Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Oct 30;25(5):1225-1240.e6.
doi: 10.1016/j.celrep.2018.10.012.

DAZL Regulates Germ Cell Survival Through a Network of PolyA-Proximal mRNA Interactions

Free PMC article

DAZL Regulates Germ Cell Survival Through a Network of PolyA-Proximal mRNA Interactions

Leah L Zagore et al. Cell Rep. .
Free PMC article


The RNA binding protein DAZL is essential for gametogenesis, but its direct in vivo functions, RNA targets, and the molecular basis for germ cell loss in Dazl-null mice are unknown. Here, we mapped transcriptome-wide DAZL-RNA interactions in vivo, revealing DAZL binding to thousands of mRNAs via polyA-proximal 3' UTR interactions. In parallel, fluorescence-activated cell sorting and RNA-seq identified mRNAs sensitive to DAZL deletion in male germ cells. Despite binding a broad set of mRNAs, integrative analyses indicate that DAZL post-transcriptionally controls only a subset of its mRNA targets, namely those corresponding to a network of genes that are critical for germ cell proliferation and survival. In addition, we provide evidence that polyA sequences have key roles in specifying DAZL-RNA interactions across the transcriptome. Our results reveal a mechanism for DAZL-RNA binding and illustrate that DAZL functions as a master regulator of a post-transcriptional mRNA program essential for germ cell survival.

Keywords: 3′ UTR; DAZL; RNA binding protein; RNA networks; germ cell survival; post-transcriptional regulation; spermatogenesis.

Conflict of interest statement


The authors declare no competing interests.


Figure 1.
Figure 1.. HITS-CLIP Identification of DAZL-RNA Contacts in the Adult Testis
(A) Autoradiograph of radiolabeled cross-linked DAZL-RNA complexes purified from adult testes. Arrowhead indicates DAZL cross-linked to minimal RNA fragments. Bracket indicates RNA fragments of 35–50 nt excised for cDNA preparation. (B) Venn diagram of overlapping CLIP reads from each replicate and identification of 11,297 genomic coordinates with overlapping CLIP reads from 3/3 biological replicates. (C) Bins of CLIP peaks normalized to RNA-seq, with GUU enrichment (green) and proportion of peaks in each bin with GUU (gray), and P values from χ2 distribution of GUU enrichment at right. (D) Top motifs present in bin 1, identified by MEME (top, middle) and DREME (bottom). (E) Top motifs associated with CIMS sites identified by MEME. (F) Top 10 6mers and 4mers associated with CIMS sites. (G) Positioning of UGUUGU, GUUG, and UGUU motifs relative to CIMS deletion sites.
Figure 2.
Figure 2.. DAZL Predominantly Binds 3′ UTRs at PolyA-Proximal Sites
(A) Distribution of BR3 CLIP sites in different genic regions. BR3 clusters mapped to regions with overlapping coding and non-coding RNAs were annotated as nc/mRNA. (B) Number of genes with ≥1 polyA site, as determined by polyA-seq analysis. (C) Metagene analysis of DAZL CLIP peaks in 20 nt bins relative to the stop codon (left, red line) or the polyA site (right, blue line). Dashed line represents the expected random distribution. (D) Top corresponds to a higher magnification view of the right part of (C). The bottom three parts show CLIP peak distribution relative to the polyA site in 3′ UTRs with 1, 2, or more peaks in the entire 3′ UTR. (E) Nucleotide frequency 500 nt upstream of polyA sites of control 3′ UTRs (no DAZL CLIP reads, top), and 3′ UTRs with DAZL-RNA interactions (bottom). (F) Motif enrichment in 3′ UTRs with and without DAZL CLIP peaks (y and x axes, respectively), where values correspond to Z scores. (G) GUU enrichment (observed/expected) in 3′ UTRs with and without Dazl CLIP peaks (green and gray lines, respectively).
Figure 3.
Figure 3.. Genetic Labeling, Isolation, and Transcriptome Profiling of GFP+ WT and Dazl KO Germ Cells
(A) IF of DDX4 and GFP in Stra8-iCre+; IRG+; Dazl2+ (Dazl WT) and Stra8-iCre+; IRG+; DazlTm1hgu/Tm1hgu (Dazl KO) testes at P6 (left and right, respectively). Sections from biological replicates are shown in the top and bottom rows. Scale bars for bottom WT panels represent 15 μM, all others represent 25 μM. (B) Flow cytometry plots of GFP and GFP+ cells from WT and KO mice. (C) Percentage of GFP+ cells isolated by FACS from P6 WT and Dazl KO mice. (D) Scatterplot of RPKM values for genes in WT and KO cells. Green and red dots denote genes with decreased and increased RNA levels in Dazl KO cells compared to WT controls. (E) Boxplot showing the distribution of RPKM values for green, red, and gray genes indicated in (D).
Figure 4.
Figure 4.. DAZL Directly Enhances mRNA Levels through PolyA-Proximal Binding
(A) Top 10 motifs (based on Z score ranking) for sequences of BR3 regions (left) or 10 nt windows centered at the 5′ end of each BR3 CLIP region. (B) Distribution of 6,465 BR3 DAZL-RNA sites from iCLIP analysis of P6 testes. (C) Venn diagram comparing genes with RNA level changes between WT and Dazl KO cells (≥2-fold, p < 0.01) and the set of 2,290 with DAZL-3′ UTR interactions. (D) Examination of GUUs in the last 250 nt of 3′ UTRs with and without BR3 sites identified by iCLIP analysis of P6 testes. (E) RPKM levels of 2,290 genes with DAZL-3′ UTR interactions in GFP+ WT and GFP+ KO cells. Green and red circles correspond to genes with RNA level differences between GFP+ WT and KO cells of ≥2-fold (adjusted p < 0.01). Blue dots correspond to genes with RPKM values that do not differ >20% between WT and KO cells. (F) Metagene analysis of the distribution of BR3 DAZL-RNA sites in 3′ UTRs of DAZL-enhanced, DAZL-repressed, and DAZL-insensitive genes (green, red, and blue, respectively). (G) Examination of 3′ UTR lengths (left) and frequency of GUU per 3′ UTR (right) for genes with increased RNA, unchanged RNA, or decreased RNA levels (green, blue, and red, respectively) in GFP+ Dazl KO cells compared to WT controls. P values (Wilcoxon rank sum test) for pairwise comparisons indicated above.
Figure 5.
Figure 5.. Rare DAZL-RNA Interactions in 5′ UTRs Are Associated with Enrichment of PolyA Tracts
(A) Comparison of BR3 CLIP tag density in the 5′ and 3′ UTRs of genes with 5′ UTR BR3 CLIP sites. Color coding represents the percentage of BR3 tags in the gene that map to either the 5′ UTR or the 3′ UTR. Dashed line denotes cutoff between genes with <50% (above) or >50% (below) of the CLIP reads in the 5′ UTR. (B–F) Representative examples of 5′ low (B–D) and 5′ high (E and F) genes with CLIP read density in BR3 regions and RNA-seq read density shown.(G) Motif enrichment in 5′ UTRs with >50% or <50% of total BR3 CLIP reads in the 5′ UTR.
Figure 6.
Figure 6.. The PolyA Tail Has an Important Role in DAZL-RNA Binding and Regulation
(A) Western blot showing dox-induced expression of DAZL in GC-1 spg cells used for RNA-seq analyses. (B) Distribution of Z scores following motif-enrichment analysis of complete regions of BR3 positions bearing overlapping CLIP reads (y axis) or 10 nt regions centered at the 5′ end of BR3 sites (x axis). (C) Distribution of BR3 CLIP sites across the genome. (D) Left: fold change in mRNA ribosome association (y axis) and total mRNA level (x axis) for 894 genes following dox induction of DAZL expression. Right: bar chart of 223 genes with ≥1.25-fold change in mRNA association with ribosomes, change in total mRNA level, or both, following DAZL expression. (E) Distribution of ribosome-protected to total mRNA (RP/T) ratios for group 1 and group 2 genes (green and red, respectively) before and after dox treatment. (F) Top 10 most-enriched 6mers in last 250 nt of groups 1 and 2 3′ UTRs. (G) Extended polyU tracts in group 1 3′ UTRs revealed by examining the distribution of the top 10 motifs (F) in each group 1 3′ UTR and merging any that overlap. (H) Luciferase activity in untreated and dox-treated cells transfected with luciferase mRNA reporters bearing polyadenylated and non-polyadenylated 3′ UTRs of the indicated genes. D’Rik corresponds to reporters bearing the 3′ UTR of D030056L22Rik. (I) Immunoprecipitation (IP) efficiency of endogenous D030056L22Rik mRNA and luciferase mRNA in dox-treated cells transfected with polyadenylated and non-polyadenylated versions of 3 different 3′ UTRs. Values in (H) and (I) are averages and SDs from 3 to 5 replicate experiments. *p < 0.05, **p < 0.01, ***p < 0.0001; n.s. denotes no statistically significant difference.
Figure 7.
Figure 7.. DAZL Binds and Regulates a Broad Network of Essential Genes Encoding Interacting Proteins
(A) Hierarchical clustering of enriched GO terms and genes associated with the set of 501 genes that are enhanced by DAZL and have DAZL-3′ UTR interactions. Subgroups of overlapping GO categories are indicated at left, while subgroups of genes are indicated with letters at bottom. (B) List of all of the genes in each letter group indicated at the bottom of (A). (C) Protein-protein interaction network for genes with DAZL-3′ UTR interactions and reduced RNA levels in Dazl KO cells compared to WT controls.

Similar articles

See all similar articles

Cited by 7 articles

See all "Cited by" articles


    1. Afgan E, Baker D, Batut B, van den Beek M, Bouvier D, Cech M, Chilton J, Clements D, Coraor N, Grüning BA, et al. (2018). The Galaxy platform for accessible, reproducible and collaborative biomedical analyses: 2018 update. Nucleic Acids Res. 46, W537–W544. - PMC - PubMed
    1. Bailey TL, Johnson J, Grant CE, and Noble WS (2015). The MEME suite. Nucleic Acids Res. 43, W39–W49. - PMC - PubMed
    1. Bicknell AA, and Ricci EP (2017). When mRNA translation meets decay. Biochem. Soc. Trans 45, 339–351. - PubMed
    1. Blume-Jensen P, Jiang G, Hyman R, Lee KF, O’Gorman S, and Hunter T (2000). Kit/stem cell factor receptor-induced activation of phosphatidylinositol 3′-kinase is essential for male fertility. Nat. Genet 24, 157–162. - PubMed
    1. Bose R, Sheng K, Moawad AR, Manku G, O’Flaherty C, Taketo T, Culty M, Fok KL, and Wing SS (2017). Ubiquitin Ligase Huwe1 Modulates Spermatogenesis by Regulating Spermatogonial Differentiation and Entry into Meiosis. Sci. Rep 7, 17759. - PMC - PubMed

Publication types

MeSH terms

LinkOut - more resources