Telmisartan But Not Valsartan Increases CaloricExpenditure and Protects Against Weight Gain

and Hepatic SteatosisKen Sugimoto, Nathan R. Qi, Ludmila Kazdova, Michal Pravenec, Toshio Ogihara, Theodore W. Kurtz

AbstractThe potential effects of angiotensin II receptor blockers (ARBs) on adipose tissue biology and body weight areof considerable interest, because these agents are frequently used to treat hypertension in patients who are prone tovisceral obesity, the metabolic syndrome, and diabetes. In rats fed a high-fat, high-carbohydrate diet, we compared theeffects of 2 ARBs, telmisartan and valsartan, on body weight, food intake, energy expenditure, fat accumulation, fat cellsize, and hepatic triglyceride levels. Telmisartan, but not valsartan, promoted increases in caloric expenditure andprotected against dietary-induced weight gain. In the telmisartan-treated rats, absolute food intake, but not food intakeadjusted for body weight, was lower than in valsartan-treated rats or controls. Telmisartan reduced the accumulation ofvisceral fat and decreased adipocyte size to a much greater extent than valsartan and was also associated with asignificant reduction in hepatic triglyceride levels. Moreover, telmisartan, but not valsartan, increased the expression ofboth nuclear-encoded and mitochondrial-encoded genes in skeletal muscle known to play important roles inmitochondrial energy metabolism. Thus, in addition to a class effect of ARBs in modulating adipocyte size, thesefindings raise the possibility that certain molecules, like telmisartan, may have a particularly strong impact on fat cellvolume and fat accumulation, as well as distinctive effects on energy metabolism, that may help protect againstdietary-induced visceral obesity and weight gain. (Hypertension. 2006;47:1003-1009.)

Key Words: liver hypertension, obesity receptors, angiotensin II

In patients with the metabolic syndrome, diabetes, or both,increased visceral fat is believed to play an important rolein the pathogenesis of insulin resistance and dyslipidemia.1Although surgical removal of subcutaneous abdominal fatmay have little or no effect on glucose or lipid metabolism,2surgical removal of intra-abdominal visceral fat can improveinsulin sensitivity.3 Moreover, pharmacological treatmentsthat redistribute fat from visceral to subcutaneous depots canalso improve insulin resistance and dyslipidemia. For exam-ple, thiazolidinedione ligands of the peroxisome proliferatoractivated receptor (PPAR) are thought to improve carbo-hydrate and lipid metabolism at least in part by promoting thedifferentiation of small adipocytes that buffer against inap-propriate deposition of fat in muscle and visceral tissues andthat are more metabolically efficient than large, hypertro-phied adipocytes.46 Given that abdominal obesity and themetabolic syndrome are very common in patients with hyper-tension, the availability of antihypertensive agents that retard theaccumulation of visceral fat, promote the formation of small,metabolically active adipocytes, and attenuate weight gain couldbe of considerable clinical value.

Recently, we and others have discovered that telmisartan, anangiotensin II receptor blocker (ARB) approved for the treat-ment of hypertension, is also a partial agonist of PPAR.7,8Whereas full agonists of PPAR, such as rosiglitazone andpioglitazone, promote weight gain while altering fat distributionand adipocyte differentiation, partial agonists (mixed agonists/antagonists) of PPAR may have the capacity to retard weightgain while promoting adipocyte differentiation.7,9 For example,we have found that telmisartan can promote adipocyte differen-tiation but also attenuate weight gain while improving glucoseand lipid metabolism in rats fed a high-fat, high-carbohydratediet.7 Sharma et al10 have reported that blockade of the angio-tensin II type 1 receptor, per se, can promote adipocyte differ-entiation and have proposed that this may contribute to theantidiabetic effects of angiotensin II receptor antagonists. It isunknown whether bifunctional molecules like telmisartan thatboth activate PPAR and block the angiotensin II receptorexert different effects on adipocyte size and on the primarydeterminants of body weight than ordinary angiotensin recep-tor blockers, such as valsartan, that lack the capacity toactivate PPAR.

Received November 4, 2005; first decision November 27, 2005; revision accepted January 13, 2006.From the Department of Laboratory Medicine (K.S., N.R.Q., T.W.K.), University of California, San Francisco; Department of Geriatric Medicine (K.S.,

T.O.), Osaka University Graduate School of Medicine, Osaka, Japan; Institute for Clinical and Experimental Medicine (L.K.), Prague, Czech Republic;Institute of Physiology (M.P.), Czech Academy of Sciences, and the Center for Applied Genomics, Prague, Czech Republic.

The first 2 authors contributed equally to these studies.Correspondence to Theodore W. Kurtz, Department of Laboratory Medicine, University of California San Francisco Clinical Laboratories, 185 Berry

St, Ste 290, San Francisco, CA 94107. E-mail KurtzT@Labmed2.ucsf.edu 2006 American Heart Association, Inc.Hypertension is available at http://www.hypertensionaha.org DOI: 10.1161/01.HYP.0000215181.60228.f7

1003

In the current studies, we investigated the effects of telmisar-tan and valsartan on energy intake, energy expenditure, motoractivity, body weight, fat accumulation, and fat cell size in ratsfed a high-fat, high-carbohydrate diet. Telmisartan, but notvalsartan, attenuated weight gain and promoted increasesin energy expenditure in the absence of changes in motoractivity. Absolute food intake, but not food intake adjustedfor body weight, was lower in the telmisartan-treated rats.Moreover, previous studies in pair-fed animals have shownthat telmisartan can attenuate weight gain in the absence ofany effects on absolute food intake.7 Thus, the telmisartan-induced attenuation of weight gain may be related to anintrinsic increase in caloric expenditure and not only todecreases in appetite or food absorption or to increases inenergy expenditure related to changes in motor activity.Telmisartan reduced the accumulation of visceral and subcu-taneous fat and decreased adipocyte size to a much greaterextent than valsartan and also significantly reduced hepatictriglyceride levels. These findings suggest that the effects oftelmisartan on energy expenditure, adipocyte size, and vis-ceral fat stores may not only involve blockade of the type 1angiotensin II receptor but perhaps other mechanisms as welland should motivate future studies of its impact on molecularmechanisms regulating dietary-induced weight gain and fatmetabolism.

MethodsMale Sprague Dawley rats were obtained from Charles RiverLaboratories (Wilmington, MA) and fed tap water and a high-fat,high-carbohydrate diet (Teklad TD03293, containing 60% fructose,10% lard, and 0.06% magnesium) starting at 6 weeks of age for 3months. The rats were individually housed and randomized to 3groups (10 rats in each group) to receive either telmisartan indrinking water (9.7 mol/kg body weight [BW] per day), valsartanin drinking water (9.7 mol/kg BW per day), or drinking water alonethroughout the entire study. These doses are approximately equiva-lent to 5 mg/kg BW per day and are not extremely high relative todoses given in humans. Water intakes and body weights weremeasured daily. To ensure delivery of the correct drug doses, thedrug concentrations were adjusted in the drinking water each weekbased on the average water consumption and body weights in eachgroup. We measured daily food intake for each rat by subtracting theamount of food remaining in the cage each day from the measuredamount of food provided each day. The average daily food intake foreach rat was then calculated by averaging all of the daily intakemeasurements obtained over the entire course of the study. The ratswere housed in a constant air-conditioned environment with a 12 to12 hour day-night cycle. Room temperature in the housing room andduring the indirect calorimetry measurements was maintained be-tween 19 and 21C. Indirect calorimetry measurements of energyexpenditure were performed during the course of the study asdescribed below. After 12 weeks, rats were anesthetized usingpentobarbital, and fat and liver tissues were excised and weighed.Dissection of fat pads was performed by a single person who had noknowledge of the drug treatment groups.

In a separate experiment, groups (5 rats per group) of 10-weekold Sprague Dawley rats were fed the same diet and treated withtelmisartan, valsartan, or vehicle for 10 days via gavage at the samedoses as described above. Soleus muscle tissue was then collectedfor measurement of mitochondria-related gene expression levels asdescribed below.

Energy Expenditure and Motor ActivityOxygen consumption (VO2), carbon dioxide production (VCO2), andtotal motor activity were measured at 1 and 4 weeks in the nonfasting

state, at 7 weeks in the fasting state, and at 8 weeks after fasting andrefeeding using an indirect calorimeter equipped with an opticalbeam activity monitoring system (Oxymax Opto-M3, ColumbusInstruments). Rats were weighed each time before the measurementsand placed into the sealed chambers at 5:00 to 5:30 PM. Themeasurements were carried out continuously for 24 hours. Duringthis time, the rats were provided their respective water solutionscontaining the appropriate treatment drug and fed the high-fat,high-carbohydrate diet (except when tested under fasting condi-tions). The calorimeter was routinely calibrated each time before use,both the VO2 and VCO2 were sampled at 30 seconds for every 15minutes, and the motor activity was counted every second. The airflow rate through the chambers was fixed between 1.95 and 2.75L/min as needed to insure that the oxygen differential betweeninflow and outflow was 0.3% at resting conditions as recom-mended in the technical specifications of the equipment manufac-turer. Energy expenditure was calculated as 3.91 VO21.10VCO21.93 N (urinary nitrogen), where N is assumed negligible.11Respiratory quotient was calculated as a ratio of VCO2 to VO2. Theresults for both food intake and energy expenditure were scaledaccording to body weight or three-quarter power of body weight. Thegroup trends and findings were similar with or without three-quarterpower scaling, and, for the sake of simplicity, results are presented usingthe body weight correction without three-quarter power scaling.

Adipocyte Number and VolumeAdipocyte number and volume were determined based on themethod of Fine and DiGirolamo.12 A portion of adipose tissue (200to 300 mg) dissected from the distal end of the right epidydimal fatpad was minced and digested in Krebs Ringer bicarbonate buffer (pH7.4) that contains 4% serum albumin and 1 mg/mL of type Icollagenase (Sigma) at 37C for 30 to 45 minutes. The digestedtissue suspension was centrifuged and washed, and the adipocyteswere collected. Six aliquots of adipocytes evenly suspended in KrebsRinger bicarbonate 4% BSA without collagenase were aspirated into6 hematocrit capillary tubes, respectively, and centrifuged in ahematocrit centrifuge (Adams MHCT II). Lipocrit of each tube wascalculated by dividing the volume of packed adipocytes into the totalvolume of suspension buffer in the tube; lipocrit of each rat wastaken from the average readings of 6 tubes. The remaining adipo-cytes were stained in 1% methylene blue, and 5 aliquots of evenlysuspended cells were examined under a microscope equipped with astage micrometer. The mean diameter of adipocytes was calculatedfrom 600 cells using digital images taken at randomly chosenfields in which all of the cells were accounted for in the fieldmeasurement. Adipocyte number was determined as the adipocytevolume estimated from the lipocrit divided by the mean adipocytevolume calculated from the average diameter of adipocytes.

Mitochondria-Related Gene Expression LevelsGiven the central role of mitochondria in energy metabolism,real-time polymerase chain reaction (PCR) analysis was used toquantify expression levels of several key genes related to mitochon-dria function in muscle: (1) mitochondrial transcription factor A(TFAM), a nuclear-encoded transcription factor that regulates mito-chondrial DNA expression levels, mitochondria copy number, andmitochondria function13,14 and (2) cytochrome C oxidase subunit 1(MTCO1), a mitochondrial-encoded enzyme involved in oxidativephosphorylation. The primers for the MTCO1 gene were designed toavoid amplifying mitochondrial DNAlike sequences in genomicDNA as recommended by Garnier et al15 using an annealingtemperature of 60C. The upstream primer for MTCO1 gene expres-sion analysis was agc agg aat agt agg gac agc, and the downstreamprimer was tga gag aag tag tag gac ggc. For amplification of the geneencoding TFAM at an annealing temperature of 53C, the upstreamprimer sequence was caa ggg aaa ttg aag ctt gt, and the downstreamprimer sequence was cat ttg ctc ttc cca aga ct. Gene expression levelswere assessed in the 10-weekold rats that had been fed a high-fructose and high-fat diet and treated with telmisartan, valsartan, or

1004 Hypertension May 2006

vehicle for 10 days. Total RNA was extracted using Trizol reagent(Invitrogen), and cDNA was prepared and analyzed by real-timePCR testing using QuantiTect SYBR Green reagents (Qiagen, Inc)on an Opticon continuous fluorescence detector (MJ Research) asdescribed previously.16 Gene expression levels were normalizedrelative to the expression of -actin, which served as the internalcontrol, with results being determined in triplicate. The results in thetelmisartan-treated rats and valsartan-treated rats were expressed asfold increases relative to the mean normalized expression level of thecontrol rats that was arbitrarily defined as 1.

Biochemical Measurements in Serumand Liver TissueSerum samples for glucose, insulin, triglycerides, leptin, and adi-ponectin were obtained in the nonfasting state and measured asdescribed previously.16 Frozen liver tissues were powdered underliquid N2 and extracted in chloroform-methanol (2:1 v/v). Afteradding 2% KH2PO4, the solution was centrifuged, and the organicphase was removed and evaporated to dryness under N2. The residuewas then dissolved in isopropyl alcohol, and triglyceride content wasdetermined by enzymatic assay (Pliva-Lachema).

Statistical AnalysisData are expressed as mean SEM. Statistical analysis of metabolicdata were performed using ANOVA and the StudentNewmanKeuls test for multiple group comparisons or with equivalentnonparametric tests as required. Statistical analysis of the geneexpression data were performed using the REST XL program, whichtests for significance by a randomization procedure.17

ResultsEffects on Body Weight Gain and OrganFat AccumulationAll of the rats gained weight during the course of the study.However, the weight gain in rats treated with telmisartan wassignificantly less than in rats treated with valsartan or incontrols. Although absolute food intake was lower in thetelmisartan-treated rats, food intake adjusted for body weightwas similar in all 3 of the groups (Figure 1). Telmisartan-treated rats and, to lesser extent, valsartan-treated rats accu-mulated significantly less inguinal subcutaneous fat andepididymal visceral fat compared with control, untreated rats(Figure 2). Serum leptin levels reflected the degree of fataccumulation and were reduced to a greater extent in thetelmisartan-treated rats than in the valsartan-treated rats (Figure2). Despite significantly lower fat mass in the telmisartan-treatedrats, no differences were detected in serum adiponectin levelsamong the 3 groups (data not shown). Serum glucose levelswere lower in the telmisartan-treated rats, 2008 mg/dL, com-pared with the valsartan-treated rats, 23614 mg/dL (P0.05),and control rats, 23811 mg/dL (P0.05). Insulin levels in thetelmisartan-treated rats, 7.91 ng/mL, were lower than in thecontrols, 12.81.5 ng/mL (P0.05) and also tended to belower than in the valsartan-treated rats, 10.61 ng/mL,although the telmisartan versus valsartan comparison did not

Figure 1. BW and food intake in ratstreated with telmisartan, valsartan, or inuntreated controls. (A) BW. (B) Averagedaily food intake throughout the studyexpressed as g/100 g BW each day.Adjustment of food intake according tothree-quarter power of BW yielded simi-lar results in the group comparisons.Solid symbols, telmisartan group; graysymbols, valsartan group; open symbols,controls. *ANOVA, P0.001 and Stu-dentNewmanKeuls, P0.05 vs valsar-tan group and control group.

Figure 2. Fat mass and serum leptin lev-els in rats treated with telmisartan, val-sartan, or in untreated controls. (A) Sub-cutaneous inguinal fat. (B) Visceralepididymal fat. (C) Serum leptin levels.*ANOVA, P0.001 and StudentNew-manKeuls, P0.05 vs valsartan groupand control group; **StudentNewmanKeuls, P0.05 vs control group.

Sugimoto et al Telmisartan and Body Weight 1005

achieve statistical significance (P0.07 in single t test).Triglyceride levels in the telmisartan-treated rats, 34055mg/dL, and the valsartan-treated rats, 37051 mg/dL, ap-peared to trend lower than in controls, 513142 mg/dL, butthe differences did not achieve statistical significance.

In telmisartan-treated rats but not in valsartan-treated rats,liver weight and hepatic triglyceride levels were significantlydecreased compared with control rats (Figure 3). Effects onheart weight, kidney weight, and weight of soleus andextensor digitorum longus muscles are presented in Table 1.In the telmisartan-treated rats, heart weights were slightlydecreased, whereas kidney weights and muscle weights wereincreased compared with the valsartan-treated rats and con-trols. Thus, the telmisartan-induced reductions in liver weightand fat weight are not simply because of a nonspecificimpairment in organ growth, because other organ weights(eg, kidney and skeletal muscle) were actually somewhatgreater in the telmisartan-treated rats.

Effects on Adipocyte Size and NumberAlthough epididymal fat pad weight was lower in telmisartan-treated rats compared with the valsartan-treated rats and con-trols, the number of adipocytes per gram of fat tissue wassignificantly greater in the telmisartan-treated rats comparedwith the other groups (Figure 4). The average adipocytevolume was markedly reduced in rats treated with telmisartancompared with controls (Figure 4). Adipocyte volume invalsartan-treated rats was also reduced compared with con-trols (Figure 4). However, the reduction in adipocyte volume inrats treated with telmisartan was greater than in rats treated with

valsartan (62.0% versus 20.7%). Thus, telmisartan appeared topromote the formation of small adipocytes while attenuatingadipocyte hypertrophy, fat mass accumulation, and weightgain. Valsartan had less pronounced effects on adipocytevolume and fat mass accumulation and did not attenuateweight gain.

Effects on Total Body Energy ExpenditureWe used indirect calorimetry to investigate whether telmis-artan might be attenuating weight gain by increasing energyexpenditure. Under all of the testing conditions, diurnalpatterns of energy expenditure and total activity showedhigher levels during the nighttime, when the rats were awakeand active, and lower levels during the daytime, when the ratswere at rest. Rats treated with telmisartan, but not valsartan,showed significant increases in total body energy expenditureboth during the nighttime and daytime compared with con-trols under all of the dietary conditions. For example, Fig-ure 5 shows the pattern of increased energy expenditure intelmisartan-treated rats compared with valsartan-treated ratsand controls under nonfasting conditions at week 4 of thestudy. Table 2 shows increased 24-hour energy expenditureresults in the telmisartan-treated rats during other feedingconditions and at different time points in the study. Increasedenergy expenditure was observed in the telmisartan-treatedrats regardless of whether the results were scaled according tobody weight or three-quarter power of body weight. Totalactivity levels were similar among the 3 groups, suggestingthat the telmisartan-induced increases in energy expenditureinvolve more than just increases in energy metabolism second-

Figure 3. Liver weight and liver triglyc-erides in rats treated with telmisartan,valsartan, or in untreated controls. (A)Liver weight. (B) Liver triglycerides.*ANOVA, P0.001 and StudentNew-manKeuls, P0.05 vs valsartan groupand control group; **ANOVA, P0.01and StudentNewmanKeuls, P0.05vs control group.

TABLE 1. Additional Organ Weights (g/100 g BW)

Group Heart Left Kidney Right Kidney Soleus Muscle EDL Muscle

Telmisartan 0.250.004* 0.390.012* 0.400.014* 0.0230.001* 0.0250.001*

Valsartan 0.260.008 0.340.011 0.350.012 0.0200.001 0.0200.001

Control 0.270.005 0.330.008 0.350.010 0.0210.001 0.0200.001

EDL indicates extensor digitorum longus.*Significantly different from valsartan and control groups, P0.05.

1006 Hypertension May 2006

ary to increased physical activity (Figure 5). Respiratory quo-tients were similar among the 3 experimental groups (data notshown) suggesting that the telmisartan-induced increases inenergy expenditure may be related to increases in both fatty acidand carbohydrate metabolism.

Effects of Mitochondrial Gene Expression LevelsTelmisartan, but not valsartan, was associated with significantincreases in the expression of genes encoding TFAM andMTCO1 in soleus muscle (Figure 6). Thus, telmisartan affectedthe expression of both nuclear and mitochondrial genes knownto play important roles in mitochondrial function.

DiscussionIn the current study, we have found that telmisartan, but notvalsartan, significantly attenuates weight gain in rats fed ahigh-fat, high-carbohydrate diet. Food intake adjusted for bodyweight was similar in all of the groups, although absolute foodintake unadjusted for body weight was lower in the telmisartan-treated rats. Previous studies in rats and mice fed high-fat dietshave shown that telmisartan may attenuate weight gain in theabsence of detectable changes in absolute food intake includingwhen animals have been pair fed to insure identical food

consumption.7,18 These observations suggest that increases inenergy expenditure, and not simply reductions in appetite, arecontributing to the ability of telmisartan to attenuate weight gain.Of course, because of the difficulties in measuring food intakewith 100% accuracy, the possibility that group differences inenergy intake might be contributing to the group differences inbody weight cannot be completely ruled out. Measurements ofcaloric expenditure using indirect calorimetry indicated thatadministration of telmisartan, but not valsartan, was associatedwith increases in energy expenditure. In addition, telmisartanwas observed to induce increases in energy expenditure andattenuate weight gain in the absence of detectable increases intotal motor activity. These observations indicate that the in-creases in energy expenditure induced by telmisartan mayinvolve more than just activity-induced increases in metabolicrate. Telmisartans weak activity on PPAR might also meritfurther metabolic investigation.7

Given that telmisartan, but not valsartan, attenuated weightgain and increased energy expenditure, the current findingsraise the possibility that the effects of telmisartan on energymetabolism may go beyond just blockade of the type 1angiotensin II receptor. Telmisartan and valsartan have sim-

Figure 4. Adipocyte volume and numberin rats treated with telmisartan, valsartan,or in untreated controls. (A) Adipocytevolume. (B) Adipocyte number. *ANOVA,P0.001 and StudentNewmanKeuls,P0.05 vs control group and valsartangroup. **P0.05 vs control group.

Figure 5. Twenty-four hour energyexpenditure and activity levels in non-fasting rats treated with telmisartan orvalsartan for 4 weeks and in untreatedcontrols. (A) Energy expenditure adjustedper 100 g BW. Adjustment of energyexpenditure according to three-quarterpower of BW yielded similar results inthe group comparisons. (B) Total activity.*ANOVA, P0.01 and StudentNewmanKeuls, P0.05 vs valsartan group andcontrol group.

Sugimoto et al Telmisartan and Body Weight 1007

ilar affinity for the type 1 angiotensin II receptor, andequimolar doses of the drugs were administered in thesestudies. However, telmisartan has a longer half-life thanvalsartan, and other pharmacological and pharmacokineticdifferences exist between these molecules as well. Thus, it ispossible that the greater metabolic impact of telmisartancompared with valsartan might also be related to greater invivo blockade of the type 1 angiotensin II receptor bytelmisartan than valsartan at the doses we administered.

In rats treated with either telmisartan or valsartan and fedthe high-fat, high-carbohydrate diet, adipocyte size and fataccumulation were reduced compared with untreated controlrats. These observations are consistent with the hypothesis ofSharma et al10 that angiotensin II receptor blockade per semay affect adipocyte biology and promote the formation ofsmall, metabolically efficient adipocytes. In the currentstudy, however, the effects of telmisartan on adipocyte sizeand visceral fat stores were much greater than those ofvalsartan. In previous studies, telmisartan was shown tohave greater effects than valsartan on adipocyte differen-tiation and on adipocyte glucose transport, although bothdrugs were tested at concentrations clearly exceeding thoserequired to block angiotensin II receptors.19 These findingsraise the possibility that, in addition to a class effect ofARBs in modulating adipocyte size, certain molecules, liketelmisartan, may have particularly robust effects on fat cellsize and metabolism that involve more than just angioten-sin II receptor blockade.

Recently, we and others have observed that the ARBtelmisartan is a partial agonist (mixed agonist/antagonist) of

PPAR, an intracellular receptor that is a major regulator offat cell differentiation.7,8 The other clinically approved ARBsappear to have relatively little or no effect on PPAR activity,with the exception of irbesartan and a metabolite of losartan,both of which are much less potent activators of PPAR thantelmisartan.7,8,20 In contrast to full agonists of PPAR, such asrosiglitazone or pioglitazone, which promote weight gain,recent studies have suggested that partial agonists (mixedagonists/antagonists) of PPAR may have the capacity toimprove glucose and lipid metabolism while attenuatingweight gain.9,18 Whereas full activation of PPAR maypromote adipocyte hypertrophy and weight gain, as well asadipocyte differentiation, partial activation of PPAR mayprotect against dietary-induced adipocyte hypertrophy andweight gain while promoting formation of small, metabol-ically efficient adipocytes.9,21 This raises the possibilitythatthe effects of telmisartan on weight gain and adipocytesize might be related, at least in part, to its ability to functionas a partial agonist of PPAR.

In the current studies, we also found that telmisartan, butnot valsartan, increases the expression of genes for both anuclear-encoded transcription factor (TFAM) that regu-lates mitochondrial function and for a mitochondrial-encoded protein (MTCO1) involved in oxidative phosphor-ylation. In comparison to conventional full agonists ofPPAR, such as the thiazolidinediones, partial agonists ofPPAR, like telmisartan, may have the ability to prefer-entially recruit certain transcriptional coactivators that areparticularly important in regulating genes that controlmitochondrial function and energy metabolism.18,22 Forexample, partial agonists appear to preferentially recruitPPAR coactivator 1-, a transcriptional coactivator knownto stimulate expression of TFAM, which, in turn, canincrease mitochondrial gene expression (eg, MTCO1) and,ultimately, mitochondrial biogenesis.22,23 Although theprecise cellular and molecular mechanisms that mediatethe robust effects of telmisartan on body weight, energyexpenditure, and fat metabolism remain to be determined,studies on PPAR coactivator recruitment and target gene

Figure 6. Soleus muscle expression ofthe genes encoding TFAM and MTCO1.(A) Real-time PCR analysis showingfold expression of TFAM in rats treatedwith telmisartan or valsartan for 10days relative to untreated controls. (B)Fold expression of MTCO1 in ratstreated with telmisartan or valsartan for10 days relative to untreated controls.*P0.05 vs control group and valsartangroup by the randomization test ofPfaffl et al.17

TABLE 2. Twenty-Four Hour Energy Expenditure (Kcal/kg BW/h)

GroupRegular Feeding

(Week 1)Fasting(Week 7)

Refeeding(Week 8)

Telmisartan 11.360.20* 7.350.21* 8.060.19*

Valsartan 10.340.18 6.530.24 7.230.22

Control 10.070.17 6.260.14 6.830.09

*Significantly different from valsartan and control groups, P0.05.

1008 Hypertension May 2006

expression, as well as on mitochondrial number, structure,and function, could represent potentially fruitful areas ofinvestigation in the future.

In addition to attenuating weight gain, telmisartan signifi-cantly attenuated the development of hepatic steatosis andhepatomegaly, whereas valsartan did not. The hepatic effectsof telmisartan could not be attributed to nonspecific effects onorgan weight, because telmisartan did not attenuate thegrowth of kidney or skeletal muscle. Additional studies willbe required to assess the relative contributions of hepatic andextrahepatic mechanisms that could mediate the ability oftelmisartan to attenuate dietary-induced accumulation of fatin liver. Given the frequent occurrence of fatty liver inpatients with the metabolic syndrome, the current findingsmay point to new opportunities for limiting the hepaticaccumulation of fat in hypertensive patients at increased riskfor developing hepatic steatosis.24

PerspectivesThe potential effects of ARBs on adipose tissue biology areof considerable interest, because these agents are frequentlyused to treat hypertension in patients who are prone tovisceral obesity, the metabolic syndrome, and diabetes. Asproposed by Sharma et al,10 the ability to promote thedifferentiation of small, metabolically efficient adipocytesmay represent a class effect of ARBs that could contribute tothe lower rates of new onset diabetes in patients treated withthese agents compared with other antihypertensive drugs. Thecurrent findings suggest that a subset of ARBs, like telmis-artan, may have particularly strong effects on adipocyte sizethat might go beyond just angiotensin II receptor blockadeand may also have a special capacity to increase energyexpenditure and protect against dietary-induced increases inweight gain and visceral fat accumulation. Thus, these find-ings should motivate future studies on the ability of mole-cules like telmisartan to attenuate weight gain and reduce therisk for visceral obesity in patients with hypertension.

AcknowledgmentsWe acknowledge the support by National Institutes of Health grantsHL35018, HL63709, and TW01236 (to T.W.K.); grant 1M6837805002from the Ministry of Education of the Czech Republic (to M.P.); grantAV0Z 50110509 to the Institute of Physiology; and grant 8495-3 fromthe Ministry of Health of the Czech Republic (to L.K). M.P. is aninternational research scholar of the Howard Hughes Medical Institute.

References1. Pi-Sunyer FX. The epidemiology of central fat distribution in relation to

disease. Nutr Rev. 2004;62:S120S126.2. Klein S, Fontana L, Young VL, Coggan AR, Kilo C, Patterson BW,

Mohammed BS. Absence of an effect of liposuction on insulin action andrisk factors for coronary heart disease. N Engl J Med. 2004;350:25492557.

3. Gabriely I, Barzilai N. Surgical removal of visceral adipose tissue: effectson insulin action. Curr Diab Rep. 2003;3:201206.

4. Kawai T, Takei I, Oguma Y, Ohashi N, Tokui M, Oguchi S, Katsukawa F,Hirose H, Shimada A, Watanabe K, Saruta T. Effects of troglitazone on fatdistribution in the treatment of male type 2 diabetes. Metabolism: Clin. Exp.1999;48:11021107.

5. Rasouli N, Raue U, Miles LM, Lu T, Di Gregorio GB, Elbein SC, KernPA. Pioglitazone improves insulin sensitivity through reduction in musclelipid and redistribution of lipid into adipose tissue. Am J Physiol Endo-crinol Metab. 2005;288:E930E934.

6. Okuno A, Tamemoto H, Tobe K, Ueki K, Mori Y, Iwamoto K, UmesonoK, Akanuma Y, Fujiwara T, Horikoshi H, Yazaki Y, Kadowaki T.Troglitazone increases the number of small adipocytes without the changeof white adipose tissue mass in obese Zucker rats. J Clin Invest. 1998;101:13541361.

7. Benson SC, Pershadsingh HA, Ho CI, Chittiboyina A, Desai P, PravenecM, Qi N, Wang J, Avery MA, Kurtz TW. Identification of telmisartan asa unique angiotensin II receptor antagonist with selective PPAR-modulating activity. Hypertension. 2004;43:9931002.

8. Schupp M, Janke J, Clasen R, Unger T, Kintscher U. Angiotensin type 1receptor blockers induce peroxisome proliferator-activated receptor-activity. Circulation. 2004;109:20542057.

9. Berger JP, Petro AE, Macnaul KL, Kelly LJ, Zhang BB, Richards K,Elbrecht A, Johnson BA, Zhou G, Doebber TW, Biswas C, Parikh M,Sharma N, Tanen MR, Thompson GM, Ventre J, Adams AD, Mosley R,Surwit RS, Moller DE. Distinct properties and advantages of a novelperoxisome proliferator- activated protein {} selective modulator. MolEndocrinol. 2003;17:662676.

10. Sharma AM, Janke J, Gorzelniak K, Engeli S, Luft FC. Angiotensinblockade prevents type 2 diabetes by formation of fat cells. Hypertension.2002;40:609611.

11. Simonson DC, DeFronzo RA. Indirect calorimetry: methodological andinterpretative problems. Am J Physiol. 1990;258:E399E412.

12. Fine JB, DiGirolamo M. A simple method to predict cellular density inadipocyte metabolic incubations. Int J Obes Relat Metab Disord. 1997;21:764768.

13. Kang D, Hamasaki N. Mitochondrial transcription factor a in the main-tenance of mitochondrial DNA: overview of its multiple roles. Ann N YAcad Sci. 2005;1042:101108.

14. Ekstrand MI, Falkenberg M, Rantanen A, Park CB, Gaspari M, HultenbyK, Rustin P, Gustafsson CM, Larsson NG. Mitochondrial transcriptionfactor A regulates mtDNA copy number in mammals. Hum Mol Genet.2004;13:935944.

15. Garnier A, Fortin D, Delomenie C, Momken I, Veksler V, Ventura-Clapier R. Depressed mitochondrial transcription factors and oxidativecapacity in rat failing cardiac and skeletal muscles. J Physiol. 2003;551:491501.

16. Qi NR, Wang J, Zidek V, Landa V, Mlejnek P, Kazdova L, Pravenec M,Kurtz TW. A new transgenic rat model of hepatic steatosis and themetabolic syndrome. Hypertension. 2005;45:10041011.

17. Pfaffl MW, Horgan GW, Dempfle L. Relative expression softwaretool (REST) for group-wise comparison and statistical analysis ofrelative expression results in real-time PCR. Nucleic Acids Res. 2002;30:e36.

18. Schupp M, Clemenz M, Gineste R, Witt H, Janke J, Helleboid S,Hennuyer N, Ruiz P, Unger T, Staels B, Kintscher U. Molecular charac-terization of new selective peroxisome proliferator-activated receptor {}modulators with angiotensin receptor blocking activity. Diabetes. 2005;54:34423452.

19. Fujimoto M, Masuzaki H, Tanaka T, Yasue S, Tomita T, Okazawa K,Fujikura J, Chusho H, Ebihara K, Hayashi T, Hosoda K, Nakao K. Anangiotensin II AT1 receptor antagonist, telmisartan augments glucoseuptake and GLUT4 protein expression in 3T3L1 adipocytes. FEBS Lett.2004;576:492497.

20. Krempl T, Schwedhelm E, Hornung D, Schmid B, Schieffer B, Boeger R.Identification of EXP 3179 as selective PPAR receptor agonist.Zeitschrift fur Kardiologie. 2005;94:V62.

21. Kubota N, Terauchi Y, Miki H, Tamemoto H, Yamauchi T, Komeda K,Satoh S, Nakano R, Ishii C, Sugiyama T, Eto K, Tsubamoto Y, Okuno A,Murakami K, Sekihara H, Hasegawa G, Naito M, Toyoshima Y, TanakaS, Shiota K, Kitamura T, Fujita T, Ezaki O, Aizawa S, Kadowaki T.PPAR mediates high-fat diet-induced adipocyte hypertrophy andinsulin resistance. Mol Cell. 1999;4:597609.

22. Burgermeister E, Schnoebelen A, Flament A, Benz J, Blum D, Kuhn B,Marki HP, Mizrahi J, Niesor E, Meyer M. A common theme for partialperoxisome proliferator-activated receptor- agonists: Recruitment ofPPAR- coactivator-1 , a central regulator of energy expenditure. 3rdInternational Symposium on PPARs. 2005:19 (abstract).

23. Puigserver P, Spiegelman BM. Peroxisome proliferator-activatedreceptor- coactivator 1 alpha (PGC-1 ): transcriptional coactivator andmetabolic regulator. Endocr Rev. 2003;24:7890.

24. Donati G, Stagni B, Piscaglia F, Venturoli N, Morselli-Labate AM,Rasciti L, Bolondi L. Increased prevalence of fatty liver in arterialhypertensive patients with normal liver enzymes: role of insulinresistance. Gut. 2004;53:10201023.

Sugimoto et al Telmisartan and Body Weight 1009