Sex differences in the genetics of sarcoidosis across European and African ancestry populations

med-10-1132799 May 4, 2023 Time: 14:33 # 1 TYPE O iginal Resea ch PUBLISHED 11 May 2023 DOI 10.3389/ med.2023.1132799 OPEN ACCESS EDITED BY Ying Zhou, Tongji Uni e si y, China REVIEWED BY Humbe o Ga cia-O iz, Na ional Ins i u e o Genomic Medicine (INMEGEN), Mexico Ilias C. Papanikolaou, Gene al Hospi al o Co u, G eece *CORRESPONDENCE Na alia V. Ri e a [email p o ec ed] RECEIVED 27 Decembe 2022 ACCEPTED 10 Ap il 2023 PUBLISHED 11 May 2023 CITATION Xiong Y, Kullbe g S, Ga man L, Pezan N, Ellinghaus D, Vasila V, Eklund A, Rybicki BA, Iannuzzi MC, Sch eibe S, Mülle -Que nheim J, Mon gome y CG, G unewald J, Padyuko L and Ri e a NV (2023) Sex di e ences in he gene ics o sa coidosis ac oss Eu opean and A ican ances y popula ions. F on . Med. 10:1132799. doi: 10.3389/ med.2023.1132799 COPYRIGHT © 2023 Xiong, Kullbe g, Ga man, Pezan , Ellinghaus, Vasila, Eklund, Rybicki, Iannuzzi, Sch eibe , Mülle -Que nheim, Mon gome y, G unewald, Padyuko and Ri e a. This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (CC BY). The use, dis ibu ion o ep oduc ion in o he o ums is pe mi ed, p o ided he o iginal au ho (s) and he copy igh owne (s) a e c edi ed and ha he o iginal publica ion in his jou nal is ci ed, in acco dance wi h accep ed academic p ac ice. No use, dis ibu ion o ep oduc ion is pe mi ed which does no comply wi h hese e ms. Sex di e ences in he gene ics o sa coidosis ac oss Eu opean and A ican ances y popula ions Ying Xiong1, Susanna Kullbe g1,2, Lo i Ga man3, Na han Pezan 3, Da id Ellinghaus4, Vasiliki Vasila1, Ande s Eklund2, Benjamin A. Rybicki5, Michael C. Iannuzzi6, S e an Sch eibe 4,7, Joachim Mülle -Que nheim8, Cou ney G. Mon gome y3, Johan G unewald1,2,9, Leonid Padyuko 9,10 and Na alia V. Ri e a1,9,10* 1Respi a o y Medicine Di ision, Depa men o Medicine Solna, Ka olinska Ins i u e , S ockholm, Sweden, 2Depa men o Respi a o y Medicine and Alle gy, Theme In lamma ion and Ageing, Ka olinska Uni e si y Hospi al, S ockholm, Sweden, 3Genes and Human Disease, Oklahoma Medical Resea ch Founda ion, Oklahoma Ci y, OK, Uni ed S a es, 4Ins i u e o Clinical Molecula Biology, Ch is ian-Alb ech s-Uni e si y o Kiel, Kiel, Ge many, 5Depa men o Public Heal h Sciences, Hen y Fo d Heal h Sys em, De oi , MI, Uni ed S a es, 6Zucke School o Medicine, S a en Island Uni e si y Hospi al, No hwell/Ho s a Uni e si y, S a en Island, NY, Uni ed S a es, 7Clinic o In e nal Medicine I, Uni e si y Hospi al Schleswig-Hols ein, Kiel, Ge many, 8Depa men o Pneumology, Uni e si y Medical Cen e F eibu g, Facul y o Medicine, Uni e si y o F eibu g, F eibu g im B eisgau, Ge many, 9Cen e o Molecula Medicine, Ka olinska Ins i u e , S ockholm, Sweden, 10Di ision o Rheuma ology, Depa men o Medicine Solna, Ka olinska Ins i u e , Ka olinska Uni e si y Hospi al, S ockholm, Sweden Backg ound: Sex di e ences in he suscep ibili y o sa coidosis a e unknown. The s udy aims o iden i y sex-dependen gene ic a ia ions in wo clinical sa coidosis pheno ypes: Lö g en’s synd ome (LS) and non-Lö g en’s synd ome (non-LS). Me hods: A me a-analysis o genome-wide associa ion s udies was conduc ed on Eu opeans and A ican Ame icans, o aling 10,103 indi iduals om h ee popula ion-based coho s, Sweden (n= 3,843), Ge many (n= 3,342), and he Uni ed S a es (n= 2,918), ollowed by an SNP lookup in he UK Biobank (UKB, n= 387,945). A genome-wide associa ion s udy based on Immunochip da a consis ing o 141,000 single nucleo ide polymo phisms (SNPs) was conduc ed in he sex g oups. The associa ion es was based on logis ic eg ession using he addi i e model in LS and non-LS sex g oups independen ly. Addi ionally, gene-based analysis, gene exp ession, exp ession quan i a i e ai loci (eQTL) mapping, and pa hway analysis we e pe o med o disco e unc ionally ele an mechanisms ela ed o sa coidosis and biological sex. Resul s: We iden i ied sex-dependen gene ic a ia ions in LS and non-LS sex g oups. Gene ic indings in LS sex g oups we e explici ly loca ed in he ex ended Majo His ocompa ibili y Complex (xMHC). In non-LS, gene ic di e ences in he sex g oups we e p ima ily loca ed in he MHC class II sub egion and ANXA11. Gene-based analysis and eQTL en ichmen e ealed dis inc sex-speci ic gene exp ession pa e ns in a ious issues and immune cell ypes. In LS sex g oups, a pa hway map ela ed o an igen p esen a ion machine y by IFN-gamma. In non-LS, pa hway maps ela ed o immune esponse lec in-induced complemen pa hway in males and ela ed o ma u a ion and mig a ion o dend i ic cells in skin sensi iza ion in emales we e iden i ied. F on ie s in Medicine 01 on ie sin.o g med-10-1132799 May 4, 2023 Time: 14:33 # 2 Xiong e al. 10.3389/ med.2023.1132799 Conclusion: Ou indings p o ide new e idence o a sex bias unde lying sa coidosis gene ic a chi ec u e, pa icula ly in clinical pheno ypes LS and non- LS. Biological sex likely plays a ole in disease mechanisms in sa coidosis. KEYWORDS sa coidosis, gene ics, genome-wide associa ion s udy (GWAS), me a-analysis, single nucleo ide polymo phisms, immunogene ics and HLA Backg ound Sa coidosis is a mul i-sys em disease wi h unknown e iology, cha ac e ized by non-casea ing g anulomas (1). In 90% o cases, sa coidosis a ec s he lungs and lympha ic sys em (2). Sa coidosis is p e alen wo ldwide, commonly a ec ing indi iduals be ween 20 and 50 yea s o age, wi h males diagnosed ea lie han emales (3,4). On a global scale, sa coidosis incidence and p e alence a y based on sex, age, geog aphy, and e hnici y (4). In he Uni ed S a es, he p e alence is 141 pe 100,000 in A ican–Ame icans and 49.8 pe 100,000 in Caucasians (5). In No dic coun ies such as Sweden, he p e alence o sa coidosis was epo ed be ween 152 and 215 pe 100,000 (6), whe eas in Denma k was epo ed a 77 pe 100,000 (7). The disease is he e ogeneous, as di e en mani es a ions and clinical ou comes ha e been obse ed in clinical se ings, pa icula ly among sex g oups (3,8,9). In e es ingly, sa coidosis is mo e common in emales, has a la e-onse , has a a iable disease cou se depending on he a ec ed o gan, and has a highe mo ali y a e han men (3). Sex di e ences a e also seen in ex a-pulmona y pheno ypes (10,11), including ca diac (12,13) and skin sa coidosis. Sex ho mones ha e been shown o play a ole in sa coidosis (14, 15). Fo ins ance, sa coidosis has a low disease ac i i y in p egnancy o goes in o emission. Howe e , he amelio a ing e ec is los as la es occu a e deli e y (16,17). In he lungs, sa coidosis is cha ac e ized by Lö g en’s synd ome (LS) and non-Lö g en’s synd ome (LS). LS is an acu e o m o he disease cha ac e ized by pe ia icula swelling a ound he ankles, e y hema nodosum, and bila e al hila lymphadenopa hy (18). Typically, emales wi h LS exhibi e y hema nodosum (EN), while males wi h LS ha e symp oms o bila e al ankle a h i is. Pa ien s wi h LS usually ha e a good p ognosis (19). LS o en occu s in indi iduals o Eu opean ances y and is ela i ely a e in indi iduals o A ican o igin. In non-LS, sa coidosis is cha ac e ized by a he e ogeneous disease cou se, o en wi h an insidious onse , un elen ing disease cou se, and a high isk o clinical o gan impai men , p edominan ly pulmona y ib osis (18, 20). Pa ien s wi h ib o ic sa coidosis ha e ma kedly dec eased su i al compa ed wi h he gene al popula ion (21,22). Abb e ia ions: LS, Lö g en’s synd ome; Non-LS, non-Lö g en’s synd ome; SNPs, single nucleo ide polymo phisms; eQTL, exp ession quan i a i e ai loci; UK Biobank, Uni ed Kingdom Biobank; ANXA11, Annexin A11 gene; GWAS, Genome-wide associa ion s udy; US-AA, Uni ed S a es A ican Ame icans; PC, P incipal componen ; PCA, p incipal componen analysis; IVW, in e se a iance weigh ing; OR, odds a io; CI, con idence in e al. The causes o clinical di e ences and disease cou se in sa coidosis among sex g oups a e unknown. Howe e , e idence- based ac o s, including gene ics, epigene ics, and en i onmen al exposu es, ha e been pos ula ed. Rega ding gene ics, ew s udies ha e add essed sex di e ences in he gene ic a ia ion in disease, despi e mos s udies adjus ing o sex in hei analyses (23). An excep ion is a genome-wide admix u e scan conduc ed in A ican Ame icans s a i ied by sex, iden i ying se e al gene ic a ian s associa ed wi h sa coidosis exclusi ely in emales (24). In his wo k, we sough o in es iga e he gene ic associa ions in he sex g oups and iden i y di e ences and commonali ies in Eu opean and A ican popula ion ances ies by a me a-analysis app oach and high-densi y mapping. Ma e ials and me hods Samples and s udy design The samples we e om h ee independen sa coidosis popula ion-based coho s om Sweden, Ge many, and he Uni ed S a es. All pa icipan s p o ided w i en in o med consen o he s udy. S udy p o ocols o all s udies had been app o ed by espec i e local ins i u ional boa ds. The Swedish coho The case-con ol s udy consis ed o 4,133 indi iduals. The s udy was app o ed by he local ins i u ional e iew boa d in S ockholm, Sweden. All pa icipan s p o ided w i en in o med consen and we e pe mi ed o use hei DNA o esea ch pu poses. Baseline clinical and demog aphic cha ac e is ics we e measu ed when he pa icipan s we e en olled in he s udy. Pa ien s we e en olled a he ime o disease in es iga ion a he Sa coidosis Cen um, Ka olinska Uni e si y Hospi al Solna, Sweden. The diagnosis was es ablished on adiog aphic mani es a ions, indings a b onchoscopy wi h b onchoal eola la age (BAL), including an ele a ed CD4/CD8 a io >3.5, and posi i e biopsies and in acco dance wi h he c i e ia ou lined by he Wo ld Associa ion o Sa coidosis and O he G anuloma ous Diso de s [WASOG, S a emen on sa coidosis (27)]. Sa coidosis pa ien s we e u he cha ac e ized in o wo clinical pheno ypes, LS and non-LS. LS was de ined by ypical clinical mani es a ions wi h an acu e onse o he disease, including e e , bila e al hila lymphadenopa hy on ches X- ay, bila e al ankle a h i is, and/o e y hema nodosum. Non-LS was de ined as a he e ogeneous disease cou se wi h an insidious onse , un elen ing disease cou se, and a F on ie s in Medicine 02 on ie sin.o g med-10-1132799 May 4, 2023 Time: 14:33 # 3 Xiong e al. 10.3389/ med.2023.1132799 high isk o clinical o gan impai men , such as de eloping ib osis in he lung. Heal hy con ols included 3,085 indi iduals who we e ec ui ed om wo la ge-scale epidemiological coho s. Mainly, 2,025 indi iduals we e om he En i onmen al In es iga ion o Rheuma oid a h i is (EIRA) s udy desc ibed in Kla eskog e al. (25) and 1,060 indi iduals we e om he Epidemiological in es iga ion o isk ac o s o Mul iple Scle osis (EIMS) s udy desc ibed in Heds om e al. (26). The Ge man coho The case-con ol s udy consis ed o 4,975 indi iduals, o which 413 we e non-LS (27). The desc ip ion and inclusion o hese pa ien s a e desc ibed elsewhe e (28,29). B ie ly, Ge man con ol subjec s (n= 4,498) we e de i ed om Popgen (n= 2,485) (30), and he Heinz Nixdo RECALL (HNR) s udy (n= 1,499) (31). Addi ionally, 304 con ol indi iduals o Sou h Ge man o igin we e ec ui ed om he Ba a ian Red C oss, and 210 con ol indi iduals we e ec ui ed om he Cha i é - Uni e si ä smedizin Be lin. The mean age and male pe cen age we e 62.5 yea s (SD = 12.3) and 40% male in non-LS and 57.8 yea s (SD = 12.2), and 51% male in con ols. The USA A ican Ame ican coho The case-con ol s udy consis ed o 1,657 indi iduals. The sample included 781 sa coidosis cases cha ac e ized as non-LS and 876 heal hy con ols. All indi iduals we e aken om an ex ensi e coho o A ican-Ame ican (AA) sa coidosis pa ien s and con ols assembled om a ious s udies. Fu he de ails a e a ailable in Re s. (32–35). Geno yping and quali y con ol Geno yping o sa coidosis pa ien s in he Swedish coho was pe o med using he Illumina Immunochip pla o m and was pe o med a The SNP&SEQ Technology Pla o m in Uppsala Uni e si y, Sweden, wi h Illumina In inium assay using he Immuno Bead Chip (Immunochip e sion 1) as desc ibed in Ri e a e al. (23). Heal hy con ols in he EIRA coho (n= 2,086) we e geno yped on he same pla o m a he Genome Ins i u e o Singapo e, as desc ibed p e iously (36,37). Heal hy con ols in he EIMS coho (n= 1,060) we e geno yped wi h he same SNP a ay desc ibed elsewhe e (26). B ie ly, quali y con ol il e ing h esholds we e applied using ools implemen ed in PLINK 1.09b so wa e (38,39). SNPs a call a e <95% wi h mino allele equency (MAF) >1% and call a e <99% wi h MAF ≤5% we e il e ed ou . Mo eo e , SNPs ha had Ha dy–Weinbe g Equilib ium (HWE) P<1×10−7(in he con ol g oup) we e also excluded. Indi iduals wi h missing geno ypes <97% we e also emo ed. Quali y con ol (QC) esul ed in 141,151 SNPs and 3,842 indi iduals on he Illumina Immunochip a ay. Geno ypes o he Ge man coho we e quali y con olled, as desc ibed in Fische e al. (40). Geno yping o he A ican Ame ican coho (US-AA) was pe o med a he OMRF using he Illumina HumanOmni1-Quad a ay o ∼1.1M a ian s ac oss he genome. O hese, 121,988 we e in common wi h he Immunochip pla o m and ca ied o wa d o analysis. Fu he de ails on geno yping and quali y con ol il e ing a e desc ibed in Ad ian o e al. (41). S a is ical analysis Associa ion analysis o LS and non-LS in sex g oups In each sex g oup, we examined he associa ion o single nucleo ide polymo phisms (SNPs) in LS and non-LS. Fo assessing he associa ion in LS, we applied an addi i e model using logis ic eg ession adjus ed o age and ou p incipal componen s (PCs). PCs we e de i ed om p incipal componen analysis (PCA) pe o med using EIGENSTRAT (42) so wa e on a p uned geno yped da a se . The p uning o geno ypes was pe o med using he p uning unc ion and de aul pa ame e s implemen ed in PLINK 1.90 be a so wa e (39). Fo assessing he associa ion in non-LS, we applied an addi i e model using logis ic eg ession wi hou co a ia e adjus men due o he lack o age in o ma ion ac oss he coho s. In LS and non-LS, a signi icance h eshold was de ined by genomic con ol-co ec ed P<3.5 ×10−7 (0.05/141,151 quali y-con olled SNPs) conside ing Bon e oni co ec ion. A sugges i e P<5×10−5was also conside ed o iden i y po en ial signals (43–46). Fu he mo e, o accoun o high linkage disequilib ium (LD) ( 2≥0.8) among associa ed SNPs, LD p uning was pe o med using P io i yP une e sion 0.01.4 so wa e1wi h de aul pa ame e s, which iden i ied and p io i ized SNPs in genomic loci and hus e e ed as ag-SNPs. As an explo a o y analysis, o assess he e ec o sex on gene ics, we implemen ed an in e ac ion model whe e he disease ou come was eg essed on SNP adjus ed o sex and he in e ac ion e m (SNP ×sex) as co a ia es. A signi ican h eshold o he in e ac ion e m was se a P<0.05. The in e ac ion analysis was pe o med using he glm unc ion implemen ed in PLINK 2.0 so wa e (39). Addi ionally, we also pe o med a he e ogenei y es o assess he e ec s o associa ion be ween males and emales. He e ogenei y s a is ics i.e., Coch an’s Q-s a is ic, I2s a is ic ( he pe cen age o a iabili y in he e ec sizes), and au2( he be ween-s udy a iance in ou me a-analysis) we e compu ed applying a andom e ec s model and using he ma unc ion implemen ed in he me a o R package (47). Me a-analysis o signi ican SNPs Me a-analysis was pe o med using SNP associa ion esul s in non-LS. He ein, he me a-analysis was conduc ed using he in e se a iance weigh ing (IVW) me hod implemen ed in he METAL so wa e (48). Fo each SNP, he combined gene ic e ec size (de ined by Be a), s anda d e o (SE), me a-P- alue (Pme a), o al a iabili y in e ec size, also known as he he e ogenei y index (I2), and he e ogenei y P- alue we e calcula ed. SNPs a Pme a <5×10−8we e de ined as genome-wide signi ican . Two me a-analyses we e conduc ed, (1) a me a-analysis conside ing SNP associa ions in he Eu opean ances y coho s (Sweden and Ge many) and (2) a me a-analysis conside ing SNP associa ions in he mul i-e hnic (Sweden, Ge many, and US A ican Ame ican). Besides METAL compu ing he e ogenei y index (I2), we also pe o med an independen he e ogenei y es applying he andom e ec s model using he ma unc ion implemen ed in he me a o 1h p://p io i yp une .sou ce o ge.ne / F on ie s in Medicine 03 on ie sin.o g med-10-1132799 May 4, 2023 Time: 14:33 # 4 Xiong e al. 10.3389/ med.2023.1132799 R package (47). Summa y o he e ogenei y s a is ics include Coch an’s Q-s a is ic, I2s a is ic ( he pe cen age o a iabili y in he e ec sizes), and au2( he be ween-s udy a iance in ou me a- analysis). Lookup o signi ican SNPs in he UK Biobank Using he UKB esou ce, we examined SNP associa ions a P<5×10−5using summa y s a is ics o a genome- wide associa ion s udy (GWAS) on “doc o -diagnosed sa coidosis” (Da a-Field ID: 22133), consis ing o 91,787 indi iduals (395 cases and 91,392 con ols) in he UKBB (49). Chie ly, sex- s a i ied summa y s a is ics GWAS on males (189 cases and 41,045 con ols) and emales (206 cases and 50,347 con ols) we e ob ained om h ps://gi hub.com/Nealelab/UK_Biobank_GWAS. GWAS analysis was conduc ed using an addi i e model and logis ic eg ession adjus ed by he i s 20 PCs, age, and age2. A lookup signi icance h eshold was se a P<5×10−8. Gene-based analysis A gene-based analysis was conduc ed using SNP associa ions a P<5×10−5and MAGMA (50). Gene-based signi icance was se o P<2.0 ×10−6based on Bon e oni co ec ion (0.05/24,769 genes). The numbe o genes was adop ed om he VEGAS2 so wa e de ini ion, whe e 24,769 unique genes on he 22 au osomes we e iden i ied (51). SNP en ichmen analysis ia exp ession quan i a i e ai loci (eQTLs) and pa hway analysis Exp ession quan i a i e ai loci en ichmen was pe o med using SNP associa ions a P<5×10−5and gene exp ession da a om 53 human issues om he Geno ype-Tissue Exp ession (GTEx) p ojec e sion 8 a ailable on he Func ional Mapping and Anno a ion o gene ic associa ions (FUMA) web ool (52). Addi ional eQTL en ichmen using gene exp ession o lung issue, whole blood, and immune cell ypes was conduc ed using EUGENE so wa e (53,54). A signi ican h eshold o n SNP co ela ed wi h gene exp ession (de ined as eQTL-SNP o eSNP) was se a P<0.05. Pa hway analysis was conduc ed based on SNP associa ions a P<5×10−8 o elucida e disease mechanisms using Me aCo eTM so wa e. Resul s The summa y o pheno ypes and he numbe o samples in he sex g oups o coho s in es iga ed a e shown in Table 1. Associa ion esul s The genome-wide associa ion analysis using high-densi y mapping Immunochip a ay was conduc ed in males and emales independen ly in Swedish, Ge man, and US-A ican Ame ican coho s. TABLE 1 Sample size in associa ion analysis in Sweden, Ge many, and Uni ed S a es (A); sample size in eplica ion lookup in he UK Biobank (B). (A) Coho s LS Non-LS Males Females Males Females Swedish 166 cases s. 849 HC 132 cases s. 2,236 HC 263 cases s. 855 HC 181 cases s. 2,245 HC Ge man 470 cases s. 1,750 HC 732 cases s. 1,692 HC USA-AA 322 cases s. 497 HC 951 cases s. 1,148 HC (B) Males Females 344 cases s. 166,644 con ols 361 cases and 193,792 con ols HC, heal hy con ols. LS sex g oups In he Swedish coho (Supplemen a y Tables 1,2), 542 SNPs in LS males and 617 SNPs in LS emales we e iden i ied a P<3.5 ×10−7. Rs2187668 (OR = 4.715, SE = 0.1983, P= 6.02 ×10−14) loca ed in HLA-DQA1 was he op signal iden i ied in males, and s9268219 (OR = 3.667, SE = 0.1586, P= 3.33 ×10−15) loca ed in C6o 10 was he op signal iden i ied in emales. A e conside ing linkage disequilib ium (LD) among associa ed signals, 77 SNPs in males and 101 SNPs in emales emained, which we e de ined as ag-SNPs. The op esul s o LS sex g oups (Table 2) a e illus a ed in he o es plo (Figu e 1). Manha an and QQ plo s a e shown in Supplemen a y Figu e 1. He e ogenei y s a is ics a e a ailable in Supplemen a y Table 2X. Condi ioning o he op SNP s2187668 in males iden i ied 12 SNPs a P<5×10−5. The mos signi ican SNP was s9296079 (OR = 7.476, SE = 0.410, P= 9.51 ×10−7), loca ed a 7.8 kb 50o HLA-DPB2. Simila ly, condi ioning o s9268219 in emales iden i ied 8 SNPs a P<5×10−5. The mos signi ican was s29243 (OR = 10.980, SE = 0.514, P= 6.04 ×10−6) loca ed in GABBR1. Non-MHC SNPs a P<5×10−5we e iden i ied nea LRRTM4 in males and CAST, LNPEP, TNIP1, CSMD1, B4GALNT1, SBNO2, and ABCG1 in emales (Supplemen a y Table 3). Non-LS sex g oups In he Swedish coho (Supplemen a y Tables 4,5), 5 SNPs in males a P<3.5 ×10−7and 1 SNP in emales a P= 6.73 ×10−7 we e iden i ied. The op signals we e s1049550 (OR = 0.5231, SE = 0.1126, P= 1.82 ×10−8) loca ed in ANXA11 in males and s1964995 (OR = 0.5314, SE = 0.1198, P= 6.73 ×10−7) loca ed a 36 kb o he 30o HLA-DRB5, in emales. A e LD assessmen , 3 SNPs in males de ined as ag-SNPs emained. In he Ge man coho (Supplemen a y Tables 6,7), s1964995 (OR = 0.5452, SE = 0.08223, P= 3.40 ×10−11) in males and s4502931 (OR = 0.5899, SE = 0.06549, P= 4.36 ×10−14) in emales we e iden i ied as he op signals. In he US A ican Ame ican coho (Supplemen a y Tables 8,9), s9271640 (OR = 1.771, SE = 0.1097, P= 3.62 ×10−7) in males and s1964995 (OR = 0.5452, SE = 0.08223, P= 3.40 ×10−11) in emales we e iden i ied as he op signals. F on ie s in Medicine 04 on ie sin.o g med-10-1132799 May 4, 2023 Time: 14:33 # 5 Xiong e al. 10.3389/ med.2023.1132799 TABLE 2 The op ag-SNPs (PGC <3.5 ×10−7) associa ed wi h LS males and emales in he Swedish coho . SNP CHR BP (hg19) Alleles Re Seq genes Sex g oup CA CAF OR 95% CI PGC s2187668 6 32,605,884 A/G HLA-DQA1 LS-M A 0.3557 4.7150 (3.1966, 6.9547) 6.02E-14 LS-F A 0.3557 3.4630 (2.5643, 4.6768) 6.57E-15 s3130288 6 32,096,001 A/C ATF6B LS-M A 0.3456 4.4560 (3.0489, 6.5124) 1.29E-13 LS-F 0.3456 3.5090 (2.58, 4.7725) 1.39E-14 s2105902 6 32,395,698 A/T 12kb 50o HLA-DRA LS-M A 0.4010 4.2330 (2.9272, 6.1214) 1.87E-13 LS-F 0.4010 3.1160 (2.32, 4.1851) 3.76E-13 s3135382 6 32,383,441 C/A 8.5kb 50o BTNL2 LS-M C 0.4446 3.9120 (2.749, 5.567) 3.47E-13 LS-F 0.4460 2.9310 (2.2128, 3.8823) 5.50E-13 s3131643 6 31,442,782 A/G 2.6kb 30o HCG26 LS-M A 0.3788 3.9640 (2.7471, 5.7201) 1.59E-12 LS-F 0.3788 2.8890 (2.1476, 3.8864) 1.57E-11 s4143332 6 31,348,365 A/G 19kb 50o MICA LS-M A 0.3389 4.1600 (2.8303, 6.1145) 3.32E-12 LS-F 0.3389 3.3970 (2.5095, 4.5985) 2.66E-14 s2071278 6 32,165,444 G/A NOTCH4 LS-M G 0.3725 3.8000 (2.6308, 5.4888) 8.55E-12 LS-F 0.3725 3.1160 (2.3032, 4.2156) 1.36E-12 s3094049 6 30,359,360 A/G 45kb 30o RPP21 LS-M A 0.2903 4.1510 (2.7873, 6.1819) 1.78E-11 LS-F 0.2903 3.2150 (2.3198, 4.4557) 1.53E-11 s2524069 6 31,244,789 T/A 4.9kb 50o HLA-C LS-M T 0.3540 3.6090 (2.5168, 5.1752) 2.11E-11 LS-F 0.3540 2.7470 (2.0481, 3.6845) 8.72E-11 s2736157 6 31,600,820 G/A PRRC2A LS-M G 0.4161 3.4380 (2.4292, 4.8657) 2.22E-11 LS-F 0.4161 2.7580 (2.0587, 3.6949) 6.19E-11 s9268219 6 32,284,108 C/A C6o 10 LS-M C 0.3356 4.4090 (3.0003, 6.4792) 4.10E-13 LS-F 0.3356 3.6670 (2.6872, 5.004) 3.33E-15 s1634721 6 30,977,680 A/G PBMUCL1 LS-M A 0.3221 3.4810 (2.372, 5.1085) 9.45E-10 LS-F 0.3221 3.5930 (2.6382, 4.8934) 6.08E-15 s3132449 6 31,626,013 A/G 25bp 30o APOM LS-M A 0.3418 3.9780 (2.7203, 5.8173) 8.08E-12 LS-F 0.3418 3.4680 (2.5569, 4.7038) 1.47E-14 s1064627 6 30,698,541 G/A FLOT1 LS-M G 0.354 2.9360 (2.0263, 4.2541) 4.61E-08 LS-F 0.354 3.3110 (2.4618, 4.4532) 2.64E-14 s9266669 6 31,348,077 A/G 19kb 50o MICA LS-M A 0.3625 3.3530 (2.3232, 4.8393) 5.56E-10 LS-F 0.3625 3.3570 (2.4872, 4.531) 2.73E-14 s886423 6 30,782,205 G/C 70kb 50o DDR1 LS-M G 0.3322 2.8220 (1.9374, 4.1107) 2.11E-07 LS-F 0.3322 3.2850 (2.4358, 4.4303) 6.69E-14 s3129963 6 32,380,208 G/A 5.3kb 50o BTNL2 LS-M G 0.4513 3.7770 (2.6521, 5.3791) 1.53E-12 LS-F 0.4513 3.0010 (2.2666, 3.9734) 1.61E-13 s2233974 6 31,080,016 C/G C6o 15 LS-M C 0.3793 3.3470 (2.325, 4.8184) 4.47E-10 LS-F 0.3793 3.1680 (2.3592, 4.2542) 1.66E-13 s3130477 6 31,428,920 G/A 2kb 50o HCP5 LS-M G 0.3389 4.0890 (2.7858, 6.0019) 5.00E-12 LS-F 0.3389 3.2220 (2.3816, 4.359) 2.91E-13 SNP, single nucleo ide polymo phism; BP, ch omosome base pai s a e based on human genome assembly e sion 19 (hg19); CA, coded allele; CAF, coded allele equency; OR, odds a io; SE, s anda d e o o odds a io; PGC, genomic con olled P- alue. The op associa ion esul s o each coho a e summa ized in Table 3. Manha an and QQ plo s o non-LS sex g oups a e shown in Supplemen a y Figu es 2–4. In e ac ion analysis wi h sex Resul s om he in e ac ion analysis in he LS and non- LS sex g oups in he Swedish coho showed signi ican SNPs in e ac ing wi h he sex a iable. In LS sex g oups (Supplemen a y Table 10), he mos signi ican in e ac ing SNPs we e s2853973 (Pin = 3.8 ×10−4) in males and s1470410 (Pin = 5.44 ×10−4) in emales. In non-LS sex g oups (Supplemen a y Table 11), he mos signi ican in e ac ing SNPs we e s10940422 (Pin = 4.18 ×10−4) in males and s12432418 (Pin = 4.85 ×10−5) in emales. F on ie s in Medicine 05 on ie sin.o g med-10-1132799 May 4, 2023 Time: 14:33 # 6 Xiong e al. 10.3389/ med.2023.1132799 FIGURE 1 Fo es plo o he 15 op common SNPs in he LS sex g oups in he Swedish coho . Me a-analysis in non-LS Non-LS sex g oups Me a-analysis in he Eu opean coho s (Sweden and Ge many) (Supplemen a y Tables 12,13) iden i ied 57 SNPs in males and 112 SNPs in emales a Pme a <5×10−8. Top signal s1964995 loca ed 36 kb om he 30o HLA-DRB5 was he same in males (OR = 1.715, SE = 0.0656, Pme a = 3.92 ×10−18) and emales (OR = 1.7304, SE = 0.0598, Pme a = 5 ×10−20). A o es plo illus a ing he op indings is shown in Figu e 2. Manha an and Q-Q plo s o he me a-analysis a e shown in Supplemen a y Figu e 5. He e ogenei y s a is ics in he SWE-GER a e shown in Supplemen a y Table 13X. A compa ison o associa ions a Pme a <5×10−8 showed 44 SNPs sha ed be ween non-LS males and emales, 17 SNPs exclusi ely associa ed wi h non-LS males, and 215 SNPs exclusi ely associa ed wi h non-LS emales. Mos SNPs we e in he MHC egion. Non-MHC SNPs include s694739 (OR = 1.3626, SE = 0.063, Pme a = 9.21 ×10−07) loca ed a 7.9 kbp om he 30o PRDX5 associa ed wi h non-LS males, and s2573346 (OR = 0.7712, SE = 0.0564, Pme a = 4.19 ×10−06) in ANXA11 associa ed wi h non- LS emales. Resul s om a me a-analysis in he mul i-e hnic coho s (Sweden, Ge many, and US-A ican Ame ican) (Supplemen a y Tables 14,15) iden i ied 12 SNPs in non-LS males and 49 in non- LS emales (Table 3). Top SNPs we e s1964995 (OR = 1.7736, SE = 0.0587, Pme a = 1.54 ×10−22) loca ed 36 kb om 30o HLA-DRB5 in males, and s2395153 (Be a = 0.6565, SE = 0.0477, Pme a = 1.11 ×10−18) loca ed a 5.9 kb om 50o C6o 10 in emales. The op me a-SNPs a e shown in Table 4. Fo es plo s (Figu e 2) illus a e he op common me a-SNPs in he Eu opean and mul i-ances y coho g oups. Manha an plo s o me a-analysis in mul i-e hnic g oups a e shown in Supplemen a y Figu e 6. He e ogenei y s a is ics in SWE-GER-USA a e shown in Supplemen a y Table 15X. A compa ison analysis a Pme a <5×10−8showed a sha ed SNP be ween non-LS males and emales, 11 SNPs exclusi ely associa ed wi h non-LS males, and 48 SNPs exclusi ely associa ed wi h non-LS emales. Mos associa ed SNPs we e in he MHC egion. Non-MHC SNPs we e obse ed in non-LS emales and included s7813186, s1049550, and s7133604 (Da a no shown). Lookup o me a-SNPs in he UK Biobank Genome-wide associa ion s udy o “Doc o diagnosed sa coidosis” in he UKB was used as a alida ion s ep. SNP- lookup showed se e al sex-speci ic SNPs in LS and non-LS a P<5×10−5. Chie ly, 223 SNPs associa ed wi h LS males and 209 SNPs associa ed wi h LS emales we e alida ed a P<5×10−8 (Supplemen a y Tables 16,17). In non-LS, 11 non-LS males and 31 SNPs non-LS emales in he SWE-GER coho s we e alida ed F on ie s in Medicine 06 on ie sin.o g med-10-1132799 May 4, 2023 Time: 14:33 # 7 Xiong e al. 10.3389/ med.2023.1132799 TABLE 3 The op ag-SNPs (PGC <3.5 ×10−7) associa ed wi h non-LS males and emales in Swedish, Ge man, and USA-AA coho s. Males SNP CHR BP (hg19) Alleles Re Seq genes Coho CA CAF OR 95% CI PGC s1049550 10 81,926,702 A/G ANXA11 SWE A 0.2962 0.5231 (0.4195, 0.6523) 1.82E-08 GER A 0.3845 0.7325 (0.6281, 0.8543) 3.63E-04 USA-AA A 0.1508 0.7724 (0.5841, 1.0215) 7.70E-02 s2239802 6 32,411,846 C/G HLA-DRA SWE C 0.3375 1.7790 (1.4295, 2.214) 4.52E-07 GER C 0.2283 1.4480 (1.2336, 1.6997) 4.75E-05 USA-AA C 0.3321 1.5140 (1.2288, 1.8655) 1.41E-04 s7197 6 32,412,580 A/G HLA-DRA SWE A 0.2885 1.7610 (1.4056, 2.2063) 1.50E-06 GER A 0.1901 1.4210 (1.1976, 1.6862) 2.96E-04 USA-AA A 0.2253 1.6350 (1.2895, 2.0731) 7.32E-05 s1964995 6 32,449,411 G/A 36kb 30o HLA-DRB5 SWE G 0.2995 0.6033 (0.4872, 0.747) 5.74E-06 GER G 0.3664 0.5452 (0.464, 0.6406) 3.40E-11 USA-AA G 0.2314 0.5568 (0.431, 0.7194) 1.21E-05 s3830135 6 32,548,464 A/G HLA-DRB1 SWE A 0.0529 0.3543 (0.2278, 0.5512) 6.71E-06 GER A 0.0980 0.4374 (0.3212, 0.5956) 2.40E-06 USA-AA C 0.0702 0.5942 (0.3897, 0.906) 1.82E-02 s3131283 6 32,119,898 A/G 177bp 50o PRRT1 SWE A 0.1971 1.8100 (1.4054, 2.3312) 6.85E-06 GER A 0.1417 1.3790 (1.1382, 1.6707) 3.21E-03 USA-AA C 0.0256 1.7040 (0.9263, 3.1348) 9.41E-02 s4530903 6 32,581,889 A/G 23kb 50o HLA-DQA1 SWE A 0.0529 0.3590 (0.2308, 0.5585) 8.79E-06 GER A 0.0973 0.4282 (0.3133, 0.5853) 1.74E-06 USA-AA T 0.0702 0.5942 (0.3897, 0.906) 1.82E-02 s1800684 6 32,151,994 T/A AGER SWE T 0.1982 1.7920 (1.3917, 2.3076) 9.75E-06 GER A 0.1428 1.3810 (1.1401, 1.6729) 3.00E-03 USA-AA A 0.0293 1.6280 (0.9369, 2.8289) 9.13E-02 s3104402 6 32,681,676 A/C 27kb 50o HLA-DQA2 SWE A 0.1050 2.0860 (1.5013, 2.8984) 1.83E-05 GER T 0.0689 1.3750 (1.0539, 1.794) 3.50E-02 USA-AA G 0.0153 2.3790 (1.0553, 5.3629) 4.12E-02 s1044506 6 32,172,065 A/C NOTCH4 SWE A 0.2016 1.7870 (1.3878, 2.3011) 1.08E-05 GER T 0.1441 1.3560 (1.1203, 1.6413) 4.97E-03 USA-AA G 0.0263 1.9640 (1.0695, 3.6067) 3.35E-02 Females s1964995 6 32,449,411 G/A 36kb 30o HLA-DRB5 SWE G 0.2995 0.5314 (0.4202, 0.6721) 6.73E-07 GER G 0.3611 0.5942 (0.519, 0.6804) 1.65E-12 USA-AA G 0.2408 0.8051 (0.6963, 0.9309) 3.77E-03 s3129727 6 32,679,690 A/G 29kb 50o HLA-DQA2 SWE A 0.0474 3.0760 (2.0025, 4.7251) 1.37E-06 GER A 0.0268 1.9580 (1.3773, 2.7836) 4.53E-04 USA-AA T 0.0217 1.6270 (1.0686, 2.4773) 2.47E-02 s3998158 6 32,681,992 G/A 27kb 50o HLA-DQA2 SWE G 0.1700 0.5047 (0.3771, 0.6755) 1.49E-05 GER C 0.1990 0.5811 (0.4908, 0.6881) 3.67E-09 USA-AA C 0.1665 1.0890 (0.925, 1.2821) 3.10E-01 s2395153 6 32,345,595 C/G 5.9kb 50o C6o 10 SWE C 0.2748 0.5947 (0.4694, 0.7535) 5.05E-05 GER C 0.3391 0.6137 (0.5347, 0.7044) 7.74E-11 USA-AA G 0.2175 0.7410 (0.6372, 0.8618) 1.18E-04 (Con inued) F on ie s in Medicine 07 on ie sin.o g med-10-1132799 May 4, 2023 Time: 14:33 # 8 Xiong e al. 10.3389/ med.2023.1132799 TABLE 3 (Con inued) Females SNP CHR BP (hg19) Alleles Re Seq genes Coho CA CAF OR 95% CI PGC s2076529 6 32,363,955 G/A BTNL2 SWE G 0.3176 0.6105 (0.4859, 0.7672) 6.66E-05 GER C 0.3752 0.6458 (0.5659, 0.7371) 1.22E-09 USA-AA C 0.3056 0.8406 (0.7363, 0.9598) 1.11E-02 s9268472 6 32,355,605 A/G 6.9kb 30o BTNL2 SWE A 0.3176 0.6107 (0.486, 0.7674) 6.71E-05 GER T 0.3754 0.6446 (0.5648, 0.7357) 1.04E-09 USA-AA A 0.3063 0.8471 (0.7421, 0.967) 1.51E-02 s9275580 6 32,679,462 G/A 30kb 50o HLA-DQA2 SWE G 0.1791 0.5543 (0.4192, 0.7329) 9.63E-05 GER C 0.2102 0.5519 (0.4666, 0.6528) 7.77E-11 USA-AA G 0.2458 1.1700 (1.0138, 1.3503) 3.35E-02 s3129882 6 32,409,530 G/A HLA-DRA SWE G 0.5146 1.5580 (1.263, 1.922) 9.81E-05 GER G 0.4499 1.1700 (1.0356, 1.322) 1.85E-02 USA-AA A 0.4586 1.0660 (0.9446, 1.203) 3.05E-01 SNP, single nucleo ide polymo phism; BP, ch omosome base pai s a e based on human genome assembly e sion 19 (hg19); CA, coded allele; CAF, coded allele equency; OR, odds a io; SE, s anda d e o o odds a io; PGC, genomic con olled P- alue. FIGURE 2 Fo es plo s o he 15 op common me a-SNPs in he non-LS sex g oups in he Eu opean g oup (Swedish-Ge man) and he mul i-e hnic g oup (Swedish-Ge man-Uni ed S a es-A ican Ame ican). a P<5×10−8(Supplemen a y Tables 18,19). In he mul i- e hnic coho s (SWE-GER-USA-AA), 2 SNPs in non-LS males and 8 SNPs in non-LS emales we e obse ed (Supplemen a y Tables 20,21). Gene-based analysis The gene-based analysis e ealed signi ican genomic loci associa ed wi h sex g oups in LS and non-LS (Supplemen a y Tables 22,23). The mos signi ican genomic locus was HLA-DRA (P= 4.36 ×10−14) in LS males and FKBPL (P= 1.39 ×10−14) in LS emales a a gene-based P<2.0 ×10−6(0.05/24,769 genes). In he Eu opean coho s (SWE-GER), 15 genes associa ed wi h non-LS males and 18 genes associa ed wi h non-LS emales we e iden i ied, whe e he mos signi ican genomic locus was HLA-DRA (P= 5.16 ×10−15) in non-LS males and HLA-DRB1 (P= 2.42 ×10−13) in non-LS emales. In he mul i-e hnic coho s (SWE-GER-USA-AA), 5 genes associa ed wi h non-LS males and 8 genes associa ed wi h non-LS emales we e iden i ied. The op genes we e HLA-DRA (P= 1.88 ×10−14) in non-LS males and C6o 10 (P= 6.47 ×10−12) in non-LS emales. A compa ison o gene-based analysis in LS and non-LS sex g oups ac oss coho s (Supplemen a y Table 24) e ealed di e ences in gene-based associa ions. In LS, gene-based associa ions showed 77 genes sha ed among sex g oups, 11 F on ie s in Medicine 08 on ie sin.o g med-10-1132799 May 4, 2023 Time: 14:33 # 9 Xiong e al. 10.3389/ med.2023.1132799 TABLE 4 The op signi ican SNPs (Pme a <5e-8) o GWAS me a-analysis on non-LS sex g oups. Sex g oup (Coho s) SNP CHR BP Alleles Re Seq_ genes CA OR 95% CI P- alue Di ec ion He -I2He -PVal Non-LS males (SWE-GER) s1964995 6 32,449,411 G/A 36kb 30o HLA-DRB5 A 1.7681 (1.5548, 2.0107) 3.92E-18 ++ 0 0.4583 s2395153 6 32,345,595 C/G 5.9kb 50o C6o 10 C 0.5861 (0.5134, 0.6692) 2.61E-15 – 0 0.5916 s2213585 6 32,413,150 G/A 323bp 30o HLA-DRA A 0.6315 (0.5618, 0.71) 1.35E-14 – 0 0.3240 s7195 6 32,412,539 A/G HLA-DRA A 1.5830 (1.4082, 1.7795) 1.46E-14 ++ 14.5 0.2795 s9271588 6 32,590,953 G/A 14kb 50o HLA-DQA1 A 1.6153 (1.4291, 1.8258) 1.72E-14 ++ 0 0.5981 s2213586 6 32,413,094 A/G 267bp 30o HLA-DRA A 1.5785 (1.4046, 1.7742) 1.93E-14 ++ 10 0.2918 s2227139 6 32,413,459 G/A 632bp 30o HLA-DRA A 0.6335 (0.5637, 0.712) 1.93E-14 – 10 0.2918 s7192 6 32,411,646 A/C HLA-DRA A 1.5763 (1.4023, 1.7721) 2.54E-14 ++ 0.3 0.3165 s4373382 6 32,350,868 C/A 11kb 50o C6o 10 A 1.6291 (1.4337, 1.8512) 6.84E-14 ++ 0 0.6069 s2076529 6 32,363,955 G/A BTNL2 A 1.6268 (1.4322, 1.8479) 7.03E-14 ++ 0 0.5625 Non-LS emales (SWE-GER) s1964995 6 32,449,411 G/A 36kb 30o HLA-DRB5 A 1.7305 (1.5391, 1.9457) 5E-20 ++ 0 0.4192 s4502931 6 32,380,782 A/T 5.9kb 50o BTNL2 A 0.6033 (0.5401, 0.674) 3.82E-19 – 0 0.4968 s2395153 6 32,345,595 C/G 5.9kb 50o C6o 10 C 0.6088 (0.5405, 0.6859) 3.14E-16 – 0 0.8219 s9275582 6 32,680,070 A/G 29kb 50o HLA-DQA2 A 0.5499 (0.4757, 0.6358) 6.15E-16 – 0 0.7935 s2647012 6 32,664,458 A/G 30kb 50o HLA-DQB1 A 1.5502 (1.3899, 1.7291) 3.74E-15 ++ 0 0.5473 s9275393 6 32,669,439 A/G 35kb 50o HLA-DQB1 A 0.5890 (0.5156, 0.6728) 6.22E-15 – 0 0.3517 s2294878 6 32,367,795 A/C BTNL2 A 0.6563 (0.5872, 0.7337) 1.27E-13 – 0 0.4173 s4530903 6 32,581,889 A/G 23kb 50o HLA-DQA1 A 0.4500 (0.364, 0.5564) 1.68E-13 – 0 0.5299 s2858332 6 32,68,1161 A/C 28kb 50o HLA-DQA2 A 0.6747 (0.6043, 0.7535) 2.83E-12 – 0 0.6205 s2858867 6 32,575,325 G/A 18kb 50o HLA-DRB1 A 0.6810 (0.6105, 0.7597) 5.54E-12 – 0 0.7201 Non-LS males (SWE-GER-USA- AA) s1964995 6 32,449,411 G/A 36kb 30o HLA-DRB5 A 1.7736 (1.5809, 1.9899) 1.54E-22 +++ 0 0.7552 s2239802 6 32,411,846 C/G HLA-DRA C 1.5360 (1.3764, 1.7143) 1.88E-14 +++ 19.6 0.2883 s4530903 6 32,581,889 A/G 23kb 50o HLA-DQA1 A 0.4478 (0.36, 0.5571) 5.40E-13 — 27.7 0.2506 s477515 6 32,569, 691 A/G 12kb 50o HLA-DRB1 A 0.6866 (0.607, 0.7767) 2.21E-09 — 0 0.4141 s389883 6 31,947,460 C/A STK19 A 0.7073 (0.6302, 0.7939) 4.02E-09 — 0 0.3934 (Con inued) F on ie s in Medicine 09 on ie sin.o g med-10-1132799 May 4, 2023 Time: 14:33 # 16 Xiong e al. 10.3389/ med.2023.1132799 73. Zhou Y, Shen L, Zhang Y, Jiang D, Li H. Human leukocy e an igen-A, -B, and -DRB1 alleles and sa coidosis in Chinese Han subjec s. Hum Immunol. (2011) 72:571–5. 74. Wenne s om A, Pie inalho A, Vauhkonen H, Lah ela L, Palikhe A, Hedman J, e al. HLA-DRB1 allele equencies and C4 copy numbe a ia ion in Finnish sa coidosis pa ien s and associa ions wi h disease p ognosis. Hum Immunol. (2012) 73:93–100. 75. an Moo sel CHM, Pe ek, M., Ri e a, NV. Un a elling he gene ic basis o sa coidosis. Eu Respi Monog . (2022) 96:41–56. 76. Rub so a K, Ma ack P, Rub so A. Sexual dimo phism in au oimmuni y. J Clin In es ig. (2015) 125:2187–93. doi: 10.1172/JCI78082 77. Ko s en P, Tampe B, Konig MF, Nikipho ou E. Sa coidosis and au oimmune diseases: di e ences, simila i ies and o e laps. Cu Opin Pulm Med. (2018) 24(5):504–12. 78. Ho mann S, F anke A, Fische A, Jacobs G, No hnagel M, Gaede KI. Genome- wide associa ion s udy iden i ies ANXA11 as a new suscep ibili y locus o sa coidosis. Na Gene . (2008) 40:1103–6. doi: 10.1038/ng.198 79. Le in AM, Iannuzzi MC, Mon gome y CG, T udeau S, Da a I, McKeigue P, e al. Associa ion o ANXA11 gene ic a ia ion wi h sa coidosis in A ican Ame icans and Eu opean Ame icans. Genes Immun. (2013) 14:13–8. 80. Iacca ino L, Ghi a dello A, Cano a M, Zen M, Be io S, Nalo o L, e al. An i-annexins au oan ibodies: hei ole as bioma ke s o au oimmune diseases. Au oimmun Re . (2011) 10:553–8. doi: 10.1016/j.au e .2011.04.007 81. Moss, SE, Mo gan, RO. The annexins. Genome Biol. (2004) 5:219. doi: 10.1186/ gb-2004-5-4-219 82. B uschi M, Pe e o A, Vaglio A, San ucci L, Candiano G, Ghigge i G. Annexin A1 and au oimmuni y: om basic science o clinical applica ions. In J Mol Sci. (2018) 19:1348. 83. Cas o F, Ca doso A, Gonçal es R, Se e K, Oli ei a M. In e e on-gamma a he c oss oads o umo immune su eillance o e asion. F on Immunol. (2018) 9:847. doi: 10.3389/ immu.2018.00847 84. Dempsey O, Pa e son E, Ke K, Denison A. Sa coidosis. BMJ. (2009) 339:b3206. doi: 10.1136/bmj.b3206 85. Miedema J, Kaise Y, B oos C, Wijsenbeek M, G unewald J, Kool M. Th17-lineage cells in pulmona y sa coidosis and Lo g en’s synd ome: iend o oe? J Au oimmun. (2018) 87:82–96. doi: 10.1016/j.jau .2017. 12.012 86. G unewald J, Eklund A. Sex-speci ic mani es a ions o Lo g en’s synd ome. Am J Respi C i Ca e Med. (2007) 175:40–4. doi: 10.1164/ ccm.200608-1 197OC 87. Fische A, Valen ony e R, Nebel A, No hnagel M, Mulle -Que nheim J, Schu mann M, e al. Female-speci ic associa ion o C-C chemokine ecep o 5 gene polymo phisms wi h Lo g en’s synd ome. J Mol Med. (2008) 86:553–61. doi: 10.1007/ s00109-008-0315-5 88. Co es A, B own M. P omise and pi alls o he immunochip. A h i is Res The . (2011) 13:101. doi: 10.1186/a 3204 F on ie s in Medicine 16 on ie sin.o g