Killer cell immunoglobulin-like receptors (KIR) and human leukocyte antigen-C (HLA-C) allorecognition patterns in women with endometriosis

frequency distributions of HLA-C alleles among endometriosis and control groups. The demographic results of endometriosis and control groups are shown in Table  1. HLA-C allele frequencies in endometriosis patients (n = 147, 294 alleles) and control patients (n = 117, 234 alleles) were determined using a sequence-based typing method. The presence of HLA-C*03:03:01 significantly increased the risk of endometri- osis with p = 0.0473 [Odds Ratio (OR) = 2.811, 95% confidence interval (CI) = 1.021–7.738] and the statistical power was 43.8% (Table  2). After multiple test analyses using Bonferroni correction, the association was not significant. frequency distributions of HLA-C group among endometriosis and control groups. We evaluated whether the HLA-C group C1 (HLA-C1) and HLA-C group C2 (HLA-C2) were associated with Characteristics Control n = 117 (%) Endometriosis n = 147 (%) p value Age a 38.44 (7.47) 36.08 (6.55) 0.012 BMIa, kg/m2 23.01 (4.47) 21.58 (3.47) 0.0032 Age of menarchea 12.53 (1.21) 12.80 (1.52) 0.5968 Duration of Menstrual cyclea 27.95 (4.95) 28.45 (2.97) 0.1392 Dysmenorrheab, n (%) 73 (62.39) 112 (76.19) 0.015 Table 1. Patient demographic results. Abbreviations: BMI, body mass index; SD, standard deviation Mean (SD) for continuous variables. n (%) for discontinuous variables. aMann-Whitney test. bχ 2 test. 3Scientific RepoRtS | (2020) 10:4897 | https://doi.org/10.1038/s41598-020-61702-y www.nature.com/scientificreportswww.nature.com/scientificreports/ endometriosis. Analysis revealed no significant differences in HLA-C1 and HLA-C2 frequencies in the endome- triosis and control groups (Table 3). frequency distributions of KIR genotypes among endometriosis and control groups. Using sequence-specific PCR amplification, we analyzed the KIR genotypes in the endometriosis and control groups. The frequencies of the KIR genotypes in women with endometriosis and controls and their statistical associations are presented in Table 4. The presence of KIR2DS2 significantly reduced the risk of endometriosis with p = 0.0394 [(OR) = 0.5577, 95% CI = 0.3251–0.9569] and the statistical power was 68.6%. After multiple test analyses using Bonferroni correction, the association was not significant. The two groups showed no significant differences in the remaining KIR genotypes. HLA-C Control (n = 117, 234 alleles) Endometriosis (n = 147, 294 alleles) OR 95% CI P valuen % n % C*01:02:01 56 23.9 59 20.1 0.798 0.5273 to 1.208 0.2907 C*01:08 0 0.0 1 0.3 2.397 0.09712 to 59.16 1 C*02:02:02:01 0 0.0 2 0.7 4.009 0.1914 to 83.97 0.5055 C*03:02:01 24 10.3 33 11.2 1.106 0.6342 to 1.930 0.7786 C*03:03:01 5 2.1 17 5.8 2.811 1.021 to 7.738 0.0473* C*03:04:01:01 30 12.8 34 11.6 0.8892 0.5265 to 1.502 0.6886 C*03:04:04 0 0.0 4 1.4 7.265 0.3889 to 135.7 0.1333 C*04:01:01:01 10 4.3 12 4.1 0.9532 0.4043 to 2.247 1 C*04:03 4 1.7 4 1.4 0.7931 0.1962 to 3.207 0.7375 C*06:02:01:01 7 3.0 5 1.7 0.561 0.1757 to 1.792 0.3849 C*07:01:01:01 1 0.4 0 0.0 0.2643 0.01071 to 6.523 0.4432 C*07:02:01:01 39 16.7 61 20.7 1.309 0.8391 to 2.042 0.264 C*07:04:01 1 0.4 1 0.3 0.7952 0.04944 to 12.79 1 C*07:359 1 0.4 0 0.0 0.2643 0.01071 to 6.523 0.4432 C*08:01:01 21 9.0 20 6.8 0.7404 0.3911 to 1.401 0.4138 C*08:03:01 0 0.0 1 0.3 2.397 0.09712 to 59.16 1 C*12:02:02 15 6.4 14 4.8 0.73 0.3449 to 1.545 0.4456 C*12:03:01:01 3 1.3 1 0.3 0.2628 0.02714 to 2.544 0.3269 C*14:02:01 6 2.6 9 3.1 1.2 0.4208 to 3.422 0.7975 C*14:02:03 0 0.0 1 0.3 2.397 0.09712 to 59.16 1 C*15:02:01 8 3.4 14 4.8 1.413 0.5822 to 3.427 0.5151 C*15:05:01 1 0.4 0 0.0 0.2643 0.01071 to 6.523 0.4432 C*16:02:01 1 0.4 0 0.0 0.2643 0.01071 to 6.523 0.4432 C*16:04:01 1 0.4 1 0.3 0.7952 0.04944 to 12.79 1 Table 2. Distribution of the HLA-C alleles in the endometriosis and control groups. Each HLA allele has four unique sets denoted by different numbers that are separated by a colon. The first two digits often correspond to the serological antigen; the two digits after the first colon denote the subtypes and order in the genome from the IMGT/HLA Database (www.ebi.ac.uk/imgt/hla/). The differences in HLA-C allele frequencies between the endometriosis and control groups were analyzed using the Fisher’s exact test. Significance was set at a P value < 0.05 and the statistical power was 43.8% calculated by G*Power. OR indicates odds ratio. CI indicates confidence interval. HLA-C Control (n = 117) Endometriosis (n = 147) OR 95% CI p valuen % n % C1 115 98.3 146 99.3 2.539 0.2273 to 28.37 0.5859 C2 29 24.8 36 24.5 0.9842 0.5602 to 1.729 1 C1C1 88 75.2 111 75.5 1.016 0.5783 to 1.785 1 C2C2 2 1.7 1 0.7 0.3938 0.03525 to 4.400 0.5859 C1C2 27 23.1 35 23.8 1.042 0.5869 to 1.849 1 Table 3. Distribution of HLA-C ligand in endometriosis and control groups. Two-sided Fisher’s exact test was used to estimate the differences between endometriosis and control groups. n: number of cases with relevant genotypes, OR: odds ratio, CI: confidence interval, Significance was set at a P value < 0.05. 4Scientific RepoRtS | (2020) 10:4897 | https://doi.org/10.1038/s41598-020-61702-y www.nature.com/scientificreportswww.nature.com/scientificreports/ frequency distributions of KIR haplotypes among endometriosis and control groups. The fre- quencies of the KIR haplotypes in women with endometriosis and controls and their statistical associations are presented in Table 5. The χ2 value was calculated by Hardy-Weinberg analysis (χ2 > 3.841 showed the subgroup was deviating from the Hardy–Weinberg equilibrium). We revealed differences between the endometriosis and control cohorts. The number of KIR centromeric A/A haplotypes was increased in the endometriosis group than controls with p = 0.0394 [(OR) = 1.793, 95% CI = 1.045–3.076] and the statistical power was 68.6%. combinations of K iR and their HLA- c ligands. The frequencies of KIR and their HLA-C ligands were analyzed for their statistical associations with endometriosis (Table  6). HLA-C C1 groups are recognized by KIR2DL2/2DS2 and KIR2DL3, while HLA-C C2 groups are recognized by KIR2DL1/2DS1. Moreover, KIR2DL2/2DL3 also binds to some HLA-C C2 molecules, and KIR2DS4 binds to some HLA-C1 and HLA-C2 31,34–36. The molecular interactions of KIR gene-HLA-ligands were calculated from the KIR frequency of a total number of HLA ligands. The total number of HLA ligands is shown in Table 3. We calculated the KIR frequency in the combination of different HA ligands. Analysis of various KIR-HLA-C combinations revealed no significant dif- ferences between the endometriosis and control cohorts (Table 6). The frequency of KIR haplotypes and HLA-C combinations also showed no significant differences between the endometriosis and control groups (Table 7).

Several factors are involved in the pathogenesis of endometriosis including genetic, neuroendocrine, and immu- nological factors 43–45. Abnormal immune responses are recognized in endometriosis patients, including excessive inflammatory cytokine secretion, autoantibody production, and NK cell regulation 17,18. Endometriosis shares similar characteristics with autoimmune diseases11,12,17,18. HLA-C affects viral infections and is implicated in sev- eral human autoimmune diseases22. HLA-C*06:02 is associated with severe early-onset psoriasis. HLA-C *12:02 was found to be associated with increased susceptibility to Crohn’s disease20. HLA-C*03 restricts the cytotoxic CD8+ T-cell responses during the Epstein-Barr virus and human immunodeficiency virus infections, as well as during co-infection with the influenza virus and the Sendai virus. Herein, we analyzed the associations between HLA-C alleles and endometriosis. Consequently, the presence of HLA-C*03:03:01 increased the risk of endome- triosis in Asian women (Table 2). However, after multiple test analyses using Bonferroni correction, the associa- tion was not significant. Previous studies reported no association between HLA genotypes and endometriosis in Caucasian women with endometriosis and controls, as assessed by serological typing 46–48. A serological study showed increased frequencies of the HLA-B*54 and HLA-C*07 alleles in Japanese patients with endometriosis 49. In a recent study, PCR-restriction fragment length polymorphism analysis revealed that the HLA-DRB1* 14:03 and HLA-DQB1*03:01 alleles are associated with endometriosis in Japanese women50,51. Another study reported an Inhibitory KIR Control (n = 117) Endometriosis (n = 147) OR 95% CI p valuen % n % KIR2DL1 114 97.4 147 100.0 9.017 0.4608 to 176.5 0.0858 KIR2DL2 37 31.6 31 21.1 0.5778 0.3314 to 1.007 0.0653 KIR2DL3 114 97.4 146 99.3 3.842 0.3942 to 37.45 0.3252 KIR2DL4 117 100.0 147 100.0 — — — KIR2DL5 52 44.4 49 33.3 0.625 0.3788 to 1.031 0.0747 KIR3DL1 115 98.3 145 98.6 1.261 0.1748 to 9.093 1 KIR3DL2 117 100.0 147 100.0 — — — KIR3DL3 117 100.0 147 100.0 — — — Activating KIR KIR2DS1 39 33.3 46 31.3 0.9109 0.5421 to 1.531 0.7911 KIR2DS2 41 35.0 34 23.1 0.5577 0.3251 to 0.9569 0.0394* KIR2DS3 29 24.8 25 17.0 0.6218 0.3409 to 1.134 0.1273 KIR2DS4# 113 96.6 143 97.3 1.265 0.3096 to 5.173 0.7358 KIR2DS4f 89 76.1 117 79.6 1.227 0.6840 to 2.201 0.5503 KIR2DS4d 60 51.3 76 51.7 1.017 0.6255 to 1.653 1 KIR2DS5 24 20.5 25 17.0 0.7941 0.4264 to 1.479 0.5249 KIR3DS1 44 37.6 48 32.7 0.8044 0.4836 to 1.338 0.4365 Pseudogene KIR2DP1 114 97.4 147 100.0 9.017 0.4608 to 176.5 0.0858 KIR3DP1 117 100.0 147 100.0 — — — Table 4. Genetic association between KIR gene in endometriosis and control groups. #The gene was considered positive if either of the two forms were present. KIR2DS4f - full-length KIR2DS4 allele variant. KIR2DS4d - deleted KIR2DS4 allele variant. Two-sided Fisher’s exact test was used to estimate the differences between endometriosis and control groups. n: number of cases with relevant genotypes, OR: odds ratio, CI: confidence interval, *versus controls, p < 0.05 and the statistical power was 68.6% calculated by G*Power. 5Scientific RepoRtS | (2020) 10:4897 | https://doi.org/10.1038/s41598-020-61702-y www.nature.com/scientificreportswww.nature.com/scientificreports/ association between the HLA-A*24, HLA-B*07:02, HLA-C*07:02, and HLA-DRB1*01:01 haplotypes and endome- triosis in Japanese women52. A previous study showed that HLA-DRB1 alleles were not associated with endome- triosis in Polish women53. A literature search identified similar reports from China, which showed an association between endometriosis and the HLA-B*46, HLA-DRB1*15, and HLA-DQA1*0401 alleles54–56. The reasons under- lying the discrepancies observed among these studies are unclear; however, the results may have been influenced by differences in the ethnicities of the women in the study cohorts and the differences in the detection methods. The frequency of KIR3DS1 was significantly lower in endometriosis patients compared to control patients57. Moreover, the protective effect of the KIR2DS5 gene was observed in endometriosis patients58. Moreover, Nowak et al. showed that the protective effect of KIR2DS5 was present only in the women who harbored the HLA-C C2 group59. Our current findings revealed that a lower proportion of endometriosis groups, which was characterized by the presence of activating KIR2DS2 compared to the control groups (Table 4). Previous studies have shown that KIR2DL2 is in the linkage disequilibrium with KIR2DS2, which caused the relative activation of KIR recep- tor, which is responsible for the loss of recognition of HLA-C 60–62. The different ethnic populations showed dif- ferent values of KIR polymorphisms in elucidating genetic relationships among human populations63. Moreover, HLA genotyping is traditionally performed using a serological method. Therefore, these discrepancies can be influenced by ethnic or assay methods as they do for discrepancies among HLA alleles. NK cell activity has been reported to be a crucial factor in the recognition and lysis of endometrial cells. NK cell activity and quantity were found to be controversial in women with endometriosis relative to controls 15,64–68. The observed increase in the proportion of CD158a+ (KIR2DL1) NK cells in the peripheral blood and peritoneal fluid in endometriosis patients suggested reduced NK cell cytotoxicity in endometriosis69. Moreover, the decrease Centromeric Control (n = 117) Endometriosis (n = 147) OR 95% CI p valuen % n % Cen-A/A 76 65.0 113 76.9 1.793 1.045 to 3.076 0.0394* Cen-A/B 38 32.5 33 22.4 0.6018 0.3480 to 1.041 0.0713 Cen-B/B 3 2.6 1 0.7 0.2603 0.02670 to 2.537 0.3252 χ 2 0.47 0.73 Telomeric Tel-A/A 71 60.7 96 65.3 1.22 0.7374 to 2.017 0.4442 Tel-A/B 42 35.9 47 32.0 0.8393 0.5026 to 1.402 0.5151 Tel-B/B 4 3.4 4 2.7 0.7902 0.1933 to 3.230 0.7358 χ 2 0.55 0.39 Table 5. Frequency of centromeric and telomeric KIR haplotypes in endometriosis and control groups. Two- sided Fisher’s exact test was used to estimate the differences between endometriosis and control groups. n: number of cases with relevant genotypes, OR: odds ratio, CI: confidence interval, χ2 value was calculated by Hardy-Weinberg analysis (χ2 > 3.841 showed the subgroup was deviating from the Hardy–Weinberg equilibrium). *versus controls, P < 0.05 and the statistical power was 68.6% calculated by G*Power. Inhibitory KIR-ligand association Control Endometriosis OR 95% CI p valuen % n % KIR2DL1-HLA-C1/C2 27 100.0 35 100.0 — — — KIR2DL1-HLA-C2/C2 1 50.0 1 100.0 3 0.05947 to 151.3 1 KIR2DL2-HLA-C1/C1 25 28.4 20 18.0 0.5538 0.2834 to 1.083 0.0902 KIR2DL2-HLA-C1/C2 10 37.0 11 31.4 0.7792 0.2704 to 2.245 0.7876 KIR2DL3-HLA-C1/C1 86 97.7 110 99.1 2.558 0.2280 to 28.70 0.5847 KIR2DL3-HLA-C1/C2 26 96.3 35 100.0 4.019 0.1573 to 102.7 0.4355 Activating KIR-ligand association KIR2DS1-HLA-C1/C2 5 18.5 13 37.1 2.6 0.7918 to 8.538 0.1593 KIR2DS1-HLA-C2/C2 2 100.0 0 0.0 0.0667 0.0008081 to 5.5 0.3333 KIR2DS2-HLA-C1/C1 28 31.8 23 20.7 0.5601 0.2947 to 1.064 0.1016 KIR2DS2-HLA-C1/C2 11 40.7 11 31.4 0.6667 0.2337 to 1.902 0.5932 Table 6. Distribution of molecular interactions of KIR gene-HLA-ligands in endometriosis and control groups. The molecular interactions of KIR gene-HLA-ligands were shown in the frequency of the KIR gene of HLA- ligands, which was calculated from the KIR frequency of the total number of HLA ligands. The total number of HLA ligands is shown in Table 3. Two-sided Fisher’s exact test was used to estimate the differences between endometriosis and control groups. n: number of cases with relevant genotypes, OR: odds ratio, CI: confidence interval. 6Scientific RepoRtS | (2020) 10:4897 | https://doi.org/10.1038/s41598-020-61702-y www.nature.com/scientificreportswww.nature.com/scientificreports/ in the proportion of NK cells facilitates endometrial cell invasion and persistent growth in endometriosis 70. Maeda et al. demonstrated that the increase in KIR2DL1 levels in women with pelvic endometriosis can inhibit NK cell activity66. However, the activation of chronic NK cell activation can also play a role in endometriosis71. The upregulation of KIR2DS1 expression in peritoneal fluid has been detected in women with endometriosis72. Recently, bioinformatic analysis showed that KIR2DS2 expression is upregulated in the secretory phase of endo- metrium in women with endometriosis relative to control women73. Thus, the regulation of NK cell activity is a complex process and can affect the pathogenesis of endometriosis74. The final activation status of functional NK cells depends on the homeostasis of all NK cell activation/inhibitory receptors and the corresponding ligands. Then, NK cells are regulated in endometrium in women with endometriosis. Further studies will be worth elu- cidating the functional relevance of the presence of these receptors and ligand proteins in the endometrium. The

of this study is that only the stage III or stage IV endometriosis patients were enrolled to investigate the genetic associations. Thus, our results did not show any association between genes and severity of disease. Further clinically relevant studies on the severity of disease and genetic associations are required. Our current findings demonstrated the association between KIR polymorphisms and HLA-C genotypes with endometriosis in women. It is the first study addressing KIR polymorphisms and HLA-C genotypes of women with stage III or IV endometriosis in Han Chinese women. The results suggested that HLA-C and KIR genotypes influence the susceptibility for endometriosis. Further studies should investigate the role of NK cells in the patho- genesis of endometriosis.

patients and controls. Han Chinese women were categorized into the endometriosis (n = 147) and control (n = 117) groups. Endometriosis was diagnosed via laparoscopic examination and confirmed via histological assessment. A total of 147 women were classified under stage III or IV endometriosis in accordance with the Revised American Society for Reproductive Medicine Classification. Women in the control group underwent benign gynecological surgery and showed no evidence of endometriosis including myoma, teratoma, serous cys- tadenoma, ovarian cyst, ovarian stroma, dermoid cyst, mucinous cystadenoma, paratubal cyst, follicular cyst, simple cyst, hydrosalpinx, corpus luteum cyst, fibrous adhesion, and struma ovarii. Considering that autoimmune disorders are associated with HLA-C alleles and KIR genotypes, the exclusion criteria comprised autoimmune disorders. The protocol was approved by the Institutional Review Board of the Taipei Medical University Hospital, and all participants have informed consent. All experiments were performed in accordance with relevant guide- lines and regulations. DnA extraction and HLA-C and KIR genotype analysis. Genomic DNA was extracted using a DNA whole-blood kit following the manufacturer’s instructions (Kurabo Industries, Osaka, Japan). HLA-C was gen- otyped using an HLAssure™ SE sequence-based typing kit (TBG Biotechnology Corp, Queensland, Australia), which was designed to determine HLA-C alleles via polymerase chain reaction (PCR) amplification using a KIR haplotypes HLA-C genotypes Control Endometriosis OR 95% CI P valuen % n % Cen-A/A C1/C1 60 68.2 88 79.3 1.786 0.9397 to 3.393 0.1016 Cen-A/A C1/C2 16 59.3 24 68.6 1.5 0.5257 to 4.280 0.5932 Cen-A/A C2/C2 0 0.0 1 100.0 15 0.1818 to 1238 0.3333 Cen-A/B C1/C1 26 29.5 22 19.8 0.5895 0.3065 to 1.134 0.1338 Cen-A/B C1/C2 10 37.0 11 31.4 0.7792 0.2704 to 2.245 0.7876 Cen-A/B C2/C2 2 100.0 0 0.0 0.0667 0.0008081 to 5.5 0.3333 Cen-B/B C1/C1 2 2.3 1 0.9 0.3909 0.03484 to 4.385 0.5847 Cen-B/B C1/C2 1 3.7 0 0.0 0.2488 0.009738 to 6.358 0.4355 Cen-B/B C2/C2 0 0.0 0 0.0 — — — Tel-A/A C1/C1 50 56.8 74 66.7 1.52 0.8529 to 2.709 0.1853 Tel-A/A C1/C2 21 77.8 21 60.0 0.4286 0.1382 to 1.329 0.1758 Tel-A/A C2/C2 0 0.0 1 100.0 15 0.1818 to 1238 0.3333 Tel-A/B C1/C1 35 39.8 33 29.7 0.6407 0.3551 to 1.156 0.1755 Tel-A/B C1/C2 6 22.2 14 40.0 2.333 0.7523 to 7.237 0.1758 Tel-A/B C2/C2 1 50.0 0 0.0 0.3333 0.0066 to 16.82 1 Tel-B/B C1/C1 3 3.4 4 3.6 1.059 0.2307 to 4.863 1 Tel-B/B C1/C2 0 0.0 0 0.0 — — — Tel-B/B C2/C2 1 50.0 0 0.0 0.3333 0.0066 to 16.82 1 Table 7. Distribution of molecular interactions of KIR haplotypes-HLA-ligands in endometriosis and control groups. The molecular interactions of KIR haplotypes-HLA-ligands were shown in the frequency of KIR haplotypes of HLA-ligands, which was calculated from the KIR haplotypes frequency of the total number of HLA ligands. The total number of HLA ligands is shown in Table 3. Two-sided Fisher’s exact test was used to estimate the differences between endometriosis and control groups. n: number of cases with relevant genotypes, OR: odds ratio, CI: confidence interval. 7Scientific RepoRtS | (2020) 10:4897 | https://doi.org/10.1038/s41598-020-61702-y www.nature.com/scientificreportswww.nature.com/scientificreports/ sequence-based typing method. Sequence data were processed using allele-typing software (AccuType™) to iden- tify the HLA-C alleles. The genotypes of the KIR genes were analyzed using the Lifecodes KIR-sequence-specific oligonucleotide (SSO) typing kit (Immucor Transplant Diagnostics, Inc., Stamford, USA) to identify the KIR loci amplified in the sample. The presence or absence of the 16 KIR genes was determined using 20 different oligonucleotide probes targeting known KIR genes (KIR3DL3 as positive control, KIR2DL1, KIR2DL2*001-3/5, KIR2DL2* 004, KIR2DL3, KIR2DL4, KIR2DL5, KIR2DP1, KIR3DL1, KIR3DL2, KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4*whole exon 4, KIR2DS4* whole exon 5, KIR2DS4* -deleted exon 5, KIR2DS5, KIR3DS1, KIR3DS1*049N, and KIR3DP1). The amplicons were analyzed on a Luminex instrument according to the manu- facturer’s instructions. The characteristics of full-length and truncated forms of KIR2DS4 were determined using the following three probes; probe 45: KIR2DS4* all full-length, probe 175: 2DS4* full-length Exon 5, and probe 234: 2DS4*deletion Exon 5. KIR genes are divided into centromeric and telomeric haplotypes 75. In short, centro- meric A/A haplotypes contained KIR2DL3 but not with KIR2DL2 and/or KIR2DS2, centromeric A/B haplotypes contained KIR2DL3 with KIR2DL2 and/or KIR2DS2, and centromeric B/B haplotypes contained KIR2DL2 and/ or KIR2DS2 but not KIR2DL3. Meanwhile, telomeric A/A haplotypes contained KIR3DL1 and KIR2DS4 but not KIR3DS1 or KIR2DS1, telomeric A/B haplotypes contained KIR3DL1 and KIR2DS4 with KIR3DS1 and/or KIR2DS1, and telomeric B/B haplotypes lacked KIR3DL1 and/or KIR2DS4 76. Statistical analyses. HLA-C allele frequencies, the genotypes of the KIR genes and KIR-HLA-C pair fre- quency in endometriosis patients and control women were compared using the Fisher’s exact test. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using the GraphPad Prism software (California, USA). P value <0.05 was considered statistically significant. Multiple tests were analyzed by the Bonferroni correc- tion using the GraphPad Prism software. The normality was analyzed by the Kolmogorov–Smirnov test using IBM SPSS statistics version 22 (New Y ork, USA). The continuous variables of patient demographic results were analyzed by the Mann–Whitney test using the GraphPad Prism software. The discontinuous variable of dys- menorrhea was analyzed by χ 2 test using the GraphPad Prism software. The statistical power was analyzed by the G*Power version 3.1.9.477. The χ2 value was calculated by Hardy–Weinberg analysis (χ 2 > 3.841 showed the subgroup was deviating from the Hardy–Weinberg equilibrium). Received: 26 September 2019; Accepted: 24 February 2020; Published: xx xx xxxx

1. Rogers, P . A. et al. Priorities for endometriosis research: recommendations from an international consensus workshop. Reprod. Sci. 16, 335–346, https://doi.org/10.1177/1933719108330568 (2009). 2. Giudice, L. C. Clinical practice. Endometriosis. N. Engl. J. Med. 362, 2389–2398, https://doi.org/10.1056/NEJMcp1000274 (2010). 3. Giudice, L. C. & Kao, L. C. Endometriosis. Lancet 364, 1789–1799, https://doi.org/10.1016/S0140-6736(04)17403-5 (2004). 4. Nnoaham, K. E. et al. Impact of endometriosis on quality of life and work productivity: a multicenter study across ten countries. Fertil. Steril. 96, 366–373 e368, https://doi.org/10.1016/j.fertnstert.2011.05.090 (2011). 5. Simpson, J. L., Elias, S., Malinak, L. R. & Buttram, V . C. Jr. Heritable aspects of endometriosis. I. Genetic studies. Am. J. Obstet. Gynecol. 137, 327–331 (1980). 6. Moen, M. H. & Magnus, P . The familial risk of endometriosis. Acta Obstet. Gynecol. Scand. 72, 560–564 (1993). 7. Kennedy, S., Mardon, H. & Barlow, D. Familial endometriosis. J. Assist. Reprod. Genet. 12, 32–34 (1995). 8. Treloar, S. A., O’Connor, D. T., O’Connor, V . M. & Martin, N. G. Genetic influences on endometriosis in an Australian twin sample. [email protected]. Fertil. Steril. 71, 701–710 (1999). 9. Bischoff, F . Z. & Simpson, J. L. Heritability and molecular genetic studies of endometriosis. Hum. Reprod. Update 6, 37–44 (2000). 10. Senturk, L. M. & Arici, A. Immunology of endometriosis. J. Reprod. Immunol. 43, 67–83 (1999). 11. Gleicher, N., el-Roeiy, A., Confino, E. & Friberg, J. Is endometriosis an autoimmune disease? Obstet. Gynecol. 70, 115–122 (1987). 12. Gleicher, N. Autoimmunity and reproductive failure. Ann. N. Y. Acad. Sci. 626, 537–544 (1991). 13. Ho, H. N. et al. Peritoneal natural killer cytotoxicity and CD25+ CD3+ lymphocyte subpopulation are decreased in women with stage III-IV endometriosis. Hum. Reprod. 10, 2671–2675 (1995). 14. Vigano, P . et al. Deficient antie ndometrium lymphocyte-mediated cytotoxicity in patients with endometriosis. Fertil. Steril. 56, 894–899 (1991). 15. Oosterlynck, D. J., Cornillie, F . J., Waer, M., Vandeputte, M. & Koninckx, P . R. Women with endometriosis show a defect in natural killer activity resulting in a decreased cytotoxicity to autologous endometrium. Fertil. Steril. 56, 45–51 (1991). 16. Oosterlynck, D. J., Meuleman, C., Waer, M., Vandeputte, M. & Koninckx, P . R. The natural killer activity of peritoneal fluid lymphocytes is decreased in women with endometriosis. Fertil. Steril. 58, 290–295 (1992). 17. Nothnick, W . B. Treating endometriosis as an autoimmune disease. Fertil. Steril. 76, 223–231 (2001). 18. Witz, C. A. Pathogenesis of endometriosis. Gynecol. Obstet. Invest. 53(Suppl 1), 52–62 (2002). 19. Shiina, T., Hosomichi, K., Inoko, H. & Kulski, J. K. The HLA genomic loci map: expression, interaction, diversity and disease. J. Hum. Genet. 54, 15–39, https://doi.org/10.1038/jhg.2008.5 (2009). 20. Matzaraki, V ., Kumar, V ., Wijmenga, C. & Zhernakova, A. The MHC locus and genetic susceptibility to autoimmune and infectious diseases. Genome Biol. 18, 76, https://doi.org/10.1186/s13059-017-1207-1 (2017). 21. Zhou, F . et al. Deep sequencing of the MHC region in the Chinese population contributes to studies of complex disease. Nat. Genet. 48, 740–746, https://doi.org/10.1038/ng.3576 (2016). 22. Blais, M. E., Dong, T. & Rowland-Jones, S. HLA-C as a mediator of natural killer and T-cell activation: spectator or key player? Immunology 133, 1–7, https://doi.org/10.1111/j.1365-2567.2011.03422.x (2011). 23. Collins, K. L., Chen, B. K., Kalams, S. A., Walker, B. D. & Baltimore, D. HIV-1 Nef protein protects infected primary cells against killing by cytotoxic T lymphocytes. Nat. 391, 397–401, https://doi.org/10.1038/34929 (1998). 24. Fellay, J. et al. A whole-genome association study of major determinants for host control of HIV-1. Sci. 317, 944–947, https://doi. org/10.1126/science.1143767 (2007). 25. Ho, P . Y . et al. Investigating the role of the HLA-Cw*06 and HLA-DRB1 genes in susceptibility to psoriatic arthritis: comparison with psoriasis and undifferentiated inflammatory arthritis. Ann. Rheum. Dis. 67, 677–682, https://doi.org/10.1136/ard.2007.071399 (2008). 26. Simmonds, M. J. et al. A novel and major association of HLA-C in Graves’ disease that eclipses the classical HLA-DRB1 effect. Hum. Mol. Genet. 16, 2149–2153, https://doi.org/10.1093/hmg/ddm165 (2007). 8Scientific RepoRtS | (2020) 10:4897 | https://doi.org/10.1038/s41598-020-61702-y www.nature.com/scientificreportswww.nature.com/scientificreports/ 27. Vivier, E. et al . Innate or adaptive immunity? The example of natural killer cells. Sci. 331, 44–49, https://doi.org/10.1126/ science.1198687 (2011). 28. Paul, S. & Lal, G. The Molecular Mechanism of Natural Killer Cells Function and Its Importance in Cancer Immunotherapy. Front. Immunol. 8, 1124, https://doi.org/10.3389/fimmu.2017.01124 (2017). 29. Middleton, D. & Gonzelez, F . The extensive polymorphism of KIR genes. Immunology 129, 8–19, https://doi.org/10.1111/j.1365- 2567.2009.03208.x (2010). 30. Falco, M., Moretta, L., Moretta, A. & Bottino, C. KIR and KIR ligand polymorphism: a new area for clinical applications? Tissue Antigens 82, 363–373, https://doi.org/10.1111/tan.12262 (2013). 31. Kulkarni, S., Martin, M. P . & Carrington, M. The Yin and Y ang of HLA and KIR in human disease. Semin. Immunol. 20, 343–352, https://doi.org/10.1016/j.smim.2008.06.003 (2008). 32. Rajalingam, R. Human diversity of killer cell immunoglobulin-like receptors and disease. Korean J. Hematol. 46, 216–228, https:// doi.org/10.5045/kjh.2011.46.4.216 (2011). 33. Ruggeri, L. et al. Donor natural killer cell allorecognition of missing self in haploidentical hematopoietic transplantation for acute myeloid leukemia: challenging its predictive value. Blood 110, 433–440, https://doi.org/10.1182/blood-2006-07-038687 (2007). 34. Graef, T. et al. KIR2DS4 is a product of gene conversion with KIR3DL2 that introduced specificity for HLA-A*11 while diminishing avidity for HLA-C. J. Exp. Med. 206, 2557–2572, https://doi.org/10.1084/jem.20091010 (2009). 35. Cella, M., Longo, A., Ferrara, G. B., Strominger, J. L. & Colonna, M. NK3-specific natural killer cells are selectively inhibited by Bw4- positive HLA alleles with isoleucine 80. J. Exp. Med. 180, 1235–1242, https://doi.org/10.1084/jem.180.4.1235 (1994). 36. Cariani, E. et al . HLA and killer immunoglobulin-like receptor genes as outcome predictors of hepatitis C virus-related hepatocellular carcinoma. Clin. Cancer Res. 19, 5465–5473, https://doi.org/10.1158/1078-0432.CCR-13-0986 (2013). 37. Y en, J. H. et al. Major histocompatibility complex class I-recognizing receptors are disease risk genes in rheumatoid arthritis. J. Exp. Med. 193, 1159–1167 (2001). 38. Martin, M. P . et al. Cutting edge: susceptibility to psoriatic arthritis: influence of activating killer Ig-like receptor genes in the absence of specific HLA-C alleles. J. Immunol. 169, 2818–2822 (2002). 39. van der Slik, A. R. et al . KIR in type 1 diabetes: disparate distribution of activating and inhibitory natural killer cell receptors in patients versus HLA-matched control subjects. Diabetes 52, 2639–2642 (2003). 40. Shastry, A. et al. Combination of KIR 2DL2 and HLA-C1 (Asn 80) confers susceptibility to type 1 diabetes in Latvians. Int. J. Immunogenet. 35, 439–446, https://doi.org/10.1111/j.1744-313X.2008.00804.x (2008). 41. Middleton, D., Diler, A. S., Meenagh, A., Sleator, C. & Gourraud, P . A. Killer immunoglobulin-like receptors (KIR2DL2 and/or KIR2DS2) in presence of their ligand (HLA-C1 group) protect against chronic myeloid leukaemia. Tissue Antigens 73, 553–560, https://doi.org/10.1111/j.1399-0039.2009.01235.x (2009). 42. Boyton, R. J. et al. HLA-C and killer cell immunoglobulin-like receptor genes in idiopathic bronchiectasis. Am. J. Respir. Crit. Care Med. 173, 327–333, https://doi.org/10.1164/rccm.200501-124OC (2006). 43. Montgomery, G. W . et al. The search for genes contributing to endometriosis risk. Hum. Reprod. Update 14, 447–457, https://doi. org/10.1093/humupd/dmn016 (2008). 44. Tariverdian, N. et al . Neuroendocrine-immune disequilibrium and endometriosis: an interdisciplinary approach. Semin. Immunopathol. 29, 193–210, https://doi.org/10.1007/s00281-007-0077-0 (2007). 45. Matarese, G., De Placido, G., Nikas, Y . & Alviggi, C. Pathogenesis of endometriosis: natural immunity dysfunction or autoimmune disease? Trends Mol. Med. 9, 223–228 (2003). 46. Moen, M., Bratlie, A. & Moen, T. Distribution of HLA-antigens among patients with endometriosis. Acta Obstet. Gynecol. Scand. Suppl. 123, 25–27 (1984). 47. Simpson, J. L., Malinak, L. R., Elias, S., Carson, S. A. & Radvany, R. A. HLA associations in endometriosis. Am. J. Obstet. Gynecol. 148, 395–397 (1984). 48. Maxwell, C., Kilpatrick, D. C., Haining, R. & Smith, S. K. No HLA-DR specificity is associated with endometriosis. Tissue Antigens 34, 145–147 (1989). 49. Ishii, K. et al. Studies on the human leukocyte antigen class I antigens in Japanese patients with macroscopically diagnosed endometriosis. Gynecol Obstet Invest 54, 150–153, 67882 (2002). 50. Ishii, K., Takakuwa, K., Mitsui, T. & Tanaka, K. Studies on the human leukocyte antigen-DR in patients with endometriosis: genotyping of HLA-DRB1 alleles. Hum. Reprod. 17, 560–563 (2002). 51. Ishii, K., Takakuwa, K., Kashima, K., Tamura, M. & Tanaka, K. Associations between patients with endometriosis and HLA class II; the analysis of HLA-DQB1 and HLA-DPB1 genotypes. Hum. Reprod. 18, 985–989 (2003). 52. Kitawaki, J. et al. Association of HLA class I and class II alleles with susceptibility to endometriosis. Hum. Immunol. 63, 1033–1038 (2002). 53. Roszkowski, P . I. et al. Susceptibility to ovarian endometriosis in Polish population is not associated with HLA-DRB1 alleles. Hum. Reprod. 20, 970–973, https://doi.org/10.1093/humrep/deh715 (2005). 54. Wang, X., Lin, Q. & Guo, S. Study on polymorphism of human leukocyte antigen I in patients with endometriosis. Zhonghua Fu Chan Ke Za Zhi 36, 150–152 (2001). 55. Wang, X. et al. Study on polymorphism of human leukocyte antigen-DRB1 allele in patients with endometriosis. Zhonghua Fu Chan Ke Za Zhi 37, 346–348 (2002). 56. Zong, L., He, Y . & Pan, D. Association of human leukocyte antigen-DQA1 with endometriosis of women in southern China. Zhonghua Fu Chan Ke Za Zhi 36, 405–407 (2001). 57. Kitawaki, J. et al. Association of killer cell immunoglobulin-like receptor genotypes with susceptibility to endometriosis. Am. J. Reprod. Immunol. 58, 481–486, https://doi.org/10.1111/j.1600-0897.2007.00533.x (2007). 58. Nowak, I. et al. Does the KIR2DS5 gene protect from some human diseases? PLoS One 5, e12381, https://doi.org/10.1371/journal. pone.0012381 (2010). 59. Nowak, I. et al. KIR2DS5 in the presence of HLA-C C2 protects against endometriosis. Immunogenetics 67, 203–209, https://doi. org/10.1007/s00251-015-0828-3 (2015). 60. Single, R. M., Martin, M. P ., Meyer, D., Gao, X. & Carrington, M. Methods for assessing gene content diversity of KIR with examples from a global set of populations. Immunogenetics 60, 711–725, https://doi.org/10.1007/s00251-008-0331-1 (2008). 61. Moesta, A. K. & Parham, P . Diverse functionality among human NK cell receptors for the C1 epitope of HLA-C: KIR2DS2, KIR2DL2, and KIR2DL3. Front. Immunol. 3, 336, https://doi.org/10.3389/fimmu.2012.00336 (2012). 62. Gourraud, P . A., Meenagh, A., Cambon-Thomsen, A. & Middleton, D. Linkage disequilibrium organization of the human KIR superlocus: implications for KIR data analyses. Immunogenetics 62, 729–740, https://doi.org/10.1007/s00251-010-0478-4 (2010). 63. NurWaliyuddin, H. Z. et al. Ancient Genetic Signatures of Orang Asli Revealed by Killer Immunoglobulin-Like Receptor Gene Polymorphisms. PLoS One 10, e0141536, https://doi.org/10.1371/journal.pone.0141536 (2015). 64. Tanaka, E., Sendo, F ., Kawagoe, S. & Hiroi, M. Decreased natural killer cell activity in women with endometriosis. Gynecol. Obstet. Invest. 34, 27–30 (1992). 65. Jeung, I., Cheon, K. & Kim, M. R. Decreased Cytotoxicity of Peripheral and Peritoneal Natural Killer Cell in Endometriosis. Biomed. Res. Int. 2016, 2916070, https://doi.org/10.1155/2016/2916070 (2016). 9Scientific RepoRtS | (2020) 10:4897 | https://doi.org/10.1038/s41598-020-61702-y www.nature.com/scientificreportswww.nature.com/scientificreports/ 66. Matsuoka, S. et al. Expression of inhibitory-motif killer immunoglobulin-like receptor, KIR2DL1, is increased in natural killer cells from women with pelvic endometriosis. Am. J. Reprod. Immunol. 53, 249–254, https://doi.org/10.1111/j.1600-0897.2005.00271.x (2005). 67. Oosterlynck, D. J., Meuleman, C., Lacquet, F . A., Waer, M. & Koninckx, P . R. Flow cytometry analysis of lymphocyte subpopulations in peritoneal fluid of women with endometriosis. Am. J. Reprod. Immunol. 31, 25–31 (1994). 68. Berkkanoglu, M. & Arici, A. Immunology and endometriosis. Am. J. Reprod. Immunol. 50, 48–59 (2003). 69. Zhang, C. et al. Killer immunoglobulin-like receptor and human leukocyte antigen expression as immunodiagnostic parameters for pelvic endometriosis. Am. J. Reprod. Immunol. 55, 106–114, https://doi.org/10.1111/j.1600-0897.2005.00332.x (2006). 70. Herington, J. L., Bruner-Tran, K. L., Lucas, J. A. & Osteen, K. G. Immune interactions in endometriosis. Expert. Rev. Clin. Immunol. 7, 611–626, https://doi.org/10.1586/eci.11.53 (2011). 71. Dias, J. A. Jr. et al. Patients with endometriosis of the rectosigmoid have a higher percentage of natural killer cells in peripheral blood. J. Minim. Invasive Gynecol. 19, 317–324, https://doi.org/10.1016/j.jmig.2011.12.021 (2012). 72. Wu, M. Y . et al. Increase in the expression of killer cell inhibitory receptors on peritoneal natural killer cells in women with endometriosis. Fertil. Steril. 74, 1187–1191 (2000). 73. Drury, J. A. et al. The dynamic changes in the number of uterine natural killer cells are specific to the eutopic but not to the ectopic endometrium in women and in a baboon model of endometriosis. Reprod. Biol. Endocrinol. 16, 67, https://doi.org/10.1186/s12958- 018-0385-3 (2018). 74. Thiruchelvam, U., Wingfield, M. & O’Farrelly, C. Natural Killer Cells: Key Players in Endometriosis. Am. J. Reprod. Immunol. 74, 291–301, https://doi.org/10.1111/aji.12408 (2015). 75. Cooley, S. et al. Donors with group B KIR haplotypes improve relapse-free survival after unrelated hematopoietic cell transplantation for acute myelogenous leukemia. Blood 113, 726–732, https://doi.org/10.1182/blood-2008-07-171926 (2009). 76. Umemura, T. et al. KIR3DL1-HLA-Bw4 combination and IL28B polymorphism predict response to Peg-IFN and ribavirin with and without telaprevir in chronic hepatitis C. Hum. Immunol. 75, 822–826, https://doi.org/10.1016/j.humimm.2014.06.003 (2014). 77. Faul, F ., Erdfelder, E., Buchner, A. & Lang, A. G. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav. Res. Methods 41, 1149–1160, https://doi.org/10.3758/BRM.41.4.1149 (2009).

This work was supported by the Ministry of Science and Technology (grant number 104-2314-B-038- 063-MY2, grant number 106-2314-B-038-072, grant number 107-2314-B-038-006, grant number 108-2314- B-038-003), Academia Sinica (grant number BM10501010036, grant number BM10601010024, grant number BM10701010027), National Health Research Institute (grant number MG-105-SP-07, grant number MG- 106-SP-07, grant number MG-107-SP-07) (CRT), and Ministry of Science and Technology (grant number 107- 2314-B-009-006) (YCC). This work was financially supported by the Center for Intelligent Drug Systems and Smart Bio-devices (IDS 2B) from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan. Author contributions Y .-C.C. designed the study, performed experiments, analysed the data and wrote the manuscripts, C.-H.C., M.-J.C., C.-W .C., P .-H.C, M.-H.Y ., Y .-J.C., E.-M.T., P .-S.Y . and S.-Y .L. enrolled patients, and C.-R.T. guided the experimental design, enrolled patients and wrote the manuscripts. competing interests The authors declare no competing interests. Additional information Correspondence and requests for materials should be addressed to C.-R.T. Reprints and permissions information is available at www.nature.com/reprints. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2020