Meromit Singer,Chao Wang,Le Cong,Nemanja D. Marjanovic,Monika S. Kowalczyk,Huiyuan Zhang,Jackson Nyman,Kaori Sakuishi,Sema Kurtuluş,David Gennert,Junrong Xia,John Kwon,James Nevin,Rebecca H. Herbst,Itai Yanai,Orit Rozenblatt‐Rosen,Vijay K. Kuchroo,Aviv Regev,Ana C. Anderson
(Cell 166, 1500–1511; September 8, 2016) In our paper we have identified distinct gene modules for T cell activation and dysfunction that are uncoupled at the single-cell level in tumor-infiltrating lymphocytes and shown that the zinc regulator proteins, metallothioneins (MT), promote T cell dysfunction. It has come to our attention that the single-cell RNA-seq data we showed in Figure 5 contains an error: out of the 8 WT plates (from 4 WT mice) and 8 MT-/- plates (from 5 MT-/- mice) sequenced and analyzed, one WT plate was switched with one MT-/- plate. We identified the problem when an independent group analyzed our deposited data and contacted us asking about unexpected results from those two plates. To ensure accuracy of all our deposited data, we have reconstructed the full-length transcripts from our single-cell RNA-seq and population RNA-seq data for the region of MT1 and MT2. It was necessary to do so as the MT-/- mouse was generated with the insertion of stop codons into both MT1 and MT2 exons (PMID: 8290567), yielding full-length transcripts. This endeavor unequivocally determined the genotype directly from the data, and it confirmed the switching of 1 WT plate with 1 MT-/- plate, while all other data we deposited has the correct genotype designation. We have now performed re-analysis of all affected figure panels (Figure 5) and found that this correction does not alter our findings. We have made the necessary changes including updating the text and figure to reflect this correction in the pdf version of the paper. We have also corrected the metadata deposited in GEO. In addition, we have corrected the STAR Methods to report the correct number of genes analyzed in the single-cell RNA-Seq experiments (7,955 genes, instead of the 9,863 genes reported in the original version), and to clarify that in this experiment we have included also cells from WT and MT-/- that expressed the pmel-1 transgene, that encodes the T cell receptor specific for the melanoma antigen pmel-17. The RNA-seq analyses yielded similar results regardless of whether the cells from the pmel-1 transgenic mice were included or not. We apologize for any inconvenience that these errors may have caused.Figure 5The Dysfunction and Activation Transcriptional Programs Are Negatively Correlated at the Single-Cell Level (Original)View Large Image Figure ViewerDownload Hi-res image Download (PPT) A Distinct Gene Module for Dysfunction Uncoupled from Activation in Tumor-Infiltrating T CellsSinger et al.CellSeptember 08, 2016In BriefSingle-cell profiling of tumor-infiltrating lymphocytes identifies critical regulators of T cell dysfunction in cancer, opening new avenues for targeting dysfunctional T cell states while leaving activation programs intact. Full-Text PDF Open Archive