Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Jun 29;9:1378.
doi: 10.3389/fimmu.2018.01378. eCollection 2018.

Optimized Peptide-MHC Multimer Protocols for Detection and Isolation of Autoimmune T-Cells

Affiliations
Free PMC article

Optimized Peptide-MHC Multimer Protocols for Detection and Isolation of Autoimmune T-Cells

Garry Dolton et al. Front Immunol. .
Free PMC article

Abstract

Peptide-MHC (pMHC) multimers have become the "gold standard" for the detection and isolation of antigen-specific T-cells but recent evidence shows that normal use of these reagents can miss fully functional T-cells that bear T-cell receptors (TCRs) with low affinity for cognate antigen. This issue is particularly pronounced for anticancer and autoimmune T-cells as self-reactive T-cell populations are enriched for low-affinity TCRs due to the removal of cells with higher affinity receptors by immune tolerance mechanisms. Here, we stained a wide variety of self-reactive human T-cells using regular pMHC staining and an optimized technique that included: (i) protein kinase inhibitor (PKI), to prevent TCR triggering and internalization, and (ii) anti-fluorochrome antibody, to reduce reagent dissociation during washing steps. Lymphocytes derived from the peripheral blood of type 1 diabetes patients were stained with pMHC multimers made with epitopes from preproinsulin (PPI), insulin-β chain, glutamic acid decarboxylase 65 (GAD65), or glucose-6-phospate catalytic subunit-related protein (IGRP) presented by disease-risk allelles HLA A*02:01 or HLA*24:02. Samples from ankylosing spondylitis patients were stained with a multimerized epitope from vasoactive intestinal polypeptide receptor 1 (VIPR1) presented by HLA B*27:05. Optimized procedures stained an average of 40.5-fold (p = 0.01, range between 1.4 and 198) more cells than could be detected without the inclusion of PKI and cross-linking anti-fluorochrome antibody. Higher order pMHC dextramers recovered more cells than pMHC tetramers in parallel assays, and standard staining protocols with pMHC tetramers routinely recovered less cells than functional assays. HLA A*02:01-restricted PPI-specific and HLA B*27:05-restricted VIPR1-specific T-cell clones generated using the optimized procedure could not be stained by standard pMHC tetramer staining. However, these clones responded well to exogenously supplied peptide and endogenously processed and presented epitopes. We also showed that anti-fluorochrome antibody-conjugated magnetic beads enhanced staining of self-reactive T-cells that could not be stained using standard protocols, thus enabling rapid ex vivo isolation of autoimmune T-cells. We, therefore, conclude that regular pMHC tetramer staining is generally unsuitable for recovering self-reactive T-cells from clinical samples and recommend the use of the optimized protocols described herein.

Keywords: T-cell; ankylosing spondylitis; autoimmune disease; cancer epitope; dextramer; peptide–MHC multimer; tetramer; type 1 diabetes.

Figures

Figure 1
Figure 1
Self-reactive T-cells have low-affinity T-cell receptors (TCRs). Anti-pathogen TCRs tend to bind to cognate Peptide–MHC (pMHC) with relatively high affinity (KD < 20 μM). The rigors of central tolerance ensure that autoimmune TCRs bind with much weaker affinity. Anticancer TCRs tend to sit between these two extremes. This schematic depicts a general overview of what is normally observed. A minority of TCRs do not obey these general rules. Cancer TCRs that bind to unstable pMHC have been described can have higher affinites (KD ~ 15 µM) (16). TCRs specific for cancer neoantigens (non-self) may behave more like pathogen-specific TCR. The arrows at the top indicate the rough affinity detection threshold of TCRs amenable to staining with standard pMHC multimer staining and an optimized procedure including protein kinase inhibitor and antibody cross-linking. Optimized staining with pMHC tetramers is believed to detect almost all antigen-specific T-cells. An optimized dextramer is only required when staining the very weakest of functional autoimmune T-cells.
Figure 2
Figure 2
Enhanced detection of autoimmune T-cells from a type I diabetic patient using optimal tetramer staining techniques. (A) Standard and optimal tetramer staining approaches to detect antigen-specific T-cells. Optimal staining uses the protein kinase inhibitor (PKI) Dasatinib to treat T-cells before staining with tetramer and then an unconjugated anti-fluorochrome Ab. We have previously described the use of conjugated secondary antibodies to bind the primary antibody, adding further fluorescence to peptide–MHC multimers labeled T-cells, but for the purpose of this study only the primary cross-linking antibody was used. T-cell receptor (TCR). (B) Sorted CD8 T-cells from a HLA A*02:01+ patient with type I diabetes were stimulated with a peptide from CMV (pp65495–503, NLVPMVATV) and then stained 2 weeks later with irrelevant (hTERT540–548, ILAKFLHWL) or CMV tetramers using standard conditions. (C) From the same patient in (B) CD8 T-cells were stimulated with peptide (HLVEALYLV) from insulin-β chain10–18 or glutamate decarboxylase 65 (GAD65114–123, VMNILLQYVV). 2 weeks post-stimulation reactivity toward the respective peptide (100 nM) was assessed by co-incubation with TAPI-0 and detection of CD107a and TNF (left panel). The T-cell lines were stained with PE-conjugated irrelevant (as above) and respective autoimmune tetramers using standard or optimal conditions (right panel). The percentage of cells residing in each gate are shown.
Figure 3
Figure 3
Detection of preproinsulin (PPI) T-cells is further improved using dextramers under optimal staining conditions. (A) Schematic representation of tetramers and dextramers showing the number of streptavidins (typically, 6–7 for dextramers), peptide–MHC (pMHCs), and phycoerythrins per unit of reagent. pMHC to streptavidin molar ratios of 4:1 and 3:1 were used to assemble tetramers and dextramers, respectively. T-cell receptor (TCR). (B) Sorted CD8 T-cells from a HLA A*02:01+ patient with type I diabetes were stimulated with a peptide from PPI (PPI15–24) and assessed 2 weeks later for reactivity against the PPI peptide (100 nM) by co-incubation with TAPI-0 and detection of CD107a and TNF. (C) The T-cells were also stained with irrelevant (hTERT540–548, ILAKFLHWL) or PPI tetramers (upper panel) and dextramers (lower panel) using standard (pMHC multimer alone) and optimal protocols (PKI + anti-PE Ab). The percentage of cells residing in each gate are shown. The red box indicates cells that were sorted by flow cytometry and shown in Figure 4.
Figure 4
Figure 4
Optimal staining with preproinsulin (PPI) tetramers allowed functional T-cells to be isolated. Peripheral blood mononuclear cells (PBMCs) from a HLA A*02:01+ patient with type I diabetes were stimulated with PPI (PPI15–24) peptide (Figure 3A), which stained with PPI tetramer and dextramer under optimal conditions (PKI + anti-PE Ab). Tetramer+ cells were sorted by flow cytometry (Figure 3B) and expanded with PHA and irradiated allogeneic PBMCs. (A) The expanded cells were stained with irrelevant (ILAKFLHWL, hTERT540–548) and PPI tetramers using standard (tetramer alone) and optimal (as above) protocols. The percentage of cells residing in each gate is shown. (B) Reactivity of the enriched T-cells for PPI peptide was assessed by co-incubation with TAPI-0 and detection of CD107a and TNF. (C) Clone GD.PPI.1 grown from the enriched line was stained with irrelevant (hTERT) and PPI tetramers using standard and optimal protocols. The MFI of staining is shown on the histogram according to the key. (D) Overnight activation assay and MIP-1β enzyme-linked immunosorbent assay, with PHA as a positive control, PPI peptide (10 nM), and K562s transduced with genes for HLA A*02:01 (A2) ± PPI cDNA (PPI).
Figure 5
Figure 5
Anti-fluorochrome magnetic microbeads enhance the staining of T-cells with tetramers. (A) The ability of a primary (1°) unconjugated anti-fluorochrome antibody (Ab) to stabilize tetramer was also tested in the form of Ab-conjugated magnetic microbeads. (B) HLA A*02:01-restricted Melanoma reactive clone CR.NLS.3 was stained with irrelevant ILAKFLHWL (hTERT540–548) and index NLSALGIFST [Insulin-like growth factor 2 mRNA binding protein 2 (IMP2367–376)], PE-conjugated tetramers using the conditions shown. All samples were treated with PKI apart from the standard conditions. The anti-PE magnetic microbeads were used as recommended by the manufacturer (Miltenyi Biotech), 20 µL per 100 µL of staining volume for up to 1 × 107 cells, or bead volume dilutions thereof. The MFI of staining is shown for each condition.
Figure 6
Figure 6
Dextramers used with magnetic based purification allowed autoimmune T-cells to be isolated directly ex vivo. (A) Peripheral blood mononuclear cells from a HLA A*02:01+ patient with type I diabetes were treated with PKI and then stained with insulin-β chain10–18 (HLVEALYLV) PE-conjugated dextramers. Post staining the cells were labeled with anti-PE antibody-conjugated microbeads and magnetically enriched. T-cells were grown in vitro for 3 weeks then stained with PE-conjugated insulin-β chain tetramer and dextramers, with hTERT540–548 (ILAKFLHWL) multimers as an irrelevant control. Staining was performed under standard (multimer alone) or optimized (PKI + anti-PE Ab) conditions. The dotted line depicts the baseline for staining based on irrelevant multimers, with the percentage of cells staining above this shown for each condition. (B) The magnetically enriched T-cell line generated in A was tested for reactivity against insulin-β chain peptide by incubation with TAPI-0 and staining for TNF and CD107a. The percentage of cells residing in each gate is shown. (C,D) CD8 T-cells from an HLA A*02:01+A*24:02+ type 1 diabetes patient were enriched using the same approach shown in A, using either HLA A*24:02-PPI3–11 (LWMRLLPLL) dextramer (C) or HLA A*02:01-IGRP265–273 (VLFGLGFAI) dextramer (D). Two weeks post expansion the cells were stained with their respective tetramers. hTERT tetramers were used as an irrelevant control. The percent of gated cells is shown and the mean fluorescence intensity for the PPI or IGRP staining displayed.
Figure 7
Figure 7
HLA B*27:05 vasoactive intestinal polypeptide receptor 1 (VIPR1) tetramers used with optimal staining conditions allowed autoimmune T-cells to be detected and isolated from patients with ankylosing spondylitis. (A) Peripheral blood mononuclear cells (PBMCs) from a HLA B*27:05 patient with ankylosing spondylitis were stained with irrelevant (HIV p24 gag263–272; KRWILLGLNK) and VIPR1400–408 (RRKWRRWHL) PE-conjugated tetramers directly ex vivo under standard (tetramer alone) and optimal (PKI + anti-PE Ab) conditions. (B) PBMCs for the same patient in A were PKI treated and stained with PE-conjugated VIPR1400–408 tetramers then labeled with anti-PE antibody-conjugated microbeads for magnetic purification. After 3 weeks of culture, the line was stained with irrelevant (p24 gag) and VIPR1 tetramer under optimal staining conditions. (C) Using the same approach as in B from a second patient. Cells were stained 3 weeks post purification with irrelevant (HIV p24 gag263–272) and VIPR1400–408 tetramers using standard and optimal staining protocols.
Figure 8
Figure 8
CD8 clones isolated from an ankylosing spondylitis patient using optimal tetramer staining are fully functional. (A) CD8 T-cell clones grown from T-cell lines (Figure 7) generated following tetramer enrichment using an optimal magnetic bead protocol (Figure 5). GD.AS69, GD.AS2, and GD.Russ2 were stained with HIV p24 gag263–272 (KRWILLGLNK) and VIPR1400–408 (RRKWRRWHL) PE-conjugated HLA B*27:05 tetramers, using standard (tetramer alone) and optimized (PKI + anti-PE Ab) protocols, according to the key. The MFI of staining is shown. (B) Functional testing of clones GD.AS69 and GD.AS2 using a TAPI-0 assay with CD107a and TNF Abs. Patient autologous lymphoblastoid cell line (LCL) was used to present HLA B*27:05 irrelevant peptide (DRASFIKNL from the α2 domain of collagen type VI114–122) and cognate VIPR1400–408 peptide (RRKWRRWHL). Autologous LCL were lentivirally transduced with genes for the α2 chain of collagen type VI (control protein) and vasoactive intestinal polypeptide receptor 1 (VIPR1) and used in the activation assay. Percentage reactivity is shown for the CD107a+ TNF+ gate.

Similar articles

See all similar articles

Cited by 7 articles

See all "Cited by" articles

References

    1. Bridgeman JS, Sewell AK, Miles JJ, Price DA, Cole DK. Structural and biophysical determinants of alphabeta T-cell antigen recognition. Immunology (2012) 135(1):9–18.10.1111/j.1365-2567.2011.03515.x - DOI - PMC - PubMed
    1. Cole DK, Laugel B, Clement M, Price DA, Wooldridge L, Sewell AK. The molecular determinants of CD8 co-receptor function. Immunology (2012) 137(2):139–48.10.1111/j.1365-2567.2012.03625.x - DOI - PMC - PubMed
    1. Rudolph MG, Stanfield RL, Wilson IA. How TCRs bind MHCs, peptides, and coreceptors. Annu Rev Immunol (2006) 24:419–66.10.1146/annurev.immunol.23.021704.115658 - DOI - PubMed
    1. Stepanek O, Prabhakar AS, Osswald C, King CG, Bulek A, Naeher D, et al. Coreceptor scanning by the T cell receptor provides a mechanism for T cell tolerance. Cell (2014) 159(2):333–45.10.1016/j.cell.2014.08.042 - DOI - PMC - PubMed
    1. Van Laethem F, Sarafova SD, Park JH, Tai X, Pobezinsky L, Guinter TI, et al. Deletion of CD4 and CD8 coreceptors permits generation of alphabetaT cells that recognize antigens independently of the MHC. Immunity (2007) 27(5):735–50.10.1016/j.immuni.2007.10.007 - DOI - PubMed

LinkOut - more resources

Feedback