UAP56 couples piRNA clusters to the perinuclear transposon silencing machinery - PubMed (original) (raw)

. 2012 Nov 9;151(4):871-884.

doi: 10.1016/j.cell.2012.09.040.

Jie Wang 2, Jia Xu 2, Zhao Zhang 1, Birgit S Koppetsch 1, Nadine Schultz 1, Thom Vreven 2, Carine Meignin 3, Ilan Davis 4, Phillip D Zamore 5, Zhiping Weng 6, William E Theurkauf 7

Affiliations

UAP56 couples piRNA clusters to the perinuclear transposon silencing machinery

Fan Zhang et al. Cell. 2012.

Abstract

piRNAs silence transposons during germline development. In Drosophila, transcripts from heterochromatic clusters are processed into primary piRNAs in the perinuclear nuage. The nuclear DEAD box protein UAP56 has been previously implicated in mRNA splicing and export, whereas the DEAD box protein Vasa has an established role in piRNA production and localizes to nuage with the piRNA binding PIWI proteins Ago3 and Aub. We show that UAP56 colocalizes with the cluster-associated HP1 variant Rhino, that nuage granules containing Vasa localize directly across the nuclear envelope from cluster foci containing UAP56 and Rhino, and that cluster transcripts immunoprecipitate with both Vasa and UAP56. Significantly, a charge-substitution mutation that alters a conserved surface residue in UAP56 disrupts colocalization with Rhino, germline piRNA production, transposon silencing, and perinuclear localization of Vasa. We therefore propose that UAP56 and Vasa function in a piRNA-processing compartment that spans the nuclear envelope.

Copyright © 2012 Elsevier Inc. All rights reserved.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Organization of the piRNA biogenesis machinery

A, B. UAP56, Rhi and Vasa localization in a stage 8 nurse cell. Single channel images are shown on left and a 3-color merged image is show on the right (color assignments as indicated). The line in the merged image indicates the position of a line scan for fluorescence intensity (B). The nuclear foci contain overlapping peaks of UAP56 and Rhi, and Vasa-GFP signal accumulates across the nuclear envelope from foci that are closely associated with the periphery. Scale bars = 2 µm C. Scatter plot comparing Rhi signal in foci at the nuclear periphery with Vasa signal in adjacent nuage foci. The diagonal indicates identical signal levels. R is the Pearson correlation coefficient, and the p value is based on Pearson’s product moment correlation coefficient and follows a t distribution. 135 pairs of foci from 25 nuclei were quantified. D. In the oocyte, Rhi localizes to nuclear foci but UAP56 does not co-localize with these foci. Rhi and UAP56 are shown separately on the left, and a merged image with Rhi in green and UAP56 in red is on the right. Scale bars = 5 µm. Also see Figure S1.

Figure 2

Figure 2. The E245K substitution and rhi mutations disrupt UAP56 localization to nuclear foci

A. UAP56 (red), Rhi (green), and nuclear pores (cyan) in wild type, uap56 and rhi mutant ovaries. UAP56 and Rhi are shown with nuclear pores (NP) in the first two images in each row. A merged image of Rhi, UAP56 and nuclear pores is show in the last panel of each row. In stage 4 uap56sz15′/uap5628 egg chambers, Rhi localizes to foci but UAP56 is dispersed in the nucleoplasm. In rhi mutations, Rhi protein is not detected and UAP56 fails to localize to foci. Scale bars are 2µm. B. Localization of transgenic wild type UAP56-Venus and sz-Venus fusion proteins. Egg chambers expressing the transgenes were Immunolabeled for of Rhi (left panel, green in right panel). Fusion protein distribution is show in the middle panel and in red in the right panel. sz-Venus, which carries the E245K substitution found in the uap56sz15′ allele, fails to co-localize with Rhi foci. Scale bars are 2µm. Also see Figure S2.

Figure 3

Figure 3. UAP56 is required for PIWI protein localization to nuage

A. Aub, Ago3 and Piwi localization in Drosophila nurse cells. Aub, Ago3, and Piwi are in green and DNA is in blue. Aub and Ago3 localized at perinuclear nuage in wild type nurse cells, while Piwi is in nuclei of both nurse cells and follicle cells. In uap56sz15′/uap5628 mutants, Aub and Ago3 localization to nuage is disrupted, but Piwi localizes to the nucleus. B. Aub and Ago3 nuage localization is rescued by the wild type UAP56-Venus transgene, but not the not by the mutant sz-Venus transgene. Gray scale image show Aub and Ago3 and merged images show Aub and Ago3 in green and DNA in blue. Scale bars are 2µm. Also see Figure S3.

Figure 4

Figure 4. uap56, rhi and vasa mutations disrupt transposon silencing but do not alter gene expression

Whole genome tiling arrays were used to assay gene and transposon expression in wild type, uap56sz15′ / uap5628, vasD5/vasPH, and rhi02086/ rhiKG00910 ovaries. A. A Genome Brower screen shot showing a region on chromosome 2L containing the nature transposon, Blood and Ent1 gene. In uap56, vas and rhi mutants, Blood expression is increased dramatically, while the Ent1 levels are comparable to control. B. Scatter plots comparing the expression of transposons in mutants relative to wild type. The diagonal indicates identical expression levels. All 3 mutants show significant (FDR<0.05) over-expression of a subset of transposon families. Over-expression of selected transposon families was confirmed by northern blotting and qPCR (Figure S4B and C). C. Scatter plots comparing the expression of protein-coding genes mutant and wild type ovaries. None of the mutations lead to a significant change (FDR<0.05) in protein coding genes expression. The uap56, rhi and vas transcripts are highlighted, as indicated by the legend. Also see Figure S4.

Figure 5

Figure 5. piRNA expression in uap56, rhi and vasa mutant ovaries

A. Scatter plots comparing the abundance of piRNAs mapping to germline enriched group 1 transposons (black), soma enriched group 3 transposons (red) and class 2 transposons, which show a sense strand bias (green). All three mutations reduced piRNA from germline specific Group 1 transposon families. B. Scatter plots comparing the abundance of cluster piRNAs in 3 mutants and wild type ovaries. Each point represents piRNAs from a single cluster. All three mutations significantly reduced piRNAs to the germline clusters, including the major dual strand cluster at 42AB (red). By contrast, uap56 and rhi mutants do not reduced piRNA to unistrand cluster 2 (green) or flam (blue). Mutations in vas reduce piRNAs linked to cluster 2, but not flam. C. Pair-wise comparison of cluster piRNAs in mutant ovaries. Cluster expression is highly correlated in all three mutants, with rhi and uap56 showing almost identical patterns of cluster piRNA expression (R=0.96). D. Genic piRNA expression in mutant ovaries. Scatter plots compare the abundance of piRNAs linked to protein coding genes in mutant ovaries relative to wild type controls. Both uap56 and rhi mutants lead to a significant increase in ectopic piRNAs from protein coding genes. R-values are shown at upper left corner. RNA sequencing data for vas and rhi are from Malone et al. (2009) and Klattenhff et al. (2009). Also see Figure S5.

Figure 6

Figure 6. piRNA expression and precursor binding to UAP56

A. piRNAs and UAP56-associated RNAs mapping to the 1/42AB on chromosome 2R. The locus has many embedded transposable elements and is flanked by protein coding genes, Pld and Jing. In wild type, piRNAs map to both the plus and minus strands of the 42AB cluster (pink tracks), and are dramatically reduced in uap56sz15′ / uap5628 mutants (purple tracks). Wild type UAP56-Venus and sz-Venus transgenes were expressed in w1 ovaries and immunoprecipitated with anti-FLAG. Bound RNAs were quantified by strand-specific RNA-Seq. The UAP-Venus input signal is in light brown, the UAP-Venus IP signal is in orange, the sz-Venus input is in light green, and the sz-Venus IP signal is dark green. All signals are normalized to ribosomal RNAs. Cluster transcripts were highly enriched over input by both wild type UAP56-Venus and sz-Venus IP immunoprecipitation. By contrast, the neighboring gene, Pld, showed a modest enrichment. The sz-Venus mutant showed somewhat diminished enrichment relative to wild type, for both cluster and gene transcripts.. B. Cluster specific piRNAs (pink) accumulate anti-sense to the embedded transposon fragments (brown, direction of transcription indicated by arrowheads). UAP56 bound precursor RNAs are derived from the same regions, but do not show clear strand bias (UAP-Venus RIP). Also see Figure S6.

Figure 7

Figure 7. UAP56 structure and function in a piRNA processing compartment

A. Drosophila UAP56 amino acid sequence treaded into a high-resolution crystal structure of human UAP56. The N-terminal and C-terminal domains are connected by a flexible linker (green), and ATP and RNA binding motifs (orange) are located at the interface between the 2 domains. Glutamate 245 (red), which is changed to a lysine in uap56sz15′, is located in a β sheet in the N-terminal domain that is attached to the linker. This negatively charged residue also appears to form a salt bridge with Arg236 (blue), which is located in an α-helix in the N-terminal domain. B. A model for UAP56 function within piRNA processing compartment that spans the nuclear envelope. (1) UAP56 associates with nascent transcripts from dual-strand piRNA clusters, which bind the HP1 homolog Rhi. We speculate that UAP56 interacts with clusters through a domain defined by residue 245E. (2) UAP56-cluster transcript complexes interact with the nuclear pore, which triggers RNA release and export. (3) Vas binds cluster transcripts within the pore, or as they emerge from the pore and enter nuage, and then delivers these RNAs to the processing machinery. Also see Figure S7.

Comment in

Similar articles

Cited by

References

    1. Aravin AA, Hannon GJ, Brennecke J. The Piwi-piRNA pathway provides an adaptive defense in the transposon arms race. Science (New York, NY) 2007;318:761–764. - PubMed
    1. Bennetzen JL. Transposable element contributions to plant gene and genome evolution. Plant Mol Biol. 2000;42:251–269. - PubMed
    1. Bergman CM, Quesneville H, Anxolabehere D, Ashburner M. Recurrent insertion and duplication generate networks of transposable element sequences in the Drosophila melanogaster genome. Genome Biol. 2006;7:R112. - PMC - PubMed
    1. Brennecke J, Aravin AA, Stark A, Dus M, Kellis M, Sachidanandam R, Hannon GJ. Discrete small RNA-generating loci as master regulators of transposon activity in Drosophila. Cell. 2007;128:1089–1103. - PubMed
    1. Britten RJ. Transposable element insertions have strongly affected human evolution. Proc Natl Acad Sci U S A. 2010;107:19945–19948. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources