Laser capture microdissection for gene expression analysis

Modeling zero inflation is not necessary for spatial transcriptomics

BACKGROUND

Spatial transcriptomics are a set of new technologies that profile gene expression on tissues with spatial localization information. With technological advances, recent spatial transcriptomics data are often in the form of sparse counts with an excessive amount of zero values.

RESULTS

We perform a comprehensive analysis on 20 spatial transcriptomics datasets collected from 11 distinct technologies to characterize the distributional properties of the expression count data and understand the statistical nature of the zero values. Across datasets, we show that a substantial fraction of genes displays overdispersion and/or zero inflation that cannot be accounted for by a Poisson model, with genes displaying overdispersion substantially overlapped with genes displaying zero inflation. In addition, we find that either the Poisson or the negative binomial model is sufficient for modeling the majority of genes across most spatial transcriptomics technologies. We further show major sources of overdispersion and zero inflation in spatial transcriptomics including gene expression heterogeneity across tissue locations and spatial distribution of cell types. In particular, when we focus on a relatively homogeneous set of tissue locations or control for cell type compositions, the number of detected overdispersed and/or zero-inflated genes is substantially reduced, and a simple Poisson model is often sufficient to fit the gene expression data there.

CONCLUSIONS

Our study provides the first comprehensive evidence that excessive zeros in spatial transcriptomics are not due to zero inflation, supporting the use of count models without a zero inflation component for modeling spatial transcriptomics.

Laser Capture Microdissection of Tissue Sections for High-Throughput RNA Analysis

The heterogeneous nature of most human organs and tissues represents a common challenge when analyzing specific structures or cells. Laser capture microdissection (LCM) enables isolation of pure cells from a mixed population of cells or tissue samples via usage of laser energy. Combined with high-throughput gene or protein techniques, compartment specific analysis elucidating the role of specialized cell types in physiological or pathophysiological activity can be performed. This chapter describes the crucial steps that have to be taken into consideration when designing and conducting a LCM project. Detailed protocols describing the workflow from project planning to high-throughput analysis of LCM material used in our laboratory are provided. Routinely occurring challenges and appropriate solutions, e.g., when working with fibrotic tissue are described.

Pregnancy and spontaneous fetal loss: A pig perspective

Pigs have a unique, non-invasive epitheliochorial placenta where maternal and fetal layers lay in apposition. Indentation of fetal capillaries into the trophoblasts and maternal capillaries into the uterine epithelium reduce the distance between the fetal and maternal blood, ensuring nutrient transfer for proper conceptus development. Another unique feature of pig pregnancy is conceptus-mediated immune cell enrichment during the early stages of conceptus attachment (around gestation Day 15). This period coincides with the development of vasculature networks at the maternal-fetal interface, which is critical for successful conceptus growth. Specific chemokines, their receptors, and chemokine decoy receptor networks coordinate this immune cell enrichment and the positioning at the maternal-fetal interface. The recruited immune cells, in turn, adopt a specialized phenotype to support key processes of maternal-fetal adaptations, including tolerance to the semi-allogeneic fetus and supporting vascularization. Disturbance in coordinated cross talk between the conceptus and maternal endometrium is an important mechanism associated with spontaneous fetal loss. The exact mechanism of fetal loss is still not yet identified, although research in the last two decades point to various factors including genetics, nutrition, uterine capacity, placental efficiency, and imbalanced immune factors at the maternal-fetal interface. In this review, we summarize some of the recent advances in endometrial immune cell functions and their regulation. We also provide insights into endometrial/placental transcriptome, microRNA biology, and extravesicular transport across the maternal-fetal interface, as well as their potential implications in porcine pregnancy success or failure.

Altered expression of chemokines and their receptors at porcine maternal-fetal interface during early and mid-gestational fetal loss

Chemokines play a significant role in pregnancy, especially during embryonic attachment and placental development. During early pregnancy, immune cells are recruited extensively to the endometrium in several species including pigs. However, this recruitment is solely mediated by the presence of the conceptus in pigs making it a unique feature compared with other species (humans, primates and mice). To understand the biological significance of chemokine expression and immune cell recruitment in the context of fetal loss, we investigate a well-characterized porcine fetal loss model during the window of early pregnancy at gestational day (gd) 20 and mid-pregnancy (gd50). These periods coincide with 25-40 % of conceptus loss. Using targeted quantitative polymerase chain reaction and Western blot approaches, we screened a specific set of chemokines. Comparisons were made with endometrial lymphocytes (ENDO LY), endometrium and chorioallantoic membranes (CAM) associated with spontaneously arresting and healthy conceptus attachment sites (CAS). mRNA expression studies revealed an increased expression of CXCR3 and CCR5 in ENDO LY and of CXCL10, CXCR3, CCL5 and CCR5 in the endometrium associated with arresting CAS at gd20. DARC was decreased in the endometrium at gd50. CCL1 was increased in CAM associated with arresting CAS at gd50. Some of these differences were also noted at the protein level (CXCL10, CXCR3, CCL5 and CCR5) in the endometrium and CAM. CD45+ immunohistochemistry demonstrated a significantly higher localization in ENDO LY in the endometrium associated with healthy versus arresting counterparts. Most of these differences were observed in early pregnancy and might contribute towards a shift in immune cell functions.

Approaches for targeted proteomics and its potential applications in neuroscience

An extensive guide on practicable and significant quantitative proteomic approaches in neuroscience research is important not only because of the existing overwhelming limitations but also for gaining valuable understanding into brain function and deciphering proteomics from the workbench to the bedside. Early methodologies to understand the functioning of biological systems are now improving with high-throughput technologies, which allow analysis of various samples concurrently, or of thousand of analytes in a particular sample. Quantitative proteomic approaches include both gel-based and non-gel-based methods that can be further divided into different labelling approaches. This review will emphasize the role of existing technologies, their advantages and disadvantages, as well as their applications in neuroscience. This review will also discuss advanced approaches for targeted proteomics using isotope-coded affinity tag (ICAT) coupled with laser capture microdissection (LCM) followed by liquid chromatography tandem mass spectrometric (LC-MS/MS) analysis. This technology can further be extended to single cell proteomics in other areas of biological sciences and can be combined with other 'omics' approaches to reveal the mechanism of a cellular alterations. This approach may lead to further investigation in basic biology, disease analysis and surveillance, as well as drug discovery. Although numerous challenges still exist, we are confident that this approach will increase the understanding of pathological mechanisms involved in neuroendocrinology, neuropsychiatric and neurodegenerative disorders by delivering protein biomarker signatures for brain dysfunction.

Adaptation of Laser Microdissection Technique for the Study of a Spontaneous Metastatic Mammary Carcinoma Mouse Model by NanoString Technologies

Laser capture microdissection (LCM) of tissue is an established tool in medical research for collection of distinguished cell populations under direct microscopic visualization for molecular analysis. LCM samples have been successfully analyzed in a number of genomic and proteomic downstream molecular applications. However, LCM sample collection and preparation procedure has to be adapted to each downstream analysis platform. In this present manuscript we describe in detail the adaptation of LCM methodology for the collection and preparation of fresh frozen samples for NanoString analysis based on a study of a model of mouse mammary gland carcinoma and its lung metastasis. Our adaptation of LCM sample preparation and workflow to the requirements of the NanoString platform allowed acquiring samples with high RNA quality. The NanoString analysis of such samples provided sensitive detection of genes of interest and their associated molecular pathways. NanoString is a reliable gene expression analysis platform that can be effectively coupled with LCM.

Gene expression analysis of bone metastasis and circulating tumor cells from metastatic castrate-resistant prostate cancer patients

Background

Characterization of genes linked to bone metastasis is critical for identification of novel prognostic or predictive biomarkers and potential therapeutic targets in metastatic castrate-resistant prostate cancer (mCRPC). Although bone marrow core biopsies (BMBx) can be obtained for gene profiling, the procedure itself is invasive and uncommon practice in mCRPC patients. Conversely, circulating tumor cells (CTCs), which are likely to stem from bone metastases, can be isolated from blood. The goals of this exploratory study were to establish a sensitive methodology to analyze gene expression in BMBx and CTCs, and to determine whether the presence or absence of detectable gene expression is concordant in matching samples from mCRPC patients.

Methods

The CellSearch^® platform was used to enrich and enumerate CTCs. Low numbers of PC3 prostate cancer (PCa) cells were spiked into normal blood to assess cell recovery rate. RNA extracted from recovered PC3 cells was amplified using an Eberwine-based procedure to obtain antisense mRNA (aRNA), and assess the linearity of the RNA amplification method. In this pilot study, RNAs extracted from CTCs and PCa cells microdissected from formalin-fixed paraffin-embedded BMBx, were amplified to obtain aRNA and assess the expression of eight genes functionally relevant to PCa bone metastasis using RT-PCR.

Results

RNAs were successfully extracted from as few as 1–5 PCa cells in blood samples. The relative expression levels of reference genes were maintained after RNA amplification. The integrity of the amplified RNA was also demonstrated by RT-PCR analysis using primer sets that target the 5′-end, middle, and 3′-end of reference mRNA. We found that in 21 out of 28 comparisons, the presence or absence of detectable gene expression in CTCs and PCa cells microdissected from single bone lesions of the same patients was concordant.

Conclusions

This exploratory analysis suggests that aRNA amplification through in vitro transcription may be useful as a method to detect gene expression in small numbers of CTCs and tumor cells microdissected from bone metastatic lesions. In some cases, gene expression in CTCs and BMBxs was not concordant, raising questions about using CTC gene expression to make clinical decisions.

Electronic supplementary material

The online version of this article (doi:10.1186/s12967-016-0829-5) contains supplementary material, which is available to authorized users.