The dual role of Spn-E in supporting heterotypic ping-pong piRNA amplification in silkworms

Abstract

The PIWI-interacting RNA (piRNA) pathway plays a crucial role in silencing transposons in the germline. piRNA-guided target cleavage by PIWI proteins triggers the biogenesis of new piRNAs from the cleaved RNA fragments. This process, known as the ping-pong cycle, is mediated by the two PIWI proteins, Siwi and BmAgo3, in silkworms. However, the detailed molecular mechanism of the ping-pong cycle remains largely unclear. Here, we show that Spindle-E (Spn-E), a putative ATP-dependent RNA helicase, is essential for BmAgo3-dependent production of Siwi-bound piRNAs in the ping-pong cycle and that this function of Spn-E requires its ATPase activity. Moreover, Spn-E acts to suppress homotypic Siwi-Siwi ping-pong, but this function of Spn-E is independent of its ATPase activity. These results highlight the dual role of Spn-E in facilitating proper heterotypic ping-pong in silkworms.

Keywords: BmAgo3; Ping-Pong Cycle; Siwi; Spn-E; piRNA.

Figure 1. The ATPase-deficient Spn-E-EQ forms aggregates with BmAgo3.

(A) CBB staining of immunoprecipitated BmAgo3 complexes from BmN4 cells treated with dsRNA targeting Rluc or Siwi. Rluc; Renilla luciferase, control. HC, IgG heavy chain; LC, IgG light chain. (B) Subcellular localization of BmAgo3, Spn-E, and FLAG-DDX43 in BmN4 cells treated with dsRNA targeting Rluc or Siwi. Rluc; Renilla luciferase, control. Scale bar, 5 μm. (C) Subcellular localization of FLAG-Spn-E (wild type or EQ) and BmAgo3 in BmN4 cells treated with dsRNA targeting the Spn-E 3′ UTR. Scale bar, 5 μm. (D) Western blot analysis of immunopurified FLAG-Spn-E (wild-type or EQ) complexes from BmN4 cells treated with dsRNA targeting the Spn-E 3′ UTR. Quantification data from three independent experiments are shown in Fig. EV1A. (E) Northern blot analysis of piR1712, piR2986, and piR484 in BmN4 cells co-transfected with the FLAG-Spn-E (wild type or EQ) plasmid and dsRNA targeting the Spn-E 3′ UTR. Quantification data for mature piRNA and pre-piRNA signals from four independent experiments are shown in Fig. EV1F,G, respectively. Source data are available online for this figure.

Figure 2. Depletion of Spn-E decreases BmAgo3-dependent production of Siwi-bound piRNAs while increasing Siwi–Siwi homotypic ping-pong.

(A) MA plot showing piRNA expression changes between control KD (Rluc RNAi) and Spn-E KD using dsRNA targeting the Spn-E 3′ UTR. Each dot represents one piRNA. Based on the changes in expression, piRNAs were divided into three groups: “increased” (red, n = 825), “unchanged” (green, n = 824), and “decreased” (blue, n = 825). (B) Relative fractions of 1U10A, 1U but not 10A, 10A but not 1U, and neither 1U nor 10A piRNAs of each group defined in (A). (C) Schematic representation of A → S, S → A, and S → S piRNAs. (D) Changes in the expression of A → S, S → A, and S → S piRNAs in Spn-E KD relative to control KD (Rluc RNAi). Two different dsRNAs targeting the Spn-E CDS and 3′ UTR were used. (E) Changes in the expression of A → S, S → A, and S → S piRNAs in the indicated conditions relative to control KD (Rluc RNAi). The Spn-E RNAi (UTR) data in (D) are included for comparison. (F) Summary of the results from the Spn-E KD rescue experiment. Spn-E KD decreases A → S piRNAs and increases S → S piRNAs. A → S piRNA production requires the ATPase activity of Spn-E, whereas the suppression of S → S piRNA production does not. Source data are available online for this figure.

Figure 3. Artificial Siwi-bound piRNA production is impaired by Spn-E KD but not by DDX43 KD.

(A) Schematic representation of the piR484-A reporter. piR484-A is an artificial A → S piRNA with a chimeric sequence of piR484 and piR2986. The details are shown in Fig. EV3A. (B) Siwi was immunoprecipitated from BmN4 cells co-transfected with the piR484-A reporter plasmid and dsRNA targeting the indicated genes for RNAi. Immunoprecipitated Siwi and Siwi-bound piR484-A were detected by western blotting and northern blotting, respectively. Rluc; Renilla luciferase, control. (C) Schematic representation of the production of piggyBac piRNA reporter-derived artificial Siwi-bound piRNAs. Left: The circular plasmid produces a reporter RNA with a long 3′ region. After cleavage by piggyBac piRNA-loaded BmAgo3, the 3′ end of the Siwi-loaded reporter RNA is processed by an unknown downstream cleavage event and/or trimming by Trimmer. Right: After cleavage by piggyBac piRNA-loaded BmAgo3, the reporter RNA from the linearized plasmid is expected to be 30-nt long, which is similar in length to that of endogenous piRNAs. Consequently, additional 3′ end processing is unnecessary in theory. (D) Siwi was immunoprecipitated from BmN4 cells co-transfected with a circular or linearized piggyBac piRNA reporter plasmid and dsRNA targeting the indicated genes for RNAi. Immunoprecipitated Siwi and reporter-derived artificial Siwi-bound piRNAs (art-Siwi piRNAs) were detected by western blotting and northern blotting, respectively. Rluc; Renilla luciferase, control. (E) KD rescue experiment of Spn-E using the piggyBac piRNA reporter. Siwi was immunoprecipitated from BmN4 cells co-transfected with a circular piggyBac piRNA reporter plasmid, the FLAG-Spn-E expression plasmid, and dsRNA targeting the Spn-E 3′ UTR. Immunoprecipitated Siwi and reporter-derived artificial Siwi-bound piRNAs (art-Siwi piRNAs) were detected by western blotting and northern blotting, respectively. Rluc; Renilla luciferase, control. Source data are available online for this figure.

Figure EV1. Characterization of Spn-E-EQ aggregates and BmAgo3-cleavage fragment release assay.

(A) Quantification of Siwi and BmAgo3 signals co-immunoprecipitated with FLAG-Spn-E (wild type or EQ) in Fig. 1D. Relative co-immunoprecipitated levels normalized to the Spn-E IP level are shown. Data are mean ± s.d. of three independent experiments (biological replicates). Statistical analysis was performed using a two-sided Student’s paired t-test, with p-values adjusted using the Holm method. NS not significant. (B) Western blot analysis of whole cell lysates of BmN4 cells treated with dsRNA for Rluc (control) or DDX6. The anti-DDX6 antibody successfully detected endogenous DDX6. (C) Subcellular localization of FLAG-Spn-E-EQ, DDX6, a P-body marker protein, and BmAgo3 in BmN4 cells. Scale bar, 5 μm. (D) Subcellular localization of FLAG-Spn-E-EQ, BmAgo3 and Siwi in BmN4 cells. Scale bar, 5 μm. (E) Tandem IP experiment on BmN4 cells co-transfected with the FLAG-Spn-E-EQ expression plasmid and dsRNA targeting the Spn-E 3′ UTR. FLAG-Spn-E-EQ was first immunoprecipitated with the FLAG tag, and the resulting immunopurified complex was then subjected to a second IP with normal rabbit IgG (Contl-IgG) or an anti-BmAgo3 antibody. Siwi in the second immunoprecipitate was detected by western blotting with or without RNase A treatment. (F) Quantification of mature piRNA signals in Fig. 1E. Relative expression levels normalized to those under Spn-E-WT expression are shown. Data are mean ± s.d. of four independent experiments (biological replicates). Statistical analysis was performed using a two-sided Student’s paired t-test, with p-values adjusted using the Holm method (*p  =  0.023). (G) Quantification of pre-piRNA signals in Fig. 1E. Relative expression levels normalized to those under Spn-E-WT expression are shown. Data are mean ± s.d. of four independent experiments (biological replicates). Statistical analysis was performed using a two-sided Student’s paired t-test, with p-values adjusted using the Holm method (*p  =  0.025, **p  =  0.003). (H) CBB staining of purified recombinant Spn-E and DDX43 proteins (rSpn-E and rDDX43, indicated by arrowheads) used for the in vitro cleavage fragment release assay. (I) Top: Schematic representation of the in vitro cleavage fragment release assay. To detect cleaved 5′ and 3′ fragments separately, the target RNA was radiolabeled at different positions (*). The 5′ or internally radiolabeled target RNAs were subjected to a cleavage assay using BmAgo3 immunoprecipitates. After the reaction, the bead fraction was incubated with rSpn-E or rDDX43 in the presence of ATP. Bottom: The cleaved fragments in the supernatant and bead fractions were detected by autoradiography. (J) Thin-layer chromatography for the detection of ATPase activity of the recombinant Spn-E protein. (K) Subcellular localization of FLAG-DDX43 (wild type or DA, an ATPase-deficient mutant) and BmAgo3 in BmN4 cells. Scale bar, 5 μm. (L) Subcellular localization of FLAG-Spn-E-EQ, HA-DDX43, and BmAgo3 in BmN4 cells. Scale bar, 5 μm.

Figure EV2. Changes in the expression of TE-mapped piRNAs upon KD of Spn-E or DDX43.

(A) Western blot analysis of whole cell lysates from BmN4 cells treated with dsRNA for Rluc (control), Spn-E, or DDX43. Two different dsRNAs (Spn-E: CDS and 3′ UTR; DDX43: two different regions of the CDS, #1 and #2) were used for RNAi. Rluc; Renilla luciferase. (B) Quantitative real-time PCR analysis of the expression of DDX43 in BmN4 cells treated with dsRNA for Rluc (control), Spn-E, or DDX43. Relative mRNA expression levels normalized to those of rp49 are shown. Rluc; Renilla luciferase. (C) MA plots showing piRNA expression changes for each TE between the control KD (Rluc RNAi) and Spn-E or DDX43 KD. Two different dsRNAs (CDS and 3′ UTR) were used for Spn-E RNAi. Each dot represents one TE. TEs with increased piRNA production in the Spn-E RNAi (UTR) are colored red. (D) MA plot showing piRNA expression changes between the control KD (Rluc RNAi) and Spn-E KD using dsRNA targeting the Spn-E CDS. Each dot represents one piRNA. Based on the three groups defined in Fig. 2A, piRNAs were color-coded as follows: “increased” (red, n = 825), “unchanged” (green, n = 824), and “decreased” (blue, n = 825). (E) MA plot showing piRNA expression changes between control KD (Rluc RNAi) and Spn-E KD using dsRNA targeting the Spn-E CDS. Each dot represents one piRNA. Based on the changes in expression, piRNAs were divided into three groups: “increased” (red, n = 805), “unchanged” (green, n = 805), and “decreased” (blue, n = 805). (F) Relative fractions of 1U10A, 1U but not 10A, 10A but not 1U, and neither 1U nor 10A piRNAs of each group in (E). (G) Scatter plots showing the PIWI binding bias of 1U but not 10A, 10A but not 1U, or 1U10A piRNAs (x-axis) and that of their putative partner piRNAs in the ping-pong cycle (y axis). (H) Changes in the expression of A → S, S → A, and S → S piRNAs in DDX43 KD relative to control KD (Rluc RNAi). Two different dsRNAs targeting the DDX43 CDS were used for RNAi. Rluc; Renilla luciferase.

Figure EV3. Construction of the piR484-A reporter system and Spn-E KD rescue experiment using the piR484-A reporter.

(A) Schematic explanation of the construction of the piR484-A reporter. piR484 is an S → A piRNA, and the sequence including the 5′ 10 nt and the upstream region of piR484 was replaced with the corresponding region of piR2986, an A → S piRNA, resulting in an artificial A → S piRNA, piR484-A. The downstream sequence of piR484 that contains a Siwi-bound piRNA target sequence was used without modification. (B) Quantitative real-time PCR analysis of the expression of DDX43 in the reporter experiments in Fig. 3B,D. Relative mRNA expression levels normalized to those of rp49 are shown. Rluc; Renilla luciferase, control. (C) KD rescue experiment of Spn-E using the piR484-A reporter. Siwi was immunoprecipitated from BmN4 cells co-transfected with the piR484-A reporter plasmid, the FLAG-Spn-E expression plasmid, and dsRNA targeting the Spn-E 3′ UTR. Immunoprecipitated Siwi and Siwi-bound piR484-A were detected by western blotting and northern blotting, respectively. Rluc; Renilla luciferase, control.