Cell Type-specific Gene Expression Profiles Research Articles

Semi-reference based cell type deconvolution with application to human metastatic cancers.

Bulk RNA-seq experiments, commonly used to discern gene expression changes across conditions, often neglect critical cell type-specific information due to their focus on average transcript abundance. Recognizing cell type contribution is crucial to understanding phenotype and disease variations. The advent of single-cell RNA sequencing has allowed detailed examination of cellular heterogeneity; however, the cost and analytic caveat prohibits such sequencing for a large number of samples. We introduce a novel deconvolution approach, SECRET, that employs cell type-specific gene expression profiles from single-cell RNA-seq to accurately estimate cell type proportions from bulk RNA-seq data. Notably, SECRET can adapt to scenarios where the cell type present in the bulk data is unrepresented in the reference, thereby offering increased flexibility in reference selection. SECRET has demonstrated superior accuracy compared to existing methods using synthetic data and has identified unknown tissue-specific cell types in real human metastatic cancers. Its versatility makes it broadly applicable across various human cancer studies.

Reconstruction of the tumor spatial microenvironment along the malignant-boundary-nonmalignant axis

Although advances in spatial transcriptomics (ST) enlarge to unveil spatial landscape of tissues, it remains challenging to delineate pathology-relevant and cellular localizations, and interactions exclusive to a spatial niche (e.g., tumor boundary). Here, we develop Cottrazm, integrating ST with hematoxylin and eosin histological image, and single-cell transcriptomics to delineate the tumor boundary connecting malignant and non-malignant cell spots in tumor tissues, deconvolute cell-type composition at spatial location, and reconstruct cell type-specific gene expression profiles at sub-spot level. We validate the performance of Cottrazm along the malignant-boundary-nonmalignant spatial axis. We identify specific macrophage and fibroblast subtypes localized around tumor boundary that interacted with tumor cells to generate a structural boundary, which limits T cell infiltration and promotes immune exclusion in tumor microenvironment. In this work, Cottrazm provides an integrated tool framework to dissect the tumor spatial microenvironment and facilitates the discovery of functional biological insights, thereby identifying therapeutic targets in oncologic ST datasets.

Genomic glucocorticoid action in embryonic mouse neural stem cells

Prenatal exposure to synthetic glucocorticoids (sGCs) reprograms brain development and predisposes the developing fetus towards potential adverse neurodevelopmental outcomes. Using a mouse model of sGC administration, previous studies show that these changes are accompanied by sexually dimorphic alterations in the transcriptome of neural stem and progenitor cells (NSPCs) derived from the embryonic telencephalon. Because cell type-specific gene expression profiles tightly regulate cell fate decisions and are controlled by a flexible landscape of chromatin domains upon which transcription factors and enhancer elements act, we multiplexed data from four genome-wide assays: RNA-seq, ATAC-seq (assay for transposase accessible chromatin followed by genome wide sequencing), dual cross-linking ChIP-seq (chromatin immunoprecipitation followed by genome wide sequencing), and microarray gene expression to identify novel relationships between gene regulation, chromatin structure, and genomic glucocorticoid receptor (GR) action in NSPCs. These data reveal that GR binds preferentially to predetermined regions of accessible chromatin to influence gene programming and cell fate decisions. In addition, we identify SOX2 as a transcription factor that impacts the genomic response of select GR target genes to sGCs (i.e., dexamethasone) in NSPCs.

PATH-30. BIOLOGIC AND CLINICAL SIGNIFICANCE OF NEOPLASTIC AND IMMUNE CELL STATES IN GROUP3 AND GROUP 4 MEDULLOBLASTOMA

Abstract Medulloblastoma (MB) is a heterogeneous malignant childhood brain cancer, classified into four molecular subgroups, WNT, SHH, Group3, and Group4, based on transcriptional, genetic, and clinical features. We aimed to understand the tumor microenvironment of Group 3 and Group 4 tumors utilizing single-cell transcriptome data. We identified four neoplastic cell states, including stemness (TNF-α), proliferative, mesenchymal and quiescence, and 11 broad lymphoid and myeloid cell types from single-cell transcriptomes. We applied CIBERSORTx to enumerate cell type proportion and CODEFACS on 73 bulk MB RNAseq profiles to determine cell type-specific abundance and gene expression profiles. We noted significant associations (both positive and negative) between the abundance of neoplastic cell states and microenvironmental cell types in both group3 and group4 MB. For example, the invasive cell state negatively correlated with dendritic cells in Group3 and Group4 tumors, but the invasive cell state was oppositely correlated with a macrophage subset in group 3 versus group 4 tumors, Survival analysis showed a significant association of higher proliferative cell state with poorer survival in Group3 MB while in group4 elevated invasive cell state, as well as elevated α, NK, and CD4-T cell abundance showed trends for poorer prognosis. Using Ecotyper, we determined cellular ecotypes based on cell type-specific transcriptome states. An ecotype containing B cell, hematopoietic, dendritic, and proliferative cell states was associated with poor prognosis in group3 and group4 tumors. We further analyzed ligand-receptor interaction between cell type states of poor prognosis ecotype; and found several interactions between ligands from immune cells (SEMA4G, LAMB2) and receptors on neoplastic cells (PLXNB2). Conversely, extracellular matrix ligands on neoplastic cells (COL1A1/2 and FN1) were found on this analysis potentially associated with integrin receptors on immune cells. Overall, our study highlights the potential biologic and clinical significance of neoplastic and immune cell states in group3/4 MB.

SCADIE: simultaneous estimation of cell type proportions and cell type-specific gene expressions using SCAD-based iterative estimating procedure

A challenge in bulk gene differential expression analysis is to differentiate changes due to cell type-specific gene expression and cell type proportions. SCADIE is an iterative algorithm that simultaneously estimates cell type-specific gene expression profiles and cell type proportions, and performs cell type-specific differential expression analysis at the group level. Through its unique penalty and objective function, SCADIE more accurately identifies cell type-specific differentially expressed genes than existing methods, including those that may be missed from single cell RNA-Seq data. SCADIE has robust performance with respect to the choice of deconvolution methods and the sources and quality of input data.

DNA Methyltransferase 3b Accelerates the Process of Atherosclerosis.

Background DNA methylation plays a key role in establishing cell type–specific gene expression profiles and patterns in atherosclerosis. The underlying mechanism remains unclear. Previous studies have shown that DNA methyltransferase 3b (DNMT3b) may play an important role in atherosclerosis. This study aimed to establish the regulatory role of DNMT3b in the development of atherosclerosis. Methods We constructed a viral vector carrying Dnmt3b shRNA to transduce ApoE−/− mice. Meanwhile, healthy human peripheral blood Treg cells were treated with inhibitor of DNMT3b (AZA and EGCG) or transduced with DNMT3b shRNA. Results It showed that Dnmt3b silencing attenuated atherosclerosis, including decreased lesion size and macrophage content and increased collagen and smooth muscle cells content in ApoE−/− mice. To further investigate the possible mechanisms, combined with previous studies by our group, we showed that Foxp3-TSDR methylation level was significantly reduced Foxp3 expression and peripheral blood Treg levels were significantly increased by Dnmt3b shRNA vector transduction in animals committed to western diet for 12 and 18 weeks. Consistently, inhibition of DNMT3b (AZA and EGCG) decreased the expression levels of DNMT3b, which can increase the expression levels of FOXP3, and increase the levels of TGF-β and IL-10 and decrease the levels of IL-β and IFN-γ. After transduction with DNMT3b shRNA, the effect was more obvious. Conclusions DNMT3b accelerated atherosclerosis, and may be associated with FOXP3 hypermethylation status in regulatory T cells.

Immune cell gene expression signatures in diffuse glioma are associated with IDH mutation status, patient outcome and malignant cell state, and highlight the importance of specific cell subsets in glioma biology

The tumor micro-environment (TME) plays an important role in various cancers, including gliomas. We estimated immune cell type-specific gene expression profiles in 3 large clinically annotated glioma datasets using CIBERSORTx and LM22/LM10 blood-based immune signatures and found that the proportions and estimated gene expression patterns of specific immune cells significantly varied according to IDH mutation status. When IDH-WT and IDH-MUT tumors were considered separately, cluster-of-cluster analyses of immune cell gene expression identified groups with distinct survival outcomes. We confirmed and extended these findings by applying a signature matrix derived from single-cell RNA-sequencing data derived from 19 glioma tumor samples to the bulk profiling data, validating findings from the LM22/LM10 results. To link immune cell signatures with outcomes in checkpoint therapy, we then showed a significant association of monocytic lineage cell gene expression clusters with patient survival and with mesenchymal gene expression scores. Integrating immune cell-based gene expression with previously described malignant cell states in glioma demonstrated that macrophage M0 abundance significantly correlated with mesenchymal state in IDH-WT gliomas, with evidence of a previously implicated role of the Oncostatin-M receptor and macrophages in the mesenchymal state. Among IDH-WT tumors that were enriched for the mesenchymal cell state, the estimated M0 macrophage expression signature coordinately also trended to a mesenchymal signature. We also examined IDH-MUT tumors stratified by 1p/19q status, showing that a mesenchymal gene expression signature the M0 macrophage fraction was enriched in IDH-MUT, non-codeleted tumors. Overall, these results highlight the biological and clinical significance of the immune cell environment related to IDH mutation status, patient prognosis and the mesenchymal state in diffuse gliomas.

Deconvolving Clinically Relevant Cellular Immune Cross-talk from Bulk Gene Expression Using CODEFACS and LIRICS Stratifies Patients with Melanoma to Anti-PD-1 Therapy.

This work presents two new computational methods that can deconvolve a large collection of bulk tumor gene expression profiles into their respective cell type-specific gene expression profiles and identify cell type-specific ligand-receptor interactions predictive of response to immune-checkpoint blockade therapy. This article is highlighted in the In This Issue feature, p. 873.

Integrative analysis of the 3D genome structure reveals that CTCF maintains the properties of mouse female germline stem cells

The three-dimensional configuration of the genome ensures cell type-specific gene expression profiles by placing genes and regulatory elements in close spatial proximity. Here, we used in situ high-throughput chromosome conformation (in situ Hi-C), RNA sequencing (RNA-seq) and chromatin immunoprecipitation sequencing (ChIP-seq) to characterize the high-order chromatin structure signature of female germline stem cells (FGSCs) and identify its regulating key factor based on the data-driven of multiple omics data. By comparison with pluripotent stem cells (PSCs), adult stem cells (ASCs), and somatic cells at three major levels of chromatin architecture, A/B compartments, topologically associating domains, and chromatin loops, the chromatin architecture of FGSCs was most similar to that of other ASCs and largely different from that of PSCs and somatic cells. After integrative analysis of the three-dimensional chromatin structure, active compartment-associating loops (aCALs) were identified as a signature of high-order chromatin organization in FGSCs, which revealed that CCCTC-binding factor was a major factor to maintain the properties of FGSCs through regulation of aCALs. We found FGSCs belong to ASCs at chromatin structure level and characterized aCALs as the high-order chromatin structure signature of FGSCs. Furthermore, CTCF was identified to play a key role in regulating aCALS to maintain the biological functions of FGSCs. These data provide a valuable resource for future studies of the features of chromatin organization in mammalian stem cells and further understanding of the fundamental characteristics of FGSCs.

Exploring the Diverse Functional and Regulatory Consequences of Alternative Splicing in Development and Disease.

Alternative splicing is a fundamental mechanism of eukaryotic RNA regulation that increases the transcriptomic and proteomic complexity within an organism. Moreover, alternative splicing provides a framework for generating unique yet complex tissue- and cell type-specific gene expression profiles, despite using a limited number of genes. Recent efforts to understand the negative consequences of aberrant splicing have increased our understanding of developmental and neurodegenerative diseases such as spinal muscular atrophy, frontotemporal dementia and Parkinsonism linked to chromosome 17, myotonic dystrophy, and amyotrophic lateral sclerosis. Moreover, these studies have led to the development of innovative therapeutic treatments for diseases caused by aberrant splicing, also known as spliceopathies. Despite this, a paucity of information exists on the physiological roles and specific functions of distinct transcript spliceforms for a given gene. Here, we will highlight work that has specifically explored the distinct functions of protein-coding spliceforms during development. Moreover, we will discuss the use of alternative splicing of noncoding exons to regulate the stability and localization of RNA transcripts.

Gene expression profiling of epidermal cell types in C. elegans using Targeted DamID.

The epidermis of Caenorhabditis elegans is an essential tissue for survival as it contributes to the formation of the cuticle barrier, as well as facilitates developmental progression and animal growth. Most of the epidermis consists of the hyp7 hypodermal syncytium, the nuclei of which are largely generated by the seam cells that exhibit stem cell-like behaviour during development. How the seam cell progenitors differ transcriptionally from the differentiated hypodermis is poorly understood. Here, we introduce Targeted DamID (TaDa) in C. elegans as a method for identifying genes expressed within a tissue of interest without cell isolation. We show that TaDa signal enrichment profiles can be used to identify genes transcribed in the epidermis and use this method to resolve differences in gene expression between the seam cells and the hypodermis. We finally predict and functionally validate new transcription and chromatin factors acting in seam cell development. These findings provide insights into cell-type-specific gene expression profiles likely associated with epidermal cell fate patterning.

Cell type diversity in scallop adductor muscles revealed by single-cell RNA-Seq

Studies on cell atlas in marine invertebrates provide a better understanding of cell types, stem cell maintenance, and lineages of cell differentiation. To investigate the molecular features of various cell types in molluscan muscles, we performed single-cell RNA sequencing (scRNA-seq) to map cell types in scallop adductor muscles. We uncovered the cell type-specific features of 20 cell clusters defined by the expression of multiple specific molecular markers. These cell clusters are mainly classified into four broad classes, including mesenchymal stem cells, muscle cells, neurons, and haemolymph cells. In particular, we identified a diverse repertoire of neurons in the striated adductor muscle, but not in the smooth muscle. We further reconstructed the cell differentiation events using all the cell clusters by single-cell pseudotemporal trajectories. By integrating dual BrdU-PCNA immunodetection, neuron-specific staining and electron microscopy observation, we showed the spatial distribution of mesenchymal stem cells and neurons in striated adductor muscle of scallops. The present findings will not only be useful to address the cell type-specific gene expression profiles in scallop muscles, but also provide valuable resources for cross-species comparison of marine organisms.

Abstract 3443: Atlas of clinically-distinct cell states and cellular ecosystems across human solid tumors

Abstract Tumors are complex ecosystems consisting of malignant, immune, and stromal elements whose dynamic interactions drive patient survival and response to therapy. A comprehensive understanding of the diversity of cellular states within the tumor microenvironment (TME), and their patterns of co-occurrence, could provide new diagnostic tools for improved disease management and novel targets for therapeutic intervention. To address this challenge, we developed EcoTyper, a novel machine learning framework for large-scale identification of TME cell states and their co-association patterns from bulk, single-cell, and spatially resolved tumor expression data. EcoTyper starts by “purifying” cell type-specific gene expression profiles of epithelial cells, immune, and stromal cell types from bulk tissue transcriptomes using CIBERSORTx (Newman et al., Nat Biotechnol 2019). It then identifies transcriptional states for each cell type, validates them in scRNA-seq data, and uncovers co-occurrence patterns between cell states in order to define tumor cellular ecosystems. Applied to 6,475 tumor and adjacent normal samples from solid tumor types profiled by The Cancer Genome Atlas (TCGA), EcoTyper identified robust transcriptional states across 12 major cell types, including epithelial, fibroblast, endothelial, and 9 immune subsets. These states included both known and novel cellular phenotypes, nearly all of which could be validated in a compendium of scRNA-seq tumor atlases spanning ~140,000 cells. Most cell states were specific to neoplastic tissue, ubiquitous across tumor types, and significantly associated with overall survival, both in TCGA and in 9,062 held-out tumor specimens (Gentles/Newman et al., Nat Medicine 2015). We found that specific cell states co-occur in distinct cellular communities with characteristic patterns of ligand-receptor interactions, genomic features, clinical outcomes, and spatial organization. One such ecosystem defined a normal-like state that was strongly enriched in non-malignant samples. Others delineated novel pro- and anti-tumor inflammatory environments involving specific fibroblast, endothelial, and immune cell transcriptional programs. In summary, large-scale deconvolution of cell type-specific transcriptomes across thousands of solid tumors revealed a comprehensive atlas of TME cell states and cellular ecosystems. Our results provide a high-resolution portrait of cellular heterogeneity in the TME across multiple solid tumor types, with implications for novel diagnostics and immunotherapeutic targets.

SCDC: bulk gene expression deconvolution by multiple single-cell RNA sequencing references.

Recent advances in single-cell RNA sequencing (scRNA-seq) enable characterization of transcriptomic profiles with single-cell resolution and circumvent averaging artifacts associated with traditional bulk RNA sequencing (RNA-seq) data. Here, we propose SCDC, a deconvolution method for bulk RNA-seq that leverages cell-type specific gene expression profiles from multiple scRNA-seq reference datasets. SCDC adopts an ENSEMBLE method to integrate deconvolution results from different scRNA-seq datasets that are produced in different laboratories and at different times, implicitly addressing the problem of batch-effect confounding. SCDC is benchmarked against existing methods using both in silico generated pseudo-bulk samples and experimentally mixed cell lines, whose known cell-type compositions serve as ground truths. We show that SCDC outperforms existing methods with improved accuracy of cell-type decomposition under both settings. To illustrate how the ENSEMBLE framework performs in complex tissues under different scenarios, we further apply our method to a human pancreatic islet dataset and a mouse mammary gland dataset. SCDC returns results that are more consistent with experimental designs and that reproduce more significant associations between cell-type proportions and measured phenotypes.

Sequencing the Human Brain at Single-Cell Resolution

This review summarizes some of the most recent studies on single-cell resolution sequencing of the post-mortem human brain and the application of these techniques for the study of psychiatric and neurological disorders. Over the last several years, researchers have optimized single-cell transcriptome and genome sequencing in post-mortem human brain tissue. This has given us unprecedented access to cell-type specific gene expression profiles and somatic mutations unique to pathological states. Additionally, we can now measure epigenetic information from individual brain cells and with advanced statistical approaches, we can focus on the key cell-types underlying psychiatric and other brain phenotypes. A new era of cell-type specific and single-cell resolution studies of the human brain is underway. With proper application of rapidly advancing laboratory and data analysis techniques, we should witness important advances in our understanding of molecular changes associated with psychiatric and neurological phenotypes.

A Microglial Signature Directing Human Aging and Neurodegeneration-Related Gene Networks.

Aging is regarded as a major risk factor for neurodegenerative diseases. Thus, a better understanding of the similarities between the aging process and neurodegenerative diseases at the cellular and molecular level may reveal better understanding of this detrimental relationship. In the present study, we mined publicly available gene expression datasets from healthy individuals and patients affected by neurodegenerative diseases (Alzheimer’s disease, Parkinson’s disease, and Huntington’s disease) across a broad age spectrum and compared those with mouse aging and mouse cell-type specific gene expression profiles. We performed weighted gene co-expression network analysis (WGCNA) and found a gene network strongly related with both aging and neurodegenerative diseases. This network was significantly enriched with a microglial signature as imputed from cell type-specific sequencing data. Since mouse models are extensively used for the study of human diseases, we further compared these human gene regulatory networks with age-specific mouse brain transcriptomes. We discovered significantly preserved networks with both human aging and human disease and identified 17 shared genes in the top-ranked immune/microglia module, among which we found five human hub genes TYROBP, FCER1G, ITGB2, MYO1F, PTPRC, and two mouse hub genes Trem2 and C1qa. Taken together, these results support the hypothesis that microglia are key players involved in human aging and neurodegenerative diseases, and suggest that mouse models should be appropriate for studying these microglial changes in human.

Altered social behavior in mice carrying a cortical Foxp2 deletion.

Genetic disruptions of the forkhead box transcription factor FOXP2 in humans cause an autosomal-dominant speech and language disorder. While FOXP2 expression pattern are highly conserved, its role in specific brain areas for mammalian social behaviors remains largely unknown. Here we studied mice carrying a homozygous cortical Foxp2 deletion. The postnatal development and gross morphological architecture of mutant mice was indistinguishable from wildtype (WT) littermates. Unbiased behavioral profiling of adult mice revealed abnormalities in approach behavior towards conspecifics as well as in the reciprocal responses of WT interaction partners. Furthermore mutant mice showed alterations in acoustical parameters of ultrasonic vocalizations, which also differed in function of the social context. Cell type-specific gene expression profiling of cortical pyramidal neurons revealed aberrant regulation of genes involved in social behavior. In particular Foxp2 mutants showed the downregulation of Mint2 (Apba2), a gene involved in approach behavior in mice and autism spectrum disorder in humans. Taken together these data demonstrate that cortical Foxp2 is required for normal social behaviors in mice.

Incidental Diagnosis of Parathyroid Lesions by Preoperative Use of Next-Generation Molecular Testing.

Parathyroid glands can mimic thyroid follicular lesions on fine-needle aspiration (FNA) cytology and can lead to unnecessary or incorrect surgery. Newer molecular panel tests using next-generation sequencing (NGS) include analysis of cell type-specific gene expression profiles such as parathyroid. The study aim is to determine the frequency and clinical implications of parathyroid tissue identification by molecular testing in cytologically indeterminate "thyroid" lesions. Molecular analysis of indeterminate thyroid FNA specimens is obtained routinely and relies on amplification-based NGS inclusive of PTH-specific expression profiles. For this study, we retrospectively examined the clinical data and management of patients with molecular results positive for PTH expression from May 2014 until May 2016. Among 4765 consecutive patients with indeterminate cytology for a presumed thyroid nodule, NGS instead indicated a parathyroid lesion in 20 patients (0.42%). The clinical data of 15 patients were available, and the subsequent clinical management was altered in 93% (14/15 patients), including five (33%) eucalcemic patients who could avoid unnecessary surgery. Primary hyperparathyroidism was not suspected in seven patients until the molecular analysis results, and primary hyperparathyroidism was diagnosed in one (14%). During parathyroid exploration, most patients (6/8, 75%) required concurrent thyroidectomy or lobectomy, but thyroid preservation was still possible in two patients. A parathyroid gland was histologically confirmed in 89%. In 0.42% of patients with indeterminate cytology results, next-generation molecular results will indicate the presence of a parathyroid lesion. When this occurs, it is accurate and can robustly impact clinical management (93%).

Identifying Differentially Expressed Genes Using Fluorescence-Activated Cell Sorting (FACS) and RNA Sequencing from Low Input Samples.

Cell type-specific gene expression profiles are useful for understanding genes that are important for the development of different tissues and organs. Here, we describe how to perform fluorescence-activated cell sorting (FACS) on Arabidopsis root protoplasts to isolate specific cell types in the root. We then detail how to extract and process RNA from a very low number of cells (≥40 cells) for RNA sequencing (RNA seq). Finally, we describe how to process RNA seq data using TopHat and how to identify differentially expressed genes using PoissonSeq.

Estimation of immune cell content in tumour tissue using single-cell RNA-seq data

As interactions between the immune system and tumour cells are governed by a complex network of cell–cell interactions, knowing the specific immune cell composition of a solid tumour may be essential to predict a patient’s response to immunotherapy. Here, we analyse in depth how to derive the cellular composition of a solid tumour from bulk gene expression data by mathematical deconvolution, using indication-specific and cell type-specific reference gene expression profiles (RGEPs) from tumour-derived single-cell RNA sequencing data. We demonstrate that tumour-derived RGEPs are essential for the successful deconvolution and that RGEPs from peripheral blood are insufficient. We distinguish nine major cell types, as well as three T cell subtypes. Using the tumour-derived RGEPs, we can estimate the content of many tumours associated immune and stromal cell types, their therapeutically relevant ratios, as well as an improved gene expression profile of the malignant cells.