Mesenchymal stromal cells: a key player in ‘innate tolerance’?

Abstract

The existence of a mesenchymal stromal cell (MSC) population with the main property of physically supporting parenchymal tissues has long been recognized in virtually all organs. However, it was only recently that MSC have been identified as playing a novel role in modulating inflammation. It has been extensively documented that, under particular circumstances, MSC potently impair virtually all cells of the immune system, including antigen-presenting-cells.

Mesenchymal stromal cell (MSC) -mediated immunosuppression is non-cognate dependent and non-antigen-specific. The effector mechanisms prevalently involve soluble factors that are used by other immunomodulatory populations that are also recruited by the MSC. Mesenchymal stromal cells expand and activate regulatory T cells and interfere with the maturation and function of antigen-presenting cells (APC). The interaction between MSC and haemopoietic stroma is fundamental because MSC depend on the presence of inflammatory molecules produced by monocytes/macrophages to become immunosuppressive. The inflammatory profile to which MSC are exposed determines their immunomodulatory properties, because only in the presence of cytokines like interferon-γ (IFN-γ) or tumour necrosis factor-α (TNF-α) do MSC become immunosuppressive (‘licensing’). Alternative stimulations polarize MSC towards a pro-inflammatory activity. More study of the physiological significance of the immunomodulatory activity is needed to better clarify their key role among the effectors of innate tolerance.

It is not surprising that MSC have generated enormous interest for therapeutic applications. Their properties have been extensively and successfully tested in animal models and in the clinical setting on a variety of autoimmune and alloimmune diseases but the modalities of the therapeutic efficacy remain to be elucidated.

Mesenchymal stromal/stem cells

Although the existence of a population of MSC has long been recognized in many adult tissues, it was only recently that these cells received centre-stage attention. The characterization of MSC within the bone marrow, initially described in the 1960s by Friedenstein et al.,^1,2 paved the way to a number of studies that identified in this population a large proportion of self-renewing progenitors capable of differentiating into adipocytes, osteoblasts and chondrocytes.^3–5 Since then, MSC with similar phenotypes and properties have been isolated from a number of other sources, including cord blood, adipose tissue, muscle and liver.^6–8 These findings led to the use of the acronym MSC to indicate mesenchymal stem cells, irrespective of their source, differentiation stage and function.

In contrast to haemopoietic stem cells, the absence of an in vivo assay for quantifying their stemness/multipotency has hindered the identification of markers that can convincingly distinguish primitive stem cells from progenitors and the even less defined fibroblasts. Human MSC are reported as expressing CD105, CD73, CD90, CD44, CD71 and Stro-1, as well as the adhesion molecules CD106 [vascular cell adhesion molecule 1 (VCAM-1)], CD166, CD54 [intercellular adhesion molecule 1 (ICAM-1)] and CD29, in the absence of any haemopoietic markers.^9–12 The identity of murine MSC has progressed recently. Apart from a set of non lineage-restricted molecules expressed at different degrees (CD29, CD44, CD40e, CD105, CD106, CD73, CD166 and Sca-1),^6,13,14 an MSC phenotype has been identified based on the ability of MSC to differentiate into haemopoietic niche cells after in vivo transplantation. Three main phenotypic profiles have been proposed: PDGFRα⁺ Sca-1⁺ CD45⁻ TER119⁻,¹⁵ the isolated expression of CD146¹⁶ and the expression of nestin.¹⁷ These markers allow us to prospectively isolate a subset of MSC capable of favouring haemopoietic reconstitution after haemopoietic stem cell (HSC) transplantation. In a series of experiments, Mendez-Ferrer et al.¹⁷ showed that, whereas parathormone administration (which increases the numbers of HSC) doubles the number of bone marrow nestin⁺ MSC, the in vivo depletion of the same cell type rapidly reduces HSC content in the bone marrow. In all of these studies, MSC were localized in the peri-vascular region in a quiescent state. The function of MSC in the bone marrow is not limited to regulating self-renewal and differentiation of HSC but is also primarily involved in their homing and mobilization into the peripheral blood both in normal¹⁸ and malignant¹⁹ conditions.

MSC deliver potent immunosuppressive effects

It has been extensively documented that, under particular circumstances, MSC effectively impair T, B and natural killer (NK) cells as well as APC, hence raising enormous interest for their potential therapeutic application.^20–23 The immunosuppressive capacity of MSC on T-cell proliferation has been demonstrated in different experimental conditions irrespective of antigen-specific or mitogenic stimulation. The fact that CD4⁺ and CD8⁺ T cells and naive or memory T cells can be equally immunosuppressed²⁰ indicates that the effect of MSC on T lymphocytes is a non-selective process. The inhibitory effect of MSC on T cells is directed mainly at the cell proliferation stage by targeting the inhibition of cyclin D2, which leads the T cells into cell cycle arrest anergy.²⁴ Not only is the effect non-antigen specific, but it is also cognate-independent because there is no need for MHC identity between MSC and the target immune effector.

The same inhibitory activity has been observed on virtually any cell of the immune system. B lymphocytes do not proliferate nor differentiate into immunoglobulin-producing cells if stimulated in the presence of MSC.²⁴ Studies investigating the relationship between MSC and NK cells provided further insight into the immunomodulatory activity of MSC whereby a two-way regulatory activity interaction seems to take place. Overall, MSCs were shown to inhibit the proliferation, IFN-γ production and cytotoxicity of in vitro interleukin-2 (IL-2) or IL-15-stimulated NK cells. However, some of the cell receptors displayed by NK cells, such as NKp30, NKG2D, CD226 (DNAM-1) and leucocyte function-associated antigen-1 (LFA-1), can bind to molecular ligands expressed by MSC [such as CD155 (PVR), CD112 (Nectin-2) and ICAM-1] and trigger the elimination of MSC themselves. Down-regulating the expression of these ligands on the MSC surface is sufficient to reduce NK-cell activation.^23,25 Recently, Crop et al.,²⁶ reported the lysis of human MSC by NK cells, highlighting the need for better understanding of this interaction ahead of the clinical application of MSC.

The non-specific inhibitory effects of MSC has also been observed on the in vitro differentiation of naive CD4⁺ T cells into T helper type 17 (Th17) cells as well on their production of IL-17, IL-22, IFN-γ and TNF-α.²² Also, the function of T cells expressing T-cell receptor-γδ is impaired by MSC.²¹

Mechanisms of immunosuppression

A number of mechanisms have been implicated in MSC-mediated immunomodulation (Fig. 1). There is now consensus that the secretion of soluble factors is fundamental in MSC activity. Some soluble factors are constitutively secreted by MSC whereas others are induced when MSC are exposed to specific inflammatory environments. It is unlikely that a single molecule is responsible for the effect, because the selective inactivation of only one is not sufficient to turn the immunosuppressive activity off. Furthermore, there are differences among species, at least between mouse and humans. In human MSC one of the most prominent mechanism is the one mediated by indoleamine 2-3-dioxygenase, which depletes the cellular microenvironment of the essential amino acid tryptophan, required for T-cell proliferation.²⁷ In contrast, murine MSC deliver their inhibitory activity especially via inducible nitric oxide synthase (iNOS) while rat MSC use preferentially haem-oxygenase 1. However, other molecules have been clearly demonstrated to be involved and they comprise transforming growth factor-β₁, hepatocyte growth factor, prostaglandin E₂ and soluble HLA-G.^28,29 The most recent report based on gene expression profiling of human MSC, has revealed that galectin-1, highly expressed intracellularly and at the cell surface of MSC, is released in a soluble form and mediates immunosuppression. A stable knockdown of galectin-1 resulted in a significant reduction of the immunomodulatory properties on T cells but not on non-alloreactive NK cells.³⁰ The reasons for such selectivity have not been clarified.

Figure 1.

Mesenchymal stromal cells (MSC) are effectors and regulators of innate tolerance. In the presence of inflammatory stimuli MSC are ‘licensed’ to acquire non-specific immunosuppressive properties that are mediated by a number of partly redundant soluble molecules. MSC also facilitate the recruitment of other immunoregulatory networks that are similarly activated following acute inflammation, amongst these monocytes/macrophages with an anti-inflammatory profile (M2) as well as regulatory T (Treg) cells. The composition of the inflammatory milieu is also fundamental to skew natural killer T (NKT) cells towards an immmunostimulating or anti-inflammatory phenotype. In contrast to NKT cells, MSC and M1-2 cells, Treg cells require the initial activation through the antigen receptor. Ag, antigen; HGF, hepatocyte growth factor; HO-1, haem-oxygenase 1; IDO, indoleamine 2-3-dioxygenase; IFN-γ, interferon-γ; iNKT, invariant NK T cells; iNOS, inducible nitric oxide synthase; PGE₂, prostaglandin E₂; TGF-β, transforming growth factor-β; TLR, Toll-like receptor; TNF-α, tumour necrosis factor-α.

In the presence of an inflammatory environment containing IFN-γ, TNF-α and IL-1β, MSC produce high levels of the chemokines CXCL-9 and CXCL-10 in response to which T cells migrate to the vicinity of MSC, where high levels of iNOS favour the inhibition of T cells. Acting either separately or in combination, pro-inflammatory cytokines drive the up-regulation of ICAM-1, VCAM-1, HLA class I and class II molecules and the inhibitor ligand B7-H1 and these might further potentiate MSC function.³¹

The notion that most effector mechanisms are exerted by the secretion of soluble factors has led to testing the possibility of re-creating an immunomodulatory niche by using MSC-conditioned medium. The earliest report showed that the administration of MSC-derived molecules, either as an intravenous bolus or extracorporeal perfusion with a bioreactor containing MSC, could provide a significant improvement in the survival of rats undergoing fulminant hepatic failure. A protein array screen revealed a large fraction of these molecules to be chemotactic cytokines or chemokines.³² The MSC-conditioned medium therapy resulted in a 90% reduction of apoptotic hepatocellular death and a threefold increment in the number of proliferating hepatocytes with improved animal survival.³³ However, it should be noted that the factors involved in immunosuppression exert their activity in a short-range fashion, making it difficult, if not impossible, to reproduce the same magnitude of activity by injecting MSC-conditioned media. Furthermore, as discussed later, the inflammatory environment is particularly important in shaping the functional profile of MSC and appears to be crucial also for the therapeutic success.

MSC recruit further immune regulatory networks

There are at least two reasons accounting for the potency of MSC-mediated immunosuppression. One is the co-operation/synergism of the various soluble factors identified and described in the previous section. The other aspect, which is gaining support, is that MSC can recruit other immunoregulatory networks. Early in vitro studies in both murine and human MSC have shown that the inhibitory effect is not dependent on CD4⁺ CD25⁺ regulatory T (Treg) cells, because removing Treg cells in culture did not prevent immunosuppression.^20,34 However, it has subsequently been found that MSC can increase regulatory T cells when co-cultured with CD4⁺ cells in vitro.³⁵ Systemic administration of MSC has been observed to protect the airways from allergen-induced pathology by inducing CD4⁺ FoxP3⁺ Treg cells and modulated cell-mediated responses at a local and systemic level, decreasing IL-4 but increasing IL-10 in bronchial fluid and from allergen-stimulated splenocytes. In this experimental system the use of metronomic doses of cyclophosphamide, which reduce Treg-cell responses, reduced the beneficial effect of MSC. Further evidence of Treg-cell activation has been achieved in solid organ transplantation whereby the administration of MSC was observed to favour the differentiation of donor-specific Treg cells.^36–40 In models of autoimmune diseases, MSC effectively prevent the bone and cartilage damage produced by collagen-induced arthritis and such an effect is associated with the in vivo induction of antigen-specific Treg cells.⁴¹ Similarly, human MSC stimulate IL-10-producing T cells and FoxP3⁺ CD4⁺ CD25⁺ T cells, with the capacity to suppress collagen-specific T-cell responses.⁴² Moreover, non-classical CD8⁺ Treg cells have been identified as a result of co-culture of peripheral blood mononuclear cells with MSC.⁴³

The activation of Treg cells may have negative implications in the therapeutic field because of the well-known facilitating effect on tumour escape from immunosurveillance.⁴⁴ It has been observed that MSC inhibit T-cell migration to breast cancer cells and inhibit NK-cell and cytotoxic T-lymphocyte functions.⁴⁵ There is also some suggestion that patients treated with MSC for their graft-versus-host leukaemia have an increased leukaemia relapse rate because of the impairment of graft-versus-leukaemia.⁴⁶

Further pathways mediating immune tolerance can be recruited and activated by MSC and one of the most important is the involvement of monocytes. There is plenty of evidence that MSC inhibit the differentiation of monocytes into dendritic cells and impair their ability to stimulate allogeneic T cells.^47–49 Of particular relevance is the demonstration that monocytes/macrophages are essential for the delivery of MSC-mediated immunosuppression. The modalities of such interaction are several and diverse. The MSC induce dendritic cells to acquire a tolerogenic profile characterized by the up-regulation of IL-10 and the inhibition of TNF-α and IFN-γ production.⁴⁷ Similarly, under particular conditions, MSC skew the inflammatory phenotype of macrophages by converting pro-inflammatory M1-type cells into a more anti-inflammatory M2-type subset.⁵⁰ When MSC are co-cultured with thioglycollate-elicited peritoneal macrophages in the presence of lipopolysaccharide, the production of the pro-inflammatory cytokines IFN-γ, TNF-α, IL-6 and IL-12p70 is markedly suppressed whereas the production of both IL-12p40 and the anti-inflammatory cytokine IL-10 is increased.⁵¹ A key role in the inflammatory switch is played by prostaglandin E₂ because cyclo-oxygenase-2 inhibitors negatively affect such MSC function. The effect of MSC on macrophages was confirmed in vivo in at least two experimental systems. In one case, MSC rendered macrophages highly susceptible to infection with Trypanosoma cruzi, increasing more than fivefold the rate of intracellular infection.⁵² In another model, the beneficial effect of MSC on sepsis was associated with the recruitment of IL-10-producing macrophages.⁵⁰

MSC have been shown to recruit macrophages/monocytes and endothelial lineage cells into the inflammation site by releasing paracrine factors⁵³ and to inhibit the migration of neutrophils by modulating macrophage cytokine release.⁵⁰

The activity of MSC on monocytes/macrophages appears to be a fundamental component in MSC-mediated immunosuppression. It was initially observed that suppression of in vitro mitogen-induced T-cell proliferation by human MSC was profoundly impaired after the removal of monocytes in culture.⁵⁴ The prominent role of macrophages was similarly observed in vitro whereby macrophage depletion or pre-treatment with antibodies specific for IL-10 or IL-10 receptor reduced the therapeutic action on sepsis.⁵⁰ Macrophage polarization might account also for the tissue repair activity of MSC in a number of various conditions. In fact, it is well known that modulation of macrophages favours the conditions for a reparative state. Approaches similar to the infusion of MSC have recently been tested to promote the resolution of inflammation and elicit myocardial infarct repair. Following in vivo uptake of phosphatidylserine-presenting liposomes by macrophages, the cells secreted high levels of anti-inflammatory cytokines and prevented ventricular dilatation and remodelling.⁵⁵

Monocytes/macrophages are not exclusively a crucial effector arm among MSC weaponry but they play a decisive role in enabling MSC to acquire their immunosuppressive properties. The concept of MSC ‘licensing’ will be explained in the next section.

Finally, the effects of MSC have also been investigated on invariant NK T cells. Invariant NKT cells represent another small subset of T cells with regulatory function and characterized by the expression of an invariant T-cell receptor-α chain (Vα14Jα18) which recognizes a non-polymorphic MHC class I-like antigen-presenting molecule (CD1d). The NKT cells can produce both Th1-type and Th2-type cytokines and have been shown to control autoimmune, allergic and anti-tumour immune responses, as well as those against infectious agents. Prigione et al.²¹ showed that human MSC inhibit invariant NKT expansion in vitro. This inhibition can significantly be counteracted by inhibiting prostaglandin E₂ synthesis. The information provided by this study is very limited however, because although MSC can inhibit the proliferation of virtually any cell type, the effects on their functions differ and understanding the activity is especially important in the case of NKT which, like monocytes/macrophages, can be alternatively activated towards a pro-inflammatory or anti-inflammatory profile.

MSC ‘licensing’

It is now clear that the surrounding environment has a vital effect on MSC immunosuppressive activity. Mesenchymal stromal cells are not constitutively inhibitory, but they acquire their immunosuppressive functions after being exposed to specific inflammatory milieux. This important principle stemmed from the observation that neutralizing antibodies against IFN-γ can revert the suppressive effect of MSC in vitro.⁵⁶ Therefore, a ‘licensing’ step is fundamental to induce MSC-mediated immunosuppression. The role of IFN-γ is more complex than just being an activating agent because its levels and the contemporary presence of other cytokines can affect the functional profile of MSC differently. Paradoxically, IFN-γ can enable MSC to act as APC^57,58 and stimulate the generation of antigen-specific cytotoxic CD8⁺ T cells in vivo. However, the acquisition of antigen-presenting properties occurs at low levels of IFN-γ, and as soon as they increase, MSC become immunosuppressive. It should be noted that the physiological relevance of MSC as APC is unclear and many of the studies remain observational and sometimes biased by the lack of proper controls.

Further inflammatory cytokines, such as TNF-α or IL-1β, take part in licensing MSC immunosuppression⁵⁹ and in different combinations can produce different effects. Although IFN-γ alone is sufficient to induce indoleamine 2-3-dioxygenase, in combination with TNF-α, it induces hepatocyte growth factor production and the two cytokines act synergistically in the induction of cyclo-oxygenase-2.³¹

To make things more complicated, not all inflammatory milieux produce the same outcome. Some studies have indicated an important role for toll-like receptors (TLR), membrane-spanning, non-catalytic receptors that recognize structurally conserved molecules derived from microbes and mediate the activation of immune responses of both innate and adaptive types. Mesenchymal stromal cells express a large number of TLR, the stimulation of which has been shown to profoundly affect MSC immunomodulatory properties as well as their migratory phenotype. It is well established that MSC express a number of TLR at both RNA and protein levels. High mRNA expression of TLR1, TLR2, TLR3, TLR4, TLR5 and TLR6 has been consistently detected, whereas TLR2, TLR3, TLR4, TLR7 and TLR9 expression has been reported by flow cytometry. Unfortunately, there is little consensus about the pattern of their expression in MSC with the major confounding factors being the heterogeneity of MSC preparations and the modality of TLR detection.

The expression of TLR on MSC has also been functionally assessed. Although TLR3 and TLR4 binding antagonises MSC immunosuppressive activity,⁶⁰ the stimulation of the same receptor on isolated MSC before their use in culture boosts MSC immunosuppressive activity.⁶¹ It has been shown that TLR3 and TLR4 activation induces the production of pro-inflammatory mediators, such as IL-1, IL-6, IL-8 and CCL5 together with the expression of iNOS, and TNF-related apoptosis-inducing ligand (TRAIL).⁶² It should be noted however that the time of exposure to TLR ligands and the concomitant presence of other cytokines are likely to add layers of complexity. Following low-level, short-term TLR-priming, Waterman⁶³ observed opposing effects of TLR3 or TLR4 stimulation. Last, targeting TLR2 results in the up-regulation of galectin-3, known to modulate T-cell proliferation.⁶⁴

MSC ‘licensing’ and its impact on therapeutics

The possibility of alternating the immunomodulatory properties of MSC depending on the inflammatory environment to which they are exposed has profound implications on how to harness their therapeutic potentials. The studies conducted to investigate MSC therapeutics in graft-versus-host disease (GvHD) are fully consistent with the biological features so far identified. Irrespective of the animal models used, MSC are effective at treating GvHD only when administered at fairly specific intervals.^65–67 At these time-points, the levels of inflammatory cytokines like IFN-γ are particularly high and therefore more likely to promote MSC immunosuppressive activity.

The clinical studies are fully in accord with these data. Although the administration of MSC at the time of HSC transplant did not change the frequency of acute or chronic GvHD,⁶⁸ their administration at the time of full-blown disease resulted in the control of the disease in a large proportion of cases.⁶⁹ Such a concept should also be instrumental in identifying which inflammatory disease could be more amenable to be treated by MSC.

The cytokine environments of acute and chronic inflammation are so different that it would be naive to expect that the administration of MSC produced only beneficial consequences. Our data on the use of MSC in an animal model of inflammatory arthritis indicate that, although MSC are extremely effective at ameliorating an acute form of collagen-induced arthritis, they can expedite disease onset and progression of the chronic form (Williams R and Dazzi F, unpublished data). Similarly, in a preliminary cohort of 32 patients with acute and chronic GvHD, we have observed that the response rate to MSC infusion varies widely between the two groups (56% in acute versus 3% in chronic GvHD) (Innes A and Dazzi F, unpublished data).

Stroma as ‘innate’ modulator of immune responses

Once the integrity of a tissue is disturbed, either by extrinsic or intrinsic elements, the tissue reacts with the initiation of an inflammatory process aiming to regain the tissue homeostasis. Immunocompetent cells like macrophages and dendritic cells have conventionally fulfilled the role as the tissue sentinels activated through TLR molecules.⁷⁰ It is becoming clear that, besides these ‘conventionally immunocompetent cells’, MSC also participate in this ‘innate tissue surveillance’ process. The notion that MSC can be polarized into opposing inflammatory modulators makes them a further key player in stromal physiology. In fact, stromal cells with properties similar, if not identical, to the ‘conventional’ MSC have been identified in virtually every tissue where they are often referred to as ‘fibroblasts’.⁷¹ Despite the attempts delivered by scientific societies to define MSC according to arbitrarily created consensus platforms, it is becoming clear that the operational definitions based on phenotypic markers, immunosuppressive functions and differentiation potential fail to distinguish a specific entity or, alternatively, they validate the idea that all stromal cells of mesenchymal origin are MSC.^72,73

If we accept that a stromal cell network exists and regulates immune reponses in every tissue, the physiological significance of the data that we summarized in this review becomes more meaningful. There is also an impressive parallel in terms of functions and recruitment modalities with stromal cells of haemopoietic origin, i.e. macrophages/monocytes. Although in a simplified approach, it has been established that stimulation of monocytes with specific cytokines or TLR agonists polarizes them into a classical M1 pro-inflammatory phenotype, whereas others promote their alternative M2 phenotype associated with anti-inflammatory and tissue repair activity.⁷⁴ Furthermore, the delivery of immunosuppression is, like MSC, non-cognate dependent and non-antigen specific. Therefore, one could propose the presence of a cellular compartment, triggered and modulated by non-polymorphic stimuli controlling immune responses in an antigen-independent fashion. The modalities of this tolerance induction might be considered as mirroring innate immunity and so be described as ‘innate tolerance’.

CD1d-restricted immune responses should also be considered within such a group of tolerance effectors. CD1d is a non-classical major histocompatibility class 1-like molecule that primarily presents either microbial or endogenous glycolipid antigens to T cells involved in innate immunity. CD1d-restricted T cells comprise NKT cells and a subpopulation of γδ T cells expressing the Vγ4 T-cell receptor. In particular, activated NKT cells secrete large quantities of cytokines that both help control infection and modulate the developing adaptive immune response. However, NKT cells can also promote Treg-cell activation⁷⁵ and the chronic in vivo stimulation of NKT often leads to a Th2 bias in the immune response and promotes the generation of tolerogenic dendritic cells. Furthermore, with similar modalities to MSC and macrophages, reagents have been identified that, by interacting with CD1d, differently bias Th-cell responses.⁷⁶

One of the best examples in which effectors of such ‘innate tolerance’ are actively recruited is cancer. Tumour cells evade immune system recognition not only by mutating antigenic epitopes initially recognized by host immune surveillance, but also and especially by creating an environment that is extremely potent at inhibiting immune responses in a non-specific fashion. Fibroblasts⁷⁷ and immunosuppressive myelomonocytic cells⁷⁸ heavily infiltrate the tumour process and facilitate the activation of ‘adaptive tolerance’ effectors like Treg cells.⁴⁵ Within this context, it is plausible to surmise a major role of MSC because of their ability to polarize and activate immunosuppressive networks as summarized in this review. This hypothesis gains support also by a recent set of data elegantly generated using a transgenic mouse in which stromal cells could be depleted. The depletion of cells expressing fibroblast activation protein-α caused rapid hypoxic necrosis of both cancer and stromal cells in immunogenic tumours by a process involving IFN-γ and TNF-α.⁷⁹

Mesenchymal stromal cells can also contribute to the tumour-related immune impairment because they produce TGF-β, which can suppress or alter the activation, maturation and differentiation of both innate and adaptive immune cells.⁸⁰ In addition, TGF-β has an important role in the differentiation and induction of Treg cells. Furthermore, in the presence of IL-6, also produced by MSC, TGF-β induces the differentiation of IL-17-producing CD4⁺ Th17 cells, which may have tumour-promoting activities.⁸¹

Concluding remarks

An interesting proposal for a ‘tissue-based’ approach to the regulation of the immune response has been recently put forward by Matzinger and Kamala.⁸² It is proposed that ‘the class of an immune response is primarily tailored to fit the tissue in which it occurs rather than to fit the damaging cause’. This could also suggest that specific tissues use their intrinsic physiological properties as a starting point to establish control over an ongoing local immune responses aiming ultimately, to restore the balance of tissue function. Within the immune system there are many cells with regulatory function, aiming to keep the immune response under a balanced activity.⁸³

Mesenchymal stromal cells have been described as present in many tissues and current literature shows that they can establish connection and modulate the activity of many cells of the immune system. In line with the initial idea that MSC have an active role in promoting the innate tissue surveillance and also have an important part in the control of exacerbated tissue immune responses; we could say that the immunosupressive effect of MSC is focused on restoring tissue homeostasis or, that it is aimed at restoring ‘tissue innate tolerance’ and this, as has previously been suggested, could be a property shared by all stromal cells.^72,84 Considering the immnuomodulating properties of MSCs discussed above; we would like to suggest that, among other cells that constitute the tissue's basic architecture MSC have the role of setting the background and actively participate in bringing together cells involved in the local tissue immune response aiming to maintain tissue homeostasis.

Disclosure

The authors declare no conflict of interest.