Supplementary Table S1. Primers used in this study. Supplementary Table S2. A/B Coverage of TCR-α (A) and TCR-β (B) variable genes with clonotype-specific primers. Supplementary Table S3. Characteristics of cloned TCRs. Supplementary Figure S4. Flow cytometric sorting of pp65-specific CD8+ T cells from CMV-seropositive donor ID3 after one week of expansion. Supplementary Table S5. Synthetic peptides used in this study. Supplementary Figure S6. Functionality of TCRCD8-NY#5 in CD4+ and CD8+ T cells. Supplementary Figure S7. Functionality of NY-ESO-1-specific CD4-TCRs in CD8+ T cells. Supplementary Table S8. HLA haplotypes from healthy donors and NSCLC patients. Supplementary Methods. Detailed materials and methods.

Abstract The determination of the epitope specificity of disease-associated T-cell responses is relevant for the development of biomarkers and targeted immunotherapies against cancer, autoimmune, and infectious diseases. The lack of known T-cell epitopes and corresponding T-cell receptors (TCR) for novel antigens hinders the efficient development and monitoring of new therapies. We developed an integrated approach for the systematic retrieval and functional characterization of TCRs from single antigen-reactive T cells that includes the identification of epitope specificity. This is accomplished through the rapid cloning of full-length TCR-α and TCR-β chains directly from single antigen-specific CD8+ or CD4+ T lymphocytes. The functional validation of cloned TCRs is conducted using in vitro–transcribed RNA transfer for expression of TCRs in T cells and HLA molecules in antigen-presenting cells. This method avoids the work and bias associated with repetitive cycles of in vitro T-cell stimulation, and enables fast characterization of antigen-specific T-cell responses. We applied this strategy to viral and tumor-associated antigens (TAA), resulting in the retrieval of 56 unique functional antigen-specific TCRs from human CD8+ and CD4+ T cells (13 specific for CMV-pp65, 16 specific for the well-known TAA NY-ESO-1, and 27 for the novel TAA TPTE), which are directed against 39 different epitopes. The proof-of-concept studies with TAAs NY-ESO-1 and TPTE revealed multiple novel TCR specificities. Our approach enables the rational development of immunotherapy strategies by providing antigen-specific TCRs and immunogenic epitopes. Cancer Immunol Res; 2(12); 1230–44. ©2014 AACR.

Donor lymphocyte infusions (DLI) are used to resolve mixed T-cell chimerism (TCC) after allo-SCT despite a substantial risk of GVHD. We analyzed the impact of prophylactic CD8-depleted (CD8(depl)) DLI in 20 recipients of anti-CD52 alemtuzumab in vivo T-cell-depleted allografts with declining donor TCC after day +60. A total of 13 patients received CD8(depl) DLI and 7 patients did not. All but one of the DLI patients converted to complete donor T-cell chimeras, whereas only one non-DLI patient converted spontaneously. DLI induced transient acute GVHD in five and extensive chronic GVHD in two patients. These data suggest the use of CD8(depl) DLI as an effective treatment for mixed TCC, particularly in patients at high risk for GVHD. We also observed that the majority of reconstituting donor-derived T cells after alemtuzumab conditioning were CD52-negative. CD8(depl) DLI significantly increased the proportion of CD52-positive CD4 T cells, whereby their beneficial effect on reconstituting the post-transplant T-cell repertoire was shown.

The human CD52 molecule is the target of the monoclonal antibody Alemtuzumab, which is used for treating patients with chemo-refractory chronic lymphocytic leukemia as well as for T cell depletion (TCD) in the context of allogeneic hematopoietic stem cell transplantation (HSCT). The molecule is expressed on the surface of lymphocytes, dendritic cells and to a lesser extent on blood-derived monocytes. Previously, investigators have demonstrated that the surface expression of CD52 on T cells is down-regulated after in vitro incubation with Alemtuzumab. By treating purified human CD4 T cells over 4 hours with 10 μg/mL Alemtuzumab in medium supplemented with 10% human AB serum in vitro, we observed a strong decrease of CD52 expression by flow cytometry with a maximum 3–7 days after incubation. The CD52 down-regulation was also found at weaker intensity on CD8 T cells. From previous studies in chronic lymphocytic leukemia patients, it is known that Alemtuzumab treatment also leads to a down-regulation of CD52 on T cells in vivo. However, similar experiments have not been performed in allogeneic HSCT patients receiving Alemtuzumab in vivo for T cell depletion. We therefore analyzed the expression of CD52 on human peripheral blood mononuclear cells isolated at repeated time points from 22 allogeneic HSCT patients after reduced-intensity conditioning with fludarabine and melphalan and in vivo T cell depletion with Alemtuzumab (100 mg). Half of the patients received prophylactic CD8-depleted donor lymphocyte infusions (DLI) to promote immune reconstitution. By flow cytometry, we observed that the CD52 expression on monocytes, B cells, and natural killer cells remained unaltered after transplantation and was not influenced by the application of DLI. In contrast, the majority of CD4 T cells were CD52-negative (median, 72%) after transplantation and they remained CD52-negative in patients who did not receive DLI throughout the first year after HSCT. The permanent lack of CD52 expression could not be explained by a continuous effect of Alemtuzumab, because earlier studies have shown that the antibody is not present in active plasma concentrations beyond day +60 after HSCT. In contrast, patients receiving CD8-depleted DLI demonstrated a significant increase in the proportion of CD52-positive CD4 T cells. In three of our patients (DLI: n=2, non-DLI: n=1) we analyzed the donor chimerism of CD52-positive and CD52-negative CD4 T cells sorted with high purity by flow cytometry. Three months after HSCT (before DLI), the proportion of donor T cells was clearly higher among the CD52-negative compared to the small proportion of CD52-positive cells in all patients (44% vs. 10%, 83% vs. 0%, and 100% vs. 40%). In the patient who did not receive DLI, the donor T cell chimerism remained mixed in the CD52-negative and CD52-positive fractions on days 200 (CD52-negative: 95%; CD52-positive: 15%) and 350 (CD52-negative: 92%; CD52-positive: 65%). In contrast, the two patients receiving CD8-depleted DLI showed a strong increase in the proportion of CD52-positive CD4 T cells that were of complete donor origin. Altogether, CD52 is permanently down-regulated in reconstituting CD4 T cells following HSCT with an Alemtuzumab-based TCD regimen unless DLI are applied. Our data support the idea of an active mechanism for CD52 down-regulation in CD4 T cells that is not related to B cells and natural killer cells and that appears to differently affect donor and host T cells, respectively.

The interpretation of the results obtained from immunomonitoring of clinical trials is a difficult task due to the variety of methods and protocols available to detect vaccine-specific T-cell responses. This heterogeneity as well as the lack of standards has led to significant scepticism towards published results. In February 2005, a working group was therefore founded under the aegis of the Association for Immunotherapy of Cancer (“CIMT”) in order to compare techniques and protocols applied for the enumeration of antigen-specific T-cell responses. Here we present the results from two consecutive phases of an international inter-laboratory testing project referred to as the “CIMT monitoring panel”. A total of 13 centers from six European countries participated in the study in which pre-tested PBMC samples, synthetic peptides and PE-conjugated HLA-tetramers were prepared centrally and distributed to participants. All were asked to determine the number of antigen-specific T-cells in each sample using tetramer staining and one functional assay. The results of the first testing round revealed that the total number of cells analyzed was the most important determinant for the sensitive detection of antigen-specific CD8 + T-cells by tetramer staining. Analysis by ELISPOT was influenced by a combination of cell number and a resting phase after thawing of peripheral blood mononuclear cells. Therefore, the experiments were repeated in a second phase but now the participants were asked to change their protocols according to the new guidelines distilled from the results of the first phase. The recommendations improved the number of antigen-specific T-cell responses that were detected and decreased the variability between the laboratories. We conclude that a two-step approach in inter-laboratory testing allows the identification of distinct variables that influence the sensitivity of different T-cell assays and to formally show that a defined correction to the protocols successfully increases the sensitivity and reduces the inter-center variability. Such “two-step” inter-laboratory projects could define rational bases for accepted international guidelines and thereby lead to the harmonization of the techniques used for immune monitoring.

Upon stimulation with a wide range of concentrations of CpG oligodeoxynucleotide 2216 (CpG 2216), plasmacytoid DC are induced to produce type I IFN (IFN-alpha/beta). In contrast, CpG 1668 shows a bell-shaped dose-response correlation, i.e. only intermediate but not high doses of CpG 1668 induce IFN-alpha/beta. interestingly, high-dose CpG 1668 completely inhibited IFN-alpha responses induced by CpG 2216. Experiments using supernatant of high-dose CpG-1668-treated cells indicated that secreted inhibitor(s) mediated the IFN-alpha shut-off. Among modulating cytokines, IL-10 turned out to be one important negative regulator. In line with this, supernatants of IL-10-deficient DC cultures stimulated with high-dose CpG 1668 did not inhibit IFN-alpha production. interestingly, high-dose CpG 1668 also inhibited IFN-alpha responses induced by the DNA-encoded mouse cytomegalovirus, whereas IFN-alpha responses induced by negative-strand RNA-encoded vesicular stomatitis virus were only marginally affected. Experiments with DC cultures devoid of TLR9 indicated that TLR9 was critically required to mediate stimulatory and modulatory signals by low and high concentrations of CpG 1668, respectively. Analysis of purified DC subsets showed that conventional DC were the main IL-10 producers, whereas plasmacytoid DC hardly produced any IL-10.

Current methods for the detection and isolation of antigen-specific CD4+ and CD8+ T cells require the availability of peptide/MHC multimers or are restricted to cells that produce cytokines after antigen contact. Here we show that de novo cell surface expression of the TNF-receptor family member CD137 (4-1BB) identifies recently activated, but not resting, human CD4+ and CD8+ memory T cells. Maximum CD137 expression level is uniformly observed in both T-cell subsets at 24h after stimulation with antigen. In experiments with CMV and EBV-reactive T cells, we confirmed the specificity of CD137 expression by co-staining with peptide/HLA tetramers. Substantial proportions of CD137+ T cells did not produce IFN-γ, suggesting that CD137 detects a broader repertoire of antigen-specific T cells. Activated CD137+ T cells could be easily purified by MACS and expanded in vitro thereafter. This CD137-based enrichment method was capable of isolating 2-fold higher numbers of anti-viral CD4+ and CD8+ T cells compared to the IFN-γ secretion assay. In conclusion, antigen-triggered CD137 expression allows the rapid detection and sorting of virus-reactive CD4+ and CD8+ T cells. The CD137 assay is most attractive for the simultaneous targeting of anti-viral T helper and effector cells in monitoring studies and adoptive immunotherapy trials.

Acute graft-versus-host disease (aGVHD) is a life-threatening complication after solid-organ transplantation, which is mediated by host-reactive donor T cells emigrating from the allograft. We report on two liver transplant recipients who developed an almost complete donor chimerism in peripheral blood and bone marrow-infiltrating T cells during aGVHD. By analyzing these T cells directly ex vivo, we found that they died by apoptosis over time without evidence of rejection by host T cells. The host-versus-donor reactivity was selectively impaired, as anti-third-party and antiviral T cells were still detectable in the host repertoire. These findings support the acquired donor-specific allotolerance concept previously established in animal transplantation studies. We also observed that the resolution of aGVHD was not accompanied by an expansion of circulating immunosuppressive CD4/CD25/FoxP3-positive T cells. In fact, graft-versus-host-reactive T cells were controlled by an alternative negative regulatory pathway, executed by the programmed death (PD)-1 receptor and its ligand PD-L1. We found high PD-1 expression on donor CD4 and CD8 T cells. In addition, blocking PD-L1 on host-derived cells significantly enhanced alloreactivity by CD8 T cells in vitro. We suggest the interference with the PD-1/PD-L1 pathway as a therapeutic strategy to control graft-versus-host-reactive T cells in allograft recipients.

As shown previously, encapsulation of a peptide derived from tyrosinase-related protein2 (TRP2) into liposomes (artificial virus envelope (AVE) 3) resulted in combination with CpG-oligodeoxynucleotides in the induction of higher numbers of antigen-specific T cells compared to vaccination with free TRP2.Here, we present further data with regard to optimal antigen dose, the relevance of vaccine injection site and on the T cell stimulatory synergism of liposomal adjuvant combinations.Compared to an aqueous solution liposomal TRP2 was more potent in the induction of TRP2-specific T cells at an optimal dose but showed a narrow dose optimum with profoundly impaired T cell responses at higher vaccine doses.Higher T cell numbers were induced when mice were vaccinated into their hint foodpads compared to intradermal vaccination, the site used routinely in murine tumor vaccination models.A synergistic adjuvant effect was observed when CpG-oligodeoxynucleotides were admixed with liposomal monophosphoryl lipid A (MPLA) and the lipopeptide Pam 3 Cys, respectively.In summary our data demonstrate that liposomes as carriers for peptide-antigen and adjuvant induce a strong antigen-specific T cell response and are superior over vaccine formulations composed of free peptide and adjuvant.

Skin is the most frequently affected organ in acute graft versus host disease (GVHD). Data from murine studies support the hypothesis that the interaction of residing host Langerhans cells of the epidermis (LC) and donor T cells is crucial for the initiation of acute GVHD. Donor T cells are also necessary to induce the switch of LC from host to donor origin after allogeneic stem cell transplantation (SCT). In an ongoing clinical protocol applying alemtuzumab-based T cell depleted (TCD) allogeneic SCT (Meyer, Blood 2007; 109:374), we observed acute skin GVHD occurring early after transplantation. We therefore intended to analyse the LC chimerism in patients undergoing this protocol. So far, LC-chimerism analysis in humans has been performed by the detection of the Y-chromosome restricting it to sex-mismatched donor/recipient pairs. Here we introduce a new method to isolate LC from small skin samples at high purity for a sensitive STR-based chimerism analysis of general applicability. Epidermal skin layers were obtained from 6 mm punch biopsies by dispase I digestion. A small slice of epidermis was used for immunofluorescent staining. The remaining sample was digested by trypsin, and CD1a/MHC-class II-positive LC were sorted by flow cytometry. This approach resulted in a mean purity of > 96% with skin of healthy individuals. However, the density of LC early after SCT following non-TCD myeloablative regimens had previously been shown to be much lower compared to healthy individuals. By CD1a-staining, we were able to show that this is also the case after TCD reduced intensity SCT. Nevertheless, LC could be purified in all of 8 analyzed patients. The isolated LC numbers ranged from 10 to more than 1000. In 4 patients we performed a re-analysis of the isolated cells by flow cytometry and confirmed a purity exceeding 97%. We obtained reliable results for LC chimerism in 6 of 8 patients. After the RNA-isolation protocol was further improved, we were able to detect signals even with 35 isolated LC in patient MZ-47. In two patients, the majority of isolated LC were of donor origin whereas the other 4 patients had predominantly host LC (patients' characteristics and chimerism results are summarized in Table 1). None of the patients developed spontaneous acute GVHD so far. For patients MZ-37 and MZ-43 LC-chimerism was also performed after day +50 post HSCT and showed a switch to >97% donor chimerism at that time. In summary, we have established a sensitive method that enables the chimerism analysis on highly purified LC independent of sex-mismatched donor/recipient pairs. Our results on a few patients' samples can not yet be related to clinical events. This assay, however, allows the comprehensive investigation of the chimerism of LC and potentially of other tissue-resident antigen presenting cells to study their impact on GVHD in humans.

Allogeneic hematopoietic stem cell transplantation (SCT) regimens incorporating the lymphocytotoxic CD52 antibody alemtuzumab demonstrate efficient engraftment and reduced graft-versus-host disease (GVHD). However, these protocols substantially impair posttransplantation antiviral and antitumor immunity. To accelerate immune reconstitution after alemtuzumab-based reduced-intensity SCT, we administered prophylactic CD8-depleted donor lymphocyte infusions (DLIs) starting on days 60 and 120 after transplantation. DLIs were processed in an immunomagnetic good manufacturing practice depletion procedure resulting in a 2.5- to 6-log reduction in CD8 T cells. Of 23 high-risk patients with hematologic malignancies, 11 received a total of 21 CD8-depleted DLIs. Five patients developed transient grade I acute GVHD following transfer. Only 2 patients with HLA-C-mismatched donors showed grade II and III acute GVHD and subsequently progressed to limited chronic GVHD. Following DLIs, 4 patients with declining hematopoietic donor chimerism converted to full chimeras. A 2.1-fold median increase of circulating CD4 T cells was observed within 2 weeks after infusion. Non-DLI patients did not show a comparable rise in CD4 counts. Four patients demonstrated enhanced frequencies of cytomegalovirus-specific CD4 and CD8 T cells following transfer. Our results suggest that prophylactic CD8-depleted DLIs accelerate immune reconstitution after lymphodepleted HLA-matched SCT and carry a low risk of inducing severe GVHD.

Current methods for the detection and isolation of antigen-specific CD4+ and CD8+ T cells require the availability of peptide/MHC multimers or are restricted to cells that produce cytokines after antigen contact. We have recently reported that de novo cell surface expression of the TNF receptor family member CD137 (4-1BB) identifies currently activated, but not resting, human alloreactive CD8+ T cells. This observation allowed us to develop a CD137-based technology for the depletion of alloreactive CD8+ T cells in vitro (Wehler et al. Blood 2007; 109:365–373). More recently, a similar approach has been described that uses activation-induced CD137 expression for the detection and enrichment of antigen-specific CD8+ T cells (Wolfl et al. Blood 2007; 110:201–210). In the current study we complement this work and demonstrate the transient up-regulation of CD137 directly on activated cytomegalovirus (CMV) or Epstein-Barr virus specific CD8+ T cells using peptide/HLA tetramer staining of PBMC from seropositive healthy individuals. Antigen-triggered CD137 expression was first detectable upon 6h of stimulation, and reached peak intensity at 24h, allowing the determination of a clear-cut population of CD137+ T cells at this time point. Most importantly, we also observed a similar CD137 expression kinetics (i.e. low baseline, maximum at 24h) on virus-specific CD4+ T cells upon activation with CMVpp65 peptides. The median frequencies of CMVpp65-reactive CD137+ cells measured ex vivo in 4 different CMV+ healthy donors after 24h of stimulation were 3.6% (range, 1.1–6.7) in CD8+ T cells and 2.7% (range, 0.8–6.4) in CD4+ T cells, respectively. We also analyzed PBMC derived from the same donors and left unstimulated, as well as PBMC from CMV-seronegative donors (n=3) stimulated with CMV peptides. None of these samples contained more than 0.3% CD137+ cells per total CD4+ and CD8+ T cells, thereby confirming the specificity of antigen-induced CD137 expression. We next established a two-step in vitro approach allowing the activation and subsequent CD137-based immunomagnetic cell sorting of virus-reactive CD4+ and CD8+ T cells at the same time. We demonstrated the suitability of this assay to isolate CMVpp65-reactive CD4+ and CD8+ T cells from PBMC of 6 CMV+ healthy individuals. Enriched fractions had a median purity of CD137+ cells of 69.4% (range 12.7–94.1) among CD4+ T cells and 70.6% (range, 28.5–93.4) among CD8+ T cells, respectively. The CD137+ populations could be expanded in vitro and showed CMVpp65-specific cytokine production by CD4+ and CD8+ T cells as well as a strong enrichment of CMVpp65/HLA tetramer-binding CD8+ T cells. We finally compared the efficiency of the CD137 assay with the IFN-γ secretion assay to isolate CMVpp65-specific CD4+ and CD8+ T cells from PBMC. Although both methods were performed at optimal conditions, the numbers of CD137+ T cells measured before and after enrichment were approximately 2-fold higher than those of IFN-γ+ cells (n=5), suggesting that CD137 might detect a broader repertoire of virus-reactive T cells. In conclusion, activation-induced CD137 expression provides a means for the rapid detection and isolation of viable virus-reactive CD4+ and CD8+ T cells. The CD137 assay is most attractive for the simultaneous targeting of both T-cell subsets in monitoring studies and adoptive immunotherapy trials.

The induction of a potent and specific T cell response is a major challenge in the development of efficacious cancer vaccine strategies. We applied a novel liposomal formulation (AVE3) for efficient delivery of antigenic peptides into APCs of the skin. These liposomes resulted in a long-lasting deposition of encapsulated compounds at the injection site and the draining lymph nodes. Using a peptide from the melanocyte differentiation antigen tyrosinase-related protein (TRP2) 2 we could show that vaccination with liposome-encapsulated peptide in combination with oligodeoxynucleotides containing unmethylated CpG motifs (CpG ODNs) as adjuvant leads to the induction of tumor cell-specific cytotoxic T cells. The most potent immune response was observed when both, TRP2 peptide and CpG ODNs, were encapsulated into AVE3. Importantly, in contrast to vaccination with free TRP2 liposomal TRP2 peptide generated T cells which respond to 1000-fold lower antigen concentration. Using the poorly immunogenic B16 melanoma model we could demonstrate that vaccination with liposomal TRP2 peptide plus CpG ODNs but not vaccination with free peptide or adjuvant alone resulted in tumor protection in subcutaneous and metastatic tumor models. In summary, vaccination with liposome-encapsulated peptide antigen and CpG ODN allows for the in vivo loading and activation of DC, thereby generating reactive CTL populations even against poorly immunogenic self-peptide presenting tumors resulting in a potent anti-tumor immune response.