Protocol Improvement for RNA Extraction From Compromised Frozen Specimens Generated in Austere Conditions: A Path Forward to Transcriptomics-Pathology Systems Integration.

Nabarun Chakraborty,Hisham Nachabe,Candace Moyler,Connie W Schmitt,Rasha Hammamieh,Stephen Butler,Cary L Honnold,Aarti Gautam

doi:10.3389/fmolb.2020.00142

Nabarun Chakraborty, Hisham Nachabe + Show 6 more

Open Access

https://doi.org/10.3389/fmolb.2020.00142

Copy DOI

Abstract

At the heart of the phenome-to-genome approach is high throughput assays, which are liable to produce false results. This risk can be mitigated by minimizing the sample bias, specifically, recycling the same tissue specimen for both phenotypic and genotypic investigations. Therefore, our aim is to suggest a methodology of obtaining robust results from frozen specimens of compromised quality, particularly if the sample is produced in conditions with limited resources. For example, generating samples at the International Space Station (ISS) is challenging because the time and laboratory footprint allotted to a project can get expensive. In an effort to be economical with available resources, snap-frozen euthanized mice are the straightforward solution; however, this method increases the risk of temperature abuse during the thawing process at the beginning of the tissue collection. We found that prolonged immersion of snap frozen mouse carcass in 10% neutral buffered formalin at 4°C yielded minimal microscopic signs of ice crystallization and delivered tissues with histomorphology that is optimal for hematoxylin and eosin (H&E) staining and fixation on glass slides. We further optimized a method to sequester the tissue specimen from the H&E slides using an incubator shaker. Using this method, we were able to recover an optimal amount of RNA that could be used for downstream transcriptomics assays. Overall, we demonstrated a protocol that enables us to maximize scientific values from tissues collected in austere condition. Furthermore, our protocol can suggest an improvement in the spatial resolution of transcriptomic assays.

Highlights

The value added from big data is in generating information and, knowledge by integrating a large amount of reads derived from multiple disparate assay types (Marx, 2013a; Andreu-Perez et al, 2015; Gligorijevicet al., 2016; Huang et al, 2017)
Dissecting a partially damaged whole tissue into multiple fragments for gene expression and histopathological assessment may confound the analysis, if a pathological lesion does not extend uniformly across all the fragments, or if there are subtle spatial variations in the tissue, as has been shown for tumors (Yan et al, 2015). This drawback attributed to the intrinsic heterogeneity of tissues that can potentially be mitigated by integrating the omics data with other “conventional” readouts (Ellinger-Ziegelbauer et al, 2011), such as histopathology analysis
Multiple assay data generated from the same tissue specimen will improve spatial resolution of tissue’s molecular landscape, enhance data integrative structure, and reduce biases in generating multidimensional information, which is the goal of making big data free from false results (Marx, 2013b; Andreu-Perez et al, 2015; Gligorijevicet al., 2016; Huang et al, 2017)

Summary

Introduction

The value added from big data is in generating information and, knowledge by integrating a large amount of reads derived from multiple disparate assay types (Marx, 2013a; Andreu-Perez et al, 2015; Gligorijevicet al., 2016; Huang et al, 2017). Dissecting a partially damaged whole tissue into multiple fragments for gene expression and histopathological assessment may confound the analysis, if a pathological lesion does not extend uniformly across all the fragments, or if there are subtle spatial variations in the tissue, as has been shown for tumors (Yan et al, 2015) This drawback attributed to the intrinsic heterogeneity of tissues that can potentially be mitigated by integrating the omics data with other “conventional” readouts (Ellinger-Ziegelbauer et al, 2011), such as histopathology analysis. An omics-pathology integration approach could be the most effective for phenometo-genome interpretation if omics assays are conducted using the particular tissue specimen, where injury signatures are informed by histopathology image analysis (Pathak and Dave, 2014; Yu et al, 2017) This approach would be an ideal phenome-togenome approach, where omics readouts could be directly informed by the visualized phenotypes (Yee, 2009; Gligorijevicet al., 2016). Multiple assay data generated from the same tissue specimen will improve spatial resolution of tissue’s molecular landscape, enhance data integrative structure, and reduce biases in generating multidimensional information, which is the goal of making big data free from false results (Marx, 2013b; Andreu-Perez et al, 2015; Gligorijevicet al., 2016; Huang et al, 2017)

Objectives

Methods

Results