Robust detection of RNA biomarkers for drug safety assessment in preclinical animal models by fully automated RNAscope ® 2 . 5 LS Assay

Robust assays to evaluate biomarkers in tissue are needed for preclinical safety assessment and toxicity studies. Here we present the application of the fully automated RNAscope® 2.5 LS Assay on Leica BOND RX for RNA in situ hybridization in formalin-fixed paraffin-embedded (FFPE) tissues from three commonly used animal models (rat, cynomolgus monkey, and dog). We demonstrate robust assay performance with high signal-tonoise ratio and well-maintained morphology in 25 different tissues from each species. Based on these tests, we provide recommendations for proper control gene for sample qualification of each tissue type, as well as optimal pretreatment protocol selection. For specific target RNA markers, we successfully detected cell type specific markers such as CD31 (PECAM1) and CD68, proliferation marker Ki-67 (MKI67), and cell cycle marker Cyclin E1 (CCNE1), as well as apoptosis-related molecules Puma (BBC3), Fas (CD95), and DR5 (TNFRSF10B). This study demonstrates that the RNAscope® 2.5 LS Assay can be an attractive platform for biomarker analysis in tissues for preclinical safety assessment and general animal studies. Introduction Preclinical drug safety assessment in animal models has been well established as a routine laboratory practice to evaluate the pathological alterations induced by novel therapeutic agents1. This preliminary evaluation serves a major role in the development of new treatments prior to trials in humans. Histopathological techniques have been traditionally applied for general morphological evaluation by hematoxylin and eosin (H&E) stain, and also for evaluation of specific biomarkers by immunohistochemistry (IHC). IHC assay has been commonly applied to assess therapeutic targets and toxicity-related biomarkers. However, consistent and systematic application of IHC techniques has been hindered by inconsistent performance of various antibody clones, time-consuming antibody development and validation, and general lack of reagents for some animal models. RNA in situ hybridization (ISH) technology presents an attractive alternative method for pathological evaluation of biomarkers in tissues from various preclinical animal models, because nucleic acid-based probes specific for any biomarkers associated with drug toxicity or mechanism can be developed and validated rapidly. The RNAscope® technology, an advanced platform for in situ RNA detection, enables detection of almost any RNA biomarker with single-molecule detection sensitivity and high specificity in formalin-fixed paraffin-embedded (FFPE) tissues2. It provides a universal solution to characterize tissue distribution of drug targets and biomarkers in a highly specific and sensitive manner, without the need to wait for antibody development and validation. The RNAscope® assay can be performed in fully automated staining systems, including Ventana Discovery XT, Ventana Discovery ULTRA, and Leica BOND RX instruments. The assay allows visualization of each individual RNA molecule as a punctate dot under a standard bright field microscope. The RNA dots can be quantified by counting the number of signal dots in individual cells, either manually or by image analysis tools, including HALOTM (Indica Labs) and SpotStudioTM (ACD) software. In this study, we demonstrate the feasibility Preclinical drug safety assessment in animal models is used to evaluate the pathological effects induced by novel therapeutic agents. Here we present the use of the RNAscope® 2.5 LS Assay for the evaluation of biomarkers in tissues from three preclinical animal models. In this study we: • Identify optimal pretreatment conditions for different tissues in different species • Provide recommendations for control gene selection for tissue qualification • Detect specific RNA markers in various FFPE tissues from multiple species 1 1Advanced Cell Diagnostics, Inc 3960 Point Eden Way Hayward, CA, USA 94545 2Drug Safety Research and Development, Pfizer Global Research and Development, Groton, CT 06340, USA 2 Application Note of evaluating RNA biomarkers in 25 types of tissues from three commonly used animal models using the RNAscope® 2.5 LS Reagent Kit-BROWN on the Leica BOND RX instrument. Robust RNA detection was achieved following a standard protocol in almost all of the tissues tested, with minor alterations to the prestaining conditions for a few tissues. Here, we identified the threshold of pretreatment needed for different tissue types. We also provide recommendations for control probes to be applied for tissue qualification, and present the evaluation of RNA biomarkers including cell type specific markers (CD68 and endothelial marker PECAM1), proliferation marker Ki-67 (MKI67), and cell cycle marker Cyclin E1 (CCNE1). Overall, our study shows that the fully automated RNAscope® 2.5 LS Assay is capable of detecting a broad range of RNA targets in all major tissue types with little to no optimization needed, and thus well suited for the histopathological evaluation of biomarkers in the assessment of drug-derived toxicity in various tissues and animal models. Materials and Methods FFPE tissues Multiple tissues from three commonly used animals (rat, dog, and cynomolgus monkey) were harvested using a standard protocol at the drug safety research and development laboratory of PfizerGroton (Table 1). Tissues were cut into 3 mm thickness then fixed in 10% neutral-buffered formalin (NBF) for 24-48 hours. Fixed tissues were dehydrated in a graded series of ethanol and xylene, followed by infiltration of melted paraffin at 56°C in an automated processor. Tissue microarrays (TMAs) were constructed, sectioned at a thickness of 5 μm and mounted on the SuperFrost® Plus slides (Fisherbrand Cat # 12-550-15). Automated RNAscope® 2.5 LS assay Ready-to-use reagents from RNAscope® 2.5 LS Reagent Kit-BROWN were loaded onto the Leica BOND RX instrument according to the user manual (Doc. No. 322100-USM). FFPE tissue sections were baked and deparaffinized on the instrument, followed by epitope retrieval (using Leica Epitope Retrieval Buffer 2 at 95°C or at 88°C for 15 min) and protease treatment (15 min at 40°C). Probe hybridization, signal amplification, colorimetric detection, and counterstaining were subsequently performed. A schematic of the RNAscope® 2.5 LS Assay workflow on Leica BOND RX is presented in Figure 1A. RNAscope® probes Control probes of low-, medium-, and high-expressing housekeeping genes (POLR2A, PPIB, and UBC, respectively) were designed and tested for tissues from each species (Table 2). Because the sequences of the human probes for housekeeping genes are over 95% homologous to the respective target mRNA sequences of cynomolgus monkey, human probes were used to test samples of cynomolgus monkey. The bacterial probe DapB was used as a negative control. Probes for the cell type biomarkers, proliferation markers, and apoptosis-related molecules used in this study were designed for each species. As summarized in Table 3, speciesspecific target probes were tested for all RNA targets except two genes, CD68 and KI67, for which human probes were used to detect cynomolgus monkey genes, due to 90-95% homology between the probe sequence and target mRNA sequence. Image acquisition and data analysis Images were acquired using a Leica Biosystems Aperio AT2 Digital Pathology Scanner. RNA markers were analyzed based on the average RNA dot number per cell. RNA quantity was scored based on manual counting described as follows. Staining results were categorized into five grades according to the number of dots visualized under a bright-field microscope. 0: No staining or less than 1 dot to every 10 cells (40X magnification); 1+: 1-3 dots/cell (visible at 20-40X magnification); 2+: 4-10 dots/cell, very few dot clusters (visible at 20-40X magnification); 3+: >10 dots/cell, and less than 10% positive cells have dot clusters (visible at 20X magnification); and 4+: >10 dots/cell, and more than 10% positive cells have dot clusters (visible at 20X magnification). Results Optimal pretreatment condition for different tissues in different species The standard protocol of RNAscope® 2.5 LS Reagent Kit is designed to work for the majority of FFPE tissues. In this study, to achieve optimal detection of RNA molecules in each tissue type, we compared two different pretreatment conditions, standard and mild, with a modification in the epitope retrieval step (Figure 1A). The standard Animal Models: Rat, Dog, and Cynomolgus Hematopoietic system Thymus, Lymph Node, Spleen, Tonsil GI tract Esophagus, Stomach, Duodenum, Jejunum, Colon Urinary tract Kidney, Urinary bladder Reproductive system Epididymis, Prostate, Testis, Ovary Skin/soft tissues Skin, Skeletal muscle Endocrine glands/ exocrine glands Liver, Pancreas, Adrenal gland Respiratory system Lung, Bronchus Nervous system Spinal cord, Retina Cardiovascular system Heart TABLE 1. Tissue types from three commonly used animal models.