TY - JOUR
AU - Han, Yanling
AB - Abstract Understanding the regulation of cell-cell interactions during the formation of compact myocardial structures is important for achieving true cardiac regeneration through enhancing the integration of stem cell-derived cardiomyocytes into the recipient myocardium. In this study, we found that cellular repressor of E1A-stimulated genes 1 (CREG1) is highly expressed in both embryonic and adult hearts. Gain- and loss-of-function analyses demonstrated that CREG1 is required for differentiation of mouse embryonic stem (ES) cell into cardiomyocytes and the formation of cohesive myocardium-like structures in a cell-autonomous fashion. Furthermore, CREG1 directly interacts with Sec8 of the exocyst complex, which tethers vesicles to the plasma membrane. Site-directed mutagenesis and rescue of CREG1 knockout ES cells showed that CREG1 binding to Sec8 is required for cardiomyocyte differentiation and cohesion. Mechanistically, CREG1, Sec8, and N-cadherin colocalize at intercalated discs in vivo and are enriched at cell-cell junctions in cultured cardiomyocytes. CREG1 overexpression enhances the assembly of adherens and gap junctions. By contrast, its knockout inhibits the Sec8-N-cadherin interaction and induces their degradation. These results suggest that the CREG1 binding to Sec8 enhances the assembly of intercellular junctions and promotes cardiomyogenesis. Schematic model for a role of CREG1 in the formation of intercalated discs between cardiomyocytes. During cardiomyocyte differentiation from embryonic stem cells, CREG1 is upregulated and binds to Sec8 of the exocyst complex. The CREG1-exocyst interaction increases the delivery of N-cadherin from intracellular compartments, such as Golgi and recycling endosomes (ER), to the cell-cell adhesion sites, thereby promoting the formation of intercalated discs. Open in new tabDownload slide Open in new tabDownload slide Intercalated discs, Exocyst, Embryonic stem cells, Cardiac differentiation Significance Statement Understanding the regulation of cell-cell interactions during the formation of compact myocardial structures is important for achieving true cardiac regeneration through enhancing the integration of stem cell-derived cardiomyocytes into the recipient myocardium. In this study, we uncovered a novel mechanism whereby CREG1 promotes cardiomyogenesis and myocyte interactions by binding to the exocyst. The findings are important to our understanding of the formation of intercalated discs that connect myocytes in the myocardium. Furthermore, CREG1 gene transfer may improve the integration of stem cell-derived cardiomyocytes into the recipient heart. Introduction Heart failure due to myocardial infarction is a leading cause of mortality in developed nations. Given that adult cardiomyocytes have a very limited ability to regenerate, the use of stem cell-derived cardiomyocytes to repair the damaged heart becomes a new therapeutic paradigm. In order to generate a large number of cardiomyocytes at the appropriate differentiation stage for transplantation and to ensure their integration in the host heart, it is important to understand the mechanisms responsible for cardiomyogenic differentiation of pluripotent stem cells. Formation of new cell-cell contacts by reconstructing intercalated discs is a prerequisite of cardiomyocyte integration. Previous studies have shown that embryonic stem (ES) cell-derived cardiomyocytes often fail to form cell-cell junctions with endogenous myocytes after being engrafted in normal cardiac muscle. Instead, they were isolated by a scar tissue capsule, thus precluding the formation of a new functional myocardial syncytium [1-5]. In the heart, contraction forces developed by cardiomyocytes are transmitted across the plasma membrane through intercalated discs, which enable the coordinated contraction of cardiac muscle. As an important component of the intercalated disc, the adherens junction (AJ) consists of N-cadherin and adaptor proteins of the catenin family, which link to the actin cytoskeleton. Embryos lacking N-cadherin die at midgestation from heart tube abnormalities [6]. In chimeric mice, the mutant cardiomyocytes cannot form cell-cell junctions with the wild-type counterparts and are excluded from the myocardium [7]. When cultured in vitro, N-cadherin-null cardiomyocytes fail to align myofibrils through regions of cell-cell contacts, resulting in their random orientation. In addition, gap junctions (GJs) are disrupted in the mutant myocytes likely because connexin43 fails to be targeted to AJs for GJ assembly [8-11]. Therefore, the expression level of N-cadherin at the myocyte junctions is vital to the morphological and functional integrity of cardiac muscle. However, it remains largely unknown how the assembly of these intercellular junctions is regulated. Elucidating the underlying molecular mechanisms is critical for designing strategies to improve the integration of stem cell-derived cardiomyocytes for cardiac repair. Cellular repressor of E1A-stimulated genes 1 (CREG1) was discovered in yeast two-hybrid screening of a Drosophila cDNA library and was initially suggested to be a transcription repressor that antagonizes transcription activation and cellular transformation induced by the adenovirus E1A oncoprotein [12]. Subsequently, it was found to be a small glycoprotein containing 220 amino acid residues and three consensus N-glycosylation sites (Asn 160, 193, and 216 in human CREG1) [13, 14]. Crystal structure analysis has revealed that CREG1 exists as a homodimer and the three glycosylation sites are all exposed on the surface [15]. The glycosylation is required for CREG1 binding to its putative receptor mannose 6-phosphate/insulin-like growth factor 2 receptor (IGF2R) [14]. In cultured cells, overexpression of CREG1 inhibited proliferation and enhanced differentiation [13, 16, 17]. Yet the physiological function of CREG1 has not been reported. Our genetic analysis revealed that global disruption of the murine Creg1 gene leads to early embryonic death around E7.5 (unpublished data). In an attempt to investigate the underlying mechanisms using ES cell-differentiated embryoid bodies (EBs), we found by accident, that overexpression of CREG1 dramatically enhances ES cell differentiation into cardiomyocytes. Loss-of-function analysis using Creg1 knockout ES cells confirmed an essential role for CREG1 in ES cell cardiomyogenesis. Of note, we identified a novel interaction between CREG1 and the exocyst component Sec8, which is required for cardiomyocyte differentiation and the assembly of intercellular junctions. These results are important not only for our understanding of cardiomyogenesis and CREG1 functions but also for the development of therapeutic strategies to improve cardiomyocyte transplantation. Materials and Methods ES Cell Culture and Embryoid Body Differentiation Creg1 heterozygous ES cell lines were generated using R1 ES cells by homologous recombination. Homozygous clones were obtained by selection with high concentrations of G418. The polymerase chain reaction (PCR) primer sequences used for genotyping are 5′-TGTCGGGAACTGTGACCAAG (forward) and 5′-CTTTAGTTGTTGAAATCTGTG (reverse) for the wild-type allele and 5′-CTCAGCCTTGGGGGTGCTGGGAAGA (forward) and 5′-TCGTCGTGACCCATGGCGATGCCTG (reverse) for the knockout allele. The neomycin-resistant cassette, which is flanked by 2 flippase recognition target (FLP) sites, was removed by transient transfection with flippase-GFP. ES cells were maintained on feeders of mitomycin-treated mouse embryonic fibroblasts or STO (SIM mouse embryonic fibroblasts selected for 6-thioguanine and ouabain resistance) cells. EB differentiation was initiated from ES cell aggregates in suspension culture, as previously described [18]. For cardiomyogenic differentiation, 5-day EBs were allowed to attach on laminin-111 (50 μg/ml)-coated glass coverslips or culture dishes and cultured for an additional 4–14 days. Formation of spontaneously beating clusters/foci was monitored daily by phase-contrast microscopy. Antibodies and Plasmids Rabbit anti-CREG1 and mCherry polyclonal and monoclonal antibodies (pAb and mAb) were from Abcam (Cambridge, MA). Goat anti-CREG1 pAb and mouse anti-CREG1 mAb were from R&D Systems (Minneapolis, MN). Mouse anti-Connexin43 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mAbs were from EMD Millipore (Billerica, MA). Rabbit anti-actin pAb and mouse anti-sarcomeric actinin and N-cadherin mAbs were from Sigma (St. Louis, MO). Mouse anti-N-cadherin mAb was from Life Technologies (Grand Island, NY). Mouse anti-sarcomeric myosin heavy chain mAb was from Hybridoma Bank (Iowa City, Iowa). Rabbit anti-cardiac troponin I was from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-myosin light chain 3 pAb was from GeneTex (Irvine, CA). Mouse anti-His-tag mAb was from BioLegend (San Diego, CA). Mouse anti-Sec6, Sec8, and Exo70 mAbs were described previously [19]. Rabbit anti-cleaved caspase 3 pAb was from Cell Signaling Technology (Danvers, MA). Cy3-, Cy5-, and horseradish peroxidase-conjugated secondary antibodies were from Jackson ImmunoResearch (West Grove, PA). Alexa 488-conjugated secondary antibodies were from Life Technologies. The cDNAs for human wild-type CREG1 was provided by Dr. Grace Gill of Tufts University (Boston, MA) and subcloned into pcDNA3.1 with a His-tag and pCNX2-IRES-green fluorescence protein (GFP). Glycosylation mutant (N160A/N193A/N216A), Sec8-binding mutant (Δ141–144 and D141A/P142A) CREG1 were generated in pcDNA3.1 using the QuickChange II mutagenesis kit (Aligent Technologies, Santa Clara, CA) and subcloned into pCNX2-IRES-GFP. Generation of CREG1 recombinant adenoviral vectors and infectious viral particles was reported previously [20]. paMHC-mCherry-Rex-Blasticidin and paMHCeGFP-Rex-Neo were obtained from Addgene (Cambridge, MA) [21]. All constructs were confirmed by DNA sequencing. The construction of GST-Sec6 and GST-Sec8 plasmids were described previously [22]. Sec8 shRNA and scrambled control plasmids were purchased from Origene (Rockville, MD). The shRNA sequence that produced the most effective Sec8 knockdown is GTTCTTCAGCAGAACTTGACCAACATCAC. Stable Transfection of ES Cells ES cells were transfected with the corresponding plasmids using the jetPRIME transfection reagent (Polyplus Transfection SA, New York, NY). Stable ES cell clones were selected based antibiotic resistance and GFP/mCherry/RFP fluorescence and expanded on STO feeder cells. Production of Recombinant Proteins HEK293 cells were transiently transfected with His-tagged wild type and mutant CREG1 using the jetPRIME reagent. After 2 days, conditioned media and cell lysates were harvested. His-CREG1 was purified by chromatography using Ni-NTA agarose beads (Qiagen, Valencia, CA). GST-Sec6/8 fusion proteins were expressed in bacteria and isolated by affinity chromatography using GSH-agarose beads. In vitro binding assay was performed at room temperature for 1 hour as described previously [23]. Isolation of Cardiomyocytes from Attached Embryoid Bodies and Neonatal Mice Spontaneously beating clusters/foci within individual EBs were isolated by mechanical dissection. Cells were then dissociated into single cells using 1 mg/ml collagenase IV and cardiomyocytes were separated on discontinuous Percoll gradients (top density 1.059 g/ml, bottom density 1.082 g/ml). Cardiomyocytes were then cultured on fibronectin (25 μg/ml)-coated glass coverslips for 24–48 hours. The purity of myocytes was determined by immunostaining for cardiac troponin I and sarcomeric myosin heavy chain. Neonatal cardiomyocytes were isolated by trypsin digestion from 1-day-old mice and grown as described [24, 25]. Myocytes were infected with adenovirus expressing CREG1 or the control virus for 48 hours and then fixed for immunofluorescence microscopy. GFP expression revealed >96% infection efficiency after exposure of cells to the virus for 48 hours. Immunofluorescence EBs and isolated cardiomyocytes cultured on glass coverslips were fixed with 3% paraformaldehyde and immunostaining was carried out as described [26]. Slides were examined with a Nikon inverted fluorescence microscope (Eclipse TE2000), and digital images were acquired with an Orca-03 cooled charge-coupled-device (CCD) camera (Hamamatsu) controlled by IP Lab software (Scanalytics). Immunoprecipitation and Immunoblotting Attached EBs and cardiomyocyte cultures were washed twice in phosphate-buffered saline (PBS) and lysed in radioimmunoprecipitation assay (RIPA) buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% NP-40, 0.25% sodium deoxycholate) containing protease and phosphatase inhibitor cocktails (Sigma). For immunoblot analysis of cardiac marker proteins, cells were lysed in SDS lysis buffer containing 10 mM Tris (pH 7.4), 1% SDS and protease and phosphatase inhibitors. Immunoprecipitation and immunoblotting were performed as described previously [27]. Statistical Analysis Statistical analysis was performed using SigmaStat software. Statistical significance was evaluated by one-way analysis of variance and Student's t test. A p-value less than 0.05 was considered significant. Error bars on all figures are represented as a standard deviation of the mean. Blots shown are representative of two to four independent repeats. Results CREG1 Expression Is High in Developing Hearts and Correlates with Cardiomyogenesis Previous studies showed that overexpression of CREG1 induced differentiation of human embryonal carcinoma cells and vascular smooth muscle cells [13, 17]. To explore its physiological functions, we analyzed its expression at embryonic and adult stages in mice. In E15.5 embryos, the CREG1 protein is highly expressed in the heart. Its expression levels are lower in the lung, kidney, muscle, brain and liver (Fig. 1A). In adult hearts, CREG1 is mainly localized to intercalated discs where the adherent junction receptor N-cadherin is highly enriched (Fig. 1B). To examine the relationship between CREG1 expression and cardiac differentiation, we employed the EB cardiogenesis model, which is often used to study early cardiomyogenic events [28, 29]. EBs cultured in suspension for 5 days form endoderm, a polarized epiblast epithelium and a proamniotic-like cavity [18, 27]. They were then allowed to attach on tissue culture dishes or glass coverslips and cultured for additional 3–7 days. The CREG1 protein was detected at low levels in 5 + 3 day EBs and significantly upregulated at 5 + 5 and 5 + 7 days (Fig. 1C). The upregulation of CREG1 is concurrent with the appearance of spontaneously beating foci in attached EBs and closely correlates with that of the cardiac markers the myosin heavy chain, cardiac actinin, troponin I, and myosin light chain 3. Immunostaining of 5 + 7 day attached EBs revealed that CREG1 is mainly expressed in cardiomyocytes but not the surrounding cells (Fig. 1D). Of note, it is primarily localized to myocyte junctions and subplasma membrane vesicular compartments. These results suggest that CREG1 may be involved in cardiomyogenic differentiation and intercellular junction assembly. Figure 1 Open in new tabDownload slide CREG1 is highly expressed in the heart and its protein levels correlate with cardiomyogenic differentiation. (A): Immunoblot analysis of mouse embryonic (E15.5) tissues revealed high levels of CREG1 expression in the heart. GAPDH served as a loading control. (B): Adult mouse hearts were immunostained for CREG1 and N-cad. Myocytes were identified by sarcomeric actinin and cTnI immunofluorescence. Nuclei were counterstained with 2-(4-Amidinophenyl) -6-indolecarbamidine dihydrochloride (DAPI). CREG1 is highly enriched at intercalated discs. (C): Mouse ES cell-derived embryoid bodies (EBs) were cultured in suspension for 5 days (5 + 0) and allowed to attach to tissue culture dishes for additional 3 (5 + 3), 5 (5 + 5), and 7 (5 + 7) days. The attached EBs were lysed and subjected to immunoblotting for CREG1, the cardiomyocyte markers MHC, cardiac actinin, cTnI, and MYL3. Actin was used as a loading control. Expression of CREG1 correlates with those of cardiomyogenic differentiation markers. (D): 5 + 7 day normal EBs were immunostained for CREG1 and cTnI. CREG1 is localized to cell-cell junctions and intracellular vesicular compartments in a punctate pattern in cardiomyocyte aggregates. The inserts are 2× magnifications of the boxed areas. Abbreviations: CREG1: cellular repressor of E1A-stimulated genes 1; cTnI, cardiac troponin I; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MHC, myosin heavy chain; MYL3, myosin light chain 3; N-cad, N-cadherin. Figure 1 Open in new tabDownload slide CREG1 is highly expressed in the heart and its protein levels correlate with cardiomyogenic differentiation. (A): Immunoblot analysis of mouse embryonic (E15.5) tissues revealed high levels of CREG1 expression in the heart. GAPDH served as a loading control. (B): Adult mouse hearts were immunostained for CREG1 and N-cad. Myocytes were identified by sarcomeric actinin and cTnI immunofluorescence. Nuclei were counterstained with 2-(4-Amidinophenyl) -6-indolecarbamidine dihydrochloride (DAPI). CREG1 is highly enriched at intercalated discs. (C): Mouse ES cell-derived embryoid bodies (EBs) were cultured in suspension for 5 days (5 + 0) and allowed to attach to tissue culture dishes for additional 3 (5 + 3), 5 (5 + 5), and 7 (5 + 7) days. The attached EBs were lysed and subjected to immunoblotting for CREG1, the cardiomyocyte markers MHC, cardiac actinin, cTnI, and MYL3. Actin was used as a loading control. Expression of CREG1 correlates with those of cardiomyogenic differentiation markers. (D): 5 + 7 day normal EBs were immunostained for CREG1 and cTnI. CREG1 is localized to cell-cell junctions and intracellular vesicular compartments in a punctate pattern in cardiomyocyte aggregates. The inserts are 2× magnifications of the boxed areas. Abbreviations: CREG1: cellular repressor of E1A-stimulated genes 1; cTnI, cardiac troponin I; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MHC, myosin heavy chain; MYL3, myosin light chain 3; N-cad, N-cadherin. CREG1 Promotes Cardiomyogenic Differentiation and the Formation of Intercellular Junctions To elucidate the role of CREG1 in cardiomyogenic differentiation, we performed gain- and loss-of-function experiments. Stable overexpression of CREG1 markedly increases spontaneously beating EBs at 5 + 4, 5 + 7, and 5 + 14 days (Fig. 2A). Interestingly, the CREG1-overexpressing myocytes often form a large contracting mass projecting upward with a well-defined boundary, which could only be observed occasionally in EBs transfected with the control vector. Biochemical analysis confirmed that cardiomyogenic differentiation is enhanced in 5 + 7 day EBs transfected with CREG1 (Fig. 2B). Of note, immunofluorescence microscopy revealed that CREG1-overexpressing myocytes are organized into three-dimensional, cohesive myocardium-like structures while the control cells form sheets with myocytes extending in all directions (Fig. 2C). We also cultured EBs in suspension for 9 days and analyzed cardiomyogenic differentiation. Vigorous beating was observed in ∼36% of the EBs overexpressing CREG1 (Supporting Information Video 1), whereas none of the control EBs displayed spontaneous contraction. Immunostaining for troponin I showed a large cluster of cardiomyocytes in the EB center (Supporting Information Fig. S1). These results suggest that increased expression of CREG1 promotes cardiomyocyte differentiation and cohesion. Figure 2 Open in new tabDownload slide Overexpression of CREG1 promotes cardiomyogenic differentiation, intercellular junction assembly, and formation of cohesive myocardium-like structures. (A): EBs stably transfected with CREG1 or the control vector were cultured in suspension for 5 days and then allowed to attach on laminin-coated culture dishes for additional 4–14 days. Spontaneously beating EBs were counted and plotted as a percentage of total EBs examined (n = 7 with a total of 351–423 EBs counted for each group, p < .01). (B): 5 + 7 day EBs were analyzed by immunoblotting for the expression of CREG1 and cardiac marker proteins. Actin served as a loading control. Overexpression of CREG1 enhances cardiomyogenic differentiation. (C): 5 + 7 day EBs were immunostained for cardiac actinin. Overexpression of CREG1 promotes the formation of cohesive myocardium-like structures. (D): Cardiomyocytes were isolated from attached EBs and cultured on fibronectin-coated glass coverslips for 24 hours. Cells were coimmunostained for N-cadherin and the MHC or connexin43 and cardiac troponin I. Overexpression of CREG1 increased N-cadherin and connexin43 immunofluorescence at myocyte junctions (arrows). (E): Pairs of myocytes with linear staining of N-cadherin or connexin43 at their junctions were counted and plotted as a percentage of total myocyte pairs examined. N = 4 with 91–96 pairs of myocytes for each group. *, p < .001. (F): Mouse neonatal cardiomyocytes were infected with Ad-CREG1 or the control virus for 48 hours. Cells were fixed and immunostained for N-cadherin and cardiac troponin I. In control myocytes, N-cadherin was localized at both cell-cell junctions and intracellularly adjacent to cell-cell contacts (arrows). In Ad-CREG1 infected cells, N-cadherin accumulation at cell-cell junctions was significantly increased (arrows). Abbreviations: Ad-CREG1, adeno-CREG1; CREG1: cellular repressor of E1A-stimulated genes 1; EBs, embryoid bodies; MHC, myosin heavy chain; MYL3, myosin light chain 3. Figure 2 Open in new tabDownload slide Overexpression of CREG1 promotes cardiomyogenic differentiation, intercellular junction assembly, and formation of cohesive myocardium-like structures. (A): EBs stably transfected with CREG1 or the control vector were cultured in suspension for 5 days and then allowed to attach on laminin-coated culture dishes for additional 4–14 days. Spontaneously beating EBs were counted and plotted as a percentage of total EBs examined (n = 7 with a total of 351–423 EBs counted for each group, p < .01). (B): 5 + 7 day EBs were analyzed by immunoblotting for the expression of CREG1 and cardiac marker proteins. Actin served as a loading control. Overexpression of CREG1 enhances cardiomyogenic differentiation. (C): 5 + 7 day EBs were immunostained for cardiac actinin. Overexpression of CREG1 promotes the formation of cohesive myocardium-like structures. (D): Cardiomyocytes were isolated from attached EBs and cultured on fibronectin-coated glass coverslips for 24 hours. Cells were coimmunostained for N-cadherin and the MHC or connexin43 and cardiac troponin I. Overexpression of CREG1 increased N-cadherin and connexin43 immunofluorescence at myocyte junctions (arrows). (E): Pairs of myocytes with linear staining of N-cadherin or connexin43 at their junctions were counted and plotted as a percentage of total myocyte pairs examined. N = 4 with 91–96 pairs of myocytes for each group. *, p < .001. (F): Mouse neonatal cardiomyocytes were infected with Ad-CREG1 or the control virus for 48 hours. Cells were fixed and immunostained for N-cadherin and cardiac troponin I. In control myocytes, N-cadherin was localized at both cell-cell junctions and intracellularly adjacent to cell-cell contacts (arrows). In Ad-CREG1 infected cells, N-cadherin accumulation at cell-cell junctions was significantly increased (arrows). Abbreviations: Ad-CREG1, adeno-CREG1; CREG1: cellular repressor of E1A-stimulated genes 1; EBs, embryoid bodies; MHC, myosin heavy chain; MYL3, myosin light chain 3. The formation of cohesive myocardium is mainly mediated by cell-cell adhesions between cardiomyocytes. To test whether CREG1 promotes cell-cell adhesions and the formation of intercellular junctions in cardiomyocytes, we isolated cardiomyocytes from attached EBs and cultured them on fibronectin-coated glass coverslips for 24 hours. Immunofluorescence microscopy showed that overexpression of CREG1 markedly increases the accumulation of the adherens junction receptor N-cadherin and the gap junction protein connexin43 at the cell-cell adhesion site (Fig. 2D, 2E). To ensure that CREG1 regulation of the cell-cell interaction is not specific to ES cell-derived cardiomyocytes, we isolated myocytes from 1-day neonatal mouse hearts. Adenovirus-mediated delivery of CREG1 to cardiomyocytes significantly increased the accumulation of N-cadherin at cell-cell junctions as compared to the control virus-infected cells (Fig. 2F). In addition, overexpression of CREG1 in neonatal cardiomyocytes improved cell alignment. Altogether, these results demonstrate that overexpression of CREG1 promotes cardiomyocyte interactions and cell-cell junction formation. Next, we addressed whether CREG1 is required for cardiomyogenic differentiation. The Creg1 gene was disrupted in 129 ES cells by replacing exons 2 and 3 with a neomycin-resistant cassette flanked by 2 flippase recognition target (FRT) sites (will be reported separately). Heterozygous ES cells were selected at high concentrations of G418 for homozygous clones. This was verified by RT-PCR genotyping and immunoblotting (Fig. 3A, 3C). The neomycin-resistant cassette was then removed by transient transfection with a flippase plasmid to facilitate rescue experiments. As shown in Figure 3B and 3C, ablation of Creg1 markedly reduces the percentage of spontaneously beating EBs and the expression of cardiac markers. Continuous monitoring of spontaneous EB contraction revealed that most of the Creg1−/− beating EBs last only for 1 day. This is likely caused by cessation of contraction because no detachment of the marked beating foci was observed (data not shown). By contrast, the wild-type counterparts can contract for 7 days or more (Fig. 3D). Strikingly, Creg1-null myocytes assume an elongated shape and do not form cell-cell adhesions (Fig. 3E). RT-PCR analysis for cardiac- and skeletal muscle-specific markers confirmed that these spindle-shaped cells are indeed cardiomyocytes but not skeletal muscle (Supporting Information Fig. S2). These results suggest that CREG1 is critical for cardiomyocyte differentiation and cell-cell adhesion. Figure 3 Open in new tabDownload slide Ablation of CREG1 inhibits cardiomyocyte differentiation and cohesion. (A): Five-day EBs were genotyped by reverse transcription-polymerase chain reaction (RT-PCR) showing the wild-type and KO bands. (B): Wild-type (+/+) and Creg1-null (−/−) EBs were cultured in suspension for 5 days and attached on dishes for additional 7 days. Spontaneously beating EBs were counted and plotted as a percentage of total EBs examined (n = 5 with a total of 302–376 EBs counted for each group, p < .01). (C): The above EBs were subjected to immunoblot analysis for CREG1 and cardiomyogenic differentiation markers. Actin served as a loading control. Ablation of CREG1 inhibits cardiomyogenic differentiation. (D): Spontaneous beating clusters/foci were marked and monitored by phase microscopy. The duration of beating EBs was plotted. The duration of beating is significantly shorter after CREG1 ablation. 25 beating foci were monitored for each group. N = 4. (E): 5 + 7 day EBs were immunostained for the cardiac MHC. Ablation of CREG1 not only reduces the number of MHC-positive myocytes but also prevents the formation of myocyte clusters. Abbreviations: CREG1: cellular repressor of E1A-stimulated genes 1; EBs, embryoid bodies; KO, knockout; MHC, myosin heavy chain; MYL3, myosin light chain 3. Figure 3 Open in new tabDownload slide Ablation of CREG1 inhibits cardiomyocyte differentiation and cohesion. (A): Five-day EBs were genotyped by reverse transcription-polymerase chain reaction (RT-PCR) showing the wild-type and KO bands. (B): Wild-type (+/+) and Creg1-null (−/−) EBs were cultured in suspension for 5 days and attached on dishes for additional 7 days. Spontaneously beating EBs were counted and plotted as a percentage of total EBs examined (n = 5 with a total of 302–376 EBs counted for each group, p < .01). (C): The above EBs were subjected to immunoblot analysis for CREG1 and cardiomyogenic differentiation markers. Actin served as a loading control. Ablation of CREG1 inhibits cardiomyogenic differentiation. (D): Spontaneous beating clusters/foci were marked and monitored by phase microscopy. The duration of beating EBs was plotted. The duration of beating is significantly shorter after CREG1 ablation. 25 beating foci were monitored for each group. N = 4. (E): 5 + 7 day EBs were immunostained for the cardiac MHC. Ablation of CREG1 not only reduces the number of MHC-positive myocytes but also prevents the formation of myocyte clusters. Abbreviations: CREG1: cellular repressor of E1A-stimulated genes 1; EBs, embryoid bodies; KO, knockout; MHC, myosin heavy chain; MYL3, myosin light chain 3. CREG1 Cell-Autonomously Regulates Cardiomyocyte Differentiation CREG1 primarily resides intracellularly in a diffuse or a punctate pattern but can also be secreted out of the cells [13, 30]. In addition, EB cultures contain many cell types and it is unknown whether the reduced cardiomyocyte differentiation after CREG1 ablation is due to loss of CREG1 in cardiomyocytes/progenitors or in the surrounding cells. If CREG1 acts as a secreted factor and/or loss of CREG1 in the surrounding nonmuscle cells affects cardiomyogenesis, formation of chimeric EBs by mixing the wild-type with Creg1-null ES cells should rescue the mutant phenotype. To test this hypothesis, we generated stable wild-type ES cell clones expressing GFP under the control of the α-myosin heavy chain promoter (αMHC) and Creg1−/− αMHC-mCherry clones. These two types of ES cells were mixed in 1:1 ratio in hanging drops to form chimeric EBs as described previously [26]. Immunofluorescence analysis showed that 84% of cardiomyocytes differentiated in chimeric EBs were GFP-positive wild-type and only 16% were mCherry-expressing Creg1−/− myocytes (Fig. 4B, 4C). We also created chimeric EBs by mixing unlabeled wild-type with Creg1−/− αMHC-mCherry ES cells. Immunostaining for the cardiac myosin heavy chain and mCherry confirmed a reduction in cardiomyocyte numbers derived from Creg1−/− ES cells (Fig. 4D). These results suggest that CREG1 regulates cardiomyogenic differentiation in a cell-autonomous fashion. Environmental cues are unlikely to play a significant role in CREG1 loss-induced cardiomyogenic defects. Figure 4 Open in new tabDownload slide Chimeric analysis. (A): A diagram depicts chimeric EB analysis. (B): WT ES cells stably transfected with GFP under the control of the αMHC were mixed in 1:1 ration with Creg1−/− ES cells transfected with αMHC-mCherry and cultured in hanging drops for 1 day followed by suspension culture for 4 days. The chimeric EBs were then cultured on glass coverslips for 7 days and immunostained for GFP and mCherry. Arrows indicate nuclear accumulation of mCherry in Creg1−/− cardiomyocytes. (C): GFP-positive wild-type and mCherry-positive Creg1−/− myocytes in beating EBs were counted and plotted as a percentage of total fluorescent cells (n = 7, p < .01). (D): 5 + 7 day Creg1−/−-αMHC-mCherry ↔ unlabeled wild-type (1:1 mixing) chimeric EBs were immunostained for cardiac myosin heavy chain and mCherry. Arrows indicate mCherry-positive Creg1−/− cardiomyocytes. Abbreviations: EB, embryoid body; GFP, green fluorescence protein; αMHC, α-myosin heavy chain promoter; WT, wild-type. Figure 4 Open in new tabDownload slide Chimeric analysis. (A): A diagram depicts chimeric EB analysis. (B): WT ES cells stably transfected with GFP under the control of the αMHC were mixed in 1:1 ration with Creg1−/− ES cells transfected with αMHC-mCherry and cultured in hanging drops for 1 day followed by suspension culture for 4 days. The chimeric EBs were then cultured on glass coverslips for 7 days and immunostained for GFP and mCherry. Arrows indicate nuclear accumulation of mCherry in Creg1−/− cardiomyocytes. (C): GFP-positive wild-type and mCherry-positive Creg1−/− myocytes in beating EBs were counted and plotted as a percentage of total fluorescent cells (n = 7, p < .01). (D): 5 + 7 day Creg1−/−-αMHC-mCherry ↔ unlabeled wild-type (1:1 mixing) chimeric EBs were immunostained for cardiac myosin heavy chain and mCherry. Arrows indicate mCherry-positive Creg1−/− cardiomyocytes. Abbreviations: EB, embryoid body; GFP, green fluorescence protein; αMHC, α-myosin heavy chain promoter; WT, wild-type. Visceral endoderm has been shown to induce cardiac differentiation both in vitro and in vivo [31, 32]. To further determine whether CREG1 regulates cardiomyogenesis by exerting its effects on germ layer formation, we analyzed EBs cultured in suspension for 7 days after they differentiated into cysts comprising endoderm and a columnar epiblast epithelium; similar to pregastrulation embryos. Phase-contrast microscopy showed no difference in germ layer formation between wild-type and Creg1−/− EBs (Supporting Information Fig. S3A, S3B). RT-PCR analysis revealed a decrease in the expression of the cardiac progenitor markers Mef2C and Nkx2-5 (Supporting Information Fig. S3C–S3E). These results further support the notion that CREG1 acts in a cell-intrinsic way to regulate differentiation and/or survival of cardiac progenitor cells. CREG1 Interacts with Sec8 To seek proteins that interact with CREG1 which may mediate its cardiomyogenic activity, we performed affinity pull-down assays using mouse embryonic heart lysates and His-tagged CREG1 bound to Ni-NTA agarose beads. Mass spectrometric analysis of silver stained gels identified IGF2R (insulin-like growth factor 2 receptor) and the exocyst components Sec8 and Sec6 as CREG1 binding proteins (Fig. 5A). Coimmunoprecipitation confirmed CREG1 binding to Sec6 and Sec8, which in turn binds to N-cadherin (Fig. 5B). In postnatal day 14 (P14) mouse hearts, CREG1, Sec8, and N-cadherin are all localized at intercalated discs, and to a lesser extent an intracellular compartment (Fig. 5C). Given that the exocyst regulates polarized exocytosis, these results suggest that the CREG1-Sec6/8 interaction may be involved in protein trafficking from intracellular compartments to the cell-cell adhesion site. To determine whether CREG1 directly binds to Sec8 and/or Sec6, we produced recombinant His-tagged CREG1 in HEK293 cells and GST-fused Sec8 and Sec6 in bacteria. In vitro reciprocal pull-down assay demonstrated that CREG1 directly binds to Sec8 but not Sec6 (Fig. 5D, 5E). Therefore, Sec6 may interact with CREG1 indirectly through Sec8. Figure 5 Open in new tabDownload slide CREG1 interacts with Sec8. (A): Homogenates of E15.5 mouse embryonic hearts were subjected to affinity pull-down assays using His-CREG1 bound to Ni-NTA-agarose beads. Mass spectrometry identified that CREG1 interacts with IGF2R, Sec8, and Sec6. (B): Heart homogenates were immunoprecipitated with antibodies to Sce6, Sec8, or control IgG. The immunoprecipitates were analyzed by immunoblotting for CREG1. CREG1 binds to Sec6 and Sec8. Coimmunoprecipitation revealed that Sec8 also binds to N-cadherin. (C): Immunofluorescence micrographs show that CREG1 colocalizes with N-cadherin and Sec8 at intercalated discs between myocytes in P14 mouse hearts (arrows). (D): His-tagged wild-type CREG1 was incubated with GST-tagged Sec6, Sec8, or GST alone for 1 hour and then affinity purified using GSH-agarose beads. Immunoblot analysis using antibodies to His-tag and CREG1 showed that CREG1 directly binds to Sec8 but not Sec6. (E): A reciprocal affinity binding assay confirmed that His-tagged CREG1 binds to the GST-Sec8 fusion protein. (F): A diagram shows CREG1 mutagenesis. (G): GST-Sec8 was incubated with His-tagged wild-type and mutant CREG1 for 1 hour and precipitated with Ni-NTA agarose beads. Bead-bound Sec8 was analyzed by immunoblotting. Deletion of amino acids 141–144 or mutations of both aspartic acid at 141 and proline at 142 to alanine (D141A/P142A) abolished CREG1 binding to Sec8. However, mutations of three N-glycosylation sites to alanine (GM) had no effect. Abbreviations: CREG1: cellular repressor of E1A-stimulated genes 1; DA, aspartic acid to alanine mutation; GM, glycosylation mutant; GSH, glutathione; GST, glutathione S-transferase; IB, immunoblotting; IGF2R, insulin-like growth factor 2 receptor; IP, immunoprecipitation; N-cad, N-cadherin; Ni-NTA, nickel-nitrilotriacetic acid; PA, proline to alanine mutation; WT, wild-type. Figure 5 Open in new tabDownload slide CREG1 interacts with Sec8. (A): Homogenates of E15.5 mouse embryonic hearts were subjected to affinity pull-down assays using His-CREG1 bound to Ni-NTA-agarose beads. Mass spectrometry identified that CREG1 interacts with IGF2R, Sec8, and Sec6. (B): Heart homogenates were immunoprecipitated with antibodies to Sce6, Sec8, or control IgG. The immunoprecipitates were analyzed by immunoblotting for CREG1. CREG1 binds to Sec6 and Sec8. Coimmunoprecipitation revealed that Sec8 also binds to N-cadherin. (C): Immunofluorescence micrographs show that CREG1 colocalizes with N-cadherin and Sec8 at intercalated discs between myocytes in P14 mouse hearts (arrows). (D): His-tagged wild-type CREG1 was incubated with GST-tagged Sec6, Sec8, or GST alone for 1 hour and then affinity purified using GSH-agarose beads. Immunoblot analysis using antibodies to His-tag and CREG1 showed that CREG1 directly binds to Sec8 but not Sec6. (E): A reciprocal affinity binding assay confirmed that His-tagged CREG1 binds to the GST-Sec8 fusion protein. (F): A diagram shows CREG1 mutagenesis. (G): GST-Sec8 was incubated with His-tagged wild-type and mutant CREG1 for 1 hour and precipitated with Ni-NTA agarose beads. Bead-bound Sec8 was analyzed by immunoblotting. Deletion of amino acids 141–144 or mutations of both aspartic acid at 141 and proline at 142 to alanine (D141A/P142A) abolished CREG1 binding to Sec8. However, mutations of three N-glycosylation sites to alanine (GM) had no effect. Abbreviations: CREG1: cellular repressor of E1A-stimulated genes 1; DA, aspartic acid to alanine mutation; GM, glycosylation mutant; GSH, glutathione; GST, glutathione S-transferase; IB, immunoblotting; IGF2R, insulin-like growth factor 2 receptor; IP, immunoprecipitation; N-cad, N-cadherin; Ni-NTA, nickel-nitrilotriacetic acid; PA, proline to alanine mutation; WT, wild-type. As mentioned above, CREG1 appears to interact with IGF2R. It has been previously demonstrated that human CREG1 contains three N-linked glycosylation sites (N160, N193, and N216) and mutation of these sites to alanine abolishes CREG1 binding to IGF2R [14, 15]. In addition, deletion of amino acid residues 141–144 of the loop region following helix α3 greatly reduced its inhibitory effect on cell proliferation [15]. To explore whether these sites are involved in Sec8 binding, we performed mutagenesis analysis. We found that mutation of all three glycosylation sites to alanine does not affect CREG1 binding to Sec8, although it reduces the molecular size of CREG1 from ∼34 kDa to ∼24 kDa (Fig. 5G). However, deletion of residues 141–144 eliminates the CREG1-Sec8 interaction in an in vitro pull-down assay. Among these four residues deleted, D141 and P142 are conserved in Drosophila, mouse, and human [12]. Mutation of these two residues to alanine also prevents CREG1 from binding to Sec8 (Fig. 5G). These results demonstrate a novel interaction between CREG1 and Sec8 mediated by the amino acid residues D141 and P142 in CREG1. CREG1 Promotes Cardiomyogenic Differentiation Through Its Interaction with Sec8 To determine whether CREG1 binding to Sec8 is required for cardiomyogenesis, we stably transfected Creg1−/− ES cells with GFP, wild-type CREG1 or the Sec8-binding mutants (Fig. 6A). Analysis of EB cardiogenesis on day 5 + 7 demonstrated that rescue with wild-type CREG1 markedly increases spontaneously beating EBs and the expression of cardiac marker proteins (Fig. 6B, 6C). By contrast, both Sec8-binding mutants (Δ141–144 and D141A/P142A) show significantly reduced ability to rescue cardiomyogenic differentiation. In addition, wild-type CREG1-rescued cardiomyocytes formed compact and cohesive myocardium-like structures, whereas Creg1−/− myocytes expressing CREG1 Δ141–144 or D141A/P142A were scattered and failed to form cell-cell adhesions (Fig. 6D and data not shown). These results suggest that the CREG1-Sec8 interaction is required for cardiomyogenic differentiation and cohesion. Figure 6 Open in new tabDownload slide CREG binding to Sec8 is required for cardiomyogenic differentiation and the formation of cohesive myocyte clusters. (A): Immunoblots show stable reconstitution of Creg1−/− EBs with wild-type and Sec8-binding mutant CREG1 [Δ141–144 (ΔM) and D141A/P142A]. The expression of endogenous CREG1 is lower than that of exogenous wild-type CREG1 introduced into Creg1−/− EBs. (B): EBs were cultured in suspension for 5 days and allowed to attach on tissue culture dishes for additional 7 days. Spontaneously, beating EBs were counted and plotted as a percentage of total EBs examined (n = 6, with a total of 375–394 EBs counted for each group, p < .01). Rescue of Creg1−/− EBs with wild-type CREG1 significantly increased cardiomyogenic differentiation while the mutants deficient in Sec8 binding had little if any effect. (C): Immunoblot analysis on 5 + 7 day EBs for cardiac markers confirmed rescue of cardiomyogenic differentiation in Creg1−/− EBs reconstituted with wild-type but not Sec8-binding mutant CREG1. (D): Immunostaining for the sarcomeric MHC revealed that wild-type CREG1-rescued cardiomyocytes form compact myocardium-like structures. By contrast, Sec8-binding mutant CREG1 (Δ141–144)-reconstituted cardiomyocytes are spindle shaped and do not form cell-cell contacts. (E): Lysates of isolated cardiomyocytes were immunoprecipitated with Sec8 monoclonal antibody followed by immunoblotting for N-cadherin and Sec8. One tenth of input lysate was subjected to direct immunoblot analysis. Ablation of CREG1 reduced N-cadherin and Sec8 and inhibited the interaction of Sec8 with N-cadherin. Abbreviations: CREG1: cellular repressor of E1A-stimulated genes 1; DA, aspartic acid to alanine mutation; EBs, embryoid bodies; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GFP, green fluorescence protein; IB, immunoblotting; MHC, myosin heavy chain; PA, proline to alanine mutation. Figure 6 Open in new tabDownload slide CREG binding to Sec8 is required for cardiomyogenic differentiation and the formation of cohesive myocyte clusters. (A): Immunoblots show stable reconstitution of Creg1−/− EBs with wild-type and Sec8-binding mutant CREG1 [Δ141–144 (ΔM) and D141A/P142A]. The expression of endogenous CREG1 is lower than that of exogenous wild-type CREG1 introduced into Creg1−/− EBs. (B): EBs were cultured in suspension for 5 days and allowed to attach on tissue culture dishes for additional 7 days. Spontaneously, beating EBs were counted and plotted as a percentage of total EBs examined (n = 6, with a total of 375–394 EBs counted for each group, p < .01). Rescue of Creg1−/− EBs with wild-type CREG1 significantly increased cardiomyogenic differentiation while the mutants deficient in Sec8 binding had little if any effect. (C): Immunoblot analysis on 5 + 7 day EBs for cardiac markers confirmed rescue of cardiomyogenic differentiation in Creg1−/− EBs reconstituted with wild-type but not Sec8-binding mutant CREG1. (D): Immunostaining for the sarcomeric MHC revealed that wild-type CREG1-rescued cardiomyocytes form compact myocardium-like structures. By contrast, Sec8-binding mutant CREG1 (Δ141–144)-reconstituted cardiomyocytes are spindle shaped and do not form cell-cell contacts. (E): Lysates of isolated cardiomyocytes were immunoprecipitated with Sec8 monoclonal antibody followed by immunoblotting for N-cadherin and Sec8. One tenth of input lysate was subjected to direct immunoblot analysis. Ablation of CREG1 reduced N-cadherin and Sec8 and inhibited the interaction of Sec8 with N-cadherin. Abbreviations: CREG1: cellular repressor of E1A-stimulated genes 1; DA, aspartic acid to alanine mutation; EBs, embryoid bodies; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GFP, green fluorescence protein; IB, immunoblotting; MHC, myosin heavy chain; PA, proline to alanine mutation. To test whether CREG1 promotes the assembly of adherens junctions by regulating the binding of Sec8 to N-cadherin, we performed coimmunoprecipitation on wild-type and Creg1−/− cardiomyocytes. N-cadherin is readily detected in Sec8 immunoprecipitates of the wild-type cardiomyocytes but it is absent from the Creg1−/− myocytes (Fig. 5E). Direct immunoblot analysis demonstrated a sixfold decrease of Sec8 and a twofold decrease of N-cadherin in the absence of CREG1. These results suggest that CREG1 binding to Sec8 promotes the interaction of Sec8 with E-cadherin and enhances the stability of these two proteins. Knockdown of Sec8 Causes N-Cadherin Degradation and Inhibits Cardiomyocyte Differentiation and Cohesion Sec8 is a subunit of the exocyst complex, which is recruited to nascent sites of cell-cell adhesions and is important for the assembly of intercellular junctions in epithelial cells [33]. However, the role of the exocyst complex and its subunits in cardiomyocytes has not been reported. Since CREG1 binding to Sec8 is required for EB cardiomyogenesis, depletion of Sec8 is expected to suppress cardiomyogenic differentiation. To test this hypothesis, we stably transfected ES cells with shRNAs targeting mouse Sec8 (Fig. 7A). Knockdown of Sec8 in 5 + 7 day EBs reduced the protein but not the mRNA level of the exocyst subunit Sec6, suggesting that the stability of Sec6 depends on Sec8 and they may exist in the same complex (Fig. 7A and Supporting Information Fig. S4). Given that Sec8 also interacted with N-cadherin in coimmunoprecipitation assay (Fig. 5B), we analyzed N-cadherin by immunoblotting. We observed a significant reduction of intact N-cadherin and increased degradation products in Sec8 knockdown EBs (Fig. 7B). However, its mRNA level is significantly increased (Supporting Information Fig. S4), which may be compensation for accelerated protein degradation. In GFP or the scrambled shRNA controls, foci of spontaneous contraction were observed in ∼40% attached EBs, whereas depletion of Sec8 reduces the number of beating EBs to 18% of the control level (Fig. 7C). Immunoblot analysis revealed decreased expression of cardiac marker proteins after Sec8 depletion, confirming that cardiomyogenic differentiation is inhibited (Fig. 7D). Similar to Creg1−/− EBs, the duration of spontaneous contraction of Sec8 knockdown EBs is usually 1 day whereas beating of control EBs persists for 7 days or more (Fig. 7E). Furthermore, depletion of Sec8 inhibited cell-cell adhesions between cardiomyocytes and the formation of cohesive myocyte clusters (Fig. 7F). These results suggest that Sec8 plays a critical role in cardiomyogenic differentiation and cell-cell adhesions between myocytes. Figure 7 Open in new tabDownload slide Knockdown of Sec8 inhibits cardiomyogenic differentiation. (A): 5 + 7 day EBs stably transfected with GFP or Sec8 shRNAs were analyzed by immunoblotting. Actin served as a loading control. Densitometry showed that Sec8 was reduced by 95% in clone D1 compared with the GFP control. KD of Sec8 reduced Sec6 levels. (B): Immunoblotting showed that shRNA-mediated depletion of Sec8 results in N-cadherin degradation. (C): Sec8 KD (clone D1) led to a significant reduction of spontaneously beating EBs compared with the GFP control (n = 4 with a total of 400 EBs counted for each group, p < .01). (D): 5 + 7 day EBs were analyzed by immunoblotting for the expression of cardiac markers. Sec8 KD inhibited cardiomyogenic differentiation. (E): Spontaneous beating foci were monitored by phase microscopy and the duration of beating EBs was plotted. The duration of beating is significantly shorter after depletion of Sec8. Twenty-five beating foci were monitored for each group. N = 4. (F): 5 + 7 day EBs were immunostained for the myosin heavy chain. Sec8 KD inhibited intercellular adhesions between cardiomyocytes. Abbreviations: EB, embryoid body; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GFP, green fluorescence protein; KD, knockdown; MHC, myosin heavy chain; MYL3, myosin light chain 3. Figure 7 Open in new tabDownload slide Knockdown of Sec8 inhibits cardiomyogenic differentiation. (A): 5 + 7 day EBs stably transfected with GFP or Sec8 shRNAs were analyzed by immunoblotting. Actin served as a loading control. Densitometry showed that Sec8 was reduced by 95% in clone D1 compared with the GFP control. KD of Sec8 reduced Sec6 levels. (B): Immunoblotting showed that shRNA-mediated depletion of Sec8 results in N-cadherin degradation. (C): Sec8 KD (clone D1) led to a significant reduction of spontaneously beating EBs compared with the GFP control (n = 4 with a total of 400 EBs counted for each group, p < .01). (D): 5 + 7 day EBs were analyzed by immunoblotting for the expression of cardiac markers. Sec8 KD inhibited cardiomyogenic differentiation. (E): Spontaneous beating foci were monitored by phase microscopy and the duration of beating EBs was plotted. The duration of beating is significantly shorter after depletion of Sec8. Twenty-five beating foci were monitored for each group. N = 4. (F): 5 + 7 day EBs were immunostained for the myosin heavy chain. Sec8 KD inhibited intercellular adhesions between cardiomyocytes. Abbreviations: EB, embryoid body; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GFP, green fluorescence protein; KD, knockdown; MHC, myosin heavy chain; MYL3, myosin light chain 3. Discussion In this work, we demonstrate that CREG1 is a novel regulator of cardiomyogenic differentiation and cell-cell adhesion. CREG1 overexpression promotes ES cell differentiation into cardiomyocytes, whereas its knockout has inhibitory effects. Chimeric analysis reveals that CREG1 promotes cardiomyogenesis in a cell-autonomous fashion. Furthermore, we identified a new protein interaction between CREG1 and the exocyst subunit Sec8. Disruption of this interaction abolishes CREG1-induced cardiomyogenic differentiation and cell-cell adhesions. These results uncover a novel mechanism of cardiomyogenesis whereby CREG1 binding to the exocyst regulates N-cadherin stability and cell-cell adhesions. CREG1 is a mysterious glycoprotein protein that contains a signal sequence at its N-terminus necessary for entry into the endoplasmic reticulum (ER) and the Golgi apparatus [13]. It can be detected in the cell culture medium and the signal sequence is required for secretion. CREG1 conditioned media can induce differentiation of NTERA-2 teratocarcinoma cells, suggesting that CREG1 may act as a soluble factor to exert its biological effects [13]. To test this hypothesis in EB cardiomyogenesis, we performed chimeric EB analysis by mixing the same number of genetically labeled wild-type and Creg1−/− ES cells. We found that wild-type cells cannot rescue cardiomyogenic differentiation of Creg1−/− ES cells. A caveat of the chimeric assay is fusion of wild-type and mutant cells. It has been reported that mesenchymal stem cells (bone marrow-derived cells are prone to fusion) can fuse with cardiomyocytes after injected into the myocardium albeit in low frequencies [34-38]. In our chimeric cultures, Creg1−/− cells may fuse with wild-type cells at the ES aggregate and/or the epiblast stage. To address this issue, we mixed wild-type ES cells expressing GFP under the control of the PGK promoter (high GFP expression in ES cells and epiblast) with Creg1−/− ES cells stably transfected with PGK-FRP in hanging drops in 1:1 ratio. After 24 hours, the chimeric EBs were cultured in suspension for 5 days. Five-micrometer-thick cryosections were cut from different blocks. Fluorescent microscopy revealed that 0.36% of epiblast cells (from which cardiomyocytes are derived) were labeled with both GFP and RFP (with a total of 1,475 cells were counted). Even if cell fusion occurs at a low frequency in the chimeric culture, acquiring the wild-type Creg1 alleles in the fusion cells should rescue the cardiomyogenic defect. However, mCherry-positive cardiomyocytes were much fewer than the wild-type in both types of chimeric cultures. This result argues for a cell-autonomous mechanism in cardiomyocytes and/or their progenitors. Recently, CREG1 has also been found to reside in lysosomes where it is processed by cathepsin-mediated cleavage in fibroblasts [30]. Additionally, it has been colocalized with its putative receptor IGF2R on the plasma membrane in epithelial cells and smooth muscle [39]. In the present study, CREG1 was observed mainly at cell-cell junctions and an intracellular vesicular compartment in cardiomyocytes. Despite these spatial variations, CREG1 is clearly associated with intracellular membrane compartments and/or the plasma membrane, suggesting that it may be involved in intracellular trafficking. To explore the mechanisms of cardiomyogenesis regulated by CREG1, we performed affinity pull-down assay followed by mass spectrometry and identified Sec8/Exoc4 of the exocyst as a novel binding partner of CREG1. The interaction between CREG1 and Sec8 was confirmed by coimmunoprecipitation, colocalization and in vitro binding assay using recombinant proteins. Furthermore, site-directed mutagenesis revealed that the amino acid residues D141 and P142 in CREG1 are required for Sec8 binding. These two amino acid residues are conserved in Drosophila, Xenopus, mouse, and human and are in a loop region that covers the putative flavin mononucleotide (FMN)-binding site [15]. In this study, we reconstituted Creg1−/− ES cells with wild-type CREG1 or the Sec8-binding mutants (Δ141–144 and D141A/P142A). Analysis of EB cardiomyogenesis suggests that CREG1 binding to Sec8 is required for cardiomyogenic differentiation and the formation of cohesive myocardium-like structures. This notion is supported by the Sec8 knockdown experiment demonstrating a reduction in the expression of cardiac markers and failure to form cell-cell adhesions. The results of our mutagenesis analysis corroborate with the previous finding that deletion of CREG1 residues 141–144 abolishes its growth inhibitory activity in NTERA-2 teratocarcinoma cells [15]. Whether this growth inhibitory effect is also mediated by the CREG1-Sec8 interaction warrants further investigation. How the CREG1-Sec8 interaction regulates cardiomyocyte junction formation remains unknown. Sec8 is a subunit of the exocyst which is an evolutionarily conserved octameric protein complex consisting of Sec3, Sec5, Sec6, Sec8, Sec10, Sec15, Exo70, and Exo84. The exocyst is localized to intracellular vesicles and specific regions of the plasma membrane. It mediates the delivery of lipids and proteins necessary for polarized membrane growth and for the formation and remodeling of junctional complexes [40-43]. In Madin-Darby canine kidney (MDCK) epithelial cells, the exocyst complex is recruited to nascent sites of cell-cell contact in response to E-cadherin-mediated adhesive interactions. This event is an important early step in the assembly of intercellular junctions [44]. In both Drosophila and mammals, the exocyst subunit Sec15 has also been shown to interact with the small GTPase Rab11 in endosomes [45-49]. This interaction appears to be important for endocytic recycling in epithelial cells and plays an essential role in adherens junction assembly and dynamics [50, 51]. In contrast to Drosophila and mammalian epithelial cells, the exocyst complex has not been characterized in cardiomyocytes. In the present study, we show that both CREG1 and Sec8 are colocalized with N-cadherin to intercalated discs and an intracellular vesicular compartment. Sec8 binds to N-cadherin in coimmunoprecipitation and shRNA-mediated depletion of Sec8 leads to degradation of N-cadherin and impaired cell-cell adhesions. These results suggest that Sec8 may be required for the targeting of N-cadherin to cell-cell adhesion sites for adherens junction assembly and/or maintenance during cardiomyogenesis. We have further shown that knockout of CREG1 prevents Sec8 binding to N-cadherin. Therefore, it appears that the interaction of CREG1 with Sec8 promote its binding to N-cadherin and facilitate their targeting to cell-cell junctions. However, clear evidence of CREG1 localization to the exocyst to tether N-cadherin-containing vesicles to the cell-cell junction is needed to support the notion that CREG1 promotes adherens junction assembly by regulating exocyst-mediated N-cadherin trafficking to the cell-cell junction. A new technology using ultrasensitive TIRFM (total internal reflection fluorescence microscopy) to visualize exocyst-mediated vesicle tethering to and fusion with the plasma membrane has enabled us to address this issue in the future [52]. In this study, we provide experimental evidence that CREG1 is critical for cardiomyogenic differentiation as well as the formation of adherens and gap junctions. It remains unclear, however, what the relationship is between the formation of intercellular junctions and the expression of cardiomyogenic markers. In the absence of CREG1, the transcription factors Mef2C and Nkx2-5 expressed in cardiac progenitor cells are reduced, suggesting that CREG1 regulates the population of cardiac progenitor cells although CREG1 is not a transcription factor per se. One possibility is that CREG1 may be required for the survival of cardiac progenitor cells. Indeed, increased caspase-3 activation was observed in Creg1−/− cardiomyocytes when cultured on laminin or fibronectin substrate (Supporting Information Fig. 5). This result suggests that apoptosis may contribute, at least in part, to the reduced expression of cardiac transcription factors and myogenic markers during the differentiation of Creg1−/− EBs. Another possibility is that CREG1 may regulate cardiomyogenesis through N-cadherin-mediated cell-cell adhesion. Early studies have demonstrated that N-cadherin is required for skeletal muscle differentiation [53, 54]. In cultured chicken heart-forming mesoderm, N-cadherin neutralizing antibody suppressed the expression of sarcomeric actinin and myosin [55]. A recent report has shown that N-cadherin in highly expressed in cell-cell junctions of the cardiac progenitor cells in the mouse anterior heart field [56]. Conditional ablation of N-cadherin in the anterior heart field significantly inhibited the proliferation and expansion of the cardiac progenitor cells. Whether CREG1 regulates the expansion of cardiac progenitor cells through N-cadherin-mediated cell-cell interaction warrants further investigation. Conclusion In summary, we demonstrated that CREG1 promotes the differentiation of ES cells to cardiomyocytes and cultivates the formation of cohesive myocardium-like structures by gain- and loss-of-function analyses. Of note, we identified a novel interaction between CREG1 and the exocyst protein Sec8. Site-directed mutagenesis and rescue experiments revealed an essential role of the CREG1-Sec8 interaction in EB cardiomyogenesis. Analysis of cardiomyocytes isolated from CREG1-overexpressing EBs suggests that CREG1 promotes the assembly of adherens and gap junctions. In addition, CREG1 is required for the binding of Sec8 to and the stability of the adherens junction receptor N-cadherin. These findings provide new insights into the biological function of CREG1 and exocyst regulation. They are important to our understanding of the formation of intercalated discs that connect myocytes in the myocardium. Acknowledgements This work was supported by a grant to Yaling Han from the National Natural Science Foundation of China (81130072) and a Grant-in-Aid from American Heart Association to Shaohua Li. J.L., Y.Q., and S.L. contributed equally to this work. Author Contributions J.L., C.Y., and X.T.: collection and assembly of data, data analysis and interpretation; Y.Q.: collection and assembly of data, data analysis and interpretation, final approval of the manuscript; S.L.: conception and design, data analysis and interpretation, manuscript writing, final approval of the manuscript, financial support; S.-C.H.: provision of study materials; S.S. and J.H.: collection and assembly of data, manuscript writing; S.A.R.: administrative support, final approval of the manuscript; L.Y.L.: administrative and financial support, final approval of the manuscript; Y.H.: data analysis and interpretation, final approval of the manuscript, financial support. Disclosure of Potential Conflicts of Interest The authors indicate no potential conflicts of interest. References 1 Laflamme MA , Chen KY, Naumova AV et al. Cardiomyocytes derived from human embryonic stem cells in pro-survival factors enhance function of infarcted rat hearts . Nat Biotechnol 2007 ; 25 : 1015 – 1024 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Dai W , Field LJ, Rubart M et al. Survival and maturation of human embryonic stem cell-derived cardiomyocytes in rat hearts . J Mol Cell Cardiol 2007 ; 43 : 504 – 516 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Laflamme MA , Gold J, Xu C et al. Formation of human myocardium in the rat heart from human embryonic stem cells . Am J Pathol 2005 ; 167 : 663 – 671 . Google Scholar Crossref Search ADS PubMed WorldCat 4 van Laake LW , Passier R, Monshouwer-Kloots J et al. Human embryonic stem cell-derived cardiomyocytes survive and mature in the mouse heart and transiently improve function after myocardial infarction . Stem Cell Res 2007 ; 1 : 9 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Passier R , van Laake LW, Mummery CL. Stem-cell-based therapy and lessons from the heart . Nature 2008 ; 453 : 322 – 329 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Radice GL , Rayburn H, Matsunami H et al. Developmental defects in mouse embryos lacking N-cadherin . Dev Biol 1997 ; 181 : 64 – 78 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Kostetskii I , Moore R, Kemler R et al. Differential adhesion leads to segregation and exclusion of N-cadherin-deficient cells in chimeric embryos . Dev Biol 2001 ; 234 : 72 – 79 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Luo Y , Radice GL. Cadherin-mediated adhesion is essential for myofibril continuity across the plasma membrane but not for assembly of the contractile apparatus . J Cell Sci 2003 ; 116 : 1471 – 1479 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Kostetskii I , Li J, Xiong Y et al. Induced deletion of the N-cadherin gene in the heart leads to dissolution of the intercalated disc structure . Circ Res 2005 ; 96 : 346 – 354 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Li J , Patel VV, Kostetskii I et al. Cardiac-specific loss of N-cadherin leads to alteration in connexins with conduction slowing and arrhythmogenesis . Circ Res 2005 ; 97 : 474 – 481 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Shaw RM , Fay AJ, Puthenveedu MA et al. Microtubule plus-end-tracking proteins target gap junctions directly from the cell interior to adherens junctions . Cell 2007 ; 128 : 547 – 560 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Veal E , Eisenstein M, Tseng ZH et al. A cellular repressor of E1A-stimulated genes that inhibits activation by E2F . Mol Cell Biol 1998 ; 18 : 5032 – 5041 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Veal E , Groisman R, Eisenstein M et al. The secreted glycoprotein CREG enhances differentiation of NTERA-2 human embryonal carcinoma cells . Oncogene 2000 ; 19 : 2120 – 2128 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Di Bacco A , Gill G. The secreted glycoprotein CREG inhibits cell growth dependent on the mannose-6-phosphate/insulin-like growth factor II receptor . Oncogene 2003 ; 22 : 5436 – 5445 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Sacher M , Di Bacco A, Lunin VV et al. The crystal structure of CREG, a secreted glycoprotein involved in cellular growth and differentiation . Proc Natl Acad Sci USA 2005 ; 102 : 18326 – 18331 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Han YL , Liu HW, Kang J et al. Expression of cellular repressor of E1A-stimulated genes in vascular smooth muscle cells of different phenotypes . Zhonghua Yi Xue Za Zhi 2005 ; 85 : 49 – 53 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 17 Han Y , Deng J, Guo L et al. CREG promotes a mature smooth muscle cell phenotype and reduces neointimal formation in balloon-injured rat carotid artery . Cardiovasc Res 2008 ; 78 : 597 – 604 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Li S , Yurchenco PD. Matrix assembly, cell polarization, and cell survival: Analysis of peri-implantation development with cultured embryonic stem cells . Methods Mol Biol 2006 ; 329 : 113 – 125 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 19 Hsu SC , Hazuka CD, Roth R et al. Subunit composition, protein interactions, and structures of the mammalian brain sec6/8 complex and septin filaments . Neuron 1998 ; 20 : 1111 – 1122 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Han Y , Guo L, Yan C et al. Adenovirus-mediated intra-arterial delivery of cellular repressor of E1A-stimulated genes inhibits neointima formation in rabbits after balloon injury . J Vasc Surg 2008 ; 48 : 201 – 209 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Kita-Matsuo H , Barcova M, Prigozhina N et al. Lentiviral vectors and protocols for creation of stable hESC lines for fluorescent tracking and drug resistance selection of cardiomyocytes . PLoS One 2009 ; 4 : e5046 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Vega IE , Hsu SC. The exocyst complex associates with microtubules to mediate vesicle targeting and neurite outgrowth . J Neurosci 2001 ; 21 : 3839 – 3848 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Zhang X , Bi E, Novick P et al. Cdc42 interacts with the exocyst and regulates polarized secretion . J Biol Chem 2001 ; 276 : 46745 – 46750 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Simpson P , Savion S. Differentiation of rat myocytes in single cell cultures with and without proliferating nonmyocardial cells. Cross-striations, ultrastructure, and chronotropic response to isoproterenol . Circ Res 1982 ; 50 : 101 – 116 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Merle PL , Feige JJ, Verdetti J. Basic fibroblast growth factor activates calcium channels in neonatal rat cardiomyocytes . J Biol Chem 1995 ; 270 : 17361 – 17367 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Qi Y , Liu J, Wu X et al. Cdc42 controls vascular network assembly through protein kinase Ciota during embryonic vasculogenesis . Arterioscler Thromb Vasc Biol 2011 ; 31 : 1861 – 1870 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Li S , Harrison D, Carbonetto S et al. Matrix assembly, regulation, and survival functions of laminin and its receptors in embryonic stem cell differentiation . J Cell Biol 2002 ; 157 : 1279 – 1290 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Moretti A , Caron L, Nakano A et al. Multipotent embryonic isl1+ progenitor cells lead to cardiac, smooth muscle, and endothelial cell diversification . Cell 2006 ; 127 : 1151 – 1165 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Wu SM , Chien KR, Mummery C. Origins and fates of cardiovascular progenitor cells . Cell 2008 ; 132 : 537 – 543 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Schahs P , Weidinger P, Probst OC et al. Cellular repressor of E1A-stimulated genes is a bona fide lysosomal protein which undergoes proteolytic maturation during its biosynthesis . Exp Cell Res 2008 ; 314 : 3036 – 3047 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Mummery C , Ward-van Oostwaard D, Doevendans P et al. Differentiation of human embryonic stem cells to cardiomyocytes: Role of coculture with visceral endoderm-like cells . Circulation 2003 ; 107 : 2733 – 2740 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Nijmeijer RM , Leeuwis JW, DeLisio A et al. Visceral endoderm induces specification of cardiomyocytes in mice . Stem Cell Res 2009 ; 3 : 170 – 178 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Andersen NJ , Yeaman C. Sec3-containing exocyst complex is required for desmosome assembly in mammalian epithelial cells . Mol Biol Cell 2010 ; 21 : 152 – 164 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Alvarez-Dolado M , Pardal R, Garcia-Verdugo JM et al. Fusion of bone-marrow-derived cells with Purkinje neurons, cardiomyocytes and hepatocytes . Nature 2003 ; 425 : 968 – 973 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Andrade J , Lam JT, Zamora M et al. Predominant fusion of bone marrow-derived cardiomyocytes . Cardiovasc Res 2005 ; 68 : 387 – 393 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Noiseux N , Gnecchi M, Lopez-Ilasaca M et al. Mesenchymal stem cells overexpressing Akt dramatically repair infarcted myocardium and improve cardiac function despite infrequent cellular fusion or differentiation . Mol Ther 2006 ; 14 : 840 – 850 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Nygren JM , Liuba K, Breitbach M et al. Myeloid and lymphoid contribution to non-haematopoietic lineages through irradiation-induced heterotypic cell fusion . Nat Cell Biol 2008 ; 10 : 584 – 592 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Freeman BT , Kouris NA, Ogle BM. Tracking fusion of human mesenchymal stem cells after transplantation to the heart . Stem Cells Transl Med 2015 ; 4 : 685 – 694 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Gordon PV , Paxton JB, Fox NS. The cellular repressor of E1A-stimulated genes mediates glucocorticoid-induced loss of the type-2 IGF receptor in ileal epithelial cells . J Endocrinol 2005 ; 185 : 265 – 273 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Hazuka CD , Foletti DL, Hsu SC et al. The sec6/8 complex is located at neurite outgrowth and axonal synapse-assembly domains . J Neurosci 1999 ; 19 : 1324 – 1334 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Lalli G , Hall A. Ral GTPases regulate neurite branching through GAP-43 and the exocyst complex . J Cell Biol 2005 ; 171 : 857 – 869 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Lipschutz JH , Guo W, O'brien LE et al. Exocyst is involved in cystogenesis and tubulogenesis and acts by modulating synthesis and delivery of basolateral plasma membrane and secretory proteins . Mol Biol Cell 2000 ; 11 : 4259 – 4275 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Matern HT , Yeaman C, Nelson WJ et al. The Sec6/8 complex in mammalian cells: Characterization of mammalian Sec3, subunit interactions, and expression of subunits in polarized cells . Proc Natl Acad Sci USA 2001 ; 98 : 9648 – 9653 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Yeaman C , Grindstaff KK, Nelson WJ. Mechanism of recruiting Sec6/8 (exocyst) complex to the apical junctional complex during polarization of epithelial cells . J Cell Sci 2004 ; 117 : 559 – 570 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Zhang XM , Ellis S, Sriratana A et al. Sec15 is an effector for the Rab11 GTPase in mammalian cells . J Biol Chem 2004 ; 279 : 43027 – 43034 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Wu S , Mehta SQ, Pichaud F et al. Sec15 interacts with Rab11 via a novel domain and affects Rab11 localization in vivo . Nat Struct Mol Biol 2005 ; 12 : 879 – 885 . Google Scholar Crossref Search ADS PubMed WorldCat 47 Guichard A , McGillivray SM, Cruz-Moreno B et al. Anthrax toxins cooperatively inhibit endocytic recycling by the Rab11/Sec15 exocyst . Nature 2010 ; 467 : 854 – 858 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Das A , Guo W. Rabs and the exocyst in ciliogenesis, tubulogenesis and beyond . Trends Cell Biol 2011 ; 21 : 383 – 386 . Google Scholar Crossref Search ADS PubMed WorldCat 49 Takahashi S , Kubo K, Waguri S et al. Rab11 regulates exocytosis of recycling vesicles at the plasma membrane . J Cell Sci 2012 ; 125 : 4049 – 4057 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 50 Linford A , Yoshimura S, Nunes Bastos R et al. Rab14 and its exchange factor FAM116 link endocytic recycling and adherens junction stability in migrating cells . Dev Cell 2012 ; 22 : 952 – 966 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Ivanov AI , Naydenov NG. Dynamics and regulation of epithelial adherens junctions: Recent discoveries and controversies . Int Rev Cell Mol Biol 2013 ; 303 : 27 – 99 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Rivera-Molina F , Toomre D. Live-cell imaging of exocyst links its spatiotemporal dynamics to various stages of vesicle fusion . J Cell Biol 2013 ; 201 : 673 – 680 . Google Scholar Crossref Search ADS PubMed WorldCat 53 Redfield A , Nieman MT, Knudsen KA. Cadherins promote skeletal muscle differentiation in three-dimensional cultures . J Cell Biol 1997 ; 138 : 1323 – 1331 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Goichberg P , Geiger B. Direct involvement of N-cadherin-mediated signaling in muscle differentiation . Mol Biol Cell 1998 ; 9 : 3119 – 3131 . Google Scholar Crossref Search ADS PubMed WorldCat 55 Imanaka-Yoshida K , Knudsen KA, Linask KK. N-cadherin is required for the differentiation and initial myofibrillogenesis of chick cardiomyocytes . Cell Motil Cytoskeleton 1998 ; 39 : 52 – 62 . Google Scholar Crossref Search ADS PubMed WorldCat 56 Soh BS , Buac K, Xu H et al. N-cadherin prevents the premature differentiation of anterior heart field progenitors in the pharyngeal mesodermal microenvironment . Cell Res 2014 ; 24 : 1420 – 1432 . Google Scholar Crossref Search ADS PubMed WorldCat Author notes Jie Liu, Yanmei Qi and Shaohua Li contributed equally to this work. © 2016 AlphaMed Press This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - CREG1 Interacts with Sec8 to Promote Cardiomyogenic Differentiation and Cell-Cell Adhesion
JF - Stem Cells
DO - 10.1002/stem.2434
DA - 2016-11-01
UR - https://www.deepdyve.com/lp/oxford-university-press/creg1-interacts-with-sec8-to-promote-cardiomyogenic-differentiation-61pH5iv07G
SP - 2648
EP - 2660
VL - 34
IS - 11
DP - DeepDyve
ER -