Spatial Organization Of Cells Research Articles

Abstract P20: Unveiling the Influence of Tumor Microenvironment and Spatial Heterogeneity on Temozolomide Resistance in Glioblastoma using an Advanced Human in vitro Model of the Blood-brain Barrier and Glioblastoma

Abstract Glioblastoma (GBM) is the most common primary malignant brain cancer in adults, with a dismal prognosis of 15 months. This is despite decades of research and improvements in surgery and radiotherapy. Temozolomide (TMZ) is the only first-in-line chemotherapeutic; however, resistance is frequent and multifactorial. While many molecular and genetic factors have been linked to TMZ resistance, the role of the solid tumor morphology and the tumor microenvironment (TME), particularly the blood-brain barrier (BBB), is unknown. Complex 3D in vitro models are able to recapitulate aspects of clinical disease such as the tumor microenvironment, such that research into the role of the TME and how it affects tumor biology and therapy resistance can be studied more easily and accurately. We have developed a complex 3D in vitro model for GBM and its surrounding BBB, that recapitulates important clinical features such as a dense tumor core with tumor cells that invade along the perivascular space; and a perfusable BBB with a physiological permeability with morphology that is altered in the presence of a tumor spheroid. We demonstrate that TMZ sensitivity decreases with increasing cancer cell spatial organization, and that the BBB can contribute to TMZ resistance. Proteomic analysis of cultured microtissues using low volume sample workflows revealed proteins linked to aggressive tumors and TMZ resistance. Notably, proteomic analysis revealed a potential role for cell cycle proteins in TMZ resistance. Preliminary data with tumor spheroids formed by patient derived GBM progenitor cells revealed a similar trend of an increase in proteins associated with tumor aggressiveness upon co-culture with a BBB. The data demonstrate the utility of combining complex physiological in vitro models and next generation mass spectrometry for interrogating the tumour microenvironment, tumor biology and accurate in vitro therapy validation. Citation Format: Maxine SY Lam, Joey JY Aw, Damien Tan, Ragavi Vijayakumar, Hui Yi Grace Lim, Swathi Yada, Qing You Pang, Nick Barker, Carol Tang, Beng Ti Ang, Radoslaw M. Sobota, Andrea Pavesi. Unveiling the Influence of Tumor Microenvironment and Spatial Heterogeneity on Temozolomide Resistance in Glioblastoma using an Advanced Human in vitro Model of the Blood-brain Barrier and Glioblastoma [abstract]. In: Proceedings of Frontiers in Cancer Science; 2023 Nov 6-8; Singapore. Philadelphia (PA): AACR; Cancer Res 2024;84(8_Suppl):Abstract nr P20.

First person – Anne Rosfelter

ABSTRACT First Person is a series of interviews with the first authors of a selection of papers published in Journal of Cell Science, helping researchers promote themselves alongside their papers. Anne Rosfelter is first author on ‘ Reduction of cortical pulling at mitotic entry facilitates aster centration’, published in JCS. Anne conducted the research described in this article while a PhD student in Alex McDougall's lab at Laboratoire Biologie du Développement (LBDV), Villefranche-sur-Mer, France. She is now a post-doctoral researcher in the lab of Yu-Chiun Wang at the RIKEN Center for Biosystems Dynamics Research (BDR), Kobe, Japan, investigating the spatial organization of cells and tissue through studying the cytoskeleton.

Differential proteomics profile of microcapillary networks in response to sound pattern-driven local cell density enhancement

Spatial cell organization and biofabrication of microcapillary networks in vitro has a great potential in tissue engineering and regenerative medicine. This study explores the impact of local cell density enhancement achieved through an innovative sound-based patterning on microcapillary networks formation and their proteomic profile. Human umbilical vein endothelial cells (HUVEC) and human pericytes from placenta (hPC-PL) were mixed in a fibrin suspension. The mild effect of sound-induced hydrodynamic forces condensed cells into architected geometries showing good fidelity to the numerical simulation of the physical process. Local cell density increased significantly within the patterned areas and the capillary-like structures formed following the cell density gradient. Over five days, these patterns were well-maintained, resulting in concentric circles and honeycomb-like structures.Proteomic analysis of the pre-condensed cells cultured for 5 days, revealed over 900 differentially expressed proteins when cells were preassembled through mild-hydrodynamic forces. Gene ontology (GO) enrichment analysis identified cellular components, molecular functions, and biological processes that were up- and down-regulated, providing insights regarding molecular processes influenced by the local density enhancement. Furthermore, we employed Ingenuity Pathway Analysis (IPA) to identify altered pathways and predict upstream regulators. Notably, VEGF-A emerged as one of the most prominent upstream regulators.Accordingly, this study initiates the unraveling of the changes in microcapillary networks at both molecular and proteins level induced by cell condensation obtained through sound patterning. The findings provide valuable insights for further investigation into sound patterning as a biofabrication technique for creating more complex microcapillary networks and advancing in vitro models.

Abstract 1145: Comparative spatial transcriptomic analysis of tumor microenvironment responses to conventional chemotherapy and immunotherapy in triple-negative breast cancer

Abstract Background: In triple negative breast cancer (TNBC), pembrolizumab in combination with cytotoxic chemotherapy has proven its efficacy and improved survival outcome. Previous studies demonstrated that PD-L1 expression, tumor mutational burden as well as certain neo-antigens were associated with response to pembrolizumab but there were no biomarkers predicting the response of pembrolizumab of TNBC in neoadjuvant setting. Therefore, tumor microenvironment (TME) evaluation would be warranted to understand the mechanism of action of pembrolizumab in TNBC. Methods: We employed the Xenium In Situ platform, a cutting-edge spatial transcriptomics tool, capable of mapping hundreds of transcripts within individual cells. Using a 380-plex gene panel, we analyzed formalin-fixed, paraffin-embedded TNBC sections treated with weekly paclitaxel with carboplatin followed by doxorubicin with cyclophosphamide (CTx) with/without pembrolizumab in neoadjuvant setting. We analyzed detailed cellular compositions of tumors, encompassing immune and stromal cells, across over one million cells and we attempted to identify various cell types and states, previously characterized through single-cell RNA sequencing, and assess their spatial heterogeneity within the TME with regard to treatment strategy and the response to treatment. Results: We collected 19 serial samples from 13 patients including 12 pre-treatment and seven post-treatment samples. Among seven post-treatment samples, one was collected after three weeks of the first treatment and one was after disease relapse. Of 13 patients, seven patients had received pembrolizumab with cytotoxic chemotherapy and four out of seven have achieved pathologic complete response (pCR). In six patients treated with only cytotoxic chemotherapy, only two achieved pCR. Our analysis revealed significant alterations in the spatial organization of cells in both the pembrolizumab with CTx and CTx-only groups. Notably, an increase in CD8 T cells and plasma cells following treatments was pronounced in post-treatment samples from patients in the pembrolizumab with CTx group. We also observed treatment-specific variations in ligand-receptor interactions between immune cells and cancer cells. Conclusion: This study provides an in-depth analysis of the TME in TNBCs in response to pembrolizumab with CTx. Utilizing the Xenium In Situ platform, we uncovered significant disparities in the immune cell population according to the application of pembrolizumab. Our findings underscore the complexity of TME responses and highlight the potential of spatial transcriptomics in informing tailored treatment strategies for TNBC. Citation Format: Do-Eon Gu, Jiyeon Hyeon, Eun Seop Seo, Hae Hyun Jung, Jonghan Yu, Seok Won Kim, Young-Hyuck Im, Ji-Yeon Kim, Woong-Yang Park. Comparative spatial transcriptomic analysis of tumor microenvironment responses to conventional chemotherapy and immunotherapy in triple-negative breast 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 1145.

Abstract 7394: Delineating spatial expression signatures of lung adenocarcinoma subtypes from spatial single-cell transcriptomics using graph neural networks

Abstract Lung Adenocarcinoma (LUAD) subtypes including complex acinar (CA), micropapillary, and solid regions are characterized by distinct morphological and biological features that bear significant implications for prognosis and treatment strategies. While histology-based subtype prediction has provided a structural understanding, it overlooks crucial biological insights related to molecular mechanisms and the influences of tumor microenvironments (TMEs). Recent advances in spatially resolved single-cell profiling enable precise measurements of gene expression at the cellular and sub-cellular levels within intact tissues. We hypothesize that spatial gene expression (SGE) patterns can effectively identify LUAD subtypes and characterize the influence of TMEs. Thus, to address the limitations of histology-based subtyping of LUAD, we developed an AI-driven approach that leverages SMI samples to predict LUAD subtypes and elucidate their underlying molecular landscape and associated TMEs. For this study, we leverage the NanoString CosMx Spatial Molecular Imaging (SMI) NSCLC FFPE datasets comprising of 8 LUAD tissue samples, annotated by pathologists at UPMC for subtype identification. The dataset consists of approximately 98k cells per sample with 18 cell types and 960 genes per cell. We encoded the spatial organization of cells as a graph with nodes representing cells with expression features and edges based on k-nearest neighbors of 30 cells. We considered the prediction of micropapillary, CA, and solid for each tumor cell and developed a Graph Convolutional Network (GCN) with semi-supervised training to account for the unlabeled nodes. A subgraph splitting strategy with cross-validation was developed to train on localized spatial regions and ensure balanced label proportions in each subgraph. Our GCN achieved 92.63% test accuracy, surpassing non-spatial deep learning models (77.45%). Notably, explanation analysis using the GNNExplainer algorithm of our trained model identified high-attribution cells in CA and solid TME, revealing unique differentially expressed genes (DEGs) - JUNB in CA, and VIM, GPX1, PSAP, IFI27 in solid tumors, compared to low-attribution cells within the same subtype. Further analysis revealed an enrichment of EMT and interferon alpha/gamma, suggesting a more inflamed phenotype than low-attribution cells. Differential expression analysis of high-attribution CA vs solid cells identified significant upregulation of NDRG1 in CA and COL1A1 in solid. Functional enrichment analysis of these DEGs revealed PI3k-Akt in CA and MAPK signaling pathway in solid, suggesting an aggressive solid TME and CA with high survival and metastasis. We further identified a distinct neutrophil-enriched TME associated with CA and revealed cell-cell communication through MIF interactions with the CD74/CXCR4 and CD44/CD74 axes. Citation Format: Arun Das, Md Musaddaqul Hasib, Zhentao Liu, Wen Meng, Yu-Chiao Chiu, Gabriel Sica, Shou-Jiang Gao. Delineating spatial expression signatures of lung adenocarcinoma subtypes from spatial single-cell transcriptomics using graph neural networks [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 7394.

Abstract 3648: The spatial organization of cells expressing MYC and BCL2 affects immune microenvironment composition and prognosis in DLBCL

Abstract Point process analyses are widely used in ecology and geography to understand the impact and relevance of spatial distribution of points in a complex system. To date, only few studies have explored point process analyses in the context of tumor heterogeneity. Malignancies often show variable patterns of cellular distribution, and the relationship of these topographic variables with underlying biological processes and clinical outcomes is not well understood. Here, we performed spatial point process analyses in diffuse large B-cell lymphoma (DLBCL), leveraging advances in multiplexed immunohistochemistry and cellular phenotyping, with downstream multi-omic and clinicopathological analyses. We have demonstrated that survival in DLBCL is strongly associated with the fraction of malignant cells showing co-expression of the oncogenes MYC and BCL2 in the absence of BCL6 (M+2+6-). These cells display non-random spatial organization within tumor infiltrates, deviating from a poisson distribution. We therefore modeled the spatial organization of M+2+6- cells within DLBCL, using x-y coordinate information from multiplexed fluorescent immunohistochemistry (mfIHC) in four independent cohorts (N=449 patients). We derived spatial point patterns, upon which Geyers point process models were applied. We observed that patients could be divided into two groups based on these: one group showed “clustered” spatial organization, while the other displayed a “dispersed” M+2+6- cell distribution. Cases with predominantly “dispersed” M+2+6- cells showed shorter overall survival in all analyzed cohorts (P < 0.05 in 4/4 cohorts). We performed multi-omic analyses to identify potential biological explanations for this association between topography and clinical outcome. Using digital spatial profiling of the transcriptome from CD20, CD3 and CD68 compartments in 47 DLBCL samples, we observed that lymphomas with an M+2+6- ‘dispersed’ phenotype had an immunologically cold microenvironment enriched in Tregs and exhausted CD4+ and CD8+ T cells. Through an integration with single-cell transcriptomic analyses (N=22 samples) we observed that malignant B-cells from cases with M+2+6- ‘dispersed’ phenotype expressed genes implicated in cell migration and adhesion, but also in potentially targetable immune checkpoints such as LAG3. In summary, by integrating multiplexed immunophenotyping and point process analysis with molecular profiling in the setting of DLBCL, we show that a dispersed pattern of distribution of the M+2+6- subclonal fraction uniquely confers poor prognosis, and associates with a potentially reversible immune cold microenvironment composition. This represents the first demonstration that the spatial distribution of malignant cell subpopulations in lymphoma can embody biological and clinical significance. Citation Format: Shruti Shridar, Michal Marek Hoppe, Patrick Jaynes, Gayatri Kumar, Siddham Jasoria, Ziwei Meng, Vaibhav Rajan, Kasthuri Kannan, Claudio Tripodo, Anand Devaprasath Jeyasekharan. The spatial organization of cells expressing MYC and BCL2 affects immune microenvironment composition and prognosis in DLBCL [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 3648.

Deciphering metabolic crosstalk in context: lessons from inflammatory diseases.

Metabolism plays a crucial role in regulating the function of immune cells in both health and disease, with altered metabolism contributing to the pathogenesis of cancer and many inflammatory diseases. The local microenvironment has a profound impact on the metabolism of immune cells. Therefore, immunological and metabolic heterogeneity as well as the spatial organization of cells in tissues should be taken into account when studying immunometabolism. Here, we highlight challenges of investigating metabolic communication. Additionally, we review the capabilities and limitations of current technologies for studying metabolism in inflamed microenvironments, including single-cell omics techniques, flow cytometry-based methods (Met-Flow, single-cell energetic metabolism by profiling translation inhibition (SCENITH)), cytometry by time of flight (CyTOF), cellular indexing of transcriptomes and epitopes by sequencing (CITE-Seq), and mass spectrometry imaging. Considering the importance of metabolism in regulating immune cells in diseased states, we also discuss the applications of metabolomics in clinical research, as well as some hurdles to overcome to implement these techniques in standard clinical practice. Finally, we provide a flowchart to assist scientists in designing effective strategies to unravel immunometabolism in disease-relevant contexts.

Automatic 3D cell segmentation of fruit parenchyma tissue from X-ray micro CT images using deep learning

BackgroundHigh quality 3D information of the microscopic plant tissue morphology—the spatial organization of cells and intercellular spaces in tissues—helps in understanding physiological processes in a wide variety of plants and tissues. X-ray micro-CT is a valuable tool that is becoming increasingly available in plant research to obtain 3D microstructural information of the intercellular pore space and individual pore sizes and shapes of tissues. However, individual cell morphology is difficult to retrieve from micro-CT as cells cannot be segmented properly due to negligible density differences at cell-to-cell interfaces. To address this, deep learning-based models were trained and tested to segment individual cells using X-ray micro-CT images of parenchyma tissue samples from apple and pear fruit with different cell and porosity characteristics.ResultsThe best segmentation model achieved an Aggregated Jaccard Index (AJI) of 0.86 and 0.73 for apple and pear tissue, respectively, which is an improvement over the current benchmark method that achieved AJIs of 0.73 and 0.67. Furthermore, the neural network was able to detect other plant tissue structures such as vascular bundles and stone cell clusters (brachysclereids), of which the latter were shown to strongly influence the spatial organization of pear cells. Based on the AJIs, apple tissue was found to be easier to segment, as the porosity and specific surface area of the pore space are higher and lower, respectively, compared to pear tissue. Moreover, samples with lower pore network connectivity, proved very difficult to segment.ConclusionsThe proposed method can be used to automatically quantify 3D cell morphology of plant tissue from micro-CT instead of opting for laborious manual annotations or less accurate segmentation approaches. In case fruit tissue porosity or pore network connectivity is too low or the specific surface area of the pore space too high, native X-ray micro-CT is unable to provide proper marker points of cell outlines, and one should rely on more elaborate contrast-enhancing scan protocols.

The diversification of methods for studying cell-cell interactions and communication.

No cell lives in a vacuum, and the molecular interactions between cells define most phenotypes. Transcriptomics provides rich information to infer cell-cell interactions and communication, thus accelerating the discovery of the roles of cells within their communities. Such research relies heavily on algorithms that infer which cells are interacting and the ligands and receptors involved. Specific pressures on different research niches are driving the evolution of next-generation computational tools, enabling new conceptual opportunities and technological advances. More sophisticated algorithms now account for the heterogeneity and spatial organization of cells, multiple ligand types and intracellular signalling events, and enable the use of larger and more complex datasets, including single-cell and spatial transcriptomics. Similarly, new high-throughput experimental methods are increasing the number and resolution of interactions that can be analysed simultaneously. Here, we explore recent progress in cell-cell interaction research and highlight the diversification of the next generation of tools, which have yielded a rich ecosystem of tools for different applications and are enabling invaluable discoveries.

Agent-based modeling of stress anisotropy driven nematic ordering in growing biofilms.

Living active collectives have evolved with remarkable self-patterning capabilities to adapt to the physical and biological constraints crucial for their growth and survival. However, the intricate process by which complex multicellular patterns emerge from a single founder cell remains elusive. In this study, we utilize an agent-based model, validated through single-cell microscopy imaging, to track the three-dimensional (3D) morphodynamics of cells within growing bacterial biofilms encased by agarose gels. The confined growth conditions give rise to a spatiotemporally heterogeneous stress landscape within the biofilm. In the core of the biofilm, where high hydrostatic and low shear stresses prevail, cell packing appears disordered. In contrast, near the gel-cell interface, a state of high shear stress and low hydrostatic stress emerges, driving nematic ordering, albeit with a time delay inherent to shear stress relaxation. Strikingly, we observe a robust spatiotemporal correlation between stress anisotropy and nematic ordering within these confined biofilms. This correlation suggests a mechanism whereby stress anisotropy plays a pivotal role in governing the spatial organization of cells. The reciprocity between stress anisotropy and cell ordering in confined biofilms opens new avenues for innovative 3D mechanically guided patterning techniques for living active collectives, which hold significant promise for a wide array of environmental and biomedical applications.

Deep learning of cell spatial organizations identifies clinically relevant insights in tissue images

Recent advancements in tissue imaging techniques have facilitated the visualization and identification of various cell types within physiological and pathological contexts. Despite the emergence of cell-cell interaction studies, there is a lack of methods for evaluating individual spatial interactions. In this study, we introduce Ceograph, a cell spatial organization-based graph convolutional network designed to analyze cell spatial organization (for example,. the cell spatial distribution, morphology, proximity, and interactions) derived from pathology images. Ceograph identifies key cell spatial organization features by accurately predicting their influence on patient clinical outcomes. In patients with oral potentially malignant disorders, our model highlights reduced structural concordance and increased closeness in epithelial substrata as driving features for an elevated risk of malignant transformation. In lung cancer patients, Ceograph detects elongated tumor nuclei and diminished stroma-stroma closeness as biomarkers for insensitivity to EGFR tyrosine kinase inhibitors. With its potential to predict various clinical outcomes, Ceograph offers a deeper understanding of biological processes and supports the development of personalized therapeutic strategies.

CNSC-44. BIOLOGICAL PATTERN DISCOVERY IN GLIOBLASTOMA

Abstract Glioblastoma(GBM - IDH wildtype) is an adult glioma, showing abysmal prognosis. Sequencing technologies show us the existence of a neurodevelopmental frame-work, upon which heterogeneous molecular patterns are overlaid. The tumor's functional capacity is the net result of extrinsic+intrinsic molecular factors interacting with each other. These factors affect how cells communicate and organize themselves into complex functional architectures that support their evolution, survival and resistance. Understanding the phenotypic significance of a tumor's architecture, will allow us to understand these processes and compute complex patterns in GBM that are "biologically-relevant". First, we showed, as a proof-of-concept, that "biologically-relevant" signatures are imprinted within the spatial organization of cells using spatial pixel analysis. We used a phase-contrast image dataset of glioblastoma stem cells grown in culture, imaged 4-12hrly, over 12-16 days. We applied 29 hand-engineered pixel features per image, deriving spatial pixel signatures for each of our 17’601 phase-contrast images. Using different computational analytical methods and gene expression from matched bulk RNA datasets, we showed that spatial pixel patterning follows biologically relevant phenotypes. We found samples of images with high PC2 scores had higher mesenchymal/microglia scores, as compared to samples of images with low PC2 scores, which showed higher neurodevelopmental signatures. In addition, we found that mathematical algorithms describing entropy, homogeneity, contrast and complexity were clearly enriched in specific biological groups. Hence, our study showed that the organizational/architectural patterns of biological entities show preservation of fundamental organizational principles. Building upon the above data (which is currently a manuscript under preparation), we will apply these principles towards pattern discovery of GBM tissues at multiple architectural hierarchies (i.e subcellular, cellular, stroma) using different GBM models, imaging and spatial transcriptomics. If structure equates to function, understanding GBM architecture will have important applications in surgical tool developments and understanding relevant mechanistic patterns in GBM.

Spatial-linked alignment tool (SLAT) for aligning heterogenous slices

Spatially resolved omics technologies reveal the spatial organization of cells in various biological systems. Here we propose SLAT (Spatially-Linked Alignment Tool), a graph-based algorithm for efficient and effective alignment of spatial slices. Adopting a graph adversarial matching strategy, SLAT is the first algorithm capable of aligning heterogenous spatial data across distinct technologies and modalities. Systematic benchmarks demonstrate SLAT’s superior precision, robustness, and speed over existing state-of-the-arts. Applications to multiple real-world datasets further show SLAT’s utility in enhancing cell-typing resolution, integrating multiple modalities for regulatory inference, and mapping fine-scale spatial-temporal changes during development. The full SLAT package is available at https://github.com/gao-lab/SLAT.

Unveiling the Influence of Tumor Microenvironment and Spatial Heterogeneity on Temozolomide Resistance in Glioblastoma Using an Advanced Human In Vitro Model of the Blood-Brain Barrier and Glioblastoma.

Glioblastoma (GBM) is the most common primary malignant brain cancer in adults with a dismal prognosis. Temozolomide (TMZ) is the first-in-line chemotherapeutic; however, resistance is frequent and multifactorial. While many molecular and genetic factors have been linked to TMZ resistance, the role of the solid tumor morphology and the tumor microenvironment, particularly the blood-brain barrier (BBB), is unknown. Here, the authorsinvestigate these using a complex in vitro model for GBM and its surrounding BBB. The model recapitulates important clinical features such as a dense tumor core with tumor cells that invade along the perivascular space; and a perfusable BBB with a physiological permeability and morphology that is altered in the presence of a tumor spheroid. It isdemonstrated that TMZ sensitivity decreases with increasing cancer cell spatial organization, and that the BBB can contribute to TMZ resistance. Proteomic analysis with next-generation low volume sample workflows of these cultured microtissues revealed potential clinically relevant proteins involved in tumor aggressiveness and TMZ resistance, demonstrating the utility of complex in vitro models for interrogating the tumor microenvironment and therapy validation.

Abstract P1185: Spatial Transcriptome Profiling In The Mouse Heart With Single-Cell Resolution

Background: The spatial organization of the diverse cells in the heart shapes biological function. For instance, during the inflammatory and fibrotic processes that are central to heart disease, many cells act mainly within their local microenvironment. Recent studies using single-cell RNA sequencing (scRNA-seq) have provided a detailed list of the cell types and states involved in development, homeostasis and in cardiovascular diseases. However, scRNA-seq requires tissue dissociation, which removes spatial information and thereby limits our ability to determine how cells interact with each other to maintain health or cause disease. As a consequence, the precise spatial organization of cell types and states in the mammalian heart remains unknown. We sought to address this knowledge gap by using multiplexed error-robust fluorescence in situ hybridization (MERFISH), a method that enables simultaneous identification and localization of single transcripts from hundreds of different genes in intact tissue sections. Methods/Results: We adapted MERFISH and developed a new single-cell image segmentation method to creating a single-cell atlas of the mouse heart. First, we designed and constructed a 128-gene MERFISH probe library including cell type marker genes and also genes covering a broad range of cardiac signaling pathways. We then optimized sample preparation, including cell boundary staining and tissue clearing. Once optimized, we used this approach to image 10 μm-thick sections from healthy mouse heart, resolving tens of thousands of cells and millions of transcripts. Our data thus constitute the first single-cell spatial transcriptome atlas of the mouse heart. Conclusion: Through our newly developed cardiac MERFISH platform, we identified tens of thousands of cells and labeled millions of individual RNA transcripts in intact sections of mouse heart. Our cellular transcriptome profile maps enable new insights into how the organization of the heart shapes its biological function.

Cellular-scale proximity labeling for recording cell spatial organization in mouse tissues.

Proximity labeling has emerged as a powerful strategy for interrogating cell-cell interactions. However, the nanometer-scale labeling radius impedes the use of current methods for indirect cell communications and makes recording cell spatial organization in tissue samples difficult. Here, we develop quinone methide-assisted identification of cell spatial organization (QMID), a chemical strategy with the labeling radius matching the cell dimension. The activating enzyme is installed on the surface of bait cells, which produces QM electrophiles that can diffuse across micrometers and label proximal prey cells independent of cell-cell contacts. In cell coculture, QMID reveals gene expression of macrophages that are regulated by spatial proximity to tumor cells. Furthermore, QMID enables labeling and isolation of proximal cells of CD4+ and CD8+ T cells in the mouse spleen, and subsequent single-cell RNA sequencing uncovers distinctive cell populations and gene expression patterns within the immune niches of specific T cell subtypes. QMID should facilitate dissecting cell spatial organization in various tissues.

Abstract 5885: A pathway-centric approach to characterising tumour heterogeneity and cell diversity across multiple cancer types

Abstract Biological tissues are composed of heterogeneous populations intricately organized in 3D architectures, yet cell type composition and spatial organization remain largely unknown for most tissue types. Single-cell sequencing analysis provides a systematic and quantitative approach to identify cell types and determine their composition in tissues. However, the spatial organization of cells and cell-cell interactions that are critical for tissue function are lost when cells are dissociated from tissue. This is especially important when working with the complex microenvironment of a tumour where such information is needed to fully understand patient prognosis. Complex diseases such as cancer can be better understood from the perspective of dysregulated pathways, rather than as a disease resulting from alterations of individual genes. Recently, spatially resolved, single-cell imaging platforms have provided a necessary solution to bridge the spatial information gap evident in previous transcriptomic methodologies. Vizgen’s MERSCOPE platform, built on multiplexed error robust in situ hybridisation (MERFISH) technology, enables the direct profiling of the spatial organisation of intact tissue with subcellular resolution. Here, we present a Pan-Cancer approach to characterise various cancers in human clinical samples with MERSCOPE. Using a 500-gene panel that includes the canonical signalling pathways of cancer, cancer type-specific genes, key immune genes, proto-oncogenes, and tumour-suppressor genes we demonstrate MERSCOPE’s ability to spatially profile gene expression across multiple tumour samples, including breast, colon, prostate, ovarian, lung, and skin cancer. To confirm the specificity and sensitivity of our pathway-focused gene panel, we assessed how tumour cell clusters in each dataset expressed different cancer type-specific markers. As expected, we found specific cancer types enriched in corresponding marker genes. To further evaluate how individual cell types in each tumour are dysregulated by cancer, we compared non-tumour cell clusters to similarly annotated clusters derived from single-cell RNA-sequencing data of normal tissue. Notably, the non-tumour cells in these datasets exhibit higher expression of stress and inflammation pathways, while immune cell clusters demonstrate higher immune activity and higher immune exhaustion compared to those in normal tissue. These results demonstrate the power of the MERSCOPE platform to generate individualised, accurate cell atlases from patient-derived tumours and to enable further insights into the relationship between genomic profiles, dysregulated pathways, and disease phenotype. Such network-centric approaches are critical for identifying genotypic causes of diseases, classifying disease subtypes, and identifying drug targets. Citation Format: Leiam Colbert, Benjamin Patterson, Cheng Yi Chen, Nicolas Fernandez, Jiang He. A pathway-centric approach to characterising tumour heterogeneity and cell diversity across multiple cancer types [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5885.

Abstract 4632: Spatial whole transcriptome analysis of differential expression for biomarker discovery in colorectal cancer

Abstract Background: Colorectal cancer (CRC) is one of the leading causes of cancer death, which is frequently caused by genetic or epigenetic changes. The study of tumor microenvironment (TME) profiling may provide a better knowledge of spatial gene expression patterns of known CRC markers, visualize the cell composition and localization, and serve to aid the understanding of the mechanisms of tumor formation. BioChain Institute, Inc. is currently investigating the TME profiling of CRC through tissue-wide, whole transcriptome analysis for the ability to discover potential therapeutic targets. Methods: BioChain’s formalin-fixed paraffin-embedded (FFPE) tissue samples from 3 primary colon adenocarcinoma tumors (PT), and their matched adjacent normal tissues (PN), were used to investigate the TME. The images of Hematoxylin and Eosin (H&E)-stained tissue sections were annotated by our pathologist. The spatial whole transcriptome was analyzed using 10x Genomics Visium Spatial Gene Expression for FFPE. This assay provided about 5,000 spatially barcoded spots with mRNA-binding oligonucleotides to capture the gene expression. The integration of the imaging and sequencing data enabled a spatially resolved transcriptome of CRC heterogeneity. Results: The annotations of both PT and PN tissue samples were found to be correlated with the spatial gene expression clustering. With the overlaying of total gene counts for each spot on the H&E-stained tissue image, about 16,000-17,000 genes were detected in each sample. The median of ~5,500 and ~1,600 genes per spot were expressed in PT and PN samples respectively. The study of gene expression levels revealed the overexpression of several well-known CRC markers (APOE, SCD, IGFBP3, TIMP1, SPARC) and the down-regulation of numerous markers (DES, CA1, KLF4) in PT samples significantly. Conclusion: Our spatial analysis results provide insight into tumor heterogeneity, spatial organization of cells within the TME and biomarker identification. BioChain is in a unique position to facilitate researchers with a plethora of oncologic and other diseased tissue types, as well as the spatial whole transcriptome analysis. Our study enables the understanding of the TME, discovery of potential diagnostic and prognostic targets, especially for the development of personalized therapy. Citation Format: Elim Cheung, Vidyodhaya Sundaram, Baowen Zhang, Lutong Zhang, Tong Lu. Spatial whole transcriptome analysis of differential expression for biomarker discovery in colorectal cancer. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4632.

Abstract 5627: Spatial protein profiling by cyclic immunofluorescence to interpret and improve bulk tumor-based predictor of response to chemotherapy with bevacizumab in neoadjuvant breast cancer treatment

Abstract Introduction: A limited number of drugs are available for use in breast cancer patients, and several are not in practical use due to the lack of adequate biomarkers. We have recently demonstrated the feasibility of using machine learning on molecular data from bulk tumor analysis to create a nine-protein signature named VEGF-inhibition Response Predictor (ViRP) for selecting BC patients for treatment with chemotherapy and bevacizumab. The ViRP score is currently being validated in the NAPEER+ clinical trial (EudraCT 2021-005850-27). Increasing evidence suggests that spatial organization of cells within the tumor microenvironment influences survival and response to therapy in numerous cancer types. In methods based on bulk tumor analysis all tumor cells are profiled simultaneously with both colocalized and distant stroma and immune cells. We are thus pursuing information on spatial organization of cellular phenotypes expressing selected cancer related proteins including our nine ViRP proteins. Methods: From the NeoAva (NCT00773695) clinical trial evaluating the effect of bevacizumab in combination with neoadjuvant chemotherapy (n=132 pts), FFPE tissue sections from patients before, during, and after treatment were made. Cyclic immunofluorescence (cyCIF) was used to profile the spatial expression of 32 cancer-signaling and 32 immune-related proteins, comprising our nine ViRP proteins, on FFPE tissue sections from selected patients (n = 20). The Galaxy-ME platform was used for image processing and downstream analysis of spatial protein profiling. Results: Use of cyCIF for spatial analysis enabled for evaluation of malignant cells in the context of surrounding microenvironmental cells, including immune cells. We found that cell type-specific protein abundance and subcellular localization formed a highly heterogenous pattern in the tissue. This was particularly evident for the nine ViRP proteins, and differences in expression between tumor cell populations will be further elucidated. Among the patients selected for cyCIF analysis, 4 were chosen based on misclassification by the ViRP signature. Ongoing studies focus on revealing spatial expression patterns to optimize the ViRP biomarker and explore why misclassification occurs. Furthermore, the observed molecular biology of the evolving tissues under treatment in responding and non-responding patients may reveal new biomarkers indicative of treatment response or resistance. Conclusion: We observe that the expression of proteins in tumor tissues is highly heterogeneous, and thus include numerous features not captured by bulk tumor analysis. Future development of new predictive tools and biomarkers that integrate molecular data which is multiparametric and spatial will set the stage for a new class of biomarkers in cancer diagnostics. Citation Format: Mads Haugland Haugen, David Kilburn, Hongli Ma, Cameron Watson, Allison Creason, Dong Zhang, Maria Aa Dahle, Ole Christian Lingjaerde, Marianne L. Smebye, Oeystein Garred, Mette S. Foersund, Mai T. Nguyen, Gunhild M. Maelandsmo, Gordon Mills, Olav Engebraaten. Spatial protein profiling by cyclic immunofluorescence to interpret and improve bulk tumor-based predictor of response to chemotherapy with bevacizumab in neoadjuvant breast cancer treatment. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5627.

Abstract 4916: Spatial analysis of local drug induced changes in tumor microenvironment predicts effective treatment combinations in breast cancer

Abstract Anticancer therapeutics primarily designed to target tumor intrinsic mechanisms may also affect components of tumor microenvironment (TME) - immune cells and non-immune stroma. Recent literature reenforces the concept that complex interactions between drugs, neoplastic cells and cells of TME determine the efficacy of anticancer therapies. Systems understanding of these interactions can serve to predict effective treatment combinations simultaneously attacking tumor cell vulnerabilities, enhancing immune surveillance, and mitigating stromal mediators of resistance. The key challenge is to find such TME-modulating combinations in a fast and more informative way. We have developed an integrated analytical platform termed Multiplex Implantable Microdevice Assay (MIMA) to rapidly decompose cancer complexity in drug response and find biomarkers with predictive value for combination therapy efficacy including immunotherapy efficacy. The system deploys a (i) miniaturized implantable microdevice for localized intratumoral drug delivery and (ii) multiplex immunostaining to measure 30+ proteins in single cells at each drug well. Computational analyses of local drug-induced changes provide information about the composition, functional state and spatial cell organization of the tumor and associated TME ultimately bringing new insights into drug mechanisms of action. We used MIMA in genetically engineered mouse models of breast cancer to evaluate effects of five targeted anticancer agents (olaparib, palbociclib, venetoclax, panobinostat, lenvatinib) and two chemotherapies (doxorubicin, paclitaxel) and predicted synergistic antitumor effects with anti-PD-1, anti-CD40, anti-CSF1R immunotherapies and vasculature modulating agents. Some of the most effective combinations were not reported before. A pan-HDAC inhibitor, panobinostat, that synergized with anti-PD-1, induced immunogenic cell death and infiltration of antigen presenting neutrophils. Further longitudinal spatial analyses revealed that mechanisms of resistance co-emerged with these response phenotypes and became prominent over time. We measured fibroblast, protumorigenic macrophage and cytotoxic B cell recruitment associated with heavy deposition of collagen VI, increased immune suppression and emergence of invasive and cancer stem cells. Combination with stroma modulating agent, losartan, improved the efficacy of panobinostat/anti-PD-1 in systemic studies implying that normalization of non-immune stroma might favorably alter aspects of TME for immune and targeted therapy efficacy. All in all, MIMA may represent a new approach to predict effective combination regimens for individual cancer patients. Extended MIMA use and computational modeling of the spatial cell patterns could provide actionable information for development of effective drug doses and schedules. Citation Format: Zuzana Tatarova, Dylan C. Blumberg, Gordon B. Mills, Lisa M. Coussens, Oliver Jonas, Joe W. Gray. Spatial analysis of local drug induced changes in tumor microenvironment predicts effective treatment combinations in breast cancer. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4916.