Supporting Information Billerbeck et al. 10.1073/pnas.0914839107 SI Methods ELISpot. To analyze HCV-specific cytokine–secreting CD8+ T Antibodies and Staining. Anti–CD8-PerCP, anti–CD8-PE, anti– cells, ELISpot assays were performed with intrahepatic and pe- + CD8-PerCP-Cy5.5, anti–CD8-Alexa Fluor 700–anti–CD4-PerCP, ripheral CD8 T cells from chronically HCV-infected patients. + anti–CD4-FITC, anti–CD4-APC, anti–IFN-γ-FITC, anti–TCRγδ- Intrahepatic and peripheral CD8 T cells were isolated and PE, anti–TCRαβ-PE, anti–TNF-α,isotypePE,isotypeFITC,and expanded in an antigen nonspecific way (2 to 4 weeks). IL-17 γ isotypeAPC were obtainedfrom BD Pharmingen. Anti–CD3-FITC/ ELISpot kits (R & D Systems) and IFN- ELISpot kits (Diac- ’ APC, anti–IL-22-APC, anti–CCR6-APC, anti–CCR2-APC, anti– lone) were used according to the manufacturers instructions, × 5 CXCR6-PE, anti–CXCR3-FITC, anti–CCR7-FITC, biotinylated and 1 10 cells were plated per well. Peptides were added at a fi μ polyclonal goat anti–IL-23R, and anti–IFN-γ-FITC were obtained nal concentration of 10 g/mL per peptide and were combined – in pools of 10 peptides, with each peptide being contained in two from R & D Systems; and anti CD161-APC from Miltenyi Biotech. fi Anti–IL-17A-PE, anti–IL-17A-Alexa 647, anti–TCRαβ-APC, anti– different pools in a matrix setup, allowing the identi cation of CD127-PE, anti–RORγT-PE, and anti–IL18R-FITC were from single positive peptides. Tests with peptide pools were per- formed in duplicates; two wells without peptides served as neg- eBioscience. Anti–CD161-PE, anti–CD3-ECD, and anti–TCRγδ- ative controls; four wells each stimulated with PMA/ionomycin FITC were obtained from Immunotech/Beckman Coulter. Anti– served as positive controls. Responses were considered positive if granzyme B-APC and anti–CD3-PacificOrangewereobtainedfrom the number of spots per well was at least three times as high as Caltag Laboratories and anti–Vα24-FITC from Serotec. Anti- – γ the mean of negative controls (typically 0 to 2 spots per well). Runx2 was obtained from Santa Cruz Biotechnology. Anti ROR T- Overlapping peptides derived from HCV strain H77 (genotype Alexa 488 was obtained from Cambridge Biosciences. FITC- or 1a) spanning 18 aa and overlapping by 11 aa were obtained from APC-conjugated antibodies to CD45RA, CD45RO, CD38, CD62L, β Beiresources (National Institutes of Health Biodefense and CD56, CD103, TGF RII, PD-1, ICOS, GITR, CD38, CD62L, Emerging Infections Research Resources Repository). CCR5, CXCR4, CCR9, CD85j, CD27, and CD28 for studies in Fig. − S4 were obtained from BD Pharmingen and R & D Systems. Ex- ELISA. Purified CD161 or CD161+ CD8+ T cells (1 × 105 per well) travidin-PE was obtained from Sigma. were cultured in a 96-well plate and stimulated with PMA/ + For immunofluorescence experiments, CD8 T cells were iso- ionomycin for 48 h. The concentrations of IL-17 and IFN-γ in the lated using an EasySep CD8 negative selection kit and positive PE supernatants were analyzed using human IL-17 or INF-γ Duoset selection beads (Stem Cell Technologies) following anti-IL18R or ELISA (R & D Systems) according to the manufacturer’s in- anti-CD161 staining. Cells were plated in eight-chamber slides structions. coated with CC2 (Nunc), fixed with methanol/acetone, air dried, blocked with 1% BSA, and washed with PBS/Tween 1% solution Microarray Data Analysis. Gene expression profiles for three before staining. For intracellular staining by flow cytometry per- sorted CD8 cell subsets from each of four healthy donors meabilization buffers from BD Biosciences and eBiosciences were were obtained using Illumina Sentrix Human Whole Genome used according to the manufacturer’s instructions. 6 BeadChips. Raw data were exported from the BeadStudio CD8 selection and polyclonal expansion of PBMCs. PBMCs (4 × 106) software (v3.3.7; Illumina) for further processing and analysis wereresuspendedin5mLPBSsolutionandincubatedwithCD8 using the R statistical software and BioConductor packages (1– Dynabeads (Invitrogen) for 20 min at 4 °C. Bound CD4+ Tcells 5). Signal data and detection scores were extracted for each of were isolated by a particle magnetic concentrator. The purity of the 12 samples. Signal data were background corrected using CD4+ Tcellswas>95% by FACS analysis. The CD4+ Tcells array-specific measures of background intensity based on were plated into one well of a 24-well plate (Greiner) in 1 mL negative control probes before being transformed and nor- complete medium (RPMI medium 1640 containing 10% FCS, malized using the vs.n2 BioConductor package. After filtering out probes not detected in any of the 12 samples 1% streptomycin/penicillin, and 1.5% 1 M Hepes containing 100 < U/mL IL-2; Hoffmann-La Roche), 0.04 g/mL anti–human CD3 (detection score 0.95), 32,159 probes were retained for further monoclonal antibody (Immunotech), and 2 × 106 irradiated analysis. Statistical testing was performed using the Linear Models for autologous PBMCs as feeder cells. Twice per week, 1 mL of Microarray Analysis (limma) BioConductor package. Three con- medium was exchanged and 100 U/mL IL-2 was added. After 3 + trasts were generated corresponding to the possible cell type com- weeks, the expanded CD8 PBMCs and intrahepatic lympho- − parisons (CD161++ vs. CD161+, CD161++ vs. CD161 ,CD161+ cytes were analyzed. − vs. CD161 ), each using a paired design. Raw P values were cor- CD8 selection and polyclonal expansion of LILs. Briefly, the liver biopsy rected for multiple testing using the false discovery rate controlling specimens were homogenized using a 70-m Dounce tissue grinder procedure of Benjamini and Hochberg. Gene details were added (BD Biosciences). Cell suspensions were incubated with CD8 fi fi + to nal probe lists using the relevant annotation le (HumanWG- Dynabeads (Invitrogen) for 20 min at 4 °C. Bound CD8 cells were 6_V3_0_R0_11282955_A.txt) downloaded from the Illumina Web isolated using a particle magnetic concentrator. The purity of site (http://www.illumina.com). CD8+ cells was >95% by FACS analysis. The intrahepatic CD4+ T + cells were then expanded as described earlier for CD8 PBMCs. RNA Isolation, cDNA Synthesis, and Quantitative RT-PCR. RNA was + Importantly, the expansion of peripheral and intrahepatic CD8 T isolated from sorted cells using the miRNeasy Mini Kit (Qia- cells from a given patient was always performed in parallel and for gen). cDNA synthesis was performed using a mixture of random the same time before analysis to obtain comparable results for and oligo(dT) primers and SuperScript III reverse transcriptase both compartments. Tonsil/kidney–derived lymphocytes were (Invitogen). Quantitative real-time PCR reactions were carried prepared similarly, although without CD8+ T cell selection and out using a LightCycler 480 (Roche Diagnostics) and a com- culture, and joint-derived lymphocytes prepared without the ho- bination of preoptimized Qiagen QuantiTect Primer Assays mogenization step. [IL-23R (Hs_IL23R_1_SG) and HPRT1 as an internal control Billerbeck et al. www.pnas.org/cgi/content/short/0914839107 1of14 (Hs_HPRT1_1_SG)] and primers designed using Primer36 (RORc, manufacturer’s instructions, with cycling conditions as follows: 5′-AGAAGGACAGGGAGCCAAG-3′ for forward and 5′-CAA- 95 °C for 15 min and 40 cycles of 94 °C for 15 s, 54 °C for 30 s, and GGGATCACTTCAATTTGTG-3′ for reverse). LightCycler 480 72 °C for 30 s. Fold expression changes were determined by the 2- SYBR Green I Master Mix (Roche) was used according to the ΔΔCT method (7). 1. R Development Core Team (2007) R: a language and environment for statistical com- 5. Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and puting (R Foundation, Vienna). powerful approach to multiple testing. J R Stat Soc, B 57:289–300. 2. Gentleman RC, et al. (2004) Bioconductor: open software development for computational 6. Rozen S, Skaletsky H (2000) Primer3 on the WWW for general users and for biology and bioinformatics. Genome Biol 5:R80. biologist programmers. In Bioinformatics methods and protocols: methods in 3. Huber W, von Heydebreck A, Sültmann H, Poustka A, Vingron M (2002) Variance molecular biology, eds Krawetz S, Misener S (Humana Press, Totowa, NJ), pp stabilization applied to microarray data calibration and to the quantification of 365–386. differential expression. Bioinformatics 18 (suppl 1):S96–S104. 7. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using 4. Smyth GK (2005) Limma: linear models for microarray data. In Bioinformatics and real-time quantitative PCR and the 2(-delta delta C(T)) method. Methods 25: computational biology solutions using R and Bioconductor, ed Gentleman R (Springer- – Verlag, New York), pp 397–420. 402 408. 29.1 2.37 12.3 0.72 5 A 105 10 8.58 4 104 10 3 103 10 CD8 <FITC-A> IFN- <APC-A> CD161 2 102 10 0 0 CD161++gate 86.5 0.47 62.1 6.45 2 3 4 5 0102 103 104 105 010 10 10 10 <PE-A> <PE-A> B 7.64 0.57 27.7 1.42 5 10 105 4 8.13 10 104 3 10 103 CD4 <APC-A> <FITC-A> IFN- CD161 2 2 10 10 CD161+gate 0 0 91.5 0.3 65.1 5.83 2 3 4 5 010 10 10 10 0102 103 104 105 <PE-A> <PE-A> IL-17 IL-17 C 10.0 p<0.0001 10.0 p=n.s. 7.5 7.5 CD4 5.0 5.0 2.5 2.5 % IL-17+ T cells % IL-17+ T cells 0.0 0.0 CD161+ CD161- CD8 CD4 D 2500 200 200 CD161++ CD161+ CD161- 2000 150 150 1500 0.2 0.9 0.41 100 Unstim. # Cells # Cells 100 # Cells 1000 50 50 500 0 0 0 0 1 2 3 4 100 101 102 103 104 100 101 102 103 104 10 10 10 10 10 FL1-H: IFN-G FITC FL1-H: IFN-G FITC FL1-H: IFN-G FITC Fig.