CROSS-REFERENCE TO RELATED APPLICATION This application claims the benefit of U.S. Provisional Application No 62/587346, filed November 16, 2017, the entire contents of which are incorporated herein by reference.

STATEMENT OF GOVERNMENT LICENSE RIGHTS

This invention was made with Government support under UM1AI109565, HHSN272200700046C, and R01 AI108839 awarded by the National Institutes of Health The Government has certain rights in the invention.

BACKGROUND

T cells with the primary purpose of suppressing other T cells through both contact-mediated and soluble mechanisms are referred to as regulatory' T cells (Tregs). Their primary' function serves to prevent autoimmune reactivity' to self-antigens by conferring peripheral tolerance after education in the thymus (thymic Tregs) Regulatory T cells can also he induced in the periphery (peripheral Tregs). These peripheral Tregs are not specific to autoantigens, but instead are specific to exogenous antigens found in allergens and pathogens. Regulatory T cells are also implicated m the suppression of anti tumor responses intended to be directed against self-antigens carried on proliferating host cancer cells. Absence or dysregulation of Tregs can conversely fail to protect against autoimmune responses by failing to suppress autoreactive effector T cell responses (Teffs). Regulatory T cells exert their suppression by both inhibiting the proliferation of Teffs as well as suppressing cytokine production in their target population.

Regulatory T cells were first identified as CD4+ T ceils carrying abundant amounts of the alpha chain of the high-affinity IL-2 receptor (CD25). While increased expression of CD25 can be used to identify Tregs, CD25 is also expressed on large numbers of effector ( 1)4 T cells because IL-2 is essential for T cell survival and proliferation. Tims, CD25 can also be upregulated on activated CD4+ T cells, making the separation of regulatory T cells and activated T effector cells difficult based on the presence of CD25. High purity is the aim for therapeutic Tregs with the potential to treat autoimmune diseases like rheumatoid arthritis (RA), multiple sclerosis (MS), type 1 diabetes (T1D), and psoriasis. Other diseases in which isolation of high purity Tregs is of relevance include graft versus host disease (GvHD) and allergic diseases, where effector T cells responses are implicated in the rejection of grafts or the induction of allergy. In order to effectively control these responses, regulatory T cells need to be isolated without any contaminating effector T cells that would exacerbate the disease response.

The narrow edge on which the balance between regulator' and effector T cell responses sits for any given disease is further complicated by the lack of distinction between thymic Tregs (tTregs) and peripheral Tregs (pTregs) in humans, and more particularly by the lack of a singularly expressed molecule that is limited to exclusively Tregs. While Foxp3 is considered to be the master transcriptional regulator of Treg programming, recently activated T cells in humans express Foxp3 causing confusion as to the purpose of Foxp3 and whether Foxp3 expression on activated T cells confers similar suppressive capabilities as seen in Tregs. In addition, Tregs can be identified on the basis of high levels of CD25, with the limitations discussed above. While these molecules are commonly used to discriminate Tregs from conventional T cells, they are constitutively expressed and cannot be used to discriminate Tregs that are specific for a particular antigen from those that are not. The isolation of live regulatory T cells is also difficult as Foxp3 staining requires fixation and permeabilization of the T cells. Furthermore, Foxp3 expression merely indicates the functionality, but not specificity of these cells. By identifying costimulatory molecules expressed after ex vivo stimulation, it is possible to separate antigen-specific Tregs from non-specific Tregs, as well as antigen-specific effector T cells. For example, CD 137 (4-1 BB) is a molecule that has been noted to discriminate antigen-specific Tregs and CD8+ T cells. However, some have argued that CD137 expression is no more effective at discriminating antigen-specific Tregs than CD25. This lead to the discovery that CD154 (CD40L) expressing effector T cells must be removed prior to CD137 enrichment, because CD154+ effector T ceils begin to express CD137 over extended periods of activation and serve as contamination to the CD137+ Tregs. There are claims that antigen-specific Tregs express CD137 at shorter intervals (4 hours) whereas antigen-specific effector T cells only begin to express CD137 at (12-16 hours). In other studies, it has been shown that the expression of CD137 on CD 154+ Teffs occurs at a shorter interval (6-8 hours) when no artificial costimulation, such as anti-CD28, is added. The difference in rate of expression is also attributed to the source of the antigen, as peptide stimulation induces CD 154 and CD 137 expression at shorter intervals than protein antigens that are required to undergo more intense antigen processing. While some groups have indicated that CD 154 is expressed by Tregs, others have found no evidence for this claim in humans and, thus, perhaps it is a phenomenon limited to mice.

While the separation of Tregs from Teffs is essential for many applications, a negative selection must be used in order to unequivocally ensure removal of contaminating Teffs especially considering that many activation-induced costimulatory molecules have shared expression between the two groups. In tins context, Teffs are defined as all T cells that do not belong to the Treg category', which is indicated by stable expression of the transcription factor Foxp3, along with high levels of CD25 and little to no CD127. These Teffs typically exhibit their effector function through the secretion of cytokines such as IL-2, IL-4, IL-9, IL-17, IL-22, and IFNy. The Teffs may or may not be antigen-specific. The secretion of IL-10 alone without subsequent Foxp3 expression does not exclude a T cell from the Teff distinction, as this type of suppressive capability can arise independently of Foxp3 expression but still serves as an essential suppressive T cell subset.

Accordingly, despite the advances in the art, there remains a need to simply, reliably, and accurately, identify and isolate Treg cells from biological samples and avoid contamination from conventional T cells. The present disclosure addresses this and related needs.

SUMMARY

This summary' is provided to introduce a selection of concepts in a simplified form that are further described below' in the Detailed Description. This summar ' is not intended to identify key features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

In one aspect, the disclosure provides a method of detecting an activated regulatory' T (Treg) cell. In some embodiments, the method comprises:

obtaining a sample comprising regulatory' T (Treg) cells that have been exposed to an antigen and/or activator in a manner such that the sample may contain activated Treg cells;

contacting the sample with a first detection molecule that specifically binds to a latent TίϊRb complex protein and a second detection molecule that specifically binds to a co-stimulatory marker; and

detecting a cell m the sample that is bound by the first detection molecule and the second detection molecule, thereby detecting an activated Treg cell. In another aspect, the disclosure provides a method of producing an enriched population of activated T regulator}' (Treg) cells. In some embodiments, the method comprises:

incubating a sample comprising regulatory T (Treg) cells with an antigen of interest and/or an activator in a manner and amount such that at least a portion of the Treg cells are activated;

contacting the sample with a first enrichment molecule that specifically binds to a latent TGFp complex protein and a second enrichment molecule that specifically binds to a co-stimulatory marker; and

enriching for cells that are bound to the first enrichment molecule and the second enrichment molecule, thereby producing an enriched population of activated Treg cells.

In another aspect, the disclosure provides a cell in the enriched population produced from the method described herein.

In another aspect, the disclosure provides a method of treating a condition treatable by the presence of an activated Treg. In some embodiments, the method comprises administering the cell described herein to a subject in need thereof.

In another aspect, the disclosure provides a method for monitoring a T regulator}' (Treg) cell response to potential exposure of a subject to an antigen. In some embodiments, the method comprises:

contacting a sample with a first detection molecule that specifically binds to a latent TGFp complex protein and a second detection molecule that specifically binds to a co-stimulator}' marker, wherein the sample comprises T cells and was obtained from a subject suspected of being exposed to an antigen; and

detecting or quantifying cells in the sample that are bound by both the first detection molecule and the second detection molecule, wherein the absence, presence, or relative abundance of cells bound by both the first detection molecule and the second detection molecule is indicati ve of a state of the Treg cell response.

In another aspect, the disclosure provides a method for monitoring the sensitivity of a subject to an antigen of interest. In some embodiments, the method comprises:

contacting a sample comprising T regulatory (Treg) cells with a first detection molecule that specifically binds to a latent TGFfl complex protein and a second detection molecule that specifically binds to a co-stimulatory marker, wherein the Treg cells have been exposed to an antigen of interest and the sample was obtained from the subject; detecting or quantifying the cells in the sample that are bound by both the first detection molecule and the second detection molecule; and

comparing the number of cells in the sample that are bound by both the first detection molecule and the second detection molecule to an established threshold to determine the sensitivity of the subject to the antigen of interest.

In another aspect, the disclosure provides a method of screening for T regulatory (Treg) cell stimulatory epitopes from an allergen or antigen of interest. In some embodiments, the method composes:

obtaining a sample comprising Treg ceils;

exposing the Treg cells to an epitope derived from an antigen of interest in an MHC context;

contacting the Treg cells with a first detection molecule that specifically binds to a latent TGFfi complex protein and a second detection molecule that specifically binds to a co-stimulatory marker; and

quantifying the relative abundance of cells in the sample that are bound by both the first detection molecule and the second detection molecule, wherein a high relative abundance indicates that the epitope stimulates development of activated Treg cells.

In another aspect, the disclosure pro vides a method of identifying the MHC Class II molecule that binds to a Treg stimulatory epitope. In some embodiments, the method comprises:

obtaining a sample comprising Treg cells;

exposing the Treg cells to an epitope derived from an antigen of interest in an MHC context;

contacting the Treg cells with a first detection molecule that specifically binds to a latent TGFfi complex protein and a second detection molecule that specifically binds to a co-stimulatory marker;

quantifying the relative abundance of cells in the sample that are bound by both the first detection molecule and the second detection molecule, wherein a high relative abundance indicates that the epitope stimulates development of activated Treg cells; and characterizing the MHC molecule complexed with the epitope that resulted in a high abundance of cells bound by both the first detection molecule and the second detection molecule. In yet another aspect, the disclosure provides a method of testing the effect of a putative therapeutic compound on an activated T regulatory (Treg) cell. In some embodiments, the method comprises exposing a ceil described herein to the therapeutic compound.

DESCRIPTION OF THE DRAWINGS

The foregoing aspects and many of the attendant advantages of this invention will become more readily appreciated as the same become better understood by reference to the following detailed description, when taken m conjunction with the accompanying drawings, wherein:

FIGURES 1A and IB illustrate that de novo expression of GARP/LAP within CD137 distinguishes highly activated Tregs. Shown are flow cytometric output following polyclonal T cell stimulation of PBMCs for 18 hours. It is shown that antigen- specific Tregs express GARP/LAP and 0X40 in CD137hi, but not CDI37mid compartments. GARP and 0X40 expression, and thus antigen specificity is limited to only the CD137hi, CD154neg compartment. (FIGURE 1 A illustrates the lack of GARP, LAP, and 0X40 in the CDl37mid compartment, while FIGURE IB illustrates the presence of these molecules in the CD137hi compartment).

FIGURE 2 illustrates that antigen-specific Tregs express GARP/LAP and 0X40 in CD137hi, but not CD137mid compartments. Shown are flow cytometric output following polyclonal T cell stimulation of PBMCs for 18 hours. It is shown that antigen- specific Tregs express GARP/LAP and 0X40 in CD137hi, but not CD137mid compartments. GARP and 0X40 expression, and thus antigen specificity, is limited to only the CD137hi, CD154neg compartment. (Inset box "A" illustrates the lack of GARP, LAP, and 0X40 in the CDI 37mid compartment, while inset box "B" illustrates the presence of these molecules in the CD I37hi compartment).

FIGURE 3A and 3B graphically illustrate that CD137+ Tregs expressing GARP/LAP exhibit higher levels of activation in multiple disease models. Patient PBMCs were isolated from various disease states and stimulated with relevant peptide antigen pools m cancer (tumor-associated antigens), autoimmunity (T1D pancreatic islet antigens), allerg} (Ain g 1 alder antigen), and vaccine (HA1 influenza antigen) for 16 hours. Cultures were then enriched for CD 137 and assessed for GARP/LAP expression, as well as other Treg-related activation markers (FIGURE 3A). Across all antigen models, GARP/LAP+ Tregs expressed increased activation as compared to CD137+ Tregs. In a polyclonal aCD3 aCD28 stimulation, the majority of the cells expressed GARP/LAP and no significant differences were seen in the activation states of the GDI 37+ and GARP/LAP+ Tregs. Significance was determined by paired T test where **** indicates p<0.0001.

FIGURE 4 illustrates that optimal CD 137 and GARP/LAP expression occurs at 16 hours of ex vivo stimulation with peptide antigen. PBMCs from a healthy patient were stimulated with influenza HA1 peptide antigen for 0, 2, 6, 8, 12, 16 and 18 hours to determine optimal expression. Although Foxp3+ CD25+ Tregs begin to express slight levels of CD137 and GARP/LAP at 6 hours, expression plateaus at 12-16 hours. Similar trends can be seen when tracking percent expression within memory CD4+ T cells. We also tracked Ki67 expression over the same times and found there was no change, indicating that ex vivo stimulation did not induce cell division. When observing phenotypic markers of Tregs like CD39, we also saw no changes in expression over the same timeframe (data not shown).

FIGURES 5A-5D illustrate that GARP/LAP expression is necessary for detection of highly Foxp3-ennched antigen-specific Tregs. Representative gating strategies for live ceil analysis of highly-enriched antigen-specific regulatory T cells. PBMCs were stimulated for 18 hours with polyclonal stimulation and cells were then enriched for GARP and/or LAP expression by magnetic bead enrichment. Using GARP/LAP expression in conjunction with other activation-induced markers (FIGURE 5 A-CD 137, FIGURE 5B-OX40, FIGURE 5C-CD27, FIGURE 5D-CD25) highly enriched antigen- specific Foxp3+ Tregs can be isolated by sorting or magnetic enrichment (Foxp3 purity can be seen in corresponding bottom panels).

FIGURE 6 graphically illustrates that polyclonal activation of human PBMCs induces enriched regulatory T cell activation GARP+ 0X40+ and CD137+ CD25+ expression. Three patients had PBMCs collected and stimulated for 18 hours at 37°C m the presence of TCR-nonspecific polyclonal stimulation. Expression of regulatory T cell markers was then assessed by flow cytometry as a percentage of parent CD4+ CD45RA- memory T cell populations. As a control, we also assayed the patients with influenza hemagglutinin peptide pools (HA1 ) and DMSO. When removing CD 154+ effectors prior to the CD137+ CD25+ analysis, the polyclonal stimulation reaches a lower percent of activation for regulatory T ceils, with a higher background. When using just GARP and 0X40 expression with no CD 154 removal, the polyclonal stimulation is higher than CD137+ CD25+ CD154- analysis and has less background in the DMSO control. The GARP+ 0X40+ compartment is also enriched for Foxp3 expression (-95% Foxp3+) as compared to the CD137+ CD25+ CD 154- compartment (-90% Foxp3+) (data not shown).

FIGURES 7A and 7B illustrate that CD 154+ T effectors do not express GARP/LAP after ex vivo stimulation. PBMCs were isolated from a grass allergic patient and stimulated for 6 or 18 hours with grass allergen Phlp peptide pools or tetanus toxoid peptide pools. FIGURE 7A: GARP/LAP expression is not observed during the 6 hour stimulation in effectors (CD 154+ CD137+) or Tregs (CD154- CD 137+) at 6 hours. FIGURE 7B: After 18 hours, GARP/LAP expression is induced by both model antigens, however it is only observed on Tregs (CD154- CD 137+) and not Teffs (CD154+ CD 137 i.

FIGURES 8A and 8B graphically illustrates that CD137+ regulatory T cells are distinct in their gene expression from CD154+ effector T cells, and GARP+ CD137+ regulatory T cells are distinct from GARP- CD 137 cells. Patients (h=10) were drawn before and after influenza vaccination and PBMCs were restimulated with HA1 peptide pools for 18 hours. Cells were then enriched for CD154 and CD137 magnetically and sorted by flow' cytometry. Sorted populations included CD154+ CD137- effector T cells (blue), GARP+ CD137+ regulator}' T cells (red), and GARP- CD137+ regulator}' T cells (green). Sorted populations then undement bulk RNAseq and transcript profiles were clustered by principal component analysis (PCA) (FIGURE 8A). Significantly upreguiated and differentially expressed regulator}' T cell-associated genes m GARP+ CD137+ Tregs (green) and GARP- CD137+ Tregs (red) were plotted (FIGURE 8B). Ail genes displayed were significantly different in expression, as noted by the doited line representing p=0.G5.

FIGURES 9A-9D graphically illustrate that GARP/LAP + CD 137+ regulatory T cells have increased regulatory potential in peanut allergic patients. Peanut allergic patients (n=3) were drawn and PBMCs were restimulated with Arali peptide pools and enriched for CD137 magnetically. Enriched cells were then sorted by GARP+ and GARP- surface expression and analyzed for RNAseq. Increased co-inhibitory expression of CTLA-4 (p<0.01) (FIGURE 9B) and LAG-3 ip 0.05 } (FIGURE 9C) 'as seen in GARP+ populations, as 'ell as inhibitory cytokine IL-10 (p<0.05) (FIGURE 9A) expression when measured by paired T test analysis. Additionally, increased expression of CCR8 (p<0.Gl) (FIGURE 9D) was observed wiien measured by paired T test analysis.

FIGURES 10A-10D illustrate that GARP/LAP+ Treg cell lines isolated from patients stain more effectively with cognate tetramer than CD137+ Treg cell lines. GARP+ GDI 37+ cell lines isolated from patients stain more effectively with cognate tetramer after in vitro culture than GARP- CD137+ ceil lines. GARP+ CD137+ and GARP- CD137+ regulatory T cells were sorted after ex vivo stimulation of alder allergic patients with Alngl p48 peptide. After two weeks m vitro culture, the lines were stained with cognate tetramer and analyzed by flow' cytometry. GARP+ CD137+ cell lines (FIGURE 10A) stain more effectively with Alngl p48 tetramer than GARP- CD137+ cell lines (FIGURE 10B) from the same patient. When comparing tetramer MFI of alder allergic patients (n=3), GARP/LAP+ Treg lines had significantly higher staining (FIGURE 10C). The same result was demonstrated in influenza HA 1 -specific Treg lines sorted for GARP/LAP expression (FIGURE 10D). * represents p<0.05 as measured by paired T test analysis.

FIGURE 11A and TIB graphically illustrate increased antigen-specificity in the GARP+ CD137+ population as compared to the GARP- CD137+ population when assessed by the number of ceils secreting the inhibitory cytokine IL-10. FIGURE 11 A: PBMCs isolated from a patient were stimulated with a tumor-associated antigen (TAA) peptide pool. Cells were magnetically enriched for GARP and LAP and CD 137 and were then sorted and cultured for 2 weeks in vitro. After culture, the cells were then restimulated with either the relevant TAA peptide pool (black bars) or an irrelevant influenza (HA1) peptide pool (gray bars) for 18 hours. Cells were then plated into an ELISpot specific for IL-10 and IFNy to assess their antigen-specificity in response to relevant and irrelevant peptide restimulation. GARP+ CD137+ lines had a significant increase (p<0.00l) in the number of cells with the ability to secrete IL-10 with relevant peptide stimulation as compared to irrelevant stimulation using T test analysis, but showed no significant difference in their ability to secret IFNy, indicating antigen- specificity and effector cytokine specificity. FIGURE 11B: PBMCs isolated from a patient were stimulated with an autoimmune peptide pool including antigens specific for type 1 diabetes (T1D). Cells were magnetically enriched for GARP and LAP and CD137 and were then sorted and cultured for 2 weeks in vitro. After culture, the cells were then restimulated with either the relevant T1D peptide pool (black bars) or an irrelevant influenza (HA1) peptide pool (gray bars) for 18 hours. Ceils were then plated into an ELISpot specific for IL-10 to assess their antigen-specificity in response to relevant and irrelevant peptide restimulation. GARP+ CD137+ fines had a significant increase (p<0.001) in the number of cells with the ability' to secrete IL-10 with relevant peptide stimulation as compared to irrelevant stimulation using T test analysis, while GARP- CD137+ fines showed no significant difference m IL-10-secreting cells when comparing irrelevant and relevant peptide stimulation.

FIGURE 12 illustrates that sorted GARP/LAP+ cell lines respond to antigen- specific restimulation through the upregulation of costimulatory molecules. GARP/LAP+ CD137+ cell lines respond to antigen-specific restimulation through the upregulation of Treg-associated costimulatory' molecules while GARP- CD137+ cell lines do not. PBMCs isolated from a patient were stimulated with an autoimmune peptide pool including antigens specific for type 1 diabetes (T1D). Cells were magnetically enriched for GARP and LAP and CD 137 and were then sorted and cultured for 2 weeks in vitro. After culture, the cells were then restimulated with either the relevant T1D peptide pool (solid line) or an irrelevant influenza (HA! ) peptide pool (dashed line) for 18 hours. Upregulation of costimulatory molecules CD 137 and 0X40, and activation molecule CD69 were seen only in the relevant restimulation of the GARP/LAP+ CD137+ lines.

FIGURES 13A and 13B graphically illustrate that GARP/LAP+ Tregs proliferate in response to vaccination whereas CD137+ Tregs do not. Correlations were plotted between percent Ki67 expression and either percent CD 137 expression (FIGURE. 13A) or percent GARP/LAP expression (FIGURE 13B). Eight patients were assayed for regulatory _' T cell proliferation two weeks after flu vaccination by restimulation with HA1 influenza peptide. Percent expression of GARP/LAP expression has been correlated with either CD137 alone or CD137+ GARP/LAP+. The correlation is extremely significant w'hen using GARP/LAP, but nonexistent in the CD137 alone group when compared to % of cells in these groups expressing Kr67, a marker we use here as a surrogate for recent antigen exposure from the vaccination, and therefore antigen specificity.

FIGURES 14A-14C graphically illustrate that allergen-specific GARP/LAP+ Tregs proliferate m response to seasonal allergen exposure more effectively than CD137+ Tregs alone. Increased Ki67 expression in GARP+ Tregs indicated increased antigen- specificity' after natural allergen exposure. Patient blood was collected before the alder allergen season and two weeks post-peak alder allergen exposure to assess changes in Ki67 expression in antigen-specific regulator _} T cell populations (FIGURE 14A). Within CD137+ regulatory T cells alone, there was no noticeable difference between Ki67 expression (FIGURE 14B), however GARP+ CD137+ regulator T cells showed a marked increase in Ki67 expression after alder allergen exposure (p=0.018, paired T test, n=6) (FIGURE 14C).

FIGURE 15 graphically illustrates that GARP and LAP are coexpressed within CD 137+ CD 154- Tregs. PBMCs from a healthy patient were stimulated polyclonally for 18 hours and enriched for GARP/LAP/CD137. We observed that within CD137+ Tregs, GARP and LAP are reliably co-expressed as parts of the latent TGFfl complex.

FIGURE 16 graphically illustrates that increased Foxp3 MFI in GARP/LAP+ Tregs indicates increased activation status. CD137+ CD25+ Tregs were compared to CD137+ GARP/LAP+ Tregs for Foxp3 MFI in a group of matched patients before and after influenza vaccination. 10 patients had blood drawn before influenza vaccination and 2 weeks post vaccination (n=2G). PBMCs were stimulated for 18 hours ex vivo with HA1 peptides and enriched for CD137/GARP/LAP. Ceils were then permeabilized and stained forFoxp3 expression. Significance was determined by paired T test with pO.OOOl.

FIGURES 17A-17C illustrate that single cell RNAseq shows that GARP/LAP+ Tregs have a mutually exclusive TCR repertoire as compared to CD 154+ Teffs. 48 CD 154+ Teffs and 48 CD 137+ GARP/LAP+ Tregs were sorted from a grass-allergic patient during season after 18 hour stimulation with Phi p 5a and Phi p 5b peptides. Using scRNAseq, we were able to effectively cluster the Teffs (orange) and Tregs (blue) using PCA analysis (FIGURE 17A). Using a volcano plot to express significantly upregulated genes m both the Tregs and Teffs, we found the CD 154 (CD40LG), IL-4, 1L-5, and IL-13 were upregulated in the effector ceils, and Foxp3, CD25 (IL2RA), and GARP (LRRC32) were upregulated in the Tregs (FIGURE 17B) When comparin expansion within the TCR repertoires of both groups, we observed mutual exclusivity of both TCR chains of both Teffs and Tregs, with increased expansion in the Teffs (FIGURE 17C).

FIGURE 18 graphically illustrates that increased proliferation in GARP/LAP+ Tregs m response to influenza vaccination only observed in peptide treatment groups. Correlations were plotted between percent Ki67 expression and either percent CD137 expression or percent G ARP/LAP expression. Eight patients were assayed for regulatory T cell proliferation two weeks after flu vaccination by restimulation with HA1 influenza peptide. Percent expression of GARP/LAP expression has been correlated with either CD137 alone or CDI37+ GARP/LAP+. The correlation is extremely significant when using GARP/LAP, but nonexistent m the CD 137 alone group when compared to % of ceils in these groups expressing Ki67, a marker we use here as a surrogate for recent antigen exposure from the vaccination. These correlations are not observed in the control groups.

FIGURES 19A and 19B graphically illustrate the use of a novel sequential CD154/GARP/CD137 assay to monitor the ex vivo frequency, surface markers and phenotypes of antigen-specific T cells. Sequential CD154/GARP/CD137 assays were carried out with peptide pools derived from TAA (FIGURE 19A) or self-antigens (FIGURE 19B). In this case, the self-antigen pool included only peptides from islet antigens, GAD65 and IGRP. Subjects tested include an HC individual (top row); and a renal cell carcinoma patient sampled before ICI therapy with nivolumab (middle row) and after two cycles of ICI (bottom row). The patient was assessed as having stable disease, but was taken off therapy because of complications due to a preexisting condition. Asterisks indicate massive increases in CD8+ Teff cells reactive with tumor- and auto- antigens.

DETAILED DESCRIPTION

Direct analysis and monitoring of immune cell populations in treated patients is essential and indispensable in clinical trials for immunotherapy, from cancer to allergy. The shift seen in the past decade from bulk population analyses towards antigen-specific analysis of immune ceil populations has created a deiuge of immune activation assays that claim superiorit' over alternate techniques. While effector T cells activation has been well-studied in terms of antigen-specificity, only more recently have similar techniques been applied in the quest for an activation marker that could serve as a panacea for regulatory' T cell (Treg) identification in antigen-specific populations.

An ideal isolation of antigen-specific regulatory T cells (Tregs) would involve a single positive selection step for a marker expressed exclusively by Tregs. Considering that much of the axis between CD154 and CD137 expression on Teffs and Tregs is poorly understood, the present disclosure describes the use of an alternate cell surface molecule to detect of antigen-specific Tregs that is closely linked to the induction and function of Tregs. As described in more detail below, the latent TGFft complex includes the pro-TGF receptor GARP (LRRC32) as well as the latency associated peptide (LAP) and is exclusively expressed by antigen-specific Tregs and not antigen-specific Teffs designated by CD 154 expression GARP expression is directly controlled by Foxp3 transcriptional programming, providing a direct link between the prevalence of this complex exclusively on activated Tregs While TGFfl is known to be a critical cytokine in the induction of Tregs, as well as an important inhibitory molecule in Treg suppressive capability, the latent complex (GARP and LAP) also selectively identifies antigen- specific Tregs m humans. GARP and LAP can be used in conjunction with costimulatory molecules like CD137, 0X40, CD27, ICOS, and CD25 to selectively identify and isolate highly pure populations of Tregs for immune monitoring or therapeutic applications. This development establishes that antigen-specificity in Tregs resides within the GARP+ Tregs and assays that address this subset of T cells do not need initial depletion of the CD 154+ Teff cells.

In accordance with the foregoing, the present disclosure provides methods and compositions for detecting, monitoring, enriching for, isolating, and using activated T regulator}' (Treg) cells. These and other facets of the disclosure encompass embodiments where the Tregs are antigen-specific.

As described above, regulatory T cells (also referred to herein as Tregs) refer to a subpopulation of T cells of the immune system. Previously referred to in the art as suppressor T cells, Tregs work to modulate immune responses to maintain tolerance to self-antigens and to prevent autoimmune disease. They are also known to have an effect to suppress or prevent inappropriate responses to allergens. They generally work by affecting a downregulation in the induction and proliferation of effector T cells Teffs.

In one aspect, the disclosure provides a method of detecting an activated regulator}' T (Treg) cell. The method comprises obtaining a sample comprising regulator _}' T (Treg) cells that have been exposed to an antigen and/or activator in a manner such that the sample may contain activated Treg cells; contacting the sample with a first detection molecule that specifically binds to a latent TORb complex protein and a second detection molecule that specifically binds to a co-stimulatory marker; and detecting a cell in the sample that is bound by the first detection molecule and the second detection molecule, thereby detecting an activated Treg cell.

The exposure of the Tregs to an antigen and/or acti vator allows any Treg cells in the sample to progress to activated Tregs if the conditions are appropriate. In some embodiments, the Tregs have been exposed to an antigen of interest, which facilitates detection of activated Tregs that are specific for the antigen of interest. The antigen of interest is typically a peptide or proteinaceous antigen. The exposure of the antigen of interest to the Tregs occurs "in a manner such that the sample may contain activated Treg ceils", which refers to the appropriate presentation of a peptide epitope to the Treg within an peptide/MHC complex. The peptide/MHC complex can be expressed by an antigen presenting cell (APC) where the APC has process the antigen and loaded the peptide into its own MHC, according to processes well-understood in the art. In such embodiments, APCs that have been exposed to the antigen of interest, have been exposed to the Tregs under conditions sufficient for the ceils to come into contact. The peptide MHC complex, if properly matched to the Treg, can stimulate activation of those Tregs that are specific for the peptide loaded onto the MHC. In other embodiments, the peptides/MHC complexes are not expressed on APCs, but rather are loaded onto and complexed with MHC monomer or multimers, according to approaches known in the art, to produce soluble peptide/MHC complexes that can also stimulate activation in Tregs that are matched to the MHC type and specifically recognize the epitope in the MHC context.

In other embodiments, the Treg cells have been exposed to an activator that, with sufficient quantities and conditions, can produce non-specific activation of Treg cells. The activator is a composition comprising one or more costimulatory molecules that bind to and activate one or more co-stimulatory receptors selected from CD 137, 0X40 (CD 134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and CD25. In some embodiments, the activator comprises costimulatory molecules that bind to and activate one or more receptors selected from CDS, CD2, and CD28. The co-stimulator molecules can be ligands or antibodies that bind to and activate signalin cascades associated with the above receptor molecules. Ligands and antibodies (or functional fragments thereof) are well-known in the art.

The one or more co-stimulatory molecules can be administered as a cocktail that, in sufficient concentration, can stimulate non-antigen specific activation of Tregs the sample. In some embodiments, the activator is contacted to the sample with the antigen of interest (as described above), where the activator is formulated to merely enhance the activation of antigen-specific Tregs and not the general population of the Tregs (i.e., not in an antigen-independent fashion). Application of an exemplary activator cocktail for enhanced antigen-specific activation is described in more detail below.

As used herein, the term "latent TOTb complex protein" includes any one of GARP (LRRC32), LAP (latent-associated peptide), and TGFfll-4. This disclosure encompasses embodiments where the first detection molecule can specifically bind to any one of the above latent TORb complex proteins. Considering that these protein associate in a complex, in other embodiments, the first detection molecule can bind to a complex or association of any two or more of the above proteins. In some embodiments, additional detection molecules are used that bind to additional latent TORb complex proteins (or complexes thereof) that are different than the latent TORb complex proteins (or complexes thereof) that are bound by the first detection molecule.

The second detection molecule specifically binds to a co-stimulatory marker on a Treg ceil that can indicate progression to an activated state. The co-stimulatory marker can he, for example, receptors specific or substantially unique to activated Treg cells, which typically are receptors that would bind to cognate ligands on, e.g., APCs. In some embodiments, the co-stimulatory marker can be selected from CD 137, 0X40 (CD 134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and CD25.

In some embodiments, the method further comprises contacting the sample with a third detection molecule that specifically binds to CD4, and wherein detecting an activated T regulatory cell comprises detecting a cell in the sample that is bound by the first detection molecule, the second detection molecule, and the third detection molecule. It is the combination of the detectable presence of the latent TίϊRb complex protein, the co-stimulatory molecule, and in some cases, CD4, that provides an accurate indication of activation of the Treg cells. When the cells are produces as the result of exposure to a particular antigen, the activated cells are specific for the antigen. One distinct advantage, as described in more detail herein, use of these combinations to characterize the (potentially antigen-specific) activated Tregs is that the status of the Treg cells does not have to be monitored, depleted, or otherwise discounted based on CD154 activation, which is currently required in other methods.

Potential structures for detection molecules will now be described. As used herein, the terms "detection molecule" and "enrichment molecule" (see below) refer to any molecule having an ability to bind to a specific target molecule (e.g., a marker of interest such as a latent TORb complex protein or a eo-stimulatory molecule) with a specific affinity' (i.e., detectable over background). Exemplary categories of detection/enrichment molecules that can be used in the context of the present disclosure include antibodies, antibody derivatives (also referred to as "antibody-like molecules"), functional target molecule-binding portions of antibodies or antibody-like molecules, peptides that specifically interact with a particular target antigen/marker (e.g., peptibodies), receptor molecules that specifically interact with a particular target antigen/marker, functional target/marker-binding portions of proteins that comprise a ligand-binding portion of a receptor that specifically binds a particular target/marker, ligands themselves (or moieties that incorporate a ligand), antigen/marker-binding scaffolds (e.g., DARPms, HEAT repeat proteins, ARM repeat proteins, tetratricopeptide repeal proteins, and other scaffolds based on naturally occurring repeat proteins, etc., (see, e.g., Boersma and Pluckthun, Curr. Opin. Biotechnol. 22: 849-857, 2011, and references cited therein, incorporated herein by reference), aptamers or target/marker-binding portions thereof.

In some embodiments, the detection/enrichment molecule is an antibody. As used herein, the term "antibody” encompasses antibodies and antibody fragments thereof, derived from any antibody-producing mammal (e.g., mouse, rat, rabbit, and primate including human), that specifically bind to a target molecule of interest (i.e., latent TGF complex protein, co-stimulatory marker, etc.) Exemplary antibodies include polyclonal, monoclonal and recombinant antibodies; multispecific antibodies (e.g., bi specific antibodies); humanized antibodies; murine antibodies; chimeric, mouse-human, mouse- primate, primate-human monoclonal antibodies; and anti-idiotype antibodies. The antigen-binding molecule can be any intact antibody molecule or fragment thereof (e.g., with a functional antigen-binding domain).

An antibody fragment is a portion derived from or related to a full-length antibody, preferably including the complementarity-determining regions (CDRs), antigen binding regions, or variable regions thereof. Illustrative examples of antibody fragments useful in the present disclosure include Fab, Fab', F(ab)2, F(ab')2 and Fv fragments, scFv fragments, diabodies, linear antibodies, single-chain antibody molecules, multispecific antibodies formed from antibody fragments, and the like. A "single-chain Fv" or "scFv" antibody fragment comprises the VJJ and Vp domains of an antibody, wherein these domains are present in a single polypeptide chain. The Fv polypeptide can further comprise a polypeptide linker between the VJJ and V domains, which enables the scFv to form the desired structure for antigen binding. Antibody fragments can be produced recombinanily, or through enzymatic digestion.

Antibodies can be further modified to suit various uses. For example, if the detection markers are intended for administration to a subject, a chimeric antibody can be used to minimize antigenicity of the binding molecule itself. A chimeric antibody is a recombinant protein that contains the variable domains and complementarity-determining regions (CDRs) derived from a non-human species (e.g., rodent) antibody, while the remainder of the antibody molecule is derived from a human antibody. A "humanized antibody" is a chimeric antibody that comprises a minimal sequence that conforms to specific complementanty-determimng regions derived from non-human immunoglobulin that is transplanted into a human antibody framework. Humanized antibodies are typically recombinant proteins in which only the antibody complementarity-determining regions (CDRs) are of non-human origin.

Production of antibodies can be accomplished using any technique commonly known in the art. For example, the production of a polyclonal antibody can be accomplished by administering an immunogen containing the target molecule of interest to an antibody-producing animal. For example, the target molecule of interest can be administered to a mammal (e.g., a rat, a mouse, a rabbit, a chicken, cattle, a monkey, a pig, a horse, a sheep, a goat, a dog, a cat, a guinea pig, a hamster) or a bird (e.g., a chicken) so as to induce production of a serum containing an target molecule-specific polyclonal antibody. The target molecule can be administered in combination with other components known to facilitate induction of a B-cell response, such as any appropriate adjuvant known in the art. Furthermore, the polyclonal antibody reagent can be further processed to remove or subtract any antibody members that have unacceptable affinity for antigens that are not the antigen of interest. Tire resulting polyclonal antibody reagent will exhibit enhanced specificity for the target molecule and are useful for detection and quantification purposes. Many approaches for adsorption of polyclonal antibody reagents to reduce cross-reactivity exist, are familiar to persons of ordinary skill in the art, and are encompassed by the present disclosure.

Monoclonal antibodies can be prepared using a wide variety of techniques known in the art including the use of hybridoma, recombinant, and phage display technologies, or a combination thereof. For example, monoclonal antibodies can be produced using hybridoma techniques including those known in the art and taught, for example, in Harlow' et al., Antibodies: A Laboratory Manual, (Cold Spring Harbor Laboratory' Press, 2nd ed. 1988); Hammerling et ah, in: Monoclonal Antibodies and T-Cell Hybridomas 563-681 (Elsevier, N.Y., 1981), incorporated herein by reference in their entireties. The term "monoclonal antibody" refers to an antibody that is derived from a single clone. including any eukaryotic, prokaryotic, or phage clone, and not the method by which it is produced. Methods for producing and screening for specific antibodies using hybridoma technology are routine and well-known in the art.

Antibody fragments that recognize specific epitopes can be generated by any technique known to those of skill in the art. For example, Fab and F(ab')2 fragments of the invention can be produced by proteolytic cleavage of immunoglobulin molecules, using enzymes such as papain (to produce Fab fragments) or pepsin (to produce F(ab')2 fragments). F(ab¾ fragments contain the variable region, the light chain constant region and the CHI domain of the heavy chain. Further, the antibodies of the present invention can also be generated using various phage display methods known in the art.

As used herein, the term "aptamer" refers to oligonucleic or peptide molecules that can hind to specific target molecules. Nucleic acid aptamers usually are short strands of oligonucleotides that exhibit specific binding properties. They are typically produced through several rounds of in vitro selection or systematic evolution by exponential enrichment protocols to select for the best binding properties, including avidity and selectivity. One type of useful nucleic acid aptamers are thioaptamers, in which some or all of the non-bridging oxygen atoms of phosphodiester bonds have been replaced with sulfur atoms, which increases binding energies with proteins and slows degradation caused by nuclease enzymes. In some embodiments, nucleic acid aptamers contain modified bases that possess altered side-chains that can facilitate the aptamer/target binding.

Peptide aptamers are protein molecules that often contain a peptide loop attached at both ends to a protamersem scaffold. The loop typically has between 10 and 20 amino acids long, and the scaffold is typically any protein that is soluble and compact. One example of the protein scaffold is Thioredoxin-A, wherein the loop structure can be inserted within the reducing active site. Peptide aptamers can be generated/selected from various types of libraries, such as phage display, mRNA display, ribosome display, bacterial display and yeast display libraries.

In some embodiments, the antigen-binding molecule is a receptor molecule or comprises a binding domain of a receptor molecule. The receptor molecule can be any receptor known that can specifically bind the target molecule as the ligand. In one embodiment, the antigen- binding molecule is or contains a protein binding domain that enables the detection of the target molecule. In yet another embodiment, the antigen-binding molecule can be or comprise a ligand or portion of a ligand that is specific for a receptor or a binding domain of a protein, which receptor or a binding domain of a protein would then serve as the target molecule.

As used herein, the term "selectively binds" refers to the ability of the antigen binding molecule to bind to the target molecule, without significant binding to other unrelated molecules, under standard conditions known m the art. Tire antigen-binding molecule can bind to other peptides, polypeptides, or proteins, but with lower affinity as determined by, e.g., immunoassays, BIAcore, or other assays known in the art. However, antigen-binding molecule preferably does not cross-react with other antigens.

The detection reagents are typically delectably labelled. Conventional labels that can be used are well-known in the art. Typically, the first and second (and third, etc.) detection molecules are each labeled with a different label such that they can be distinguishable from each other. The detectable labels facilitate efficient detection of binding of the first and second (and third, etc.) binding molecules to their intended targets on the Treg ceils.

In method to observe the detectable labels can be used. One exemplary method for detecting the binding is the use of fluorescence-activated cell sorting (FACS), which can detect different fluorescent labels on cells to permit detection and sorting of cells that are positive for the latent TGFf) complex proiein(s) and co-stimulatory marker. Another exemplary method for detecting binding is the use of mass cytometry (CyTOF). In CyTOF, the binding molecules are labeled with isotopicaliy pure elements produce signature signals in a time-of-flight mass spectrometer. Treg cells that have the first and second detection molecules bound thereto will produce a distinct signature as compared to Treg cells that have only one, or ceils that have neither, of the detection molecules bound thereto. Both FACS and CyTOF can also be used to collect (e.g., enrich, isolate, etc.) the cells. An advantage of FACS is that the cells can remain alive.

In some embodiments, the method comprises enriching for activated Tregs based on the binding status detected in the method. This can be accomplished, for example, using FACS that sorts and collects cells with a particular fluorescence pattern based on the binding of the detection molecules. As used herein, "enriching" means increasing the relative proportion of the activated Tregs as compared to the sample. In some embodiments, the activated Tregs are substantially isolated from other cells, such as quiescent/nafve Tregs. In this context, "substantially" implies near complete but does not require complete isolation. Some amount of cells that are not activated Tregs are permitted so long as the proportion remains small (e.g., 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2 %, 1%, or less) of the batch.

In some embodiments, the sample can be obtained from or derived from a subject. The subject can be any vertebrate animal that produces Treg cells, e.g., human, mouse, rat, rabbit, dog, cat, horse, cow, goat, and the like. Tire sample can be, or be derived from, whole blood, tissue, lavage, tumor biopsy, and the like.

In some embodiments, the method further comprises an initial step of obtaining or deri ving the sample of Treg cells from a subject.

In some embodiments, the method comprises the active step of contacting the initial sample comprising Treg cells with sufficient antigen or activator in a manner and amount such that at least a portion of the Treg ceils are activated. Exposure to antigen of interest and activator compositions are described above. Furthermore, a description of an illustrative approach to contacting the sample comprising Tregs with antigen of interest and activator to activated Tregs is provided below.

In some embodiments, the method further comprises contacting the cells with a co-stimulatory molecule, as described above.

In another aspect, the disclosure provides a method of producing an enriched population of activated T regulatory (Treg) cells. The method comprises incubating a sample comprising Treg cells with an antigen of interest and/or an activator in a manner and amount such that at least a portion of the Treg cells are activated; contacting the sample with a first enrichment molecule that specifically binds to a latent TGFp complex protein and a second enrichment molecule that specifically binds to a co-stimuiatory marker; and enriching for cells that are bound to the first enrichment molecule arid the second enrichment molecule, thereby producing an enriched population of activated Treg cells.

Elements of the method, such as incubation with or exposure to the antigen of interest and/or the activator, the antigen of interest and/or the activator elements themselves, the latent TGFp complex protein, the co-stimulatory marker, steps of enrichment, arid the like, are discussed in more detail above and are applicable to this aspect of the disclosure. The term "enrichment molecule" invokes the same structure implied by "detection molecule" as used and described above, but is so named to invoke the intended function of the molecules. In some embodiments, the method further comprises an initial step of obtaining the sample of sample of Treg cells from a subject. Such a sample can be, or be derived from, whole blood, tissue, lavage, tumor biopsy, and the like, which is obtained from a subject.

In some embodiments, the sample is incubated with an antigen of interest, as described above, which results in antigen-specificity of the activated Tregs. In some embodiments, the sample is incubated with an activator, as described above, in a manner that induced antigen-independent activation of the Tregs. In yet further embodiments, the sample is incubated with an antigen of interest and sufficient activator, in a manner that further promotes the activation of antigen -sped fie Tregs but does not result in substantial activation of Tregs that are not specific for the antigen of interest.

The enriched Tregs can be administered to a subject in need thereof, for example, to treat or ameliorate a condition characterized by inappropriate or excessive Teff response to an antigen. Such conditions include allergies, autoimmune disease, graft vs host reactivity, and the like. Typically, the enriched Tregs are MHC (HLA) matched to the subject. In some embodiments, the subject from whom the sample of Tregs was originally obtained is the same subject receiving administration of the enriched activated Tregs.

In other aspects, the disclosure also provides the activated Treg cell produced from the above methods. The activated Treg ceil can be specific for an antigen of interest used to produce the activated Treg. Additionally, the Treg can be in an enriched or isolated population of activated Tregs with the same antigen specificity.

As indicated above, the disclosure also encompasses methods of using the disclosed activated Treg cells. In one aspect, the disclosure provides a method of treating a condition treatable by administration of an activated Treg. Tire method comprises administering the disclosed Tregs. In some embodiments, the condition is an allergy, autoimmune disease, graft vs host reactivity, and the like. The Tregs can be rationally produced to be specific for an antigen of interest that is determined to be a causal component of the condition to be treated. As used herein, the terms "treat" or "treatment" refer to therapeutic interventions to stop, ameliorate, or manage a disease, disorder, or condition. Therapeutic or prophylactic/preventive benefits include improved clinical outcome; lessening or alleviation of symptoms associated with a condition; decreased occurrence of symptoms; improved quality of life; longer disease-free status; diminishment of extent of disease, stabilization of disease state; delay of disease progression; remission; survival; prolonged survival; or any combination thereof.

The disclosed strategies for detecting and enriching activated Tregs based on a combination of markers (which do not require any detection or depletion of CD ! 54) can incorporated into various investigatory or diagnostic-type methods.

For example, m one aspect, the disclosure provides a method for monitoring a T regulatory (Treg) cell response to potential antigens. The method comprises contacting a sample with a first detection molecule that specifically binds to a latent TGFj] complex protein and a second detection molecule that specifically binds to a co stimulatory marker. The sample comprises T cells and was obtained from a subject suspected of being exposed to an antigen. The method also comprises detecting or quantifying cells in the sample that are bound by both the first detection molecule and the second detection molecule. Cells that are bound by the first detection molecule and the second detection molecule are established as being activated Tregs. Thus, the absence, presence, or relative abundance of cells bound by both the first detection molecule and the second detection molecule is indicative of a state of the Treg cell response.

This method can be applied, for example, on a sample obtained or derived from a subject after the subject has been administered or has otherwise been exposed to an antigen of interest. The method can determine whether the subject generates Tregs in response to the administration or exposure. Alternatively, the sample could comprise naive T cells and the method further comprises incubating the sample with an antigen of interest, as described above. For example, this can be applied to determine the responsivity of Treg development and activation to allergens, autoantigens, and/or cancer antigens, and the like. The presence, absence, or level of activated, antigen-specific Tregs is indicative of the subject's capacity to generate Tregs specific for the antigen of interest.

In another aspect, the disclosure provides a method for monitoring the sensitivity of a subject to an antigen of interest. The method comprises contacting a sample comprising Treg cells with a first detection molecule that specifically binds to a latent TGFp complex protein and a second detection molecule that specifically binds to a co-stimulatory marker, wherein the Treg cells have been exposed to an antigen of interest and the sample was obtained from the subject; detecting or quantifying the cells in the sample that are bound by both the first detection molecule and the second detection molecule; comparing the number of cells in the sample that are bound by both the fust detection molecule and the second detection molecule to an established threshold to determine the sensitivity of the subject to the antigen of interest.

The Treg cells can be been exposed to an antigen of interest, as described above, in vivo, or ex vivo after the same was obtained.

In some embodiments, a relative increase in the number of cells in the sample that are bound by both the first detection molecule and the second detection molecules compared to the established threshold indicates a low sensitivity of the subject to the antigen of interest. The threshold can be established at an earlier time point from the subject using the same method. A relative increase in cells that are bound by both the first detection molecule and the second detection molecule over the established threshold indicates a decreasing sensitivity to the antigen of interest. Alternatively, a relative decrease in cells that are bound by both the first detection molecule and the second detection molecule over the established threshold indicates an increasing sensitivity to the antigen of interest. In some embodiments, the threshold is established during or prior to administration of the therapeutic treatment and the sensitivity of the subject to the antigen of interest at a later time point is indicative of the efficacy of the therapeutic treatment.

In another aspect, the disclosure provides a method of screening for T regulatory (Treg) cell stimulatory epitopes from an allergen or antigen of interest. The method comprises steps of: obtaining a sample comprising Treg cells; exposing the Treg cells to an epitope derived from an antigen of interest in an MHC context; contacting the Treg cells with a first detection molecule that specifically binds to a latent TGFfi complex protein and a second detection molecule that specifically binds to a co-stimulatory marker; and quantifying the relative abundance of cells in the sample that are bound by both the first detection molecule and the second detection molecule. A high relative abundance indicates that the epitope stimulates development of activated Treg cells.

Exposing Treg cells to an epitope is generally described above, and can be mediated by antigen presenting cells or can be accomplished with soluble epitope-loaded MHC complexes. The method can further comprise exposing the Tregs to an appropriate amount of activator to facilitate activation of antigen-specific Tregs, but not all Tregs generally. Elements, such as the activator are described above.

The method can be repeated multiple times against a panel of different epitopes. For example, multiple epitopes derived from the same antigen can be tested to determine the most (or least) stimulating epitope from the antigen for purposes of producing Tregs. Again, the antigens can be selected from allergen, autoantigen, cancer antigen, and the like.

In another aspect, the disclosure provides a method of identifying an MHC molecule/complex that binds to a Treg-stimulatory epitope. The method comprises: obtaining a sample comprising Treg cells; exposing the Treg cells to an epitope derived from an antigen of interest in an MHC context; contacting the Treg cells with a first detection molecule that specifically binds to a latent TORb complex protein and a second detection molecule that specifically binds to a co stimulatory marker; quantifying the relative abundance of cells in the sample that are bound by both the first detection molecule and the second detection molecule, wherein a high relative abundance indicates that the epitope stimulates development of activated Treg cells; and characterizing the MHC molecule complexed with the epitope that resulted in a high abundance of cells bound by both the first detection molecule and the second detection molecule.

For example, the Tregs can be exposed separately to a panel of different APCs that have been allowed to process and display an epitope. After quantification of the resulting activated Tregs, the MHC from the APCs resulting in the best activation can be characterized.

In yet another aspect, the disclosure provides for methods of testing effect of a putative therapeutic compound on a Treg ceil, including on the development of a Treg ceil into an activated state. The method comprises exposing the disclosed Treg ceil (before or after activation) and monitoring the effect of the therapeutic compound.

For any of these applications, the latent TGFf> complex protein and co-stimuiatory marker, activator, or any other elements not explicitly described are defined previously in other contexts, and are understood to be equally applicable here.

Unless specifically defined herein, all terms used herein have the same meaning as they would to one skilled in the art of the present invention. Practitioners are particularly directed to Sambrook I., et al. (eds.) Molecular Cloning: A Laboratory Manual, 3rd ed., Cold Spring Harbor Press, Plainsview, New York (2001); Ausubel, F.M., et al. (eds.), Current Protocols in Molecular Biolog ^< , John Wiley & Sons, New York (2010); and Coligan, J.E., et al. (eds.). Current Protocols in Immunology, John Wiley & Sons, New York (2010) for definitions and terms of art. The use of the term "or" in the claims is used to mean "and/or" unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive, although the disclosure supports a definition that refers to only alternatives and "and/or."

Following long-standing patent law, the words "a" and "an," when used in conjunction with the word "comprising" in the claims or specification, denotes one or more, unless specifically noted.

Unless the context clearly requires otherwise, throughout the description and the claims, the words "comprise," "comprising," and the like, are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is to indicate, m the sense of "including, but not limited to." Words using the singular or plural number also include the plural and singular number, respectively. Additionally, the words "herein," "above," and "below',” and words of similar import, when used in this application, shall refer to this application as a whole and not to any particular portions of the application. Unless stated otherwise, the term "about" implies minor variation around the stated value of no more than 10% (above or below').

Disclosed are materials, compositions, and devices that can be used for, can be used in conjunction with, can be used in preparation for, or are products of the disclosed methods and compositions. It is understood that, when combinations, subsets, interactions, groups, etc., of these materials are disclosed, each of various individual and collective combinations is specifically contemplated, even though specific reference to each and every' single combination and permutation of these compounds may not be explicitly disclosed. This concept applies to all aspects of this disclosure including, but not limited to, steps in the described methods. Thus, specific elements of any foregoing embodiments can be combined or substituted for elements m other embodiments. For example, if there are a variety of additional steps that can be performed, it is understood that each of these additional steps can be performed with any specific method steps or combination of method steps of the disclosed methods, and that each such combination or subset of combinations is specifically contemplated and should be considered disclosed. Additionally, it is understood that the embodiments described herein can be implemented using any suitable material such as those described elsewhere herein or as known in the art.

Publications cited herein and the subject matter for winch they are cited are hereby specifically incorporated by reference in their entireties. The following examples are provided for the purpose of illustrating, not limiting, the disclosure.

EXAMPLES

Example 1

This Example describes an exemplary protocol for performing an antigen-specific

T regulatory ceil assay of the present disclosure (also referred to as ASTRA). Use of the ASTRA approach in combination with multicolor flow cytometry can be a valuable tool for detecting and monitoring of CD4+ regulatory _' T cell response to a particular antigen, determining which epitopes or antigen fragments that bind to MHC class II molecule to elicit a regulatory T cell response, and for the development of specific immunotherapies.

Unlike other assays, the ASTRA method allows simultaneous assessment of other ceil phenotypes or functions, and is compatible with downstream RNA-based assays and preserves cell viability. Accordingly, the ASTRA method allows for tracking antigen- specific T regulatory cells and can be applied for diverse applications, such as:

methods for monitoring antigen-specific T regulatory cell responses;

methods for isolating and purifying live antigen-specific T regulator) _' cells;

methods for identifying MHC class Il-restricted immune epitopes of a predetermined polypeptide antigen that elicit a regulatory' T cell response;

methods for determining the MHC class II molecule that binds to the regulatory' T cell epitopes; and

methods for testing effect of drug compound on live antigen-specific T regulatory cells.

To illustrate the efficacy of the ASTRA method, human T cells were stimulated and characterized according to the below exemplary protocol:

1. PBMCs were isolated from whole blood using density gradient centrifugation.

2. PBMCs (l-2x l0 ⁷ cells) were then stimulated for 18 hr in RPMI 1640 supplemented with 5% AB serum, with the antigen (peptide, protein, vaccine, crude extract, or stimulatory cocktail) at 37°C and 5% C0 ₂ .

3. After stimulation, ceils were centrifuged at 1200 RPM for 5 minutes.

4. Enriching antibodies (a-GARP, a-LAP) were added to remaining cells at a volume of 3 mE per 50 million cells. 5. Cells were vortexed and resuspended and allowed to incubate for 15-20 minutes at room temperature in the dark.

6. Cells were washed with 4 L of PBS and centrifuged at 1200 RPM for 5 minutes and decanted.

7. 35 pL of anti-fluorophore microbeads were added to each tube per 50 million cells. Cells were then resuspended and vortexed, and allowed to incubate for 10- 15 minutes at RT in the dark.

8. Ceils were washed with 4 mL of PBS and centrifuged at 1200 RPM for 5 minutes and decanted.

9. During step 6, 1 mL of PBS was added to the appropriate number of columns or tubes loaded onto a magnet. Flow through was collected in a new- tube.

10. Ceils were resuspended m 1 mL. of PBS per 50 million cells. Resuspended cells were added to a magnetic column or tube and allow to completely pass through column or settle while the flow through was collected in the same tube as step 7.

11. Another 1 mL of PBS was added to the column and three drops were allowed to fall into the flow through tube while still attached to the magnet. After three drops, the column was transferred to a new tube and the remaining volume of PBS was allowed to flow through the column.

12. After column the column was eluted, 1 mL of PBS was added to the liberated column and plunged at a rate of 2 drops/second into the elution tube.

13. Elution column was centrifuged at 1200 RPM for 5 minutes and decanted.

14. Surface staining antibody cocktail (including at minimum GARP and/or LAP) w¾s then added to remaining volume of cells and allowed to incubate for 15-20 minutes at room temperature in the dark. Surface panel addressed by the cocktail can optionally include CD154, CD69, ICOS for live-cell analysis of highly -enriched antigen- specific regulatory T cells. For fixation with Foxp3, Ki67 and Helios staining, proceed to step 15. Otherwise, analyze live ceils after step 15.

15. Cells were washed w'ith 4 mL of PBS and centrifuged at 1200 RPM for 5 minutes and decanted.

16. 200 u.L of FoxP3 fixation buffer (4x dilution of stock) was added to resuspended cells and incubated for 1 hour at 4°C in the dark (Alternately, 20-30 minutes at RT in the dark has been shown to produce similar results) 17. Fixed cells were washed with 4 mL of lx permeabilization/wash buffer (20x stock in ultrapure water) and centrifuged for 5 minutes at 1500 RPM and decanted carefully. It is noted that fixed ceils tend to peilet less well, but viability is no longer a concern.

18. Nuclear/intracellular staining antibody cocktail was added to the remaining volume of cells. Cells were vortexed and incubated at 4°C for 18-24 hours. Shorter incubations times can be used with discretion.

19. Before analysis, cells were vrashed with 4 mL of lx permeabilizationAvash buffer and centrifuged for 5 minutes at 1500 RPM and decanted carefully. Cells were then washed with 4 mL of PBS and centrifuged for 5 minutes at 1500 RPM again (cells not in wash buffer will no longer remain permeabilized).

20. Ceils were resuspended in 200 pL of PBS and analyzed via flow cytometr'.

The cells detected and sorted by flow cytometry were confirmed to be highly enriched antigen-specific Foxp3-t- Treg cells (see, e.g., disclosure of Example 2, below), demonstrating that the ASTRA approach results in enriched, live Tregs specific for a stimulating antigen that can then be isolated by cell sorting or alternatively using magnetic enrichment. Thus, using GARP/LAP expression in conjunction with other activation-induced markers (e.g., CD137, 0X40, CD27, CD25) can be used to detect highly enriched antigen-specific Foxp3+ Tregs.

Example 2

This Example describes the further development of the ASTRA approach that utilizes combinations of CD 137 and TORb complex components GARP and LAP as markers to isolate and characterize antigen specific Tregs.

Abstract

Alterations in the number and function of regulatory' T cells (Tregs) have been implicated in many diseases such as autoimmunity, graft versus host disease, allergy and cancer. Information about their frequencies, phenotypes, and functional capacities is essential to understand the mechanisms of protective immunity' or immunopathology. In this study we report a new flow' cytometric assay that enables the direct ex vivo characterization and isolation of antigen-specific Treg cells. This method relies on the de novo expression of the latent TOEb complex (GARP/LAP) within GDI 37+ CD4+ T cells following brief antigen stimulation. While induction of CD 137 denotes Foxp3-stable Tregs, we show that antigen-specific Treg cells specifically fall into GARP/LAP+ CD137+ compartment. Importantly, surface staining and enrichment of GARP/LAP allows for purification and isolation of live, antigen-specific Tregs. Quantitative and qualitative assessment of these CD4+ Tregs cells that respond to any antigen of interest will provide new avenues for cellular immunotherapies. Antigen-specific markers for regulatory Tregs have long been sought after to provide accurate assessments of their frequency and phenotype in autoimmunity, infectious disease, allergy, and cancer. Here we demonstrate that using surface-expressed GARP/LAP, we can efficiently enrich Tregs based on their reactivity to specific antigens. These Tregs can be effectively characterized as expressing higher levels Foxp3, along with other Treg-associated markers, and can be sorted, cultured, and confirmed with epitope-specific tetramer staining. By analyzing antigen-specific Treg populations as opposed to bulk Treg populations, nuanced characteristics of regulator ' subsets can be ascertained which would otherwise be occluded by non-specific cellular contamination

Introduction

Regulatory CD4+ T cells (Tregs) play a critical role in immunity and tolerance to both self and foreign antigens. Starting with the discovery' and characterization of the transcription factor Foxp3, the continued study of Tregs and their role in autoimmunity, cancer, allergy, and infectious disease has become central to the field of immunology. Whether elicited by allergy immunotherapy or dysregulated in autoimmune disease, the ability to identify Tregs based on antigen-specificity affords the ability to view the central players in tolerance separate from the amalgamation of the immunological milieu. Recent advances in major histocompatibility class II (MHCII) tetramer technology ⁷ have allowed for the pinpointing of antigen-specific T cell responses, but with the limitation of epitope MHC restriction in both mice and humans.

The advent of assays using endogenous costimuiatory upregulation upon TCR activation have allowed us to use markers like CD 154 (CD40L) and CD69 to capture antigen specificity of effector T cells (Teffs), free from the restriction of MHC, These markers have been well characterized and translated into robust assays, however they are not without their own shortcomings. While similar markers for antigen-specificity have been sought after to assess Tregs, there is growing evidence that none of the costimuiatory molecules used to designate antigen-specificity within the Teffs can be used to distinguish Treg subsets without first parsing the two groups. Recent w'ork has shown that CD 154 is unable to track Tregs m humans and should be used to remove Teffs from the Treg isolation. Instead these groups use CD 137 (4- IBB) as a marker for highly stable Foxp3+ Tregs. Other makers, such as the costimulatory molecule and CD137 family member 0X40, surface components of the latent TGFft receptor GARP and LAP (latency associated peptide), and IL-1 receptor type i/II (CD121a/b) have been shown to be enriched on activated regulatory T ceils in humans. These stimulation-based assays have the advantage of elucidating epitope-specific differences between Teff and Treg specificities without the restriction of patient HLA allele differences as well as HLA binding differences. MHC II tetramers have long been considered the best indicator of antigen-specificity in Teffs, however their compatibility with Tregs still remains murky, an issue made even more unclear by unanswered questions involving the elicitation of peripheral Tregs (pTregs) and thymic Tregs (tTregs) from naive subsets in humans and nebulous results involving the ability of singular epitopes to dictate multiple T cell fates. It has been shown that the TCR repertoire of Tregs is largely non-overlapping with the conventional subset, and of significantly higher affinity.

Here, we show' an assay that has been developed to capture antigen-specific Tregs using surface-expressed molecules upreguiated upon activation with specific antigen that are not influenced by bystander activation. By using CD137 and latent TGFjt complex components (GARP and LAP), we are able to reliably identify antigen-specific Tregs in multiple disease states (allergy, autoimmunity, cancer) as w-ell as in healthy individuals. The Tregs can be isolated from the circulation and analyzed ex vivo as well as sorted into and maintained as T cell lines. The identified cells are bona fide Tregs, expressing Foxp3, as well as other common regulatory _' molecules CD25, Helios, and CTLA-4. Upon isolation and culture of antigen-specific Tregs, tetramer can be used to validate the epitope-specific nature of the cells, as well as antigen-specific induction of TL-10 and re expression of relevant costimulatory molecules after resting. mRNA sequencing also provides insight into the phenotype of these cells in comparison to other antigen-specific effector T cell subsets. By using latent TGFf) components GARP and LAP as markers of antigen specificity', as w'ell as costimulatory molecule CD 137 as a marker of demethylated TSDR in regulatory T cells, it is possible to enumerate and isolate highly enriched, Foxp3-stable, antigen-specific regulatory T cells from human patients. These antigen-specific Tregs can be used as an immune-monitoring technique or as a therapeutic target for disease treatment. Results

Activation-induced regulatory T cells have high levels of de novo GARP/LAP expression

When attempting to capture antigen-specific Tregs in allergic patients during alder allergy season, we found that alder antigen pooled tetramers were unable to detect Foxp3+ Tregs (FIGURE 1A). We ran a CD154 upregulation assay on the same patient and found that CD 154 did not capture Foxp3+ Tregs and resembled the allergic effector phenotype observed in the ex vivo tetramer stain, aside from upregulated costimulatory molecules. This led us to conclude that tetramer staining perhaps biased towards effector T cells, like the CD154 assay, leaving the antigen-specific regulator' T cells undetected. Building upon earlier work done m characterizing the distinction between effector T cells and regulatory T cells we began looking at the CD137 expression on PBMCs from the same alder allergic patient. We observed a distinct population expressing GARP and LAP, and looked at eostimulatory activation on this population and found it had sharply upregulated features of activation (FIGURE IB). We also observed eostimulatory molecule 0X40 along with GARP/LAP and noticed it was largely absent on CD137mid regulatory T cells (FIGURE 2; inset box A), but highly upregulated on only the regulatory T cells expressing the highest abundance of CD137 (FIGURE 2; inset box B). We also found that other TNSFRSF protein family members, of which CD137 and 0X40 are members of, were also expressed in highest abundance on these Tregs. The clustering of multiple TNSFRSF eostimulatory molecules (CD137, 0X40, CD27) along with the possibly functionally polarizing latent TOEb complex components (GARP, LAP) led us to further investigate these cells as highly-activated regulatory T cells induced by antigen- specific stimulation. We also observed that both components of the latent TGFfi complex, GARP and LAP, were reliably coexpressed on CD 137+ CD 154- regulatory T cells (FIGURE 15). This allowed us to co-stain for them to increase fluorescence intensity during flow cytometry as well as enrichment efficiency.

We then stained for CD137+ Tregs expressing GARP/LAP in PCMC's from multiple disease models and found that they exhibit higher levels of activation. Patient PBMCs were isolated from various disease states and stimulated with relevant peptide antigen pools in cancer (tumor-associated antigens), autoimmunity (T1D pancreatic islet antigens), allergy (Ain g 1 alder antigen), and vaccine (HAI influenza antigen) for 16 hours. Cultures were then enriched for CD 137 and assessed for GARP/LAP expression, as well as other Treg-related activation markers (FIGURE 3 A). Across all antigen models, GARP/LAP+ Tregs expressed increased activation as compared to CD137+ Tregs. In a polyclonal aCD3 aCD28 stimulation, the majority of the cells expressed G ARP/LAP and no significant differences were seen in the activation states of the CD137+ and GARP/LAP+ Tregs. We then determined that optimal CD137 and GARP/LAP expression occurs beginning at 16 hours of ex vivo stimulation with peptide antigen. PBMCs from a healthy patient were stimulated with influenza HAf peptide antigen for 0, 2, 6, 8, 12, 16 and 18 hours to determine optimal expression. Although Foxp3+ CD25+ Tregs begin to express slight levels of CD137 and GARP/LAP at 6 hours, expression plateaus at 12-16 hours. Similar trends can be seen when tracking percent expression within memory CD4+ T cells. We also tracked K167 expression over the same times and found there was no change, indicating that ex vivo stimulation did not induce cell division. When observin phenotypic markers of Tregs like CD39, we also saw no changes m expression over the same timeframe (data not shown).

When we assessed the GARP/LAP+ Tregs after TCR-nonspecific polyclonal stimulation for 18 hours, we found that within the memory CD4 T cells (CD4+ CD45RA- ) GARP/LAP could be used in conjunction with multiple costimulatory molecules to enrich an extremely pure population of Foxp3+ regulatory T ceils. Using either CD137 (FIGURE 5A) or 0X40 (FIGURE 5B) in tandem with the latent TGFfl complex achieves a purity of close to 95% Foxp3+ CD25+ Tregs. GARP/LAP could also be faithfully used w th either CD25 or CD27 (FIGURES 5C and 5D) although these markers can be nonspecifically upregulated during stimulation. After devising this simple gating strategy to isolate and enumerate GARP/LAP+ Tregs after ex vivo stimulation, w-e compared control stimulations to antigen-specific stimulations in three patients. Using both a polyclonal, TCR-nonspecific stimulation, alon with an influenza vaccine we observed increased induction of GARP/LAP as compared to CD 137 in the polyclonal stimulation group, and decreased expression of GARP/LAP as compared to CD137 in the negative control (DMSO) group (FIGURE 6). Tills indicated that GARP/LAP in conjunction with CD 137 was more effective at capturing regulatory T cells, while also having decreased background. This claim was further solidified in the influenza-specific stimulation with hemagglutinin (HA1) peptide stimulation, where there seemed to be less background and variability when GARP/LAP w ^? as used with CD137. While the specificity of these influenza-specific Tregs was not yet confirmed, 3% of the CD4+ memory compartment seemed to be a more reasonable measure of HA1 -specific Tregs, as compared to the 15- 20% seen in the CD137+ CD25+ group. Our assumption was that there was a considerable amount of natural or bystander-induced activation among the CD137+ Tregs which was not eliminated by the ubiquitously high expression of CD25 on all regulatory T cell populations. The most interesting development was the discovery that while CD154- CD137+ CD25+ regulator).' T cells had a Foxp3 MFI significantly less than that seen in GARP/LAP+ CD137+ regulator _)' T cells, indicating increased activation and functionality as Tregs (FIGURE 16). Taken altogether with these data, the CD137+ CD154- compartment contains regulatory T cells with highly demethylated TSDR regions, but we contest that antigen-specificity is conferred by the expression of the latent TORb complex components GARP and LAP.

G ARP/LAP + regulatory T cells do not upregulate CD 154

Antigen-specific Tregs have been shown to be identifiable by a lack of CD154 expression, distinct from the effector subset, as well as expression of the costimulatory molecule CD137. CD137 expression is markedly slower in upregulation after antigen stimulation than CD154, a phenomenon which has not been fully explored. This is effect is most likely due to the decreased proliferation of Tregs ex vivo as compared to Teffs, requiring different costimulatory molecules to maintain interaction of the TCR with the pMHC. We sought to observe whether G ARP/LAP expression was limited to just regulator _} ' T cells, as its expression has been tied to Foxp3 activation. Using two antigen- specific peptide stimulation models, tetanus toxoid and major grass allergens, we were able to show that CD 154+ Teffs do not upregulate GARP/LAP after 18 hour stimulation (FIGURES 7 A and 7B). It was first observed that 6 hour stimulation with both peptide pools failed to upregulate GARP/LAP in either the CD 154+ Teffs or the CD137+ Tregs (FIGURE 7A). With 18 hours stimulation the upregulation of GARP/LAP was only observed in the CD137+ CD154- compartment, allowing for the identification of Foxp3+ regulatory T cells (FIGURE 7B). CD154- CD137- populations are shown m both antigen models to display background expression in non-activated memory T cells. Lack of GARP/LAP expression in the effector T cells allow' us to faithfully use it as a marker of antigen-specific regulatory T cells without the removal of effector T ceils that could potential] _)' be contaminating the population through shared molecule expression.

GARP/LAP r regulatory T cells are transcriptionally distinct from regulatory T cells lacking surface-expressed GARP/LAP As CD 137, 0X40, and other TNFRSF family members also serve as activation- induced costimulatory molecules on effector T cells, we focused our efforts on determining the role GARP/LAP expression may have on driving Treg development and homeostasis. TOEb has been well described as an essential cytokine for the induction and maintenance of regulatory T cell populations and is not a canonical costimulatory molecule like other TNFSRF family members. We again used influenza vaccination as a robust model to assess the role GARP/LAP plays among the CD137+ regulatory T cells by sorting both GARP/LAP- and GARP/LAP+ antigen-specific Tregs within the CD137+ CD154- compartment in patients who had recently undergone influenza vaccination. These cells transcriptional profiles were then assessed using bulk RNA sequencing and dimensionality ⁷ reduction was surveyed by principle component analysis (FIGURES 8A and 8B). Both the CD 154+ and CD 154- are clearly distinct from one another transcriptionally. The interesting thing to in terms of regulatory T cells was that the GDI 54- CD137+ GARP/LAP+ cells were just as starkly different from the GDI 54- CD137+ GARP/LAP- grouping as they were from the CD154+ grouping. Following m the vein of RNAseq, we decided to look at peanut allergic patients and compare GARP+ CD137+ populations to GARP- CD137+ populations as we had done for the patients receiving influenza vaccination (FIGURES 9A-9D). The GARP positivity within the CD 137+ CD 154- compartment clearly defined cells of a more regulatory ⁷ phenotype, with significant increases in the number of IE- 10, CTLA-4, and LAG-3 transcripts observed, ail essential molecules for regulatory function. We also observed a significant increase m the chemokine receptor associated with regulatory T cell homing, CCR8, when comparing GARP/LAP+ cells to GARP/LAP- cells (data not shown). Taken together, this data indicated that there was increased capacity for regulatory function in GARP/LAP+ Tregs as compared to CD137+ Tregs.

We also used single cell RNAseq to determine if the same results seen in bulk sequencing could be observed at the single cell level (FIGURES 17A-17C). Using a patient who was grass allergic during grass pollen season, we were able to effectively sort 47 Teffs based on CD154 (CD40LG) expression, as well as 48 Tregs based on GARP/LAP expression. The individual Teffs and Tregs were mostly completely distinct in their clustering by PCA (FIGURE 17A), aside from some overlap of a small number of Tregs that clustered with Teffs. Using index sorting, we were able to discern that these ceils had low MFI values for GARP/LAP expression, most likely leading to their incorrect identification as Tregs. The volcano plot of the most significant differences in genes upregulated in Teffs vs. Tregs clearly illustrates canonical regulatory T cell markers Foxp3, IL2RA (CD25), and LRRC32 (GARP), while the allergic Teffs show' a very typical phenotype of IL-4, IL-5, and IL-13 as well as the delineating marker CD40LG (CD154) (FIGURE 17B). Most interestingly, using TCR repertoire analysis of alpha and beta chains, we could dearly see that while both Teffs and Tregs have clonally expanded populations of CD4+ T cells, there was no clonal overlap between the two groups (FIGURE 17C). This indicates to use that these cells are either responding to completely different epitopes within the allergen, or that Tregs against the allergen are not induced peripherally from chronic stimulation of the effector pool. More single cell sequencing of multiple patients will better codify Teff and Treg specificities to various antigens.

GARP /LAP + upregulation denotes antigen-specificity among regulatory T cells

After confirming that GDI 37+ GARP/LAP+ Tregs have a distinct RNAseq profile from CD 137+ GARP/LAP- Tregs across multiple disease models, as well as increased capacity to suppress immune responses based on cytokine and coinhibitory molecules, we sought to confirm antigen-specificity m the regulatory T cells using HLA class II tetramers. While many studies in mice have used tetramer to identify regulatory T cells, in humans ex vivo tetramer staining seems to favor effector T cell identification over regulator}' T cells. This could be a result of a number of issues: increased Teff frequency in peripheral blood vs. tissue, differences in TCR density, differences in TCR- pMHC affinity, or decreased presence of co-inhibitory molecules that contact HLA class II. In order to circumvent this issue, we opted to sort Tregs into cell lines based on GARP/LAP expression and assess their tetramer binding after two weeks of in vitro expansion. Using a singular immunodominant epitope from the alder pollen antigen Ain g 1 we stimulated PBMCs from an HLA DBR1*1501 alder allergic patient and sorted CD 137+ Tregs for GARP/LAP expression. After in vitro expansion we stained both GDI 37+ GARP/LAP + and GDI 37+ GARP/LAP- lines with cognate tetramer to determine antigen specificity in the lines. We observed increased tetramer staining in the GARP/LAP+ lines (FIGURE 10A) as compared to GARP/LAP- lines (FIGURE 10B) as illustrated in the representative example, however many lines failed to stain with tetramer for unknown reasons. This could be due to the limitations listed above, or perhaps because the epitope-specificity differences between regulator ' and effector T cells have not yet been explored. When comparing tetramer MFi of alder allergic patients (n=3), GA1U ^} /LAP+ Treg lines had significantly higher staining (FIGURE 10C). The same result was demonstrated in influenza HA1 -specific Treg lines sorted for GARP/LAP expression (FIGURE 10D). While we attempted to use characterized, immunodominant epitopes as antigens, these epitopes are observed in effector responses, which may not overlap with regulatory responses, especially depending on various HLA alleles.

In order to alternately assess antigen-specificity in in vitro lines we used an ELISpot assay with mixed restimulations to observe Tregs secreting IL-10 in an antigen- specific manner. After stimulating in vitro GARP/LAP+ CD 137+ lines with matched feeders and relevant or irrelevant peptide pools to the original stimulating antigen a significant increase in the number of IL-lO-secreting Tregs can be seen in response to the tumor-associated antigen peptide pool (relevant) in comparison to the influenza- associated peptide pool (irrelevant). This difference is not observed when assaying for cells secreting IFNy, indicating that not only is the effect antigen-specific in the context of stimulus, but also specific to regulatory T cells secreting the suppressive cytokine IL- 10 (FIGURE 1 1 A). We also observed that the IL- 10-specific effect is seen only with relevant restimulation of type 1 diabetes (T1D) antigens m GARP/LAP+ CD137+ cell lines, and is not observed in the GARP/LAP- CD137+ cell fines from the same patient (FIGURE 1 1 B). Taken together, these two panels illustrate antigen-specificity observed only in the GARP/LAP+ Tregs, m an effect that is only observed through IL-10 secretion. These observations are in line with RNAseq data from above, as well as the tetramer data in allergic patients.

In our final in vitro assessment of GARP/LAP+ Treg lines, we show that antigen- specificity can also be exhibited through upregulation of relevant regulatory T cell costimulatory molecules after incubation with cognate peptide. In vitro cells lines were created and maintained in the same fashion as mentioned above, after which these lines were restimulated with either matched APCs as well as relevant antigen in the form of T1D peptides or irrelevant antigen in the form of HA1 influenza peptides (FIGURE 12) While the irrelevant and relevant peptide stimulations appear to do very little to restimulate the expression of CD137, OX-40, or CD69 in GARP/LAP- CD137+ regulatory' T cell lines, there is obvious restimulation of the costimulatory ligands in the relevant vs. irrelevant restimulation in the GARP/LAP+ CD137+ regulatory' T cell lines. This further solidifies the claim that antigen-specificity' within the CD137+ regulatory T cells resides solely in the GARP/LAP+ population, as only this population can re-express relevant costimulatory molecules to regulatory T cells in an antigen-specific fashion with relevant T1D antigen stimulation. As we had effectively convinced ourselves that the GARP/LAP+ population of Tregs were indeed antigen-specific to a significant degree more than the GARP/LAP- population, we moved on to in vivo models to determine whether the antigen-specificity of these cells could be observed directly in the periphery of humans undergoing antigen exposure.

GARP/LAP regulatory T cells are activated by natural exposure to antigen

To assess the activation of the antigen-specific Tregs in vivo , we opted to draw healthy patients before and after receiving the influenza vaccination in attempt to observe antigen-specific stimulation of regulator T cells. The flu vaccination model allowed us to more accurately assess the antigen load administered and the time of exposure, as well as providing a much stronger response for observation. We tracked the hemagglutinin- specific effector and regulator}' responses of human patients (n=10) before and after immunization with the vaccine (FIGURES 13A and 13B). We found that the ratio of antigen-specific effectors to antigen-specific Tregs increased significantly after immunization (p=0.0016), with marked expansion of antigen-specific CD4+ T cell effectors (Data not shown). When CD137 alone was used as an antigen-specific marker for activation and was compared to CD154 expression in the effector population, there was no change in the ratio of effectors to Tregs before and after vaccination and the number of Tregs outweighed by tenfold. To determine if the CD 137+ or the CD 137+ GARP/LAP+ were tracking antigen-specificity more effectively, we looked at post vaccination Ki67 expression within both populations. We avoid an in vitro approach by using Ki67 as a marker for ceils undergoing division in response to natural activation. The Ki67 percent expression correlated with the GARP percent expression within the CD137 compartment very well, yielding an R = 0.88 and a P = 0.006. When the same Ki67 expression was correlated to CD137 percent expression m post vaccine patients, the correlation was nonexistent (R ^{2 :::} 0.17, P = 0.31 ). We also assessed control responses in the same patients using a mock DMSO stimulation and ultimately saw no correlation between either the CD 137+ or GARP/LAP + groups (FIGURE 18). This led us to believe that the antigen specificity resided in the CD154- CD137+ GARP/LAP+ Tregs, and that perhaps CD137 alone was insufficient alone to identify antigen-specificity. We were also able to observe that antigen-specificity resides within the population of activation-induced GARP/LAP+ regulatory T cells in a seasonal allergy model. By using alder pollen allergic patients and a natural seasonal exposure to alder pollen, we were able to assess antigen-specific T cell responses to alder antigen before and after the peak of the season. We drew allergic patients in the winter preceding the season, and then tracked the alder pollen counts until the peak season was observed (FIGURE 14A). We then redrew the same patients two weeks following the peak alder season (FIGURE 14B and 14C). When assessing the Foxp3+ Ki67+ populations within the CD137+ compartment, there was no increase of proliferating Tregs observed in allergic patients (n=6) after seasonal alder pollen exposure. On the contrar', when assessing the GARP/LAP+ population within CD137 Tregs, there was a significant increase in the Foxp3+ Ki67+ cells after seasonal pollen exposure. This indicated to us that the GARP/LAP-t- regulatory T cell population contains antigen-specific regulatory T cells that respond and proliferate in matched allerg}' patients who are naturally exposed to alder pollen. Taken altogether with the in vitro data generated from the GARP/LAP+ T cell lines and the transcriptional profiling of GARP/LAP+ Tregs vs. GARP/LAP- Tregs, we contest that while CD137+ regulatory T cells stably express Foxp3 and have a highly demethylated TSDR, only activation-induced expression of the latent TGFp complex components GARP and LAP denotes hue antigen-specificity amongst Tregs.

Discussion

The direct detection of live, antigen-specific regulatory T cells in humans would confer immediate benefits to the medical and research communities, as both a monitoring technique and a therapeutic treatment. The discovery of Foxp3 realized a long held believe that "suppressor cells" existed within the human body as a part of the immune system with the role of quelling overzealous or unnecessary immune responses. The newly coined "regulatory T cells" could be identified by increased expression of the high affinity IL-2 receptor alpha (CD25) and lack of expression of the IL-7 receptor (CD127). Researchers have since used these surface expressed markers as a live cell method of sorting regulatory T ceils enriched for Foxp3 expression. While these markers serve well as indicators of Foxp3+ Tregs, they make no distinction between the entirety of Tregs observed in a patient and just those specific for an antigen of interest. Foxp3 is similar in that it detects all regulatory T cells within a patient, but not those specific to certain antigens. It also requires fixation and nuclear staining, dashing all chances of further analysis of the detected Tregs By identifying the expression of activation-induced molecules that correspond only to antigen-specific regulator} T cells, we are able to effectively isolate only Tregs specific for an antigen of interest. We demonstrate the universality of the approach by detecting and characterizing these cells across a broad cohort of patients, including allergic, autoimmune, cancer, infectious disease, and healthy patients.

The assay requires no intracellular staining or Golgi -blocking treatments, as earlier versions of CD 154 assays required. There is no artificial stimulation required in the form of anti-CD28, as seen m recent CD137 assays. There is no need to remove effector T cells prior to regulatory T cell detection, as GARP/LAP is not expressed on effector T ceils, unlike CD137. Cells can be sorted or sequenced following detection because there is no necessity to stain for intracellular antigens. Tire single, positive- enrichment step allows for a simple assay that imposes no undue stress upon the cells by subjecting them to multiple enrichment steps. GARP/LAP serves as an indicator of antigen-specificity in Tregs and denotes stable Foxp3 expression, while CD137 and other TNFSRF costimulatory molecules like 0X40 are upregulated by highly TSDR demethyiated Tregs after activation in preparation for interaction with APCs, presumably to propagate the regulatory response as well as dampen signals to APCs in order to suppress reactivity in the environment. Deregulation of certain costimulatory molecules during antigen stimulation can be affected by bystander effects due to cytokine signaling (IL-2, IL-7, !L-33) and other growth factors present in the stimulation milieu. Members of the tumor necrosis factor receptor family (TNFR) of costimulatory molecules such as CD 137 and 0X40 have been shown to have direct binding of STATS that is IL-2 driven, indicating that there is a possibility that CD 137 upregulation is partially dependent on IL- 2 signaling.

While the method outlined here has been largely used for immune-monitoring purposes in patients undergoing therapy, we envision it as a multifaceted technique that will also be used itself to create therapies for many common diseases. By sorting and expanding antigen-specific regulatory T cells, patients with autoimmune disease could potentially be treated in a manner that would either increase the frequency of regulatory T cells required to quell self-reactive responses, or perhaps provide a gene therapy target for dysregulated regulatory T cells to allow them to properly function once again in autoimmune patients. Conversely, as a cancer therapeutic target, these antigen-specific Tregs could be edited to remove an anergy and restore their ability to target mutated cells by turning off some of their suppressive functions and implementing effector ones. Ultimately, the advantageousness of the technique derives from the ability to target antigen-specific, rather than bulk populations. This decreases the possibility for nonspecific, off-target effects as seen in checkpoint blockade therapies, while increasing the potency of the response, ultimately requiring less cells due to the increased specificity. In vriro-induced Tregs (iTregs) possibly fail to correct autoimmune reactions, not because they lack the density to occupy the compartment, but most likely because they fail to properly engage with presented autoantigens and induce contact-mediated suppression, as well as become properly activated via TCR to cany out suppressive tasks. There are also risks with iTregs converting back into more pathogenic phenotypes due to decreased Foxp3 stability. By allowing for the detection of antigen-specific regulatory T cells in human patients that are naturally occurring rather than induced in vitro , we allow for the advancement of their understanding and role in human disease as well as a potential therapeutic target to treat numerous autoimmune diseases and cancers.

Methods

Cell stimulation

Blood draws from patients were arranged by the Benaroya Research Institute clinical core in conjunction with the Virginia Mason Hospital and Medical Center. Blood arrived within 24 hours of draw and w¾s stored in the dark at room temperature. PBMCs were then isolated from whole blood using ficoll separation. The ficoll preparation was centrifuged at 1800 RPM for 20 minutes and the bully coat was gently removed using a transfer pipet. Red blood cells were lysed using an in-house hemolytic buffer after 10 minutes of incubation at room temperature. Cells were then washed and counted. PBMCs were first incubated with anti CD40 at a concentration of 1 uL per 10 ⁶ cells. After a 15 minute incubation, cells were plated at a density of 10 x 1G ⁶ PBMCs per mL and with peptide at a final concentration of 10 ug/mL. After 18 hours of incubation at 37 degrees cells were harvested and washed.

Staining of CD 154/CD 137/GARP/LAP, cell-surface, and intracellular antigens

Harvested cells w¾re stained for CD 154 (PE-CF594) and enriched using anti-PE magnetic beads and Miltenyi MS columns. After the first round of enrichment, the enriched effectors were stained for surface markers. The flow through from the CD154 enrichment was then stained for GARP (PE) and LAP (PE) and magnetically enriched using the same approach. Cells were then stained for surface markers and fixed and stained for transcription factors. For nuclear staining, the eBioscienee permeabilization/fixation kit was used.

Flow cytometry and data analysis

Flow cytometry experiments consisted of twelve-color to sixteen-color panels. We acquired data on an LSRFortessa flow cytometer for analysis (BD Biosciences), and analyzed it in FlowJo version 10.0.7 (Tree Star). We performed cell-sorting experiments on a FACSAria Fusion cytometer (BD Biosciences). We collected at least 150,000 events for antigen-specific assays.

RNA-seq library generation sequencing, and alignment

For bulk library preparation, samples of 100 CD154+, CD137+, or GARP/LAP+ CD4+ memory T-cells were sorted. 10 samples from 10 unique donors were prepared (both pre and post influenza vaccination). Samples were sorted into SMARTer v4 lysis reagents (Clontech). Cells were lysed and cDNA was synthesized. After amplification, sequencing libraries w¾re prepared using the Nextera XT DNA Library Preparation Kit (Illumina).

Single-read sequencing of libraries was performed on either on a HiScanSQ with 100-base pair long reads (5 Cl preparations) or a HiSeq2500 sequencer (Illumina, 12 Cl preparations and all bulk RNA-seq data with 58-base pair long reads, using TruSeq v3 Cluster and SBS kits (Illumina) with a target depth of >2.5M reads. Base-calling was performed automatically in Illumina BaseSpace after sequencing; FASTQ reads were trimmed in a local Galaxy server in two steps: 1) hard-trimming to remove 1 3'-end base (FASTQ Trimmer tool, v.1.0.0); 2) quality trimming from both ends until minimum base quality for each read > 30 (FASTQ Quality Trimmer tool, v.1.0.0) (Langmead, B., et al., Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 10, R25 (2009); Giardme, B. et al. Galaxy: A platform for interactive large- scale genome analysis. Genome Res. 15, 1451-1455 (2005)). Reads were aligned to the UCSC Human genome assembly version 19 in Galaxy using bowtie and TopHat (Tophat for Illumina tool, v.1.5.) (Trapnell, C., et al, TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1 105-1111 (2009).). Read counts per Ensemhl gene ID were estimated in Galaxy using htseq-count (htseq-count tool, n.0.4.1 ) (Anders, S., et al., HTSeq-a Python framework to work with high-throughput sequencing data. J Bioinformatics 31, 166- 169 (2015)). To determine the endogenous TCR sequences of single cells from RNA-seq read, we used a previously described approach (Cerosaletti, K. et al, Single-Cell RNA Sequencing Reveals Expanded Clones of Islet Antigen-Reactive CD4(+) T Cells in Peripheral Blood of Subjects with Type 1 Diabetes. J Immunol. 199, 323-335 (2017)) Bnefly, we utilized methods for genome-independent (de novo) assembly to construct a set of overlapping DNA segments (or contigs) (PMID: 21572440). MiXCR analysis of these contigs was used to identify productive TCR rearrangements.

Creation of in vitro Treg lines

We plated GARP/LAP+/- cell fractions purified by cell sorting into 96-well plates, with each well receiving 5-10 cells in 100 pL RPMT media plus 10% human serum. 10,000 irradiated feeders per well were then added with IL-2 and PHA at a final well volume of 200 uL. Cells were cultured for 14 days at 37°C and 5% CO _? .. 100 uL of media per well was replaced at 7 days. Proliferating wells were counted and assessed for viability before being used in assays.

Cytokine analysts by ELISpot

We cultured cell lines at 37 ^C C for up to 24 b with a ratio of 10: 1 cells to irradiated, matched feeder cells with peptide. After 24 h of co-culture, we plated 10,000 cells per well in the capture antibody coated and blocked 96 well plate. Capture antibodies include TL-10, ITNg, and TI.,-4 from ELISpot kits (U-CyTecb Biosciences). Stimulated cells were incubated at 37 °C for another 24 h and then washed and developed.

EXAMPLE 3

This Example describes application of the disclosed ASTRA approach to profile and characterize T cells that are specific for tumor or auto-antigens in an attempt to better study T cell signatures in a relevant context, i.e., during cancer treatment.

Rationale

A relevant application of the ASTRA method disclosed herein is to elucidate the relationship between tumor- and auto-antigen-specific T cells in response to immune checkpoint inhibitor (ICI) therapy. Although immune-related adverse events (irAE) have been associated with favorable clinical responses to ICI, the relationship between these events is imperfect. Measuring antigen-specific T cell responses, which are proximal pharmacodynamics markers of immune responses, will likely clarify the relationship between anti-tumor and autoimmune responses, and will enable better management of these events clinically during ICI.

The ASTRA assay disclosed herein is applied to identify, enumerate and characterize tumor- and auto- antigen-specific T cells m parallel. In a subset of subjects, these cells are isolated using the same approach and then perform low input bulk RNA- seq to identify antigen -sped lie T cell signatures and determine how they change in response to ICI therapy.

Cohort selection & Clinical Outcome Measures

For this Example, samples are obtained using our "Immune Checkpoint Inhibitor (ICI) Cancer Research Protocol" (ICICR). The samples tested are from visits before and 1-3 months after initiation of CPI therapy. Testing both pre-and post-treatment visits enables better understanding of the relationship of antigen-specific T cell signatures to therapy. Ideally, 30 Million PBMCs are collected from each subject and each time point to run all experiments. Frozen PBMC are used for all assays, allowing performance of assays for each subject's samples on the same day. It is proposed to study patients with and without irAE in order to determine if the responses m individuals with irAE differ from those without, or by type of irAE. Prior therapy of cancer patients also has the potential to confound investigations, therefore this variable is considered when selecting subjects and during analyses.

Experimental Design

1 - Characterization of T cells specific for tumor-antigens

The ASTRA assay as described herein is used to monitor the ex vivo frequency, surface markers and phenotypes of tumor-specific T cells in samples from patients prior to and after ICI therapy. A major advantage of this assay is that multiple possible peptide epitopes can be pooled, and then analyzed in side-by-side assays. For example, we have successfully experimented with peptide pools comprising up to -1,000 different peptides. Although responses to both conventional tumor associated antigens (TAA) and neoantigens (i.e., created by tumor-specific mutations) have been linked to ICI therapy, this study focuses on TAA, as they are not individualized and are therefore more broadly applicable. We have developed a pool of overlapping peptides derived from TAA, MAGE-A1, MAGE- A3, MAGE-A4, NY-ESOl, Survivm, MUC-1, CEA, Tyrosinase, GplOO and MART-1. Most of these molecules are differentiation or germline antigens shared by multiple tumor types. After isolation, enriched cells are characterized for a variety of markers, for example, using markers of maturation/differentiation (CD4, CD27, CD45RB, CD45RA), exhaustion (TIM-3, LAG-3, PD-1, CD49b), activation (CD38, CD69, ICOS), and horning (CCR4, CXCR3, CCR6). Co-enriched antigen-specific CD4+ Treg cells and CD8+ Teff cells can be specifically characterized using markers of maturation/differentiation (CD4, CD27, CD45RA), exhaustion (TIM-3, LAG-3, PD-1, CD49b, KLRGI , CD 160) and activation (CD38, CD25, CD69, OX-40). Anti-CD 14, CD 19 and Via-Probe reagent (BD Biosciences) can be used as a dump gate. The frequency of antigen specific T cells will be determined by the formula F = n/N, where n designates the number of cells in the bound fraction after enrichment and N is the total number of CD4+ or CD8+ T cells (calculated as 100 ^c the number of CD4+ or CD8+ T cells in 1/100th unenriched fraction that was saved for analysis). This multi-parameter immune profiling can be carried out with a FACS Fusion flow cytometer, such that cells of interest can be simultaneously sorted for subsequent use in transcript profiling.

As an example, the assay was performed with tumor antigens and T cells from a healthy control individual and a cancer patient before and after ICI therapy using CD137 and GARP as markers for antigen-specific activation of Treg cells. See FIGURE 19A. Of note, is the massive tumor-antigen specific CD8+ T cell response after ICI therapy. The antigen-specific results were reminiscent of the global increase in CD8+ T cells observed by others following ICI therapy.

2 - Characterization of T cells specific for self-antigens

A similar approach is used from the same PBMC samples to characterize auto antigen responses. Altogether, these combined experiments allow integration of all data and generate a comprehensive map of the immune landscape in untreated and ICI treated cancer patients. One focus is on peptides spanning antigens present in organs or tissues that have been implicated ICI irAE. Specifically, islet antigens (ICI-mduced T1D and pancreatitis), thyroid antigens (ICI-induced thyroiditis), synovium antigens (ICI-mduced RA and polyarthritis) and skin antigens (ICI-induced vitiligo and pemphigus) are contemplated in this screen. The self-antigen pool therefore includes overlapping peptides derived from insulin, IGRJP, GAD65, thyroxin, thyroglobulin, vimentin, alpha enolase, desmoglein 3, and tyrosinase. An example of the assay with self- antigens and T cells from a healthy control individual and a cancer patient before and after ICI therapy is shown in FIGURE 19B. Strikingly, there was a massive increase in CD8+ Teff cells reactive with self-antigens (islet antigens only, in this case) m cells in the cancer patient after ICI-therapy, in parallel with the tumor-antigen specific CD8+ T ceils. Together, these results suggest that this patient had both anti-tumor and anti-islet CDS T cell responses triggered by ICI therapy. It is noteworthy that autoimmune diabetes can be triggered by anti -PD- 1 ICI therapy.

The preliminary results disclosed herein demonstrate the utility of the disclosed ASTRA approach to assaying activated, antigen-specific Treg cells in a wide variety of contexts. Specifically, the results show that antigen-specific Treg cells can be detected and monitored in the context of a cancer treatment to determine the effects on immune- related adverse events connected with the cancer treatment.

EXEMPLARY EMBODIMENTS

Exemplary, non-limiting embodiments are described below:

Embodiment 1. A method of detecting an activated regulator' T (Treg) cell, comprising:

obtaining a sample comprising regulatory T (Treg) cells that have been exposed to an antigen and/or activator in a manner such that the sample may contain activated Treg ceils

contacting the sample with a first detection molecule that specifically binds to a latent TGFfi complex protein and a second detection molecule that specifically binds to a co-stimulatory marker; and

detecting a ceil in the sample that is bound by the first detection molecule and the second detection molecule, thereby detecting an activated Treg cell.

Embodiment 2. The method of embodiment 1, wherein the Treg cells have been exposed to antigen presenting cells that display a peptide antigen in a peptide/MHC complex.

Embodiment 3. The method of embodiment 1, wherein the Treg cells have been exposed to a soluble peptide-MHC monomer or mul timer complex.

Embodiment 4. The method of embodiment 1 , 2, or 3, wherein the activator is a composition comprising one or more costimulatory molecules that bind to and activate one or more co-stimulatory receptors selected from CD137, 0X40 (CD 134), CD27, CD69, ICOS (CD278), CDS, CD49d, CD2, and CD25. Embodiment 5. The method of one of embodiments 1-4, wherein the latent TOEb complex protein is GARP (LRRC32), LAP (latent-associated peptide), TGF[> I -4. or a complex of any thereof.

Embodiment 6. The method of one of embodiments 1-5, wherein the eo-stimulatory marker is selected from CD 137, 0X40 (CD 134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and CD25.

Embodiment ?. The method of one of embodiments 1-6, wherein the method further comprises contacting the sample with a third detection molecule that specifically binds to CD4, and wherein detecting an activated T regulatory cell comprises detecting a cell in the sample that is bound by the first detection molecule, the second detection molecule, and the third detection molecule.

Embodiment s. The method of one of embodiments 1-7, wherein one or more of the first detection molecule, the second detection molecule, and the third detection molecule is an antibody, antibody-like molecule, aptamer, or a functional antigen-binding fragment or domain thereof.

Embodiment 9. The method of embodiment 8, wherein the antibody-like molecule is a single-chain antibody, a bispecific antibody, a Fab fragment, or a F(ah) ₂ fragment.

Embodiment 10. The method of embodiment 9, wherein the single-chain antibody is a single chain variable fragment (scFv), single-chain Fab fragment (scFab), VJJH fragment, V _N AR, or nanobody.

Embodiment 11. The method of one of embodiments 1 -10, wherein each of the first detection molecule and the second detection molecule is detectably labeled with mutually distinguishable labels.

Embodiment 12. The method of one of embodiments 1-11, wherein each of the first detection molecule, the second detection molecule, and the third detection molecule is detectably labeled with mutually distinguishable labels.

Embodiment 13. The method of one of embodiments 1-12, wherein detecting binding of the first detection molecule to the cell, binding of the second detection molecule to the cell, and/or binding of the third detection molecule to the cell comprises use of fluorescence-activated cell sorting (FACS) or mass cytometr ' (CyTOF).

Embodiment 14. The method of one of embodiments 1-13, further comprising enriching for the activated Treg cell. Embodiment 15. The method of one of embodiments 1 -14, further comprising isolating the activated Treg cell.

Embodiment 16. The method of one of embodiments 1 -15, wherein the sample is a biological sample from a subject, such as blood, tissue, lavage, tumor, or is derived therefrom.

Embodiment 17. The method of one of embodiments 1 -16, further comprising an initial step of obtaining the sample of Treg cells from a subject.

Embodiment 18. The method of one of embodiments 1-17, wherein obtaining a sample comprises a step of contacting an initial sample comprising Treg cells with sufficient antigen or activator in a manner and amount such that at least a portion of the Treg cells are activated.

Embodiment 19. The method of embodiment 18, comprising contacting the cells with the activator.

Embodiment 20. The method one of embodiments 18-19, wherein the activator is a composition comprising one or more costimulatory molecules that bind to and activate one or more co-stimulatory receptors selected CD137, 0X40 (CD 134), CD27, CD69, !CQS (CD278), CDS, CD49d, CD2, and CD25.

Embodiment 21. A method of producing an enriched population of activated T regulatory' (Treg) cells, comprising:

incubating a sample comprising regulatory T (Treg) cells with an antigen of interest and/or an activator in a manner and amount such that at least a portion of the Treg ceils are activated;

contacting the sample with a first enrichment molecule that specifically binds to a latent TίϊRb complex protein and a second enrichment molecule that specifically binds to a co-stimulatory marker; and

enriching for cells that are bound to the first enrichment molecule and the second enrichment molecule, thereby producing an enriched population of activated Treg cells.

Embodiment 22. The method of embodiment 21 , wherein the antigen of interest is a peptide that is complexed with MHC on an antigen presenting cell.

Embodiment 23. The method of embodiment 21, wherein the antigen of interest is a peptide complexed with a MHC monomer or multimer to form a soluble complex. Embodiment 24. The method of one of embodiments 21-23, wherein the activator is a composition comprising one or more costimulatory molecules that bind to and activate one or more co-stimulatory receptors selected from CD137, 0X40 (CD 134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and (1)25

Embodiment 25. The method of one of embodiments 21-24, wherein the latent TORb complex protein is GARP (LRRC32), LAP (latent-associated peptide), TGFf)I-4, or a complex of any thereof.

Embodiment 26. The method of one of embodiments 21-25, wherein the costimulatory marker is selected from CD137, 0X40 (CD134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and CD25.

Embodiment 27. The method of one of embodiments 21-26, wherein the method further comprises contacting the sample with a third enrichment molecule that specifically binds to CD4, and enriching for cells that are bound by the first enrichment molecule, the second enrichment molecule, and the third enrichment molecule.

Embodiment 28. The method of one of embodiments 21-27, wherein one or more of the first enrichment molecule, the second enrichment molecule, and the third enrichment molecule is an antibody, antibody-like molecule, aptamer, or a functional antigen-binding fragment or domain thereof.

Embodiment 29. The method of embodiment 28, wherein the antibody -like molecule is a single-chain antibody, a bispecific antibody, a Fab fragment, or a Fiabjy fragment.

Embodiment 30. The method of embodiment 29, wherein the single-chain antibody is a single chain variable fragment (scFv), single-chain Fab fragment (scFab), V _H H fragment, V^AR, or nanobody.

Embodiment 31. The method of one of embodiments 21 -30, wherein the first enrichment molecule and the second enrichment molecule are each detectably labeled with mutually distinguishable labels.

Embodiment 32. The method of one of embodiments 27-31 , wherein the first enrichment molecule, the second enrichment molecule, and the third enrichment molecule are each detectably labeled with mutually distinguishable labels.

Embodiment 33. The method of one of embodiments 21 -32, wherein the enriching step comprises detecting binding of the first enrichment molecule to the cell and binding of the second enrichment molecule to the cell, or detecting binding of the first enrichment molecule to the cell, binding of the second enrichment molecule to the cell, and binding of the third enrichment molecule to the cell, using fluorescence- activated cell sorting (FACS).

Embodiment 34. The method of one of embodiments 21 -33, further comprising an initial step of obtaining the sample of Treg cells from a subject.

Embodiment 35. The method of one of embodiments 21-34, wherein the sample is a biological sample from a subject, such as blood, tissue, lavage, tumor, or is derived therefrom.

Embodiment 36. The method of one of embodiments 21-35, comprising incubating the sample with the activator.

Embodiment 37. The method of embodiment 36, wherein the activator is a composition comprising one or more costimulatory molecules that bind to and activate one or more co-stimulatory receptors selected from CD137, 0X40 (CD134), CD27, ICOS (CD278), CD3, CD49d, CD2, and CD25.

Embodiment 38. The method of one of embodiments 21-37, further comprising administering one or more activated Treg cells of the enriched population to a subject in need thereof.

Embodiment 39. The method of embodiment 38, wherein the subject is the same subject from whom the sample was obtained or derived, or is HLA matched to the subject from whom the sample was obtained or derived.

Embodiment 40. A cell m the enriched population produced from the method of any one of embodiments 21-37.

Embodiment 41. A method of treating a condition treatable by the presence of an activated Treg, comprising administering the cell of embodiment 40 to a subject in need thereof.

Embodiment 42. The method of embodiment 41, wherein the condition is an autoimmune disease or an allergy.

Embodiment 43. A method for monitoring a T regulator' (Treg) cell response to potential exposure of a subject to an antigen, comprising:

contacting a sample with a first detection molecule that specifically binds to a latent TGFfi complex protein and a second detection molecule that specifically binds to a co-stimulatory marker, wherein the sample comprises T cells and was obtained from a subject suspected of being exposed to an antigen; and detecting or quantifying cells in the sample that are hound by both the first detection molecule and the second detection molecule, wherein the absence, presence, or relative abundance of cells bound by both the first detection molecule and the second detection molecule is indicative of a state of the Treg cell response.

Embodiment 44. The method of embodiment 43, wherein the latent TORb complex protein is GARP (LRRC32), LAP (latent-associated peptide), TGFpl-4, or a complex of any thereof.

Embodiment 45. The method of embodiment 43 or 44, wherein the costimulatory marker is selected from CD137, 0X40 (CD134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and CD25.

Embodiment 46. A method for monitoring the sensitivity of a subject to an antigen of interest, comprising:

contacting a sample comprising T regulatory (Treg) cells with a first detection molecule that specifically binds to a latent TOTb complex protein and a second detection molecule that specifically binds to a co-stimulatory marker, wherein the Treg cells have been exposed to an antigen of interest and the sample was obtained from the subject; detecting or quantifying the cells in the sample that are bound by both the first detection molecule and the second detection molecule,

comparing the number of cells in the sample that are bound by both the first detection molecule and the second detection molecule to an established threshold to determine the sensitivity of the subject to the antigen of interest.

Embodiment 47. The method of embodiment 46, wherein the latent TOEb complex protein is GARP (LRRC32), LAP (latent-associated peptide), TORb I -4. or a complex of any thereof

Embodiment 48. The method of embodiment 46 or 47, wherein the costimulatory marker is selected from CD137, 0X40 (CD134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and CD25.

Embodiment 49. The method of one of embodiments 46-48, wherein a relative increase in the number of cells m the sample that are bound by both the first detection molecule and the second detection molecules compared to the established threshold indicates a low sensitivity of the subject to the antigen of interest.

Embodiment 50. The method of one of embodiments 46-49, wherein the threshold is established at an earlier time point from the subject using the same method, and wherein a relative increase in cells that are bound by both the first detection molecule and the second detection molecule over the established threshold indicates a decreasing sensitivity to the antigen of interest, and wherein a relative decrease in cells that are bound by both the first detection molecule and the second detection molecule over the established threshold indicates an increasing sensitivity to the antigen of interest.

Embodiment S l. The method of embodiment 50, wherein the threshold is established during or prior to administration of the therapeutic treatment and the sensitivity of the subject to the antigen of interest at a later time point is indicative of the efficacy of the therapeutic treatment.

Embodiment 52. A method of screening for T regulator) _' (Treg) cell stimulatory epitopes from an allergen or antigen of interest, comprising:

obtaining a sample comprising Treg ceils;

exposing the Treg cells to an epitope derived from an antigen of interest in an MHC context;

contacting the Treg cells with a first detection molecule that specifically binds to a latent TGF[3 complex protein and a second detection molecule that specifically binds to a eo-stimulatory marker; and

quantifying the relative abundance of cells in the sample that are bound by both the first detection molecule and the second detection molecule, wherein a high relative abundance indicates that the epitope stimulates development of activated Treg cells.

Embodiment 53. The method of embodiment 52, wherein exposing the Treg cells to an epitope comprises exposing the Treg cells to antigen presenting cells that display the epitope m a peptide/MHC complex.

Embodiment 54. The method of embodiment 52, wherein exposing the Treg cells to an epitope comprises exposing the Treg cells to soluble peptide-MHC monomer or multimer complexes that comprise the epitope loaded onto the MHC monomer or multimer in a complex.

Embodiment 55. The method of one of embodiments 52-54, further comprising exposing the Treg ceils to a composition comprising one or more costimulatory molecules that bind to and activate one or more co-stimulatory receptors selected from CD137, 0X40 (CD134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and CD25. Embodiment 56. The method of one of embodiments 52-55, wherein the latent TGFB complex protein is GARP (LRRC32), LAP (latent-associated peptide), TGFjll -4, or a complex of any thereof.

Embodiment 57. The method of one of embodiments 52-56, wherein the costimulatory marker is selected from CD137, 0X40 (CD134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and CD25.

Embodiment 58. The method of one of embodiments 52-57, further comprising repeating the method for one or more different epitopes derived from the same antigen of interest to determine the epitopes that stimulate the relatively increased levels of activated Treg cells.

Embodiment 59. A method of identifying the MHC Class 11 molecule that binds to a Treg stimulatory epitope, comprising:

obtaining a sample comprising Treg cells;

exposing the Treg cells to an epitope derived from an antigen of interest in an MHC context;

contacting the Treg cells with a first detection molecule that specifically binds to a latent TORb complex protein and a second detection molecule that specifically binds to a co-stimulatory marker;

quantifying the relative abundance of cells in the sample that are bound by both the first detection molecule and the second detection molecule, wherein a high relative abundance indicates that the epitope stimulates development of activated Treg cells; and characterizing the MHC molecule complexed with the epitope that resulted in a high abundance of cells bound by both the first detection molecule and the second detection molecule.

Embodiment 60. The method of embodiment 59, wherein exposing the Treg cells to an epitope comprises exposing the Treg cells to an antigen presenting cell that displays the epitope in a peptide/MHC complex.

Embodiment 61. The method of embodiment 60 or 61, wherein the method further comprises isolating the antigen presenting cell that produced a high abundance of ceils bound by both the first detection molecule and the second detection molecule; and

Embodiment 62. The method of one of embodiments 59-61, wherein exposing the Treg cells to an epitope comprises exposing the Treg cells to soluble peptide-MHC monomer or mul timer complexes that comprise the epitope loaded onto the MHC monomer or multimer in a complex.

Embodiment 63. The method of one of embodiments 59-62, further comprising exposing the Treg cells to a composition comprising one or more costimulatory molecules that bind to and activate one or more eo-stimulatory receptors selected from CD137, 0X40 (CD134), CD27, CD69, ICOS (CD278), CDS, CD49d, CD2, and CD25

Embodiment 64. The method of one of embodiments 59-63, wherein the latent T(3Rb complex protein is GARP (LRRC32), LAP (latent-associated peptide), TGFfd -4, or a complex of any thereof.

Embodiment 65. The method of one of embodiments 59-64, wherein the costimulatory molecule is selected from CD137, 0X40 (CD134), CD27, CD69, ICOS (CD278), CD3, CD49d, CD2, and CD25.

Embodiment 66. A method of testing the effect of a putative therapeutic compound on an activated T regulatory (Treg) cell, comprising exposing the cell of embodiment 40 to the therapeutic compound.

While illustrative embodiments have been illustrated and described, it will be appreciated that various changes can be made therein without departing from the spirit and scope of the invention.