Erik Sahai

Although studies of tumor invasion have mainly emphasized the evolution of genetic and epigenetic heterogeneity, cellular behaviors also depend on the physical microenvironment. Yet, the tumor microenvironment is both complex and continuously remodeled, making it hard to rigorously test its causal influence on tumor dissemination. Here, we adopted a reductionist approach to studying tumor cell invasion in microengineered environments of tunable patterned geometries. We find that not only the presence of obstacles but more unexpectedly their spatial disorder causes a drastic shift from a collective to a single-cell mode of invasion—comparable in strength to loss of cell-cell adhesion. Combining live-imaging and perturbation experiments with minimal biophysical modeling, we demonstrate that cell dissemination results both from local geometrical constraints and a global integration of spatial disorder over time. We identify generic universality classes of invasion dynamics dependent on environmental geometry, enabling us to identify quantitative physical principles for tumor dissemination.

Background: Changes in the extracellular matrix (ECM) are a recognised feature of aggressive prostate cancer, but they are not exploited in clinical decision-making. We aimed to develop automated quantitative ECM parameters to facilitate risk stratification for localised prostate cancer. Methods: 378 quantitative ECM parameters were derived from picrosirius red-stained diagnostic prostate biopsies in a cohort of 422 patients, matched 1:1 for recurrence, recruited to the CHHiP (Conventional or Hypofractionated High Dose Intensity Modulated Radiotherapy in Prostate Cancer) trial of radiotherapy fractionation for localised prostate cancer. These ECM parameters comprehensively described fibre architecture, gaps and ECM texture. Machine learning models at the level of both individual image tiles and patients defined how ECM parameters related to tumour versus normal prostate, Gleason grade group and recurrence. Shapley analysis was used to interpret ECM feature importance and develop signatures associated with recurrence. Results: Specific ECM patterns identified tumour versus normal prostate, Gleason pattern 4 versus 3 and recurrence. ECM patterns associated with recurrence were enriched in Gleason 4+3 patients, versus Gleason 3+4 patients. Shapley analysis revealed that biopsies from patients with recurrence had smaller more elongated gaps between fibres, with finer grained ECM texture and lower ECM homogeneity than less recurrent regions. Interpretation: Quantitative automated analysis of ECM architecture can inform probability of prostate cancer recurrence after radiotherapy; Features relating to ECM gap size and texture are of particular relevance. Funding: This work was funded by Prostate Cancer Research, Cancer Research UK and the Francis Crick Institute.

High content imaging is being applied to achieve quantitative fluorescence readouts in increasingly complex 3-dimensional (3D) cell culture models such as spheroids and organoids. Compared to conventional 2D assays, 3D assays better represent biological heterogeneity but require more complex sample preparation, 3D imaging and 3D image analysis that can affect the accuracy and precision of such assays. We used spheroids formed from the NRAS-activated melanoma cell line 19161 modified to express an ERK kinase translocation reporter (KTR) as an exemplar 3D phenotypic assay carried out in 96-well plates. The spheroids were treated with the ERK activator TPA and a range of concentrations of the MEK inhibitor Binimetinib. 3D live-cell imaging with sub-cellular spatial resolution was performed using a dual-view oblique plane microscope (dOPM) – a form of single-objective light-sheet microscope – and the experiment was performed separately at 4 different institutes. The results were analysed using an identical 3D analysis pipeline and parameters. We assessed the variation in assay readout using a linear mixed effects model. Random variance at the well level was negligible (SD = 0.0048 relative to range of KTR biosensor readout at reference site of 0.17), indicating low technical noise. Treatment effects were dose-dependent and highly statistically significant compared to DMSO control across all sites (Dunnett-corrected p < 0.001). The range in KTR readout between the minimum (3.5 μM Binimetinib) and maximum (100 nM TPA) treatments varied between 59 to 96% relative to the reference site. Measured bias in KTR readout between sites was between 6 and 12% of the range of the reference site. This study quantifies the reproducibility of a 3D live spheroid-based assay employing a fluorescence biosensor requiring readout out at the per-cell level using the dOPM platform and discusses areas where experimental protocol could be improved in the future to further improve reproducibility.

BioStudies – one package for all the data supporting a study

Non-small cell lung cancer (NSCLC) exhibits disparate responses to anti-angiogenic therapies between its two major histologic subtypes, lung adenocarcinoma (LUAD) and squamous cell carcinoma (LUSC), suggesting a histotype-dependent angiogenesis regulation. Tumor-associated fibroblasts (TAFs) exhibit an activated/myofibroblast-like phenotype in NSCLC, and are emerging as major regulators of tumor progression; yet, their role in controlling angiogenesis in NSCLC remains undefined. Here we analyzed angiogenesis/hypoxia markers in NSCLC, and combined transcriptomics (bulk RNA-seq, scRNA-seq), angiogenesis arrays, genetic perturbations and functional in vitro and in vivo assays to dissect the histotype-dependent production of pro-angiogenic factors in TAFs. We observed greater angiogenesis and reduced necrosis/hypoxia in LUAD compared to LUSC across multiple patient cohorts. The LUAD-TAF secretome was primed for angiogenesis through SMAD3-dependent overproduction of key regulators, particularly TIMP-1 and VEGF-A. We also uncovered a novel function for TIMP-1 in promoting endothelial cell hyperbranching over basal VEGF signaling. Conversely, LUSC-TAFs displayed diminished angiogenic effects despite upregulating HIF-1α and a hypoxia-associated transcriptional signature, owing to their lower SMAD3 and compensatory increase in SMAD2. Our study unveils the critical influence of TAFs in shaping the distinct angiogenic landscapes in LUAD and LUSC through the opposing SMAD2/3 regulation of TIMP-1, VEGF-A and hypoxia signaling. These results also highlight the therapeutic potential of targeting stromal SMAD3/TIMP-1 in LUAD or microenvironmental stressors such as hypoxia and acidosis in LUSC. In addition, these findings provide a biological framework for understanding the histotype-dependent patterns of dissemination, immune evasion, and response to anti-angiogenic therapies in NSCLC.

The bone marrow haematopoietic niche is composed of a diverse array of cell types and extracellular matrix components that together support healthy haematopoiesis. However, live imaging of the bone marrow microenvironment is hampered by tissue accessibility limitations. Using intravital imaging through a titanium imaging window, we investigated the dynamics of human haematopoietic cells and mesenchymal stromal cells within an ectopically implanted humanised scaffold in an immunodeficient murine host. These cell populations expand and differentiate over time, accompanied by progressive remodelling of the scaffold. We observe migration of murine endothelial cells into the scaffold, leading to the formation of a vascular network during the initial development of the humanised niche. Subsequently, the dense collagen matrix that makes up the implanted niche is altered and larger gaps form in regions populated by mesenchymal stroma cells. Collectively, our findings demonstrate dynamic remodelling of the extracellular milieu that supports haematopoietic cell development and establish a platform for longitudinal, in vivo investigation of these processes. Altogether, we describe a novel model that aligns with the 3R guiding principles and enables real-time assessment of bone marrow cell dynamics in vivo.

Oblique plane microscopy (OPM) is a form of light-sheet fluorescence microscopy (LSFM) employing a single microscope objective at the sample for both fluorescence excitation and detection. Dual-view OPM (dOPM) is an optically folded form of OPM. We present an improved dOPM system employing a 60×/1.2NA water immersion primary objective and measure the spatial resolution and fluorescence collection efficiency for illumination angles of 35° and 45° with respect to the coverslip. Illumination at 35° provides slightly better lateral resolution and collection efficiency. Collection efficiency measurements are compared to a full vectorial raytracing simulation of the system. Using a light-sheet angle of 35°, the median bead FWHM for 100 nm diameter fluorescent beads in x, y, and z and the optical sectioning strength were measured over a volume of 100 × 100 × 100 μm3 to be 0.29, 0.31, 0.83, and 2.45–3.00 μm, respectively when the two dOPM views are fused. We demonstrate less photobleaching in time-lapse dOPM of live mEmerald-expressing organoids compared to widefield epi-fluorescence z-stack imaging under the condition of equal detected fluorescence signal from a point object in focus. We demonstrate dOPM for multifield-of-view 3D imaging of biological samples in 96-well plates and apply it to imaging cells in collagen gel and quantifying the FUCCI cell-cycle reporter to provide drug dose–response curves in spheroids. We also use it to perform time-lapse multifield-of-view imaging and demonstrate the detection of organoid lumen closure and reopening, organoid migration within a collagen gel and observing dynamic events in arrays of ex vivo tissue slices.

Resistance to targeted therapies develops in the majority of patients, with numerous different mechanisms having been documented. This is frequently linked to intra-tumour heterogeneity and cancer cells transitioning into different states. In this study, we demonstrate cooperativity between co-occurring resistant states in a single tumour. Using BRAF mutant melanoma as a paradigm, we generate three different resistant states within a single model and demonstrate that they have varying differentiation states, capacities to migrate, and have very few common additional therapeutic vulnerabilities. Through a combination of experiments, including using Cre-mediated recombination to generate heterogeneity in existing tumours, and in silico modelling, we show that intra-tumour heterogeneity is the most favoured state for therapy resistant tumours. This is under-pinned by signalling between different melanoma states, with YAP1 active cells providing supporting signals for other cells but inhibiting their own proliferation. Optimal disease control requires targeting both the YAP1 active cell state and the inter-cellular communication networks. We identify the histone demethylase inhibitor GSK-J4 as being particularly effective in targeting both features of resistant tumours, and demonstrate its ability to control melanoma with multiple concurrent resistance mechanisms.

Summary: Stromal barriers exclude CD8+ T cells from accessing cancer cells and hamper immune-mediated tumour control. Through multi-pronged analysis of tumours that transition from immune inflamed to immune excluded, we reveal that the formation of stromal barriers is associated with the acquisition of neuronal gene expression programmes in cancer-associated fibroblasts (CAFs), including TUBB3 expression. This leads to neuronal mimicry, with stromal barrier formation underpinned by coordinated transient bursts of intracellular calcium release, similar to those observed in neuronal tissue. Blockade of calcium release through either pharmacological or molecular interventions, such as STC2 depletion, prevents CAF alignment and the build-up of CD8+ T cells at stromal boundaries. Nintedanib treatment prevents neuronal mimicry and restores immune-mediated tumour control. Thus, we uncover unexpected mimicry of neuronal behaviour in CAFs, document the mechanism by which it leads to immune exclusion, and identify ways to prevent the induction of neuronal mimicry and restore immune-mediated tumour control. Graphical

Radiotherapy is known to cause changes in the tumour stroma which can undermine treatment efficacy. Our understanding of this process has historically centred around effects driven by Transforming Growth Factor-beta (TGF-β) and alpha-smooth muscle actin (α-SMA)+ fibroblasts. Here, we identified a rapid expansion of podoplanin (PDPN)+ fibroblasts following radiotherapy in breast, head and neck and melanoma tumours. This fibrosis was not dependent on TGF-β, but was downstream of a radiotherapy-induced adaptive immune response. CD8+ T-cells entering the tumour after radiation were sequestered at the interface between residual tumour cells and PDPN+ fibroblasts and failed to enter the tumour core. Genetic deletion of PDPN in fibroblasts impacted their cytoskeleton and ability to organise extracellular matrix. This was associated with increased CD8+ T-cell entry and spontaneous tumour regression. Overall, we identify a mechanism whereby PDPN+ fibrosis limits immune-mediated radiation cell kill and demonstrate that disruption of PDPN signalling favours tumour control. Significance: In this study we show that rapid podoplanin (PDPN)+ fibroblast expansion following radiotherapy limits immune-mediated radiation cell kill. Targeting PDPN and associated downstream signalling improves tumour control and is a promising strategy in combination with radiotherapy. Graphical