, Landauer N, Stoddard J, et al. (2008) Rhesus monkeys and humans share common susceptibility genes for age-related macular disease. Hum Mol Genet 17: 2673–2680.
- 55. Negri DR, Baroncelli S, Catone S, Comini A, Michelini Z, et al. (2004) Protective efficacy of a multicomponent vector vaccine in cynomolgus monkeys after intrarectal simian immunodeficiency virus challenge. J Gen Virol 85: 1191–1201.
- 56. Ten Haaft P, Verstrepen B, Uberla K, Rosenwirth B, Heeney J (1998) A pathogenic threshold of virus load defined in simian immunodeficiency virus- or simian-human immunodeficiency virus-infected macaques. J Virol 72: 10281–10285.
- 57. Untergasser A, Nijveen H, Rao X, Bisseling T, Geurts R, et al. (2007) Primer3Plus, an enhanced web interface to Primer3. Nucleic Acids Res 35: W71–74.
- 58. Sato K, Hamada M, Asai K, Mituyama T (2009) CENTROIDFOLD: a web server for RNA secondary structure prediction. Nucleic Acids Res 37: W277–280.
- 59. Gruber AR, Lorenz R, Bernhart SH, Neubock R, Hofacker IL (2008) The Vienna RNA websuite. Nucleic Acids Res 36: W70–74.