Exhaustive T-cell repertoire sequencing of human peripheral blood samples reveals signatures of antigen selection and a directly measured repertoire size of at least 1 million clonotypes

Abstract

Massively parallel sequencing is a useful approach for characterizing T-cell receptor diversity. However, immune receptors are extraordinarily difficult sequencing targets because any given receptor variant may be present in very low abundance and may differ legitimately by only a single nucleotide. We show that the sensitivity of sequence-based repertoire profiling is limited by both sequencing depth and sequencing accuracy. At two timepoints, 1 wk apart, we isolated bulk PBMC plus naïve (CD45RA+/CD45RO-) and memory (CD45RA-/CD45RO+) T-cell subsets from a healthy donor. From T-cell receptor beta chain (TCRB) mRNA we constructed and sequenced multiple libraries to obtain a total of 1.7 billion paired sequence reads. The sequencing error rate was determined empirically and used to inform a high stringency data filtering procedure. The error filtered data yielded 1,061,522 distinct TCRB nucleotide sequences from this subject which establishes a new, directly measured, lower limit on individual T-cell repertoire size and provides a useful reference set of sequences for repertoire analysis. TCRB nucleotide sequences obtained from two additional donors were compared to those from the first donor and revealed limited sharing (up to 1.1%) of nucleotide sequences among donors, but substantially higher sharing (up to 14.2%) of inferred amino acid sequences. For each donor, shared amino acid sequences were encoded by a much larger diversity of nucleotide sequences than were unshared amino acid sequences. We also observed a highly statistically significant association between numbers of shared sequences and shared HLA class I alleles.

Figure 1.

Distinct TCRB sequences. (A) Sequence coverage of distinct TRBJ sequences observed upon alignment of Illumina reads quality filtered to >99.9% predicted accuracy (Q30). There was substantial coverage, up to several million-fold, of the 13 human TRBJ segments, but thousands of artifactual distinct TRBJ sequences were also observed, arising from residual sequencing error. (B) Restricting the data set to D96 effectively retains all real TRBJ sequences and excludes all of the artifactual TRBJ sequences observed when no coverage restriction is applied (D100).

Figure 2.

TCRB diversity in peripheral blood samples. (A) Rarefaction curves plotted using all data from all libraries from blood draw 1 (round symbols) and blood draw 2, taken 1 wk later (square symbols) from donor 1. Random resampling was done in triplicate, and error bars are contained within symbols. Both curves plateau, suggesting that more sequencing of either sample would not be expected to produce many new sequences. However, this tendency toward leveling is a property of rarefaction curves, and the plateau is not informative regarding absolute abundance. Hence, rarefaction curves must be interpreted with caution. (B) Accumulation analysis provides a more meaningful measure of saturation. Here we show the number of new distinct sequences found in each library (x-axis) against the total number of distinct sequences from the blood sample as a whole. The TCRB diversity within blood draw 1 from donor 1 appears to have been captured, since analyzing additional libraries would not be expected to yield many new TCRB sequences. In contrast to the rarefaction curve present in the previous panel, library-based accumulation shows that the diversity of blood draw 2 is similar to that of blood draw 1 but has not yet been fully captured. (C) Despite saturation of blood draw 1, sequences found in blood draw 2 only partially overlap, indicating that there is considerable un-sampled diversity within the peripheral blood TCRB repertoire of this individual.

Figure 3.

TCRB sequence diversity in the naïve and memory compartments of a deeply sequenced individual. (A) Frequency distributions of CD45RA+/CD45RO− (naïve) and CD45RA−/CD45RO+ (memory) T-cell subsets isolated by FACS from PBMCs from a separate blood sample from donor 1, taken on day 1. The similarity of the two curves reflects the presence of high diversity within both subsets, although this is slightly greater for the naïve subset, as evident from the extended tail of the distribution. There are, in both cases, a small number of extreme copy clonotypes and a relative clonotype abundance varying over four orders of magnitude. (B) There is more overlap with the deeply sequenced sorted cells for the memory versus the naïve subset, but even the overlap of the memory subset is modest. This is consistent with a large total repertoire size that can be only partially captured in a given blood draw. (C) Of the sequences that were shared between sorted naïve and memory cells on day 1, there was a preferential transition to the memory subset on day 8.