Langenbeck's Archives of Surgery

, Volume 396, Issue 4, pp 489–497

Mesenchymal stem cells and progenitor cells in connective tissue engineering and regenerative medicine: is there a future for transplantation?


    • Leibniz Research Laboratories for Biotechnology and Artificial Organs (LEBAO)Hannover Medical School
  • Cornelia Kasper
    • Institut für Technische ChemieLeibniz Universität Hannover
  • Ralf Hass
    • Laboratory of Biochemistry and Tumor Biology, Department of Obstetrics and GynecologyHannover Medical School
  • Axel Haverich
    • Leibniz Research Laboratories for Biotechnology and Artificial Organs (LEBAO)Hannover Medical School
    • Department of Cardiac, Thoracic, Transplantation and Vascular SurgeryHannover Medical School
Review Article

DOI: 10.1007/s00423-011-0762-2

Cite this article as:
Hilfiker, A., Kasper, C., Hass, R. et al. Langenbecks Arch Surg (2011) 396: 489. doi:10.1007/s00423-011-0762-2



Transplantation surgery suffers from a shortage of donor organs worldwide. Cell injection and tissue engineering (TE), thus emerge as alternative therapy options. The purpose of this article is to review the progress of TE technology, focusing on mesenchymal stem cells (MSC) as a cell source for artificial functional tissue.


MSC from many different sources can be minimally invasively harvested: peripheral blood, fat tissue, bone marrow, amniotic fluid, cord blood. In comparison to embryonic stem cells (ESC), there are no ethical concerns; MSC can be extracted from autologous or allogenic tissue and cause an immune modulatory effect by suppressing the graft-versus-host reaction (GvHD). Furthermore, MSC do not develop into teratomas when transplanted, a consequence observed with ESC and iPS cells.


MSC as multipotent cells are capable of differentiating into mesodermal and non-mesodermal lineages. However, further studies must be performed to elucidate the differentiation capacity of MSC from different sources, and to understand the involved pathways and processes. Already, MSC have been successfully applied in clinical trials, e.g., to heal large bone defects, cartilage lesions, spinal cord injuries, cardiovascular diseases, hematological pathologies, osteogenesis imperfecta, and GvHD. A detailed understanding of the behavior and homing of MSC is desirable to enlarge the clinical application spectrum of MSC towards the in vitro generation of functional tissue for implantation, for example, resilient cartilage, contractile myocardial replacement tissue, and bioartificial heart valves.


Tissue engineeringMSCConnective tissueTransplantation


Throughout many centuries, manhood has been attracted by the idea of replacing diseased limbs and organs. However, not until the twentieth century, based on far-reaching discoveries, has organ transplantation became a real therapy option. First, the discovery of the ABO blood system by Karl Landsteiner in 1900; second, the description of HLA by Dausset in the 1950s; third, the discovery of penicillin by Alexander Fleming in 1928 and the triumphal procession of antibiotics into the clinic; and fourth, the discovery of immunosuppressive agents and treatments, i.e., Ciclosporin A, tacrolimus, prednisone, and irradiation.

In 1979, the first report on the clinical use of Ciclosporin A in renal allograft recipients [1] was published representing the beginning of successful transplant surgery. From this point on, patients had a good chance to fight graft rejection and were surviving more than 20 years with transplanted organs.

However, already in the late 1980s, it became clear that organ transplantation would be limited due to the shortage of donor organs relative to an increasing number of people in need, ultimately requiring the development of new strategies. Thus, a new technology emerged, termed “Tissue Engineering”, focusing on the development of functional substitutes for damaged tissue by combining principles of biology and engineering. This technology is based on a general principle (Fig. 1). For the generation of a functional tissue, three cornerstones are needed: (1) A matrix providing a three-dimensional body supporting adhesion, proliferation, and function of appropriate cells; (2) cells, preferentially of autologous origin, providing required tissue specific function at a physiological level; and (3) signals and stimuli mimicking an in vivo-like environment for in vitro differentiation, maturation, and functionalization of the tissue-engineered construct. Since the basic idea of tissue engineering (TE) was born, many reports of successfully generated organ parts were published, for example, biological heart valves [28], cartilage [912], blood vessels [1323], skin [2426], bone [2729], and tendon [30, 31].
Fig. 1

Construction of functional tissue in vitro

Research laboratories focusing on matrices have been investigating the use of either biological matrices or synthetic materials, biodegradable or non-degradable, depending on the application. The choice of appropriate cell sources, in addition to choosing an appropriate matrix, is often very difficult. The cells should be of autologous origin or non-immunogenic, highly proliferative, and either capable of maintaining its differentiation state or capable of being differentiated into the requested cell type, preferentially from a pluripotent state. Since differentiated cells, for example, isolated primary chondrocytes, dedifferentiate after relatively few passages, most researchers concentrate on stem cells as a source for TE. Although in recent reports, induced pluripotent stem cells appear to be a suitable cell source (no ethical concerns), tissue-derived stem cells without the risk of teratoma formation when transplanted, is the focus of TE technology.

Overview of MSC sources

As a natural source in the field of regenerative medicine, mesenchymal stem cells (MSC) could provide a preferred tool for cell replacement approaches or tissue engineering. MSC were first isolated from the bone marrow (BM-MSC) stem cell niche [32]. Thereafter, a variety of different tissues have been characterized as potential sources for MSC (Fig. 2). These tissues include adipose tissue and peripheral blood; however, extensive research over the past years has revealed that cells with morphological and functional characteristics similar to BM-MSC, can be identified in a large number of organs or tissues [33]. Despite having different origins, these MSC populations maintain cell biological properties typically associated with stem cells. These include continuous cell cycle progression for self-renewal and the potential to differentiate into highly specialized cell types of the mesodermal phenotype [34, 35] including chondroblast, osteoblast, and adipocyte lineages [36] (Fig. 3). Interestingly, BM-MSC have also been reported to be inducible via the ectodermal or endodermal germline demonstrating the expression of neuron-like factors [37, 38], insulin production [39], or hepatic lineage-associated genes [40], respectively. In addition to these general stem cell properties, the International Society for Cellular Therapy proposed a more specific panel of markers for the characterization of MSC [36]. Due to the failure to identify a certain unique MSC cell-surface molecule, a set of minimal criteria for MSC was recommended, which includes the capability of adherence to plastic surfaces and the expression of the cell surface markers CD44, CD73, CD90, and CD105 with a concomitant absence of CD14, CD19, CD34, CD45, and HLA-DR expression.
Fig. 2

Predominant sources of human mesenchymal stem cells (MSC)

Based upon these cell biological properties and markers, recent research has identified another source of MSC, originating in birth-associated tissues including human amniotic fluid-derived MSC, placenta-derived MSC, cord blood-derived MSC, and umbilical cord-derived MSC (UC-MSC) [41, 42] (Fig. 2). In particular, the umbilical cord harbors other components as it consists of a squamous epithelium covering two arteries and one vein, surrounded by a mucous connective tissue, rich in hyaluronic acid, also known as Wharton's jelly. Whereas capillaries and lymphatic vessels are not found in umbilical cord tissue [43], respectively, MSC-like cells have been isolated from different compartments of the umbilical cord, i.e., the epithelium [44], subendothelium of the umbilical vein [45, 46], and perivascular region [47], although with various degrees of reproducibility. Therefore, most studies so far in birth-associated tissues have used UC-MSC derived from whole umbilical cord or from Wharton's jelly [4850]. In addition, the umbilical cord tissue appears to be ethically non-controversial, reliable, and an easily accessible source for the enrichment of cell populations with MSC-like properties [43, 5153]. In contrast, bone marrow aspiration for the isolation of BM-MSC represents an invasive procedure, and the portion of BM-MSC in the mononuclear cell fraction is comparably very small. Typically, the amount of BM-MSC varies between 0.001% and 0.01% of the total mononuclear cell population in the bone marrow [34]. Such low frequency of BM-MSC implies an extensive in vitro maintenance and expansion of these stem cells, thus increasing the risk of differentiation induction and epigenetic alterations. Moreover, previous work has demonstrated that the quality of BM-MSC declines with progressive age [54] and in the presence of certain degenerative diseases [55]. Conversely, the umbilical cord tissue enables the acquisition and collection of an exponentially larger yield of MSC suggesting that MSC derived from embryonic or post-natal tissues, such as UC-MSC, may represent a useful alternative with certain advantages as compared to MSC from adult tissues, such as bone marrow or adipose tissue, since UC-MSC provide an easily accessible, more consistent, and richer source of MSC [56]. Moreover, separation of MSC subpopulations of different function and size within the UC-MSC culture using the counterflow centrifugal elutriation (CCE) technique has been successful [57]. In addition, the collection of large MSC with slowed cell cycle progression and enhanced expression of aging markers such as senescence-associated ß-galactosidase, CCE-associated cell separation also resulted in the accumulation of a smaller-sized UC-MSC population. This population maintains a more juvenile, highly proliferative phenotype with a self-renewal capacity along with the presence of various MSC markers. These characteristics may further support the use of this small-sized UC-MSC subpopulation as a valuable tool in regenerative medicine [57].

Differentiation capacity of MSC

Dominici et al. defined the minimal criteria for MSC: differentiation into osteoblasts, chondrocytes, and adipocytes. In this respect, characterization of MSC plasticity, along these pathways has been well established in vitro [36]. The key factors for successful in vitro MSC proliferation and differentiation are the culture conditions, which need to be accurately defined. These conditions are prerequisites for the required development of standardized protocols for MSC differentiation, and their establishment is crucial for subsequent therapeutic application. MSC expansion must conform to GMP guidelines including both serum- and geno-free, well-defined culture conditions [5861]. Currently, a large selection of research grade differentiation media is commercially available.

The differentitation potential of MSC from various tissues (e.g., bone marrow, adipose tissue, umbilical cord blood, periosteum and muscle synovial membranes, Wharton's jelly, dental pulp) has been described in several review papers. Based on these reviews, the most common sources for this purpose are bone marrow (BM) and adipose tissue [53, 56, 6264]. Bone marrow is considered the “gold standard” for the isolation of multipotent MSC, although the harvesting of BM is a highly invasive procedure carrying a potential risk for infection. Moreover, the number, differentiation capacity, and the life span of MSC isolated from BM decline with increasing age [65]. Adipose tissue is an alternative source that can be obtained by a less invasive method and in larger quantities as compared to BM. These cells can be isolated from cosmetic liposuctions or fat grafts in large numbers and can be grown easily under standard tissue culture conditions. Furthermore, the multilineage differentiation capacity of the cells has been confirmed [66].

Chondrogenic differentiation

The chondrogenic differentiation of MSC is typically detected by the formation of cell spheres in pellet culture expressing type II collagen in the extracellular matrix. The medium commonly contains different combinations of bioactive substances such as transforming growth factor (TGF-beta), dexamethasone, fibroblast growth factor, and insulin-like growth factor, bone morphogenetic proteins (BMPs), and ascorbic acid [6769]. The chondrogenic differentiation is commonly detected by measurement of extracellular matrix components such as glycosaminoglycan, immunohistological stainings for collagen II and aggrecan, and the confirmation of typical markers by PCR analyses.

Adipogenic differentiation

The adipogenic differentiation capacity of MSC from different sources has been reviewed by Kuhbier et al. and Gomillon et al. [70, 71]. MSC differentiation into adipocytes involves a number of steps, and the process can be monitored by cell morphological changes, extracellular matrix formation, and the accumulation of intracellular lipid droplets. These droplets are commonly detected by oil red staining. For adipogenic differentiation, the medium is supplemented with dexamethasone, insulin, 3-isobutyl-1-methylxanthine, or transferrin [72].

Osteogenic differentiation

For osteogenic differentiation of MSC in vitro, a confluent monolayer of MSC are most commonly cultured with medium containing dexamethasone, beta-glycerophosphate, and ascorbic acid for several weeks. Furthermore, many studies show positive effects on differentiation with vitamin D, TGF-beta, and BMP supplementation [61, 73]. Expression of alkaline phosphatase and mineralization assayed by von Kossa or alizarin red staining, in addition to the identification of typical osteogenic marker proteins (e.g., bone sialoprotein, osteocalcin, RUNX2) by PCR analyses, confirm osteogenic differentiation.
Fig. 3

Differentiation potential of umbilical cord tissue-derived MSC according to minimal criteria [36]. a Osteogenic differentiation (von Kossa stain). Bone mineralization is represented by dark precipitate. b Adipogenic differentiation (Oil red stain). Accumulated lipid droplets are stained in red. c Chondrogenic (Alcian Blue stain). Extracellular matrix components are stained in green/blue. Scale bars 500 μm (a), 100 μm (b), and 200 μm (c)

Further differentiation

Moreover, a variety of studies have demonstrated that MSC may also generate mature cells typically arising from endoderm [74, 75] and ectoderm [7678]. The plasticity of MSC is documented by a rapidly growing number of studies on selectively differentiated MSC into cell types such as oligodendrocyte progenitor cells [79], dopaminergic neurons [80], hepatocytes [75], and insulin-producing cells [81, 82]. The controversial topic of MSC neural transdifferentiation induction in vitro, has been extensively reviewed elsewhere [83]. The reports on differentiation of MSC into functional cardiomyocytes are inconsistent. Wang et al. reported that MSC derived from Wharton's jelly demonstrate cardiomyocyte morphology and express specific markers (N-cadherin and cardiac troponin) after induction by 5-azacytidine or cardiomyocyte-conditioned medium [84]. However, no clear evidence of cardiomyocyte differentiation could be observed in the study published my Martin-Rendon et al. [85].

MSC in transplantation

MSC can be derived from a large variety of different organs and tissues as discussed above. Multipotent capacities have been described for these cells whereby the originating tissue heterogeneity, harboring an appropriate tissue-specific stem cell niche, may also explain controversial findings reported in the literature. Thus, the different MSC populations suggest distinct cell biological properties (i.e., differences in the proliferative and migratory capacity) including stem cell functions, which need to be addressed in order to consider MSC as a potential therapeutic tool in transplantations, both in vitro and in vivo.

Various studies have demonstrated that MSC in general, represent a useful cell type for replacements or transplantations due to unique properties and functions. Importantly, MSC do not form teratomas after reimplantation in contrast to other stem cell populations, for example, ESC. Moreover, MSC exhibit strong immunosuppressive and immunomodulatory properties, which may play an important role in modifying graft-versus-host reactions during allogenic transplantations [8688]. Indeed, two studies have demonstrated the capability of MSC to attenuate graft-versus-host disease (GvHD). First, Le Blanc et al. successfully transplanted haploidentical MSC in patients with severe treatment-resistant grade IV acute GvHD of the gut and liver. However, the complex mechanisms of immune reconstitution after MSC infusion remain unknown [89]. Second, application of approximately 1 × 106 MSC/kg in a long-term study of patients suffering from steroid-refractory grades III to IV GvHD was successful and revealed a promising therapeutic approach, despite two deaths of unknown etiology [90]. Ultimately, MSC from a single source should be useful beyond autologous transplantations and could provide a therapeutic potential in allogenic transplantations.

To date, MSC have been successfully applied in clinical trials to treat large bone defects [91, 92], cartilage lesions [91, 93], spinal cord injuries [94], cardiovascular diseases [95, 96], hematological pathologies [97, 98], and osteogenesis imperfecta [99]. The therapeutical benefits observed after in vivo administration of MSC are not exclusively restricted to their direct differentiation into cells of appropriate host tissue. MSC can also develop a tropism for tumor cells as demonstrated for brain tumors such as gliomas. Following inoculation of MSC into the contralateral hemisphere of a glioma in rats, the cells began to migrate towards the tumor through the corpus callosum. Moreover, intratumoral injection of MSC resulted in a significant inhibition of tumor growth and increased the survival of these rats [100]. Similar observations were made in the human system where MSC were shown to integrate into human gliomas after intravascular or local injection [101]. Although the molecular basis for this tropism of human MSC for human gliomas is poorly understood, it is hypothesized that these effects are a result of a crosstalk between cytokines, growth factors, and receptors. Malignant glioma cells have been demonstrated to recruit mesenchymal progenitor cells by secreting angiogenic factors, such as VEGF [102], and cytokines, including IL-8 [103]. In addition, matrix metalloproteases and tissue inhibitor of metalloproteases have been associated with MSC migration to injured tissues [104]. In this context, more detailed studies of UC-MSC demonstrated that these cells had a significantly enhanced capacity to migrate towards glioma cells, as compared to BM-MSC due to an elevated expression of IL-8 receptors and CXC chemokine receptors-1 and −2 on UC-MSC [105].

Finally, transplanted MSC may also modulate certain functions of surrounding cells and tissues. Indeed, bioactive substances secreted by the MSC are of major importance in enhancing migration, proliferation, and differentiation of neighboring tissue-resident progenitor cells, diminishing apoptosis, and promoting angiogenesis and/or hematopoiesis [106109].

Summary and outlook

The published data to date, suggest that MSC, as multipotent cells, are capable of differentiating into mesodermal and non-mesodermal lineages. However, existing data, comparing the differentiation potential of MSC, should be carefully interpreted, as further studies must be performed to elucidate the differentiation capacity of MSC from different sources and to understand the involved pathways and processes [110]. In addition, a detailed understanding of the behavior and homing of MSC in vivo is required before MSC can be considered for clinical applications as safe therapeutics [111]. Furthermore, a greater knowledge of MSC biology will also enhance the clinical application spectrum towards the in vitro generation of difficult functional tissue for implantation, for example, resilient cartilage, contractile myocardial replacement tissue, and bioartificial heart valves.

Conflicts of interest


Copyright information

© Springer-Verlag 2011