The laser capture microdissection technique enables isolation of specific cells and tissues and thereby removes the noise of gene expression from other cells or tissues in the gene expression profile. A protocol developed for laser microdissection of human gonocytes was adjusted and optimised to isolate juvenile zebrafish gonads.

The juvenile zebrafish gonad is not morphologically distinguishable when using dehydrated cryosections on membrane slides and a specific staining method is necessary to identify the gonads. The protocol setup in this study allows staining, identification, isolation and subsequent RNA purification and amplification of gonads from individual juvenile zebrafish thereby enabling gonadal gene expression profiling.

Zebrafish were embedded in Tissue-Tech Optimal Cutting Temperature (OCT) compound (Sakura Fintek Europe, Zoeterwonde, NL), rapidly frozen in isopentan on dryice and stored at -80°C until further analysis. The frozen tissue was cut in 20 μm serial sections on a Cryostat and either mounted on RNase free membrane slides (Molecular Machines & Industries, Glatbrugg, Switzerland) for microdissection or on Superfrost slides for HE staining. The membrane slides were fixed immediately in 75% ethanol for 10 min. at room temperature, stored in 100% ethanol in -80°C and was then stained with NBT BCIP as described previously [11]. In short, membrane slides were placed: 10 sec in incubation buffer, 90-120 sec. in NBT BCIP solution [262.5 μg/mL p-Nitro-Blue tetrazolium chloride; 225 μg/mL 5-Bromo-4-chloro-3-indolyl phosphate dipotassium salt (both from Sigma-Aldrich, St Louis, MO, USA); dissolved in 70% and 100% dimethyl formamide, respectively], 10 sec in DEPC water, followed by dehydration in ethanol (10 sec 62% ethanol, 2× 10 sec 96% ethanol and 2× 10 sec 100% ethanol). The stained germ cells and the surrounding area corresponding to the gonads were dissected using Olympus SmartCut microdissection system according to manufacturers instructions. Lysis buffer from RNAqueous-Micro Kit (Ambion, Austin, TX, USA) was added to the microdissected tissue as fast as possible and samples were stored at -80°C until further analysis. RNA was purified using RNAqueous-Micro Kit according to the manufacturers instructions for microdissected tissue. The quality of the RNA was analysed using Agilent Bioanalyser 2100 and Agilent RNA 6000 Pico Kit (Agilent Technologies, Santa Clara, CA, USA). The RNA was amplified using MessageAmpII™ II aRNA Amplification Kit in two rounds. After both rounds of amplification, samples were quantified using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA).

cDNA synthesis was performed using 0.5 μg amplified RNA and 50 ng/μl random hexamer primers in a final volume of 20 μl. cDNA control was performed without RNA. Reverse transcriptase polymerase chain reaction (RT-PCR) using 1 μl cDNA and gene specific primers placed just upstream of the polyA site was performed in (final concentrations): 10 mM Tris-HCl, pH 8.3; 50 mM KCl; 1.8 mM MgCl₂; 0.1% Triton X-100; 0.0005% gelatine; 250 μM dNTP and 1 pmol/μl primer. Specific primers targeting vasa (fwd: CATCGCATAGGAAGAACTGGA; rev: GGCTCATCGCTCTTGAAGGAT) and β-actin (fwd: AGTGCGACGTGGACATCCGTA; rev: GCACTTCCTGTGGACGATGGA) were designed to span intron-exon boundaries. Cycle conditions were: one cycle of 2 min at 95°C; 40 cycles of 30 sec at 95°C, 1 min at 62°C, 1 min at 72°C and finally one cycle of 5 min at 72°C. PCR products were run on 1% agarose gels and visualized by ethidium bromide staining. Bands from each primer combination were excised and sequenced for verification (DNA Technology, Aarhus, Denmark).

Probes for ISH were prepared by RT-PCR amplification of vasa transcripts and reamplification of PCR fragments using nested primers specific to the fragments with an added T3-promotor sequence in combination with the T7-extended downstream primer (AATTAACCCTCACTAAAGGGCTGTGGGAACACC and TAATACGACTCACTATAGGGCCCTTCCGCGGAGT; T3 and T7 promoter sequences are in bold). PCR conditions were: 5 min 95°C; 5 cycles of 30 sec 95°C, 1 min 45°C, 1 min 72°C; 20 cycles of 30 sec 95°C, 1 min 65°C, 1 min 72°C and finally 5 min 72°C. The resulting PCR product was purified on a 1% agarose gel and sequenced. Aliquots of 200 ng were used for in vitro transcription labelling with biotin-16-UTP (Roche Diagnostics GmbH, Germany), using the MEGAscript-T3 (sense) or MEGAscript-T7 (anti-sense) kits, as described by the manufacturer (Ambion/ABI, Austin, TX, USA). To estimate quantity and labelling efficiencies, aliquots of the labelled RNA product were analysed by agarose gel (1%) electrophoresis. Bands from each primer combination were excised and sequenced for verification (DNA Technology, Aarhus, Denmark).

The morphology of cryo-sectioned dehydrated tissue is inferior compared to tissue in fixative and when membrane slides used for microdissection are used instead of glass slides, the morphology is further impaired. Therefore, it is difficult to distinguish between different organs in juvenile zebrafish and a specific staining method to the tissue of interest is absolutely necessary. In a previous study, Sonne et al. (2009) stained gonocytes in human testis for expression of alkaline phosphatase visualized with NBT BCIP and used the microdissected cells for microarray analysis [18]. As the expression of alkaline phosphatase is conserved in zebrafish gonocytes, we adapted the method to stain fetal germ cells in gonads of juvenile zebrafish. Staining of zebrafish cryosections with NBT BCIP showed a clear colouring of the gonad and HE staining of the next serial cryocut confirmed the location of the gonad (Figure (Figure2).2). This ensured that when zebrafish cryosections are stained with NBT BCIP it is in fact the germ cells of the gonad that are specifically stained and this allows microdissection of cryosections with impaired morphology. The NBT BCIP staining was specific to the gonad, however, in all investigated fish we also observed staining of the eyes inside the natural black pigmentation (Figure 3A-B). The morphology of the NBT BCIP stained cryosections was compatible with laser microdissection and yielded RNA of a quality suitable for further analysis including RT-PCR. However, to obtain enough tissue for RNA purification it was necessary to microdissect serial cryocuts corresponding to the entire gonad of each individual fish (Figure 3C-D), which amounted to approximately 2 × 10⁶ μm². After RNA purification and two rounds of linear amplification a total amount of 21 -180 μg RNA was available from each gonad. To ensure that the microdissected tissue indeed was from the gonads, expression of vasa was investigated by RT-PCR. Figure Figure44 shows that vasa was expressed in the three samples of microdissected gonads whereas it was not detected in microdissected non-gonadal tissue, which primarily consisted of muscle tissue. To our knowledge, this is the first study allowing isolation of the gonad from individual juvenile zebrafish. However, a recent study by Vinas & Piferrer (2008) described laser microdissection of different spermatogenic stages from Mayer haematoxylin stained testis of adult sea bass males [19].

The best morphology and the highest RNA quality were obtained when the alkaline phosphatase staining protocol was applied to cryocuts fixed in 75% ethanol [11]. However, the staining itself resulted in a rapid decrease in RNA quality and in this study we performed the microdissection as fast as possible after staining in order to minimise RNA degradation. The quality of the purified RNA from microdissected gonad was analysed using BioAnalyzer picogel, which allows a visual inspection of RNA integrity and generates ribosomal RNA ratios, a RNA Integrity Number (RIN) that is a standardised RNA quality control (Figure (Figure5).5). In this study the visual inspection of the picogel was weighted more important than the RIN value as the RIN application states that accurate values cannot be obtained below concentrations of 25 ng/μl. We measured concentrations of only 10-20 ng RNA/μl after the first round of linear application using a Nanodrop spectrophotometer, which means that the concentration in the sample before amplification is much lower most likely in the pg/μl range and therefore not within the limits that allow determination of accurate RIN values. Both the visual inspection of the picogel and the RIN values (3.5-5.5) indicated some degradation of RNA. However, according to earlier studies, it is possible to investigate gene expression profiles obtained from partially degraded RNA, which still have visible ribosomal bands and obtain reliable results if used carefully [3,9,10,12,20]. The linear amplification method used in this study has the advantage compared to exponential RNA amplification methods, such as reverse transcriptase-PCR, that it maintains the representation of the starting RNA population [6,21,22]. However, when comparing to the exponential PCR based approach the amplification is not as efficient. Generally, amplification of microdissected tissue that is partially degraded does inevitable result in RNA truncated corresponding to a loss of the 5'-end of the transcript [23] and therefore all primers used in this study lies within the 3'-end of the gene.