Introduction
MSCs can be defined as multipotent stem cells which hold the abilities of self-renewal and differentiation into multiple types of cells including osteoblasts, chondrocytes, adipocytes, endothelial cells, myocytes and so on. The isolation of MSCs dates back to 1970s, when Friedenstein isolated fibroblast-like, plastic-adherent cells that could form fibroblastic colonies
in vitro [
1]. Since then, MSCs have received much attention in regenerative medicine due to easy isolation, low immunogenicity, multi-differentiation potential, immunomodulatory properties and lack of ethical controversy.
Studies have shown that MSCs exhibited therapeutic potentials in animal models of acute lung injury [
2], myocardial infarction [
3-
5], acute renal injury [
6], liver injury [
7] and sepsis [
8]. Migration of injected MSCs to the damaged tissues is a key step for these cells to participate in tissue repair. For example, when injected intravenously, MSCs appeared to preferentially home to site of inflammation [
9,
10]. Mechanisms regarding how MSCs migrate into the injured tissue have been a focus of many studies. Some important regulatory factors have been identified, including chemoattractants, growth factors and receptors, and endogenous wound electric fields (EFs). This review aims to present a brief summary about the regulatory factors and intracellular signaling pathways of migration of MSCs . Several discrepancies between studies are also discussed.
Chemokines and receptors
SDF-1–CXCR4 axis
Stromal-derived factor 1(SDF-1), first recognized as a lymphocyte and monocyte specific chemoattractant [
11], belongs to a large family of chemotactic cytokines called “chemokines.” SDF-1 plays a pivotal role in mobilization and homing of stem cells. SDF-1 induced the migration of rat bone marrow MSCs and human umbilical cord blood MSCs in a dose-dependent manner [
12-
14]. MSCs express C-X-C chemokine receptor type 4 (CXCR4), the receptor for SDF-1, and it has been implicated in several studies that SDF-1-induced migration of MSCs was inhibited by the CXCR4-specific antagonist AMD3100 [
14-
17], indicating a key role of SDF-1-CXCR4 axis in migration of MSCs. In an
in vivo study, homing of transplanted MSCs to injured sites in the brain was shown to be partly mediated by SDF-1-CXCR4 interaction [
13].
However, the role of SDF-1-CXCR4 axis in mediating MSC migration was doubted in another study. In an animal model of myocardial infarction, blockade of CXCR4 had little effect on migration of injected bone marrow MSCs into the infracted myocardium [
18]. The precise reason is not clear and may be the varilabe expression level5of CXCR4 on cultured MSCs used by different research groups. Some studies reported that CXCR4 was present at the surface of a small subset of MSCs [
15,
19], and others even showed that CXCR4 was usually absent on the surface of culture-expanded MSCs [
20-
22]. More studies including careful control of MSC culture conditions are needed to clarify this issue.
LPA and LPA1 axis
Lysophosphatidic acid (LPA) is a small bioactive phospholipid, of which the signal is mediated in part, through G protein-coupled receptor, LPA
1. LPA
1 expession was found on human MSCs [
23] and LPA has been shown to play a role in MSC migration. Synovial fluid from rheumatoid arthritis patients greatly stimulated migration of human bone marrow MSCs, which was completely abrogated by pretreatment of the cells with the LPA receptor antagonist Ki16425 [
24]. Similarly, Song
et al. [
25] showed that LPA-stimulated migration of human BMSCs was mediated through an LPA
1-dependent manner. However, using transwell assays, Jaganathan
et al. [
26] have found that culture-expanded human MSCs migrated poorly toward LPA. The underlying molecular mechanism associated with the differential effects of LPA on MSC migration needs further exploration.
Inflammatory factors
Several inflammatory factors have been shown to be involved in MSC migration. Tumor necrosis factor-α (TNF-α) is an important pro-inflammatory cytokine which is present at most sites of injury with inflammation. It has been shown that TNF-α was a potent regulator of MSC migration
in vitro [
27,
28]. Migration assays demonstrated that TNF-α had a chemotactic effect on MSCs, while this effects was inhibited by a specific neutralizing antibody to TNF receptor II [
29].
Macrophage migration inhibitory factor (MIF) is a pro-inflammatory cytokine. Recombinant MIF inhibited
in vitro chemokinesis of MSCs, while the small-molecule MIF antagonist, (S,R)-3-(4-hydroxyphenyl)-4,5-dihydro-5-isoxazole acetic acid methylester restored MSC migration, thus indicating a regulatory role of MIF in MSC migration [
30].
Extracellular high mobility group box 1 (HMGB1) is a cytokine that plays a role in inflammation, tissue injury and regeneration. Using transwell experiments, Meng
et al. [
31] showed that HMGB1 could act as a chemoattractant for MSCs in a dose-dependent manner, although the precise signaling cascades were not clear.
Monocyte chemotactic protein-1 (MCP-1) is a chemokine which is involved in recruitment and activation of macrophages during inflammation.
In vitro analysis suggested that MCP-1 enhanced MSCs migration [
32]. Moreover, it has been revealed in several
in vivo studies that MCP-1 stimulated MSC migration to ischemia in the rat brain [
33], as well as to breast tumor [
34] and gliomas [
35].
Another inflammatory factor interleukin 8 could increase the expression of SDF-1 by MSCs through activation of the protein kinase C (PKC) zeta isoform, while PKC is required for in vitro MSC migration in response to tumor conditioned medium [
36].
Growth factor-receptor axes
Several growth factors, including hepatocyte growth factor (HGF), epidermal growth factor receptor, platelet-derived growth factor, insulin-like growth factor and angiopoietin-1, have been reported to act as chemoattractants for MSCs [
37-
41]. HGF, for example, could strongly attract human MSCs which expressed functional c-met receptors, while the specific c-met blocking agent K-252a significantly inhibited such a chemotactic response, suggesting an important role of HGF-c-met axis in migration of MSCs [
16]. What’s more, research has shown that the expression level of HGF was upregulated at sites of tissue damage [
42,
43], so it is possible that HGF-induced chemoattraction may direct MSCs into the HGF-rich environment of injured sites.
It was found that receptors for several growth factors were found on MSCs, including c-met, epidermal growth factor receptor, platelet-derived growth factor receptor, insulin-like growth factor 1 receptor and Ang-1 receptor [
37]. Thus, multiple growth factor-receptor axes were implicated in the signaling of MSC migration.
Granulocyte colony-stimulating factor (G-CSF), firstly identified as a growth factor which regulates neutrophilic granulocyte proliferation, maturation and functional activation [
44], is a potent stimulator for bone marrow stem cells to migrate out of the bone marrow to sites of injury. Administration of G-CSF in rodent models of ischemic stroke significantly increased the number of stem cells mobilized from the bone marrow to site of ischemic injury and improved stroke-induced neurological deficits [
45,
46]. In another study which used a mouse model of myocardial infarction, infusion of G-CSF increased the level of mobilized bone marrow stem cells up to 250% [
47]. Circulating levels of G-CSF in healthy adults is lower than 30 pg/mL, while this level could reach as high as 2000 pg/mL in conditions of biological stress such as systemic infection [
48], indicating that G-CSF has a significant role in the pathophysiological process. However, it is still not fully understood how G-CSF and the mobilized stem cells exert their respective influence. More studies are needed to clarify the mechanisms.
Endogenous EFs and stem cell migration
Endogenous EFs are instantaneously generated when an injury disrupts an epithelial layer, and could last days to weeks until the wound heals [
49]. In recent years, a series of studies have proved that the endogenous EFs play a very important role in guiding wound healing [
49-
52]. Wound-induced EFs in rat cornea control the healing process of cornea wounds, and pharmacological disruption of these EFs disrupts wound healing [
53].
Several studies have revealed the effect of electric fields on adult stem cells, including MSCs. Cultured murine adipose-derived stromal cells migrated toward the cathode under direct current EFs in a dose-dependent manner [
54]. Besides, EFs could induce orientation of MSCs perpendicular to the EF direction [
55,
56]. Another study showed that EFs could induce directional migration of neural stem cells
in vitro.
Multiple signaling pathways
Extracellualr signals are transduced through intracellular sinaling pathways to induce or modulate migration of MSCs. Multiple signaling pathways are involved in the process, which are discussed below.
PI-3K/AKT signaling pathway
Phosphoinositide 3-kinase (PI-3K)/Akt signaling pathway has been indicated to be involved in migration of multiple types of cells. In response to chemoattractant gradients or electric stimulation, PI-3K is activated and polarized to the leading edge of cells, and cells then polarize and migrate directedly [
57]. Using Transwell Migration System, it has been demonstrated that genetically modified MSCs overexpressing Snail showed stronger migration capacities, while disruption of the PI-3K-dependent pathway using specific PI-3K inhibitor, wortmannin, brought on reduction in Snail-mediated migration of MSCs [
58]. TNF-α mediated migration of MSCs was dependent on the p38 mitogen-activated protein (MAP) kinases (MAPK) and PI3K/Akt signaling pathway [
29]. Similarly, SDF-1α or basic fibroblast growth factor-induced migration of MSCs was attenuated by PI3K/Akt inhibitor LY294002 or wortmannin [
12,
14,
59].
PI-3K signaling pathway plays a central role in electric field-directed cell migration, as is indicated by the fact that directed migration of many cell types were significantly inhibited when treated with PI-3K inhibitor [
49,
60–
63]. Under EFs, human adipose tissue-derived mesenchymal stem cells (ASCs) migrated faster toward the cathode, while PI-3K inhibitor LY294002 decreased the migration speed [
54].
MAPK/ERK1/2 signaling pathway
MAPK signaling pathway plays a role in migration of MSCs. SDF-1-induced cell mobilization required activation of extracellular signal-regulated kinase 1/2 (ERK1/2), one type of MAPK [
62,
63]. TNF-α mediated MSC migration requires p38 MAPK signaling [
29]. Activation of ERK and p38 MAPK was involved in stable thromboxane A2 mimetic U46619-stimulated migration of human MSCs, and the migration process was abrogated with pharmaceutical inhibition of ERK or p38 MAPK [
64]. Li
et al. [
65] have reported that SDF-1α-induced migration of MSCs was significantly attenuated using ERK inhibitor PD98059, or p38 MAPK inhibition with SB203580. Tamama
et al. [
41] found that epidermal growth factor (EGF) caused robust phosphorylation of ERK in rat and immortalized human bone marrow MSCs and stimulated motility of these cells. Taken together, these results suggested that MAPK signaling play an important role in migration of MSCs.
However, several studies have found that inhibition of MAPK signaling showed little effect on migration of MSCs. It was shown that inhibition of P38 or P42/44 MAPK inhibitor using SB203580 or PD98059 respectively had little effect on Snail-mediated human bone marrow MSCs migration [
58]. Kang
et al. [
66] showed that platelet-derived growth factor (PDGF) -induced migration of ASCs was not affected by a pretreatment with MEK inhibitor U0126 and p38 MAPK inhibitor SB202190. More studies are needed to further investigate the role of MAPK signaling in migration of MSCs.
Other signaling pathways
Rho GTPases have been shown to be critical regulators of migration of hematopoietic progenitor cells (HPCs) [
67]. Lee
et al. [
68] have demonstrated that RhoA-Rho signaling was required in LPA induced migration of human ASCs. They performed pull-down assays using Rhotekin, which selectively binds to GTP-bound active form of RhoA, to assess the effects of LPA on RhoA activities and found that LPA increased the level of active form of RhoA in the ASCs. What’s more, inhibition of RhoA with Y27632 significantly attenuated the LPA-induced migration of ASCs, thus suggesting that RhoA-Rho kinase-dependent pathway play an important role in LPA-induced migration of ASCs.
It is well known that Wnt signaling plays a role in the metastasis of many kinds of cancer cells [
69,
70]. Using transwell migration and wound healing assays, Shang
et al. [
71] have shown that Wnt3a promoted the migration capacity of rat MSCs, and migration of MSCs was significantly reduced using Wnt3a antibody added to the Wnt3a-Conditioned Medium (Wnt3a-CM), indicating that Wnt signaling is capable of promoting the migration of MSCs.
Summary
Therapeutic potentials of MSCs have been demonstrated in an increasing number of studies. Migration is a fundamental cellular function that enables MSCs to leave the bone marrow and relocate to distant injured tissues, where they participate in the repair process. Many studies have been focused on the mechanisms of migration of MSCs and some key molecules and signaling pathways have been revealed, as discussed above. A number of chemoattractants and receptors, signaling pathways as well as endogenous EFs. were reported to be important regulators in migration of MSCs, as summarized in Fig. 1.
Despite those previously described molecules, the precise mechanisms of MSC migration are still not clear. To make the issue even more complicated, conflicting results have arisen regarding some important molecules involved in the migration process of MSCs, such as SDF-1/CXCR4 axis and MAPK/ERK pathway. Several reasons may account for the inconsistency between studies, including the heterogenecity and sources of MSCs used in each study [
72]. MSCs used in studies were isolated from different tissues, including bone marrow, umbilical cord, adipose tissue, umbilical cord blood and placenta and so on. The precise differences of migratory capacity between different sources of MSCs have not been fully defined. In addition, both the confluency and passage number of cultured MSCs have been shown to be important factors influencing migration of MSCs [
72-
75].
Several implications for future studies may be drawn from the above context. (1) Special attention should be paid to the source and culture conditions of MSCs used in each study, since heterogeneity between MSCs regarding sources and passage numbers may have significant impact on migration capacity of these cells; (2) Special attention should be paid to interactions between different signaling pathways in migration of MSCs, since the migration process is controlled by a complicated signal network, including chemical, electrical and possibly mechanical regulators.
Understanding the molecular mechanisms of MSC migration will be of great importance in clinical settings. It will help us to optimize the migration or targeting of transplanted MSCs as well as endogenous MSCs toward injured tissues, enabling a greater number of MSCs participating in the healing process, through regulation of the multiple factors involved in the migration process.
Higher Education Press and Springer-Verlag Berlin Heidelberg