Insights and modulation of RNA polymerase–dependent R-loop and dsRNA in Fanconi anemia hematopoietic stem cells
Michihiro Hashimoto, Xiaomin Feng, Jie Bai, Huimin Zeng, Tian Li, Jue Li, Terumasa Umemoto, Paul R. Andreassen, Gang Huang
Michihiro Hashimoto, Xiaomin Feng, Jie Bai, Huimin Zeng, Tian Li, Jue Li, Terumasa Umemoto, Paul R. Andreassen, Gang Huang
View: Text | PDF
Research Article Cell biology Hematology

Insights and modulation of RNA polymerase–dependent R-loop and dsRNA in Fanconi anemia hematopoietic stem cells

Abstract

Fanconi anemia (FA) is the most common BM failure (BMF) syndrome. FA genes have a role in suppressing DNA-RNA hybrids, termed R-loops, which can be generated via transcription mediated by RNA polymerase (RNAP). How these processes, including a role in fate determination of hematopoietic stem cells (HSCs), are related to BMF is largely unknown. Single FA gene KO in mice does not recapitulate most phenotypes observed in patients with FA. Thus, we generated a mouse model for FA by introducing heterozygous Setd2, which restricts RNAP-dependent transcription. We showed that FA patient–derived cells and Setd2+/– Fanca–/– HSCs share increased R-loop and dsRNA levels and a ribosomal biogenesis defect. Further, Setd2+/– Fanca–/– HSCs displayed cell cycle arrest, mitotic errors, and BMF phenotypes. Importantly, utilizing our Setd2+/– Fanca–/– mice, we discovered that Juglone, a pan-RNAP inhibitor, reduces R-loop and dsRNA and reverses ribosomal biogenesis defects and mitotic errors, thereby rescuing BMF. This study establishes a mouse model that underscores a key role for R-loop formation, ribosomal biogenesis defects, and mitotic errors in HSCs in driving BMF in FA. We also introduce a potential therapeutic avenue based upon pan-inhibition of RNAPs utilizing Juglone.

Authors

Michihiro Hashimoto, Xiaomin Feng, Jie Bai, Huimin Zeng, Tian Li, Jue Li, Terumasa Umemoto, Paul R. Andreassen, Gang Huang

×

Figure 1

Human FANCA-deficient cells from a patient with FA display accumulation of R-loops and dsRNA and errors in ribosomal biogenesis.

Options: View larger image (or click on image) Download as PowerPoint
Human FANCA-deficient cells from a patient with FA display accumulation ...
(A) Visualization of R-loops detected by S9.6 mAb (red) and dsRNA detected by rJ2 mAb (green) and via immunofluorescence microscopy in FANCA-corrected or -deficient cells from a patient with FA. Nuclei shown by counterstaining with DAPI (blue). Original magnification is ×63. (B and C) Quantification of R-loops (B) and dsRNA (C) based upon the area integrated intensity measured by ImageJ (NIH). One-tailed Student’s t test; n = 37 for each condition. (D) Quantification of the expression of ribosomal genes using qRT-PCR. One-tailed Student’s t test; n = 3 for each condition. (E) DNA dot blots detecting R-loop levels using S9.6 antibody in FANCA-deficient cells reconstituted with 2 RNases (RNASEH1 or DICER1) or vector alone (Vec). (F) Quantification of R-loop levels shown in E. One-way ANOVA with Tukey’s multiple-comparison test; n = 3 for each condition. (G) RNA dot blots detecting levels of dsRNA in FANCA-deficient cells reconstituted with 2 RNases (RNASEH1 or DICER1) or Vec. n = 3 for each condition. (H) Quantification of dsRNA levels shown in G. One-way ANOVA with Tukey’s multiple-comparison test; n = 3 for each condition. (I) Quantification of specific rRNA forms in FANCA-deficient cells containing empty vector with or without overexpression of RNASE H1 or DICER1. One-way ANOVA with Tukey’s multiple-comparison test; n = 3 for each condition. (J) Visual summary. (K and L) For patient-derived cells, we utilized FANCA-deficient GM6935 cells (84). (K) Relative expression levels of SETD2, ALDH2, ADH5, and TP53 in healthy donors (n = 11) and FA patients (n = 21) were analyzed using normalized microarray data (GSE16334). Expression values presented as log2-transformed intensities. One-tailed Student’s t test. (L) Reactome pathway enrichment analysis comparing HD and FA gene clusters performed using compareCluster function (fun = “enrichPathway”, pvalueCutoff = 0.05) from clusterProfiler and ReactomePA. Top 10 enriched pathways are shown in dot plot: color indicates adjusted P value; dot size represents gene ratio. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.