scholarly journals Single cell RNA sequencing of calvarial and long bone endocortical cells

2019 ◽  
Author(s):  
Ugur M. Ayturk ◽  
Joseph P. Scollan ◽  
Alexander Vesprey ◽  
Christina M. Jacobsen ◽  
Paola Divieti Pajevic ◽  
...  
Keyword(s):  
Gene Expression ◽  
Single Cell ◽  
Cultured Cells ◽  
Neutralizing Antibody ◽  
Long Bone ◽  
Appendicular Skeleton ◽  
Rna Seq ◽  
Isolated Cells

ABSTRACTSingle cell RNA-seq (scRNA-seq) is emerging as a powerful technology to examine transcriptomes of individual cells. We determined whether scRNA-seq could be used to detect the effect of environmental and pharmacologic perturbations on osteoblasts. We began with a commonly used in vitro system in which freshly isolated neonatal mouse calvarial cells are expanded and induced to produce a mineralized matrix. We used scRNA-seq to compare the relative cell type abundances and the transcriptomes of freshly isolated cells to those that had been cultured for 12 days in vitro. We observed that the percentage of macrophage-like cells increased from 6% in freshly isolated calvarial cells to 34% in cultured cells. We also found that Bglap transcripts were abundant in freshly isolated osteoblasts but nearly undetectable in the cultured calvarial cells. Thus, scRNA-seq revealed significant differences between heterogeneity of cells in vivo and in vitro. We next performed scRNA-seq on freshly recovered long bone endocortical cells from mice that received either vehicle or Sclerostin-neutralizing antibody for 1 week. Bone anabolism-associated transcripts were also not significantly increased in immature and mature osteoblasts recovered from Sclerostin-neutralizing antibody treated mice; this is likely a consequence of being underpowered to detect modest changes in gene expression, since only 7% of the sequenced endocortical cells were osteoblasts, and a limited portion of their transcriptomes were sampled. We conclude that scRNA-seq can detect changes in cell abundance, identity, and gene expression in skeletally derived cells. In order to detect modest changes in osteoblast gene expression at the single cell level in the appendicular skeleton, larger numbers of osteoblasts from endocortical bone are required.

scholarly journals Hematopoietic Stem Cells Respond Heterogeneously to Inflammatory Signals in an Age-Dependent Manner

Blood ◽  
2016 ◽  
Vol 128 (22) ◽  
pp. 566-566
Author(s):  
Arnav Mehta ◽  
Mati Mann ◽  
Monika Kowalczyk ◽  
Carl de Boer ◽  
Jun Wang ◽  
...  
Keyword(s):  
Gene Expression ◽  
Single Cell ◽  
Immune Cell ◽  
Transcriptional Regulators ◽  
Rna Seq ◽  
Hematopoietic Stem ◽  
Aged Mice ◽  
Inflammatory Signals

Abstract Hematopoietic stem cells (HSCs) have the unique responsibility to produce balanced immune cell output throughout an organism's life. Importantly, they must do so robustly despite a plethora of external stress, including frequent inflammatory challenge. With age, the accumulation of these stresses leads to impaired HSC function and myeloid-biased output. Aged HSCs are also more prone to pathological hematopoiesis, such as myeloproliferative disorder, leukemia and autoimmune diseases. However, little is known about the subcellular mechanisms that govern the inflammatory response of HSCs with age, which in turn might contribute to pathologic transformation. We show that young hematopoietic stem and progenitor cells (HSPCs) demonstrate a robust transcriptional response to toll-like receptor (TLR) ligands. Interestingly, this response is similar to that of mature immune cell types such as dendritic cells. Using single-cell proteomic assays, we found that young HSPCs secrete a diverse array of myeloid and lymphoid cytokines. However, when challenged with TLR ligands in vivo, young mice acutely increase myeloid-biased output but return rapidly to baseline hematopoietic output of both lymphoid and myeloid cells. Moreover, inflammatory challenge of young long-term HSCs in vitro did not perturb the function and output of these cells in bone marrow reconstitution experiments. In contrast to their counterparts from young mice, we found HSPCs obtained from aged mice have a diminished ability to secrete cytokines in response to TLR ligands. Furthermore, they secrete a homogenous subset of myeloid-biased cytokines. When challenged with TLR ligands in vivo, aged mice acutely increased myeloid output and maintain elevated myeloid output for several months implying memory of the inflammatory challenge. Consistent with this, we also found that pre-stimulation of aged HSCs prior to bone marrow transplant results in a sustained increase in myeloid output compared to unstimulated aged HSCs. To elucidate the differential heterogeneity between young and aged HSPCs in response to TLR signaling, we next performed single-cell RNA sequencing (RNA-seq) experiments. We found that the sustained myeloid output in aged mice after TLR stimulation is largely due to expansion of a myeloid-biased HSC subset in the aged HSC pool. By characterizing the gene expression networks that define these myeloid-biased HSCs under stimulation conditions, we were then able to identify a myeloid-biased HSC subset in both the unperturbed young and aged HSC pools. Moreover, we found that these cells are more abundant in aged mice at steady-state, and that these HSCs demonstrate a unique response to inflammatory challenge. We further identify putative transcriptional regulators, including Klf4, Klf5, Ikzf1 and Stat3, among others, that define gene expression in these myeloid-biased HSCs. We further show that loss of function of these factors can differentially alter myeloid output in young and aged mice both in vitro and in vivo. Our results demonstrate that there is a differential response of young and aged HSCs to inflammatory signals. Using single-cell RNA-seq and protein secretion studies, we elucidate the molecular heterogeneity of the HSC pool at steady state and with TLR stimulation. By resolving heterogeneous subsets of cells in both the young and aged HSC pool, and by uncovering the transcriptional regulators that influence their function, we thus propose a new model of inflammatory hematopoiesis that may have implications to understanding age-related defects in immune development. Disclosures No relevant conflicts of interest to declare.

scholarly journals Enhancing droplet-based single-nucleus RNA-seq resolution using the semi-supervised machine learning classifier DIEM

2019 ◽  
Author(s):  
Marcus Alvarez ◽  
Elior Rahmani ◽  
Brandon Jew ◽  
Kristina M. Garske ◽  
Zong Miao ◽  
...  
Keyword(s):  
Gene Expression ◽  
Single Cell ◽  
Cell Types ◽  
Supervised Machine Learning ◽  
Data Sets ◽  
Rna Seq ◽  
Novel Approach ◽  
Single Nucleus ◽  
Downstream Analysis

AbstractSingle-nucleus RNA sequencing (snRNA-seq) measures gene expression in individual nuclei instead of cells, allowing for unbiased cell type characterization in solid tissues. Contrary to single-cell RNA seq (scRNA-seq), we observe that snRNA-seq is commonly subject to contamination by high amounts of extranuclear background RNA, which can lead to identification of spurious cell types in downstream clustering analyses if overlooked. We present a novel approach to remove debris-contaminated droplets in snRNA-seq experiments, called Debris Identification using Expectation Maximization (DIEM). Our likelihood-based approach models the gene expression distribution of debris and cell types, which are estimated using EM. We evaluated DIEM using three snRNA-seq data sets: 1) human differentiating preadipocytes in vitro, 2) fresh mouse brain tissue, and 3) human frozen adipose tissue (AT) from six individuals. All three data sets showed various degrees of extranuclear RNA contamination. We observed that existing methods fail to account for contaminated droplets and led to spurious cell types. When compared to filtering using these state of the art methods, DIEM better removed droplets containing high levels of extranuclear RNA and led to higher quality clusters. Although DIEM was designed for snRNA-seq data, we also successfully applied DIEM to single-cell data. To conclude, our novel method DIEM removes debris-contaminated droplets from single-cell-based data fast and effectively, leading to cleaner downstream analysis. Our code is freely available for use at https://github.com/marcalva/diem.

scholarly journals Single Cell RNA-Seq Characterises Pre-Leukemic Transformation Driven By CEBPA N321D in the Hoxb8-FL Cell Line

Blood ◽  
2018 ◽  
Vol 132 (Supplement 1) ◽  
pp. 3887-3887
Author(s):  
Moosa Qureshi ◽  
Fernando Calero-Nieto ◽  
Iwo Kucinski ◽  
Sarah Kinston ◽  
George Giotopoulos ◽  
...  
Keyword(s):  
Dendritic Cell ◽  
Single Cell ◽  
Cell Line ◽  
Single Cells ◽  
Mutant Form ◽  
Rna Seq ◽  
Wild Type ◽  
Leukemic Transformation

Abstract The C/EBPα transcription factor plays a pivotal role in myeloid differentiation and E2F-mediated cell cycle regulation. Although CEBPA mutations are common in acute myeloid leukaemia (AML), little is known regarding pre-leukemic alterations caused by mutated CEBPA. Here, we investigated early events involved in pre-leukemic transformation driven by CEBPA N321D in the LMPP-like cell line Hoxb8-FL (Redecke et al., Nat Methods 2013), which can be maintained in vitro as a self-renewing LMPP population using Flt3L and estradiol, as well as differentiated both in vitro and in vivo into myeloid and lymphoid cell types. Hoxb8-FL cells were retrovirally transduced with Empty Vector (EV), wild-type CEBPA (CEBPA WT) or its N321D mutant form (CEBPA N321D). CEBPA WT-transduced cells showed increased expression of cd11b and SIRPα and downregulation of c-kit, suggesting that wild-type CEBPA was sufficient to promote differentiation even under LMPP growth conditions. Interestingly, we did not observe the same phenotype in CEBPA N321D-transduced cells. Upon withdrawal of estradiol, both EV and CEBPA WT-transduced cells differentiated rapidly into a conventional dendritic cell (cDC) phenotype by day 7 and died within 12 days. By contrast, CEBPA N321D-transduced cells continued to grow for in excess of 56 days, with an initial cDC phenotype but by day 30 demonstrating a plasmacytoid dendritic cell precursor phenotype. CEBPA N321D-transduced cells were morphologically distinct from EV-transduced cells. To test leukemogenic potential in vivo, we performed transplantation experiments in lethally irradiated mice. Serial monitoring of peripheral blood demonstrated that Hoxb8-FL derived cells had disappeared by 4 weeks, and did not reappear. However, at 6 months CEBPA N321D-transduced cells could still be detected in bone marrow in contrast to EV-transduced cells but without any leukemic phenotype. To identify early events involved in pre-leukemic transformation, the differentiation profiles of EV, CEBPA WT and CEBPA N321D-transduced cells were examined with single cell RNA-seq (scRNA-seq). 576 single cells were taken from 3 biological replicates at days 0 and 5 post-differentiation, and analysed using the Automated Single-Cell Analysis Pipeline (Gardeux et al., Bioinformatics 2017). Visualisation by t-SNE (Fig 1) demonstrated: (i) CEBPA WT-transduced cells formed a distinct cluster at day 0 before withdrawal of estradiol; (ii) CEBPA N321D-transduced cells separated from EV and CEBPA WT-transduced cells after 5 days of differentiation, (iii) two subpopulations could be identified within the CEBPA N321D-transduced cells at day 5, with a cluster of five CEBPA N321D-transduced single cells distributed amongst or very close to the day 0 non-differentiated cells. Differential expression analysis identified 224 genes upregulated and 633 genes downregulated specifically in the CEBPA N321D-transduced cells when compared to EV cells after 5 days of differentiation. This gene expression signature revealed that CEBPA N321D-transduced cells switched on a HSC/MEP/CMP transcriptional program and switched off a myeloid dendritic cell program. Finally, in order to further dissect the effect of the N321D mutation, the binding profile of endogenous and CEBPA N321D was compared by ChIP-seq before and after 5 days of differentiation. Integration with scRNA-seq data identified 160 genes specifically downregulated in CEBPA N321D-transduced cells which were associated with the binding of the mutant protein. This list of genes included genes previously implicated in dendritic cell differentiation (such as NOTCH2, JAK2), as well as a number of genes not previously implicated in the evolution of AML, representing potentially novel therapeutic targets. Disclosures No relevant conflicts of interest to declare.

scholarly journals Gene Expression Profile of the Human Colorectal Carcinoma LoVo Cells Treated With Sporamin and Thapsigargin

2021 ◽  
Vol 11 ◽  
Author(s):  
Chun Yang ◽  
Si-Jia Chen ◽  
Bo-Wen Chen ◽  
Kai-Wen Zhang ◽  
Jing-Jie Zhang ◽  
...  
Keyword(s):  
Gene Expression ◽  
Gene Expression Profile ◽  
Expression Profile ◽  
Ipomoea Batatas ◽  
Expression Profiles ◽  
Rna Seq ◽  
Ppi Network ◽  
Illumina Platform

Sporamin, a proteinase inhibitor isolated from the sweet potato (Ipomoea batatas), has shown promising anticancer effect against colorectal cancer (CRC) in vitro and in vivo but its mechanisms of action are poorly understood. In the present study, high throughput RNA sequencing (RNA-seq) technology was applied to explore the transcriptomic changes induced by sporamin in the presence of thapsigargin (TG), a non-12-O-tetradecanolphorbol-13-acetate type cancer promoter, in the LoVo human CRC cells. Cellular total RNA was extracted from the cells after they were treated with vehicle (CTL), 1 μM of thapsigargin (TG), or 1 μM of TG plus 30 μM of sporamin (TGSP) for 24 h. The migratory capacity of the cells was determined by wound healing assay. The gene expression profiles of the cells were determined by RNA-seq on an Illumina platform. GO enrichment analysis, KEGG pathway analysis, protein-protein interaction (PPI) network construction, and transcription factors (TF) prediction were all performed based on the differentially expressed genes (DEGs) across groups with a series of bioinformatics tools. Finally, the effect and potential molecular targets of the sporamin at the transcriptome level were evaluated. Sporamin significantly inhibited the migration of cells induced by TG. Among the 17915 genes detected in RNA-seq, 46 DEGs were attributable to the effect of sporamin. RT-PCR experiment validated that the expression of RGPD2, SULT1A3, and BIVM-ERCC5 were up-regulated while NYP4R, FOXN1, PAK6, and CEACAM20 were down-regulated. Sporamin enhanced the mineral absorption pathway, worm longevity regulating pathway, and pyrimidine metabolism pathway. Two TFs (SMIM11A and ATOH8) were down-regulated by sporamin. HMOX1 (up-regulated) and NME1-NME2 (down-regulated) were the main nodes in a PPI network consisting of 16 DEGs that were modulated by sporamin in the presence of TG. Sporamin could favorably alter the gene expression profile of CRC cells, up-regulating the genes that contribute to the homeostasis of intracellular metal ions and the activities of essential enzymes and DNA damage repairment. More studies are warranted to verify its effect on specific genes and delineate the mechanism of action implicated in the process.

scholarly journals Analysis of theIn VivoTranscriptome ofBordetella pertussisduring Infection of Mice

mSphere ◽  
2019 ◽  
Vol 4 (2) ◽  
Author(s):  
Ting Y. Wong ◽  
Jesse M. Hall ◽  
Evan S. Nowak ◽  
Dylan T. Boehm ◽  
Laura A. Gonyar ◽  
...  
Keyword(s):  
Gene Expression ◽  
Iron Acquisition ◽  
Expression Profiles ◽  
Gene Expression Profiles ◽  
Growth Conditions ◽  
Rna Seq ◽  
Content Type ◽  
Murine Lung

ABSTRACTBordetella pertussiscauses the disease whooping cough through coordinated control of virulence factors by theBordetellavirulence gene system. Microarrays and, more recently, RNA sequencing (RNA-seq) have been used to describein vitrogene expression profiles ofB. pertussisand other pathogens. In previous studies, we have analyzed thein vitrogene expression profiles ofB. pertussis, and we hypothesize that the infection transcriptome profilein vivois significantly different from that under laboratory growth conditions. To study the infection transcriptome ofB. pertussis, we developed a simple filtration technique for isolation of bacteria from infected lungs. The work flow involves filtering the bacteria out of the lung homogenate using a 5-μm-pore-size syringe filter. The captured bacteria are then lysed to isolate RNA for Illumina library preparation and RNA-seq analysis. Upon comparing thein vitroandin vivogene expression profiles, we identified 351 and 255 genes as activated and repressed, respectively, during murine lung infection. As expected, numerous genes associated with virulent-phase growth were activated in the murine host, including pertussis toxin (PT), the PT secretion apparatus, and the type III secretion system. A significant number of genes encoding iron acquisition and heme uptake proteins were highly expressed during infection, supporting iron acquisition as critical forB. pertussissurvivalin vivo. Numerous metabolic genes were repressed during infection. Overall, these data shed light on the gene expression profile ofB. pertussisduring infection, and this method will facilitate efforts to understand how this pathogen causes infection.IMPORTANCEIn vitrogrowth conditions for bacteria do not fully recapitulate the host environment. RNA sequencing transcriptome analysis allows for the characterization of the infection gene expression profiles of pathogens in complex environments. Isolation of the pathogen from infected tissues is critical because of the large amounts of host RNA present in crude lysates of infected organs. A filtration method was developed that enabled enrichment of the pathogen RNA for RNA-seq analysis. The resulting data describe the “infection transcriptome” ofB. pertussisin the murine lung. This strategy can be utilized for pathogens in other hosts and, thus, expand our knowledge of what bacteria express during infection.

243. In vitro culture significantly reduces differences in gene expression profiles between myometrial and fibroid smooth muscle cells

2005 ◽  
Vol 17 (9) ◽  
pp. 96
Author(s):  
M. Zaitseva ◽  
P. A. W. Rogers
Keyword(s):  
Gene Expression ◽  
Smooth Muscle ◽  
Smooth Muscle Cells ◽  
Cultured Cells ◽  
Expression Profiles ◽  
Gene Expression Profiles ◽  
Muscle Cells ◽  
Compare Gene Expression

Fibroids are benign neoplasms of the smooth muscle cells of the uterus. Cultured myometrial (M) and fibroid (F) smooth muscle cells (SMC) have been widely used as a model for the study of fibroid growth. Although it has been shown that FSMC can behave differently in culture to MSMC, it is not clear how relevant the cultured cells and their responses are to the in-vivo situation. The aim of the present study was to compare gene expression profiles of M and F tissue to cells isolated from the same tissue and cultured for up to 3 passages. M and F were collected from hysterectomy specimens (n = 6), part was snap frozen for RNA and the rest used to isolate SMC, which were cultured for 3 passages and RNA was collected at passage 0 (P0) and 3 (P3). 36 microarrays were performed on 8K human cDNA slides, 6 per each specimen (3 for M and 3 for F: tissue, cell at P0 and P3) against reference RNA. Analysis revealed significant differences between tissues and cultured cells. Independent clustering assigned tissues versus cells into two distinct groups based on their expression profiles. Parametric ANOVA with Benjamini-Hochberg correction and post-hoc testing was used to determine similarities and differences between tissues and cells. 128 genes were found to be statistically different between M and F tissue, 66 between MSMC and FSMC at P0, and only 9 at P3. More than 1100 genes were significantly changed between tissues and cultured cells, with 648 genes common between both M and F cells at P0 and P3. Similar numbers of genes were up regulated as were down regulated. Expression profiles of genes of interest including estrogen receptor α and progesterone receptor were also validated using real-time PCR. This is the first study to compare gene expression of in vivo and in vitro fibroid and myometrial SMC. The results demonstrate that large changes occur in SMC gene expression in culture, reducing differences between myometrial and fibroid cells. This study indicates that results of in vitro studies should be interpreted with caution as many genes have an altered gene expression profile in culture.

The VSL#3 probiotic formula induces mucin gene expression and secretion in colonic epithelial cells

2007 ◽  
Vol 292 (1) ◽  
pp. G315-G322 ◽  
Author(s):  
C. Caballero-Franco ◽  
K. Keller ◽  
C. De Simone ◽  
K. Chadee
Keyword(s):  
Gene Expression ◽  
Epithelial Cells ◽  
Cultured Cells ◽  
In Vivo Studies ◽  
Lactobacillus Species ◽  
Mucin Secretion ◽  
Colonic Epithelial Cells ◽  
Dnase Treatment

Several studies have stressed the importance of the microbiota in the maintenance of the gastrointestinal epithelium. Administration of probiotic bacteria, supplements composed of microbiota constituents, was previously shown to diminish symptoms in patients suffering from inflammatory bowel diseases. This raises the possibility that probiotics may play an active role in enhancing the intestinal barrier at the mucosal surface. In this study, we investigated whether the clinically tested VSL#3 probiotic formula and/or its secreted components can augment the protective mucus layer in vivo and in vitro. For in vivo studies, Wistar rats were orally administered the probiotic mixture VSL#3 on a daily basis for seven days. After treatment, basal luminal mucin content increased by 60%. In addition, we exposed isolated rat colonic loops to the VSL#3 probiotic formula, which significantly stimulated colonic mucin (MUC) secretion and MUC2 gene expression; however, MUC1 and MUC3 gene expression were only slightly elevated. The effect of the VSL#3 mucin secretagogue was also tested in vitro by use of LS 174T colonic epithelial cells. In contrast to the animal studies, cultured cells incubated with VSL#3 bacteria did not exhibit increased mucin secretion. However, the bacterial secreted products contained in the conditioned media stimulated a remarkable mucin secretion effect. Among the three bacterial groups ( Lactobacilli, Bifidobacteria, and Streptococci) contained in VSL#3, the Lactobacillus species were the strongest potentiator of mucin secretion in vitro. A preliminary characterization of the putative mucin secretagogue suggested that it was a heat-resistant soluble compound, which is not sensitive to protease and DNase treatment. These findings contribute to a better understanding of the complex and beneficial interaction between colonic epithelial cells and intestinal bacteria.

scholarly journals Rat Cytomegalovirus Gene Expression in Cardiac Allograft Recipients Is Tissue Specific and Does Not Parallel the Profiles Detected In Vitro

2007 ◽  
Vol 81 (8) ◽  
pp. 3816-3826 ◽  
Author(s):  
Daniel N. Streblow ◽  
Koen W. R. van Cleef ◽  
Craig N. Kreklywich ◽  
Christine Meyer ◽  
Patricia Smith ◽  
...  
Keyword(s):  
Gene Expression ◽  
Rat Heart ◽  
Cultured Cells ◽  
Viral Gene Expression ◽  
Viral Gene ◽  
Viral Mrna ◽  
Tissue Specific ◽  
Viral Genes

ABSTRACT Rat cytomegalovirus (RCMV) is a β-herpesvirus with a 230-kbp genome containing over 167 open reading frames (ORFs). RCMV gene expression is tightly regulated in cultured cells, occurring in three distinct kinetic classes (immediate early, early, and late). However, the extent of viral-gene expression in vivo and its relationship to the in vitro expression are unknown. In this study, we used RCMV-specific DNA microarrays to investigate the viral transcriptional profiles in cultured, RCMV-infected endothelial cells, fibroblasts, and aortic smooth muscle cells and to compare these profiles to those found in tissues from RCMV-infected rat heart transplant recipients. In cultured cells, RCMV expresses approximately 95% of the known viral ORFs with few differences between cell types. By contrast, in vivo viral-gene expression in tissues from rat heart allograft recipients is highly restricted. In the tissues studied, a total of 80 viral genes expressing levels twice above background (5,000 to 10,000 copies per μg total RNA) were detected. In each tissue type, there were a number of genes expressed exclusively in that tissue. Although viral mRNA and genomic DNA levels were lower in the spleen than in submandibular glands, the number of individual viral genes expressed was higher in the spleen (60 versus 41). This finding suggests that the number of viral genes expressed is specific to a given tissue and is not dependent upon the viral load or viral mRNA levels. Our results demonstrate that the profiles, as well as the amplitude, of viral-gene expression are tissue specific and are dramatically different from those in infected cultured cells, indicating that RCMV gene expression in vitro does not reflect viral-gene expression in vivo.

248 Immunotherapy persister cells uncovered by dynamic single-cell RNA-sequencing

2020 ◽  
Vol 8 (Suppl 3) ◽  
pp. A268-A269
Author(s):  
Kartik Sehgal ◽  
Andrew Portell ◽  
Elena Ivanova ◽  
Patrick Lizotte ◽  
Navin Mahadevan ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Single Cell ◽  
Rna Sequencing ◽  
Ex Vivo ◽  
Interferon Γ ◽  
Rna Seq ◽  
Persister Cells ◽  
Pre Treatment

BackgroundTo understand fundamental mechanisms of immune escape, we leveraged our functional ex vivo platform of murine derived organotypic tumor spheroids (DOTS)1 to determine if drug-tolerant persister cells analogous to oncogene targeted therapies limit efficacy of programmed death (PD)-1 blockade, and to identify therapeutic vulnerabilities to overcome anti-PD-1 (αPD-1) resistance.MethodsMurine syngeneic cancer models with well-characterized response to αPD-1 therapy were chosen: MC38 (sensitive) and CT26 (partially resistant). Bulk and single-cell (sc) RNA-sequencing (RNA-seq) were performed on αPD-1 treated DOTS. In vitro culture studies were conducted with or without cytokines (100 ng/ml) or drugs (500 nM). In vivo studies in mice bearing MC38 or CT26 tumors evaluated the combinatorial strategy with PD-1 blockade. We further evaluated our findings in scRNA-seq of an αPD-1 refractory colorectal cancer (CRC) patient tumor.2ResultsBulk RNA-seq of αPD-1 treated DOTS revealed a mesenchymal resistant phenotype with upregulated TNF-α/NFκB signaling (figure 1). scRNA-seq further identified a discrete sub-population of immunotherapy persister cells (IPCs). These cells expressed a stem-like phenotype including downregulation of E2F targets indicative of quiescence, suppression of interferon-γ response genes, induction of hybrid epithelial-to-mesenchymal state, and active IL-6 signaling (figure 1). Ly6a/stem cell antigen-1 (Sca-1) and Snai1 were found to be differentially upregulated in IPCs resistant to PD-1 blockade (not shown). Sca-1 positivity was confirmed in pre-existing tumor populations in vitro (figure 2). When enriched via sorting, these cells remained more persistently Sca-1+ at 96 hours in culture of CT26 compared to MC38 cells, related to increased autocrine IL-6 production by CT26 Sca-1+ cells. Indeed, IL-6 supplementation was capable of expanding Sca-1+ cells in culture (figure 2). Sca-1+ cells expressing ovalbumin peptide were refractory to OT-1 T cell mediated killing and failed to upregulate MHC class-1 antigen presentation (H-2Kb) in response to IL-6, in contrast to interferon-γ (not shown). Analysis of RNA-seq data further identified Birc2/3 as potential targets limiting TNF-mediated apoptosis of these cells (not shown). Notably, Birc2/3 antagonism depleted Sca-1+ IPCs in vitro and significantly potentiated the impact of PD-1 blockade in vivo in MC38, and less robustly in CT26 (figure 3). Evaluation in a microsatellite-instability high CRC patient identified a pre-existent IPC subpopulation within the αPD-1 refractory pre-treatment tumor, with high SNAI1 expression compared to CRC samples in TCGA (figure 4).Abstract 248 Figure 1Bulk and single-cell (sc) RNA-sequencing (RNA-seq) of MDOTS identifies an anti-PD-1 (αPD-1) resistant subpopulation of persister cells. IgG= isotype controlAbstract 248 Figure 2Pre-existent population of stem cell antigen-1 (Sca-1)+ cells expands in response to interleukin-6 (IL-6), as characterized by flow cytometry evaluation in murine syngeneic cancer models at baseline and after purification by fluorescence-activated cell sorting (FACS). H = hoursAbstract 248 Figure 3Combination of anti-PD-1 therapy with Birc2/3 antagonism increases tumor responses and improves survival. CR = complete responseAbstract 248 Figure 4Single-cell RNA-sequencing (scRNA-seq) of a pre-treatment microsatellite-instability (MSI-H) colorectal cancer (CRC) patient tumor, refractory to anti-PD-1 (αPD-1) therapy, reveals presence of SNAI1-high immunotherapy persister cellsConclusionsHigh-resolution functional ex vivo profiling identified Sca-1+/Snai1high stem-like ‘immunotherapy persister cells‘ and uncovered their anti-apoptotic dependencies targetable with Birc2/3 antagonism to augment αPD-1 efficacy.Ethics ApprovalThis study was approved by the Dana-Farber Animal Care and Use Committee and Novartis Institutional Animal Care and Use Committee. Informed written consent to participate in Dana-Farber/Harvard Cancer Center institutional review board (IRB)-approved research protocols was obtained from the human subject. A copy of the written consent is available for review by the Editor of this journal. The study was conducted per the WMA Declaration of Helsinki and IRB-approved protocols.ReferencesJenkins RW, Aref AR, Lizotte PH, Ivanova E, Stinson S, Zhou CW, et al. Ex Vivo Profiling of PD-1 Blockade using organotypic tumor spheroids. Cancer Discov. 2018;8(2):196–668 215.Gurjao C, Liu D, Hofree M, AlDubayan SH, Wakiro I, Su MJ, et al. intrinsic resistance to immune checkpoint blockade in a mismatch repair-deficient colorectal cancer. Cancer Immunol Res 2019;7(8):1230–6.

Sign in / Sign up

Export Citation Format

Share Document