Minor variant of rs 16827043 in the iron regulator hemojuvelin gene (HJV) contributes to hypertension: The TAMRISK study

Mino a ian o s 16827043 in he i on egula o hemoju elin gene (HJV) con ibu es o hype ension The TAMRISK s udy Seppo T. Nikka i, MD, PhD a,b , Anni-Lau a Vis o, MD a , Ki si M. Mää ä, MD a , Ta ja A. Kunnas, PhD a,∗ Abs ac I is known ha i on o e load may lead o an inc eased isk o many diseases. Acco ding o GWAS s udies, i on egula o y p o ein HFE gene a ian H63D ( s1799945) was associa ed wi h hype ension, an obse a ion which we we e able o confi m also in ou TAMRISK coho . Thus, i is possible ha abno mali ies in i on homeos asis may p edispose o hype ension. This p omp ed us o s udy whe he he e is an associa ion be ween hype ension and ano he i on o e load-associa ed gene, hemoju elin (HJV), which has 2 common polymo phic si es ( s 16827043, s7536827). The s udy included 336 hype ensi e cases and 480 con ols. All pa icipan s we e 50- yea -old Finnish men and women, and he da a was collec ed om he Tampe e adul popula ion ca dio ascula isk s udy (TAMRISK). Geno ypes we e de e mined using Compe i i e Allelic Specific PCR (KASP). We ound ha he mino a ian o he HJV polymo phic si e s16827043 (G-allele) is a s a is ically significan ac o associa ed wi h hype ension among 50 yea -old indi iduals compa ed wi h he AA geno ype ca ie s (OR=1.66, 95% CI: 1.06 –2.60, P=0.03). The isk was e en highe when o e weigh subjec s (BMI>30) we e excluded om he analyses. Fo he o he polymo phic a ian s7536827, associa ion wi h hype ension was ound only among no mal o sligh ly o e weigh A-allele ca ie s. In conclusion, HJV gene ic a ian s we e associa ed wi h essen ial hype ension in Finnish subjec s om he TAMRISK coho . P e ious s udies oge he wi h he p esen one indica e ha indi iduals wi h possible dys egula ion o i on me abolism may ha e highe isk o hype ension han hose wi h no mal i on homeos asis. Abb e ia ions: BMI =body mass index, CI =confidence in e al, HFE =his ocompa ibili y complex class I-like ansmemb ane p o ein (hemoch oma osis p o ein), HJV =hemoju elin (p e iously HFE2), PCR =polyme ase chain eac ion, PHE =pe iodic heal h examina ion. Keywo ds: gene ic a ian s, HJV, hype ension, i on 1. In oduc ion I on me abolism has been s udied ex ensi ely o unde s and how he body main ains i on homeos asis. I is now known ha hepcidin plays an impo an ole on in es inal i on abso p ion and i on ecycling in mac ophages. Hepcidin unc ions by dec easing he amoun o i on eleased om mac ophages o abso bed om in es ine. I is also known ha hepcidin defiency may lead o se e e i on o e load in mul iple o gans. [1–4] Hepcidin syn hesis is complexly egula ed by di e en p o eins and pa hways. [5] Two impo an ansmemb ane p o eins, HFE (his ocompa ibili y complex class I-like ansmemb ane p o ein) and HJV (hemoju elin), a e key modula o s o hepcidin exp ession. HJV ac s as a bone-mo phogene ic p o ein (BMP) co- ep esso , d i ing hepcidin ansc ip ion ia he BMP-SMAD signaling cascade. [6] Using a mouse model, Ken e al [7] showed ha HFE and HJV ope a e in he same pa hway o egula ion o hepcidin exp ession and i on me abolism. Recen ly, Wu e al [8] showed ha HJV is he key egula o o hepcidin and ha HFE ac s in an HJV-dependen manne . Some mu a ions in hese 2 p o eins ha e been associa ed wi h se e e i on o e load in pa ien s wi h he edi a y haemoch oma osis. [9,10] Dys egula ion in i on homeos asis may also lead o mild i on o e load, which has no been gene ally aken in o accoun . Two p e ious GWAS s udies [11,12] ha e ound an associa ion be ween HFE (H63D) gene ic a ian and hype ension. We we e able o eplica e his associa ion in he TAMRISK coho and showed ha ca ie s o he mu a ion had highe isk o hype ension han hose wi hou his mu a ion. [13] Mu a ions in he HJV gene ha lead o se e e i on o e load a e a e, al hough o e 40 mu a ions o HJV ha e been ecog- nized. [14] The e o e, we analyzed 2 ela i ely equen a ian s o his gene ha ha e been shown o media e dys unc ional i on Edi o : Ming Zhang. Funding: Compe i i e esea ch unding o he Pi kanmaa Hospi al Dis ic unded his wo k. The au ho s ha e no conflic s o in e es o disclose. a Depa men o Medical Biochemis y, Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, Finland, b Fimlab labo a o ies, Tampe e, Finland. ∗ Co espondence: Ta ja A. Kunnas, Depa men o Medical Biochemis y, Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, Finland (e-mail: a ja.kunnas@u a.fi). Copy igh ©2017 he Au ho (s). Published by Wol e s Kluwe Heal h, Inc. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion-Non Comme cial-No De i a i es License 4.0 (CCBY-NC- ND), whe e i is pe missible o download and sha e he wo k p o ided i is p ope ly ci ed. The wo k canno be changed in any way o used comme cially wi hou pe mission om he jou nal. Medicine (2017) 96:5(e6052) Recei ed: 22 Augus 2016 / Recei ed in final o m: 23 Decembe 2016 / Accep ed: 13 Janua y 2017 h p://dx.doi.o g/10.1097/MD.0000000000006052 Obse a ional S udy Medicine® OPEN 1 egula ion. [15] In he p esen s udy, we wan ed o examine a possible associa ion be ween hese HJV gene ic a ian s and hype ension in he Finnish TAMRISK coho . 2. Ma e ials and me hods 2.1. S udy popula ion The Tampe e adul popula ion ca dio ascula isk s udy (TAM- RISK) is a p ospec i e, longi udinal popula ion-based heal h su ey s udy in Tampe e, a ci y in sou he n Finland wi h a popula ion o 210 000. The da a o he TAMRISK s udy was collec ed om he pe iodic heal h examina ions (PHE) done o 50-yea -old men and women li ing in Tampe e. The PHEincluded one 60-minu e session wi h a public heal h nu se a he cen e ’s heal h examina ion uni as p e iously desc ibed. [13,16] TAMRISK da a includes in o ma ion o isk ac o s o hype ension: blood p essu e, weigh , amily his o y o ca dio ascula diseases, lipid alues and smoking, diabe es, and exe cise habi s. Physical ac i i y was defined as imes o exe cise/ week (enhanced b ea hing and swea ing). Cu en and p e ious diseases we e iden ified based on sel - epo o diagnosis by a physician, including hype ension. Cases in his s udy we e he subjec s who had hype ension and/o CAD a he age o 50 yea s as diagnosed by a physician by no mal heal hca e p ocedu es. Fo mos pa ien s, physicians diagnose hype ension when blood p essu e eadings a e consis en ly 140/90mm Hg o abo e. Fo each case, a leas 1 no mo ensi e con ol wi h he same sex and simila smoking habi s we e chosen om a PHE coho (n=6000). Smoking s a us was e alua ed based on sel - epo ing. Using he pa ien ’s na ional iden i y code, da a on hospi al- iza ions including ICD-10 codes o discha ge diagnoses we e ob ained om he Finnish Na ional Hospi al Discha ge Regis y (HILMO) main ained by he Na ional Ins i u e o Heal h and Wel a e. P e alence o ischemic hea diseases (I20-I25) we e ollowed up om 2005 o 2014 un il he subjec s we e on he a e age 60 yea s old. Buccal swabs o DNA ex ac ion and a pe missions o m o use PHE da a we e collec ed by mail sepa a ely o he physical examina ion. The DNA samples we e collec ed du ing yea s 2006–2010. In o med consen was ob ained om all pa ic- ipan s. The E hics Commi ees o he Tampe e Uni e si y Hospi al and he Ci y o Tampe e app o ed he s udy. 2.2. Geno yping DNA was ex ac ed om buccal swabs using a comme cial ki (Qiagen Inc., Valencia, CA). Geno yping was pe o med using KASP (Compe i i e Allelic Specific Amplifica ion) geno yp- ing se ices a KBioscience Ins i u e, UK. De ails o his me hod can be ob ained om h ps://www.lgcg oup.com/geno yping/. 2.3. S a is ical analysis S a is ical analyses we e pe o med using IBM SPSS S a is ics 23, and Ha dy–Weinbe g equilib ium o he geno ypes was calcula ed using OEGE (online encyclopedia calcula o o gene ic epidemi- ology s udies). [17] T- es and 1-way ANOVA o con inuous a iables (clinical cha ac e is ics) and chi-squa e es o ca ego i- cal a iables (hype ension, co ona y a e y disease) we e applied o he compa ison o HJV geno ype g oups. Associa ions o he geno yped HJV gene a ian s wi h hype ension/co ona y a e y disease wi h isk ac o s (BMI, glucose, choles e ol and gende ) we e analyzed using logis ic eg ession analysis. P- alues less han 0.05 we e conside ed significan . 3. Resul s Clinical cha ac e is ics o he s udy popula ion a he age o 50 yea s a e p esen ed in Table 1. B iefly, he case g oup comp ised 336 hype ensi e cases and con ol g oup 480 no mo ensi e subjec s wi h he same sex dis ibu ion and simila smoking habi s. A o al o 78 subjec s we e ound o ha e co ona y a e y disease when ollowed up o 60 yea s o age. Geno yping was success ul in 808 subjec s o he HJV s16827043 (A>G) and in 795 subjec s o he s7536827 (A> T). The measu ed geno ype equencies we e no significan ly di e en om he expec a ions o Ha dy–Weinbe g equilib ium (x 2 =0.22 o s16827043 and x 2 =0.09 o s7536827). The clinical cha ac e is ics acco ding o di e en geno ypes a e shown in Tables 2 and 3. Table 1 Clinical cha ac e is ics o cases and con ols o he s udy popula ion and subpopula ion wi h BMI <30. S udy popula ion Subpopula ion wi h BMI<30 Cases (n=336) Con ols (n=480) P(Cases n=239) Con ols (n=399) P Age, y 50±050±050±050±0 BMI, kg/m 2 28.8±5.1 25.5 ±3.7 <0.001 25.9±2.8 24.8±2.7 <0.001 Hemoglobin 147.0±13.4 145.4±13.2 0.165 146.3±13.0 144.8±13.1 0.240 Choles e ol, mmol/L 5.38±0.99 5.37 ±0.88 0.887 5.42±1.02 5.37 ±0.88 0.510 LDL choles e ol, mmol/L 3.16±0.88 3.17 ±0.82 0.838 3.15±0.92 3.16 ±0.83 0.921 Glucose, mmol/L 5.17±1.29 4.86 ±0.53 <0.001 5.11±1.45 4.85±0.53 0.001 Sys olic blood p essu e, mm Hg 142.7±16.6 129.3±14.8 <0.000 142.7±17.0 128.4±14.3 <0.000 Dias olic blood p essu e, mm Hg 92.8±8.8 84.4±9.1 <0.000 92.7±8.9 83.8±8.7 <0.000 Hype ension % 100 0 100 0 Diabe es % 12.8 0 <0.000 9.6 0 <0.000 Myoca dial in a c ion % 3.6 0 <0.000 3.2 0 0.001 Exe cise, a leas wice a week, % 62.8 57.2 0.135 62.8 55.5 0.102 Family his o y o hype ension % 71.6 42.5 <0.000 74.7 41.5 <0.000 Gende , male, % 58.4 56.8 0.319 58.3 64.1 0.165 Da a is p esen ed as mean ±SD BMI =body mass index, LDL =low densi y lipop o ein, SD=s anda d de ia ion. Nikka i e al. Medicine (2017) 96:5 Medicine 2 Fo HJV polymo phism s16827043, he e we e only 3 indi iduals who we e homozygous o he GG geno ype and hey we e combined o he GA geno ype g oup. A he age o 50 yea s, 58 o 111 G-allele ca ie s (52,2%) had diagnosed hype ension compa ed o 287 o 697 (41.3%) o hose homozygous o he wild ype (AA), espec i ely (P=0.041). Also, dias olic blood p essu e eadings we e significan ly highe among G-allele ca ie s (Table 2). When he isk o hype ension was analyzed by logis ic eg ession using HJV a ian s, BMI, glucose, choles e ol, and gende as explainable a iables, OR o HJV G-allele ca ie s was 1.66 (P=0.03, 95% CI: 1.06–2.60), o BMI 1.19 (P<0.001, 95% CI: 1.15–1.24), o glucose 1.49 (P< 0.001, 95% CI: 1.16–1.92), o choles e ol 0.99 (P=0.96, 95% CI: 0.84–1.18), and o gende 1.50 (P=0.02, 95% CI: 1.07–2.09) compa ed wi h AA geno ype. In o de o exclude he s ong e ec o BMI on hype ension, we also analyzed a subpopula ion o he s udy pa icipan s wi h no mal o only sligh ly ele a ed BMI. When o e weigh pa icipan s (BMI>30) we e excluded om he analyses, he isk o hype ension among G-allele ca ie s emained significan (OR=1.80, 95% CI 1.09–2.98, P=0.02). Adjus ed and unadjus ed esul s a e in Table 4. No s a is ically significan associa ion wi h co ona y a e y disease was ound (Table 2). Fo he o he polymo phic si e ( s7536927, T>A), no associa ion be ween geno ypes o combined alleles and hype - ension was ound in he whole s udy popula ion (P=0.29 o geno ypes and P=0.24 o A-allele ca ie s). Howe e , when subjec s we e s a ified acco ding o weigh , ca ie s o he A-allele whose BMI was less han 30 had mo e o en hype ension compa ed wi h hose homozygous o he wild ype (TT) (OR=1.56, P=0.04, 95% CI 1.01–2.39) (Table 4.) As wi h he s7536827, no associa ion wi h co ona y a e y disease was ound (Table 2). 4. Discussion The esul s o he p esen s udy sugges ha gene ic a ia ion in he HJV gene is significan ly associa ed wi h hype ension in a 50- yea -old Finnish popula ion. P e ious s udies o he HJV and HFE gene ic polymo phisms ha e concen a ed mainly on hemoch oma osis. Ou esul s and hose o o he s indica e ha dis u bances in i on me abolism may also inc ease he isk o hype ension. [18,19] We ound no impac o he 2 s udied HJV polymo phisms on co ona y a e y disease. In his pape , we show ha he mino allele G o he HJV a ian s16827043 was associa ed wi h hype ension. In addi ion, he associa ion be ween he mino allele and hype ension was e en s onge among no mal- o only sligh ly o e weigh subjec s. Fo he o he HJV polymo phic si e s7536827, no s a is ically significan associa ion wi h hype ension was ound. Howe e , when T allele ca ie s we e combined and obese subjec s we e excluded, also his gene ic a ia ion associa ed wi h hype ension. Al hough he mechanism is no known, ou esul oge he wi h p e ious ones p o ides a u he link be ween i on me abolism and hype ension. [11–13,20] Mild i on o e load is one possible explana ion o highe blood p essu e o ca ie s o he HJV mino a ian s, as has been sugges ed o H63D. [13] Bo h o hese memb ane ecep o s a e in ol ed in pa hways leading o hepcidin ansc ip ion. A ecen Table 2 Clinical cha ac e is ics o he s udy popula ion s a ified acco ding o HJV s 16827043 geno ypes. HJV PP s16827043 AA AG GG AA s AG s GG AA s (AG+ GG) n a 50 697 108 3 Hype ension % 41.3 51.0 100 0.024 0.041 Co ona y a e y disease, %, n=78 9.9 8.3 0 0.745 0.729 Sys olic blood p essu e, mm Hg 134. 7±16.8 137.7 ±16.7 136.7±13.3 0.237 0.095 Dias olic blood p essu e, mm Hg 87.7±9.8 89.6±9.7 98.9±7.2 0.035 0.033 Body mass index, kg/m 2 26.9±4.7 27.1 ±4.2 25.5±2.3 0.784 0.746 Choles e ol, mmol/L 5.35±0.99 5.55 ±0.90 5.53±0.97 0.151 0.052 Glucose, mmol/L 5.03±1.20 5.02 ±0.68 4.83±0.25 0.953 0.898 HJV =hemoju elin (p e iously HFE2). P alues om he chi-squa e es o ca ego ical a iables and 1 way ANOVA o T- es o con inuous a iables. P alues <0.05 a e in bold. Table 3 Clinical cha ac e is ics o he s udy popula ion s a ified acco ding o HJV s 7536827 geno ypes. HJV PPP s7536827 AA AT TT AA s AT s TT AA s (AT+TT) TT s (AA+AT) n a 50 186 397 212 Hype ension % 39.2 45.5 40.9 0.293 0.602 0.241 Co ona y a e y disease, %, n=78 8.1 10.4 9.4 0.680 0.161 1.000 Sys olic blood p essu e, mm Hg 134.9±16.3 135.5±16.6 134.4±17.7 0.738 0.898 0.475 Dias olic blood p essu e, mm Hg 87.6±10.2 88.5±9.7 87.4±9.6 0.362 0.529 0.302 BMI, kg/m 2 26.6±4.8 27.1±4.6 26.9±4.5 0.406 0.250 0.767 Choles e ol, mmol/L 5.37±0.95 5.34±1.05 5.46±0.89 0.340 0.846 0.151 Glucose, mmol/L 5.03±0.91 5.01±1.19 5.07±1.24 0.841 0.949 0.567 BMI =body mass index. P alues om he chi-squa e es o ca ego ical a iables and 1-way ANOVA o T- es o con inuous a iables. P alues <0.05 a e in bold. Nikka i e al. Medicine (2017) 96:5 www.md-jou nal.com 3 s udy has epo ed ha HJV unc ions as enhance o i on signaling o hepcidin. Since hepcidin is he main i on egula o y ho mone, dis u bances in pa hways a ec ing hepcidin exp ession may block i s unc ion as a eedback inhibi o o i on abso p ion. Howe e , only comple e loss o hepcidin in humans is esponsible o a e ye se e e o ms o massi e body i on o e load. [21] In Finland, es s o assessing body i on le els (se um e i in and ans e in sa u a ion) a e no ou inely measu ed and he e o e a limi a ion o he TAMRISK s udy popula ion is he lack o hese sa u a ion ma ke s. Howe e , i has p e iously been published ha men wi h essen ial hype ension had g ea e i on s o es han no mo ensi e con ols. [20] I is known ha p e alence o hype ension is highe in obese han in lean popula ions and he e is nea ly linea ela ionship be ween BMI and blood p essu e. [22] I has p e iously been shown ha he e is a gene-en i onmen al associa ion be ween hype en- sion genes and BMI. Thus, obesi y will o e ide he gene ic e ec . [13,23] The associa ion o HJV and hype ension was simila o e en s onge o no mal weigh and sligh ly o e weigh subjec s han in obese subjec s, confi ming he ole o HJV. Al hough he mechanism is no ye known, ou esul s sugges ha he mino allele G o he HJV a ian ( s16827043) is associa ed wi h highe isk o hype ension a he age o 50 yea s, compa ed wi h he AA-geno ype ca ie s. In addi ion, we ound ha among no mal o sligh ly o e weigh indi iduals, also T-allele o he HJV s7536827 inc eases he isk o hype ension. I is he e o e possible ha dys egula ion in i on me abolism is a significan ac o behind hype ension. Abno mali ies in i on homeos asis and ela ionship be ween i on me abolism and hype ension wa an u he s udies. Acknowledgmen s The au ho s hank all o he pa icipan s o he TAMRISK s udy and Mi ka Pie iläinen and Ulla Saa ijoki o hei skil ul echnical assis ance. Re e ences [1] C owno e B, Co ey C. He edi a y hemoch oma osis. Am Fam Physician 2013;87:183–90. [2] Means RTJ . Hepcidin and i on egula ion in heal h and disease. Am J Med Sci 2013;345:57–60. [3] Lo eal O, Haziza-Pigeon C, T oadec M, e al. Hepcidin in i on me abolism. Cu P o ein Pep ide Sci 2005;6:279–91. [4] Nicolas G, Chau e C, Via e L, e al. The gene encoding he i on egula o y pep ide hepcidin is egula ed by anemia, hypoxia and inflamma ion. J Clin In es 2002;110:1037–44. [5] Sil a B, Faus ino P. An o e iew o molecula basis o i on me abolism egula ion and he associa ed pa hologies. Biochimica e Biophysica Ac a 2015;1852:1347–59. [6] Babi JL, Huang FW, W igh ing DM, e al. Bone mo phpgene ic p o ein signaling by hemoju elin egula es hepcidin exp ession. Na Gene 2006;38:531–9. [7] Ken P, Wilkinson N, Cons an e M, e al. H e and Hj exhibi o e lapping unc ions o i on signaling o hepcidin. J Mol Med 2015;93:489–98. [8] Wu Q, Wang H, An P, e al. HJV and HFE play dis inc oles in egula ing hepcidin. An ioxid Redox Signal 2015;22:1325–36. [9] Fede J, Gni ke A, Thomas W, e al. A no el MCH class I-like gene is mu a ed in pa ien s wi h he edi a y haemoch oma osis. Na u e Gene 1996;13:399–408. [10] Papanikolaou G, Samuels ME, Ludwig EH, e al. Mu a ions in HFE2 cause i on o e load in ch omosome 1q-linked ju enile hemoch oma o- sis. Na Gene 2004;36:77–82. [11] Eh e GB, Mun oe PB, Rice KM, e al. Gene ic a ian s in no el pa hways influence blood p essu e and ca dio ascula disease isk. Na u e 2011;478:103–9. [12] Lu X, Wang L, Lin X, e al. Genome-wide associa ion s udy in Chinese iden ifies no el logi o blood p essu e and hype ension. Hum Mol Gene 2015;24:865–74. [13] Mää ä KM, Nikka i ST, Kunnas TA. Gene ic a ian coding o i on egula o y p o ein HFE con ibu es o hype ension, he TAMRISK s udy. Medicine (Bal imo e) 2015;94:e464. [14] Co e A, Canali S, Babi J. Hemoju elin and bone mo phogene ic p o ein (BMP) signaling in i on homeos asis. F on Pha macol 2014;5: 104. [15] Mile J, Dehais V, Bou gain C, e al. Common a ian s in he BMP2, BMP4 and HJV genes o he hepcidin egula ion pa hway modula e HFE hemoch oma osis pene ance. Am J Hum Gene 2007;81:799–807. [16] Mää ä KM, Nikka i ST, Läh eelä KH, e al. A unc ional a ian in he se ine- h eonine kinase coding gene is associa ed wi h hype ension: a case-con ol s udy in a Finnish popula ion, he Tampe e adul popula ion ca dio ascula isk s udy. J Hype ension 2013;31:516–20. [17] Rod igues S, Gaun TR, Day INM. Ha dy–Weinbe g equilib ium es ing o biological asce ainmen o Mendelian andomiza ion s udies. Am J Epidemiol 2009;169:505–14. [18] Na eka A, Olds RL, Lau MW, e al. Ele a ed blood p essu e: ou amily’s aul ? The gene ics o essen ial hype ension. Wo d J Ca diol 2014;6:327–37. [19] Valen i L, Malobe i A, Signo ini S, e al. I on s o es, hepcidin, and ao ic s i ness in indi iduals wi h hype ension. Plos One 2015;10:e0134635. [20] Pipe no A, T ombini P, Gelosa M, e al. Inc eased se um e i in is common in men wi h essen ial hype ension. J Hype ens 2002;20: 1513–8. [21] Pie angelo A. Gene ics, gene ic es ing, and managemen o hemoch o- ma osis: 15 yea s since hepcidin. Gas oen e ology 2015;149:1240–51. [22] Weinbe ge MH, Finebe g NS, Finebe g SE, e al. Sal sensi i i y, pulse p essu e, and dea h in no mal and hype ensi e humans. Hype ension 2001;37:429–32. [23] Ma eau J-B, Sass C, Pfis e M, e al. The Leu554Phe polymo phism in he E-selec in gene is associa ed wi h blood p essu e in o e weigh people. J Hype ens 2004;22:305–11. Table 4 Unadjus ed and adjus ed OR esul s ob ained om logis ic eg ession analysis o hype ension. Uni a ia e model ∗ Mul i a ia e model † OR (95% CI) POR (95% CI) P All pa icipan s s16827043, G-allele s AA 1.54 (1.02–2.32) 0.041 1.66 (1.06–2.60) 0.028 s7536827, A-allele s TT 1.26 (0.88–1.69) 0.247 1.36 (0.94–1.98) 0.103 BMI<30 kg/m 2 OR (95% CI) POR (95% CI) P s16827043, G-allele s AA 1.70 (1.05–2.75) 0.032 1.80 (1.09–2.98) 0.022 s7536827, A-allele s TT 1.48 (0.99–2.22) 0.056 1.56 (1.01–2.39) 0.044 BMI=body mass index, CI=confidence in e al, OR =odds a io. ∗ Uni a ia e model wi h HJV SNP alone † Mul i a ia e 2 model wi h HJV SNP, BMI, glucose, choles e ol, and gende as explainable a iables 4 Nikka i e al. Medicine (2017) 96:5 Medicine