Análise de KIR e HLA em alta resolução em uma população euro-descendente e em pacientes com carcinoma mamário esporádico
Resumo
Resumo: As células assassinas naturais (ou NK, do inglês natural killer) são importantes efetores da imunidade inata e adaptativa, atuando no combate a células infectadas e células tumorais. Os genes KIR (killer-cell immunoglobulinlike receptors) codificam um grupo de receptores que modulam as respostas dessas células através do reconhecimento de antígenos leucocitários humanos (HLA, do inglês, human leukocyte antigens). Combinações de variantes KIR-HLA já foram associadas com diversas doenças, incluindo vários tipos de câncer, doenças autoimunes e infecciosas. No entanto, devido à complexidade estrutural da família de genes KIR, poucos estudos atingiram um nível de profundidade de análise que permitiu explorar a variação alélica e de número de cópias desses genes em larga escala. Com o objetivo de preencher essa lacuna, nosso primeiro objetivo foi descrever a variação de KIR em alta resolução em uma grande amostra populacional, aplicando sequenciamento de nova geração em 2130 Euro-descendentes dos Estados Unidos. A determinação precisa do número de cópias de KIR nos permitiu identificar um conjunto de haplótipos KIR incomuns, responsáveis por 5,2% da variação estrutural total. Observamos que KIR2DL4 foi o gene moldura que mais variou em número de cópias (6,5% de todos os indivíduos). Além disso, identificamos 250 alelos com resolução de 5 dígitos, dos quais 90 têm frequências >ou igual a 1%. Encontramos padrões de sequência que eram consistentes com a presença de novos alelos em 398 (18,7%) indivíduos e contextualizamos vários polimorfismos de nucleotídeo únicos (SNP) órfãos dentro do complexo. Também identificamos uma nova variante de KIR2DL1 (Pro151Arg), a qual demonstramos por dinâmica molecular que esta substituição possivelmente afeta a interação com HLA-C. No contexto de KIR e doenças, nós escolhemos o câncer de mama como foco do nosso estuado devido à falta de estudos que analisam KIR e HLA de classe I e classe II nessa doença. O câncer de mama (BC) é a neoplasia mais comum entre as mulheres, com tumores que variam em agressividade, tipo histológico, subtipo molecular e desenvolvimento de metástases entre outros. Neste trabalho, analisamos o DNA de 550 mulheres com câncer de mama esporádico e 747 controles. Nós identificamos que a presença de três cópias de KIR2DL5 estava associada à proteção (OR = 0,21, pcorr = 0,027), enquanto a presença de três cópias de KIR3DL1S1 foi associada ao risco aumentado para BC (OR = 2,44, pcorr = 0,009). Também identificamos que o par KIR2DL3*001+HLA-C1 estava significativamente reduzido em pacientes com o subtipo HER2+ não luminal (ER e PR negativo) em comparação com todos os outros subtipos (OR = 0,23, pcorr = 0,016), além de uma forte associação de KIR2DS1*002+HLA-C2 em pacientes com tumores triplo negativos em comparação com todos os outros subtipos (OR = 3,34, pcorr < 1,4 x10-4). Por fim, observamos um forte efeito protetor do par KIR2DL1*001+HLAC2, significativamente aumentado em pacientes com BC de grau I em comparação com os graus mais agressivos II+III (OR = 0,25, pcorr = 0,038). Nós também exploramos a diversidade de HLA classe I e classe II em BC e seus subtipos. HLA-C*12:03 estava significativamente mais frequente em pacientes com câncer de mama triplo negativo (OR = 2.7, p = 0.0025, pcorr = 0.033), bem como variações em amino ácidos nas posições 11 e 12 de HLA-B. HLADPB1* 17:01 foi associado à proteção ao desenvolvimento de metástases em linfonodos (OR = 0.21, p = 0.0019, pcorr = 0.037). E por fim, a presença do amino ácido arginina na posição 151 de HLA-A estava associada ao desenvolvimento de câncer de mama grau 3 (OR = 3.36, p = 0.0016, pcorr = 0.016). Já os aminoácidos lisina e ácido aspártico nas posições 9 e 11 de HLA-DRB1 estavam associadas à proteção contra o câncer de mama de grau 3 (OR = 0.13, p = 0.0033, pcorr = 0.033). Nenhum estudo anterior explorou a variação estrutural e de sequência de KIR em nenhuma população ou doença com a profundidade que apresentamos neste trabalho. Demonstramos que a junção do sequenciamento de alto rendimento com ferramentas computacionais de última geração permite a exploração de todos os aspectos da variação de KIR, incluindo a determinação da diversidade de haplótipos em nível populacional, melhorando a compreensão dessa complexa família gênica e fornecendo uma referência importante para estudos futuros. Abstract: Natural killer cells (or NK) are important effectors of innate and adaptive immunity, acting in the fight against infected and tumor cells. The KIR genes (killer-cell immunoglobulin-like receptors) encode a group of receptors that modulate the responses of these cells through the recognition of human leukocyte antigens (HLA). Combinations of KIR-HLA variants have been associated with several diseases, including many types of cancer, autoimmune and infectious diseases. However, due to the structural complexity of the KIR gene family, few studies have reached a level of analysis depth that allowed exploring the allelic and copy number variation of these genes on a large scale. In order to fill this gap, our first objective was to describe the KIR variation in high resolution in a large population sample, applying next-generation sequencing in 2130 Euro-descendants from the United States. Precise determination of the KIR copy number allowed us to identify a set of unusual KIR haplotypes, responsible for 5.2% of the total structural variation. We observed that KIR2DL4 was the framework gene that most varied in number of copies (6.5% of all individuals). In addition, we identified 250 alleles with 5-digit resolution, of which 90 have frequencies > or equal to 1%. We also found sequence patterns that were consistent with the presence of new alleles in 398 (18.7%) individuals and contextualized several orphan single nucleotide polymorphisms (SNP) within the complex. We also identified a new variant of KIR2DL1 (Pro151Arg), which we demonstrated by molecular dynamics that this substitution possibly affects the interaction with HLA-C. In the context of KIR and disease, we chose breast cancer as the focus of our study due to the lack of studies that analyze KIR and HLA class I and class II in this disease. Breast cancer (BC) is the most common malignancy among women, with tumors that vary in aggressiveness, histological type, molecular subtype, and development of metastases, among others. In this work, we analyzed the DNA of 550 women with sporadic breast cancer and 747 controls. We found that the presence of three copies of KIR2DL5 was associated with protection (OR = 0.21, pcorr = 0.027), while the presence of three copies of KIR3DL1S1 was associated with increased risk for BC (OR = 2.44, pcorr = 0.009). We also identified that the KIR2DL3*001+HLA-C1 pair was significantly reduced in patients with the non-luminal HER2+ subtype (ER and PR negative) compared to all other subtypes (OR = 0.23, pcorr = 0.016), in addition to a strong association of KIR2DS1*002+HLA-C2 in patients with triple negative tumors compared to all other subtypes (OR = 3.34, pcorr < 1.4 x10-4). Finally, we observed a strong protective effect of the KIR2DL1*001+HLA-C2 pair, significantly increased in patients with grade I BC compared to the more aggressive grades II+III (OR = 0.25, pcorr = 0.038). We also explored the diversity of HLA class I and class II in BC and their subtypes. HLA-C*12:03 was observed significantly more frequent in patients with triple negative breast cancer (OR = 2.7, p = 0.0025, pcorr = 0.033), as well as variations in amino acids at positions 11 and 12 of HLA-B. HLADPB1* 17:01 was associated with protection against the development of lymph node metastases (OR = 0.21, p = 0.0019, pcorr = 0.037). Finally, the presence of the amino acid arginine at position 151 of HLA-A was associated with the development of grade III breast cancer (OR = 3.36, p = 0.0016, pcorr = 0.016). The amino acids lysine and aspartic acid at positions 9 and 11 of HLA-DRB1 were associated with protection against grade III breast cancer (OR = 0.13, p = 0.0033, pcorr = 0.033). No previous study has explored the structural and sequence variation of KIR in any population or disease to the depth that we have presented in this work. We demonstrate that the combination of high-throughput sequencing with state-of-the-art computational tools allows for the exploration of all aspects of KIR variation, including the determination of haplotype diversity at the population level, improving the understanding of this complex gene family, and providing an important reference for future studies.
Collections
- Teses [117]