The regulation of acetylation is central for the epigenetic control of lineage-specific gene expression and determines cell fate decisions. We provide evidence that the inhibition of histone deacetylases (HDACs) blocks the endothelial differentiation of adult progenitor cells. To define the mechanisms by which HDAC inhibition prevents endothelial differentiation, we determined the expression of homeobox transcription factors and demonstrated that HoxA9 expression is down-regulated by HDAC inhibitors. The causal involvement of HoxA9 in the endothelial differentiation of adult progenitor cells is supported by the finding that HoxA9 overexpression partially rescued the endothelial differentiation blockade induced by HDAC inhibitors. Knockdown and overexpression studies revealed that HoxA9 acts as a master switch to regulate the expression of prototypical endothelial-committed genes such as endothelial nitric oxide synthase, VEGF-R2, and VE-cadherin, and mediates the shear stress–induced maturation of endothelial cells. Consistently, HoxA9-deficient mice exhibited lower numbers of endothelial progenitor cells and showed an impaired postnatal neovascularization capacity after the induction of ischemia. Thus, HoxA9 is regulated by HDACs and is critical for postnatal neovascularization.

Endothelial progenitor cells (EPCs) can originate from bone marrow–derived progenitor cells, circulate with the blood on mobilization from the bone marrow, and home to sites of active vessel growth, a process termed “adult vasculogenesis” (1, 2). The recruitment of EPCs is involved in tumor vascularization (3, 4) and contributes to ischemia-triggered neovascularization (58). Although the exact characterization of the EPCs is not entirely clear, various studies suggest that EPC develop from common endothelial and hematopoietic precursor cells, so-called adult hemangioblasts (1, 2, 9). The molecular mechanisms directing endothelial differentiation from stem or progenitor cells, however, are incompletely understood.

As a vehicle for modulating gene expression, chromatin structure remodeling plays a central role in normal development, the physiological differentiation of cells, and both embryonic and adult stem cell functions (10, 11). Indeed, the acetylation of histones is part of the complex epigenetic regulatory process determining lineage-specific gene expression and cell fate decisions by altering the local structure of chromatin (12). Previous reports suggest that the global deacetylation of histones is necessary for in vitro differentiation of embryonic stem (ES) cells (13) and oligodendrocyte lineage progression (14). The interplay between histone acetyltransferases and histone deacetylases (HDACs) is a key regulator in the dynamics of chromatin structure and function. The family of HDACs comprises at least 17 genes that are classified into three groups. Class I HDACs (HDAC1, HDAC2, HDAC3, and HDAC8) and class II HDACs (HDAC4, HDAC5, HDAC6, HDAC7, HDAC9, and HDAC10) can deacetylate histone tails and target other cellular proteins (15). Class III HDACs (Sirtuins) were identified on the basis of their similarity with sir2, a yeast transcription repressor requiring NAD+ as a cofactor (16). Importantly, the inhibition of HDAC was shown not only to block postnatal vessel growth in an animal model of tumor vascularization (17), but also to down-regulate the endothelial nitric oxide synthase (eNOS; reference 18). Thus, we hypothesized that HDAC activity may be required for endothelial differentiation of progenitor cells. Because HDAC inhibitors blocked endothelial differentiation, we further explored the down-stream mechanisms, thereby focusing on the acetylation-dependent regulation of homeobox genes (Hox's).

Hox's encode transcriptional regulatory proteins, which are characterized by a common 60-amino acid DNA-binding motif and regulate differentiation during embryonic development and tissue morphogenesis (19). Members of the Hox family of homeodomain transcription factors play important roles in the embryonic development of the cardiovascular system and also regulate angiogenesis in the adult organism (for review see reference 20). Several Hox transcription factors (e.g., HoxD3, HoxC6, and HoxC8) modulate the expression of integrins, adhesion molecules, and extracellular matrix proteins in mature endothelial cells (21, 22), whereas HoxB5 appears to be involved in the in vitro differentiation of embryonic precursor cells toward endothelial lineage (23). HoxA9, which is important for myeloid, erythroid, and lymphoid hematopoiesis (24, 25) and stem cell expansion (26), is also particularly essential for the migration and tube-forming capacity of mature endothelial cells (27) and, thus, could serve as a switch toward endothelial commitment during progenitor cell maturation.

Our present data demonstrate that HDAC inhibition abrogates the endothelial differentiation of progenitor cells and reduces the expression of the homeobox transcription factor HoxA9. Knockdown and overexpression studies revealed that HoxA9 is a critical regulator of postnatal neovascularization and acts as a master switch to direct expression of the endothelial-committed genes.

Results

HDAC inhibitors abrogate endothelial differentiation of progenitor cells

To test the involvement of HDAC activity in endothelial lineage progression, we investigated the effects of HDAC inhibitors on EPCs derived from peripheral blood mononuclear cells (MNCs; references 28, 29). Adherent DiI-Ac-LDL–labeled EPCs express the endothelial marker proteins von Willebrand factor (vWF), CD105, and VE-cadherin and bind lectin (Fig. S1; references 28, 29). Four structurally unrelated HDAC inhibitors—butyrate (BuA), MS-275, Trichostatin A (TSA), and valproate—inhibited the generation of EPCs (Fig. 1 

, a–c; and Fig. S2). Moreover, HDAC inhibitors selectively reduced the expression of VEGF-R2 during the culture of MNCs in endothelial differentiation medium, whereas the proportion of CD45+ cells and CD14+ monocytes remained unchanged (Fig. 1 d). Similarly, HDAC inhibitors also prevented the endothelial differentiation of bone marrow–derived human CD34+ hematopoietic progenitor or murine Sca-1+/lin cells, both of which were exposed to the same endothelial differentiation conditions (Fig. 1, e and f). Control experiments confirmed that the reduction in numbers of EPCs was not related to an antiadhesive or proapoptotic effect of the HDAC inhibitors, because integrin expression (Fig. 1 g) and EPC adhesion and apoptosis rates (Figs. S3 and S4, respectively) were not affected. In addition, caspase inhibitor treatment did not reverse the effect of HDAC inhibitors (Fig. 1 c). In summary, these data suggest that HDAC inhibitors interfere with the ex vivo endothelial lineage progression of circulating peripheral blood- or bone marrow–derived progenitor cells.

To investigate the effect of HDAC inhibitors in a physiological model of vasculogenesis and angiogenesis, we performed an ex vivo analysis of new vessel growth in the allantois assay. The HDAC inhibitors MS-275 (3 μM) and BuA (2 mM) profoundly blocked the formation of a vascular network from embryonic progenitor cells (Fig. 1 h and not depicted), indicating that HDAC activity is required for ex vivo vessel growth.

HDAC inhibition down-regulates HoxA9 expression and EPC formation

Because homeobox transcription factors play important roles in the embryonic development of the cardiovascular system and in neovascularization in the adult organism (20, 27), we postulated that the expression of Hox proteins might be altered by HDAC inhibition. BuA or MS-275 time- and dose-dependently reduced mRNA expression (Fig. 2 

a) and protein levels (17 ± 3% of control; Fig. 2 b) of HoxA9. Moreover, the down-regulation of HDAC1 by small interfering RNA (siRNA) reduced HoxA9 expression (Fig. 2 c). In contrast, the homeodomain transcription factor HoxD9 was not regulated (Fig. 2 b) and HDAC inhibitors caused only a minor reduction in mRNA levels of HoxB5 (77 ± 11%; Fig. 2 a), indicating that HoxA9 is rather specifically regulated by HDAC.

HoxA9 is essential for postnatal neovascularization

Next, we determined the role of HoxA9 in postnatal neovascularization using a previously described hind limb ischemia model and measured the recovery of limb blood flow in WT versus heterozygote HoxA9+/− and homozygote HoxA9−/− mice. As depicted in Fig. 3 

a, HoxA9+/− and HoxA9−/− mice showed a severely impaired recovery of blood flow as evidenced by a significantly higher incidence of limb necrosis. Limb perfusion as assessed by laser Doppler 2 wk after the induction of ischemia was significantly reduced in HoxA9−/− as compared with WT mice (26.9 ± 14.3% compared with WT; P < 0.05). Consistently, HoxA9−/− mice exhibited extensive thigh muscle necrosis and a reduced number of arterioles (Fig. 3, b and c), demonstrating that HoxA9 is essential for postnatal neovascularization and blood flow recovery after hind limb ischemia.

HoxA9 contributes to the endothelial commitment of progenitor cells

To test the hypothesis that HoxA9 is important for endothelial commitment, we measured HoxA9 expression during endothelial differentiation of peripheral blood–derived progenitor cells. HoxA9 expression increased during differentiation of peripheral blood–derived progenitors toward an endothelial phenotype (Fig. 4 

a). The increase in HoxA9 expression during endothelial differentiation and maturation was paralleled by an enhanced expression of the endothelial marker proteins eNOS and VEGF-R2 in peripheral blood–derived progenitor cells (reference 30; unpublished data). Additionally, we determined the expression of HoxA9 during the endothelial differentiation of ES cells as a model of in vitro endothelial differentiation. Similar to the finding in peripheral blood–derived adult EPCs, the endothelial differentiation of ES cells was associated with an increased expression of HoxA9 that coincided with the increase in endothelial marker genes such as eNOS and VEGF-R2 (Fig. 4, b and c). Thus, elevated HoxA9 gene expression is associated with an increased endothelial commitment, as determined by the expression of endothelial marker genes.

We transduced blood-derived MNCs with a HoxA9 adenovirus to test whether modulation of HoxA9 expression is capable of regulating the number of EPCs. HoxA9 overexpression increased the yield of EPCs significantly (Fig. 4 d). Moreover, we determined the number of EPCs in homozygote and heterozygote HoxA9-deficient mice. HoxA9−/− and HoxA9+/− mice showed a significantly reduced number of Dil-ac-LDL+ adherent cells and outgrowing EPC colonies (Fig. 4, e and f). These data indicate that HoxA9 plays a critical role in the endothelial commitment of adult progenitor cells.

HoxA9 regulates eNOS, VEGF-R2, and VE-cadherin gene expression

To further substantiate that HoxA9 plays a critical role in the expression of endothelial genes, we determined whether HoxA9 is required for the maintenance of endothelial marker gene expression in human umbilical vein endothelial cells (HUVECs) as a model for mature endothelial cells. The inhibition of HoxA9 expression by siRNA reduced the mRNA expression of the endothelial marker genes eNOS, VEGF-R2, and VE-cadherin in mature endothelial cells (Fig. 5 

, a and b). Moreover, HoxA9 siRNA inhibited eNOS protein expression (Fig. 5 c) and the generation of nitric oxide, as measured by DAF-2DA staining, from 46 ± 6 arbitrary units in scrambled oligonucleotide-treated to 11 ± 3 arbitrary units in siRNA-treated endothelial cells (n = 4, P < 0.001). HoxA9-deficient mice consistently showed a reduced eNOS protein expression in the heart (Fig. 5 d). In contrast, overexpression of HoxA9 enhanced the expression of eNOS (Fig. 5 e). These data demonstrate that HoxA9 regulates the expression of endothelial marker genes in vitro and in vivo.

To assess whether HoxA9 interacts with the eNOS, VEGF-R2 and VE-cadherin promoters, we performed chromatin immunoprecipitation assays. After cross-linking, immunoprecipitates of endogenous HoxA9 were subjected to PCR using primers directed against the eNOS, VEGF-R2, and VE-cadherin promoters. Endogenous HoxA9 bound to the eNOS and the VEGF-R2 promoters (Fig. 6 

a), whereas HoxA9 did not interact with the VE-cadherin promoter, at least not within the specific region −1163 to −380 bp upstream of ATG investigated (Fig. 6 a). To test the specificity of the binding of HoxA9 to the eNOS and VEGF-R2 promoters, HUVECs were transiently transfected with WT HoxA9 or a HoxA9 mutant (mt) lacking the DNA-binding domain (Δ206-272 [reference 27]; Fig. 6 b). Overexpressed WT HoxA9 bound to the eNOS and VEGF-R2 promoters, whereas the HoxA9 mt showed no binding (Fig. 6 c), confirming the specificity of the interaction. To determine the promoter activation by HoxA9, we used reporter gene constructs driven by the eNOS, VEGF-R2, or VE-cadherin promoters. HoxA9 significantly increased eNOS, VEGF-R2, and VE-cadherin promoter activation (Fig. 6, d–f). Thus, HoxA9 regulates the transcription of prototypic endothelial marker genes in endothelial cells.

Physiological levels of shear stress induce endothelial maturation of progenitor cells in a HoxA9-dependent manner

EPCs not only home to sites of ischemia, but also to denuded arteries, thereby promoting the recovery of the endothelial monolayer after a denuding injury. Immediately after attachment, progenitor cells are exposed to laminar blood flow (3133). Because laminar blood flow increases the expression of various endothelial genes such as eNOS (for review see reference 34) and VEGF-R2 (35, 36), we hypothesized that shear stress may enhance the endothelial maturation of progenitor cells by the up-regulation of HoxA9. Indeed, shear stress stimulated the expression of HoxA9 in a time- and dose-dependent manner (Fig. 7 

a). Moreover, the exposure of peripheral blood MNC–derived EPCs to shear stress increased the expression of VEGF-R2 (Fig. 7 b), eNOS, and the expression of a VEGF-R2 promoter–driven reporter gene (unpublished data). Shear stress also enhanced VEGF-R2 expression in isolated CD34+ hematopoietic progenitor cells (Fig. 7 b). Concomitantly, the expression of the pan-leukocyte marker protein CD45 decreased (unpublished data), which was consistent with a maturation toward the endothelial lineage. In contrast, when macrophages were exposed to shear stress, no increase in VEGF-R2 expression was noted (Fig. 7 b), indicating that the capacity of shear stress to induce endothelial lineage marker expression is restricted to progenitor cells and mature endothelial cells. Next, we determined whether the shear stress–induced expression of endothelial-committed genes is indeed mediated via HoxA9. A blockade of HoxA9 gene expression by siRNA abrogated the shear stress–stimulated increase in eNOS and VEGF-R2 expression (Fig. 7 c), documenting that shear stress requires HoxA9 to up-regulate endothelial marker genes.

HoxA9 overexpression partially rescues the HDAC inhibitor–mediated blockade of endothelial differentiation

To finally determine whether the HDAC inhibitor–mediated down-regulation of HoxA9 is indeed causally involved in the reduction of endothelial commitment and down-regulation of endothelial marker genes by HDAC inhibitor treatment, we overexpressed HoxA9 by adenoviral transduction. Indeed, HoxA9 overexpression partially reversed the down-regulation of the number of EPCs by HDAC inhibition (Fig. 8 

a). To determine whether the HDAC inhibitor–induced down-regulation of eNOS (18) is mediated by transcriptional repression of HoxA9, we overexpressed HoxA9 before treatment of the cells with HDAC inhibitors. Indeed, the profound down-regulation of eNOS protein expression by HDAC inhibitors was prevented by overexpression of HoxA9 (Fig. 8 b) suggesting that the transcriptional repression of HoxA9 by HDAC inhibitors may contribute to eNOS down-regulation and inhibition of endothelial commitment.

Discussion

This study demonstrates that the endothelial lineage commitment of circulating blood- or bone marrow–derived progenitor cells requires HDAC activity. The inhibition of HDACs decreased the expression of the transcription factor HoxA9 and reduced the number of endothelial cells derived from different progenitor cell sources. HoxA9 regulates various typical endothelial marker proteins, which are also important for the functional activity of endothelial cells. HoxA9 deficiency reduces endothelial lineage commitment and results in severe impairment of neovascularization, demonstrating an important role of HoxA9 in postnatal neovascularization.

HDACs comprise at least 17 genes, of which HDAC1, HDAC3, and SIRT1 are expressed in human peripheral blood–derived EPCs (unpublished data). The pharmacological inhibition of class I and II HDACs by structurally different pharmacological HDAC inhibitors abrogated HoxA9 expression and the endothelial commitment of progenitor cells. Furthermore, a specific down-regulation of HDAC1 by siRNA reduced the expression of HoxA9. In contrast, HoxD9 and HoxB5 were not significantly regulated by HDAC inhibitors, implicating a specific dependency of HoxA9 transcription on HDAC activity. HDACs are a component of the ALL-1 supercomplex, which binds to the HoxA9 promoter and is required for HoxA9 transcription (37). Thus, one may speculate that HDAC activity is necessary for the ALL-1 supercomplex to allow for transcription of HoxA9 (37).

Our data demonstrate that HoxA9 is required for postnatal neovascularization after ischemia. Previous studies additionally documented that the inhibition of HDAC blocks tumor angiogenesis (17). However, the importance of HDAC and HoxA9 for the embryonic development of vascular structures in vivo is unclear. In fact, several lines of evidence indicate that HDAC1 and HoxA9 are not essential for embryonic vessel formation. In contrast to VEGF gene deficiency, which is lethal because of impaired vessel formation even at the level of haploinsufficiency (38), or VEGF-R2 knockout mice, which die between embryonic days 8.5 and 9.5 from the lack of any vascular formation (39), no abnormalities with respect to embryonic vascularization were obvious in HDAC1−/− or HoxA9−/− mice (40, 41). In our experiments, however, HDAC inhibitors abrogated embryonic angiogenesis and vasculogenesis in the allantois ex vivo assay. This discrepancy might well be rationalized by the broad spectrum inhibitory effect of the pharmacological HDAC inhibitors used in the present study and may indicate that the specific lack of HDAC1 might be compensated for by the various other members of the HDAC family in vivo. Likewise, the lack of a severe embryonic phenotype of HoxA9−/− mice might also be caused by compensation by other members of the Hox family such as HoxB5 during embryonic development in vivo (23). However, our data clearly indicate that HoxA9 is required for adult vasculogenesis. Indeed, HoxA9−/− and HoxA9+/− mice showed a severe impairment of endothelial colony formation, blood flow recovery, and functional regeneration after the induction of ischemia. Thus, one may speculate that mechanisms regulating postnatal neovascularization are not necessarily identical to the mechanisms operational during embryonic vascular development.

Why is HoxA9 important for postnatal neovascularization? In addition to the requirement of HoxA9 for endothelial progenitor colony formation and adult vasculogenesis, our data indicate that HoxA9 directly regulates a variety of key endothelial genes that are involved in the functional maturation and activity of endothelial cells. All three HoxA9 target genes, eNOS (42, 43), VEGF-R2 (39), and VE-cadherin (44), are crucial for angiogenesis. Moreover, the reduction of HoxA9 additionally results in the reduced expression of EphB4 and, thus, inhibits endothelial migration in vitro (27). The reduction of these HoxA9 regulated genes in vivo (as shown for the eNOS) may contribute to the severe neovascularization defect of HoxA9-deficient mice. In addition, HoxA9 regulates myeloid and lymphoid hematopoiesis (25) and, therefore, may affect inflammation-mediated angiogenesis. However, a reduction of myeloid cells was detected in homozygote, but not in heterozygote, mice (25), whereas in our study heterozygote mice showed a severe impairment of neovascularization and endothelial colony forming activity. Furthermore, the overall peripheral white blood cell counts after the induction of ischemia were not notably different between HoxA9-deficient mice and WT littermates (unpublished data). Thus, the impaired neovascularization capacity of HoxA9−/− and HoxA9+/− mice is most likely not related to a general imbalance of inflammatory cells.

Finally, our data provide novel insights into the prototypic physiological mechanism inducing the maturation of endothelial cells. Shear stress is not only one of the most powerful antiatherosclerotic factors, but is also an absolute prerequisite for a functionally intact endothelial monolayer. We demonstrate that the up-regulation of the endothelial signature gene pattern is dependent on HoxA9. Not only did shear stress increase the expression of endothelial marker proteins such as eNOS and VEGF-R2, but it also concomitantly down-regulated the pan-leukocyte marker CD45, which is expressed on hematopoietic progenitor cells. These data clearly indicate that shear stress is capable of promoting the commitment of progenitor cells to an endothelial phenotype. The stimulation of HoxA9 expression and subsequent endothelial maturation by shear stress may considerably contribute to a timely recovery of the endothelial monolayer after injury to prevent atherosclerotic lesion development and restenosis formation, respectively (32, 33, 45, 46).

This study demonstrates that the HDAC-dependent transcription of HoxA9 is necessary for ex vivo maturation of progenitor cells toward the endothelial lineage. These findings not only contribute to a better understanding of postnatal endothelial maturation, but, given the pivotal role of EPCs for neovascularization of ischemic tissue and reendothelialization, may also provide important therapeutic targets. Of note, the inhibition of adult vasculogenesis by HDAC inhibitors may also contribute to the reported profound antitumor activity of HDAC inhibitors (15).

Materials And Methods

Isolation and ex vivo endothelial differentiation of EPCs

Ex vivo EPC differentiation from circulating MNCs was assayed as described previously (28). In brief, MNCs were isolated by density gradient centrifugation with Biocoll separating solution (density 1.077; Biochrom AG) from human peripheral blood buffy coats. Immediately after isolation, 8 × 106 MNC/ml of medium was plated on culture dishes coated with human fibronectin and maintained in endothelial basal medium (EBM; Cambrex) supplemented with EGM SingleQuots (Cambrex) and 20% FCS (GIBCO BRL). After 3 d in culture, nonadherent cells were removed by thorough washing with PBS. Adherent cells were stained with 2.4 μg/ml Dil-Ac-LDL (Harbor Bio-Products) at 37°C for 1 h and fixed with 4% paraformaldehyde for 10 min. Purified human CD34+ BM MNCs (BMCs) were purchased from Cambrex and were differentiated to EPCs under the same conditions as described above.

For the preparation of Sca-1+/lin BMC-derived EPCs, murine BM was isolated from the hindlimbs of 6–8-wk-old female C57BL/6 mice. Then, cells were filtrated using a 40-μm pore size cell strainer. After blocking with purified anti–mouse CD16/CD32 (FcγIII/II receptor) antibodies (1:100; BD Biosciences), washed cells were incubated with a biotinylated anti–lineage marker antibody cocktail (Becton Dickinson) and antibiotin microbeads (Miltenyi Biotec), and isolated with an automated magnetic cell sorting device (autoMACS; Miltenyi Biotec). Separated lin BMCs were incubated with microbeads directly conjugated to anti–Sca-1 antibodies, and Sca-1+/lin BMCs were isolated by a second run through the autoMACS. To assess the capacities of these immature BM stem cells, 0.5–1.0 × 106 pooled Sca-1+/lin BMCs were plated on 24-well culture dishes coated with human fibronectin and maintained in EBM supplemented with EGM SingleQuots and 20% FCS. After 3 d in culture, adherent Sca-1+/lin BMCs were stained with Dil-Ac-LDL and were fixed. Then, cells were stained for vWF (Acris Antibodies) and the nuclear marker TO-PRO-3 iodide (Molecular Probes), or against 10 μg/ml lectin by incubating with FITC-labeled Ulex europaeus agglutinin I (Sigma-Aldrich) for 1 h. Staining for both vWF or lectin, and Dil-Ac-LDL, was evaluated by confocal microscopy.

For the isolation of spleen-derived EPCs, murine MNCs were isolated from homogenized splenic tissue derived from HoxA9−/− (provided by H. Jeffrey Lawrence, University of California, San Francisco, Veteran's Administration Medical Center, San Francisco, CA) or WT littermates by density gradient centrifugation with Biocoll separating solution. 4 × 106 MNCs were plated on fibronectin-coated 24-well plates in 0.5 ml EBM supplemented with EGM SingleQuots and 20% FCS and were stained as described above. Outgrowing colonies were detected after 10 d.

Cell culture

Pooled HUVECs were purchased from CellSystems and cultured as previously described (27). HUVECs were exposed to laminar flow in a cone-and-plate apparatus as previously described (47). For shear stress experiments, human CD34+ hematopoietic progenitor cells were purified from MNCs by positive selection with anti-CD34 microbeads (Miltenyi Biotec; reference 28). For the cultivation of macrophages, CD14+ monocytes were purified from MNCs by positive selection with anti-CD14 microbeads (Miltenyi Biotec) using a magnetic cell sorter (Miltenyi Biotec). Purity assessed by FACS analysis was >95%. CD14+ monocytes were incubated in RPMI 1640 with 10% FCS in the presence of 50 ng/ml M-CSF to induce macrophage differentiation (29).

Allantois assay

Allantoides were isolated as previously described (48). In brief, embryos at 8.5 d postcoitum were dissected in PBS at 4°C, and the allantoides were excised and seeded into four-chambered culture slides (Nalgene Nunc International) containing 0.4 ml DMEM with 10% FCS, 2 mmol/liter L-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin (all reagents obtained from Life Technologies). Explants were cultured at 37°C in a 5% CO2 incubator for 18–20 h in the presence or absence of the HDAC inhibitors. The allantois cultures were fixed for 20 min at 25°C with 4% paraformaldehyde, washed twice in PBS, permeabilized for 15 min at 25°C with 0.02% Triton X-100 in PBS, blocked for 1 h at 25°C with 3% BSA (Sigma-Aldrich) in PBS, and stained with PECAM-1 antibody (MEC13.3; reference 49).

Western blot analysis

Cells were lysed with buffer (20 mmol/liter Tris, pH 7.4, 150 mmol/liter NaCl, 1 mmol/liter EDTA, 1 mmol/liter EGTA, 1% Triton X-100, 2.5 mmol/liter sodium pyrophosphate, 1 mmol/liter β-glycerophosphate, 1 mmol/liter Na3VO4, 1 μg/ml leupeptin, 1 mmol/liter PMSF) for 15 min on ice. After centrifuging at 20,000 g for 15 min at 4°C, protein content was measured according to the Bradford method. Homogenates (40 μg per lane) were separated on SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes, which were then incubated with antibodies to p21 (BD Biosciences), HDAC1 (Abcam), HoxA9, and HoxD9 (Santa Cruz Biotechnology, Inc.), K9-K14-diacetylated histone H3 (Upstate Biotechnology), total histone H3 (Cell Signaling), extracellular signal–related kinase 1/2 (Cell Signaling), eNOS (BD Biosciences), or tubulin (NeoMarkers).

RT-PCR

Total RNA was isolated and either subjected to conventional RT-PCR or quantitative RT-PCR using the oligonucleotide primers summarized in Tables S1 and S2 . Quantification of mRNA was performed in a one-step RT-PCR reaction using the LightCycler (Roche Diagnostics) real-time thermocycler according to the manufacturer's instructions. Amplification was performed with 40 cycles at an annealing temperature of 61°C. Copy numbers were calculated by the instrument software (Roche Diagnostics) from standard curves of an in vitro–transcribed IL-10 cytokine primer mix (Light Cycler control kit RNA; Roche Diagnostics). The specificity of the amplification reaction was determined by a melting curve analysis.

Generation of recombinant adenovirus and adenoviral infection

HoxA9 (transcript variant 1) was amplified by RT-PCR and cloned into a shuttle pAd Track-CMV vector. This plasmid was linearized by digesting with restriction endonuclease PmeI and subsequently cotransformed into Escherichia coli BJ5183 cells with an adenoviral backbone pAdEasy-1 plasmid (all plasmids and E. coli cells were a gift from B. Vogelstein, Howard Hughes Medical Institute, Johns Hopkins Medical Institutions, Baltimore, MD). Recombinants were selected by kanamycin resistance. Finally, recombinants were transfected into HEK293 cells. Recombinant adenoviruses were generated within 7–10 d. The lacZ gene codes for the enzyme β-galactosidase. The lacZ adenovirus was used as control.

Peripheral blood MNCs (4 × 106 cells/1-cm well) were resuspended in 2.5 ml RPMI 1640 (GIBCO BRL) with 10% FCS and preincubated for 30 min with a mixture of adenovirus, 10 μl Antennapedia peptide (RQIKIWFQNRRMKWKK; 2.5 mM; Biosyntan), and 100 μl Optimem (Life Technologies). After 24 h, 4 × 106 cells were resuspended in 1 ml EBM supplemented with EGM SingleQuots and 20% FCS and plated on fibronectin-coated wells.

Endothelial differentiation of ES cells

CJ7 ES cells, a 129/Sv-derived cell line, were cultivated as previously described (50). To initiate ES cell differentiation and embryonic body formation, ES cells were trypsinized and suspended in IMDM (Life Technologies) with 15% FBS, 10 μg/ml insulin (Sigma-Aldrich), 100 U/ml penicillin, 100 μg/ml streptomycin, 450 μmol/liter monothioglycerol, and endothelial differentiation promoting growth factors including 50 ng/ml recombinant human VEGF (PeproTech), 2 U/ml recombinant human erythropoietin (Cilag AG), 100 ng/ml human basic fibroblast growth factor (Genzyme), and 10 ng/ml murine interleukin 6 (Genzyme). After 7 d, ES cell–derived endothelial cells were collected by anti-CD31 immunomagnetic selection (51).

Reporter gene assay

Reporter gene constructs were previously described (1.6-kb human eNOS promoter fragment [reference 52]; 4-kb human VEGF-R2 promoter [reference 36]). The VE-cadherin promoter (3,032-kb fragment: −2928/+104) was cloned into KpnI–XhoI restriction sites in pGl3 enhancer plasmid (Promega). 3.5 × 105 HUVECs were transiently transfected with 3 μg plasmid DNA using 18 μl Superfect (QIAGEN) as previously described (47). After incubation, cells were lysed with lysis buffer (Promega), and luciferase activity was measured using the Luciferase System (Promega) with a luminometer (model Luminat LB 9501; Berthold).

Murine ischemic hind limb model

The effect of HoxA9 on ischemia-induced neovascularization was investigated in a murine model of hind limb ischemia. The present study was performed with the permission of the State of Hesse (Regierungspräsidium Darmstadt), according to section 8 of the German Law for the Protection of Animals, and conforms to the Guide for the Care and Use of Laboratory Animals measurements. In brief, the proximal portion of the femoral artery, including the superficial and the deep branch, as well as the distal portion of the saphenous artery, was ligated and all arterial branches between the ligations were obliterated using an electrical coagulator. The overlying skin was closed using three surgical staples. 2 wk later, we determined the morphology of the limb and measured the ischemic (right) to normal (left) limb blood flow ratio using a laser Doppler blood flow meter (model moorLDI-Mark 2; Moor Instruments). Before initiating scanning, mice were placed on a heating pad at 37°C to minimize variations in temperature. After the recording of complete scan laser Doppler color images, the perfusion of the ischemic and nonischemic limb was calculated on the basis of colored histogram pixels. To minimize variables, including ambient light and temperature, calculated perfusion was expressed as the ratio of ischemic to nonischemic hind limb perfusion.

For morphological analysis, 8-μm frozen sections of the adductor and semimembranous muscles were used. Myocyte membranes were stained using an antibody to laminin (rabbit) followed by anti–rabbit-Alexa 488. Conductance vessels in the adductor and semimembranous muscles were identified by size (>20 μm) and staining using a Cy3 labeled mouse monoclonal antibody for smooth muscle actin (Sigma-Aldrich).

Statistics

Data are expressed as mean ± SEM or as indicated in the figure legends. Two treatment groups were compared with the independent samples t test, and three or more groups by one-way analysis of variance followed by post-hoc analysis adjusted with a least significant difference correction for multiple comparisons (SPSS Inc.). Results were considered statistically significant when P < 0.05.

Online supplemental material

Fig. S1 shows a representative flow cytometric characterization of endothelial marker expression in adherent human peripheral blood MNC–derived EPCs. Fig. S2 presents a comparison of the effect of structural analoga without HDAC inhibitory capacity (2 mM acetate and 1 mM valpromide) and with HDAC inhibitory capacity (2 mM BuA and 1 mM valproate) on the ex vivo formation of EPCs from human peripheral blood MNCs during 72 h of incubation under endothelial differentiation conditions. Fig. S3 shows cell matrix adhesion. Fig. S4 presents the effect of a 72-h incubation of peripheral blood MNCs with 2 mM BuA, 10 μM MS-275, or 1 mM valproate (VPA) under endothelial differentiation conditions on apoptosis. Table S1 summarizes the sequences of the oligonucleotide primer pairs (forward and reverse) used for RT-PCR. Table S2 summarizes the sequences of the oligonucleotide primer pairs (forward and reverse) used for quantitative RT-PCR (Light Cycler). Online supplemental material is available.

Acknowledgments

We are grateful to Rebeca Salguero Palacios, Melanie Näher, Iris Stügelmaier, Andrea Knau, Marion Muhly-Reinholz, Tino Röxe, and Dorit Lüthje for expert technical assistance.

This work was supported by grants to S. Dimmeler from the Deutsche Forschungsgemeinschaft (Di600/2-5 and FOR501[Di600/6-1]) and to L. Rössig from the Deutsche Krebshilfe (106177). This work emanates from the European Vascular Genomics Network, a Network of Excellence supported by the European Union's sixth Framework Programme for Research Priority 1, “Life sciences, genomics, and biotechnology for health” (contract LSHM-CT-2003-503254). K.-i. Sasaki was in part supported by the Japan Heart Foundation/Bayer Yakuhin Research Grant Abroad.

The authors have no conflicting financial interests.

References

References
1
Carmeliet, P.
2000
. Mechanisms of angiogenesis and arteriogenesis.
Nat. Med.
6
:
389
–395.
2
Shi, Q., S. Rafii, M.H. Wu, E.S. Wijelath, C. Yu, A. Ishida, Y. Fujita, S. Kothari, R. Mohle, L.R. Sauvage, et al.
1998
. Evidence for circulating bone marrow-derived endothelial cells.
Blood.
92
:
362
–367.
3
Lyden, D., K. Hattori, S. Dias, C. Costa, P. Blaikie, L. Butros, A. Chadburn, B. Heissig, W. Marks, L. Witte, et al.
2001
. Impaired recruitment of bone-marrow-derived endothelial and hematopoietic precursor cells blocks tumor angiogenesis and growth.
Nat. Med.
7
:
1194
–1201.
4
Stoll, B.R., C. Migliorini, A. Kadambi, L.L. Munn, and R.K. Jain.
2003
. A mathematical model of the contribution of endothelial progenitor cells to angiogenesis in tumors: implications for antiangiogenic therapy.
Blood.
102
:
2555
–2561.
5
Asahara, T., H. Masuda, T. Takahashi, C. Kalka, C. Pastore, M. Silver, M. Kearne, M. Magner, and J.M. Isner.
1999
. Bone marrow origin of endothelial progenitor cells responsible for postnatal vasculogenesis in physiological and pathological neovascularization.
Circ. Res.
85
:
221
–228.
6
Kalka, C., H. Masuda, T. Takahashi, W.M. Kalka-Moll, M. Silver, M. Kearney, T. Li, J.M. Isner, and T. Asahara.
2000
. Transplantation of ex vivo expanded endothelial progenitor cells for therapeutic neovascularization.
Proc. Natl. Acad. Sci. USA.
97
:
3422
–3427.
7
Murohara, T., H. Ikeda, J. Duan, S. Shintani, K. Sasaki, H. Eguchi, I. Onitsuka, K. Matsui, and T. Imaizumi.
2000
. Transplanted cord blood-derived endothelial precursor cells augment postnatal neovascularization.
J. Clin. Invest.
105
:
1527
–1536.
8
Takahashi, T., C. Kalka, H. Masuda, D. Chen, M. Silver, M. Kearney, M. Magner, J.M. Isner, and T. Asahara.
1999
. Ischemia- and cytokine-induced mobilization of bone marrow-derived endothelial progenitor cells for neovascularization.
Nat. Med.
5
:
434
–438.
9
Choi, K., M. Kennedy, A. Kazarov, J.C. Papadimitriou, and G. Keller.
1998
. A common precursor for hematopoietic and endothelial cells.
Development.
125
:
725
–732.
10
Cerny, J., and P.J. Quesenberry.
2004
. Chromatin remodeling and stem cell theory of relativity.
J. Cell. Physiol.
201
:
1
–16.
11
Valk-Lingbeek, M.E., S.W. Bruggeman, and M. van Lohuizen.
2004
. Stem cells and cancer; the polycomb connection.
Cell.
118
:
409
–418.
12
Jenuwein, T., and C.D. Allis.
2001
. Translating the histone code.
Science.
293
:
1074
–1080.
13
Lee, J.H., S.R. Hart, and D.G. Skalnik.
2004
. Histone deacetylase activity is required for embryonic stem cell differentiation.
Genesis.
38
:
32
–38.
14
Marin-Husstege, M., M. Muggironi, A. Liu, and P. Casaccia-Bonnefil.
2002
. Histone deacetylase activity is necessary for oligodendrocyte lineage progression.
J. Neurosci.
22
:
10333
–10345.
15
Johnstone, R.W.
2002
. Histone-deacetylase inhibitors: novel drugs for the treatment of cancer.
Nat. Rev. Drug Discov.
1
:
287
–299.
16
Imai, S., C.M. Armstrong, M. Kaeberlein, and L. Guarente.
2000
. Transcriptional silencing and longevity protein Sir2 is an NAD-dependent histone deacetylase.
Nature.
403
:
795
–800.
17
Kim, M.S., H.J. Kwon, Y.M. Lee, J.H. Baek, J.E. Jang, S.W. Lee, E.J. Moon, H.S. Kim, S.K. Lee, H.Y. Chung, et al.
2001
. Histone deacetylases induce angiogenesis by negative regulation of tumor suppressor genes.
Nat. Med.
7
:
437
–443.
18
Rössig, L., H. Li, B. Fisslthaler, C. Urbich, I. Fleming, U. Forstermann, A.M. Zeiher, and S. Dimmeler.
2002
. Inhibitors of histone deacetylation downregulate the expression of endothelial nitric oxide synthase and compromise endothelial cell function in vasorelaxation and angiogenesis.
Circ. Res.
91
:
837
–844.
19
Abate-Shen, C.
2002
. Deregulated homeobox gene expression in cancer: cause or consequence?
Nat. Rev. Cancer.
2
:
777
–785.
20
Gorski, D.H., and K. Walsh.
2000
. The role of homeobox genes in vascular remodeling and angiogenesis.
Circ. Res.
87
:
865
–872.
21
Boudreau, N.J., and J.A. Varner.
2004
. The homeobox transcription factor Hox D3 promotes integrin alpha5beta1 expression and function during angiogenesis.
J. Biol. Chem.
279
:
4862
–4868.
22
Jones, F.S., B.D. Holst, O. Minowa, E.M. De Robertis, and G.M. Edelman.
1993
. Binding and transcriptional activation of the promoter for the neural cell adhesion molecule by HoxC6 (Hox-3.3).
Proc. Natl. Acad. Sci. USA.
90
:
6557
–6561.
23
Wu, Y., M. Moser, V.L. Bautch, and C. Patterson.
2003
. HoxB5 is an upstream transcriptional switch for differentiation of the vascular endothelium from precursor cells.
Mol. Cell. Biol.
23
:
5680
–5691.
24
Izon, D.J., S. Rozenfeld, S.T. Fong, L. Komuves, C. Largman, and H.J. Lawrence.
1998
. Loss of function of the homeobox gene Hoxa-9 perturbs early T-cell development and induces apoptosis in primitive thymocytes.
Blood.
92
:
383
–393.
25
Lawrence, H.J., C.D. Helgason, G. Sauvageau, S. Fong, D.J. Izon, R.K. Humphries, and C. Largman.
1997
. Mice bearing a targeted interruption of the homeobox gene HOXA9 have defects in myeloid, erythroid, and lymphoid hematopoiesis.
Blood.
89
:
1922
–1930.
26
Thorsteinsdottir, U., A. Mamo, E. Kroon, L. Jerome, J. Bijl, H.J. Lawrence, K. Humphries, and G. Sauvageau.
2002
. Overexpression of the myeloid leukemia-associated Hoxa9 gene in bone marrow cells induces stem cell expansion.
Blood.
99
:
121
–129.
27
Brühl, T., C. Urbich, D. Aicher, A. Acker-Palmer, A.M. Zeiher, and S. Dimmeler.
2004
. Homeobox A9 transcriptionally regulates the EphB4 receptor to modulate endothelial cell migration and tube formation.
Circ. Res.
94
:
743
–751.
28
Dimmeler, S., A. Aicher, M. Vasa, C. Mildner-Rihm, K. Adler, M. Tiemann, H. Rutten, S. Fichtlscherer, H. Martin, and A.M. Zeiher.
2001
. HMG-CoA reductase inhibitors (statins) increase endothelial progenitor cells via the PI 3-kinase/Akt pathway.
J. Clin. Invest.
108
:
391
–397.
29
Urbich, C., C. Heeschen, A. Aicher, E. Dernbach, A.M. Zeiher, and S. Dimmeler.
2003
. Relevance of monocytic features for neovascularization capacity of circulating endothelial progenitor cells.
Circulation.
108
:
2511
–2516.
30
Dernbach, E., C. Urbich, R.P. Brandes, W.K. Hofmann, A.M. Zeiher, and S. Dimmeler.
2004
. Anti-oxidative stress-associated genes in circulating progenitor cells: evidence for enhanced resistance against oxidative stress.
Blood.
104
:
3591
–3597.
31
Yamamoto, K., T. Takahashi, T. Asahara, N. Ohura, T. Sokabe, A. Kamiya, and J. Ando.
2003
. Proliferation, diferentiation, and tube formation by endothelial progenitor cells in response to shear stress.
J. Appl. Physiol.
95
:
2081
–2088.
32
Fujiyama, S., K. Amano, K. Uehira, M. Yoshida, Y. Nishiwaki, Y. Nozawa, D. Jin, S. Takai, M. Miyazaki, K. Egashira, et al.
2003
. Bone marrow monocyte lineage cells adhere on injured endothelium in a monocyte chemoattractant protein-1-dependent manner and accelerate reendothelialization as endothelial progenitor cells.
Circ. Res.
93
:
980
–989.
33
Walter, D.H., K. Rittig, F. Bahlmann, R. Kirchmair, M. Silver, R. Murayama, H. Nishimura, D.W. Losordo, T. Asahara, and J.M. Isner.
2002
. Statin therapy accelerates reendothelialisation: a novel effect involving mobilisation and incorporation of bone marrow-derived endothelial progenitor cells.
Circulation.
105
:
3017
–3024.
34
Traub, O., and B.C. Berk.
1998
. Laminar shear stress: mechanisms by which endothelial cells transduce an atheroprotective force.
Arterioscler. Thromb. Vasc. Biol.
18
:
677
–685.
35
Staalesen, T., B. Risberg, and E. Mattsson.
2002
. The kinase insert domain-containing receptor (KDR) is regulated by shear stress.
Scand. Cardiovasc. J.
36
:
368
–372.
36
Urbich, C., M. Stein, K. Reisinger, R. Kaufmann, S. Dimmeler, and J. Gille.
2003
. Fluid shear stress-induced transcriptional activation of the vascular endothelial growth factor receptor-2 gene requires Sp1-dependent DNA binding.
FEBS Lett.
535
:
87
–93.
37
Nakamura, T., T. Mori, S. Tada, W. Krajewski, T. Rozovskaia, R. Wassell, G. Dubois, A. Mazo, C.M. Croce, and E. Canaani.
2002
. ALL-1 is a histone methyltransferase that assembles a supercomplex of proteins involved in transcriptional regulation.
Mol. Cell.
10
:
1119
–1128.
38
Carmeliet, P., V. Ferreira, G. Breier, S. Pollefeyt, L. Kieckens, M. Gertsenstein, M. Fahrig, A. Vandenhoeck, K. Harpal, C. Eberhardt, et al.
1996
. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele.
Nature.
380
:
435
–439.
39
Shalaby, F., J. Rossant, T.P. Yamaguchi, M. Gertsenstein, X.F. Wu, M.L. Breitman, and A.C. Schuh.
1995
. Failure of blood-island formation and vasculogenesis in Flk-1-deficient mice.
Nature.
376
:
62
–66.
40
Chen, F., and M.R. Capecchi.
1997
. Targeted mutations in hoxa-9 and hoxb-9 reveal synergistic interactions.
Dev. Biol.
181
:
186
–196.
41
Lagger, G., D. O'Carroll, M. Rembold, H. Khier, J. Tischler, G. Weitzer, B. Schuettengruber, C. Hauser, R. Brunmeir, T. Jenuwein, and C. Seiser.
2002
. Essential function of histone deacetylase 1 in proliferation control and CDK inhibitor repression.
EMBO J.
21
:
2672
–2681.
42
Murohara, T., T. Asahara, M. Silver, C. Bauters, H. Masuda, C. Kalka, M. Kearney, D. Chen, J.F. Symes, M.C. Fishman, et al.
1998
. Nitric oxide synthase modulates angiogenesis in response to tissue ischemia.
J. Clin. Invest.
101
:
2567
–2578.
43
Aicher, A., C. Heeschen, C. Mildner-Rihm, C. Urbich, C. Ihling, K. Technau-Ihling, A.M. Zeiher, and S. Dimmeler.
2003
. Essential role of endothelial nitric oxide synthase for mobilization of stem and progenitor cells.
Nat. Med.
9
:
1370
–1376.
44
Carmeliet, P., M.G. Lampugnani, L. Moons, F. Breviario, V. Compernolle, F. Bono, G. Balconi, R. Spagnuolo, B. Oostuyse, M. Dewerchin, et al.
1999
. Targeted deficiency or cytosolic truncation of the VE-cadherin gene in mice impairs VEGF-mediated endothelial survival and angiogenesis.
Cell.
98
:
147
–157.
45
Griese, D.P., A. Ehsan, L.G. Melo, D. Kong, L. Zhang, M.J. Mann, R.E. Pratt, R.C. Mulligan, and V.J. Dzau.
2003
. Isolation and transplantation of autologous circulating endothelial cells into denuded vessels and prosthetic grafts. Implications for cell-based vascular therapy.
Circulation.
108
:
2710
–2715.
46
Werner, N., S. Junk, U. Laufs, A. Link, K. Walenta, M. Bohm, and G. Nickenig.
2003
. Intravenous transfusion of endothelial progenitor cells reduces neointima formation after vascular injury.
Circ. Res.
93
:
e17
–e24.
47
Dimmeler, S., B. Fisslthaler, I. Fleming, C. Hermann, R. Busse, and A.M. Zeiher.
1999
. Activation of nitric oxide synthase in endothelial cells via Akt-dependent phosphorylation.
Nature.
399
:
601
–605.
48
Drake, C.J., and P.A. Fleming.
2000
. Vasculogenesis in the day 6.5 to 9.5 mouse embryo.
Blood.
95
:
1671
–1679.
49
Vecchi, A., C. Garlanda, M.G. Lampugnani, M. Resnati, C. Matteucci, A. Stoppacciaro, H. Schnurch, W. Risau, L. Ruco, A. Mantovani, et al.
1994
. Monoclonal antibodies specific for endothelial cells of mouse blood vessels. Their application in the identification of adult and embryonic endothelium.
Eur. J. Cell Biol.
63
:
247
–254.
50
Balconi, G., R. Spagnuolo, and E. Dejana.
2000
. Development of endothelial cell lines from embryonic stem cells: a tool for studying genetically manipulated endothelial cells in vitro.
Arterioscler. Thromb. Vasc. Biol.
20
:
1443
–1451.
51
Dong, Q.G., S. Bernasconi, S. Lostaglio, R.W. De Calmanovici, I. Martin-Padura, F. Breviario, C. Garlanda, S. Ramponi, A. Mantovani, and A. Vecchi.
1997
. A general strategy for isolation of endothelial cells from murine tissues. Characterization of two endothelial cell lines from the murine lung and subcutaneous sponge implants.
Arterioscler. Thromb. Vasc. Biol.
17
:
1599
–1604.
52
Li, H., S.A. Oehrlein, T. Wallerath, I. Ihrig-Biedert, P. Wohlfart, T. Ulshofer, T. Jessen, T. Herget, U. Forstermann, and H. Kleinert.
1998
. Activation of protein kinase C alpha and/or epsilon enhances transcription of the human endothelial nitric oxide synthase gene.
Mol. Pharmacol.
53
:
630
–637.

Abbreviations used: BMC, BM cells; butyrate, BuA; EBM, endothelial basal medium; eNOS, endothelial nitric oxide synthase; EPC, endothelial progenitor cell; ES, embryonic stem; HDAC, histone deacetylase; Hox, Homeobox gene; HUVEC, human umbilical vein endothelial cell; MNC, mononuclear cell; mt, mutant; siRNA, small interfering RNA; TSA, Trichostatin A; vWF, von Willebrand factor.

L. Rössig, C. Urbich, and T. Brühl contributed equally to this work.

Supplementary data