Mesenchymal stem cells (MSCs) have been isolated from various tissues such as bone marrow, adipose tissue, placenta, intestine, nervous tissue and heart (1, 2). MSCs are clinically important due to their ability to self-renew, differentiate into a variety of different cell types, and regulate inflammatory and immunomodulatory processes (3–6). The secretome of MSCs, which contains a variety of bioactive substances such as growth factors, cytokines, chemokines and microvesicles determines their paracrine effects which, in turn, are determined and moderated by the cellular milieu (3, 4, 7–10). Many investigations have shown that altering biological, pharmacological, and/or biophysical parameters affects MSC differentiation capabilities and the bioactive factors that they secrete, which potentially increases their therapeutic potential (11, 12). This process of priming, or licensing, of MSCs can be used to up-regulate their in vivo therapeutic functions by increasing their anti-inflammatory and immunomodulatory profiles, angiogenic properties, ability to migrate and prolonging their survival (13–15). Various treatments such as the application of cytokines, growth factors, pharmacological drugs, biomaterials and maintenance under different culture conditions, including hypoxia, have been used to prime MSCs (16–20). Significantly, different priming techniques have different effects in terms of the cytokines, chemokines and growth factors produced by the MSCs, which ultimately affects their role in modulating inflammation and immune responses (21–23).

The most common sources for equine MSCs are bone marrow and adipose tissue, both of which require sedation of the horse and an invasive extraction of tissue. However, a minimally invasive microbiopsy-based method for the isolation of MSCs from equine skeletal muscle, that does not require sedation, has been developed in which the ease of harvesting, and rapid expansion of the cells, combine to make these muscle-derived MSCs (M-MSCs) very appealing for autologous and allogeneic therapies (24). Given that the strongest potential for MSCs as a cell-based therapy lies in their abilities to dampen inflammation and modulate the immune response (25–27), we sought to better understand the anti-inflammatory and immunomodulatory potential of the M-MSCs by examining their responses to the inflammatory cytokines tumor necrosis factor-α (TNF-α) and interferon-γ (IFN-γ), the immunostimulant polyinosinic:polycytidylic acid (poly(I:C)) and by heat-shocking. TNF-α and IFN-γ are inflammatory cytokines produced by activated T cells (28) that have been shown to improve the anti-inflammatory and immunomodulatory functions of MSCs across a number of species, including horse and dog (22, 29–33). Poly(I:C) is an immunostimulant used to simulate infection by double-stranded RNA viruses (34). It binds to the Toll-like receptor-3 (TLR-3) (35) which is expressed by B lymphocytes, dendritic cells and macrophages (36), and when used to prime MSCs from horse and human has been shown to improve their immunomodulatory activity (37–40). Exposing cells to elevated temperatures induces the production of heat-shock proteins which is an evolutionarily highly conserved response to cellular stress (41). In human, dog, and rat, heat-shock proteins also play an important role against inflammatory conditions such as acute lung injury, acute spinal cord trauma and osteoarthritis by inducing the expression of anti-inflammatory cytokines (42–45). Thus, we explored the possibility of increasing the production of anti-inflammatory factors by the M-MSCs via heat-shocking.

All of this research was performed at The University of Queensland (UQ) using established cell lines (provided by co-authors Justine Ceusters and Didier Serteyn) and so UQ Animal Ethics approval was not required.

The microbiopsy procedures that were performed on horses to isolate the M-MSCs that were subsequently donated for this study were approved by the Animal Ethical Commission of the University of Liège.

MSCs are a promising treatment option for immune-mediated and inflammatory diseases, and as a stimulant for tissue regeneration through various mechanisms such as angiogenesis, immunosuppressive and anti-inflammatory actions, and the stimulation of cell migration, proliferation, and differentiation (47, 48). The production of bioactive substances such as growth factors, chemokines, and immunomodulatory compounds by MSCs is primarily responsible for their beneficial effects (49).

In this study, we observed that M-MSCs derived from three horses of mixed sex and two different breeds, but of a similar age, exhibited heterogeneity in their constitutive gene expression profiles, before priming with various techniques. This variability was evident in the production of inflammatory factors, anti-inflammatory factors, growth factors, expression of Toll-like receptors and MHC-I. This finding is important because it highlights the inherent heterogeneity of M-MSCs derived from different horses which has previously been described for bone marrow-derived MSCs (BM-MSCs) (50, 51). The significance of this observation lies in several key aspects including variability in therapeutic potential: for instance, the heterogeneity in the production of inflammatory factors, anti-inflammatory factors, and growth factors suggests that unprimed M-MSCs from different horses may respond differently to inflammatory environments in vivo and consequently have varying therapeutic effects.

MSCs, including those from the horse, are frequently primed to increase their immunomodulatory and anti-inflammatory properties (16, 23, 37, 41, 52–55). However, there are significant translational barriers, especially in the horse, due to the lack of standardized priming techniques and a limited understanding of the responses of MSCs from different tissues, and individuals, to priming (56). This study examined the effects of three different priming methods on M-MSCs – cytokine stimulation, poly(I:C) activation, and heat shock treatment – all of which have previously demonstrated potential in enhancing MSC immunomodulation and anti-inflammatory functions in horses and humans (16, 23, 52–54, 57–59).

In this study, cytokine priming increased the expression of HO-1 (Horse 2) and TSG-6 (all horses) which have been associated with reduced inflammation in rodents and human (59–62), and downregulated the expression of the pro-inflammatory cytokines COX-2 (Horse 1) and IL-8 (Horse 1). The downregulation of COX-2 expression in Horse 1 is in contrast to the other two horses where cytokine priming increased the expression of COX-2, which has also been consistently reported in equine BM-MSCs primed with TNF-α and IFN-γ (16, 52).

Cytokine priming significantly increased the expression of IL-6 in M-MSCs from all three horses, which is consistent with observations in equine BM-MSCs (16, 23, 52, 54). While the role of IL-6 in equine immunomodulation is not fully understood, in contrast to its role in human BM-MSCs where it suppresses the proliferation of stimulated T lymphocytes (63), this does not appear to be the case in equine BM-MSCs (64, 65). In our study, only M-MSCs from Horse 1 expressed IL-6 constitutively, while equine BM-MSCs, adipose tissue-derived MSCs (AT-MSCs) and endometrium-derived MSCs (E-MSCs) all express IL-6 constitutively, albeit at varying levels (66). Constitutive expression of IL-6 by human BM-MSCs stimulates their proliferation (63, 67), inhibits their differentiation along the adipogenic and chondrogenic pathways, prevents them from undergoing apoptosis, and increases the rate of their migration in an in vitro scratch assay (67). This disparity in constitutive IL-6 expression between the M-MSCs from different individuals, and in comparison with equine BM-, AT- and E-MSCs, is therefore intriguing given its potential roles, based on human studies, in regulating MSC proliferation, survival and the maintenance of the undifferentiated state.

We also show that cytokine priming significantly upregulates VEGF and FGF-2 in all three horses, and LOXL-2 in one of the horses (Horse 1). VEGF and FGF-2 play a central role in angiogenesis during tissue regeneration (68, 69), while FGF-2 is also involved in wound healing and tissue repair (70). LOXL-2 encodes an enzyme involved in the development of the extracellular matrix by catalyzing the crosslinking of collagen and elastin fibers (71) and, by extension, is involved in maintaining the structural integrity of the connective tissue proper that forms during wound healing (72). This suggests that cytokine-primed M-MSCs may have enhanced tissue repair potential compared to non-primed M-MSCs.

Toll-like receptors (TLRs) are localized to neutrophils, macrophages, dendritic cells and natural killer cells where they play a key role in activating the innate immune response against viral, bacterial and fungal pathogens (73), and their roles in MSC biology is an area of growing research interest. TLRs 2 and 4 are expressed on the cell surface where they recognize lipids and lipoproteins, and lipopolysaccharides, respectively, in microbial membranes (73). TLRs 3 and 9 are expressed intracellularly; TLR-3 recognizes viral double-stranded RNA, while TLR-9 responds to bacterial and viral DNA (73). Human MSCs from BM, AT and umbilical cord blood (UCB) express high levels of TLR-3 and TLR-4, and low levels of TLR-2 and TLR-9 (74), while MSCs derived from Wharton’s jelly (WJ) lack expression of TLR-4 (75). In response to inflammation, BM- and AT-MSCs upregulate TLR-3 expression, while only the BM-MSCs upregulated TLR4 (75). Taken together, these data demonstrate that for human MSCs, tissue source influences TLR expression profiles and, consequently, MSC responses to inflammation. The data for equine MSCs is less comprehensive, but expression of TLR-3 in equine MSCs after priming with poly(I:C) improves the bactericidal activity of the MSCs (76), and their immunomodulatory ability in decreasing T lymphocyte proliferation (37). In our study, cytokine- and poly(I:C)-primed M-MSCs from all three horses exhibit increased expression of TLR-3. TLR-2 has been implicated in shifting murine BM-MSCs toward a pro-inflammatory phenotype (77). In our study, cytokine priming significantly decreased the expression of TLR-2 in Horses 2 and 3. Expression of TLR-9 by human BM-MSCs induces the production of pro-inflammatory cytokines (78, 79). In our M-MSCs, none of the treatments increased the expression of TLR-9 above constitutive levels in any of the horses which, in combination with the upregulation of the anti-inflammatory TLR-3 in all of the horses, and a downregulation of the pro-inflammatory TLR-2 in horses 2 and 3, supports further investigation into the use of cytokine-primed M-MSCs in managing inflammation in the horse.

Human and veterinary clinical interest in MSCs has been principally directed at the use of allogeneic MSCs as a readily accessible, off-the-shelf supply of MSCs. The advantage of allogeneic cells over the use of autologous cells is that fully characterized cells are available immediately to treat acute injuries or conditions, whereas the collection and expansion in culture of patient-specific autologous cells takes time, during which inflammation associated with the acute injury or condition can cause further tissue damage. Autologous cells may also not be as effective when collected from aged or diseased individuals (80). MSCs from different species are typically considered immune privileged due to their ability to evade immune detection by allogeneic hosts; however, a more accurate description is that they are immune evasive, exerting their immunomodulatory and anti-inflammatory effects before ultimately being cleared by the host’s immune system (81). This ability to evade immune detection has been largely attributed to low expression of MHC-I, and low or absent expression of MHC-II (81). Equine BM-, AT-, peripheral blood (PB) and amnion-derived MSCs constitutively express MHC-I (16, 23, 50–53, 82), as we observed in this study with M-MSCs. Since priming is required to stimulate the immunomodulatory and anti-inflammatory activities of MSCs, we wished to determine if any of the priming methods that we used would induce a change in MHC-I expression. Across the different treatments, the expression of MHC-I remained unchanged from constitutive levels in all three horses. The effects of specifically cytokine priming on the expression of MHC-I by equine BM-MSCs are varied with some studies reporting an increase in MHC-I expression levels and others noting no change; however, this may be due to the concentrations of TNF-α and IFN-γ used, and the duration of priming, in the different studies since Cequier and colleagues (52), and Barrachina et al. (16) both of whom did not see an upregulation of MHC-I expression, primed the BM-MSCs with 5 ng/μl of each of TNF-α and IFN-γ for only 12 h, while an earlier study by Barrachina and colleagues (23), primed with 20 ng/μl of TNF-α and 50 ng/μl IFN-γ for 72 h and did observe an increase in MHC-I expression. We primed the M-MSCs with 50 ng/μl of each of TNF-α and IFN-γ for 72 h and still did not see an increase in MHC-I expression suggesting that in an inflammatory environment in vivo, the M-MSCs are perhaps less likely to upregulate MHC-I expression than BM-MSCs.

We did not detect constitutive expression of MHC-II by M-MSCs from any of the horses in this study, which is consistent with an analysis on AT- and PB-MSCs (50), and some studies on BM-MSCs, where the expression of MHC-II has been found to be absent, present or variable between individual horses (50–52, 82). As with MHC-I expression, the effect of priming with TNF-α and IFN-γ on MHC-II expression by equine BM-MSCs differs between studies, but appears to be independent of the concentrations of TNF-α and IFN-γ used, and the length of exposure: for example, both Barrachina and colleagues (16) and Cequier et al. (52) used 5 ng/μl of each of TNF-α and IFN-γ for 12 h and observed an upregulation, and no change, to MHC-II levels, respectively. Whether equine BM-MSCs, or for that matter MSCs from any tissue source, express MHC-II constitutively, and/or upregulate its expression in response to inflammatory stimuli, is difficult to predict and this is clearly illustrated in the study by Schnabel and colleagues (51) where the proportion of BM-MSCs that constitutively expressed MHC-II for each of 10 thoroughbreds ranged from 0.18%, through 32.42, to 98.35%.

Heat-shocking significantly upregulated the expression of MHC-II to detectable levels in all of the horses. Heat stress induces the production of heat shock proteins (HSPs) that function as molecular chaperones, enabling proteins that have become unfolded or misfolded in response to heat stress to refold into their normal, functional state (83). HSPs also function in antigen processing and in the transfer of antigenic peptides to MHC-I and MHC-II molecules (83) which may go some way to explaining the induced expression of MHC-II in the M-MSCs after heat-shocking.

MSCs exert their immunomodulatory effects, in particular the suppression of T cell proliferation, via paracrine signaling by IDO and iNOS (84–89). MSCs from mouse, rat, rabbit and hamster suppress T cell proliferation via iNOS signaling, while in human, monkey, pig, dog and cow this role is performed by IDO (27). Equine BM-MSCs primed with TNF-α and IFN-γ upregulate their expression of both IDO and iNOS (16, 23, 52, 54). In a study comparing the responses of equine AT-, CT-, BM- and CB-MSCs to mitogen-stimulated T cells producing high levels of TNF-α and IFN-γ, Carrade and colleagues (65) found that only the BM- and CB-MSCs produced nitric oxide, which is synthesized by iNOS from L-arginine, and none of the MSCs produced IDO when exposed to the stimulated T cells (65). In our study, M-MSCs did not express IDO or iNOS constitutively, or after priming, similar to AT- and CT-MSCs that do not produce NO or express IDO when stimulated (64, 65). Carrade Holt and colleagues (64) also determined that MSCs from different tissue sources regulate T cell proliferation via different mechanisms, including apoptosis and cell cycle arrest, that are independent of IDO and iNOS activity, and our M-MSCs appear to similarly activate IDO- and iNOS-independent pathways in response to stimulation with TNF-α and IFN-γ.

In summary, our work highlights a key role for cytokine priming in enhancing the anti-inflammatory, immunomodulatory, and regenerative potential of M-MSCs. However, the heterogeneity between M-MSCs from different individuals poses challenges for standardizing M-MSC-based therapies, as the variability in cell properties could lead to inconsistent treatment outcomes. The main limitation of this study is that it examined M-MSCs from three horses; therefore, the optimisation of therapeutic outcomes will require the characterization of M-MSCs from a variety of individuals to guide the application of appropriate priming strategies to ensure the desired M-MSC secretome for the clinical objective.