A simple and rapid method for combining fluorescent in situ RNA hybridization (FISH) and immunofluorescence in the C. elegans germline

Graphical abstract


Background
Tracking expression and subcellular localization of RNAs and proteins during development provides important clues to understand their biological and physiological function. In conventional in situ RNA hybridization methods, sliced tissues are incubated with biotin or digoxygenin-labeled RNA probes that are synthesized by PCR, and then endogenous RNAs:RNA probe hybrids are visualized by antibodies against biotin or digoxygenin ( Fig. 1A and B). This two-step procedure exhibited low sensitivity and non-specific probe binding. Therefore, a highly sensitive RNA detection method was required to analyze the in vivo subcellular localization of endogenous RNAs with better preservation of the fine structure.
Recently, an advanced fluorescent in situ RNA hybridization (henceforth called "FISH") method using fluorescent (e.g., Alexa 594, Alexa 488, Cy3, or Cy5)-labelled small RNA molecule ($20 nucleotides) probes has been developed ( Fig. 1C and D) in multiple organisms, including C. elegans whole worms and embryos [1], yeast [2], zebrafish embryos [3], and Drosophila egg chamber [4]. Particularly, this FISH method enables the analysis of several different RNAs simultaneously using multi-color fluorescence and the analysis of transcriptional activity at the cellular level. Although this method has made a significant progression in analyzing the expression pattern and subcellular localization of the specific RNAs, combining FISH with immunofluorescence (henceforth called "IF") method to detect RNAs and their associated proteins simultaneously is a practically challenge due to different hybridization conditions (Saline-sodium citrate (SSC) for FISH; Phosphate-buffered saline (PBS) for IF), different reaction temperatures (!30 C, usually with formamide for FISH; 4-24 C, no harsh chemicals for IF), and RNase sensitivity [5] (Table 1). To date, this combining method has been developed only in Drosophila egg chamber [4] and embryos [5]. Therefore, developing the FISH/IF combining protocol will lead to advanced biochemical, cellular, and functional analysis of RNA:protein complex in the specific cells or tissues.
The nematode Caenorhabditis elegans (C. elegans) is an attractive model organism for study of basic biological and biomedical sciences. Particularly, RNA regulations by RNA-binding factors (e.g., RNA-binding proteins and microRNAs) have been studied intensively in the C. elegans germline [6,7]. These RNA regulations control germline stem cell (GSC) maintenance, mitosis/meiosis decision, and sex determination [8]. Aberrant regulations result in sterility and germline tumors [7]. Therefore, the simultaneous analysis of RNAs and RNA-binding proteins in the specific cell types and tissues provides a powerful tool for investigating its cellular and physiological function in vivo. To date, in C. elegans, FISH method has been developed in only whole worms and embryos [1], but not in germlines. Moreover, FISH/IF combining protocol has not yet been developed because the reaction conditions for the two procedures conflict in multiple ways (Table 1). We here present a simple protocol to visualize RNAs and proteins simultaneously using FISH and IF in the same C. elegans germline. This method enables detection of RNA/protein complex with better preservation of the fine structure at the cellular level.

Probe design and synthesis
Stellaris FISH probes were designed as described by Ji and van Oudenaarden [1]. Briefly, FISH probe length is 17-22 base pairs, while probe spacing is no less than 2 base pairs. Probe GC contents are about 45%, and the number of probes for each target RNA are between 30 and 96 depending on target transcript length. Typically, 48 probes are used to ensure good signal quality. A web-based probe design software is available at: http://www.biosearchtech.com/stellarisdesigner/ C. elegans culture and synchronize worm 1. Worms are cultured on 100 mm NGM (Nematode Growth Media) agar plates seeded with OP50 E. coli bacteria at permissive temperature (20 C), as described [9].
2. 4-5 days after plating, (when there are many adult worms on NGM agar plates), wash worms off NGM agar plates with 5 ml of M9 buffer per plate and transfer them into a 15 ml conical tube.
3. Pellet worms by table-top centrifugation at 1000 rpm for 1 min and carefully remove the supernatant.
4. Add 10 ml bleaching solution to the conical tube with the worm pellet. The subcellular localization of ssq-1 (sperm-specific family, class Q) mRNA in C. elegans spermatocytes. Primary spermatocyte (left) and secondary spermatocyte (right). Scale bar: 3 mm.

5.
Wait until over 90% of the worms are lysed and several embryos remain. (typically, this takes about 5-6 min).
Note: Vortexing every one minute helps the bleach lysis. Bleaching longer than 10 min will damage embryos.
6. Spin down the embryos at 1000 rpm for 1 min and remove the bleaching solution. 7. Wash three times with 10 ml of M9 buffer. 8. After the last centrifugation, add 10 ml of M9 solution to the tube with embryos and allow the embryos to hatch in 20 C incubator by locking the tube gently.
9. About 12 h later, spin down worms at 1000 rpm for 1 min, remove the supernatant, and monitor hatched L1 worms by placing 1 drop of the samples on a slide glass and estimate the number of hatched L1 worms.
10. Plate about 200 L1 worms on an NGM agar plate seeded with OP50 E. coli bacteria. 11. Incubate them for three days at 20 C.

Germline dissection and fixation
12. Transfer about 50 adult worms to a glass dish using a platinum worm picker. Note: Alternatively, wash worms off a plate with 1.5 ml of PTW/Levamisole and transfer to a glass dish (60 Â 15 mm), filled 5 ml PTW/Levamisole solution.
13. Once worms are paralyzed (1-2 min), cut off head or tail parts using a disposable scalpel under a dissecting microscope (Fig. 2).
Note: Dissecting 50 gonads within 15 min will make better germline morphologies. 14. Transfer dissected worms to a 1.5 ml microcentrifuge tube. 15. Spin down at about 8000 rpm for 2-3 s using a mini-centrifuge and carefully remove the supernatant using a micropipette. 16. Add 200 ml of 3% paraformaldehyde (PFA) fixation solution to the microcentrifuge tube and incubate 10-30 min at room temperature.
Note: For tumorous germlines, 30-min fixation will help to make better germline morphologies. 19. After the last centrifugation, carefully remove the PTW solution, add 200 ml of ice-cold 100% methanol to the microcentrifuge tube, and incubate for 10 min at À20 C. Note: Alternatively, fixed gonads can be stored in cold methanol at À20 C for a few days.  Note: To distinguish FISH and IF signals, use secondary antibodies conjugated with different fluorochromes from that used for FISH probe (i.e., Cy3 (Red) for FISH and Alexa 488 (Green) for IF).
34. Carefully remove the supernatant by spinning down at about 8000 rpm for 2-3 s using a mini-centrifuge.
35. Add DAPI solution (100 ng/mL) to the microcentrifuge tube and incubate them for 10 min at room temperature.
36. Carefully remove the supernatant by spinning down about 8000 rpm for 2-3 min using a mini-centrifuge, and then wash three times using PTW/BSA solution for at least 30 min (10-min interval).

Image acquisition
37. Completely dissolve 2% agar using a microwave. 38. Drop 2% agar liquid on the center of glass slide, and immediately place another glass slide on the glass slide with agar drop.
39. After gelation, carefully remove the upper glass slide, and then transfer the stained germline samples onto the agar pad on glass slide and immediately aspirate the residual liquid using a mouth pipette.
40. Drop 10 ml of antifade mounting solution (e.g., VECTASHIELD) onto the stained germline samples and carefully spread them to avoid the overlap of samples. 41. Carefully place a coverslip over the top to put the sample between the agar pad and coverslip. Note: If necessary, seal the coverslip with nail polish to prevent evaporation of antifade mounting solution.
42. Take pictures using a fluorescence microscopy and analyze the results.

Method validation
In this report, we first developed a simple combining protocol by adapting the FISH labeling procedure from previous protocols [1] and by optimizing reaction conditions for both FISH and IF using C. elegans germline (Table 1). Current FISH and IF procedures usually take three or four days in C. elegans whole worms [1]. Furthermore, it has been generally accepted that combining RNA FISH with IF can be technically challenging because of RNase contamination in IF reaction solutions (especially crude antibodies). However, our combined protocol can be done in one or two days in the dissected gonads, depending on antibody sensitivity. In addition, tissue samples that were fixed with 3-4% paraformaldehyde and/or methanol retain more RNAs compared to other conventional fixatives, such as Carnoy's solution (60% ethanol, 30% chloroform and 10% glacial acetic acid), and therefore show strong and stable FISH signals [10][11][12]. We used this approach to examine the expression of msp mRNAs and MSP proteins (Fig. 3). The MSPs (Major Sperm Proteins) are encoded in the C. elegans genome by a multigene family in small clusters at three chromosomal loci [13]. In spite of their dispersed locations, all of the MSP genes are expressed exclusively in the spermatogenic germline, but not in the oogenic germlines. This protocol was thus tested in three different germlines: wild-type hermaphrodite with both sperm and oocytes (Fig. 3A), wild-type male germlines with only spermatogenic germ cells (Fig. 3B), and fog-3(q520) mutant germlines [14] with only oogenic germ cells as a negative control (Fig. 3C). FOG-3 is critical for sperm fate specification [14,15]. For FISH experiment, we used Stellar FISH probes for msp gene as previously described [1,16]. For both FISH and IF, we tested three different fixation conditions: (i) 4% paraformaldehyde/1X PBS followed by 70% Ethanol, (ii) 3% paraformaldehyde/0.25% glutaraldehyde/0.1 M K 2 HPO 4 (pH7.2), and (iii) 3% paraformaldehyde/0.1 M K 2 HPO 4 (pH 7.2) followed by 100% Methanol. We found that only protocol (iii) reserved both signals: msp RNAs (Red in Fig. 3A and B) and MSP proteins (Green in Fig. 3A and B) were highly accumulated in both the late pachytene meiotic germ cells and primary (1 ) spermatocytes ( Fig. 3A and B). Notably, MSP proteins persisted in secondary (2 ) spermatocytes and mature sperm (Fig. 3B), whereas msp mRNAs were dramatically decreased in secondary spermatocytes and were undetectable in mature sperm (Fig. 3B). We also obtained similar results with another RNA probes for sperm-specific gene, ssq-1 (Sperm-Specific family, class Q) [16] (data not shown). Their specificities were determined in the feminized fog-3(q520) mutant germline: no msp RNAs and MSP proteins were detected.
We next tested two different protocols to find the better condition: (i) IF followed by FISH and (ii) FISH followed by IF. We found that protocol (i) resulted in a decrease of the IF signals, indicating that FISH hybridization condition (SSC buffer, higher hybridization temperature, and formamide chemical) may distort IF signals. However, protocol (ii) reserved both signals ( Fig. 3A and B). The advantage of this protocol is three-fold: (1) A simplified FISH protocol with better preservation of the fine structure, (2) Reduced non-specific somatic autofluorescence: FISH and IF in the extruded germlines eliminate autofluorescence signals from somatic tissues (e.g., intestine), and (3) Combined FISH and immunofluorescence in the same germline. Therefore, our combining protocol may significantly contribute to the study of RNA/protein regulation during C. elegans germline development. However, this FISH/IF combining protocol is limited to the C. elegans germline. Therefore, finding the simple combining labeling protocol in C. elegans whole worms and embryos will provide a benefit to better understand the mechanism of RNA/protein regulation in other tissues.

Conflict of interests
The authors declare no conflict of interest.