Heterogeneity assessment of functional T cell avidity

The potency of cellular immune responses strongly depends on T cell avidity to antigen. Yet, functional avidity measurements are rarely performed in patients, mainly due to the technical challenges of characterizing heterogeneous T cells. The mean functional T cell avidity can be determined by the IFN-γ Elispot assay, with titrated amounts of peptide. Using this assay, we developed a method revealing the heterogeneity of functional avidity, represented by the steepness/hillslope of the peptide titration curve, documented by proof of principle experiments and mathematical modeling. Our data show that not only natural polyclonal CD8 T cell populations from cancer patients, but also monoclonal T cells differ strongly in their heterogeneity of functional avidity. Interestingly, clones and polyclonal cells displayed comparable ranges of heterogeneity. We conclude that besides the mean functional avidity, it is feasible and useful to determine its heterogeneity (hillslope) for characterizing T cell responses in basic research and patient investigation.


Supplementary Tables
Supplementary Table 1│Overview of all T cell clones tested by the IFN-γ Elispot assay. EC 50 , Hillslope, R square (as a measure of the goodness of curve fit) and maximal numbers of spots were determined by Elispot assays as described in Methods. Purification of CD8 T cells. 10x10 6 PBMCs were thawed, washed, and resuspended with 80 µl cell sorting buffer (PBS, 0.2 % BSA, 5 µM EDTA), kept on ice. 20 µl of CD8 microbeads (Miltenyi, 130-045-201) were added, mixed and incubated for 30 min (mixing every 10 min) in the fridge. Cells were washed by adding 2 ml of cell sorting buffer, centrifuged at 1500 rpm for 5 minutes and resuspended in 500 µl of cell sorting buffer. MS columns (Miltenyi) were loaded with a maximal number of 10 7 labeled cells or 2x10 8 total cells. 15 ml Falcon tubes were labeled as CD8-and placed below the columns to collect the flow-through of unlabeled CD8-cells. The columns were rinsed once with 500 µl of cell sorting buffer, before the addition of the cell suspension. The volume passed through entirely and the column was washed three times by adding 500 µl of cell sorting buffer. The columns were removed from the separators and placed onto Falcon tubes. 2 ml of cell sorting buffer was added to the columns and the plunger was pushed firmly into the columns to flush the positive-labeled cells. The CD8-and the CD8+ cells fractions were centrifuged and washed after with 2 ml RPMI medium supplemented with 8% Human Serum to determine the cell number.

Supplementary
Expansion of CD8 T cell clones. Upon expansion, the specificity of each generated T cell clone was verified by multimer staining. T cell clones were expanded by periodic (every 14-21 d) restimulation with PHA, human IL-2, and irradiated feeder cells. Monoclonality was confirmed for the majority of the clones used in this study, by sequencing of TCR-α and -β chains, and by FACS staining with Vβ specific antibodies (data not shown).

In vitro stimulation (IVS) of purified CD8 T cells with peptide for expansion of the antigen specific T cells.
Because the available blood volume and cell numbers from melanoma patients were low, we first expanded them by in vitro stimulation (IVS) for 12-14 days. The CD8-fraction was irradiated with a dose of 30 Grays and resuspended with fresh RPMI + 8% Human Serum. Melan-A analog peptide (ELAGIGILTV, stock 1mg/ml), Cytomegalovirus peptide (CMV; NLVPMVATV, stock 1mg/ml), Epstein-Barr virus peptide (EBV; GLCTLVAML, stock 1mg/ml), and Influenza virus peptide (FLU; GILGFVFTL, stock 1mg/ml), were added respectively, at the final concentration of 1 µM. CD8 T cells were added and incubated at 37°C. After 24 hours, IL-2 was added, starting with 50 U/ml, and increased to 100 U/ml on the second day, and from the next day on, with 150 U/ml. The culture medium was changed twice per week. After 12-14 days, the antigen specificity was determined by tetramer FACS staining, and cultures were used for Elispot assays. We studied the impact of in vitro stimulation (IVS) on the functional avidity and heterogeneity of the expanded cells, in comparison to direct ex vivo analysis. Analyzing 11 healthy donors with different specificities for CMV, EBV and FLU virus we found that the EC 50 values in all cases were significantly increased (Supplementary Fig. 6a). Therefore, IVS increased the functional avidity of the expanded cells, likely through enhanced activation and co-receptor activity [1][2][3] . In contrast, the hillslope values stayed similar or changed to variable degrees, such that the difference between direct ex vivo analysis and after IVS was not statistically significant (Supplementary Fig. 6b), with some populations showing increased and others decreased hillslopes after IVS. Overall, IVS clearly impacted on the T cell's avidity, suggesting that future studies should preferably be done directly ex vivo, albeit this requires large blood volumes.
Chromium release cytotoxicity assay. Chromium release assays were performed using radioactive 51 Chromium. For peptide titration experiments, 51 Cr-labeled TAP2/2-deficient T2 cells (HLA-A*0201+) were pulsed with serial dilutions from 10 −6 M to 10 −13 M of Melan-A EAAGIGILTV peptide. The effector cell to target cell (E:T) ratio was 10:1. After 4 hours incubation, supernatants were analyzed in a TopCount NXT benchtop microplate scintillation and luminescence counter. The percentage of specific lysis was calculated as 100 x (experimental -spontaneous release) / (total -spontaneous release). The cytotoxicity assays were performed three times.
IFN-γ Elispot assay. 25 µl of 35% ethanol/well was put in 96 well PVDF plates (MSIPS4510, Millipore) and incubated at room temperature (RT) for 30 seconds. The wells were emptied by flicking the plate over a sink and gently tapping on absorbent paper. The plates were thoroughly washed 3x with 100 µl 1X PBS per well. 100 µl of diluted capture antibody (100 µl into 10 ml 1X PBS) was added to every well and the plates were covered and incubated at 4 o C overnight. The wells were washed as previously, once with 100 µl 1X PBS. 100 µl of culture medium with 10% serum was added to every well and the plates were incubated at RT for 2 hours. The wells were washed with 100 µl 1X PBS, and 300 antigen specific CD8 T cells in 50 µl of RPMI with 8% Human serum were plated with 50 µl of 20'000 T2 cells per well and the addition of 1 µΜ of the native Melan A peptide (EAA, stock 1mg/ml), in a total volume of 200 µl per well. The plates were incubated at 37 o C in a CO 2 incubator for 20 hours. With the aim to always plate 300 antigen specific T cells per well, the polyclonal T cell populations were first characterized with fluorescent tetramers by flow cytometry, and the numbers of cells were adjusted according to the percent of tetramerpositive T cells. The next day the plates were incubated with 100 µl of 0.05% PBS-Tween solution (wash buffer) per well at 4 o C for 10 min and subsequently 3x with wash buffer. 100 µl of diluted detection antibody (100 µl into 10 ml Dilution buffer) was added to every well and incubated at RT for 1 hour 30 min. The plates were washed 3x with 100 µl of wash buffer and 100 µl of diluted Streptavidin -AP conjugate was added to every well. After an incubation of 1 hour at RT, the plates were washed 3x with wash buffer and 3x with distilled water (both sides of the membrane, peeling off the plate bottom). 100 µl of BCIP/NBT buffer was added to every well and the plates were incubated for 5-15 min, monitoring spot formation visually throughout the incubation to assess sufficient color development. The wells were emptied and both sides of the membrane were rinsed 3x with distilled water. The frequency of the resulting colored spots corresponding to the cytokine producing cells, was determined using the Elispot Bioreader 5000. All Elispot experiments were performed in triplicates. For clones, the experiments were performed 2-3 times. The total polyclonal populations were only tested once because of the limited cell availability. The standard deviations from the triplicate Elispot cultures (Figure 5a-f) were relatively small. Furthermore, we always tested a clone of a well-known functional avidity as a control in parallel to the patients' polyclonal populations, and only used the results from those experiments in which the controls showed proper reproducibility.