The influence of preparatory fixing and staining protocols and the subsequent UV laser microdissection of murine brain tissue sections (post mortem) on the RNA quality is described below. The RIN numbers of cryo-asserved untreated brain sections were compared with sections that were treated and isolated according to standard protocol. Figure 1 shows the sample preparation workflow from cryosection to RIN analysis.
Cryosection with subsequent staining and fixing
To compare untreated tissue samples with fixed and stained tissue samples, three cryo-asserved murine brains of known RNA integrity (RIN) were used. After cryo-asservation at –80 °C, these were transferred to a cryomicrotome and incubated at –35 °C for 45 min. Then the tissue was equilibrated at –18 °C for 30 min and cut into 12 µm sections at –18 °C by cryosectioning. The murine brain sections were divided into two groups which were treated alternately. One (the control group) was directly frozen at –80 °C. The other group was applied to an RNase-free, UV-treated PEN slide. This was done by briefly thawing the sections at room temperature and fixing them for 2 min in 75 % ethanol cooled to –20 °C. After fixation, they were immediately stained with drops of sterile-filtered cresyl violet (CV) dye solution (1 % cresyl violet in 100 % ethanol) and incubated for 45 sec with slight to and fro movement. After this staining procedure, the dye solution was briefly drained off and the slide was immersed in an ascending sequence of ethanol concentrations (75 %, 95 % and 100 %) for 4 sec each and then finally fixed in 100 % ethanol (test group). Following this fixing and staining procedure, the slide was dried in a drying chamber with silica gel for 45 min at room temperature and then stored at –80 °C in the dry box pending laser microdissection.
UV laser microdissection
After fixing and staining, the applied tissue was isolated using the gentle and contamination-free technique of UV laser microdissection. This was done by marking the tissue and dividing it into rectangles using the 10x objective before completely ablating it with the “Draw + Cut” mode using sufficient energy. The dissectates were collected in 0.5 ml RNase-free caps to allow purification of the RNA content. These were stored at –80 °C pending RNA isolation.
Parallel RNA isolation
Both groups – the untreated samples stored at –80 °C, and the fixed, stained and laser microdissected samples – were thawed in parallel and treated with the same reagents simultaneously. The RNeasy Mini Kit® of the Qiagen company was used for purification, following the instructions of the manufacturer. The purified RNA was eluted with 30 µl RNase-free water into a fresh RNase-free test tube.
Comparison of the RIN values regarding RNA quality
To compare the RNA quality of native cryo tissue sections with that of CV-stained, EtOH-fixed, laser microdissected tissue sections, RNA of the differently treated samples was applied to the same RNA nanochip (Agilent Bioanalyzer). The RNA nanochip separates RNA samples by capillary gel electrophoresis and grades the quality of the samples according to the way they move in the chip. The quality is expressed as a RIN value: a RIN number of 10 indicates completely intact RNA and a RIN number of 1 completely degraded RNA.
The RIN analysis conducted here (sample measurement in Fig. 2) showed that the RIN values of RNA from directly frozen native tissue were not much different from those treated according to the described UV-LMD standard protocol.
The isolated RNA samples were then divided into two groups and one half was precipitated. After RNA precipitation (according to the instructions given in “Improved quantitative real-time RT-PCR for expression profiling of individual cells”, Liss, 2002), the RIN values of the tissue samples from native, precipitated cryo tissue were slightly lower than the RIN value of RNA from CV-stained, EtOH-fixed, UV-LMD-treated tissue. The difference can be attributed to the poly-I of the precipitation mixture – some of these short-strand RNA molecules are purified and detected with the other RNA during the RIN measurement in the chip and counted as part of the degraded proportion. In qPCR reactions the poly-I does not show a negative influence. Therefore, fixation, staining and UV laser microdissection of the brain sections with or without RNA precipitation do not impair RNA quality. Figure 2 shows a comparison of the RNA quality of native, fixed and stained murine brain sections.
Altogether, it was shown that UV laser microdissection with the method described (Gründemann et al., http://www.ncbi.nlm.nih.gov/pubmed/21761319) does not significantly impair the RNA quality of cryo-asserved brain tissue. Such techniques offer an appropriate method of sample preparation and non-destructive isolation of single cells to obtain RNA of unchanged quality.