Here, we describe an efficient high throughput in situ hybridization (ISH) method for visualizing patterns of mRNA expression in developing fetal mouse prostate tissue sections. The method can be easily adapted to visualize mRNA expression patterns in other mouse tissues or in tissues from other species.
Development of the lower urogenital tract (LUT) is an intricate process. This complexity is evidenced during formation of the prostate from the fetal male urethra, which relies on androgenic signals and epithelial-mesenchymal interactions1,2. Understanding the molecular mechanisms responsible for prostate development may reveal growth mechanisms that are inappropriately reawakened later in life to give rise to prostate diseases such as benign prostatic hyperplasia and prostate cancer.
The developing LUT is anatomically complex. By the time prostatic budding begins on 16.5 days post conception (dpc), numerous cell types are present. Vasculature, nerves and smooth muscle reside within the mesenchymal stroma3. This stroma surrounds a multilayered epithelium and gives rise to the fetal prostate through androgen receptor-dependent paracrine signals4. The identity of the stromal androgen receptor-responsive genes required for prostate development and the mechanism by which prostate ductal epithelium forms in response to these genes is not fully understood. The ability to precisely identify cell types and localize expression of specific factors within them is imperative to further understand prostate development. In situ hybridization (ISH) allows for localization of mRNAs within a tissue. Thus, this method can be used to identify pattern and timing of expression of signaling molecules and their receptors, thereby elucidating potential prostate developmental regulators.
Here, we describe a high throughput ISH technique to identify mRNA expression patterns in the fetal mouse LUT using vibrating microtome-cut sections. This method offers several advantages over other ISH protocols. Performing ISH on thin sections adhered to a slide is technically difficult; cryosections frequently have poor structural quality while both cryosections and paraffin sections often result in weak signal resolution. Performing ISH on whole mount tissues can result in probe trapping. In contrast, our high throughput technique utilizes thick-cut sections that reveal detailed tissue architecture. Modified microfuge tubes allow easy handling of sections during the ISH procedure. A maximum of 4 mRNA transcripts can be screened from a single 17.5dpc LUT with up to 24 mRNA transcripts detected in a single run, thereby reducing cost and maximizing efficiency. This method allows multiple treatment groups to be processed identically and as a single unit, thereby removing any bias for interpreting data. Most pertinently for prostate researchers, this method provides a spatial and temporal location of low and high abundance mRNA transcripts in the fetal mouse urethra that gives rise to the prostate ductal network.
1. Synthesis of a Digoxigenin-11-UTP-Labeled Riboprobe from a PCR-Generated Template
2. Preparation of Vibrating Microtome Blade (Based on Previously Described Protocol)7
3. Dissection, Storage, and Preparation of Urogenital Tissues for Sectioning
4. Embedding Urogenital Tissue in Agarose
5. Sectioning Urogenital Tissue With a Vibrating Microtome (Based on Previously Described Protocol)7
6. Sample Basket Preparation For In Situ Hybridization
7. Embryo Powder Preparation For In Situ Hybridization (Based on Previously Described Protocol)8
8. In Situ Hybridization Day 1
9. In Situ Hybridization Day 2
10. In Situ Hybridization Day 3
11. Representative Results:
The spatial orientation of LUT tissue in agarose determines the plane of tissue sections. For sagittal sections, at least two-thirds of the bladder is excised and the remaining LUT tissue is embedded in agar such that the urethral midline is parallel to the flat surface of the agarose plug (Figure 1A). Minor adjustments in the tissue plane can be made by beveling the flat edge of the agarose plug. A representative sagittal section from a 17.5dpc male LUT tissue that is oriented in this plane is shown in Figure 1B.
Sample baskets protect the delicate tissue sections from loss and from accumulating dust and particulate matter during the multi-day ISH procedure. Sample baskets are prepared by melting polyester mesh to the cut end of a 1.5mL microfuge tube (Figure 1C). A small hole is pierced into the lid of each sample basket to facilitate solution flow into and out of the baskets. Sample baskets are suspended in ISH solutions by placing them in 12mm holes drilled into 24-well plate lids (Figure 1D). The modified plate lids support baskets when they are transferred between 24-well plates during solution changes.
It is challenging to limit non-specific background staining during the long incubation periods required for detection of low abundance mRNAs. The addition of 0.2mM sodium azide to sample buffers and their subsequent filtration through 0.22μm filters appeared to limit background staining (Figure 2). Using the method described here, there does not appear to be visible differences in background staining when samples are incubated in color development solution for a prolonged time periods (Figure 3).
Figure 1. Preparation of a mouse lower urogenital (LUT) tract tissue section and microcentrifuge tube a basket for ISH. A LUT containing part of the bladder, pelvic urethra and associated Wolffian and Mϋllerian duct-derived structure) is embedded in a cylindrical plug of 4% low-melt agarose. (A) The plug is glued to a specimen mounting disk and (B) cut into 50μm sections with a vibrating microtome. (C) An LUT section is transferred into a microcentrifuge tube basket that is prepared by piercing a hole into the tube lid and fusing polyester mesh to the cut bottom end of the tube. (D) The microcentrifuge tube is inserted into 12mm holes drilled into a 24-well plate lid so that tissue sections are suspended in buffer solution during the ISH protocol. Arrowheads indicate the LUT tissue in the agarose plug.
Figure 2. Incorporation of 0.2mM sodium azide into 0.22μm filtered solutions improves tissue quality and reduces background staining. 17.5dpc male mouse lower urogenital tracts (LUTs) were sectioned in a sagittal plane to a thickness of 50μm. Tissue sections were stained by ISH using a probe directed against twist homolog 1. Buffers used for ISH were either (A) 0.22μm filtered and supplemented with 0.2mM sodium azide (NaAz) or (B) unfiltered and not supplemented with NaAz. Arrowheads indicate background staining. Images were captured at the same magnification. Results are representative staining patterns for n = 3 litter-independent mice.
Figure 3. Background staining intensity does not appear to increase with prolonged color development. 17.5dpc male mouse lower urogenital tracts (LUTs) were sectioned in a sagittal plane to a thickness of 50μm. Sections were stained by ISH by incubating them in chromagen staining solution for (A) 9.5h using a probe that recognizes the high abundance transcript estrogen-related receptor gamma (Esrrg), (B) for 43.5h using a probe that recognizes the medium abundance transcript bromodomain adjacent to zinc finger domain, 2A (Baz2a), or (C) for 236h using a probe that recognizes the low abundance transcript wingless-type MMTV integration site family, member 10a (Wnt10a). Images were captured at the same magnification. Results are representative staining patterns for n = 3 litter-independent mice.
Using the method described here, it is possible to detect mRNAs in all of the major cell types and tissue compartments of the fetal male and female mouse LUTs including the mesenchymal pads, urothelium, smooth muscle, prostatic buds, ejaculatory duct, and vagina. The 50μm sections used in this protocol have the advantage of being thick enough to resolve tissue architecture (such as blood vessels) but are thin enough to avoid probe trapping, which is a methodological problem commonly encountered during whole-mount ISH. Every new riboprobe is assessed on positive control tissue, where staining has been assessed in a previous published study. We ensure that staining patterns are specific in these tissues. Another advantage of our method is that patterns of multiple mRNAs can be assessed in adjacent tissue sections from the same LUT tissue. Furthermore, this method can be coupled with immunohistochemical techniques to visualize cell specific protein markers and identify cell types stained by the ISH protocol. This method is incompatible with traditional counterstaining methods, such as nuclear counterstaining with Fast Red or hematoxylin. However, we have successfully counterstained samples with fluorescent nuclear stains, including 4′,6-diamidino-2-phenylindole (DAPI) and propidium iodide.
The method described here includes several improvements over one that was previously described7. Almost all of the buffers and reagent stock solutions in the current manuscript are prepared in advance and can be stored for long periods of time, which increases efficiency. The addition of 0.2mM sodium azide to most reagents and their filtration through 0.22μm membranes greatly diminishes non-specific background staining, increases reagent shelf-life, and minimizes accumulation of particulate matter on the tissue sections during the ISH procedure. This manuscript also describes the manufacture of permeable microcentrifuge tube baskets that contain tissue sections during ISH and a modified multi-well culture plate lid that holds the baskets during buffer changes. We found that these devices, which are easily manufactured in the lab, reduce sample loss, minimize tissue section damage during processing, and improve efficiency. Furthermore, the use of a sealable plastic container as a humidity chamber helps to protect against evaporation. This is especially important for tissue sections in peripheral sample wells, where the so-called ‘edge effect’ can introduce an experimental variable. While using humidity chambers, we have not observed appreciable staining quality differences in peripheral versus inner sample wells.
While this protocol is an efficient mechanism to study mRNA expression patterns in mouse LUT tissues, there are some limitations with this method. Efficiency of mRNA detection varies among riboprobes and absolute abundance of mRNAs cannot be determined. Another limitation is that staining patterns can vary depending on the section plane. To minimize these limitations, we assess each mRNA pattern in multiple sections from multiple litter-independent fetuses.
This method can be optimized for visualizing mRNA expression patterns in other mouse tissue types or other species. Such a modification requires that the microtome blade amplitude and speed be optimized during tissue sectioning and that proteinase K concentration be optimized during the ISH procedure. Furthermore, it is necessary to pre-absorb the anti-digoxigenin antibody with embryo powder from the same species of tissue that is being assessed by ISH. Although this method is not useful for tissue sections thinner than 40 microns, because the sections disintegrate during ISH staining, it can be adapted for use in high-throughput whole-mount ISH staining. We have used this method successfully to conduct whole-mount ISH on fetal and neonatal mouse prostate as well as fetal gonad and kidney. For whole-mount ISH staining, it is necessary to optimize proteinase K tissue digestion as well as the quantity and duration of washes following overnight antibody incubation to reduce background staining caused by probe trapping.
The authors have nothing to disclose.
The authors would like to thank Dr. Lan Yi, Cancer Institute of New Jersey, for technical assistance in preparing tissue baskets. This work was funded by National Institutes of Health Grants DK083425 and DK070219.
Name of the reagent | Company | Catalogue number |
---|---|---|
Anti-Digoxigenin antibody, Fab fragments | Roche Applied Science | 11214667001 |
Blocking reagent | Roche Applied Science | 11096176001 |
BM Purple AP substrate, precipitating | Roche Applied Science | 11442074001 |
Bovine Serum Albumin | Fisher Scientific | BP1600-100 |
Cell culture plate, 24 well | Corning | 3524 |
Digoxigenin 11-UTP | Roche Applied Science | 1277073910 |
dNTPs | Roche Applied Science | 11969064001 |
Double-edged razor blade | Wilkinson Sword | Classic Model |
Eliminase RNase remover | Decon Laboratories | 1102 |
Formamide | Sigma | F5786-1L |
Gel extraction kit | Qiagen | 28704 |
Glutaraldehyde, 25% solution in H2O | Sigma | G6257-100ML |
Heparin, sodium salt | Sigma | H3393 |
Hydrogen peroxide, 30% solution in H2O | Fisher Scientific | BP2633-500 |
Levamisole | Sigma | L9756 |
Loctite 404 quick set instant adhesive | Henkel Corp. | 46551 |
Magnesium chloride | Fisher Scientific | M33-500 |
Maleic acid | Sigma | M0375-500G |
Microcentrifuge tubes, 1.5mL | Biologix Research Company | BP337-100 |
Millicell culture plate insert | Millipore | PICM01250 |
Molecular grinding resin | G-Biosciences | 786-138PR |
Paraformaldehyde, 4% solution in phosphate buffered saline | Affymetrix | 19943 |
Phosphate buffered saline, without Ca & Mg | MP Biomedicals | ICN1760420 |
Polyester mesh, 33 micron, 12” x 24” | Small Parts Inc | CMY-0033-D |
Proteinase K solution, 20mg/ml | Amresco | E195-5ML |
QIAshredder Columns | Qiagen | 79654 |
Q solution | Qiagen | Provided with Taq DNA polymerase |
RNase | Sigma | R6513 |
RNase inhibitor | Roche Applied Science | 03335399001 |
RNeasy mini kit | Qiagen | 74104 |
RNEASY mini kit | Qiagen | 74104 |
RQ1 RNase-free DNase | Promega | M6101 |
SeaPlaque low-melt agarose | Lonza | 50101 |
Sheep serum | Sigma | S2263-500mL |
Stericup filter unit, 0.22μm, polyethersulfone, 500mL | Millipore | SCGPU05RE |
Sodium azide, granular | Fisher Scientific | S227I-100 |
Sodium chloride | Fisher Scientific | BP358-212 |
Sodium dodecyl sulfate | Fisher Scientific | S529-500 |
SSC, 20X solution | Research Products International | S24022-4000.0 |
SuperScript III first-strand synthesis system | Invitrogen | 18080-051 |
T7 RNA polymerase | Roche Applied Science | 10881767001 |
Taq DNA polymerase | Qiagen | 201203 |
Tris-HCl | Fisher Scientific | BP153-1 |
Tween 20 | Fisher Scientific | BP337-100 |
Vibrating microtome with deluxe specimen bath | Leica Microsystems | VT1000A |
Yeast tRNA | Roche Applied Science | 109495 |