Histone deacetylases (HDACs) have a central role in the regulation of gene expression. Here we investigated whether HDAC7 has an impact on embryonic stem (ES) cell differentiation into smooth muscle cells (SMCs). ES cells were seeded on collagen-IV-coated flasks and cultured in the absence of leukemia inhibitory factor in differentiation medium to induce SMC differentiation. Western blots and double-immunofluorescence staining demonstrated that HDAC7 has a parallel expression pattern with SMC marker genes. In ex vivo culture of embryonic cells from SM22-LacZ transgenic mice, overexpression of HDAC7 significantly increased β-galactosidase-positive cell numbers and enzyme activity, indicating its crucial role in SMC differentiation during embryonic development. We found that HDAC7 undergoes alternative splicing during ES cell differentiation. Platelet-derived growth factor enhanced ES cell differentiation into SMCs through upregulation of HDAC7 splicing. Further experiments revealed that HDAC7 splicing induced SMC differentiation through modulation of the SRF-myocardin complex. These findings suggest that HDAC7 splicing is important for SMC differentiation and vessel formation in embryonic development.
Embryonic stem (ES) cells have the remarkable capability of differentiating into specific cell lineages, such as cardiovascular cells, in response to different stimuli in vitro (Keller, 1995; Smith, 2001). Both endothelial and vascular smooth muscle lineages can develop from a common progenitor, the vascular progenitor cell (Yamashita et al., 2000). Vascular endothelial growth factor (VEGF) and platelet-derived growth factor BB (PDGF-BB) have distinct roles in such progenitor cell differentiation (Gerber et al., 1998; Hellstrom et al., 1999). In particular, VEGF increases endothelial cell differentiation, whereas PDGF induces SMC differentiation (Gerecht-Nir et al., 2003). However, the underlying mechanism for vascular cell differentiation is still unclear.
Vascular SMCs have a crucial function in the embryonic development and physiological maintenance of the cardiovascular system (Gittenberger-de Groot et al., 1999). In adults, recent experimental and human data suggest a potential contribution of vascular progenitors, which originate from the circulation and the perivascular adventitia, to the differentiation of SMCs found within atherosclerotic lesions (Campbell and Campbell, 1994; Hu et al., 2004; Mathur and Martin, 2004; Xu et al., 2003). Obviously, vascular progenitor cells that can differentiate into either endothelial cells or SMCs could contribute to both vessel development in embryos and vascular disease in adults. Thus, the mechanisms that regulate SMC differentiation need to be elucidated.
SMC differentiation is a complicated and inadequately defined process. A number of factors are involved in this process, such as serum-response factor (SRF), myocardin, and myocyte enhancer factor 2 (MEF2). SRF has a key role during SMC differentiation by activating the transcription of an array of muscle-specific genes (Cao et al., 2005). Myocardin is an SRF cofactor expressed specifically in smooth muscle and cardiomyocytes throughout embryonic development and adulthood (Wang et al., 2001). MEF2C is mainly expressed in vessel SMCs throughout embryonic development (Lin et al., 1998).
The homeostasis of histone acetylation and deacetylation is known to have a central role in the regulation of gene expression, through the modulation of chromosome assembly or disassembly and through co-operation with other transcription factors (Wu et al., 2001; Yang and Seto, 2003). Histone deacetylases (HDACs) are part of transcriptional corepressor complexes and are key regulators in the differentiation of stem cells towards a specific cell lineage (Kato et al., 2004; Sterner and Berger, 2000). Three different classes of human HDACs have been defined based on their homology to HDACs found in Saccharomyces cerevisiae (Taunton et al., 1996; Verdin et al., 2003). The class II HDACs: HDAC4, HDAC5, HDAC6, HDAC7, HDAC9 and HDAC10 contain both nuclear localisation and export signals, trafficking between the cytoplasm and nucleus (Wang and Yang, 2001). Class II HDACs are reported to be important for cell differentiation in tissues such as muscles (Deng et al., 2005; Dressel et al., 2001; Karamboulas et al., 2006; Lu et al., 2000; McKinsey et al., 2000; McKinsey et al., 2001; Miyake et al., 2003; Vega et al., 2004; Zhang et al., 2002). HDAC7 is a member of the human class II HDAC family. However, there is no reported direct link between HDAC7 and SMC differentiation. This study aimed to investigate the role of HDAC7 in SMC differentiation from ES cells. We found that HDAC7 expression and alternative splicing correlated with ES cell differentiation towards SMCs.
HDAC7 is upregulated during SMC differentiation
Our previous study demonstrated that laminar flow induced ES cell differentiation toward endothelial cells, and highlighted the involvement of HDAC3 (Zeng et al., 2006). We found that there was a concomitant decrease of HDAC7 and SMC marker expression by shear stress (supplementary material Fig. S1). Thus, we wondered whether HDAC7 was important during ES cell differentiation, especially towards SMCs. Experiments were performed to elucidate a potential link between HDAC7 and SMC marker expression.
ES cells were seeded onto collagen-IV-coated flasks and cultured in the absence of leukaemia inhibitory factor (LIF) for 1 to 9 days to stimulate SMC differentiation, as previously described (Xiao et al., 2007). Western blot analysis showed that HDAC7, smooth muscle actin (SMA) and calponin were expressed at very low levels in undifferentiated and early stage differentiating ES cells. Following 5-7 days in culture, the expression of HDAC7, SMA and calponin was increased in a similar pattern (Fig. 1A). In addition, high levels of these proteins were detected in mature SMCs (Fig. 1A). Similar results were also obtained at the RNA level, showing a parallel expression of HDAC7 and SMC markers (Myocardin, SMA, SM22a, calponin and SMMHC) (Fig. 1B). Double-immunofluorescence staining showed that most HDAC7-positive cells were also positive for SMA and calponin. HDAC7 colocalized with SMA or calponin in the cytoplasm of differentiated SMCs and was scarcely apparent within the nucleus (Fig. 1C). By contrast, a high number of mature SMCs exhibited positive HDAC7 staining, found mainly in the nucleus (Fig. 1D). Moreover, enforced expression of HDAC7 by adenoviral gene transfer induced SMA and calponin expression in a dose-dependent manner in differentiated ES cells (Fig. 1E). These results suggest that HDAC7 is involved in ES cell differentiation toward SMCs.
HDAC7 is necessary for SMC differentiation
To test whether HDAC7 is necessary for SMC differentiation, HDAC7 siRNA knockdown experiments were performed. As expected, HDAC7 siRNA downregulated the expression of SMC markers such as SMA, calponin and SMMHC at the protein level (Fig. 2A) and RNA level (SMA, SM22a, calponin, SMMHC) (Fig. 2B). HDAC7 short hairpin (sh) RNA lentiviral plasmid transfer was also used to confirm the downregulation of HDAC7. As shown in Fig. 2A (right panel), shRNA lentiviral plasmid suppressed HDAC7 expression, concomitant with that of the SMC markers. SMA-Luc and SM22-Luc reporter assays also demonstrated downregulation of promoter activity after knockdown of HDAC7 (Fig. 2C).
HDAC7 is involved in SMC differentiation during embryo development
Next, we examined whether HDAC7 played a crucial role in SMC differentiation during embryonic development. Cells were isolated from SM22-LacZ mice at embryonic day (E)10-12 (expressing LacZ gene under the control of SM22 promoter) and cultured on collagen-IV-coated plates or slides in differentiation medium. The effects of HDAC7 overexpression by adenoviral gene transfer and siRNA knockdown on SMC differentiation were determined. X-gal staining showed that Ad-HDAC7 markedly increased the number of positive cells and the staining intensity in single cells (Fig. 3A, left panel). The dye was then eluted and quantified relative to protein content. Statistical analysis showed a significant increase in X-gal staining in Ad-HDAC7-infected cells compared with that in control cells (Fig. 3A, right panel). Western blot analysis further confirmed that increased levels of HDAC7 correlated with enhanced expression of SMA, and β-galactosidase (Fig. 3B, left panel). Moreover, HDAC7 siRNA downregulated protein levels (Fig. 3B, right panel). These results suggest that HDAC7 is involved in SMC differentiation during embryonic development.
HDAC7 undergoes splicing during ES cell differentiation
To facilitate this study, HDAC7 was amplified and cloned from predifferentiated ES cell mRNA by RT-PCR. Interestingly, most of the clones contained an intron of 57 bases between the start codon, ATG and the second ATG codon, which contained three stop codons disrupting the open reading frame from the first start codon. This resulted in alternative translation from the second ATG codon, giving rise to a short HDAC7 isoform lacking the first 22 N-terminal amino acids (Fig. 4A).
To investigate whether HDAC7 underwent alternative splicing during ES cell differentiation, mRNA was extracted from differentiating cells at various time points and RT-PCR was conducted using specific primers flanking the intron sequence (Fig. 4B). Undifferentiated ES cells expressed only the HDAC7 isoform containing the intron, whereas culture on collagen-IV-coated plates in the absence of LIF for 3 days resulted in the appearance of a spliced isoform lacking the intron. The ratio of spliced to unspliced isoform increased as the differentiation process proceeded (Fig. 4B). Interestingly, mature SMCs mainly expressed the spliced HDAC7 isoform (Fig. 4B), suggesting that HDAC7 splicing is correlated with SMC differentiation. This notion was further supported by the fact that increased levels of the spliced isoform of HDAC7 were observed in ES-derived SMCs during progressive stages of differentiation (Xiao et al., 2007; Xiao et al., 2006), as indicated by RT-PCR analysis (Fig. 4C).
HDAC7 splicing promotes ES cell differentiation toward SMCs
To further investigate the role of HDAC7 splicing in ES cell differentiation, three different HDAC7 cDNAs were cloned (Fig. 5A) into the modified pShuttle2-Flag-HA vector, designated pShuttle2-HDAC7, pShuttle2-HDAC7-1, pShuttle2-HDAC7-2, respectively. The pShuttle2-HDAC7 is the unspliced isoform, containing the intron and giving rise to a short HA-tagged HDAC7 lacking the first 22 amino acids, as pShuttle2-HDAC7-2 does. However, after splicing, pShuttle2-HDAC7 will give rise to both Flag and HA-tagged full-length HDAC7, as pShuttle2-HDAC7-1 does (Fig. 5B). When the unspliced HDAC7 vector was transferred into mature SMCs by adenoviral gene transfer, both Flag and HA tags were detected, indicating that HDAC7 underwent splicing in mature SMCs (Fig. 5C). Also, the spliced HDAC7 isoform was cloned into a pShuttle vector lacking the HA tag.
Transient co-transfection assays revealed that only the spliced HDAC7 isoform stimulated SMA and SM22 reporter gene expression in differentiated ES cells; the unspliced isoform had little impact, whereas the short isoform exerted an inhibitory effect (Fig. 5D). In contrast to ES cells, both the unspliced and spliced HDAC7 vectors stimulated reporter gene expression in mature SMCs, the effect of which was most prominent in the spliced vector. Unspliced HDAC7 was seen to have this effect as a result of splicing in mature SMCs (Fig. 5E). These results indicate that HDAC7 splicing is necessary to induce SMC differentiation from ES cells. Further experiments were performed to assess the deacetylase activities of the short and full-length HDAC7. ES cells were infected with Ad-tTA, Ad-HDAC7-1, Ad-HDAC7-2, and the deacetylase activity was detected 48 hours later. Both isoforms maintained their deacetylase activity at similar levels (Fig. 5F), indicating that lack of the 22 amino acids in the N-terminal does not affect the deacetylase activity of the short of HDAC7, and suggesting that the 22 amino acids may exert other functions.
Furthermore, we wondered whether the short HDAC7 (HDAC7-2) had any effect on the function of the full-length HDAC7-1. ES cells were cotransfected with pShuttle2-HDAC7-1 expressing only Flag tag and different amounts of pShuttle2-HDAC7-2. Data in Fig. 5G demonstrate that HDAC7-2 suppressed HDAC7-1-induced SM22-Luc reporter gene expression in a dose-dependent manner. Additionally, overexpression of HDAC7-2 did not affect the HDAC7-1 protein level (Fig. 5G, lower panel). These results indicate that HDAC7-2 has a dominant-negative effect on spliced HDAC7-mediated gene expression in SMCs.
PDGF-BB-enhanced SMC differentiation is mediated through HDAC7
As PDGF was reported to be a key factor in SMC differentiation, we hypothesized that there might be a link between PDGF stimulation and HDAC7 induction. Indeed, PDGF upregulated HDAC7, SMA and calponin protein levels in a similar manner (Fig. 6A). Moreover, PDGF enhanced the expression of HDAC7 reporter genes in both differentiated ES cells and SMCs (Fig. 6B), indicating that PDGF induces HDAC7 transcription.
Further experiments were performed to detect the effect of PDGF on HDAC7 splicing. PDGF significantly enhanced the effect of the unspliced HDAC7 plasmid on SM22 reporter gene expression in ES cells to a level comparable with that of the spliced plasmid (Fig. 6C), suggesting that PDGF increases HDAC7 splicing. RT-PCR indicated that splicing of HDAC7 increased significantly under PDGF treatment (Fig. 6D). To confirm this, enforced expression of unspliced HDAC7 was introduced to ES cells by adenoviral gene transfer, followed by PDGF treatment. As shown in Fig. 6E, the Flag-tagged HDAC7 (associated with expression of the spliced isoform), was highly upregulated after PDGF treatment. No significant differences were observed in HDAC7 expression levels in the presence of PDGF, which may be due to overexpression of exogenous HDAC7 (Fig. 6E). Finally, HDAC7 siRNA ablated PDGF-BB-induced SM22-Luc reporter gene expression in differentiated ES cells (Fig. 6F). Taken together, these results suggest that PDGF may promote ES cell differentiation into SMCs through upregulation of HDAC7 transcription and splicing.
HDAC7 isoform lacking the first 22 amino acids degrades MEF2C via the proteasome
To test whether different HDAC7 isoforms possess different cellular localization, the HDAC7-2 and HDAC7-1 vectors were introduced into SMCs by adenoviral gene transfer. As shown in Fig. 7A, the short HDAC7-2 mainly localized to the cytoplasm, whereas the spliced form localized to both the cytoplasm, and more abundantly, to the nucleus (Fig. 7B). To confirm this result, the cytoplasmic and nuclear components were fractionated (Fig. 7C). Interestingly, overexpression of the short HDAC7 increased the accumulation of endogenous HDAC7 (mainly spliced) in the nucleus of SMCs (Fig. 7C). These data suggest that different HDAC7 isoforms exist in different cellular compartments and might have different interactions with other proteins, such as transcription factors.
To elucidate whether different HDAC7 isoforms have different affinities to transcription factors, we determined the effect of spliced and short forms of HDAC7 on MEF2C, a transcription factor for SMC differentiation. MEF2C protein level was significantly decreased in the presence of the short HDAC7, but not in the presence of the spliced HDAC7 isoform (Fig. 7D) (supplementary material Fig. S2). However, the cellular localization of MEF2C seemed unchanged under HDAC7-1 and HDAC7-2 overexpression (supplementary material Fig. S2).
To discover how HDAC7-2 modulates MEF2C levels, we then performed RT-PCR in cells overexpressing HDAC7-1 and HDAC7-2. The data shown in supplementary material Fig. S3 indicate that neither HDAC7-1 nor HDAC7-2 has any effect on MEF2C mRNA levels, suggesting that HDAC7-2 decreases the MEF2C protein level through post-translational degradation. Co-immunoprecipitation experiments in SMCs revealed that the short HDAC7 isoform bound to MEF2C, whereas the spliced isoform did not (Fig. 7E). Similar results were obtained in differentiated ES cells (data not shown). The decrease in MEF2C induced by the short HDAC7 isoform could be ablated by the presence of proteasome inhibitor MG-132 (Fig. 7F). Further experiments showed that overexpression of HDAC7-2 increased the ubiquitylation of MEF2C (Fig. 7G). These results suggest that the short HDAC7 modulates MEF2C through proteasome-mediated degradation and has no effect on MEF2C cellular compartmentalization.
HDAC7 splicing induced SMC differentiation through modulation of the SRF-myocardin complex
Further experiments were performed to detect the effects of HDAC7 isoforms on SRF, another key transcription factor for SMC differentiation. Both HDAC7 isoforms did not suppress SRF at either the protein (Fig. 8A) or RNA (supplementary material Fig. S3) level. In contrast to MEF2C, spliced HDAC7 bound to SRF whereas short HDAC7 did not (Fig. 8A). It is reported that SRF activates genes involved in SMC differentiation and proliferation by recruiting myocardin to form a complex in the SMC marker gene promoter region (Wang et al., 2001). To test whether the different HDAC7 isoforms affected the recruitment of SRF and myocardin in the SMC marker gene promoter region, chromatin immunoprecipitation (ChIP) assays were performed. Spliced HDAC7 increased myocardin binding to SM22 and the calponin (CNN1) gene promoters, whereas the short form decreased both SRF and myocardin binding to these promoters (Fig. 8B). A deletion of the CArG box in the SM22 promoter abolished HDAC7-1-induced SM22-Luc reporter gene expression (supplementary material Fig. S4). These findings suggest that the spliced HDAC7 stimulates SM22 and calponin gene transcription during SMC differentiation by increasing the recruitment of myocardin to the CArG box in the promoter region.
As the spliced HDAC7 could associate with SRF, we wondered whether HDAC7 was also recruited to the promoter and affected the histone acetylation status. ChIP assays did not show HDAC7 binding to the promoters (data not shown). Unlike the short HDAC7, overexpression of spliced HDAC7 did not suppress the acetylation of Lys9 and dimethylation of Lys4 on Histone 3 tails (AcH3 and H3K4DM) in the promoter regions of SM22 and the calponin gene in ES cells (Fig. 8C).
Here, we demonstrated for the first time that HDAC7 undergoes alternative splicing during ES cell differentiation, and mediates SMC differentiation through modulation of the SRF-myocardin complex. PDGF-BB promotes ES cell differentiation into SMCs through the upregulation of HDAC7 transcription and splicing. Moreover, we provide evidence that the additional 22 amino acids in the N-terminus of spliced HDAC7 affect its cellular localization and influence the affinity of HDAC7 for associated proteins. These findings enhance our knowledge of SMC differentiation and provide useful insights into the mechanisms of vessel development.
We observed a clear correlation between HDAC7 expression and SMC differentiation from ES cells or ex vivo cultures of embryonic cells. Upregulation of HDAC7 expression increased the expression of SMC markers, whereas downregulation significantly decreased SMC marker expression, indicating a crucial role for HDAC7 in SMC differentiation. A recent study showed that HDAC7-knockout mice had fewer SMA-positive cells in the vessel wall of embryos, and the staining of SMA in the dorsal aorta was less prominent. Moreover, fewer than normal SMCs were observed surrounding dorsal aortae in the knockout embryos at E11, indicating that HDAC7 has an essential role in SMCs. However, in the same study, the histological sections of mouse embryos harbouring the HDAC7 mutation expressing LacZ did not show β-galactosidase staining in SMCs of the ascending aorta (Chang et al., 2006). This discrepancy might arise from the deletion of intron 2. It is possible that there are a number of important regulatory enhancer-like regions existing in intron 2 (19,729 bp), which are necessary for HDAC7 expression in SMCs.
An important finding of this study is that HDAC7 undergoes alternative splicing during ES cell differentiation into SMCs. HDAC7 mRNA from undifferentiated ES cells contains a 57 bp intron, in which three stop codons disrupt the open reading frame from the initiation ATG codon, resulting in alternative translation from a second ATG codon and giving rise to a short HDAC7 isoform lacking 22 N-terminal amino acids. During differentiation towards the SMC lineage, the intron is excised, giving rise to the full-length HDAC7 protein. Mature SMCs predominantly expressed the spliced HDAC7 isoform. The removal of 22 amino acid residues is likely to alter the conformation of HDAC7 and therefore alter its interactions with other proteins. HDAC7 is a shuttle protein, trafficking between the cytoplasm and the nucleus during cell differentiation (Kao et al., 2001; Karvonen et al., 2006). The full-length HDAC7 (spliced isoform) exists in both the cytoplasm and nucleus in mature SMCs. However, the short HDAC7 protein, which is translated from unspliced HDAC7 mRNA, predominantly exists in the cytoplasm. One possible explanation is that the short HDAC7 has a higher affinity for the anchorage proteins in the cytoplasm, but a lower affinity for trafficking-associated motor molecules than the spliced isoform. Thus, the overexpression of the short HDAC7 isoform occupies all the binding sites of the anchorage proteins, driving the accumulation of endogenous spliced HDAC7 in the nucleus. However, the association of HDAC7 with anchorage proteins might be necessary for HDAC7 activation. Hence, during overexpression of short HDAC7, endogenous HDAC7 accumulates in the nucleus, but is inactive. Importantly, both isoforms maintain their deacetylase activity, indicating that the N-terminus is responsible for the different localizations and functions.
Previous reports identified PDGF-BB as a key regulator for SMC differentiation (Nishishita and Lin, 2004). Recent reports indicate that vascular progenitor cells are able to differentiate into contractile-type SMCs in the absence of VEGF or into synthetic-type SMCs in the presence of PDGF-BB, respectively (Miyata et al., 2005). However, reports have also shown that treatment of SMCs with PDGF-BB dramatically reduced SMC α-actin synthesis in the prolonged absence of serum (Holycross et al., 1992), suppressing SMC differentiation (Wang et al., 2004). Our experiments showed that SMC differentiation from ES cells was enhanced through short stimulations with PDGF-BB (within 24 hours) in the absence of serum, indicating that the duration of serum deprivation and the proliferation stage of cells are important parameters in SMC differentiation mediated by PDGF-BB. This highlights the fact that PDGF-BB has different functions in mature SMCs and ES cells. In our previous study, proteomic analysis indicated that ES-derived SMCs are different from mature SMCs, and that PDGF-BB increased SMC marker gene expression (Yin et al., 2006). Therefore, it seems that PDGF-BB can initiate and enhance stem or progenitor cell differentiation towards SMCs through regulation of SMC-specific transcription factors. In this study, we showed that PDGF-BB induced SMC differentiation through upregulation of HDAC7 transcription and its alternative splicing.
Similarly to other HDACs, HDAC7 can also associate with transcription factors and other HDACs, modulating the acetylation status of transcription factors. Different isoforms might have different affinities for these proteins, eliciting different roles during ES cell differentiation. MEF2C is a key transcription factor for SMC differentiation. MEF2C has a specific temporal and spatial expression in the embryo and participates in vascular development; mice lacking MEF2C have no differentiated SMCs in the vasculature (Lin et al., 1998). HDAC7 represses MEF2C-dependent transcription via a physical interaction with the MADS domain of MEF2C. This was mapped to residues 72-172 of HDAC7, a region conserved among three classes of HDACs: HDAC4, HDAC5 and HDAC7 (Dressel et al., 2001; Kao et al., 2001). In our study, spliced HDAC7 does not bind to MEF2C and has no effect on the MEF2C protein level. However, the short HDAC7 binds to MEF2C and induces MEF2C degradation via the proteasome. The short HDAC7 protein might deacetylate MEF2C and expose lysine residues for ubiquitylation, because the ubiquitylation of MEF2C was increased in cells overexpressing HDAC7-2. Enhanced ubiquitylation was also observed in the input samples after HDAC7 treatment. Ubiquitylation and acetylation can occur at the same lysine residues in target proteins, thus acetylation will block the lysine residue, preventing ubiquitylation. Overexpression of HDAC7 might deacetylate these acetylated lysine residues, exposing them for ubiquitylation. Therefore, ubiquitylated proteins will be increased in the presence of HDAC7.
SRF is a key transcription factor in SMC differentiation, and it binds to CArG box DNA to recruit downstream accessory factors to regulate SMC transcription (McDonald et al., 2006). The coactivator myocardin interacts with SRF through a basic and glutamine-rich domain near the N-terminus (Wang et al., 2004) and increases association of SRF with methylated histone and CArG box chromatin during activation of SMC gene expression (McDonald et al., 2006). Moreover, myocardin might also utilize a specific epigenetic element (H3K4dMe) to control SRF association with CArG box chromatin (McDonald et al., 2006). Spliced HDAC7 binds to SRF and increases the recruitment of myocardin to SRF. ChIP assays clearly show the increase of myocardin binding to SM22 and calponin gene promoters in cells overexpressing HDAC7-1. The deletion of the CarG boxes abolished HDAC7-1-induced SM22-luc reporter gene expression, suggesting the involvement of SRF in HDAC7-mediated SMC differentiation. These findings support the idea that splicing of HDAC7 induces SMC marker expression through modulation of the SRF-myocardin complex. The short HDAC7 isoform significantly suppressed the binding of SRF and myocardin to the SM22 and calponin gene promoters, although endogenous spliced HDAC7 accumulated in the nucleus in the presence of exogenous short HDAC7. This provides further evidence that short HDAC7 prevents HDAC7 activation in the cytoplasm, but how HDAC7 is anchored and activated in cytoplasm needs more detailed investigation.
Moreover, future studies are necessary to shed light on the binding of SRF to spliced HDAC7, and to provide answers to a number of questions such as: (1) does spliced HDAC7 deacetylate the lysine residues of SRF, altering the conformation of SRF and increasing the affinity of SRF to myocardin? (2) How does the exogenous short HDAC7 increase indirectly the deacetylation and dimethylation of histone H3 in SM22 and calponin promoter, because this isoform does not localize to the nucleus?
In summary, HDAC7 mRNA undergoes alternative splicing during ES cell differentiation towards a SMC lineage. At an early stage, HDAC7 is expressed as a partially spliced isoform, which contains a 57 bp intron altering the open reading frame from the primary ATG codon. This gives rise to a short HDAC7 isoform, transcribed from the second ATG codon and resulting in an HDAC7 isoform lacking the first 22 amino acids. When triggered by PDGF or other stimuli, HDAC7 mRNA undergoes splicing to remove the intron, giving rise to full-length HDAC7. The spliced HDAC7 will be activated in the cytoplasm and translocated into the nucleus, where it associates with and modulates SRF, increasing SRF binding to myocardin and the recruitment of SRF-myocardin complex to the SM22 promoter. Thus, the overall effect is to drive SMC marker gene expression and cell differentiation towards an SMC lineage. By contrast, the short HDAC7 isoform binds to MEF2C, leading to MEF2C degradation via the proteasome. The short HDAC7 is predominantly localized in the cytoplasm, which may prevent activation of spliced HDAC7 in the cytoplasm. The overall effect prevents SMC gene expression, favouring progenitor cell differentiation towards other cell lineages. This hypothesis is illustrated in Fig. 9.
Vascular SMCs have a crucial role in both physiological maintenance of the cardiovascular system during embryonic development, and in the pathophysiology of vascular diseases, such as atherosclerosis, in adults (Gittenberger-de Groot et al., 1999; Hu et al., 2002; Liu et al., 2004). In this study, we found that alternative splicing of HDAC7 has an important role in determining ES cell differentiation towards a SMC lineage. Thus, targeting HDAC7 splicing will provide a new therapeutic strategy for intervention in vascular diseases.
Materials and Methods
Goat anti-HDAC7 (C-18, sc-11491), rabbit anti-HDAC7 (sc-1142), goat anti-MEF2C (sc-13268), rabbit ant-Ub (FL-76, sc-9133), goat myocardin (sc-34238), rabbit anti-SRF (sc-335) and rabbit anti-histone H4 (H-97, Sc-10810)] antibodies were purchased from Santa Cruz Biotech (Santa Cruz, CA). Mouse anti-α-tubulin (Clone B-5-1-2, T 5168), rabbit-anti-HDAC7 (KG-17, H 2662), mouse anti-HA (Clone HA-7, H9658) and monoclonal anti-HA-agarose-conjugated antibody (A2095) were from Sigma (St Louis, MO). Two sets of antibodies against calponin (Sigma, C2687 and Abcam, ab46794) and smooth muscle myosin heavy chain (SMMHC) (Sigma, M7786 and AbD Serotec, AHP1117) were used in this study. Antibodies against monoclonal anti-α smooth muscle actin (SMA) (Sigma, Clone 1A4, A5228), and Flag (F9291) were from Sigma. Antibody against β-galactosidase (beta-Gal) (rabbit) was from Delta Biolabs. Antibodies against acetyl histone H3 and dimethylated Lys4 histone H3 were from Upstate. All antibodies were raised in the mouse except those indicated. All secondary antibodies were from Dakocytomation, (Glostrup, Denmark). The deacetylase activity was detected with an HDAC activity assay kit (colorimetric) (ab1432) from Abcam.
Mouse ES cells (ES-D3 cell line, CRL-1934; ATCC, Manassas, VA) were cultured in gelatin-coated flasks in Dulbecco's modified essential medium (DMEM) (ATCC) supplemented with 10% fetal bovine serum (FBS) (ATCC), 10 ng/ml LIF (Chemicon), 0.1 mM 2-mercaptoethanol, 100 U/ml penicillin, 100 μg/ml streptomycin in a humidified incubator supplemented with 5% CO2, split at a 1:6 ratio every other day. Cell passages 3-20 were used in this study. Mature SMCs were isolated from mouse aorta as described previously (Leitges et al., 2001), and maintained in DMEM supplemented with 10% FCS, 100 U/ml penicillin, 100 μg/ml streptomycin, split at a 1:3 ratio every 3 days. HEK293 cell line was purchased from ATCC and cultured according to the company's recommendation.
ES cell differentiation
For differentiation, ES cells were seeded on mouse collagen IV (5 μg/ml)-coated flasks or plates in differentiation medium [DM, MEM alpha medium (Gibco) supplemented with 10% FBS (Gibco, lot 3095073K), 0.05 mM 2-mercaptoethanol, 100 U/ml penicillin, and 100 μg/ml streptomycin] for 1 to 9 days before further treatment. The medium was refreshed every other day. Shear experiments were performed as described previously (Zeng et al., 2006). For PDGF treatment, the pretreated ES cells were cultured in serum-free MEM alpha medium supplemented with 1% bovine serum albumin (BSA), 10 ng/ml insulin (Sigma), 0.05 mM 2-mercaptoethanol, 100 U/ml penicillin, and 100 μg/ml streptomycin for 1 hour, followed by the addition of 25 ng/ml PDGF-BB (Sigma) and further incubation for 3, 6 or 24 hours.
Cells were harvested and lysed in lysis buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA pH 8.0) supplemented with protease inhibitors and 0.5% Triton X-100 by sonication for whole cell lysate, or with hypotonic buffer (10 mM HEPES-KOH pH 7.2, 1.5 mM MgCl2, 10 mM KCl) and high-salt buffer (20 mM HEPES-KOH pH 7.2, 25% glycerol, 1.5 mM MgCl2, 420 mM KCl, 0.2 mM EDTA) supplemented with protease inhibitors and 0.5% NP-40 for nuclear and cytoplasmic fractions.
Indirect immunofluorescent assay
ES cells were seeded on collagen-IV-coated slides and cultured in DM. SMCs were seeded on gelatin-coated slides and cultured up to 70% confluence. An indirect immunofluorescent assay was performed. Goat anti-HDAC7 and FITC-conjugated donkey anti-goat IgG antibodies were used for HDAC7 staining, whereas Cy3-conjugated mouse anti-SMA antibody and mouse anti-calponin or TRITC-conjugated swine anti-mouse IgG antibodies were used for staining SMA and calponin, respectively. The cells were mounted with fluorescent mounting medium and observed under a fluorescence microscope, and SP5 confocal microscope. Images were assessed by Zeiss Axioplan 2 Imaging microscope with Plan-NEOPLUAR ×40/0.75 objective lenses, AxioCam camera and Axiovision software at room temperature, and were processed using Adobe Photoshop software.
RNA extraction and reverse transcriptase-polymerase chain reaction (RT-PCR)
Total RNA was extracted using the RNeasy Mini Kit (Qiagen) according to the manufacturer's protocol. 2 μg RNA were reverse transcribed into cDNA with random primer by MMLV reverse transcriptase (RT) (Promega). 50 ng cDNA (relative to RNA amount) was amplified by standard PCR with Taq DNA polymerase (Invitrogen) and primers.
Full-length mouse HDAC7 cDNA fragment was obtained by RT-PCR amplification from 3 day pre-differentiated ES cell RNA with the primer set HDAC7c-1 and HDAC7c-2 (Fig. 6A; supplementary material Table S1), and subcloned into KpnI/XbaI sites of the modified pShuttle2-Flag-HA vector, designated pshuttle2-HDAC7. The spliced HDAC7, designated as pshuttle2-HDAC7-1 and the short HDAC7, designated as pshuttle2-HDAC7-2 isoforms were created by PCR-based mutagenesis with specific primer sets (Fig. 6A; supplementary material Table S1). HDAC7, SM22 and SMA gene promoter sequences were amplified from human genomic DNA as follows: 458 bp HDAC7, 5′-AAGCCAGCAAGATCCTCATTG-3′ and 5′-ACAGATGGCCGTGAGGTCATG-3′ (Acc. No. AK036586); 415 bp SMA 5′-ACGGCCGCCTCCTCTTCCTC-3′ and 5′-GCCCAGCTTCGTCGTATTCC-3′ (BC064800); 507 bp SM22, 5′-GCAGTCCAAAATTGAGAAGA-3′ and 5′-CTGTTGCTGCCCATTTGAAG-3′ (BC003795); 384 bp MHC, 5′-AGGCAGACCTCATGCAGCTC-3′ and 5′-GAGCTTGGCTTTGACAGCAC-3′ (BC026142); 343 bp-actin, 5′-CACAACTGGGACGACATGGAG-3′and 5′-TTCATGAGGTAGTCAGTCTGG-3′ (M12481); 842 bp Calponin, 5′-TAACCGAGGTCCTGCCTACG-3′ and 5′-TGTGGGTGGGCTCACTCAGC-3′ (Z19542); 1055 bp HDAC7-P, 5′-CTAGACAAGCTTACAGAGAGAGGGAGCAGG-3′ and 5′-CACTCCCTCGAGGACAGTCTGTGGCTG-3′ (AC004466). HDAC7, SM22 and SMA gene promoter sequences were amplified from mouse genomic DNA as follows: 1113 bp SMA-P, 5′-TGCATGAGCCGTGGGAG-3′ (16-32 bp); 5′-ACTTACCCTGACAGCGAC-3′ (1128-1111 bp) (M57409); 1350 bp SM22-P, 5′-TTCAGGACGTAATCAGTG-3′ (4-21 bp); 5′-AGCTTCGGTGTCTGGGCTG-3′ (1371-1353 bp) (AH003214).
Promoter sequences were cloned into pGL3-luc basic vector (Promega), designated pGL3-HDAC7-Luc, pGL3-SM22-Luc and pGL3-SMA-Luc, respectively. SM22 reporter construct with deleted CArG-boxes was generated (Primer set: forward, 5′-ACCGGAAAGACACCAAGTTGG-3′; reverse, 5′-ACCAGCCTGTGTGGAGTGAG-3′). All the constructs were verified by DNA sequencing.
For transient transfection, the ES cells were cultured on collagen-IV-coated 12-well plate for 3 days, then transfected with reporter gene (0.33 μg/well) alone or together with expression plasmid (0.16 μg/well), using Fugene-6-Reagent (Roche Molecular Biochemicals), according to the manufacturer's instructions. Renilla luciferase (0.1 μg/well) was included in all transfection assays as internal control. The modified pShuttle2 vector was used as mock control. Luciferase and Renilla activities were detected 48 hours after transfection using a standard protocol. Relative luciferase unit (RLU) was defined as the ratio of Firefly versus Renilla with that of the control (set as 1.0). For verification of HDAC7 expression vectors, HEK293 cells were transfected with 1 μg/well of HDAC7 isoforms in six-well plates with Fugene 6 reagent, followed by western blot analysis.
For transient co-transfection experiments, different amount of pShuttle-HDAC7-2 expression plasmid (0, 0.5, 1.0 and 1.5 μg per 1×106 cells) with pShuttle-HDAC7-1 (0.5 μg per 1×106 cells) were introduced into ES cells by nucleofector II (Amaxa, Germany) with mouse ES cell nucleofection kit (Amaxa, VPH-1001) and using program A-30 according to the manufacturer's instructions. The appropriate amount of empty vector pShuttle was included as plasmid amount compensation and internal control. Nucleofected cells were plated in dishes coated with 5 μg/ml collagen IV and cultured for 2 days in DM. Total protein were harvested and subjected to western blot analysis.
The HDAC7 siRNA (sc-35547) and control CTL3 siRNA (sc-62166) were purchased from Santa Cruz Biotech, or control siRNA (Cat. No: 4611) was purchased from Ambion Ltd (Huntingdon, UK). ES cells were cultured on collagen IV-coated six-well plates for 5 days; 80 nM HDAC7 siRNA or CTL3 siRNA oligonucleotides were introduced into each well with siIMPORTER transfection reagent (Millipore) alone or together with reporter plasmids pGL3-SM22-Luc, and pGL3-SMA-Luc (0.17 μg each reporter per well) according to the protocol provided. All siRNA experiments were carried out in triplicate. Control siRNA-transfected cells were included as controls, and 0.05 μg of Renilla luciferase was used as internal control. All transfections were carried out in triplicate. The cells were harvested 72 hours after transfection and the luciferase activity assay or western blot analysis was performed. HDAC7 was also suppressed by HDAC7 shRNA lentiviral plasmids transfer (NM_019572 Sigma), according to the protocol provided.
Adenoviral DNA transfer
Ad-HDAC7 viral DNA was constructed from the corresponding plasmids into BD Adeno-X Viral DNA system (Clontech), the resulting viral particles were created and amplified in HEK293 cells with protocol provided. For adenoviral DNA transfer experiments, ES cells were seeded on collagen IV-coated dishes and cultured in DM for 3 days, then infected with Ad-HDAC7 at specific MOI (multiplicity of infection) as indicated in figures and further incubated in DM for 2 days, followed by western blot analysis. The empty virus Ad-tTA was used as a control virus and to compensate the MOI.
Ex vivo embryonic cell culture
E10-E12 embryos were harvested from SM22-LacZ mice (Moessler et al., 1996) and cells were dispensed through 1 ml syringe and No.19G needle. The cells were then seeded in collagen-IV-coated dishes in DM and incubated for 24 hours. After treatment with trypsin, 5×105 cells/2 ml were then seeded in collagen-IV-coated six-well plates in DM for another 24 hours. The cells were infected with Ad-HDAC7 virus at 10 MOI or transfected with HDAC7 siRNA, and incubated for 72 hours before X-gal staining and western blot analysis were performed. Ad-tTA virus and control siRNA were used as controls. Images were assessed and processed as described above except ×20/0.5 objective lenses were used instead. After the pictures were taken, the dye was extracted from the stained samples with DMSO, followed by detection of absorbance at 590 nm and protein level quantification. The readout of A590nm was normalized per μg protein with that of Ad-tTA set as 1.0. All animal experiments were performed according to protocols approved by the Institutional Committee for Use and Care of Laboratory Animals.
Cells were infected with Ad-HDAC7-1, HDAC7-2 or Ad-tTA. 48 hours after infection, the cells were lysed by rotation for 1 hour at 4°C. 1 mg whole lysate was subjected to a standard co-immunoprecipitation procedure. Lysates were pre-cleared with normal IgG and then were incubated with HA or MEF2C antibodies for 2 hours at 4°C and precipitated by incubation for a further 2 hours with protein-G-Sepharose beads. Precipitated proteins were resolved by SDS gel electrophoresis and subsequently immunoblotted with MEF2C, SRF, ubiquitin and HA antibodies.
Chromatin immunoprecipitation (ChIP)
Differentiated ES cells were infected with Ad-HDAC7-1, HDAC7-2 or Ad-tTA. The ChIP assays were performed as described previously (Zampetaki et al., 2007). In brief, ES cells were treated with 1% (v/v) formaldehyde at room temperature for 10 min and then quenched with glycine at room temperature. The medium was removed, and cells were harvested for sonication. The sheared samples were diluted into 1 ml immunoprecipitation buffer containing 25 mM Tris-HCl, pH 7.2, 0.1% NP-40, 150 mM NaCl, 1 mM EDTA, and immunoprecipitation was conducted with antibodies raised against myocardin, SRF, acetyl-histone H3, or dimethylated Lys4-histone H3, together with single-strand salmon sperm DNA saturated with Protein-G-Sepharose beads. Normal IgG was used as a control. The immunoprecipitates were eluted from the beads using 100 μl elution buffer (50 mM NaHCO3, 1% SDS). A total of 200 μl proteinase K solution was added to a total elution volume of 300 μl and incubated at 60°C overnight. DNA was extracted, purified, and then used to amplify target sequences by PCR. The primers used to amplify the promoter regions were: SM22 ChIP forward, 5′-GGATTTGGGGAATCCTGTCT-3′; SM22 ChIP reverse, 5′-GGGAAGGAGAGAACCCTCAG-3′; calponin ChIP forward 5′-CAGACCAGCCTGATTTGGAT-3′; and calponin ChIP reverse, 5′-AGGAATAGGCAGGATGCTCA-3′. Aliquots of chromatin were also analysed before immunoprecipitation and served as an input control. SM22 exon 5 forward, 5′-CTTCACAGACAGCCAACTGC-3′ and SM22 exon 5 reverse, 5′-CCCATGTGCAGTCATCTTTG-3′ primers amplify a part of the mouse SM22 exon 5 and were used as a negative control for the assay. The data obtained from four independent experiments were quantified, using Photoshop. Fold of relative binding was defined as the ratio of band intensity in ChIP samples to that in the Input sample with that of Ad-tTA group set as 1.0.
Primers were designed using Primer Express software (Applied Biosystems) and a published sequence for the mouse genes: SMA, 5′-TCCTGACGCTGAAGTATCCGAT-3′ and 5′-GGCCACACGAAGCTCGTTATAG-3′ (X13297); SM22, 5′-GATATGGCAGCAGTGCAGAG-3′ and 5′-AGTTGGCTGTCTGTGAAGTC-3′ (MMU36588); Calponin, 5′-GGTCCTGCCTACGGCTTGTC-3′ and 5′-TCGCAAAGAATGATCCCGTC-3′ (NM009922); MHC, 5′-AAGCAGCCAGCATCAAGGAG-3′ and 5′-AGCTCTGCCATGTCCTCCAC-3′ (NM013607); Myocardin, 5′-TCAATGAGAAGATCGCTCTCCG-3′ and 5′-GTCATCCTCAAAGGCGAATGC-3′ (NM145136); HDAC7, 5′-CCCAGTGTGCTCTACATTTCCC-3′ and 5′-CACGTTGACATTGAAGCCCTC-3′ (AF207749). For each gene, SYBR Green (Applied Biosystems) was used in place of a labelled probe. 18S ribosomal RNA was used as the internal control for each gene and target gene was amplified in duplex in PCR mixtures (25 μl final volume) containing 12.5 μl Sybr® Green PCR Master Mix (Applied Biosystems), cDNA template (20 ng) and optimised primers. PCR thermal cycle parameters were: 2 minutes at 50°C, 10 minutes at 95°C and 40 cycles of 95°C for 15 seconds and 60°C for 1 minute. Reactions were performed, and fluorescence was monitored in an ABI Prism 7000 Sequence Detector system (Applied Biosystems). Relative mRNA expression level was defined as the ratio of target gene expression level to 18S rRNA expression level with that of the control sample set as 1.0.
Data are expressed as the mean ± s.e.m. and were analyzed with a two-tailed Student's t-test for two-groups or pair-wise comparisons. A value of P<0.05 was considered to be significant.
We thank John Paul Kirton for critical reading of the manuscript. This work was supported by Grants from British Heart Foundation and Oak Foundation.
Supplementary material available online at http://jcs.biologists.org/cgi/content/full/122/4/460/DC1
↵* These authors contributed equally to this work
- Accepted October 14, 2008.
- © The Company of Biologists Limited 2009