Fig 1.
Schematic phylogenetic tree of Eukaryota at the phylum level.
(A) Phyla where HEN1 protein was functionally studied appear in purple. Phyla where HEN1 homologs are found in the genome but have not been functionally studied yet appears in red. The sponges (Porifera) and comb jellies (Ctenophora) were illustrated as polytomy due to the current uncertainty regarding their relative phylogenetic positions. (B) A table indicating the type of sRNAs methylated by HEN1 in respective species is provided.
Fig 2.
In Nematostella vectensis the miRNAs are frequently methylated and methylation frequencies are stable during development.
(A) Scatter plot presenting the change in normalized read counts of individual miRNAs in control and periodate treated libraries. In blue, miRNAs whose levels changed less than two-fold. In green, miRNAs whose levels dropped two-fold or more. The majority of the miRNAs showed little to no change after periodate treatment. The non-methylated spike-ins are indicated as red dots in the scatterplot to demonstrate the efficiency of the periodate treatment. The axes are scaled to normalized read counts (NRC). The data represents the mean of three independent biological replicates. (B) Heatmap displaying log2-fold change of miRNA read counts between periodate treated and untreated samples in late planula (LP), primary polyp [9], adult female and male (AF and AM). The data is divided into two major clusters of heavily and weakly methylated miRNAs based on the fold change. To reduce noise, lowly expressed miRNAs (less than 50 read counts for a individual miRNA) were excluded from this analysis. (C) Depletion of miRNA* upon periodate treatment. The scatterplot represents the fold change in read counts of guide and star sequences of individual miRNAs before and after periodate treatment. Ratio of fold changes equal or larger than 1.5 are indicated in green. The miRNA* of miR-2022, miR-2025 and miR-2026 showed significantly higher fold-change compared to their guide sequences. Guides of moderately and weakly methylated miRNAs such as miR-2027 and miR-2028 showed a similar fold change to their stars. The results are presented in a box-plot in (D) showing the overall higher fold-change for star sequences compared to their guides. The box plot presenting the mean fold change for miRNA and miRNA* analyzed from planula larvae, primary polyps, adult male and female. P < 0.00001, Mann-Whitney test, miRNA n = 31.
Fig 3.
HEN1 is essential for Nematostella development.
(A-B) Schematic diagram of MO targeting region on the hen1 gene of Nematostella. The MO is designed to target hen1 exon-intron junction located in methyltransferase domain (green). This MO impaired the splicing by deleting 3rd exon of hen1. The splicing variation was validated by PCR. Due to deletion of 3rd exon, the band in HEN1 MO-injected embryos shifted down. In contrast, the bands in control MO-injected embryos and wildtype presented the expected size. (C-D) Animals injected with control MO developed to primary polyps after 7 dpf. In contrast, animals injected with HEN1 MO stopped developing prior to metamorphosis (D) ~90% of HEN1 depleted animals did not reach primary polyp stage at 7 dpf, triplicates, n = 300, ***P < 0.001 (Student’s t-test). (E) The timeline of Nematostella development and the relative progress of control and HEN1 MO-injected animals. B = Blastula; G = Gastrula; EP = Early Planula; LP = Late Planula; M = Metamorphosis; PP = Primary Polyp.
Fig 4.
HEN1 is required for miRNA stability.
(A) Change in miRNA read counts after HEN1 depletion is depicted by scatter plot. Each dot represents the read counts of an individual miRNA before and after HEN1 knockdown. miRNAs that showed a depletion ≥ two-fold change are indicated in green. The data represent the mean of two independent biological replicates. (B) The relative abundance of miRNA read counts of HEN1 MO vs. control MO are presented in a bar plot. A significant reduction of miRNA read counts is noted in HEN1 MO, (P < .0001, Wilcoxon signed-rank test). The data represents the mean of two independent biological replicates ± SD. (C) The levels of miR-2025, miR-2026, miR-2027 and miR-2028, before and after HEN1 knockdown measured by qPCR using LNA primers. As opposed to miR-2027 and miR-2028, the abundance of mir-2025 and mir-2026 decreased in HEN1 knockdown. The data represents the mean of minimum five independent biological replicates ± SD. ** P ≤ 0.005. * P ≤ 0.05, (Student’s t-test). (D) Loss of HEN1 results in shortened Nematostella miRNAs. Total reads mapped to guide miRNAs are analysed based on their isoform sizes ranging from 18 to 24 nt. the read counts were presented as the percentage of nucleotide length distribution. In HEN1 MO data the miRNA length is reduced at the percentage of 22 and 23 nt sized isoforms, as indicated with red arrows. (E) miRNAs from HEN1 MO showed accumulation of shorter isoforms. The ratio of percentage of read counts of the most abundant mature miRNA isoform and an isoform shorter by one nucleotide than dominant mature miRNA (“Mature -1nt”) was calculated between the HEN1 MO and control MO the experiment was performed in duplicates and the result is significant at P ≤ 0.01 (Mann-Whitney test). (F) Mean lengths of individual miRNAs were compared between the control MO and HEN1 MO, the data are presented by scatter plot. miRNAs showed in red dots exhibited decrease in mean length in HEN1 MO vs. control MO (P ≤ 0.003, Wilcoxon Signed-Rank Test).
Fig 5.
HEN1 is required for piRNA stability.
(A) The piRNA read counts were normalized, each dot represents the abundance of an individual piRNA analyzed from HEN1 MO vs. control MO. piRNAs that showed a depletion ≥ two fold are indicated in green. The data represent the mean of two independent biological replicates. (B) The Relative abundance of Nematostella piRNAs between HEN1 MO and control MO, in HEN1 MO the piRNA read counts were significantly reduced (P < .0001, Wilcoxon Signed-Rank Test). (C) Scatter plot presenting the change in normalized read counts of individual piRNAs upon periodate treatment. ~90% of piRNAs remained unchanged in abundance upon periodate treatment (indicated in blue). (D) The percentage of nucleotide length distribution plotted for piRNAs reads mapped at 22–28 nt in length. In HEN1 MO data the percentage of 27 and 28 nt sizes was reduced (indicated with red arrows) compared to the control.
Fig 6.
miRNA and piRNA biogenesis components are essential for Nematostella development.
(A) Western blot analysis using custom Nematostella Dicer1 antibody confirms the protein levels were reduced by the knockdown (B) Schematic diagram of MO targeting region on the PIWI2 gene of Nematostella. The MO is designed to target PIWI2 exon-intron junction. This MO impaired the splicing by 3rd intron retention of PIWI2 gene. The splicing variation was validated by PCR. Due to intron retention, the PCR product in PIWI2 MO-injected embryos shifted its size. In contrast, the bands in control MO-injected embryos and wildtype presented the expected size. (C) Animals injected with control MO developed to primary polyps after 10 dpf. In contrast, animals injected with Dicer1 MO stopped developing prior to metamorphosis (D) ~76% of Dicer1 depleted animals did not reach primary polyp stage at 10 dpf, experiment performed in triplicates, n = 300, ***P < 0.005 (Student’s t-test). (E) Animals injected with control MO developed to primary polyps after 9 dpf. In contrast, animals injected with PIWI2 MO stopped developing prior to metamorphosis (F) ~90% of PIWI2 depleted animals did not reach primary polyp stage at 9 dpf, triplicates, n = 300, ***P < 0.005) (Student’s t-test).
Fig 7.
Nematostella Dicer1 and PIWI2 knockdown affects sRNA biogenesis.
In the scatter plots in panels A, C, E and G the sRNA read counts were normalized, each dot represents the abundance of an individual sRNA analyzed from morphants vs. control. sRNAs that showed a depletion greater than two fold are indicated in green. (A) miRNAs in Dicer1 MO vs. control MO (B) Relative abundance of Nematostella miRNAs between Dicer1 MO and control MO, in Dicer1 MO the miRNA read counts were significantly reduced (P < 0.0001, Wilcoxon signed-rank test). (C) piRNAs in Dicer1 MO vs. control MO. (D) Relative abundance of Nematostella piRNAs between Dicer1 MO and control MO, in Dicer1 MO the piRNA read counts were not significantly reduced (P = 0.32218, Wilcoxon signed-rank test). (E) miRNAs in Piwi2 MO vs. control. (F) Relative abundance of Nematostella miRNAs between Piwi2 MO and control MO, in Piwi2 MO the miRNA read counts were significantly reduced (P < 0.0001, Wilcoxon signed-rank test). (G) piRNAs in Piwi2 MO vs. control MO. (H) Relative abundance of Nematostella piRNAs between Piwi2 MO and control MO, in Piwi2 MO the piRNA read counts were significantly reduced (P < 0.0001, Wilcoxon signed-rank test).
Fig 8.
A putative schematic representation of Nematostella miRNA biogenesis and methylation of guide miRNA.
This scheme is based on the results of the current work as well as results of previous studies in Bilateria [2, 32, 66, 67]. The model suggests that after strand selection by the AGO occurs, the guide strand is methylated by HEN1. When methylation does not occur the guide strand is degraded by exonucleases.