Single-cell and single-nucleus RNA-sequencing from paired normal-adenocarcinoma lung samples provide both common and discordant biological insights
Sebastien Renaut, Victoria Saavedra Armero, Dominique K. Boudreau, Nathalie Gaudreault, Patrice Desmeules, Sébastien Thériault, Patrick Mathieu, Philippe Joubert, Yohan Bosse
Abstract
Whether single-cell RNA-sequencing (scRNA-seq) captures the same biological information as single-nucleus RNA-sequencing (snRNA-seq) remains uncertain and likely to be context-dependent. Herein, a head-to-head comparison was performed in matched normal-adenocarcinoma human lung samples to assess biological insights derived from scRNA-seq versus snRNA-seq and better understand the cellular transition that occurs from normal to tumoral tissue. Here, the transcriptome of 160,621 cells/nuclei was obtained. In non-tumor lung, cell type proportions varied widely between scRNA-seq and snRNA-seq with a predominance of immune cells in the former (81.5%) and epithelial cells (69.9%) in the later. Similar results were observed in adenocarcinomas, in addition to an overall increase in cell type heterogeneity and a greater prevalence of copy number variants in cells of epithelial origin, which suggests malignant assignment. The cell type transition that occurs from normal lung tissue to adenocarcinoma was not always concordant whether cells or nuclei were examined.
Introduction
Single-cell sequencing (scRNA-seq) has the ability to inspect the cellular heterogeneity of tissue and cancer with unprecedented details, and as such provides important insights into the cellular origin and cell-specific molecular defects that play a role in disease pathogenesis [1–4]. However, given the pace at which the field is evolving, uncertainties remain with respect to the design and analysis of single-cell transcriptomic datasets in order to gain the most from biological samples. Fresh biospecimens are generally prioritized for cell viability and greater yield of high-quality cells. For tissues, scRNA-seq requires disaggregating the tissue to release individual cells into a single-cell suspension. Differences in dissociation and sample preparation efficiency across cell types are known to affect RNA integrity and can skew cell type proportions. A well-known instance of dissociation bias is observed in human lung tissue, where dissociation of fresh tumor (biopsies or resected specimens) commonly results in a majority of immune cells being sequenced [5–7]. While the aforementioned cell-type dissociation bias can be partly alleviated by enriching the epithelial cell fraction using EPCAM-based cell sorting [6], single cell preparation protocols may also affect cell viability and introduce transcriptional signatures associated with dissociation and stress responses [6,8,9].
Materials and methods
Ethics statement
All patients provided written informed consent, and the ethics committee of the IUCPQ-UL approved the study.
Patients and samples
Lung samples were collected from four patients that underwent curative intent primary lung cancer surgery at the Institut universitaire de cardiologie et de pneumologie de Québec–Université Laval (IUCPQ-UL) in 2021–2023, henceforth referred to patient 1, 2, 3 and 4. The four patients were self-reported white French Canadian (European ancestry) with no prior chemotherapy and/or radiation therapy, and all patients were between the age of 59 and 69, former smokers with adenocarcinomas (See Fig 1 for overview of experimental design, and Table A in S1 Text for detailed clinical characteristics of patients).
Results
Experimental design
Four patients, two tissue types (Normal/Tumor) and three experimental methods (scRNA-seq, snRNA-seq & immune-depleted scRNA-seq, hereafter labelled as Cell, Nucleus and Immune-depleted cell) were processed for a total of twenty-four samples. The experimental design is presented in Fig 1. The four patients underwent lung cancer surgery with pathologically confirmed LUAD (Fig 1A). The clinical characteristics of patients are detailed in Table A in S1 Text. Both LUAD and normal lung specimens were obtained from each patient (Fig 1B). Fresh tissues were immediately processed for scRNA-seq and adjacent samples were flashed frozen and stored at -80°C until further processing for snRNA-seq (Fig 1C). The single cell suspensions dissociated from fresh tissues were also submitted to CD45+ immune depletion, leading to three cell suspensions per specimen and thus six per patient (Fig 1D). The characteristics of samples and cell/nucleus suspensions are presented in Table B in S1 Text. Single cell suspensions were converted to libraries using the 10x Genomics workflow (Fig 1E) and sequenced on an Illumina NextSeq 2000 aiming for ~10,000 cells or nuclei per sample (Fig 1F). We partitioned the analysis by focusing on 1) normal lung tissues, 2) LUAD tissues, 3) paired normal-adenocarcinoma lung samples, and 4) immune-depleted samples (Fig 1G).
Discussion
In this study we generated a dataset of 160,621 cells/nuclei showing commonalities and discordances in biological insights derived from single-cell and single-nucleus RNA-sequencing of paired normal-adenocarcinoma human lung specimens. A distinct portrait of cellular composition was observed per experimental methods that favors scRNA-seq of fresh samples to map the immune landscape of lung adenocarcinoma. On the other hand, snRNA-seq of frozen samples surpassed the relative merits of scRNA-seq to obtain a dataset with cell type proportion that match tissue content and to provide a more cost-effective approach for research applications necessitating a higher number of epithelial and cancer cells (see Table I in S1 Text for a summary of the benefits of each method). In these paired lung samples, we identified gene expression and cell type transitions from normal to tumoral tissue that were not always concordant whether cells or nuclei were examined. The most striking difference was the ligand-receptor interactions that varied more across methods (cells vs. nuclei) rather than tissue types (Normal vs. Tumor). Immune cell depletion partly alleviated some of the differences in cell type composition between cells and nuclei, but at the detriment of inducing a stress response and affecting the transcriptome biological signal.
Acknowledgments
The authors would like to thank the research staff at the IUCPQ biobank for their valuable assistance.
Citation: Renaut S, Saavedra Armero V, Boudreau DK, Gaudreault N, Desmeules P, Thériault S, et al. (2024) Single-cell and single-nucleus RNA-sequencing from paired normal-adenocarcinoma lung samples provide both common and discordant biological insights. PLoS Genet 20(5): e1011301. https://doi.org/10.1371/journal.pgen.1011301
Editor: Yan Tang, Brigham and Women’s Hospital Department of Medicine, UNITED STATES
Received: December 13, 2023; Accepted: May 13, 2024; Published: May 30, 2024
Copyright: © 2024 Renaut et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The datasets generated by Cellranger are available as open-access downloadable files on Zenodo (10.5281/zenodo.11205626). All analytical codes used to produce the results of this study are available at https://github.com/Yohan-Bosse-Lab/scRNA.
Funding: This work was supported by the IUCPQ Foundation and a generous donation from Mr. Normand Lord. P.M. is the recipient of the Joseph C. Edwards Foundation granted to Université Laval. P.J. is the recipient of a Junior 2 Clinical Research Scholar award from the Fonds de recherche Québec - Santé (FRQS). Y.B. holds a Canada Research Chair in Genomics of Heart and Lung Diseases. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1011301#abstract0
