Dominant regulation of long-term allograft survival is mediated by microRNA-142

Organ transplantation is often lifesaving, but the long-term deleterious effects of combinatorial immunosuppression regimens and allograft failure cause significant morbidity and mortality. Long-term graft survival in the absence of continuing immunosuppression, defined as operational tolerance, has never been described in the context of multiple major histocompatibility complex (MHC) mismatches. Here, we show that miR-142 deficiency leads to indefinite allograft survival in a fully MHC mismatched murine cardiac transplant model in the absence of exogenous immu -nosuppression. We demonstrate that the cause of indefinite allograft survival in the absence of miR-142 maps specifically to the T cell compartment. Of therapeutic relevance, temporal deletion of miR-142 in adult mice prior to transplantation of a fully MHC mismatched skin allograft resulted in prolonged allograft survival. Mechanistically, miR-142 directly targets Tgfbr1 for repression in regulatory T cells (T REG ). This leads to increased T REG sensitivity to transforming growth factor – beta and promotes transplant tolerance via an augmented peripheral T REG response in the absence of miR-142. These data identify manipulation of miR-142 as a promising ap-proach for the induction of tolerance in human transplantation.


| INTRODUC TI ON
Solid allograft transplantation can be lifesaving at the point of organ failure. However, long-term allograft and patient survival depends on sustained drug-induced immunosuppression. Acute cellular rejection (ACR) occurs in around 25% of heart transplant recipients in the first-year posttransplant despite optimal levels of currently available immunosuppressive therapy, accounting for 10% of mortality in the first-year posttransplant and increasing the risk of 5-year mortality. 1 Effector CD4 + T cells (T EFF ) are critical mediators of allograft responses resulting in allograft rejection. Induction of tolerance is a major goal in transplantation, potentially allowing withdrawal of immunosuppression and indefinite allograft survival. 2 Regulatory T cells (T REG ) are a subset of CD4 + T cells that exert dominant suppression of T EFF responses. 3 Peripherally induced T REG cells are critically dependent on transforming growth factor -beta (TGF-β) for the induction of their key transcription factor, FoxP3. 4 The vital role of T REGS in transplant tolerance is apparent where deletion of T REGS in mice already tolerant of kidney allografts leads to allograft rejection. 5 In human studies, the expression of FoxP3 in transplant infiltrating T cells is associated with donor-specific hyporesponsiveness and improved graft histological findings. 6 MicroRNAs (miRNAs) are short noncoding RNAs that regulate gene expression posttranscriptionally by binding to target sequences on mRNAs, leading to their degradation and/or translational inhibition. 7 The role of miRNAs in the regulation of both innate and adaptive immune responses is increasingly recognized, with aberrant expression of miRNAs contributing to autoimmune diseases and malignancies. 7,8 Recently, we demonstrated a critical role for  in regulating peripheral immune tolerance via its action in CD4 + T cells. 9 MiR-142 is highly expressed in cells of hematopoietic lineages, including CD4 + T cells 10 and is conserved across vertebrates, including between humans and mice, supporting the validity of murine models in the investigation of miR-142 function in human disease and therapeutic development. 11 This microRNA exists as two functional isoforms (miR-142-3p and miR-142-5p), capable of binding to different mRNA targets but processed from a single pre-miRNA hairpin. In the context of acute cellular rejection, levels of miR-142-3p and miR-142-5p are consistently increased in cardiac allograft tissue. 12 Serum miR-142-3p levels are also significantly higher in the context of biopsy proven ACR in heart transplant recipients. 13 MiR-142 over-expression is highly predictive of acute T cell-mediated rejection in renal allograft biopsies and is increased in peripheral blood mononuclear cells (PBMC) of patients with chronic antibody-mediated rejection. 14,15 However, miR-142 has also been shown to be up-regulated in B lymphocyte subsets of PBMCs in operationally tolerant renal transplant patients. 16,17 Whether miR-142 actually plays a functional role in transplant rejection and what that role may be, has not previously been explored.

| Heterotopic heart transplantation
BALB/c (H-2 d ) cardiac allografts were transplanted heterotopically into C57BL/6 (H-2 b ) recipients that is, a major MHC mismatch (class I and II). 18 Daily palpation of the recipient abdomen for the heterotopic heartbeat, monitoring for signs of slowing or reduced impulse was performed. Allograft rejection was defined as complete cessation of the heterotopic heartbeat. A standard measure of indefinite allograft surgical has been defined as allograft survival of over 100 days. 19,20
Only ER T2 Cre −/+ animals were utilized in experiments.
100 μg tamoxifen (Sigma # 156738) diluted in sunflower oil, warmed to 37°C were administered intraperitoneally once daily on 3 consecutive days to ER T2 Cre x Mir142 fl/fl mice at 5-6 weeks of age.

| Skin transplantation
Skin transplantation was performed as previously described. 21 Ears of euthanized donor mice were disinfected with 70% ethanol, excised, and split into ventral and dorsal halves. Ventral tissue was kept in phosphate-buffered saline (PBS) on ice prior to implantation and the collagenous ventral flap discarded. Recipient mice were anesthetized by 3% isoflurane inhalation in 100% oxygen at a flow rate of 2 L/min, then maintained using 1.5%-2% isoflurane at 2 L/min intraoperatively. The recipient site was shaved and swabbed with 70% ethanol. A 1-1.5 cm incision was made over the back. The skin graft was then placed atop the graft bed and wrapped in a sterile bandage. The bandage was secured with a single 2-0 silk suture and removed 6-7 days posttransplant.

| Real-time PCR
RNA was extracted using the RNeasy Micro Kit (Qiagen, Hilden, Germany). cDNA was prepared using TaqMan advanced miRNA cDNA synthesis kit (Thermo Fisher Scientific, Waltham, MA). TaqMan fast advanced master mix and TaqMan advanced microRNA Assays were used (Thermo Fisher Scientific) and run on a 7900HT Fast Real-Time PCR system (Thermo Fisher Scientific). U6 small nuclear RNA, miR-191-5p and miR-361 were used as endogenous controls. Sera from wild-type recipients of fully allogeneic allografts and naïve animals were used as positive and negative controls respectively.

| Histology
Samples were fixed in 10% neutral buffered formalin for 48 hours before paraffin-embedding and sectioning. Histological scoring was performed blinded according to the International Society for Heart and Lung Transplantation standardizing grading criteria for the diagnosis of heart rejection 22 and the Banff 2007 working classification of skin allograft pathology. 23 Microscopy was performed with an Olympus BX51 microscope.

| Hematoxylin & eosin staining
5 mm thick sections were dewaxed, rehydrated with water, then stained with Mayer's hematoxylin for 5 minutes, washed and incubated with 1% Eosin stain for 5 minutes, rinsed briefly with tap water, dehydrated rapidly through graded methanol, cleared in Xylene and mounted in DPX.
HRP-conjugated goat anti-rabbit IgG was used for secondary detection (GE Healthcare, Chicago, IL). Blots were developed using enhanced chemiluminenscence (Thermo Scientific/Pearce). Intensity was quantified using Genetools software (Syngne) and expressed as fold increase above basal level.

| T cell intrinsic miR-142 deficiency prolongs skin allograft survival
Given the striking results observed in our heterotopic heart trans-  Figure 2F). In addition, more extensive FoxP3 expression was observed in Cd4 cre+ Mir142 fl/fl recipient cardiac allografts taken from the model outlined in Figure 1F ( Figure   S4). These data suggest that in response to exposure to an allograft, an enhanced T REG response is generated by the absence of miR-142 isoforms.  Figure 3C).

Deletion of miR-142 either constitutively or conditionally via
As before, we found a significant increase in the proportion of CD4 + CD25 + T REGS ( Figure 3D) and decrease in T EFF/MEM populations in ER T2 Cre + Mir142 fl/fl mice when compared with control mice at day 8 in secondary lymphoid tissue sites ( Figure 3D,E). Directly ex vivo ICC revealed significantly reduced CD4 + T cell production of IFN-γ and IL-17 production at day 8 ( Figure 3F). Similarly, donor-specific antibody responses were significantly reduced in ER T2 Cre + Mir142 fl/fl mice when compared with control mice ( Figure 3G).

| Expression of the miR-142-3p target Tgfbr1 is increased in miR-142 deficient T REG and augments sensitivity to TGF-β signaling
In all the transplantation models we had explored, an enhanced T REGS response was apparent in the context of T cell miR-142 deficiency. Therefore, we reasoned that the absence of miR-142 in the context of exposure to an allograft may promote a more tolerogenic environment, via mechanisms which augment T REG development and subsequent dominant tolerance.
Among factors which promote T REG development and function, the TGF-β signaling pathway plays an important role, not least by promoting the expression of Foxp3. 29,30 TGF-β signals via a heterodimeric receptor composed of ALK5 (Tgfbr1) and the subunit Tgfbr2.

| D ISCUSS I ON
We have demonstrated that the absence of miR-142 induces indefinite cardiac allograft survival across full MHC (Class I and II) mismatches, caused by a T cell-specific mechanism, which promotes dominant tolerance and suppression of effector CD4 + T cell responses via the enhancement of the peripheral T REG response.
Diminished T cell numbers in secondary lymphoid organs have been observed in Mir142 −/− mice. 37,38 We also found this to be the case in our constitutive and T cell-conditional knockout lines at baseline (data not shown). However, CD4 and CD8 deficient mice as well as Rag −/− mice, which lack T and B cells, can vigorously reject skin allografts at the same rate as WT mice. [39][40][41][42] B6 µMT −/− mice are B cell-deficient with a paucity of CD4 + T cells yet are still capable of rejecting skin and heart allografts. 41,43 Importantly, as few as 1 × 10 3 donor-specific T cells are capable of rejecting skin allografts. 25,44 The quantitatively similar responses to the allograft in our mice, which are only partially T cell-deficient and possess T cells in their peripheral tissues, indicates that this is unlikely to contribute significantly to the observed long-term transplant survival.
The strength of T EFF/MEM responses in both lymphoid and nonlymphoid tissues is a barrier to transplant tolerance. 45,46 The presence and number of alloantigen-specific T EFF/MEM cells in humans correlates with worse transplant outcomes. 47 Despite current immunosuppressive regimens capable of non-specific T cell depletion, certain T EFF/MEM subsets have been found to survive treatment with the commonly used depleting induction agents, Campath (anti-CD52 mAb) or anti-thymocyte globulin. Indeed, these cells expand and proliferate in the "empty" host to play a prominent role in acute rejection. 48 Other attempts to augment allograft survival using  17,57 Several studies have shown that miR-142 plays a critical and highly conserved role in regulating TGF-β receptor signaling in vertebrates. [31][32][33][34][35][36] Upregulation of Tgfbr1 expression in CD4 + CD25 + FoxP3 + T REGS from mice with T REG -conditional deletion F I G U R E 3 Prolonged skin allograft survival in ER T2 Cre + Mir142 fl/fl mice. A, Kaplan Meier allograft survival chart of control vs ER T2 Cre + Mir142 fl/fl littermates (n > 8 per group, Mantel-Cox test ****P < .0001). B, Macroscopic appearance of control vs ER T2 Cre + Mir142 fl/fl skin allografts at days 8 and 12. C, Hematoxylin and eosin (H&E) staining of formalin fixed, paraffin embedded (FFPE) sections from allografts from ER T2 Cre + Mir142 fl/fl and control recipients at day 8 posttransplant (magnification ×5 and ×20; scale bars 1000 and 100 µm) (left); Banff grading of cell-mediated skin allograft rejection 23 (n = 5; Student's t test; ***P < .001) (right). D, Percentage of T REG (CD4 + CD25 + ) (n = 4; Mann-Whitney U test, **P < .01). E, Percentage of T EFF/MEM (CD4 + CD25 − CD44 high CD62L low ) cells (left) and percentage of T NAIVE (CD4 + CD25 − CD62L high CD44 low ) (right) (n = 4; Mann-Whitney U test, *P < .05). F, Direct ex vivo interferon-gamma (IFNγ) and interleukin (IL)-17 cytokine responses in CD3 + CD4 + and CD3 + CD8 + T cells at day 8 posttransplant (n = 4 per group; *P < .05, **P < .01; unpaired two-tailed t test). G, Donor-specific antibody responses, control vs ER T2 Cre + Mir142 fl/fl at day 8 posttransplant (n > 8 per group; **P < .01; unpaired two-tailed t test) of Mir142, in combination with the highly conserved seed sequences for miR-142-3p in the 3′UTR of Tgfbr1 make this a plausible target for miR-142-3p in T REG . 11 In contrast to our previous findings, where we found that the absence of miR-142 specifically in FoxP3 + T REGS results in a lethal, multi-system autoimmune disease and impaired T REG suppressive function, 9 here we found that the absence of miR-142 in Pde3b. Whether this difference reflects the differential expression of the -3p and -5p strand of miR-142 in separate T REG subsets (eg, thymic T REG vs peripheral T REG ) in response to self vs nonself-antigens remains to be determined.
In summary, miRNAs have emerged as key regulators of numerous biological processes. Our data identify that miR-142 plays a critical role in solid organ transplantation, modulating expression of its target Tgfbr1 in T REGS , thereby modulating T REG sensitivity to TGF-β and promoting transplant tolerance via augmented T REG development. These findings suggest that targeted T REG therapy aimed at manipulation of miR-142 and its target TGFBR1 could have therapeutic potential in improving allograft survival.