thymosin β4 facilitates epicardial neovascularization of the injured adult heart

Ann. N.Y. Acad. Sci. ISSN 0077-8923

ANNALS OF THE NEW YORK ACADEMY OF SCIENCESIssue: Thymosins in Health and Disease

Thymosin β4 facilitates epicardial neovascularizationof the injured adult heart

Nicola Smart,1 Catherine A. Risebro,1 James E. Clark,2 Elisabeth Ehler,3 Lucile Miquerol,4

Alex Rossdeutsch,1 Michael S. Marber,2 and Paul R. Riley1

1Molecular Medicine Unit, UCL Institute of Child Health, London, UK. 2Cardiovascular Division, Kings College London, RayneInstitute, St Thomas’ Hospital, London, UK. 3The Randall Centre of Cell and Molecular Biophysics and The CardiovascularDivision, King’s College, London, UK. 4Institut de Biologie du Developpement de Marseille-Luminy, Universite de laMediterranee, Marseille, France

Address for correspondence: Paul R. Riley, Molecular Medicine Unit, UCL Institute of Child Health, London, United [email protected]

Ischemic heart disease complicated by coronary artery occlusion causes myocardial infarction (MI), which isthe major cause of morbidity and mortality in humans (http://www.who.int/cardiovascular_diseases/resources/atlas/en/index.html). After MI the human heart has an impaired capacity to regenerate and, despite the high preva-lence of cardiovascular disease worldwide, there is currently only limited insight into how to stimulate repair ofthe injured adult heart from its component parts. Efficient cardiac regeneration requires the replacement of lostcardiomyocytes, formation of new coronary blood vessels, and appropriate modulation of inflammation to preventmaladaptive remodeling, fibrosis/scarring, and consequent cardiac dysfunction. Here we show that thymosin β4(Tβ4) promotes new vasculature in both the intact and injured mammalian heart. We demonstrate that limitedEPDC-derived endothelial-restricted neovascularization constitutes suboptimal “endogenous repair,” following in-jury, which is significantly augmented by Tβ4 to increase and stabilize the vascular plexus via collateral vesselgrowth. As such, we identify Tβ4 as a facilitator of cardiac neovascularization and highlight adult EPDCs as residentprogenitors which, when instructed by Tβ4, have the capacity to sustain the myocardium after ischemic damage.

Keywords: thymosin; heart; epicardium; EPDCs; myocardial infarction; neovascularization; regeneration

Introduction

Individuals surviving acute myocardial infarction(MI) are still at risk of death from heart failureand dysrhythmia. This occurs because necrotic my-ocardium is replaced by noncontractile, poorly con-ducting, fibrotic scar tissue, although spared car-diac muscle undergoes hypertrophy in an attemptto restore overall contractile function.1 Ultimately,these processes lead to pathological remodelingof the heart characterized by dilatation, impairedejection of blood, and a propensity to sustainedmalignant electrical rhythms;2 the clinical manifes-tations of which include progressive heart failure,sudden death, or cardiac rupture. Approaches toprevent or mitigate the effects of remodeling post-MI have focused on replacement of damaged my-ocardium with healthy myocytes and the induction

of vessel formation to sustain both new and re-tained cardiac muscle. To this end significant efforthas been invested in cell transplantation strategieswith autologous bone marrow derived stem cells(reviewed in Ref. 3) and in the search for embry-onic4–6 or adult cardiac progenitor cells7 which mayreplace damaged muscle and/or contribute to neo-vascularization. Key to success of the latter is theidentification of factors that may induce endoge-nous progenitor cells to initiate myocardial repairand collateral vessel growth.

Previously, ectopic administration of thymosin�4 (T�4) in a mouse model of MI was shownto reduce scarring and improve cardiac functionvia Akt-induced cardiomyocyte survival.8 We re-cently revealed a distinct mechanism for T�4 ac-tion via the induction of adult epicardium-derivedcells (EPDCs) ex vivo to form vascular precursors

doi: 10.1111/j.1749-6632.2010.05478.xAnn. N.Y. Acad. Sci. 1194 (2010) 97–104 c© 2010 New York Academy of Sciences. 97

Tβ4 and epicardial neovascularization Smart et al.

with the potential for neovascularization.9 However,neither of these studies addressed an in vivo role forT�4 in the key cellular events required to repairand/or regenerate the injured heart, and our earlierstudy fell short of realizing the lineage potential ofadult epicardium-derived progenitors in the heartproper.9

Here we demonstrate in a mouse MI model thatT�4 activates adult EPDCs to promote a robust andstable vascular response to injury. The stimulatoryeffect of T�4 post-MI occurs against a backgroundof a suboptimal EPDC response such that, in theabsence of T�4 treatment, EPDCs contribute anendothelial-cell restricted capillary plexus, whichlacks smooth muscle support. Hence the ability ofT�4 to induce collateral growth via EPDCs ensuresvascular stability and prolonged neovascularizationto sustain the surviving myocardium postinjury.

Results

Tβ4 induces neovascularization in the intactadult heartThe minimum requirement for T�4 to promoteEPDC-derived vascular endothelial and smoothmuscle cells ex vivo, was reported previously.9 To in-vestigate whether T�4 can stimulate bona fide newvessel growth in vivo we first examined hearts from again of function (intact heart) mouse model, estab-lished by intraperitoneal injection of T�4 (150 �gin 0.1 mL PBS) or vehicle (0.1 mL PBS) into wild-type adult mice (males 25–30 g) every 2 days for upto 1 week or every 3 days for up to 4 weeks.

Hearts were assessed using a combination ofWestern and immunofluorescence analyses for vas-cular markers and evidence of new coronary arteriesat 2, 4, 7, 14, and 28 days posttreatment. Endothe-lial markers Tie2 (2.0-fold), PECAM (9.3-fold), andVEGF (4.8-fold) and the smooth muscle markerSM�A (9.8-fold) were significantly increased fol-lowing 2 days of T�4 treatment compared to con-trols (Fig. 1A; Fig. S1A and S1B). The T�4-inducedincrease in vascular markers persisted throughoutthe duration of the experiment and was accom-panied by a significant increase in proliferation asdetermined by elevated levels (9.1-fold) of phospho-histone H3 (P-HH3; Figs. 1A and S1B). Immunoflu-orescence on T�4-treated hearts showed strong re-gionalized staining for PECAM and SM�A, in anexpanded subepicardial space and immediate un-

derlying myocardium, with an increase in the num-ber of coronary vessels after 28 days as indicatedby endothelial cell lined arterioles (Fig. 1B and 1C)surrounded by smooth muscle (Fig. 1D and 1E).The vasculature observed following T�4 treatmentis referred to as newly formed based on the factthat neither expansion of the subepicardium nor anassociated vascular network, were observed in con-trol hearts. Instead this region in controls was en-tirely superficial with existing coronary vessels char-acteristically located deep within the myocardium(Fig. 1C and 1E).

These studies suggest that not only can T�4 pro-mote neovascularization in vivo but that this re-sponse can occur in the absence of injury.

Neovascularization is optimized in an injurysettingWe next investigated whether T�4 could promoteneovascularization in an injury model of MI (bycoronary artery ligation) in adult mice (n = 27 MIsin total), treated with either T�4 (n = 13) or vehicle(n = 14; injection regimen as for gain of function)and whether this may be enhanced as comparedto the intact gain of function model. T�4-treated,infarcted hearts revealed an increase in cell prolifer-ation in the epicardium and subepicardial space de-termined by the presence of Ki67+ cells after 7 dayspostinfarct (Fig. S2A–S2C). The elevated cell num-ber was accompanied by significantly increasedPECAM and SM�A protein expression (9.6-foldand 8.3-fold increases, respectively, at d7; Figs. 2Aand S1C). The incidence of vascular endothelial andsmooth muscle cells in the subepicardial space ofT�4-treated infarcted hearts by day 7 (Fig. 2B–2E)was increased beyond that observed in intact gainof function hearts at an equivalent stage (data notshown) and comparable to that observed in in-tact hearts after 28 days of T�4 treatment (com-pare Fig. 2B and 2D with Fig. 1B). Injury alone(vehicle-treated post-MI) induced an endogenousendothelial (PECAM+) response (Fig. 2C); how-ever, there was no equivalent response at the level ofsmooth muscle cells (Fig. 2E). Extensive smoothmuscle cell (SM�A+) migration and differentia-tion was only established following treatment withT�4 (Fig. 2D). Smooth muscle collateral growthmay, therefore, explain the beneficial effects of T�4treatment post-MI,8 as compared to the relatively

98 Ann. N.Y. Acad. Sci. 1194 (2010) 97–104 c© 2010 New York Academy of Sciences.

Smart et al. Tβ4 and epicardial neovascularization

Figure 1. T�4 promotes vasculogenesis in the intact adult heart. Western analyses (A) and immunofluorescence(B–E) for vascular markers on adult intact mouse hearts treated with either T�4 or vehicle (co). Tie2, PECAM, andVEGF are upregulated in intact hearts following 2 days of T�4 treatment, accompanied by an increase in levels ofSM�A and P-HH3; Western samples from control and T�4 treated hearts were run on the same gel/marker andseparated for presentation (A). After 28 days of T�4 treatment, PECAM and SM�A positive cells reveal new capillariesdeveloping extensively throughout the subepicardial space and underlying myocardium compared to vehicle-treatedcontrols (highlighted by arrowheads; B–E); white asterisks in (B) and (C) denote background autofluoresence in themyocardium; inset of vessel in (C) illustrates specific PECAM staining of coronary endothelium. Co, control; ep,epicardium; my, myocardium; ses, subepicardial space.

unstable, endothelial-restricted, endogenous re-sponse (Fig. 2C and 2E).

Tβ4-stimulated adult epicardium contributesvascular precursor cells in vivoTo assess whether T�4-induced coronary vesselsmight be epicardial in origin, we examined theincidence of PECAM+ and SM�A+ cells in lin-eage trace hearts, derived from crosses betweenGata5-Cre transgenic mice10 and a R26R-EYFP re-porter strain11 (Gata5-Cre/+; R26R-EYFP/+ mice,hitherto referred to as Gata5-EYFP). A region ofthe Gata5 promoter has previously been shownto preferentially drive Cre expression in the pro-epicardium and epicardial derivatives during de-velopment without affecting myocardial cells12 andthe Gata5-Cre transgenic mice were used success-fully to generate mice with an epicardial dele-tion of �-catenin12 and the PDGF receptor �.13

Here we demonstrate that Gata5-EYFP can act asa lineage trace for EPDCs in the adult heart both

ex vivo (Fig. 3A–3D) and in vivo (Fig. 3E–3J).Specificity of the trace was confirmed by the factthat EYFP+ cells were not present throughout themyocardium, but restricted to a subset of isolatedmyocardial cells (presumably of embryological epi-cardial origin) and vascular derivatives (Fig. 3F and3G). Moreover, the major population of EYFP+ cells(identified using a mouse monoclonal �-EGFP an-tibody which detects EYFP: �-YFP) was not onlylocalized to the outer cell layer overlying the ven-tricles but also contained rare, isolated cells whichcostained for the fetal epicardial markers Tbx18 andWT-1 (Fig. 3H–3J).

In lineage trace control hearts, EPDCs were alsoobserved to mobilize from the epicardium post-MI and migrate into the underlying ventricularmyocardium (Fig. 3I–3L) as evidence of an endoge-nous, albeit inefficient, EPDC-response to injury.In addition, we also detected EYFP+ cells costainedwith procollagen type I as a marker for fibroblasts(Fig. S3A–S3C; including fibroblast cell numbers:

Ann. N.Y. Acad. Sci. 1194 (2010) 97–104 c© 2010 New York Academy of Sciences. 99


Figure 2. T�4 promotes vasculogenesis in the injuredadult heart. PECAM and SM�A protein levels are ele-vated post MI following 2 days of T�4 treatment com-pared to vehicletreated controls, and this increase inboth PECAM and SM�A persists through to day 7post-MI (A). After 7 days of T�4 treatment postin-farct PECAM+ vessels (B) are observed surrounding thescarred myocardium; a comparatively reduced endoge-nous response to MI occurs even in control, as revealed byepicardial expansion and increase in PECAM+ vessels (C;white asterisk highlights autofluoresence background inthe myocardium). A significant number of SM�A+ cellsare observed delaminating from the epicardium and mi-grating into the underlying myocardium (highlightedby white arrow) following T�4 treatment (D) as com-pared with control-treated hearts where relatively few(preexisting) SM�A+ cells are observed highlighted bywhite arrowheads (E). co, control; ep, epicardium; my,myocardium; sc, scar tissue; v, vehicle.

Fig. S3D) consistent with our previous observations,in T�4-stimulated explant cultures, of a restora-tion of embryonic pluripotency to activated adultEPDCs.9

Analysis of T�4-induced coronary vessels inGata5-EYFP lineage trace hearts confirmed a signif-icant mobilization of EPDCs following T�4 treat-ment and the presence of clusters of small prolifera-tive EYFP+ cells (Fig. 4A and 4B), located proximal

to, and in contact with, established PECAM+ andSM�A+ vessels as evidence of an ongoing contri-bution of EPDCs to existing vasculature (Fig. 4Cand 4D). However, EYFP+ cell contribution wasinsufficient to account for the full extent of newvessel formation in the presence of T�4; not onlywere PECAM+ (Fig. 4A–4C) or SM�A+ (Fig. 4D)EPDCs rounded in appearance and lacking ma-ture endothelial or smooth muscle cell morphology,but the vascular plexus formed in the subepicardialspace was devoid of EYFP+ cells (not shown). Thelack of direct adult EPDC contribution to the ma-ture endothelial lineage in vivo is consistent with therestricted potential (smooth muscle, fibroblast, andmyocardial cells) recently attributed to embryologi-cal EPDCs in the developing heart.14 BrdU labelingrevealed an absence of proliferating PECAM+ cellsat the site of native vessels (data not shown), sug-gesting vascular expansion by T�4 was due, at leastin part, to de novo vasculogenesis, consistent withthat observed in the expanded subepicardial region(Fig. 2B).

Collectively these findings suggested optimalT�4-induced neovascularization in the injury set-ting, and were confirmed by a quantitative assess-ment of T�4-induced coronary vasculature for boththe gain of function and injury models. Countsof rhodamine–dextran–perfused vessels in Con-nexin40 (Cx40)-EGFP transgenic mice, which la-bels all coronary arteries,15 revealed a significant1.2-fold increase in numbers of perfused coronaryvessels (smooth muscle-lined arterioles) following28 days of T�4 treatment, compared to controls(70.70 ± 3.62, T�4 vs. 57.40 ± 2.97, control (co);number of perfused vessels per section ± SEM, n =3; 60 fields imaged at five separate comparable levelsthrough the heart, P = 0.008; Fig. 4E). An assess-ment of PECAM+ cells in infarcted hearts revealed ahighly significant, 3.5-fold increase in the number ofendothelial cells in T�4 versus vehicle treated afteronly 4 days (150.33 ± 10.94, T�4 vs. 42.25 ± 5.17,control; n = 3; number of cells per field ± SEM,P = 9.09 × 10−9; Fig. 4F). Image J analyses fur-ther revealed a significant twofold increase in vesselarea after 7 days of T�4 treatment post-MI (10.67 ±0.55% vessel area/field, T�4 vs. 6.69 ± 0.60%, co;n = 7; 6 fields per heart at comparable levels; P =9.45 × 10−6; Fig. 4G). Moreover, in support of theobserved injury-induced endogenous neovascular-ization, the overall increase in vessel/arteriole area



Figure 3. A Gata5-EYFP lineage trace of EPDCs in the adult heart. In Gata5-EYFP adult heart explants, emergingEPDCs are EYFP+ (A); differentiated EPDCs retain EYFP expression, as represented by a SM�A+ vascular smoothmuscle cell (B–D); 2-�m scale bar in panel (D) applies to panels (B–D). Transverse sections of Gata5-EYFP adultmouse heart (E–J), EYFP fluorescence highlights cells residing in the outer epicardial layer and subepicardial space(E), and detects epicardial derivatives residing in the underlying myocardium (highlighted by white arrowheads; F)and coronary vessels (G); 40-�m scale bar in E applies to panel F. Lineage traced cells were stained with a mousemonoclonal �-EGFP antibody which recognizes EYFP (�-YFP) to exclude autofluorescence (H–J) and costained witheither �-Tbx18 (I) or �-WT-1 (J) antibodies to reveal EYFP+ cells specifically in the outer epicardium and isolatedcells coexpressing epicardial markers (highlighted by white arrowheads, I, J); 40-�m scale bar in H also appliesto panels I and J. Following MI, an expansion of EYFP+ EPDCs precedes their delamination from the epicardium(highlighted by white arrows) and migration toward the scar tissue (K); this response occurs even in the absence ofT�4, as illustrated in a control heart (L–N; panels grouped by black box); 20-�m scale bar in panel (N) applies topanels (L–N). cv, coronary vessel; ep, epicardium; EPDCs, epicardium-derived cells; my, myocardium; sc, scar; ses,subepicardial space.



Figure 4. Activated EPDCs associate with and contributeto de novo coronary vasculature. Gata5-EYFP lineagetrace analysis reveals EPDC association with and con-tribution to the coronary vasculature (A–C, grouped byblack box; D). Clusters of small EYFP+ cells (green) wereobserved adjacent to, and in contact with, newly formedPECAM+ vessels (red) and merged panels show EYFP+

cells costained with PECAM (yellow; C; highlighted bywhite arrowheads). SM�A+ cells (green) contributed toEYFP+ vessels (�-EGFP to detect EYFP; staining in red),costaining (yellow) reveals EPDC-origin of a proportionof the smooth muscle cells (D; costained cells highlightedby white arrowheads). Quantitative analysis of the

in vehicle-treated hearts was dependent upon theseverity of injury (4.75% in mild injury vs. 11.38%in severe injury; % vessel area/field) measured asinfarct area/total LV area (mean score of infarct areaover 10 sections at comparable levels; Fig. 4H).

T�4 initiated a significant vascular response in theintact heart, which was further enhanced followinginjury to give rise to a subepicardial network of arte-rioles and functional (perfused) vessels in vivo; thusT�4 acts synergistically with injury-induced vascu-logenic signaling. EPDCs are activated as an endoge-nous response to injury but, although they are ob-served to contribute vascular “progenitors” to newor existing coronary vessels (Fig. 4C and 4D), theydo not account for the full extent of T�4-inducedneovascularization. Further studies are required todetermine whether the epicardium acts as a “niche”for bona fide (non-EPDC) endothelial progenitorcells or whether other lineages in the adult heartmay contribute to the vascular response.

Discussion

We reveal that T�4 induces a significant vasculo-genic response following MI and that adult EPDCsare mobilized to contribute to the new coronaryvessels in vivo. PECAM+ endothelial cells, arisingfrom the epicardium and extending as a networkinto the subepicardial space, are clearly evident in

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−number of rhodamine-conjugated dextran (Dextran-TRITC) -perfused coronary arteries in intact Cx40-EGFPadult hearts (E; lower two fluorescence panels) following28-day treatment with T�4 or vehicle (co), illustrating asignificant increase in stable vessels following T�4 treat-ment ∗P <0.01 (E), PECAM+ cell counts (F) on MI heartsfollowing 4-day treatment with T�4 or vehicle (co) ∗∗P

< 0.001 and vessel area quantification (performed us-ing Image J; G) after 7-day treatment ∗∗∗P < 0.0001 (allP-values calculated by Student’s t-test). Following MI,an endogenous response induces new vessel formation(G; co histogram) which correlates with the extent of in-jury (H); the degree of new vessel formation is illustratedfor a single example of MI at day 7 which resulted inmild injury alongside a single example of a severe injury(H; error bars excluded since n = 1 as representatives ofeach severity index). ca, coronary artery; co, control; cv,coronary vein; ep, epicardium; my, myocardium; sc, scartissue; ses, subepicardial space; v, vehicle.



vehicle-treated hearts as evidence of an injury-induced vascular response. However, we failed toobserve smooth muscle collateral growth in control-injured hearts. By contrast T�4 was not only ableto induce a significant endothelial cell response, butalso associated smooth muscle cell differentiation,with SM�A+ cells intimately associated with thenewly developing coronaries. We propose, there-fore, that collateral growth, following T�4 admin-istration ensures the stability of the de novo vessels,optimizing the neovascularization relative to thatderived from injury alone. The importance of col-lateral growth is reflected in the shortcomings of an-giogenic therapy and clinical trials for patients withinoperable ischemic heart disease, whereby, admin-istration of growth factors, VEGF or FGF, either inthe form of purified protein or gene therapy failed,largely due to their inability to promote smoothmuscle support and arteriogenesis (reviewed in Ref.16). The ability of T�4 to induce new coronary ves-sels, which are both durable and perfused, therefore,is an essential component of cardiac repair giventhe need to sustain resident healthy myocardiumduring compensatory cardiac function and estab-lish a stable vascular plexus to support myocardialregeneration.

The application of T�4-induced neovasculariza-tion and stimulated EPDCs, facilitating survival, re-covery and regenerative replacement of destroyedmyocardium is a significant step toward therapy foracute MI in humans.

Acknowledgments

This research was supported by the British HeartFoundation and Medical Research Council. Weare very grateful to Pilar Ruiz-Lozano and FrankCostantini for generously providing the Gata5-Creand R26R-EYFP mouse strains, respectively. Wethank Jon Epstein for comments on the manuscript.

Supporting Information

Additional Supporting Information may be foundin the online version of this article:

Figure S1. Western blots for vehicle control injec-tions over 2–7 days and scanning densitometry forall Western analyses. Western blots on heart extractsfrom 8-week-old adult mice injected with vehicle(0.1 mL PBS) or T�4 (150 �g in 0.1 mL PBS) af-

ter 2, 4, and 7 days. Western analyses show thatvascular (Tie2, PECAM, VEGF, and SM�A) andproliferation (P-HH3) markers were unchanged invehicle-treated mice over the 7-day time course (A).Equivalent loading was ensured and normalizedagainst GAPDH levels. Western blots showing thetime course of vehicle treatment from 7 to 28 daysand of T�4 treatment from 2 to 28 days are shownin Figures 1 and 2. Scanning densitometry (Image J)as a quantitative representation of all Western blotsnormalized against GAPDH levels. Histograms re-late to Western blots shown in Fig. 1(A) and 1(B)Fig. 2(A) 2(C); gray and black bars indicate controland T�4 treatments, respectively. All scans are ofsingle bands.

Figure S2. T�4 induces significant proliferationof EPDCs. Following MI, an increase in proliferationof cells (determined by Ki67 immunofluorescence)within the epicardium and subepicardial space wasobserved in T�4-treated hearts (A); highlighted bywhite arrowheads and shown at higher resolutionin (B); only one or two isolated Ki67+ nuclei wereever observed in the underlying myocardium (A)or the epicardium/subepicardial space of controlhearts (C). Ki67+ nuclei were detected in EPDCsup to day 7 post-MI, but were not detected at day14 (not shown), indicating that proliferation occursfollowing MI as a component of the initial epicar-dial repair response. Co, control; ep, epicardium;my, myocardium; sc, scar tissue; ses, subepicardialspace.

Figure S3. The epicardium gives rise to fibrob-lasts following MI. Consistent with an activation ofadult EPDCs, EYFP+ cells (A) costained with pro-collagen type I (B; highlighted by white arrowheadsin C), indicatve of fibroblasts, were detected in theexpanded subepicardial space at day 7 post-MI; 20-�m scale bar in panel (A) applies to panels (A–C).Fibroblast number did not vary significantly withT�4 treatment, compared with control: P = 0.57;Student’s t-test (D).

Please note: Wiley-Blackwell are not responsible forthe content or functionality of any supporting ma-terials supplied by the authors. Any queries (otherthan missing material) should be directed to thecorresponding author for the article.

Conflicts of interest

The authors declare no conflicts of interest.



References

1. Poss, K.D. 2007. Getting to the heart of regeneration in

zebrafish. Semin. Cell Dev. Biol. 18: 36–45.

2. Cavasin, M.A. 2006. Therapeutic potential of thymosin-

beta4 and its derivative N-acetyl-seryl-aspartyl-lysyl-

proline (Ac-SDKP) in cardiac healing after infarction.

Am. J. Cardiovasc. Drugs 6: 305–311.

3. Srivastava, D. & K.N. Ivey. 2006. Potential of stem-cell-

based therapies for heart disease. Nature 441: 1097–1099.

4. Moretti, A., L. Caron, A. Nakano, et al. 2006. Multipo-

tent embryonic Isl1(+) progenitor cells lead to cardiac,

smooth muscle, and endothelial cell diversification. Cell

127: 1151–1165.

5. Wu, S.M., Y. Fujiwara, S.M. Cibulsky, et al. 2006. Devel-

opmental origin of a bipotential myocardial and smooth

muscle cell precursor in the mammalian heart. Cell 127:

1137–1150.

6. Prall, O.W., M.K. Menon, M.J. Solloway, et al. 2007.

An nkx2-5/bmp2/smad1 negative feedback loop con-

trols heart progenitor specification and proliferation.

Cell 128: 947–959.

7. Laugwitz, K.L., A. Moretti, J. Lam, et al. 2005. Postnatal

isl1+ cardioblasts enter fully differentiated cardiomy-

ocyte lineages. Nature 433: 647–653.

8. Bock-Marquette, I., A. Saxena, M.D. White, et al. 2004.

Thymosin beta4 activates integrin-linked kinase and

promotes cardiac cell migration, survival and cardiac

repair. Nature 432: 466–472.

9. Smart, N., C.A. Risebro, A.A. Melville, et al. 2007. Thy-

mosin beta4 induces adult epicardial progenitor mobi-

lization and neovascularization. Nature 445: 177–182.

10. Merki, E., M. Zamora, A. Raya, et al. 2005. Epicardial

retinoid X receptor alpha is required for myocardial

growth and coronary artery formation. Proc. Natl. Acad.

Sci. USA 102: 18455–18460.

11. Srinivas, S., T. Watanabe, C.S. Lin, et al. 2001. Cre re-

porter strains produced by targeted insertion of EYFP

and ECFP into the ROSA26 locus BMC. Dev. Biol. 1: 4.

12. Zamora, M., J. Manner & P. Ruiz-Lozano. 2007.

Epicardium-derived progenitor cells require beta-

catenin for coronary artery formation. Proc. Natl. Acad.

Sci. USA 104: 18109–18114.

13. Mellgren, A.M., C.L. Smith, G.S. Olsen, et al. 2008.

Platelet-derived growth factor receptor {beta} signal-

ing is required for efficient epicardial cell migration and

development of two distinct coronary vascular smooth

muscle cell populations. Circ. Res. 103: 1392–1401.

14. Cai, C.L., J.C. Martin, Y. Sun, et al. 2008. A myocardial

lineage derives from Tbx18 epicardial cells. Nature 454:

104–108.

15. Miquerol, L., S. Meysen, M. Mangoni, et al. 2004.

Architectural and functional asymmetry of the His-

Purkinje system of the murine heart. Cardiovasc. Res. 63:

77–86.

16. Tirziu, D. & M. Simons. 2005. Angiogenesis in the human

heart: gene and cell therapy. Angiogenesis 8: 241–251.


thymosin β4 facilitates epicardial neovascularization of the injured adult heart

Documents