Abstract
Because miR-181a has been described to alter T cell activation, we hypothesized that manipulation of miR-181a expression in donor T cells may alter acute graft-versus-host disease (aGvHD) after allogeneic bone marrow transplantation (BMT). We therefore analyzed the impact of enhanced and reduced miR-181a expression in donor T cells on aGvHD induction by lentiviral gene transfer into primary T cells and using miR-181a/b-1−/− T cells, respectively. BMT-recipient mice receiving donor T cells with enhanced miR-181a expression showed no signs of aGvHD and survived for the time of follow-up, whereas T cells lacking miR-181a/b-1 accelerated aGvHD. In line with these data, analysis of donor T cells in blood, secondary lymphoid organs, and target organs of aGvHD after BMT showed significantly reduced numbers of miR-181a–transduced T cells, as compared with controls. In addition, expansion of activated T cells with enhanced miR-181a expression was reduced in vitro and in vivo. We further show that anti-apoptotic BCL-2 protein expression is reduced in murine and human T cells upon overexpression of miR-181a, suggesting that regulation of BCL-2-expression by miR-181a may contribute to altered alloreactivity of T cells in aGvHD. These data indicate that proteins regulated by miR-181a may be therapeutic targets for aGvHD prevention.
Introduction
Allogeneic stem cell or bone marrow (BM) transplantation (BMT) is the most effective and often unique curative therapy for a variety of high-risk hematological malignancies, BM failure syndromes, and congenital immune deficiencies. However, graft-versus-host disease (GvHD) remains the most frequent severe complication in BMT, contributing largely to nonrelapse mortality. Acute GvHD (aGvHD) is initiated by an aggressive immune response of donor-derived alloreactive T cells directed against host tissues. After allopriming, donor T cells migrate to typical GvHD target organs such as skin, liver, and/or gut, inducing a strong immune response, which ultimately leads to organ damage and/or failure (1). Accordingly, aGvHD prevention and treatment are performed with immunosuppressive drugs such as corticosteroids, calcineurin inhibitors, and different Abs, among others. However, steroid refractory intestinal aGvHD still has a mortality rate of close to 100%. Therefore, new therapeutic strategies are urgently needed.
MicroRNAs (miRNA) are small noncoding RNAs involved in posttranscriptional regulation of gene expression (2, 3). Despite many efforts, however, the precise physiological function of individual miRNAs and their pathophysiological contribution to different disease states still remain elusive. In aGvHD, a role for miRNAs in allogeneic donor T cells has recently been described. Overexpression of miR-155 in donor T cells leads to increased GvHD mortality, whereas miR-155 deficiency prevents aGvHD (4). Furthermore, increased miR-146a expression in donor T cells prevents from GvHD development by targeting TNFR-associated factor 6, leading to reduced TNF transcription (5). It has also been shown that blocking the miR17-92 cluster alleviates aGvHD while preserving the graft-versus-leukemia (GvL) effect (6).
The miR-181 family encompasses six miRNAs (miR-181a1, miR-181a2, miR-181b1, miR-181b2, miR-181c, miR-181d) encoded in three paralogs (7). The mature miR-181a1 and miR-181a2 as well as miR-181b1 and miR-181b2 are identical, and miR-181a and miR-181b share identical seed sequences. Li et al. (8) reported that miR-181a regulates T cell sensitivity to Ag by downregulation of several phosphatases downstream of the TCR. The authors demonstrated increased Ca influx, enhanced TCR signaling strength, and elevated levels of phosphorylated signaling intermediates upon TCR activation in miR-181a– overexpressing thymocytes. Overexpression of miR-181a in adult human peripheral T cells resulted in increased Ca flux (9). Conversely, thymocytes from mice lacking miR-181a/b-1 displayed a defect in Ca signaling, whereas peripheral T cells from these only showed mild aberrations in baseline Ca flux (10, 11). Accordingly, miR-181a has been shown to be involved in positive selection of double-positive thymocytes, regulation of central and peripheral tolerance, and production of invariant NK cells (10–13). Interestingly, compound deletion of miR-181 paralogs reduces survival of affected mice, and mice deficient for all three miR-181 clusters could not be obtained (13). Furthermore, inhibition of NOTCH-induced T-lineage acute lymphoblastic leukemia by reduced expression of miR-181a has been described (14). However, the impact of miR-181a overexpression on peripheral T cell expansion and function has not yet been reported.
Because miR-181a modulates TCR signaling, we hypothesized that it may also modulate the alloresponse of donor T cells in aGvHD. We therefore used allogeneic donor T cells with enhanced or reduced expression of miR-181a due to lentiviral transduction or knockout of miR-181a/b-1 (10), respectively, to analyze miR-181a gain- and loss-of-function phenotypes in a murine model of aGvHD. Interestingly, endogenous miR-181a is downregulated upon TCR activation in vitro. However, increased and reduced expression of miR-181a in alloreactive donor T cells abrogate and accelerate aGvHD in recipient mice, respectively. Accordingly, expansion of activated T cells overexpressing miR-181a is reduced in vitro and in vivo. Furthermore, expression of anti-apoptotic proteins BCL-2 and MCL-1 as well as proapoptotic BAX is diminished in T cells upon overexpression of miR-181a, suggesting that regulation of mitochondrial BCL-2 family members by miR-181a may contribute to altered alloreactivity of T cells in aGvHD. These data provide a rationale to search for druggable miR-181a targets relevant for T cell function to be evaluated for prevention and/or therapy of aGvHD in preclinical models.
Materials and Methods
Mice
Recipient B6D2F1/Crl (BDF1, H-2bxd) mice were purchased from Charles River Laboratory (Sulzfeld, Germany). Donor C57BL/6J-IghaThy1aGpi1a/J (C57BL/6, H-2b, Thy1.1+) and C57BL/6.SJL-PtprcaPep3b/BoyJ (H-2b, Ly5.1+) mice were bred at central animal facility of Hannover Medical School. miR-181a/b-1−/− and miR-181a/b-1+/− mice have been described elsewhere (10) and were also bred at central animal facility of Hannover Medical School. All mice used in the experiments were 8–10 wk old. All animal studies were performed under institutional and governmental directives and were approved by the Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit (permit number 09/1685).
Cell culture and cytokines
HEK293T cells were maintained in DMEM (Life Technologies, Eggenstein, Germany) with 10% FCS (PAA Laboratories, Pasching, Austria), and 1% penicillin/streptomycin (Life Technologies, Eggenstein, Germany). Primary T cells were harvested from donor C57BL/6 mice (8–10 wk old), and pan-T cells were isolated using Milteny Biotec MACS kits (Bergisch Gladbach, Germany). Subsequently, T cells were cultured in IL-2–supplemented (100 U/ml) RPMI 1640 (Roche, Basel, Switzerland) medium with 10% FCS and 1% penicillin/streptomycin for 24 h before exposure to lentiviral particles.
DNA constructs
Lentiviral gene transfer was used for stable expression of miR-181a in T cells. To generate stable miRNA gain-of-function phenotypes, we cloned a lentiviral vector harboring an H1-promoter-driven miR-181a within a miR-30 backbone cassette in the 3′ long terminal repeat corresponding to our short hairpin RNA-encoding lentiviral vectors previously described (15). The location in the U3 region of the 3′ long terminal repeat leads to duplication of the miR-181a expression cassette during reverse transcription. Lentiviral transduction of target cells will result in long-term miR-181a transcription. The empty vector served as a control. Enhanced GFP (eGFP) gene is encoded in all miRNA-lentiviral vectors as a reporter gene.
Lentiviral transduction
Lentiviral transduction was performed as previously described (15). Briefly, lentiviral particles were produced by transfection of HEK293T cells. Murine primary T cells were activated by surface-coated anti-CD3 and anti-CD28 Abs and IL-2 (100 U/ml) for 24 h prior to the lentiviral transduction. Activated T cells (2 × 106) were transduced in lentiviral supernatant supplemented with protamine sulfate. After 24 h, cells were washed with PBS and further cultured in RPMI 1640 medium supplemented with IL-2 for additional 24 h. Mean transduction efficacies were high with 66 and 47% for control and miR-181a, respectively, as evaluated by flow cytometry of eGFP expression. RNA was collected for miRNA expression analysis.
Quantitative real-time PCR
Total RNA from mouse T cells was prepared using TRIzol (Invitrogen Life Technologies, Carlsbad, CA). Expression of miR-181a was determined by miR quantitative RT-PCR (qRT-PCR) using miRNA specific looped reverse-transcriptase primers and TaqMan probes (hsa-miR-181a, ID 000480), as recommended by the manufacturer (Applied Biosystems, Foster City, CA). Normalization was performed using the 2-ΔΔ cycle threshold method relative to U6 small nuclear RNA (snRNA).
Allogeneic BMT and GvHD induction
Allogeneic BMT (C57BL/6→BDF1) was performed, as previously described (16). Donor BM cells were isolated from femura and tibiae. For T cell isolation, T cells were stained with anti-CD3 biotin and were separated with streptavidin-magnetic beads by using Miltenyi Biotec MACS purification kit. Recipient BDF1 mice received 11 Gy total body irradiation ([137Cs]γ-source). Transduced C57BL/6 donor T cells and T cell–depleted (TCD)-BM cells were adoptively transplanted into recipients i.v. within 24 h after irradiation. Three days after transduction with either mir-181a or control lentiviruses, 2 × 106 FACS-sorted GFP+ donor T cells were used in BMT experiments to induce GvHD in recipient mice. For all other transfer experiments, 1 × 106 GFP+ donor T cells transduced with either miR-181a or control lentiviruses were used. Individual conditions in experiments are stated in the figure legends. Antibiotic treatment (Cotrimoxazol, Ratiopharm, Ulm, Germany) was provided in the first 2 wk posttransplantation. Survival of BMT recipients was monitored daily. Weight and clinical signs of aGvHD were scored two times per week (17).
Abs and flow cytometry
mAbs used for flow cytometry (FACS) were purchased from BD Pharmingen (San Jose, CA) and eBiosciences (San Diego, CA): CD3-PE (clone 17a2), CD4-PerCP (clone RM4-5), CD8a-allophycocyanin/Cy7 (clone 53-6.7), CD45.1 (Ly5.1)-allophycocyanin (clone A20), CD90.1 (Thy1.1)-PE/Cy7 (clone OX-7), and eGFP-Alexa Fluor 488 (polyclonal). Flow cytometric analysis was performed on a BD Pharmingen LSRII (San Jose, CA). FlowJo software (Ashland, OR) was used for data analysis.
In vitro and in vivo proliferation assay
T cell proliferation in vitro was determined by thymidine incorporation. In brief, T cells were stimulated with Dynabeads Mouse T-Activator CD3/CD28 (Invitrogen Life Technologies, Carlsbad, CA) for 2 d. Culture medium supplemented with 3H-thymidine (0.8 μCi/well) was added for additional 18-h incubation. MicroBeta radioactivity counter (PerkinElmer, Waltham, MA) was used to detect the thymidine uptake. To analyze T cell proliferation in vivo, recipient mice were given BrdU (1 mg/ml) in drinking water for 7 d after transplantation. Donor T cells were isolated from recipients to detect the BrdU incorporation using FACS analysis. Labeling with Cell Proliferation Dye eFluor670 (eBioscience, San Diego, CA) at a concentration of 5 μM or CFSE (Invitrogen-Life Technologies) at a concentration of 10 μM was used to monitor proliferation of eGFP+ T cells in vivo.
Donor T cell isolation from lymphoid organs and GvHD-target organs
Donor T cells were harvested from C57BL/6-Ly5.1 or -Thy1.1 mice for lentiviral transduction. Lethally irradiated BDF1 recipients received allogeneic donor BM cells and a 1:1 mixture of miR-181a–overexpressing or control-transduced T cells (with congenic markers Thy1.1+ or Ly5.1+) i.v. At day 7 after BMT, lymphocytes from recipient peripheral blood, lymph nodes (LN), spleen, BM, liver, small intestine, and colon were isolated. In brief, peripheral blood from hepatic vena cava or heart was collected directly into Heparin blood collection tubes. After RBC lysis, LN (including mesenteric, inguinal, axillary, and cervical LN) were mashed in PBS supplemented with 5% FCS. Spleens were mashed and RBC lysis was next performed to exclude erythrocytes. BM cells from femurs and tibias were flushed out by syringes filled with PBS supplemented with 5% FCS. RBC lysis was applied to remove RBCs. Liver tissues were sliced into 1- to 2-mm3 pieces in PBS and were treated with histolyticum collagenase (200 U/ml) and pancreas DNase I (100 U/ml) for 60 min at 37°C. After incubation, cells were filtered and were subjected to Percoll gradient separation. The interphase was recovered as the fraction of hepatic lymphocytes. Small intestines and colons (without Peyer’s patches) were opened longitudinally by slicing on the lumen. Fragmented tissues were incubated in RPMI 1640 culture medium supplemented with 5 mM EDTA for 20 min at 37°C. Cells were filtered, centrifuged, and resuspended in RPMI 1640 culture medium for Percoll gradient separation. The interphase was recovered as the fraction of intestinal intraepithelial lymphocytes. To collect lamina propria lymphocytes, tissue pieces were incubated in HBSS medium supplemented with 2 mM EDTA and 10% FCS for 30 min at 37°C. The tissues were next incubated in RPMI 1640 medium supplemented with collagenase (100 U/ml) for additional 30 min at 37°C. Cell suspensions were collected for the Percoll gradient separation. The interphase was recovered as the fraction of lamina propria lymphocytes. All samples were filtered through nylon meshes to create single-cell suspensions for FACS analysis.
Cell lysis and Western blot
Cell lysis and Western blot were carried out as previously described (18). Anti–BCL-2 (2876), BAX (2772), BCL-XL (2762), MCL-1 (4572), BIM (C34C5), COX IV (6B3), and GAPDH (14C10, 2118) Abs were purchased from Cell Signaling Technology (Danvers, MA). Anti-BAD (62465) and COXIV (6B3, 33985) Abs were purchased from Abcam (Cambridge, U.K.). HRP-conjugated secondary Ab was obtained from Roche (Basel, Switzerland). Immunoreactive bands were visualized using the Western Lightning Plus-ECL Detection System (Perkin Elmer, Waltham, MA) and were imaged by exposure to x-ray film. Densitometric analysis of x-ray films was performed using Image Studio Lite Software (LI-COR Biosciences, Lincoln, NE). The intensity ratio of the protein of interest band to the COX IV or GAPDH band (loading control) was calculated to measure changes in protein levels.
Statistical analysis
Statistical analysis was performed with Excel (Microsoft, Redmond, WA) and Prism 4 (GraphPad). Significances were calculated by using Mann–Whitney U test. The p values ≤0.05 were considered statistically significant. Survival data were analyzed using log-rank test (GraphPad, La Jolla, CA).
Results
Endogenous and transgenic miR-181a expression in primary murine T cells
Because miR-181a has been shown to modulate TCR signaling strength, we analyzed its expression in resting and activated T cells after stimulation with anti-CD3, anti-CD28 Abs, and IL-2 by miR qRT-PCR. miR-181a expression in both CD4+ and CD8+ cells continuously declines during stimulation and drops to ∼10% of the initial level after 4 d (Fig. 1A). For stable overexpression of miR-181a, we developed a protocol for lentiviral transduction of primary T cells, as schematically shown in Fig. 1B. T cells were harvested and stimulated for 24 h before transduction with recombinant lentiviruses encoding miR-181a (Fig. 1C). Two days after lentiviral transduction, the transduction efficacy into primary T cells was >40% for miR-181a and >60% for the control virus (Fig. 1D); miR qRT-PCR revealed an increase of miR-181a expression by ∼230-fold as compared with controls (Fig. 1E). In contrast, miR-181a expression is reduced in primary T cells from miR-181a/b-1−/− as compared with heterozygous miR-181a/b-1+/− mice (Fig. 1F). Residual levels of miR-181a in T cells isolated from miR-181a/b-1−/− mice detected in this assay might be expressed from the miR-181a/b-2 gene, although it is widely assumed that at least in thymocytes the miR-181a/b-2 gene does not significantly contribute to expression of miR-181a (14). Alternatively, we cannot exclude that in the absence of miR-181a this assay might detect miR-181c, which displays a 1-nt difference when compared with miR-181a. We next overexpressed miR-181a in heterozygous and miR-181a/b-1−/− T cells and analyzed its expression levels normalized to U6 snRNA (ΔCT; Fig. 1G). Whereas normalized miR-181a expression differs strongly between control-transduced heterozygous and miR-181a/b-1−/− T cells, as seen before, this difference is abolished upon lentiviral overexpression of miR-181a, resulting in a nearly identical miR-181a/U6 snRNA ratio. Based on these results, we used heterozygous T cells for lentiviral overexpression and refer T cells overexpressing miR-181a and those from miR-181a/b-1−/− mice as miR-181a high and miR-181 low cells, respectively.
miR-181a expression in murine T cells. (A) miR-181a expression in purified CD4+ and CD8+ T cells stimulated with anti-CD3, anti-CD28 Abs, and IL-2 for 8–96 h. miR RT-PCR data are shown as mean ± SEM from three independent experiments. (B) Scheme of lentiviral gene transfer into primary T cells. (C) Cartoon of the lentiviral transgene plasmid encoding the miR-181a expression cassette. (D) Transduction efficacy in murine primary T cells 2 d after lentiviral transduction. Data represent the mean ratio of GFP+ cells ± SEM from three independent experiments. (E) Expression of miR-181a in transduced T cells 2 d after lentiviral transduction. Data represent the mean of miR-181a expression as compared with U6 snRNA ± SEM from three independent experiments. (F) Expression of miR-181a in miR-181alow T cells. T cells from miR-181a/b-1+/− mice served as control. Data represent the mean of miR-181a expression as compared with U6 snRNA ± SEM from three independent experiments. (G) Differential expression of miR-181a in heterozygous (miR-181a/b-1+/−) and miR-181alow (miR-181a/b-1−/−) T cells after miR-181a overexpression. T cells harvested from heterozygous and miR-181alow mice were lentiviral transduced with either control- or miR-181a–encoding lentiviruses. miR qRT-PCR was employed to determine miR-181a and U6 snRNA expression in the GFP+ sorted T cells. Data represent mean ∆cycle threshold of miR-181a and U6 snRNA expression ± SEM from two independent experiments, each in duplicate (total n = 4). Higher ∆cycle thresholds indicate lower miR-181a expression (***p ≤ 0.001).
miR-181a expression in murine T cells. (A) miR-181a expression in purified CD4+ and CD8+ T cells stimulated with anti-CD3, anti-CD28 Abs, and IL-2 for 8–96 h. miR RT-PCR data are shown as mean ± SEM from three independent experiments. (B) Scheme of lentiviral gene transfer into primary T cells. (C) Cartoon of the lentiviral transgene plasmid encoding the miR-181a expression cassette. (D) Transduction efficacy in murine primary T cells 2 d after lentiviral transduction. Data represent the mean ratio of GFP+ cells ± SEM from three independent experiments. (E) Expression of miR-181a in transduced T cells 2 d after lentiviral transduction. Data represent the mean of miR-181a expression as compared with U6 snRNA ± SEM from three independent experiments. (F) Expression of miR-181a in miR-181alow T cells. T cells from miR-181a/b-1+/− mice served as control. Data represent the mean of miR-181a expression as compared with U6 snRNA ± SEM from three independent experiments. (G) Differential expression of miR-181a in heterozygous (miR-181a/b-1+/−) and miR-181alow (miR-181a/b-1−/−) T cells after miR-181a overexpression. T cells harvested from heterozygous and miR-181alow mice were lentiviral transduced with either control- or miR-181a–encoding lentiviruses. miR qRT-PCR was employed to determine miR-181a and U6 snRNA expression in the GFP+ sorted T cells. Data represent mean ∆cycle threshold of miR-181a and U6 snRNA expression ± SEM from two independent experiments, each in duplicate (total n = 4). Higher ∆cycle thresholds indicate lower miR-181a expression (***p ≤ 0.001).
Impact of miR-181a expression in donor T cells on aGvHD
To study the impact of miR-181a on the development of aGvHD, control and miR-181a high donor T cells were transplanted into lethally irradiated BDF1 recipients together with TCD-BM, and mice were clinically monitored over time. Both groups of mice, receiving either miR-181a high donor T cells or TCD-BM, did not develop aGvHD, and all mice survived during the time of these experiments (Fig. 2A). In contrast, all mice transplanted with control-transduced T cells developed severe aGvHD, as reflected by an increased clinical GvHD score, and died within 10 wk after BMT (Fig. 2A, 2C). Accordingly, mice transplanted with either TCD-BM– or miR-181a–transduced T cells recovered from weight loss after transplantation, whereas mice transplanted with control-transduced donor T cells did not (Fig. 2B).
Course of acute GvHD induced by miR-181a high and low donor T cells. Graphs show (A) survival, (B) body weight, and (C) clinical score of BMT recipients (BDF1) transplanted with donor (C57BL/6) TCD-BM alone or in combination with 2 × 106 FACS-sorted GFP+ control or 2 × 106 miR-181ahigh T cells. Data are pooled from two independent experiments. Log-rank test was used for statistical survival analyses (***p ≤ 0.001). (D–F) Show survival, body weight, and clinical score of recipients (BDF1) adoptively transplanted with donor (C57BL/6) TCD-BM alone or in combination with 1 × 106 miR-181a/b-1+/− control or 1 × 106 miR-181alow T cells. Data are summarized from four independent experiments. Log-rank test was used for statistical survival analyses (*p ≤ 0.05). (G) Recovery of transduced donor T cells from peripheral blood at day 40 after transplantation. Data show the mean ratio ± SEM of eGFP+ T cells from two independent experiments.
Course of acute GvHD induced by miR-181a high and low donor T cells. Graphs show (A) survival, (B) body weight, and (C) clinical score of BMT recipients (BDF1) transplanted with donor (C57BL/6) TCD-BM alone or in combination with 2 × 106 FACS-sorted GFP+ control or 2 × 106 miR-181ahigh T cells. Data are pooled from two independent experiments. Log-rank test was used for statistical survival analyses (***p ≤ 0.001). (D–F) Show survival, body weight, and clinical score of recipients (BDF1) adoptively transplanted with donor (C57BL/6) TCD-BM alone or in combination with 1 × 106 miR-181a/b-1+/− control or 1 × 106 miR-181alow T cells. Data are summarized from four independent experiments. Log-rank test was used for statistical survival analyses (*p ≤ 0.05). (G) Recovery of transduced donor T cells from peripheral blood at day 40 after transplantation. Data show the mean ratio ± SEM of eGFP+ T cells from two independent experiments.
We then analyzed miR-181a loss-of-function phenotypes and performed identical experiments with miR-181a low donor T cells. As shown in Fig. 2D, miR-181a low T cells showed aGvHD development with a similar clinical GvHD score (Fig. 2F) as compared with controls, but a significantly shorter survival (Fig. 2D). Accordingly, both groups of mice suffering from aGvHD did not recover from irradiation-induced weight loss, whereas mice transplanted with TCD-BM only regained their body weight within 3 wk (Fig. 2E).
Finally, we analyzed the recovery of miR-181a high and control-transduced T cells in peripheral blood after transplantation. At day 40 after BMT, only very few miR-181a high CD4+ and CD8+ cells were detectable by flow cytometry by eGFP fluorescence. In contrast, the levels of eGFP+ control-transduced CD4+ and CD8+ are detectable with 26 and 54% of CD4+ and CD8+ cells, respectively (Fig. 2G). These data suggest that overexpression of miR-181a results in limited proliferation and/or survival of transplanted T cells.
Effects of miR-181a expression levels on expansion of primary T cells in vitro and in vivo
To study potential mechanisms leading to prevention of aGvHD by overexpression of miR-181a, we next analyzed the proliferation of miR-181a high T cells in vitro. As shown in Fig. 3A, control-transduced and miR-181a high cells showed almost identical expansion kinetics up to 15 d after lentiviral transduction in cultures supplemented with IL-2 only. In contrast, restimulation after 5 d with anti-CD3 and anti-CD28 beads revealed a dose-dependent reduction in proliferating cells in the presence of high miR-181a expression as compared with controls (Fig. 3B). These data indicate that overexpression of miR-181a interferes with T cell expansion after recurrent stimulation. Given that these cells retained their capacity to expand without restimulation, these data suggest that ectopic expression of miR-181a affects T cell survival rather than proliferation.
Cell expansion of miR-181a high T cells in vitro. (A) Cell numbers of FACS-sorted eGFP+ transduced T cells cultured in IL-2–supplemented medium. Total cell numbers were counted over time, and data represent the mean ± SEM from two independent experiments. (B) Proliferation of transduced T cells after restimulation (after 5 d). Lentivirally transduced FACS-sorted eGFP+ T cells were cultured with IL-2 for 5 d only before restimulation with increasing concentrations of anti-CD3 and anti-CD28 beads. Thymidine (0.8 μCi) was added at 18 h prior to the detection. Data represent mean thymidine incorporation ± SEM from two independent experiments, each in triplicate (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001).
Cell expansion of miR-181a high T cells in vitro. (A) Cell numbers of FACS-sorted eGFP+ transduced T cells cultured in IL-2–supplemented medium. Total cell numbers were counted over time, and data represent the mean ± SEM from two independent experiments. (B) Proliferation of transduced T cells after restimulation (after 5 d). Lentivirally transduced FACS-sorted eGFP+ T cells were cultured with IL-2 for 5 d only before restimulation with increasing concentrations of anti-CD3 and anti-CD28 beads. Thymidine (0.8 μCi) was added at 18 h prior to the detection. Data represent mean thymidine incorporation ± SEM from two independent experiments, each in triplicate (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001).
To track allogeneic T cells in vivo, we used congenic markers (Thy1 and Ly5) of C57BL/6 donor T cells to differentially monitor control and miR-181a high cells (Fig. 4A). As shown in Fig. 4B, the ratio of miR-181a high CD4+ and CD8+ T cells is reduced by ∼44–74% as compared with controls in peripheral LN, spleen, BM, and blood 7 d after transplantation. Similar but less pronounced results were found for recovery of control and miR-181a high CD4+ and CD8+ cells from aGvHD target organs such as liver, small intestine, and colon (Fig. 4C). Analysis of cell proliferation in vivo by eFluor labeling (Supplemental Fig. 1) did not reveal significant differences in proliferation of miR-181a high as compared with control-transduced T cells after T cell transfer. Accordingly, proliferation of miR-181a low T cells was also not reduced in this assay (Supplemental Fig. 2). These data suggest that the small differences in proliferation rates may not be sufficient to explain the observed reduction of miR-181a high T cell expansion upon activation, thus further supporting the hypothesis that miR-181a controls T cell survival in this context.
Homing of donor T cells to secondary lymphoid organs and GvHD targets organs. (A) Cartoon of the experimental setup. Equal amounts (1 × 106 each) of donor eGFP+ control- and miR-181a–transduced T cells were adoptively transferred into recipients (without prior sorting), and donor T cells were recovered from recipients’ secondary lymphoid organs and GvHD target organs after 7 d. (B) Ratios of the absolute cell numbers of CD4+ (left) and CD8+ (right) eGFP+ miR-181ahigh T cells against control T cells (miR-181ahigh:control) in secondary lymphoid organs. Total lymphocytes were harvested from LN, spleen (SPL), BM, and blood at day 7 after transplantation. The absolute cell number of donor control (GFP+Ly5.1+) and miR-181ahigh (GFP+Thy1.1+) T cells from each organ was determined by FACS analysis. (C) Ratios of the absolute cell numbers of CD4+ (left) and CD8+ (right) eGFP+ miR-181ahigh T cells against control T cells (miR-181ahigh:control) in GvHD target organs liver, small intestine (SI), and colon (CO), including fractions of intraepithelial lymphocytes (IEL) and lamina propria lymphocytes (LPL). Total lymphocytes were harvested from individual organ after 7 d, and the absolute cell number of donor control (GFP+Ly5.1+) and miR-181ahigh (GFP+Thy1.1+) T cells was determined by FACS analysis. Data represent the mean ± SEM for both input cells (before transplantation) and output cells (after isolation from each organ) from two independent experiments (n = 4). Ratio <1 indicates less presence of miR-181ahigh T cells than control T cells in individual organ.
Homing of donor T cells to secondary lymphoid organs and GvHD targets organs. (A) Cartoon of the experimental setup. Equal amounts (1 × 106 each) of donor eGFP+ control- and miR-181a–transduced T cells were adoptively transferred into recipients (without prior sorting), and donor T cells were recovered from recipients’ secondary lymphoid organs and GvHD target organs after 7 d. (B) Ratios of the absolute cell numbers of CD4+ (left) and CD8+ (right) eGFP+ miR-181ahigh T cells against control T cells (miR-181ahigh:control) in secondary lymphoid organs. Total lymphocytes were harvested from LN, spleen (SPL), BM, and blood at day 7 after transplantation. The absolute cell number of donor control (GFP+Ly5.1+) and miR-181ahigh (GFP+Thy1.1+) T cells from each organ was determined by FACS analysis. (C) Ratios of the absolute cell numbers of CD4+ (left) and CD8+ (right) eGFP+ miR-181ahigh T cells against control T cells (miR-181ahigh:control) in GvHD target organs liver, small intestine (SI), and colon (CO), including fractions of intraepithelial lymphocytes (IEL) and lamina propria lymphocytes (LPL). Total lymphocytes were harvested from individual organ after 7 d, and the absolute cell number of donor control (GFP+Ly5.1+) and miR-181ahigh (GFP+Thy1.1+) T cells was determined by FACS analysis. Data represent the mean ± SEM for both input cells (before transplantation) and output cells (after isolation from each organ) from two independent experiments (n = 4). Ratio <1 indicates less presence of miR-181ahigh T cells than control T cells in individual organ.
miR-181a targets in T cells
Several miR-181a targets have been identified that may be relevant for T cell expansion upon activation, including PTEN (13) and BCL-2 (19). Therefore, we analyzed the impact of miR-181a overexpression on BCL-2 protein expression in murine and human T cell lines. As shown in Fig. 5, overexpression of miR-181a reduces BCL-2 protein expression in murine EL4 and human JURKAT cells by ∼50 and 20%, respectively. We next analyzed protein expression of other pro- and anti-apoptotic BCL-2 family members in JURKAT cells upon miR-181a overexpression. As shown in Fig. 5B, protein expression of BCL-2, MCL-1, and BAX is slightly reduced, whereas that of BCL-XL, BAD, and BIM remained unchanged. Beside BCL-2, regulation of MCL-1 and BAX protein expression by miR-181a has also been shown in astrocytes and breast cancer cells, respectively (20, 21). These data suggest that regulation of the intrinsic apoptotic pathway by miR-181a may contribute to the inhibitory effects of miR-181a overexpression on T cell alloreactivity, although the precise mechanism and its contribution to the effects observed remain to be determined.
Effects of miR-181a overexpression on protein expression of BCL-2 family members. (A) Murine EL4 T cells were lentivirally transduced, and FACS-sorted eGFP+ cells were analyzed for BCL-2 expression by Western blotting. Densitometry analysis indicates relative BCL-2 protein levels from one representative of two independent experiments. (B) Human Jurkat T cells were lentivirally transduced, and FACS-sorted eGFP+ cells were analyzed for BCL-2, BCL-XL, MCL-1, BAD, BAX, and BIM expression by Western blotting. Densitometry analysis indicates relative protein levels from one representative of three independent experiments.
Effects of miR-181a overexpression on protein expression of BCL-2 family members. (A) Murine EL4 T cells were lentivirally transduced, and FACS-sorted eGFP+ cells were analyzed for BCL-2 expression by Western blotting. Densitometry analysis indicates relative BCL-2 protein levels from one representative of two independent experiments. (B) Human Jurkat T cells were lentivirally transduced, and FACS-sorted eGFP+ cells were analyzed for BCL-2, BCL-XL, MCL-1, BAD, BAX, and BIM expression by Western blotting. Densitometry analysis indicates relative protein levels from one representative of three independent experiments.
Discussion
In this work, we demonstrate that overexpression of miR-181a in primary T cells inhibits their response to nonspecific stimulation in vitro and their alloreactivity in vivo. Whereas numbers of control and miR-181a high T cells are almost identical in cultures without further stimulation after lentiviral gene transfer (Fig. 3A), recurrent stimulation via TCR and IL-2 restricts expansion of miR-181a high cells in a dose-dependent manner (Fig. 3B). This effect is even stronger in vivo 40 d after allogeneic BMT (Fig. 2G), and this might be due to the fact that over time and in the GvHD situation, an outgrowth of dominant T cell clones over disadvantaged (miR-181ahigh) clones occurs. Interestingly, endogenous miR-181a expression is markedly reduced over time in both CD4+ and CD8+ cells upon unspecific stimulation (Fig. 1A). These data are in line with data from stimulated human T cells (22), suggesting that downregulation of miR-181a after T cell stimulation may have some physiological relevance for T cell expansion and function. Accordingly, in vivo recovery of miR-181a high T cells from both secondary lymphoid organs and GvHD target organs is impaired as compared with controls (Fig. 4). Although we could not demonstrate reduced proliferation of miR-181a high T cells in vivo, we cannot exclude such an effect contributing to the observed phenotype. However, reduced recovery of miR-181 high T cells could also be due to limited survival after lentiviral transduction. In line with these results, overexpression of miR-181a in donor T cells prevents otherwise lethal GvHD, and the clinical course of mice transplanted with miR-181a high donor T cells is indistinguishable from that of mice transplanted with TCD-BM grafts (Fig. 2A–C). In contrast, miR-181a low donor T cells accelerate aGvHD with shorter survival of recipient mice (Fig. 2D). These data clearly indicate that constitutive overexpression of miR-181a in T cells interferes with T cell expansion upon stimulation in vitro and in vivo.
The molecular mechanism as to how miR-181a reduces T cell expansion after stimulation is currently not completely understood. It is well established that modulation of expression of miR-181 affects TCR signaling strength, most prominently demonstrated at the level of Ca flux (8, 10, 11). Li et al. (8) identified tyrosine phosphatases SHP-2 and PTPN-22 and ERK-specific phosphatases DUSP6 and DUSP5 as miR-181a targets in T cells. These data are in line with enhanced propagation of TCR signal transduction by repression of phosphatases downstream of the TCR. On the other side, BCL-2, an anti-apoptotic protein in the mitochondrial outer membrane, has been identified as a miR-181a target in T cells (19). In addition, other members of the BCL-2 family of intrinsic apoptosis regulators such as the proapoptotic BAX and the anti-apoptotic MCL-1 have been found to be regulated by miR-181a in astrocytes and breast cancer cells (20, 21). In line with these data, we in this study demonstrate in murine and/or human miR-181a–transduced T cells some reduction of BCL-2, MCL-1, and BAX protein expression. These data may suggest that miR-181a could contribute to T cell activation and expansion by regulation of intrinsic apoptosis. Because the impact of mitochondrial proteins on clonal expansion of activated T cells has been described (23), the regulation of different targets in functionally different pathways by miR-181a (phosphatases downstream of TCR, regulators of mitochondrial apoptosis, and PTEN among others) may result in a more complex phenotype, although the relative importance of each individual pathway remains to be determined. Interestingly, Wu et al. (6) observed BCL-2 and PTEN as a target of the miR-17-92 cluster. Whereas the threshold for TCR signaling may be lowered by miR-181a, its expression has to be limited to maintain the balance of pro- and anti-apoptotic signals in activated T cells. These combined effects may explain the activation-dependent inhibition of T cell expansion and homing in vitro and in vivo.
Furthermore, the fine-tuning of intrinsic apoptosis depending on TCR signaling as described by Wensveen et al. (23) may be important for the impact of miR-181a expression on both GvHD and GvL effects. Although miR-181a expression is reduced at least upon stimulation with anti-CD3/CD28 beads (Fig. 1A) (22), specific effects of allostimulation on miR-181a expression have not yet been described. Interestingly, Sun et al. (24) recently reported a set of 44 miRNAs specifically enriched after allo- as compared with anti-CD3/CD28 and syngeneic stimulation using AGO-CLIP-CHIP. Therefore, differential miRNA expression may be important for the balance of GvHD and GvL effects in miR-181a high cells depending on the TCR signaling strength and kinetic of allostimulation as compared with potentially weaker and later signals relevant for GvL effects.
In summary, our data indicate that understanding miR-181a function in T cells may provide a new approach to identify druggable targets for prevention and/or treatment of aGvHD. Although suitable tools to transiently and specifically enhance miRNA expression in T cells are not yet available, increased expression and enhanced function of miR-181a may induce depletion of activated alloreactive T cells. This approach needs extensive further evaluation in preclinical models to add new therapeutic options to prevent and treat acute GvHD in the future.
Acknowledgements
We acknowledge the assistance of the Cell Sorting Core Facility of the Hannover Medical School, supported in part by Braukmann-Wittenberg-Herz-Stiftung and German Research Foundation. We thank Natalia Ziętara and Marcin Łyszkiewicz for providing miR-181a/b-1−/− and miR-181a/b-1+/− mice.
Footnotes
This work was supported by Deutsche Forschungsgemeinschaft Grants SFB738-A7 (to A.K.), SFB-738-A8 (to M.E. and C.K.), SFB738-B5 (to R.F.), KR2320/2-1 (to A.K.), KR2320/3-1 (to A.K.), and EXC62 (to A.K.); Unit 6.7 (to M.E. and M.S.); “Rebirth,” Deutsche Krebshilfe 109451 (to M.E., M.S., and C.K.); and H.W. and J. Hector-Stiftung M49 (to M.S. and M.E.).
The online version of this article contains supplemental material.
References
Disclosures
The authors have no financial conflicts of interest.