PD-1 re-expression is differentially regulated by IL-12 vs IFNα in CD8 T cells

TUMOR_immunotherapyDownregulation of immune functions following responses to pathogen infections is critical for limiting damage to the host by the immune system.  T cell activity is known to be downregulated by a variety of negative regulatory mechanisms including negative checkpoint regulatory proteins, a family of CD28-related molecules.  PD-1 is one such molecule that is transiently expressed on activated T cells.   The ligands for PD-1 are PD-L1 and PD-L2, members of the B7 family of molecules which are upregulated on antigen presenting cells and tumor cells.  Interaction of PD-1 with its ligand leads to inhibition of TCR-mediated signaling via recruitment of SHP1 and SHP2 phosphatases to the TCR synapse.  In the August 2013 edition of The Journal of Immunology, Gerner et al., demonstrate that CD8 T cells initially activated in the presence of IL-12 and IFNα differentially re-express PD-1 upon antigen restimulation.

Cytokines play roles in regulation of nearly every aspect of immune responses.  The cytokine milieu present during T cell activation directs differentiation into the different functional classes of CD4 T helper or CD8 T cells.  This study sought to determine the differences in anti-tumor CD8 T cell effector functions mediated when T cells are activated in the presence of various cytokines.  IL-12 and IFNα activate both overlapping and distinct gene programs and promote cytotoxic CD8 T cell responses.  Thus, these cytokines were chosen for comparison in this study.

In this system, CD8+ OT-1 cells were activated ex-vivo in the presence of either IL-12 or IFNα, and transferred into B16-OVA tumor-bearing mice.  T cells activated in the presence of IL-12 were found to mediate tumor-growth inhibition significantly better than if they had been activated in the presence of IFNα.  Over time in tumor-bearing mice, transferred IFNα-matured OT-1 cells were observed to decline in number and lost the ability to produce IFNγ ex vivo upon restimulation, indicating these cells may be exhausted.

Because PD-1 is known to be a marker and mediator of T cell exhaustion, PD-1 expression was examined.  Initial induction levels of PD-1 were comparable on OT-1 cells following ex vivo activation with IFNα or IL-12.  Following transfer into tumor-bearing mice, PD-1 levels declined over time on both types of cells isolated from the spleen and on IL-12 matured cells isolated from the tumor.  However, PD-1 expression was high on transferred IFNα-matured cells when isolated from the tumor.  Similar results were seen when cells were transferred into mice that subsequently received an injection of the OVA peptide.  Thus, CD8+ T cells matured in the presence of IFNα appear to re-express significantly higher levels of PD-1 upon antigen restimulation than IL-12 matured T cells.

PD-1 and PD-L1 targeting with inhibitory antibodies have emerged as promising avenues in tumor immunotherapy.  In this study, anti-PD-1 antibody administration had no additional anti-tumor effect in mice that received IL-12-matured T cells, while in mice that received IFNα-matured T cells, anti-PD-1 antibodies led to inhibition of tumor-growth to a level similar to that in mice that had received IL-12-matured T cells.  Thus, the relatively poor ability of IFNα-matured T cells to efficiently inhibit tumor growth appears to be largely due to PD-1 upregulation.  Finally, when T cells were matured with both IL-12 and IFNα, the effect of IL-12 was dominant.

Many questions remain regarding the mechanisms mediating PD-1 re-expression in IFNα vs. IL-12 matured T cells.  However, since IL-12 activity was dominant over IFNα on regulating PD-1 expression, IL-12 administration during immunotherapy regimens may enhance anti-tumor T cell responses by blocking the mechanisms by which IFNα enhances PD-1 re-expression.

Further Reading:

Cutting Edge: IL-12 and Type I IFN Differentially Program CD8 T Cells for Programmed Death 1 Re-expression Levels and Tumor Control.  Gerner MY, Heltemes-Harris LM, Fife BT, Mescher MF. J Immunol. 2013 Aug 1;191(3):1011-5. doi: 10.4049/jimmunol.1300652. Epub 2013 Jun 26.

Using Mass Spectrometry for Mass T cell Epitope Discovery

Time of Flight Mass Cytometry (CyTOF) is a relatively new multiparametric technology that is far outpacing standard fluorescence-based flow cytometry in the number of parameters that can be simultaneously assessed on a single cell.  In CyTOF, rare transition element isotope-conjugated antibodies are used to label cellular antigens of interest, the magnitude of which is then quantitated by a time of flight mass cytometer, as discussed previously. Previous studies assessing 34 cell surface and intracellular proteins by this technology demonstrated the existence of high dimensional complexity in the heterogeneity of human bone marrow and CD8+ T cell populations.  In a July 2013 article in Nature Biotechonology, Newell et al., move CyTOF and the field of immunology another technological step forward by utilizing CyTOF to measure the frequencies of Rotavirus antigen-specific T cells in human peripheral blood mononuclear cells (PBMCs) and jejunal tissue with peptide-MHC tetramers.

In CyTOF, the theoretical maximum number of simultaneously assessable parameters is 100-200 depending on the instrument.  This vastly outnumbers the assessable parameters of standard fluorescence-based flow cytometry.  To date however, only approximately 40 metal ions have been utilized for antibody labeling, and the development of further metal-chelating technologies is awaited in order to utilize the maximum capacity of the CyTOF instrument.  In the current study, the authors circumvent this limitation by using a “bar-coding” methodology in which a variant combination of three out of ten metal ions are used for labeling each tetramer, allowing for up to 120 different metal combinations.

In this study, the authors sought to identify Rotavirus epitopes recognized by human CD8+ T cells in the context of the MHC class I allele, HLA-A*0201.  To date, only two Rotavirus epitopes recognized by T cells have been identified, and little is known about the phenotypic and functional diversity of antigen-specific T cells for any particular pathogen.  The technical difficulties in proper epitope prediction along with the limited number of cells attainable from human blood samples contribute to these issues.  Thus, this method represents a huge leap forward in the potential to identify significantly more antigen-specific T cell epitopes and to extensively classify these cells functionally.  Using an MHC-prediction algorithm, 77 possible Rotavirus peptides were identified that bound to HLA-A*0201.  An additional 32 positive and negative control tetramers were added for a total of 109 labeled tetramers used to stain each sample simultaneously.  This was further combined with 23-27 metal-chelated antibodies specific for cell surface and intracellular antigens to phenotypically characterize the T cells. A specialized Matlab script was used to analyze the high-dimensional data obtained following mass spectrometry of PBMC and jejunal samples.

On average, CD8 T cell populations specific for two Rotavirus-peptides plus 6-7 peptides from other viruses including influenza, EBV, and CMV, were identified on average across PBMCs from the 17 healthy donors analyzed.  These antigen-specific T cell populations were further phenotypically characterized by expression of surface and intracellular markers.   CD8 T cells specific for six Rotavirus epitopes that included the two previously identified epitopes, were recurrently detected in PBMCs from at least two individuals.  Of these, CD8 cells specific for a Rotavirus peptide from the VP3 protein were most common among healthy donor PBMCs and were phenotypically unique, being of the effector memory subtype compared with a central memory phenotype typical of the T cells specific for the other Rotavirus peptides.  VP3-specific T cells were also uniquely present in jejunal tissue obtained from obese patients that had undergone gastric bypass surgeries.  Thus, this methodology discovered at least 4 new Rotavirus peptides as well as unique characteristics of the different antigen-specific CD8 T cell populations.

In summary, this methodology of combining CyTOF technology with tetramer “bar-coding” paves the way for a vast expansion over fluorescent-based flow cytometry techniques for identifying antigen-specific T cell populations.  As vaccine strategies are an ongoing goal for treatment and prevention of infectious diseases and cancer, it is important to not only identify the peptides that can elicit T cell responses, but also functionally characterize these T cells in order to maximally promote desired immune responses.

Further  Reading:

Combinatorial tetramer staining and mass cytometry analysis facilitate T-cell epitope mapping and characterization.  Newell EW, Sigal N, Nair N, Kidd BA, Greenberg HB, Davis MM. Nat Biotechnol. 2013 Jul;31(7):623-9. doi: 10.1038/nbt.2593. Epub 2013 Jun 9.

Cracking the code of human T-cell immunity.  Harvey CJ, Wucherpfennig KW. Nat Biotechnol. 2013 Jul 9;31(7):609-10. doi: 10.1038/nbt.2626.

Cytometry by time-of-flight shows combinatorial cytokine expression and virus-specific cell niches within a continuum of CD8+ T cell phenotypes. Immunity. 2012 Jan 27;36(1):142-52. doi: 10.1016/j.immuni.2012.01.002.

Single-cell mass cytometry of differential immune and drug responses across a human hematopoietic continuum.  Bendall SC, Simonds EF, Qiu P, Amir el-AD, Krutzik PO, Finck R, Bruggner RV, Melamed R, Trejo A, Ornatsky OI, Balderas RS, Plevritis SK, Sachs K, Pe’er D, Tanner SD, Nolan GP. Science. 2011 May 6;332(6030):687-96. doi: 10.1126/science.1198704.

A bifunctional FoxP3+ regulatory T cell subset converts to pro-inflammatory helper T cells

Recently a number of studies have arisen characterizing Tregulatory cellvarious functional subsets of CD4+ FoxP3+ regulatory T cells (TREGS), as well as their plasticity and ability to differentiate into other TH subtypes.  For instance, TREGS that express RORγt were found to be the specific TREG subset that promotes pro-tumor immune functions in colorectal cancer patients.  In a recent article in Immunity, Sharma et al. identify another TREG subset: FoxP3+ TREGS that loose expression of Eos convert to a pro-inflammatory helper subtype that promotes naïve CD8+ T cells differentiation into potent effectors.

Eos is a transcription factor in the Ikaros family, and acts as an obligate co-repressor in complex with FoxP3 to inhibit expression of FoxP3-repressed genes.  In a quest to understand why TREGS in inflammatory environments were observed to become pro-inflammatory without losing FoxP3 expression, Sharma et al. examined the expression of Eos in FoxP3+ TREGS under inflammatory conditions.

Conversion of FoxP3+ TREGS into an inflammatory phenotype was demonstrated by acquired expression of IL-2, IL-17, and CD40L in the draining lymph nodes of a vaccination site compared with FoxP3+ TREGS at distant lymph nodes that did not gain this function.  In these converted inflammatory FoxP3+ TREGS, expression of Eos was rapidly lost.  IL-6 was required for downregulation of Eos, as TREGS in mice lacking IL-6 did not lose Eos expression under the same conditions.  However, IL-6 alone was insufficient for Eos downregulation, which also required interactions with MHC class II on activated dendritic cells.  Loss of Eos expression was furthermore shown to be required for acquisition of the pro-inflammatory phenotype, as TREGS with forced overexpression of Eos did not undergo this conversion.

Interestingly, not all FoxP3+ TREGS were equivalent in their propensity to lose Eos expression and become pro-inflammatory.  Thymic FoxP3+ TREGS were assessed for stability of Eos under treatment with cyclohexamide. CD38+CD69+CD103 TREGS were “Eos-labile” and specifically lost Eos expression within one hour of cyclohexamide treatment, while CD38CD69CD103+ TREGS maintained Eos expression.  Expression of other markers associated with FoxP3+ TREGS including CD25 and CTLA-4 were equivalent between these two phenotypes highlighting the inability of using these TREG markers to discriminate between these populations.  When these FoxP3+ TREGS were sorted into CD38+CD103and CD38CD103+ subsets and transferred into mice, followed by the vaccination schema, only CD38+CD103 TREGS lost Eos expression and gained CD40L and IL-2 expression. The Eos-labile TREGS do however have characteristic suppressive functions when examined in several models including protection from colitis in a Rag-deficient CD45RBHI effector cell-driven autoimmune colitis model and in vitro suppression of T cell proliferation driven by anti-CD3.

Because the Eos-labile subset was observed in the thymus as part of the natural TREG repertoire, the authors examined the signals required for development of this subset.  Again, IL-6 was required as this subset did not arise in IL-6-/- mice.  Epigenetic analysis of DNA methylation patterns comparing these FoxP3+ TREGS subsets revealed distinctive patterns of methylation yet these subsets were still much more closely related to each other as compared with FoxP3 CD4+ T cells.  Future studies will be needed to determine the nature of these epigenetic differences and which signals are controlled by IL-6.

Interestingly, the authors explored the functional contribution of the Eos-labile pro-inflammatory TREGS subset on CD8+ priming in the vaccination model.  Depletion of TREGS resulted in loss of CD8+ T cell proliferation and granzyme B expression as well as loss of CD86 upregulation on DCs, while adding back just the Eos-labile subset or IL-2 plus CD40-agonist antibodies rescued these defects.  The Eos-labile subset did not however, contribute to reactivation of memory CD4+ T cells, and thus these cells appear to play a specific role in the initial priming stages of naïve T cell activation.  Thus, despite having regulatory activity, these cells are critical in priming CD8+ T cell responses by supplying IL-2 and CD40L signals.

However, indoleamine 2,3-dioxygenase (IDO) was able to block Eos downregulation and acquisition of IL-2, IL-17, and CD40L expression.  Importantly, in a murine tumor vaccination model, blocking IDO was important for FoxP3+ inflammatory TREG induction and acquisition of anti-tumor effector CD8+ T cell responses.  The mechanism of IDO inhibition of Eos downregulation was found to be at least in part, dependent on the antagonization of the IL-6-STAT3 pathway by IDO-mediated production of kynurenine-pathway metabolites which activate the aryl hydrocarbon receptor (AhR).  Interestingly, different AhR ligands have been previously shown to differentially regulate induction of TH17 cells vs. TREGS (Quintana et al.), and kyenurine was a TREG inducing AhR ligand (Mezrich et al.).  Additionally, the contrasting effects of IL-6 and IDO will be an important factor in priming immune cell responses.

Overall, this thorough investigation identified the mechanisms that induce and inhibit this newly defined Eos-labile TREG subset that maintains FoxP3 expression and has typical suppressive TREG activity, yet is critically important in priming effector T cell immune responses.  Future studies will be needed to address how these cells balance regulatory and priming activities as well as the relationships between this subset and the many other TREG subsets described.

An inherently bifunctional subset of foxp3(+) T helper cells is controlled by the transcription factor eos.   Sharma MD, Huang L, Choi JH, Lee EJ, Wilson JM, Lemos H, Pan F, Blazar BR, Pardoll DM, Mellor AL, Shi H, Munn DH. Immunity. 2013 May 23;38(5):998-1012. doi: 10.1016/j.immuni.2013.01.013. Epub 2013 May 16.

Eos, goddess of treg cell reprogramming.  Rieder SA, Shevach EM. Immunity. 2013 May 23;38(5):849-50. doi: 10.1016/j.immuni.2013.05.001.

Control of T(reg) and T(H)17 cell differentiation by the aryl hydrocarbon receptor.  Quintana FJ, Basso AS, Iglesias AH, Korn T, Farez MF, Bettelli E, Caccamo M, Oukka M, Weiner HL. Nature. 2008 May 1;453(7191):65-71. doi: 10.1038/nature06880. Epub 2008 Mar 23.

An interaction between kynurenine and the aryl hydrocarbon receptor can generate regulatory T cells.  Mezrich JD, Fechner JH, Zhang X, Johnson BP, Burlingham WJ, Bradfield CA. J Immunol. 2010 Sep 15;185(6):3190-8. doi: 10.4049/jimmunol.0903670. Epub 2010 Aug 18.

Optimizing Assays to Find Rare Antigen-Specific T cells in Cryopreserved PBMCs

Immunomonitoring of T cell based immune responses spans a wide range of therapeutic applications such as infectious and autoimmune diseases and is particularly important for vaccine research. Regardless of the therapeutic application, immunomonitoring can be a daunting task due to the variability of methods and protocols available. There are several commonly used functional assays for the enumeration of antigen specific CD8+ T cells and there is great variability in the protocols that are used for these assays. Thus, making it increasing difficult to thoroughly interpret data obtained from multi-center clinical trials and to compare results between laboratories. In order to address some of the issues associated with immunomonitoring of clinical trials, the Association for Immunotherapy of Cancer (CIMT) formed a CIMT monitoring panel tasked to standardize protocols for assaying T cell antigen immune responses. Thirteen centers from 6 different European countries participated in this study. They were given the same samples and asked to determine the number of antigen specific T cells and assess their antigen specific function using tetramer staining and a functional assay of their choice. Common techniques used for monitoring antigen induced immune responses included ELISPOT assays, HLA-multimer staining and intracellular cytokine staining (ICS).

Pre-tested samples of peripheral blood mononuclear cells (PBMC), synthetic peptides, and PE-conjugated HLA-tetramers were distributed to each center. Using HLA-typed healthy volunteers, PBMCs were isolated by Ficoll density gradient separation. Each sample was tested for T cell reactivity against CMV and influenza. All centers received an HLA-A negative control as well as HLA-A positive samples consisting of a combination of CMV and influenza reactive PBMCs. The study comprised of 2 phases; Phase I consisted of all centers performing the assays with their commonly used protocols, and in Phase II each center received optimized protocols based on the findings from Phase I.

For Phase I’s tetramer-staining assay, the laboratories could choose to stain samples with antibodies (Ab) for CD8+ alone, CD3+CD8+, or CD4+ CD8+ and use their preferred Ab clone, fluorescent dye, and Ab concentration. For the functional assays synthetic peptides were provided and each group could choose either the INF-γ ELISPOT assay, FACS-based intracellular INF-γ staining or both with their antigen concentration of choice ranging from 1-10 g/ml. To reduce variability in FACS analysis, sample plots were provided as well as gate settings and quadrants. Tetramer-staining data reported included; number of viable cells post-thawing, cytometer model, number of lymphocytes and/or CD8+ cells analyzed. Data was presented as percent of tetramer-positive cells among CD8+, CD3+CD8+, or CD4+ lymphocytes depending on what antibody cocktail was chosen. For the functional assays each center reported the type of ELISPOT plates used, reagents and conditions used, and number cells tested.

Tetramer results from the Phase I study showed the number CD8+ cells analyzed significantly affected the sensitivity of tetramer staining. Antigen-specific T cell reactivity when less than 30,000 CD8+ T cells were counted resulted in only 70% responsiveness detected. In contrast, when more than 30,000 CD8+ cells were counted, an 89% response was observed. Although, when antigen-specific T cells were present at high frequencies the number of counted cells did not matter. Interestingly, Ab clone variability, Ab concentration, or cytometer type did not result in any significant differences. Thus, the main factors affecting antigen-specific T cell reactivity by tetramer staining is the number of CD8+ cells used. For Phase II it was then recommended at least 1 x106 PBMCs are used for this assay.

The majority of groups chose the INF-γ ELISPOTas their functional assay. Results showed a large amount of heterogeneity between the centers. Some centers included a resting phase after thawing the cells, of 2-20 hours, resulting in 73% positive reactivity (number of spot forming cells per seeded PBMC). In contrast, not allowing a resting phase resulted in only detecting 30% of the positive cells. Additionally, intra-center replicate reproducibility was significantly affected by the number of replicates used, where duplicates often failed the Student t test and triplicates were sufficient to reach statistical significance. Addition of allogenic-APCs for binging and presentation of the synthetic peptides was found to have a negative effect on detection response (28% of all responses vs. 58%). When looking at the number of cells seeded per well, those with more than 4 x105 PBMC detected 71% positive samples and those with less than 4 x105 only detected 43%. Granted, when antigen specific T cells were available at high frequencies the number of counted cells did not affect the response rates. Consequently, Phase II’s minimum requirements for the INF-γ ELISPOT protocol included: (1) triplicates should be performed for each test antigen (2) avoid using allogenic-APCs (3) include a resting phase (4) use over 4 x 105 PBMCs per well.

Another interesting finding from this study was that lab experience in performing these assays had no effect on the performance of the assays compared to labs that had just adopted the techniques. Further highlighting the importance of developing standardized protocols for immunomonitoring assays. This study did not however, address specific detection limits for the ELISPOT assays, the variability between ELISPOT plate readers, nor serum source effects on background and specificity. In addition, it was not reported whether live/dead cell stains where included in the tetramer assays and how combinations of these may have had an effect on the sensitivity of the assay.

Overall, this study identified several factors that should be generally implemented when performing tetramer staining and INF-γ ELISPOT assays with cryopreserved PBMC samples. Furthermore, these protocol modifications are particularly important when assaying antigen-specific T cell populations present at low frequencies.


The CIMT-monitoring panel: a two-step approach to harmonize the enumeration of antigen-specific CD8+ T lymphocytes by structural and functional assays. Britten CM, Gouttefangeas C, Welters MJ, Pawelec G, Koch S, Ottensmeier C, Mander A, Walter S, Paschen A, Müller-Berghaus J, Haas I, Mackensen A, Køllgaard T, thor Straten P, Schmitt M, Giannopoulos K, Maier R, Veelken H, Bertinetti C, Konur A, Huber C, Stevanović S, Wölfel T, van der Burg SH. Cancer Immunol Immunother. 2008 Mar;57(3):289-302. Epub 2007 Aug 25.