Cell Sorting Research Articles

Endovascular biopsy in neurointerventional surgery: A systematic review.

Endothelial cells (ECs) continuously line the cerebrovasculature. Molecular aberrations in the ECs are hallmarks and contributory factors to the development of cerebrovascular diseases, including intracranial aneurysms and arteriovenous malformations (AVMs). Endovascular biopsy has been introduced as a method to harvest ECs and obtain relevant biologic information. We aimed to summarize the literature on endovascular biopsy in neurointerventional surgery. We conducted a comprehensive literature search in multiple databases, identifying eligible studies focusing on neurosurgical applications of endovascular biopsy. The systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The relevant information was collected, including study characteristics, biopsy techniques, and key findings. Nine studies met the inclusion criteria and were included. The studies involved the collection of ECs using various endovascular devices including coils, guide wires, different stents, and forceps. Endothelial-enrichment techniques, such fluorescence-activated cell sorting (FACS), collected ECs and facilitated downstream applications of bulk or single-cell RNA sequencing (scRNAseq). The studies provided insights into gene expression profiles and identified potential biomarkers associated with intracranial aneurysms. However, challenges were observed in obtaining an adequate number of ECs and identifying consistent biomarkers. Endovascular biopsy of endothelial cells (ECs) in cerebrovascular pathologies shows promise for gene expression profiling. However, many studies have been limited in sample size and underpowered to identify "signature genes" for aneurysm growth or rupture. Advancements in minimally invasive biopsy methods have potential to facilitate applications of precision medicine in the treatment of cerebrovascular disorders.

Abstract 291: Unveiling heterogeneity of cancer-associated fibroblasts (CAFs) across multiple solid cancer types

Abstract Background: Stromal cells, including CAFs, are a heterogeneous group of cells in the tumor microenvironment. Many models describe their heterogeneity like the classification into 3 main subtypes in breast cancer: myCAF (myofibroblasts), iCAF (inflammatory CAF), and PVL (perivascular-like) cells. However, a pan-cancer model has not been established to exhaustively characterize stromal cell heterogeneity, undermining efforts in targeting CAFs with anti-tumor therapies. We investigated the heterogeneity of stromal cells in diverse diagnoses with single-cell RNA sequencing (scRNA-seq) data and bulk RNA-seq of sorted stromal cells to define consistent subtypes. Design: First, a meta-cohort of 11 scRNA-seq open-source datasets (>500,000 cells, >100 patients, 6 diagnoses) was analyzed. We also generated a scRNA-seq dataset of CD90+ cells derived from tumor tissue of an ovarian cancer patient. Additionally, bulk RNA-seq of 6 PVL, 2 iCAF, and 4 myCAF sorted cell samples across four solid cancers was performed. scRNA-seq data was analyzed with the SCANPY Python package. Results: Stromal cell clusters were defined as positive for PDGFRB, THY1 (CD90), and COL1A1 expression, and negative for the expression of known non-stromal and endothelial cell markers, by separately analyzing the open-source scRNA-seq datasets. Analysis of stromal cell heterogeneity revealed that the CAFs were consistently classified into 3 subtypes across all datasets: myCAF (high expression: FAP and ACTA2), PVL (high expression: MCAM (CD146) and ACTA2), and iCAF (high expression: APOD and cytokines IL6, CXCL12, CXCL2). Differential gene expression analysis of the scRNA-seq meta-cohort allowed us to develop functional gene expression signatures to define each subtype. All scRNA-sequenced CD90+ cells from ovarian cancer tissue were classified into the CAF subtypes. The myCAF and PVL signatures were specifically upregulated in the corresponding cell types (p<0.01, Mann-Whitney U test). While the iCAF signature was enriched in both iCAFs and myCAFs, myCAF and PVL signatures were downregulated in iCAFs. Finally, we developed a fluorescence-activated cell sorting (FACS) panel to separate CAF subtypes, using negative markers (CD45, CD31, EPCAM), a positive marker for stromal cells (CD90) and cell-type surface markers (CD146, FAP, CD91, CD44) for myCAF, iCAF and PVL cells. The high expression of CAF-specific genes in the sorted CAF populations denoted high sample purity. Importantly, the 3 CAF subpopulations separated by gating based on different levels of cell-type surface markers showed expression patterns congruent with the developed signatures across diagnoses. Summary: Altogether, the consistent classification of stromal cells with our cell-specific expression profiles demonstrated that the 3 CAF subtypes are conserved in many cancer types, offering groundwork for a pan-cancer model of stromal cell heterogeneity. Citation Format: Andrey Kravets, Boris Shpak, Anastasia Zotova, Maria Savchenko, Sofya Kust, Olga Golubeva, Mary Abdou, Fedor Grigoryev, Irina Bulusheva, Beniamin Sargsyan, Polina Pavlovich, Vlad Maximov, Vladimir Kushnarev, Michael Goldberg, Nikita Kotlov, Alexander Bagaev. Unveiling heterogeneity of cancer-associated fibroblasts (CAFs) across multiple solid cancer types [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 291.

Abstract 1598: Individualized longitudinal assessment reveals therapy-induced tumor and microenvironment dynamics in preclinical pancreatic cancer models

Abstract Background & Aim: Progression and therapeutic resistance in pancreatic cancer depends on the ability of cancer cells to adopt specific cell states. These phenotypes largely depend on environmental tissue niches. Here, we investigated therapy-induced tumor dynamics in vivo at single-cell spatial and temporal resolution in different subtypes of PDAC using several genetically engineered mouse models, which reflect distinct human PDAC ecosystems. Methods & Results: We established orthotopic tumor models of pancreatic cancer with distinct tumor microenvironment (TME) formation (reactive vs deserted stroma), which were monitored for tumor growth to define tumor response patterns under MAPK inhibition. Serial ultrasound-guided biopsy sampling was performed to collect tissues from the same tumors at pre-treatment, responsive and recurrence periods.Transcriptomic analysis of biopsies using the Mfuzz algorithm delineated highly dynamic gene and pathway activities during treatment. Reactive, but not deserted, TME showed an upregulation of immune-related pathways (antigen presentation, T cell receptor, NK cytotoxicity) immediately after MAPK inhibition in regressed tumors that dropped dramatically during recurrence of tumors, in which upregulation of TGF-b signaling was observed. In the deserted stroma model, on the other hand, an enrichment in drug metabolic pathways during recurrence was observed.CAF cluster deconvolution analysis revealed an increase in a LRRC15+ CAF cluster, which was TGF-b-driven and immunosuppressive, during recurrence period only in the reactive model. The dynamic changes of immune cells and CAFs were validated at protein level by multiplex immunofluorescent imaging. Cmputational spatial analysis revealed a tumor-encapsulating and CD8+ cytotoxic T cell-repelling pattern exhibited by LRRC15+FAP+ CAFs, implying a potential role of this CAF subset in T cell suppression during recurrence in reactive subtype. Using spectral cell sorting with CellView image technology, specific CAF subsets, tumor cells as well as immune cell subsets based on 30+ markers and cell morphology were isolated, characterized and assessed by ex vivo functional assays to confirm the immunosuppressive and tumor protective mechanism mediated by CAFs. Conclusion: Longitudinal in vivo assessment of distinct PDAC subtypes provides deep mechanistic insights to accurately unlock specific response patterns of tumors with heterogeneous TME dynamics upon therapeutic perturbation, leveraging identification of key nodes for intervention. Citation Format: Jens T. Siveke, Rui Fang, Phyllis Fung-Yi Cheung, Jiajin Yang, Juanfei Peng, Kristina Althoff, Konstantinos Savvatakis. Individualized longitudinal assessment reveals therapy-induced tumor and microenvironment dynamics in preclinical pancreatic cancer models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1598.

Abstract 4216: Indication agnostic evaluation of CAR T cell function using patient-derived immunocompetent tumoroid modelling platform

Abstract Introduction: Engineered T cells redirected towards tumor-specific cytotoxicity is a promising immunotherapeutic modality that has been well characterized for efficacy against B cell malignancies and evaluated in multiple tumor types. However, the development and optimization of CAR T cell therapies requires a rigorous preclinical model for the evaluation of CAR T cell function, tumor killing, cytokine production, and memory responses. Experimental Procedure: We are evaluating our model platform Custom Organoid Modelling Platform for Accurate and Speedy SolutionS (COMPASS 2) as a comprehensive platform for testing developmental and functional stages of CAR T cells including activation, memory, and exhaustion. We are performing qualitative and quantitative analyses of surface markers to recapitulate the complexity of human cancer. Additionally, since CAR T cells mediate crosstalk with other immune components in the TME, we also evaluate and monitor the presence of B cells, NK cells, macrophages, and endogenous T cells. Triple Negative Breast Cancer (TNBC); Immunophenotyping (IP); Immunofluorescence (IF); Fluorescence Activated Cell Sorting (FACS); T regulatory (Treg). Summary of Results: Our evaluation of COMPASS 2 for functional testing of CAR T cells demonstrates the feasibility of the platform in preclinical CAR T cell evaluation. We showed T cell infiltration and analyzed cytokine production that may alter tumor cells and TME. Crosstalk with other immune cells will be evaluated. We are currently evaluating the tumor volume changes using an EGFRvIII-targeted CART cell therapy on TNBC model systems, followed by apoptosis and EGFR analysis using FACS. Conclusion: Our COMPASS 2 model platform enables studies that require recapitulating hurdles to CAR therapy, including antigen escape, T cell dysfunction and trafficking; to reveal an essential role of the endogenous immune response mediating the antitumor effect following CAR therapy. Citation Format: Damieanus O. Ochola, Huckie Dell Mundo, Preethi Samuel, Natalia Korkorina, Bhuvanesh Dave. Indication agnostic evaluation of CAR T cell function using patient-derived immunocompetent tumoroid modelling platform [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4216.

Endoplasmic reticulum stress drives macrophages to produce IL-33 to favor Th2 polarization in the airways.

Interleukin (IL)-33 is a key driver of T helper 2 (Th2) cell polarization. Endoplasmic reticulum (ER) stress plays a role in the skewed T cell activation. The objective of this project is to elucidate the role of IL-33 derived from macrophages in inducing Th2 polarization in the airways. In this study, bronchoalveolar lavage fluids (BALF) were collected from patients with asthma and healthy control subjects. Macrophages were isolated from the BALF by flow cytometry cell sorting. An asthmatic mouse model was established using the ovalbumin/alum protocol. The results showed that increased IL33 gene activity and ER stress-related molecules in BALF-derived M2a macrophages was observed in asthmatic patients. Levels of IL33 gene activity in M2a cells were positively correlated with levels of asthma response in asthma patients. Sensitization exacerbated the ER stress in the airway macrophages, which increased the expression of IL-33 in macrophages of airway in sensitized mice. Conditional ablation of Il33 or Perk or Atf4 genes in macrophages prevented induction of airway allergy in mice. In conclusion, asthma airway macrophages express high levels of IL-33 and at high ER stress status. Inhibition of IL-33 or ER stress in macrophages can effectively alleviate experimental asthma.

Abstract 1271: PMEPA1 acts as a switch to modulate cooperative cellular invasion and drive NSCLC tumor progression

Abstract Metastatic disease drives cancer patient mortality. One primary mode of metastasis is collective invasion, whereby cohesive groups of cells invade into the adjacent stroma while maintaining cell-cell contacts. Importantly, these cellular packs harbor genetically and phenotypically heterogeneous subpopulations that cooperate to drive invasion. To deconstruct how phenotypic cellular heterogeneity facilitates distinct molecular profiles within a single tumor, we established the technique Spatiotemporal Cellular and Genomic Analysis (SaGA). SaGA is an image-guided approach that optically highlights phenotypically defined cell(s) for live fluorescence-activated cell sorting and analysis. To elucidate the transcriptional heterogeneity across invasive packs within a 3D physiologically-relevant microenvironment, we combined SaGA with single cell RNA sequencing for isolation and analysis of actively invading cells. This novel single cell transcriptomics approach reveals that - during cellular invasion into the microenvironment - the follower subpopulation supplies TGFβ to heterogeneously stimulate both autocrine canonical and paracrine noncanonical signaling pathways. This cooperative exchange facilitates a symbiotic relationship between tumor subpopulations in non-small cell lung carcinoma both in vitro and in vivo. In vitro, we determine that PMEPA1 protein levels mitigate TGFβ downstream effector signaling within heterogeneous tumor cells to modulate JAG1 expression and activity. In vivo, we show that a reduction in JAG1 levels decreases primary tumor cell invasion and progression. We mechanistically define a TGFβ-PMEPA1-JAG1 signaling axis across leader and follower subpopulations critical for tumor cell invasion and cancer progression. Together, these data highlight the importance of cellular heterogeneity and subpopulation communication required for tumor progression. Citation Format: Tala O. Khatib, Brian A. Pedro, Christina M. Knippler, Veronika Y. Matsuk, Sarah F. Webster, Sergei Bombin, Bhakti Dwivedi, Jeanne Kowalski, Henry R. Johnston, Janna K. Mouw, Adam I. Marcus. PMEPA1 acts as a switch to modulate cooperative cellular invasion and drive NSCLC tumor progression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1271.

Abstract 287: Analysis for the role of antigen-presenting cancer-associated fibroblasts in tumor microenvironment of colorectal cancer

Abstract A new subset of cancer associated fibroblasts (CAFs) called antigen presenting CAF (apCAF) was recently identified, and modulation of the tumor immunity might be associated with tumor progression. We examined the markers of apCAF, such as IRF5, HLA-DRA and HLA-DQA1 by bioinformatics analyses of human colorectal cancer (CRC) datasets. The expression of apCAF markers were upregulated in CAFs isolated from human CRCs and liver metastases compared with their normal counterparts. Kaplan-Meir analysis of 172 human CRC samples clarified that the patients with apCAF marker positive expression in the tumor stroma had worse prognosis than that with negative expression. In addition, the origin of apCAFs was reported to be mesothelial cells, therefore, we established cancer-associated mesothelial cells (CAmes) from ascites of human CRC patients with peritoneal dissemination. Consistently, the expression of antigen-presentation markers was highly upregulated in CAmes, however, allo-mixed lymphocyte reaction analysis demonstrated that their stimulating activity of CD4+ T lymphocytes was diminished. In an in vivo study, we applied a rectal orthotopic transplant model with mouse tumor organoid (MTO) to investigate the effect of apCAF on tumor progressions. Using fluorescence-activated cell sorting (FACS), we examined the ratios of apCAF, myCAF, and iCAF in orthotopic rectal tumors. The percentage of apCAF was highest in the 6-week-tumors and the percentage of myCAF was highest in the 9-week-tumors. Interestingly, the percentage of apCAF was higher in liver metastases than in primary tumor, consistent with bioinformatics analysis. The apCAF increased during the stage of tumor growth, and the percentage of myCAF increase in the later stage of tumor progression with fibrosis. In summary, apCAF might be associated with tumor progression in CRC. Citation Format: Yasuhiro Fukui, Hiroaki Kasashima, Zizhou Wang, Ken Yonemitsu, Kishu Kitayama, Yuichiro Miki, Mami Yoshii, Tatsunari Fukuoka, Tatsuro Tamura, Masatsune Shibutani, Takahiro Toyokawa, Hiroaki Tanaka, Shigeru Lee, Masakazu Yashiro, Kiyoshi Maeda. Analysis for the role of antigen-presenting cancer-associated fibroblasts in tumor microenvironment of colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 287.

Abstract 2760: Novel GPCR recombinant monoclonal antibodies for cancer biology research

Abstract Recombinant technology is becoming the preferred production mechanism for creation of antibody reagents used in academic and clinical cancer biology research. Recombinant antibodies (rAbs) are superior to standard polyclonal and hybridoma-generated monoclonal reagents as they offer reproducible performance and consistent, scalable supply. A major focus of antibody manufacturers is to leverage rAb production strategies to generate antibodies against challenging targets for which there are few, if any, reliable antibodies. G-protein coupled receptors (GPCRs) are encoded by more than 820 human genes and represent the largest membrane protein superfamily in the genome. These seven-pass transmembrane domain receptors are involved in a multitude of physiological and pathophysiological processes related to various chronic diseases, including an array of malignancies. Perhaps 30-40% of all therapeutics are directed against GPCR activity, and ~350 of the non-olfactory GPCRs are thought to be potential candidates for drug development. However, for a slew of technical reasons, the development and validation of high-quality antibodies against GPCRs has proven difficult. In an effort to facilitate further research into GPCRs, GeneTex is combining its recombinant antibody production platform with enhanced validation techniques to make specific monoclonal antibodies against GPCRs with direct relevance to cancer and other diseases. GeneTex’s rAb production protocol employs a multi-parameter fluorescence-activated single cell sorting (FACS)-based methodology to identify and select antigen-specific IgG+ memory B cells from an immunized rabbit (Starkie et al., 2016). The heavy and light chain variable region genes from single cells are amplified, cloned, and co-expressed in mammalian cells to generate a monoclonal IgG. Application-specific testing is performed in conjunction with knockout/knockdown (KO/KD), differential expression comparison in cells and tissues, cell fraction enrichment (i.e., membrane extracts versus whole cell extracts), and other methodologies. Antibodies directed against extracellular GPCR domains are validated for specificity using GPCR arrays (CDI Labs, Baltimore, MD). This workflow has resulted in the production of specific antibodies for a number of GPCRs studied by cancer biologists. These include novel recombinant monoclonal antibodies to the dopamine D2 receptor (DRD2), retinoic acid-induced protein 3 (RAI3), CXC motif chemokine receptor 2 (CXCR2), CXCR4, and CXCR7. All of these antibodies were validated for at least two applications, while the specificity of three (i.e., RAI3, CXCR4, and CXCR7) is supported by KO/KD data. GPCR array data argues for the specificity of the other two antibodies (i.e., DRD2 and CXCR2). This same production/validation approach will be applied as GeneTex targets the remaining GPCRs. Citation Format: Alexander Ball, Yung Lin Hsieh, Chun Kai Huang, Sega Huang, Shang Ru Wu, Yue Yun Chang, Jiming Liu, Po Shin Peng, Chia Yi Lin. Novel GPCR recombinant monoclonal antibodies for cancer biology research [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2760.

Abstract 3772: Tissue-niche-based and cell-type-selective in-depth proteomics

Abstract Introduction: The development, physiology function, and pathogenesis of the tissues are intricately orchestrated by the spatial organization of diverse cell populations. Proteins are the fundamental components within individual cells that control every function. Therefore, by integrating both information on cellular location and molecular profiling, spatial proteomics has emerged as an indispensable field for understanding mechanisms governing tissue homeostasis, disease progression, and therapeutic responses. Here, I introduce a spatial proteomics method designed to target selective cell phenotypes within definite tissue niches for in-depth spatial biology analysis. Method: In this approach, we developed a sequential pipeline highlighted with immunofluorescence staining, microscopic photobleaching, fluorescence cell sorting, and liquid chromatography-mass spectrometry (LC-MS) analysis. The process begins with staining 400-μm-thick tissue macrosections with fluorescence-conjugated antibodies to visualize specific cell types of interest. We employed two different fluorescence-conjugated antibodies of the same clone (e.g., Alexa488-anti-CD11c and Alexa647-anti-CD11c antibodies) for the staining on every targeted cell (e.g., CD11c+ immune cells). We then introduced “photobleaching barcodes” using a confocal microscope to photobleach/exhaust either one of the fluorescence signals stained on cells by exposing it to the matched laser. This allows us to create differentiable fluorescence barcodes that label cells at distinct regions, incorporating cell spatial information into our fluorescence detection. After tissue dissociation, barcoded cells were sorted into different groups which were also classified by their original locations in the macrosection. Finally, protein extraction and LC-MS analysis were conducted on the collected cells to enable comprehensive spatial proteomic analysis. Results: The initial application in investigating dendritic cell (DC) subsets in mouse spleen during lipopolysaccharide (LPS)-induced inflammation reveals significant proteome differences among three splenic DC subsets, categorized by their locations in or outside spleen T-cell zones as well as control DCs. Our result aligns with previously published proteome data in splenic DCs, underscoring the feasibility and reliability of our technology. Currently, we are in the process of evaluating breast cancer cell heterogeneity in human biopsy specimens using our approach to profile the tumor-immune microenvironment. Conclusion: Spatial proteomic findings from both applications are poised to discover potential drug targets that benefit treatment efficacy for inflammatory diseases and breast cancer. This technology features deep protein profiling for cell-type-specific spatial proteomics and is designed for broad applications across diverse tissue specimens in various diseases. Citation Format: Yi-Chien Wu, Elie Abi Khalil, Samuel Weng, Steve Lee. Tissue-niche-based and cell-type-selective in-depth proteomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3772.

Abstract 3533: Spatially Resolved Laser-Activated Cell Sorting (SLACS) and AI integration for enhanced pathological analysis and biomarker discovery

Abstract This study focuses on the integration of Artificial Intelligence (AI) with Spatially Resolved Laser-Activated Cell Sorting (SLACS) for enhanced pathological analysis and biomarker discovery. The primary approach involves using AI to automatically analyze slide images, identifying critical areas associated with diseases such as cancer or cells emitting specific signals. AI algorithms efficiently segment these significant regions or cellular groups, facilitating targeted analysis. Following the AI-driven segmentation, SLACS comes into play, isolating the identified cells from the slide for further examination. This precise sorting allows for the focused study of cells relevant to disease processes, enhancing the efficiency and accuracy of pathological analysis. The separated samples obtained through SLACS are then subjected to detailed analysis. This step is crucial for understanding the cellular mechanisms at play in disease states and for the identification of potential biomarkers. These biomarkers are vital for developing targeted therapeutic strategies and advancing personalized medicine. By combining AI's advanced image processing and pattern recognition capabilities with the precision of SLACS, this integrated approach significantly improves the identification and analysis of disease-related cellular characteristics. It streamlines the process of isolating and studying specific cell populations, contributing to a deeper understanding of disease mechanisms and progression. In conclusion, the synergistic application of AI and SLACS represents a significant advancement in pathological research and biomarker discovery. This methodology not only enhances the accuracy of disease marker identification but also opens new avenues for research and clinical application. Citation Format: Haewook Jang, Amos Chungwon Lee, Sumin Lee, Sunghoon Kwon. Spatially Resolved Laser-Activated Cell Sorting (SLACS) and AI integration for enhanced pathological analysis and biomarker discovery [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3533.

Abstract 4264: The TWEAK-Fn14-NF-kB signaling axis enhances expression of integrin AVB3 in a post-chemotherapy tumor microenvironment

Abstract High grade serous ovarian cancer (HGSOC) patients initially respond to chemotherapy but a majority relapse with chemoresistant disease. Research supports the notion that tumor-initiating cells (TICs), a subpopulation of drug resistant stem-like tumor cells, are responsible for facilitating relapse and therapies targeting these fecund cells may prolong remission. The tumor microenvironment (TME) plays a crucial role in therapy resistance, and tumor relapse. Cytotoxic chemotherapies have been shown to modify the stroma of the TME which may favor the survival of TICs displaying specific adhesion proteins. We and others have shown that NF-kB transcription factors RelA and RelB respond to signals from the ovarian TME to promote proliferation, chemoresistance, and survival. Our results further show that RelA and RelB regulate expression of specific integrins. Recently we found that TNF-like inducer of apoptosis (TWEAK), a cytokine involved in tissue repair, and its only known receptor Fn14, are enriched following chemotherapy, leading to enhanced chemoresistance and survival of TICs. TWEAK also induces expression of specific integrins in ovarian TICs. Therefore, we hypothesize that NF-kB mediated integrin expression enables adhesion and survival of TICs to a chemotherapy modified extracellular matrix (ECM) to facilitate tumor repopulation and relapse. Using magnetic activated cell sorting (MACS) for CD117+ TICs, we found that a significant majority are enriched for integrin AVB3. Supporting a role for TWEAK-Fn14-NF-kB in mediating integrin expression we further show that integrin AVB3, but not AVB5 or A5B1, was significantly increased in CD117+ TICs in the presence of TWEAK and was dependent on Fn14 as demonstrated through knockdown experiments. RNA-sequencing analysis of tumor cells grown in spheroid conditions with RelA or RelB shRNA knockdown relative to control, show ITGAV is dependent on RelA while ITGB3 expression is dependent on RelB. Taken together, our findings suggest CD117+ TICs have enhanced expression of AVB3 and is further induced by TWEAK mediated NF-kB activity for adaptation to a stressful post-chemotherapy environment. Ongoing in vitro and in vivo studies are investigating the therapeutic benefit of targeting AVB3 or TWEAK in CD117+ TICs for prevention of adhesion and tumor repopulation following chemotherapy. These studies will uncover interactions between TICs and ECM through their unique expression of integrins and clarifies the role of NF-kB transcription factors RelA and RelB in regulating TIC survival in the post-chemotherapy TME. Citation Format: Omar Lujano Olazaba, Mikella Robinson, Cecilia Gallo, Mena Shammas, Katelyn Shelby, Luisjesus Cruz, Samuel F. Gilbert, Carrie D. House. The TWEAK-Fn14-NF-kB signaling axis enhances expression of integrin AVB3 in a post-chemotherapy tumor microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4264.

Abstract 298: Contribution of cancer associated fibroblasts to treatment response and resistance in high-risk multifocal prostate cancer

Abstract Patients with multifocal, locally advanced prostate cancer are at the highest risk of recurrence and death. Neoadjuvant therapies have limited response rates in this patient cohort and there remains a critical need to develop curative treatments. Prostate cancer (PC) shows marked heterogeneity, which is not limited to tumor cells but extends across immune and stromal compartments and has been linked to metastatic disease and therapeutic resistance. Cancer associated fibroblasts (CAFs) have been proposed to play a critical role in PC progression and invasion, but it is still unclear how CAFs interact with tumor cells. We evaluated CAFs from prostatectomy specimens as drivers of resistance to neoadjuvant androgen receptor signaling inhibitor (ARSi) and chemotherapy and assessed treatment response in co-culture systems in the context of CAF presence. We have utilized a novel radio-pathology tool that integrates PSMA PET/MRI scans to identify regions of interest that associate with heterogenous treatment response. MRI scans were utilized to print patient-specific 3D molds to micro dissect live tissue for subsequent cell sorting and molecular analyses. We have established a multi-parameter flow cytometry panel to characterize primary prostate fibroblasts from tumor foci and adjacent normal prostate tissue. In addition, co-culture assays were performed including immortalized CAF (hPrCSC-44) and MSC-derived fibroblast. To examine the tumor-promoting role of CAFs, fibroblasts were co-cultured with androgen-sensitive 3D tumor PC spheroids (LNCaP, C42B and LAPC4) in the integrated microfluidic STACKs platform that allows for the assessment of spatio-temporal interaction in a variety of culture microenvironments to model complex interactions. Cytotoxicity in 3D PC spheroids was assessed after Apalutamide, Darolutamide, or Docetaxel treatment by confocal microscopy. Integrated DNA, RNA, and radiology analysis of tumor biopsies have identified gene signatures associated with early biochemical recurrence. Most biopsies had low tumor purity, evidence of the cytotoxic impact of chemohormonal therapy while high tumor purity predicted high tumor expression. Poor clinical outcome was significantly associated with elevated stromal scores while epithelial and stromal enrichment were inversely correlated (rho= -0.80, p< 5 × 10−16). The presence of CAFs significantly decreased Docetaxel (C42B only vs C42B+CAF, 86.07% vs. 44.57%, respectively, p<0.001; LAPC4 only vs LAPC4+CAF, 70.98% vs 26.64%, respectively, p<0.001) and Darolutamide-induced (C42B only vs C42B+CAF, 52.59% vs 37.87%, respectively, p=0.0407). In conclusion, recurrent PC is linked to increased stromal content and the presence of CAFs supported tumor survival in response to ARSi and Docetaxel treatment. The molecular drivers of this phenomenon are currently being investigated. Citation Format: Nikolett Lupsa, Erika Heninger, Adeline Ding, Shannon R. Reese, Xavier T. Hazelberg, Aaron M. LeBeau, Brian P. Johnson, Peter P. Geiger, David J. Beebe, David A. Quigley, Joshua M. Lang. Contribution of cancer associated fibroblasts to treatment response and resistance in high-risk multifocal prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 298.

Abstract 3771: 3D correlated multi-scale cellular and molecular architecture of solid tumors

Abstract A solid tumor presents heterogeneous 3D structures at tissue, cell, and molecule levels. To develop advanced cancer diagnostics and therapies, it is important to understand in-depth cellular and molecular architecture of a tumor at multiple scales. Here we introduce novel optical microscopy methodologies for in-depth 3D spatial analysis of tumor tissues: 1) LED photobleaching-mediated 3D multiplex immunofluorescence (IF) microscopy, 2) Correlated 3D multi-resolution optical microscopy, and 3) 3D tissue region-based cell type-selective multi-omics. We built a high-power LED light irradiator with a broad emission spectrum (490-700 nm) that bleaches a broad wavelength of fluorescence signals simultaneously throughout an optically cleared, 400 µm-thick tumor tissue. Cyclic workflow involving IF staining, optical tissue clearing, 3D confocal microscopy, and LED photobleaching enables visualization of multiple cell types in the same whole 400 µm-thick tumor tissue after computational 3D image registration. The constructed multiplex image offers comprehensive 3D cellular landscape of a tumor. We also developed a method for tracking regions of interest (ROIs) in the middle of a large tumor tissue using a UV-activated visible dye in light sheet microscopy. Light sheet microscopy allows imaging an optically cleared large-sized tumor tissue at a low spatial, tissue resolution. To mark the ROI positions on a tumor, the dye is infused into agarose gel surrounding the tumor tissue and be activated and present purple-colored lines where are exposed to UV light sheets in light sheet microscopy. The tumor tissue was physically sectioned at the marked ROI positions using a vibratome after reversing tissue clearing process. Sequential IF staining and high-resolution confocal microscopy permit for visualizing detailed cellular compositions of the ROIs in the tumor. The tissue and cellular-resolution 3D images from light sheet and confocal microscopy are correlatively integrated to display ‘Google Earth’-like multi-scale view of a tumor tissue. To further analyze the ROIs of a tumor at molecular level, we have developed optical microscope-mediated fluorescence lithography methods that allow fluorescently labeling of a selected cell type in ROIs of an optically cleared 3D tumor tissue. After wash free from clearing solution and dissociation of the tumor into cells, fluorescence-marked, intact cells from the ROIs are collected through flow cell sorting. Proteins and mRNAs are extracted from the collected cells and sent to LC-MS/MS analysis and RNA sequencing. The spatial multi-omics method provides in-depth molecular information of selected cells in the ROIs of a tumor. We believe all these 3D microscopy methods will provide powerful spatial analysis tools for better understanding cellular and molecular architecture of solid tumors. Citation Format: Steve Seung-Young Lee, Yi-Chien Wu, Jingtian Zheng, Elie Abi Khalil, Samuel Wang, Alexander Lippert, Joshua Plank. 3D correlated multi-scale cellular and molecular architecture of solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3771.

Abstract 3523: Self-supervised foundation model captures high-dimensional morphology data from single cell brightfield images

Abstract Our knowledge of cancer biology has advanced with the characterization of distinct cell types and cell states within heterogeneous tumor environments. Many methods for imaging and sorting tumor cells require biomarker labels that alter cell characteristics and create a selection bias. The use of label-free single-cell methods should further improve cancer studies involving viable, unperturbed cells for downstream assays such as RNA-Seq, cell culture expansion, and functional assays. To eliminate biomarkers and enable broader assessment of cells, we developed REM-I, a platform that characterizes and sorts unlabeled single cells based on high-dimensional morphology. Cells are captured with brightfield imaging and processed in real-time by self-supervised deep learning models to generate quantitative AI embeddings representative of cell morphology. A significant technical challenge in building REM-I was developing an AI model based on extracted features from cell images without prior knowledge of cell types, cell preparation, or other application-specific knowledge. Accordingly, we developed the Human Foundation Model (HFM), a hybrid architecture that combines self-supervised learning (SSL) and morphometrics (computer vision) to extract 115 dimensional embeddings representing cell morphology from high-resolution REM-I cell images. SSL produces a foundation model with high generalization capacity that enables hypothesis-free sample exploration and efficient generation of application-specific models. Meanwhile, computer vision extracts features that represent measurable and interpretable concepts (e.g., cell size, shape, texture, intensity). The training process for the HFM self-supervised backbone model utilizes the discriminatory power of supervised tasks. Using synthetic cells and three cancer cell lines, we trained and then validated the reproducibility and generalization capabilities of the resulting model. Results show combining deep learning and morphometrics improve interpretability of data and enable rapid characterization and classification of tumor cells with high accuracy. We also report on the Deepcell® Axon data suite, a tool to analyze data and customize reusable workflows. This includes the ability to store and manage data, visualize high dimensional data as low-dimensional projections, and train classifiers to identify and sort cell populations. To enable compatibility with user-preferred downstream analysis pipelines, Axon provides data export options for images, plots, and embeddings. Our approach allows users of all skill levels to access and interpret AI-enabled morphological profiling. Applications of REM-I include hypothesis-free evaluation of heterogeneous tumor samples, label-free cancer cell enrichment, characterization of distinct cell states, and multi-omic integration. Citation Format: Kiran Saini, Senzeyu Zhang, Ryan Carelli, Kevin B. Jacobs, Amy Wong-Thai, Cris Luengo, Vivian Lu, Anastasia Mavropoulos, Andreja Jovic, Jeanette Mei, Thomas Vollbrecht, Stephane C. Boutet, Mahyar Salek, Maddison Masaeli. Self-supervised foundation model captures high-dimensional morphology data from single cell brightfield images [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3523.

Optoelectronic Tweezers Micro-Well System for Highly Efficient Single-Cell Trapping, Dynamic Sorting, and Retrieval.

Single-cell arrays have emerged as a versatile method for executing single-cell manipulations across an array of biological applications. In this paper, an innovative microfluidic platform is unveiled that utilizes optoelectronic tweezers (OETs) to array and sort individual cells at a flow rate of 20µLmin-1. This platform is also adept at executing dielectrophoresis (DEP)-based, light-guided single-cell retrievals from designated micro-wells. This presents a compelling non-contact method for the rapid and straightforward sorting of cells that are hard to distinguish. Within this system, cells are individually confined to micro-wells, achieving an impressive high single-cell capture rate exceeding 91.9%. The roles of illuminating patterns, flow velocities, and applied electrical voltages are delved intoin enhancing the single-cell capture rate. By integrating the OET system with the micro-well arrays, the device showcases adaptability and a plethora of functions. It can concurrently trap and segregate specific cells, guided by their dielectric signatures. Experimental results, derived from a mixed sample of HepG2 and L-O2 cells, reveal a sorting accuracy for L-O2 cells surpassing 91%. Fluorescence markers allow for the identification of sequestered, fluorescence-tagged HepG2 cells, which can subsequently be selectively released within the chip. This platform's rapidity in capturing and releasing individual cells augments its potential for future biological research and applications.

An open-source FACS automation system for high-throughput cell biology.

Recent advances in gene editing are enabling the engineering of cells with an unprecedented level of scale. To capitalize on this opportunity, new methods are needed to accelerate the different steps required to manufacture and handle engineered cells. Here, we describe the development of an integrated software and hardware platform to automate Fluorescence-Activated Cell Sorting (FACS), a central step for the selection of cells displaying desired molecular attributes. Sorting large numbers of samples is laborious, and, to date, no automated system exists to sequentially manage FACS samples, likely owing to the need to tailor sorting conditions ("gating") to each individual sample. Our platform is built around a commercial instrument and integrates the handling and transfer of samples to and from the instrument, autonomous control of the instrument's software, and the algorithmic generation of sorting gates, resulting in walkaway functionality. Automation eliminates operator errors, standardizes gating conditions by eliminating operator-to-operator variations, and reduces hands-on labor by 93%. Moreover, our strategy for automating the operation of a commercial instrument control software in the absence of an Application Program Interface (API) exemplifies a universal solution for other instruments that lack an API. Our software and hardware designs are fully open-source and include step-by-step build documentation to contribute to a growing open ecosystem of tools for high-throughput cell biology.

Dielectrophoretic cell sorting with high velocity enabled by two-layer sidewall microelectrodes extending along the entire channel

Dielectrophoresis has been widely applied in microfluidic cell separation, and it enables label-free and size-independent cell sorting based on their dielectric fingerprints which is prerequisite for downstream biomedical analysis. The linear flow velocity and cross-sectional area are the two key factors proportional to the flow rate and thus directly related to the sample processing throughput. Aiming at higher flow rate, the state-of-art techniques mostly focus on the cross-sectional area, either by using wide chamber with angled electrode digits or 3D volumetric electrodes with expanded channel depth. On the other hand, the velocity is always carefully limited because cells have very squeezed time to interact with the DEP force that drives them to migrate, and the flow drag induced by high velocity potentially poses adverse effect against the DEP-driven cell migration. In this paper, we present DEP-activated cell sorting at high-velocity being 1–2 orders of magnitude higher than the existing DEP separators, which is enabled by easily-fabricated two-layer 3D sidewall electrodes extending along the entire flow channel. Such electrodes project DEP force fully decoupled from the flow drag, and acting on cells continuously, allowing sufficient time for cell migration. The two-layer design leads to cell migration across a shorter distance between the upper and lower layers, with a unique bridge adaptor evolved at the downstream for the collection of the cells vertically separated. The DEP device achieves separation of live and dead yeast cells at high velocity up to 78.3 mm/s, more than a full order of magnitude higher compared to any reported DEP platforms. This velocity corresponds to a competitive flow rate of 3.1 mL/h given the fairly small cross-sectional area, where the averaged collection rate is over 95 %. For the separation between the live and dead Hela cells, our device enables a high velocity up to 50.5 mm/s with an averaged collection rate of ∼85 %. A high throughput of 1.79 × 1 0^(5) cells/min for the live and dead Hela cell separation has been achieved with 89.7 % collection for live cells and a velocity of 30.3 mm/s.

Genetic Knock-Ins of Endogenous Fluorescent Tags in RAW 264.7 Murine Macrophages Using CRISPR/Cas9 Genome Editing.

CRISPR/Cas9 genome editing is a widely used tool for creating genetic knock-ins, which allow for endogenous tagging of genes. This is in contrast with random insertion using viral vectors, where expression of the inserted transgene changes the total copy number of a gene in a cell and does not reflect the endogenous chromatin environment or any trans-acting regulation experienced at a locus. There are very few protocols for endogenous fluorescent tagging in macrophages. Here, we describe a protocol to design and test CRISPR guide RNAs and donor plasmids, to transfect them into RAW 264.7 mouse macrophage-like cells using the Neon transfection system and to grow up clonal populations of cells containing the endogenous knock-in at various loci. We have used this protocol to create endogenous fluorescent knock-ins in at least six loci, including both endogenously tagging genes and inserting transgenes in the Rosa26 and Tigre safe harbor loci. This protocol uses circular plasmid DNA as the donor template and delivers the sgRNA and Cas9 as an all-in-one expression plasmid. We designed this protocol for fluorescent protein knock-ins; it is best used when positive clones can be identified by fluorescence. However, it may be possible to adapt the protocol for non-fluorescent knock-ins. This protocol allows for the fairly straightforward creation of clonal populations of macrophages with tags at the endogenous loci of genes. We also describe how to set up imaging experiments in 24-well plates to track fluorescence in the edited cells over time. Key features • CRISPR knock-in of fluorescent proteins in RAW 264.7 mouse macrophages at diverse genomic loci. • This protocol is optimized for the use of the Neon transfection system. • Includes instructions for growing up edited clonal populations from single cells with one single-cell sorting step and efficient growth in conditioned media after cell sorting. • Designed for knocking in fluorescent proteins and screening transfected cells by FACS, but modification for non-fluorescent knock-ins may be possible.

Analysis of the types and functions of CD34+ cells in full-thickness skin defect wounds of normal mice and diabetic mice by single-cell RNA sequencing

Objective: To analyze the types and functions of CD34+ cells in full-thickness skin defect wounds of normal mice and diabetic mice by single-cell RNA sequencing. Methods: This study was an experimental study. The CD34+ cell lineage tracing mouse was produced, and the visualization of CD34+ cells under the fluorescent condition was realized. Six male CD34+ cell lineage tracing mice aged 7-8 weeks (designated as diabetic group) were intraperitoneally injected with streptozotocin to establish a diabetic model, and full-thickness skin defect wounds were prepared on their backs when they reached 13 weeks old. Another 6 male CD34+ cell lineage tracing mice aged 13 weeks (designated as control group) were also subjected to full-thickness skin defect wounds on their backs. On post-injury day (PID) 4, wound tissue was collected from 3 mice in control group and 2 mice in diabetic group, and digested to prepare single-cell suspensions. CD34+ cells were screened using fluorescence-activated cell sorting, followed by single-cell RNA sequencing. The Seurat 4.0.2 program in the R programming language was utilized for dimensionality reduction, visualization, and cell clustering analysis of CD34+ cell types, and to screen and annotate the marker genes for each CD34+ cell subpopulation. Kyoto encyclopedia of genes and genomes (KEGG) and gene ontology (GO) enrichment analysis was performed to analyze the differentially expressed genes (DEGs) of CD34+ fibroblasts (Fbs), smooth muscle cells (SMCs), keratinocytes (KCs), and chondrocyte-like cells (CLCs) in the wound tissue of two groups of mice for exploring cellular functions. Results: On PID 4, CD34+ cells in the wound tissue of both groups of mice were consisted of 7 cell types, specifically endothelial cells, Fbs, KCs, macrophages, T cells, SMCs, and CLCs. Among these, Fbs were further classified into 5 subpopulations. Compared with those in control group, the proportions of CD34+ endothelial cells, Fbs subpopulation 1, Fbs subpopulation 4, KCs, and CLCs in the wound tissue of mice were increased in diabetic group, while the proportions of CD34+ Fbs subpopulation 2, Fbs subpopulation 3, and SMCs were decreased. The marker genes for annotating CD34+ CLCs, endothelial cells, Fbs subpopulation 1, Fbs subpopulation 2, Fbs subpopulation 3, Fbs subpopulation 4, Fbs subpopulation 5, KCs, macrophages, SMCs, and T cells were respectively metastasis-associated lung adenocarcinoma transcript 1, fatty acid binding protein 4, Gremlin 1, complement component 4B, H19 imprinted maternally expressed transcript, Dickkopf Wnt signaling pathway inhibitor 2, fibromodulin, keratin 5, CD74 molecule, regulator of G protein signaling 5, and inducible T-cell co-stimulator molecule. KEGG and GO enrichment analysis revealed that, compared with those in control group, DEGs with significant differential expression (SDE) in CD34+ Fbs from the wound tissue of mice in diabetic group on PID 4 were significantly enriched in terms related to inflammatory response, extracellular matrix (ECM) organization, regulation of cell proliferation, and aging (with Pvalues all <0.05), DEGs with SDE in CD34+ SMCs were significantly enriched in terms related to cell migration, apoptotic process, positive regulation of transcription, and phagosome (with P values all <0.05), DEGs with SDE in CD34+ KCs were significantly enriched in terms related to mitochondrial function, transcription, and neurodegenerative diseases (with P values all <0.05), and DEGs with SDE in CD34+ CLCs were significantly enriched in terms related to rhythm regulation, ECM, and viral infection (with P values all <0.05). Conclusions: CD34+ cells display high heterogeneity in the healing process of full-thickness skin defect wounds in both normal mice and diabetic mice. The significantly enriched functions of DEGs with SDE in CD34+ cell subpopulations in the wound tissue of the two mouse groups are closely related to the wound healing process.

Lack of ATP2B1 in CD4+ T Cells Causes Colitis.

The ATP2B1 gene encodes for a calcium pump, which plays a role in removing Ca2+ from cells and maintaining intracellular Ca2+ homeostasis. Reduction of the intracellular Ca2+ concentration in CD4+ T cells is thought to reduce the severity of colitis, while elevation of Ca2+ in CD4+ T cells induces T cell hyperactivity. Our aim was to clarify the role of ATP2B1 in CD4+ T cells and in inflammatory bowel disease development. A murine CD4+ T cell-specific knockout (KO) of ATP2B1 was created using a Cre-loxP system. CD4+ T cells were isolated from thymus, spleen, and blood using fluorescence-activated cell sorting. To quantify messenger RNA levels, quantitative real-time polymerase chain reaction was performed. Although the percentages of CD4+ T cells in both KO mouse spleen and blood decreased compared with those of the control samples, both T-bet (a T helper 1 [Th1] activity marker) and GATA3 (a Th2 activity marker) expression levels were further increased in KO mouse blood CD4+ T cells (vs control blood). Diarrhea and colonic wall thickening (with mucosal changes, including crypt distortion) were seen in KO mice but not in control mice. Prior to diarrhea onset, the KO mouse colon length was already noted to be shorter, and the KO mouse stool water and lipid content were higher than that of the control mice. Tumor necrosis factor α and gp91 expressions were increased in KO mouse colon. Lack of ATP2B1 in CD4+ T cells leads to Th1 and Th2 activation, which contributes to colitis via elevation of tumor necrosis factor α and oxidative stress.