Visualization and quantification of rDNA instabilities in mammalian cells and mouse models

Abstract

Ribosomal DNA (rDNA) encodes the 18S, 5.8S, and 28S rRNA, accounting for ∼70% of cellular transcription. Despite its essential role and links to cancer and aging, quantifying rDNA instability in mammals remains challenging due to its repetitive organization and inherent heterogeneity. Here, we developed a murine rDNA FISH probe and genomic tools tailored for laboratory mouse strains. The results confirmed rDNA cluster locations, revealed substantial inter- and intra-strain as well as intercellular heterogeneity in rDNA organization within inbred mice and unstressed cells, and identified sources of spontaneous and replication-associated DNA double-strand breaks in the rDNA transcription termination region. Using mouse embryonic stem cells, we showed that BRCA1-mediated homologous recombination promotes rDNA instability, the non-homologous end joining factor XRCC1, but not Ku, suppresses intra-cluster deletions, and ATM kinase preserves rDNA cluster stability. Together, these findings establish a platform and tools for studying rDNA instability in animal models relevant to aging and cancer research.

Graphical Abstract

Figure 1.

Detecting murine rDNA with an IGS-ETS FISH probe. (A) Schematic of a single murine rDNA repeat. The 10.7 kb IGS-ETS FISH probe was generated from EcoRI and MluI digestion and indicated in the diagram (see also Supplementary Fig. S1A). Representative rDNA FISH image of a metaphase C57BL/6 (B) and 129/Sv (C) ES cell. Quantification of the percentage of metaphases with given rDNA clusters in n = 55 C57BL/6 (D) and n = 81 129/Sv (E) ES cells. (F) Model (left) indicates the location of murine rDNA relative to telomere, minor satellite, and major satellite. Right, representative four-color FISH images of four independent chromosomes stained with DAPI, rDNA, minor satellite, and major satellite. n = 10 metaphases were analyzed for each strain, with the same rDNA location observed. Scale bar = 10 µm.

Figure 2.

Distribution of rDNA arrays in 129/Sv strain ES cells. (A, B) A representative set of sequential rDNA FISH and SKY analyses of 129/Sv ES cell metaphases. A-left: Karyotyping; A-right: rDNA FISH; B-left: corresponding SKY paint; and B-right: chromosome grid of the SKY with rDNA-containing chromosome boxed. N = 24 metaphases were analyzed with a consistent pattern. (C) Representative rDNA FISH and chromosome paint analyses with custom paint to Chr12, 15, 16, 18, and 19. The number of metaphase cells analyzed is shown at the bottom. (D) Representative rDNA FISH with the large clusters (red circles) and small clusters (yellow circles) marked. The estimated percentage of total rDNA signal on each chromosome was determined via co-FISH and chromosome paint. The bars indicate the mean and the standard deviation (SD) of n = 20 metaphases. Bottom: summary of the percentage of rDNA signal on Chr12, 16, 18, and 19, and the corresponding number of repeats and the cluster length, assuming ∼270 copies of the 45 kb rDNA gene are present in each cell. Sorted by rDNA cluster size from big to small. The numbers indicate the mean ± SD. Scale bar = 10 µm.

Figure 3.

Distribution of rDNA clusters in the C57BL/6 ES cells. (A, B) A representative set of sequential rDNA FISH and SKY analyses of C57BL/6 ES cell metaphases. A-left: Karyotyping; A-right: rDNA FISH; B-left: corresponding SKY paint; and B-right: chromosome grid of the SKY with rDNA-containing chromosome boxed. N = 8 metaphases were analyzed with a consistent pattern. (C) Representative rDNA FISH and chromosome paint analyses with custom paint to Chr12, 15, 16, 18, and 19. The number of metaphase cells analyzed is shown at the bottom. (D) Representative rDNA FISH with the large clusters (red circles) and small clusters (yellow circles) marked. The estimated percentage of total rDNA signal on each chromosome was determined via co-FISH and chromosome paint. The bars indicate the mean and the standard deviation (SD) of n = 20 metaphases. Bottom: summary of the percentage of rDNA signal on Chr12, 15, 16, 18, and 19 and the corresponding number of repeats and the cluster length, assuming ∼234 copies of 45 kb rDNA gene are present in each cell. Sorted by rDNA cluster size from big to small. The numbers indicate the mean ± SD. Scale bar = 10 µm.

Figure 4.

Heterogeneity in rDNA cluster size on Chr15 and Chr18 among inbred C57BL/6 mice. (A) Representative images (top) and quantification (bottom) of rDNA FISH on chromosome 15, co-stained with Chr15 paint and DAPI. Examples of metaphases exhibit either (from left to right) two small rDNA clusters (Chr15 SS), one small and one large (Chr15 SL), or two large rDNA clusters (Chr15 LL) on Chr15. At least 80 metaphases were analyzed per mouse and showed a consistent pattern (80/80). (B) Representative image (left) and quantification (right) of the rDNA FISH on chromosome 18 and DAPI. At least 80 metaphases were analyzed per mouse and showed a consistent pattern. Insets in both panels (A) and (B) display rDNA FISH signals without chromosome paint to better illustrate rDNA signal intensity. (C) Analyses of rDNA cluster heterogeneity on Chr15 in activated splenic B cells (88 metaphases) and T cells (86 metaphases) from the same mouse. The number of metaphases with LL and SL rDNA FISH configurations was plotted by cell type. (D) Distribution of Chr15 and Chr18 heterogeneity among 13 inbred C57BL/6 mice obtained from Taconic Farms (2024–2025). (A, B) Data represent mean ± SD of 10 metaphases per mouse. Scale bar = 10 µm. (E) Distribution of rDNA cluster numbers observed in 3 out of the 13 WT C57BL/6 mice with Chr15 SS (BL6-1), SL (BL6-5), and LL (BL6-10) rDNA patterns. 30 metaphases were analyzed for each mouse. The Kolmogorov–Smirnov test was used to compare two distributions. n.s. = not significant.

Figure 5.

Single-ended replication-associated DNA DSBs in rDNA. (A) Structure of 1× murine rDNA repeat. The directions of 45S rRNA transcription and rDNA replication are indicated by orange and purple arrows, respectively. The relative positions of TTF1 binding sites (white boxes, T1–T10) and regions with spontaneous DSBs detected by END-seq (red boxes, R1–R4) are shown at the bottom. (B) Normalized read density (RPM) from END-seq in quiescent (G1) and activated (S) murine B cells. (C) END-seq reveals asymmetric DSBs in activated B cells. Top strand reads represent the right ends of DSBs (red bars), and bottom strand reads represent DSB left ends (blue bars). Four right-end peaks are observed downstream of the rDNA replication termination site (shaded red) and GC stretch (shaded blue). Corresponding genomic DNA sequences with annotations are displayed above.

Figure 6.

Asymmetric RPA association at rDNA DSBs. (A) Overview of strand-specific RPA ChIP-seq and END-seq at TTF1 binding sites (T1–T10), with a magnified view highlighting strand-specific RPA ChIP signals. The relative genomic position within the rDNA repeat is indicated at the top of each panel. (B) Proposed working model for the generation of single-ended DNA DSBs on the lagging strand during rDNA replication. Blue lines represent the rDNA template, and red lines depict newly synthesized rRNA. The replication fork proceeds from right to left, with the leading strand on top and the lagging strand at the bottom. Replication of the leading strand terminates near the GC-rich Sal box, as previously mapped. A single-ended DSB forms when lagging-strand priming is delayed, resulting in extended regions of RPA-coated single-stranded DNA. The putative break site is indicated by a scissor.

Figure 7.

The impact of DNA repair and response deficiency on rDNA stabilities measured by FISH. (A) Frequency of different rDNA cluster numbers observed in WT TC1 mouse ES cells and cells with mutant DNA repair pathway proteins. At least 50 metaphases were analyzed for each genotype. The Kolmogorov–Smirnov test was used to compare two distributions, and only statistically significant differences were marked. (B) Total rDNA fluorescent intensity in arbitrary units (a.u.) measured by FISH in WT/TC1/129/Sv and indicated mutant mouse ES cells (all with 129/Sv background). At least 30 metaphases were measured. (C) PFGE analysis of PmeI-digested Atm^+/+ (TC1 129/Sv) and Atm^−/− ES cells DNA probed with an rDNA probe (see Supplementary Fig. S1E). M: H. Wingei size marker. (D) rDNA FISH signal (a.u.) of metaphases (n = 30) from Atm^+/+ and Atm^−/− v-abl kinase-transformed murine B cells. Color indicates three independent repeats. (E) rDNA FISH signal (a.u.) of interphases (n = 30) derived from Atm^+/+ and Atm^−/− v-abl kinase-transformed murine B cells. (F) rDNA FISH signal (a.u.) of WT (TC1) or Atm^−/− ES cells, and WT (TC1) ES cells treated with 15 µM ATM inhibitor KU55933 for 72 h. (G) The ddPCR estimated rDNA copy number per haploid genome in WT129/Sv, Atm^−/−, Xrcc4^−/−, and C57BL/6 mouse ES cells. Mean copy numbers were calculated from 3–4 replicates. For panels (B, D–F), the bars represent mean ± SEM. For panel (G), the bars represent mean ± SD. P-value was calculated via the unpaired Student’s t-test between mutants and WT. n.s. P > .05; *: P < .05; ***: P < .001; and ****: P < .0001. Scale bar = 10 µm.