JAK2/STAT3 Directs Cardiomyogenesis Within Murine Embryonic Stem Cells In Vitro
http://www.100md.com
《干细胞学杂志》
a Department of Cell Biology, Georgetown University Medical Center, Washington, DC, USA;
b Thomas Jefferson School of Technology, Alexandria, VA, USA
Key Words. Embryonic stem cells ? Cardiac development ? Signal transduction ? JAK2/STAT3
Correspondence: G. Ian Gallicano, Ph.D., Georgetown University Medical Institute, Department of Cell Biology, 3900 Reservoir Road NW, Room NE203, Washington, DC 20007, USA. Telephone: 202-687-0228; Fax: 202-687-1823; e-mail: gig@georgetown.edu
ABSTRACT
The first organ to form in mammals is the heart. In murine development, at approximately E7.5 to E8.0, distinct mesodermal cells form bilateral cardiac progenitor cells that proceed to fuse and form a primitive heart tube . During tube formation, distinctive beating of cardiac cells within the developing heart begins at approximately E8.0 . Although overall molecular control of heart tube morphogenesis has been investigated in great detail both in vivo and in vitro, few investigations exist with regard to the initial regulatory mechanisms that differentiate progenitor cells into beating cardiomyocytes. Some obvious reasons for this lack of information have been the difficulty of procuring mammalian embryos at precardiomyocyte stages as well as the difficulty in analyzing the few cardiomyocytes available from early heart tubes. Recently, however, a cardiac in vitro model system was developed using embryonic stem (ES) cells. Manipulating ES cells to form embryoid bodies (EBs) that consistently give rise to beating cardiomyocytes has been shown to be relatively straight-forward and highly reproducible . More important, the electrophysiology of these beating cells has been documented and verified as closely mimicking cardiomyocyte electrophysiology .
To determine the signal transduction pathways that drive cardiomyocyte differentiation, we subjected lysates from beating and nonbeating areas of EBs (procured using a microsurgical protocol similar to the one we previously described ) to a kinase expression screen. Out of 75 kinases tested, JAK2 demonstrated the highest level of expression (>70%) within beating ES-derived cardiomyocytes compared with nonbeating ES cells.
JAK2 is a member of the Janus kinase family, which includes JAK1-3 and TYK2. A primary function of these kinases is to regulate gene transcription by phosphorylating a family of transcription factors called signal transducers and activators of transcription (STATs). Seven STAT family members exist, all of which have been investigated in many in vitro and in vivo systems, including knockout mouse models . Members of this signal transduction pathway have been implicated as a major constituent of adult cardiac pathophysiology, including pressure overload–induced cardiac hypertrophy , increased sensitivity to inflammation, cardiac fibrosis, and heart failure , as well as a paradoxical cardioprotection by ischemic preconditioning and cardiac dysfunction induced by ischemia/reperfusion .
The conventional knockout of JAK2 resulted in embryonic lethality at embryonic days 11–12 (E11–E12) due to a significant reduction in erythropoiesis . Also observed was a significant delay in heart morphogenesis , but this phenotype was not explored in depth in those knockout embryos.
Ablation of STAT3 in mice led to embryonic lethality at E6.5–E7.0, which is, coincidentally, immediately before cardiomyocyte formation . Moreover, the heart-specific STAT3 knockouts produced by conventional cre/lox technology led to higher susceptibility to heart failure in juvenile mice . It must be noted, however, that STAT3 was not analyzed during cardiomyocyte differentiation in the study by Jacoby et al. because the cre in the heart-specific knockout was driven by the -myosin heavy chain promoter, which is expressed at low levels before birth, followed by higher levels long after cardiomyocyte differentiation.
Consequently, based on our kinase expression screen implicating JAK2, the JAK2/STAT3 knockout data , and the evidence that JAK2/STAT3 is clearly involved in remodeling diseased adult cardiomyocytes, we hypothesized that distinct regulation of specific components must occur during initial modeling of cardiac progenitor cells, driving them down the cardiomyocyte differentiation pathway.
To test this hypothesis, specific stages of EB differentiation were analyzed by using highly specific molecular, biochemical, pharmacological, and microscopic analyses. We found that the JAK2/STAT3 pathway is essential for proper gene expression and development of beating cardiomyocytes from ES cells.
MATERIALS AND METHODS
JAK2 Expression and Activity Is Elevated in Beating Areas
Comprehensive analysis of beating and nonbeating areas from CCE EBs was determined molecularly by RT-PCR using primers specific for cardiac genes (data not shown; see for similar results in EBs derived from R1 ES cells). Beating areas always expressed all cardiac-specific genes tested, whereas expression of cardiac-specific genes from nonbeating areas was highly variable . The raw data from our proteomic screen (Kinexus Inc., Vancouver, Canada) revealed a dramatic increase of JAK2 in beating areas compared with nonbeating areas (Figs. 1A–C). Western blot analyses in our laboratory verified the kinase expression data from the Kinexus Inc. proteomics report (Fig. 1D).
Figure 1. Graphs show raw and normalized data taken directly from the Kinexus protein kinase screen. (A): Kinexus Inc. (http://www.kinexus.ca/) uses a highly sensitive imaging system with a 16-bit camera (Bio-Rad Fluor-S Max Multi-Imager) in combination with quantitation software (Bio-Rad Quantity One) to quantify and analyze the chemiluminescent samples. The resulting trace quantity for each band scanned at the maximum scan time is termed the raw data. Because the relationship between scan time and band intensity is linear over the quantifiable range of the signal intensity, the raw data from the scans are normalized to 60 seconds (counts per minute ) for uniformity. (B):. After normalization, data are converted to percentages by subtracting the control (nonbeating) normalized CPM from the experimental (beating) normalized CPM, followed by dividing the difference by the control (nonbeating) normalized CPM and multiplying by 100. (C): The actual JAK2 bands from the Kinexus report show a stronger signal in beating areas compared with nonbeating areas. The red lines in (C) are generated and used by the computer to find the correct bands. (D): Our follow-up Western confirmed the Kinexus data. A ratio of JAK2 protein levels was calculated by normalizing to tubulin (from the same blot), showing that there is 1.6 times (65%) more JAK2 in beating areas than nonbeating areas. (E, F): Confocal microscopy of EBs showed higher levels of JAK2 in beating areas, represented by (F) anticardiac troponin, compared with nonbeating areas subjected to anti-JAK2. Arrows point to JAK2 on cell membranes. (G): Confocal microscopy of EBs clearly showed high amounts of activated pJAK2 on membranes (arrows) in cardiac troponin–positive cells. (H): Western blot shows clear increase in active JAK2 by two times in beating areas. Hela cells serve as a positive control. Scale bar in (E) = 25 μm. Abbreviations: B, beating; NB, nonbeating.
JAK2 localization in beating areas was additionally resolved using confocal microscopy. Antibodies directed against JAK2 or its physiologically active counterpart, phosphorylated JAK2 (pJAK2) , followed by fluorescently labeled secondary antibodies, resulted in a strong, highly specific signal in cells residing in beating areas (Figs. 1E–G). Confocal microscopy clearly showed distinct staining for JAK2 and pJAK2 on or near the plasma membrane in newly differentiated, cardiac troponin T (Tnnt2)–positive, beating cardiomyocytes (~12–18 hours after onset of beating; Figs. 1E–G) . This high level of staining at or near the membrane was rarely seen in cells residing in nonbeating areas. Some areas that were clearly not beating occasionally stained brightly for JAK2; however, the staining pattern in most of these cells was diffuse and cytoplasmic. pJAK2 also was occasionally observed in nonbeating areas (Figs. 1E–G). Western blot analysis supported our confocal data, because pJAK2 was found to be highly elevated in beating areas compared with nonbeating areas (Fig. 1H), additionally supporting the fact that JAK2 was indeed active in beating areas.
JAK2 Inhibition and Activation Alters Cardiomyocyte Differentiation
To determine the physiological relevance of JAK2 during cardiomyogenesis, we first inhibited JAK2 function using the relatively specific pharmacological agent to JAK2, AG490. EBs were subjected to medium containing various concentrations of AG490, including a concentration known to be specific for JAK2. Addition of AG490 before beating significantly decreased the number of EBs containing at least one beating area compared with controls (Fig. 2A). Control EBs (DMSO treated and untreated) always showed over time a gradual increase in the number of EBs that had at least one beating area. Usually by day 7 after plating, 75%–95% of the control EBs contained at least one beating area. However, in a dose-dependent manner, AG490 significantly (p < .05) inhibited cardiomyocyte differentiation through JAK2 as measured by STAT3 phosphorylation and beating morphology (Fig. 2A). Addition of AG490 to the medium after beating had commenced showed no effect compared with control EBs (data not shown). Interestingly, beating was virtually wiped out when EBs were subjected to AG490 at levels (10 μM) that inhibit JAK2 and other kinases.
Figure 2. (A): Graph shows dosage affects of AG490 on ES cell differentiation into beating cardiomyocytes. On average for this experiment, approximately 20% of untreated and DMSO-treated control EBs began beating 7 days after plating, gradually rising each day until day 11, when the number of EBs beating plateaued at approximately 80%–90%. In EBs subjected to AG490 at concentrations specific for JAK2 (5 μM) , beating was reduced by approximately 50% overall, whereas AG490 at 1/10 concentration (500 nM) resulted in only 20% fewer EBs beating compared with controls. At concentrations (10 μM) known to inhibit other kinases (e.g., phosphatidylinositol 3, unknowns), beating is virtually wiped out. Inset shows the state of phosphorylated STAT3 from ES cells pretreated with specific concentrations of AG490. Phosphorylation declines as AG490 increases. (B): Total number of beating foci were counted and averaged. Untreated control EBs on average had one beating foci per EB beginning 7 days after plating, followed by a trend upward to six to seven beating foci by day 12. AG879, a drug structurally similar to AG490 but not known to inhibit JAK2, also showed a similar trend as untreated controls. AG490-treated EBs, however, had on average 0.5 beating areas per EB. Data are from at least three separate experiments. All data are presented as mean ± standard error of the mean for all EBs treated at each time point. Shown in the inset of each figure is the value n = X number of EBs observed. Abbreviations: DMSO, dimethyl sulfoxide; EB, embryoid body; ES, embryonic stem.
Closer analysis of each differentiated EB revealed that the average number of beating areas was altered when treated with specific concentrations of AG490 (Fig. 2B). As described previously , the entire EB rarely beats. Only distinct areas or foci differentiate into cardiomyocytes that beat at relatively constant rates for at least 10–12 days. Although untreated control EBs and EBs subjected to AG897, an agent chemically similar to AG490 but not shown to inhibit JAK2, resulted in progressively more beating foci over time (Fig. 2B), inhibition of JAK2 significantly decreased the average number of beating foci in EBs. Antagonizing JAK2 with AG490 did not affect beat rate in the few EBs that were beating. Controls and treated EBs all beat at an average of approximately 63 beats per minute (bpm); 62.3 ± 4.0 bpm for controls and 61 ± 3.0 bpm for AG490-treated EBs.
We confirmed these results and accumulated new data using dom/neg and dom/pos JAK2 constructs. Stable JAK2 dom/neg (provided by Dr. James Ihle) ES cell lines were generated by simultaneously electroporating a ptkNEO construct, followed by G418 selection (Fig. 3A). In the loss-of-function analyses, JAK2 dom/neg expression (Fig. 3B) resulted in a marked decrease over time in EBs that were beating compared with untreated or empty vector controls (75%–98% at days 8 through 13). The average number of beating areas was also significantly (p < .05) decreased in dom/neg cell lines compared with control EBs (Fig. 3C). An intriguing observation, however, was the fact that although inhibiting JAK2 resulted in fewer beating areas, many EBs were still capable of generating beating cardiomyocytes. Underlying reasons for these results may lie in one of two explanations: JAK2 dom/neg construct may not completely inhibit wild-type JAK2, or other compensatory mechanisms may come into play. The latter may provide an explanation as to why JAK2 knockout mice contain hearts, albeit with defects .
Figure 3. (A): Stable expression of the JAK2 dom/neg construct (KE15) in ES cells was detected by RT-PCR. The Jak2 (KE15) construct was produced by mutating amino acid 882 from K to E. This mutation disrupted an MboII restriction site, which enabled detection of endogenous JAK2 and the KE15 construct using RT-PCR, which produced an endogenous 179-bp JAK2 product (primer sequences: 5'TTCGGGAGTGTG-GAGATGT and 3'CTGTAGCACACTCCCTTC). Using the restriction enzyme MboII, wild-type JAK2 PCR product was cleaved into three fragments of 22, 78, and 79 bp (two larger bands appear as one). Dom/neg JAK2 yielded two fragments of 100 and 79 bp. (B): Dom/neg JAK2 resulted in a 25%–30% decrease in beating up to day 11. (C): Dom/neg JAK2 also significantly affected the number of beating foci per EB. By day 12, control (empty Neo cassette) EBs contained eight to nine beating foci per EB, whereas dom/neg JAK2-expressing EBs contained approximately three to four beating areas per EB. (D): Graph shows the effect of dom/pos JAK2 (D4). EBs transiently transfected with a D4 contained significantly more (p > .001, on average seven to eight) beating areas than GFP-transfected or untransfected EBs by day 7 after plating. This difference was maintained for at least 4–5 days. (E): Western blot shows expression of D4. No expression was seen in untransfected ES cell (ES) or beating EBs (B). D4 expression was observed with 1.0 and 1.5 μM cDNA. Stable lines (a–d) rarely differentiated into cardiomyocytes. We attributed this result to perpetual activation of STAT3, which promoted ES cell self-renewal. Except where shown, n 48 for each data point. Abbreviations: dom/neg, dominant/negative; dom/pos, dominant/positive; EB, embryoid body; ES, embryonic stem; GFP, green fluorescent protein; RT-PCR, reverse transcription–polymerase chain reaction.
Equally intriguing were the gain-of-function data obtained from EBs expressing dom/pos JAK2. Based on the data thus far (Figs. 1, 2), if JAK2 was a critical component of cardiac differentiation, we expected to observe an increase in beating areas in EBs expressing dom/pos JAK2 compared with controls. In fact, we clearly observed almost all of the dom/pos EBs containing significantly (p < .05) more beating areas beginning as soon as 2 days after transient transfection of JAK2 dom/pos cDNA into EBs (Figs. 3D–F). Consequently, when the proteomics, confocal images, drug, and molecular data are taken together (and upon closer inspection of the images in JAK2 knockout mouse literature), JAK2 clearly seems to play a role in normal early cardiac development, although it is possible that other compensatory mechanisms (e.g., other JAKs SRC family members) may substitute for or bypass JAK2 to eventually allow the heart to form (albeit delayed significantly and less efficiently) .
STAT3 Expression and Activity Is Elevated in Beating Areas
JAK2 phosphorylates STAT3, resulting in STAT3 translocation into the nucleus, where it acts as a transcription factor . To determine whether STAT3 was directly involved in initial stages of cardiac differentiation, we first used confocal microscopy to follow STAT3 in beating and nonbeating areas within EBs (Figs. 4A–I). In beating, Tnnt2+ areas, distinctly higher levels of STAT3 staining were evident, including within nuclei (Figs. 4A–C); in contrast, STAT3 was detectable only at relatively low levels in most nonbeating, Tnnt2– areas.
Figure 4. Confocal microscopy reveals expression and activity of STAT3 in beating foci within embryoid bodies. (A–C): STAT3 (Rhodamine; arrows in A) is elevated in beating cardiomyocytes derived from embryonic stem cells, as shown by anti-STAT3 antibodies. Arrows in (B) show cardiac troponin T (Tnnt2) staining (fluorescein isothiocyanate) representing cardiomyocyte differentiation in cells observed in (A). (C): DAPI staining reveals each nuclei within the field of view of (A, B). (D–F): When phosphorylated on Y705, STAT3 translocates to the nucleus and becomes transcriptionally active. Using an antibody directed against Y705-phosphorylated STAT3, staining (red) was clearly observed within nuclei (arrowheads in D) of Tnnt2+ cells (green staining cells in E). Dual label of (E) anti-pSTAT3 and Tnnt2 revealed in more detail pSTAT3 residing within each cardiomyocyte nucleus (arrowheads). Nucleoli within each nucleus were devoid of pSTAT3 (thin arrows). (G–H): Within a nonbeating area, arrowheads in (G, I) point to STAT3-free nuclei, whereas in (H), anti-Tnnt2 antibodies show no signal. Scale bar in (A) = 10 μm for A–I.
To determine whether STAT3 was physiologically active, we used an antibody specific for the phosphorylated form of STAT3 (pSTAT3) . In all cases, pSTAT3 was observed within the nucleus of Tnnt2+ cardiomyocytes (Figs. 4D–F), whereas only occasionally was it found in cells that did not stain for Tnnt2 or were not beating (Figs. 4G–I). Confocal data were confirmed by Western blot analysis of STAT3, which showed levels of STAT3 approximately threefold higher and active pSTAT3 approximately twofold higher in cells of beating areas compared with cells of nonbeating areas (Figs. 5A, 5B).
Figure 5. Western blots provided evidence confirming our confocal data. (A): STAT3 increased by close to three times in beating areas, whereas (B) pSTAT3 increased by at least two times. (C): Western blot shows that STAT3 is active in undifferentiated ES cells, as demonstrated by its phosphorylation. Phosphorylation levels drop beginning 2 days after inducing the differentiation process. (D): Phosphorylated STAT3 returns 5 days after plating EBs, when beating begins in wild-type EBs. By day 14, both EB beating and pSTAT3 are at their peak. Abbreviations: B, beating; EB, embryoid body; ES, embryonic stem; NB, nonbeating.
In the developing mouse embryo, migration of cardiomyocyte precursors begins at E6.5 and progresses quickly, with the commencement of beating at E8. Thus, in our in vitro model of this process, we expected to observe STAT3 activation (albeit at very low levels at first) occurring a few days before the onset of beating. To test this hypothesis, we followed our standard procedure for differentiating ES cells, taking protein samples from entire EBs each day during the differentiation process. Although we found detectable levels of STAT3 throughout differentiation, levels of pSTAT3 that were high in undifferentiated ES cells dropped dramatically as soon as day 2 of the differentiation process (Fig. 5C). Interestingly, pSTAT3 levels were detectable again in EBs (entire EB tested) approximately 1–3 days before the average onset of beating within EBs (Fig. 5D). This observation fit nicely with the known temporal activation of STAT3 compared with the onset of beating in vivo .
STAT3 Inhibition Restricts Cardiomyocyte Differentiation
To determine the function of STAT3 in cardiomyocyte formation, we made stable ES lines expressing a STAT3 dom/neg construct (STAT3?; Fig. 6A) . EBs expressing high (clones 7A, 8B) and medium (clone 10D) levels of the STAT3? resulted in only 12.0% ± 4.3% of EBs containing at least one beating area, an observation that lasted over 21 days after starting the differentiation process (up to day 12 graphed; Fig. 6B). In contrast, control EBs (untreated and NEO stable lines) containing beating areas peaked at days 9 through 12 at approximately 95%, remaining at that level until day 21. Interestingly, low STAT3? expression resulted in approximately 50%–75% EBs containing at least one beating area (Fig. 4B; ES line 8A). The fact that the average number of beating areas per EB (Fig. 6C) and the rate at which they were beating were also significantly decreased by STAT3? (Fig.6D) strongly suggests that STAT3 is a key component in the cardiomyogenesis pathway.
Figure 6. STAT3 function was analyzed molecularly using dominant/negative STAT3. (A): ES cell lines were electroporated STAT3? and CMV Neo in a 10:1 ratio, followed by selection with G418 to produce STAT? stable ES cell lines. STAT3? is detectable as a smaller band than STAT3 by SDS-PAGE. Out of 10 clones, four revealed different levels of expression. Based on levels of expression attained by normalizing band density to endogenous STAT3, lines 7A and 8B were considered high expressers, whereas 10D was considered a medium expresser and 8A a low expresser. (B): Graph shows affects of STAT3? expression on cardiomyocyte differentiation from ES cells. On average, approximately 25%–30% of nonelectroporated or CMV neoelectroporated control EBs began beating 7 days after plating, gradually rising each day until day 11, when the number of EBs beating plateaued at approximately 80%–90%. Only 15%–16% of high and medium STAT3? expressers showed signs of differentiation by day 12. The low expresser line (8A) resulted in significant decrease in differentiation (p > .001); however, significantly more EBs were found beating in line 8A than higher STAT3? expressing lines. (C): Total number of beating foci were counted and averaged. Untreated control EBs on average had 2.5 ± 0.6 beating foci per EB 7 days after plating, followed by a trend upward to 5.8 ± 1.1 beating foci by day 12. Each ES cell line showed a significant decrease in beating areas per EB. (D): Beat rate also was affected depending on levels of expression, approximately 70 beats per minute in controls versus 20 to 22 beats per minute in the highest expressing lines. Data are from at least three separate experiments. All data are presented as mean ± standard error of the mean for all EBs treated at each time point. n 48 EBs observed for all data points. Abbreviations: CMV, cytomegalovirus; EB, embryoid body; ES, embryonic stem.
Upon closer inspection of wild-type ES cells analyzed by Western blots probed with STAT3 antibodies, we observed an endogenous band at approximately 83 kD appearing 2–3 days after inducing ES cells to differentiate (Fig. 7A). This band was the same size as STAT3? observed in our ES cell lines stably expressing STAT3? (Fig. 6A). RT-PCR analysis verified putative endogenous STAT3? expression soon after EBs began to differentiate (2-day EB; Fig. 7B). We used the four stable ES cell lines as positive controls, all of which showed STAT3? expression. The STAT3? band was not detected in undifferentiated EBs.
Figure 7. (A): Endogenous STAT3? was detected 2–3 days after ES cells were processed for differentiation. STAT3? fell below levels of detection at approximately day 6, a few days before the average onset of beating in EBs. (B): Reverse transcription–polymerase chain reaction of undifferentiated ES cells, EBs differentiated for 2–3 days (when the Western in A detected STAT3?), and each stable ES cell line revealed that STAT3? was indeed expressed endogenously early on in the differentiation process and in our ES cell lines. Abbreviations: EB, embryoid body; ES, embryonic stem.
STAT3 Inhibition Suppresses Key Cardiac Gene Expression
Finally, Western blot and RT-PCR analyses were used to resolve why our ES cell lines expressing exogenous STAT3? were not completing their differentiation process (beating). Antibodies to alpha-cardiac actin, Nkx2.5, and GATA-4 were used as probes on nonbeating, beating, and STAT3?-expressing ES lines. Although isolated beating areas were positive for all cardiac proteins tested, high- and medium-expressing STAT3? ES lines (line 7A, 8B, and 10D) showed expression only for GATA-4. Low STAT3?-expressing lines (line 8A) expressed cardiac markers, albeit at lower levels (Fig. 8).
Figure 8. Western blot analyses revealed the loss of expression of cardiac-specific genes in STAT3? stably transected ES cells. (A): Cardiac actin and Nkx2.5 were detected in all isolated beating areas and in line 8A, whereas in the medium- and high-expressing lines and in isolated nonbeating areas, these proteins were not detected. (B): Superarray kit containing cardiac-specific and control mRNA was used for semiquantitative reverse transcription–polymerase chain reaction. Expression of many cardiac genes was markedly decreased in NB areas and STAT3? stable lines (1 subunit of the L-type calcium channel; DHP). (C): Expression of C-EBP? and , transcription factors reliant on STAT3 for transcription, was evident in isolated beating areas but was markedly reduced in undifferentiated, NB areas and high STAT3?-expressing ES cell lines. n 48 for each graphed data point. Abbreviations: EB, embryoid body; ES, embryonic stem; NB, nonbeating.
Semiquantitative RT-PCR confirmed our Western blot analyses and added new data as primers to Tnnt2, TnnC, Nkx2.5, cardiac actin, and 1 subunit of the L-type Ca2+ channel (DHP) were used to determine expression of possible downstream genes affected by the suppression of STAT3. Expression of specific cardiac genes was markedly lower in STAT3? EBs differentiated for at least 7–10 days after plating (Fig. 8B). Interestingly, promoter sequence analysis of the DHP gene, cardiac actin, and Nkx2.5 revealed putative STAT3 DNA binding sites, suggesting that STAT3? inhibition of STAT3 could lower or even prevent expression of these cardiac genes (Table 1).
Table 1. Promoters containing STAT and/or C/EBP sites
Closer inspection of promoter regions for many cardiac genes revealed that DNA binding sites were common for the interleukin (IL)-6 acute-phase response gene C/EBP, which has been shown to be reliant on STAT3 for expression . Semiquantitative RT-PCR analysis of STAT3? ES cell lines (lines 7A through 10D) revealed a loss or reduction in C/EBP? and expression (Fig. 8C). Table 1 summarizes cardiac genes with STAT3 or C/EBP regulatory elements. The fact that C/EBP? and sites are found within many cardiac genes and that STAT3? expression leads to the loss of these specific cardiac genes strongly suggests that expression of many cardiac genes is dependent on active STAT3 or C/EBP.
CONCLUSION
K.F. and G.R. contributed equally to this study. We would like to thank Dr. Robert Lechleider, Dr. Chris Taylor, and Tammy Gallicano for their critical reading of the manuscript. We also thank Alex C. Potocki for his confocal microscopy guidance. Work in this investigation was supported by the Transgenic Shared Resource under the directorship of G.I.G., who is supported by Cancer Center Support Grant CA51008-13. This work was primarily supported by grant HL70204-01 from the NIH and grant O265429U from the American Heart Association, both awarded to G.I.G.
REFERENCES
Arai A, Yamamoto K, Toyama J. Murine cardiac progenitor cells require visceral embryonic endoderm and primitive streak for terminal differentiation. Dev Dyn 1997;210:344–353.
Ji RP, Phoon CKL, Aristizabal O et al. Onset of cardiac function during early mouse embryogenesis coincides with entry of primitive erythroblasts into the embryo proper. Circ Res 2003;92:133–135.
Wobus AM, Wallukat G, Hescheler J et al. Pluripotent mouse embryonic stem cells are able to differentiate into cardiomyocytes expressing chronotropic responses to adrenergic and cholinergic agents and Ca2+ channel blockers. Differentiation 1991;48:173–182.
Metzger JM, Lin WI, Samuelson LC. Vital staining of cardiac myocytes during embryonic stem cell cardiogenesis in vitro. Circ Res 1996;78:547–552.
Hescheler J, Fleischmann BK, Lentini S et al. Embryonic stem cells: a model to study structural and functional properties in cardiomyogenesis. Cardiovasc Res 1997;36:149–162.
Xu C, Liguori G, Adamson E et al. Specific arrest of cardiogenesis in cultured embryonic stem cells lacking cripto-1. Dev Biol 1998;196:237–241.
Meyer N, Jaconi M, Landopoulou A et al. A fluorescent reporter gene as a marker for ventricular specification in ES-derived cardiac cells. FEBS Lett 2000;478:151–158.
Zhou X, Quann E, Gallicano GI. Differentiation of nonbeating embryonic stem cells into beating cardiomyocytes is dependent on downregulation of PKC? and in concert with upregulation of PKC. Dev Biol 2003;255:407–422.
Maltsev VA, Wobus AM, Rohwedel J et al. Cardiomyocytes differentiated in vitro from embryonic stem cells developmentally express cardiac-specific genes and ionic currents. Circ Res 1994;75:233–244.
Durbin JE, Hackenmiller R, Simon MC et al. Targeted disruption of the mouse Stat1 gene results in compromised innate immunity to viral disease. Cell 1996;84:443–450.
Kaplan MH, Schindler U, Smiley ST et al. Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity 1996;4:313–319.
Kaplan MH, Sun YL, Hoey T et al. Impaired IL-12 responses and enhanced development of Th2 cells in Stat4-deficient mice. Nature 1996;382:174–177.
Meraz MA, White JM, Sheehan KC et al. Targeted disruption of the mouse Stat1 gene in mice reveals unexpected physiologic specificity in the JAK-STAT signalling pathway. Cell 1996;84:431–442.
Thierfelder WE, van Deursen JM, Yamamoto K et al. Requirement for Stat4 in interleukin-12-mediated responses of natural killer and T cells. Nature 1996;382:171–174.
Liu X, Robinson GW, Wagner KU et al. Stat5 is mandatory for adult mammary glad development and lactogenesis. Genes Dev 1997;11:179–186.
Udy GB, Towers RP, Snell RG et al. Requirement of Stat5b for sexual dimorphism of body growth rates and liver gene expression. Proc Natl Acad Sci U S A 1997;94:7239–7244.
Imada K, Leonard WJ. The Jak-STAT pathway. Mol Immunol 2000;37:1–11.
Kunisada K, Tone E, Fujio Y et al. Activation of gp130 transduces hypertrophic signals via STAT3 in cardiac myocytes. Circulation 1998;98:346–352.
Booz GW, Day JNE, Baker KM. Interplay between the cardiac renin angiotensin system and JAK-STAT signaling: role in cardiac hy pertrophy, ischemia/reperfusion dysfunction, and heart failure. J Mol Cell Cardiol 2002;34:1443–1453.
Jacoby JJ, Kalinowski A, Liu M et al. Cardiomyocyte-restricted knockout of STAT3 results in higher sensitivity to inflammation, cardiac fibrosis, and heart failure with advanced age. Proc Natl Acad Sci U S A 2003;100:12929–12934.
Hattori R, Maulik N, Otani H et al. Role of STAT3 in ischemic preconditioning. J Mol Cell Cardiol 2001;33:1929–1936.
Negoro S, Kunisada K, Tone E et al. Activation of JAK/STAT pathway transduces cytoprotective signal in rat acute myocardial infarction. Cardiovasc Res 2000;47:797–805.
Xuan YT, Guo Y, Han H et al. An essential role of the JAK-STAT pathway in ischemic preconditioning. Proc Natl Acad Sci U S A 2001;98:9050–9055.
Mascareno E, El-Shafei M, Maulik N et al. JAK/STAT signaling is associated with cardiac dysfunction during ischemia and reperfusion. Circulation 2001;104:325–329.
Bolli R, Dawn B, Xuan Y.-T. Role of the JAK-STAT pathway in protection against myocardial ischemia/reperfusion injury. Trends Cardiovasc Med 2003;13:72–79.
Neubauer H, Cumano A, Muller M et al. Jak2 deficiency defines an essential developmental checkpoint in definitive hematopoiesis. Cell 1998;93:397–409.
Parganas E, Wang D, Stravopodis D et al. Jak2 is essential for signaling through a variety of cytokine receptors. Cell 1998;93:385–395.
Kotenko SV, Izotova LS, Pollack BP et al. Other kinases can substitute for Jak2 in signal transduction by interferon-gamma. J Biol Chem 1996;271:17174–17182.
Caldenhoven E, van Dijk TB, Solari R et al. STAT3beta, a splice variant of transcription factor STAT3, is a dominant negative regulator of transcription. J Biol Chem 1996;271:13221–13227.
Feng J, Witthuhn BA, Matsuda T et al. Activation of Jak2 catalytic activity requires phosphorylation of Y1007 in the kinase activation loop. Mol Cell Biol 1997;17:2497–2501.
Zhong H, Murphy TJ, Minneman KP. Activation of signal transducers and activators of transcription by alpha(1A)-adrenergic receptor stimulation in PC12 cells. Mol Pharmacol 2000;57:961–967.
Duncan SA, Zhong Z, Wen Z et al. STAT signaling is active during early mammalian development. Dev Dyn 1997;280:190–198.
Catlett-Falcone R, Landowski TH, Oshiro MM et al. Constitutive activation of Stat3 signaling confers resistance to apoptosis in human U266 myeloma cells. Immunity 1999;10:105–115.
Cantwell CA, Sterneck E, Johnson PF. Interleukin-6-specific activation of the C/EBPdelta gene in hepatocytes is mediated by Stat3 and Sp1. Mol Cell Biol 1998;18:2108–2117.
Niehof M, Streetz K, Rakemann T et al. Interleukin-6-induced tethering of STAT3 to the LAP/C/EBP? promoter suggests a new mechanism of transcriptional regulation by STAT3. J Biol Chem 2001;276:9016–9027.
Ernst M, Oates A, Dunn AR. gp130-mediated signal transduction in embryonic stem cells involves activation of Jak and Ras/mitogen-activated protein kinase pathways. J Biol Chem 1996;271:30136–30143.
Niwa H, Burdon T, Chambers I et al. Self-renewal of pluripotent embryonic stem cells is mediated via activation of STAT3. Genes Dev 1998;12:2048–2060.
Raz R, Lee C-K, Cannizzaro LA et al. Essential role of STAT3 for embryonic stem cell pluripotency. Proc Natl Acad Sci U S A 1999;96:2846–2851.
Zhang Y, Turkson J, Carter-Su C et al. Activation of Stat3 in v-Src-transformed fibroblasts requires cooperation of Jak1 kinase activity. J Biol Chem 2000;275:24935–24944.
Reecy JM, Li X, Yamada M et al. Identification of upstream regulatory regions in the heart-expressed homeobox gene Nkx2-5. Development 1999;126:839–849.
Liu L, Fan I, El-Zaru MR et al. Regulation of DHP receptor expression by elements in the 5'-flanking sequence. Am J Physiol Heart Circ Physiol 2000;278:H1153–H1162.
Chen CY, Schwartz RJ. Recruitment of the tinman homolog Nkx-2.5 by serum response factor activates cardiac alpha-actin gene transcription. Mol Cell Biol 1996;16:6372–6384.
Lyons I, Pasons LM, Hartley L et al. Myogenic and morphogenetic defects in the heart tubes of murine embryos lacking the homeo box gene Nkx2-5. Genes Dev 1995;9:1654–1666.
Kunisada K, Negoro S, Tone E et al. Signal transducer and activator of transcription 3 in the heart transduces not only a hypertrophic signal but a protective signal against doxorubicin-induced cardiomyopathy. Proc Natl Acad Sci U S A 2000;97:315–319.
Takeda K, Noguchi K, Shi W et al. Targeted disruption of the mouse Stat3 gene leads to early embryonic lethality. Proc Natl Acad Sci U S A 1997;94:3801–3804.
Valdembri D, Serini G, Vacca A et al. In vivo activation of JAK2/ STAT-3 pathway during angiogenesis induced by GM-CSF. FASEB J 2002;16:225–227.
Lumelsky N, Blondel O, Laeng P et al. Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science 2001;292:1389–1394.
Mazurova Y. New therapeutic approaches for the treatment of Huntington’s disease. Acta Medica 2001;44:119–123.
Vogel G. Stem cell research: rat brains respond to embryonic stem cells. Science 2002;295:254–255.
Quitschke WW, DePonti-Zilli L, Lin ZY et al. Identification of two nuclear factor-binding domains on the chicken cardiac actin promoter: implications for regulation of the gene. Mol Cell Biol 1989;9:3218–3230.
Wang Q, Sigmund CD, Lin JJ-C. Identification of cis elements in the cardiac troponin T gene conferring specific expression in cardiac muscle of transgenic mice. Circ Res 2000;86:478–484.
Parmacek MS, Vora AJ, Shen T et al. Identification and characterization of a cardiac-specific transcriptional regulatory element in the slow/cardiac troponin gene. Mol Cell Biol 1992;12:1967–1976.
Meydan N, Grunberger T, Dadi H et al. Inhibition of acute lymphoblastic leukaemia by a Jak-2 inhibitor. Nature 1996;379:645–648.(Kara Foshaya, Gisella Rod)
b Thomas Jefferson School of Technology, Alexandria, VA, USA
Key Words. Embryonic stem cells ? Cardiac development ? Signal transduction ? JAK2/STAT3
Correspondence: G. Ian Gallicano, Ph.D., Georgetown University Medical Institute, Department of Cell Biology, 3900 Reservoir Road NW, Room NE203, Washington, DC 20007, USA. Telephone: 202-687-0228; Fax: 202-687-1823; e-mail: gig@georgetown.edu
ABSTRACT
The first organ to form in mammals is the heart. In murine development, at approximately E7.5 to E8.0, distinct mesodermal cells form bilateral cardiac progenitor cells that proceed to fuse and form a primitive heart tube . During tube formation, distinctive beating of cardiac cells within the developing heart begins at approximately E8.0 . Although overall molecular control of heart tube morphogenesis has been investigated in great detail both in vivo and in vitro, few investigations exist with regard to the initial regulatory mechanisms that differentiate progenitor cells into beating cardiomyocytes. Some obvious reasons for this lack of information have been the difficulty of procuring mammalian embryos at precardiomyocyte stages as well as the difficulty in analyzing the few cardiomyocytes available from early heart tubes. Recently, however, a cardiac in vitro model system was developed using embryonic stem (ES) cells. Manipulating ES cells to form embryoid bodies (EBs) that consistently give rise to beating cardiomyocytes has been shown to be relatively straight-forward and highly reproducible . More important, the electrophysiology of these beating cells has been documented and verified as closely mimicking cardiomyocyte electrophysiology .
To determine the signal transduction pathways that drive cardiomyocyte differentiation, we subjected lysates from beating and nonbeating areas of EBs (procured using a microsurgical protocol similar to the one we previously described ) to a kinase expression screen. Out of 75 kinases tested, JAK2 demonstrated the highest level of expression (>70%) within beating ES-derived cardiomyocytes compared with nonbeating ES cells.
JAK2 is a member of the Janus kinase family, which includes JAK1-3 and TYK2. A primary function of these kinases is to regulate gene transcription by phosphorylating a family of transcription factors called signal transducers and activators of transcription (STATs). Seven STAT family members exist, all of which have been investigated in many in vitro and in vivo systems, including knockout mouse models . Members of this signal transduction pathway have been implicated as a major constituent of adult cardiac pathophysiology, including pressure overload–induced cardiac hypertrophy , increased sensitivity to inflammation, cardiac fibrosis, and heart failure , as well as a paradoxical cardioprotection by ischemic preconditioning and cardiac dysfunction induced by ischemia/reperfusion .
The conventional knockout of JAK2 resulted in embryonic lethality at embryonic days 11–12 (E11–E12) due to a significant reduction in erythropoiesis . Also observed was a significant delay in heart morphogenesis , but this phenotype was not explored in depth in those knockout embryos.
Ablation of STAT3 in mice led to embryonic lethality at E6.5–E7.0, which is, coincidentally, immediately before cardiomyocyte formation . Moreover, the heart-specific STAT3 knockouts produced by conventional cre/lox technology led to higher susceptibility to heart failure in juvenile mice . It must be noted, however, that STAT3 was not analyzed during cardiomyocyte differentiation in the study by Jacoby et al. because the cre in the heart-specific knockout was driven by the -myosin heavy chain promoter, which is expressed at low levels before birth, followed by higher levels long after cardiomyocyte differentiation.
Consequently, based on our kinase expression screen implicating JAK2, the JAK2/STAT3 knockout data , and the evidence that JAK2/STAT3 is clearly involved in remodeling diseased adult cardiomyocytes, we hypothesized that distinct regulation of specific components must occur during initial modeling of cardiac progenitor cells, driving them down the cardiomyocyte differentiation pathway.
To test this hypothesis, specific stages of EB differentiation were analyzed by using highly specific molecular, biochemical, pharmacological, and microscopic analyses. We found that the JAK2/STAT3 pathway is essential for proper gene expression and development of beating cardiomyocytes from ES cells.
MATERIALS AND METHODS
JAK2 Expression and Activity Is Elevated in Beating Areas
Comprehensive analysis of beating and nonbeating areas from CCE EBs was determined molecularly by RT-PCR using primers specific for cardiac genes (data not shown; see for similar results in EBs derived from R1 ES cells). Beating areas always expressed all cardiac-specific genes tested, whereas expression of cardiac-specific genes from nonbeating areas was highly variable . The raw data from our proteomic screen (Kinexus Inc., Vancouver, Canada) revealed a dramatic increase of JAK2 in beating areas compared with nonbeating areas (Figs. 1A–C). Western blot analyses in our laboratory verified the kinase expression data from the Kinexus Inc. proteomics report (Fig. 1D).
Figure 1. Graphs show raw and normalized data taken directly from the Kinexus protein kinase screen. (A): Kinexus Inc. (http://www.kinexus.ca/) uses a highly sensitive imaging system with a 16-bit camera (Bio-Rad Fluor-S Max Multi-Imager) in combination with quantitation software (Bio-Rad Quantity One) to quantify and analyze the chemiluminescent samples. The resulting trace quantity for each band scanned at the maximum scan time is termed the raw data. Because the relationship between scan time and band intensity is linear over the quantifiable range of the signal intensity, the raw data from the scans are normalized to 60 seconds (counts per minute ) for uniformity. (B):. After normalization, data are converted to percentages by subtracting the control (nonbeating) normalized CPM from the experimental (beating) normalized CPM, followed by dividing the difference by the control (nonbeating) normalized CPM and multiplying by 100. (C): The actual JAK2 bands from the Kinexus report show a stronger signal in beating areas compared with nonbeating areas. The red lines in (C) are generated and used by the computer to find the correct bands. (D): Our follow-up Western confirmed the Kinexus data. A ratio of JAK2 protein levels was calculated by normalizing to tubulin (from the same blot), showing that there is 1.6 times (65%) more JAK2 in beating areas than nonbeating areas. (E, F): Confocal microscopy of EBs showed higher levels of JAK2 in beating areas, represented by (F) anticardiac troponin, compared with nonbeating areas subjected to anti-JAK2. Arrows point to JAK2 on cell membranes. (G): Confocal microscopy of EBs clearly showed high amounts of activated pJAK2 on membranes (arrows) in cardiac troponin–positive cells. (H): Western blot shows clear increase in active JAK2 by two times in beating areas. Hela cells serve as a positive control. Scale bar in (E) = 25 μm. Abbreviations: B, beating; NB, nonbeating.
JAK2 localization in beating areas was additionally resolved using confocal microscopy. Antibodies directed against JAK2 or its physiologically active counterpart, phosphorylated JAK2 (pJAK2) , followed by fluorescently labeled secondary antibodies, resulted in a strong, highly specific signal in cells residing in beating areas (Figs. 1E–G). Confocal microscopy clearly showed distinct staining for JAK2 and pJAK2 on or near the plasma membrane in newly differentiated, cardiac troponin T (Tnnt2)–positive, beating cardiomyocytes (~12–18 hours after onset of beating; Figs. 1E–G) . This high level of staining at or near the membrane was rarely seen in cells residing in nonbeating areas. Some areas that were clearly not beating occasionally stained brightly for JAK2; however, the staining pattern in most of these cells was diffuse and cytoplasmic. pJAK2 also was occasionally observed in nonbeating areas (Figs. 1E–G). Western blot analysis supported our confocal data, because pJAK2 was found to be highly elevated in beating areas compared with nonbeating areas (Fig. 1H), additionally supporting the fact that JAK2 was indeed active in beating areas.
JAK2 Inhibition and Activation Alters Cardiomyocyte Differentiation
To determine the physiological relevance of JAK2 during cardiomyogenesis, we first inhibited JAK2 function using the relatively specific pharmacological agent to JAK2, AG490. EBs were subjected to medium containing various concentrations of AG490, including a concentration known to be specific for JAK2. Addition of AG490 before beating significantly decreased the number of EBs containing at least one beating area compared with controls (Fig. 2A). Control EBs (DMSO treated and untreated) always showed over time a gradual increase in the number of EBs that had at least one beating area. Usually by day 7 after plating, 75%–95% of the control EBs contained at least one beating area. However, in a dose-dependent manner, AG490 significantly (p < .05) inhibited cardiomyocyte differentiation through JAK2 as measured by STAT3 phosphorylation and beating morphology (Fig. 2A). Addition of AG490 to the medium after beating had commenced showed no effect compared with control EBs (data not shown). Interestingly, beating was virtually wiped out when EBs were subjected to AG490 at levels (10 μM) that inhibit JAK2 and other kinases.
Figure 2. (A): Graph shows dosage affects of AG490 on ES cell differentiation into beating cardiomyocytes. On average for this experiment, approximately 20% of untreated and DMSO-treated control EBs began beating 7 days after plating, gradually rising each day until day 11, when the number of EBs beating plateaued at approximately 80%–90%. In EBs subjected to AG490 at concentrations specific for JAK2 (5 μM) , beating was reduced by approximately 50% overall, whereas AG490 at 1/10 concentration (500 nM) resulted in only 20% fewer EBs beating compared with controls. At concentrations (10 μM) known to inhibit other kinases (e.g., phosphatidylinositol 3, unknowns), beating is virtually wiped out. Inset shows the state of phosphorylated STAT3 from ES cells pretreated with specific concentrations of AG490. Phosphorylation declines as AG490 increases. (B): Total number of beating foci were counted and averaged. Untreated control EBs on average had one beating foci per EB beginning 7 days after plating, followed by a trend upward to six to seven beating foci by day 12. AG879, a drug structurally similar to AG490 but not known to inhibit JAK2, also showed a similar trend as untreated controls. AG490-treated EBs, however, had on average 0.5 beating areas per EB. Data are from at least three separate experiments. All data are presented as mean ± standard error of the mean for all EBs treated at each time point. Shown in the inset of each figure is the value n = X number of EBs observed. Abbreviations: DMSO, dimethyl sulfoxide; EB, embryoid body; ES, embryonic stem.
Closer analysis of each differentiated EB revealed that the average number of beating areas was altered when treated with specific concentrations of AG490 (Fig. 2B). As described previously , the entire EB rarely beats. Only distinct areas or foci differentiate into cardiomyocytes that beat at relatively constant rates for at least 10–12 days. Although untreated control EBs and EBs subjected to AG897, an agent chemically similar to AG490 but not shown to inhibit JAK2, resulted in progressively more beating foci over time (Fig. 2B), inhibition of JAK2 significantly decreased the average number of beating foci in EBs. Antagonizing JAK2 with AG490 did not affect beat rate in the few EBs that were beating. Controls and treated EBs all beat at an average of approximately 63 beats per minute (bpm); 62.3 ± 4.0 bpm for controls and 61 ± 3.0 bpm for AG490-treated EBs.
We confirmed these results and accumulated new data using dom/neg and dom/pos JAK2 constructs. Stable JAK2 dom/neg (provided by Dr. James Ihle) ES cell lines were generated by simultaneously electroporating a ptkNEO construct, followed by G418 selection (Fig. 3A). In the loss-of-function analyses, JAK2 dom/neg expression (Fig. 3B) resulted in a marked decrease over time in EBs that were beating compared with untreated or empty vector controls (75%–98% at days 8 through 13). The average number of beating areas was also significantly (p < .05) decreased in dom/neg cell lines compared with control EBs (Fig. 3C). An intriguing observation, however, was the fact that although inhibiting JAK2 resulted in fewer beating areas, many EBs were still capable of generating beating cardiomyocytes. Underlying reasons for these results may lie in one of two explanations: JAK2 dom/neg construct may not completely inhibit wild-type JAK2, or other compensatory mechanisms may come into play. The latter may provide an explanation as to why JAK2 knockout mice contain hearts, albeit with defects .
Figure 3. (A): Stable expression of the JAK2 dom/neg construct (KE15) in ES cells was detected by RT-PCR. The Jak2 (KE15) construct was produced by mutating amino acid 882 from K to E. This mutation disrupted an MboII restriction site, which enabled detection of endogenous JAK2 and the KE15 construct using RT-PCR, which produced an endogenous 179-bp JAK2 product (primer sequences: 5'TTCGGGAGTGTG-GAGATGT and 3'CTGTAGCACACTCCCTTC). Using the restriction enzyme MboII, wild-type JAK2 PCR product was cleaved into three fragments of 22, 78, and 79 bp (two larger bands appear as one). Dom/neg JAK2 yielded two fragments of 100 and 79 bp. (B): Dom/neg JAK2 resulted in a 25%–30% decrease in beating up to day 11. (C): Dom/neg JAK2 also significantly affected the number of beating foci per EB. By day 12, control (empty Neo cassette) EBs contained eight to nine beating foci per EB, whereas dom/neg JAK2-expressing EBs contained approximately three to four beating areas per EB. (D): Graph shows the effect of dom/pos JAK2 (D4). EBs transiently transfected with a D4 contained significantly more (p > .001, on average seven to eight) beating areas than GFP-transfected or untransfected EBs by day 7 after plating. This difference was maintained for at least 4–5 days. (E): Western blot shows expression of D4. No expression was seen in untransfected ES cell (ES) or beating EBs (B). D4 expression was observed with 1.0 and 1.5 μM cDNA. Stable lines (a–d) rarely differentiated into cardiomyocytes. We attributed this result to perpetual activation of STAT3, which promoted ES cell self-renewal. Except where shown, n 48 for each data point. Abbreviations: dom/neg, dominant/negative; dom/pos, dominant/positive; EB, embryoid body; ES, embryonic stem; GFP, green fluorescent protein; RT-PCR, reverse transcription–polymerase chain reaction.
Equally intriguing were the gain-of-function data obtained from EBs expressing dom/pos JAK2. Based on the data thus far (Figs. 1, 2), if JAK2 was a critical component of cardiac differentiation, we expected to observe an increase in beating areas in EBs expressing dom/pos JAK2 compared with controls. In fact, we clearly observed almost all of the dom/pos EBs containing significantly (p < .05) more beating areas beginning as soon as 2 days after transient transfection of JAK2 dom/pos cDNA into EBs (Figs. 3D–F). Consequently, when the proteomics, confocal images, drug, and molecular data are taken together (and upon closer inspection of the images in JAK2 knockout mouse literature), JAK2 clearly seems to play a role in normal early cardiac development, although it is possible that other compensatory mechanisms (e.g., other JAKs SRC family members) may substitute for or bypass JAK2 to eventually allow the heart to form (albeit delayed significantly and less efficiently) .
STAT3 Expression and Activity Is Elevated in Beating Areas
JAK2 phosphorylates STAT3, resulting in STAT3 translocation into the nucleus, where it acts as a transcription factor . To determine whether STAT3 was directly involved in initial stages of cardiac differentiation, we first used confocal microscopy to follow STAT3 in beating and nonbeating areas within EBs (Figs. 4A–I). In beating, Tnnt2+ areas, distinctly higher levels of STAT3 staining were evident, including within nuclei (Figs. 4A–C); in contrast, STAT3 was detectable only at relatively low levels in most nonbeating, Tnnt2– areas.
Figure 4. Confocal microscopy reveals expression and activity of STAT3 in beating foci within embryoid bodies. (A–C): STAT3 (Rhodamine; arrows in A) is elevated in beating cardiomyocytes derived from embryonic stem cells, as shown by anti-STAT3 antibodies. Arrows in (B) show cardiac troponin T (Tnnt2) staining (fluorescein isothiocyanate) representing cardiomyocyte differentiation in cells observed in (A). (C): DAPI staining reveals each nuclei within the field of view of (A, B). (D–F): When phosphorylated on Y705, STAT3 translocates to the nucleus and becomes transcriptionally active. Using an antibody directed against Y705-phosphorylated STAT3, staining (red) was clearly observed within nuclei (arrowheads in D) of Tnnt2+ cells (green staining cells in E). Dual label of (E) anti-pSTAT3 and Tnnt2 revealed in more detail pSTAT3 residing within each cardiomyocyte nucleus (arrowheads). Nucleoli within each nucleus were devoid of pSTAT3 (thin arrows). (G–H): Within a nonbeating area, arrowheads in (G, I) point to STAT3-free nuclei, whereas in (H), anti-Tnnt2 antibodies show no signal. Scale bar in (A) = 10 μm for A–I.
To determine whether STAT3 was physiologically active, we used an antibody specific for the phosphorylated form of STAT3 (pSTAT3) . In all cases, pSTAT3 was observed within the nucleus of Tnnt2+ cardiomyocytes (Figs. 4D–F), whereas only occasionally was it found in cells that did not stain for Tnnt2 or were not beating (Figs. 4G–I). Confocal data were confirmed by Western blot analysis of STAT3, which showed levels of STAT3 approximately threefold higher and active pSTAT3 approximately twofold higher in cells of beating areas compared with cells of nonbeating areas (Figs. 5A, 5B).
Figure 5. Western blots provided evidence confirming our confocal data. (A): STAT3 increased by close to three times in beating areas, whereas (B) pSTAT3 increased by at least two times. (C): Western blot shows that STAT3 is active in undifferentiated ES cells, as demonstrated by its phosphorylation. Phosphorylation levels drop beginning 2 days after inducing the differentiation process. (D): Phosphorylated STAT3 returns 5 days after plating EBs, when beating begins in wild-type EBs. By day 14, both EB beating and pSTAT3 are at their peak. Abbreviations: B, beating; EB, embryoid body; ES, embryonic stem; NB, nonbeating.
In the developing mouse embryo, migration of cardiomyocyte precursors begins at E6.5 and progresses quickly, with the commencement of beating at E8. Thus, in our in vitro model of this process, we expected to observe STAT3 activation (albeit at very low levels at first) occurring a few days before the onset of beating. To test this hypothesis, we followed our standard procedure for differentiating ES cells, taking protein samples from entire EBs each day during the differentiation process. Although we found detectable levels of STAT3 throughout differentiation, levels of pSTAT3 that were high in undifferentiated ES cells dropped dramatically as soon as day 2 of the differentiation process (Fig. 5C). Interestingly, pSTAT3 levels were detectable again in EBs (entire EB tested) approximately 1–3 days before the average onset of beating within EBs (Fig. 5D). This observation fit nicely with the known temporal activation of STAT3 compared with the onset of beating in vivo .
STAT3 Inhibition Restricts Cardiomyocyte Differentiation
To determine the function of STAT3 in cardiomyocyte formation, we made stable ES lines expressing a STAT3 dom/neg construct (STAT3?; Fig. 6A) . EBs expressing high (clones 7A, 8B) and medium (clone 10D) levels of the STAT3? resulted in only 12.0% ± 4.3% of EBs containing at least one beating area, an observation that lasted over 21 days after starting the differentiation process (up to day 12 graphed; Fig. 6B). In contrast, control EBs (untreated and NEO stable lines) containing beating areas peaked at days 9 through 12 at approximately 95%, remaining at that level until day 21. Interestingly, low STAT3? expression resulted in approximately 50%–75% EBs containing at least one beating area (Fig. 4B; ES line 8A). The fact that the average number of beating areas per EB (Fig. 6C) and the rate at which they were beating were also significantly decreased by STAT3? (Fig.6D) strongly suggests that STAT3 is a key component in the cardiomyogenesis pathway.
Figure 6. STAT3 function was analyzed molecularly using dominant/negative STAT3. (A): ES cell lines were electroporated STAT3? and CMV Neo in a 10:1 ratio, followed by selection with G418 to produce STAT? stable ES cell lines. STAT3? is detectable as a smaller band than STAT3 by SDS-PAGE. Out of 10 clones, four revealed different levels of expression. Based on levels of expression attained by normalizing band density to endogenous STAT3, lines 7A and 8B were considered high expressers, whereas 10D was considered a medium expresser and 8A a low expresser. (B): Graph shows affects of STAT3? expression on cardiomyocyte differentiation from ES cells. On average, approximately 25%–30% of nonelectroporated or CMV neoelectroporated control EBs began beating 7 days after plating, gradually rising each day until day 11, when the number of EBs beating plateaued at approximately 80%–90%. Only 15%–16% of high and medium STAT3? expressers showed signs of differentiation by day 12. The low expresser line (8A) resulted in significant decrease in differentiation (p > .001); however, significantly more EBs were found beating in line 8A than higher STAT3? expressing lines. (C): Total number of beating foci were counted and averaged. Untreated control EBs on average had 2.5 ± 0.6 beating foci per EB 7 days after plating, followed by a trend upward to 5.8 ± 1.1 beating foci by day 12. Each ES cell line showed a significant decrease in beating areas per EB. (D): Beat rate also was affected depending on levels of expression, approximately 70 beats per minute in controls versus 20 to 22 beats per minute in the highest expressing lines. Data are from at least three separate experiments. All data are presented as mean ± standard error of the mean for all EBs treated at each time point. n 48 EBs observed for all data points. Abbreviations: CMV, cytomegalovirus; EB, embryoid body; ES, embryonic stem.
Upon closer inspection of wild-type ES cells analyzed by Western blots probed with STAT3 antibodies, we observed an endogenous band at approximately 83 kD appearing 2–3 days after inducing ES cells to differentiate (Fig. 7A). This band was the same size as STAT3? observed in our ES cell lines stably expressing STAT3? (Fig. 6A). RT-PCR analysis verified putative endogenous STAT3? expression soon after EBs began to differentiate (2-day EB; Fig. 7B). We used the four stable ES cell lines as positive controls, all of which showed STAT3? expression. The STAT3? band was not detected in undifferentiated EBs.
Figure 7. (A): Endogenous STAT3? was detected 2–3 days after ES cells were processed for differentiation. STAT3? fell below levels of detection at approximately day 6, a few days before the average onset of beating in EBs. (B): Reverse transcription–polymerase chain reaction of undifferentiated ES cells, EBs differentiated for 2–3 days (when the Western in A detected STAT3?), and each stable ES cell line revealed that STAT3? was indeed expressed endogenously early on in the differentiation process and in our ES cell lines. Abbreviations: EB, embryoid body; ES, embryonic stem.
STAT3 Inhibition Suppresses Key Cardiac Gene Expression
Finally, Western blot and RT-PCR analyses were used to resolve why our ES cell lines expressing exogenous STAT3? were not completing their differentiation process (beating). Antibodies to alpha-cardiac actin, Nkx2.5, and GATA-4 were used as probes on nonbeating, beating, and STAT3?-expressing ES lines. Although isolated beating areas were positive for all cardiac proteins tested, high- and medium-expressing STAT3? ES lines (line 7A, 8B, and 10D) showed expression only for GATA-4. Low STAT3?-expressing lines (line 8A) expressed cardiac markers, albeit at lower levels (Fig. 8).
Figure 8. Western blot analyses revealed the loss of expression of cardiac-specific genes in STAT3? stably transected ES cells. (A): Cardiac actin and Nkx2.5 were detected in all isolated beating areas and in line 8A, whereas in the medium- and high-expressing lines and in isolated nonbeating areas, these proteins were not detected. (B): Superarray kit containing cardiac-specific and control mRNA was used for semiquantitative reverse transcription–polymerase chain reaction. Expression of many cardiac genes was markedly decreased in NB areas and STAT3? stable lines (1 subunit of the L-type calcium channel; DHP). (C): Expression of C-EBP? and , transcription factors reliant on STAT3 for transcription, was evident in isolated beating areas but was markedly reduced in undifferentiated, NB areas and high STAT3?-expressing ES cell lines. n 48 for each graphed data point. Abbreviations: EB, embryoid body; ES, embryonic stem; NB, nonbeating.
Semiquantitative RT-PCR confirmed our Western blot analyses and added new data as primers to Tnnt2, TnnC, Nkx2.5, cardiac actin, and 1 subunit of the L-type Ca2+ channel (DHP) were used to determine expression of possible downstream genes affected by the suppression of STAT3. Expression of specific cardiac genes was markedly lower in STAT3? EBs differentiated for at least 7–10 days after plating (Fig. 8B). Interestingly, promoter sequence analysis of the DHP gene, cardiac actin, and Nkx2.5 revealed putative STAT3 DNA binding sites, suggesting that STAT3? inhibition of STAT3 could lower or even prevent expression of these cardiac genes (Table 1).
Table 1. Promoters containing STAT and/or C/EBP sites
Closer inspection of promoter regions for many cardiac genes revealed that DNA binding sites were common for the interleukin (IL)-6 acute-phase response gene C/EBP, which has been shown to be reliant on STAT3 for expression . Semiquantitative RT-PCR analysis of STAT3? ES cell lines (lines 7A through 10D) revealed a loss or reduction in C/EBP? and expression (Fig. 8C). Table 1 summarizes cardiac genes with STAT3 or C/EBP regulatory elements. The fact that C/EBP? and sites are found within many cardiac genes and that STAT3? expression leads to the loss of these specific cardiac genes strongly suggests that expression of many cardiac genes is dependent on active STAT3 or C/EBP.
CONCLUSION
K.F. and G.R. contributed equally to this study. We would like to thank Dr. Robert Lechleider, Dr. Chris Taylor, and Tammy Gallicano for their critical reading of the manuscript. We also thank Alex C. Potocki for his confocal microscopy guidance. Work in this investigation was supported by the Transgenic Shared Resource under the directorship of G.I.G., who is supported by Cancer Center Support Grant CA51008-13. This work was primarily supported by grant HL70204-01 from the NIH and grant O265429U from the American Heart Association, both awarded to G.I.G.
REFERENCES
Arai A, Yamamoto K, Toyama J. Murine cardiac progenitor cells require visceral embryonic endoderm and primitive streak for terminal differentiation. Dev Dyn 1997;210:344–353.
Ji RP, Phoon CKL, Aristizabal O et al. Onset of cardiac function during early mouse embryogenesis coincides with entry of primitive erythroblasts into the embryo proper. Circ Res 2003;92:133–135.
Wobus AM, Wallukat G, Hescheler J et al. Pluripotent mouse embryonic stem cells are able to differentiate into cardiomyocytes expressing chronotropic responses to adrenergic and cholinergic agents and Ca2+ channel blockers. Differentiation 1991;48:173–182.
Metzger JM, Lin WI, Samuelson LC. Vital staining of cardiac myocytes during embryonic stem cell cardiogenesis in vitro. Circ Res 1996;78:547–552.
Hescheler J, Fleischmann BK, Lentini S et al. Embryonic stem cells: a model to study structural and functional properties in cardiomyogenesis. Cardiovasc Res 1997;36:149–162.
Xu C, Liguori G, Adamson E et al. Specific arrest of cardiogenesis in cultured embryonic stem cells lacking cripto-1. Dev Biol 1998;196:237–241.
Meyer N, Jaconi M, Landopoulou A et al. A fluorescent reporter gene as a marker for ventricular specification in ES-derived cardiac cells. FEBS Lett 2000;478:151–158.
Zhou X, Quann E, Gallicano GI. Differentiation of nonbeating embryonic stem cells into beating cardiomyocytes is dependent on downregulation of PKC? and in concert with upregulation of PKC. Dev Biol 2003;255:407–422.
Maltsev VA, Wobus AM, Rohwedel J et al. Cardiomyocytes differentiated in vitro from embryonic stem cells developmentally express cardiac-specific genes and ionic currents. Circ Res 1994;75:233–244.
Durbin JE, Hackenmiller R, Simon MC et al. Targeted disruption of the mouse Stat1 gene results in compromised innate immunity to viral disease. Cell 1996;84:443–450.
Kaplan MH, Schindler U, Smiley ST et al. Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity 1996;4:313–319.
Kaplan MH, Sun YL, Hoey T et al. Impaired IL-12 responses and enhanced development of Th2 cells in Stat4-deficient mice. Nature 1996;382:174–177.
Meraz MA, White JM, Sheehan KC et al. Targeted disruption of the mouse Stat1 gene in mice reveals unexpected physiologic specificity in the JAK-STAT signalling pathway. Cell 1996;84:431–442.
Thierfelder WE, van Deursen JM, Yamamoto K et al. Requirement for Stat4 in interleukin-12-mediated responses of natural killer and T cells. Nature 1996;382:171–174.
Liu X, Robinson GW, Wagner KU et al. Stat5 is mandatory for adult mammary glad development and lactogenesis. Genes Dev 1997;11:179–186.
Udy GB, Towers RP, Snell RG et al. Requirement of Stat5b for sexual dimorphism of body growth rates and liver gene expression. Proc Natl Acad Sci U S A 1997;94:7239–7244.
Imada K, Leonard WJ. The Jak-STAT pathway. Mol Immunol 2000;37:1–11.
Kunisada K, Tone E, Fujio Y et al. Activation of gp130 transduces hypertrophic signals via STAT3 in cardiac myocytes. Circulation 1998;98:346–352.
Booz GW, Day JNE, Baker KM. Interplay between the cardiac renin angiotensin system and JAK-STAT signaling: role in cardiac hy pertrophy, ischemia/reperfusion dysfunction, and heart failure. J Mol Cell Cardiol 2002;34:1443–1453.
Jacoby JJ, Kalinowski A, Liu M et al. Cardiomyocyte-restricted knockout of STAT3 results in higher sensitivity to inflammation, cardiac fibrosis, and heart failure with advanced age. Proc Natl Acad Sci U S A 2003;100:12929–12934.
Hattori R, Maulik N, Otani H et al. Role of STAT3 in ischemic preconditioning. J Mol Cell Cardiol 2001;33:1929–1936.
Negoro S, Kunisada K, Tone E et al. Activation of JAK/STAT pathway transduces cytoprotective signal in rat acute myocardial infarction. Cardiovasc Res 2000;47:797–805.
Xuan YT, Guo Y, Han H et al. An essential role of the JAK-STAT pathway in ischemic preconditioning. Proc Natl Acad Sci U S A 2001;98:9050–9055.
Mascareno E, El-Shafei M, Maulik N et al. JAK/STAT signaling is associated with cardiac dysfunction during ischemia and reperfusion. Circulation 2001;104:325–329.
Bolli R, Dawn B, Xuan Y.-T. Role of the JAK-STAT pathway in protection against myocardial ischemia/reperfusion injury. Trends Cardiovasc Med 2003;13:72–79.
Neubauer H, Cumano A, Muller M et al. Jak2 deficiency defines an essential developmental checkpoint in definitive hematopoiesis. Cell 1998;93:397–409.
Parganas E, Wang D, Stravopodis D et al. Jak2 is essential for signaling through a variety of cytokine receptors. Cell 1998;93:385–395.
Kotenko SV, Izotova LS, Pollack BP et al. Other kinases can substitute for Jak2 in signal transduction by interferon-gamma. J Biol Chem 1996;271:17174–17182.
Caldenhoven E, van Dijk TB, Solari R et al. STAT3beta, a splice variant of transcription factor STAT3, is a dominant negative regulator of transcription. J Biol Chem 1996;271:13221–13227.
Feng J, Witthuhn BA, Matsuda T et al. Activation of Jak2 catalytic activity requires phosphorylation of Y1007 in the kinase activation loop. Mol Cell Biol 1997;17:2497–2501.
Zhong H, Murphy TJ, Minneman KP. Activation of signal transducers and activators of transcription by alpha(1A)-adrenergic receptor stimulation in PC12 cells. Mol Pharmacol 2000;57:961–967.
Duncan SA, Zhong Z, Wen Z et al. STAT signaling is active during early mammalian development. Dev Dyn 1997;280:190–198.
Catlett-Falcone R, Landowski TH, Oshiro MM et al. Constitutive activation of Stat3 signaling confers resistance to apoptosis in human U266 myeloma cells. Immunity 1999;10:105–115.
Cantwell CA, Sterneck E, Johnson PF. Interleukin-6-specific activation of the C/EBPdelta gene in hepatocytes is mediated by Stat3 and Sp1. Mol Cell Biol 1998;18:2108–2117.
Niehof M, Streetz K, Rakemann T et al. Interleukin-6-induced tethering of STAT3 to the LAP/C/EBP? promoter suggests a new mechanism of transcriptional regulation by STAT3. J Biol Chem 2001;276:9016–9027.
Ernst M, Oates A, Dunn AR. gp130-mediated signal transduction in embryonic stem cells involves activation of Jak and Ras/mitogen-activated protein kinase pathways. J Biol Chem 1996;271:30136–30143.
Niwa H, Burdon T, Chambers I et al. Self-renewal of pluripotent embryonic stem cells is mediated via activation of STAT3. Genes Dev 1998;12:2048–2060.
Raz R, Lee C-K, Cannizzaro LA et al. Essential role of STAT3 for embryonic stem cell pluripotency. Proc Natl Acad Sci U S A 1999;96:2846–2851.
Zhang Y, Turkson J, Carter-Su C et al. Activation of Stat3 in v-Src-transformed fibroblasts requires cooperation of Jak1 kinase activity. J Biol Chem 2000;275:24935–24944.
Reecy JM, Li X, Yamada M et al. Identification of upstream regulatory regions in the heart-expressed homeobox gene Nkx2-5. Development 1999;126:839–849.
Liu L, Fan I, El-Zaru MR et al. Regulation of DHP receptor expression by elements in the 5'-flanking sequence. Am J Physiol Heart Circ Physiol 2000;278:H1153–H1162.
Chen CY, Schwartz RJ. Recruitment of the tinman homolog Nkx-2.5 by serum response factor activates cardiac alpha-actin gene transcription. Mol Cell Biol 1996;16:6372–6384.
Lyons I, Pasons LM, Hartley L et al. Myogenic and morphogenetic defects in the heart tubes of murine embryos lacking the homeo box gene Nkx2-5. Genes Dev 1995;9:1654–1666.
Kunisada K, Negoro S, Tone E et al. Signal transducer and activator of transcription 3 in the heart transduces not only a hypertrophic signal but a protective signal against doxorubicin-induced cardiomyopathy. Proc Natl Acad Sci U S A 2000;97:315–319.
Takeda K, Noguchi K, Shi W et al. Targeted disruption of the mouse Stat3 gene leads to early embryonic lethality. Proc Natl Acad Sci U S A 1997;94:3801–3804.
Valdembri D, Serini G, Vacca A et al. In vivo activation of JAK2/ STAT-3 pathway during angiogenesis induced by GM-CSF. FASEB J 2002;16:225–227.
Lumelsky N, Blondel O, Laeng P et al. Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science 2001;292:1389–1394.
Mazurova Y. New therapeutic approaches for the treatment of Huntington’s disease. Acta Medica 2001;44:119–123.
Vogel G. Stem cell research: rat brains respond to embryonic stem cells. Science 2002;295:254–255.
Quitschke WW, DePonti-Zilli L, Lin ZY et al. Identification of two nuclear factor-binding domains on the chicken cardiac actin promoter: implications for regulation of the gene. Mol Cell Biol 1989;9:3218–3230.
Wang Q, Sigmund CD, Lin JJ-C. Identification of cis elements in the cardiac troponin T gene conferring specific expression in cardiac muscle of transgenic mice. Circ Res 2000;86:478–484.
Parmacek MS, Vora AJ, Shen T et al. Identification and characterization of a cardiac-specific transcriptional regulatory element in the slow/cardiac troponin gene. Mol Cell Biol 1992;12:1967–1976.
Meydan N, Grunberger T, Dadi H et al. Inhibition of acute lymphoblastic leukaemia by a Jak-2 inhibitor. Nature 1996;379:645–648.(Kara Foshaya, Gisella Rod)