Tyramide signal amplification coupled with multiple immunolabeling and RNAScope in situ hybridization in formaldehyde‐fixed paraffin‐embedded human fetal brain

Abstract

Several strategies have been recently introduced to improve the practicality of multiple immunolabeling and RNA in situ hybridization protocols. Tyramide signal amplification (TSA) is a powerful method used to improve the detection sensitivity of immunohistochemistry. RNAScope is a novel commercially available in situ hybridization assay for the detection of RNA expression. In this work, we describe the use of TSA and RNAScope in situ hybridization as extremely sensitive and specific methods for the evaluation of protein and RNA expression in formaldehyde‐fixed paraffin‐embedded human fetal brain sections. These two techniques, when properly optimized, were highly compatible with routine formaldehyde‐fixed paraffin‐embedded tissue that preserves the best morphological characteristics of delicate fetal brain samples, enabling an unparalleled ability to simultaneously visualize the expression of multiple protein and mRNA of genes that are sparsely expressed in the human fetal telencephalon.

RNAscope can be combined with immunofluorescence in precious human fetal brain sections to reveal information about localization of transcription factor expression. ZIC4 and PAX6 occupy separate domains across the forebrain, whereas ZIC4 and FOXP2 show clear co‐localization in the dorsal thalamus.

1. INTRODUCTION

Immunohistochemistry (IHC) and RNA in situ hybridization (ISH) are very important techniques in the field of developmental neurobiology that allow us to assess the expression of protein and mRNA for a wide variety of transcription factors and discrete signaling molecules (e.g., SHH and FGF). However, the technical complexity, insufficient sensitivity, and specificity of the routine application of these two methods make their use very limited for some histological analyses, particularly post‐mortem human embryonic and fetal tissue.

The development of appropriate multiple immunofluorescence labeling methods is one of the main obstacles in applying IHC. The selection of the appropriate combination of primary antibodies becomes problematical when the primary antibodies are raised in the same host species, which leads to cross‐reactivity of secondary antibodies with each of the primary antibodies (Tóth & Mezey, 2007). A second general problem is the insufficient sensitivity of primary antibodies to target antigens that are present in small or barely detectable amounts, leading to poor signal visualization for these antigens (Van der Loos 2007; Warford et al. 2014). However, relatively novel hybrid detection strategies have been recently introduced to improve the practicality of multiple immunolabeling methods and circumvent some of these technical obstacles. Some of these strategies involve elution of antibodies between rounds of staining, using HRP polymer conjugated secondary antibodies and the tyramide signal amplification method which covalently binds fluorescent tags to the tissue at the site of antigen detection, increasing sensitivity, and allowing the washing away of antibodies while leaving staining behind before further immunostaining (for more detailed information on the principle and history of these methods see Hunyady et al. 1996; Shi et al. 1999; Tóth & Mezey, 2007; Stack et al. 2014; Goto et al. 2015).

Unlike other RNA analysis techniques (e.g., RT‐PCR), in situ analysis of RNA biomarkers allows spatial descriptive analysis of RNA expression in the developing fetal brain and is important for verifying the results of single‐cell RNA transcriptomics. For example, investigating the graded and compartmented expression of genes (transcription factors, morphogens, and receptors) in developing cerebral cortex permits us to understand how the dorsal pallium is stereotypically divided into functionally distinct areas even at the earliest stages (Alzu'bi et al. 2017a; Clowry et al. 2018; Molnár et al. 2019). However, a long‐standing criticism of routine RNA ISH techniques is the lack of sensitivity and specificity, especially for genes expressed at very low levels (e.g., genes for morphogens and their receptors); therefore, the use of these techniques in the field of developmental neurobiology has been challenging. Recently, Advanced Cell Diagnostics, Inc., (Hayward, California, USA) presented a novel RNA ISH technology (RNAScope in situ hybridization) that implement a unique strategy of probe formulation. Pairs of probes that bind adjacent RNA sequences must hybridize independently to their respective targets for subsequent signal amplification to occur. This pairing requirement greatly increases the signal‐to‐noise ratio and ensures high specificity because of the low probability of two different probes binding adjacent non‐specific sequences (for more detailed information on the principle of this technology see Wang et al. 2012).

RNAScope technology has proved to be a reliable method in many areas of biological and pathological research giving reproducible and quantitative results (Wang et al. 2014; Bingham et al. 2017; Chan et al. 2018; Jolly et al. 2019; De Biase et al. 2021). Further, the gentle tissue permeabilization technique used in this technology, and the possibility for using a fluorescent assay allow the combination of RNAScope in situ hybridization with immunohistochemistry, enabling simultaneous visualization of mRNA and protein antigens (Wang et al. 2012; Grabinski et al. 2015; Kann & Krauss 2019; Annese et al. 2020). However, the use of RNAScope technology, specifically in combination with immunohistochemistry, in human fetal samples has not been extensively evaluated. In this study, we have evaluated the feasibility of multiple immunolabeling (with TSA method) and RNAScope (including coupling with immunofluorescence) techniques in formalin‐fixed, paraffin‐embedded sections of human fetal forebrain.

2. METHODS AND MATERIALS

2.1. Human fetal brains and ethical approval

Human fetal tissue from terminated pregnancies was obtained from the joint MRC/Wellcome Trust‐funded Human Developmental Biology Resource (HDBR, http://www.hdbr.org; Gerrelli et al. 2015) based in Newcastle University and the Institute of Child Health University College London. All tissue used in this study was collected by the Newcastle center with appropriate maternal consent and approval from the Newcastle and North Tyneside NHS Health Authority Research Ethics Committee (REC reference 18/NE/0290). Both centers are licensed by the UK Human Tissue Authority (license numbers 12,534 and 1220). All embryonic and fetal samples used in this study were obtained from RU486 induced medical terminations and ranged in age from 8 to 19 post‐conceptional weeks (PCW). Ages were estimated from foot, and heel to knee length measurements according to Hern (1984). In addition, samples were karyotyped. Full details are provided in Table 1.

TABLE 1.

Details of embryonic/fetal samples

Featured in figure	Estimated age postconceptional weeks	Foot length (mm)	Knee heel Length (mm)	Karyotype
1	8	5	6	46XX
2	19	34	58	46XX
3	12	15	25	46XX
3	10	9	11	46XY
4	8	5	8	46XY
5	10	9.5	15	46XY

2.2. Tissue processing and sectioning

Fetal brains were rapidly isolated and fixed for at least 24 h at 4°C in 4% formaldehyde (from paraformaldehyde powder, Sigma Aldrich, Poole, UK) in 0.1 M phosphate‐buffered saline (PBS). 8 PCW embryos were similarly fixed and sectioned whole. The exact post‐mortem interval to fixation is unknown but is in the range of 1–12 h. After fixation, embryos, whole, or half fetal brains were dehydrated in graded ethanol (70% for 15 min, 100% for 45 min, 2 × 100% for 60 min) at room temperature (RT). The larger fixed brains were dissected into blocks of approximately equal size, with the number depending on the size of the brain. Blocks were then incubated in xylene for 2 h before embedding in paraffin (Excelsior AS Tissue Processor, Thermo Scientific). Brain tissue blocks were cut into 8 μm thick sections (Leica RM 2235 microtome) mounted onto Superfrost+ slides (Thermo Fisher Scientific) and used for immunostaining and RNAScope in situ hybridization.

2.3. Multiple immunofluorescence

This was performed using the method employed by Tóth & Mezey (2007) in adult rodent models with slight modification. In the first round of staining, paraffin sections were first dewaxed in xylene (2 × 5 min) and then rehydrated via four changes of graded ethanol (100%, 100%, 95%, and 70%). Endogenous peroxidase activity was blocked by treatment with methanol peroxide (3 ml of 30% hydrogen peroxide +180 ml methanol) for 10 min. For antigen retrieval, sections were rinsed under tap water and boiled in a microwave in 10 mM citrate buffer pH 6 for 10 min. Sections were then incubated with the appropriate normal 10% blocking serum (species in which secondary antibody was raised) in Tris‐buffered saline pH 7.4 (TBS) for 10 min at RT before incubation with the primary antibody (diluted in 10% normal blocking serum) overnight at 4°C. Details of all the primary antibodies used in this study can be found in Table 2. Then, sections were washed 2 × 5 min in TBS and incubated with HRP polymer‐conjugated secondary antibody for 30 min (ImmPRESS™ HRP IgG [Peroxidase] Polymer Detection Kit, Vector Labs) at RT, washed 2 × 5 min in TBS and incubated in the dark for 10 min at RT with fluorescein tyramide diluted at 1/500 in 1X Amplification buffer (Tyramide Signal Amplification [TSA™] fluorescein plus system reagent, Perkin Elmer). Tyramide reacts with HRP to leave fluorescent tags covalently bound to the section.

TABLE 2.

Primary antibodies

Primary antibody to	Species	Dilution	Supplier	RRID number (where available)
FOXP2	Mouse monoclonal	1/50	Santa Cruz, Heidelberg, Germany.	AB_2721204
OTX2	Mouse monoclonal	1/200	Santa Cruz.	Catalog No. SC‐514195
ROBO1	Rabbit polyclonal	1/1500	Abcam, Cambridge, UK	AB_449561
SCGN	Rabbit polyclonal	1/500	Sigma‐Aldrich, Poole, UK	AB_1079874
DLX2	Mouse monoclonal	1/200	Santa Cruz	Catalog No. SC‐393879
GAD67	Mouse polyclonal	1/1000	Merck Millipore, Watford, UK	AB_2278725
PAX6	Rabbit polyclonal	1/500	Cambridge Bioscience, Cambridge, UK.	AB_2565003

Prior to starting the second round of staining, sections were first washed in TBS (2 × 5 min) and boiled in 10 mM citrate buffer (in microwave for 10 min) to remove all antibodies and unbound fluorescein from the first round. Sections were then incubated in 10% normal serum before incubating with the second primary antibody (Table 2) for 2 h at RT. Following washing (2 × 5 min in TBS), sections were again incubated with ready to use HRP‐conjugated secondary antibody for 30 min at RT, washed 2 × 5 min in TBS, then incubated in the dark for 10 min at RT with CY3 tyramide diluted at 1/500 in 1X Amplification buffer (Tyramide Signal Amplification [TSA™] CY3 plus system reagent, Perkin Elmer). The same steps can be repeated for a third round of staining (not reported here but see Alzu'bi et al. 2017b) using CY5 Tyramide (Tyramide Signal Amplification [TSA™] CY5 plus system reagent, Perkin Elmer). Sections were washed (in three changes of TBS for 5 min each) before applying 4′,6‐diamidino‐2‐phenylindole dihydrochloride (DAPI; Thermo Fisher Scientific) and mounted using Vectashield Hardset Mounting Medium (Vector Labs).

2.4. RNAScope in situ hybridization

Manual RNAScope assays were performed using The BaseScope™ Reagent Kit v2‐RED (Catalog No. 323600, ACD Bio Techne) according to the manufacturer's protocol. The routine RNAScope assay consists of four main steps, the first step is the sections pre‐treatment to allow the access of the target probe (conceptualized as a “Z”), In the second step, multiple tandem probes are hybridized in pairs (“ZZ”) to multiple RNA targets. In the third step, the detection probes are pre‐amplified by an adapter linked to several amplifiers containing multiple chromogenic labels. In the last step, signal detection is carried out by developing a chromogen to produce small punctate red dots (Wang et al. 2012). In this study, each sample was quality controlled for RNA integrity with a probe specific to the ubiquitously expressed reference gene GAPDH. We have previously demonstrated by qPCR that in human fetal samples GAPDH mRNA is expressed at the same high level across all regions of cortex (Harkin et al. 2017). Negative control background staining was evaluated using a probe specific to the bacterial DapB gene.

2.5. Section pre‐treatment

Paraffin sections were first baked on a heating pad for 10 min at 60°C, dewaxed in Xylene (2 × 5 min), then incubated in 100% ethanol (2 × 1 min) and air‐dried for 5–10 min at room temperature (RT). Sections were then covered with drops of 30% hydrogen peroxide solution for 10 min at room temperature and washed in distilled water by moving the slides rack up and down three to five times. For target retrieval, sections were boiled with the manufacturer's target retrieval buffer (ACD Biotechne) for 20 min at 90–100°C using a Cookworks vegetable steamer (Argos). Individual tissue sections were isolated on slides using a hydrophobic barrier pen. To increase target accessibility, protease digestion was then carried out by incubating sections with protease plus solution (ACD Biotechne) at 40°C for 30 min (any oven or incubator maintain a temperature between 37–40°C is also suitable). Sections were finally washed with distilled water before proceeding to probe hybridization.

2.6. Probe hybridization and signal amplification

Sections were incubated with the target gene's probe for 120 min at 40°C. Details of all the probes used in this study are found in Table 3. The slides were washed 2 × 2 min in 1X wash buffer at RT, and the hybridized signals were then amplified by six consecutive signal amplification steps using six different solutions provided by the manufacturer (Hybridize Amp 1–6, ACD Biotechne). In detail, sections were incubated with Hybridize Amp 1 for 30 min at 40°C, Hybridize Amp 2 for 15 min at 40°C, Hybridize Amp 3 for 30 min at 40°C, Hybridize Amp 4 for 15 min at 40°C, Hybridize Amp 5 for 30 min at RT, Hybridize Amp 6 for 15 min at RT. Sections were washed 2 × 2 min in 1X wash buffer (ACD Biotechne) at RT after each signal amplification step.

TABLE 3.

RNAScope probes used

Probe	Catalog number
Hs‐CHRNA4	498,331
Hs‐CHRNA5	482,401
Hs‐CHRNA7	310,101
Hs‐CHRNB2	498,351
Hs‐SHH	600,951
Hs‐ZIC4	525,661
Hs‐GAPDH	310,321
Negative control probe‐DapB	310,043

2.7. Signal detection

For signal detection, sections were incubated for 10 min at RT in red solution (RNAScope 2.5 HD Assay–RED, ACD Biotechne) freshly prepared following the manufacturer instructions (mixing 1 volume of RED‐B to 60 volumes of RED‐A to make the total volume needed to cover the sections). Slides were then rinsed in distilled water for 2 min, counterstained with 10%–20% hematoxylin (using a higher concentration of hematoxylin can obscure any positive signal). Slides should be dried for 10–15 min in a 60°C dry oven and mounted using DPX (Sigma‐Aldrich). Positive signals are indicated by red chromogenic dots in the cytoplasm or nucleus (Figures 3 and 4).

FIGURE 3.

(a–c) Chromogenic RNAScope in situ hybridization for nAChR subunits mRNA in the cortical plate of 12 PCW fetal brain. CHRNA4 was relatively highly expressed compared with moderate expression for CHRNA5 and low expression for CHRNA7. (d) Very strong expression was detected for the positive control reference gene GAPDH. (e) No expression was detected for the negative control dapB gene. (f–k) show results from conventional in situ hybridization. Comparison of sense/anti‐sense staining (f, g) suggests expression of CHRNA4, particularly in the CP and VZ but it would not be easy to localize expression to individual cells as staining is diffuse. A similar result was obtained for CHRNA7 (j, k) whereas for CHRNA5, sense stained more strongly than antisense suggesting the experiment has failed, or a transcript from the reverse strand was detected. CP, cortical plate; IZ, intermediate zone; MZ, marginal zone; SVZ, subventricular zone; VZ, ventricular zone. (a–c) adapted from Alzu'bi et al. (2020) under a creative commons license. Scale bars = 100 μm

FIGURE 4.

(a) Chromogenic RNAScope in situ hybridization for SHH mRNA in sagittal section of 8 PCW fetal brain. SHH expression was detected in ganglionic eminences (GE) and preoptic area (POA). Distinctive expression was also detected at the zona limitans intrathalamica (ZLI) a signaling center and a restrictive border between the thalamus and the prethalamus. (b) Coronal section of 8 PCW fetal brain counterstained with H&E, SHH was also highly expressed in population of cells in the globus pallidus (GP). Scale bars = A, 200 μm; A inset, 50 μm; B, 300 μm; B inset, 100 μm

2.8. RNAScope fluorescent in situ hybridization coupled with immunofluorescence

To avoid RNA degradation during the immunofluorescence steps, in situ hybridization was carried out first using RNAScope Multiplex Fluorescent Reagent Kit v2 Assay (ACD). Sections pre‐treatment and probe hybridization steps were performed as described above. Unlike RNAScope chromogenic assay, the signal amplification in RNAScope fluorescent multiplex v2 assay only included three steps: AMP1 for 30 min at 40°C, AMP2 for 30 min at 40°C and AMP3 for 15 min at 40°C. The sections were then incubated with C1‐HRP (as the probe, ZIC4 used in this case, was a channel 1 probe) at 40°C for 15 min and washed 2 × 2 min in 1X wash buffer at RT. The hybridized signals were detected by incubating the sections with Cy3 diluted in 1X amplification buffer (1/500, Tyramide Signal Amplification Cy3 plus system reagent, Perkin Elmer) at 40°C for 30 min. Sections were then washed and incubated with an HRP blocker for 15 min at 40°C before proceeding to the immunofluorescence.

For immunofluorescence, sections were first boiled in 10 mM citrate buffer pH 6 (in microwave for 10 min) to remove unbound tyramide from the first round (RNAScope fluorescent in situ hybridization), and incubated in 10% normal serum before incubating with the primary antibody of the target antigen for 2 h at RT. Sections were then washed 2 × 5 min in TBS and incubated with HRP polymer‐conjugated secondary antibody for 30 min at RT. Signals were detected by incubating the sections with fluorescein tyramide diluted at 1/500 in 1X Amplification buffer (Tyramide Signal Amplification fluorescein plus system reagent, Perkin Elmer) in dark for 10 min. Sections were then washed (2 × 5 min) before applying 4′,6‐diamidino‐2‐phenylindole dihydrochloride (DAPI; Thermo Fisher Scientific) and mounted using Vectashield Hardset Mounting Medium (Vector Labs).

2.9. Image acquisition

RNAScope in situ hybridization images were captured using Leica SCN400 Slide Scanner. Fluorescent images were obtained with a Zeiss Axioimager Z2 apotome. Processing of images, which included only adjustment of brightness and sharpness, was achieved using the Adobe Photoshop CS6 software.

3. RESULTS AND DISCUSSION

3.1. Multiple immunofluorescence

In Figures 1 and 2, we present examples of applying a combination of three complex detection strategies in multiple immunofluorescence labeling: elution of antibodies between rounds of immunostaining, HRP polymer‐conjugated secondary antibodies, and tyramide signal amplification on sections from 8, 10, and 19 PCW brains. Compartmentalization and cellular identities of different parts of the fetal brain, identified by the expressions of specific antigens, were clearly visualized. Various combinations of primary antibodies from the same and different host species (mouse/mouse, rat/rat, and mouse/goat) that detect either nuclear or cytoplasmic antigens were used here. A polymer‐enhanced detection system (ImmPRESS™ HRP IgG [Peroxidase] Polymer Detection Kit, Vector Labs) that contains HRP‐conjugated secondary antibody was applied. For signal detection, the TSA method was employed from Perkin Elmer according to the manufacturer's instructions. However, we further diluted the fluorescein tyramide in 1X amplification buffer to 1/500 dilution instead of the recommended 1/50; this was optimized after testing a range of dilutions from 1;100 to 1:1000. This step was found to be critical for achieving a good balance of signal amplification and background suppression. Elution of the bound primary and secondary antibodies after the first round of staining was achieved by boiling the sections with the same buffer used for antigen retrieval (10 mM citrate buffer, pH 6.0) in the microwave for 10 min. The heat treatment using the citrate buffer did not significantly reduce fluorescent signal from the first round, confirming that HRP‐activated tyramide binds covalently and efficiently to electron‐rich amino acids of proteins at the site of the immunoreaction making it resistant to the citrate treatment (Bobrow et al. 1989; Hasui & Murata, 2005; Shojaeian et al. 2018).

FIGURE 1.

Double immunofluorescence labeling in formaldehyde‐fixed paraffin‐embedded human 8 and 10 PCW fetal brain sections. (a) Double labeling with two antibodies originating from the same host species (mouse): FOXP2 (fluorescein, green) and OTX2 (Cy3, red). The two transcription factors show largely complementary expression in the ganglionic eminences (GE) and thalamus (Th) although with some overlap (yellow) in the dorsal thalamic subventricular zone. The higher magnification inset shows a mosaic of FOXP2+, OTX2+, and double‐labeled (yellow) cells throughout the ventricular (VZ) and subventricular zones (SVZ) although with different proportions in each, whereas the mantle zone is comprised almost entirely of FOXP2+ cells. (b) Double labeling with two antibodies originating from rat; anti‐SCGN (fluorescein, green) and anti‐ROBO1 (Cy3, red) adapted from Alzu'bi et al. (2019) under a creative commons license. The two cytoplasmic markers show extensive colocalization in thalamic neurons and their afferents in the internal capsule (IC) and cortical subplate and Intermediate zone (IZ, yellow). Nuclei counterstained with DAPI (blue). ChP, choroid plexus. Scale bar = 1 mm

FIGURE 2.

(a–c) Double labeling with two antibodies originating from the same host species (mouse): GAD67 (fluorescein, green) and DLX2 (Cy3, red) in the cortical plate of 19 PCW human fetal brain. (d) At higher magnification, co‐localization of the cytoplasmic enzyme GAD67 with the nuclear transcription factor DLX2 in GABAergic neurons is clearly demonstrated in nearly all cases. Nuclei counterstained with DAPI (blue). Adapted from Alzu'bi and Clowry (2019) under a creative commons license. Scale bars = , a‐c200 μm d, 100 μm

The use of these three strategies represents a significant improvement in multiple immunofluorescence labeling, profoundly improving the practicality of this method by giving bright images of antigen labeling we previously struggled to detect using fluorescently tagged secondary antibodies. The development of compact HRP polymer‐conjugated secondary antibodies, to introduce large number of peroxidase molecules, has remarkably enhanced the detection sensitivity of this method ( Shi et al. 1999; Shojaeian et al. 2018). This system also significantly simplified the staining procedures; it is ready to use and provides faster staining steps compared with using biotinylated secondary antibodies followed by treatment with avidin‐biotin complex (ABC). Additionally, this system lowers the cost by permitting further dilution of expensive primary antibodies.

The intermediary antibodies elution method in combination with TSA methods provided a remedy for one of the major problems in using multiple immunofluorescence labeling (Tóth & Mezey 2007). These methods enable the researcher to choose the primary antibodies for the target antigen regardless of their host species and without the fear of cross‐reactivity (Tóth & Mezey 2007; Pirici et al. 2009; Zhang et al. 2017). In addition, TSA offers sensitivity 10–200 times that of standard IHC, which enhances the signal visualization of the target antigens (Shojaeian et al. 2018) even if they are expressed in barely detectable amounts. The commercially available TSA kits provide enough reagents for staining 50–150 slides, using 1/500 dilution and not 1/50 dilution as recommended, of fluorescein tyramide in 1X amplification buffer. This allows us to use the kit for large number of slides, considering that the 1X amplification buffer can be replaced in the lab with a solution of tris buffered saline pH 7.4 containing 0.003% hydrogen peroxide.

3.2. Single and multiplexed RNAScope with immunofluorescence

Figure 3a–c illustrates employing RNAScope RNA ISH for detection of expression of three nicotinic acetylcholine (ACh) receptor subunit genes in the cortical plate of a 12 PCW brain. These receptors were expressed in relatively different levels from high for CHRNA4, to moderate for CHRNA5, and low for CHRNA7. This was correlated with expression observed by RNAseq of age‐matched samples from human fetal brain (Alzu'bi et al. 2020). Using previously described methods of probe manufacture and non‐radioactive tissue in situ hybridization histology (Bayatti et al. 2008) we had been able to observe approximately similar patterns of expression, but the staining had been more diffuse, with higher background and was strictly a qualitative observation at best. For CHRNA5 stronger staining was observed in some cells with the sense probe, rather than the anti‐sense probe (Figure 3f–k). This could be because a transcript from the reverse strand was detected. Although originally this was checked for during probe design, the most recent annotation of the GENCODE dataset (GENCODE—Human Release 39 (gencodegenes.org) does reveal potential transcripts on the reverse strand of the CHRNA5 gene. RNAScope detection provides a punctate staining with low background that permits a quantitative analysis of expression levels (Chermahini et al. 2019; Jolly et al. 2019; Annese et al. 2020)

Figure 4 illustrates the chromogenic RNA detection of mRNA for the diffusible morphogen Sonic Hedgehog (SHH), important in patterning the developing fetal forebrain (Kiecker and Lumsden, 2004; Bardet et al., 2010; Alvarez‐Bolado et al., 2012). Very discrete expression in signaling centers, predicted from observations of other vertebrate brains, was observed. In both Figures 3 and 4, dilute hemotoxylin was used as a counterstain. It is purple/pink shade can obscure low levels of red RNAScope staining. We have recently changed to toluidine blue as our preferred counterstain.

In Figure 5, we present two examples of coupling RNAScope fluorescent in‐situ hybridization with immunofluorescence. These RNAs in situ hybridization experiments were performed according to the manufacturer's instructions (Wang et al. 2012) with a slight modification described in the methods. Expression of mRNA for the transcription factor ZIC4, for which there is currently no antibody available, is compared with immunofluorescence for transcription factors PAX6 (Figure 5a) and FOXP2 (Figure 5b). PAX6 and ZIC4 are largely expressed in different locations, whereas FOXP2 and ZIC4 show extensive co‐localization in the thalamus, in particular (Figure 5c). The combination of the RNAScope and immunohistochemistry methods have been recently implemented in several types of tissues using various approaches and strategies, this includes works on thick free‐floating sections of adult rat brain (Grabinski et al. 2015) mounted skeletal myofibers from adult mouse (Kann and Krauss 2019), and human formalin‐fixed, paraffin‐embedded sections of diffuse large B‐cell lymphomas (Annese et al. 2020). Here we propose dual RNAScope ISH–IHC is also an extremely powerful method to study human neurogenesis, because it allows researchers to visualize simultaneous co‐localization of gene and protein expression, and changes in expression, to specific cell populations across the developing fetal brain.

FIGURE 5.

RNAscope fluorescent in situ hybridization coupled with Immunofluorescence on sections from 10 PCW human fetal brain. (a) Double labeling for ZIC4 mRNA (Cy3, red) and PAX6 protein (fluorescein, green) showing largely complementary patterns of expression. (b) Double labeling for ZIC4 mRNA (Cy3, red) and FOXP2 protein (fluorescein, green) showing co‐localization of expression in the thalamus (c, yellow). Scale bars = B (and for a) 2 mm; C 300 μm

RNAScope represents a valuable addition to RNA ISH methodology in the field of developmental neurobiology. This technique has many advantages‐ it is highly compatible with routine formalin‐fixed, paraffin‐embedded fetal brains; sensitive enough, with remarkable background suppression, to allow detection of genes that are expressed at low levels like receptors and morphogens. This means that it can be used both to confirm expression of gene expression patterns revealed by single‐cell RNAseq analysis but also detect expression of genes revealed by whole tissue RNAseq for which single‐cell RNAseq is not sufficiently sensitive. Finally, it is a time saving method that can be performed in 1 day compared with routine RNA ISH which is time‐consuming and labor‐intensive (3–4 day protocol).

There is one drawback to using the RNAScope approach. The reagents and technology for making the probes are copyrighted and are expensive to purchase. However, the apparently guaranteed success of the method so far, in our hands, saves us a lot of time and money when considering the high failure rates we have encountered in using conventional methods. In addition, as described above, most of the incubation steps for protease digestion, probe hybridization, and signal amplification are required to be performed under strictly controlled conditions of temperature and humidity in a special oven sold by the manufacturer. Although we have found the oven easy to use and to give excellent results, any oven or incubator that can maintain a temperature between 37 and 40°C was found to work perfectly well.

Alzu’bi, AP , Sankar, N , Crosier, M , Kerwin, J , Clowry, GJ. (2022) Tyramide signal amplification coupled with multiple immunolabeling and RNAScope in situ hybridization in formaldehyde‐fixed paraffin‐embedded human fetal brain. Journal of Anatomy, 241, 33–41. 10.1111/joa.13644

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.