Preparation of Adult Mouse Skeletal Tissue Sections for RNA In Situ Hybridization

Abstract

The RNA in situ hybridization assay is essential in many studies to evaluate gene expression in vivo. It consists of generating tissue sections and subsequently hybridizing these sections with RNA probes. Keeping RNA intact is a challenge while harvesting tissue samples, processing through embedding, sectioning them, and conditioning the sections for hybridization. These challenges are particularly strong for adult skeletal tissues due to their copious, dense, and mineralized extracellular matrices. Here, we describe a method optimized to successfully hybridize RNA species, even of low abundance, in adult mouse bone and cartilage samples. This method involves tissue fixation with paraformaldehyde, demineralization with Morse’s solution and paraffin embedding, all of which can be completed in 4 days. Sections are then generated and hybridized using a 1-day standard protocol. Sections prepared using this method are compatible with immunostaining and standard staining procedures for skeletal tissues.

1. Introduction

The adult skeleton is mainly composed of bone and cartilage. These two distinct tissues are rich in extracellular matrix and sparse in cells. Their matrix consists of a fibrillar collagen network that entraps a dense gel of proteoglycans in the case of cartilage and that is mineralized in the case of bone. The activities of chondrocytes, osteoblasts, and osteoclasts are tightly controlled under homeostatic healthy conditions according to mechanical loading, hormonal influences and aging, and they can be drastically altered under pathological, genetic, and environmental conditions. Bone and cartilage diseases, namely, osteoarthritis and osteoporosis, affect a large subset of the adult and aging human population [1]. Research on these and other diseases often involves animal models. Assessing gene expression is key to analyzing cell activities in these models. This can be done using qRT-PCR and bulk sequencing of RNA extracted from tissues, using single-cell RNA sequencing, and using RNA in situ hybridization. The latter method involves hybridization of target RNAs in tissue sections with labeled antisense RNA probes. It very advantageously complements the others by allowing for visualization of gene expression in individual cells in morphologically preserved tissue. This is especially important for tissues that house several cell types and in which each cell type population is composed of cells at multiple differentiation stages or activity levels, such as bone, growth plate and articular cartilage.

Standard tissue processing and RNA in situ hybridization methods have proven to be challenging for adult bone and cartilage due to the low cellularity of the tissues, their dense extracellular matrix, and the mineralization of this matrix in the case of bone. Demineralization of skeletal samples is recommended for tissue, cellular, and molecular analysis of paraffin or frozen sections. This process typically involves incubation of the samples for extended periods of time in solutions at acidic pH or containing the cation chelator ethylenediaminetetraacetic acid (EDTA), increasing the chance of RNA degradation due to the release of RNases from possibly incompletely fixed tissue areas. In this chapter, we describe a quick and RNA-friendly method for sample processing and RNA in situ hybridization. Adult mouse skeletal tissues are fixed using 4% paraformaldehyde for one or two days, demineralized using Morse’s solution for another one or two days, and embedded in paraffin (see Note 1). Sections are then made and subjected to a one-day RNA in situ hybridization assay (RNAscope technology, Advanced Cell Diagnostics). Thus, results can be obtained within a week. We show that this method fully preserves RNA integrity in both cartilage and bone tissues. It robustly detects mRNA targets of both high and low cellular abundance.

2. Materials

2.1. Reagents and Solutions

1. Diethyl pyrocarbonate (DEPC).

2. RNase-free water, commercially available or prepared in house using 0.1% DEPC. Add 1 ml of DEPC to 1000 ml MilliQ water. Stir vigorously until all DEPC globules disappear. Let sit at room temperature for 1 h to overnight. Autoclave for 45 min to fully inactivate DEPC.

3. RNaseZap™. This RNase decontamination solution is recommended to inactivate any RNase traces from glassware, plastic surfaces, countertops, and pipettes. Apply RNaseZap solution to the surfaces to be cleaned and then rinse with RNase-free water.

4. Phosphate buffered saline (PBS) prepared in RNase-free water.

5. 4% paraformaldehyde (PFA) diluted in RNase-free PBS from 16% stock.

6. Sodium citrate.

7. Formic acid.

8. Morse’s solution (10% sodium citrate and 22.5% formic acid). Dissolve 100 g sodium citrate in 500 ml RNase-free water. Add 240 ml 94.5% formic acid stock (VWR, Cat. No. 97064-708). Add RNase-free water up to 1 l. Prepare no more than 3 days before use.

9. Xylene.

10. 100% ethanol.

2.2. Histology and Other Supplies

1. Dissection tools: scissors, forceps, and scalpels.

2. Insulin syringes and needles.

3. Tissue processing/embedding cassettes.

4. Screw-capped bottles with large opening (allowing the entry of embedding cassettes).

5. Thermometer.

6. Paraffin wax.

7. SuperFrost Plus slides.

8. ImmEdge™ hydrophobic barrier pen.

9. Tissue-Tek manual slide staining set, including white and green (xylene resistant) dishes, and two vertical 24-slide racks.

10. Square-mouth glass Wheaton staining jars (10-slide capacity).

11. VectaMount™ permanent mounting medium (Vector Laboratories, Cat. No., H-5000).

12. Cover glasses (24 × 50 mm).

2.3. Equipment

1. Microtome.

2. Water bath or incubator capable of holding temperature at 40 ± 1 °C.

3. Drying oven capable of holding temperature at 60 ± 1 °C.

4. Bright-field microscope and accessories.

5. Hybridization oven and slide tray set (Advance Cell Diagnostics [ACD] or equivalent).

2.4. RNAscope Reagents (ACD)

1. RNAscope 2.5 HD duplex detection kit (chromogenic) (for duplex detection), or RNAscope 2.5 HD Assay—RED (for single detection).

2. RNAscope probes for positive and negative controls and for RNAs of interest.

3. RNAscope hydrogen peroxide.

4. Wash buffer kit.

5. Custom pretreatment reagent.

3. Methods

3.1. Preparation of Tissue Sections

RNase-free conditions are not required for steps 1–7. Speed is the most important aspect in this part of the protocol. Proceed with one animal at a time and get the tissues of interest dissected out and immersed into fixative as quickly as possible (within 10 min), otherwise dying cells will release RNases, which will quickly destroy RNAs before the tissue is fully fixed.

1. Euthanize one mouse at a time via CO₂ asphyxiation followed by cervical dislocation. Immediately after death is confirmed, skin the mouse where appropriate and dissect out the desired skeletal elements. Cut off as much muscle and other nonskeletal tissues as possible and suitable.

2. Bore a few holes or make small slits in large bones and joints to ensure that the fixative and wash solutions rapidly access internal structures, such as bone marrow and synovial cavities.

3. Using an insulin syringe, inject small amounts of 4% PFA into remaining muscles to ensure better penetration of the fixative through the whole tissue.

4. Place the samples into tissue-embedding cassettes and submerge the cassettes in 4% PFA solution in a screw-capped bottle. Place the bottles on a rocker or rotator and incubate at room temperature for 24–48 h (depending on mouse age and tissue size).

5. Wash the samples for three times 15 min in RNase-free PBS.

6. Fill the bottles with Morse’s solution (about 30 ml per sample). Allow samples to decalcify at room temperature for 24–48 h (see Notes 2 and 3).

7. Wash the samples for three times 15 min in RNase-free PBS.

8. Dehydrate samples in a graded ethanol series (70%, 80%, 90%, 95%, 100%, and 100%) followed by two changes in xylene, and two changes in paraffin. Incubate the samples in each solution for at least 1 h (see Note 4).

9. Embed the tissues in paraffin blocks using a standard method.

10. Cut 4 to 6-pm-thick tissue sections with a microtome (see Note 5). Float sections in a freshly cleaned DEPC-water bath at 40–45 °C until they have smoothed out on the water’s surface (see Note 6).

11. Carefully lift the sections onto Superfrost Plus slides (see Note 7). Position the section as close to the middle of the slide as possible on both axes (see Note 8).

12. Dry the slides at 45 °C on a cleaned slide warmer until the sections have fully spread (about 2 h), and then dry them overnight at room temperature on a clean surface covered with 2–3 layers of Kimwipes in a protected location (e.g., an empty drawer or a cleaned slide dryer).

13. Store the slides in a slide box for up to 3 months at 4 °C (see Notes 9 and 10).

3.2. RNA In Situ Hybridization

The RNA in situ assay itself could in principle be carried out following any previously described protocol for radioactive or fluorescent probes [2-4]. We routinely use the RNAscope method according the manufacturer’s instructions, but with the following adaptations:

1. Prepare all solutions in RNase-free water (see Subheading 2.1).

2. Replace the heated target retrieval and protease steps with an incubation of the slides with ACD Custom Pretreatment Reagent for 30 min at 40 °C. This solution is specifically designed for tissue sections that easily detach from the slides during target retrieval at high temperature (98–100 °C).

3.3. Image Acquisition and Data Analysis

Image the slides under regular bright-field microscopy conditions. Signals generated with the RNAscope method will appear as colored dots within cells counterstained with hematoxylin (light purple). Signal colors (purplish red, bluish green, etc.) will depend on the types of probes selected for the assay. Densitometry software (e.g., NIH ImageJ) can be used to quantify RNA signals.

Figure 1 illustrates the superiority of Morse’s solution over EDTA solution to preserve RNA integrity during skeletal tissue demineralization.

Fig. 1.

Comparison of RNA signals obtained following tissue demineralization with 15% EDTA solution (pH 7.4) for 1 week (top panels) or with Morse’s solution for 1 day (bottom panels). Parasagittal sections of 3-month-old mouse knees were hybridized with ACD positive control probes for Ppib RNA (bluish green signal) and Polr2a RNA (purplish red). Ppib (encoding peptidyl-prolyl cis-trans isomerase B) is expressed ubiquitously at a moderate-high level and Polr2a (encoding the largest subunit of RNA polymerase II) at a moderate-low level. Signals for Polr2a RNA are visible in only a few cells and those for Ppib RNA are visible in hardly any cell in EDTA-treated samples. Signals for both RNAs are much stronger and are visible in many more cells in Morse’s solution-treated samples

Figure 2 compares RNA signals obtained for three marker genes for skeletal tissues: Acan, a gene highly and specifically expressed in chondrocytes (encoding the core protein of aggrecan); Runx2, a gene expressed at a low level in prehypertrophic and hypertrophic growth plate chondrocytes and in osteoblasts (encoding the RUNT-domain transcription factor 2); and Dmp1, a gene specifically and substantially expressed in growth plate terminal chondrocytes, osteoblasts, and osteocytes (encoding the dentin matrix protein 1).

Fig. 2.

Comparison of signals obtained for Acan, Runx2, and Dmp1 RNAs. A 3-month-old mouse knee was fixed for 2 days and demineralized in Morse’s solution for the next 2 days. Left panels, low-magnification images of parasagittal sections through the tibial proximal epiphysis, including articular cartilage (AC), secondary ossification center (SOC), growth plate cartilage (GP), and primary ossification center (POC). Middle and right panels, high-magnification images of regions boxed in the left panels. For Runx2, which shows very low expression, a few cells are presented at even higher magnification to show individual RNA signal dots. The signals for all three RNA probes are seen in purplish red. As expected, articular cartilage and growth plate cartilage show strong expression of Acan, but none in bone and bone marrow. Runx2 is detected at a low level in growth plate chondrocytes and in bone and bone marrow cells. Dmp1 is highly expressed in terminal chondrocytes of the growth plate, but not in less mature cells and articular chondrocytes. It is also strongly expressed in osteoblasts and osteocytes