Monocyte Chemoattractant Protein-Induced Protein 1 (MCPIP1) Enhances Angiogenic and Cardiomyogenic Potential of Murine Bone Marrow-Derived Mesenchymal Stem Cells

The current evidence suggests that beneficial effects of mesenchymal stem cells (MSCs) toward myocardial repair are largely due to paracrine actions of several factors. Although Monocyte chemoattractant protein-induced protein 1 (MCPIP1) is involved in the regulation of inflammatory response, apoptosis and angiogenesis, whether MCPIP1 plays any role in stem cell-induced cardiac repair has never been examined. By employing retroviral (RV)-transduced overexpression of MCPIP1, we investigated the impact of MCPIP1 on viability, apoptosis, proliferation, metabolic activity, proteome, secretome and differentiation capacity of murine bone marrow (BM) - derived MSCs. MCPIP1 overexpression enhanced angiogenic and cardiac differentiation of MSCs compared with controls as indicated by elevated expression of genes accompanying angiogenesis and cardiomyogenesis in vitro. The proangiogenic activity of MCPIP1-overexpressing MSCs (MCPIP1-MSCs) was also confirmed by increased capillary-like structure formation under several culture conditions. This increase in differentiation capacity was associated with decreased proliferation of MCPIP1-MSCs when compared with controls. MCPIP1-MSCs also expressed increased levels of proteins involved in angiogenesis, autophagy, and induction of differentiation, but not adverse inflammatory agents. We conclude that MCPIP1 enhances endothelial and cardiac differentiation of MSCs. Thus, modulating MCPIP1 expression may be a novel approach useful for enhancing the immune-regulatory, anti-apoptotic, anti-inflammatory and regenerative capacity of BM-derived MSCs for myocardial repair and regeneration of ischemic tissues.


Introduction
Bone marrow (BM)-derived mesenchymal stem cells (MSCs) have been used extensively for tissue regeneration due to their broad differentiation potential, low immunogenicity and innate anti-inflammatory properties [1][2][3]. It has also been reported that a fraction of MSCs are able to differentiate into cardiomyocytes and endothelial cells following stimulation in vitro or in vivo [4][5][6][7][8][9]. Several recent studies indicate that therapy with BM-derived MSCs improves left ventricular (LV) function and myocardial perfusion after myocardial infarction (MI) [1,[10][11][12]. However, the benefits of MSC therapy for cardiac repair in vivo has been variable [1,10].
Monocyte Chemoattractant Protein-1-Induced Protein 1 (MCPIP1; Zc3h12a) has been identified in human macrophages following stimulation with interleukin 1β (IL-1β) [21]. Although the highest level of MCPIP1 has been found in leukocytes, it may also be expressed in other cell types [21]. MCPIP1 has been shown to be induced by several proinflammatory agents and cytokines, and may act as a macrophage activator and negative regulator of oxidative stress and inflammatory gene expression [22]. Moreover, overexpression of MCPIP1 in these cells significantly decreased promoter activity of tumor necrosis factor α (TNF-α) and inducible nitric-oxide synthase (iNOS) in a dose-dependent manner, indicating anti-inflammatory properties [22]. Interestingly, it has recently been shown that MCPIP-1 exhibited RNAse activity due to the presence of a PilT N-terminus (PIN) domain and may play a regulatory (including anti-inflammatory) role via mRNA and pre-miRNA decay in different cell types, including non-hematopoietic cells [21,23,24]. Importantly, Niu and colleagues have reported that MCPIP1 may be involved in enhancing angiogenic activity and secretion of several proangiogenic factors in human umbilical vein endothelial cells (HUVECs) [25]. Moreover, the specific knockdown of MCPIP1 in these cells resulted in suppression of VEGF and hypoxia-inducible factor-1α (HIF-1α), which are important agents involved in proangiogenic and antioxidative cell responses [25]. Notably, studies employing transgenic mice with myocardial MCPIP1 expression (under an α-MHC promoter) have reported extensive attenuation of inflammatory response-related cardiac dysfunction [26]. Data from several laboratories strongly indicate that MCPIP1 may reside in both cytoplasmic as well as nuclear compartments with distinct roles in different cell types [21,23,27].
Although the impact of MCPIP1 on angiogenic potential, survival and inflammatory response was studied in several mature cell types, the role of MCPIP1 in stem cells, such as MSCs has never been investigated. Thus, in this study we evaluated the impact of MCPIP1 on several functions of MSCs, including cell viability and apoptosis, proliferation, metabolic activity, transcriptomic, proteome and secretome profiles, as well as angiogenic and cardiomyogenic differentiation capacity.

Animals
Four-to-six-week-old C57Bl/6 mice were used for experiments. All procedures were performed in accordance with the approval of the Ethical Committee on Animal Testing at the Jagiellonian University (JU) in Krakow (approval number: 31/2012). C57Bl/6 mice were supplied for experiments by Charles River (Wilmington MA, USA) and were subsequently held for 7 day quarantine, prior to their experimental use, in Animal Facility of Department of Biophysics at the Faculty of Biochemistry, Biophysics and Biotechnology JU in Krakow. Mice were handled accordingly to standard animal procedures complied in controlled animal facilities. Up to 5 mice of the same gander were kept per one cage and were constantly controlled for health conditions, supplied with water and dry fodder containing balanced nutrition and were handled accordingly to the standard regulations. Animals were sacrificed by intraperitoneal injection of lethal dose of pentobarbital (Morbital; Biowet, Pulawy, Poland; 100mg/kg b.w.) representing acceptable method for mice euthanasia and accordingly to the ethical approval listed above. Tibias and femurs were harvested immediately following the animal euthanasia.

Isolation and culture of MSCs
Bone marrow cells were harvested by flushing cavities of tibia and femur bones with DMEM/ F12 medium (Sigma-Aldrich). Cells were centrifuged, re-suspended in complete medium (DMEM/F12 with 10% FBS, Sigma-Aldrich; and penicillin/ streptomycin, Gibco, Life Technologies) and seeded into a Primaria culture flask (BD Falcon) at a density of 25x10 6 nucleated cells/ 75cm 2 . Flushed bones were additionally fragmented and enzymatically digested with collagenases type I and II (1 mg/ml; Sigma Aldrich) for 1.5h at 37°C. Released cells were washed and added equally to flasks containing flushed BM cells. Cells were cultured in standard conditions for 72h and non-adherent cells were removed. Cells were passaged with 0.25% trypsin/ EDTA (Gibco, Life Technologies) when confluence of cells reached close to 90%.

Retrovirus generation
Retroviral vectors containing the following plasmids were prepared in our laboratory and used in the study: pMX-MCPIP1 coding for MCPIP1, pMX-GFP (control plasmid containing insert with enhanced green fluorescence protein (GFP)), and pMX-Puro (empty plasmid, devoid of any insert). Retrovirus packaging was performed in a modified HEK293 (Phoenix Amphotropic) as described earlier [28].

Retroviral transduction
To obtain MCPIP1 overexpressing MSCs, retrovirus-mediated gene transfer based on pMX-Puro system was performed [29]. MSCs on passage 3 to 4 were transduced with vectors: pMX-MCPIP1, pMX-GFP or pMX-Puro. Each vector was added to 0.5x10 6 MSCs in the presence of 4 μg/ml polybrene (Merck Millipore) in complete growth medium (MOI = 7). The medium was changed after 24h. In order to increase the population of MCPIP1-overexpressing cells, transduction was repeated after 48h.

Viability, proliferation and metabolic activity assessment
The viability and metabolic status of MSCs were measured by MTT assay and ATP concentration, respectively. Tests were performed at 48 and 72h after second transduction.
Infinite 200 microplate reader (Tecan Group Ltd.) at a wavelength of 570 nm with background subtraction at 650 nm.

ATP concentration measurement
MSCs were seeded on 96-well white plates to obtain 90% of confluence in each time point and the ATP Lite Luminescence assay kit was performed according to manufacturer's instructions (Perkin Elmer). Luminescence was measured using the SpectraFluor Plus (Tecan Group Ltd) microplate reader.

Proliferation ratio assessment
The relative proliferation ratio of MSCs was evaluated at 72h after the second transduction. MSCs were counted with the Countess II Automated Cell Counter (Life technologies).

Western blotting
Expression of MCPIP1 at the protein level was detected by western blotting [30]. Briefly, MCPIP1 was assessed using primary rabbit anti-MCPIP1 antibody [21]. As a loading control, primary mouse anti-actin antibody (Sigma-Aldrich) was used. Detailed procedure is included in S1 Appendix (Extended Materials and Methods).

Necrosis and apoptosis detection
Assessment of necrosis and apoptosis was done with Annexin V Apoptosis Detection Kit (BD Biosciences) and by Vybrant FAM Caspase-3 and 7-Assay Kit (Life Technologies) with flow cytometry. Staining was performed according to manufacturer's protocols and analyzed using using LSR II flow cytometer and FACS Diva software (Becton Dickinson).

Global proteomic analysis-sample processing
MSCs (1.7 x 10 6 ) from all groups were lysed in 4% SDS and 0.1 M DTT in Tris-HCl (BioShop, pH 7.6) and sonicated for 10 min. The samples were further incubated at 95°C for 5 min and centrifuged at 30,000g for 30 min at 20°C. Then, the proteins were digested using the filter aided sample preparation (FASP) technique [31] (detailed procedures including liquid chromatography and tandem mass spectrometry as well as analysis of proteomic data are included in S1 Appendix (Extended Materials and Methods).

Capillary-like tube formation assay
Twenty four-well plates were coated with Matrigel Matrix Grow Factor Reduced (BD Pharmingen) (100μl/well) and incubated at 37°C for 30 min. MCPIP1-expressing MSCs and control MSCs were seeded at a density of 1x10 5 cells/well in the EGM-2MV medium (Clonetics, Lonza). HUVECs were used as a positive control, while freshly isolated total nucleated cells (TNCs) from murine BM were utilized as a negative control. Cells were incubated for 12 or 18h at 37°C. Tube formation was investigated every 2h and images were collected with an Olympus IMT-2 microscope equipped with a CCD camera. The results were computed as absolute number of capillary-like structures and branches per field formed by cultured cell groups at different time points along the assay.
Real-time PCR was performed using Sybr Green qPCR Master Mix (EURx), cDNA template (10ng), forward and reverse primer (1μM; Genomed). The sequences of primers are included in S1 Appendix (Extended Materials and Methods). Relative quantification of gene expression was calculated using the comparative C t method. The relative quantitative value of the target normalized to an endogenous control (β2-microglobulin gene) and relative to a calibrator was expressed as 2 -ΔΔCt (i.e.-fold difference).

Immunocytochemistry
Immunocytochemistry staining was performed to evaluate: i) Gata-4 and troponin T-C; and ii) Gata-2 and VE-cadherin expression in MSC-derived cardiac and endothelial cells, respectively. Cells were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, washed, and stained with primary anti-Gata-4 and anti-troponin T-C antibodies or with anti-Gata-2 and anti-VE-cadherin antibodies followed by secondary antibodies and DAPI. Cells were analyzed with a Leica DM IRE2 (Ver. 4000) fluorescent microscope. A detailed description of the staining procedure and used antibodies are included in S1 Appendix (Extended Materials and Methods).

Secretome analysis
Western blotting (semi-quantitative). MCPIP1-expressing MSCs and control cells at 72h following the second transduction were cultured in EGM-2MV medium for 10 days and subsequently were washed twice with DMEM/F12 and further cultured in DMEM/F12 with 0.5% BSA (Sigma-Aldrich) for 24h. Cell culture supernatants were collected and frozen at -80°C until analysis. Analysis of expression of 53 angiogenesis-related proteins was carried out using the Proteome Profiler Mouse Angiogenesis Array Kit (R&D Systems) according to the manufacturer's protocol. The average signals (pixel density) computed from duplicate spots representing each angiogenesis-related protein were determined by Quantity One software (BioRad). An average background signal was subtracted from each spot during the analysis.
Multiplex analysis (quantitative). In addition, cell culture supernatants were evaluated for the presence of selected angiogenesis-related factors with a Luminex-based platform. Quantitative analysis of concentrations of endoglin, endothelin, VEGF-α, HGF, SDF-1, MCP-1 and IL-1β was performed using the Milliplex MAP Kit (Mouse Angiogenesis/Growth Factor Magnetic Bead Panel; EMD Millipore) according to the manufacturer's protocol. Protein concentrations were read on a Luminex platform (Austin). Median Fluorescent Intensity (MFI) for each protein was analyzed using 5-parameter logistic or spline curve-fitting method for calculating concentrations in tested samples by in Luminex IS 2.3 Software (Austin).

Statistical analysis
All experiments were carried out in triplicates (except global proteome analysis which was repeated twice). In every repetition of the experiment was used primary culture of MSCs at passage 3-4. Results are presented as mean values ± standard deviations (SD). Paired Student's t tests were used to compare MCPIP1-overexpressing MSCs and Puro-treated cells. Untreated Control cells were used in all experiments as a negative control. A P value <0.05 was considered as statistically significant. All statistical analyses were performed using the Origin (ver. 9.1) statistical software (Microcal Software).

MCPIP1 is efficiently overexpressed in primary MSCs following retroviral transduction
To evaluate the impact of MCPIP1 on functions of BM-derived MSCs, cells were transduced with retroviral vectors carrying the sequence coding MCPIP1 (MCPIP1-MSCs). Empty vectortreated (Puro) or untreated (Control) MSCs were used as controls. In order to confirm expression of MCPIP1 at the mRNA level, total RNA was isolated from MSCs at 48 and 72 hours after transduction. The relative level of expression of mRNA for MCPIP1 was computed in comparison with Puro-treated cells (set as 1) (S1C Fig To investigate the potential impact of MCPIP1 on MSC viability and metabolic activity, several functional tests were done at the indicated time points post transduction. Thus, MTT assay was used to measure cytotoxic effects of MCPIP1 at 48 and 72h following viral transduction. Relative viability was computed when compared with Puro-treated cells (shown as 1). We did not observe any notable impairment of cell viability after transduction with retroviral vectors at each time point (S1E Fig). The effect of MCPIP1 on MSC metabolic activity was evaluated by the ATP measuring assay. Relative metabolic activity was also computed in comparison to Puro-treated cells (set as 1). We did not find a significant impact of MCPIP1 protein on metabolic activity of MSCs at any time-point (S1F Fig).
The transduction efficiency was evaluated following MSC treatment with a retroviral vector carrying GFP for 72h by applying both fluorescent microscopy and flow cytometry (S1A and S1B Fig). A total of 60.0 ± 16.5% of transduced adherent fraction of cultured BM cells expressed GFP, corresponding to transduction efficiency within the entire fraction of cells (S1B Fig). Thus, because of the highest expression of MCPIP1 observed at 72h post transduction that was not accompanied by significant impairment in viability or metabolic activity of MCPIP1-overexpressing MSCs, the cells at this time-point after transduction were subsequently used for further experiments examining the impact of MCPIP1 on several MSC functions including SC-related gene expression as well as angiogenic and cardiomyogenic differentiation potential in vitro.

MCPIP1 does not affect antigenic phenotype and viability but decreases MSC proliferation
To evaluate the impact of MCPIP1 on MSC viability and apoptosis activation, several assays were done at 72h post transduction. Apoptosis and necrosis were examined by phosphatydyloserine detection and 7-aminoactinomycin D staining as well as by evaluation of caspase-3 and -7 activation. No major impact of MCPIP1 on MSC viability was noted. A small fraction of cells (less than 5%) undergoing cell death was detected within in all groups of MSCs ( Fig 1A).
We also evaluated the impact of MCPIP1 on proliferation of MSCs at 72h after transduction. Interestingly, we found that MCPIP1 overexpression was associated with a significant decrease in MSC proliferation rate when compared with both control groups ( Fig 1B). However, the proliferative activity of MSCs overexpressing MCPIP1 was still sustained at a relatively high level (80.19% ± 7.34 when compared with Puro-treated cells). Importantly, we did not observe any major differences in morphology between MCPIP1-overexpressing MSCs and control cells (Fig 1B).
To investigate the potential influence of MCPIP1 overexpression on expression of classical MSC markers and the content of major antigenically-defined MSC subpopulations, we evaluated the multiantigenic profile of transduced cells by immunostaining against CD45, CD90.2, CD105 and Sca-1 antigens followed by flow cytometric analysis ( Fig 1C). We found that similar to control cells, MCPIP1-overexpressing MSCs consist of three main subpopulations including CD90 + /CD105 -, CD105 + /Sca-1 + and CD90 + /Sca-1 + that represent 82.7±5.2%; 88.1±4.4% and 92.4±2.0% of CD45-negative MSC population, respectively ( Fig 1C). Furthermore, no major effect on expression of MSC-related surface antigens was observed in cells with MCPIP1 overexpression. Thus, the content of main subsets of MSCs was comparable between MCPIP1-overexpressing cells and both control groups ( Fig 1C).

MCPIP1 decreases expression of pluripotency-associated genes and enhances differentiation capacity of MSCs
The expression of pluripotency markers such as Oct-3/4A, Klf-4, c-Myc and Sox2 was evaluated at the mRNA level by real-time RT-PCR (Fig 2A) to assess whether MCPIP1 may be involved in regulation of stem cell-related genes. The results showed that MCPIP1 overexpression was associated with a decreased level of Oct-4 and Klf-4 (transcription factors maintaining stemness) in comparison with Puro-treated and untreated Control MSCs (Fig 2A).
Importantly, the Matrigel assay showed that the decrease in SC-related gene expression was accompanied by enhanced angiogenic differentiation capacity of MSCs expressing MCPIP1 when compared with control groups of MSCs (Fig 2C and 2D). In quantitative analysis (S1B Fig), we observed faster and greater formation of capillary-like structures by MSCs expressing MCPIP1 when compared with control cells (Fig 2C and 2D, S1 Table). The results suggest a role for MCPIP1 in enforcing the differentiation capacity of MSCs.
Interestingly, the proteomic data may also suggest that MCPIP1 expression favors decreased proliferative status of MSCs along with promoting their differentiation ability. Mass spectroscopy was used to evaluate the global proteome of all examined groups of MSCs. In each analyzed sample were detected more than 2600 proteins (Fig 2B). We found several proteins uniquely expressed or with elevated concentration in MCPIP1-expressing MSCs when compared with Puro-treated cells, including proteins involved in cell proliferation, differentiation, Wnt/β-catenin signaling as well as regulating processes of apoptosis and autophagy (Table 1). Interestingly, several proteins negatively regulating cell proliferation were present in this group including testin and protein phosphatase 1G implicated in negative regulation of cell proliferation and cell cycle arrest, respectively (Table 1). Moreover, we identified several proteins not present in MCPIP1-overexpressing MSCs when compared with Puro cells, including proteins negatively regulating Notch signaling as well as involved in apoptosis and intracellular reorganization (Table 2). Thus, the proteomic profile of MCPIP1-overexpressing MSCs corresponds to their lower proliferative activity and enhanced differentiation capacity.

MCPIP1 enhances the angiogenic potential of MSCs following stimulation accompanied by an increase in expression of proangiogenic factors
After establishing that MCPIP1-MSCs at 72h after transduction exhibited greater proangiogenic activity, we launched a long-term culture of these cells in proangiogenic medium (EGM-2MV) and assessed the angiogenic differentiation after 5d and 10d of culture (Fig 3). The expression of proangiogenic genes was investigated at mRNA level by real time RT-PCR and we found elevated expression of Gata-2, vWF and VE-cadherin after 5 days of angiogenic differentiation peaking at 10 days of culture in MCPIP1-MSCs when compared with Puro-treated cells (Fig 3A). Enhanced angiogenic capacity of MCPIP1-MSCs was confirmed in a direct differentiation assay followed by immunocytochemical staining for angiogenic proteins (Fig 3B  and 3C). Quantitative analyses revealed the greatest number of cells with endothelial phenotype expressing intranuclear transcription factor Gata-2 and membrane VE-cadherin within MCPIP1-MSCs when compared with Puro cells (Fig 3C). Moreover, MCPIP1-overexpressing MSCs exhibited greater angiogenic activity in the functional Matrigel assay following pre-differentiation culture in proangiogenic medium (Fig 4).  Table).
To determine whether MCPIP1-overexpressing MSCs are able to secrete any protein involved in angiogenesis that may stimulate neighboring cells in a paracrine manner, we compared the secretome of MSCs overexpressing MCPIP1 and control MSCs by using Western immunoblotting as well as quantitative analysis with the Luminex platform (Fig 5 and S3 Fig). Based on our previous experimental data indicating that MCPIP1 up-regulates angiogenesisrelated genes and promotes capillary-like tube formation especially after 10 days following endothelial stimulation, we used conditioned cell culture medium derived from MCPIP1-MSCs and control MSCs at this time point. We found that secretion of endothelin, a tissue inhibitor of metalloproteinase-1 (TIMP-1), Serpin E1, IFN-γ inducible-protein-10 (IP-10), matrix metalloproteinase-3 (MMP-3), stromal cell-derived factor 1 (SDF-1), osteopontin and insulin-like growth factor-binding protein 9 were significantly upregulated in MCPIP1-overexpressing MSCs when compared with Puro cells (Fig 5). Next, we employed the quantitative Luminex-based assay where we tested the secretion of selected proangiogenic proteins by MCPIP1-overexpressing MSCs and controls at day 0, 5 and 10 following endothelial differentiation (S3 Fig). We detected elevated concentration of endoglin in MCPIP1-overexpressing MSCs when compared with Puro-treated cells, especially after 10 days of differentiation (S3   Similar to previous results, the concentration of endothelin increased slightly after 5 and 10 days of pre-differentiation. Interestingly, in case of VEGF, we observed a decreased level of this factor along the differentiation culture in all groups. In contrast, the concentration of SDF-1 increased after 10 days of stimulation (Fig 5 and S3 Fig). Importantly, with both assays, we did not detect any measurable level of IL-1β or other inflammatory cytokines or chemokines produced by MCPIP1-overexpressing MSCs (data not shown). These data confirm the anti-inflammatory properties of MCPIP1 also in MSCs that have already been reported for other cell types [21,23]. Finally, we detected decreased concentration of MCP-1 released by MCPIP1-overexpressing MSCs when compared with Purotreated and untreated MSCs that remained consistent during differentiation culture (S3 Fig). This may suggest a negative feedback loop of MCP-1 regulation by MCPIP1 in MSCs.

MCPIP1 promotes cardiomyogenic differentiation of MSCs
To establish the impact of MCPIP1 on cardiac differentiation, MCPIP1-MSCs and control cells were differentiated into cardiomyocytes in vitro as previously described [32]. We analyzed mRNA expression for cardiac markers such as Gata-4, Nkx2.5, Myl-2 and Myh-6 after 5 and 10 days of differentiation induction (Fig 6A). We found that the expression of these genes was markedly elevated after 5 and 10 days of cardiomyogenic differentiation induction in MCPI-P1-overexpressing MSCs when compared with Puro-treated cells (Fig 6A).
Enhanced cardiomyogenic capacity of MSCs overexpressing MCPIP1 was confirmed by immunocytochemical staining for typical cardiac proteins following differentiation in cardiac medium in vitro (Fig 6B and 6C). Quantitative assessment of differentiating MSCs (after 5 and 10 days of culture) revealed greater numbers of cells with cardiac phenotype expressing intranuclear cardiac transcription factor Gata-4 and cytoplasmic structural protein troponin T-C  within MCPIP1-MSCs when compared with Puro cells (Fig 6C). Interestingly, we found increase in expression of autophagy related genes in MCPIP1-MSCs undergoing cardiac or angiogenic differentiation (Fig 7A and 7B).
Our findings indicate that MCPIP1 protein enhances not only angiogenic, but also cardiomyogenic differentiation of MSCs. Interestingly, our gene expression and proteomic studies suggest that such processes may be accompanied with intracellular reorganization via autophagy, which may represent new interesting subject for future studies and needs to be further investigated (Fig 7C).

Discussion
According to annual reports of the World Health Organization, ischemic heart disease (IHD) is one of the leading causes of death worldwide [33]. IHD may lead to acute ischemic events such as myocardial infarction (MI) resulting in massive cardiomyocyte death due to apoptosis and necrosis [34]. The profound inflammatory reaction that follows ischemia contributes progression of ventricular remodeling and eventually dysfunction, heart failure and higher mortality [35]. Therapeutic approaches have therefore been aimed at inhibiting the adverse consequences of myocyte loss and tissue remodeling. These have included both pharmacological methods and cell-based regenerative strategies with stem cells [36][37][38].
Bone marrow MSCs, a subset of adherent cells with multipotent differentiation capacity, are able to differentiate into several cell types, including cardiomyocytes and endothelial cells [39,40]. Due to this broad differentiation potential, low immunogenicity, and anti-inflammatory activities, MSCs have been widely employed for myocardial salvage, repair, and regeneration following ischemic injury [41][42][43][44].
In this study, BM-derived MSCs were genetically modified to overexpress MCPIP1, a protein exhibiting a strong silencing potential of the inflammatory response and modulation of gene expression by RNase activity [21,23,24]. MCPIP1 has also recently been implicated in adipogenesis [28] and angiogenesis [25]. Although, a role of MCPIP1 in cardiomyocyte survival was reported [26,45], the role of this protein in cardiomyogenesis from stem cells has never been examined.
In view of the role of MCPIP1 in cell survival and initiation of apoptosis [25,46,47], we evaluated the activation of proapoptotic caspase-3 and -7 as well as the extent of cells undergoing apoptosis and necrosis. By employing several independent assays, we showed that MCPIP1 overexpression did not negatively impact the viability of MSCs. Moreover, we observed normal metabolic activity based on measurement of total ATP concentration in such cells. These results indicate greater resistance of stem cells such as BM-derived MSCs against activation of apoptosis via MCPIP1 when compared with other more mature cells such as primary HUVECs and cardiomyocytes previously examined by other groups [25,46,47].
Interestingly, other studies have reported a beneficial effect of cardiac MCPIP1 expression on myocardial injury. Using a transgenic mouse model with myocardial MCPIP1 expression (under α-MHC promoter), Niu et al. reported significant attenuation of inflammation-related cardiac dysfunction [26,45]. The authors observed decreased expression of inflammatory cytokines and iNOS as well as caspase 3/7 activity and apoptosis in MCPIP1-expressing myocytes following treatment with LPS when compared with wild type animals [26,45]. These results were confirmed by other investigators reporting lower levels of inflammatory cytokines (e.g. TNF-α, IL-1β) along with higher levels of anti-inflammatory cytokines (e.g. IL-10) in myocardium of animals with MCPIP1 expression following LPS treatment [48][49][50]. This phenomenon may be related to RNase activity of MCPIP1 due to the presence of the PilT N-terminus domain which has been shown to promote degradation of mRNA for inflammatory cytokines such as IL-6 and IL-1 as well as select pre-miRNAs that regulate other factors involved in inflammation [21,23,24]. Moreover, Morimoto et al. have demonstrated a cardioprotective effect of MCP-1 (acting via stimulation of MCPIP1 expression) on myocardium following ischemia/ reperfusion (I/R) injury in mice [51,52]. Thus, the findings suggest that MSCs overexpressing MCPIP1 may have an anti-apoptotic and anti-inflammatory role on myocardium following transplantation into ischemic heart tissue.
Moreover, MCPIP1 may also be involved in deubiquitination of TRAF family proteins which are known to mediate inflammatory response [53]. Interestingly, our proteomic analysis of MSCs from all three experimental groups revealed expression of ubiquitin-conjugating enzyme E2 H involved in protein ubiquitination and ubiquitin fusion degradation protein 1 responsible for proteasome-mediated protein catabolic process in MCPIP1-overexpressing MSCs, but not in control MSCs. Thus, these data suggest that MCPIP1 expressed in MSCs may indirectly enhance ubiquitination of selected proteins resulting in their degradation. This phenomenon may also potentially be involved in inhibition of inflammatory cytokine production by MCPIP1-MSCs; however, this needs to be further evaluated. Overall, these data again indicate that MCPIP-1 may exert anti-inflammatory and anti-apoptotic properties in MSCs.
Interestingly, we found that although MCPIP1 did not notably affect morphology or antigenic phenotype of MSCs, it significantly impaired proliferation of MSCs expressing MCPIP1 when compared with Puro vector-treated cells. Consistently, global proteomic analysis revealed that MCPIP1-MSCs expressed greater levels of several proteins involved in negative regulation of cell proliferation and cell cycle arrest, including testin and protein phosphatase 1G. It is generally known that a decrease in cell proliferation may often be accompanied by induction of cell differentiation [54,55]. Thus, we expected that the effect of MCPIP1 on MSC proliferation may be related to greater differentiation status of these cells confirmed by further experiments.
It has been reported that MCPIP1 may enhance the angiogenic potential of HUVEC cells, which represent mature cells, and BM-derived monocytic cells [25,56]. Using Matrigel assay, Niu and colleagues have shown that HUVECs stimulated with MCP-1 show greater angiogenic potential. Importantly, knockdown of MCPIP1 in HUVECs resulted in suppression of VEGF and hypoxia-inducible factor-1α (HIF-1α) genes, the angiogenesis-related factors induced by MCP-1 [25]. Similarly, angiogenic differentiation of human BM-derived monocytic cells (BMNCs) with MCPIP1 expression and enhanced neovascularization after transplantation into ischemic tissues in vivo has also been reported [56]. However, the impact of MCPIP1 on the angiogenic capacity of stem cells such as MSCs has never been studied.
Our results show, for the first time, that MCPIP1 may enhance angiogenic differentiation of adult stem cells such as MSCs in vitro. Our results also reveal significant upregulation of endothelial transcription factors and structural proteins, including Gata-2 and VE-cadherin at both mRNA and protein levels in MSCs expressing MCPIP1 following culture in proangiogenic medium. Moreover, these data also establish that MCPIP1 enhances formation of capillary-like structures in Matrigel assay after proangiogenic stimulation. Interestingly, the greater angiogenic capacity of MCPIP1-MSCs was accompanied by decreased expression of early stem cellrelated genes such as Oct-4, Nanog and Sox2. Our data indicate enhanced angiogenic capacity of MSCs expressing MCPIP1 when compared with control cells confirming proangiogenic activity of MCPIP1 not only in mature cells but also in stem cells.
In the present study we also report, for the first time, that MCPIP1 expression enhances differentiation of BM-derived MSCs into cardiomyocytes in vitro. We found that MCPIP1 promoted the acquisition of a cardiac phenotype by MSCs, evidenced by upregulation of cardiomyogenesis-related transcription factors, such as Gata-4 and Nkx2.5, as well as structural proteins, including Myh-6 and Myl-2. Furthermore, we detected a greater number of cells expressing Gata-4 and troponin T-C at the protein level in MCPIP1-MSCs following 10 days of culture in cardiomyogenic medium. Our findings therefore support a role of MCPIP1 in generation of cardiomyocytes stem cells. These data also provide the rationale to develop novel strategies to transplant MSCs overexpressing MCPIP-1 into ischemic myocardium to enhance cardiac and vascular regeneration.
Current evidence from in vitro and in vivo studies strongly indicate that major mechanisms of stem cell-mediated neovascularization include not only effective stem cell retention and vascular differentiation in the infarcted heart, but also secretion of several factors impacting endogenous cells in a paracrine manner [57,58]. Interestingly, we found that MCPIP1-overexpressing MSCs secrete higher levels of several proteins involved in angiogenesis including endoglin, endothelin, TIMP-1, serpin E1, IP-10 and MMP-3, when compared with controls. Furthermore, we observed a higher level of chemokine SDF-1 secreted by MSCs expressing MCPIP1 when compared with control cells. SDF-1 has been shown to play a pivotal role in directing CXCR4 + stem cells toward the site of injury, an essential step for cardiac repair [59][60][61]. Our data suggest that production of SDF-1 by transplanted MCPIP1-MSCs may enhance homing of other endogenous stem cells to the infarcted heart and promote regeneration. Aside from serving as a homing factor, SDF-1 has also been shown to be a proangiogenic factor [62,63], and as such may increase new vessel formation when released by transplanted MCPIP1-MSCs.
Interestingly, it has been recently demonstrated that forced expression of MCPIP1 may induce autophagy in HUVECs finally leading to improvement of angiogenic potential [64]. The phenomenon may be explained by the necessary rebuilding of cell contents during differentiation via elimination of several proteins and organelles associated with metabolic changes [65][66][67]. Our global proteomic analysis showed increased expression of autophagy regulators, such as CDGSH iron-sulfur domain-containing protein 2 exclusively in MSCs expressing MCPIP1. Importantly, the expression of genes related to autophagy such as beclin 2 and Atg7 was greatly increased in MCPIP1-MSCs during their angiogenic and cardiac differentiation. Moreover, we found several proteins potentially involved in the intracellular cytoskeleton and membrane reorganization to be expressed at a higher level in MCPIP1-MSCs, including Rasrelated protein Rab-11B, tubulin alpha chain-like 3 or SLAIN motif-containing protein involved in microtubule organization. Elevated expression of proteins involved in intracellular membrane trafficking and recruiting effector molecules critical for vesicle traffic along actinor microtubule-based cytoskeletal structures, such as secretory carrier-associated proteins or Rab proteins was also reported [68]. The data may indicate that MSCs expressing MCPIP1 contain a set of molecules participating in intracellular component remodeling and vesicular transport facilitating differentiation processes in these SCs. Thus, the observed greater angiogenic and cardiac differentiation potential of MCPIP1-MSCs may be preceded by autophagy related to MCPIP1 expression and intracellular reorganization. However, this interesting phenomenon warrants further investigation.
In summary, overexpression of MCPIP1 reduces the expression of pluripotency associated markers, and increases angiogenic and cardiomyogenic potential of MSCs. MCPIP1 does not notably impact MSC viability, metabolic activity, morphology, and antigenic profile, yet reduced the proliferation rate. MCPIP1-overexpressing MSCs express several proteins potentially involved in angiogenesis, autophagy and processes accompanying cell differentiation ( Fig  7C). These results strongly indicate the beneficial impact of MCPIP1 protein on MSC features which may be utilized for potential clinical applications.  Table. Quantitative analysis of number of branches and capillaries formed by MSCs in capillary-like formation assay-number of branches and capillaries calculated per field formed by non-differentiated MSC groups (at 72h following transduction). (DOC) S2 Table. Quantitative analysis of number of branches and capillaries formed by MSCs in capillary-like formation assay-number of branches and capillaries formed by MSC groups after 5 and 10 days of endothelial stimulation. (DOC)