Downloaded by guest on September 28, 2021 a Carrasco Nancy and Kaspari R. Rachel the muscle in skeletal thermogenesis adaptive is increased mice by hypothyroid marked of phenotype lean paradoxical The www.pnas.org/cgi/doi/10.1073/pnas.2008919117 cognitive even and serum) the in untreated, TSH (6–10). if of impairment levels defects, THs of increased transport levels reduced and iodide by defined cause state (a hypothyroidism that muta- develop or NIS disorders in and iodide-deficiency tions iodide with con- patients euthyroid of a Thus, maintain effects consumption dition. to required the Sufficient are function to (2). NIS adequate leads hormones ultimately these reg- which of negatively transcription, and positively gene both where ulate to T bloodstream, receptors body, TH form, the the nuclear active throughout with in biologically tissues Once more all nega- (2–4). virtually the turn, THs reach negative in THs the by a the is, by promoted In regulated are which THs. release tively (TSH), the and hormone iodide of as biosynthesis thyroid-stimulating of (5), constituent TH (NIS) transport loop, essential symporter feedback an active sodium/iodide is the the by (1– iodine on thyroid life They relies the throughout muscle. into biosynthesis (BMR) skeletal rate TH and metabolic 4). lungs, basal regulate (CNS), also system nervous tral T thermogenesis adaptive hypothyroidism these against of strategies new understanding develop deeper to possible obesity. A it phenotype. make collectively may leaner which changes changes, a metabolic in to of hypothyroidism variety lead that a indicate in results findings mice Our muscle. processes both skeletal regulate rather, in negatively stimulation; to likely FAO Nei- are or hormones (FAO). are thyroid expression (TSH) oxidation sarcolipin hormone for acid thyroid-stimulating responsible fatty nor in catecholamines sarcolipin, increase these ther up-regulation marker associated intake, thermogenic compensatory an food adaptive with of than skeletal-muscle reduced signs leaner the to being of show part to mice in addition hypothyroid owing In thyroid. mice, active the mediates euthyroid in that protein sym- uptake membrane sodium/iodide iodide plasma the the lacking (NIS), mouse mice porter drug-free hypothyroidism: a used of we model phenomenon, a mech- this have the underlying elucidate to To anism reported controls. euthyroid been than phenotype have leaner models How- mouse basal gain. decreased hypothyroid weight ever, a and expenditure, with energy Hypothy- energy associated lower rate, typically in metabolic thermogenesis. is involved humans adaptive in mechanisms roidism notably the most on path- stud- expenditure, other the concentrated while on have signaling, combat with- focused hunger ies To and have intake satiety expenditure. studies underlying energy energy many ways in epidemic, in obesity increase increase the an concomitant comorbidities. a grow- from of its out results variety given a Brent) gain with A. worldwide, Gregory association Weight and problem its Bianco and health C. Antonio incidence major by ing reviewed a 2020; 7, is May review Obesity for (sent 2020 22, July Carrasco, Nancy by Contributed and 37203; TN Nashville, University, Vanderbilt eateto ellradMlclrPyilg,Yl colo eiie e ae,C 06510; CT Haven, New Medicine, of School Yale Physiology, Molecular and Cellular of Department etyodhroe Ts r-oohrnn [T [T tri-iodothyronine roxine (THs; hormones thyroid he 4 | )aecuilfrtedvlpeto h cen- the of development the for crucial are ]) oimidd ypre (NIS) symporter sodium/iodide a nraReyna-Neyra Andrea , | a,b,3 sarcolipin 3 rdmnnl interacts predominantly , c eateto nenlMdcn,Yl colo eiie e ae,C 06520 CT Haven, New Medicine, of School Yale Medicine, Internal of Department | odintake food a aaJung Lara , 3 | n thy- and ] a,1 ljnr al Torres-Manzo Paola Alejandra , doi:10.1073/pnas.2008919117/-/DCSupplemental at online information supporting contains article This 3 2 1 the under Published interest.y competing no declare authors at The Medicine of Angeles.y School Los Geffen California David of G.A.B., University and the Chicago; of University A.C.B., Reviewers: n ...promdrsac;RRK,AR-...... M,adNC nlzddata; analyzed N.C. paper. and A.P.T.-M., the A.P.T.-M., wrote L.J., N.C. L.J., A.R.-N., and A.R.-N., R.R.K. R.R.K., R.R.K., and research; research; performed designed S.M.H. N.C. and and R.R.K. contributions: Author trce ra elo teto saptniltreatment pro-thermogenic potential be a to as thought generally attention are of has THs deal As obesity. thermogenesis (20). for great rea- adaptive unknown a the remains of weight but attracted stimulation gain controls, above, to than failure mentioned leaner their for are son mice hypothyroid hypothyroidism that Using induce models. to other used potentially drugs in the the of without effects hypothyroidism confounding study of to possible effects it metabolic makes that the model mouse a (22), NIS encoding body lower this unexplored. for largely (18– responsible remain weight mechanisms mice the euthyroid However, than 21). genetically leaner or consistently mice—whether treatment drug hypothyroid engineered—are by contrast, induced is In hypothyroidism associated (17). their often gain is energy weight hypothyroidism thermo- with promote humans, adaptive to In stimulating rise (14–16). by trying to genesis obesity continue on combat to considered focused to in expected were have expenditure is issue adults studies number US health Many this of serious (13). and 42% 2018, a over in is as obese 12), States, United (11, the expenditure energy and owo orsodnemyb drse.Eal [email protected] Email: addressed. be may correspondence whom To Brazil. y 08060-070, do Paulo Cruzeiro Sao Science, University, Health Sul in Program Post-Graduate Interdisciplinary address: Present Germany.y 69120, Heidelberg University, Heidelberg Medicine, of School address: Present olgtnwtresfrpooigwih loss. bring weight promoting help for eventually targets may new light and to hypothyroidism adaptations in physiological occur the into that insights strik- yield display findings thermogenesis. and Our adaptive skeletal-muscle food of levels less is higher consume ingly show hypothyroidism mice we of Here, hypothyroid controls. model that euthyroid mouse than leaner have drug-free paradoxically We a loss. weight that inducing found for a targets still therapeutic is of there obesity, lack underlying under- mechanisms as our the deepened causes of concern, have standing leading findings health many recent major Although with death. a associated of strongly is States is which United gain the world, weight in the increasing around steadily is and obesity of rate The Significance ehv eeae ncot(O iefor mice (KO) knockout generated have We input energy between imbalance an of result the Obesity, NSlicense.y PNAS b eateto oeua hsooyadBiophysics, and Physiology Molecular of Department Slc5a5 y b adoM Hirabara M. Sandro , Omc,w aedemonstrated have we mice, KO . y https://www.pnas.org/lookup/suppl/ NSLts Articles Latest PNAS h gene the Slc5a5, a,c,2 | , f8 of 1

PHYSIOLOGY (23–27); as a result, much of the available information on the Slc5a5 wild-type (WT) mice were switched to an iodide-sufficient topic of thermogenesis regulation by the THs has been obtained diet (ISD; 6 ppm iodide) with the same nutrient composition as from studies in which THs were administered to mice to generate the IDD, such that these mice were used as euthyroid controls peripheral or central hyperthyroidism (23–27). Here, in con- (SI Appendix, Fig. S1A). In agreement with our previous find- trast, we have capitalized on our Slc5a5 KO mice to determine ings (20), Slc5a5 KO mice on an IDD had lower serum levels of the effects of low levels of THs on several metabolic parame- THs (T3 [Fig. 1A] and T4 [Fig. 1B]) and markedly higher serum ters. These hypothyroid mice are leaner than euthyroid controls TSH levels (Fig. 1C) than Slc5a5 WT littermate controls fed and exhibit reduced body weight, lean mass, and fat mass, an ISD. owing in part to reduced caloric intake. Strikingly, our hypothy- Although our ISD and IDD are isocaloric, hypothyroid mice roid mice up-regulate skeletal-muscle adaptive thermogenesis— failed to gain body weight after being switched to an IDD most likely as a compensatory mechanism—although they are (Fig. 1D). MRI showed that hypothyroid mice had less fat mass still cold-intolerant. These results indicate that a series of phys- (Fig. 1E) and lean mass (Fig. 1F) than euthyroid controls and iological adaptations occur in hypothyroidism, which cumula- that most of the changes in the body weight of the hypothy- tively lead to a leaner phenotype. Identifying these adapta- roid mice were likely due to differences in lean mass (Fig. 1G). tions may enable us to develop novel strategies for combating When fat and lean mass were normalized to body weight, fat obesity. mass was still reduced in hypothyroid mice, whereas the two groups had proportionally similar amounts of lean mass (SI Results Appendix, Fig. S1 B and C). Overall fat mass normalized to body The Low Caloric Intake of Hypothyroid Mice Contributes to Their weight increased in both groups over time, whereas lean mass Lean Phenotype. We investigated the metabolic alterations that decreased (SI Appendix, Fig. S1D). These results thus indicate occur in hypothyroidism, using our whole-body Slc5a5 KO mouse that some component(s) of hypothyroid physiology promote a model. These mice offer the key advantage that their hypothy- leaner phenotype. roidism is not drug-induced. We have previously shown that Body weight is determined by the balance between nutrient when Slc5a5 KO mice are fed a standard rodent-chow diet, intake and energy output. To determine what factor(s) might which supplies 40 times the recommended daily amount of be contributing to the leaner phenotype of hypothyroid mice, iodide for rodents (28), iodide is in such great excess that it we first investigated nutrient intake by monitoring the weekly enters the thyroid via routes other than NIS, which partially food consumption of both groups of mice. One week after the rescues TH biosynthesis (22). Therefore, all mice were kept switch to an IDD, hypothyroid mice consumed significantly less on a chow diet (6 parts per million [ppm] iodide) for their food than euthyroid mice at a level that remained stable, but first 8 wk of life to promote normal development. After 8 wk, lower, for the entire 2 mo of monitoring (Fig. 1H). Therefore, Slc5a5 KO mice were switched to an iodide-deficient diet (IDD; the hypothyroid state caused mice to eat less and, thus, be leaner 0.02 ppm iodide), which renders such mice hypothyroid, and than euthyroid mice.

ABC

DE F

GH

Fig. 1. Hypothyroid (Hypo) mice eat less and are leaner than euthyroid (Eu) controls. Slc5a5 WT and KO mice were fed an ISD (6 ppm iodide) and IDD (0.02 ppm iodide), respectively, for 10 wk starting at 8 wk of age, rendering WT mice Eu and KO mice Hypo. (A–C) Serum T3 (A), T4 (B), and TSH (C) levels after 10 wk on special diet. For the TSH measurements, three Eu samples are below the detection limit of the assay, as depicted by the dotted line. Eu: n = 5; Hypo: n = 4. For D–F and H, week zero is a baseline measurement performed before the diet switch. (D) Weekly body weight of Hypo and Eu littermate controls. (E and F) MRI composition of fat mass (E) and lean mass (F). (G) Average fat mass and lean mass gained over 10 wk of special diet. (H) Weekly food intake of Hypo and Eu littermate controls. Eu: n = 10; Hypo: n = 9. Data are expressed as the mean ± SEM and were analyzed by an unpaired t test (A–C and G) or a two-way ANOVA followed by a Bonferroni correction (D–F and H). **P < 0.01; ***P < 0.001; ****P < 0.0001.

2 of 8 | www.pnas.org/cgi/doi/10.1073/pnas.2008919117 Kaspari et al. Downloaded by guest on September 28, 2021 Downloaded by guest on September 28, 2021 h vrg cosec ru fmtblcraig olce t5mnitrasta hnudretannierrgeso ihasn-ucinfi.Data fit. sine-function a al. with et regression Kaspari nonlinear a underwent then that intervals 5-min at collected readings metabolic of in group each (C across analysis average cage the metabolic entire the across movement core maintain to the used it of energy split sum the and the we movement is for mice, expenditure used energy energy hypothyroid Total in components: its altered into is expenditure energy expenditure energy controls. notion in euthyroid and the differences mice supports fundamental hypothyroid between further expendi- are which energy there weight, in that decrease increased a with showed expenditure ture surpris- mice energy phenotype—but, hypothyroid revealed in expected ingly, weights—the increase analysis body an regression higher exhibit at The mice probably variable. euthyroid is that body-mass adjust mice quantitatively the cannot we hypothyroid for differences, performed mass and body expen- be to energy euthyroid related not in difference between could the of analysis diture some although ANCOVA Thus, 30). an (29, between so different S2D), very hypothyroid: Fig. were 0.0316; body slopes (euthyroid: in regression groups overlap the not did (ANCOVA). and groups covariance weight, weight, experimental of body two analysis our in an However, perform differences to for intended adjust we and in expenditure difference the energy 2 to (Fig. them contributor between substantial expenditure a energy is 1D) (Fig. groups quotients respiratory , Appendix S2C). similar Fig. (SI very Appendix, had cycles (SI groups inactive both and but less a S2B), active Fig. consumed with the both mice Consistent during S2A). hypothyroid oxygen active Fig. expenditure, the Appendix, energy both (SI during reduced mice cycles hypothyroid inactive in representative and lower is was which BMR, expenditure, of energy resting average the aeaayi.(A analysis. cage 2. Fig. 2 (Fig. cycle active the differ- during less pronounced occurring expend most ence the to with appear controls, euthyroid mice than hypothyroid energy Overall, periodicity group. active/inactive each the the analyzed in with have instead synchronized we data mice, dark of metabolic and groups inac- and both light active in indicates the capture periods better tive between shift To mice. evenly large hypothyroid split a our in is Such cycles period h. mice active 6.9 hypothyroid the remarkable that of a by rhythms shifted circadian dark-cycle were the the to prior whereas h a 2.3 midpoint, have approximately wave mice of sine euthyroid offset a activity that energy- peak to emerged the it fit rhythmicity, regression of analyzed quantitate nonlinear to rhythms we a circadian When using data the altered. expenditure that are suggest mice to results hypothyroid controls euthyroid These them hypothy- of that 2A). by to switching relative (Fig. expenditure time after in energy shifted the was that wk from mice roid is emerged 10 that analysis pattern mice cage striking metabolic of A diets. groups respective anal- their both cage metabolic on a performed ysis we expenditure, energy titate Activity. and Expenditure Energy Reduces Hypothyroidism ABCD B ogi etrudrtnigo hte—n,i o how— so, if whether—and, of understanding better a gain To two the between weight body in difference the that expect We and yohri Hp)mc xii eue nryepniueadatvt.Mc eefdterrsetv it o 0w ro oametabolic a to prior wk 10 for diets respective their fed were Mice activity. and expenditure energy reduced exhibit mice (Hypo) Hypothyroid D ersn h mean the represent and .(C (B). cycles active/inactive the across averaged and (A) analysis cage metabolic entire the across expenditure Energy B) ± E n eeaaye ya unpaired an by analyzed were and SEM A −0.0389; and A n vrgdars h cieiatv yls(D). cycles active/inactive the across averaged and ) .T quantitate To B). and IAppendix, SI .O note, Of B). oquan- To t et *P test. i.S3 the Fig. but Appendix, levels, (SI protein significant UCP1 B of statistically and higher not BAT S3A), toward was the Fig. trend difference in Appendix, a (SI higher was mice there slightly euthyroid were than levels 33). hypothyroid (mRNA) (32, (UCP1) RNA 1 ger protein adap- uncoupling tissue of protein as membrane mediator adipose mitochondrial known key inner the the brown is BAT, thermogenesis In thermogenesis: tive muscle. the skeletal adaptive investigated and that of (BAT) we mice, thermogenesis sites hypothyroid the main and between euthyroid 3A). in finely (Fig. occurs not more did differentiate controls euthyroid To the of that signif- whereas decreased icantly, temperature body core However, their hypothyroid, switch. them diet body the the core after similar before very had pre-diet—i.e., mice temperatures of groups two energy The of expenditure. component main other the thermogenesis, investigated Markedly Mice. Are Hypothyroid Thermogenesis in Adaptive Up-Regulated Skeletal-Muscle of Markers as 31). even such 21, 17, hypothyroidism, mice, (2, activity of euthyroid and signs expenditure than energy classic phenotype reduced exhibited leaner they a though hypothy- had drug-free our mice summary, In roid cycle. active the during mice fsroii a iie oso-wthseea uce:We S3 muscles: Fig. hypothyroid skeletal Appendix , of (SI diaphragms 3 slow-twitch mice the up-regulation in to (Fig. 36), expression limited increased mice (35, detected was results hypothyroid previous sarcolipin of with of muscle agreement soleus In D). the in higher 34). the (15, thermogenesis by adaptive mediating plays hydrolysis in (SERCA), role ATP important ATPase an from calcium reticulum reticulum sarco/endoplasmic sarcoplasmic muscle, skeletal the uptake calcium In into uncouples D). S3 which of sarcolipin, Fig. proteolipid BAT small the Appendix, the in (SI thermogenesis mice of hypothyroid that activation suggesting moderate mice, was euthyroid from there BAT of lipids staining fewer H&E showed than and multilocular more was mice hypothyroid move- Total thermogenesis). 2 for (Fig. used ment (i.e., temperature body hc a otiuet hi enrphenotype. leaner their mice, to contribute hypothyroid may in which mechanisms skeletal- These thermogenic of stimulation adaptive (FAO). pronounced muscle a oxidation is acid there that fatty suggest results increased of oxida- lipid indicative incomplete tion, represent that intermediates [ cycle [ Krebs more and produced 3E) mice (Fig. hypothyroid S3 of Fig. muscles Appendix , soleus (SI quadriceps their G or gastrocnemius their and ohsroii RAadpoenlvl eemarkedly were levels protein and mRNA sarcolipin Both and < .5 **P 0.05; .Hmtxlnadesn(&)sann fBTfrom BAT of staining (H&E) eosin and Hematoxylin C). H .sn [ ).Using Slc5a5 < C 14 .1 S o infiat v. vrg;E,euthyroid. Eu, average; Avg., significant. not NS, 0.01. and Omc eesice oa D orender to IDD an to switched were mice KO ]ai-oul eaoie Fg 3F (Fig. metabolites C]-acid-soluble 14 ]pliaea rcr efudta the that found we tracer, a as C]-palmitate a rsial eue nhypothyroid in reduced drastically was D) n o ohgop.Dt in Data groups. both for 7 = E and F nadto oatvt,w also we activity, to addition In ,btn xrsini either in expression no but ), NSLts Articles Latest PNAS A Ucp1 and and ,sc as such ), C 14 represent messen- C]-CO Total D) | f8 of 3 B– 2

PHYSIOLOGY ABC

DEF

Fig. 3. Hypothyroid (Hypo) mice exhibit signs of stimulated skeletal-muscle adaptive thermogenesis. Mice were fed their respective diets for 10 wk. (A) Core body temperature (temp.) measured using a rectal probe before the diet switch (Pre-diet) and then after 10 wk on special diet (Post-diet). n = 4 for both groups. (B and C) Sarcolipin mRNA (n = 3 or 4) (B) and protein expression (n = 5) (C) in the soleus muscle. (D) Quantification of C.(E and F) Amount of 14 14 14 [ C]-CO2 (E) and [ C]-acid-soluble metabolites (F) produced by solei incubated in [ C]-palmitate as a tracer. Data are expressed as the mean ± SEM and were analyzed by an unpaired t test. **P < 0.01; ***P < 0.001; ****P < 0.0001. NS, not significant. Eu, euthyroid.

To investigate whether adaptive thermogenesis is stimulated in hypothyroid mice stimulate sarcolipin expression, as TSH has in hypothyroid mice as a compensatory mechanism to help them been implicated in BAT thermogenesis (specifically, restoring maintain their core body temperature, we subjected the mice to expression of the TSH receptor to the BAT of TSH receptor a cold challenge. We placed euthyroid and hypothyroid mice at [Tshr] KO mice partially rescued thermogenesis; ref. 37). To 4 ◦C and measured their core body temperature every 15 min. ascertain whether or not elevated levels of TSH are directly The core body temperature of the hypothyroid mice decreased responsible for increased sarcolipin expression, we used another dramatically as a result of the challenge, and they were therefore mouse model of hypothyroidism: Tshr KO mice (38). As these removed from the 4 ◦C environment after 135 min (Fig. 4A). mice lack the TSH receptor, they are unable to stimulate TH syn- Infrared imaging confirmed that, after 2 h at 4 ◦C, hypothy- thesis or release, making them markedly hypothyroid. Tshr WT roid mice were clearly much colder than euthyroid controls control mice were kept on a chow diet (6 ppm iodide), whereas (Fig. 4B). These results demonstrate that hypothyroid mice lack Tshr KO mice were kept on a 6 ppm iodide, TH-supplemented the metabolic mechanisms necessary to maintain their core body diet for the first 8 wk of life to promote normal development. temperature in response to an extreme cold stress. After 8 wk, Tshr WT mice were placed on an ISD and were used Interestingly, sarcolipin expression was still much higher in as euthyroid controls, while Tshr KO mice were placed on an the soleus of hypothyroid than euthyroid mice after the cold IDD and were used as a model of hypothyroidism (SI Appendix, challenge (Fig. 4 C and D), while UCP1 levels in the BAT Fig. S5A). As expected, Tshr KO mice had reduced serum lev- of euthyroid and hypothyroid mice remained very similar (SI els of THs (T3 [SI Appendix, Fig. S5B] and T4 [SI Appendix, Appendix, Fig. S4 A and B). However, because the cold chal- Fig. S5C]) and increased serum TSH levels (SI Appendix, lenge lasted only a very short time (135 min), it is likely that Fig. S5D). Sarcolipin expression was higher in the solei of these changes in protein expression occurred at room temper- Tshr KO mice than Tshr WT mice (Fig. 5 A and B). Further- ature, prior to the cold exposure. Consistent with this notion, more, sarcolipin up-regulation in hypothyroid mice correlated sarcolipin expression levels in the soleus muscle (Fig. 4 E and with higher levels of FAO in their soleus muscle (measured by F) and UCP1 levels in the BAT (SI Appendix, Fig. S4 C and using [14C]-palmitate as a tracer). Solei from both hypothyroid D) were very similar in hypothyroid mice kept at room temper- mouse models (Tshr KO and Slc5a5 KO) produced more [14C]- 14 ature and hypothyroid mice that underwent a cold challenge. CO2 (Fig. 5C) and [ C]-acid-soluble metabolites (Fig. 5D) than These results show that hypothyroid mice were unable to main- solei from euthyroid controls. These results show that signaling tain their core body temperature when placed at 4 ◦C, even through the TSH receptor does not stimulate either sarcolipin though their UCP1 expression was similar to euthyroid controls expression or FAO. and their skeletal muscle was “primed” for adaptive thermo- It has been proposed that sympathetic nervous system (SNS) genesis due to the increased sarcolipin expression, suggesting signaling may regulate skeletal-muscle adaptive thermogenesis that other mechanisms, such as a reduced BMR, may be coun- and sarcolipin expression (39), owing to the role it plays in stim- teracting the increased adaptive thermogenesis. Furthermore, ulating adipose-tissue adaptive thermogenesis (1, 16, 27, 33, 40, shivering is the first line of defense used to maintain core body 41). In agreement with previous findings (42–44), catecholamine temperature before adaptive thermogenic mechanisms are trig- levels were higher in hypothyroid than in euthyroid mice after gered (33). As our hypothyroid mice had to be removed from the 10 wk on their respective diets (Fig. 6 A and B). To ascertain cold challenge so quickly, their shivering may be impaired. whether or not the elevated levels of catecholamines found in our hypothyroid mice were directly responsible for their increased Neither TSH nor Catecholamine Signaling Play a Key Role in sarcolipin expression, we quantitated the levels of plasma cat- Hypothyroidism-Induced Sarcolipin Expression and Increased FAO. echolamines in both groups at an earlier time point, with the Very little is known concerning the regulation of sarcolipin rationale that SNS compensation may not yet have been trig- expression. However, it is possible that the high levels of TSH gered at that point. We have previously shown that Slc5a5 KO

4 of 8 | www.pnas.org/cgi/doi/10.1073/pnas.2008919117 Kaspari et al. Downloaded by guest on September 28, 2021 Downloaded by guest on September 28, 2021 ***P Hypo: apr tal. et Kaspari (C ieetyod(Eu euthyroid mice i.5. Fig. (22 temperature room at kept mice Hypo a of underwent muscle that soleus mice of Hypo the and Quantification 4 Eu (D) at of challenge. h muscle cold 2 soleus Hypo the after a in and and expression challenge (Eu) tein cold euthyroid the a of before images mouse infrared transponders. Representative temperature (B) groups. via 4 taken to moved (temp.) then temperature were (22 They wk. temperature 10 room at housed 4. Fig. aclpnepeso,rmral,wssilu-euae nthe in occurred. up-regulated yet still not was had remarkably, compensation As expression, SNS Sarcolipin (20). that IDD indicating trols, an on 6 (Fig. levels wk S5 Fig. IDD, Appendix, 3 the (SI on roid only wk 4 after after expected, hypothyroid are mice in Data ****P unpaired correction. an Bonferroni by a analyzed by were followed mean ANOVA two-way the a as by expressed are Data (4 groups. challenge cold a underwent CD EF AB ABCD and < n < muto [ of Amount D) .0;**** P 0.001; S sntrqie o tmlto fsroii xrsinadFAO. and expression sarcolipin of stimulation for required not is TSH .Dt r xrse stemean the as expressed are Data 3. = yohri Hp)mc r eeeycl-noeat iewere Mice cold-intolerant. severely are mice (Hypo) Hypothyroid .01 S o significant. not NS, 0.0001. A and T n Omc yohri (Hypo hypothyroid mice KO and ) < 14 eesmlrt hs fteetyodcon- euthyroid the of those to similar were B) .01 yo hypothyroid. Hypo, 0.0001. C]-CO 2 ,btterpam catecholamine plasma their but E–G), t (C ◦ et *P test. ) (F C). n [ and ) ◦ ◦ )adfdterrsetv it for diets respective their fed and C) (E C. oebody Core (A) challenge. cold a for C unicto of Quantification ) Slc5a5KO 14 < ± rdcdb oe nuae n[ in incubated solei by produced (D) metabolites C]-acid-soluble aclpnpoenepeso in expression protein Sarcolipin ) .5 **P 0.05; E.Dt in Data SEM. ± E n eeaaye yaoewyAOAfloe yaTkyps o nlss *P analysis. hoc post Tukey a by followed ANOVA one-way a by analyzed were and SEM iewr hypothy- were mice T ◦ < unicto of Quantification B) ( muscle. soleus the in sarcolipin of expression Protein (A) ). .(C C. .1 ***P 0.01; E. A aclpnpro- Sarcolipin ) n eeanalyzed were n o both for 5 = o both for 5 = ◦ )o that or C) < D 0.001; and F Tshr ftecokpoen eid1adPro oatrcircadian alter to in 2 signaling Period mTOR regulate and THs 1 above, expression mentioned Period the As proteins rhythms. regulate clock to the hypothalamus suprachiasmatic of the the of in circadian located (SCN) our clock nucleus circadian shifted that master the in displayed in nals authors’ theirs, 2 those (Fig. unlike rhythms from mice, differ energy. reduced hypothyroid results conserve a that our have changes mice However, phenotypic hypothyroid despite that paradox- weight, seem show body findings they These because 31). that ical, (21, showing expenditure activity energy results running-wheel and reduced with exhibit rats consistent and mice are 2 hypothyroid findings (Fig. Our controls euthyroid roidism. less than expend activity mice less and display hypothyroid and and Our BMR energy decreased expenditure. a energy with lower associated often is hypothyroidism regulate to CNS the in intake. be food together to yet signal has they it (41), whether tissues THs determined other the in synergistically Although con- signal consumption. also SNS food and may decreased mice elevated their hypothyroid the to our tribute that in suggesting catecholamines (47–49), of levels intake food associated reduced been has with signaling SNS increased Furthermore, decreased in tions. reported occurs hypothalamic have opposite of (46) the al. levels that et inferred Coppola as be hypothyroidism, can shown It mitochon- T have uncoupling. UCP2-mediated (46) of drial increased al. production through feeding et local promote Coppola stimulates reported and was fasting (45), hypothalamus that feeding the increase of in (ARC) signaling to (mTOR) nucleus rapamycin arcuate of the consume target they mammalian because of vation part 1H leaner in (Fig. are least food mice less at gain hypothyroid not controls, that do euthyroid shown they than why have understood we well Here, not weight. is it drug-generated However, from (18, 21). hypothyroidism findings 19, of with models mouse agreement engineered genetically in and is that result reported compen- a have triggers We loss. mice weight that in promote hypothyroidism that mechanisms that satory show we Here, on depend Discussion not do FAO increased signaling. catecholamine and expression sarcolipin 6F (Fig. lites eeso [ of levels 6 (Fig. conditions these C under mice hypothyroid of muscle soleus H r e euaoso M hogotlf 13,and (1–3), life throughout BMR of regulators key are THs TadK iewr e hi epciedesfr4w,rneigWT rendering wk, 4 for diets respective their fed were mice KO and WT and —yia aiettoso hypothy- of manifestations S2A)—typical Fig. Appendix, SI Slc5a5 .I diin yohri ieehbtdincreased exhibited mice hypothyroid addition, In D). 14 Omc e nIDfi ogi oywih (20), weight body gain to fail IDD an fed mice KO C]-CO ,srnl ugsigta hypothyroidism-induced that suggesting strongly ), A and .I etal yetyodrtmdl acti- model, rat hyperthyroid centrally a In ). 2 n [ and 6E) (Fig. 14 .Coe l 5)soe htmO sig- mTOR that showed (50) al. et Cao C). ]pliaea rcr Eu tracer. a as C]-palmitate Ucp2 RAudrhptyodcondi- hypothyroid under mRNA 14 ]ai-oul metabo- C]-acid-soluble NSLts Articles Latest PNAS A. T 3 : < n n nteACto ARC the in o l groups. all for 4 = ;Hypo 4; = .5 **P 0.05; | T < : n f8 of 5 0.01; 4; =

PHYSIOLOGY ABTHs negatively regulate sarcolipin expression in skeletal mus- cle. Therefore, when TH levels are low, sarcolipin expression in the skeletal muscle increases. Consistent with this notion, Minamisawa et al. (54) proposed that THs accelerated sarcolipin mRNA degradation in the heart, and Johann et al. (25) recently reported that sarcolipin mRNA levels are lower in the soleus muscles of hyperthyroid than euthyroid mice. Finally, it has been known for decades that THs play an important role in calcium cycling in the skeletal muscle by stimulating expression of SERCA and the ryanodine receptor (Ryr) (1). Therefore, it seems highly plausible that the THs could also be involved in regulating sarcolipin expression and skeletal muscle CDmetabolism. Taken together, our findings suggest that the decreases in THs and BMR that occur in hypothyroid mice trigger compensatory mechanisms that, in turn, stimulate adaptive thermogenesis in an attempt to maintain core body temperature. These processes along with a reduced food intake ultimately result in a leaner phenotype. There is no question that the compensatory mech- anisms triggered in hypothyroidism merit deep and sustained investigation, as they may eventually open up new avenues to EFcombating obesity. Materials and Methods Animals. The Yale University Institutional Animal Care and Use Committee approved all mouse experiments. Mice were housed in a pathogen-free facility at Yale University on a 12-h light/dark cycle. All experiments were performed on male littermates housed at approximately 22 ◦C unless spec- ified otherwise. Mice were fasted overnight prior to the FAO experiments and plasma collection for norepinephrine and epinephrine. The metabolic cage analysis was performed by Vanderbilt University’s Mouse Metabolic Phenotyping Center (MMPC). Fig. 6. Catecholamines are not required for stimulation of sarcolipin Slc5a5 KO mice on the C57BL/6J background were generated as described expression and FAO. (A and B) Plasma norepinephrine (A) and epinephrine (22). We have shown that when Slc5a5 KO mice are placed on a standard (B) levels in euthyroid (Eu) and hypothyroid (Hypo) mice after 4 and 10 wk rodent-chow diet (ENVIGO catalog no. 2018S), which contains 40 times more on either ISD or IDD, respectively. n = 7 to 9 for both groups. For C–F, mice iodide than the daily recommended amount for rodents (28), iodide is in were fed their respective diets for 4 wk. (C) Sarcolipin protein expression in such great excess that it enters the thyroid via nonspecific routes, which the soleus muscle. (D) Quantification of C. n = 4 for both groups. (E and F) partially rescues TH biosynthesis (22). Therefore, Slc5a5 KO mice develop Amount of [14C]-CO (E) and [14C]-acid-soluble metabolites (F) produced by 2 normally and only become hypothyroid when placed on an IDD. To generate solei incubated with [14C]-palmitate as a tracer. n = 4 for both groups. Data Slc5a5 WT euthyroid and Slc5a5 KO hypothyroid mice, all mice were given are expressed as the mean ± SEM and were analyzed by an unpaired t test. ad libitum access to a standard chow diet (6 ppm iodide; ENVIGO catalog *P < 0.05; ***P < 0.001; ****P < 0.0001. NS, not significant. no. 2018S) for the first 8 wk of life to promote normal development. After 8 wk, Slc5a5 WT mice were switched to ad libitum access of an ISD (ENVIGO catalog no. TD.170762; 6 ppm iodide) to keep them euthyroid, whereas the ARC to promote food intake (45); therefore, it seems plau- Slc5a5 KO mice were switched to ad libitum access of an IDD (ENVIGO sible that hypothyroidism alters mTOR signaling in the SCN. catalog no. TD.120723; 0.02 ppm iodide) to render them hypothyroid (SI However, it remains to be thoroughly investigated how TH dis- Appendix, Fig. S1A). The ISD and IDD have the same nutrient composition orders bring about metabolic and behavioral changes by altering and only differ in the amount of iodide supplied. Mice were fed their respec- tive diets for either 4 or 10 wk, as specified in the text. Slc5a5 euthyroid circadian rhythms. and hypothyroid mice are referred to as “Eu” and “Hypo,” respectively, in Stimulation of adaptive thermogenesis has recently emerged the figures. as a promising approach to promoting weight loss (14–16). THs To generate Tshr euthyroid and hypothyroid mice, the following diet are generally thought to be prothermogenic (1, 41, 51–53); as regimen was followed: Tshr WT mice were given ad libitum access to a a consequence, most studies of THs have been carried out standard rodent-chow diet (6 ppm iodide; ENVIGO catalog no. 2018S) for by using systemically (25–27) or centrally (23, 24) hyperthy- the first 8 wk of life before being switched to the ISD (ENVIGO catalog roid animals. Here, in contrast, we studied hypothyroid mice, no. TD.170762; 6 ppm iodide) to keep them euthyroid. Tshr KO mice were which displayed moderate activation of BAT thermogenesis (SI given ad libitum access to a TH-supplemented diet (ENVIGO catalog no. Appendix, Fig. S3 A–D) and pronounced stimulation of skeletal- TD.170763; 6 ppm iodide and 0.33 ppm levothyroxine) for the first 8 wk of life to keep them euthyroid throughout development. After 8 wk, Tshr muscle adaptive thermogenesis (Fig. 3 B–F), but were still KO mice were switched to ad libitum access of the IDD (ENVIGO catalog severely cold-intolerant (Fig. 4 A and B). Our results indicate no. TD.120723; 0.02 ppm iodide) to render them hypothyroid (SI Appendix, that some component of hypothyroidism, such as the reduction Fig. S5A). Mice were fed their respective diets for 4 wk. Tshr euthyroid in BMR, interferes with the ability of these mice to maintain and hypothyroid mice are referred to as “EuT” and “HypoT,” respectively in their core body temperature. Furthermore, our findings suggest the figures. that skeletal-muscle adaptive thermogenesis may play a more prominent role in TH-induced temperature alterations than does Temperature Measurements. Rectal temperature measurements were per- the adipose tissue. This is in agreement with recent results pre- formed by using a BIOSEB TK8851 rodent thermometer, while transpon- der measurements were performed by using programmable tempera- sented by Johann et al. (25) and Dittner et al. (26), who reported ture transponders (IPTT-300; Bio Medic Data Systems). Transponders were that hyperthyroidism triggers pyrexia without signaling through injected at 6 wk of age into the subcutaneous region of the side body. Pre- UCP1-dependent mechanisms. diet measurements were conducted at 8 wk of age immediately prior to the Having found that neither TSH (Fig. 5) nor catecholamines switch to a special diet. Post-diet measurements were performed after 10 (Fig. 6) are involved in sarcolipin regulation, we postulate that wk of special diet.

6 of 8 | www.pnas.org/cgi/doi/10.1073/pnas.2008919117 Kaspari et al. Downloaded by guest on September 28, 2021 Downloaded by guest on September 28, 2021 1 .M peemn .S le,Oeiyadterglto feeg balance. energy of regulation the and Obesity Flier, S. J. Spiegelman, M. in B. mutation missense 11. novel A Weiss, E. R. Abdullah, A. M. Ebrhim, S. R. Watanabe, Y. 10. apr tal. et Kaspari using by generated gene were housekeeping sequences Primer the expres- conditions. of Gene an Primer-BLAST. same 1010180). expression on the no. to performed accor-under catalog normalized was (Illumina in was and System 95072) sion instructions PCR no. Real-Time manufacturer’s catalog Eco the (Quantabio with FastMix Quantabio dance manufacturer’s Green the the using SYBR SuperMix by with performed PerfeCTa DNA cDNA accordance was cDNA in qScript Complementary of 95048) Quantabio Amplification 74104). instructions. no. the and no. catalog using 15596018) (Quantabio catalog by no. kit (Qiagen synthesized catalog was Kit (Ambion (cDNA) TRIzol Mini using RNeasy at by stored the and extracted dissection was upon RNA nitrogen liquid in frozen Analysis. PCR Real-Time the to according used was were kits instructions. Panel manufacturer’s Both Bead MPTMAG-49K). no. Magnetic catalog Pituitary T (Millipore plasma Mouse measure MAP to Ser- Milliplex used Analytical The and Assay Core. Hormone VUMC vices the T by Similarly, radioimmunoassay Core. a (VUMC) using Services Center Analytical Medical and University Assay Vanderbilt Hormone the elec- by via chromatography detection liquid trochemical high-performance by epinephrine measured were and levels at norepinephrine stored spun and for quickly down were plasma Samples collection. following to immediately measurements added was glutathione Measurements. Hormone counter. scintillation a using [ of levels olwdb diino C.NO tacnetaino a de into added for was vial M incubation 2 the of in concentration compartment a isolated at an NaOH HCl. of [1– addition Ci/mL by 35 0.1 followed and at palmitate bath con- total Buffer water mM Bicarbonate shaking 0.1 Krebs–Ringer a taining a in to min moved were 60 muscles to acclimation, 30 Muscles for (Sigma; tension. K4002) Buffer constant no. Bicarbonate Krebs–Ringer catalog under oxygenated muscle an the in keep acclimated were to clip stainless-steel a to FAO. software Tools+ FLIR analysis. the image for and used Technology, fell was immedi- UltraMax temperature were with body Camera Mice core Imaging of body. their wk if side challenge 6 the 30 cold below at of the injected from region were removed subcutaneous transpon- ately which the temperature Systems), into programmable Data using age Medic by Bio (IPTT-300; min 15 ders every made were 4 to moved 22 Challenge. Cold .C otln,M aoe-eeisy .Crac,TeNa The Carrasco, N. Paroder-Belenitsky, M. metabolism. Portulano, of regulation C. hormone Thyroid 3. Brent, A. G. Liu, Y. Y. Mullur, R. regulation. 2. hormonal their and mechanisms Thermogenic Silva, E. J. 1. .J oln,S eeof uain ntesdu/oiesmotr(I)gn sa as gene (NIS) symporter sodium/iodide the in Mutations Refetoff, structural S. a Pohlenz, of Identification J. Carrasco, iodide 7. N. Levy, thyroid D. the Vieja, La of De characterization A. Ginter, and C. Cloning Levy, O. Carrasco, 6. N. Levy, O. Dai, G. 5. sodium/iodide The Carrasco, N. Amzel, M. L. Ferrandino, G. Reyna-Neyra, A. Ravera, S. 4. .M .Re-sr .D aVea .S itr .Crac,Mlclrcharacterization Molecular Carrasco, N. Ginter, S. C. Vieja, la De A. Reed-Tsur, D. M. 9. Nicola P. J. 8. Rev. (2006). 435–464 3–4 (2001). 531–543 hypothyroidism. congenital and defects transport iodide for cause hypothyroidism. congenital human causes that Na thyroid for requirement transporter. applications. clinical Physiol. and preclinical Rev. and physiology Molecular (NIS): symporter impact. medical and Mechanism 0817 (2018). 1068–1070 the Na Endocrinology a NIS, V59E of Na hypothyroidism. the in tion (1999). 476 ◦ o h rt1 ko ie fe 0w fete S rID hywere they IDD, or ISD either of wk 10 After life. of wk 18 first the for C h oesmsl a sltdfo nshtzdmc n attached and mice anesthetized from isolated was muscle soleus The SLC5A5 5–8 (2014). 355–382 94, ◦ 14 .Ifae mgn a efre yuigaFI 40Thermal T460 FLIR a using by performed was imaging Infrared C. C]-CO ◦ tal. et o odcalne oebd-eprtr measurements body-temperature Core challenge. cold a for C eei uaeefml ihcneia hypothyroidism. congenital with family Sudanese a in gene Nature 6–8 (2017). 261–289 79, 0738 (2008). 3077–3084 149, + uhri n yohri itrae eemitie at maintained were littermates hypothyroid and Euthyroid oietasotdfc:Fntoa hrceiaino oe muta- novel a of characterization Functional defect: transport Iodide , /I .Ci.Edcio.Metab. Endocrinol. Clin. J. 2 – + n [ and /I ypre ’utasae eini ain ihcongenital with patient a in region 5’-untranslated symporter 5–6 (1996). 458–460 379, −80 – ypre uatta asscneia I congenital causes that mutant symporter ise o N xrcinwr meitl flash- immediately were extraction RNA for Tissues 14 ◦ h rsraieehln lclttactcacid– tetraacetic glycol ethylene preservative The 4 ni nlss oeiehieadepinephrine and Norepinephrine analysis. until C + ]ai oul eaoie eedtrie by determined were metabolites soluble C]-acid /I Mlioectlgn.RHMG3K n TSH and RTHYMAG-30K) no. catalog (Millipore – ypre NS ucinfo nlsso mutation a of analysis from function (NIS) symporter nor Rev. Endocr. 10–10 (2011). E1100–E1107 96, 0–4 (2014). 106–149 35, ESLett. FEBS 14 ]pliaefru o6h, 6 to up for palmitate C] 14 CO 64 (1998). 36–40 429, 2 3 bopin Finally, absorption. + −80 /I a esrdby measured was – – Biochimie rnpr defect. transport ypre (NIS): symporter hso.Rev. Physiol. ◦ ni use. until C Thyroid ◦ 18S .After C. Cell 469– 81, Physiol. Annu. 104, run 86, 28, 7 .Cae,A .Bac,J okas .P etr,Hypothyroidism. Peeters, P. R. Jonklaas, J. Bianco, C. thermogene- A. adaptive Chaker, in L. advances New 17. Spiegelman, M. B. Kazak, L. Chouchani, T. E. 16. Bal C. N. induces 15. ablation UCP1 Nedergaard, J. Cannon, B. Golozoubova, severe V. and Feldmann, obesity M. of H. Prevalence Ogden, 14. L. C. Fryar, D. C. Carroll, D. obesity. M. and Hales, balance M. Energy C. Peters, C. 13. J. Wyatt, R. H. Hill, O. J. 12. 2 .Ferrandino G. 22. Patyra K. 21. Ferrandino G. 20. to Weiner J. and hormone 19. thyroid to Responsiveness Bianco, C. A. Olivares, L. E. Ueta, B. C. 18. n nltclSrie oei upre yNHGat K567and DK059637 Grants NIH by Assay supported Hormone is VUMC Core by The DK020593. Services supported 1S10RR028101-01. Vanderbilt is Analytical The and MMPC and The R.R.K.). DK059637 Grants analysis. (to Grants NIH cage DK118814-01A1 by NIH metabolic supported the 1F31 discus- Dr. performed was and helpful MMPC and work N.C.) for This Lantier, (to York). and DK041544 New Louise comments Center, Dr. Medical insightful Sinai Forbush, their Biff for The sion. Dr. Caplan J. McGuinness, P. Michael Owen Dr. ACKNOWLEDGMENTS. Availability. Data for h 7.9 and euthyroid for h 3.4 hypothyroid). quotient: respiratory VO and hypothyroid; h hypothyroid; for 2.3 for h movement: 6.3 expenditure—total and energy euthyroid for for very calculated were those parameters to other parameters. evalua- the similar of for for comparison shifts used permit phase to determined were parameters (Independently cycles metabolic other active/inactive all same of from 6.9-h tion The and phase-shifted and cycle. accordingly Active periods dark/light respectively. 2.3- 12-h the mice, as obtaining defined hypothyroid were software, and cycles 7 inactive euthyroid for Prism shifts GraphPad phase using fit function as expressed are figures all significant. in not Data indicates values. used missing was pg/mL) the assay. means (180.74 for assay the substitute the of of a In limit limit as detection detection software. the the analyses, 7 these below For Prism were respectively, GraphPad ANOVA mice, two-way on euthyroid a correction or 1C Bonferroni analysis, Fig. hoc a post by Tukey a followed by followed ANOVA way Analysis. Statistical catalog control. loading (Abcam 1:1,000 a (COXIV). or as IV used ab8245) was subunit ab33985) no. oxidase a no. c catalog Either cytochrome (Abcam ab10983). against no. GAPDH dilution catalog against no. (Abcam dilution catalog primary UCP1 1:10,000 Sigma against with (Millipore 1:10,000 probed sarcolipin and was ABT13) against membrane 1:1,000 transfer The semidry diluted 221B). SD antibodies no. Transblot catalog a mem- using (Bio-Rad nitrocellulose 1620112) cell 0.2-µm no. gel a Bis-Tris catalog to 12% (Bio-Rad Pro- transferred to brane P-40. and 4 NuPAGE Scientific) Nonidet Invitrogen Fisher 0.5% an (Thermo on and electrophoresed 11836170001), were no. teins ethylenediaminetetraacetic catalog Mini, (Roche cOmplete acid-free tablets Na cocktail mM 5 inhibitor NaCl, tease mM 150 containing 7.5), buffer (pH in at Tris·HCl homogenized mM stored were 50 tissue and BAT and dissection muscle upon skeletal Both nitrogen liquid in frozen Analysis. Blot Western oass htmct,w efre olna ersinwt sine a with regression nonlinear a performed we rhythmicity, assess To i:UP n beyond. and UCP1 sis: mammals. in thermogenesis stress thermoneutrality. thermal at from living exempt National mice by in thermogenesis 360, diet-induced abolishes Brief and obesity Data (NCHS 2020). MD, 2017-2018 Hyattsville, Statistics, States, Health United for Center adults: among obesity (2012). 132 yohriimi Na in hypothyroidism mice. in (2018). lesions 6239–6251 carcinoma-like thyroid papillary and hyperplasia, roidism, (2017). mechanisms. extrahepatic and intra- mice. in tissues adipose white of browning obesity. diet-induced of model (2011). mouse 3571–3581 a 152, in skele- and thermogenesis tissue muscle adipose tal brown between differences underlies temperature ambient (2017). 1562 ± and Tshr E.*P SEM. ata hrct-pcfi Gα thyrocyte-specific Partial al., et hri omn ttsdfie rw dps iseatvt and activity tissue adipose brown defines status hormone Thyroid al., et he n v fteTHvle o the for values TSH the of five and three S5G, Fig. Appendix, SI aclpni el dnie euao fmuscle-based of regulator identified newly a is Sarcolipin al., et Omc eeagnru itfo r er ais(Mount Davies Terry Dr. from gift generous a were mice KO ahgnsso yohriimidcdNFDi rvnby driven is NAFLD hypothyroidism-induced of Pathogenesis al., et neteeyhg itr oiespl oetlssevere forestalls supply iodide dietary high extremely An al., et l td aaaeicue nteatceand article the in included are data study All < eut eeaaye yuiga unpaired an using by analyzed were Results + .5 **P 0.05; elMetab. Cell /I ise o etr ltigwr meitl flash- immediately were blotting Western for Tissues – etaktemmeso h ..laboratory, N.C. the of members the thank We ypre NS ncotmice. knockout (NIS) symporter a.Med. Nat. elMetab. Cell < 73 (2019). 27–37 29, .1 ***P 0.01; rc al cd c.U.S.A. Sci. Acad. Natl. Proc. 5517 (2012). 1575–1579 18, 0–0 (2009). 203–209 9, c.Rep. Sci. s ecec ed orpdosthypothy- rapid-onset to leads deficiency 2 . o uhri n . h 6.8 and euthyroid for h 2.2 : < 3 NSLts Articles Latest PNAS 82 (2016). 38124 6, VO .0;****P 0.001; 4 c.Rep. Sci. 0m a,2m pro- mM 2 NaF, mM 10 , Circulation −80 39(2017). 5329 7, Lancet E9172–E9180 114, IAppendix SI < ◦ Endocrinology ni use. until C t .01 NS 0.0001. AE J. FASEB et one- test, 1550– 390, 126– 126, | f8 of 7 32, .

PHYSIOLOGY 23. M. Lopez et al., Hypothalamic AMPK and fatty acid metabolism mediate thyroid 39. M. Pant, N. C. Bal, M. Periasamy, Sarcolipin: A key thermogenic and metabolic regulation of energy balance. Nat. Med. 16, 1001–1008 (2010). regulator in skeletal muscle. Trends Endocrinol. Metab. 27, 881–892 (2016). 24. N. Martinez-Sanchez et al., Hypothalamic AMPK-ER stress-JNK1 axis mediates the cen- 40. E. S. Bachman et al., βAR signaling required for diet-induced thermogenesis and tral actions of thyroid hormones on energy balance. Cell Metab. 26, 212–229.e12 obesity resistance. Science 297, 843–845 (2002). (2017). 41. J. E. Silva, S. D. Bianco, Thyroid-adrenergic interactions: Physiological and clinical 25. K. Johann et al., Thyroid-hormone-induced browning of white adipose tissue implications. Thyroid 18, 157–165 (2008). does not contribute to thermogenesis and glucose consumption. Cell Rep. 27, 42. L. Landsberg, J. Axelrod, Influence of pituitary, thyroid, and adrenal hormones on 3385–3400.e3 (2019). norepinephrine turnover and metabolism in the rat heart. Circ. Res. 22, 559–571 26. C. Dittner, E. Lindsund, B. Cannon, J. Nedergaard, At thermoneutrality, acute (1968). thyroxine-induced thermogenesis and pyrexia are independent of UCP1. Mol. Metab. 43. T. Tu, C. W. Nash, The influence of prolonged hyper- and hypothyroid states on the 25, 20–34 (2019). noradrenaline content of rat tissues and on the accumulation and efflux rates of 27. A. Guilherme et al., Control of adipocyte thermogenesis and lipogenesis through β3- tritiated noradrenaline. Can. J. Physiol. Pharmacol. 53, 74–80 (1975). adrenergic and thyroid hormone signal integration. Cell Rep. 31, 107598 (2020). 44. P. Coulombe, J. H. Dussault, P. Walker, Plasma catecholamine concentrations in 28. National Research Council Subcommittee on Laboratory Animal Nutrition, Nutrient hyperthyroidism and hypothyroidism. Metabolism 25, 973–979 (1976). Requirements of Laboratory Animals (Nutrient Requirements of Domestic Animals, 45. L. Varela et al., Hypothalamic mTOR pathway mediates thyroid hormone-induced National Academy of Sciences, Washington, DC, ed. 4, 1995), pp. xii, 173. hyperphagia in hyperthyroidism. J. Pathol. 227, 209–222 (2012). 29. A. I. Mina et al., CalR: A web-based analysis tool for indirect calorimetry experiments. 46. A. Coppola et al., A central thermogenic-like mechanism in feeding regulation: An Cell Metab. 28, 656–666.e1 (2018). interplay between arcuate nucleus T3 and UCP2. Cell Metab. 5, 21–33 (2007). 30. J. R. Speakman, Measuring energy metabolism in the mouse-theoretical, practical, 47. G. A. Bray, Reciprocal relation of food intake and sympathetic activity: Experimental and analytical considerations. Front. Physiol. 4, 34 (2013). observations and clinical implications. Int. J. Obes. Relat. Metab. Disord. 24 (suppl. 2), 31. L. P. Klieverik et al., Thyroid hormone effects on whole-body energy homeostasis and S8–S17 (2000). tissue-specific fatty acid uptake in vivo. Endocrinology 150, 5639–5648 (2009). 48. I. S. Racotta, M. Cabanac, L. Lafrance, Dissociation of anorectic and affective 32. S. Enerback et al., Mice lacking mitochondrial uncoupling protein are cold-sensitive responses to epinephrine and glucose in rats. Physiol. Behav. 58, 125–130 (1995). but not obese. Nature 387, 90–94 (1997). 49. C. S. Mantzoros et al., Activation of β3 adrenergic receptors suppresses leptin expres- 33. B. Cannon, J. Nedergaard, Brown adipose tissue: Function and physiological sion and mediates a leptin-independent inhibition of food intake in mice. Diabetes significance. Physiol. Rev. 84, 277–359 (2004). 45, 909–914 (1996). 34. N. C. Bal, S. K. Sahoo, S. K. Maurya, M. Periasamy, The role of sarcolipin in muscle 50. R. Cao, A. Li, H. Y. Cho, B. Lee, K. Obrietan, Mammalian target of rapamycin signaling non-shivering thermogenesis. Front. Physiol. 9, 1217 (2018). modulates photic entrainment of the suprachiasmatic circadian clock. J. Neurosci. 30, 35. S. K. Maurya et al., Sarcolipin signaling promotes mitochondrial biogenesis and 6302–6314 (2010). oxidative metabolism in skeletal muscle. Cell Rep. 24, 2919–2931 (2018). 51. V. Golozoubova et al., Depressed thermogenesis but competent brown adipose tissue 36. G. J. Babu, P. Bhupathy, C. A. Carnes, G. E. Billman, M. Periasamy, Differential expres- recruitment in mice devoid of all hormone-binding thyroid hormone receptors. Mol. sion of sarcolipin protein during muscle development and cardiac pathophysiology. Endocrinol. 18, 384–401 (2004). J. Mol. Cell. Cardiol. 43, 215–222 (2007). 52. M. Lopez, C. V. Alvarez, R. Nogueiras, C. Dieguez, Energy balance regulation by 37. T. Endo, T. Kobayashi, Thyroid-stimulating hormone receptor in brown adipose tissue thyroid hormones at central level. Trends Mol. Med. 19, 418–427 (2013). is involved in the regulation of thermogenesis. Am. J. Physiol. Endocrinol. Metab. 295, 53. J. Mittag, More than fever—novel concepts in the regulation of body temperature by E514–E518 (2008). thyroid hormones. Exp. Clin. Endocrinol. Diabetes 128, 428–431 (2020). 38. R. C. Marians et al., Defining thyrotropin-dependent and -independent steps of thy- 54. S. Minamisawa et al., Post-transcriptional downregulation of sarcolipin mRNA roid hormone synthesis by using thyrotropin receptor-null mice. Proc. Natl. Acad. Sci. by triiodothyronine in the atrial myocardium. FEBS Lett. 580, 2247–2252 U.S.A. 99, 15776–15781 (2002). (2006).

8 of 8 | www.pnas.org/cgi/doi/10.1073/pnas.2008919117 Kaspari et al. Downloaded by guest on September 28, 2021