Eur J Appl Physiol

DOI 10.1007/s00421-009-1250-z

ORIGINAL ARTICLE

Resistance training improves femoral artery endothelial dysfunction in aged rats

M. Brennan Harris · Kristen N. Slack · David T. Prestosa · David J. Hryvniak

Accepted: 8 October 2009© Springer-Verlag 2009

Abstract Although endurance exercise improves age-associated endothelial dysfunction, few studies have examinedthe eVects of resistance training and the potential molecularmechanisms involved in altering vascular reactivity with age.Young (9 months) and aged (20 months) male, Fisher 344rats were divided into four groups: Young Sedentary (YS,n = 14), Young Trained (YT, n = 10), Aged Sedentary (AS,n = 12), and Aged Trained (AT, n = 10). Resistance trainingconsisted of climbing a 1 m wire ladder, at an 85° angle,3 days/week for 6 weeks with increasing weight added to thetail. Endothelial function in femoral arteries was determinedby constructing acetylcholine dose–response curves on awire myograph. Femoral artery phospho-Ser1179-eNOS,eNOS and Hsp90 expression were evaluated by Westernblot. Acetylcholine-induced vasorelaxation was signiWcantly(P < 0.05) impaired in AS compared to YS and YT but notAT compared to YS and YT. Phospho-Ser1179-eNOS andeNOS were elevated (P < 0.05) in aged animals but notchanged with resistance training. Resistance trainingincreased Hsp90 levels in both young and old animals.Therefore, resistance training improves age-associated endo-thelial dysfunction in femoral arteries without changes ineNOS phosphorylation and expression. Increased Hsp90expression, a regulator of eNOS activity and coupling, sug-gests a potential mechanism for this improvement.

Keywords Exercise · Hsp90 · eNOS · Nitric oxide

Introduction

Increasing age is associated with a reduction in endothe-lium-dependent vasorelaxation in both humans and animalmodels (Matz et al. 2000; Muller-Delp 2006). However, ithas also been shown that age-related endothelial dysfunc-tion can be reversed by endurance exercise training (Spieret al. 2004). Although aging is associated with an increasein cardiovascular disease risk, aging is also associated witha decline in quality of life (QOL) which is largely due tolosses in muscular mass and strength (Kell et al. 2001). Ithas recently been suggested that resistance exercise trainingmay be a desirable alternative for exercise in the aging pop-ulation due to the numerous beneWcial health eVects thatcannot be achieved through endurance exercise trainingsuch as increases in strength and power (Phillips 2007).However, few studies have examined the eVects of resis-tance exercise training on vascular function.

In studies of normal human subjects using handgriptraining, one study found an improvement in endothelial-mediated vasodilation (Bank et al. 1998) while two othersdid not (Franke et al. 1998; Green et al. 1994). In addition,Miyachi et al. (2004) reported that whole body resistanceexercise in young men increased carotid arterial stiVness,but in a later study they reported no change in endothelialfunction (Kawano et al. 2008). In cardiovascular disease(CVD) models where endothelial dysfunction is evident theresults have been more consistent. Three studies usinghandgrip training reported improvements in endothelium-dependent dilation in chronic heart failure patients (Hambrechtet al. 2003; Hornig et al. 1996; Katz et al. 1997) andonly one did not (Bank et al. 1998). It is interesting to notethat in the studies cited above that observed improvedendothelial function with resistance training the subjectsmean age was over 40 years old, whereas the studies that

Communicated by Dag Linnarsson.

M. B. Harris (&) · K. N. Slack · D. T. Prestosa · D. J. HryvniakMolecular and Cardiovascular Physiology Laboratory, Department of Kinesiology & Health Sciences, The College of William & Mary, Williamsburg, VA 23187-8795, USAe-mail: mbharr@wm.edu

123

Eur J Appl Physiol

observed no change or a decline in endothelial functionall had subjects with a mean age less than 40 years old.Moreover, examination of the molecular mechanisms ofimproved endothelial function in these human studies havebeen limited to pharmacological approaches due to theinvasive techniques that must be employed for biochemicaltissue analysis.

To date, there appears to be only one animal study whichhas examined the eVects of resistance exercise training onendothelium-dependent vascular function. Figard et al.(2006) reported an improvement in endothelial dysfunctionfollowing isometric strength training in ovariectomizedrats. Using pharmacological techniques, these authors sug-gested a number of potential factors to account for thisimprovement. However, no measures of changes in vas-oregulatory proteins were reported. Furthermore, the use ofisometric strength training induces hemodynamic responsesthat are quite diVerent from the more traditional dynamicstrength training programs used in humans.

One of the primary molecular mechanisms believed toplay a role in endurance exercise-induced improvement invascular function is an improvement in the production orbioavailability of nitric oxide (NO). Endurance exercisetraining in both human and animal models has been shownto increase the expression of endothelial nitric oxide syn-thase (eNOS) (Green et al. 2004; Hambrecht et al. 2003;Spier et al. 2004), the primary source of vascular endothe-lial NO production. However, despite an improvement inendothelial function following training, some studies haveobserved no increase in eNOS expression. This discrepancyappears to be dependent, in part, on the type of trainingemployed and the vascular bed examined (Muller-Delp2006) leading to the conclusion that eNOS activity is regu-lated by other mechanisms. Because of the important roleof NO in the vascular system, eNOS is tightly regulated bya number of factors including phosphorylation and protein–protein interactions (Sessa 2004). For instance, Hambrechtet al. (2003) demonstrated that endurance exerciseincreases both eNOS expression and increased eNOS phos-phorylation at Ser-1177 in vascular tissues. Phosphoryla-tion at this site increases eNOS activity and may also play arole in increasing the bioavailability of NO (Fulton et al.1999). In addition, eNOS is also regulated by interactionwith heat shock protein 90 (Hsp90) (Sessa 2004). Hsp90association with eNOS is believed to play a role in deter-mining the balance of superoxide generation and NOproduction from eNOS (Pritchard et al. 2001). We havepreviously shown that endurance exercise increases NOSactivity and Hsp90/NOS activity in skeletal muscles (Harriset al. 2008). However, it appears that no studies have exam-ined the eVects of resistance exercise training on theexpression of eNOS and Hsp90 in blood vessels or changesin eNOS phosphorylation. In light of these observations, we

hypothesized that resistance exercise training would resultin improved endothelial dependent dilation in aged animalsexhibiting endothelial dysfunction and that this improve-ment may be related to increased eNOS expression andphosphorylation at Ser-1179 and/or increased Hsp90expression in rat femoral arteries.

Methods

Animals

Experiments were approved by the Institutional AnimalCare and Use Committee of The College of William &Mary and adhered to the ACSM standards of humaneanimal experimentation. Young (9 months) and aged(20 months) male Fisher 344 rats were divided into fourgroups: Young Sedentary (YS, n = 14), Young Trained(YT, n = 10), Aged Sedentary (AS, n = 12), and AgedTrained (AT, n = 10).

Resistance training

Animals were trained using a protocol previously described(Lee et al. 2004). BrieXy, animals climbed a 1 m ladderwith 2-cm steps, inclined at 85° 3 days/week for 6 weeks. Acylinder containing weights was attached to the base of thetail and resistance was increased by adding weights to thecylinder. Rats did two climbs (two repetitions) of the ladderat each workload; 50, 75, and 100% of the maximal loadfrom their previous exercise session. In addition, if the ratcompleted both climbs at 100% of the previous maximalload they attempted an additional one or two climbs with anincreased weight (+30 g) resulting in a failed climb (total6–8 climbs/day). The initial load consisted of 50% of bodyweight. Each climb was separated by a 2 min rest interval.When necessary, a spray of cool water was used to moti-vate the animals.

Vascular reactivity

Twenty-four hours following the last exercise animals wereanesthetized with sodium pentobarbital (40 mg/kg, i.p.) andfemoral arteries were removed and placed in ice-cold salinesolution. Vessels were cleaned of adherent fat and connec-tive tissue and cut into two, 2 mm rings. The remainder ofthe femoral artery was frozen in liquid nitrogen and usedfor immunoblotting. The rings were mounted using smallwires on stainless steel holders in muscle baths on aDMT610 myograph for isometric force recording. Musclebaths were Wlled with physiological salt solution (PSS) con-sisting of: 130 mM NaCl, 4.7 mM KCl, 1.18 mM KH2PO4,1.17 mM MgSO4·7H2O, 1.6 CaCl2, 14.9 mM NaHCO3,

123

Eur J Appl Physiol

5.5 mM dextrose, and 0.03 mM Na2EDTA heated to 37°Cand aerated with 95% O2/5% CO2. The rings were stretchedto an optimal resting tension (1.5 g) and allowed to stabilizefor 1 h. After equilibration, constrictor responses to phenyl-ephrine were determined using cumulative doses (10¡7 to10¡4 M). Relaxation responses to cumulative doses of ace-tylcholine (Ach, 10¡7 to 10¡4 M) and sodium nitroprusside(SNP, 10¡9 to 3 £ 10¡4 M) were determined following pre-constriction with 10¡4 M phenylephrine with »30 min ofrecovery between each drug until the resting tension stabi-lized. The bathing medium was changed every 5 min duringthe recovery periods.

Immunoblotting

A portion of the femoral artery removed for vascular reac-tivity was snap frozen in liquid nitrogen. Tissues were thenhomogenized in lysis buVer (50 mM Tris–HCl, 0.1 mMEDTA, 0.1 mM EGTA, 1% TX-100, 1% sodium deoxycho-late, 1 mM PMSF, 1 mM Na3VO4, and protease inhibitors)using a glass–glass homogenizer. Samples were centrifugedat 10,000g for 10 min at 4°C and divided for analysis.

Protein concentrations of the homogenized femoral arterysamples were determined using the Bio-Rad detergent-com-patible protein assay. Volumes were adjusted to equalize theconcentration of the samples. 100 �g of each sample werethen loaded and run on a 10% polyacrylamide gel and trans-ferred to a nitrocellulose membrane, which was subse-quently immunoblotted with either anti-eNOS (610297, BDTransduction Laboratories), anti-phospho-Ser-1179 eNOS(612393, BD Transduction Laboratories) or anti-Hsp90 anti-bodies (610419, BD Transduction Laboratories).

Statistical analysis

Relaxation responses in femoral artery rings were calcu-lated as percent relaxation from 10¡7 M phenylephrine con-traction. EC50 was determined using a non-linear regression(P < 0.05). Maximal acetylcholine relaxation responseswere analyzed using two-way ANOVA and Bonferonnipost hoc test (P < 0.05). Animals whose femoral arterieswere non-responsive to acetylcholine-induced relaxationwhere eliminated from the analysis as the endothelium wasdamaged during the surgery.

Results from immunoblotting were determined by densi-tometric analysis of radiography Wlms using a BioRadGelDocXR. Densities for each sample were normalizedto a percentage of either YS or YT and analyzed using aTwo-way ANOVA with Bonferroni post-tests (P < 0.05).

Results

Resistance training

Six weeks of resistance exercise training resulted in a sig-niWcant increase (P < 0.05) in the amount of weight liftedby both young and aged animals from 50% of body weightpre-training to 264 § 7% (Young) and 169 § 4% (Aged) ofbody weight at the end of the training period. In addition,the Wnal amount of weight lifted expressed as a percent ofbody weight was signiWcantly diVerent (P < 0.05) betweenyoung and aged animals.

Six weeks of resistance exercise training resulted inhypertrophy of the mixed Wber type plantaris muscle asindicated by a signiWcant increase (P < 0.05) in the ratio ofplantaris weight to body weight between sedentary andtrained animals (Table 1). However, in the primarily slow-twitch soleus muscle no signiWcant diVerence (P > 0.05) inmuscle to body weight ratio was observed (Table 1).

Femoral artery vasoreactivity

Cumulative doses of phenylephrine produced increasedcontractile responses in all groups (YS, YT, AS, and AT;Fig. 1). Vasoconstrictor responses were signiWcantly diVer-ent (P < 0.05) between young and aged animals at the high-est doses (10¡4 to 10¡3), but no diVerence occurred as aresult of training. No signiWcant diVerences in the EC50 forphenylephrine was observed between the groups.

Cumulative doses of ACh resulted in signiWcant relaxa-tion in all groups (YS, YT, AS, and AT) and no diVerencewas observed in the initial tension development in the fem-oral rings in response to the bolus dose of 10¡4 PE (EC80)used for preconstriction (Fig. 2). ACh-induced vasodilationwas signiWcantly reduced (P < 0.05) in AS femoral ringsbetween 10¡6 and 10¡4 M compared to YS and YT. Fur-thermore, sensitivity to ACh in femoral rings from AS rats

Table 1 Body and muscle masses (mean § SE)

YS (n = 10) YT (n = 10) AS (n = 9) AT (n = 9)

Body mass, g 365 § 10 321 § 6 458 § 13* 442 § 8*

Soleus muscle mass, mg 124 § 4 118 § 6 134 § 6 138 § 7

Soleus-to-body mass ratio, mg/g £ 100 34.2 § 0.8 37.9 § 1.9 29.3 § 0.8 31.1 § 0.9

Plantaris muscle mass, mg 280 § 14 297 § 18 285 § 6 340 § 16

Plantaris-to-body mass ratio, mg/g £ 100 76.8 § 3.4 92.9 § 6.4** 62.8 § 2.2 79.6 § 3.9**

*P < 0.05 aged versus young, two-way ANOVA; **P < 0.05 trained versus sedentary, two-way ANOVA

123

Eur J Appl Physiol

(EC50 = 5.93 § 0.16 £ 10¡6 M) was signiWcantly lowercompared to YS (EC50 = 2.86 § 0.019 £ 10¡6 M) and AT(EC50 = 2.38 § 0.21 £ 10¡6 M) animals. ACh-inducedvasodilation was signiWcantly diVerent in AT compared toYS only at the highest doses (3 £ 10¡5 to 10¡4 M) and nosigniWcant diVerences were observed between AT and YT.In addition, no signiWcant diVerences in ACh-induced vaso-dilation was observed between YS and YT. The maximalACh-induced percent relaxation was: (1) YS (64.4.3 §3.1%), (2) YT (59.3 § 5.5%), (3) AS (39.3 § 5.3%), and(4) AT (47.0 § 4.0%).

Maximal relaxation of femoral rings to SNP at the samepreconstriction level (PE, 10–7 M) was not signiWcantlydiVerent among groups (Fig. 3). However, the sensitivity toSNP was reduced in AS (EC50, 5.41 § 1.06 £ 10¡7) and

AT (EC50, 4.13 § 1.06 £ 10¡7) compared to YS (EC50,1.42 § 1.13 £ 10¡7) and YT (EC50, 1.87 § 1.10 £ 10¡7).

eNOS, P-Ser1179-eNOS, and Hsp90 protein expression in femoral arteries

Two-way ANOVA of immunoblotting results revealed thataged animals had signiWcantly greater (P < 0.05) eNOS(Fig. 4a) and P-eNOS (Fig. 4b) protein expression in thefemoral arteries compared to young animals which wasunaVected by resistance exercise training. However, resis-tance exercise training resulted in a signiWcant (P < 0.05)increase in Hsp90 expression regardless of age (Fig. 5).

Discussion

Previous studies examining exercise-induced improve-ments in age-associated endothelial dysfunction havefocused primarily on endurance exercise training. To date,there appears to be only one study in the literature that hasexamined the eVects of resistance exercise training on age-associated endothelial dysfunction but this study did notexamine changes in eNOS expression or other regulatoryproteins (Figard et al. 2006). The results of our studyconWrm the Wndings of Figard et al. (2006) that resistanceexercise training improves age-associated endothelial dys-function in rats. Furthermore, the present study makes thefollowing new observations: (1) improvements in endo-thelial dysfunction are not associated with an increase ineNOS expression or phosphorylation and (2) resistanceexercise training results in increased Hsp90 expression inrat femoral arteries. These data provide additional insightinto the molecular mechanisms involved in age-associatedendothelial dysfunction and resistance exercise trainingimprovements.

Fig. 1 Phenylephrine-induced vasoconstriction was signiWcantlygreater at the highest doses in aged versus young animals. No signiW-cant diVerences in the EC50 for phenylephrine were observed(*P < 0.05 aged vs. young, two-way ANOVA)

-8 -7 -6 -5 -4 -3 -20

5

10

15

20

25

30

35

40

45

YS (n=10)

YT (n=10)

AS (n=9)

AT (n=9)

phenylephrine, [log M]

Dev

elo

ped

Ten

sio

n, m

N

Fig. 2 Acetylcholine-induced vasorelaxation was signiWcantly im-paired in AS versus YS and YT (P < 0.05). No signiWcant diVerencewas observed between AT and YS, YT or AS. EC50 for AS was signiW-cantly lower (P < 0.05). *P < 0.05 AS versus YS and YT (two-wayANOVA with Bonferonni post hoc test)

-8 -7 -6 -5 -4 -3

0

25

50

75

YS (n=10)

YT (n=10)

AS (n=9)

AT (n=9)

*

**

*

acetylcholine, log [mol/L]

% r

elax

atio

n

Fig. 3 Sodium nitroprusside-induced vasorelaxation was not signiW-cantly diVerent between groups (P > 0.05)

-9 -8 -7 -6 -5 -4 -3

0

10

20

30

40

50

60

70

80

90

100

110

YS (n=10)

YT (n=10)

AS (n=9)

AT (n=9)

sodium nitroprusside, log [mol/L]

%re

laxa

tio

n

123

Eur J Appl Physiol

In the Wrst series of experiments, we established thatendothelium-dependent vasodilation in rat femoral arteriesis impaired in aged versus young rats (Fig. 2). These dataare consistent with numerous studies examining age-associ-ated endothelial dysfunction in large conduit arteries of therat such as the aorta (Barton et al. 1997; Delp et al. 1995;Jasperse and Laughlin 2006; Spier et al. 1999), but it is incontrast to one study examining the eVects of age on ratfemoral arteries (Barton et al. 1997). This diVerence may bedue to the number and strain of rats used in the study. Inboth the present study and the study by Barton et al. (1997)femoral artery aging had no eVect on vasoconstrictorresponses or on the vasodilator eVects of sodium nitroprus-side (Figs. 1, 3, respectively). Because sodium nitroprus-side acts as an NO donor, these data suggest that the agedanimals in this study have no impairment in smooth func-tion and that the vasodilatory dysfunction is endotheliumdependent.

After establishing an age-associated impairment in endo-thelial function, we demonstrated that 6 weeks of progres-sive, dynamic resistance exercise training consisting ofweighted ladder climbing improved endothelium dependentdilation in rat femoral arteries. This is consistent with areview of animal models of aging and exercise by Jasperseand Laughlin (2006) which reported that endurance or aero-bic exercise training (typically consisting of treadmill run-ning) improves age-associated endothelial dysfunction in avariety of animals and vascular beds. As stated earlier, thisdata is also consistent with the Wndings of Figard et al.(2006) who reported that isometric strength trainingimproved age-associated endothelial dysfunction in rataorta. Therefore, it appears that like endurance exercisetraining, resistance exercise training can result in improvedendothelial function.

A number of cellular and molecular mechanisms are pro-posed to play a role in age-associated endothelial dysfunc-tion and subsequent improvements with exercise training.One of the primary areas of investigation examined hasbeen changes in eNOS expression and activity. eNOSexpression has been shown to increase (Cernadas et al.1998; van der Loo et al. 2000, 2005), decrease (Bartonet al. 1997; Iemitsu et al. 2006; Soucy et al. 2006) and notchange (Schulman et al. 2007; Smith et al. 2006) in bloodvessels from aged rats. Similarly exercise training has beenshown to result in variable eNOS expression patterns(Jasperse and Laughlin 2006). DiVerences in these studiesappear to be due, in part, to diVerent rat strains, diVerentage ranges, and diVerent vascular beds examined. Twostudies have speciWcally examined eNOS expression infemoral arteries from aged rats. Barton et al. (1997)reported a decrease in eNOS mRNA expression in 32–33 month old, female, Ro-Ro Wistar rats whereas van derLoo et al. (2005) reported an increase in eNOS expression

Fig. 4 eNOS and phospho-Ser1179-eNOS protein expression wassigniWcantly greater in femoral arteries from aged animals versusyoung animals (*P < 0.05, two-way ANOVA, n ¸ 7 per group). Therewas no signiWcant eVect of resistance training on femoral artery eNOSprotein expression in aged and young animals (P > 0.05)

Young Aged

Sedentary

Trained

% o

f Y

T* (aged vs. young) A

eNOS

Young Aged

Sedentary

Trained

% o

f Y

S

YS YT AS AT

B

P-eNOS

* (aged vs. young)

0

50

100

150

200

250*

0

100

200

300

YS YT AS AT

Fig. 5 Hsp90 protein expression was signiWcantly greater in femoralarteries from trained animals versus sedentary animals (*P < 0.05,two-way ANOVA, n ¸ 9 per group). There was no signiWcant eVect ofage on femoral artery Hsp90 protein expression (P > 0.05)

Young

Sedentary

Trained

% o

f Y

T **

Hsp90

Aged

Sedentary

Trained

% o

f Y

T **

YS YT AS AT

0

50

100

150

200

123

Eur J Appl Physiol

in middle-aged (18–20 months) and old (31–34 months)male, Fisher x BN rats. The results of the present studydemonstrating increased eNOS expression in male,20 month old, Fisher 344 rats are consistent with the moreclosely related model used in the study by van der Looet al. (2005). In addition, our data demonstrating nochange in eNOS expression as a result of exercise trainingis in contrast to a few studies which have demonstrated anexercise-induced increase in eNOS expression in bloodvessels from aged rats (Spier et al. 2004; Tanabe et al.2003). However, none of theses studies have speciWcallyexamined rat femoral arteries or resistance exercise train-ing. Interestingly, the results of the current study and oth-ers (Spier et al. 2004; van der Loo et al. 2005) present aparadoxical observation that aging results in an increasedeNOS expression while NO-dependent vasodilation isreduced. Furthermore, exercise training results inimproved endothelial function despite no change in eNOSexpression. Taken together these data suggest that endo-thelial derived NO is regulated by mechanisms other thaneNOS expression.

As part of this study, we also examined whether thesealterations in endothelial-dependent vasoreactivity couldbe due to changes in eNOS phosphorylation. A review bySmith and Hagen (2003) suggests that eNOS phosphory-lation declines with age. However, the results of the pres-ent study suggest that eNOS phosphorylation on Ser-1179primarily mimicked changes in eNOS expression in therat femoral artery. In addition, while some studies haveindicated that eNOS phosphorylation is increased as aresult of exercise, these studies have examined only theeVects of endurance exercise training and have evaluatedeither coronary arteries in humans (Hambrecht et al.2003) or heart tissue from swim trained rats (Iemitsu et al.2006). The present study suggests that resistance exercisetraining does not alter eNOS activity through changes ineNOS phosphorylation at Ser-1179 in rat femoral arteries.However, while Ser-1179 is the most widely studied siteof regulatory phosphorylation on eNOS there are numberof additional sites that may also be altered by resistanceexercise training and should be considered in futurestudies.

Another potential mechanism for increased endothe-lial-derived NO production is activation and regulation ofeNOS by Hsp90. As mentioned previously, Hsp90 isbelieved to play a role in determining the balance ofsuperoxide generation and NO production from eNOS(Pritchard et al. 2001). A decrease in Hsp90 associationwith eNOS has been implicated as a mechanism of endo-thelial impairment in persistent pulmonary hypertension(Konduri et al. 2007) and hypoxia (Shi et al. 2002). Morerecently, Smith et al. (2006) have also demonstrated thataging resulted in a decline in Hsp90 association with

eNOS in rat aorta. Furthermore, increasing Hsp90 expres-sion can result in enhanced eNOS activity and improvedendothelial function (Sud et al. 2007). It appears, how-ever, that no other studies have examined the eVects ofexercise training on Hsp90 expression in blood vessels.The present study provides new evidence that vascularHsp90 expression is increased following resistance exer-cise training. Although the direct interaction of Hsp90 andeNOS was not determined in this study, the observedresistance exercise training-induced increase in Hsp90expression may result in an increase in association ofHsp90 and eNOS which may have resulted in a decreasein eNOS uncoupling accounting for the improvement inendothelial function in the aged femoral arteries. Thiswould be consistent with our previous Wnding that endur-ance exercise training increases Hsp90/eNOS interactionin skeletal muscle (Harris et al. 2008). In addition, it is notpossible in the present study to determine whether theincrease in Hsp90 expression occurs in the smooth mus-cle, endothelium or both. Future studies should directlyexamine resistance exercise-induced changes in Hsp90/eNOS association and evaluate changes in NO and super-oxide generation in the aged vessels.

In summary, the present study is the Wrst to examinemolecular mechanisms that may account for the improvedendothelial function which results from resistance exercisetraining in aged rats. While previous studies examining themechanisms behind age-associated endothelial dysfunc-tion have suggested a role for impaired NO bioavailabilitydue to an increase in superoxide generation, the presentstudy suggests that the improvement in endothelial func-tion may be the result of increased Hsp90 expression.However, one limitation of the present study is the use ofthe femoral artery for the evaluation of vascular function.The femoral artery was used in this study because of itsrole as main feed artery for the hindlimb muscles whichare active during this type of exercise. Future studiesshould examine a variety of vascular beds as it has beenshown that endurance exercise training results in non-uni-form changes in endothelium-dependent dilation acrossthe vasculature (Jasperse and Laughlin 2006). The presentstudy does provide a unique focus on the use of resistanceexercise training rather than endurance exercise training totreat age-associated endothelial dysfunction. Althoughboth modes of exercise appear to be beneWcial, it is unclearwhether this is the result of diverse or common underlyingmechanisms. For instance, we have previously shown thatincreased core temperatures result in increased Hsp90 andeNOS expression as well as increased eNOS activity(Harris et al. 2003). Future studies may focus on the roleof temperature, oxidative stress, shear stress and others aspotential mechanisms for the exercise-induced adapta-tions.

123

Eur J Appl Physiol

Acknowledgments This work was supported by an American HeartAssociation Scientist Development Grant 0430157N (M.B.H.) andThe Borgenicht Program for Aging Studies and Exercise Science(M.B.H.).

References

Bank AJ, Shammas RA, Mullen K, Chuang PP (1998) EVects of short-term forearm exercise training on resistance vessel endothelialfunction in normal subjects and patients with heart failure. J CardFail 4(3):193–201

Barton M, Cosentino F, Brandes RP, Moreau P, Shaw S, Luscher TF(1997) Anatomic heterogeneity of vascular aging: Role of nitricoxide and endothelin. Hypertension 30(4):817–824

Cernadas MR, Sanchez de Miguel L, Garcia-Duran M, Gonzalez-Fernandez F, Millas I, Monton M et al (1998) Expression ofconstitutive and inducible nitric oxide synthases in the vascularwall of young and aging rats. Circ Res 83(3):279–286

Delp MD, Brown M, Laughlin MH, Hasser EM (1995) Rat aorticvasoreactivity is altered by old age and hindlimb unloading.J Appl Physiol 78(6):2079–2086

Figard H, Gaume V, Mougin F, Demougeot C, Berthelot A (2006)BeneWcial eVects of isometric strength training on endothelialdysfunction in rats. Appl Physiol Nutr Metab 31(5):621–630

Franke WD, Stephens GM, Schmid PG 3rd (1998) EVects of intenseexercise training on endothelium-dependent exercise-inducedvasodilatation. Clin Physiol 18(6):521–528

Fulton D, Gratton JP, McCabe TJ, Fontana J, Fujio Y, Walsh K et al(1999) Regulation of endothelium-derived nitric oxide productionby the protein kinase akt [published erratum appears in nature1999 aug 19;400(6746):792]. Nature 399(6736):597–601

Green DJ, Cable NT, Fox C, Rankin JM, Taylor RR (1994) ModiWca-tion of forearm resistance vessels by exercise training in youngmen. J Appl Physiol 77(4):1829–1833

Green DJ, Maiorana A, O’Driscoll G, Taylor R (2004) EVect ofexercise training on endothelium-derived nitric oxide function inhumans. J Physiol 561(Pt 1):1–25

Hambrecht R, Adams V, Erbs S, Linke A, Krankel N, Shu Y et al(2003) Regular physical activity improves endothelial func-tion in patients with coronary artery disease by increasingphosphorylation of endothelial nitric oxide synthase. Circulation107(25):3152–3158

Harris MB, Blackstone MA, Ju H, Venema VJ, Venema RC (2003)Heat-induced increases in endothelial NO synthase expressionand activity and endothelial NO release. Am J Physiol Heart CircPhysiol 285(1):H333–H340

Harris MB, Mitchell BM, Sood SG, Webb RC, Venema RC (2008)Increased nitric oxide synthase activity and Hsp90 association inskeletal muscle following chronic exercise. Eur J Appl Physiol104(5):795–802

Hornig B, Maier V, Drexler H (1996) Physical training improves endo-thelial function in patients with chronic heart failure. Circulation93(2):210–214

Iemitsu M, Maeda S, Jesmin S, Otsuki T, Miyauchi T (2006) Exercisetraining improves aging-induced downregulation of VEGFangiogenic signaling cascade in hearts. Am J Physiol Heart CircPhysiol 291(3):H1290–H1298

Jasperse JL, Laughlin MH (2006) Endothelial function and exercisetraining: evidence from studies using animal models. Med SciSports Exerc 38(3):445–454

Katz SD, Yuen J, Bijou R, LeJemtel TH (1997) Training improvesendothelium-dependent vasodilation in resistance vessels ofpatients with heart failure. J Appl Physiol 82(5):1488–1492

Kawano H, Tanimoto M, Yamamoto K, Sanada K, Gando Y, Tabata Iet al (2008) Resistance training is associated with increased arte-rial stiVness and blood pressure but does not adversely aVectendothelial function as measured by arterial reactivity to the coldpressor test. Exp Physiol 93(2):296–302

Kell RT, Bell G, Quinney A (2001) Musculoskeletal Wtness, health out-comes and quality of life. Sports Med 31(12):863–873

Konduri GG, Bakhutashvili I, Eis A, Pritchard K Jr (2007) Oxidantstress from uncoupled nitric oxide synthase impairs vasodilationin fetal lambs with persistent pulmonary hypertension. Am JPhysiol Heart Circ Physiol 292(4):H1812–H1820

Lee S, Barton ER, Sweeney HL, Farrar RP (2004) Viral expression ofinsulin-like growth factor-I enhances muscle hypertrophy inresistance-trained rats. J Appl Physiol 96(3):1097–1104

Matz RL, Schott C, Stoclet JC, Andriantsitohaina R (2000) Age-related endothelial dysfunction with respect to nitric oxide,endothelium-derived hyperpolarizing factor and cyclooxygenaseproducts. Physiol Res 49(1):11–18

Miyachi M, Kawano H, Sugawara J, Takahashi K, Hayashi K,Yamazaki K et al (2004) Unfavorable eVects of resistance train-ing on central arterial compliance: a randomized interventionstudy. Circulation 110(18):2858–2863

Muller-Delp JM (2006) Aging-induced adaptations of microvascularreactivity. Microcirculation 13(4):301–314

Phillips SM (2007) Resistance exercise: good for more than just grandmaand grandpa’s muscles. Appl Physiol Nutr Metab 32(6):1198–1205

Pritchard KA Jr, Ackerman AW, Gross ER, Stepp DW, Shi Y, FontanaJT et al (2001) Heat shock protein 90 mediates the balance ofnitric oxide and superoxide anion from endothelial nitric-oxidesynthase. J Biol Chem 276(21):17621–17624

Schulman IH, Zhou MS, Jaimes EA, Raij L (2007) Dissociationbetween metabolic and vascular insulin resistance in aging. AmJ Physiol Heart Circ Physiol 293(1):H853–H859

Sessa WC (2004) eNOS at a glance. J Cell Sci 117(Pt 12):2427–2429Shi Y, Baker JE, Zhang C, Tweddell JS, Su J, Pritchard KA Jr (2002)

Chronic hypoxia increases endothelial nitric oxide synthasegeneration of nitric oxide by increasing heat shock protein 90association and serine phosphorylation 1. Circ Res 91(4):300–306

Smith AR, Hagen TM (2003) Vascular endothelial dysfunction inaging: loss of akt-dependent endothelial nitric oxide synthasephosphorylation and partial restoration by (R)-alpha-lipoic acid.Biochem Soc Trans 31(Pt 6):1447–1449

Smith AR, Visioli F, Hagen TM (2006) Plasma membrane-associatedendothelial nitric oxide synthase and activity in aging rat aorticvascular endothelia markedly decline with age. Arch BiochemBiophys 454(1):100–105

Soucy KG, Ryoo S, Benjo A, Lim HK, Gupta G, Sohi JS et al (2006)Impaired shear stress-induced nitric oxide production throughdecreased NOS phosphorylation contributes to age-related vascularstiVness. J Appl Physiol 101(6):1751–1759

Spier SA, Laughlin MH, Delp MD (1999) EVects of acute and chronicexercise on vasoconstrictor responsiveness of rat abdominalaorta. J Appl Physiol 87(5):1752–1757

Spier SA, Delp MD, Meininger CJ, Donato AJ, Ramsey MW, Muller-Delp JM (2004) EVects of ageing and exercise training on endo-thelium-dependent vasodilatation and structure of rat skeletalmuscle arterioles. J Physiol (Lond) 556(Pt 3):947–958

Sud N, Sharma S, Wiseman DA, Harmon C, Kumar S, Venema RCet al (2007) Nitric oxide and superoxide generation from endothe-lial NOS: modulation by HSP90. Am J Physiol Lung Cell MolPhysiol 293(6):L1444–L1453

Tanabe T, Maeda S, Miyauchi T, Iemitsu M, Takanashi M, Irukayama-Tomobe Y et al (2003) Exercise training improves ageing-induced decrease in eNOS expression of the aorta. Acta PhysiolScand 178(1):3–10

123

Eur J Appl Physiol

van der Loo B, Labugger R, Skepper JN, Bachschmid M, Kilo J,Powell JM et al (2000) Enhanced peroxynitrite formation is asso-ciated with vascular aging. J Exp Med 192(12):1731–1744

van der Loo B, Bachschmid M, Labugger R, Schildknecht S, Kilo J,Hahn R et al (2005) Expression and activity patterns of nitric

oxide synthases and antioxidant enzymes reveal a substantialheterogeneity between cardiac and vascular aging in the rat.Biogerontology 6(5):325–334

123