Molecular genetics and population studies of the KIR and HLA gene complexes

经典的HLA I类分子，HLA- a， -B和-C，向CD8+ T细胞呈递抗原肽，引发适应性免疫反应。编码这些分子的基因是高度多态性的，导致在个体和群体水平上肽库的广泛多样性。HLA I类分子的肽装载主要发生在肽装载复合物（PLC）内的内质网中。Tapasin是PLC的关键成分，它通过与肽空HLA I类结合来执行肽“编辑”功能，稳定其结构，促进低亲和力肽的解离。在没有tapasin的情况下，HLA I类异体在细胞表面表达水平上有所不同。一些同种异体在tapasin缺陷细胞表面的表达水平非常低（tapasin依赖性同种异体），而其他同种异体在这些细胞上表现出正常表达（tapasin非依赖性同种异体）。tapasin依赖性（TD）的确切分子决定因素尚不清楚，尽管肽C端附近肽结合槽中的氨基酸似乎对这种现象贡献最大。例如，位于该区域的单个氨基酸变化（D116Y）将高度依赖tapasin的B4402同种异型与不依赖tapasin的B4405同种异型区分开来。Tapasin的功能可以被病毒靶向，作为下调HLA I类和逃避细胞毒性CD8+ T细胞（CTL）反应的手段。同样，在各种人类癌症中也观察到tapasin表达的缺失。因此，当tapasin功能减弱时，不依赖于tapasin的HLA I类同种异体在诱导针对病毒感染细胞或肿瘤细胞的CTL反应方面可能是有利的。tapasin对肽库的同种异体特异性调节也可能影响CTL反应的质量。我们量化了欧洲人和非裔美国人所有常见HLA同种异体的TD水平，并测试了差异HLA I类TD的功能意义及其对疾病的影响。对来自受感染对象的整个HIV-1蛋白质组的细胞毒性T细胞反应的体外检查表明，依赖tapasin的同种异体比不依赖tapasin的同种异体呈现出更有限的不同肽集，这一数据得到了计算预测的支持。这表明，tapasin依赖性的变化可能通过改变肽库大小来影响免疫反应的强度。为了支持这一模型，我们观察到携带具有更大的tapasin独立性的HLA I类基因型的个体更慢地发展为艾滋病，并保持较低的病毒载量，可能是由于肽呈现的广度增加。因此，tapasin依赖水平，像HLA合子一样，可以作为限制或扩大人类HLA- 1肽库宽度的一种手段，最终影响对病原体和疫苗的免疫反应。该地区的复杂性限制了评估KIR基因组多样性性质和程度的方法的发展。这个基因家族的成员具有很高的序列相似性。此外，在基因组水平上，就KIR基因含量而言，存在显著的多样性。因此，基于短读测序数据确定KIR基因的缺失或存在是极具挑战性的。目前，我们正在开发一种计算方法，允许基于全基因组或全外显子组测序数据预测KIR基因含量。这种方法利用相同长度的序列片段（k-mers）是独特的单个KIR基因。初步结果是有希望的，并表明基于测序数据的KIR基因的高置信度预测是可行的。这种方法将允许我们查询公开可用的数据集进行疾病关联分析。IgG亚类IgG1、IgG2、IgG3和IgG4由于其恒定H链的差异而表现出不同的功能特性，分别由单个遗传位点IGHG1、IGHG2、IGHG3和IGHG4编码。这些基因在人类第14号染色体上的IGHC基因座内形成一个约150kb的簇。恒定区Fc部分（CH2和CH3结构域）介导抗体稳定性及其效应功能，如细胞毒性、吞噬作用和补体活化。因此，该区域的多态性可能直接影响免疫应答。例如，IgG3中435位天然发生的从精氨酸到组氨酸的变化导致抗体半衰期急剧增加。恒定区域的变异主要通过血清学方法（Gm-Am同种异体）表征，核苷酸多样性信息有限。由于IGHG基因之间的高度同源性，基因组的这一区域在全基因组研究中没有很好地覆盖。我们开发了一种基于Sanger测序的基因分型方法，该方法涵盖了IGHG1， IGHG2和IGHG3基因的所有外显子。此外，我们可以区分铰链外显子拷贝数，这对于IGHG3来说是可变的。该方法已应用于几个人群，包括健康的白人、感染艾滋病毒的白人和美国黑人，以及来自南非的健康黑人。我们观察到这些组之间的频率分布有显著差异。

The classical HLA class I molecules, HLA-A, -B and -C, present antigenic peptides to CD8+ T cells, eliciting an adaptive immune response. The genes encoding these molecules are highly polymorphic, resulting in extensive diversity of the peptide repertoire, both within individuals and at the population level. Peptide loading of HLA class I molecules takes place primarily in the endoplasmic reticulum within the peptide loading complex (PLC). Tapasin is a critical component of the PLC, which performs its peptide "editing" function by association with peptide-empty HLA class I, stabilizing its structure, and promoting dissociation of low affinity peptides. HLA class I allotypes vary in level of cell surface expression in the absence of tapasin. Some allotypes are expressed at very low levels on the surface of tapasin-deficient cells (tapasin-dependent allotypes), while others exhibit normal expression on these cells (tapasin-independent allotypes). The exact molecular determinants of tapasin dependence (TD) remain unknown, although amino acids in the peptide binding groove near the peptide C terminus appear to contribute most to this phenomenon. For example, a single amino acid change (D116Y) located in this region distinguishes the highly tapasin-dependent B4402 allotype from the tapasin-independent B4405 allotype. Tapasin function can be targeted by viruses as a means of downmodulating HLA class I and evading cytotoxic CD8+ T cell (CTL) responses. Similarly, loss of tapasin expression has been observed in various human cancers. Thus, tapasin-independent HLA class I allotypes may be advantageous in terms of eliciting CTL responses against virally infected cells or tumor cells when tapasin function has been diminished. Allotype-specific regulation of the peptide repertoire by tapasin may also affect the quality of CTL responses. We have quantified the level of TD across all common HLA allotypes present in European and African Americans and tested the functional significance of differential HLA class I TD and its impact on disease. Ex vivo examination of cytotoxic T cell responses to the entire HIV-1 proteome from infected subjects indicates that tapasin-dependent allotypes present a more limited set of distinct peptides than do tapasin-independent allotypes, data supported by computational predictions. This suggests that variation in tapasin dependence may impact the strength of the immune responses by altering peptide repertoire size. In support of this model, we observed that individuals carrying HLA class I genotypes characterized by greater tapasin independence progress more slowly to AIDS and maintain lower viral loads, presumably due to increased breadth of peptide presentation. Thus, tapasin dependence level, like HLA zygosity, may serve as a means to restrict or expand breadth of the HLA-I peptide repertoire across humans, ultimately influencing immune responses to pathogens and vaccines. The development of methods for assessing the nature and extent of KIR genomic diversity has been limited by the complexity of the region. Members of this gene family share a high amount of sequence similarity. Also, there is remarkable diversity with respect to KIR gene content at the genomic level. Therefore, it is extremely challenging to determine the absence or presence of KIR genes based on short-read sequencing data. Currently, we are developing a computational approach that allows predictions for KIR gene content based on either whole-genome or whole-exome sequencing data. This approach utilizes same-length sequence fragments (k-mers) that are unique to individual KIR genes. Preliminary results are promising and show that high-confidence predictions of KIR genes based on sequencing data is feasible. This method will allow us to interrogate publicly available datasets for disease association analyses. IgG subclasses, IgG1, IgG2, IgG3, and IgG4, display distinct functional properties due to the differences in their constant H chains, encoded by individual genetic loci, IGHG1, IGHG2, IGHG3, and IGHG4, respectively. These genes form a cluster spanning approximately 150 kb region within the Ig H chain constant (IGHC) locus on human chromosome 14. The Fc portion of the constant region (CH2 and CH3 domains) mediates antibody stability and its effector functions, such as cytotoxicity, phagocytosis, and complement activation. Therefore, polymorphism in this region may directly affect immune responses. For example, the naturally occurring change from arginine to histidine at position 435 in IgG3 causes a dramatic increase in the antibody half-life. Variation in the constant regions has been characterized primarily by serological methods (Gm-Am allotypes) with limited information on nucleotide diversity. This region of the genome is not well covered in genome-wide studies due to high homology between the IGHG genes. We have developed a genotyping method based on Sanger sequencing, which covers all exons of IGHG1, IGHG2, and IGHG3 genes. In addition, we can distinguish hinge exon copy number, which is variable for IGHG3. The method has been applied to several population groups, including healthy whites, HIV infected whites and blacks from the US, as well as healthy blacks from South Africa. We have observed striking differences in frequency distributions among these groups.