The influence of mTOR inhibitors on immunity and the relationship to post-transplant malignancy

The known role of mammalian target of rapamycin (mTOR) in the immune response has been rapidly evolving, from what was once thought to be a simple immunosuppressive antiproliferative effect on T cells to a very complex central role that serves to integrate multiple signals given to T cells, B cells and antigen-presenting cells. The complexity of this topic is demonstrated by recent data suggesting that mTOR inhibition can either inhibit or promote certain aspects of immune responses, depending on the nature of the antigenic stimulus, and the environmental conditions cueing the cellular immunological players. There is even evidence that, under mTOR inhibition, an immune response to one foreign entity (for example, an organ transplant) may be simultaneously completely different to that of another (for example, tumour or microorganism). To understand how this might be possible, it is necessary to investigate the central role that mTOR seems to have in shaping the immune response. This review is aimed at examining how mTOR controls the development and function of key immune cells, and puts this information primarily in the context of organ transplant rejection and post-transplant malignancy.

A relationship between the use of general immunosuppressive drugs to prevent allograft rejection and the development of cancer after organ transplantation has been recognised for decades. Th e scientifi c transplant community has developed a growing concern about cancer, since post-transplant malignancy has emerged as a leading cause of morbidity and mortality, especially in patients who have a high or long-term exposure to immunosuppression [1,2]. Th ere are diff erent explanations for why post-transplant malignancy occurs more frequently in this pharmacologically immunosuppressed population, including enhancement of tumour invasive properties [3] and reduction in DNA repair mechanisms [4]. However, the most discussed mechanism is the seemingly obvious eff ect of suppressing the ability of immune cells to detect and eliminate cancer as it develops (immune surveillance). Although one might be intuitive to hypothesise that all immunosuppressive drugs will have the same suppressive eff ect against tumour immunity, recent research suggests that this may not be the case. New basic questions have therefore been raised, including the following: how do mammalian target of rapamycin (mTOR) inhibitors aff ect the develop ment of specifi c immune cells that are most critical to produce an eff ective anti-tumour immune response? Do the various immunosuppressive drugs aff ect these cells diff erently? Is it possible to enhance an immune response towards a tumour, while at the same time inhibiting immunity towards a transplanted allograft? Since the most common tumours that develop in transplant recipients are virally associated [5], how do mTOR inhibitors infl uence specifi c viral infections in these patients?
Certainly intriguing is the fact that specifi c immune responses formed simultaneously against an allograft and a (immunogenic) tumour may be diff erent in nature, and are likely to be altered signifi cantly by various immunosuppressive substances, depending on the answers to the questions posed above. Th e present review focuses on new evidence that the mTOR pathway is uniquely positioned to aff ect the diff erential development of lymphocyte subpopulations, as well as the maturation of antigen-presenting cell (APC) populations, all of which are critical in the formation of immune responses towards organ transplants and tumours. Interestingly, some of the aff ected immune cell types sway immunity

Abstract
The known role of mammalian target of rapamycin (mTOR) in the immune response has been rapidly evolving, from what was once thought to be a simple immunosuppressive antiproliferative eff ect on T cells to a very complex central role that serves to integrate multiple signals given to T cells, B cells and antigenpresenting cells. The complexity of this topic is demonstrated by recent data suggesting that mTOR inhibition can either inhibit or promote certain aspects of immune responses, depending on the nature of the antigenic stimulus, and the environmental conditions cueing the cellular immunological players. There is even evidence that, under mTOR inhibition, an immune response to one foreign entity (for example, an organ transplant) may be simultaneously completely diff erent to that of another (for example, tumour or microorganism). To understand how this might be possible, it is necessary to investigate the central role that mTOR seems to have in shaping the immune response. This review is aimed at examining how mTOR controls the development and function of key immune cells, and puts this information primarily in the context of organ transplant rejection and post-transplant malignancy.

Keywords: mammalian target of rapamycin, posttransplant malignancy, T cells, B cells, dendritic cells, organ transplantation
The infl uence of mTOR inhibitors on immunity and the relationship to post-transplant malignancy Edward K Geissler* towards suppressive regulation, and others enhance eff ector cell capabilities. With regard to organ trans plantation recipients, the ideal overall eff ect sought therapeutically is complex, and demanding, in that it is desirable to minimise the immune response against the allograft, while at the same time enhancing immunity against tumours (and infectious microorganisms).
Although this suggestion appears to be mutually exclusive, there are reasons to believe that the development of such a diametrically opposing response to an organ transplant and a tumour or microorganism may be possible. In this article the eff ects of mTOR inhibitors on diff erent cellular components of the immune system are reviewed. Th e aim is to put forward the latest information regarding the eff ects of mTOR inhibition on immunity in the context of reducing the complex problem of posttransplant malignancy in transplant recipients.

Immunosuppressive and immunomodulatory eff ects of mTOR inhibitors
Inhibitors of mTOR have been used for more than a decade as immunosuppressive agents to prevent organ transplant rejection. Early mechanistic studies on rapamycin pointed towards a rather simple explanation for the immunosuppressive eff ect, in that the mTOR pathway was found to be essential for the cell proliferation signal (often referred to as signal 3) triggered by IL-2, thus preventing the expansion of donor-specifi c T cells activated through the T-cell receptor [6]. Th is immunosuppressive eff ect contrasts in general with calcineurin inhibitors (CNIs), which act by preventing the initial activation (signal 1) of T cells [7]. While our overall mechanistic understanding of the basic suppressive activity of CNIs is still generally accepted, new data suggest that the suppressive eff ect of mTOR inhibition on immune cells is very complex and probably does not rest primarily on inhibition of T-cell proliferation.
Recent mechanistic investigations indicate that mTOR plays a central role in the diff erentiation of T-cell subsets, and also controls aspects of B-cell and APC development. In fact, mTOR is a critical regulator of the immune response because it acts as a central node for sensing nutrient availability, cytokine/growth factor signalling and costimulatory factors ( Figure 1). Indeed, mTOR is in a unique intracellular signalling position to integrate all of these factors so cells can eff ectively and properly balance cues from the ever-changing microenvironment, such as those induced by microbiological (for example, bacterial, viral) or allo geneic (for example, organ transplant, tumour) challenges.

Role of mTOR in immune cell development -T cells
T cells are critically involved at nearly all levels of any immune response. While the primary eff ect of mTOR inhibition on T cells was initially attributed to blockage of IL-2 proliferation-inducing signalling, hints that this is not the only eff ect have become evident. One clue was that the initial fi nding that rapamycin treatment induces T-cell anergy (lack of responsiveness) through inhibition of proliferation [8,9] was later found to be independent of this anti-proliferative eff ect, and rather to be due to a direct inhibitory eff ect on mTOR itself [10]. Subsequent investigations into the link of mTOR to T-cell metabolism, and to transcription factors that are now recognised to control T-cell subset diff erentiation, opened new views towards mTOR inhibitor eff ects on the immune response.
Regarding metabolism, mTOR's central role comes directly into play because activated lymphocytes primarily use glycolysis for energy due to their need to produce proteins, nucleotides and lipids that are essential for the generation of key biosynthetic substrates [11,12]; the shifting from mitochondrial respiration to glycolysis (referred to as the Warburg eff ect) is similar to that which occurs in cancer cells. Interestingly, mTOR as a regulator of metabolism provides links to lymphocyte activation in this context. One example is that T-cell co-stimulation via CD28 triggers the activation of signalling molecules upstream of mTOR that promote expression of necessary membrane glucose transporters. In general, one can state that inhibition of cell metabolism through mTOR leads to inhibition of T-cell-mediated immunity. Th e importance of this idea cannot be overstressed since it has been shown, for instance, that T-cell anergy is due at least in part to decreased mTOR activation [13]; if mTOR is resistant to reactivation in an anergic state, then the required metabolic machinery is not going to be available and the cell will remain anergic to otherwise stimulatory signals. Indeed, substances such as metformin and AICAR, which mimic energy depletion and activate Geissler Transplantation Research 2013, 2(Suppl 1):S2 http://www.transplantation research.com/supplements/2/S1/S2 AMPK (an inhibitor of mTOR), promote T-cell anergy [13][14][15]. Cell metabolism, mTOR and the immune response currently constitute an intense area of basic research that has substantial therapeutic potential and implications.
Th e diff erentiation of CD4 and CD8 T-cell populations has a major impact on the development of any immune response to allogeneic transplants or tumour entities. Recent data demonstrate an important role for mTOR in determining the T-cell diff erentiation pattern (Figure 2). To understand this role better, fi rst it is necessary to recognise that mTOR is part of two large complexes, referred to as mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2), where mTORC1 is directly inhibited by rapamycin while mTORC2 is only indirectly and partially inhibited with long-term exposure to the drug. Second, it is helpful to know that diff erentiation of Th 1, Th 2 and Th 17 T-helper cell subsets is regulated by the lineage-specifi c transcription factors T-bet, GATA-3 and RORγt, respectively. Considering this background informa tion, recent experimental models suggest that block ing of mTORC1 with rapamycin, or by knocking out essential components of the mTORC1, a Th 2 polarised T-cell dominance develops; whereas knocking out mTORC2 polarises the T-helper immune response towards Th 1 and Th 17 cell development [16]. Most interestingly, blocking both mTOR complexes leads to the generation of a Foxp3 + T-regulatory (Treg) cell expansion. Moreover, these Treg cells are resistant to apoptosis [17]. Indeed, Treg cells appear in general to require less mTOR activity, which is consistent with the reduced metabolic demands for these cells compared with eff ector T cells [18]. Interestingly, although Treg cells depend on IL-2 for proliferation, IL-2 stimulation results in high levels of STAT5 phosphorylation, rather than activation of mTOR [19], suggesting that diff erent T-cell subpopulations depend on alternate signalling pathways for expansion and survival. In terms of thera peutic application of mTOR inhibitors, research suggests that the diff erential eff ects summarised above depend substantially on the dose, duration and timing of the drug application [20][21][22], indicating that more is to be learnt about how best to apply mTOR inhibitors to suit the clinical purpose intended. Some of the same eff ects apply to CD8 + cells regarding mTOR dependence. For instance, activation of CD8 + cells also primarily depends on glycolysis [23], and diff er en tiation of eff ector CD8 + cells requires mTORC1-dependent T-bet expression [24]. Most critically, mTOR is involved in the transition of eff ector to memory CD8 + T cells (Figure 2), and this appears to rely on conversion of T-bet to eomesodermin transcription factor expression [24][25][26]; blocking mTOR with rapamycin has this exact eff ect, and therefore promotes the development and sustenance of memory T cells that transition effi ciently into eff ector cells highly capable of producing immune responses to, for instance, tumours [24]. Similar to Treg cells, memory CD8 + T cells depend on mitochondrial oxidative phosphory lation for energy (rather than glycolysis) and are driven by STAT5 signalling. One perplexing question is therefore whether mTOR inhibition increases immunity to viruses, bacteria and tumours, while at the same time protects organ transplants from rejection.
Recent data suggest that rapamycin treatment augments CD8 + T-cell memory responses towards viruses. Th is eff ect has been demonstrated by impressive boosting of vaccination responses both in mice [27,28] and in nonhuman primate studies [29]; in the nonhuman primate experiments, immunosuppressive doses of sirolimus promoted CD8 + T-cell memory towards vaccinia virus, while CNI use did not. Indeed, it is ironic that an immunosuppressive agent is being considered for boosting vaccination responses in humans. Another inter est ing aspect of this research is that viral infections (for example, human papillomavirus, hepatitis B/C, Epstein-Barr virus, human herpesvirus 8) are associated with the most common post-transplant malignancies, suggesting that a boost in immunity to these viruses could aff ect cancer development. Moreover, several recent experimental studies indicate that rapamycin adminis tration directly enhances memory T-cell formation against tumours [22,24]. Th is is an observation we have also been able to confi rm in the laboratory [30], and we can add that CNIs do not support memory development in our models. Th e boosting of T-cell memory with mTOR inhi bition has substantial therapeutic implications regarding the problems of viral infection and posttransplant malignancy in organ transplant recipients.
Th is leads to the question raised earlier of whether an immune response can be promoted in one foreign entity (for example, virus, bacteria or tumour cell) and yet inhibited by another (an organ transplant). An interesting experimental study from Ferrer and colleagues demonstrates that rapamycin-treated mice have protection against rejection of an OVA-expressing skin allograft, while at the same time showing a heightened CD8 + T-cell response against the same OVA epitope expressed by bacteria (Listeria monocytogenes) [31]. Th is is a critical observation, since it opens the possibility that mTOR inhibitors can enhance immunity to infectious agents without at the same time promoting the immune reaction against an organ allo graft. In fact, it can be argued that enhancement of CD4 + Treg cell and CD8 + T suppressor cell [32] responses towards allografts may provide for long-lasting protec tion and perhaps even some degree of immunological tolerance. Unfortunately, it is completely unclear why there is such a divergent response to two foreign entities expressing the same foreign protein. Does this divergence relate to the microenvironmental conditions under which allograft versus microbiological antigens are presented to the immune system, or are other factors responsible? Th is is clearly an intriguing area of research, and high lights the importance of mTOR's role in orchestrating complex immune responses.

Role of mTOR in B-cell and antigen-presenting cell development
Less information exists regarding the role of mTOR inhibitors on B cells. However, data from the mTOR hypomorph mouse (constitutive reduced, but not absent, mTOR levels) suggest that B-cell development may be even more aff ected than T cells [33]. In these mice, B-cell development in the bone marrow is partially inhibited, which was refl ected by decreased B-cell proliferation in response to antigenic stimulation and reduced antibody production capability. Interestingly, mice with B cells that overexpress mTOR because of a TSC1 deletion (TSC1 normally inhibits mTOR, see Figure 1) also demonstrate similar defects in B-cell diff erentiation and antibody production [34]. Another indication for an mTOR role comes from the fact that activated B cells, like T cells, use glycolysis as a primary source of energy [35]. Together, these early experimental indications suggest that mTOR is likely to have a signifi cant impact on B-cell activation, diff erentiation and function, but more in-depth studies are lacking to defi ne the exact role of mTOR in B-cellmediated immunity.
Finally, the mTOR pathway is also important for the diff erentiation and function of APCs. In particular, mTOR inhibition has a potent eff ect on the maturation of dendritic cells (DCs). Diff erentiation into conventional, CD8 + and plasmacytoid DCs appears to depend on mTOR. Indeed, mice with uninhibited mTOR activity (via PTEN deletion, see Figure 1) develop an abnormal highly expanded DC compartment [36], suggesting that mTOR plays a critical role in maintaining APC homeostasis in vivo. Moreover, rapamycin treatment has a profound eff ect on APC function, in that co-stimulatory molecule expression is decreased, leading to an inhibited ability for APC to stimulate T-cell activation [37,38]. Rapa mycin-treated DCs are even known to induce tolerance in animal models [39,40], possibly through their ability to promote the development of Treg cells [41]. In fact, researchers that are expanding Treg cells for the purpose of cell therapy often use rapamycin to produce a more stable Treg phenotype [42]. One should mention, however, that rapamycin has seemingly opposing eff ects on the early development of immune responses involving plasmacytoid DCs versus other DCs. While plasmacytoid DCs activated via toll-like receptors depend on mTOR to elicit type 1 interferon-based expression responses, lipopolysaccharide activation of monocytes and DCs leads to inhibition of a proinfl ammatory gene expression pattern through the mTOR pathway [43]; mTOR inhibition could thus aff ect responses to bacterial challenges, especially in immunosuppressed transplant recipients. mTOR therefore has important eff ects on APC homeostasis and development that require a great deal more research to be fully understood. Nonetheless, mTOR clearly has yet another key function in the development of immune responses.

Conclusion
What was once thought to be a simple explanation (antiproliferative eff ect) for how mTOR inhibitors reduce the immune reaction to organ allografts is now developing into a very complex explanation. One should also state that while several nonimmunological mechanistic explana tions for the anti-tumour eff ects of mTOR inhibitors have been described [44], promotion of immune responses to cancer is unexpectedly coming more into focus. Th e most recent data suggest that mTOR acts as a central node for coordinating activities of the most important cells (T cells, B cells and APCs) forming the immune response to various challenges. Interestingly, some of these eff ects inhibit an immune response, and other eff ects actually promote immunity; the setting of the antigenic challenge appears to be crucial, since energy availability, signalling cues and cell activation all converge to at least some degree upon mTOR.
What does this mean for transplant patients in terms of allograft protection (immunosuppression), viral or bacterial infection and post-transplant malignancy? Although there are no simple answers to this question, more light is being shed on the topic with intensive ongoing research. In terms of protecting allografts from rejection, mTOR inhibitors are attractive from the theoretical perspective that they might be optimal for maintaining a state of donor-specifi c regulation through promotion of tolerogenic DCs and Treg cells. Although mTOR inhibitors alone do not appear to produce tolerance in transplant recipients [45,46], perhaps strategic use of these drugs in combination with novel induction therapies or cell therapy [47,48] could yield better results. Regarding infectious complications associated with immunosuppression in organ transplantation, there is already early evidence that mTOR inhibitors might reduce the problem of some viral infections, including cytomegalovirus, human herpesvirus 8 and BK virus [49][50][51][52][53][54]. In turn, if viral infections can be decreased, mTOR inhibitors could have an indirect impact on the development of post-transplant malig nan cies. In addition, promotion of memory CD8 + T-cell responses against tumour cells could also lessen the problem of cancer in transplant recipients.
To conclude, although early evidence suggests that mTOR inhibitors have the potential to promote an immune response against an infectious microorganism or tumour entity, and can paradoxically function to inhibit immunity against an organ allograft, further research is needed to untangle the operative mechanisms and to ultimately explore the full potential of mTOR inhibitors in the setting of organ transplantation.

Competing interests
The author declares that he has no competing interests.

Declaration
This article has been published as part of Transplantation Research Volume 2 Suppl 1, 2013: Proceedings of the 12th International Symposium on Transplantation. The full contents of the supplement are available at http://www.transplantationresearch.com/supplements/2/S1. The supplement is based on the 12th International Transplantation Symposium held on 16-17 April 2012, in Istanbul, Turkey. The symposium and publication based on these proceedings were supported by an unrestricted educational grant from Pfi zer Inc., who had no infl uence on the editorial content. Editorial support was provided by Synergy, Richmond, UK and was funded by Pfi zer Inc. The articles have been through the journal's standard peer review process and the Supplement Editors declare that they have no competing interests.