Rheumatoid arthritis occurs most often in people who express HLA-DR molecules containing a five aa “shared epitope” in the β chain. These MHCII molecules preferentially bind citrullinated peptides formed by posttranslational modification of arginine. Citrullinated peptide:HLA-DR complexes may act as arthritis-initiating neo-antigens for CD4+ T cells. Here, we used fluorophore-conjugated HLA-DR tetramers containing citrullinated peptides from human cartilage intermediate layer protein, fibrinogen, vimentin, or enolase 1 to track cognate CD4+ T cells. Immunization of HLA-DR transgenic mice with citrullinated peptides from vimentin or enolase 1 failed to cause any expansion of tetramer-binding cells, whereas immunization with citrullinated peptides from cartilage intermediate layer protein or fibrinogen elicited some expansion. The expanded tetramer-binding populations, however, had lower T helper 1 and higher regulatory T cell frequencies than populations elicited by viral peptides. These results indicate that HLA-DR–bound citrullinated peptides are not neo-antigens and induce varying degrees of immune tolerance that could pose a barrier to rheumatoid arthritis.

Rheumatoid arthritis (RA) is a CD4+ T cell–mediated disease that is characterized by the presence of antibodies to proteins that include the aa citrulline (Valesini et al., 2015). Citrulline is a non-charged, non-genetically-encoded aa that is formed when protein arginine deiminases (PAD) deiminate the positive charge on arginine in proteins (Mondal and Thompson, 2019). The greatest risk factor for the development of RA is the expression of certain allelic forms of HLA-DR MHC class II (MHCII) molecules (Scherer et al., 2020). MHCII molecules are membrane-anchored heterodimers of α and β chains expressed on B cells and myeloid cells that bind peptides, usually from extracellular proteins (Germain, 1994). The top of the α and β chain heterodimer forms a binding groove that accommodates a nine aa core sequence within peptides that are 9–25 aa long (Rudolph et al., 2006). The aa at positions 1, 4, 6, and 9 of the nine-mer core are embedded in pockets in the MHCII binding groove while residues 2, 3, 5, 7, and 8 point up out of the groove (Marrack et al., 2008). The up-pointing residues of the peptide, along with residues from the sides of the MHCII groove, can be bound by a TCR on a CD4+ T cell (Marrack et al., 2008; Rudolph et al., 2006). TCR engagement causes transduction of signals that stimulate the CD4+ T cell to proliferate, differentiate, and produce cytokines that enhance the capacity of other cells of the immune system to kill microbes and tumors (Tubo and Jenkins, 2014). TCRs are generated through a process of gene segment recombination such that each nascent CD4+ T cell expresses a unique TCR (Davis et al., 1998). CD4+ T cells that by chance produce a TCR specific for an MHCII-bound peptide of host origin are subject to immune tolerance mechanisms including intrathymic deletion, functional inactivation, or suppression by regulatory T cells (Tregs; Xing and Hogquist, 2012). This process, however, can fail in some individuals, leading to autoimmunity.

Autoimmune diseases including RA are linked to certain of the many allelic forms of HLA-DR β chains (Olerup et al., 1991). RA is associated with variants that share an aa sequence (the so-called shared epitope; Malmström et al., 2017) that forms a positively charged P4 pocket that repels positively arginine but accommodates citrulline (Ting et al., 2018). A prominent model is that inflammation induces expression of PAD enzymes, which convert arginine to citrulline in many proteins (Darrah and Andrade, 2018). Some of the peptides in the modified proteins contain citrulline at P4 in a nine-mer core with P1, P6, and P9 residues that are permissive for binding to shared epitope HLA-DR molecules (Lim et al., 2021). The citrullinated peptide:HLA-DR complexes are then thought to activate cognate CD4+ T cells that are important in the pathogenesis of RA (Alivernini et al., 2022).

The fact that citrullinated peptide:HLA-DR–specific T cells can be identified in people with RA (Gerstner et al., 2020; James et al., 2014) indicates that not all such T cells are deleted during T cell development. Indeed, it could be that these T cells are immunologically naïve before RA onset because citrullinated peptide:HLA-DR complexes are not displayed in the thymus under normal conditions and are thus essentially foreign antigens. In this scenario, citrullinated peptide:HLA-DR–specific T cells are expected to be present in people who do not have RA, have a naïve phenotype, and be highly responsive to initial exposure to citrullinated peptide:HLA-DR complexes in the same way that naïve T cells respond to foreign antigens from microbes. It is possible, however, that the pre-RA repertoire of citrullinated peptide:HLA-DR–specific T cells experiences some incomplete form of immune tolerance that poses a barrier to RA. It has been difficult to distinguish between these possibilities because even foreign epitope-specific CD4+ T cells are rare and can be difficult to detect in diverse repertoires (Moon et al., 2007).

Here, we used HLA-DR4 (DR4) transgenic (Tg) mice (Ito et al., 1996), fluorophore-labeled peptide:DR4 tetramers, and a sensitive flow cytometry–based cell enrichment method (Moon et al., 2007) to assess the capacity of citrullinated peptide:DR4-specific T cells to respond to immunization with citrullinated peptides. The results show that CD4+ T cells capable of responding to certain DR4-bound citrullinated peptides are completely absent from the repertoire while other DR4-bound citrullinated peptides elicit poor CD4+ T cells responses. Thus, the citrullinated peptide:DR4-specific pre-RA T cell repertoire shows evidence of varying degrees of immune tolerance that could pose a barrier to the development of the disease.

Peptide loading of DR4 tetramers

Our goal was to use fluorophore-labeled DR4 tetramers containing citrullinated peptides to track cognate T cells by flow cytometry. DR4 molecules containing citrullinated peptides could not be produced in vitro by the standard approach of covalently attaching the peptide to the MHCII β chain via genetic means (Crawford et al., 1998; Moon et al., 2007) because codons for citrulline do not exist in eukaryotic cells (Mondal and Thompson, 2019). Accordingly, we developed a system to load citrullinated peptides into DR4 molecules. The approach relied on DR4 molecules containing the class II invariant chain-associated peptide (CLIP) peptide that normally occupies the peptide-binding groove of nascent MHCII molecules during biosynthesis (Denzin and Cresswell, 1995). HLA-DM, which is targeted to low pH sites of MHCII biosynthesis, removes CLIP and facilitates the binding of a different peptide. Some types of MHCII molecules, however, can exchange CLIP for other peptides without HLA-DM (Stebbins et al., 1995). We therefore attempted to produce an HLA-DM–independent system for the sake of simplicity. A plasmid was constructed to encode human CLIP followed by a linker with an internal thrombin cleavage site and then the DRB1*04:01 β chain. This construct was expressed in insect cells with another plasmid encoding the DRA*0101 α chain. DRA*0101/CLIP-thrombin linker-DRB1*04:01 heterodimeric monomers (CLIP(T):DR4) were purified from culture supernatants, biotinylated, and tetramerized with streptavidin (SA)-fluorophores (SA-phycoerythrin [SA-PE] or SA-allophycocyanin [SA-APC]). These tetramers were then mixed with the well-characterized DR4-binding peptide 306–318 from influenza hemagglutinin (HAp; Zanders et al., 1984) under various conditions. The resulting tetramers were used to detect cognate T cells by flow cytometry in DR4 Tg mice immunized with HAp in CFA. Spleen and lymph node cell suspensions from several mice were split into equal samples, magnetically enriched with the HAp-exchanged DR4 tetramers (Moon et al., 2009), and compared to a positive-control tetramer in which HAp was covalently attached to the DRB1*04:01 β chain. The samples were gated for viable single cells with the light scatter properties of lymphocytes that expressed CD4 but non-T cell markers CD11b, F4/80, CD11c, B220, or CD14 (Fig. 1, A and B). Tetramer-binding cells that expressed the activation marker CD44 were identified in the CD4+ population. About 3,500 CD4+ T cells were detected with the covalent HAp:DR4 tetramer (Fig. 1 C). A reagent produced by treating the CLIP-thrombin-DR4 tetramer with thrombin at pH 6 during the HAp exchange reaction detected just as many CD4+ T cells as the covalent HAp:DR4 tetramer (Fig. 1 D). Increasing the pH of the reaction to 7.4 partially decreased the number of CD4+ T cells detected by the resulting tetramer (Fig. 1 E), and omission of thrombin at pH 6 reduced it (Fig. 1 F) to the background number detected with unexchanged CLIP(T):DR4 tetramer (Fig. 1 G). These results indicated that CLIP in the CLIP(T):DR4 tetramer could be completely exchanged with HAp without HLA-DM in the presence of thrombin at pH 6.

These conditions were then used in an ELISA format to assess the binding of several citrullinated peptides to plate-bound DR4 as a preliminary step toward tetramer production. Peptides from human cartilage intermediate layer protein (CILP cit988), fibrinogen β chain (Fib cit74), vimentin (Vim cit71), and enolase 1 (Eno cit15) were chosen for study (Table S1) based on previous work identifying specific T cells in RA patients (Gerstner et al., 2020). The native versions of these peptides contain arginine at P4 of their putative nine-mer cores and thus would not be expected to bind well to DR4. The citrullinated versions in contrast were expected to bind DR4 because they contained citrulline at P4 with P1, 6, and 9 residues that are permissive for DR4 binding. Indeed, each of the citrullinated peptides bound to DR4 at least as well as HAp and Fib cit74, Vim cit71, and Eno cit15 bound to DR4 at least 10 times better than their arginine-containing versions, while CILP cit988 bound only slightly better (Fig. 2). Thus, Fib cit74, Vim cit71, and Eno cit15 bound much better to DR4 than native versions, whereas native CILP and CILP cit988 were both good binders. These results are consistent with the literature (Ting et al., 2018) and indicate that DR4 tetramers containing each of the citrullinated peptides could be produced for use in flow cytometry experiments.

The CD4+ T cell repertoire has evidence of tolerance to DR-bound citrullinated peptides

We then used a strategy to study tolerance in the citrullinated peptide:DR4-specific T cell repertoire based on previous research on mice expressing GFP as an artificial self-antigen in different cells of the body (Malhotra et al., 2016). Three forms of tolerance were identified in that study by flow cytometric measurement of GFP peptide:MHCII tetramer-binding CD4+ T cells in mice immunized with GFP. Mice that expressed GFP in many thymic antigen-presenting cells lacked GFP peptide:MHCII tetramer-binding CD4+ T cells before or after immunization, consistent with intrathymic deletion of such T cells during development. In contrast, mice that expressed GFP in rare thymic antigen-presenting cells generated fewer GFP peptide:MHCII tetramer-binding effector T cells than control mice lacking GFP. Furthermore, the expanded population was deficient in type 1 (Th1), type 17 (Th17), and follicular T helper cells (Tfh), and enriched for Treg cells and an unknown population of conventional T (Tconv) cells lacking Th1, Th17, and Tfh markers. This response pattern was associated with deletion of T cells with the highest GFP peptide:MHCII tetramer-binding capacity (Malhotra et al., 2016). Finally, mice that expressed GFP only in the cytosols of cells outside the thymus generated the same number and type of GFP peptide:MHCII tetramer-binding T cells after immunization as mice that did not express GFP, suggesting that the GFP peptide:MHCII-specific T cell repertoire did not experience any form of tolerance. If DR4-bound mouse CILP cit988, Fib cit74, Vim cit71, and Eno cit15, which are similar or identical to the human versions (Table S1), are not normally presented in DR4 Tg mice, then the cognate CD4+ T cell repertoire should not be tolerant to these ligands and immunization with citrullinated human peptides should induce robust expansion of citrullinated human peptide:DR4-specific Th1, Th17, and Tfh cells but few Treg cells.

We first assessed unmanipulated mice to look for evidence of clonal deletion of citrullinated peptide:DR4-specific T cells before immunization. This was possible because peptide:MHCII tetramer-based cell enrichment has the sensitivity to detect naïve T cells in the pre-immune repertoire (Moon et al., 2007). Single-cell suspensions were prepared from spleen and draining lymph nodes of DR4 Tg mice, stained with a 50:50 mix of two DR4 tetramers each containing the same peptide but labeled with different fluorophores, and then enriched with magnetic beads. This double staining approach was first used by Locksley and colleagues (Stetson et al., 2002) and later by us (Nelson et al., 2015) to improve the TCR specificity of the method by eliminating T cells that bind only one of the tetramers, and thus cannot be binding it by the TCR, and including T cells that bind both tetramers as expected for cells that have half their TCRs occupied by each tetramer. Earlier work showed that the limit of detection of this assay is about three cells per mouse (Moon et al., 2007). At least twice this many HAp:DR4 tetramer-binding CD4+ T cells were detected in six of six mice, with an average of about 80 cells per mouse (Fig. 3, A–C). CILP cit988:DR4 or Eno cit15:DR4-specific tetramer-binding CD4+ T cells were much rarer. Six or more CILP cit988:DR4 tetramer-binding CD4+ T cells were detected in five of seven mice with an average of only about nine cells per mouse, while Eno cit15:DR4-specific tetramer-binding CD4+ T cells were detected in only one of seven mice, making them too rare for further analysis. These results indicated that the citrullinated p:DR4-specific T cell populations were smaller than one specific for a genuine foreign peptide:DR4 ligand.

That conclusion assumes, however, that the cells of interest were genuine antigen-specific T cells that bound the tetramers via their TCRs. Several pieces of evidence indeed indicate that this is the case. First, essentially all cells in the HAp:DR4 and CILP cit988:DR4 tetramer-binding populations expressed CD4 but not CD8 (Fig. 3, B–E) as expected for cells with MHCII-restricted TCRs (Rudolph et al., 2006). As another test of TCR specificity, samples were stained simultaneously with a 50:50 mix of HAp:DR4 and CILP cit988:DR4 tetramers labeled with different fluorophores. As shown in Fig. 3 F, the CD4+ T cells that bound HAp:DR4 tetramer did not bind CILP cit988:DR4 tetramer and the other way around as expected for cells with TCRs specific for different peptide:MHCII ligands. In addition, the CILP cit988:DR4 tetramer-binding population bound less tetramer per cell than the HAp:DR4 tetramer-binding one (Fig. 3 G) as observed in other cases in which T cells with high avidity for a self-peptide:MHCII ligand were deleted from the repertoire (Moon et al., 2011; Malhotra et al., 2016). Furthermore, the HAp:DR4 tetramer-binding T cells had the CD44low phenotype expected for naïve cells that were never exposed to the peptide:MHCII ligand for which their TCRs are specific (Fig. 3 H). This was less clear for the CILP cit988:DR4 tetramer-binding T cells in most mice although the small size of this population made phenotypic assessment difficult. Thus, the combined evidence supports the conclusion that the peptide:DR4 tetramer-based enrichment method detects CD4+ T cells in a TCR-specific manner and suggests that the Eno cit15:DR4- and CILP cit988:DR4-specific T cell populations undergo complete and partial clonal deletion, respectively.

As another test of this hypothesis, we immunized DR4 Tg mice by subcutaneous injection with CILP cit988, Eno cit15, Fib cit74, or Vim cit71 peptides or viral peptides—HAp or a peptide from the envelope protein of Yellow Fever virus (YFVp; Mateus et al., 2020)—in CFA. 14 d after immunization, single-cell suspensions were prepared from spleen and draining lymph nodes and enriched with citrullinated human peptide:DR4 tetramers prepared using the conditions described in Fig. 3. Tetramer-binding cells were then stained with antibodies specific for T cells and informative cell surface proteins and analyzed by flow cytometry.

The gating scheme used to identify tetramer-binding cells is shown in Fig. 4 A for examples from mice immunized with HAp, Fib cit74, or Vim cit71 peptides. Mice immunized with HAp contained about 60,000 tetramer-binding CD44hi CD4+ T cells in their secondary lymphoid organs (SLO) and mice immunized with YFVp had about 10,000 (Fig. 4 B). This difference is likely related to differences in the number of naïve CD4+ T cells specific for these foreign peptide:DR4 ligands, a variable that is known to influence the number of effector cells produced after immunization (Nelson et al., 2015). In contrast, mice immunized with Eno cit15 contained almost no tetramer-binding cells (Fig. 4 B) as expected since such cells were essentially undetectable before immunization (Fig. 3 B). Most Vim cit71-immunized mice also lacked tetramer-binding T cells. Mice immunized with CILP cit988 or Fib cit74 contained about 6,000 or 600 tetramer-binding CD4+ T cells in their SLO, numbers which were less than those induced by foreign peptides HAp and YFVp but significantly greater than those induced by Vim cit71 or Eno cit15. Thus, although Eno cit17:DR4- and Vim cit71:DR4-specific T cells were absent, the expansion of CILP cit988:DR4- and Fib cit74:DR4-specific T cells after immunization demonstrated that some citrullinated peptide responsive cells were present, as predicted for CILP cit988 by detection of tetramer-binding T cells before immunization (Fig. 3 B).

We then assessed the T helper subset differentiation to look for tolerance effects in the tetramer-binding cells that expanded following immunization. Treg cells in the CD44hi tetramer-binding populations were identified by expression of Foxp3 and Tconv cells by lack of this transcription factor. CD44hi tetramer-binding Tconv populations were further divided into RORγt+ Th17, RORγt CXCR6+ Th1, RORγt CXCR5+ Tfh (Crotty, 2011), or RORγt CXCR5 CXCR6 cells of unknown lineage (Hong et al., 2022). RORγt was used to detect Th17 cells because it is the master transcription factor for this subset (Ivanov et al., 2006). CXCR6 was used to detect Th1 cells (Kim et al., 2001; Zander et al., 2022) rather than T-bet, the lineage-defining transcription factor for this subset (Szabo et al., 2000), because it gave better staining. Examples of the gating strategy are shown for HAp:DR4 and CILP cit988:DR4 tetramer-binding populations in Fig. 5 A. The HAp:DR4 and YFVp:DR4 tetramer-binding populations in peptide-immunized mice consisted of about 25% Tfh, 40% Th1, 15% Th17, and <2% Treg cells (Fig. 5 B) as expected for foreign peptide:MHCII-specific effector cells responding to the relevant peptide in CFA (Malhotra et al., 2016). In contrast, although the percentages of Tfh and Th17 cells in the CILP cit988:DR4 and Fib cit74:DR4 tetramer-binding populations in peptide-immunized mice were like those in the HAp:DR4 and YFVp:DR4 tetramer-binding populations, the percentages of Th1 cells were lower and the percentages of Treg cells were higher. CILP cit988:DR4 and Fib cit74 tetramer-binding populations also had significantly higher percentages of unknown lineage cells than the HAp:DR4 and YFVp:DR4 tetramer-binding populations. Thus, CILP cit988:DR4- and Fib cit74:DR4-specific T cell repertoires tended to overproduce Treg cells and underproduce Th1 cells, as previously described for T cell repertoires specific for self-peptide:MHCII complexes displayed by a minority of thymic antigen-presenting cells.

The goal of this study was to determine whether CD4+ T cells specific for DR4-bound citrullinated host peptides respond to these ligands as if they were foreign epitopes, that is, with no barrier from immune tolerance mechanisms. The initial premise was that the native versions of these peptides with an arginine at P4 could not bind to DR4 and thus could not induce tolerance in the T cells with TCRs specific for the DR4-bound citrullinated peptides that share TCR contact residues at P2, 3, 5, 7, and 8 with the native peptides. It was therefore expected that naïve DR4 Tg mice would contain naïve CD4+ T cells capable of robust clonal expansion and differentiation following immunization with the citrullinated peptides. On the contrary, we found that immunization of DR4 Tg mice failed to induce expansion of Vim cit71:DR4 or Eno cit15:DR4 tetramer-binding T cells even though both peptides bind well to DR4. The fact that Eno cit15:DR4 tetramer-binding T cells could not be detected before immunization indicates that the T cell repertoires specific for these complexes experience a potent form of tolerance, which based on previous work, was likely caused by intrathymic deletion of almost all cognate T cells (Malhotra et al., 2016). This conclusion is supported by earlier work showing that immunization of mice expressing shared epitope-containing HLA-DR molecules with Vim cit71 did not generate in vitro proliferation or cytokine production by CD4+ T cells (Becart et al., 2021; James et al., 2014). Similarly, James et al. (2014) showed that T cells from DR4 Tg mice immunized with Eno cit15 proliferated only weakly in vitro in response to this peptide. Because the native Vim and Eno peptides bind very weakly to DR4, Vim cit71:DR4 and Eno cit15:DR4 complexes are the ligands likely to be responsible for the observed T cell tolerance. Earlier findings that PAD enzymes and citrullinated proteins are present in mouse medullary thymic epithelial cells (mTECs; Engelmann et al., 2016) indicate that these are the antigen-presenting cells that cause deletion of Vim cit71:DR4- and Eno cit15:DR4-specific T cells. In any case, the near completeness of the deletion suggests that many thymic antigen-presenting cells display Vim cit71:DR4 and Eno cit15:DR4 complexes.

The Fib cit74:DR4- and CILP cit988:DR4-specific CD4+ T cell repertoires had evidence of a milder form of tolerance. As noted in other studies (James et al., 2014; Lim et al., 2021), Fib cit74:DR4- and CILP cit988:DR4-specific T cells are capable of clonal expansion, which proves that not all CD4+ T cells in these repertoires are deleted. Our direct detection of CILP cit988:DR4-specific T cells in the preimmunization repertoire cements this conclusion. The expanded Fib cit74:DR4- and CILP cit988:DR4 tetramer-binding populations, however, had a lower frequency of Th1 cells and a higher frequency of Treg and unknown lineage cells than populations induced by viral peptides. This signature was observed in a previous study for a CD4+ T cell repertoire specific for a self-peptide:MHCII complex displayed on a minority of thymic antigen-presenting cells (Malhotra et al., 2016). DR4-bound peptides from CILP and fibrinogen might be sparsely displayed by mTECs because transcripts encoding CILP and fibrinogen are at least 10 times less abundant in mTECs than those encoding vimentin and enolase 1 (Heng et al., 2008). The tolerance effects that were observed for Fib cit74:DR4-specific T cells are likely caused by Fib cit74:DR4 complexes since native Fib binds DR4 very poorly. The situation is less clear for CILP since CILP cit988:DR4 and native CILP both bind to DR4. The relatively robust clonal expansion and near normal pattern of Th differentiation exhibited by CILP cit988:DR4-specific CD4+ T cells, however, indicates that CILP cit988:DR4 and/or native CILP:DR4 complexes have only weak tolerizing effects, including deletion of T cells with TCRs with the highest avidity for these ligands.

The immune tolerance that we observed in citrullinated peptide:DR4-specific T cell repertoires could be a barrier to RA development in most individuals and could explain why it is difficult to induce the disease in most mouse strains and does not occur in most DR4+ people. A challenge for the future will be to understand the differentiation states of the citrullinated peptide:DR4-specific T cells that are not deleted and can overcome their functional limitations under certain conditions to cause RA.

Peptides

Peptides with or without N-terminal FITC groups were purchased from Genscript and reconstituted in DMSO.

Peptide exchange into CLIP(T):DR4 tetramers

Sequences encoding human CLIP (87–101) followed by linker with an internal thrombin cleavage site (PVSKMRMATPLLMQAGGGGSLVPRGSGGGG—nonamer binding core and thrombin cleavage sites underlined) were cloned upstream into an expression plasmid (Moon et al., 2007) upstream of sequences encoding the DRB1*04:01 chain followed by a 6× HIS tag. This plasmid was transfected into S2 insect cells with three other plasmids encoding the DRA*0101 chain linked to a BirA tag, BirA ligase, and Blasticidin resistance. Biotinylated CLIP(T):DR4 β chain/DR α chain monomers (CLIP(T):DR4) were produced and purified from culture supernatants as previously described (Nelson et al., 2015). Monomers were tetramerized with SA-PE or SA-APC. CLIP(T):DR4/SA-PE tetramer was mixed with peptides in a 0.2 ml PCR tube such that the final concentration of tetramer was 1 µM and peptide 100 µM. 0.1 U of restriction grade thrombin (Millipore) was added and the mixture was incubated at 37°C for 6 h. Sodium citrate buffer, pH 5.4, was then added to the mixture at a 1:5 ratio (citrate buffer:tetramer/peptide mixture) to reduce the overall pH to 6. The reaction mixture was then incubated at 37°C overnight. The mixture was then used for staining at a 20-nM final concentration.

Peptide exchange into plate-bound CLIP(T):DR4 monomers

Enzyme immunoassay/radioimmunoassay flat bottom 96-well plates (3590; Corning Costart REF) were incubated sequentially with 10 µg/ml SA, 2% BSA (A9647; Sigma-Aldrich) in PBS blocking solution, and 1.2 mg/ml CLIP(T):DR4-biotin for 1 h at 37°C with 0.5% BSA in PBS washes between steps. FITC-labeled peptides in PBS were added to wells and incubated at 37°C for 1 h. Restriction grade thrombin, 0.025 U (cat# 69671; Millipore), was then added to each well and incubated at 37°C for 1 h. Sodium citrate buffer, pH 5.4, was then added to the mixture at a 1:5 ratio (citrate buffer:tetramer/peptide mixture) to reduce the overall pH to 6 and incubated at 37°C overnight to allow the added peptides to replace CLIP in the DR4 groove. The plates were washed and incubated at 37°C for 1 h with horseradish peroxidase (HRP)–labeled FITC antibody (cat# 200-0032-037; Jackson ImmunoResearch), BioFX TMB Super Sensitive One Component HRP Microwell Substrate (Surmodics IVD), and BioFX 450 nm Liquid Stop Solution for TMB Microwell Substrates (Surmodics IVD). The optical density of the resulting color reaction was read at 405 nm.

Mice

6–8-wk-old male or female mice containing transgenes encoding the HLA-DR α 1 domain fused to the murine H-2 I-E α 2 domain and the HLA-DRB1*0401 β 1 domain fused to the H-2 I-Ed β 2 domain and lacking endogenous murine MHCII molecules (B6.129S2-H2-Ab1tm1Gru Tg(HLA-DRA/H2-Ea,HLA-DRB1*0401/H2-Eb)1Kito; Ito et al., 1996) were purchased from Taconic Laboratories. The mice were immunized subcutaneously with 10 or 100 µg of peptide emulsified in 0.20 ml of CFA. Mice were cared for by staff of the University of Minnesota Research Animal Resources in a specific pathogen–free facility. All animal studies were performed according to National Institutes of Health guidelines using protocols approved by the University of Minnesota Institutional Animal Care Committee.

Cell enrichment and flow cytometry

Single-cell suspensions were prepared from pooled mouse spleens and lymph nodes (axillary, brachial, inguinal, lumbar, caudal, and cervical) by mashing the tissue with a syringe plunger. Cells were incubated with 50 nM dasatinib (Lissina et al., 2009) at 37°C for 30 min and then stained for 1 h at room temperature with 10 nM peptide:DR4/SA-APC and peptide:DR4/SA-PE tetramers at the same time along with Fc block, CXCR5 (L138D7; BV650; BioLegend), and CXCR6 (SA051D1; BV711; BioLegend) antibodies. In experiments involving naïve mice, cells were incubated with 50 nM dasatinib throughout the 1-h tetramer incubation at room temperature. Samples were then mixed with magnetic beads conjugated with APC and PE antibodies and enriched for tetramer-binding cells on magnetized columns (Miltenyi) as described (Moon et al., 2009). In experiments involving immunized mice, enriched samples were then stained for 30 min at 4°C with antibodies specific for CD4 (GK1.5; BUV496; BD), CD8 (53-6.7; APC-ef780; BD), CD11b (M1/70; APC-ef780; eBioscience), CD11c (N418; APC-ef780; eBioscience), F4/80 (BM8; APC-ef780; eBioscience), CD44 (IM7; BUV395; BD), or B220 (RA3–6B2; APC-ef780; eBioscience), and with GhostDye Red 780 (Tonbo Biosciences) to assess viability. Cells were then fixed and permeabilized with the Foxp3/Transcription Factor Staining Buffer Set (eBioscience) and stained overnight at 4°C with antibodies specific for T-bet (eBio4B10; PE-Cy7; Invitrogen), Foxp3 (FJK-16s; AF488; eBioscience), and RORγt (Q31-378; PerCP-Cy5.5; BD). In experiments involving naïve mice, enriched samples were stained for 30 min at 4°C with antibodies specific for CD3 (17A2, BV421; BioLegend), CD4 (H129.19, BUV496; BD Biosciences), CD8 (53-6.7, PE-CF594; BD Biosciences), CD44 (IM7, FITC; BioLegend), CD11b (M1/70; APC-ef780; eBioscience), CD11c (N418; APC-ef780; eBioscience), F4/80 (BM8; APC-ef780; eBioscience), B220 (RA3–6B2; APC-ef780; eBioscience), NK1.1 (PK136, APC-ef780; Invitrogen), and with GhostDye Red 780. Fluorescence data were collected with an LSRII flow cytometer (BD) and analyzed with FlowJo software.

Statistical methods

Statistical analyses were performed with Prizm software. Outlying values were identified in the Th subset differentiation data set using the robust regression and outlier removal method with Q = 1%. Outlying values from three mice were removed from the total dataset of 33 mice.

Online supplemental material

In Table S1, peptide sequences are shown as a single-letter code with x = citrulline. Putative nine-mer core sequences are shown in yellow. Differences between human and mouse peptides are shown in red.

Data are available in the article itself and its supplementary materials.

The authors thank Jennifer Walter and Lillian R. Hilo for technical assistance, Daniel Mueller for helpful discussions, and the University of Minnesota Flow Cytometry Facility for maintenance of flow cytometers.

Grants from the National Institutes of Health (P01 AI35962 and R01 AI027998 to M.K. Jenkins: T32 HL007741 to M.K. McElwee) and the Rheumatology Research Foundation Scientist Development Award (award 889570 to S.A. Mahmud) supported this work.

Author contributions: M.K. McElwee designed and performed experiments and wrote the paper. S.A. Mahmud designed and performed experiments. T. Dileepan designed and produced peptide:DR4 tetramers. M.K. Jenkins conceptualized and oversaw experiments and wrote the paper. All authors reviewed and provided edits to the manuscript.

Alivernini
,
S.
,
G.S.
Firestein
, and
I.B.
McInnes
.
2022
.
The pathogenesis of rheumatoid arthritis
.
Immunity
.
55
:
2255
2270
.
Becart
,
S.
,
K.B.
Whittington
,
A.
Prislovsky
,
N.L.
Rao
, and
E.F.
Rosloniec
.
2021
.
The role of posttranslational modifications in generating neo-epitopes that bind to rheumatoid arthritis-associated HLA-DR alleles and promote autoimmune T cell responses
.
PLoS One
.
16
:e0245541.
Crawford
,
F.
,
H.
Kozono
,
J.
White
,
P.
Marrack
, and
J.
Kappler
.
1998
.
Detection of antigen-specific T cells with multivalent soluble class II MHC covalent peptide complexes
.
Immunity
.
8
:
675
682
.
Crotty
,
S.
2011
.
Follicular helper CD4 T cells (TFH)
.
Annu. Rev. Immunol.
29
:
621
663
.
Darrah
,
E.
, and
F.
Andrade
.
2018
.
Rheumatoid arthritis and citrullination
.
Curr. Opin. Rheumatol.
30
:
72
78
.
Davis
,
M.M.
,
J.J.
Boniface
,
Z.
Reich
,
D.
Lyons
,
J.
Hampl
,
B.
Arden
, and
Y.
Chien
.
1998
.
Ligand recognition by alpha beta T cell receptors
.
Annu. Rev. Immunol.
16
:
523
544
.
Denzin
,
L.K.
, and
P.
Cresswell
.
1995
.
HLA-DM induces CLIP dissociation from MHC class II alpha beta dimers and facilitates peptide loading
.
Cell
.
82
:
155
165
.
Engelmann
,
R.
,
A.
Biemelt
,
A.
Cordshagen
,
A.
Johl
,
D.
Kuthning
, and
B.
Müller-Hilke
.
2016
.
The prerequisites for central tolerance induction against citrullinated proteins in the mouse
.
PLoS One
.
11
:e0158773.
Germain
,
R.N.
1994
.
MHC-dependent antigen processing and peptide presentation: Providing ligands for T lymphocyte activation
.
Cell
.
76
:
287
299
.
Gerstner
,
C.
,
S.
Turcinov
,
A.H.
Hensvold
,
K.
Chemin
,
H.
Uchtenhagen
,
T.H.
Ramwadhdoebe
,
A.
Dubnovitsky
,
G.
Kozhukh
,
L.
Rönnblom
,
W.W.
Kwok
, et al
.
2020
.
Multi-HLA class II tetramer analyses of citrulline-reactive T cells and early treatment response in rheumatoid arthritis
.
BMC Immunol.
21
:
27
.
Heng
,
T.S.P.
,
M.W.
Painter
, and
Immunological Genome Project Consortium
.
2008
.
The immunological genome project: Networks of gene expression in immune cells
.
Nat. Immunol.
9
:
1091
1094
.
Hong
,
S.-W.
,
P.D.
Krueger
,
K.C.
Osum
,
T.
Dileepan
,
A.
Herman
,
D.L.
Mueller
, and
M.K.
Jenkins
.
2022
.
Immune tolerance of food is mediated by layers of CD4+ T cell dysfunction
.
Nature
.
607
:
762
768
.
Ito
,
K.
,
H.J.
Bian
,
M.
Molina
,
J.
Han
,
J.
Magram
,
E.
Saar
,
C.
Belunis
,
D.R.
Bolin
,
R.
Arceo
,
R.
Campbell
, et al
.
1996
.
HLA-DR4-IE chimeric class II transgenic, murine class II-deficient mice are susceptible to experimental allergic encephalomyelitis
.
J. Exp. Med.
183
:
2635
2644
.
Ivanov
,
I.I.
,
B.S.
McKenzie
,
L.
Zhou
,
C.E.
Tadokoro
,
A.
Lepelley
,
J.J.
Lafaille
,
D.J.
Cua
, and
D.R.
Littman
.
2006
.
The orphan nuclear receptor RORgammat directs the differentiation program of proinflammatory IL-17+ T helper cells
.
Cell
.
126
:
1121
1133
.
James
,
E.A.
,
M.
Rieck
,
J.
Pieper
,
J.A.
Gebe
,
B.B.
Yue
,
M.
Tatum
,
M.
Peda
,
C.
Sandin
,
L.
Klareskog
,
V.
Malmström
, and
J.H.
Buckner
.
2014
.
Citrulline-specific Th1 cells are increased in rheumatoid arthritis and their frequency is influenced by disease duration and therapy
.
Arthritis Rheumatol.
66
:
1712
1722
.
Kim
,
C.H.
,
E.J.
Kunkel
,
J.
Boisvert
,
B.
Johnston
,
J.J.
Campbell
,
M.C.
Genovese
,
H.B.
Greenberg
, and
E.C.
Butcher
.
2001
.
Bonzo/CXCR6 expression defines type 1-polarized T-cell subsets with extralymphoid tissue homing potential
.
J. Clin. Invest.
107
:
595
601
.
Lim
,
J.J.
,
C.M.
Jones
,
T.J.
Loh
,
Y.T.
Ting
,
P.
Zareie
,
K.L.
Loh
,
N.J.
Felix
,
A.
Suri
,
M.
McKinnon
,
F.
Stevenaert
, et al
.
2021
.
The shared susceptibility epitope of HLA-DR4 binds citrullinated self-antigens and the TCR
.
Sci. Immunol.
6
:eabe0896.
Lissina
,
A.
,
K.
Ladell
,
A.
Skowera
,
M.
Clement
,
E.
Edwards
,
R.
Seggewiss
,
H.A.
van den Berg
,
E.
Gostick
,
K.
Gallagher
,
E.
Jones
, et al
.
2009
.
Protein kinase inhibitors substantially improve the physical detection of T-cells with peptide-MHC tetramers
.
J. Immunol. Methods
.
340
:
11
24
.
Malhotra
,
D.
,
J.L.
Linehan
,
T.
Dileepan
,
Y.J.
Lee
,
W.E.
Purtha
,
J.V.
Lu
,
R.W.
Nelson
,
B.T.
Fife
,
H.T.
Orr
,
M.S.
Anderson
, et al
.
2016
.
Tolerance is established in polyclonal CD4(+) T cells by distinct mechanisms, according to self-peptide expression patterns
.
Nat. Immunol.
17
:
187
195
.
Malmström
,
V.
,
A.I.
Catrina
, and
L.
Klareskog
.
2017
.
The immunopathogenesis of seropositive rheumatoid arthritis: From triggering to targeting
.
Nat. Rev. Immunol.
17
:
60
75
.
Marrack
,
P.
,
J.P.
Scott-Browne
,
S.
Dai
,
L.
Gapin
, and
J.W.
Kappler
.
2008
.
Evolutionarily conserved amino acids that control TCR-MHC interaction
.
Annu. Rev. Immunol.
26
:
171
203
.
Mateus
,
J.
,
A.
Grifoni
,
H.
Voic
,
M.A.
Angelo
,
E.
Phillips
,
S.
Mallal
,
J.
Sidney
,
A.
Sette
, and
D.
Weiskopf
.
2020
.
Identification of novel yellow fever class II epitopes in YF-17D vaccinees
.
Viruses
.
12
:
1300
.
Mondal
,
S.
, and
P.R.
Thompson
.
2019
.
Protein arginine deiminases (PADs): Biochemistry and chemical biology of protein citrullination
.
Acc. Chem. Res.
52
:
818
832
.
Moon
,
J.J.
,
H.H.
Chu
,
J.
Hataye
,
A.J.
Pagán
,
M.
Pepper
,
J.B.
McLachlan
,
T.
Zell
, and
M.K.
Jenkins
.
2009
.
Tracking epitope-specific T cells
.
Nat. Protoc.
4
:
565
581
.
Moon
,
J.J.
,
H.H.
Chu
,
M.
Pepper
,
S.J.
McSorley
,
S.C.
Jameson
,
R.M.
Kedl
, and
M.K.
Jenkins
.
2007
.
Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude
.
Immunity
.
27
:
203
213
.
Moon
,
J.J.
,
P.
Dash
,
T.H.
Oguin
,
J.L.
McClaren
,
H.H.
Chu
,
P.G.
Thomas
, and
M.K.
Jenkins
.
2011
.
Quantitative impact of thymic selection on Foxp3+ and Foxp3- subsets of self-peptide/MHC class II-specific CD4+ T cells
Proc. Natl. Acad. Sci. USA
.
108
:
14602
14607
.
Nelson
,
R.W.
,
D.
Beisang
,
N.J.
Tubo
,
T.
Dileepan
,
D.L.
Wiesner
,
K.
Nielsen
,
M.
Wüthrich
,
B.S.
Klein
,
D.I.
Kotov
,
J.A.
Spanier
, et al
.
2015
.
T cell receptor cross-reactivity between similar foreign and self peptides influences naive cell population size and autoimmunity
.
Immunity
.
42
:
95
107
.
Olerup
,
O.
,
M.
Troye-Blomberg
,
G.M.
Schreuder
, and
E.M.
Riley
.
1991
.
HLA-DR and -DQ gene polymorphism in West Africans is twice as extensive as in north European Caucasians: Evolutionary implications
.
Proc. Natl. Acad. Sci. USA
.
88
:
8480
8484
.
Rudolph
,
M.G.
,
R.L.
Stanfield
, and
I.A.
Wilson
.
2006
.
How TCRs bind MHCs, peptides, and coreceptors
.
Annu. Rev. Immunol.
24
:
419
466
.
Scherer
,
H.U.
,
T.
Häupl
, and
G.R.
Burmester
.
2020
.
The etiology of rheumatoid arthritis
.
J. Autoimmun.
110
:
102400
.
Stebbins
,
C.C.
,
G.E.
Loss
Jr
,
C.G.
Elias
,
A.
Chervonsky
, and
A.J.
Sant
.
1995
.
The requirement for DM in class II-restricted antigen presentation and SDS-stable dimer formation is allele and species dependent
.
J. Exp. Med.
181
:
223
234
.
Stetson
,
D.B.
,
M.
Mohrs
,
V.
Mallet-Designe
,
L.
Teyton
, and
R.M.
Locksley
.
2002
.
Rapid expansion and IL-4 expression by Leishmania-specific naive helper T cells in vivo
.
Immunity
.
17
:
191
200
.
Szabo
,
S.J.
,
S.T.
Kim
,
G.L.
Costa
,
X.
Zhang
,
C.G.
Fathman
, and
L.H.
Glimcher
.
2000
.
A novel transcription factor, T-bet, directs Th1 lineage commitment
.
Cell
.
100
:
655
669
.
Ting
,
Y.T.
,
J.
Petersen
,
S.H.
Ramarathinam
,
S.W.
Scally
,
K.L.
Loh
,
R.
Thomas
,
A.
Suri
,
D.G.
Baker
,
A.W.
Purcell
,
H.H.
Reid
, and
J.
Rossjohn
.
2018
.
The interplay between citrullination and HLA-DRB1 polymorphism in shaping peptide binding hierarchies in rheumatoid arthritis
.
J. Biol. Chem.
293
:
3236
3251
.
Tubo
,
N.J.
, and
M.K.
Jenkins
.
2014
.
CD4+ T cells: Guardians of the phagosome
.
Clin. Microbiol. Rev.
27
:
200
213
.
Valesini
,
G.
,
M.C.
Gerardi
,
C.
Iannuccelli
,
V.A.
Pacucci
,
M.
Pendolino
, and
Y.
Shoenfeld
.
2015
.
Citrullination and autoimmunity
.
Autoimmun. Rev.
14
:
490
497
.
Xing
,
Y.
, and
K.A.
Hogquist
.
2012
.
T-Cell tolerance: Central and peripheral
.
Cold Spring Harb. Perspect. Biol.
4
:
a006957
.
Zander
,
R.
,
M.Y.
Kasmani
,
Y.
Chen
,
P.
Topchyan
,
J.
Shen
,
S.
Zheng
,
R.
Burns
,
J.
Ingram
,
C.
Cui
,
N.
Joshi
, et al
.
2022
.
Tfh-cell-derived interleukin 21 sustains effector CD8+ T cell responses during chronic viral infection
.
Immunity
.
55
:
475
493.e5
.
Zanders
,
E.D.
,
M.
Feldmann
,
N.
Green
, and
J.R.
Lamb
.
1984
.
Direct evaluation of antigen binding to human T lymphocyte clones: Involvement of major histocompatibility complex products in antigen binding
.
Eur. J. Immunol.
14
:
1101
1105
.

Author notes

*

M.K. McElwee, T. Dileepan, and S.A. Mahmud contributed equally to this paper.

Disclosures: The authors declare no competing interests exist.

This article is distributed under the terms of an Attribution–Noncommercial–Share Alike–No Mirror Sites license for the first six months after the publication date (see http://www.rupress.org/terms/). After six months it is available under a Creative Commons License (Attribution–Noncommercial–Share Alike 4.0 International license, as described at https://creativecommons.org/licenses/by-nc-sa/4.0/).

Supplementary data