JOURNAL OF TISSUE ENGINEERING AND REGENERATIVE MEDICINE RESEARCH J Tissue Eng Regen Med (2013) Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/term.1795
ARTICLE
Podocalyxin-like protein 1 is a relevant marker for human c-kitpos cardiac stem cells Isabel Moscoso1, Naiara Tejados2, Olga Barreiro3, Pilar Sepúlveda4, Alberto Izarra1, Enrique Calvo1, Akaitz Dorronsoro2, Juan Manuel Salcedo2, Rafael Sádaba5, Antonio Díez-Juan6, César Trigueros2 and Antonio Bernad1* 1
Cardiovascular Development and Repair, Fundación Centro Nacional de Investigaciones Cardiovasculares Carlos III, Madrid, Spain Inbiomed Foundation, San Sebastian, Spain 3 Vascular Biology and Inflammation Departments, Fundación Centro Nacional de Investigaciones Cardiovasculares Carlos III, Madrid, Spain 4 Regenerative Medicine and Heart Transplantation Unit, Instituto de Investigación Sanitaria La Fe, Valencia, Spain 5 Department of Cardiac Surgery, Hospital de Navarra, Pamplona, Spain 6 Centro de Investigación Príncipe Felipe, Valencia, Spain 2
Abstract Cardiac progenitor cells (CPCs) from adult myocardium offer an alternative cell therapy approach for ischaemic heart disease. Improved clinical performance of CPCs in clinical trials requires a comprehensive definition of their biology and specific interactions with the environment. In this work we characterize specific human CPC surface markers and study some of their related functions. c-kitpos human CPCs (hCPCs) were characterized for cell surface marker expression, pluripotency, early and late cardiac differentiation markers and therapeutic activity in a rat model of acute myocardial infarction. The results indicate that hCPCs are a mesenchymal stem cell (MSC)-like population, with a similar immunoregulatory capacity. A partial hCPC membrane proteome was analysed by liquid chromatography–mass spectrometry/mass spectrometry and 36 proteins were identified. Several, including CD26, myoferlin and podocalyxin-like protein 1 (PODXL), have been previously described in other stem-cell systems. Suppression and overexpression analysis demonstrated that PODXL regulates hCPC activation, migration and differentiation; it also modulates their local immunoregulatory capacity. Therefore, hCPCs are a resident cardiac population that shares many features with hMSCs, including their capacity for local immunoregulation. Expression of PODXL appears to favour the immature state of hCPCs, while its downregulation facilitates their differentiation. Copyright © 2013 John Wiley & Sons, Ltd. Received 28 November 2012; Revised 18 January 2013; Accepted 12 June 2013
Supporting information may be found in the online version of this article. Keywords Human cardiac progenitor cells; membrane proteome; PODXL; differentiation; migration; immunoregulation
1. Introduction The heart was long considered a terminally differentiated postmitotic organ, in which the number of cardiomyocytes was fixed at birth (Anversa et al., 2007; Leri et al., 2011). This
*Correspondence to: A. Bernad, Cardiovascular Development and Repair Department, Fundación Centro Nacional de Investigaciones Cardiovasculares Carlos III, 28029 Madrid, Spain. E-mail:
[email protected] Copyright © 2013 John Wiley & Sons, Ltd.
view has since been revised and retrospective [14C] birth dating of cardiac cells suggests an annual turnover of 0.45–1% of myocytes in the adult human heart (Bergmann et al., 2009). Several groups have demonstrated that the adult mammalian myocardium contains resident populations of cardiac progenitor cells (CPCs, also referred to as cardiac stem cells) with the potential capacity to differentiate into cardiomyocytes and other cell types (Segers and Lee, 2008). CPCs have been identified and isolated by a variety of approaches, including expression of surface markers (c-kit or sca1) and physiological properties (side population or
I. Moscoso et al.
cardiosphere-formation) (Laflamme and Murry, 2011). The fundamental properties of progenitor cells are self-renewal, clonogenicity and multipotency in vitro and in vivo. CPCs have a high proliferation and differentiation potential in vitro (Segers and Lee, 2008). The most extensively studied CPC population is defined by expression of the marker c-kit. These cells are self-renewing, clonogenic and multipotent, giving rise to myocytes, smooth muscle and endothelial cells (Anversa et al., 2007). Their identification radically changed our understanding of myocardial biology (Beltrami et al., 2003). CPCs can also be identified by their capacity to form cardiospheres. Cardiosphere-forming cells are isolated on the basis of their ability to migrate from cardiac tissue explants and form spheroids in suspension culture (Messina et al., 2004). These cardiospheres are composed of a mixture of cells (cardiosphere-derived cells, CDCs), including c-kitpos cells. CDCs can give rise to cardiomyocytes in vitro and in vivo after transplantation, and transplanted cells enhance cardiac function after infarction (Lee et al., 2011). There is therefore a need for reliable models that permit identification and tracking of CPC lineage and phenotype without resorting to transplantation or cell culture (Martin-Puig et al., 2008; Laflamme and Murry, 2011). During the last decade, cell therapy for ischemic heart failure has passed feasibility and safety tests in phase I and II clinical trials, and results to date indicate moderate benefits (Gyongyosi et al., 2009; Assmus et al., 2010). Preliminary results of trials evaluating CPC preparations have recently been published. SCIPIO (stages A and B) has demonstrated the safety and efficacy of intracoronary infusion with autologous c-kitpos CPCs for improving LV systolic function and reduced infarct size in post-MI patients 1 year after injection (Bolli et al., 2011). These promising results warrant further and larger studies. Proteomics approaches have recently been used to identify cell-surface markers of stem-cell function and fate (Cao et al., 2012; Sarkar et al., 2012). Membrane proteins can be studied by cell-surface biotinylation followed by liquid chromatography–mass spectrometry/mass spectrometry (LC-MS/MS) (Lee et al., 2009b). This approach recently identified the transferrin receptor as a candidate marker of neural stem cells (NSCs) (Cao et al., 2012) and defined the membrane proteome of undifferentiated human embryonic stem cells (hESCs) (Sarkar et al., 2012). Here, the cell-surface biotinylation approach to profile membrane proteins in hCPCs was used. A set of 36 proteins expressed in hCPCs was identified, including several previously associated with adult stem cells. Our results with hCPCs are compatible with findings in hESCs and suggest a role for podocalyxin-like 1 (PODXL) in the regulation of hCPC responses to homeostatic and injury-derived signals.
2. Materials and methods Expanded methods are available in the Supporting Information, Methods S1, and provide details of hCPC and hMSC isolation, culture and manipulation. The local ethics Copyright © 2013 John Wiley & Sons, Ltd.
committee approved animal studies. All animal procedures conformed to EU Directive 86/609/EEC and Recommendation 2007/526/EC regarding the protection of animals used for experimental and other scientific purposes, enforced in Spanish law under Real Decreto 1201/2005.
2.1. hCPC in vitro studies c-kitpos CPCs were obtained from human cardiac biopsies essentially as previously described (Messina et al., 2004), with the modifications indicated. Human mesenchymal stem cells (hMSCs) were obtained and cultured as described in the Supporting Information, Methods S1. Immunosuppressive capacity was determined by measuring the proliferation of human CD3+ lymphocytes in co-culture with hCPCs or hMSCs (0.2–1 × 10 6 lymphocytes: 1 × 105 progenitors) using the CellTrace™ carboxyfluorescein diacetate succinimidyl ester (CFSE) Cell Proliferation Kit (Invitrogen, Carlsbad, CA, USA). For cardiomyocyte-induced hCPC differentiation, neonatal rat cardiomyocytes (NRCM) were resuspended in cardiomyocyte medium containing 1 μg/ml cytosine β-D-arabinofuranoside (Sigma, St Louis, MO, USA). After 6 days, hCPCs (GFP or Cherry positive) were seeded onto the NRCMs at 9 × 104/cm2. After 11 days hCPCs were sorted and analysed. Decellularized rat heart matrix was obtained as described in Methods S1. Pluripotency gene expression was analysed with the Proteome Profiler™ Human Pluripotent Stem Cell Array Kit (ARY010; R&D Systems, Minneapolis, MN, USA).
2.2. Proteomic analysis hCPC membrane proteins in intact cells were labelled for 1 h with Sulfo-NHS-LC-Biotin (Invitrogen), and washed cells were lysed and labelled proteins extracted with streptavidin-conjugated magnetic beads (Invitrogen). Proteins were deglycosylated with PNGase F (Sigma, St Louis, MO, USA), and washed with phosphate-buffered saline (PBS) and guanidine chloride. After trypsin digestion, peptides were analysed by LC-MS/MS. Thirty-six membrane proteins were identified (from four replicates), and 21 identifications were validated by quantitative realtime polymerase chain reaction (qRT-PCR).
2.3. PODXL function in hCPCs PODXL in hCPCs was suppressed or overexpressed by lentiviral transduction (see Methods S1). The adhesion strength of hCPCs to an endothelial monolayer was studied under defined flow conditions in a parallel flow chamber, described in detail on the Glycotech website (http://www. glycotech.com). Migration on a chemotactic gradient was studied in transwell chambers (Millipore, Billerica, MA, USA). Scratch wound-healing assays were performed as detailed in Methods S1. Cells were counted and wound area measured with IMAGEJ 10.2 (National Institutes of Health, Bethesda, MD, USA). J Tissue Eng Regen Med (2013) DOI: 10.1002/term
PODXL is a marker of cardiac stem cells
2.4. Statistical analysis Results are presented as means ± standard deviation (SD). The significance of differences between groups (p ≤ 0.01 or p ≤ 0.05) was determined by Student's t-test, Mann–Whitney U test or Two-way ANOVA (Bonferroni post-test) using PRISM 5.0c (GraphPad Software, La Jolla, CA, USA) or SPSS 11.0 (Chicago, IL, USA).
3. Results and discussion 3.1. Human cardiac progenitor cells have an MSC-like surface expression profile and show immunoregulatory potential c-kitpos hCPCs were obtained from human cardiac biopsies (Messina et al., 2004) and accounted for 0.5–5% of cells. After expansion for 2 weeks, preparations were characterized by flow cytometry and qRT-PCR. The hCPCs were positive on flow cytometry for CD117, CD90, CD73, CD105, CD29 and CD13, and negative for CD45, CD34 and CD31 (Figure 1A). A qRT-PCR analysis demonstrated expression of CD166, CD29, CD304, CD49, CD73, LRRC59, SDF1, Klf4 and Bmi1, while expression of CD44, CD98, CX3CR1, SLC9A3R2, ITGA6, SELE and SSEA1 was absent or weak (Figure 1B). In addition, protein array comparison of hCPCs with a human embryonic stem cell line (hES-3) (Figure 1C) demonstrated weak hCPC expression of classical embryonic pluripotency factors (Oct4, Nanog, Sox2 and ECAD); however, hCPCs were positive for expression of GATA4, OTX2, SNAI1, FOXA2, PDX1 and VEGRF2, and expressed higher levels of Sox17. Unlike hESCs, hCPCs were negative for SSEA3 and SSEA4 (Figure 1D). The immunophenotype of hCPCs was similar to that of hMSCs (Figure 1E) and, therefore, their capacity to locally control immune and inflammatory reactions was evaluated. T lymphocytes (CD3), activated with 10 ng/ml interleukin (IL)-2 and anti-CD3/CD28 magnetic beads for 48 h and labelled with CFSE, were co-cultured with hCPCs (CPC:T lymphocyte ratio 1:2–10) for 6 days. hCPC co-culture reduced T lymphocyte proliferation by 50% (Figure 1F), compared with a 75% reduction in hMSC co-cultures. hCPCs thus have the capacity to control T lymphocyte proliferation, suggesting that they have a similar immunoregulatory capacity to hMSCs from bone marrow or adipose tissue. We next investigated the ability of hCPCs to modulate other cell types in the injured heart. Analysis of putative soluble factors involved in cardiac regeneration (Lee et al., 2011) before transplantation revealed a complex picture (Figure S1A); however, examination of infarcted rat hearts 4 weeks after injection with hCPCs showed upregulated expression of vascular endothelial growth factor (VEGF) and IL-6 (Figure S1B), as well as moderate upregulation of insulin-like growth factor (IGF)-1 and unaltered expression of hepatocyte growth factor (HGF) (not shown). These changes were accompanied by a marked increase in vascular density (Figure S1C) but no significant reduction of infarct size (Figure S1D). These results are consistent with Copyright © 2013 John Wiley & Sons, Ltd.
published findings (Mazo et al., 2012) indicating that the main action of hCPCs upon transplantation is to increase local vascular density through a paracrine action.
3.2. Identification of stem cell-associated membrane proteins in hCPCs Analysis of membrane subproteomes to identify unique markers and possible therapeutic targets is a powerful approach for improving definition and subfractionation of stem-cell populations from more committed cells. This strategy has been used for NSCs, ESCs, endothelial progenitors, myoblasts and MSCs (Carvalho et al., 2011; Cao et al., 2012; Sarkar et al., 2012). Similar proteomic analysis of the secretome has also been used to identify putative autocrine and paracrine factors for the ex vivo expansion of adult stem cells (Stastna et al., 2010). To define the hCPC immunophenotype we labelled hCPCs (four experimental replicates) with sulfo-NHS-SS-biotin and conducted an LC-MS/MS proteomics analysis of plasma membrane protein fractions (Figure S2A,B). This analysis identified 36 membrane proteins (Figure S2C, Table S1), corresponding to 26.33% of the total proteins identified. Based on a literature search, the first 10 proteins scored, which related to stem cells and cardiovascular field, were selected and studied (Table 1). The relative abundance of each protein was calculated in relation with CD73 as the reference protein, because it is a MSC marker and the most abundant protein in these cells (Figure S2B), and the results (Figure S2C) indicate that CD73, transforming growth factor (TGF)β, Ig-h3, CD49 and CD29 are the most abundant proteins. Gene expression for 13 proteins was compared by qRT-PCR between three hCPC samples (BC19, H1 and H4), hMSCs and foreskin fibroblasts (Figure 2A). Furthermore, variability among isolates was evaluated using two additional hCPC samples (H1 and H4). Several proteins were more abundant in adult stem cells than in fibroblasts, including ASPH (C-terminal aspartyl/asparaginyl betahydroxylase) but only two seemed to be expressed at higher levels by hCPCs (BC19): the serine exopeptidase CD26 and podocalyxin-like I (PODXL), although CD26 seems also to be highly expressed by fibroblasts. In other membrane– subproteome studies, myoferlin has also been detected in placenta (Posey et al., 2011), seprase and CD73 in MSCs (Carvalho et al., 2011) and CD29, CD44, CD49, CD98 and CD166 in various cancer types (Yang et al., 2011). Further characterization of CD26 and PODXL was conducted in parallel with analysis of myoferlin (MYOF), which has a demonstrated role in muscle homeostasis (Posey et al., 2011). Expression of CD26, myoferlin and PODXL in hCPCs was reconfirmed by flow cytometry (Figure 2B), which demonstrated expression of these markers and the membrane protein CD49f in > 80% of cells. Lower percentages of cells were positive for CD49d, CXCR4, Tra-1-60 and Tra-1-81. Western blot analysis confirmed higher expression of CD26, myoferlin and PODXL in hCPCs than in hMSCs and fibroblasts (Figure 2C). Expression of these three membrane proteins was absent or significantly lower in J Tissue Eng Regen Med (2013) DOI: 10.1002/term
I. Moscoso et al.
Figure 1. Human cardiac progenitor cells (hCPCs) growth and characterization. A, Flow cytometry analysis of selected membrane markers in hCPC sample BC19. B, Gene expression profiling by quantitative real-time polymerase chain reaction. C, hCPC pluripotency protein expression array; human embryonic stem cells (ESCs) cells were examined as a positive control. D, Flow cytometry analysis of SSEA3 and SSEA4 expression in hCPCs and hESCs. E, Co-stimulatory molecules expressed on hCPCs and hMSCs (flow cytometry). F, hCPC immunosuppressive capacity in vitro. carboxyfluorescein diacetate succinimidyl ester (CFSE)-loaded lymphocytes were stimulated with mitogenic factors alone (control) or in the presence of hCPCs or MSCs (n = 3; mean ± SD). Histograms show representative traces for CFSE-positive; E.I., expansion index (mean number of divisions)
mature cardiomyocytes (Figure 2C). CD26/DPP4 (dipeptidyl-peptidase 4) is an intrinsic membrane glycoprotein and serine exopeptidase that cleaves X-proline Copyright © 2013 John Wiley & Sons, Ltd.
dipeptides from polypeptide N-termini. It will be interesting to investigate whether the expression of CD26/DPP4 on CPCs is involved in cardiac homeostasis, aging and response J Tissue Eng Regen Med (2013) DOI: 10.1002/term
PODXL is a marker of cardiac stem cells Table 1. Identity and main function of selected proteins identified by liquid chromatography–mass spectrometry/mass spectrometry (LC-MS/MS) in human cardiac progenitor cells (hCPCs) (n = 4) Proteins identified
Main functions
References*
Arginyl-glycyl-aspartic acid (RGD)-containing protein Chan et al. (2003); Irigoyen et al. (2008); Binding to collagen types I, II and IV Li et al. (2008) Modulation of cell adhesion (negative regulation) Ligand recognition sequence for several integrins Role in cell–collagen interactions Mutations in this gene are associated with multiple types of corneal dystrophy CD99 Cell surface glycoprotein. Oh et al. (2007); Mamdouh et al. (2009); Involved in leukocyte migration, T-cell adhesion, Chaerkady et al. (2010) ganglioside GM1 and transmembrane protein transport, and T-cell death by a caspase-independent pathway The encoded protein may have the ability to rearrange the actin cytoskeleton Tumour suppressor in osteosarcoma CD201 Receptor for activated protein C; involved in blood coagulation Balazs et al. (2006) Mutations associated with venous thromboembolism and myocardial infarction, as well as with late fetal loss during pregnancy. Member of the G protein-coupled receptor superfamily Iida et al. (2003); Tsukada et al. (2003) G protein-coupled Produced predominantly in vascular smooth muscle cells receptor 180 May play an important role in the regulation of vascular remodelling Cardiotrophin CLCF1Member of the glycoprotein (gp) 130 cytokine family Tsai et al. (2011); Senaldi et al. (1999) Together with cardiotrophin-like cytokine factor 1 (CLCF1) forms a dimer that competes with ciliary neurotrophic factor (CNTF) for binding to the CNTFR complex, and activates the Jak-STAT signalling cascade Potent neurotrophic factor, B-cell stimulatory agent and neuroendocrine modulator Defects in CLCF1 cause cold-induced sweating syndrome 2 ASPH It is thought to play an important role in calcium homeostasis. Yuan et al. (2007); Dulhunty et al. (2009) The longest isoforms include a C-terminal aspartyl/asparaginyl beta-hydroxylase domain; epidermal growth factor (EGF)-like domains of some proteins are targets for the activity. FAM62A May play a role as calcium-regulated intrinsic membrane protein Lange et al. (2009) MMP-14 Involved in the breakdown of extracellular matrix in normal Lu et al. (2010); Shirvaikar et al. (2010); physiological processes Ries et al. (2007) Member of the membrane-type (MT)-matrix metalloproteinase (MMP) subfamily; contains a potential transmembrane domain suggesting that is expressed at the cell surface rather than secreted. MYADM Myeloid-associated differentiation marker; myeloid up Aranda et al. (2011); Pettersson et al. (2000); regulated protein Wang et al. (2007) MYADM is a novel haematopoietic-associated gene that is upregulated when multipotent cells are allowed to differentiate toward the granulocytic and monocytic lineages Rettig et al., 1994; Ghersi et al., 2006; Seprase Homodimeric integral membrane gelatinase belonging to the serine protease family Bae et al., 2008) Involved in the control of fibroblast growth or epithelial– mesenchymal interactions during development, tissue repair and epithelial carcinogenesis. Selectively expressed in reactive stromal fibroblasts of epithelial cancers, granulation tissue of healing wounds, and malignant cells of bone and soft tissue sarcomas
Transforming growth factor β ig-h3 (TGFBI)
*References are included in Supporting material online.
to acute injuries. Ferlins are calcium-sensing, C2 domaincontaining proteins that regulate myoblast fusion and are implicated in human muscle disease. Ferlins are also expressed in other tissues (e.g. endothelium and placenta), and are putative regulators of endocytosis (Posey et al., 2011). Apart from their expression in myoblasts, ferlins have not been previously identified stem cell of progenitor compartments. Myoferlin regulates intracellular trafficking, including the endocytic recycling of growth factor receptors such as IGF1R and VEGFR-2 (Demonbreun et al., 2010; Yu Copyright © 2013 John Wiley & Sons, Ltd.
et al., 2011), which are important factors in CPCs (Ellison et al., 2011; Hosoda, 2012). Myoferlin's role in CPC biology may be related to the maintenance of membrane function and turnover, although direct experimental evidence will be required to establish this. PODXL is a CD34-family celladhesion glycoprotein, abundantly expressed in the kidney glomerular epithelium (Larrucea et al., 2008). High PODXL expression is associated with poor prognosis in several human cancers, such as glioblastoma, prostate and colorectal cancers, and small cell lung carcinoma (Koch et al., 2008). In J Tissue Eng Regen Med (2013) DOI: 10.1002/term
I. Moscoso et al.
Figure 2. Identified putative stem-cell markers in human cardiac progenitor cells (hCPCs). A, Expression of genes encoding identified proteins were checked by quantitative real-time polymerase chain reaction and compared with the expression in human mesenchymal stem cells (hMSCs) and foreskin fibroblasts. Variability among isolates was evaluated in additional hCPC samples (H1 and H4) (n = 2). B, Flow cytometry confirmation of the expression of CD26, myoferlin, PODXL and other membrane markers. C, Western blot for CD26, myoferlin and PODXL in cardiomyocytes (CM), hCPC (BC19) and a control cell line (HepG2)
some cases demethylated CpG islands in the podocalyxin-like gene have been described, indicating epigenetic regulation (Koch et al., 2008). More recently PODXL has been recognized as a marker of embryonic and adult stem cells, including postnatal retina Nrl-GFP-expressing precursors and MSCs from adipose tissue, cord blood and bone marrow (Lee et al., 2009a). Moreover, erythropoietinregulated PODXL expression is proposed to mobilize erythroblasts from a hypothesized stromal niche and possibly to promote reticulocyte egress (Sathyanarayana et al., 2007). Interestingly, PODXL is the antigen targeted by the cytotoxic monoclonal antibody mAb-84 in hESCs, which triggers the formation of membrane pores by PODXL aggregates, resulting in cell death (Tan et al., 2009). mAb-84 is proposed as a tool for eliminating undifferentiated hESC remnants from batches of differentiated cells before clinical use. In addition, high expression of PODXL in an hMSC subpopulation (PODXLhi/CD49fhi) is associated with high clonogenicity and low differentiation potential (Lee et al., 2009a). PODXL is also selectively expressed in the blood-brain-barrier but not in other microvessel structures (Agarwal et al., 2010). Our results demonstrate that PODXL is preferentially expressed in hCPCs compared with hMSCs and fibroblasts, suggesting value as an hCPC biomarker.
3.3. PODXL levels influence the differentiation capacity of hCPCs Subsequent analyses focused on the potential role of PODXL in the regulation of hCPC function. To induce loss or gain of function, hCPCs were transduced with lentiviral vectors encoding PODXL shRNA or full-length PODXL to generate Copyright © 2013 John Wiley & Sons, Ltd.
hCPCsh and hCPCover cells. Inhibition and overexpression were validated by immunofluorescence (not shown) and qRT-PCR (Figure 3A), which demonstrated a ~0.3-fold expression decrease in hCPCsh cells and a ~8-fold increase in hCPCover cells. The effects of altered PODXL expression were tested with a panel of genes whose function has been related to PODXL (Dawn et al., 2006; Darmellah et al., 2009; Lee et al., 2009a; Martinez-Estrada et al., 2010; Martinez-Sales et al., 2011). Upregulation of PODXL in hCPCover cells increased the expression of CD62L and SELE, and had a weaker positive effect on the expression of SLC9A3R2 and WT1; in contrast, expression of ITGA6 was reduced. The downregulation of PODXL in hCPCsh cells moderately reduced SLC9A3R2 expression and had a stronger effect on CD62L, EZR and SELE (Figure 3A). Silencing of PODXL induced marked and significant expression of the cardiotrophin-like cytokine factor 1 gene (CLCF1) and GPR180 (Figure S3C,D), with increases of 2.4-fold and 4.4-fold, respectively; however, these genes were unaffected by PODXL overexpression. CLCF1 is one of the ligands for ciliary neurotrophic factor (CNTF), and CLCF1 mutations underlie 10% of cases of cold-induced sweating syndrome (CISS)/Crisponi syndrome (Hahn and Boman, 1993). GPR180, an orphan G-coupled receptor preferentially expressed in vascular smooth muscle cells and associated with restenosis (Tsukada et al., 2003), has not been previously reported in stem cell systems. Modulations of E-selectin, CD62L and WT1 were confirmed by immunocytochemistry (ICC) (Figure 3B). ITGA6 and CD62L are related to MSC migration to infarcted cardiac tissue (Dawn et al., 2006; Lee et al., 2009a). E-selectin is highly expressed in circulating endothelial cells in heart failure patients J Tissue Eng Regen Med (2013) DOI: 10.1002/term
PODXL is a marker of cardiac stem cells
Figure 3. Impact of PODXL inhibition and overexpression on the human cardiac progenitor cell (hCPC) profile. A, Expression of a naïve ), PODXL-depleted cells panel of genes related to podocalyxin-like protein 1 (PODXL) function in unmodified hCPCs (hCPC sh over (hCPC ) and PODXLoverexpressing cells (hCPC ) (n = 5). B, Expression of SELE, CD62L and WT1 was confirmed by over cells immunofluoresence staining (scale bar, 100 μm). C, Several genes related to adult progenitor status are upregulated in hCPC sh in non-expanded cultures (n = 5). D, After 21 days of expansion in standard medium, hCPC cultures show increased expression of sh over cardiomyocyte markers (n = 5). E, Immunofluorescent detection of the cardiac proteins Nkx2.5 and cTnT in hCPC and hCPC cells naïve , after 11 days' co-culture with neonatal rat cardiomyocytes (scale bar, 100 μm). F, Cardiomyocyte marker expression in hCPC over sh hCPC and hCPC cells expanded for 21 days on decellularized rat left ventricles or in standard culture flasks (n = 5)
(Martinez-Sales et al., 2011), EZR mediates the effects of Na+/H+ exchanger (NHE1) activation in cardiac myocytes (Darmellah et al., 2009), and WT1 is required for cardiovascular progenitor-cell formation (Martinez-Estrada et al., 2010). Gene expression of most of these proteins is upregulated in hCPCover cells. Interestingly, it has been proposed that PODXL transcription is regulated positively by WT1 and negatively by PINCH1 (Wang et al., 2011). Copyright © 2013 John Wiley & Sons, Ltd.
The reduced WT1 expression in hCPCsh cells suggests that WT1 is reciprocally regulated by PODXL. All these results suggest a role for PODXL in hCPCs compatible with all previous description in various cellular models. A large panel of genes related to functions in adult progenitor cells and mature cardiac cells was then examined. Several genes were upregulated in hCPCover cells relative to non-transduced cells (hCPCnaïve) (Figure 3C), J Tissue Eng Regen Med (2013) DOI: 10.1002/term
I. Moscoso et al.
while only MYH6 appeared to be negatively regulated in hCPCover cells. In contrast, hCPCsh cells showed a significant reduction in the expression of some (about a 40% of them) of the genes analysed, These results indicate that the expression level of PODXL influences the gene expression profile of hCPCs, and likely has an impact on their biological potential. Over-expression of PODXL seems to promote an immature (progenitor) expression profile, whereas PODXL down-modulation promotes a pro-differentiation state. This conclusion was confirmed by analysis of hCPCsh cells under sustained expansion for 21 days in normal culture medium. At the end of this period, hCPCsh cultures showed clear evidence of enhanced cardiomyogenic differentiation, with notable upregulation of MYH6 (8-fold compared with d0), cTnT (5.5-fold), Nkx2.5 (5-fold), ALB (4-fold) and ACTC1 (2.5-fold), among others (Figure 3D). However, hCPCnaïve and hCPCover cells did not show relevant variations in these genes over the expansion period. To further test the differential potential of hCPCs, co-cultures with NRCMs were established. After 2–3 weeks hCPCs were sorted [by green fluorescent protein (GFP) or Cherry fluorescence] and analyzed for expression of cardiomyocyte markers by qRT-PCR (Figure S3B). Again, inhibition of PODXL seems to favour differentiation compared with hCPCover and hCPCnaïve. Co-cultures were also analysed at end-point by ICC (Figures 3E and S3A), but no changes were found in any of the cardiac proteins analysed. As a further test of differentiation potential in a more physiological context, decellularized rat left ventricles were prepared (Ott et al., 2008) seeded with hCPCnaïve, hCPCover or hCPCsh cells. After 21 days, hCPCsh cells seeded on ventricular substrates showed expression of cardiac markers, including ACTC1, Tnn2 and MYH6, at higher levels than obtained in parallel conventional flask cultures (Figure 3F). In contrast, hCPCnaïve and hCPCover cells showed no evidence of spontaneous differentiation. These results strongly suggest that high levels of PODXL expression maintain hCPCs in the immature state, whereas its downregulation facilitates their differentiation.
3.4. PODXL promotes hCPC adhesion and migration and supports their immunosuppressive capacity To examine the effect of PODXL expression on hCPC biological functions we first analysed adhesion properties. Detachment assays were carried out to measure the strength of hCPC–endothelium interactions under a shear stress (1.8 dyn/cm2), which is within the physiological flow range found in post-capillary venules where most extravasation occurs. Under these conditions, hCPCsh cells showed weaker binding to Human Umbilical Vein Endothelial Cells (HUVEC) than hCPCnaïve and hCPCover (Figure 4A). In addition, careful inspection of cell morphology revealed marked changes in hCPCover cells (increased membrane protrusive activity and fast spreading on endothelium), whereas hCPCnaïve cells showed a less prominent phenotype and most hCPCsh cells remained round Copyright © 2013 John Wiley & Sons, Ltd.
(Figure 4B). This behaviour of hCPCover cells suggests a higher pro-migratory potential. To further characterize the phenotypes of hCPC populations their migration capacity was examined. Migration assays were performed in Boyden chambers; cells were seeded in the upper part of the chamber in medium without fetal bovine serum (FBS), and medium supplemented with FBS was added to the lower part. hCPCover cells showed a greater capacity to migrate to the lower chamber than either hCPCnaïve or hCPCsh cells, which did not differ in migration capacity (Figure 4C). In wound-healing assays, cells were grown to confluence and then monolayers were scratched and cultured in an incubator for a further 72 h. hCPCover cells recolonized the wound area faster than either hCPCnaïve or hCPCsh cells (Figure 4D,E), correlating with the cell-shape differences detected in detachment assays (Figure 4B). The cell-shape changes associated with altered PODXL expression are linked to changes in hCPC actin dynamics and adhesion and migration properties. PODXL overexpression increased hCPC membrane protrusive activity under shear flow, resulting in faster spreading and enhanced adhesion on endothelium. Overexpression of PODXL also promoted migration, in agreement with the recent description of the proadhesive properties of human PODXL, which enhances the adherence of CHO cells to immobilized ligands and increases migration and cell–cell interactions in an integrindependent manner (Larrucea et al., 2008). The expression of the PODXL gene is strongly increased during TGFβ-induced endothelial-to-mesenchymal transition (EMT) in migrating A549 cells (Meng et al., 2011). Podocalyxin silencing reduced cell-ruffle formation, spreading and migration, and affected the expression patterns of several proteins that normally change during EMT (e.g. vimentin and E-cadherin); moreover, collagen type I co-localized with podocalyxin at the leading edges of migrating cells (Meng et al., 2011). Levels of membraneexposed PODXL can be also regulated by proteolytic cleavage of the ectodomain, and the released ectodomains can interfere with signalling via intact membrane-bound PODXL (Fernandez et al., 2011). PODXL-enhanced hCPC adhesion and migration has an important impact on their immunoregulatory capacity. Immunomodulation is a characteristic of MSCs, which play specific roles in the maintenance of peripheral and transplantation tolerance, autoimmunity, tumour evasion, and fetal–maternal tolerance (Nauta and Fibbe, 2007). As hCPCs inhibit T-lymphocyte proliferation in co-culture (Figure 1F), this study analysed whether the impaired adhesion and migration by PODXL-depleted hCPCs affected their immunoregulatory potential. Whereas hCPCover cells reduced T-cell proliferation to a similar extent as hCPCnaïve cells, hCPCsh had no significant effect on lymphocyte proliferation (data not shown), suggesting a role in hCPC immunoregulatory capacity. The findings suggest that changes in PODXL expression correlate with the ability of hCPCs to attach to endothelium and migrate, and to modulate immune responses, demonstrating the immunoregulatory capacity of hCPCs (Malliaras et al., 2012). PODXL thus appears to be associated with the undifferentiated hCPC state, possibly by J Tissue Eng Regen Med (2013) DOI: 10.1002/term
PODXL is a marker of cardiac stem cells
Figure 4. Effect of podocalyxin-like protein 1 (PODXL) on human cardiac progenitor cell (hCPC) adhesion, migration and immunosuppressive capacity. A,B, Detachment assay with hCPCs bound to a tumour necrosis factor (TNF)-α-treated Human Umbilical Vein naïve sh over Endothelial Cells (HUVEC) monolayer. hCPC , hCPC or hCPC cells were allowed to adhere to the HUVECs for 5 min under 2 static conditions, and a shear stress of 1.8 dyn/cm was then applied for 20 min. A, Quantification of the remaining adhered 2 hCPCs after the assay relative to the initial number of bound. Data are means ± SD from three different fields of view (0.45 mm /field) (*p < 0.05; n = 2). B, Cell morphology parameters calculated for individual adherent cells after the detachment assay. Data are means ± SD from a representative experiment (n = 10 cells per condition; *p < 0.05, † p < 0.01). C, Migration transwell assays demonover naïve sh have a higher migration capacity than hCPC and hCPC cells. D,E, Wound-healing assay, showing rapid strate that hCPC over cells (scale bar, 200 μm). Data are means ± SD (n = 4; *p < 0.05, †p < 0.01) recolonization of the wound area by hCPC
favouring specific interactions within niches. Modulation of PODXL expression may be involved in the regulation of CPC activation, migration and differentiation (Figure S4) and therefore in the fine control of their local immunoregulatory capacity.
of Science and Innovation (SAF 2008-02099; PLE2009-0147 and PSE-010000-2009-3), Comunidad Autónoma de Madrid (P-BIO-0306-2006) and the European Commission (FP7-HEALTH2009/CARE-MI). I.M. is currently a postdoctoral fellow funded by ISCIII. Inbiomed Foundation was supported by the Obra Social KUTXA. CNIC is supported by The Spanish Ministry of Science and Innovation and the Pro-CNIC Foundation.
Acknowledgements The authors thank all members of Bernad laboratory for their helpful discussions and technical support, and especially Carmen Albo and Viral Vectors, Microscopy and Cellomics Units, at CNIC. This work was supported by grants to A.B. from the Ministry Copyright © 2013 John Wiley & Sons, Ltd.
Conflict of interest The authors have declared that there is no conflict of interest. J Tissue Eng Regen Med (2013) DOI: 10.1002/term
I. Moscoso et al.
Supporting information on the internet The following supporting information may be found in the online version of this article: Figure S1. Polymerase chain reaction expression analysis of growth factors and interleukins expressed in human cardiac progenitor cells (hCPCs). Figure S2. Analysis of human cardiac progenitor cell (hCPC) membrane subproteome Figure S3. Immunofluorescence analysis of hCPCsh and hCPCover cells. Figure S4. Hypothetical influence of the level of podocalyxinlike protein (PODXL) expression on progenitor cell state. Methods S1. Isolation and culture of human cardiac progenitor cells (hCPC). Human bone marrow mesenchymal stem cells (hMSC). Neonatal cardiomyocyte co-cultures.
Cardiomyocytes red fluorescence staining. Decellularized rat hearts. Cell surface labeling and affinity purification. Enzyme digestion, LC-MS/MS analysis and database searching. Lentiviral vector production and transduction. Flow cytometry. Immunosuppression assay in vitro. Adhesion and migration assays. Pluripotency protein array. qRT-PCR. Western blotting. Antibodies and immunofluorescence. Xenotransplantation of human CPC into immunodeficient rats. References Material and Methods. References Table 1 and Table S1. Table S1. General functions of the human cardiac progenitor cell (hCPC) membrane proteins identified by liquid chromatography–mass spectrometry/mass spectrometry (LC-MS/MS). Table S2. Primers used in quantitative real-time polymerase chain reaction analyses. Table S3. Antibodies used in flow cytometry and immunofluorescence assays.
References Agarwal N, Lippmann ES, Shusta EV. 2010; Identification and expression profiling of blood-brain barrier membrane proteins. J Neurochem 112: 625–635. Anversa P, Leri A, Rota M et al. 2007; Concise review: stem cells, myocardial regeneration, and methodological artifacts. Stem Cells 25: 589–601. Assmus B, Rolf A, Erbs S et al. 2010; Clinical outcome 2 years after intracoronary administration of bone marrow-derived progenitor cells in acute myocardial infarction. Circ Heart Fail 3: 89–96. Beltrami AP, Barlucchi L, Torella D et al. 2003; Adult cardiac stem cells are multipotent and support myocardial regeneration. Cell 114: 763–776. Bergmann O, Bhardwaj RD, Bernard S et al. 2009; Evidence for cardiomyocyte renewal in humans. Science 324: 98–102. Bolli R, Chugh AR, D'Amario D et al. 2011; Cardiac stem cells in patients with ischaemic cardiomyopathy (SCIPIO): initial results of a randomised phase 1 trial. Lancet 378: 1847–1857. Cao R, Chen K, Song Q et al. 2012; Quantitative proteomic analysis of membrane proteins involved in astroglial differentiation of neural stem cells by SILAC labeling coupled with LC-MS/MS. J Proteome Res 11: 829–838. Carvalho MM, Teixeira FG, Reis RL et al. 2011; Mesenchymal stem cells in the umbilical cord: phenotypic characterization, secretome and applications in central nervous system regenerative medicine. Curr Stem Cell Res Ther 6: 221–228. Darmellah A, Rucker-Martin C, Feuvray D. 2009; ERM proteins mediate the effects + + of Na /H exchanger (NHE1) activation in cardiac myocytes. Cardiovasc Res 81: 294–300. Dawn B, Guo Y, Rezazadeh A et al. 2006; Postinfarct cytokine therapy regenerates cardiac tissue and improves left ventricular function. Circ Res 98: 1098–1105. Demonbreun AR, Posey AD, Heretis K et al. 2010; Myoferlin is required for insulin-
Copyright © 2013 John Wiley & Sons, Ltd.
like growth factor response and muscle growth. FASEB J 24: 1284–1295. Ellison GM, Torella D, Dellegrottaglie S et al. 2011; Endogenous cardiac stem cell activation by insulin-like growth factor-1/ hepatocyte growth factor intracoronary injection fosters survival and regeneration of the infarcted pig heart. J Am Coll Cardiol 58: 977–986. Fernandez D, Larrucea S, Nowakowski A et al. 2011; Release of podocalyxin into the extracellular space. Role of metalloproteinases. Biochim Biophys Acta 1813: 1504–1510. Gyongyosi M, Lang I, Dettke M et al. 2009; Combined delivery approach of bone marrow mononuclear stem cells early and late after myocardial infarction: the MYSTAR prospective, randomized study. Nat Clin Pract Cardiovasc Med 6: 70–81. Hahn AF, Boman H. 1993; Cold-induced sweating syndrome including Crisponi Syndrome. In GeneReviews, Pagon RA, Bird TD, Dolan CR et al. (eds). Seattle, WA: University of Washington, Seattle, 1993–2013. Hosoda T. 2012; C-kit-positive cardiac stem cells and myocardial regeneration. Am J Cardiovasc Dis 2: 58–67. Koch LK, Zhou H, Ellinger J et al. 2008; Stem cell marker expression in small cell lung carcinoma and developing lung tissue. Hum Pathol 39: 1597–1605. Laflamme MA, Murry CE. 2011; Heart regeneration. Nature 473: 326–335. Larrucea S, Butta N, Arias-Salgado EG et al. 2008; Expression of podocalyxin enhances the adherence, migration, and intercellular communication of cells. Exp Cell Res 314: 2004–2015. Lee RH, Seo MJ, Pulin AA et al. 2009a; The CD34-like protein PODXL and alpha6integrin (CD49f) identify early progenitor MSCs with increased clonogenicity and migration to infarcted heart in mice. Blood 113: 816–826. Lee SK, Kim Y, Kim SS et al. 2009b; Differential expression of cell surface proteins in human bone marrow mesenchymal stem
cells cultured with or without basic fibroblast growth factor containing medium. Proteomics 9: 4389–4405. Lee ST, White AJ, Matsushita S et al. 2011; Intramyocardial injection of autologous cardiospheres or cardiosphere-derived cells preserves function and minimizes adverse ventricular remodeling in pigs with heart failure post-myocardial infarction. J Am Coll Cardiol 57: 455–465. Leri A, Kajstura J, Anversa P. 2011; Role of cardiac stem cells in cardiac pathophysiology: a paradigm shift in human myocardial biology. Circ Res 109: 941–961. Malliaras K, Li TS, Luthringer D et al. 2012; Safety and efficacy of allogeneic cell therapy in infarcted rats transplanted with mismatched cardiosphere-derived cells. Circulation 125: 100–112. Martin-Puig S, Wang Z, Chien KR. 2008; Lives of a heart cell: tracing the origins of cardiac progenitors. Cell Stem Cell 2: 320–331. Martinez-Estrada OM, Lettice LA, Essafi A et al. 2010; Wt1 is required for cardiovascular progenitor cell formation through transcriptional control of Snail and Ecadherin. Nat Genet 42: 89–93. Martinez-Sales V, Sanchez-Lazaro I, Vila V et al. 2011; Circulating endothelial cells in patients with heart failure and left ventricular dysfunction. Dis Markers 31: 75–82. Mazo M, Hernández S, Gavira JJ et al. 2012; Treatment of reperfused ischemia with adipose-derived stem cells in a preclinical Swine model of myocardial infarction. Cell Transplant 21: 2723–2733. Meng X, Ezzati P, Wilkins JA. 2011; Requirement of podocalyxin in TGF-beta induced epithelial mesenchymal transition. PLoS One 6: 1–10. Messina E, De Angelis L, Frati G et al. 2004; Isolation and expansion of adult cardiac stem cells from human and murine heart. Circ Res 95: 911–921. Nauta AJ, Fibbe WE. 2007; Immunomodulatory properties of mesenchymal stromal cells. Blood 110: 3499–3506.
J Tissue Eng Regen Med (2013) DOI: 10.1002/term
PODXL is a marker of cardiac stem cells Ott HC, Matthiesen TS, Goh SK et al. 2008; Perfusion-decellularized matrix: using nature's platform to engineer a bioartificial heart. Nat Med 14: 213–221. Posey AD Jr, Pytel P, Gardikiotes K et al. 2011; Endocytic recycling proteins EHD1 and EHD2 interact with fer-1-like-5 (Fer1L5) and mediate myoblast fusion. J Biol Chem 286: 7379–7388. Sarkar P, Collier TS, Randall SM et al. 2012; The subcellular proteome of undifferentiated human embryonic stem cells. Proteomics 12: 421–430. Sathyanarayana P, Menon MP, Bogacheva O et al. 2007; Erythropoietin modulation of
Copyright © 2013 John Wiley & Sons, Ltd.
podocalyxin and a proposed erythroblast niche. Blood 110: 509–518. Segers VF, Lee RT. 2008; Stem-cell therapy for cardiac disease. Nature 451: 937–942. Stastna M, Chimenti I, Marban E et al. 2010; Identification and functionality of proteomes secreted by rat cardiac stem cells and neonatal cardiomyocytes. Proteomics 10: 245–253. Tan HL, Fong WJ, Lee EH et al. 2009; mAb 84, a cytotoxic antibody that kills undifferentiated human embryonic stem cells via oncosis. Stem Cells 27: 1792–1801. Tsukada S, Iwai M, Nishiu J et al. 2003; Inhibition of experimental intimal thickening
in mice lacking a novel G-protein-coupled receptor. Circulation 107: 313–319. Wang D, Li Y, Wu C et al. 2011; PINCH1 is transcriptional regulator in podocytes that interacts with WT1 and represses podocalyxin expression. PLoS One 6: 1–10. Yang L, Nyalwidhe JO, Guo S et al. 2011; Targeted identification of metastasisassociated cell-surface sialoglycoproteins in prostate cancer. Mol Cell Proteomics 10: 1–16. Yu C, Sharma A, Trane A et al. 2011; Myoferlin gene silencing decreases Tie-2 expression in vitro and angiogenesis in vivo. Vascul Pharmacol 55: 26–33.
J Tissue Eng Regen Med (2013) DOI: 10.1002/term
