Annals of Medicine. 2006; 38: 144–153
ORIGINAL ARTICLE
Human mesenchymal stem cells do not differentiate into cardiomyocytes in a cardiac ischemic xenomodel
˚ NSSON-BROBERG2, KATARINA KARL-HENRIK GRINNEMO1, AGNETA MA ¨ RDELL2, ANWAR J. SIDDIQUI2, LEBLANC3,4, MATTHIAS CORBASCIO1, EVA WA 2 2 ´ N & GO ¨ RAN DELLGREN1 XIAOJIN HAO , CHRISTER SYLVE Departments of 1Cardiothoracic Surgery and Anaesthesiology, 2Cardiology, 3Clinical Immunology, 4Center for Allogenic Stem Cell Transplantation, Karolinska University Hospital, Karolinska Institutet, Stockholm, Sweden
Abstract AIM. As the capability of human mesenchymal stem cells (hMSC) to engraft, differentiate and improve myocardial function cannot be studied in humans, exploration was performed in a xenomodel. METHODS. The rats were divided into three groups depending on the type of rats used (Rowett nude (RNU) or Fischer rats +/2 immunosuppression). Different groups were treated with intramyocardial injection of hMSC (1–2 million) either directly or three days after ligation of the left anterior descending artery (LAD). Myocardial function was investigated by echocardiography. The hMSC were identified with fluorescence in situ hybridization and myocardial differentiation was assessed by immunohistochemistry. RESULTS. When hMSC were injected directly after LAD ligation they could be identified in half (8/16) of the RNU rats (without immunosuppression) at 4 weeks. When injected 3 days after LAD ligation in immunosuppressed RNU rats they were identified in all (6/6) rats at 6 weeks. The surviving hMSC showed signs of differentiation into fibroblasts. No cardiomyocyte differentiation was observed. There was no difference in myocardial function in treated animals compared to controls. CONCLUSIONS. The hMSC survived in this xenomodel up to 6 weeks. However, hMSC required implantation into immunoincompetent animals as well as immunosuppression to survive, indicating that these cells are otherwise rejected. Furthermore, these cells did not differentiate into cardiomyocytes nor did they improve heart function in this xenomodel.
Key words: Echocardiography, mesenchymal stem cell, myocardial infarction, xenotransplantation
Introduction A large myocardial infarction results in ventricular remodeling, ventricular dilatation and progressive heart failure (1). Although limited regeneration of cardiomyocytes has recently been reported in human infarcted hearts (2), it is generally acknowledged that this does not inhibit the development of fibrous scar tissue. Current treatment modalities in end-stage heart failure are limited and include medical therapy, mechanical ventricular assist devices and cardiac transplantation. Stem cell therapy may be an alternative treatment of end-stage heart failure in the future. Cell transplantation, including fibroblasts (3), smooth muscle cells (3), skeletal myoblasts
(4,5), fetal and autologous cardiomyocytes (3,6,7) and bone marrow cells (8,9) have been demonstrated not only to engraft but also to improve cardiac function after myocardial infarction in animal models. In addition, embryonic stem cells (ES) may be an alternative cell source and it has been demonstrated that ES can form stable intracardiac grafts (10). However, the anticipated immune response induced by allogeneic ES may limit their potential therapeutic applications. With this in mind, the accessibility and the pluripotent capacity of human mesenchymal stem cells (hMSC) make these cells clinically interesting (11). Furthermore, hMSC are an attractive option not
Correspondence: Go¨ran Dellgren, MD, PhD, Dept. of Cardiothoracic Surgery and Anesthesiology, Karolinska University Hospital, Karolinska Institute, S-171 76 Stockholm, Sweden. Fax: +46 8 321931. E-mail:
[email protected] (Received 30 August 2005; revision accepted 17 October 2005) ISSN 0785-3890 print/ISSN 1365-2060 online # 2006 Taylor & Francis DOI: 10.1080/07853890500422982
Human mesenchymal stem cells and cardiomyocytes
Abbreviations
Key messages
hMSC ES RNU
N
LAD FISH GFP LVEDD FAC MO SSC PBS DAPI FITC IFN-g IL-10 PGE2 NK
human mesenchymal stem cells embryonic stem cells Rowett nude rats: athymic rats with the genotype rnu/rnu left anterior descending artery fluorescent in situ hybridization technique green fluorescent protein left ventricular end diastolic dimension fractional area change magnet-optic disc sodium citrate-sodium chloride buffer phosphate-buffered saline solution 49, 6-diamidino-2-phenylindole fluorescein isothiocyanate interferon-g interleukin-10 prostaglandin E2 natural killer cells
only because they are easily harvested but also because patients can be autologously transplanted with their own stem cells. As the differentiation towards a cardiomyocyte phenotype may differ between hMSC and mesenchymal stem cells of other species that potential has to be studied specifically for hMSC. As such experimental exploration for ethical reasons cannot be done in an autologous model, a xenogenic model has to be used. We have previously shown that the hMSC induce a xenogenic immune response when transplanted into rats (12). In an attempt to circumvent these problems we have now studied engraftment, differentiation and cardiac function of hMSC implanted into infarcted immunocompetent or Tcell deficient immunoincompetent rats with or without immunosuppression.
Material and methods Experimental animals Twenty-one Fischer rats (B&K Universal AB, Sollentuna, Sweden) weighing 400–450 g and forty Rowett nude rats (RNU, genotype rnu/rnu) (Charles River Deutschland Inc, Germany) weighing 250– 300 g were used in this study. The Animal Care Committee at the Karolinska University Hospital approved all procedures. The animals received
N
145
In an ischemic xenomodel, human mesenchymal stem cells required implantation into immunoincompetent animals as well as immunosuppression to survive up to six weeks, indicating that these cells are otherwise rejected. Furthermore, these cells did not differentiate into cardiomyocytes nor improve heart function.
humane care in compliance with the ‘Guide for the Care and Use of Laboratory Animals’ prepared by the Institute of Laboratory Animal Resources, National Research and published by the National Academy Press (13). Stem cell isolation procedure To isolate hMSC, bone marrow aspirates of 5–10 ml were taken from the sternum of patients not older than 65 years of age and undergoing cardiac surgery. hMSC were isolated and cultured through seven passages as previously reported (12,14). By flow cytometry, cultured hMSC were uniformly positive for CD166, CD105, CD44, CD29, SH3 and SH4 and negative for hematopoetic markers CD14, CD34 and CD45. On induction, these cells differentiated into osteoblasts, adipocytes and chondrocytes, as previously described (14). Anesthesia and postoperative care The rats were anaesthetized with a subcutaneous injection of Midazolam (Dormicum, 5 mg/kg) (F. Hoffmann-La Roche Ltd, Switzerland), Medetomidin hydrochloride (Domitor vet, 0.1 mg/kg) (Orion Corp., Espoo, Finland), Fentanyl (0.3 mg/ kg) (B. Braun Medical AG, Seesatz, Switzerland) and subsequently endotracheally intubated. Positive-pressure ventilation was maintained at a rate of 110 cycles per minute with a tidal volume of 1.5 mL with room air supplemented by oxygen (2 L/min) using a Zoovent ventilator (Model CWC600AP, BK Universal, U.K.). The anesthesia was reversed by an intramuscular injection of Flumazenil (Lanexat, 0.1 mg/kg) (F. Hoffmann-La Roche Ltd, Switzerland), Atipamezol hydrochloride (Antisedan vet 5 mg/kg) (Orion Corp., Espoo, Finland) and Buprenorphin hydrochloride (Temgesic, 0,00 4 mg/kg) (Schering-Plough Corp., Kenilworth, UK).
146
K-H. Grinnemo et al.
Figure 1. Outline of the experimental protocols.
Myocardial scar formation and hMSC implantation Under general anesthesia a left lateral thoracotomy was performed. The left anterior descending artery (LAD) was ligated. The hMSC (1–26106) were suspended in 50 mL Perfadex (Vitrolife Sweden AB, Kungsbacka, Sweden) and subsequently injected into the myocardium surrounding the infarcted scar. The control rats were injected with 50 mL Perfadex. The rats were randomly divided into three groups (Figure 1). In group I (direct, RNU, without Tacrolimus); the hMSC were injected directly after LAD ligation in sixteen RNU rats, with six controls. In group II (3 days, RNU, with/without Tacrolimus) and group III (3 days, Fischer, with Tacrolimus);
both groups underwent injection with cells three days after LAD ligation. Eleven RNU rats (five rats in IIa and six rats in IIb), and fourteen Fischer rats (III) were injected with hMSC through a rethoracotomy, with seven control rats in II and seven in III. All animals in IIb (3 days, RNU, with Tacrolimus) and III (3 days, Fischer, with Tacrolimus) were treated with daily intramuscular injections of Tacrolimus (0.1 mg/kg/day) (Fujisawa Healthcare, Inc, Deerfield, IL). The stem cells were injected in a standardized way at several sites and there was no difference in the injection technique when injecting in the acute or subacute phase of the myocardial infarction.
Human mesenchymal stem cells and cardiomyocytes Cardiac function measurements In vivo cardiac function and dimensions were assessed by echocardiography using a Vingmed Vivid 5 (Vingmed A/S, Norway) ultrasound system equipped with a 10 MHz transducer. The animals were sedated with Medetomidin/Midazolam and examined in a supine position and subsequently anesthesia was reversed with Flumazenil/Atipamezol hydrochloride, as described above. The examinations were performed before LAD-ligation, one week after injection of hMSC/Perfadex and after four to six weeks (Figure 1). Parasternal long- and shortaxis views were obtained and stored on a magnetoptic (MO) disc. From the M-mode tracings (guided by 2-D images in the parasternal long axis), measurements of the left ventricular end diastolic dimension (LVEDD) as well as the dimensions of the septum and posterior wall were performed. To assess the contractile function of the left ventricle, the fractional area change (FAC5diastolic area – systolic area)/diastolic area) was calculated by performing planimetry in the parasternal short axis view. The measurements were made at the widest obtainable part of the left ventricle in the long axis view and at the level of the papillary muscles in the short axis view. Detection of transplanted cells and evaluation of differentiation Animals in group I were euthanized at four weeks and group II (3 days, RNU, with/without Tacrolimus) and III (3 days, Fischer, with Tacrolimus) at six weeks. The hearts were fixed in buffered formalin solution for 24 hours, and subsequently embedded in paraffin and sliced into 3 mm thick sections by the use of a microtome (Microm, HM 355S) (Microm Laborgera¨te GmbH, Walldorf, Germany). Human cardiac muscle was used as positive control. The sections were then prepared for hematoxylin-eosin staining to identify the site of injection and thereby the area where to find the implanted cells. The exogenously implanted hMSC were identified with fluorescent in situ hybridization technique (FISH), which previously has been described and demonstrated to specifically detect DNA of human origin (12). In brief the FISH method involves the following steps. The paraffin slides were baked at 60˚C for 2h, the paraffin was removed and the slides were rehydrated. The slides were further treated with citrate buffer, pepsinized followed by treatment with 2x sodium citratesodium chloride buffer (SSC), 1% formaldehyde / 50mM MgCl2/ PBS followed by a dehydration process. Then 3 mL FISH probe cocktail for human
147
DNA (Vysis Inc, Downers Grove, IL), was applied to each section. Simultaneous denaturation of probe and target DNA at 74˚C for 10 minutes and the slides were subsequently incubated overnight, to allow hybridization of probe and target DNA, in a humidified chamber at 39˚C. After repeated washes, the remaining probe molecules were stained with antifade containing 49, 6-diamidino-2-phenylindole (DAPI). Slides were then examined using an Olympus DP10 fluorescent microscope (Olympus BX60, Olympus Optical CO, Ltd, Tokyo, Japan) where the nuclei of the transplanted human cells stained red. Slides with transplanted cells, identified with FISH, were incubated with antibodies directed towards cardiac specific antigens. The sections were rinsed with phosphate-buffered saline solution (PBS) and the non-specific labeling was blocked by incubating the sections in 5% rabbit serum (X0902, Dako, Glostrup, Denmark) for 30 minutes. The slides were incubated overnight at room temperature with mouse monoclonal antibodies directed against cardiomyocyte markers desmin (clone DE-R-11, Dako, Glostrup, Denmark), sarcomeric actin (clone 5C5, Sigma-Aldrich Corporation, St.Louis, MO), myosin (clone NOQ7.5.4D, Sigma-Aldrich Corp., St.Louis, MO), tropomyosin (T9283 clone CH1, Sigma-Aldrich Corp., St.Louis, MO) and connexin 43 (clone CXN-6, Sigma-Aldrich Corp., St.Louis, MO). The sections were then incubated with fluorescence-labeled rabbit anti-mouse antibodies in the blocking solution at the dilution 1:10 fluorescein isothiocyanate [FITC]-labeled, (F0313, Dako, Glostrup, Denmark) and visualized in the fluorescence microscope. Data analysis All histological and echocardiographic analyses were done blinded. All values are presented as mean¡SD. The data collected for treated animals and controls were compared by paired Student’s t test at every time point. Data retrieved over time from echocardiography were tested with unpaired Student’s t test for each group of animals and when there was a significant difference ANOVA for repeated measurements was performed. A Pv0.05 was considered statistically significant.
Results Myocardial infarction model One control rat in group I (direct, RNU, without Tacrolimus) and II (3 days, with/without
148
K-H. Grinnemo et al.
(a)
(b)
Figure 2. A representative photomicrograph of a transected normal heart (a) and a heart one week after LAD ligation (b) in Fischer rats. Notice the large transmural infarction in the antero-lateral wall (arrow) (b). (Hematoxylin and eosin 610).
Tacrolimus), respectively, were euthanized after two days due to signs of severe heart failure. The other rats survived the entire observation period without signs of discomfort.
Figure 3. hMSC identified with FISH staining (arrows) in RNU rats four weeks after LAD-ligation, after being implanted in the acute phase of myocardial infarction (group I; direct, RNU, without Tacrolimus). The hMSC are arranged in a cluster surrounded by fibrous tissue and do not make contact with the host’s cardiomyocytes.
Histological analysis The LAD-ligation resulted in an antero-lateral transmural fibrous scar of all examined rats, visualized by hematoxylin-eosin staining (Figure 2). The implanted hMSC were detected in half (8/16) of the rat hearts in group I (direct, RNU, without Tacrolimus), in all (6/6) of the rats in group IIb but in neither of the rats in groups IIa (3 days, RNU, without Tacrolimus) or III (3 days, Fischer, with Tacrolimus). In group I (direct RNU, without Tacrolimus), the hMSC were injected directly after LAD ligation and there were only a few surviving stem cells, arranged in clusters, encapsulated by dense fibrous tissue (Figure 3). However, it seemed that the hMSC had migrated from the site of injection into the infarction area. In group IIb (3 days, RNU, with Tacrolimus) the hMSC were injected three days after LAD ligation in RNU rats treated with Tacrolimus. In these rat hearts the hMSC had migrated into the infarction area, and they were arranged in a more dispersed manner, separated from the rat cardiomyocytes by dense fibrous tissue (Figure 4). No surviving hMSC were found in the RNU rats of group IIa (3 days, RNU, without Tacrolimus) or in the Fischer rats of group III (3 days, Fischer, with Tacrolimus). All the
Human mesenchymal stem cells and cardiomyocytes
Figure 4. This photomicrograph illustrates FISH staining of hMSC (arrows) six weeks after infarction induction in RNU rats with immunosuppression (group III; 3 days, Fischer, with Tacrolimus). The hMSC seemed to have migrated more freely into the infarction tissue and are surrounded by fibrous tissue.
surviving hMSC stained negative for all the tested cardiomyocyte specific antigens as well as for connexin 43. Cardiac function Echocardiographic data, observed measurements of LVEDD and the calculated FAC, retrieved preoperatively, at 1 week and at 4 to 6 weeks are given in Table I. The LAD ligation resulted in a reproducible decrease in cardiac function with dilatation of the left ventricle in all groups. FAC decreased (0.57¡0.10 versus 0.50¡0.15, P50.001) and LVEDD increased (0.65¡0.08 versus 0.70¡0.08,
149
P50.0005) between baseline and 1 week in the whole material and there were no differences between the groups. However, at 1 week postoperatively FAC was significantly higher for stem cell treated animals in group III (3 days, Fischer, with Tacrolimus) compared to controls (0.54¡0.12 versus 0.38¡0.15, P50.026). This early difference, however, was not evident any longer at 6 weeks prior to sacrifice (0.49¡0.10 versus 0.44¡0.17, P50.45). FAC remained unchanged between 1 week and 4 to 6 weeks in the whole population (0.49¡0.15 versus 0.47¡0.11, P50.2) and regardless whether stem cells had been delivered or not. Indeed, transplantation of hMSC did not seem to reverse the dilatation, assessed by LVEDD, of the left ventricle. In contrast, the left ventricle continued to dilate between 1 and 4 to 6 weeks (0.70 versus 0.78, Pv0.001) with no significant difference between animals that received treatment compared to controls. Discussion Satellite cells, embryonic (ES) and mesenchymal stem cells (MSC) are all interesting alternatives for cellular cardiomyoplasty. Transplantation of skeletal myoblasts into ischemic myocardium, another suggested form of cellular cardiomyoplasty, has shown that these cells do not produce gap junctions, and results in electrical instability of the host myocardium (15). ES cells have a remarkable capacity in vitro to differentiate into cells derived from all three germ layers (16). However, accessibility and ethical concerns limit the use of these cells for cardiomyoplasty. The bone marrow contains MSC, which have a pluripotent capacity and can differentiate into cells
Table I. Echocardiographic data during follow-up
n
LVEDD preop
Group I. (direct, RNU, without Tacrolimus) hMSC 16 0.63¡0.08 Control 5 0.70¡0.12 Group II. a. hMSC (3 days, RNU, without Tacrolimus) b. hMSC (3 days, RNU, with Tacrolimus) Control
LVEDD 1 week
LVEDD 4–6 weeks
FAC preop
FAC 1 week
FAC 4–6 weeks
0.70¡0.08{ 0.70¡0.13{
0.79¡0.09{ 0.78¡0.10{
0.60¡0.10 0.52¡0.03
0.52¡0.20 0.47¡0.09
0.49¡0.10 0.47¡0.11
5
0.59¡0.05{*
0.74¡0.04{*
0.84¡0.03{
0.54¡0.04
0.51¡0.04{
0.47¡0.04{
6
0.68¡0.06
0.76¡0.10
0.88¡0.08
0.49¡0.06
0.45¡0.11
0.41¡0.10
6
0.69¡0.08{
0.76¡0.01
0.83¡0.06
0.47¡0.08
0.52¡0.03
0.46¡0.06
0.67¡0.08{ 0.70¡0.09
0.78¡0.09{ 0.77¡0.08
0.64¡0.07* 0.54¡0.12*{ 0.59¡0.12* 0.37¡0.15*{
Group III. (3 days, Fischer, with Tacrolimus) hMSC 14 0.66¡0.09 Control 7 0.60¡0.07
0.49¡0.10 0.44¡0.17
Echocardiographic data of the left ventricular end diastolic diameter (LVEDD) and fractional area change (FAC) during the follow-up period. hMSC5human mesenchymal stem cells. Significant difference by paired Student’s t test between preop and 1 week is indicated by * and between 1 week and 4–6 weeks is indicated by {. Significant difference by unpaired Student’s t test between treated animals and controls at any time point is indicated by {. A P-valuev0.05 was considered a statistically significant difference.
150
K-H. Grinnemo et al.
of mesenchymal lineage (17). These cells have been suggested for autologous transplantation in patients with heart failure due to ischemic heart disease. Indeed, in an autologous rat model it was demonstrated that these cells can engraft into the infarcted heart, differentiate and improve function (18). Whether these hMSC, harvested from patients with myocardial ischemia, can differentiate into cardiomyocytes or not is undetermined. Therefore, we investigated whether or not CD342 hMSC would engraft, differentiate or improve heart function in an acute cardiac ischemic rat xenogenic model as a human autologous model was not feasible. We could detect engraftment and persistence of transplanted hMSC for more than four weeks in all of the athymic RNU rats treated with Tacrolimus and half of the RNU rats not receiving immunosuppression. No human cells were detected in immunocompetent Fischer rats. The absence of hMSC in immunocompetent Fischer rats corroborates work previously reported by our group (12). Human MSC are not inherently immunostimulatory in vitro and fail to induce proliferation of allogeneic lymphocytes (19,20). We recently reported on fully mismatched allogeneic fetal liver-derived MSC transplanted into an immunocompetent fetus with osteogenesis imperfecta in the third trimester of gestation (21). No immunoreactivity was observed when patient lymphocytes were re-exposed to graft in vitro, indicating that hMSC can be tolerated when transplanted across MSC barriers in humans. Similarly, after intrauterine transplantation of hMSC into sheep, the cells persisted long term and differentiated along multiple mesenchymal lineages (22). The hMSC could be found in multiple tissues including the bone marrow and thymic epithelium. The persistence of these cells is therefore not surprising as they are present during the formation of the immune system and could participate in natural killer (NK) and T cell selection. In addition, the cells were transferred by injection into the peritoneum, which is not a particularly inflammatory environment, and they were transferred at a time of high cell growth. This environment is radically different from a newly infarcted adult myocardium where inflammatory signals are rampant. Furthermore, immunologic tolerance to mouse MSC was reported by Saito et al. in immunocompetent rats allowing engraftment in the injured myocardium (23). In these experiments cells were transplanted intravenously one week before infarction. These results are in direct contrast to the results attained in our study: however, that experiment differs from ours in two fundamental ways.
The transplantation of mouse cells into rats is a concordant xenotransplant and subsequently less immunogenic because similar species will share more common antigens. The intravenous injection of cells is a non-inflammatory event and therefore will not to the same degree lead to the expression of costimulatory factors compared to injection into infarcted myocardium. The lack of inflammatory and costimulatory signals maybe sufficient to allow for the engraftment of MSC into the bone marrow. The MSC cultured by our group are not immunostimulatory in vitro or in vivo in an allogeneic setting (19,21). Instead, the cells are immunosuppressive and reduce lymphocyte proliferation and the formation of cytotoxic T-cells when present in mixed lymphocytes cultures (14,24,25). Furthermore, preliminary results indicate that the cells are immunosuppressive in vivo and may reverse severe steroid resistant acute graft-versus-host disease after allogeneic stem cell transplantation (26). In spite of the cells’ inability to induce alloreactivity, hMSC were rejected in a xenogenic model after infusion into immunocompetent Sprage-Dawley rats after myocardial infarction (12). An immune reaction to the transplanted hMSC was supported both by the histological examination of sectioned hearts, where massive infiltration of immune cells was noted one week after injection, as well as by significant proliferation when murine lymphocytes, derived from animals that had received a MSC injection, were co-cultured with hMSC. This indicates that an immune reaction had occurred against the transplanted human cells. Rejection in a xenogeneic but not allogeneic system may be explained by the fact that although hMSC suppress the formation of CD4 and CD8 T-cells, immune responses following xenotransplantation include both acquired immunity and innate immunity, in which natural antibody, complement, NK cells and macrophages all play independent rolls (27). In RNU rats, engraftment was improved in animals treated with Tacrolimus. The observation of hMSC in immunosuppressed RNU rats but not in untreated RNU rats is disgruntling. This may to some extent be explained by the fact that RNU rats seem to be capable of producing T-cells some weeks after birth. But the most likely explanation is that the cells are in a state of rejection, which is not complete at four weeks but finished after six weeks. Not all rejection needs to be mediated by T cells and therefore there are other components of the immune system, which could be the main effector component responsible for rejection in this human-to-rat strain combination.
Human mesenchymal stem cells and cardiomyocytes The low number of surviving hMSC could then be related to rejection since none of these animals (in group I or IIa) were treated with immunosuppression. In allogeneic studies in vitro, hMSC has been shown to be able to down-modulate immune responses by increasing production of interleukin-10 (IL-10) and decreasing interferon-c (IFN-c) production partially mediated by secretion of prostaglandin E2 (PGE2) (28). Although hMSC clearly are immunomodulatory in vitro in an allogeneic setting, these mechanisms may not be sufficiently robust in a discordant xenogeneic setting. The potential number of reactive T-cell clones is much greater in discordant xenogeneic models and the incompatibility of inhibitory receptors may allow for NK-cell activation not present in the allogeneic setting. The mechanism of the hMSC immunosuppressive capacity is still unknown and may be species specific as these signals may be phylogenitically divergent. Furthermore, cell survival is not only related to immunological factors but also to how these cells are treated in general. Cell survival is heavily influenced by hypoxia, access to nutrients and other unknown physiological factors. The transplanted hMSC were found in large quantities in the immunosuppressed RNU rats where the stem cells were injected three days after LAD ligation. In these rats the hMSC did not differentiate into cardiomyocytes, nor did they improve cardiac function. This means that they could have differentiated into fibroblasts or just remained as undifferentiated MSC, which also have a fibroblast-like appearance. These results contrast with the work by other groups, who have reported extensive cardiac regeneration and improved cardiac function after direct injection of human or animal bone marrow cells into the animal heart (8,9,18). Others have shown that MSC have the ability to home to sites of tissue damage, for instance an infarcted region of the heart. Saito et al. (23) were the first to demonstrate that mouse MSC injected intravenously into infarcted rat myocardium, could be identified in the circulation and subsequently in the infarcted region. The majority of the engrafted cells were found to be positive for cardiomyocyte proteins. The MSC that had differentiated into cardiomyocytes were not only found in the border zone of the myocardial infarct region, but also in the scar tissue. It seems therefore that MSC can survive both in the scar and in the peri-infarct region. In our study, we injected hMSC into the peri-infarct region, however, even if we by mistake would have injected the cells into the infarct region this would then not explain our discrepant results with low or absent engraftment of hMSC, since these cells are
151
capable of surviving in both areas. A small amount of the surviving hMSC was found around the injection site, but the majority of the stem cells had migrated into the peri-infarct and into the infarcted region. The lack of differentiation among implanted hMSC, observed in this study, may be explained by the xenogeneic setting. Differentiation stimuli between species may differ and the hMSC might differentiate into cardiomyocytes in an autologous or allogeneic setting, however, currently it is not known which factors or cell-to-cell contact characteristics are the specific differentiation signals. An alternative explanation for the discrepant results lies in the different assays used to detect cardiomyogenic differentiation. In those studies cited above, the identification of transplanted cells relies exclusively on immunofluorescence staining to track the cells and monitor cell differentiation. In an infarcted heart there is a high level of non-specific autofluorescence and to detect the transplanted cells one must use methods, which produce intense signals that clearly can be separated from background signaling. Our experience from green fluorescent protein (GFP) staining is that these signals can be difficult to distinguish from the background noise, instead, in our experiments; we have used a FISH procedure with high sensitivity and specificity (12). In the FISH method we used a DNA probe that binds to and clearly stains total human DNA, and makes it possible to separate these transplanted human cells from rat cells. Even if these transplanted hMSC did not differentiate to cardiomyocytes, it may be possible that they improve heart function by paracrine function and by influencing the remodeling process. In contrast though, our echocardiographic data regarding the size of the left ventricle as well as the left ventricular function post myocardial infarction do not support the idea that these cells improve the remodeling process over time. In our study, animals treated with hMSC showed no improvement in cardiac function, other than maybe transient in one of the groups. Moreover, cell-treated animals did not differ from control animals in the tendency to develop a dilated left ventricle over time, which indicates no difference between groups regarding the remodeling process. Therefore, the hypothesis that the hMSC may influence the remodeling process must be questioned and needs to be further investigated. Recent studies have shown that hematopoetic stem cells or bone marrow cells, in experimental models similar to ours, do not transdifferentiate towards a cardiomyocyte phenotype (29,30). Our finding that there is engraftment although transient
152
K-H. Grinnemo et al.
of hMSC are well in line with what others have reported for bone marrow or hematopoetic stem cells (30). However, there is an important difference between hMSC and bone marrow cells since the hMSC have been shown to differentiate in vitro towards other kinds of cells, such as cartilage, adipose tissue and muscle (11). In contrast, hematopoetic stem cells or bone marrow are believed to transdifferentiate but it has never been convincingly shown to be the case. In conclusion, the hMSC survived in this xenomodel up to 6 weeks. However, these cells required implantation into immunoincompetent animals with immunosuppression to survive, indicating that these cells are undergoing rejection after being implanted. Furthermore, these cells failed to differentiate into cells similar to myocardial cells nor did they improve heart function, also shown to be the case for bone marrow cells or hematopoetic stem cells (29,30), which clearly emphasizes that further studies are needed to investigate how to improve engraftment as well as differentiation before starting clinical trials.
7.
8.
9.
10.
11.
12.
13. 14.
Acknowledgments The Swedish Medical Research Council (9515), the Swedish Heart and Lung Foundation, the Belve´n Foundation, the Tobias Foundation, the Tore Nilsson Foundation, the Swedish Cancer Society (4562-B03-XAC), the Children’s Cancer Foundation (01/039), the Swedish Research Council (K2003-32XD-14716-01A), the Stockholm Cancer Society, the Swedish Medical Society and the Karolinska Institutet supported this study.
15.
16.
17.
References 18. 1. Braunwald E, Pfeffer MA. Ventricular enlargement and remodeling following acute myocardial infarction: mechanisms and management. Am J Cardiol. 1991;68:1D–6. 2. Beltrami AP, Urbanek K, Kajstura J, Yan SM, Finato N, Bussani R, et al. Evidence that human cardiac myocytes divide after myocardial infarction. N Eng J Med. 2001; 344:1750–7. 3. Sakai T, Li R-K, Weisel RD, Mickle DA, Jia ZQ, Tomita S, et al. Fetal cell transplantation: a comparison of three cell types. J Thorac Cardiovasc Surg. 1999;118:715–25. 4. Taylor DA, Atkins BZ, Hungspreugs P, Jones TR, Reedy MC, Hutcheson KA, et al. Regenerating functional myocardium: Improved performance after skeletal myoblast transplantation. Nat Med. 1998;4:929–33. 5. Menasche P, Hagege AA, Scorsin M, Pouzet B, Desnos M, Duboc D, et al. Myoblast transplantation for heart failure. Lancet. 2001;357:279–80. 6. Scorsin M, Hagege A, Vilquin JT, Fiszman M, Marotte F, Samuel JL, et al. Comparison of the effects of fetal cardiomyocyte and skeletal myoblast transplantation on
19.
20.
21.
22.
postinfarction left ventricular function. J Thorac Cardiovasc Surg. 2000;119:1169–75. Li RK, Weisel RD, Mickle DA, Jia ZQ, Kim EJ, Sakai T, et al. Autologous porcine heart cell transplantation improved heart function after a myocardial infarction. J Thorac Cardiovasc Surg. 2000;119:62–8. Tomita S, Li RK, Weisel RD, Mickle DA, Kim EJ, Sakai T, et al. Autologous transplantation of bone marrow cells improves damaged heart function. Circulation. 1999;100 (suppl II):II247–56. Toma C, Pittenger MF, Cahill KS, Byrne BJ, Kessler PD. Human mesenchymal stem cells differentiate to a cardiomyocyte phenotype in the adult murine heart. Circulation. 2002;105:93–8. Min JY, Yang Y, Sullivan MF, Ke Q, Converso KL, Chen Y, et al. Long-term improvement of cardiac function by transplantation of embryonic stem cell in post-infarcted rats. J Thorac Cardiovasc Surg. 2003;125:361–9. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, et al. Multilineage potential of adult human mesenchymal stem cells. Science. 1999;284:143–7. Grinnemo KH, Ma˚nsson A, Dellgren G, Klingberg D, Wardell E, Drvota V, et al. Xenoreactivity and engraftment of human mesenchymal stem cells transplanted into infarcted rat myocardium. J Thorac Cardiovasc Surg. 2004;127: 1293–300. Guide for the care and use of laboratory animals. Washington [DC]: National Academy Press; 1996. 140 pages. Le Blanc K, Tammik L, Sundberg B, Haynesworth SE, Ringden O. Mesenchymal stem cells inhibit and stimulate mixed lymphocyte cultures and mitogenic responses independently of the major histocompatibility complex. Scand J Immunol. 2003;57:11–20. El Oakley RM, Ooi OC, Bongso A, Yacoub MH. Myocyte transplantation for myocardial repair: a few good cells can mend a broken heart. Ann Thorac Surg. 2001;71: 1724–33. Rohwedel J, Guan K, Hegert C, Wobus AM. Embryonic stem cells as an in vitro model for mutagenicity, cytotoxicity and embryo toxicity studies: present state and future prospects. Toxicol In Vitro. 2001;15:741–53. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, et al. Multilineage potential of adult human mesenchymal stem cells. Science. 1999;284: 143–7. Mangi AA, Noiseux N, Kong D, He H, Rezvani M, Ingwall JS, et al. Mesenchymal stem cells modified with Akt prevent remodeling and restore performance of infarcted hearts. Nat Med. 2003;9:1195–201. Le Blanc K, Tammik C, Rosendahl K, Zetterberg E, Ringden O. HLA expression and immunologic properties of differentiated and undifferentiated mesenchymal stem cells. Exp Hematol. 2003;31:890–6. Klyushnenkova E, Mosca JD, Zernetkina V, Majumdar MK, Beggs KJ, Simonetti DW, et al. T cell responses to allogeneic human mesenchymal stem cells: immunogenicity, tolerance, and suppression. J Biomed Sci. 2005;12: 47–57. Le Blanc K, Gotherstrom C, Ringden O, Hassan M, McMahon R, Horwitz E, et al. Fetal mesenchymal stem-cell engraftment in bone after in utero transplantation in a patient with severe osteogenesis imperfecta. Transplantation. 2005; 79:1607–14. Liechty KW, MacKenzie TC, Shaaban AF, Radu A, Moseley AM, Deans R, et al. Human mesenchymal stem cells engraft and demonstrate site-specific differentiation
Human mesenchymal stem cells and cardiomyocytes
23.
24.
25.
26.
after in utero transplantation in sheep. Nat Med. 2000;6: 1282–6. Saito T, Kuang JQ, Bi ttira B, Al-Khaldi A, Chiu RC. Xenotransplant cardiac chimera: immune tolerance of adult stem cells. Ann Thorac Surg. 2002;74:19–24. Le Blanc K, Rasmusson I, Gotherstrom C, Seidel C, Sundberg B, Sundin M, et al. Mesenchymal stem cells inhibit the expression of IL-2 receptor (CD25) and CD38 on phytohemagglutinin activated lymphocytes. Scand J Immunol. 2004;60:307–15. Rasmusson I, Ringde´n O, Sundberg B, Le Blanc K. Mesenchymal stem cells inhibit lymphocyte proliferation by mitogens and alloantigens by different mechanisms. Exp Cell Res. 2005;305:33–41. Le Blanc K, Rasmusson I, Sundberg B, Gotherstrom C, Hassan M, Uzunel M, et al. Treatment of severe acute
27.
28.
29.
30.
153
graft-versus-host disease with third party haploidentical mesenchymal stem cells. Lancet. 2004;363:1439–41. Zhang Z, Be´dard E, Luo Y, Wang H, Deng S, Kelvin D, et al. Animals models in xenotransplantation (review). Expert Opin Investig Drug. 2000;9:2051–68. Aggarwal S, Pittenger M. Human mesenchymal stem cells modulate allogeneic immune cell responses. Blood. 2005; 105:1815–22. Murry CE, Soonpaa MH, Reinecke H, Nakajima H, Nakajima HO, Tubart M, et al. Haematopoietic stem cells do not transdifferentiate into cardiac myocytes in myocardial infarcts. Nature. 2004;428:664–8. Nygren JM, Jovinge S, Breitbach M, Sawen P, Roll W, Hescheler J, et al. Bone marrow-derived hematopoietic cells generate cardiomyocytes at a low frequency through cell fusion, but not transdifferentiation. Nat Med. 2004;10:494–501.
