Fig 1.
Diagrammatic representation of the pMHC array.
Lipophilically dyed CD8+ T cells (0.8–1.2 x 106 per microarray) were incubated at 37°C in warm colourless X-VIVO 15 media with the pMHC array. Each pMHC spot contains 1ng of tetramer and each slide can hold up to 1,000 spots of pMHC.
Table 1.
pMHCs used on the array to detect virus and LAA-specific T cell populations within the peripheral blood CD8+ population of leukaemia patients.
Fig 2.
Direct comparison of epitope specific T cell detection by pMHC tetramers following negative CD8+ isolation by conventional techniques.
CD8+ T cells were isolated from HLA-A*0201+, normal donor buffy coat using commercially available negative isolation kits. This normal donor was CMV sero-negative, pMHC-CMV pp65 negative and pMHC-Flu M1 positive. (A) 106 cells per test were incubated with 1μg of anti-CD8-FITC antibody, the cells washed and analysed by flow cytometry using a FACScaliburTM. Method 2 gave the highest number of CD8+ cells (FL1). (B) 106 CD8 T cells isolated from the same normal donor by Method 1 or 2 were labelled with anti-CD8-FITC (y-axis) and pMHC-(Flu or CMV)-SA-PE (x-axis) and analysed by flow cytometry. We found that Method 2 StemSep negative isolation of CD8s provided the best separation of pMHC-labelled CD8 T cells, in this case with Flu M1 tetramers.
Fig 3.
Correlation between the detection of specific T cell populations using flow cytometry and pMHC arrays.
Negatively isolated CD8+ T cells from a normal donor who was pMHC-Flu M1 positive and CMV sero-negative were (A) labelled with anti-CD8-FITC (y-axis) and pMHC-(Flu or CMV)-SA-PE (x-axis) and analysed by flow cytometry. We showed that a minimum 0.7 x 106 CD8+ cells (including controls) could be used to show that the sample had Flu M1+ specific T cells but not A2/CMV pp65 specific T cells by conventional flow cytometry. CD8-FITC (FL1-H) is shown on the x-axis and pMHC-SA-PE (FL2-H) staining on the y-axis. (B) 106 CD8+ T cells/ml, from the same normal donor sample, were lipophillically dyed with DiD and incubated for 20 minutes at 37°C with a custom-made hydrogel slide. Unbound cells were washed away with warm X- VIVO. CD8+ T cells (shown stained red) are visible at the single cell level bound to the Flu tetramer, which is visualised by the 5% AlexaFluor532 conjugated to streptavidin (shown as a green spot), on the ProScanArray. Composites show the co-localisation of Flu-specific CD8+ T cells to the Flu M1 tetramer spot. Few if any T cells are bound to the CMV pp65 pMHC spot or the negative control random pMHC library (also tetramerised with 5% AF532-streptavidin in streptavidin). Limits of detection in both assays were ≤0.1% of the CD8+ population.
Fig 4.
CD8+ T cells from leukaemia patients binding to pMHC spots on the pMHC array.
(A + B) SEM showing a CD8+ positive T cell bound at the site of a tetramer spot on the pMHC array; (C) a single pit on the pMHC array suggesting that T cells dig into the hydrogel when binding to pMHCs and subsequently detach.
Fig 5.
Analysis of leukaemia patient samples on the pMHC array.
pMHC array (A) printed with 40 tetramers visualised by AlexFluor532 (shown in green, with bleaching to white at the highest intensity). Each pMHC complex were printed six times consecutively on one or two rows in two different areas of the hydrogel; Each line contains up to seven pMHC spots shown within the orange box; (B-E) white circles indicate the position of the pMHC spot on the array while CD8+ lymphocytes visualised by DiD lipophilic staining (shown in red), indicate the presence of epitope specific T cells recognising (B) FLU M1 and Tyrosinase in AML016; (C) MelanA in AML013FU; (D) MUC1 and PASD1(2) in ALL003 and (E) PASD1(5) in AML001. All epitopes described were conjugated to MHC Class I (HLA-A2) complex.
Table 2.
Results from pMHC array analysis where there was detectable binding of virus and LAAs-specific pMHCs by “untouched” CD8+ T cells purified from leukaemia patients.
Fig 6.
PASD1 expression in AML patient samples.
Immunolabelling was used to detect PASD1 protein expression in leukaemia patient samples. Haematoxylin provides the blue background stain to allow differentiation of the cell nucleus from the cytoplasm. The brown precipitate indicates a positive reaction between secondary antibody and detection reagent, at the antigen site. In this image AML004, AML008 and AML014 immunolabelled cells are shown, AML015 was one of the samples which did not have demonstrable PASD1 expression. No primary antibody was used as a negative control and actin was used as a positive control for immunolabelling.