Application of microRNA In Situ Hybridization on Long-term Stored Human Formalin-fixed Paraffin-embedded Brain Samples from Psychiatric Patients

Abstract

Here, we report that long-term stored human brain samples, formalin-fixed paraffin-embedded (FFPE) from The Brain Collection, University of Southern Denmark, Denmark ( https://www.sdu.dk/en/forskning/bridge/the-brain-collection ), can be used for in situ hybridization (ISH) analysis of selected microRNAs (miRNAs). The Human Brain Collection consists of brains from 9479 subjects who died at a Danish State Mental Hospital in the period of 1945-1982. In the present study we included tissue specimens from prefrontal cortex (PFC) and hippocampus (HIP) from 163 patients diagnosed with schizophrenia, bipolar disorder, or major depressive disorder. Initially, the Nanostring nCounter platform was used to identify miRNA candidates for ISH analysis using the miRNAscope technology. Based on the Nanostring nCounter quantifications with bulk tissue, we identified and selected 10 miRNAs from PFC (miR-9-5p, miR-29b-3p, miR-30c-5p, miR-124-3p, miR-125b-5p, miR-138-5p, miR-181a-5p, miR-224-5p, miR-302d-3p, and miR-432-5p) and 6 miRNAs from HIP (let-7a-5p, miR-7-5p, miR-124-3p, miR-127-3p, miR-145-5p and miR-149-5p). miRNAscope ISH analysis was then performed with the respective probes on 30 PFC and 30 HIP samples, respectively. In the PFC six miRNAs (miR-9-5p, miR-29b-3p, miR-124-3p, miR-125b-5p, miR-138-5p, and miR-181a-5p) were detected and four (miR-145-5p, let-7a-5p, miR-124-3p and miR-7-5p) in the HIP samples. In both brain regions miR-124-3p was the most abundantly expressed. We conclude, that the combination of the Nanostring nCounter technology and the miRNAscope analysis is a valid approach to study spatial expression of specific miRNAs in these up to 76 years old FFPE blocks. This opens a new avenue of possibilities for studying the underlying epigenetic mechanisms in mental disorders.

Keywords: Brain; FFPE; In situ hybridization; MiRNAscope; MicroRNA; Psychiatric disorders.

Fig. 1

Re-embedding old paraffin blocks. Examples of original paraffin blocks produced from 1945–1982 (A). The brain tissue specimens had been fixed and paraffin embedded and mounted onto wooden blocks. For the current study the old paraffin blocks were mounted onto modern plastic cassettes (B)

Fig. 2

microRNA expression using the Nanostring nCounter technology. miRNAs were extracted from paraffin sections and submitted to quantitative assessment using the nCounter expression panels. The standard deviation (SD) is plotted against the mean count values of the individual miRNAs obtained from the prefrontal cortex (A) and hippocampus (B) samples. The mean count values show the average expression level and the SD-values indicate the spread of the values in the sample cohort. Among highly expressed miRNAs in prefrontal cortex are miR-181a-5p, miR-124-3p, miR-302 d-3p, miR-29b-3p, and miR-9-5p, and in hippocampus are indicated let-7a-5p, miR-124-3p and miR-127-3p

Fig. 3

miRNAScope control probes in old formalin-fixed paraffin-embedded (FFPE) blocks. Serial tissue sections were obtained with FFPE samples from prefrontal cortex (panel A) and hippocampus (panel B) and stained using the automated miRNAScope technology. Representative images show staining with probes for miR-124-3p, snRNA U6 and scramble. High staining intensity (red stain) was obtained with miR-124-3p and snRNA U6 probes in both prefrontal cortex and hippocampus, and virtually no staining with the scramble probe

Fig. 4

Variation in the performance of old formalin-fixed paraffin-embedded blocks. Tissue sections were obtained from 5 (A-E) prefrontal cortex samples, and stained for miR-124-3p using automated miRNAScope, then scanned to obtain digital whole slides (upper row) from which a subset region is shown covering grey and white matter interface (lower row). Case A and D performed with highest staining intensity, B and E with intermediate staining intensity and C with the lowest miR-124-3p staining intensity

Fig. 5

Expression of selected microRNAs in prefrontal cortex, grey matter investigated using in situ Hybridization (ISH). Tissue sections from prefrontal cortex samples were stained for selected miRNAs (Table 2) using automated miRNAScope. Intense ISH signal is seen in the grey matter (GM) with the miR-124-3p probe (upper row) at both high and low magnification, whereas probes for the other miRNAs show a discrete or no signal (and therefore only shown at high magnification). Positive ISH signal (examples indicated by circles) is seen with probes against miR-9-5p, miR-29b-3p, miR-125b-5p, miR-138-5p and miR-181a-5p whereas no ISH signal is seen with probes to miR-30c-5p, miR-224-5p, miR-302 d-3p and miR-432-5p

Fig. 6

Expression of selected microRNAs in hippocampus investigated using in situ Hybridization (ISH). Tissue sections from hippocampus samples were stained with selected microRNA probes (Table 3) using automated miRNAScope. Discrete diffuse staining is seen with probes to let-7a-5p and miR-7-5p in the dentate granule cell layer. Intense ISH signal is seen with the miR-124-3p probe, whereas no ISH signal is seen with probes to miR-127-3p, miR-145-5p and miR-149-5p in the dentate granule cell layer. The miR-145-5p probe stained capillaries (c) and arteries (a) in hippocampus samples