CONCLUSION
In summary, this study expanded the known range of koffrates and provided a collection of TCR sequences which may have application for comparative studies between high and low affinity receptors.
METHODS
Study subjects
Two individuals with acutely resolving HCV infection (CL-MCRL and CL-3089) were sourced from the HITS cohort (reference) for single-cell RNA-sequencing (scRNA-seq). For each population, cells from a single timepoint following resolution of infection were selected for sorting cells for single-cell colony expansion.
Single-cell RNA-sequencing
scRNA-seq was performed as described previously19. Briefly HCV-specific CD8+ T cell populations were isolated by HLA-I Dextramer staining (HLA-A*01:01 ATDALMTGF or HLA-B*07:02 GPRLGVRAT ) (Immudex) and sorting with flow cytometry (BD FACSAria III) for plate-based Smart-Seq2 scRNA-seq35,36. Sequencing was performed with 150bp paired end sequencing on Illumina NovaSeq 6000 and NextSeq 500 platforms. Full length TCR sequences were reconstructed using the VDJPuzzle package with default parameters 22.
Reversible and non-reversible multimer preparation
Reversible MHC Streptamers were prepared by incubating 1μg of the relevant pMHC monomer conjugated to AF488 (HLA-B*07:02 GPRLGVRATor HLA-A*01:01 ATDALMTGF ) (laboratory synthesis) with 1μg Strep-Tactin-APC (IBA Lifescience, PN 6-5010-001) in 25 μl of FACS buffer (PBS/1% BSA). Non-reversible multimers were similarly prepared by incubating 1μg of the relevant pMHC monomer conjugated to biotin (laboratory synthesis) with 1 μg of Streptavidin-PE (BioLegend, PN 405245). Incubations were performed on ice in the dark for 45 minutes. Streptamers and multimers were used within 6 hours of preparation.
Koffdissociation assay using PBMC
The dissociation rate for the GPR-specific population from CL-MCRL was measured directly from cryopreserved PBMC. Cryopreserved cells were thawed and stained first with the reversible Streptamer for 20 minutes followed by surface antibody staining: CD8-eFluor450 (OKT-8) and CD19-PE-Cy5 (HIB19) for an additional 20 minutes without washing. Cells were then washed twice and stained with a HLA-B*07:02 GPRLGVRATDextramer (Immudex, Copenhagen, Denmark) for 20 minutes. Two final washes were performed, and cells were resuspended with one drop of propidium iodide (Molecular Probes) and 100 μl of FACS buffer. Data was acquired on a LSR Fortessa x20 (BD Biosciences) in a cooled tube holder. After a brief period of acquisition, a 1X biotin solution (Sigma-Aldrich) was added to dissociate the reversible Streptamer. Additional biotin solution was added as required and the tube containing cells was periodically placed on ice without sample acquisition to preserve sample availability during the dissociation process. The time between subsequent acquisitions was embedded in raw FCS files and used to determine the time since initial acquisition.
Single-cell colony expansion
Total PBMC were thawed and stained first with the relevant Dextramer conjugated to PE (Immudex), followed by Fixable yellow viability stain (Invitrogen) and finally a panel of surface antibodies: CD3-BV480 (UCHT1), CD8-APC-R700 (RPA-T8), CD19-PE-Cy5 (HIB19). Single HCV-specific CD8+ T cells (lymphocytes/singlets/live cells/CD3+/CD19/CD8+/Dextramer+) were sorted into separate wells of a 384-well flat bottom plate containing expansion media (CTS OpTmizer T Cell Expansion serum free media) (Gibco) supplemented with L-glutamine (2mM) (Sigma-Aldrich), penicillin/streptomycin (1X) (Sigma-Aldrich), IL‑2 (500IU/ml) (STEMCELL Technologies), PHA-L (1μg/μl) (Sigma‑Aldrich) and containing 1×106 gamma irradiated feeder PBMC cells from unrelated blood donors. Sorting was performed with single cell precision on a FACS Aria III flow cytometer (BD) and plates were incubated in a 37°C, 5% CO2 incubator.
On day 7 after sorting, 50μl of supplementing media containing CTS OpTmizer T Cell Expansion serum free media with L-glutamine (2 mM), penicillin/streptomycin (1X concentration), and IL-2 (500 IU/ml) was added to each well. On every subsequent 3-4 days, 50 μl of existing media was removed and replaced with an equal volume of supplementing media. If cell colonies reached 100% confluency, they were transferred to 96-well flat bottom plates (typically after three weeks) in supplementing media containing PHA-L (1μg/μl). Colonies continued to be monitored and supplemented every 3-4 days and after an additional two weeks or when 100% confluency was reached, they were cryopreserved in two identical aliquots containing 50% media, 40% FBS, and 10% DMSO.
Koff dissociation assay using colony expansions
Single-cell expansions were retrieved from cryopreservation and stained first with the reversible Streptamer for 20 minutes followed by surface antibody staining: CD8-eFluor450 (OKT-8) and CD19-PE-Cy5 (HIB19) for an additional 20 minutes without washing. Cells were then washed twice and stained with the non‑reversible multimer for 20 minutes. Two final washes were performed, and cells were resuspended with one drop of propidium iodide (molecular probes) and 100 μl of FACS buffer.
Data acquisition by flow cytometry for dissociation assay
Stained cells were transferred to 5 mL round-bottom polystyrene tubes which had been pierced with a syringe to allow delivery of a biotin solution during acquisition. Cells were kept cool at 4°C or on ice at all points during preparation and acquisition. Data were acquired on an LSR Fortessa x20 (BD Biosciences). At the beginning of data acquisition, events were recorded for up to 60 seconds followed by rapid injection of 1.5 mL PBC containing 1X biotin (Sigma-Aldrich). Samples were continuously acquired until loss of MHC monomer signal was observed and additional biotin solution was added as required.
Calculation of dissociation and koff constants
Compensated fluorescence and time values from the lymphocyte/singlet/live/CD8+/non-reversible multimer subpopulation were exported from raw FCS files using FlowJo (v10.6.1, BD) and loaded into Prism (v7.04, GraphPad). Time values (in seconds), MHC-AF488, and Strep-Tactin-APC fluorescence values were used as input for non-linear regression under ‘XY analyses’ to compute the half-life values for the associated dissociation curve. koff values were calculated by taking the reciprocal of the computed half-life in seconds and upper and lower 95% confidence intervals for half-lives were also reported.
Targeted TCR amplification of colony expansions
Total RNA was extracted from an aliquot of cells from selected single-cell colony expansions using an RNeasy micro kit (Qiagen) and stored at -20°C. RNA was reverse transcribed using the SuperScript IV Reverse Transcriptase kit (Invitrogen) according to manufacturer’s recommendations. First, template RNA was incubated with a poly dT primer (AAGCAGTGGTATCAACGCAGAGTACT30VN) and nucleotides for 5 minutes at 65°C then immediately placed on ice. Next, the mix containing reverse transcriptase and buffer was added and the mixture was incubated at 50°C for 15 minutes followed by 80°C for 10 minutes.
Two rounds of nested PCRs were performed to enrich for TCR-associated transcripts using previously published primer pools37. All reactions used identical cycling conditions consisting of an initial denaturation (95°C for one minute) followed by 35 cycles of: 95°C for 20 seconds, 52°C for 20 seconds, 72°C for 45 seconds, followed by a final extension at 72°C for 4 minutes. For the first round PCR, alpha and beta chain transcripts were amplified in a single reaction consisting of PCR buffer (1X), dNTPs (0.2 mM), TRAC external primer (0.4 μM), TRBC external primer (0.4 μM), TRAV external primer pool (0.1 μM), TRBV external primer pool (0.1 μM), Taq DNA polymerase (Qiagen), and 5ul of reverse transcribed template. For the second round nested PCR, 2.5 μl of first round product was added to a total reaction mix of 25 μl containing PCR buffer (1X), dNTPs (0.2mM) and Taq DNA polymerase (Qiagen). Separate reactions were prepared for the amplification of alpha and beta chain transcripts containing either TRAC internal primer (0.4 μM) and TRAV internal primer pool (0.1 μM), or TRBC internal primer (0.4 μM) and TRBV internal primer pool (0.1 μM).
Amplification products were validated by gel electrophoresis and purified using ExoSAP-IT Express (Applied Biosystems). Samples were submitted for Sanger sequencing at the UNSW Ramaciotti Centre for Genomics using the relevant internal constant region primer. Electropherograms after sequencing were manually inspected for clear signals. Nucleotide sequences were used to query the online IMGT/V-QUEST tool and obtain TCR gene usage and CDR3 amino acid sequences (available at http://www.imgt.org/IMGT_vquest/input).
ACKNOWLEDGEMENTS
We thank the study participants for their generous donation of samples. The HITS-p investigators include Andrew Lloyd, Lisa Maher, Kate Dolan, Paul Haber, William Rawlinson, Carla Treloar and Gregory Dore.  This research was supported from National Health and Medical Research Council of Australia (NHMRC) Project Nos. APP1121643, 1027551, 1060199, Partnership No. 1016351, and Programme Nos. 510488 and 1053206. The HITS-c cohort was supported by the UNSW Hepatitis C Vaccine Initiative and NHMRC Project Grant No. 630483. F.L., A.R.L. and R.A.B. are supported by NHMRC Research Fellowships (Numbers: 1128416, 1041897 and 1084706). F.L., A.R.L. and R.A.B. are supported by NHMRC Research Fellowships (Numbers: 1128416, 1041897 and 1084706). C.C. and J.S. are supported by Australian Government Research Training Program (RTP) Scholarships. D.H.B. was supported by the Deutsche Forschungsgemeinschaft (DFG, SFB-TRR338/1-452881907-A01). K.S. is supported by the German Federal Ministry of Education and Research (BMBF, projects 01KI2013). The authors acknowledge the facilities and technical support from the staff at the UNSW Flow Cytometry core facility located within the Mark Wainwright Analytical Centre. Sequencing was performed by staff at the Ramaciotti Centre for Genomics, UNSW. Analyses of single-cell sequencing data were performed with resources allocated from the National Computational Infrastructure, Australia.