Differential Regulation of Cellular Senescence and Differentiation by Prolyl Isomerase Pin1 in Cardiac Progenitor Cells*

Background: Pin1 is a prolyl isomerase that modulates the structure of phosphoproteins. Results: Loss of Pin1 causes cell cycle arrest and senescence, whereas Pin1 overexpression increases differentiation and inhibits senescence of cardiac progenitor cells (CPCs). Conclusion: Pin1 has pleiotropic roles in CPCs. Significance: Pin1 may be a molecular target to enhance repair, survival, and differentiation and antagonize senescence of CPCs. Autologous c-kit+ cardiac progenitor cells (CPCs) are currently used in the clinic to treat heart disease. CPC-based regeneration may be further augmented by better understanding molecular mechanisms of endogenous cardiac repair and enhancement of pro-survival signaling pathways that antagonize senescence while also increasing differentiation. The prolyl isomerase Pin1 regulates multiple signaling cascades by modulating protein folding and thereby activity and stability of phosphoproteins. In this study, we examine the heretofore unexplored role of Pin1 in CPCs. Pin1 is expressed in CPCs in vitro and in vivo and is associated with increased proliferation. Pin1 is required for cell cycle progression and loss of Pin1 causes cell cycle arrest in the G1 phase in CPCs, concomitantly associated with decreased expression of Cyclins D and B and increased expression of cell cycle inhibitors p53 and retinoblastoma (Rb). Pin1 deletion increases cellular senescence but not differentiation or cell death of CPCs. Pin1 is required for endogenous CPC response as Pin1 knock-out mice have a reduced number of proliferating CPCs after ischemic challenge. Pin1 overexpression also impairs proliferation and causes G2/M phase cell cycle arrest with concurrent down-regulation of Cyclin B, p53, and Rb. Additionally, Pin1 overexpression inhibits replicative senescence, increases differentiation, and inhibits cell death of CPCs, indicating that cell cycle arrest caused by Pin1 overexpression is a consequence of differentiation and not senescence or cell death. In conclusion, Pin1 has pleiotropic roles in CPCs and may be a molecular target to promote survival, enhance repair, improve differentiation, and antagonize senescence.

Ever since the initial discovery of c-kit ϩ cardiac progenitor cells (CPCs) 3 in the adult heart (1), many researchers have committed to understanding the biology of CPCs and deciphering mechanisms to better manipulate them. The potential therapeutic capacity of autologous CPCs derived and expanded from patients with ischemic cardiomyopathy has been proven in animal models and phase I clinical trials (1)(2)(3). However, the efficiency of CPC-mediated regeneration may be further augmented by delineating molecular mechanisms of endogenous cardiac repair and enhancing pro-survival pathways that maintain the youthfulness of stem cells, whereas also increasing differentiation. Prolyl isomerase Pin1 is one such molecule that functions as a molecular orchestrator of signaling yielding multifarious beneficial effects (4).
Pin1 regulates the stability, activity, and subcellular localization of proteins by controlling the three-dimensional structure of phosphoproteins through alterations of cis-trans isomerization upon recognition of a major regulatory phosphorylation motif (Ser/Thr-Pro) (4). Pin1 is expressed in proliferating cells and exerts wide ranging influence upon multiple regulatory molecules of the mitotic processes including cyclins and cell cycle inhibitors such as Cyclin D, Cyclin B, Cyclin E, p53, and retinoblastoma (Rb) to name a few (5)(6)(7)(8). By functioning as a molecular timer, Pin1 tightly controls cell cycle progression, cell division, and in turn also regulates cellular senescence (9 -12). Loss of Pin1 in cultured cells stabilizes p53 and induces senescence (13). Pin1 knock-out mice exhibit a premature senescent phenotype (14) and decline in Pin1 expression in the myocardium correlates with aging (4), suggesting a critical role for Pin1 as an anti-aging molecule. Lineage commitment and cell differentiation that are highly relevant in the context of stem cells are also dramatically influenced by Pin1. Pin1 maintains the balance between stem cell pluripotency, stemness, and commitment by stabilizing and activating molecules controlling self-renewal and differentiation (15)(16)(17). These provocative roles of Pin1 in the areas of proliferation, survival, senescence, and cell fate determination have tremendous impact on stem cell biology that remains unexplored in the context of CPCs. Furthermore, our recent study uncovering the role of Pin1 in regulating cardiac hypertrophy makes this molecule central to many of the most significant areas of myocardial signal transduction currently under investigation.

EXPERIMENTAL PROCEDURES
Mice Surgery and Echocardiography-All experimental procedures were performed according to the guidelines established by San Diego State University for experiments in animals and all protocols were approved by the Institutional Animal Care and Use Committee. Myocardial infarction was surgically induced in female 10-week-old Pin1 knock-out (KO) mice and wild type (WT) controls as previously described (18). Mice were anesthetized using 1-2% isoflurane and echocardiography was performed using Vevo770 High Resolution In Vivo Micro-Imaging System (Visual Sonics) as described in Ref. 4. B-mode echocardiography of the left ventricle was recorded in a parasternal long-axis view and left ventricle end-diastolic volume, and left ventricle end-systolic volumes were obtained.
Cardiac Progenitor Cell Isolation-CPCs were isolated from FVB mice using anti-c-kit antibody as described previously (18,19).
siRNA and Lentivirus-CPCs were transfected with siRNA specific to Pin1 (Invitrogen) using HiPerfect (Qiagen) following the manufacturer's instructions (18). Lentiviruses harboring EGFP or human Pin1 with EGFP were made as previously described (19). CPCs were transduced with lentiviruses at multiplicity of infection of 10 to obtain stable cell lines.
Proliferation and Differentiation Assay-CPCs were cultured in full medium consisting of DMEM/F-12 with 10% FBS, 1% penicillin-streptomycin-glutamine (PSG), 0.02 mg/ml of basic FGF, 0.04 mg/ml of EGF, 1000 units/ml of leukemia inhibitory factor (LIF) and insulin-transferrin-sodium selenite (ITS). CPC numbers were determined using MTT and CyQuant proliferation assays (Invitrogen) following the manufacturer's instructions (18,19). Viable cells were also manually counted under a microscope by staining CPCs with trypan blue and excluding blue cells. Differentiation was induced by culturing CPCs in ␣ minimal essential media and treating with dexamethasone (Dex, 10 nM) for 6 days, as described previously (19).
Cell Cycle and Cell Death Analyses-CPCs were cultured for 2 days in DMEM/F-12 media supplemented with growth factors but deprived of FBS to synchronize cells at G 0 . Medium was replaced with full medium supplemented with 10 M 5-bromo-2Ј-deoxyuridine (BrdU) to release and label cells progressing through the cell cycle. Every 12 h after release, cells were harvested, fixed in 70% ethanol, and processed for flow cytometric analyses as described in detail (20). For cell death assays, CPCs were stained with propidium iodide briefly (2 min). Flow cytometry was performed on FACSCanto (BD Biosciences) and data were processed using FlowJo software.
Clonogenicity Assay-Clonogenicity was determined in stable CPC lines wherein CPCs were single-cell sorted based on fluorescent protein expression by FACS, cultured in a 96-well plate, and observed periodically over an 8-day period. FACS was performed on FACSAria (BD Biosciences).
Immunoblot Analysis-Whole cell lysates were resolved by SDS-polyacrylamide gel electrophoresis and immunoblot analyses were performed as described (4). Immunoblot band intensities were measured using ImageJ software (NIH) (4).
RNA Extraction and Quantitative Real Time PCR Assay-Total RNA was extracted from cultured CPCs, complementary DNA was synthesized, and quantitative real time PCR (qRT-PCR) was performed on samples as described (4).
Microscopy-Immunohistochemistry on mouse heart sections and cultured CPCs was performed as described previously and images were obtained on a Leica SP2 Confocal microscope (18,19). Senescent CPCs were detected using the senescenceassociated ␤-galactosidase (SA-␤-gal) assay (Abcam number ab65351) following the manufacturer's protocol. Images were obtained on Olympus IX70 microscope.
Antibodies-Antibodies used in the study include those against Pin1 Cellular Fractionation-Cellular fractionation was performed using the PARIS kit (Invitrogen, number AM 1921) following the manufacturer's instructions.
Statistical Analysis-Data are presented as mean Ϯ S.E. Statistical analysis was performed using GraphPad Prism (GraphPad Software Inc.). Multiple group comparison was performed by one-way analysis of variance. For time course analysis of echocardiographic assessment of cardiac function, cell proliferation and cell cycle two-way analysis of variance was applied. For both procedures Bonferroni post-tests were performed. Comparisons between two groups were performed using Student's t test. p values Ͻ 0.05 were considered statistically significant.
with c-kit in hearts subjected to myocardial infarction where c-kit ϩ CPCs were detected (Fig. 1A). Pin1 was expressed in CPCs cultured under low and high serum conditions, with 1.4fold more Pin1 protein in proliferating CPCs cultured under high serum conditions (Fig. 1B). Pin1 expression declined dramatically (Ϫ60%) in CPCs treated with Dex, a nonspecific inducer of differentiation (Fig. 1C). Interestingly Pin1 was localized to both the cytosol and nucleus under normal culture conditions, but became predominantly cytosolic upon Dexinduced differentiation (Fig. 1D). Furthermore, cellular fractionation analyses revealed that Dex-induced differentiation decreased both cytosolic and nuclear Pin1 expression (Fig. 1E). Collectively, Pin1 is expressed in c-kit ϩ CPCs and may have a role in regulation of proliferation and differentiation.
Pin1 Is Required for Proliferation of CPCs-The precise role of Pin1 in CPCs was elucidated by silencing Pin1 using siRNA (siPin1). Following confirmation of successful knockdown ( Fig.  2A), proliferation rates of CPCs were determined. Loss of Pin1 attenuated CPC proliferation by 38 and 44% after 6 days in culture as determined by CyQuant and MTT assays (Fig. 2, B and C). This finding prompted examination of the exact stage of the cell cycle that Pin1 interferes with for knockdown. To this end, serum-starved CPCs synchronized in G 0 were labeled with BrdU and released to progress through the cell cycle over 36 h in culture. Pin1 silencing dramatically decreased the percentage of cells entering S phase (53%), whereas control siRNA-treated CPCs were predominantly BrdU positive (95% within 36 h in growth media) (Fig. 2D). Pin1 knockdown increased the percentage of cells that were stuck in G 1 with 80% of Pin1-silenced CPCs remaining in G 1 after 24 h compared with only 41% in control CPCs (Fig. 2E). These results indicate that Pin1 is required for cell cycle progression and loss of Pin1 causes G 1 phase cell cycle arrest.
Loss of Pin1 Causes Senescence in CPCs-Expression of cell cycle regulatory proteins established to be Pin1 targets were analyzed upon Pin1 knockdown. Cyclin D, which is crucial for progression through G 1 (21), was significantly decreased upon Pin1 silencing (Ϫ30%) (Fig. 3A). Cyclin B, required for progression into mitosis and a target of Pin1 signaling (6), was also decreased by Pin1 knockdown (Ϫ55%) (Fig. 3A). Cell cycle inhibitors p53 and Rb play critical roles in the G 1 /S checkpoint (22) and were up-regulated 4.8-and 1.7-fold, respectively, upon Pin1 silencing in CPCs (Fig. 3B). Because robust up-regulation of p53 and Rb is a trigger for senescence (23), we determined if loss of Pin1 induces cellular senescence in CPCs and treated cells with a senesecence-associated ␤-galactosidase assay (SA-␤-gal), a well established biomarker for senescence (24). Pin1  4). C-E, CPCs were treated with and without Dex (10 nM) for 3 days to induce differentiation. C, Pin1 expression decreases upon differentiation as shown by immunoblots and densitometric analyses (n ϭ 4). D, confocal microscopic images and quantitation shows diminished nuclear (blue) and increased cytosolic localization of Pin1 (red) following Dex treatment. E, cellular fractionation analyses reveal decreased Pin1 expression in whole cell lysate, cytosolic (Cyto), and nuclear (Nuc) fractions in CPCs treated with Dex. GAPDH and Lamin A/C show specificity of the cytosolic and nuclear fractions, respectively. *, p Ͻ 0.05; ***, p Ͻ 0.001. knockdown caused a 3-fold increase in the number of senescent cells (Fig. 3C). Interestingly loss of Pin1 did not increase basal cell death of CPCs and significantly decreased serum starvation-induced apoptosis (Ϫ55%) (Fig. 3D). The differentiation potential of CPCs was also determined because Pin1 is known to regulate molecules involved in cell fate determination. Pin1 silencing decreased the ability of CPCs to differentiate as determined by protein and mRNA expression of differentiation markers ␣-smooth muscle actin (␣-SMA) and cardiac troponin T, respectively following treatment with Dex (Fig. 3, E and F). Collectively, these data suggest that loss of Pin1-induced cell cycle arrest is attributable to acquisition of senescence and not to differentiation or cell death of CPCs.
Pin1 activity was inhibited by treatment with juglone to determine whether inhibition of Pin1 activity yields similar phenotypes as Pin1 knockdown. Expression of Cyclin D was significantly decreased (Ϫ29%), whereas p53 was up-regulated 1.3-fold upon Pin1 inhibition (Fig. 3H). A small but significant increase in the number of senescent cells (9.5%, Fig. 3I) was also observed, suggesting that inhibition of Pin1 activity partly recapitulates the phenotypes of Pin1 knockdown.
Loss of Pin1 Attenuates CPC Expansion after Myocardial Infarction-The impact of Pin1 depletion upon CPC expansion in vivo was determined after induction of myocardial infarction by ligation of the left anterior descending artery in adult Pin1 knock-out (KO) mice and wild type littermates (WT). The number of c-kit ϩ CPCs was markedly reduced in infarcted KO hearts compared with WT (69% decrease; Fig. 4, A and C). KO hearts also had significantly fewer (Ϫ42%) proliferating c-kit ϩ CPCs as determined by co-staining of c-kit with Ki-67, a cellular proliferation marker (Fig. 4, B and D). No significant differences in cardiac geometry and function were detected between KO and WT mice as determined by echocardiographic assessment 12 weeks after infarction (Fig. 4, E and F).

Pin1 Overexpression Causes Cell Cycle
Arrest-CPCs were stably transduced with lentiviruses overexpressing Pin1 and GFP (Pin1) or EGFP alone (EGFP) to better understand Pin1mediated regulation of CPC biology. Up-regulation of Pin1 (5-fold, Fig. 5A) attenuated CPC proliferation by 52 and 29% as determined by viable cell count and MTT assays (Fig. 5, B and  C). Detailed cell cycle analyses revealed that Pin1 overexpression did not affect cell cycle re-entry and S phase progression after synchronization as measured by BrdU incorporation (Fig.  5D). Consistently, no significant change was detected in the percentage of cells in G 1 between Pin1 and EGFP at 24 h after release (Fig. 5E). Interestingly, however, Pin1 overexpression caused a 2-fold increase in the proportion of cells in the G 2 /M phase at 36 h after release when CPCs expressing EGFP re-entered G 1 (Fig. 5E). Thus Pin1 overexpression in CPCs caused cell cycle arrest by blocking G 2 /M exit.
Pin1 Overexpression Antagonizes Senescence and Increases Differentiation of CPCs-Expression of cyclins and cell cycle inhibitors were detected upon Pin1 overexpression in CPCs. A significant decrease in Cyclin B protein level (Ϫ25%), but not Cyclin D was observed in CPCs overexpressing Pin1 (Fig. 6A), consistent with G 2 /M cell cycle arrest. Pin1 overexpression caused a significant reduction (Ϫ30%) in expression of both p53 and Rb (Fig. 6B). Because expression of molecular regulators of senescence was decreased, we determined if Pin1 can inhibit senescence. Pin1 overexpression decreased the percentage of senescent cells (Ϫ54%) relative to CPCs expressing EGFP, which expressed more senescent cells in late culture passage (Fig. 6C), suggesting a role for Pin1 in antagonizing replicative senescence. Pin1 overexpression also decreased serum deprivation-induced cell death in CPCs (Fig. 6D). Differentiation was enhanced upon Pin1 overexpression and Dex treatment in CPCs, as determined by protein and mRNA expression of ␣-SMA and cardiac troponin T, respectively (Fig. 6, E and F). Taken together, these results indicate that cell cycle arrest caused by Pin1 overexpression is a consequence of increased differentiation and not senescence or cell death.
Stemness and multipotency were also determined upon Pin1 overexpression in CPCs. Interestingly, c-kit expression was significantly decreased (Ϫ35%, Fig. 6G), consistent with increased differentiation observed upon Pin1 overepression. However, typical multipotency factors were differentially regulated upon Pin1 overexpression in CPCs. Dramatic up-regulation of KLF4 (1.5-fold) and Nanog (2.4-fold), but not c-Myc and Oct4 (Fig.  6G), was detected upon Pin1 overexpression in CPCs, suggesting that Pin1 overexpression correlates with an intermediate phenotype of stemness in CPCs. No differences in clonogenic-  ity were seen upon Pin1 overexpression in CPCs, as determined by the size and number of clones that grew from single cell sorting (Fig. 6H).

DISCUSSION
This study is the first to our knowledge that characterizes the role of Pin1 in c-kit ϩ CPCs. Pin1 expression in vitro and in vivo correlates with a highly proliferative state of CPCs. Serum starvation and Dex treatment both decrease CPC proliferation 4 and are concomitantly associated with reduced Pin1 expression (Fig. 1), suggesting a critical role for Pin1 in regulating CPC proliferation and differentiation. Consistently our data demonstrates significant reduction in CPC numbers upon loss of Pin1 in vitro (Fig. 2) and in vivo (Fig. 4). That Pin1 KO mice have dramatically fewer proliferating CPCs after myocardial infarction highlights the significance and requirement of Pin1 for endogenous CPC expansion upon ischemic challenge.
Loss of Pin1 induces G 1 phase cell cycle arrest in cultured CPCs (Fig. 2) and consequentially molecules essential for G 1 /S phase progression are dramatically affected by Pin1 silencing (Fig. 3). On the contrary, Pin1 overexpression also causes cell cycle arrest in the G 2 /M phase in CPCs (Fig. 5). This is in line with evidences in HeLa cells and Xenopus egg extracts where overexpression of Pin1 has been shown to inhibit cell proliferation (25,26). Therefore high expression of Pin1 in several human cancers could represent a compensatory negative feedback mechanism to curtail malignant proliferation (11).
Mitotic proteins such as the cyclins and cell cycle inhibitors are influenced by Pin1 signaling. Genetically engineered Pin1 KO mice exhibit a phenotype resembling Cyclin D1 null phenotypes with cell proliferative abnormalities and defective cell cycle progression (27). Pin1 directly interacts with Cyclin D to stabilize its expression by inhibition of degradation (8). On the other hand, Pin1 can also interact with Ras or Akt, and both signals are involved in transcriptional induction of Cyclin D (4,(27)(28)(29). Pin1 regulates transition through mitotic phases to allow for completion prior to progression, such as in exit from the replication checkpoint (6) and regulating activation of Cyclin B (30). By interacting with different upstream and downstream regulators of Cyclin B such as Cdc25c, Plx1, Myt1, and Wee1, Pin1 tightly controls Cyclin B activity and expression in a cell type-dependent manner, thereby accounting for declined Cyclin B expression upon manipulation of Pin1 levels (26,30,31). Pin1 interacts with cell cycle inhibitor p53 in response to stress and causes conformational alterations that inhibit the interaction between p53 and E3 ubiquitin ligase murine double  FEBRUARY 28, 2014 • VOLUME 289 • NUMBER 9 minute (MDM2). This prevents MDM2-mediated degradation of p53 thereby stabilizing the protein and activating downstream signaling cascades (7,32,33), Pin1 also modulates activity of Che1, an RNA polymerase II-binding protein that participates in p53 activation and maintenance of the G 2 /M checkpoint (34). The downstream signaling response of p53 stabilization is cell type-dependent and varies between cell cycle arrest and cell death. In the context of CPCs, loss of Pin1 increases p53 concomitantly associated with increased cellular senescence but attenuated cell death (Fig. 3). Consistent phenotypes of Cyclin D down-regulation, p53 accumulation, and acquisition of senescence are also seen upon inhibition of Pin1 activity by juglone treatment in CPCs (Fig. 3). Whether loss of function of Pin1-induced CPC senescence is mechanistically p53 dependent remains to be determined. Pin1 also directly interacts with Rb and selectively alters Rb phosphorylation from hypo-to hyperphosphorylation in tumors (35). Rb acts as a tumor suppressor protein and is involved in regulation of G 1 -S progression, senescence, and differentiation (36).

Pin1 Regulates Senescence and Differentiation
That loss of Pin1 causes senescence and Pin1 overexpression attenuates replicative senescence in CPCs is corroborative with several lines of evidence linking Pin1 to age-associated diseases. Pin1 knock-out mice have premature senescence with an early onset of age-dependent neuropathy and increased neuronal degeneration, similar to Alzheimer disease (37). Pin1 maintains telomere length (14), regulates the metabolism of reactive oxygen species in cells (38,39), modulates autophagy (40), and controls protein ubiquitination and degradation (41,42), all of which contribute to regulation of organismal and cellular lifespan. The role of Pin1 in regulating cardiac aging and age-associated cardiomyopathies is a topic of future investigation.
Our data suggests that Pin1 is regulated by differentiation and in turn modulates the ability of CPCs to differentiate and commit (Figs. 1, 3, and 6). Expression and localization of Pin1 is altered during Dex-induced differentiation in CPCs (Fig. 1). Cell confluence of CPCs is impacted by the presence or absence of Dex, however, Pin1 localization is not affected by changes in cell confluence. 4 Thus, the cytosolic localization of Pin1 observed during Dex treatment (Fig. 1D) is independent of increases in cell density and dependent mainly on Dex-induced differentiation. Pin1 plays an active role in both the cytosol and nucleus by targeting different substrates in each of the cellular compartments. Cell cycle progression, chromosomal condensation, and regulation of transcription factors c-Fos, c-Jun, NFB, and Stat3 are some of the roles Pin1 plays in the nucleus (43). It is tempting to speculate from our findings that cytosolic Pin1 may specifically have a role in the regulation of differentiation in CPCs. Several published findings link the serine threonine kinase Akt and Pin1 (4,28,44). The spatiotemporal pattern of Akt activity governs cellular fate; Akt activity is mainly cytosolic during differentiation of skeletal muscle and neural cells and our own findings demonstrate that forced nuclear localization of Akt inhibits differentiation of CPCs (45,46). Pin1 and Akt interaction in the cytosol to synergistically regulate differentiation upon Dex treatment in CPCs seems a reasonable speculation, but requires further investigation. Pin1 also modulates several other signaling molecules involved in cell fate determination. By stabilizing ␤-catenin Pin1 induces neuronal differentiation of neural progenitor cells (16). A strong correlation between Pin1 expression and Notch activity has also been reported implicating Notch as a substrate mediating effects of Pin1 signaling and regulating proliferation and lineage commitment (17).
Another striking consequence of Pin1 manipulation are the effects upon typical stemness and multipotency markers in CPCs. Expression of surface marker c-kit is dramatically decreased during both overexpression and loss of Pin1 (Figs. 3 and 6), consistent with increased differentiation and acquisition of senescence in CPCs. Pin1 is essential for maintenance of stem cell self-renewability and pluripotency in embryonic stem cells and stabilizes typical pluripotency factors such as c-Myc, KLF4, Oct4, and Nanog (15,47,48). Consistently loss of Pin1 dramatically decreases expression of c-Myc, KLF4, Oct4, and Nanog in CPCs (Fig. 3). Interestingly, however, hyperactivation of Pin1 does not cause the exact opposite and a differential regulation of multipotency factors is observed (Fig. 6).
Pin1 is touted to both enhance and suppress cell survival in a context dependent manner (25,34,39,40,49,50), however, both gain and loss of Pin1 seems to protect CPCs from serum starvation-induced apoptosis (Figs. 3 and 6). It is possible that the intracellular and environmental cues associated with cumulative effects from the regulation of myriad signaling cascades determine whether cell survival is potentiated or inhibited by Pin1 action.
Overall, this study identifies pleiotropic roles of Pin1, placing Pin1 as a master regulator of the fundamental relationships between cell proliferation, senescence, differentiation, and stemness (Fig. 7). The disparate regulation of cellular senescence and differentiation, together with the requirement of Pin1 for endogenous CPC response and phenotypic characteristics makes Pin1 a coveted molecular interventional target to promote survival, enhance myocardial regeneration, and antagonize aging. Although the simultaneous regulation of multiple signaling pathways collectively facilitates the different phenotypes seen upon Pin1 manipulation, further research is needed to determine strategies to ensure target specificity, which remains a significant challenge with Pin1.