Single Nucleotide Polymorphism-sensitive FISH Detection of Locus-specific Ribosomal RNA Transcription in Drosophila melanogaster

Summary

The protocol describes single nucleotide polymorphisms-sensitive fluorescence in situ hybridization (SNP-FISH) method to distinguish ribosomal RNA transcripts derived from the X or Y chromosome ribosomal DNA locus in Drosophila melanogaster.

Abstract

In order to ensure sufficient ribosomal RNA (rRNA) is transcribed for ribosome function, genomes contain hundreds of tandem duplications of the sequences that encode rRNAs, composing regions called ribosomal DNA (rDNA) loci. Many organisms' genomes contain more than one rDNA locus distributed across different chromosomes, and rRNA may be transcribed from multiple rDNA loci or only a single locus. Changes in rRNA transcription sources often indicate ribosomal distress. However, the homogeneity of rRNA obstructs distinguishing if rRNA is transcribed from multiple or a single rDNA locus. Here, we describe a method that uses single nucleotide polymorphism-sensitive fluorescence in situ hybridization (SNP-FISH) to distinguish rRNA transcription between Drosophila melanogaster rDNA loci on the X and Y chromosomes. This method leverages locus-specific rRNA variants to enable easy detection of the source of rRNA transcription with single-cell resolution. This assay can be applied to any Drosophila cell type and may be adapted for use in other systems.

Introduction

The 18S, 5.8S, and 28S ribosomal RNAs (rRNAs) required for ribosome function are co-transcribed in a single cistron called the 45S pre-rRNA. The three rRNAs are separated within the 45S pre-rRNA by two internal transcribed spacer sequences (ITS) and flanked by external transcribed spacer sequences (ETS) at the 5' and 3' end. All of these spacer sequences are removed from the 45S pre-rRNA for the mature rRNAs to be incorporated into ribosomes (see¹). In order to meet the high demand for rRNA production needed to support ribosome activity, all Eukaryotic genomes contain hundreds of copies of the 45S sequence. These copies are clustered together in tandem repeats, forming genomic regions called ribosomal DNA (rDNA) loci. Most Eukaryotic genomes contain multiple rDNA loci spread across separate chromosomes (see²). The high redundancy of 45S copies means there are typically many more 45S copies than necessary for transcription, and the majority of 45S copies are transcriptionally silent and buried in heterochromatin³. Transcriptional activity is not uniform across rDNA loci, and variation in rDNA locus transcription is widely observed across organisms^4,5,6, indicating that 45S transcription is differentially regulated at individual rDNA loci. Locus-wide rDNA silencing has been observed in plants, invertebrates, and vertebrates^7,8,9,10, suggesting that locus-wide rDNA silencing is a major mechanism for regulating rRNA dosage¹¹. rRNA transcription is a key regulatory step in ribosome production, and changes to rRNA transcription that modulate translational activity are an important feature of both normal differentiation and disease conditions¹². Changes in rDNA locus transcription are also associated with reductions in total rDNA copy number¹³. Thus, altered locus-specific rDNA transcription may be an important indicator of disrupted cellular physiology.

While locus-specific silencing appears to be a major source of rRNA transcriptional regulation, the mechanisms that establish this silencing and direct which rDNA loci are transcriptionally active are largely unknown. The homogeneity of rDNA sequences makes it difficult to assess individual rDNA locus transcription, preventing widespread analysis of locus-specific rDNA transcriptional activity. This challenge may be possible to overcome by taking advantage of the structural and single nucleotide sequence variation between rDNA copies within the same genome^4,5,6,13. Some of these variants may be shared among most or all copies in an individual locus, creating unique rDNA locus haplotypes of multiple variants that distinguish an rDNA locus. Indeed, recent mapping of rDNA variants to specific loci by the telomere-to-telomere consortium revealed locus-specific shared rDNA variants in the human genome¹⁴. Such rDNA locus maps may enable the identification of locus-specific transcription from sequencing datasets; however, the availability of these maps currently remains limited and are not easy to produce. Since rDNA loci only reside on the sex chromosomes in Drosophila melanogaster (one locus on each X and Y chromosome)¹⁵, the Y chromosome rDNA locus is absent from XX females. This natural isolation from the Y chromosome locus means single nucleotide polymorphism (SNP) variants between the X and Y rDNA loci can be easily identified in short-read sequencing as any variants that are present in males (XY) but absent from females (XX). Previously identified SNPs can be rapidly tested in ungenotyped Drosophila strains through simple PCR and Sanger sequencing of DNA from males and females. Furthermore, the availability of Drosophila strains with an X chromosome that has no rDNA locus¹⁶ means individual X and Y rDNA loci can be individually isolated for even more precise rDNA SNP characterization. This easy identification of SNP variants between the Drosophila rDNA loci makes it possible to determine unique X and Y rDNA loci SNP haplotypes¹³. X chromosome rDNA loci are also typically completely silent in Drosophila males, but both X chromosome loci are transcribed in females^10,17, making Drosophila a useful system to study locus-wide rDNA silencing.

Fluorescence in situ hybridization (FISH) is a powerful tool to visualize the presence of specific RNA or DNA sequences within the individual cells of a tissue. FISH labels target RNA or DNA sequences through antisense oligonucleotide probes conjugated to fluorophores. FISH probes are typically required to be ~20 nucleotides long in order to provide stable enough target binding to be visualized but probes this long can also easily bind to non-target sequences that differ by only a single nucleotide (Figure 1A). Conversely, probes short enough to confer single-nucleotide specificity do not bind stably enough for visualization. To balance specificity and stability, SNP-sensitive FISH (SNP-FISH) combines a ~26 nucleotide long antisense fluorescent probe with a non-fluorescent sense mask oligonucleotide that binds to all but the 5' end of the probe¹⁸. This mask leaves only the 10 most 5' nucleotides of the probe single-stranded and available for target binding (Figure 1B). This short single-stranded portion of the probe confers specificity to the target but prevents cross-reactivity with RNAs that have even a single mismatch with these 10 nucleotides (Figure 1B). Once the 5' end of the probe binds its target, passive strand displacement strips the mask from the probe, allowing the 3' region to bind the target and creating stable target labeling (Figure 1B)¹⁸. Therefore, SNP-FISH can differentially visualize RNAs that differ by a single SNP by using a pair of probes that have different fluorophores that each complement a different SNP but share a common mask (Figure 1C). Although this method only recruits a single fluorophore-conjugated probe to the target SNP, pooling multiple probe-mask sets to several SNPs within a shared rDNA haplotype can be used to amplify the SNP-sensitive detection of a single rRNA (Figure 1D). Furthermore, because rRNA is so abundantly transcribed, rRNA transcripts can be easily visualized in FISH experiments using a small number of fluorescent labels per RNA. This low requirement for fluorophore per RNA means that SNP-FISH can visualize rRNAs from a specific locus with only a few unique SNPs. This method can easily detect locus-specific rRNA transcription in a variety of Drosophila tissues and developmental stages¹⁷. It is particularly effective in Drosophila male germline stem cells, which change between exclusive Y-rDNA transcription and co-expression of X- and Y-chromosome rDNA loci during aging¹³. Here, we provide an SNP-FISH protocol to visualize distinct X and Y rRNA transcripts in the Drosophila testis to achieve the overall goal of assessing locus-specific rRNA silencing. This method uses four SNPs previously characterized between the rDNA loci on the X and Y chromosomes of the common y¹w¹ laboratory Drosophila strain (Figure 1D).

Protocol

1. Preparation of buffers and reagents

NOTE: RNase-free technique is to be used throughout steps 1, 3, and 4, and certified RNase-free reagents should be used.

1. In a chemical hood, prepare 40 mL of 4% formaldehyde in 1x PBS fixation solution by adding 10 mL of 16% formaldehyde and 4 mL RNase-free 10x PBS to 26 mL of RNase-free water. Distribute into 1 mL aliquots and store at -20 °C within 30 min of preparing. Fixation solution can be stored at -20 °C for up to 6 months.
   CAUTION: The solution contains formaldehyde. Handle with gloves and appropriate PPE and safely dispose of it.
2. Prepare 10 mL of hybridization buffer by mixing 1 mL of 20x RNase-free saline-sodium citrate (SSC), 1 g dextran sulfate, 100 µL of 100 mg/mL Saccharomyces cerevisiae tRNA, 100 µL of 200 mM Vanadyl ribonucleoside complex, 1 mL of 5% RNase-free BSA, 1 mL of deionized formamide combined with 6 mL of RNase-free water. Distribute into 1 mL aliquots in 1.5 mL microfuge tubes and store at -20 °C for up to 1 month.
3. Prepare 50 mL of wash buffer by adding 5 mL of 20x RNase-free SSC, 5 mL of deionized formamide, 50 µL of Triton-X to 40 mL of RNase-free water. Store at room temperature in the dark for up to 1 month.
4. Prepare 10 mL of DNA isolation buffer by adding 100 µL of 1M Tris-HCl pH 7.5-8.0, 20 µL of 0.5M EDTA, and 50 µL of 5M NaCl to 9.8 mL of water.

2. Sample genotyping for X- and Y-specific rRNA SNPs

1. Collect individual unmated homozygous X female and male harboring the same X chromosome in 0.2 mL tubes and chill on ice.
2. Combine 50 µL of DNA isolation buffer with 0.5 µL of 20 mg / mL Proteinase K per sample.
3. Add 50 µL of DNA isolation buffer / Proteinase K to the Drosophila sample and homogenize the sample using a pipette tip.
4. Incubate samples at 37°C for 30 minutes, followed by 95°C for 2 minutes. Transfer 40 mL of sample (avoid animal debris) to a clean 1.5 mL tube.
5. Separately amplify each SNP target region from each DNA sample by PCR using 10 µM concentration of the following primer pairs: 18S SNP Forward + Reverse; ITS1 SNP Forward + Reverse; 28S SNP Forward + Reverse. The primer sequence and expected PCR amplicon size are found in Table 1.
6. Use gel electrophoresis to separate the entire PCR sample on a 1% agarose 1x TAE gel to confirm the expected PCR product. Expected product sizes are listed in Table 1.
7. Isolate PCR product from the gel using any commercially available gel extraction kit.
8. Sequence PCR products by Sanger sequencing. Sequence samples in both directions using separate individual reactions for the F and R primer for each target.
9. Determine the X chromosome SNP variant at the positions in Table 2 in an unmated female DNA sample. The expected X haplotype is described in Table 2.
10. Observe SNP positions in male DNA sample sequencing chromatogram. Infer Y chromosome SNP variant based on the already determined X SNP variant at each SNP position. The expected Y haplotype is described in Table 2.

Primer name	Sequence	Expected Amplicon Size
18S SNP F	GACTACCATGGTTGCAACGG	652 bp
18S SNP R	TTCACCTCTCGCGTCGTAAT
ITS1 SNP F	CTTGCGTGTTACGGTTGTTTC	955 bp
ITS1 SNP R	ACAGCATGGACTGCGATATG
28S SNP F	ATGCGTAGAAGTGTTTGGCG	598 bp
28S SNP R	GCCGACTTCCCTTACCTACA

Table 1: Primer sequences for sequencing rDNA SNPs. Primer pairs to amplify rDNA regions that contain previously characterized SNPs in the 18S, ITS1, and 28S rDNA regions. Primer oligonucleotide sequence and expected PCR amplicon size are listed.

Haplotype	SNP Position	Sequence
X rDNA	18S	1603-ATACTTGTATTTTTTCATATG-1625
	ITS1-1	2873-CGTTAATAAATATTTGTAATT-2895
	ITS1-2	3115-GAAAATCGAAGAAACAAAATT-3137
	28S	5932-AACAAAAATGCCTAACTATAT-5954
Y rDNA	18S	1603-ATACTTGTATCTTTTCATATG-1624
	ITS1-1	2873-CGTTAATAAACATTTGTAATT-2895
	ITS1-2	3115-GAAAATCGAAAAAACAAAATT-3137
	28S	5932-AACAAAAATGGCTAACTATAT-5954

Table 2: Expected rDNA haplotypes. Expected SNPs at X and Y chromosome rDNA loci in the y¹w¹ Drosophila strain. SNP indicated in bold and underlined. Position listed based on consensus 45S rRNA sequence (Supplementary file 1).

3. Testis dissection, fixation, and permeabilization

1. Clean stereomicroscope, workstation, dissecting dish, and dissecting forceps with 70% ethanol (or another RNase treatment).
2. Under a stereomicroscope, dissect Drosophila to isolate testes in 1x RNAse-free PBS. Grasp the middle of the animal's abdomen with one pair of forceps and grasp the end of the abdomen with another pair of forceps to gently pull the animal apart. Testes will dissociate from the animal and can be further separated using forceps. Collect 30 - 40 testes per sample.
3. Using forceps, pick up testes and transfer them from the dissecting dish to a 1.5 mL microfuge tube containing 0.5 mL of 1x RNase-free PBS.
4. Aspirate PBS and add 1mL of fixation solution. Incubate at room temperature on a nutating shaker for 30 min. Aspirate fixative.
5. Wash 2x with 1 mL of 1x RNase-free PBS for 5 min each on a nutating shaker. Aspirate 1x PBS and add 1 mL of RNase-free 70% ethanol (EtOH). Incubate at 4 °C overnight on a nutating shaker.

4. SNP-sensitive ribosomal RNA FISH

1. Aspirate ethanol and add 1 mL of wash buffer. Incubate at room temp for 3 min on a nutating shaker, followed by 2 min of tube resting upright.
2. Mix probes and masks with hybridization buffer, making 100 µL of total volume per sample. Probe, mask sequences, and fluorophore labels are listed in Table 3. When using all 4 SNPs, prepare 80 µL of hybridization buffer, 1 µL each of 10 µM Y-SNP probe (4 µL total), 1 µL each of 10 µM X-SNP probe (4 µL total), and 3 µL each of 10 µM common mask (12 µL total)
3. Aspirate wash buffer, then add 100 µL of hybridization solution. Apply transparent film on the edges of the tube, wrap in aluminum foil to protect from light, and incubate in a 37 °C water bath for at least 24 h.
4. Add 1 mL of wash buffer to the sample without aspirating the hybridization solution. Incubate in a 37 °C water bath for 30 min in the dark.
5. Aspirate wash buffer, then add 1 mL of wash buffer to the sample. Incubate in a 37 °C water bath for 30 min in the dark.
6. Aspirate wash buffer and add 50 µL of mounting media. Samples can be immediately mounted on slides or stored for up to 1 week at 4 °C prior to mounting.

SNP Position	Oligonucleotide	Sequence		3’ conjugated Fluorophore
18S	X SNP probe	AAAAAATACAAGTATTTAATCACATA		Alexa 488
	Y SNP probe	AAAAGATACAAGTATTTAATCACATA		Alexa 647
	Mask	TATGTGATTAAATACT
ITS1-1	X SNP probe	AAATATTTATTAACGGTAAGGATATT		Alexa 488
	Y SNP probe	AAATGTTTATTAACGGTAAGGATATT		Alexa 647
	Mask	AATATCCTTACCGTTA
ITS1-2	X SNP probe	GTTTCTTCGATTTTCATGTTCGAAAC		Alexa 488
	Y SNP probe	GTTTTTTCGATTTTCATGTTCGAAAC		Alexa 647
	Mask	GTTTCGAACATGAAAA
28S	X SNP probe	TTAGGCATTTTTGTTTTACTTGAAAA		Alexa 488
	Y SNP probe	TTAGCCATTTTTGTTTTACTTGAAAA		Alexa 647
	Mask	TTTTCAAGTAAAACAA

Table 3: Probe and mask sequences for rDNA SNP-FISH. SNP positions are highlighted in bold. The probe portion that binds to mask oligonucleotide is underlined. Examples of suitable 3' conjugated fluorophores are listed, but any compatible fluorophores can be used.

5. Preparation of samples for imaging

1. Working under a dissecting stereomicroscope, use a wide-orifice pipette to transfer the sample to a microscope slide.
2. Using forceps, arrange testes in a line for ease to find samples when imaging (no specific orientation required). Cover sample with a coverslip. Blot coverslip edges with a tissue wipe to remove excess mounting media.
3. Seal the edges of the coverslip with nail polish. Leave slide in dark at room temp for 10 min for nail polish to dry. Slides can be used immediately for confocal imaging or be stored for up to one month at 4 °C prior to imaging^.

Results

Sequencing results from SNP genotyping are expected to detect SNP differences between the X and Y rDNA loci. These SNPs are detected by directly assessing the SNP positions in sequencing chromatograms (Figure 2). Sequencing of female samples is expected to have a single sequencing signal at the SNP position, indicating SNP homozygosity between the two X chromosomes (Figure 2A). Male sample sequencing results are expected to have a double peak at the SNP position (Figure 2B). Based on the X chromosome genotype determined from female sample sequencing, the non-X variant is inferred to be a Y chromosome rDNA SNP (i.e., X = T, Y = C for the 18S SNP sequencing in Figure 2). Alternatively, a single sequencing peak at a SNP position in the male sample would indicate homozygosity between male and female rDNA loci at that SNP position, and that position would not be usable for rRNA SNP-FISH.

Following the protocol for SNP FISH, samples can be visualized by mounting on slides and placed under a sealed cover slip, followed by confocal microscopy. Using the probes listed in Table 3, the Y-derived rRNA signal is detected by Alexa Fluor 647 emission and the X-derived rRNA signal is detected by Alexa Fluor 488 emission. rRNA signals are expected to be observed in the nucleolus, which can be identified as the DAPI-poor hole in the nucleus (Figure 3A). The nucleolus is particularly easy to identify in germ cells due to their large nucleus and nucleolus (Figure 3A). The faint signal can typically also be found in the cytoplasm (Figure 3), but this is a non-specific signal that can also be detected in samples without rRNA containing a complementary SNP (i.e., Y signal in cells lacking Y rDNA or X signal in cells lacking X rDNA; Figure 3B-C). Furthermore, artificial co-labeling of X and Y rRNA can sometimes be detected in the somatic hub, even in samples lacking either X or Y rDNA loci (Figure 3B-C, yellow arrows). Thus, only nuclear signals should be used to assess locus-specific transcription. Most cells, particularly somatic cells, are expected to only have a Y rRNA signal, indicating that rRNA is exclusively transcribed from the Y rDNA locus (Figure 3A, yellow dotted circle, and red arrow)^10,17. However, co-expression of X and Y rRNA is often detected in germ cells, particularly germline stem cells¹³ (Figure 3A, white dotted circles). X and Y rRNA signals in co-expressing cells can be adjacent and form a single nucleolus (Figure 3A top cell) or can be separate, forming two nucleoli (Figure 3A bottom cell).The examples in Figure 3 are provided from rDNA SNP-FISH using probe sets targeting only two SNPs (the two ITS1 SNPs), demonstrating only two SNPs are necessary for rDNA SNP-FISH. However, more robust signals are observed when using probes targeting all four SNPs¹³.

Negative controls are an important inclusion for this assay to confirm that probes do not cross-react with the wrong SNP target, especially when first troubleshooting the method. X chromosome negative controls include any condition that lacks X chromosome rDNA, such as males harboring an X chromosome with an rDNA deletion. The X chromosome negative controls are expected to only detect Y rRNA signal and no X rRNA (Figure 3B). Y chromosome negative controls include any condition that lacks Y chromosome rDNA. Some examples of these conditions are any XX female tissues, tissue from males lacking a Y chromosome, or males harboring a Y chromosome with an rDNA deletion¹⁵. The Y chromosome negative controls are expected to only detect X rRNA signal and have no detectable Y rRNA signal (Figure 3C). Note that these controls are not only important to determine probe specificity, but also to determine any signal background or artifacts. Any novel probes or assay conditions that provide specificity in such controls can be accurately used to detect locus-specific rDNA transcription.

Different genetic, developmental, or environmental conditions may alter the likelihood that rDNA is exclusively transcribed from the Y-chromosome rDNA locus^13,17. The frequency of rDNA transcription from a single locus or multiple loci can be quantified and directly compared between samples. In order to quantitatively compare differences in rDNA locus transcription, each cell must be individually categorized as Y-dominant, Co-dominant, or X-dominant. Cells are only categorized as Y- and X-dominant if only that respective signal is detected. Any cell with both Y and X rRNA signals is considered co-dominant, even if one signal is much weaker than the other. Thus, cells containing very strong X and Y signals are qualitatively considered the same as cells with strong Y and weak X or strong X and weak Y signals. The total percentage of all cells across all samples in each category can be compared between samples by chi-squared test. While this assay works well to qualitatively determine if a particular rDNA locus is transcribed or not, we have not observed consistent evaluations between samples when directly quantifying the fluorescence intensity of individual SNP-FISH signals. Thus, we do not recommend making quantitative assessments of FISH signal intensity between samples or the relative intensity of X and Y rDNA signals within a single cell. The source of this inconsistency in fluorescence intensity is unclear, though it may be due to the inefficient binding of masked probes that do not bind to all of the target rRNAs.

Figure 1: Diagram of SNP-FISH method. (A) Traditional FISH antisense oligonucleotide probes cross-react with non-target RNAs that differ by only one nucleotide. (B) The SNP-FISH method uses complementary mask oligonucleotide bound to the 3' region of the oligonucleotide probe. The mask leaves only 10 nucleotides free to bind to the target sequence. An unmasked probe region does not stably bind to non-target sequences that differ by one or more nucleotides. Mask dissociates from probe after probe binds to target, allowing stable binding between probe and target. (C) Differently labeled paired SNP probes combined with a common mask specifically bind targets that differ by a single nucleotide without cross-reacting. (D) Multiple probes can be pooled together to specifically label distinct rRNA haplotypes. Four SNP differences have been characterized between X and Y rDNA loci in the y¹w¹ Drosophila melanogaster strain. One SNP in the 18S rRNA, one in the 28S rRNA, and two in the ITS1 portion of the pre-rRNA. X and Y-specific rRNA haplotypes used for SNP-FISH are displayed. Probes targeting the X rDNA variant at the four rRNA SNPs are conjugated to a common fluorophore (green), and probes targeting the Y rDNA variants are conjugated to a different fluorophore (magenta). A common mask is used for each SNP (four total). SNP-specific probe binding at each SNP position on the rRNA can specifically label rRNA from the X and Y rDNA loci with up to four FISH oligonucleotide probes. Please click here to view a larger version of this figure.

Figure 2: Examples of homozygous and heterozygous rDNA SNP sequencing results. Example sequencing chromatograms from 18S SNP sequencing of DNA from (A) female and (B) male samples. 18S SNP position indicated by an asterisk (*). The single thymine (T) signal for the SNP position in the female sample indicates a thymine at the SNP position in the X chromosome rDNA locus. The double signal at the SNP position split between thymine and cytosine (C) in the male sample indicates a cytosine at the SNP position in the Y chromosome rDNA locus (inferred by knowing thymine is at the X chromosome locus). Sequencing results were viewed using ApE – a plasmid editor software¹⁹. Please click here to view a larger version of this figure.

Figure 3: Examples of SNP-FISH results in the Drosophila testis using only two SNP-FISH probe sets. SNP FISH examples in the testis of animals with (A)both X and Y rDNA,(B) only Y rDNA, or (C) only X rDNA. These examples only used probe sets labeling the two ITS1 rRNA SNPs, demonstrating that rRNA SNP FISH works with as little as two target SNPs. Germ cells are identified by their large round nucleus, and somatic cells by their smaller and less round nucleus. Germline stem cells are identified by their position directly next to the somatic hub, marked by an asterisk (*). Germ cells that transcribe rDNA from both the X and Y rDNA loci are identified by both X and Y nuclear FISH signals (White dotted circles, A-A''). Germ cells that transcribe DNA from only the Y rDNA locus only have a Y nuclear FISH signal (Yellow dotted circle). Y-only expressing somatic cells are marked with a red arrow. Artificial co-labeling of X and Y rDNA loci can be observed in the somatic hub (yellow arrow). The scale bar is 10 µm. Please click here to view a larger version of this figure.

Supplementary File 1: rRNA sequence of X chromosome locus. Please click here to download this File.

Discussion

Here, we describe a method to use SNP-FISH to distinguish 45S rRNA transcripts derived from the X and Y chromosome rDNA loci in Drosophila melanogaster tissues. The most critical step in this protocol is the accurate genotyping of 45S SNPs to be used as SNP-FISH targets. We provide primers and protocol to genotype four known Drosophila 45S SNPs, but other sequencing methods may reveal novel SNPs that could alternatively be used for the assay. Any SNP positions found to be identical between X and Y chromosome rDNA loci (i.e., there is a single sequencing peak in male DNA samples) cannot be used for SNP FISH. Some SNP positions may be found to be heterogeneous within a single rDNA locus (i.e., sequencing results do not give a single SNP variant for XX samples or only a very minor second sequencing peak in XY samples). SNPs that are heterogenous within a single rDNA locus are also not suitable for this assay since one of the variants would be shared between the two rDNA loci, making transcription from a single locus or multiple loci indistinguishable. Furthermore, due to female sperm storage²⁰, DNA isolated from mated females may contain Y chromosome rDNA. Thus, it is critical that unmated females are used for XX sequencing analysis. Importantly, this method requires at least two chromosome-specific SNPs to enable at least two locus-specific probes to bind each rRNA molecule for detection, limiting its application to rDNA loci with at least two distinguishing SNPs between them. The availability of more SNPs allows for more probes to bind each rRNA and can provide greater signal effectiveness. Interestingly, this method only labels rRNA in the nucleolus, despite two probe pairs targeting SNPs in the 18S and 28S rRNAs which are exported to the cytoplasm (rRNA containing the two ITS1 target sites only exists in the nucleolus). The lack of cytoplasmic FISH signal suggests two non-exclusive possibilities: 1) the 18S and 28S probes alone are insufficient to detect cytoplasmic rRNA, perhaps because rRNA is more dispersed throughout the cytoplasm than in the nucleolus, or 2) there is lower probe binding efficiency to rRNAs integrated into the mature ribosomal subunits than to the unprocessed 45S rRNA. Additional SNPs within the 18S and 28S rRNAs may enable SNP-sensitive FISH detection of cytoplasmic rRNAs.

Other methods to assess locus-specific rRNA transcription include rRNA sequencing approaches that differentiate the expression of rRNA variants that are assigned to specific loci⁶. Similarly, qPCR can differentially assess the expression of rRNA variants that are distinct enough to enable unique detection^5,21, though it is unclear if the silencing of these variants represents the silencing of an entire rDNA locus. While these methods can be effective in quantifying the expression of specific rRNA variants, they cannot be done with single-cell resolution. The heterogeneity of X chromosome rDNA silencing in Drosophila tissues suggests that there is strong cell-to-cell variation in rDNA locus transcription¹⁷ and highlights the need for techniques that can account for this variability. Imaging features associated with active transcription at rDNA loci, such as the H3.3 histone variant¹⁰ or the Pol I transcription factor UBF²², have been used to identify locus-specific rRNA transcription with single-cell resolution. However, both of these methods require the condensed chromosomes of mitotic cells in order to distinguish transcription occurring at any particular rDNA locus. In Drosophila cells, rDNA loci at the X and Y chromosomes may be identified by chromosome shape¹⁰, but identification of transcriptionally active rDNA loci in human cells also requires co-staining with chromosome-specific labels²². The SNP-FISH method does not require either chromosome-specific co-labeling or labeling of transcriptional markers and can be assessed in any cell cycle stage or post-mitotic cells, providing flexibility for use in diverse tissues and experimental conditions.

This rRNA SNP-FISH method may be modified for use with other SNPs that distinguish between Drosophila rRNAs or potentially be applied to SNPs that distinguish between rDNA loci in other organisms. Applying this method to organisms with more than two rDNA loci will require a locus to have a unique rDNA variant haplotype containing at least two SNPs that are not present on any other rDNA locus and that are shared between all 45S copies at that locus. This requirement means the method will only be able to specify transcription from one specific locus compared to all others (i.e., rRNA SNPs from locus 1 compared to SNPs shared by loci 2 and 3). However, if each rDNA locus has a compatible haplotype, multiple experiments could be combined to individually assess the transcription of each locus one at a time. The relatively low stringency of the hybridization temperature (37 °C) used in this protocol allows for strong probe binding, particularly given the short, unmasked portion of the probe, though hybridization at higher temperatures (50-75 °C) has been shown to enhance the signal of some FISH probes²³. Higher hybridization temperatures may enhance mask strand displacement and increase probe binding, but too high temperatures may destabilize probe-mask binding and lose SNP-specificity. For this reason, we do not expect SNP-FISH to specifically label DNA because the temperatures needed to melt double-stranded DNA targets to enable probe binding would also destabilize probe-mask binding and eliminate SNP-sensitive target specificity. Still, optimization of hybridization temperature for new rRNA SNP probes may increase the signal, particularly when only two SNP sites are available. Loss of X chromosome rDNA silencing is associated with rDNA copy number reduction in Drosophila¹³, so the development of SNP-FISH assays to characterize locus-specific rRNA transcription in other systems may serve as a useful tool to evaluate the integrity of rDNA and ribosome function. Furthermore, this method has been modified for use with a deletion variant in Drosophila, and this single structural rRNA variant was suitable for detection (perhaps due to greater target binding affinity without a probe mask)¹⁷. The use of structural rRNA variants potentially combined with SNP variants broadens the potential for application in other systems. Since the mechanisms that regulate rDNA locus transcription remain largely unclear, the flexibility and potential adaptability of rRNA SNP-FISH to other systems make it a powerful tool for future studies to investigate rRNA transcription.

Materials

Name	Company	Catalog Number	Comments
PTC Tempo 96 Thermal Cycler	Bio Rad	12015382	Any thermal cycler can be used
0.2 mL 8 strip flatcap PCR tubes	VWR	89133-912	Any compatable tubes can be used
1 kb DNA ladder	NEB	N3232S	To use when checking sequencing PCR amplicon size by gel electrophoresis
1.5 mL graduated microcentrifuge tubes	USA Scientific	1615-5510	Any certified RNase free tube will do
2 mM dNTP Mix	ThermoFisher Scientific	R0241
50x TAE Buffer	Bio Rad	1610743	Used for gel electrophoresis. Any TAE buffer can be used.
5M NaCl			Any NaCl can be used or prepared from any source
Agarose	VWR	97064-250	Any Agarose can be used
ApE - A plasmid Editor software	N/A	N/A	https://jorgensen.biology.utah.edu/wayned/ape/
Clear nail polish			Any nail polish from any retailer can be used
Cover glass, No. 1 Thickness	Thomas Scientific	6672A38
Deionized Formamide	Fisher Scientific	NC9569627
Dextran Sulfate Sodium	Sigma Aldrich	D8906-10G
DreamTaq Green PCR Master Mix	ThermoFisher Scientific	K1081
Dumont #5 forceps	Fine Science Tools	11252-20	Used for dissecting samples
EDTA (0.5M), Ph 8.0	ThermoFisher Scientific	R1021	Any comparable product can be used
Ethanol	VWR	89125-172	Any 200 proof Ethanol can be used. Used for dilution to 70% ethanol for permeabilization and cleaning for Rnase-free conditions
Ethidium Bromide	ThermoFisher Scientific	15585011
Horizontal Mini Gel Electrophoresis System	Fisher Scientific	14-955-170	Any gel electrophoresis system can be used
Kimwipes	Fisher Scientific	06-666
Micro-Test Staining Dish	Electron Microscopy Sciences	71564	Used for dissecting samples
Nutating shaker	Sigma Aldrich	BMSB3D1020	Any nutating shaker can be used
Parafilm	USA Scientific	3023-4526
Phosphate-Buffered Saline (10X) pH 7.4, RNase-free	Life Technologies	AM9624
Pierce 16% Formaldehyde (w/v), Methanol-free	Life Technologies	28908
Precision General Purpose Water Bath	Life Technologies	TSGP02	Any water bath can be used
QIAquick Gel Extraction Kit	Qiagen	28704	Any gel extraction kit can be used
Recombinant Proteinase K Solution (20 mg/mL)	Invitrogen	AM2546	Any comparable product can be used
Rnase-free UltraPure BSA	ThermoFisher Scientific	AM2618
S. cerevisiae tRNA	Sigma Aldrich	R8759
SSC (20X), RNase-free	Fisher Scientific	AM9763
Triton-X 100	Life Technologies	A16046.AE
UltaPure 1M Tris-HCl Buffer, pH 7.5	ThermoFisher Scientific	15567027	Any comparable product can be used
UltraPure DNase/RNase-Free Distilled Water	Life Technologies	10977023
Vanadyl ribonucleoside complex	NEB	S1402S
VECTASHIELD mounting media with DAPI	Vector Laboratories	H-1200-10
VWR Superfrost Microscope Slide	VWR	48311-601
y[1]w[1] Drosophila melanogaster line	Bloomington Drosophila Stock Center	1495
Zeiss LSM 980 confocal microscope	Zeiss microscopy		Any confocal microscope with compatable emission and detection can be used
Zeiss Stemi 2000-C Stereo Microscope and light source	Microscope Central	455053	Any steromicroscope can be used