Menopause modulates the circulating metabolome : evidence from a prospective cohort study

This is a sel -a chi ed e sion o an o iginal a icle. This e sion may di e om he o iginal in pagina ion and ypog aphic de ails. Au ho (s): Ti le: Yea : Ve sion: Copy igh : Righ s: Righ s u l: Please ci e he o iginal e sion: CC BY 4.0 h ps://c ea i ecommons.o g/licenses/by/4.0/ Menopause modula es he ci cula ing me abolome : e idence om a p ospec i e coho s udy © The Au ho (s) 2022. Published by Ox o d Uni e si y P ess on behal o Eu opean Socie y o Ca diology Published e sion Ka ppinen, Ja i E.; Tö mäkangas, Timo; Kujala, U ho M.; Sipilä, Sa ianna; Laukkanen, Ja i; Aukee, Pauliina; Ko anen, Vuokko; Laakkonen, Eija K. Ka ppinen, J. E., Tö mäkangas, T., Kujala, U. M., Sipilä, S., Laukkanen, J., Aukee, P., Ko anen, V., & Laakkonen, E. K. (2022). Menopause modula es he ci cula ing me abolome : e idence om a p ospec i e coho s udy. Eu opean Jou nal o P e en i e Ca diology, 29(10), 1448-1459. h ps://doi.o g/10.1093/eu jpc/zwac060 2022 ......................................................................................................................................................................................... ......................................................................................................................................................................................... Menopause modula es he ci cula ing me abolome: e idence om a p ospec i e coho s udy † Ja i E. Ka ppinen 1 , Timo Tö mäkangas 2 , U ho M. Kujala 1 , Sa ianna Sipilä 2 , Ja i Laukkanen 1,3,4 , Pauliina Aukee 5 , Vuokko Ko anen 2 , and Eija K. Laakkonen 2 * 1 Facul y o Spo and Heal h Sciences, Uni e si y o Jy äskylä, Jy äskylä, Finland; 2 Ge on ology Resea ch Cen e and Facul y o Spo and Heal h Sciences, Uni e si y o Jy äskylä, Jy äskylä, Finland; 3 Depa men o In e nal Medicine, Cen al Finland Heal h Ca e Dis ic , Jy äskylä, Finland; 4 Ins i u e o Clinical Medicine, Uni e si y o Eas e n Finland, Kuopio, Finland; and 5 Depa men o Obs e ics and Gynecology, Pel ic Floo Resea ch and The apy Uni , Cen al Finland Heal h Ca e Dis ic , Jy äskylä, Finland Recei ed 26 Janua y 2022; e ised 22 Feb ua y 2022; accep ed 17 Ma ch 2022 Aims We s udied he changes in he ci cula ing me abolome and hei ela ion o he menopausal ho monal shi in 17β-oes- adiol and ollicle-s imula ing ho mone le els among women ansi ioning om pe imenopause o ea ly pos menopause. Me hods and esul s We analysed longi udinal da a om 218 Finnish women, 35 o whom s a ed menopausal ho mone he apy du ing he s udy. The menopausal ansi ion was moni o ed wi h mens ual dia ies and se um ho mone measu emen s. The median ollow-up was 14 mon hs (in e qua ile ange: 8–20). Se um me aboli es we e quan ified wi h a ge ed nuclea magne ic esonance me abolomics. The model esul s we e adjus ed o age, ollow-up du a ion, educa ion, li es yle, and mul iple compa isons. Menopause was associa ed wi h 85 me aboli e measu es. The concen a ion o apoB (0.17 s anda d de i- a ion [SD], 99.5% confidence in e al [CI] 0.03–0.31), e y-low-densi y lipop o ein iglyce ides (0.25 SD, CI 0.05–0.45) and pa icles (0.21 SD, CI 0.05–0.36), low-densi y lipop o ein (LDL) choles e ol (0.17 SD, CI 0.01–0.34) and pa icles (0.17 SD, CI 0.03–0.31), high-densi y lipop o ein (HDL) iglyce ides (0.24 SD, CI 0.02–0.46), glyce ol (0.32 SD, CI 0.07–0.58) and leucine inc eased (0.25 SD, CI 0.02–0.49). Ci a e (−0.36 SD, CI −0.57 o −0.14) and 3-hyd oxybu y a e concen a ions dec eased (−0.46 SD, CI −0.75 o −0.17). Mos me aboli e changes we e associa ed wi h he menopausal ho monal shi . This explained 11% and 9% o he LDL choles e ol and pa icle concen a ion inc ease, espec i ely. Menopausal ho mone he apy was associa ed wi h inc eased medium- o-la ge HDL pa icle coun and dec eased small- o-medium LDL pa icle and glycine concen a ion. Conclusions Menopause is associa ed wi h p oa he ogenic ci cula ing me abolome al e a ions. Female sex ho mones le els a e con- nec ed o he al e a ions, highligh ing hei impac on women’s ca dio ascula heal h. ------------------------------------------------------------------------------------------------------------------------------------------------------------ * Co esponding au ho . Tel: +358408053588, Email: eija.k.laakkonen@jyu.fi † The wo k was pe o med a Ge on ology Resea ch Cen e and Facul y o Spo and Heal h Sciences, Uni e si y o Jy äskylä, Jy äskylä, Finland. © The Au ho (s) 2022. Published by Ox o d Uni e si y P ess on behal o Eu opean Socie y o Ca diology. 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 (h ps://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed euse, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Eu opean Jou nal o P e en i e Ca diology (2022) 00,1–12 h ps://doi.o g/10.1093/eu jpc/zwac060 FULL RESEARCH PAPER Gende -specific condi ions Downloaded om h ps://academic.oup.com/eu jpc/ad ance-a icle/doi/10.1093/eu jpc/zwac060/6580397 by Uni e si y o Jy askyla use on 13 July 2022 ......................................................................................................................................................................................... G aphical Abs ac Keywo ds Menopause •Ho mone eplacemen he apy •Ca dio ascula diseases •Me abolomics •Oes adiol 2Menopause and he ci cula ing me abolome Downloaded om h ps://academic.oup.com/eu jpc/ad ance-a icle/doi/10.1093/eu jpc/zwac060/6580397 by Uni e si y o Jy askyla use on 13 July 2022 In oduc ion Mos women ace menopause a he age o 48–52 yea s, 1 esul ing om he cessa ion o o a ian ollicula ac i i y and diagnosed 12 mon hs a e he final mens ual pe iod. 2 The cha ac e is ic meno- pausal ho monal shi consis s o a decline in 17β-oes adiol (E2) and a concomi an inc ease in ollicle-s imula ing ho mone (FSH) le- els. 1 Oes ogen-con aining menopausal ho mone he apy (MHT) alle ia es menopausal symp oms and es o es sys emic E2 wi hou lowe ing FSH o p emenopausal le els. 3 Menopause is hough o p edispose women o a he oscle o ic ca - dio ascula disease (ACVD) since hey de elop obs uc i e co ona y a e y disease 7–10 yea s la e han men, and his isk ises a e meno- pause. 4 Also, p ema u e menopause is associa ed wi h an inc eased ACVD isk. 5 The causali y be ween hese phenomena is di ficul o p o e, as he menopause-d i en me abolic changes may p edispose women o p esen ACVD a an olde age, making i di ficul o dis in- guish hei e ec s om o he ageing- ela ed changes. 6 Mo eo e , as i is gene ally no possible o p edic he iming o menopause, es ablishing app op ia e longi udinal s udy se ings is challenging. Pe imenopause ( he menopausal ansi ion phase) also di e s in iming and du a ion among indi iduals. 1 The e o e, indi idual in o ma ion on menopause p og ession benefi s he in es iga ion o menopausal e ec s. Resea ch on clinical bioma ke s suppo s he ela ionship be- ween menopause and ACVD isk. In longi udinal s udies, meno- pause is associa ed wi h inc eases in ci cula ing iglyce ide and low-densi y lipop o ein (LDL) choles e ol le els. Howe e , he e ec o menopause on high-densi y lipop o ein (HDL) choles e ol con- cen a ion and i s di ec ion is con o e sial. 7–10 The bioma ke changes may esul di ec ly om he menopausal ho monal shi o indi ec ly ia inc eased adiposi y. 11 The ole o emale sex ho - mones is s eng hened by s udies on MHT in pos menopausal wo- men. Oes ogen-only MHT lowe s LDL choles e ol and inc eases HDL choles e ol concen a ions accompanied by a ise in iglyce - ide le els when adminis e ed o ally. 12 In combina ion wi h MHT, he e ec s on HDL choles e ol a e modula ed by selec ed p oges o- gen. 12 MHT may also imp o e blood glucose egula ion, 13 eflec ing b oad sys emic e ec s on me abolism. Compa ed wi h clinical me hods, me abolomics o e a wide lens o in es iga e menopausal e ec s on he ci cula ing me abolome. Two popula ion-le el nuclea magne ic esonance (NMR) me abolo- mics s udies on menopause ha e been conduc ed. The c oss- sec ional s udy by Au o e al. 14 examined associa ions be ween age, sex, and menopause and he ci cula ing me abolome in 26 065 Finnish and Es onian indi iduals, including 10 083 women. La e , Wang e al. 15 in es iga ed c oss-sec ional associa ions in he UK among 3312 women, wi h 1492 longi udinal samples aken 2.5 yea s apa . Thei esul s we e simila ; menopause was associa ed wi h a p oa he ogenic shi in lipop o ein measu emen s and non- lipid me aboli es, such as amino acids. 14,15 Bo h s udies elied on sel - epo ed menopausal s a us and did no associa e he findings o he emale sex ho mone le els. The e o e, he e, we in es iga ed whe he he menopause- ela ed ho monal shi modula es he ci cula ing me abolome in a longi udin- al design, whe e he ollow-up o he menopausal ansi ion p og es- sion was indi idualized and moni o ed wi h epea ed FSH le el measu emen s. The p emise was ha menopause has an iden ifiable me abolomic finge p in esul ing om he shi o emale sex ho - mone le els. Me hods S udy design and pa icipan s This s udy used da a om he Es ogenic Regula ion o Muscle Apop osis (ERMA) p ospec i e coho s udy 16 and was app o ed by he e hics commi ee o he Cen al Finland Heal h Ca e Dis ic (KSSHP Dn o 8U/2014). The s udy complied wi h he Decla a ion o Helsinki, and pa - icipan s ga e in o med consen . The ec ui men was ca ied ou om 2014 o 2015. The sample was andomly d awn om he Popula ion In o ma ion Sys em. An in i a ion and a p eques ionnai e we e sen o 6878 women aged 47–55 li ing in he Jy äskylä a ea, o whom 3229 (47%) esponded. Based on he p e- ques ionnai e da a, 1627 women we e in i ed o menopausal s a us de- e mina ion. Exclusion c i e ia we e sel - epo ed body mass index .35 kg/m 2 and medical condi ions o use o medica ion a ec ing he o a ies, he ho mone o inflamma o y p ofile, o daily unc ioning. The menopausal s a us o 1393 women was de e mined, and 1158 women we e called o he Heal h and Spo s Labo a o y o he Uni e si y o Jy äskylä o physiological and psychological measu emen s. The ERMA s udy aimed o c ea e a coho o he in es iga ion o menopausal e ec s wi h a minimum influence o ageing. The e o e, 381 pe imenopausal women we e in i ed o pa icipa e in he longi udin- al Co e-ERMA s udy (Figu e 1). The fi s measu emen s we e pe o med du ing 2015 and 2016. Women kep a mens ual dia y du ing ollow-up and isi ed he labo a o y o he measu emen o FSH le els e e y 3 o 6 mon hs un il ea ly pos menopause. The ollow-up was scheduled ac- co ding o he pa icipan ’s menopausal ansi ion p og ession. The final las measu emen s we e pe o med in Decembe 2018. O he 381 pa icipan s, he ollow-up was comple ed by 234 women. One pa icipan died, and 26 d opped ou . Th ee women epo ed in- consis en MHT use and 117 women did no each pos menopause o hei menopausal s a us was unce ain a he las measu emen , leading o exclusion. F om his s udy, we excluded 12 women using lipid o glucose-lowe ing medica ion and 4 women diagnosed wi h cance . The e o e, he me abolomics analyses we e pe o med o 218 pa ici- pan s, wi h 35 (15%) s a ing MHT du ing ollow-up. Ho mone and me aboli es p ofile Blood samples we e collec ed be ween 7 and 10 a.m. a e o e nigh as - ing and p ocessed o se um collec ion wi h a s anda d p ocedu e. The samples we e aliquo ed and s o ed a –80°C un il analysis. E2 and FSH le els we e measu ed wi h IMMULITE 2000 XPi (Siemens Heal hca e Diagnos ics, UK). Me aboli es we e analysed wi h a a ge ed p o on nu- clea magne ic esonance ( 1 H-NMR) spec oscopy pla o m (Nigh ingale Heal h L d., Helsinki, Finland; bioma ke quan ifica ion e sion 2020). 17 The echnical de ails o he me hod ha e been epo ed p e iously. 18 The pla o m quan ifies 250 me aboli e measu es and 180 we e selec ed as ou comes. We le he 70 lipop o ein lipid a ios ou om he analyses due o he limi ed sample size and as hey mos ly p o ide o e lapping in- o ma ion wi h absolu e lipid concen a ions. Menopausal s a us and menopausal ho mone he apy use The 2011 S ages o Rep oduc i e Aging Wo kshop (STRAW+10) guide- lines we e used o de e mine he menopausal s a us. 2 An excep ion was ha he pa icipan s kep a mens ual dia y o a leas 3 mon hs be o e he fi s blood sampling. Pe imenopausal women we e equi ed o ha e J.E. Ka ppinen e al.3 Downloaded om h ps://academic.oup.com/eu jpc/ad ance-a icle/doi/10.1093/eu jpc/zwac060/6580397 by Uni e si y o Jy askyla use on 13 July 2022 i egula o no mens ual bleeding and FSH le els o 17–30 IU/L. Du ing ollow-up, women we e de e mined pos menopausal a e wo consecu- i e FSH measu emen s o .30 IU/L and 6 mon hs o ameno hoea. The c i e ion di e s om he guidelines whe e pos menopause begins a e 12 mon hs o ameno hoea due o p ac ical limi a ions. We decided o comp omise he leng h o he ollow-up pe iod o maximize he numbe o women wi h alid end measu emen s inside a limi ed p ojec unding pe iod. The median ameno hoea du a ion was 8.1 mon hs (in e qua ile ange: 6.5–10.4). Howe e , he absence o menses was ,6mon hsin28 pa icipan s. To accoun o he p o ocol de ia ion, we pe o med a sensi- i i y analysis wi hou hese pa icipan s. The pos menopausal s a us o 25 pa icipan s was de e mined solely based on FSH le els because o p io hys e ec omy (n=11) o ambigui y in mens ual dia y epo ing (n= 14). The FSH le els we e measu ed again a he las measu emen . MHT was que ied a each ollow-up mee ing, and MHT pa icipan s we e in i ed o he las measu emen 6 mon hs a e he ini ia ion o he ea men o allow he medica ion o exe i s e ec s. One woman epo ed s a ing MHT 17 days be o e and ano he pa icipan 2 days be- o e he las measu emen . We only classified he o me as an MHT use because he 2-day use o low dose ansde mal E2 was unlikely o a ec he s udied ou comes subs an ially. He me aboli e le els did no signifi- can ly di e om he o he pa icipan s’le els. O he 35 MHT s a e s, 27 used o al p oduc s con aining ei he E2-only (n=7), E2 in combin- a ion wi h dyd oges e one (n=16) o no e his e one ace a e (n=4). Conce ning he E2-only MHT s a e s, fi e we e use s o le ono ges el- eleasing in au e ine de ices, and wo did no need p oges ogen because o pas hys e ec omy. The es o he sample (n=8) used ans- de mal MHT, se en o whom used E2-gel wi h a le ono ges el- eleasing in au e ine de ice and one E2 and no e his e one ace a e pa ches. In 2020, 12% o Finnish women aged ≥45 yea s used MHT. 19 The e o e, he a io o MHT use s in ou sample (15%) eflec s equency o MHT use in he Finnish gene al popula ion. Educa ion le el, li es yle ac o s, and body composi ion Supplemen a y ma e ial online, Me hod Supplemen includes a de ailed desc ip ion o he s udy co a ia es. B iefly, he educa ion le el (p ima y, Figu e 1 S udy flowcha and s a is ical analysis app oach. 4Menopause and he ci cula ing me abolome Downloaded om h ps://academic.oup.com/eu jpc/ad ance-a icle/doi/10.1093/eu jpc/zwac060/6580397 by Uni e si y o Jy askyla use on 13 July 2022 seconda y, o e ia y) and li es yle ac o s we e eco ded wi h s uc- u ed ques ionnai es. Due o he low numbe o pa icipan s wi h p i- ma y educa ion, p ima y and seconda y we e combined in he s a is ical analyses. Smoking s a us was classified as ‘ne e ’,‘qui e ’,o ‘cu en smoke ’. As he numbe o cu en smoke s was low, dummy a iables we e c ea ed o desc ibe whe he he pa icipan had e e smoked and smoked cu en ly. Alcohol use was calcula ed as po ions pe week. Physical ac i i y was calcula ed as me abolic equi alen hou s pe day (MET-h/d) by assessing he du a ion, equency, and in ensi y o leisu e- ime physical ac i i y and he ime spen on ac i e commu ing. 20 Use o common oods in Finnish ood cul u e we e eco ded wi h a 45-i em ood- equency ques ionnai e and he die quali y was measu ed wi h an 11-elemen die quali y sum sco e (DQS) adap ed om a ali- da ed ool, 21 whe e a highe sco e eflec s a heal hie die . Heigh was measu ed wi h a s adiome e a he fi s measu emen . Body mass and body a pe cen age we e measu ed wi h InBody720 (Biospace, Seoul, Ko ea). S a is ical analyses The s a is ical analyses a e desc ibed in de ail in he Supplemen a y ma e ial online, Me hod Supplemen . Ques ionnai e-based da a we e missing om one pa icipan a he fi s and wo pa icipan s a he las measu emen . We comple ed he missing alues by inspec ing pa i- cipan s’answe s om o he isi s o mean impu a ion. Me aboli e da a we e nea ly comple e, and we did no impu e he a e missing alues. R e sion 4.0.0 o newe was used o s a is ical analyses unless s a ed o he wise. The p ima y esul s a e he associa ions be ween menopause and me- aboli e measu emen s in 183 women expe iencing na u al menopause, i.e. who did no s a MHT du ing ollow-up (Figu e 1). Analyses we e pe o med using linea mixed-e ec models wi h andom in e cep a e me aboli e Box-Cox ans o ma ion and s anda diza ion wi h espec o he fi s measu emen s. The me aboli es we e ea ed as ou comes and he menopausal s a us as exposu e. Two adjus ed models we e buil in addi ion o a c ude model. The fi s adjus ed model ( he main s udy e- sul s) included he co a ia es o age a he fi s measu emen , ollow-up du a ion, educa ion le el, and li es yle ac o s. In he second adjus ed model, he body a pe cen age (po en ial media o be ween menopause and me aboli es) was included as a co a ia e. The Ke -Šidák co ec ion was used o accoun o mul iple es ing. We also pe o med a sensi i i y analysis whe e we excluded he pa icipan s whose ameno hoea du - a ion was less han 6 mon hs be o e he las measu emen s. Two explo a o y analyses we e pe o med. Fi s , he di ec and indi ec associa ions ( ia body a pe cen age change) be ween he menopausal ho monal shi and me aboli e changes in he women ex- pe iencing na u al menopause (n=183) we e in es iga ed using la en change sco e modelling. Bo h E2 and FSH we e included in he model simul aneously because his combina ion be e cha ac e izes an indi i- dual’s sex ho mone p ofile as E2 le els fluc ua e du ing pe imenopause. The used model (Supplemen a y ma e ial online, Me hod Supplemen Figu e S1) can be seen as an ex ension o he pai ed - es , whe e change is con olled o me aboli e concen a ions a he fi s measu emen . The calcula ed e ec sizes a e in e p e ed simila ly o squa ed semi- pa ial co ela ions and indica e how much o he me aboli e change is explained by he menopausal ho monal shi . Mplus e sion 7.4 was used o es ima e he model pa ame e s. False disco e y a e adjus men was used o co ec o mul iple es ing. The associa ion o MHT ini i- a ion du ing ollow-up wi h he me aboli e measu emen s was s udied in he whole sample (N=218, n MHT s a e s =35) using menopausal s a- us and MHT in e ac ion as he exposu e. The model s uc u es and mul- iple es ing co ec ions we e pe o med as in he p ima y analysis. Resul s Pa icipan cha ac e is ics can be ound in Table 1. The mean age o pa icipan s was 51.7 (SD =1.9) a he fi s measu emen . Hype ension was he mos commonly diagnosed condi ion in he coho . Pa icipan s did no use lipid-lowe ing agen s, e en hough mo e han hal had ele a ed LDL choles e ol le els (.3 mmol/L). MHT s a e s we e younge han non-use s a he fi s measu e- men . Mo eo e , hey had lowe FSH le els and sys olic/dias olic blood p essu e. As o he es o he a iables, he g oups showed simila cha ac e is ics a he beginning o he s udy. The median ollow-up du a ion was 14 mon hs (in e qua ile ange: 8–20), anging om 4 mon hs o 3.5 yea s. Expec edly, E2 de- c eased and FSH inc eased in women no using MHT du ing ollow- up. Among MHT s a e s, E2 inc eased wi hou an appa en change in FSH le els. As o li es yle ac o s, alcohol use dec eased in bo h g oups and body a pe cen age inc eased by 1% o he o al sample wi h a simila end in bo h g oups. Menopause and me aboli e associa ions In Supplemen a y ma e ial online, Resul Supplemen Table S1, he ab- solu e me aboli e concen a ions du ing he s udy and he uns an- da dized me aboli e change sco es o he 183 women expe iencing na u al menopause a e shown. Menopause was asso- cia ed wi h a s a is ically significan change in 85 me aboli e measu es (Supplemen a y ma e ial online, Resul Supplemen Table S3; he key findings a e summa ized in Figu e 2). The apolipop o ein B (apoB)-con aining pa icle coun inc eased by 0.17 SD (99.95% con- fidence in e al [CI] 0.03–0.31), esul ing om he inc eased e y-low-densi y lipop o ein (VLDL; 0.21 SD, CI 0.05–0.36) and LDL pa icle coun s (0.17 SD, CI 0.03–0.31). The VLDL inc ease a- ou ed he inc ease o VLDL pa icle size (0.22 SD, CI 0.03–0.41). The apolipop o ein A-I (apoA-I) and o al HDL pa icle coun s did no change ma kedly. Howe e , i is unce ain whe he small HDL subclass is a ec ed by menopause (0.19 SD, CI −0.01 o 0.40). F om he lipid measu emen s, choles e ol concen a ions in all apoB-con aining lipop o ein classes inc eased om 0.17 o 0.20 SD. The inc ease in VLDL iglyce ide (0.25 SD, CI 0.05–0.45) and HDL iglyce ide concen a ions (0.24 SD, CI 0.02–0.46) was e en mo e p onounced. To al se um a y acid concen a ion did no change bu he a y acid p ofile shi ed om polyunsa u a ed o sa- u a ed di ec ion. As o he o he me aboli es, ci a e concen a ion dec eased by −0.36 SD (CI −0.57 o −0.14) and glyce ol concen a- ion inc eased by 0.32 SD (CI 0.07–0.58). Glu amine concen a ion dec eased by −0.46 SD (CI −0.72 o −0.20) and leucine concen a- ion inc eased by 0.25 SD (CI 0.02–0.49). Ace oace a e and 3-hyd o- xybu y a e le els dec eased by −0.44 SD (CI −0.72 o −0.15) and −0.46 (CI −0.75 o −0.17), espec i ely. When body a pe cen age was included as a co a ia e, he associa ion sizes o lipop o ein, lipid, glyce ol, and amino acid concen a ions we e oughly 0.03–0.05 SD smalle (Supplemen a y ma e ial online, Resul Supplemen Table S3). The adjus men had a negligible e ec on he associa ions o meno- pause wi h ci a e and ke one le els. The esul s we e e y simila in he sensi i i y analysis whe e he 28 pa icipan s wi h ,6 mon hs o ameno hoea we e excluded; howe e , almos all ound associa ions became sligh ly mo e obus (Supplemen a y ma e ial online, Resul Supplemen Table S6). J.E. Ka ppinen e al.5 Downloaded om h ps://academic.oup.com/eu jpc/ad ance-a icle/doi/10.1093/eu jpc/zwac060/6580397 by Uni e si y o Jy askyla use on 13 July 2022 No ably, he associa ion be ween menopause and small HDL pa icle coun was s a is ically significan (0.24 SD, CI 0.01–0.47). Menopausal ho monal shi and me aboli e changes The menopausal ho monal shi di ec ly explained he change in 64 o he 85 me aboli es iden ified as menopause- esponsi e in he p ima y analysis, wi h e ec sizes anging om 2.1 o 11.2% (Supplemen a y ma e ial online, Resul Supplemen Table S4). Based on he size o he s anda dized eg ession es ima es o he wo ho mones included in he model, he decline o E2 le els had mo e impac on he esul s han he inc eased FSH le els. The ho monal shi was no associa ed wi h indi ec me aboli e changes since he shi was no associa ed wi h body a pe cen - age change. The key findings a e summa ized in Figu e 3. The menopausal ho - monal shi di ec ly explained he concen a ion inc ease in apoB ................................................ ................................................ ..................................................................................................................................................... Table 1 Cha ac e is ics o he pa icipan s a he fi s and las measu emen s Na u al menopause (n=183) MHT s a e s (n=35) Pe imenopause Pos menopause Pe imenopause Pos menopause Age (yea s) 51.9 (1.9) 53.1 (1.9) 50.6 (1.8) 52.0 (1.9) Follow-up (days) –397 (243–603) –464 (334–707) HT use (days) –––220 (191–242) Sex ho mones E2 (nmol/L) 0.25 (0.17–0.41) 0.16 (0.11–0.27) 0.25 (0.18–0.42) 0.28 (0.20–0.61) FSH (IU/L) 30.0 (23.4–48.5) 69.8 (46.8–89.7) 23.0 (16.5–36.1) 36.1 (18.4–60.0) Educa ion P ima y 5 (3%) 5 (3%) 1 (3%) 1 (3%) Seconda y 100 (55%) 100 (55%) 15 (43%) 15 (43%) Te ia y 78 (42%) 78 (43%) 19 (54%) 19 (54%) Body composi ion Heigh (cm) 165.1 (5.7) –166.3 (5.5) – Body mass (kg) 69.5 (11.3) 70.2 (11.6) 70.4 (10.2) 70.5 (10.8) Body mass index ca ego ies Heal hy weigh (,25 kg/m 2 ) 92 (50%) 85 (46%) 20 (57%) 19 (54%) O e weigh (25–29.9 kg/m 2 ) 66 (36%) 71 (39%) 10 (29%) 12 (34%) Obesi y (≥30 kg/m 2 ) 25 (14%) 27 (15%) 5 (14%) 4 (11%) Pe cen body a (%) 30.9 (8.3) 32.1 (8.0) 30.5 (7.1) 31.4 (6.3) Smoking Ne e 122 (67%) 122 (67%) 26 (74%) 26 (74%) Qui e 49 (27%) 50 (28%) 6 (17%) 5 (14%) Smoke 12 (7%) 11 (6%) 3 (9%) 4 (11%) Alcohol use (po ions/week) 3.0 (1.0–5.0) 2.5 (1.0–5.0) 4.5 (2.5–5.8) 2.5 (1.0–4.9) Die quali y sco e (0–11) 5.7 (2.3) 5.5 (2.1) 5.5 (2.4) 5.6 (2.4) Physical ac i i y (MET-h/d) 3.6 (1.7–4.8) 3.8 (2.3–5.6) 4.5 (1.8–7.5) 4.8 (2.5–7.5) Blood p essu e Dias olic (mmHg) 85 (10) 82 (10) 80 (9) 77 (11) Sys olic (mmHg) 134 (19) 131 (18) 126 (16) 121 (14) Diagnosed hype ension 24 (13%) 29 (16%) 4 (11%) 4 (11%) ATC1 class C medica ion 25 (14%) 34 (19%) 5 (14%) 6 (17%) Dyslipidaemia and blood glucose LDL choles e ol .3 mmol/L a 99 (54%) 119 (65%) 22 (62%) 20 (57%) HDL choles e ol ,1.2 mmol/L 5 (3%) 4 (2%) 1 (3%) 0 (0%) T iglyce ides Op imal (,1.2 mmol/L) 123 (67%) 107 (58%) 24 (69%) 24 (69%) Ele a ed (.1.7 mmol/L) 22 (12%) 27 (15%) 1 (3%) 5 (14%) Fas ing glucose .6 mmol/L 8 (4%) 12 (7%) 2 (6%) 3 (9%) Da a as means (s anda d de ia ion), medians (in e qua ile ange), o coun s (%). a Calcula ed wi h F iedewald o mula. E2, 17β-oes adiol; FSH, ollicle-s imula ing ho mone; HDL, high-densi y lipop o ein; LDL, low-densi y lipop o ein; MET, me abolic equi alen o ask; MHT, menopausal ho mone he apy. 6Menopause and he ci cula ing me abolome Downloaded om h ps://academic.oup.com/eu jpc/ad ance-a icle/doi/10.1093/eu jpc/zwac060/6580397 by Uni e si y o Jy askyla use on 13 July 2022 (7.4%), LDL pa icles (8.5%), and LDL choles e ol (10.6%). The e ec sizes conce ning he o al VLDL pa icle coun (4.0%) and iglyce - ide le els (2.9%) we e mo e modes . E en hough he associa ion o menopause and he inc eased small HDL pa icle coun was incon- clusi e in he p ima y analysis, he explo a o y analysis s ongly linked hem wi h an e ec size o 10.9%. F om he esul s conce ning he associa ion o non-lipop o ein and lipid wi h menopause in he p ima y analysis, a di ec associa ion wi h he ho monal shi was confi med o glu amine (2.4%), ci a e (5.0%), 3-hyd oxybu y a e (4.4%), and ace oace a e (3.4%) concen- a ions. The la ges e ec size o he ho monal shi was ound o y osine concen a ion (7.8%), al hough menopause and y osine Figu e 2 The associa ions o he menopausal ansi ion and key me aboli e measu emen s in women expe iencing na u al menopause (n=183) adjus ed o age a he fi s measu emen , ollow-up du a ion, educa ion le el, smoking s a us, alcohol use, physical ac i i y, and die quali y. The figu e desc ibes he s anda dized change (exp essed as s anda d de ia ion [SD] uni s) in he me aboli e le els du ing ollow-up wi h espec o he fi s measu emen alues, allowing he compa ison be ween me aboli e measu es wi h di e en uni s and concen a ions. Apo, apolipop o ein; BCAA, b anched-chain amino acids; CI, confidence in e al; DHA, docosahexaenoic acid; HDL, high-densi y lipop o ein; IDL, in e media e-densi y lipop o ein; LA, linoleic acid; LDL, low-densi y lipop o ein; MUFA, monounsa u a ed a y acids; PUFA, polyunsa u a ed a y acids; SFA, sa u a ed a y acids; TG/PG, iglyce ide/phosphoglyce ide; VLDL, e y-low-densi y lipop o ein. J.E. Ka ppinen e al.7 Downloaded om h ps://academic.oup.com/eu jpc/ad ance-a icle/doi/10.1093/eu jpc/zwac060/6580397 by Uni e si y o Jy askyla use on 13 July 2022 le el associa ion did no emain s a is ically significan in he p ima y analysis a e mul iple compa ison co ec ions. Menopausal ho mone he apy and me aboli e changes In Supplemen a y ma e ial online, Resul Supplemen Table S2, he ab- solu e me aboli e concen a ions o MHT s a e s du ing he s udy including he uns anda dized change sco es a e p esen ed. MHT ini- ia ion du ing ollow-up was cha ac e ized by a dec eased apoB/ apoA-I a io (Supplemen a y ma e ial online, Resul Supplemen Table S5; he key s a is ically significan findings a e summa ized in Table 2), esul ing om he inc eased pa icle coun in medium and la ge HDL subclasses and he dec eased small and medium LDL subclass pa icle coun s. VLDL was less a ec ed by MHT han LDL and HDL. As o non-lipid me aboli es, we ound an in e se as- socia ion be ween MHT and glycine concen a ion. Adjus ing o body a pe cen age did no influence hese associa ions. Discussion This s udy in es iga ed he associa ions o menopause and ci cula ing me abolome in 183 women ansi ioning om pe imenopause o ea ly pos menopause. The s udy also explo ed whe he he meno- pausal ho monal shi explains he obse ed changes and whe he MHT ini ia ed du ing he ollow-up (n=35) was associa ed wi h me- aboli e changes. The esul s showed ha menopause-induced ho - monal shi is associa ed wi h a p oa he ogenic me abolomic finge p in . MHT specifically influences LDL, HDL, and glycine me- abolism. These findings b oadly ag ee wi h ea lie me abolomics s udies on menopause, 14,15 and now connec he p e ious and p e- sen obse a ions o he emale sex ho mone le els. Menopause modula es lipop o ein and lipid me abolism owa ds a p oa he ogenic p ofile apoB-con aining lipop o eins cause ACVD, 22 wi h LDL being he p ima y disease d i e . 23 S udies using bo h clinical me hods and me abolomics ha e iden ified LDL as one o he mos menopause- esponsi e bioma ke s. 8,9,14,15 In ag eemen wi h his, we epo inc eased apoB and LDL choles e ol and pa icle concen a ion, 7– 11% o which a e di ec ly explained by he menopausal ho monal shi . The unde lying mechanism is p obably he well-es ablished oes ogen-media ed LDL ecep o modula ion, 24,25 influencing LDL clea ance om he ci cula ion. 26 Howe e , simila o Au o e al. 14 bu con a y o Wang e al., 15 menopause did no al e LDL size dis ibu ion in ou s udy. MHT was associa ed wi h de- c eased pa icle coun s in medium and small LDL subclasses, p o id- ing u he e idence ha emale sex ho mones egula e LDL Figu e 3 The associa ions o menopausal ho monal change ( he decline in oes adiol and inc ease in ollicle-s imula ing ho mone le els) wi h key me aboli e changes. The colou s and hei in ensi y show he associa ion di ec ion and he e ec size. The models we e adjus ed o age a he fi s measu emen , ollow-up du a ion, educa ion le el, smoking s a us, alcohol use, physical ac i i y, and die quali y. False disco e y co ec ed P- alues: *≤0.05; ** ≤0.01. 3-OHB, 3-hyd oxybu y a e; Ala, alanine; Apo, apolipop o ein; BCAA, b anched-chain amino acids; DHA, docosahexaenoic acid; HDL, high-densi y lipop o ein; Gln, glu amine; Gly, glycine; GlycA, glycop o ein ace yls; His, his idine; IDL, in e media e-densi y lipop o ein; Ile, isoleucine; LA, linoleic acid; LDL, low-densi y lipop o ein; Leu, leucine; MUFA, monounsa u a ed a y acids; Val, aline; PG, phosphoglyce ides; Phe, phenylalanine; PUFA, polyunsa u a ed a y acids; SFA, sa u a ed a y acids; TG/PG, iglyce ide/phosphoglyce ide a io; Ty , y osine; VLDL, e y-low-densi y lipop o ein. 8Menopause and he ci cula ing me abolome Downloaded om h ps://academic.oup.com/eu jpc/ad ance-a icle/doi/10.1093/eu jpc/zwac060/6580397 by Uni e si y o Jy askyla use on 13 July 2022