An estimated 800,000 children are born each year with HIV-1 infection contracted from the mother. 1 This transmission of HIV-1 occurs during pregnancy, around the time of labor and delivery, and postnatally through breast-feeding. 2 Transmission rates in the absence of one or more interventions to prevent mother-to-child transmission range from 14–32% in non–breast-feeding populations and from 25–48% in breast-feeding populations. 3
Unique genetic and immunologic events distinguish antepartum transmission from subsequent transmission of HIV-1, since antepartum transmission occurs in a setting where the child shares at least half of his or her major histocompatibility (MHC) genes with the mother, and while the mother tolerates the paternally derived fetal histocompatibility molecules.
Antepartum HIV-1 transmission is potentially related to HLA class I alleles of the MHC because of their role in both determining maternal-infant compatibility and in regulating the CD8 T-cell surveillance of virally infected cells that results in either killing of virus-infected cells or the production of factors that interfere with viral infectivity and replication. 4,5 Through the avidity of the pockets in different allelic MHC molecules for particular amino acid side chains, the binding properties of MHC molecules encoded by these alleles, first, determine the individual's somatically generated repertoire of CD8 T cells by binding different self-peptides during thymic selection and, second, specify the particular viral peptides to be bound and presented to the CD8 T cells during an adaptive immune response. The importance of HLA class I alleles in defining an individual's response to HIV-1 infection is emphasized by the association of rapid HIV-1 disease progression with the HLA-B*35 family of alleles. 6–8 Conversely, HLA-B27 and HLA-B57 are associated with long-term nonprogression 7–9 apparently by binding a much broader repertoire of peptides containing immunodominant peptides derived from critical regions in the viral genome. 9,10 Interestingly, more recent molecular subtyping of the HLA class I alleles revealed that the phenotype of rapid progression to AIDS is not conferred by all alleles in the HLA-B*35 family. The prototypic HLA-B*35 allele, HLA-B*3501, was found not to be associated with rapid progression of HIV-1 infection, but rather the association with rapid progression is primarily with several of the less common HLA-B35 alleles, eg, HLA-B*3502 and B*3503 as well as with the structurally related allele HLA-B*5301. 11,12 The narrow spectrum of peptide binding of molecules encoded by these alleles 13–15 that includes few peptide motifs of HIV-1 proteins 10 suggests that the rapid progression of HIV-1 infection reflects an escape from the reduced repertoire of T cells able to recognize the small number of presented HIV-1 peptides. 9,10,16–18 Because the mothers' MHC class I alleles define a nearly unique environment of adaptive immune recognition in which the HIV-1 infection evolves, and the viruses that are transmitted to the infant have already evolved in and evaded the maternal immune response environment defined by some of the same class I HLA histocompatibility genes inherited by the infant, 19 it is possible that different HLA-B alleles influence transmission by this process.
A second way HLA class I molecules could be relevant to antepartum HIV-1 transmission involves their interaction with 2 sets of stereotyped natural killer (NK) receptors. NK receptors are a part of the innate immune response that are triggered by alterations, particularly decreases, in the expression of HLA class I molecules, the “missing self” mechanism. 20 This system mediates the inhibitory response that appears to be a critical mechanism in inducing maternal tolerance of the paternally derived fetal major histocompatibility molecules of the child, which if engrafted in different circumstances would result in rejection. 21–23 While the mechanism of this tolerance is not fully understood, it is mediated by the massive placental accumulation of NK and CD8 T cells that express NK receptors. 21–23
One family of stereotyped NK receptors, the killer inhibitor receptors (KIR), is a molecularly and genetically intricate system 24 that interacts with various polymorphic regions of HLA molecules not directly involved in peptide-binding, although the nature of many of the KIR ligands remains to be determined. The varied expression of KIR family members on different NK cells and CD8 T lymphocytes influenced by binding to MHC polymorphisms creates a somatically selected repertoire that recognizes most allelic MHC class I molecules. 25 HLA-B molecules encode a sequence motif that regulates allele-specific interaction with KIR3DL1, an inhibitory member of the KIR family. Cells infected with viruses that avoid immunosurveillance by selectively downregulating the expression of HLA-B 5 are killed by NK cells upon the diminution of tonic inhibitory signaling through their inhibitory NKR, such as KIR3DL1, following decreased HLA-B expression. NK cells have been implicated in the control of HIV-1 infection in adults by studies on HLA-B alleles. 26,27 Moreover, activated NK cells directly suppress the replication of HIV-1 by CC chemokine secretion. 28 NK receptors also relate directly to the adaptive immune response because KIRs are expressed on CD8 T cells. 29,30 Here, decreased HLA-B expression on infected cells removes the inhibitory signal transduced by KIR3DL1 and lowers the threshold for CD8 T-cell activation by viral peptides, enhancing the adaptive CD8 T-cell anti-HIV-1 response.
The consequences of both aspects of polymorphism of the HLA-B molecules, that related to peptide binding and that interacting with the KIR receptors, are both potentially relevant to mother-to-child transmission of HIV-1, but each has completely different implications in terms of the particular alleles that would be associated with transmission. The first hypothesis is that HLA alleles associated with more rapid disease progression among HIV-1-infected individuals would be associated with maternal HIV-1 transmission, and the HLA alleles of an infected infant would resemble those of the mother. A second hypothesis is that the interaction of HLA with the NK receptors of the innate immune system receptors importantly determines whether transmission to the infant occurred. Here, one would expect that the particular HLA-B alleles of the mother encoding KIR3DL1 ligands in triggering an NK response would be of primary importance, and the alleles of the infant would be largely irrelevant. We tested these 2 alternative hypotheses by using sequence-based DNA typing of cryo-preserved lymphocytes from HIV-1-infected women and their children enrolled in the Women and Infants Transmission Study (WITS) 31 to determine the influence of particular HLA-B alleles on the occurrence of HIV-1 transmission.
WITS is a multicenter, prospective cohort study. 31 The study population was composed of HIV-1-infected women enrolled in WITS who delivered between 1990–1993 and from whom cryopreserved lymphocytes were available (along with such lymphocytes from their children). Eligible cases were all those women who transmitted HIV-1 to their children, and eligible controls were those women who did not transmit. HIV-1-infected children had at least 2 positive HIV-1 cultures. HIV-1-uninfected children had no positive HIV-1 cultures and had at least 2 negative HIV-1 cultures at 1 month and at or after 6 months of age. For every case, 2 controls were randomly selected and were frequency-matched to the cases on year of delivery, study site, and ethnicity.
Maternal viral loads were determined, as described. 32 DNA isolation and sequence-based HLA-B typing was performed with a sequence-based method as described 33 and modified by using 5′YGTCGCCGBGGTCCCAGTTC-TAAAG and 3′GGAGGCCATCCCCGGCGACCTAT in-tronic primers, followed by exon-specific sequencing in both forward and reverse directions. Mothers and children typing for only 1 allele were considered homozygous.
Allele frequencies were compared by Fisher exact test uncorrected for multiple comparisons. Interactions were determined by the Breslow-Day homogeneity test. In the logistic regression, the association between each allele and transmission was adjusted for log10 plasma viral load, infant birth weight, and receipt (or lack thereof) of zidovudine monotherapy. HIV-1 RNA < 400 copies/mL was assigned a value of 400 copies/mL for statistical analysis. We performed a permutation test to assess whether the observed differences in distribution of HLA-B alleles between cases and controls was greater than might be expected due to chance alone (by testing the null hypothesis that there is no relationship between maternal HLA-B and transmission). A measure of overall imbalance between cases and controls with respect to HLA-B was defined by constructing a 2 × 2 table for each 2-digit allele and transmission status, calculating the χ2 statistic for each table, and summing them up. Then we created 10,000 simulated data sets by randomly permuting the cases/control status of the mothers in our data set and compared the imbalance statistic calculated from the observed data with the distribution of imbalance measures from our simulated data sets to see the degree to which the observed imbalance was extreme relative to that distribution.
The successfully typed study population comprised 83 transmitting mother/infant pairs and 163 matched mother/infant pairs. Cases and controls were similar in terms of all characteristics examined, except children born to cases were more likely to be low birth weight (Table 1). High maternal plasma viral loads (>10,000 copies/mL) were somewhat more common among transmitting mothers (cases) (67%) than in non-transmitting, control mothers (55%, P = 0.07). The frequency of CCR5Δ32, a variant chemokine receptor allele with a confirmed association with progression to AIDS, did not differ between cases and controls.
Mothers transmitting HIV-1 to their infants had significantly increased frequencies of HLA-B*1302, B*3501, B*3503, B*4402, and B*5001, both overall and among subsets of the population (African American and Hispanic mothers, and African American mothers alone) (Table 2). Grouping of separately identified alleles was performed to gain insight into the biologic effects and whether the alleles were operating in the same or separate individuals. More transmitting mothers had HLA-B*3501 or B*3503 than nontransmitters (18 vs. 4%; odds ratio [OR] = 4.9, P = 0.0006). The association of HLA-B*4402 with increased transmission was significant among African American and Hispanic mothers, while HLA-B*1302 was associated with increased transmission overall (OR = 6.3) and also among African American and Hispanic mothers (OR = 8.6). HLA-B*5001, found predominantly among Hispanic mothers (5%), was associated with increased transmission among this ethnic group (OR = 9.2). Overall, we detected B*3501, *3503, *1302, *4402, and *5001 in 22% of the study population. These alleles accounted for 40% of cases and 13% of controls (OR = 4.5). Among African Americans and Hispanics, the presence of these alleles increased the odds of transmission 7-fold.
To identify additional HLA-B associations, mothers with HLA-B*35 alleles were eliminated from the analysis, revealing that B*5301 was significantly associated with decreased transmission (Table 2). Grouping of B*4901 and B*5301 demonstrated that the effect of each allele was separately associated with decreased transmission (Table 2). Similarly, the association with increased transmission was observed in separate individuals with alleles HLA-B*1302, *4402, and *5001. A permutation test performed to address the issue of multiple comparisons showed that the observed difference in HLA-B alleles between transmitting and nontransmitting mothers would only occur 0.43% of the time by chance (P = 0.0043). A complete list of the results for all maternal alleles is shown in Table 3.
Analysis of the infant's HLA-B locus alleles revealed that only HLA-B*1402 was marginally associated with increased maternal-infant transmission (P = 0.05) (Table 3). None of the alleles found to be relevant to transmission in the mother had a detectable effect on transmission when they occurred in the infant (Table 3).
Although higher maternal viral load is associated with a greater risk of HIV-1 transmission, 32 upon stratifying by maternal viral load (Table 4), the association of HLA-B*3501 or B*3503 with increased transmission was found primarily among cases with low viral loads (<10,000 copies/mL; median = 3,595 copies/mL, OR = 25.2, P = 0.0001). Overall, 48% of cases with low viral loads had 1 of the 5 susceptibility alleles, compared with 8% of controls (OR = 10.4, P = 0.0001). Reciprocally, 25% of controls with high viral loads had B*4901 and B*5301, compared with 5% of cases (P = 0.003).
Logistic regression models confirmed that transmission was associated with the HLA-B alleles identified by univariate analysis (Table 5). Notably, among mothers with low viral loads, only maternal HLA B*3501 remained associated with transmission (OR = 18.6), while in mothers with high viral loads B*5301 was independently associated with a lower transmission risk (P = 0.04) (Table 5). Using the approach of MacDonald et al, 34 logistic regression analyses indicated no significant association of HIV-1 transmission with sharing of HLA-B alleles between mother and infant (data not shown).
Of the 83 cases, 63 could be classified 35 according to the presumed timing of transmission (early transmission [in utero]: 10 cases, vs. late transmission [at the end of pregnancy or during parturition]: 53 cases). Although the proportions of HLA-B*3501 and B*3503 were equal in early vs. late transmitting cases (Table 6), 5 of the 10 cases of early transmission bore either HLA-B*1302 or B*4402, while 8% of controls and 19% of cases with late transmission had either of these 2 alleles (P = 0.002 ), suggesting an association between HLA-B*1302 or B*4402 with early transmission.
Vertical transmission of HIV-1 during pregnancy likely involves several immune mechanisms operating at different levels. Our findings suggest that, in addition to previously identified risk factors for vertical transmission of HIV-1, there is a novel and important role for the innate immune recognition of HLA-B molecules in the mother. Transmitting mothers were more likely to have certain HLA-B alleles (B*3501, B*3503, B*1302, B*4402, and B*5001), overall, in specific subsets of the population (African American and Hispanic mothers), or both, although it should be cautioned that the approach of grouping alleles can lead to somewhat inflated probability values. Reciprocally, B*4901 and B*5301 were each associated with a lower risk of mother-to-child transmission of HIV-1. The presence or absence of these HLA-B alleles in the child was not associated with transmission, nor was maternal-child concordance for these or other HLA-B alleles significantly associated with transmission. Intriguingly, the association of HLA-B*3501 or B*3503 with increased transmission was found primarily in mothers who had low viral loads, while the reduced transmission associated with HLA-B*4901 or B*5301 alleles was predominantly found among mothers with high viral loads.
HLA-B*3501 primarily differs from the HLA-B*3502 and B*3503 alleles because the “F” pocket encoded by the latter alleles that binds the C-terminus of the presented peptide is smaller and further restricts the number of amino acids that can bind. B*5301 also differs from HLA-B*3501 at positions 77, 80, 81, 82, and 83 that similarly define a portion of the “F” pocket 4,14 (Figs. 1 and 2). The pattern of the association of HLA-B*35 alleles and the structurally related HLA-B*5301 allele with increased maternal-infant HIV-1 transmission differed markedly from the pattern of these alleles that has been associated with rapid progression of HIV-1 infection in adults. For example, the allele B*3501, not associated with rapid progression, 12 is strongly associated with heightened vertical transmission. Additionally, B*5301, associated with rapid progression, 12 is protective against maternal-infant transmission. Moreover, the other alleles identified as associated with either a higher or lower risk of mother-to-child transmission (B*1301, B*4402, B*5001, and B*4901) have not been associated with a more rapid rate of HIV-1 disease progression. These associations are inconsistent with the hypothesis that maternal-infant transmission of HIV-1 is an extension of the inadequate handling of the HIV-1 infection by the elements of the mother's adaptive immune response against the virus determined by the peptide-binding properties of HLA-B alleles that result in rapid progression of the HIV-1 infection. Additionally, the finding that the HLA-B allele effect was most striking in mothers with low viral loads and that transmitting HLA B*3501 or B*3503 mothers did not exhibit significantly higher viral loads or decreased CD4 counts (Table 1, and data not shown) further argued against the interpretation that transmission is primarily an extension of inadequate maternal control of the HIV-1 infection.
There is a striking symmetric structural relationship between the pairs of protective and susceptibility alleles (HLA-B*5301 and B*3501; (HLA-B*4901 and B*5001), with each paired protective and susceptibility allele differing from the other only at positions 77, 80, 81, 82, and 83 4,14 (Figs. 1 and 2). These results suggest this region encodes a polymorphic structural determinant that could be the molecular basis of the effect of these alleles on vertical HIV-1 transmission. Some of these residues in this sequence motif form part of the C-terminus peptide pocket. 4,14 Insight into differences in this pocket between B*5301 and B*3501 alleles benefit from a series of elegant studies showing that both are sub-Saharan alleles, with HLA-B*5301 more common among West Africans, in whom it confers malaria resistance by binding a trophozoite peptide through accommodating additional C-terminal residues. 14,15 Thus reduced vertical HIV-1 transmission in B*5301 mothers might reflect a slightly broader antiviral T-cell repertoire, or the absence of other constraining features of the B*3501 F pocket. 14
However, a second function of the sequence motif from 77 to 83 is to encode ligands regulating allele-specific interaction of HLA-B molecules with KIR3DL1. 5,29,30,36 KIR3DL1 recognizes 4 similar ligand structures on HLA-B molecules that differ in their capacity to inhibit cytolysis, each expressing the serologic HLA-Bw4 marker. 29,30 Stronger inhibition of NK-mediated cytotoxicity is provided by alleles with isoleucine at 80 (ile-80, N-I-A-L-R) compared with alleles with threonine (thr-80, N-T-A-L-R) at this position. HLA-B alleles lacking the HLA-Bw4 marker (S-N-L-R-G) cannot react with KIR3DL1 and lack inhibitory functions. 36 The expression of these ligands by particular HLA-B molecules is relevant to the findings of the present study since the HLA-B*3501 and B*5001alleles, as well as HLA-B*3503, each associated with increased transmission of HIV-1 from mother to infant, are S-N-L-R-G alleles that do not encode a KIR3DL1 ligand. Conversely, HLA-B*4901 and *5301, associated with decreased HV-1 transmission, are ile-80, N-I-A-L-R alleles that encode a strong KIR3DL1 ligand (Fig. 2). Indeed, fine differences in KIR3DL1 recognition appear more important than peptide-binding properties; since among mothers who did not receive zidovudine monotherapy, when all HLA-B alleles were categorized only by the presence of amino acid sequences at residue 77–83, those with ile-80 (N-I-A-L-R) alleles had a lower risk of transmission (OR = 0.55), while mothers with N-T-A-L-R alleles had a higher risk of transmission (OR = 2.05, Table 5 and Fig. 2), as were the HLA-B alleles characterized by the S-N-L-R-G motif that do not encode a KIR3DL1 ligand. These observations suggest that the ability of an HLA-B molecule to bind and signal strongly though KIR3DL1 is associated with reduced likelihood of vertical transmission. Supporting this interpretation, NK receptor engagement has been implicated in the control of HIV-1 infection through studies of the HLA-B allele frequencies. 26,27 Moreover, NK cells directly suppress HIV-1 replication by CC chemokine secretion. 28
The massive placental accumulation of NK and CD8 T cells with NK receptors that mediate tolerance of the fetal allograft through incompletely defined pathways 21–23 provides an appropriately situated population of cells with the potential to potently suppress placental HIV-1 replication via this proposed mechanism. We postulate that this cell population plays a critical role in suppressing HIV-1 transmission, but that its participation in inhibiting vertical transmission depends on the mother's set of HLA-alleles, and whether they bear the appropriate ligands to engage NK receptors. It is possible that activation of this inhibitory mechanism is either an intrinsic part of the tolerance response driven by the presence of the placenta, or perhaps and more likely, a specific response to the lowering of MHC class I molecules induced by HIV-1 infection. Interestingly, the pattern of MHC class I lowering induced by HIV-1, involving decreased HLA-A and HLA-B, but not HLA-C gene products, mimics the pattern of HLA class I expression found on trophoblast. 37 This pattern of selective reduction of HLA-B and HLA-A molecules induced by HIV-1 might explain why the HLA-B-encoded KIR ligands could play a more important role in this situation than those encoded by HLA-C. Because of the importance of the NK cell receptor class I interaction in controlling the development of the NK cell repertoire during ontogeny of the individual, 25 we hypothesize that because the HLA-B*3501 or B*5001 alleles do not encode KIR3DL1 ligands, the placental NK and CD8 T-cell repertoires of mothers with these alleles have not acquired the ability to perceive the reduced (missing-self) expression of these HLA-B alleles 25 that occurs in HIV-1-infected CD4 T cells or monocytes. 37 This results in the failure to trigger adequately an additional NKR-dependent event in the fetomaternal interface critical to suppressing HIV-1 transmission. This event could depend on a pure NK cell pathway, such as an induction of sufficient CC chemokine release to effectively diminish viral replication, 28 or on a CD8 T-cell pathway in view of the expression of NK receptors on CD8 T cells 29,30 where removal of the inhibitory signal transduced by KIR3DL1 by lowered expression of HLA-B on an infected cell in turn lowers the threshold for CD8 T-cell activation by viral peptides presented by the remaining HLA molecules, potentially enhancing the adaptive CD8 T-cell anti-HIV-1 response.
The structural implications of the association of HLA-B*1302 and B*4402 alleles with increased vertical transmission are less clear but a similar mechanism involving weaker KIR3DL1 engagement is suggested because the same thr-80 KIR3DL1 ligand is encoded by both HLA-B*1302 and B*4402 alleles. We speculate that this ligand might not adequately activate a KIR3DL1-dependent placental mechanism that thwarts in utero transmission, but additional information on the role of these alleles is needed.
Although the study population for our analyses incorporated data from all HIV-1-infected women who delivered prior to the widespread utilization of any intervention to prevent mother-to-child transmission of HIV-1, and for whom cryo-preserved lymphocytes were available, the study population was still small relative to the numbers of polymorphic HLA-B alleles (in excess of 600), and confirmation of these results in a larger study is warranted. Since several of the alleles are increased in sub-Saharan populations and HLA-B*13 alleles are increased in Southeast Asia populations, studies based in these regions may offer excellent opportunities for extension of these results. KIR with activating functions, eg, KIR3DS1, 27 KIR3DL1 polymorphisms, KIR specific for HLA-C polymorphisms, as well as HLA-C polymorphisms, may also be relevant to the role of NK cells in influencing vertical transmission and these each require further study to determine other potentially parallel instances of an interplay between adaptive and innate immunity in the vertical transmission of HIV-1. In addition it is likely that other mechanisms involving alleles of different HLA loci may be relevant to vertical transmission, in view of reports describing associations with maternal-infant concordance and on the presence of particular MHC alleles in infants that could influence the infant's response to HIV-1. 34,38,39 However, an important implication of the present observations on the role of NK receptors in vertical transmission is that additional work on these innate immune response receptors and the function of placental NK and CD8 T cells may provide insights leading to the development of novel therapeutic and vaccine strategies to reduce maternal-infant HIV-1 transmission through mechanisms that augment innate immune function.
We especially thank Dr. Mary Carrington for her helpful comments and Dr. Young Chan and Ms. Tara Jackson for their assistance in the early stages of this work. Principal investigators, study coordinators, program officers, and funding of WITS include: Clemente Diaz, Edna Pacheco-Acosta (University of Puerto Rico, San Juan, PR; U01 AI 34858); Ruth Tuomala, Ellen Cooper, Donna Mesthene (Boston/Worcester Site, Boston, MA; 9U01 DA 15054); Philip La Russa, Alice Higgins (Columbia Presbyterian Hospital, New York, NY; U01 DA 15053); Sheldon Landesman, Edward Handelsman, Gail Moroso (State University of New York, Brooklyn, NY; U01 HD 36117); Kenneth Rich, Delmyra Turpin (University of Illinois at Chicago, Chicago, IL; U01 AI 34841); William Shearer, Susan Pacheco, Norma Cooper (Baylor College of Medicine, Houston, TX; U01 HD 41983); Joana Rosario (National Institute of Allergy and Infectious Diseases, Bethesda, MD); Robert Nugent, (National Institute of Child Health and Human Development, Bethesda, MD); Vincent Smeriglio, Katherine Davenny (National Institute on Drug Abuse, Bethesda, MD); and Bruce Thompson (Clinical Trials & Surveys Corp., Baltimore, MD, N01 AI 85339). Scientific Leadership Core: Kenneth Rich (PI), Delmyra Turpin (Study Coordinator) (1 U01 AI 50274-01).
2. Kourtis AP, Bulterys M, Nesheim SR, et al. Understanding the timing of HIV transmission from mother to infant. JAMA
3. De Cock KM, Fowler MG, Mercier E, et al. Prevention of mother-to-child HIV transmission in resource-poor countries: translating research into policy and practice. JAMA
4. Little AM, Parham P. Polymorphism and evolution of HLA class I and II genes and molecules. Rev Immunogenet
5. Vilches C, Parham P. KIR: diverse, rapidly evolving receptors of innate and adaptive immunity. Annu Rev Immunol
6. Itescu S, Mathur-Wagh U, Skovron ML, et al. HLA-B35 is associated with accelerated progression to AIDS. J Acquir Immune Defic Syndr
7. Kaslow RA, Carrington M, Apple R, et al. Influence of combinations of human major histocompatibility complex genes on the course of HIV-1 infection. Nat Med
8. Winchester RJ, Charron D, Louie L, et al. The role of HLA in influencing the time of development of a particular outcome of HIV-1 infection. EDK, Paris, France: Proceedings of the Twelfth International Histocompatibility Workshop and Conference 1997.
9. Goulder PJ, Phillips RE, Colbert RA, et al. Late escape from an immunodominant cytotoxic T-lymphocyte response associated with progression to AIDS. Nat Med
10. Nelson GW, Kaslow R, Mann DL. Frequency of HLA allele-specific peptide motifs in HIV-1 proteins correlates with the allele's association with relative rates of disease progression after HIV-1 infection. Proc Natl Acad Sci U S A
11. Carrington M, Nelson GW, Martin MP, et al. HLA and HIV-1: heterozygote advantage and B*35-Cw*04 disadvantage. Science
12. Gao X, Nelson GW, Karacki P, et al. Effect of a single amino acid change in MHC class I molecules on the rate of progression to AIDS. N Engl J Med
13. Rammensee H, Bachmann J, Emmerich NP, et al. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics
14. Smith KJ, Reid SW, Harlos K, et al. Bound water structure and polymorphic amino acids act together to allow the binding of different peptides to MHC class I HLA-B53. Immunity
15. Hill AV, Elvin J, Willis AC, et al. Molecular analysis of the association of HLA-B53 and resistance to severe malaria. Nature
16. Kelleher AD, Long C, Holmes EC, et al. Clustered mutations in HIV-1 gag are consistently required for escape from HLA-B27-restricted cytotoxic T lymphocyte responses. J Exp Med
17. McMichael AJ, Rowland-Jones SL. Cellular immune responses to HIV. Nature
18. Goulder PJ, Brander C, Tang Y, et al. Evolution and transmission of stable CTL escape mutations in HIV infection. Nature
19. Goulder PJ, Jeena P, Tudor-Williams G, et al. Paediatric HIV infection: correlates of protective immunity and global perspectives in prevention and management. Br Med Bull
20. Ljunggren HG, Karre K. In search of the “missing self”: MHC molecules and NK cell recognition. Immunol Today
21. Tafuri A, Alferink J, Moller P, et al. T cell awareness of paternal alloantigens during pregnancy. Science
22. Drake PM, Gunn MD, Charo IF, et al. Human placental cytotrophoblasts attract monocytes and CD56(bright) natural killer cells via the actions of monocyte inflammatory protein 1alpha. J Exp Med
23. King A, Allan DS, Bowen M, et al. HLA-E is expressed on trophoblast and interacts with CD94/NKG2 receptors on decidual NK cells. Eur J Immunol
24. Hsu KC, Chida S, Geraghty DE, et al. The killer cell immunoglobulin-like receptor (KIR) genomic region: gene-order, haplotypes and allelic polymorphism. Immunol Rev
25. Valiante NM, Lienert K, Shilling HG, et al. Killer cell receptors: keeping pace with MHC class I evolution. Immunol Rev
26. Flores-Villanueva PO, Yunis EJ, Delgado JC, et al. Control of HIV-1 viremia and protection from AIDS are associated with HLA-Bw4 homozygosity. Proc Natl Acad Sci U S A
27. Martin MP, Gao X, Lee JH, et al. Epistatic interaction between KIR3DS1 and HLA-B delays the progression to AIDS. Nat Genet
28. Kottilil S, Chun TW, Moir S, et al. Innate immunity in human immunodeficiency virus infection: effect of viremia on natural killer cell function. J Infect Dis
29. Lanier LL. NK cell receptors. Annu Rev Immunol
30. Raulet DH, Held W. Natural killer cell receptors: the offs and ons of NK cell recognition. Cell
31. Sheon AR, Fox HE, Rich KC, et al. The Women and Infants Transmission Study of Maternal-Infant HIV Transmission: study design methods, and baseline data. J Womens Health
32. Garcia PM, Kalish LA, Pitt J, et al. Maternal levels of plasma human immunodeficiency virus type 1 RNA and the risk of perinatal transmission. Women and Infants Transmission Study Group. N Engl J Med
33. Johnston-Dow L, Bengtson A, Scheltinga S, et al. DNA sequencing based typing of HLA-A. EDK, Paris, France: Proceedings of the Twelfth International Histocompatibility Workshop and Conference 1997;1:250–254.
34. MacDonald KS, Embree J, Njenga S, et al. Mother-child class I HLA concordance increases perinatal human immunodeficiency virus type 1 transmission. J Infect Dis
35. Bryson YJ, Luzuriaga K, Sullivan JL, et al. Proposed definitions for in utero versus intrapartum transmission of HIV-1. N Engl J Med
36. Gumperz JE, Barber LD, Valiante NM, et al. Conserved and variable residues within the Bw4 motif of HLA-B make separable contributions to recognition by the NKB1 killer cell-inhibitory receptor. J Immunol
37. Cohen GB, Gandhi RT, Davis DM, et al. The selective downregulation of class I major histocompatibility complex proteins by HIV-1 protects HIV-infected cells from NK cells. Immunity
38. Winchester R, Chen Y, Rose S, et al. MHC class II DR alleles DRB1*1501 and those encoding HLA-DR13 are preferentially associated with a diminution in maternally transmitted HIV-1 infection in different ethnic groups: determination by an automated sequence-based typing method. Proc Natl Acad Sci U S A
39. Kuhn L, Coutsoudis A, Moodley D, et al. T-helper cell responses to HIV envelope peptides in cord blood: protection against intrapartum and breast-feeding transmission. AIDS
Keywords:© 2004 Lippincott Williams & Wilkins, Inc.
HIV-1 vertical transmission; HLA-B alleles; viral load; NK receptor; KIR3DL1 receptor