Publications

The development of retinoblastoma is thought to require pathological genetic changes in both alleles of the RB1 gene. However, cases exist where RB1 mutations are undetectable, suggesting alternative pathways to malignancy. We used whole-genome sequencing (WGS) and transcriptomics to investigate the landscape of sporadic retinoblastomas derived from twenty patients, sought RB1 and other driver mutations and investigated mutational signatures. At least one RB1 mutation was identified in all retinoblastomas, including new mutations in addition to those previously identified by clinical screening. Ten tumours carried structural rearrangements involving RB1 ranging from relatively simple to extremely complex rearrangement patterns, including a chromothripsis-like pattern in one tumour. Bilateral tumours obtained from one patient harboured conserved germline but divergent somatic RB1 mutations, indicating independent evolution. Mutational signature analysis showed predominance of signatures associated with cell division, an absence of ultraviolet-related DNA damage and a profound platinum-related mutational signature in a chemotherapy-exposed tumour. Most RB1 mutations are identifiable by clinical screening. However, the increased resolution and ability to detect otherwise elusive rearrangements by WGS have important repercussions on clinical management and advice on recurrence risks.

Multiparameter tissue imaging enables analysis of cell-cell interactions in situ, the cellular basis for tissue structure, and novel cell types that are spatially restricted, giving clues to biological mechanisms behind tissue homeostasis and disease. Here, we streamlined and simplified the multiplexed imaging method CO-Detection by indEXing (CODEX) by validating 58 unique oligonucleotide barcodes that can be conjugated to antibodies. We showed that barcoded antibodies retained their specificity for staining cognate targets in human tissue. Antibodies were visualized one at a time by adding a fluorescently labeled oligonucleotide complementary to oligonucleotide barcode, imaging, stripping, and repeating this cycle. With this we developed a panel of 46 antibodies that was used to stain five human lymphoid tissues: three tonsils, a spleen, and a LN. To analyze the data produced, an image processing and analysis pipeline was developed that enabled single-cell analysis on the data, including unsupervised clustering, that revealed 31 cell types across all tissues. We compared cell-type compositions within and directly surrounding follicles from the different lymphoid organs and evaluated cell-cell density correlations. This sequential oligonucleotide exchange technique enables a facile imaging of tissues that leverages pre-existing imaging infrastructure to decrease the barriers to broad use of multiplexed imaging.

An ever-increasing array of imaging technologies are being used in the study of complex biological samples, each of which provides complementary, occasionally overlapping information at different length scales and spatial resolutions. It is important to understand the information provided by one technique in the context of the other to achieve a more holistic overview of such complex samples. One way to achieve this is to use annotations from one modality to investigate additional modalities. For microscopy-based techniques, these annotations could be manually generated using digital pathology software or automatically generated by machine learning (including deep learning) methods. Here, we present a generic method for using annotations from one microscopy modality to extract information from complementary modalities. We also present a fast, general, multimodal registration workflow [evaluated on multiple mass spectrometry imaging (MSI) modalities, matrix-assisted laser desorption/ionization, desorption electrospray ionization, and rapid evaporative ionization mass spectrometry] for automatic alignment of complex data sets, demonstrating an order of magnitude speed-up compared to previously published work. To demonstrate the power of the annotation transfer and multimodal registration workflows, we combine MSI, histological staining (such as hematoxylin and eosin), and deep learning (automatic annotation of histology images) to investigate a pancreatic cancer mouse model. Neoplastic pancreatic tissue regions, which were histologically indistinguishable from one another, were observed to be metabolically different. We demonstrate the use of the proposed methods to better understand tumor heterogeneity and the tumor microenvironment by transferring machine learning results freely between the two modalities.

Matrix-assisted laser desorption ionization (MALDI) operated at atmospheric pressure has been shown to be a promising technique for mass spectrometry imaging of biological tissues at high spatial resolution. Recent studies have shown several orders of magnitude improvement in sensitivity afforded by coupling with a low-temperature plasma (LTP) for postionization. In this work we report the first results from "matrix-free" imaging using our atmospheric pressure (AP) transmission mode (TM) (MA)LDI source with LTP postionization. Direct MSI analysis of murine testis with no sample preparation after tissue sectioning enabled imaging of a range of lipid classes at pixel sizes of 25 μm. We compared results from the matrix-free methods with MALDI experiments in which the matrix was applied on top, underneath, or layered as a sandwich. The sandwich preparation was found to lead to ion yields approximately 2- or 3-fold higher than the other methods, indicating that the addition of a light absorbing matrix remains beneficial. Nonetheless, LDI methods confer a range of advantages, and the sensitivity improvements provided by postionization strategies are a promising step toward high-efficiency laser sampling under ambient conditions.

Methods for highly multiplexed RNA imaging are limited in spatial resolution, and thus in their ability to localize transcripts to nanoscale and subcellular compartments. We adapt expansion microscopy, which physically expands biological specimens, for long-read untargeted and targeted in situ RNA sequencing. We applied untargeted expansion sequencing (ExSeq) to mouse brain, yielding readout of thousands of genes, including splice variants and novel transcripts. Targeted ExSeq yielded nanoscale-resolution maps of RNAs throughout dendrites and spines in neurons of the mouse hippocampus, revealing patterns across multiple cell types; layer-specific cell types across mouse visual cortex; and the organization and position-dependent states of tumor and immune cells in a human metastatic breast cancer biopsy. Thus ExSeq enables highly multiplexed mapping of RNAs, from nanoscale to system scale.

Summary: In situ sequencing of physically expanded specimens enables multiplexed mapping of RNAs at nanoscale, subcellular resolution.