A statistical framework for multiparameter analysis at the single-cell level

Phenotypic characterization of individual cells provides crucial insights into intercellular heterogeneity and enables access to information that is unavailable from ensemble averaged, bulk cell analyses. Single-cell studies have attracted significant interest in recent years and spurred the development of a variety of commercially available and research-grade technologies. To quantify cell-to-cell variability of cell populations, we have developed an experimental platform for real-time measurements of oxygen consumption (OC) kinetics at the single-cell level. Unique challenges inherent to these single-cell measurements arise, and no existing data analysis methodology is available to address them. Here we present a data processing and analysis method that addresses challenges encountered with this unique type of data in order to extract biologically relevant information. We applied the method to analyze OC profiles obtained with single cells of two different cell lines derived from metaplastic and dysplastic human Barrett's esophageal epithelium. In terms of method development, three main challenges were considered for this heterogeneous dynamic system: (i) high levels of noise, (ii) the lack of a priori knowledge of single-cell dynamics, and (iii) the role of intercellular variability within and across cell types. Several strategies and solutions to address each of these three challenges are presented. The features such as slopes, intercepts, breakpoint or change-point were extracted for every OC profile and compared across individual cells and cell types. The results demonstrated that the extracted features facilitated exposition of subtle differences between individual cells and their responses to cell-cell interactions. With minor modifications, this method can be used to process and analyze data from other acquisition and experimental modalities at the single-cell level, providing a valuable statistical framework for single-cell analysis.     

Proteomic screening of a cell line model of esophageal carcinogenesis identifies cathepsin D and aldo-keto reductase 1C2 and 1B10 dysregulation in Barrett's esophagus and esophageal adenocarcinoma

Esophageal adenocarcinoma (EA) incidence is increasing rapidly and is associated with a poor prognosis. Identifying biomarkers of disease development and progression would be invaluable tools to inform clinical practice. Two-dimensional polyacrylamide gel electrophoresis was used to screen 10 esophageal cell lines representing distinct stages in the development of esophageal cancer. Thirty-three proteins were identified by MALDI-TOF-MS which demonstrated differences in expression across the cell lines. Western blotting and qRT-PCR confirmed increased cathepsin D and aldo-keto reductases 1C2 and 1B10 expression in metaplastic and dysplastic cell lines. Expression of these proteins was further assessed in esophageal epithelium from patients with nonerosive (NERD) and erosive gastro-esophageal reflux disease, Barrett's esophagus (BE) and EA. When compared with normal epithelium of NERD patients, (i) cathepsin D mRNA levels demonstrated a stepwise increase in expression (p<0.05) in erosive, metaplastic and EA tissue; (ii) AKR1B10 expression increased (p<0.05) 3- and 9-fold in erosive and Barrett's epithelium, respectively; and (iii) AKR1C2 levels increased (p<0.05) in erosive and Barrett's epithelium, but were reduced (p<0.05) in EA. These proteins may contribute to disease development via effects on apoptosis, transport of bile acids and retinoid metabolism and should be considered as candidates for further mechanistic and clinical investigations.     

Computational Analysis of AMPK-Mediated Neuroprotection Suggests Acute Excitotoxic Bioenergetics and Glucose Dynamics Are Regulated by a Minimal Set of Critical Reactions

Loss of ionic homeostasis during excitotoxic stress depletes ATP levels and activates the AMP-activated protein kinase (AMPK), re-establishing energy production by increased expression of glucose transporters on the plasma membrane. Here, we develop a computational model to test whether this AMPK-mediated glucose import can rapidly restore ATP levels following a transient excitotoxic insult. We demonstrate that a highly compact model, comprising a minimal set of critical reactions, can closely resemble the rapid dynamics and cell-to-cell heterogeneity of ATP levels and AMPK activity, as confirmed by single-cell fluorescence microscopy in rat primary cerebellar neurons exposed to glutamate excitotoxicity. The model further correctly predicted an excitotoxicity-induced elevation of intracellular glucose, and well resembled the delayed recovery and cell-to-cell heterogeneity of experimentally measured glucose dynamics. The model also predicted necrotic bioenergetic collapse and altered calcium dynamics following more severe excitotoxic insults. In conclusion, our data suggest that a minimal set of critical reactions may determine the acute bioenergetic response to transient excitotoxicity and that an AMPK-mediated increase in intracellular glucose may be sufficient to rapidly recover ATP levels following an excitotoxic insult.     

DNA damage levels are raised in Barrett's oesophageal mucosa relative to the squamous epithelium of the oesophagus.

Barrett's oesophagus (BE) is a pre-malignant metaplastic tissue predisposing to oesophageal adenocarcinoma (EC), and gastro-oesophageal reflux is a risk factor for both conditions. Reflux of acid and bile can cause mucosal injury and initiate chronic inflammation. These processes can induce DNA damage, possibly via an oxidative stress mechanism, thus increasing the likelihood of progression from Barrett's metaplasia to dysplasia and finally carcinoma. The comet assay was optimized for the detection of DNA damage (strand breaks and alkali-labile sites) in oesophageal biopsies, including incorporation of the DNA repair enzyme Fapy-DNA glycosylase (Fpg). Fpg allows the detection of 8-hydroxy-2-deoxyguanosine (8-OHdG) sites, a known pro-mutagenic DNA lesion. BE patients were recruited from BE surveillance clinics and oesophageal biopsies collected at endoscopy. Comet analysis revealed significantly increased (p < 0.001) DNA damage in Barrett's epithelium compared with matched squamous epithelium, with median % tail DNA values of 25.1% (first to third quartile 21.7-29.6%) and 18.6% (first to third quartile 16.9-21.4%), respectively. The median % tail DNA was up to 70% higher in the matched BE tissue compared with squamous epithelium from the same patient. Fpg sensitive sites were demonstrated in both tissue types at similar levels. The raised level of DNA damage in the premalignant BE may contribute to the accumulation of genetic alterations occurring during progression to EC. Understanding these underlying mechanisms provides a basis for cancer prevention strategies in BE patients.     

DNA damage levels are raised in Barrett's oesophageal mucosa relative to the squamous epithelium of the oesophagus

Barrett's oesophagus (BE) is a pre-malignant metaplastic tissue predisposing to oesophageal adenocarcinoma (EC), and gastro-oesophageal reflux is a risk factor for both conditions. Reflux of acid and bile can cause mucosal injury and initiate chronic inflammation. These processes can induce DNA damage, possibly via an oxidative stress mechanism, thus increasing the likelihood of progression from Barrett's metaplasia to dysplasia and finally carcinoma. The comet assay was optimized for the detection of DNA damage (strand breaks and alkali-labile sites) in oesophageal biopsies, including incorporation of the DNA repair enzyme Fapy-DNA glycosylase (Fpg). Fpg allows the detection of 8-hydroxy-2-deoxyguanosine (8-OHdG) sites, a known pro-mutagenic DNA lesion. BE patients were recruited from BE surveillance clinics and oesophageal biopsies collected at endoscopy. Comet analysis revealed significantly increased (p<0.001) DNA damage in Barrett's epithelium compared with matched squamous epithelium, with median % tail DNA values of 25.1% (first to third quartile 21.7–29.6%) and 18.6% (first to third quartile 16.9–21.4%), respectively. The median % tail DNA was up to 70% higher in the matched BE tissue compared with squamous epithelium from the same patient. Fpg sensitive sites were demonstrated in both tissue types at similar levels. The raised level of DNA damage in the premalignant BE may contribute to the accumulation of genetic alterations occurring during progression to EC. Understanding these underlying mechanisms provides a basis for cancer prevention strategies in BE patients.     

Robust Inference of Cell-to-Cell Expression Variations from Single- and K-Cell Profiling

Quantifying heterogeneity in gene expression among single cells can reveal information inaccessible to cell-population averaged measurements. However, the expression level of many genes in single cells fall below the detection limit of even the most sensitive technologies currently available. One proposed approach to overcome this challenge is to measure random pools of k cells (e.g., 10) to increase sensitivity, followed by computational “deconvolution” of cellular heterogeneity parameters (CHPs), such as the biological variance of single-cell expression levels. Existing approaches infer CHPs using either single-cell or k-cell data alone, and typically within a single population of cells. However, integrating both single- and k-cell data may reap additional benefits, and quantifying differences in CHPs across cell populations or conditions could reveal novel biological information. Here we present a Bayesian approach that can utilize single-cell, k-cell, or both simultaneously to infer CHPs within a single condition or their differences across two conditions. Using simulated as well as experimentally generated single- and k-cell data, we found situations where each data type would offer advantages, but using both together can improve precision and better reconcile CHP information contained in single- and k-cell data. We illustrate the utility of our approach by applying it to jointly generated single- and k-cell data to reveal CHP differences in several key inflammatory genes between resting and inflammatory cytokine-activated human macrophages, delineating differences in the distribution of ‘ON’ versus ‘OFF’ cells and in continuous variation of expression level among cells. Our approach thus offers a practical and robust framework to assess and compare cellular heterogeneity within and across biological conditions using modern multiplexed technologies.     

Mitochondrial dysfunction accounts for the stochastic heterogeneity in telomere-dependent senescence

Aging is an inherently stochastic process, and its hallmark is heterogeneity between organisms, cell types, and clonal populations, even in identical environments. The replicative lifespan of primary human cells is telomere dependent; however, its heterogeneity is not understood. We show that mitochondrial superoxide production increases with replicative age in human fibroblasts despite an adaptive UCP-2–dependent mitochondrial uncoupling. This mitochondrial dysfunction is accompanied by compromised [Ca²⁺]_i homeostasis and other indicators of a retrograde response in senescent cells. Replicative senescence of human fibroblasts is delayed by mild mitochondrial uncoupling. Uncoupling reduces mitochondrial superoxide generation, slows down telomere shortening, and delays formation of telomeric γ-H2A.X foci. This indicates mitochondrial production of reactive oxygen species (ROS) as one of the causes of replicative senescence. By sorting early senescent (SES) cells from young proliferating fibroblast cultures, we show that SES cells have higher ROS levels, dysfunctional mitochondria, shorter telomeres, and telomeric γ-H2A.X foci. We propose that mitochondrial ROS is a major determinant of telomere-dependent senescence at the single-cell level that is responsible for cell-to-cell variation in replicative lifespan.     

Single-Cell Measurements of Enzyme Levels as a Predictive Tool for Cellular Fates during Organic Acid Production

Organic acids derived from engineered microbes can replace fossil-derived chemicals in many applications. Fungal hosts are preferred for organic acid production because they tolerate lignocellulosic hydrolysates and low pH, allowing economic production and recovery of the free acid. However, cell death caused by cytosolic acidification constrains productivity. Cytosolic acidification affects cells asynchronously, suggesting that there is an underlying cell-to-cell heterogeneity in acid productivity and/or in resistance to toxicity. We used fluorescence microscopy to investigate the relationship between enzyme concentration, cytosolic pH, and viability at the single-cell level in Saccharomyces cerevisiae engineered to synthesize xylonic acid. We found that cultures producing xylonic acid accumulate cells with cytosolic pH below 5 (referred to here as “acidified”). Using live-cell time courses, we found that the probability of acidification was related to the initial levels of xylose dehydrogenase and sharply increased from 0.2 to 0.8 with just a 60% increase in enzyme abundance (Hill coefficient, >6). This “switch-like” relationship likely results from an enzyme level threshold above which the produced acid overwhelms the cell''s pH buffering capacity. Consistent with this hypothesis, we showed that expression of xylose dehydrogenase from a chromosomal locus yields ∼20 times fewer acidified cells and ∼2-fold more xylonic acid relative to expression of the enzyme from a plasmid with variable copy number. These results suggest that strategies that further reduce cell-to-cell heterogeneity in enzyme levels could result in additional gains in xylonic acid productivity. Our results demonstrate a generalizable approach that takes advantage of the cell-to-cell variation of a clonal population to uncover causal relationships in the toxicity of engineered pathways.     

基于单细胞转录组的视网膜母细胞瘤进展特征分析

视网膜母细胞瘤（retinoblastoma,RB）是婴幼儿最常见的眼内恶性肿瘤。在RB进展过程中的关键致病因素目前尚不十分清楚。因此,识别与RB进展密切相关的基因能为病情诊断及基因治疗提供重要信息。然而,肿瘤组织具有很强的细胞异质性,不同病理状态下的细胞,其功能及基因表达都可能呈现显著的差异。本研究从公共基因表达数据库（gene expression omnibus,GEO）下载了1例4个月肿瘤患者和1例2年患者的肿瘤及癌旁组织的单细胞转录组测序数据,从单细胞水平解析不同患病时长的RB肿瘤转录图谱,鉴定与RB进展有潜在关联的细胞亚群及基因集。结果显示,肿瘤组织与癌旁组织在单细胞转录图谱上具有整体的一致性,但视锥前体G1期细胞群、G2期细胞群以及小胶质细胞群在肿瘤与癌旁组织中的分布比例存在明显差异。进一步分析了这3种细胞群在RB肿瘤进展过程中的作用。研究发现,在RB肿瘤的早期阶段,视锥前体细胞在G1期异常增殖,随着RB肿瘤的进展,视锥前体G2期细胞比例显著增加。同时,RB进展过程的小胶质细胞群差异分析结果显示,主要参与免疫应答的关键基因包括RPL23、B2M、HLA家族基因。本研究可为RB发病机制及进展研究提供更多新视角和数据资源。     

Cellular noise regulons underlie fluctuations in Saccharomyces cerevisiae

Stochasticity is a hallmark of cellular processes, and different classes of genes show large differences in their cell-to-cell variability (noise). To decipher the sources and consequences of this noise, we systematically measured pairwise correlations between large numbers of genes, including those with high variability. We find that there is substantial pathway variability shared across similarly regulated genes. This induces quantitative correlations in the expression of functionally related genes such as those involved in the Msn2/4 stress response pathway, amino-acid?biosynthesis, and mitochondrial maintenance. Bioinformatic analyses and genetic perturbations suggest that fluctuations in PKA and Tor signaling contribute to pathway-specific variability. Our results argue that a limited number of well-delineated "noise regulons" operate across a yeast cell and that such coordinated fluctuations enable a stochastic but coherent induction of functionally related genes. Finally, we show that pathway noise is a quantitative tool for exploring pathway features and regulatory relationships in un-stimulated systems.     

单细胞测序技术及其在乳腺癌研究中的应用

乳腺癌是起源于乳腺各级导管和乳腺上皮细胞,由增生到不典型增生而逐步发展成原位癌、早期浸润癌至浸润性癌的一种恶性肿瘤。传统高通量测序技术对乳腺癌的研究主要是鉴定与乳腺癌发生发展相关的"驱动基因",但是对于乳腺癌基因组结构变化以及亚克隆的鉴定等存在一定的局限性,并且忽略了乳腺癌肿瘤细胞之间的异质性。近年来兴起的单细胞测序技术是以单个细胞为研究对象,对基因拷贝和基因表达等数据进行分析,探究乳腺癌的细胞组成、细胞状态和细胞命运,绘制乳腺癌生态系统图谱,对临床患者进行精准分层,为实现个体化治疗提供支持。同时,还可以揭示乳腺癌与T细胞、巨噬细胞等免疫细胞间的相关性,为发现乳腺癌新的治疗靶点、免疫检查点等提供参考。本文对单细胞测序技术及其在乳腺癌研究中的应用和研究进展进行了综述,以期为单细胞测序技术的进一步发展提供参考,同时为理解乳腺癌的发病机制和免疫治疗提供基础支持。     

Medoidshift clustering applied to genomic bulk tumor data

Despite the enormous medical impact of cancers and intensive study of their biology, detailed characterization of tumor growth and development remains elusive. This difficulty occurs in large part because of enormous heterogeneity in the molecular mechanisms of cancer progression, both tumor-to-tumor and cell-to-cell in single tumors. Advances in genomic technologies, especially at the single-cell level, are improving the situation, but these approaches are held back by limitations of the biotechnologies for gathering genomic data from heterogeneous cell populations and the computational methods for making sense of those data. One popular way to gain the advantages of whole-genome methods without the cost of single-cell genomics has been the use of computational deconvolution (unmixing) methods to reconstruct clonal heterogeneity from bulk genomic data. These methods, too, are limited by the difficulty of inferring genomic profiles of rare or subtly varying clonal subpopulations from bulk data, a problem that can be computationally reduced to that of reconstructing the geometry of point clouds of tumor samples in a genome space. Here, we present a new method to improve that reconstruction by better identifying subspaces corresponding to tumors produced from mixtures of distinct combinations of clonal subpopulations. We develop a nonparametric clustering method based on medoidshift clustering for identifying subgroups of tumors expected to correspond to distinct trajectories of evolutionary progression. We show on synthetic and real tumor copy-number data that this new method substantially improves our ability to resolve discrete tumor subgroups, a key step in the process of accurately deconvolving tumor genomic data and inferring clonal heterogeneity from bulk data.