- Email: [email protected]
The assessment of the in vivo to in vitro cellular transition of human umbilical cord multipotent stromal cells
Placenta 36 (2015) 232e239
Contents lists available at ScienceDirect
Placenta journal homepage: www.elsevier.com/locate/placenta
The assessment of the in vivo to in vitro cellular transition of human umbilical cord multipotent stromal cells H. Coskun a, A. Can b, * a
Ankara University Biotechnology Institute, Tandogan, Besevler, 06110 Ankara, Turkey Ankara University School of Medicine, Department of Histology and Embryology, Laboratory for Stem Cells and Reproductive Medicine, Sihhiye, Ankara 06100, Turkey b
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 28 November 2014
Introduction: Human umbilical cord stroma is a rich source of primitive multipotent stromal cells (hUCMSCs). However, the methods for hUC-MSC isolation and propagation remain controversial and vary among laboratories. Our group previously demonstrated that two cell types emerge upon enzymatic isolation of hUC-MSCs, which subsequently undergo a transition towards a fibroblastoid phenotype in later passages. The aim of this study was to further analyse cultured hUC-MSCs by evaluating the cytoskeletal and cell adhesion proteins and by comparing the remodelling of those proteins in umbilical cord sections to determine the cell alterations due to enzymatic and explant methods. Methods: Tissue sections and cultured cells isolated by enzymatic or explant methods were analysed morphologically and by labelling cytokeratin, vimentin, alpha-smooth muscle actin, E-cadherin and Ncadherin profiles. Results: The present observations confirmed that wide, flat cells (type-1) share myofibroblastic features, appear exclusively in enzymatically isolated early cultures; gradually diminish or are replaced by fibroblastoid cells (type-2) in later passages. In contrast, the explant method does not result in the existence of type-1 cells in vitro. Among the tested CK subtypes, CK18 expression is upregulated, whereas CK19 expression is downregulated upon culturing after both protocols. Vimentin and a-SMA, as the major intermediate filaments of hUC-MSCs were found unaltered throughout the culturing period regardless of the cell isolation technique used. Discussion: The data presented confirm and further elucidate the previously observed phenotypic change in hUC-MSCs as illustrated by alterations in structural proteins during enzymatic isolation and subsequent culturing of cells compared with in situ equivalents. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Umbilical cord Mesenchymal stem cell Vimentin Cytokeratin Alpha-smooth muscle actin Cadherin
1. Introduction Over the last decade, human umbilical cord (hUC) matrix, which is generally known as Wharton's jelly, has been the focus of many studies, with special emphasis on the spatial distribution and functional structure of stromal cells, namely, human umbilical cord multipotent stromal cells (hUC-MSCs) and the surrounding extracellular matrix, which serves as a niche for the residing cells [1]. The expression of certain embryonic and mesenchymal stem cell markers in these cells provides suggestive evidence regarding their multipotency [2,3]. Recently, this potential has ultimately led many researchers to initiate the evaluation of these cells in clinical trials,
* Corresponding author. Tel.: þ90 312 5958169; fax: þ90 312 3106370. E-mail address: [email protected] (A. Can). http://dx.doi.org/10.1016/j.placenta.2014.11.024 0143-4004/© 2014 Elsevier Ltd. All rights reserved.
particularly for immunomodulation in inflammatory diseases, such as liver cirrhosis [4,5], systemic lupus erythematous (SLE) [6], and type 1 [7] and type 2 diabetes [8]. While hUC-MSCs still require further characterisation, these cells have been successfully differentiated in vitro into various cell types derived from three germ layers (reviewed by Kim et al., 2013 [9]). Although more than 50 clinical trials are currently underway (www.clinicaltrials.gov), only a few studies [10e12] have specifically focused on the cell surface alterations when cells are either enzymatically extracted from the tissue of origin or mechanically dissected and then induced to migrate from tissue explants in vitro conditions, which would allow cells to subsequently proliferate, stay viable and maintain multipotency for prolonged periods. Tissue resident hUC-MSCs principally resemble naïve mesenchymal fibroblasts found in almost all connective tissues. However, ultrastructural studies have indicated that hUC-MSCs also possess
H. Coskun, A. Can / Placenta 36 (2015) 232e239
some cellular characteristics belonging to myofibroblasts and to smooth muscle cells [11,13,14]. Freshly isolated and plated cells primarily display a fibroblastic appearance over the culture period. However, our group has previously reported that two distinct cell phenotypes (i.e., type-1 and type-2) immediately emerge upon primary monolayer cell culturing, starting from the first passage [1,15,16]. Type-1 cells exhibit flat, wide cytoplasm corresponding to their myofibroblastic nature, whereas type-2 cells appear to have a fibroblastoid phenotype. In general, type-1 cells dominate in early passages, whereas later passages are practically composed of only type-2 cells. Therefore, we hypothesised that a cellular transition or replacement rapidly occurs upon extracting cells from their native environment and continues for a certain period in vitro until cells accommodate solely to the culture conditions. The aims of this study were to determine whether i) cell isolation and early propagation techniques have an effect on the appearance of these two phenotypes and ii) expression of major intracytoplasmic filamentous proteins intrinsically found in hUC-MSCs residing in different umbilical cord subcompartments undergo alterations due to the extraction and subsequent propagation of cells in vitro. For this purpose, we evaluated stromal cells in the adjacent subcompartments of cord sections and in subsequent cultures (P0eP7) regarding their major intracellular cytoskeletal and intercellular adhesion proteins, i.e., cytokeratin (CK) types, vimentin (vim), asmooth muscle actin (a-SMA), F-actin and cadherin family CAMs, such as E-cadherin (E-cad) and N-cadherin (N-cad). The abundance of type-1 cells especially in enzymatically-isolated cells, which gradually diminish in subsequent passages, suggests that cellecell and cell-matrix forces during cell isolation procedures strictly interact with cytoskeletal network and CAMs that ultimately result dramatic changes in cell phenotype. 2. Materials & methods 2.1. Processing of hUCs for microscopy, isolation, and culture of hUC-MSCs Umbilical cords were obtained from full-term infants by Caesarean sections (n ¼ 6). Ethical approval was obtained from the Institutional Ethical Review Board (approval # 18-578-12, 2012). All chemicals and reagents used in this study were purchased from SigmaeAldrich (St. Louis, MO, USA) unless stated otherwise. 15 cmlong cords were transferred to the laboratory in Leibovit's 15 media supplemented with 3% (w/v) penicillin-streptomycin and amphotericin-B (5 mg/mL). A block of cord from each sample was fixed in 4% (w/v) paraformaldehyde (PFA) for 24 h, immersed in 1.2 M sucrose solution as a cryoprotectant, and finally frozen at 60 C for cryosectioning (8 mm thick). For cell isolation, the cord stroma was initially freed from the umbilical vessels and the amniotic membrane. Enzymatic cell isolation or mechanical dissection/explant cell culture techniques were performed on the remaining stromal tissue. The enzymatic method, which was based on the use of 0.1% collagenase type I and 2% dispase, was previously described in detail [17]. The explant method was implemented by plating the mechanically dissected stromal
233
tissue fragments onto polystyrene tissue culture flasks and then allowing these fragments to attach for 24 h, followed by resuspension in culture medium. Following the log phase and confluence in primary culture (P0), hUC-MSCs were subcultured (P1) approximately every 8e10 days in DMEM-Ham's F12 culture media, containing 10% FCS, up to the seventh passage (P7). In addition to culturing the cells in polystyrene flasks, cells were also cultivated on poly L-lysine-coated glass coverslips for immunostaining and high-resolution confocal microscopy. At around 4e5th day of each passage, cells were evaluated for morphological and immunofluorescent protein analyses (see below).
2.2. Immunocytochemistry and microscopic analysis Single- and double-labelling procedures were performed using a series of antibodies against major cytoskeletal proteins and CAMs. Cell monolayers on glass coverslips were fixed with 3.5% PFA at room temperature (RT) for 30 min. Fixed cells were incubated with 1% bovine serum albumin (BSA) in PBS with 0.1% Triton X-100 at RT and were kept in PBS at 4 C until processing. Mouse monoclonal antibodies against pan-CK (CK types 4, 5, 6, 8, 10, 13, 18), CK18, CK19, vimentin, a-SMA, Ecadherin, N-cadherin (Invitrogen, NY, USA) and FITC-phalloidin (35 mg/mL in PBS for 60 min) were applied. Cy3-conjugated goat anti-mouse IgG (Jackson Immunoresearch Laboratories, USA) and FITC-conjugated goat anti-mouse IgG (Jackson Immunoresearch Laboratories, USA) served as secondary antibodies. 7Aminoactinomycine D (7-AAD) (100 mM for 20 min) or Hoechst 33258 (1 mg/mL in 1:1 PBS/glycerol mounting medium) were used for nuclear labelling. Unless otherwise indicated, all antibodies were diluted 1:100 in PBS and incubated for 90 min at 37 C in a humidified chamber. All labelled and stained cells were examined using a Carl Zeiss LSM 510 confocal laser scanning microscope (Jena, Germany), equipped with a Plan-Neofluar 40/1.3 NA oil objective, 488-nm argon ion and 543-nm green helium neon laser lines. The detection parameters, including laser intensity, amplifier offset, amplifier gain, and pinhole diameter were optimised and fixed at the same values for all specimens. Three-dimensional images were reconstructed by consecutive optical sections of various thicknesses (0.25e0.38 mm) using LSM 510 v.3.8 software. Semi-quantitative evaluations of signals were independently performed by the authors. The expression of CK, vimentin and aSMA was defined as relatively diffuse cytoplasmic staining in cord sections and as characteristic cytoplasmic filamentous patterns in cultured cells. Positive cadherin expression was defined as tiny dots and rods in cord sections and as distinctive membranous rims in cultured cells. The varying signal intensities were analysed using ImageJ v.3.91 software. The formula used to calculate the corrected total cell fluorescence (CTCF) was CTCF ¼ Integrated density (area of selected cell mean fluorescence of background readings) (Fig. 1). The CTCF results were converted to ratiometric values to compare signals as fold increases. The percentage of positively stained cultured cells was scored as one positive (10e25% of cells were positively stained), two positive (26%e50% of cells were positively stained), three positive (more than 50% of the cells were positively stained) and four positive (almost all cells were positively stained).
2.3. Statistical analysis The statistical analysis was performed using Statistical Package for Social Sciences v15.0 software (SPSS Inc., USA). Statistical methods included an analysis of variance (ANOVA) for comparing multiple groups. The correlation between two groups was assessed using Pearson's correlation analysis. p < 0.05 was considered statistically significant.
Fig. 1. Quantification of total fluorescence (B) in each raw figure (A) was calculated using a formula named as “corrected total cell fluorescence (CTCF)”. The average of values emitted from three different areas on background fluorescence (C) was used in the formula as mean fluorescence of background readings (shown by three circles).
234
H. Coskun, A. Can / Placenta 36 (2015) 232e239
3. Results 3.1. Phenotypic evaluation of hUC-MSCs in culture Enzymatically dissociated or mechanically chopped cord explants were initially grown in polystyrene T75 culture flasks until reaching 80% confluency (P0). Type-1 cells (Fig. 2A) were frequently encountered among type-2 cells immediately after inoculating the enzymatically dissociated cells (Fig. 2D), persisted in early cultures (P1e3), then gradually decreased (Fig. 2E), and almost disappeared in later passages (Fig. 2F and G). In contrast, only fibroblastoid type2 cells (Fig. 2A) dispersed in explant cultures (Fig. 1H) and persisted in the same form throughout all passages (P1e7) (Fig. 2IeK). Essentially, no cell phenotype difference was noted in later passages between enzymatically and mechanically dissociated cultures (compare Fig. 2F, G with Fig. 2J, K). Consistent with our previous experiments and reports, both group of cells were found positive (>90%) for common MSC cell surface markers, namely, CD73, CD90, and CD105, albeit in varying ratios, and were found negative for CD31 and CD45, as detected by flow cytometry (data not shown).
3.2. Cytoskeletal filament distribution in cord stromal cells Representative tissue compartments from cord sections were designated as amniotic membrane (AM), subamniotic stroma (SAS), intervascular stroma (IVS), and perivascular stroma (PVS), as illustrated in Fig. 3. As previously noted, we confirmed the
myofibroblastic features of UC stromal cells by evaluating the aSMA and vimentin filaments as the most common myofibroblast markers [18]. Second, the expression of pan-CK, and certain CK subtypes (CK18, CK19) (Fig. 3 and Table 1) and F-actin were evaluated. Both cell types displayed well-organised F-actin filaments (Fig. 2B and C). CKs were predominantly found in the AM and in the PVS, whereas lower quantities of CKs were found in the SAS and in the IVS (Fig. 3). As one of the components of the pan-CK antibody cocktail, CK18 alone was faint in the PVS, IVS, and SAS, and slightly higher in the AM (Fig. 3). In contrast, CK19, which was not included in the pan-CK antibody cocktail, was found strongly positive in the AM and SAS, extremely faint in the IVS and completely absent in the PVS (Fig. 3). Vimentin appeared as the most abundant intermediate filament protein in the entire IVS and PVS but not in the AM or SAS. aSMA was detected in all compartments, including the SAS, IVS and PVS, with the exception of AM.
3.3. Cytoskeletal filament distribution in cultured hUC-MSCs To address whether the isolation and/or transfer of cells/explants to culture plates initiates alterations in any of the cytoskeletal networks, we differentially analysed the cultured cells through passages 0e7. Representative confocal images are illustrated in Fig. 4, and semi-quantitative evaluations are provided in Table 1. Pan-CK was strongly positive in both explants and enzymaticallyisolated cells. Only type-1 cells were found positive for CK18. Interestingly, type-2 cells were almost devoid of both CK18 and CK19 (Fig. 4). Throughout the seven passages, all cells were strongly
Fig. 2. Morphological features of cultured hUC-MSCs detected by relief contrast (A, DeK) and confocal microscopy (B and C). Upon isolation of cells by the enzymatic dissociation of cords (Enz), mixtures of two distinct cell phenotypes e type-1 as myofibroblastoid (solid line-bordered cells in A; and arrowheads in DeF) and type-2 as fibroblastoid (dotted linebordered cells in A) were encountered (DeG). In contrast, only one cell phenotype (type-2) was observed in explant cultures (Exp) starting from P0 (H) throughout the entire passages (IeK). Extended F-actin filaments (stress fibres) were detected as a distinctive feature in flat and wide cytoplasm of type-1 cells implying a typical myofibroblast (B). F-actin filaments in type-2 cells appeared to form fibroblast-like organisation (C). Scale bars ¼ 50 mm.
H. Coskun, A. Can / Placenta 36 (2015) 232e239
235
Fig. 3. Cytoskeletal filament distribution (green signals) observed subcompartments of UC stromal cells. PanCK were predominantly found in AM and PVS, lower quantity in SAS and IVS. CK18 was almost absent in SAS, IVS and PVS, slightly positive in AM. In contrast, CK19 was strongly positive in AM and SAS cells, totally absent in IVS and PVS. Vimentin was found dominant in the entire IVS and PVS, but not found AM and SA. aSMA was detected in all compartments including SAS, IVS and PVS with the exception of AM. Red signal ¼ Nucleus (7AAD). Signal intensity was calculated using CTCF formula (see, Fig. 1 and Materials & Methods) and is calculated results were converted to ratiometric values to compare signals as fold increase as illustrated in bar graph. For each group of marker, signals were set relative to the level of lowest signal that were normalised to 1. Note the extensive fold increase in vimentin signal (57.3 fold increase) in PVS. Asterisks refer to significantly different signals (p < 0.05) from the remaining ones within the same group. Scale bars ¼ 50 mm.
236
H. Coskun, A. Can / Placenta 36 (2015) 232e239
Table 1 Abundance of cytoskeletal filaments in cultured hUC-MSCs through passages. P0
P2
Enza PanCK CK18 CK19 Vim aSMA a b c
c
þþþ þþ þþ þþþþ þþþþ
P4
P7
Expb
Enz
Exp
Enz
Exp
Enz
Exp
þþ þþ þ þþþþ þþþþ
þþþþ þþ þ þþþþ þþþþ
þþþ þþ þ þþþþ þþþþ
þþþþ þþþ þ þþþþ þþþ
þþþþ þþþ þ þþþþ þþþþ
þþþ þþþ þ þþþ þþþ
þþþ þþþ þ þþþþ þþþþ
Enzymatic isolation method. Mechanical dissection/explant method. See, Materials & methods for the evaluation protocol.
positive for vimentin and aSMA, regardless of the isolation technique used (Fig. 4). Cells extruded from tissue explants proved that all explants originated from the UC stroma and not from the AM because the explants stained positive for vimentin and aSMA. No significant staining intensity difference per cell was noted for any cytoskeletal element throughout the culturing period. 3.4. Cadherin expression of hUC-MSCs in cord sections and in cultures E-cadherin, which is an epithelial marker, was used to confirm that no epithelial cell contamination existed in the cultures. In the cord sections, AM was positive for E-cadherin, as anticipated,
Fig. 4. Intermediate filaments (CKs and vimentin) and aSMA distribution (green signals) in enzymatic and mechanical dissection/explant cultures. PanCK and CK18 were highly expressed in type-1 cells through P1eP7. CK18 was detected exclusively in type-1 cells; CK19 showed no positivity in any of the cell types, except mitotically active cells (arrowhead). Vimentin and aSMA were strongly expressed in both cell types throughout the entire passages. Red signal ¼ Nucleus (7AAD). Scale bars ¼ 50 mm.
H. Coskun, A. Can / Placenta 36 (2015) 232e239
whereas all stromal cells were entirely negative (Fig. 5). N-cadherin, which is a mesenchymal cell adhesion molecule [19,20], was found expressed in cells that resided in the IVS and in the PVS (Fig. 5). Cultured hUC-MSCs did not express E-cadherin after any of the cell isolation procedures. Interestingly, several membranelocalised N-cadherin molecules were found assembled, particularly in prolonged cultures (12e15 days after plating), and with significantly higher frequencies in explant cultures (Fig. 5). 4. Discussion Present data demonstrate that variations in cytoskeletal protein expressions in isolated hUC-MSCs strictly depend on the cell isolation technique, passage number and the tissue subcompartment from which part the umbilical cord was excised. Human UCMSCs represent an important multipotent stromal cell population that show promising therapeutic effects on clinical applications. Nonetheless, certain discrepancies in the hUC-MSC literature still exist in descriptions of the biological properties of MSCs.
237
Additionally, their overall therapeutic effects may be shadowed in part by the existence of aberrant cell populations or deviations that the tissue-resident cell possess in a given primary culture, including variations in function [21,22]. For instance, one of the most remarkable intermediate filament proteins, desmin, is consistently expressed in all stromal cells in vivo [11,16,23]; however, its expression gradually diminishes when cells are cultured in vitro [11,16,17]. Therefore, one can predict that the hUC-MSC isolation method, particularly the use of proteases to digest the extracellular matrix for enriching the stromal cells, is detrimental to hUC-MSC functions due to the non-specific degradation of certain cell surface receptors and interconnected intracytoplasmic domains [24,25], whereas the explant culture, which allows cells to proliferate and disperse from the existing tissue pieces, avoids such potential artefacts. As shown in the present study, the enzymatic isolation approach significantly induced the existence of myofibroblastic cells called type-1 cells. To offer a more physiological microenvironment while cells are being extracted from a highly rich collagen fibre and hyaluronic acid matrix, several UC-MSC
Fig. 5. E-cadherin and N-cadherin profiles (green signals) in cord stroma (upper row) and cultured hUC-MSCs (lower row). Faint E-cadherin was exclusively detected in AM, no staining was noted in the entire stroma. N-cadherin was mainly localized in IVS and PVS cells, while AM and SAS was totally lacking N-cadherin expression. No E-cadherin was noted in enzymatic or mechanical dissection/explant cultures or in isolated cells. Interestingly, N-cadherin expression was detected strongly positive in prolonged explant cultures (12e15 days) compared to that in enzymatically-isolated cells. Red signal ¼ Nucleus (7AAD). Scale bars ¼ 50 mm.
238
H. Coskun, A. Can / Placenta 36 (2015) 232e239
isolation techniques have been reported so far that considerably vary both in the use enzyme formulations and the mechanical splitting of the stromal tissue. In recent years, if the amniotic membrane and vessels are carefully removed to avoid any other cell contamination, then the explant method has been shown to be superior to the enzymatic method in many aspects for achieving higher cell proliferation rates, purity, and speed, with a lower risk of microbial contamination and, more importantly, clinical compatibility [2,10,12,26e29]. Margossian et al. recently assessed the differences in MSC markers between P0 and P1 cultures and found that CD29, CD44, CD73, CD90, CD105, CD164 and CD166 were significantly elevated at P1 compared with primary cultures (P0). Nevertheless, these researchers did not report any significant change in those markers arising from enzymatic or explant cultures at P1 [10]. More drastically, these researchers demonstrated the absence of CD271, which is one of the most selective markers for MSCs, after enzymatic culture, although CD271-positive IVS cells were found in situ. Similarly, Schugar et al. demonstrated the discordance in CD146, CD44, CD105 and CD73 levels before and after the isolation of cells with different enzyme formulations [12]. Garzon et al. recently demonstrated that hUC-MSCs belonging to different umbilical cord subcompartments and in subsequent cultures express different amount of CK filament subtypes in Ref. [30]. More than 20 different types of CK filaments exist in eukaryotic cells and are primarily categorised in two classes, type I CKs and type II CKs. Cytokeratins are usually found in pairs, which are composed of a type I cytokeratin and a type II cytokeratin in the cytoplasm. Among these cytokeratins, CK8 and CK18 are expressed in human myofibroblastic cells [31]. Several lines of human transformed fibroblasts also express CK 8 and CK18 [32]. In this study, we demonstrated that CK18 and CK19 have extremely low expression or absent in stroma-resident cells. In fact, all CK subtypes were relatively less abundant in the entire stroma compared with AM cells, which are true epithelial cells bearing junctional complexes and basement membranes. The Pan-CK antibody, which contains CK4, 5, 6, 8, 10, 13, and 18 subtypes, stained positive in all stromal cells and the highest staining was observed in the PVS suggesting that this positivity does not primarily originate from CK18 and CK19 but may originate from the expression of CK8 and other CK subtypes. In contrast, CK19 appears to be the dominant type of CK in the AM. Type-1 cells are rich in CK18, whereas type-2 cells, regardless of the isolation technique used, are devoid of CK18, a finding that is similar to that observed in tissue sections. The finding that CK18-positive staining is strongly found in type-1 cells but is lacking in tissue-resident hUC-MSCs supports the hypothesis that these cells are aberrant cell types that temporally express abundant CK18 upon enzymatic isolation. Although both type of cells possessed highly organised F-actin filaments, the typical orientation of these microfilaments demonstrated a clear morphological difference between those two cell types. The observation that both fibroblasts and myofibroblasts possess extensive stress fibres [18] that enable cells to firmly adhere to the substrate suggests that these cells can transdifferentiate into each other. Thus, the entire population of cells observed in cultures throughout all passages can be considered highly plastic and can adjust to the microenvironmental conditions. Because the culture conditions used in this study favour the proliferation of fibroblastoid cell types, the transdifferentiation of myofibroblastic type1 cells into a fibroblastic phenotype is an anticipated phenomenon. If this transdifferentiation is the case, then before this cell conversion, myofibroblasts will express high levels of CK18, which may be considered a differentiation marker of type-1 cells to type-2 cells. The widespread distribution and high intensity of the aSMA protein in type-1 cells also confirms that these cells are giant myofibroblasts that are occasionally encountered in cultures. In contrast, the
intense and widespread vimentin filaments detected before and after isolation clearly indicated that all cultured cells originated from cord stroma, which is an extraembryonic mesenchymal tissue, and not from the amniotic membrane, which originates from epiblast-amnioblast cells. Cadherin family proteins play important roles in cell adhesion, forming adherens junctions to bind cells together within tissues. So far, over 80 types of cadherins in humans have been identified and sequenced. Thus, similar to CKs, certain types of cadherin molecules designate the type or origin of cells [33,34]. Epithelial cells are generally positive for E-cadherin, whereas N-cadherin, which is known as a ‘mesenchymal cadherin’, dominates during the epithelial-to-mesenchymal transition [19]. In the present study, we evaluated the N-cadherin profiles concomitantly in cord sections and cultured cells to examine whether the in situ expression of Ncadherin undergoes a transition due to isolation and to the subsequent culturing conditions. N-cadherin expression was intrinsically expressed in resident cells of the IVS and PVS and in cultured cells after being synthesised in the cytoplasm and assembled in cell membranes. Interestingly, explant cultures provided significantly higher N-cadherin expression, which may be considered a positive sign of functional characteristics of a mesenchymal stem cell. Cells containing a specific cadherin subtype tend to cluster together and to exclude the other cadherin types, both in cell culture and during development [35]. Thus, using an explant culture system seems to provide a more physiological milieu to harvest intact cells compared with enzymatic isolation. 5. Conclusions The following is a summary of our findings: i) Type-1 cells, which were strongly proven as myofibroblasts, appeared mainly in enzymatically isolated early cultures. ii) The dominant fibroblastoid cells were consistently obtained by the explant culturing method. iii) Both procedures did not generally alter the CK distribution upon extraction. However, among the tested subtypes of CKs, CK18 expression was upregulated, whereas CK19 expression was downregulated upon monolayer cell culturing. Vimentin and a-SMA, which are the major intermediate filament proteins of hUC-MSCs particularly localised in the IVS and in the PVS, were found unaltered throughout the culturing period, regardless of the cell isolation technique used. iv) E-cadherin was consistently absent in all stromal cells both in vivo and in vitro, whereas N-cadherin was exclusively noted in the IVS and PVS sections, and noted in later passages with significantly higher expression under explant culture conditions. The data presented in this study represent a typical phenotypic change during isolation and culturing of hUC-MSCs compared with in situ equivalents. Although the possibility of cell contamination cannot be completely excluded during any of the tissue dissociation and subsequent culturing techniques, cells of interest can be purified over time by favouring the most suitable culture media, specific supplements and cell attachment conditions. Therefore, optimal cell isolation techniques and culture conditions should be precisely considered, particularly when preparing clinical-grade hUC-MSCs to be allogeneically transferred back in vivo. Conflict of interest The authors report no conflict of interest. Acknowledgements The authors would like to thank Prof. Fadıl Kara, M.D. who kindly provided umbilical cords and the mothers who were
H. Coskun, A. Can / Placenta 36 (2015) 232e239
involved in sample donation. This work was supported by funding Ankara University Scientific Research Project 13L3330018 (to HC and AC).
References [1] Can A, Karahuseyinoglu S. Concise review: human umbilical cord stroma with regard to the source of fetus-derived stem cells. Stem Cells 2007;25(11): 2886e95. [2] La Rocca G, Anzalone R, Corrao S, Magno F, Loria T, Lo Iacono M, et al. Isolation and characterization of Oct-4þ/HLA-Gþ mesenchymal stem cells from human umbilical cord matrix: differentiation potential and detection of new markers. Histochem Cell Biol 2009;131(2):267e82. [3] Fong CY, Chak LL, Biswas A, Tan JH, Gauthaman K, Chan WK, et al. Human Wharton's jelly stem cells have unique transcriptome profiles compared to human embryonic stem cells and other mesenchymal stem cells. Stem Cell Rev 2011;7(1):1e16. [4] Wang L, Li J, Liu H, Li Y, Fu J, Sun Y, et al. Pilot study of umbilical cord-derived mesenchymal stem cell transfusion in patients with primary biliary cirrhosis. J Gastroenterol Hepatol 2013;28(Suppl. 1):85e92. [5] Zhang Z, Lin H, Shi M, Xu R, Fu J, Lv J, et al. Human umbilical cord mesenchymal stem cells improve liver function and ascites in decompensated liver cirrhosis patients. J Gastroenterol Hepatol 2012;27(Suppl. 2):112e20. [6] Sun L, Wang D, Liang J, Zhang H, Feng X, Wang H, et al. Umbilical cord mesenchymal stem cell transplantation in severe and refractory systemic lupus erythematosus. Arthritis Rheum 2010;62(8):2467e75. [7] Hu J, Yu X, Wang Z, Wang F, Wang L, Gao H, et al. Long term effects of the implantation of Wharton's jelly-derived mesenchymal stem cells from the umbilical cord for newly-onset type 1 diabetes mellitus. Endocr J 2013;60(3): 347e57. [8] Liu X, Zheng P, Wang X, Dai G, Cheng H, Zhang Z, et al. A preliminary evaluation of efficacy and safety of Wharton's jelly mesenchymal stem cell transplantation in patients with type 2 diabetes mellitus. Stem Cell Res Ther 2014;5(2):57. [9] Kim DW, Staples M, Shinozuka K, Pantcheva P, Kang SD, Borlongan CV. Wharton's jelly-derived mesenchymal stem cells: phenotypic characterization and optimizing their therapeutic potential for clinical applications. Int J Mol Sci 2013;14(6):11692e712. [10] Margossian T, Reppel L, Makdissy N, Stoltz JF, Bensoussan D, Huselstein C. Mesenchymal stem cells derived from Wharton's jelly: comparative phenotype analysis between tissue and in vitro expansion. Biomed Mater Eng 2012;22(4):243e54. [11] Ryu YJ, Seol HS, Cho TJ, Kwon TJ, Jang SJ, Cho J. Comparison of the ultrastructural and immunophenotypic characteristics of human umbilical cordderived mesenchymal stromal cells and in situ cells in Wharton's jelly. Ultrastruct Pathol 2013;37(3):196e203. [12] Schugar RC, Chirieleison SM, Wescoe KE, Schmidt BT, Askew Y, Nance JJ, et al. High harvest yield, high expansion, and phenotype stability of CD146 mesenchymal stromal cells from whole primitive human umbilical cord tissue. J Biomed Biotechnol 2009;2009:789526. [13] Takechi K, Kuwabara Y, Mizuno M. Ultrastructural and immunohistochemical studies of Wharton's jelly umbilical cord cells. Placenta 1993;14(2):235e45. [14] Kobayashi K, Kubota T, Aso T. Study on myofibroblast differentiation in the stromal cells of Wharton's jelly: expression and localization of alpha-smooth muscle actin. Early Hum Dev 1998;51(3):223e33. [15] Balci D, Can A. The assessment of cryopreservation conditions for human umbilical cord stroma-derived mesenchymal stem cells towards a potential use for stem cell banking. Curr Stem Cell Res Ther 2013;8(1):60e72.
239
[16] Karahuseyinoglu S, Cinar O, Kilic E, Kara F, Akay GG, Demiralp DO, et al. Biology of stem cells in human umbilical cord stroma: in situ and in vitro surveys. Stem Cells 2007;25(2):319e31. [17] Can A, Balci D. Isolation, culture, and characterization of human umbilical cord stroma-derived mesenchymal stem cells. Methods Mol Biol 2011;698:51e62. [18] Tomasek JJ, Gabbiani G, Hinz B, Chaponnier C, Brown RA. Myofibroblasts and mechano-regulation of connective tissue remodelling. Nat Rev Mol Cell Biol 2002;3(5):349e63. [19] van Roy F. Beyond E-cadherin: roles of other cadherin superfamily members in cancer. Nat Rev Cancer 2014;14(2):121e34. [20] Lamouille S, Xu J, Derynck R. Molecular mechanisms of epithelialmesenchymal transition. Nat Rev Mol Cell Biol 2014;15(3):178e96. [21] Meng X, Ichim TE, Zhong J, Rogers A, Yin Z, Jackson J, et al. Endometrial regenerative cells: a novel stem cell population. J Transl Med 2007;5:57. [22] Majore I, Moretti P, Hass R, Kasper C. Identification of subpopulations in mesenchymal stem cell-like cultures from human umbilical cord. Cell Commun Signal 2009;7:6. [23] Nanaev AK, Kohnen G, Milovanov AP, Domogatsky SP, Kaufmann P. Stromal differentiation and architecture of the human umbilical cord. Placenta 1997;18(1):53e64. [24] Abuzakouk M, Feighery C, O'Farrelly C. Collagenase and dispase enzymes disrupt lymphocyte surface molecules. J Immunol Methods 1996;194(2): 211e6. [25] Autengruber A, Gereke M, Hansen G, Hennig C, Bruder D. Impact of enzymatic tissue disintegration on the level of surface molecule expression and immune cell function. Eur J Microbiol Immunol 2012;2(2):112e20. [26] Petsa A, Gargani S, Felesakis A, Grigoriadis N, Grigoriadis I. Effectiveness of protocol for the isolation of Wharton's jelly stem cells in large-scale applications. In Vitro Cell Dev Biol Anim 2009;45(10):573e6. [27] Magin AS, Korfer NR, Partenheimer H, Lange C, Zander A, Noll T. Primary cells as feeder cells for coculture expansion of human hematopoietic stem cells from umbilical cord bloodea comparative study. Stem Cells Dev 2009;18(1): 173e86. [28] Salehinejad P, Alitheen NB, Ali AM, Omar AR, Mohit M, Janzamin E, et al. Comparison of different methods for the isolation of mesenchymal stem cells from human umbilical cord Wharton's jelly. In Vitro Cell Dev Biol Anim 2012;48(2):75e83. [29] Ishige I, Nagamura-Inoue T, Honda MJ, Harnprasopwat R, Kido M, Sugimoto M, et al. Comparison of mesenchymal stem cells derived from arterial, venous, and Wharton's jelly explants of human umbilical cord. Int J Hematol 2009;90(2):261e9. [30] Garzon I, Alfonso-Rodriguez CA, Martinez-Gomez C, Carriel V, MartinPiedra MA, Fernandez-Valades R, et al. Expression of epithelial markers by human umbilical cord stem cells. A topographical analysis. Placenta 2014. [31] Bozhok Iu M, Bannikov GA, Tavokina LV, Svitkina TM, Troianovskii SM. Local expression of cytokeratins 8, 17 and 18 in the mesenchyme and smooth muscles in the early stages of human organogenesis. Ontogenez 1989;20(3): 250e7. [32] Knapp AC, Franke WW. Spontaneous losses of control of cytokeratin gene expression in transformed, non-epithelial human cells occurring at different levels of regulation. Cell 1989;59(1):67e79. [33] Hazan RB, Qiao R, Keren R, Badano I, Suyama K. Cadherin switch in tumor progression. Ann N Y Acad Sci 2004;1014:155e63. [34] Araki K, Shimura T, Suzuki H, Tsutsumi S, Wada W, Yajima T, et al. E/N-cadherin switch mediates cancer progression via TGF-beta-induced epithelial-tomesenchymal transition in extrahepatic cholangiocarcinoma. Br J Cancer 2011;105(12):1885e93. [35] Bello SM, Millo H, Rajebhosale M, Price SR. Catenin-dependent cadherin function drives divisional segregation of spinal motor neurons. J Neurosci: Off J Soc Neurosci 2012;32(2):490e505.
Contents lists available at ScienceDirect
Placenta journal homepage: www.elsevier.com/locate/placenta
The assessment of the in vivo to in vitro cellular transition of human umbilical cord multipotent stromal cells H. Coskun a, A. Can b, * a
Ankara University Biotechnology Institute, Tandogan, Besevler, 06110 Ankara, Turkey Ankara University School of Medicine, Department of Histology and Embryology, Laboratory for Stem Cells and Reproductive Medicine, Sihhiye, Ankara 06100, Turkey b
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 28 November 2014
Introduction: Human umbilical cord stroma is a rich source of primitive multipotent stromal cells (hUCMSCs). However, the methods for hUC-MSC isolation and propagation remain controversial and vary among laboratories. Our group previously demonstrated that two cell types emerge upon enzymatic isolation of hUC-MSCs, which subsequently undergo a transition towards a fibroblastoid phenotype in later passages. The aim of this study was to further analyse cultured hUC-MSCs by evaluating the cytoskeletal and cell adhesion proteins and by comparing the remodelling of those proteins in umbilical cord sections to determine the cell alterations due to enzymatic and explant methods. Methods: Tissue sections and cultured cells isolated by enzymatic or explant methods were analysed morphologically and by labelling cytokeratin, vimentin, alpha-smooth muscle actin, E-cadherin and Ncadherin profiles. Results: The present observations confirmed that wide, flat cells (type-1) share myofibroblastic features, appear exclusively in enzymatically isolated early cultures; gradually diminish or are replaced by fibroblastoid cells (type-2) in later passages. In contrast, the explant method does not result in the existence of type-1 cells in vitro. Among the tested CK subtypes, CK18 expression is upregulated, whereas CK19 expression is downregulated upon culturing after both protocols. Vimentin and a-SMA, as the major intermediate filaments of hUC-MSCs were found unaltered throughout the culturing period regardless of the cell isolation technique used. Discussion: The data presented confirm and further elucidate the previously observed phenotypic change in hUC-MSCs as illustrated by alterations in structural proteins during enzymatic isolation and subsequent culturing of cells compared with in situ equivalents. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Umbilical cord Mesenchymal stem cell Vimentin Cytokeratin Alpha-smooth muscle actin Cadherin
1. Introduction Over the last decade, human umbilical cord (hUC) matrix, which is generally known as Wharton's jelly, has been the focus of many studies, with special emphasis on the spatial distribution and functional structure of stromal cells, namely, human umbilical cord multipotent stromal cells (hUC-MSCs) and the surrounding extracellular matrix, which serves as a niche for the residing cells [1]. The expression of certain embryonic and mesenchymal stem cell markers in these cells provides suggestive evidence regarding their multipotency [2,3]. Recently, this potential has ultimately led many researchers to initiate the evaluation of these cells in clinical trials,
* Corresponding author. Tel.: þ90 312 5958169; fax: þ90 312 3106370. E-mail address: [email protected] (A. Can). http://dx.doi.org/10.1016/j.placenta.2014.11.024 0143-4004/© 2014 Elsevier Ltd. All rights reserved.
particularly for immunomodulation in inflammatory diseases, such as liver cirrhosis [4,5], systemic lupus erythematous (SLE) [6], and type 1 [7] and type 2 diabetes [8]. While hUC-MSCs still require further characterisation, these cells have been successfully differentiated in vitro into various cell types derived from three germ layers (reviewed by Kim et al., 2013 [9]). Although more than 50 clinical trials are currently underway (www.clinicaltrials.gov), only a few studies [10e12] have specifically focused on the cell surface alterations when cells are either enzymatically extracted from the tissue of origin or mechanically dissected and then induced to migrate from tissue explants in vitro conditions, which would allow cells to subsequently proliferate, stay viable and maintain multipotency for prolonged periods. Tissue resident hUC-MSCs principally resemble naïve mesenchymal fibroblasts found in almost all connective tissues. However, ultrastructural studies have indicated that hUC-MSCs also possess
H. Coskun, A. Can / Placenta 36 (2015) 232e239
some cellular characteristics belonging to myofibroblasts and to smooth muscle cells [11,13,14]. Freshly isolated and plated cells primarily display a fibroblastic appearance over the culture period. However, our group has previously reported that two distinct cell phenotypes (i.e., type-1 and type-2) immediately emerge upon primary monolayer cell culturing, starting from the first passage [1,15,16]. Type-1 cells exhibit flat, wide cytoplasm corresponding to their myofibroblastic nature, whereas type-2 cells appear to have a fibroblastoid phenotype. In general, type-1 cells dominate in early passages, whereas later passages are practically composed of only type-2 cells. Therefore, we hypothesised that a cellular transition or replacement rapidly occurs upon extracting cells from their native environment and continues for a certain period in vitro until cells accommodate solely to the culture conditions. The aims of this study were to determine whether i) cell isolation and early propagation techniques have an effect on the appearance of these two phenotypes and ii) expression of major intracytoplasmic filamentous proteins intrinsically found in hUC-MSCs residing in different umbilical cord subcompartments undergo alterations due to the extraction and subsequent propagation of cells in vitro. For this purpose, we evaluated stromal cells in the adjacent subcompartments of cord sections and in subsequent cultures (P0eP7) regarding their major intracellular cytoskeletal and intercellular adhesion proteins, i.e., cytokeratin (CK) types, vimentin (vim), asmooth muscle actin (a-SMA), F-actin and cadherin family CAMs, such as E-cadherin (E-cad) and N-cadherin (N-cad). The abundance of type-1 cells especially in enzymatically-isolated cells, which gradually diminish in subsequent passages, suggests that cellecell and cell-matrix forces during cell isolation procedures strictly interact with cytoskeletal network and CAMs that ultimately result dramatic changes in cell phenotype. 2. Materials & methods 2.1. Processing of hUCs for microscopy, isolation, and culture of hUC-MSCs Umbilical cords were obtained from full-term infants by Caesarean sections (n ¼ 6). Ethical approval was obtained from the Institutional Ethical Review Board (approval # 18-578-12, 2012). All chemicals and reagents used in this study were purchased from SigmaeAldrich (St. Louis, MO, USA) unless stated otherwise. 15 cmlong cords were transferred to the laboratory in Leibovit's 15 media supplemented with 3% (w/v) penicillin-streptomycin and amphotericin-B (5 mg/mL). A block of cord from each sample was fixed in 4% (w/v) paraformaldehyde (PFA) for 24 h, immersed in 1.2 M sucrose solution as a cryoprotectant, and finally frozen at 60 C for cryosectioning (8 mm thick). For cell isolation, the cord stroma was initially freed from the umbilical vessels and the amniotic membrane. Enzymatic cell isolation or mechanical dissection/explant cell culture techniques were performed on the remaining stromal tissue. The enzymatic method, which was based on the use of 0.1% collagenase type I and 2% dispase, was previously described in detail [17]. The explant method was implemented by plating the mechanically dissected stromal
233
tissue fragments onto polystyrene tissue culture flasks and then allowing these fragments to attach for 24 h, followed by resuspension in culture medium. Following the log phase and confluence in primary culture (P0), hUC-MSCs were subcultured (P1) approximately every 8e10 days in DMEM-Ham's F12 culture media, containing 10% FCS, up to the seventh passage (P7). In addition to culturing the cells in polystyrene flasks, cells were also cultivated on poly L-lysine-coated glass coverslips for immunostaining and high-resolution confocal microscopy. At around 4e5th day of each passage, cells were evaluated for morphological and immunofluorescent protein analyses (see below).
2.2. Immunocytochemistry and microscopic analysis Single- and double-labelling procedures were performed using a series of antibodies against major cytoskeletal proteins and CAMs. Cell monolayers on glass coverslips were fixed with 3.5% PFA at room temperature (RT) for 30 min. Fixed cells were incubated with 1% bovine serum albumin (BSA) in PBS with 0.1% Triton X-100 at RT and were kept in PBS at 4 C until processing. Mouse monoclonal antibodies against pan-CK (CK types 4, 5, 6, 8, 10, 13, 18), CK18, CK19, vimentin, a-SMA, Ecadherin, N-cadherin (Invitrogen, NY, USA) and FITC-phalloidin (35 mg/mL in PBS for 60 min) were applied. Cy3-conjugated goat anti-mouse IgG (Jackson Immunoresearch Laboratories, USA) and FITC-conjugated goat anti-mouse IgG (Jackson Immunoresearch Laboratories, USA) served as secondary antibodies. 7Aminoactinomycine D (7-AAD) (100 mM for 20 min) or Hoechst 33258 (1 mg/mL in 1:1 PBS/glycerol mounting medium) were used for nuclear labelling. Unless otherwise indicated, all antibodies were diluted 1:100 in PBS and incubated for 90 min at 37 C in a humidified chamber. All labelled and stained cells were examined using a Carl Zeiss LSM 510 confocal laser scanning microscope (Jena, Germany), equipped with a Plan-Neofluar 40/1.3 NA oil objective, 488-nm argon ion and 543-nm green helium neon laser lines. The detection parameters, including laser intensity, amplifier offset, amplifier gain, and pinhole diameter were optimised and fixed at the same values for all specimens. Three-dimensional images were reconstructed by consecutive optical sections of various thicknesses (0.25e0.38 mm) using LSM 510 v.3.8 software. Semi-quantitative evaluations of signals were independently performed by the authors. The expression of CK, vimentin and aSMA was defined as relatively diffuse cytoplasmic staining in cord sections and as characteristic cytoplasmic filamentous patterns in cultured cells. Positive cadherin expression was defined as tiny dots and rods in cord sections and as distinctive membranous rims in cultured cells. The varying signal intensities were analysed using ImageJ v.3.91 software. The formula used to calculate the corrected total cell fluorescence (CTCF) was CTCF ¼ Integrated density (area of selected cell mean fluorescence of background readings) (Fig. 1). The CTCF results were converted to ratiometric values to compare signals as fold increases. The percentage of positively stained cultured cells was scored as one positive (10e25% of cells were positively stained), two positive (26%e50% of cells were positively stained), three positive (more than 50% of the cells were positively stained) and four positive (almost all cells were positively stained).
2.3. Statistical analysis The statistical analysis was performed using Statistical Package for Social Sciences v15.0 software (SPSS Inc., USA). Statistical methods included an analysis of variance (ANOVA) for comparing multiple groups. The correlation between two groups was assessed using Pearson's correlation analysis. p < 0.05 was considered statistically significant.
Fig. 1. Quantification of total fluorescence (B) in each raw figure (A) was calculated using a formula named as “corrected total cell fluorescence (CTCF)”. The average of values emitted from three different areas on background fluorescence (C) was used in the formula as mean fluorescence of background readings (shown by three circles).
234
H. Coskun, A. Can / Placenta 36 (2015) 232e239
3. Results 3.1. Phenotypic evaluation of hUC-MSCs in culture Enzymatically dissociated or mechanically chopped cord explants were initially grown in polystyrene T75 culture flasks until reaching 80% confluency (P0). Type-1 cells (Fig. 2A) were frequently encountered among type-2 cells immediately after inoculating the enzymatically dissociated cells (Fig. 2D), persisted in early cultures (P1e3), then gradually decreased (Fig. 2E), and almost disappeared in later passages (Fig. 2F and G). In contrast, only fibroblastoid type2 cells (Fig. 2A) dispersed in explant cultures (Fig. 1H) and persisted in the same form throughout all passages (P1e7) (Fig. 2IeK). Essentially, no cell phenotype difference was noted in later passages between enzymatically and mechanically dissociated cultures (compare Fig. 2F, G with Fig. 2J, K). Consistent with our previous experiments and reports, both group of cells were found positive (>90%) for common MSC cell surface markers, namely, CD73, CD90, and CD105, albeit in varying ratios, and were found negative for CD31 and CD45, as detected by flow cytometry (data not shown).
3.2. Cytoskeletal filament distribution in cord stromal cells Representative tissue compartments from cord sections were designated as amniotic membrane (AM), subamniotic stroma (SAS), intervascular stroma (IVS), and perivascular stroma (PVS), as illustrated in Fig. 3. As previously noted, we confirmed the
myofibroblastic features of UC stromal cells by evaluating the aSMA and vimentin filaments as the most common myofibroblast markers [18]. Second, the expression of pan-CK, and certain CK subtypes (CK18, CK19) (Fig. 3 and Table 1) and F-actin were evaluated. Both cell types displayed well-organised F-actin filaments (Fig. 2B and C). CKs were predominantly found in the AM and in the PVS, whereas lower quantities of CKs were found in the SAS and in the IVS (Fig. 3). As one of the components of the pan-CK antibody cocktail, CK18 alone was faint in the PVS, IVS, and SAS, and slightly higher in the AM (Fig. 3). In contrast, CK19, which was not included in the pan-CK antibody cocktail, was found strongly positive in the AM and SAS, extremely faint in the IVS and completely absent in the PVS (Fig. 3). Vimentin appeared as the most abundant intermediate filament protein in the entire IVS and PVS but not in the AM or SAS. aSMA was detected in all compartments, including the SAS, IVS and PVS, with the exception of AM.
3.3. Cytoskeletal filament distribution in cultured hUC-MSCs To address whether the isolation and/or transfer of cells/explants to culture plates initiates alterations in any of the cytoskeletal networks, we differentially analysed the cultured cells through passages 0e7. Representative confocal images are illustrated in Fig. 4, and semi-quantitative evaluations are provided in Table 1. Pan-CK was strongly positive in both explants and enzymaticallyisolated cells. Only type-1 cells were found positive for CK18. Interestingly, type-2 cells were almost devoid of both CK18 and CK19 (Fig. 4). Throughout the seven passages, all cells were strongly
Fig. 2. Morphological features of cultured hUC-MSCs detected by relief contrast (A, DeK) and confocal microscopy (B and C). Upon isolation of cells by the enzymatic dissociation of cords (Enz), mixtures of two distinct cell phenotypes e type-1 as myofibroblastoid (solid line-bordered cells in A; and arrowheads in DeF) and type-2 as fibroblastoid (dotted linebordered cells in A) were encountered (DeG). In contrast, only one cell phenotype (type-2) was observed in explant cultures (Exp) starting from P0 (H) throughout the entire passages (IeK). Extended F-actin filaments (stress fibres) were detected as a distinctive feature in flat and wide cytoplasm of type-1 cells implying a typical myofibroblast (B). F-actin filaments in type-2 cells appeared to form fibroblast-like organisation (C). Scale bars ¼ 50 mm.
H. Coskun, A. Can / Placenta 36 (2015) 232e239
235
Fig. 3. Cytoskeletal filament distribution (green signals) observed subcompartments of UC stromal cells. PanCK were predominantly found in AM and PVS, lower quantity in SAS and IVS. CK18 was almost absent in SAS, IVS and PVS, slightly positive in AM. In contrast, CK19 was strongly positive in AM and SAS cells, totally absent in IVS and PVS. Vimentin was found dominant in the entire IVS and PVS, but not found AM and SA. aSMA was detected in all compartments including SAS, IVS and PVS with the exception of AM. Red signal ¼ Nucleus (7AAD). Signal intensity was calculated using CTCF formula (see, Fig. 1 and Materials & Methods) and is calculated results were converted to ratiometric values to compare signals as fold increase as illustrated in bar graph. For each group of marker, signals were set relative to the level of lowest signal that were normalised to 1. Note the extensive fold increase in vimentin signal (57.3 fold increase) in PVS. Asterisks refer to significantly different signals (p < 0.05) from the remaining ones within the same group. Scale bars ¼ 50 mm.
236
H. Coskun, A. Can / Placenta 36 (2015) 232e239
Table 1 Abundance of cytoskeletal filaments in cultured hUC-MSCs through passages. P0
P2
Enza PanCK CK18 CK19 Vim aSMA a b c
c
þþþ þþ þþ þþþþ þþþþ
P4
P7
Expb
Enz
Exp
Enz
Exp
Enz
Exp
þþ þþ þ þþþþ þþþþ
þþþþ þþ þ þþþþ þþþþ
þþþ þþ þ þþþþ þþþþ
þþþþ þþþ þ þþþþ þþþ
þþþþ þþþ þ þþþþ þþþþ
þþþ þþþ þ þþþ þþþ
þþþ þþþ þ þþþþ þþþþ
Enzymatic isolation method. Mechanical dissection/explant method. See, Materials & methods for the evaluation protocol.
positive for vimentin and aSMA, regardless of the isolation technique used (Fig. 4). Cells extruded from tissue explants proved that all explants originated from the UC stroma and not from the AM because the explants stained positive for vimentin and aSMA. No significant staining intensity difference per cell was noted for any cytoskeletal element throughout the culturing period. 3.4. Cadherin expression of hUC-MSCs in cord sections and in cultures E-cadherin, which is an epithelial marker, was used to confirm that no epithelial cell contamination existed in the cultures. In the cord sections, AM was positive for E-cadherin, as anticipated,
Fig. 4. Intermediate filaments (CKs and vimentin) and aSMA distribution (green signals) in enzymatic and mechanical dissection/explant cultures. PanCK and CK18 were highly expressed in type-1 cells through P1eP7. CK18 was detected exclusively in type-1 cells; CK19 showed no positivity in any of the cell types, except mitotically active cells (arrowhead). Vimentin and aSMA were strongly expressed in both cell types throughout the entire passages. Red signal ¼ Nucleus (7AAD). Scale bars ¼ 50 mm.
H. Coskun, A. Can / Placenta 36 (2015) 232e239
whereas all stromal cells were entirely negative (Fig. 5). N-cadherin, which is a mesenchymal cell adhesion molecule [19,20], was found expressed in cells that resided in the IVS and in the PVS (Fig. 5). Cultured hUC-MSCs did not express E-cadherin after any of the cell isolation procedures. Interestingly, several membranelocalised N-cadherin molecules were found assembled, particularly in prolonged cultures (12e15 days after plating), and with significantly higher frequencies in explant cultures (Fig. 5). 4. Discussion Present data demonstrate that variations in cytoskeletal protein expressions in isolated hUC-MSCs strictly depend on the cell isolation technique, passage number and the tissue subcompartment from which part the umbilical cord was excised. Human UCMSCs represent an important multipotent stromal cell population that show promising therapeutic effects on clinical applications. Nonetheless, certain discrepancies in the hUC-MSC literature still exist in descriptions of the biological properties of MSCs.
237
Additionally, their overall therapeutic effects may be shadowed in part by the existence of aberrant cell populations or deviations that the tissue-resident cell possess in a given primary culture, including variations in function [21,22]. For instance, one of the most remarkable intermediate filament proteins, desmin, is consistently expressed in all stromal cells in vivo [11,16,23]; however, its expression gradually diminishes when cells are cultured in vitro [11,16,17]. Therefore, one can predict that the hUC-MSC isolation method, particularly the use of proteases to digest the extracellular matrix for enriching the stromal cells, is detrimental to hUC-MSC functions due to the non-specific degradation of certain cell surface receptors and interconnected intracytoplasmic domains [24,25], whereas the explant culture, which allows cells to proliferate and disperse from the existing tissue pieces, avoids such potential artefacts. As shown in the present study, the enzymatic isolation approach significantly induced the existence of myofibroblastic cells called type-1 cells. To offer a more physiological microenvironment while cells are being extracted from a highly rich collagen fibre and hyaluronic acid matrix, several UC-MSC
Fig. 5. E-cadherin and N-cadherin profiles (green signals) in cord stroma (upper row) and cultured hUC-MSCs (lower row). Faint E-cadherin was exclusively detected in AM, no staining was noted in the entire stroma. N-cadherin was mainly localized in IVS and PVS cells, while AM and SAS was totally lacking N-cadherin expression. No E-cadherin was noted in enzymatic or mechanical dissection/explant cultures or in isolated cells. Interestingly, N-cadherin expression was detected strongly positive in prolonged explant cultures (12e15 days) compared to that in enzymatically-isolated cells. Red signal ¼ Nucleus (7AAD). Scale bars ¼ 50 mm.
238
H. Coskun, A. Can / Placenta 36 (2015) 232e239
isolation techniques have been reported so far that considerably vary both in the use enzyme formulations and the mechanical splitting of the stromal tissue. In recent years, if the amniotic membrane and vessels are carefully removed to avoid any other cell contamination, then the explant method has been shown to be superior to the enzymatic method in many aspects for achieving higher cell proliferation rates, purity, and speed, with a lower risk of microbial contamination and, more importantly, clinical compatibility [2,10,12,26e29]. Margossian et al. recently assessed the differences in MSC markers between P0 and P1 cultures and found that CD29, CD44, CD73, CD90, CD105, CD164 and CD166 were significantly elevated at P1 compared with primary cultures (P0). Nevertheless, these researchers did not report any significant change in those markers arising from enzymatic or explant cultures at P1 [10]. More drastically, these researchers demonstrated the absence of CD271, which is one of the most selective markers for MSCs, after enzymatic culture, although CD271-positive IVS cells were found in situ. Similarly, Schugar et al. demonstrated the discordance in CD146, CD44, CD105 and CD73 levels before and after the isolation of cells with different enzyme formulations [12]. Garzon et al. recently demonstrated that hUC-MSCs belonging to different umbilical cord subcompartments and in subsequent cultures express different amount of CK filament subtypes in Ref. [30]. More than 20 different types of CK filaments exist in eukaryotic cells and are primarily categorised in two classes, type I CKs and type II CKs. Cytokeratins are usually found in pairs, which are composed of a type I cytokeratin and a type II cytokeratin in the cytoplasm. Among these cytokeratins, CK8 and CK18 are expressed in human myofibroblastic cells [31]. Several lines of human transformed fibroblasts also express CK 8 and CK18 [32]. In this study, we demonstrated that CK18 and CK19 have extremely low expression or absent in stroma-resident cells. In fact, all CK subtypes were relatively less abundant in the entire stroma compared with AM cells, which are true epithelial cells bearing junctional complexes and basement membranes. The Pan-CK antibody, which contains CK4, 5, 6, 8, 10, 13, and 18 subtypes, stained positive in all stromal cells and the highest staining was observed in the PVS suggesting that this positivity does not primarily originate from CK18 and CK19 but may originate from the expression of CK8 and other CK subtypes. In contrast, CK19 appears to be the dominant type of CK in the AM. Type-1 cells are rich in CK18, whereas type-2 cells, regardless of the isolation technique used, are devoid of CK18, a finding that is similar to that observed in tissue sections. The finding that CK18-positive staining is strongly found in type-1 cells but is lacking in tissue-resident hUC-MSCs supports the hypothesis that these cells are aberrant cell types that temporally express abundant CK18 upon enzymatic isolation. Although both type of cells possessed highly organised F-actin filaments, the typical orientation of these microfilaments demonstrated a clear morphological difference between those two cell types. The observation that both fibroblasts and myofibroblasts possess extensive stress fibres [18] that enable cells to firmly adhere to the substrate suggests that these cells can transdifferentiate into each other. Thus, the entire population of cells observed in cultures throughout all passages can be considered highly plastic and can adjust to the microenvironmental conditions. Because the culture conditions used in this study favour the proliferation of fibroblastoid cell types, the transdifferentiation of myofibroblastic type1 cells into a fibroblastic phenotype is an anticipated phenomenon. If this transdifferentiation is the case, then before this cell conversion, myofibroblasts will express high levels of CK18, which may be considered a differentiation marker of type-1 cells to type-2 cells. The widespread distribution and high intensity of the aSMA protein in type-1 cells also confirms that these cells are giant myofibroblasts that are occasionally encountered in cultures. In contrast, the
intense and widespread vimentin filaments detected before and after isolation clearly indicated that all cultured cells originated from cord stroma, which is an extraembryonic mesenchymal tissue, and not from the amniotic membrane, which originates from epiblast-amnioblast cells. Cadherin family proteins play important roles in cell adhesion, forming adherens junctions to bind cells together within tissues. So far, over 80 types of cadherins in humans have been identified and sequenced. Thus, similar to CKs, certain types of cadherin molecules designate the type or origin of cells [33,34]. Epithelial cells are generally positive for E-cadherin, whereas N-cadherin, which is known as a ‘mesenchymal cadherin’, dominates during the epithelial-to-mesenchymal transition [19]. In the present study, we evaluated the N-cadherin profiles concomitantly in cord sections and cultured cells to examine whether the in situ expression of Ncadherin undergoes a transition due to isolation and to the subsequent culturing conditions. N-cadherin expression was intrinsically expressed in resident cells of the IVS and PVS and in cultured cells after being synthesised in the cytoplasm and assembled in cell membranes. Interestingly, explant cultures provided significantly higher N-cadherin expression, which may be considered a positive sign of functional characteristics of a mesenchymal stem cell. Cells containing a specific cadherin subtype tend to cluster together and to exclude the other cadherin types, both in cell culture and during development [35]. Thus, using an explant culture system seems to provide a more physiological milieu to harvest intact cells compared with enzymatic isolation. 5. Conclusions The following is a summary of our findings: i) Type-1 cells, which were strongly proven as myofibroblasts, appeared mainly in enzymatically isolated early cultures. ii) The dominant fibroblastoid cells were consistently obtained by the explant culturing method. iii) Both procedures did not generally alter the CK distribution upon extraction. However, among the tested subtypes of CKs, CK18 expression was upregulated, whereas CK19 expression was downregulated upon monolayer cell culturing. Vimentin and a-SMA, which are the major intermediate filament proteins of hUC-MSCs particularly localised in the IVS and in the PVS, were found unaltered throughout the culturing period, regardless of the cell isolation technique used. iv) E-cadherin was consistently absent in all stromal cells both in vivo and in vitro, whereas N-cadherin was exclusively noted in the IVS and PVS sections, and noted in later passages with significantly higher expression under explant culture conditions. The data presented in this study represent a typical phenotypic change during isolation and culturing of hUC-MSCs compared with in situ equivalents. Although the possibility of cell contamination cannot be completely excluded during any of the tissue dissociation and subsequent culturing techniques, cells of interest can be purified over time by favouring the most suitable culture media, specific supplements and cell attachment conditions. Therefore, optimal cell isolation techniques and culture conditions should be precisely considered, particularly when preparing clinical-grade hUC-MSCs to be allogeneically transferred back in vivo. Conflict of interest The authors report no conflict of interest. Acknowledgements The authors would like to thank Prof. Fadıl Kara, M.D. who kindly provided umbilical cords and the mothers who were
H. Coskun, A. Can / Placenta 36 (2015) 232e239
involved in sample donation. This work was supported by funding Ankara University Scientific Research Project 13L3330018 (to HC and AC).
References [1] Can A, Karahuseyinoglu S. Concise review: human umbilical cord stroma with regard to the source of fetus-derived stem cells. Stem Cells 2007;25(11): 2886e95. [2] La Rocca G, Anzalone R, Corrao S, Magno F, Loria T, Lo Iacono M, et al. Isolation and characterization of Oct-4þ/HLA-Gþ mesenchymal stem cells from human umbilical cord matrix: differentiation potential and detection of new markers. Histochem Cell Biol 2009;131(2):267e82. [3] Fong CY, Chak LL, Biswas A, Tan JH, Gauthaman K, Chan WK, et al. Human Wharton's jelly stem cells have unique transcriptome profiles compared to human embryonic stem cells and other mesenchymal stem cells. Stem Cell Rev 2011;7(1):1e16. [4] Wang L, Li J, Liu H, Li Y, Fu J, Sun Y, et al. Pilot study of umbilical cord-derived mesenchymal stem cell transfusion in patients with primary biliary cirrhosis. J Gastroenterol Hepatol 2013;28(Suppl. 1):85e92. [5] Zhang Z, Lin H, Shi M, Xu R, Fu J, Lv J, et al. Human umbilical cord mesenchymal stem cells improve liver function and ascites in decompensated liver cirrhosis patients. J Gastroenterol Hepatol 2012;27(Suppl. 2):112e20. [6] Sun L, Wang D, Liang J, Zhang H, Feng X, Wang H, et al. Umbilical cord mesenchymal stem cell transplantation in severe and refractory systemic lupus erythematosus. Arthritis Rheum 2010;62(8):2467e75. [7] Hu J, Yu X, Wang Z, Wang F, Wang L, Gao H, et al. Long term effects of the implantation of Wharton's jelly-derived mesenchymal stem cells from the umbilical cord for newly-onset type 1 diabetes mellitus. Endocr J 2013;60(3): 347e57. [8] Liu X, Zheng P, Wang X, Dai G, Cheng H, Zhang Z, et al. A preliminary evaluation of efficacy and safety of Wharton's jelly mesenchymal stem cell transplantation in patients with type 2 diabetes mellitus. Stem Cell Res Ther 2014;5(2):57. [9] Kim DW, Staples M, Shinozuka K, Pantcheva P, Kang SD, Borlongan CV. Wharton's jelly-derived mesenchymal stem cells: phenotypic characterization and optimizing their therapeutic potential for clinical applications. Int J Mol Sci 2013;14(6):11692e712. [10] Margossian T, Reppel L, Makdissy N, Stoltz JF, Bensoussan D, Huselstein C. Mesenchymal stem cells derived from Wharton's jelly: comparative phenotype analysis between tissue and in vitro expansion. Biomed Mater Eng 2012;22(4):243e54. [11] Ryu YJ, Seol HS, Cho TJ, Kwon TJ, Jang SJ, Cho J. Comparison of the ultrastructural and immunophenotypic characteristics of human umbilical cordderived mesenchymal stromal cells and in situ cells in Wharton's jelly. Ultrastruct Pathol 2013;37(3):196e203. [12] Schugar RC, Chirieleison SM, Wescoe KE, Schmidt BT, Askew Y, Nance JJ, et al. High harvest yield, high expansion, and phenotype stability of CD146 mesenchymal stromal cells from whole primitive human umbilical cord tissue. J Biomed Biotechnol 2009;2009:789526. [13] Takechi K, Kuwabara Y, Mizuno M. Ultrastructural and immunohistochemical studies of Wharton's jelly umbilical cord cells. Placenta 1993;14(2):235e45. [14] Kobayashi K, Kubota T, Aso T. Study on myofibroblast differentiation in the stromal cells of Wharton's jelly: expression and localization of alpha-smooth muscle actin. Early Hum Dev 1998;51(3):223e33. [15] Balci D, Can A. The assessment of cryopreservation conditions for human umbilical cord stroma-derived mesenchymal stem cells towards a potential use for stem cell banking. Curr Stem Cell Res Ther 2013;8(1):60e72.
239
[16] Karahuseyinoglu S, Cinar O, Kilic E, Kara F, Akay GG, Demiralp DO, et al. Biology of stem cells in human umbilical cord stroma: in situ and in vitro surveys. Stem Cells 2007;25(2):319e31. [17] Can A, Balci D. Isolation, culture, and characterization of human umbilical cord stroma-derived mesenchymal stem cells. Methods Mol Biol 2011;698:51e62. [18] Tomasek JJ, Gabbiani G, Hinz B, Chaponnier C, Brown RA. Myofibroblasts and mechano-regulation of connective tissue remodelling. Nat Rev Mol Cell Biol 2002;3(5):349e63. [19] van Roy F. Beyond E-cadherin: roles of other cadherin superfamily members in cancer. Nat Rev Cancer 2014;14(2):121e34. [20] Lamouille S, Xu J, Derynck R. Molecular mechanisms of epithelialmesenchymal transition. Nat Rev Mol Cell Biol 2014;15(3):178e96. [21] Meng X, Ichim TE, Zhong J, Rogers A, Yin Z, Jackson J, et al. Endometrial regenerative cells: a novel stem cell population. J Transl Med 2007;5:57. [22] Majore I, Moretti P, Hass R, Kasper C. Identification of subpopulations in mesenchymal stem cell-like cultures from human umbilical cord. Cell Commun Signal 2009;7:6. [23] Nanaev AK, Kohnen G, Milovanov AP, Domogatsky SP, Kaufmann P. Stromal differentiation and architecture of the human umbilical cord. Placenta 1997;18(1):53e64. [24] Abuzakouk M, Feighery C, O'Farrelly C. Collagenase and dispase enzymes disrupt lymphocyte surface molecules. J Immunol Methods 1996;194(2): 211e6. [25] Autengruber A, Gereke M, Hansen G, Hennig C, Bruder D. Impact of enzymatic tissue disintegration on the level of surface molecule expression and immune cell function. Eur J Microbiol Immunol 2012;2(2):112e20. [26] Petsa A, Gargani S, Felesakis A, Grigoriadis N, Grigoriadis I. Effectiveness of protocol for the isolation of Wharton's jelly stem cells in large-scale applications. In Vitro Cell Dev Biol Anim 2009;45(10):573e6. [27] Magin AS, Korfer NR, Partenheimer H, Lange C, Zander A, Noll T. Primary cells as feeder cells for coculture expansion of human hematopoietic stem cells from umbilical cord bloodea comparative study. Stem Cells Dev 2009;18(1): 173e86. [28] Salehinejad P, Alitheen NB, Ali AM, Omar AR, Mohit M, Janzamin E, et al. Comparison of different methods for the isolation of mesenchymal stem cells from human umbilical cord Wharton's jelly. In Vitro Cell Dev Biol Anim 2012;48(2):75e83. [29] Ishige I, Nagamura-Inoue T, Honda MJ, Harnprasopwat R, Kido M, Sugimoto M, et al. Comparison of mesenchymal stem cells derived from arterial, venous, and Wharton's jelly explants of human umbilical cord. Int J Hematol 2009;90(2):261e9. [30] Garzon I, Alfonso-Rodriguez CA, Martinez-Gomez C, Carriel V, MartinPiedra MA, Fernandez-Valades R, et al. Expression of epithelial markers by human umbilical cord stem cells. A topographical analysis. Placenta 2014. [31] Bozhok Iu M, Bannikov GA, Tavokina LV, Svitkina TM, Troianovskii SM. Local expression of cytokeratins 8, 17 and 18 in the mesenchyme and smooth muscles in the early stages of human organogenesis. Ontogenez 1989;20(3): 250e7. [32] Knapp AC, Franke WW. Spontaneous losses of control of cytokeratin gene expression in transformed, non-epithelial human cells occurring at different levels of regulation. Cell 1989;59(1):67e79. [33] Hazan RB, Qiao R, Keren R, Badano I, Suyama K. Cadherin switch in tumor progression. Ann N Y Acad Sci 2004;1014:155e63. [34] Araki K, Shimura T, Suzuki H, Tsutsumi S, Wada W, Yajima T, et al. E/N-cadherin switch mediates cancer progression via TGF-beta-induced epithelial-tomesenchymal transition in extrahepatic cholangiocarcinoma. Br J Cancer 2011;105(12):1885e93. [35] Bello SM, Millo H, Rajebhosale M, Price SR. Catenin-dependent cadherin function drives divisional segregation of spinal motor neurons. J Neurosci: Off J Soc Neurosci 2012;32(2):490e505.