http://vmj.sagepub.com/Vascular Medicine

http://vmj.sagepub.com/content/14/1/29The online version of this article can be found at:

 DOI: 10.1177/1358863X08096666

2009 14: 29Vasc MedLosordo

Willem-Peter T Ruifrok, Rudolf A de Boer, Atsushi Iwakura, Marcy Silver, Kengo Kusano, Rene A Tio and Douglas Wischemia

Estradiol-induced, endothelial progenitor cell-mediated neovascularization in male mice with hind-limb  

Published by:

http://www.sagepublications.com

On behalf of: 

  Society for Vascular Medicine

can be found at:Vascular MedicineAdditional services and information for     

  http://vmj.sagepub.com/cgi/alertsEmail Alerts:

 

http://vmj.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://vmj.sagepub.com/content/14/1/29.refs.htmlCitations:  

by guest on May 18, 2011vmj.sagepub.comDownloaded from

Estradiol-induced, endothelial progenitor cell-mediatedneovascularization in male mice with hind-limb ischemia

Willem-Peter T Ruifrok1,2, Rudolf A de Boer1, Atsushi Iwakura2, Marcy Silver2, Kengo Kusano2, Rene A Tio1 andDouglas W Losordo3

Abstract: We investigated whether administration of estradiol to male mice aug-

ments mobilization of bone marrow-derived endothelial progenitor cells (EPC) and

incorporation into foci of neovascularization after hind-limb ischemia, thereby con-

tributing to blood flow restoration. Mice were randomized and implanted with pla-

cebo pellets or pellets containing low-dose estradiol (0.39 mg) or high-dose estradiol

(1.7 mg). Hind-limb ischemia was induced by unilateral resection of the left femoral

artery 1 week after pellet implantation, then EPC mobilization and functional recovery

was evaluated. EPC recruitment was assessed in mice transplanted with bone mar-

row from transgenic donors expressing β-galactosidase driven by the Tie-2 promoter.

EPC culture assay performed 2 weeks after pellet implantation revealed a significantly

greater (p < 0.05) number of circulating EPCs in the high-dose estradiol group than in

the low-dose estradiol and placebo groups. At 3 and 4 weeks after induction of hind-

limb ischemia, perfusion was significantly greater (p < 0.05) in high-dose estradiol

mice than in mice implanted with the low-dose estradiol or placebo pellets. At 1 and

4 weeks after hind-limb ischemia surgery, more bone marrow-derived EPCs, identi-

fied as β-galactosidase-positive cells, were observed in ischemic regions from high-

dose estradiol animals than in low-dose (p < 0.05) or placebo groups (p < 0.05). These

results indicate that estradiol dose-dependently increases the levels of EPCs in

peripheral blood in male animals, improves the recovery of blood flow, and

decreases limb necrosis after hind-limb ischemia, and that this enhancement occurs,

in part, through augmentation of EPC mobilization and greater incorporation of bone

marrow-derived EPCs into foci of neovascularization.

Key words: angiogenesis; endothelial progenitor cell; estrogen; neovascularization

Introduction

Since the discovery of endothelial progenitor cells(EPCs) in 1997,1 many studies have investigatedtheir potential angiogenic capacity.2–6 The relation-ships between EPCs and various growth factors,cytokines, and hormones, including estrogen,7

have been studied in an effort to identify their poten-tial for therapeutic angiogenesis. The emergingconsensus of these studies is that blood-vessel regen-

eration evolves not only from the migration andproliferation of endothelial cells adjacent to thesite of injury (a process called angiogenesis), butthat circulating EPCs also contribute to the forma-tion of new blood vessels.1,8–10 The mechanisms thatregulate EPC-mediated neovascularization areslowly being characterized.5,11,12

Estrogen regulates angiogenesis by influencingthe activity of endothelial cells. Under normal phys-iological conditions, angiogenesis in the uterus isassociated with fluctuations of circulating estradiol(E2) and other sex steroids.13–17 The well-established link between estrogen and breast tumorinvasiveness also suggests that estrogen exerts a pro-angiogenic effect because the formation of metasta-ses is dependent on angiogenesis.18 Studies in micehave revealed that E2 accelerates endothelial recov-ery after arterial injury,19,20 and E2 has been shownto prevent endothelial cell apoptosis5,21,22 and toexert other positive effects on mature endothelialcells.

1Department of Cardiology, University Medical CenterGroningen, University of Groningen, Groningen, TheNetherlands; 2Division of Cardiovascular Research, StElizabeth’s Medical Center, Tufts University Medical School,Boston, MA, USA; 3Feinberg Cardiovascular ResearchInstitute, Northwestern University Feinberg School ofMedicine, Chicago, IL, USA

Correspondence to: DW Losordo, Feinberg CardiovascularResearch Institute, Northwestern University Feinberg Schoolof Medicine, 303 E Chicago Ave., Tarry 12-703, Chicago, IL60611, USA. Email: d-losordo@northwestern.edu

Vascular Medicine 2009; 14: 29–36

© 2009 SAGE Publications, Los Angeles, London, New Delhi and Singapore 10.1177/1358863X08096666

by guest on May 18, 2011vmj.sagepub.comDownloaded from

To date, experiments in animal models that haveinvestigated the effects of E2 on EPCs have beenperformed exclusively in female animals,5,12,22 andit is not known whether these results can be extrap-olated to males. Furthermore, the potential contri-bution of EPC-mediated neovascularization tofunctional recovery at the ischemic site continuesto be debated. This study sought to determinewhether estrogen mobilizes EPCs from the bonemarrow after hind-limb ischemia (HLI) in male ani-mals, and whether the mobilized EPCs subsequentlycontribute to post-ischemic neovascularization.

Methods

Animal modelsAll experiments were conducted according to proto-cols approved by the Animal Care and Use Com-mittee of St Elizabeth’s Medical Center, Boston,MA and conform to the Guide for the Care andUse of Laboratory Animals by the US NationalInstitutes of Health. Experiments were performedin male FVB wild type (WT) mice (Jackson Labo-ratory, Bar Harbor, ME, USA), aged 6 weeks. HLIwas induced by unilateral resection of the left femo-ral artery from the proximal end of the femoralartery up to the distal portion of the saphenousvein. The femoral artery and all side-brancheswere dissected and excised, then the overlying skinwas closed using a surgical stapler. After surgery,the mice were kept on a warm plate until fullyawake. For bone marrow transplant experiments,mice were irradiated (16 G) to destroy endogenousbone marrow and transplanted with a minimum of3 × 106 bone marrow cells from age-matched maletransgenic donors that expressed β-galactosidase;β-galactosidase transcription was regulated by theendothelial cell-specific Tie-2 promoter. Mice wereanesthetized by intraperitoneal injection with a mix-ture of ketamine (80 mg/kg) and xylazine (4 mg/kg)and sacrificed with an overdose of the same anes-thetic. Animals were housed under standard con-ditions with ad libitum food and water. Over theduration of the experiments, the mortality rate was6% for intact mice, 10% for mice that underwentHLI surgery alone, and 33% for mice that under-went both bone marrow transplantation and HLIsurgery.

DesignThree experiments were performed. For each exper-iment, non-castrated WTmale mice were implantedwith 90-day release E2 pellets containing high(1.7 mg/pellet) or low (0.39 mg/pellet) doses of E2(17β-estradiol), or placebo (Innovative Research ofAmerica, Sarasota, FL, USA). In a previous study

of experimental carotid artery injury, serum E2levels were approximately 300-fold greater in ovari-ectomized female mice implanted with the high-dose E2 pellet than in mice implanted with theplacebo pellet.5 Pellets were implanted subcutane-ously as described elsewhere.12

The first experiment was performed with 60 mice(n = 20 in each treatment group). Mice wereimplanted at Week 0, and five mice from eachgroup were sacrificed at weekly intervals over thenext 4 weeks. Blood was collected by intracardiacpuncture, and the number of circulating EPCs wasassessed via the EPC culture assay.In the second experiment, 60 mice (n = 20 in each

treatment group) were implanted at Week –1, HLIwas induced at week 0, and five mice from eachgroup were sacrificed at weekly intervals over thenext 4 weeks. At sacrifice, all ischemic, non-necrotic tissue was collected and stained with fluo-rescent CD31 antibodies to assess capillary density;blood was collected by intracardiac puncture, andthe number of circulating EPCs was assessed viathe EPC culture assay. Functional recovery wasevaluated from Laser Doppler Perfusion Imaging(LDPI) measurements taken before HLI surgeryand at sacrifice, and limb necrosis was assessed visu-ally as previously described.23

For the third experiment, 42 mice underwentbone marrow transplantation surgery at Week –5,and the bone marrow was allowed to regenerateover 4 weeks. Surviving mice (n = 14 in each treat-ment group) were implanted at Week –1, HLIsurgery was performed at Week 0, and mice weresacrificed at Week 1 and Week 4. After sacrifice,all ischemic, non-necrotic tissue was collected andstained for expression of β-galactosidase (a specificmarker for cells derived from donor bone marrow)and CD31 (a specific marker for endothelial cells) toidentify bone marrow-derived EPCs at the sites ofneovascularization.

EPC culture assayTotal mononuclear cells were isolated from 1 ml ofarterial blood by density gradient centrifugationusing a Histopaque-1083 (Sigma-Aldrich Inc.,St Louis, MO, USA), then cultured in phenol red-free endothelial cell basal medium (Clonetics Corp.,San Diego, CA, USA) supplemented with 5% fetalbovine serum, antibiotics, and growth factors onfour-well glass slides coated with rat plasma vitro-nectin (Sigma-Aldrich Inc.) in 0.5% gelatin solution.After 4 days, EPCs were recognized as attachedspindle-shaped cells and assayed by costainingwith acetylated LDL-DiI (Biomedical Technolo-gies, Stoughton, MA, USA) and FITC-conjugatedBandeiraea simplicifolia lectin I (Vector Laborato-ries, Burlingame, CA, USA). Fifteen randomly

30 WT Ruifrok et al.

Vascular Medicine 2009; 14: 29–36

by guest on May 18, 2011vmj.sagepub.comDownloaded from

selected fields on each slide were viewed with fluo-rescence microscopy at 20× magnification, anddouble-positive cells were identified as EPCs24 andcounted; EPC counts were presented as the averagenumber of EPCs/mm2.

PerfusionPerfusion in both ischemic (left) and non-ischemic(right) limbs was quantified via LDPI (Moor Instru-ments Inc, Wilmington, DE, USA) before HLI sur-gery and at predetermined time points afterward.Three consecutive measurements were obtained overthe affected region (leg and foot) for each time point.Color-coded images were recorded, and the averageperfusion in each (i.e. ischemic and non-ischemic)foot was calculated. To account for variations inambient light, temperature, and other conditions,perfusion was presented as the ratio of ischemic(left) to non-ischemic (right) measurements.

Capillary densityTissue samples were mounted and stained for CD31expression, and one slide for each mouse was exam-ined with light microscopy at 20× magnification.Capillaries positively stained for CD31 (an endothe-lial cell-specific marker) were counted in five ran-domly chosen microscopic fields on each slide, andcapillary density was presented as the average num-ber of capillaries/mm2.

Bone marrow-derived cellsTissue samples from WT mice transplanted withbone marrow from transgenic donors were mountedand stained for both β-galactosidase and CD31expression. Two slides for each mouse were viewedunder fluorescent microscopy at 20× magnification.Five randomly chosen fields were examined for eachslide, and cells that were positively stained for bothmarkers were identified and counted; counts werepresented as the average number of double-positive cells/mm2.

StatisticsResults are reported as mean ± standard error of themean (SEM), and statistical analyses were per-formed with SPSS 12.0 for Windows. Comparisonsbetween groups were analyzed via ANOVA withDunnett’s correction, and a p-value of less than0.05 was considered significant.

Results

Circulating EPC countsWithin the low- and high-dose E2 treatment groups,mice with HLI displayed significantly greater EPC

counts at all weeks than those measured in mice thatdid not undergo HLI surgery (p < 0.05); EPCcounts in animals implanted with the high-dose E2pellet peaked at Week 2 in both the presence andabsence of HLI. In the absence of HLI, the numberof circulating EPCs at Weeks 1 and 2 was signifi-cantly higher in animals treated with the high-doseE2 pellet than in the low-dose (p < 0.05) or placebo-treatment groups (p < 0.05) (Figure 1A), and thedifference between the high-dose group and the pla-cebo group remained significant at Week 3(p < 0.05); there was no significant differencebetween EPC counts in the low-dose and placebogroups at any time point. After induction of HLI,EPC counts were significantly greater in mice thathad received the high-dose pellet than in placebo-treated mice at Week 2 (p < 0.05 versus placebo)and Week 3 (p < 0.01 for high-dose and p < 0.05 for

Figure 1 Endothelial progenitor cell counts in miceimplanted with pellets containing 1.7 mg estradiol (highdose), 0.39 mg estradiol (low dose), or placebo. (A) Intactmice were implanted at Week 0 and the EPC culture assaywas performed 1–4 weeks later. (B) Mice were implantedat Week –1, hind-limb ischemia was induced at Week 0,and the EPC culture assay was performed 1–4 weekslater. *p < 0.05 versus the low-dose group, ‡p < 0.05 ver-sus the placebo group.

Estradiol modulates EPC-mediated neovascularization in males 31

Vascular Medicine 2009; 14: 29–36

by guest on May 18, 2011vmj.sagepub.comDownloaded from

low-dose versus placebo); EPC counts in the high-dose animals were greater than those in the low-dosegroup, but the difference was not statistically signif-icant (Figure 1B).

Perfusion and limb necrosisAt 3 and 4 weeks after HLI surgery, high-dose E2treatment was associated with significantly greaterrestoration of perfusion than was observed in ani-mals implanted with either the low-dose E2 or pla-cebo pellets (Figure 2B). The ratio of perfusion inthe ischemic and non-ischemic limbs was 0.82 ±0.03 in Week 3 and 0.88 ± 0.11 in Week 4 amonganimals that received the high-dose E2 pellet, com-pared to 0.44 ± 0.03 (p < 0.01) and 0.47 ± 0.01(p < 0.05), respectively, in animals that receivedthe placebo pellet. The differences between low-dose E2 and placebo treatment were not significantat any time point. High-dose E2 treatment was also

associated with a lower proportion of limb necrosisat Week 4 (0%) than was observed in animalsimplanted with the placebo pellet (50%); limb necro-sis in the low-dose E2 group was 22%.

Capillary densityAt 2, 3 and 4 weeks after induction of HLI, capillarydensity was significantly higher (p < 0.05) in micetreated with a high dose of E2 than in animalsimplanted with the placebo pellet (Figure 3), andthe differences between the high-dose and low-doseE2 treatment groups at Weeks 2 and 4 were alsosignificant (p < 0.05).

Bone marrow-derived cellsAt 1 and 4 weeks after induction of HLI in mice thathad undergone bone marrow transplantation sur-gery, animals implanted with the high-dose E2 pelletdisplayed significantly more β-galactosidase-positive(i.e. bonemarrow-derived) cells at the site of ischemiathan were observed in either the low-dose group orplacebo-treated animals (Figure 4A); the differencebetween the low-dose E2 and placebo-treatmentgroups was significant at Week 1. The number ofβ-galactosidase-positive cells at the site of ischemiadeclined between Weeks 1 and 4 in all treatmentgroups.Cells that stained positive for both β-galacto-

sidase and CD31 (representative of bone marrow-derived capillary density) were significantly morecommon 1 week after induction of HLI in micetreated with a high dose of E2 than in low-dose E2or placebo-treated animals, but this differenceabated by Week 4, when the prevalence of double-positive cells was less than 25 cells/mm2 in all groups

Figure 2 Recovery of perfusion in mice implanted withpellets containing 1.7 mg estradiol (high dose), 0.39 mgestradiol (low dose), or placebo. Mice were implanted atWeek -1, hind-limb ischemia was induced at Week 0, andlaser Doppler perfusion imaging (LDPI) was performed1–4 weeks later. (A) Representative LDPI images foreach treatment group at various time points; severelyrestricted limb blood flow is indicated in blue, and redindicates normal limb perfusion. (B) The ratio of perfu-sion recorded in the ischemic and normal limbs.‡p < 0.05 versus the placebo group.

Figure 3 Capillary density in mice implanted with pel-lets containing 1.7 mg estradiol (high dose), 0.39 mgestradiol (low dose), or placebo. Mice were implanted atWeek –1, and hind-limb ischemia was induced at Week 0.Tissue samples were obtained from mice sacrificed1–4 weeks later and stained for expression of CD31 (anendothelial cell-specific marker). *p < 0.05 versus thelow-dose group, ‡p < 0.05 versus the placebo group.

32 WT Ruifrok et al.

Vascular Medicine 2009; 14: 29–36

by guest on May 18, 2011vmj.sagepub.comDownloaded from

(Figure 4B). The prevalence of double-positive cellsdid not differ significantly between the low-dose E2and placebo groups.

Discussion

Peripheral artery disease affects 8–12 million peoplein the United States.25,26 The prevalence of periph-eral artery disease is lower in premenopausal

women than in men,27 and one factor believed tocontribute to this difference is hormonal status.The potential protective role of E2 in peripheraland cardiovascular disease is implied by results inanimal models and findings from extensive observa-tional human studies.28–30 Administration of E2yields profound effects in both ischemia-reperfusion and chronic heart failure models,31,32

and high doses of E2 attenuated the developmentof heart failure in a rat model of myocardialinfarction.33,34 The acute protective effects of E2

Figure 4 Incorporation of bone marrow-derived endothelial cells at the sites of neovascularization in mice implantedwith pellets containing 1.7 mg estradiol (high dose), 0.39 mg estradiol (low dose), or placebo. Mice underwent trans-plantation surgery at Week –5, were implanted at Week –1, and hind-limb ischemia was induced at Week 0. Tissue sam-ples were obtained from mice sacrificed 1 and 4 weeks later and stained for expression of CD31 (an endothelial cell-specific marker) and β-galactosidase (a specific marker for cells derived from donor bone marrow). (A) Representativeslides quantitatively stained for β-galactosidase expression; positively stained cells are blue. (B) The prevalence of cellsstained positive for β-galactosidase expression. (C) Representative slides quantitatively stained for CD31 andβ-galactosidase expression; the three panels display (from top to bottom) endothelial cells (green), bone marrow-derived cells (red), and bone marrow-derived endothelial cells (yellow). (D) The prevalence of cells stained positive forboth CD31 and β-galactosidase expression. *p < 0.05 versus the low-dose group, ‡p < 0.05 versus the placebo group.

Estradiol modulates EPC-mediated neovascularization in males 33

Vascular Medicine 2009; 14: 29–36

by guest on May 18, 2011vmj.sagepub.comDownloaded from

treatment likely evolve, at least in part, through theinhibition of apoptosis.21,35

To date, only female animals have been used inanimal-model investigations of the potential benefi-cial effects of E2 for peripheral arterial disease, sowe sought to extend these studies to male animals byadministering E2 to male mice with HLI. Surgicalinduction of HLI is a common and reliable methodfor modeling peripheral arterial disease,36–40 andbecause ischemia occurs in the limb, rather than inthe heart, functional recovery can be easily moni-tored by using LDPI to measure blood flow. Func-tional data is much more difficult to obtain in mod-els of cardiovascular diseases such as myocardialischemia-reperfusion injury.Our results indicate that the beneficial effects of

E2 administration observed in female animals withHLI are reproducible in males. Four weeks afterHLI surgery, mice administered high doses of E2had recovered nearly 90% of non-ischemic perfusionand displayed no necrosis, compared to 47% recov-ery and 50% necrosis in placebo-treated animals;capillary density was also greater with high-doseE2 treatment than with placebo. Furthermore, theeffects appeared to be mediated by EPCs; E2 admin-istration dose-dependently mobilized EPCs fromthe bone marrow, and the mobilized EPCs partici-pated in neovascularization. The prevalence of bonemarrow-derived cells in the ischemic region wasgreater in animals that received the high-dose E2pellet than in the placebo group at Weeks 1 and 4,and cells expressing both β-galactosidase andCD31 were more common in the high-dose groupat Week 1. The difference in double-positive cellprevalence abated by Week 4 because the activity ofβ-galactosidase decreased over time.The involvement of EPCs in angiogenesis has been

a popular topic of research, andmany studies indicatethat strategies involving the use of EPCs for therapeu-tic angiogenesis are feasible. Many growth factors,cytokines, and hormones can influence EPC mobili-zation and homing,41 including vascular endothelialgrowth factor,4,24,42–44 stromal cell-derived factor 1,45

statins,46 and erythropoietin,47–49 and the therapeuticpotential of these agents for enhancing EPC-mediated angiogenesis is being evaluated. The effectsof E2 are dependent on EPC incorporation into thesites of neovascularization and are mediated byendothelial nitric oxide synthase,5 and matrixmetalloproteinase-9.12 Estradiol also reduces myo-cyte apoptosis in vivo and in vitro via estrogenreceptor- and phospho-inositide-3 kinase/akt-dependent pathways.21,34 Estrogen receptors α and βhave been extensively investigated and are vital forthe incorporation of bone marrow-derived EPCs inischemic tissue; the role of the α receptor appears tobe more prominent.11,50

We acknowledge that the EPC culture assay doesnot completely exclude cells of a non-endotheliallineage, and that FACS analysis could provide addi-tional specificity. However, the maximum amountof blood that can be taken from a mouse (1 ml) isnot sufficient for both the EPC culture assay andFACS analysis, and this limitation can be overcomeonly by pooling blood samples from several mice,which would reduce the statistical power of subse-quent analyses. The limited amount of bloodobtainable from each mouse also precluded themeasurement of circulating E2 levels. Finally, thesmall number of animals (five to seven per treat-ment group per time point) yielded a large distribu-tion of measurements for each time point, whichmay have occasionally prevented the differencesbetween treatment groups from reaching statisticalsignificance.In conclusion, our study shows that E2 dose-

dependently increases the level of circulating EPCsin the peripheral blood of both intact male mice andmale mice with HLI, and that E2-mobilized EPCshome to the foci of neovascularization during HLIto promote recovery. Furthermore, E2 was alsoassociated with improved blood-flow recovery,greater capillary density, and less limb necrosis,with optimal effects observed 2–3 weeks after theonset of HLI. These findings suggest that systemicE2 administration may increase EPC-mediated neo-vascularization under ischemic conditions causedby peripheral artery disease.

Acknowledgements

Funding: National Institutes of Health (HL-53354,HL-77428, HL-63414, HL-57516, HL-80137 andHL-P01-66957, all to DWL); Netherlands HeartFoundation (to WTR).All authors report no conflicts of interest.

References

1 Asahara, T, Murohara, T, Sullivan, A, et al. Isolation ofputative progenitor endothelial cells for angiogenesis. Sci-ence 1997; 275: 964–967.

2 Amrani, DL, Port, S. Cardiovascular disease: potentialimpact of stem cell therapy. Expert Rev Cardiovasc Ther2003; 1: 453–461.

3 Dobert, N, Britten, M, Assmus, B, et al. Transplantation ofprogenitor cells after reperfused acute myocardial infarc-tion: evaluation of perfusion and myocardial viability withFDG-PET and thallium SPECT. Eur J Nucl Med MolImaging 2004; 31: 1146–1151.

4 Iwaguro, H, Yamaguchi, J, Kalka, C, et al. Endothelialprogenitor cell vascular endothelial growth factor gene

34 WT Ruifrok et al.

Vascular Medicine 2009; 14: 29–36

by guest on May 18, 2011vmj.sagepub.comDownloaded from

transfer for vascular regeneration. Circulation 2002; 105:732–738.

5 Iwakura, A, Luedemann, C, Shastry, S, et al. Estrogen-mediated, endothelial nitric oxide synthase-dependentmobilization of bone marrow-derived endothelial progeni-tor cells contributes to reendothelialization after arterialinjury. Circulation 2003; 108: 3115–3121.

6 Kudo, FA, Nishibe, T, Nishibe, M, Yasuda, K. Autolo-gous transplantation of peripheral blood endothelial pro-genitor cells (CD34+) for therapeutic angiogenesis inpatients with critical limb ischemia. Int Angiol 2003; 22:344–348.

7 Krasinski, K, Spyridopoulos, I, Asahara, T, van der Zee,R, Isner, JM, Losordo, DW. Estradiol accelerates func-tional endothelial recovery after arterial injury. Circulation1997; 95: 1768–1772.

8 Reyes, M, Dudek, A, Jahagirdar, B, Koodie, L, Marker,PH, Verfaillie, CM. Origin of endothelial progenitors inhuman postnatal bone marrow. J Clin Invest 2002; 109:337–346.

9 Shi, Q, Rafii, S, Wu, MH, et al. Evidence for circulatingbone marrow-derived endothelial cells. Blood 1998; 92:362–367.

10 Takahashi, T, Kalka, C, Masuda, H, et al. Ischemia- andcytokine-induced mobilization of bone marrow-derivedendothelial progenitor cells for neovascularization. NatMed 1999; 5: 434–438.

11 Hamada, H, Kim, MK, Iwakura, A, et al. Estrogen recep-tors alpha and beta mediate contribution of bone marrow-derived endothelial progenitor cells to functional recoveryafter myocardial infarction. Circulation 2006; 114: 2261–2270.

12 Iwakura, A, Shastry, S, Luedemann, C, et al. Estradiolenhances recovery after myocardial infarction by augment-ing incorporation of bone marrow-derived endothelialprogenitor cells into sites of ischemia-induced neovascular-ization via endothelial nitric oxide synthase-mediated acti-vation of matrix metalloproteinase-9. Circulation 2006;113: 1605–1614.

13 Shifren, JL, Tseng, JF, Zaloudek, CJ, et al. Ovarian steroidregulation of vascular endothelial growth factor in thehuman endometrium: implications for angiogenesis duringthe menstrual cycle and in the pathogenesis of endometri-osis. J Clin Endocrinol Metab 1996; 81: 3112–3118.

14 Findlay, JK. Angiogenesis in reproductive tissues. J Endo-crinol 1986; 111: 357–366.

15 Koos, RD, Le Maire, WJ. Factors that may regulate thegrowth and regression of blood vessels in the ovary. SeminReprod Endocrinol 1983; 1: 295–307.

16 Reynolds, LP, Killielea, SD, Redmer, DA. Angiogenesis inthe female reproductive system. FASEB J 1992; 6: 886–892.

17 Goede, V, Schmidt, T, Kimmina, S, Kozian, D, Augustin,HG. Analysis of blood vessel maturation process duringcyclic ovarian angiogenesis. Lab Invest 1998; 78: 1385–1394.

18 Losordo, DW, Isner, JM. Estrogen and angiogenesis: areview. Arterioscler Thromb Vasc Biol 2001; 21: 6–12.

19 Garcia, PM, Gimenez, J, Bonacasa, B, et al. 17beta-estradiol exerts a beneficial effect on coronary vascularremodeling in the early stages of hypertension in spontane-ously hypertensive rats. Menopause 2005; 12: 453–459.

20 Miller, AP, Feng, W, Xing, D, et al. Estrogen modulatesinflammatory mediator expression and neutrophil chemo-taxis in injured arteries. Circulation 2004; 110: 1664–1669.

21 Patten, RD, Pourati, I, Aronovitz, MJ, et al. 17beta-estradiol reduces cardiomyocyte apoptosis in vivo and invitro via activation of phospho-inositide-3 kinase/Akt sig-naling. Circ Res 2004; 95: 692–699.

22 Strehlow, K, Werner, N, Berweiler, J, et al. Estrogenincreases bone marrow-derived endothelial progenitor cellproduction and diminishes neointima formation. Circula-tion 2003; 107: 3059–3065.

23 Goukassian, DA, Qin, G, Dolan, C, et al. Tumor necrosisfactor-alpha receptor p75 is required in ischemia-inducedneovascularization. Circulation 2007; 115: 752–762.

24 Asahara, T, Takahashi, T, Masuda, H, et al. VEGF contri-butes to postnatal neovascularization by mobilizing bonemarrow-derived endothelial progenitor cells. EMBO J1999; 18: 3964–3972.

25 Hirsch, AT, Criqui, MH, Treat-Jacobson, D, et al. Periph-eral arterial disease detection, awareness, and treatment inprimary care. JAMA 2001; 286: 1317–1324.

26 Criqui, MH, Vargas, V, Denenberg, JO, et al. Ethnicity andperipheral arterial disease: the San Diego Population Study.Circulation 2005; 112: 2703–2707.

27 Selvin, E, Erlinger, TP. Prevalence of and risk factors forperipheral arterial disease in the United States: resultsfrom the National Health and Nutrition Examination Sur-vey, 1999–2000. Circulation 2004; 110: 738–743.

28 Mendelsohn, ME, Karas, RH. The protective effects ofestrogen on the cardiovascular system. N Engl J Med1999; 340: 1801–1811.

29 Nikolic, I, Liu, D, Bell, JA, Collins, J, Steenbergen, C,Murphy, E. Treatment with an estrogen receptor-beta-selective agonist is cardioprotective. J Mol Cell Car-diol 2007; 42: 769–780.

30 Urata, Y, Ihara, Y, Murata, H, et al. 17Beta-estradiol pro-tects against oxidative stress-induced cell death through theglutathione/glutaredoxin-dependent redox regulation ofAkt in myocardiac H9c2 cells. J Biol Chem 2006; 281:13092–13102.

31 Shinohara, T, Takahashi, N, Ooie, T, et al. Estrogen inhi-bits hyperthermia-induced expression of heat-shock protein72 and cardioprotection against ischemia/reperfusion injuryin female rat heart. J Mol Cell Cardiol 2004; 37: 1053–1061.

32 Zhai, P, Eurell, TE, Cotthaus, R, Jeffery, EH, Bahr, JM,Gross, DR. Effect of estrogen on global myocardialischemia-reperfusion injury in female rats. Am J Physiol2000; 279: H2766–H2775.

33 Beer, S, Reincke, M, Kral, M, et al. High-dose 17beta-estradiol treatment prevents development of heart failurepost-myocardial infarction in the rat. Basic Res Cardiol2007; 102: 9–18.

34 Satoh, M, Matter, CM, Ogita, H, et al. Inhibition ofapoptosis-regulated signaling kinase-1 and prevention ofcongestive heart failure by estrogen. Circulation 2007; 115:3197–3204.

35 Kim, JK, Pedram, A, Razandi, M, Levin, ER. Estrogenprevents cardiomyocyte apoptosis through inhibition ofreactive oxygen species and differential regulation of p38kinase isoforms. J Biol Chem 2006; 281: 6760–6767.

36 Hopkins, SP, Bulgrin, JP, Sims, RL, Bowman, B,Donovan, DL, Schmidt, SP. Controlled delivery of vascu-lar endothelial growth factor promotes neovascularization

Estradiol modulates EPC-mediated neovascularization in males 35

Vascular Medicine 2009; 14: 29–36

by guest on May 18, 2011vmj.sagepub.comDownloaded from

and maintains limb function in a rabbit model of ischemia.J Vasc Surg 1998; 27: 886–894 discussion 95.

37 Kyriakides, ZS, Petinakis, P, Kaklamanis, L, et al. Genderdoes not influence angiogenesis and arteriogenesis in a rab-bit model of chronic hind limb ischemia. Int J Cardiol 2003;92: 83–91.

38 Kyriakides, ZS, Petinakis, P, Kaklamanis, L, et al. Intra-muscular administration of estrogen may promote angio-genesis and perfusion in a rabbit model of chronic limbischemia. Cardiovasc Res 2001; 49: 626–633.

39 Senthilkumar, A, Smith, RD, Khitha, J, et al. Sildenafilpromotes ischemia-induced angiogenesis through a PKG-dependent pathway. Arterioscler Thromb Vasc Biol 2007;27: 1947–1954.

40 Tiidus, PM, Deller, M, Bombardier, E, Gul, M, Liu, XL.Estrogen supplementation failed to attenuate biochemicalindices of neutrophil infiltration or damage in rat skeletalmuscles following ischemia. Biol Res 2005; 38: 213–223.

41 Di Stefano, R, Santoni, T, Barsotti, MC, et al. Differentgrowth conditions for peripheral blood endothelial progeni-tors. Cardiovasc Radiat Med 2002; 3: 172–175.

42 Kalka, C, Masuda, H, Takahashi, T, et al. Vascular endo-thelial growth factor(165) gene transfer augments circulat-ing endothelial progenitor cells in human subjects. Circ Res2000; 86: 1198–1202.

43 Silvestre, JS, Tamarat, R, Ebrahimian, TG, et al. Vascularendothelial growth factor-B promotes in vivo angiogenesis.Circ Res 2003; 93: 114–123.

44 Yoon, YS, Losordo, DW. All in the family: VEGF-B joinsthe ranks of proangiogenic cytokines. Circ Res 2003; 93:87–90.

45 Yamaguchi, J, Kusano, KF, Masuo, O, et al. Stromal cell-derived factor-1 effects on ex vivo expanded endothelialprogenitor cell recruitment for ischemic neovascularization.Circulation 2003; 107: 1322–1328.

46 Llevadot, J, Murasawa, S, Kureishi, Y, et al. HMG-CoAreductase inhibitor mobilizes bone marrow-derived endo-thelial progenitor cells. J Clin Invest 2001; 108: 399–405.

47 Bahlmann, FH, De Groot, K, Spandau, JM, et al. Erythro-poietin regulates endothelial progenitor cells. Blood 2004;103: 921–926.

48 Heeschen, C, Aicher, A, Lehmann, R, et al. Erythropoietinis a potent physiologic stimulus for endothelial progenitorcell mobilization. Blood 2003; 102: 1340–1346.

49 Westenbrink, BD, Lipsic, E, van der Meer, P, et al. Eryth-ropoietin improves cardiac function through endothelialprogenitor cell and vascular endothelial growth factormediated neovascularization. Eur Heart J 2007; 28: 2018–2027.

50 Booth, EA, Obeid, NR, Lucchesi, BR. Activation of estro-gen receptor-alpha protects the in vivo rabbit heart fromischemia-reperfusion injury. Am J Physiol 2005; 289:H2039–H2047.

36 WT Ruifrok et al.

Vascular Medicine 2009; 14: 29–36

by guest on May 18, 2011vmj.sagepub.comDownloaded from