Pdsg1 and Pdsg2, novel proteins involved in developmental genome remodelling in Paramecium

Abstract

The epigenetic influence of maternal cells on the development of their progeny has long been studied in various eukaryotes. Multicellular organisms usually provide their zygotes not only with nutrients but also with functional elements required for proper development, such as coding and non-coding RNAs. These maternally deposited RNAs exhibit a variety of functions, from regulating gene expression to assuring genome integrity. In ciliates, such as Paramecium these RNAs participate in the programming of large-scale genome reorganization during development, distinguishing germline-limited DNA, which is excised, from somatic-destined DNA. Only a handful of proteins playing roles in this process have been identified so far, including typical RNAi-derived factors such as Dicer-like and Piwi proteins. Here we report and characterize two novel proteins, Pdsg1 and Pdsg2 (Paramecium protein involved in Development of the Somatic Genome 1 and 2), involved in Paramecium genome reorganization. We show that these proteins are necessary for the excision of germline-limited DNA during development and the survival of sexual progeny. Knockdown of PDSG1 and PDSG2 genes affects the populations of small RNAs known to be involved in the programming of DNA elimination (scanRNAs and iesRNAs) and chromatin modification patterns during development. Our results suggest an association between RNA-mediated trans-generational epigenetic signal and chromatin modifications in the process of Paramecium genome reorganization.

Figure 1. Effects of PDSG1 and PDSG2 silencing on progeny survival and IES excision.

(A) Survival test. Graphic representation of percent of normally dividing (white), sick (grey) and dead (black) progeny cells. The silencing of PDSG1 and PDSG2 was lethal in 95% and 97% of cells, respectively. NOWA1-KD is a positive control. Empty vector (EV) is a negative control. (B) PDSG1 and PDSG2 silencing efficiency was assayed by Northern blot by comparing the control (EV, empty vector) and silenced cultures. Three developmental points were analysed during sexual cycles. Early developmental stage (E) includes cells undergoing meiosis and 50% of cells present fragmented old MAC. Middle developmental stage (M) presents 100% of the cells with fragmented old MAC. Late developmental stage (L) includes cells with fragmented old MAC and a substantial number of cells with evident new developing MAC. (C) Effect of PDSG1 and PDSG2 silencing on transposon elimination. Macronulcear DNA was extracted form PDSG1-KD and PDSG2-KD cultures and analyzed for the retention of Sardine and Thon transposons using specific probes. Quantification signal of two classes of transposons was normalized to the mitochondrial DNA probe (mtDNA). * A two-tailed Student's t-test was used to assess statistical significance of the differences in the mean (values of bars), and an asterisk is shown if the p-value from this test is <0.05. For Sardine elements p-values are: PDSG1-KD vs Ctrl: 0.054; PDSG2-KD vs Ctrl: 2.7e-3. For Thon elements p-values are: PDSG1-KD vs Ctrl: 8.8e-5; PDSG2-KD vs Ctrl: 2.3e-4. Probe against Actin gene was used as an additional loading control. (D) IES retention PCR. Excision of 5 mcIES (1–5) and 2 non-mcIESs (6–7) are shown. Upper band represents IES+, lower band represents DNA with excised IES (IES-). IES: 1 (mtA promoter IES); 2 (51G4404); 3 (51A6649); 4 (51A2591); 5 (51G2832); 6 (51G1413); 7 (51A1835).

Figure 2. Subcellular localization of Pdsg1-GFP and Pdsg2-GFP fusion proteins during development.

1. Graphical representation of C-terminally tagged PDSG1 with GFP. 2. Graphical representation of N-terminally tagged PDSG2 with GFP. (A–E) Localization pattern of PDSG1-GFP. (F–J) Localization pattern of Pdsg2-GFP. (A, F) Vegetative cells with the intact MAC. (B, C, G, H) beginning of old MAC fragmentation that represents early development. (D, I) Middle stage of development with fragmented MAC. (E, J) Late development when new MAC is formed while the fragments of the old MAC are still present in the cytoplasm. Magenta: DAPI; green: GFP; white arrow: old macronucleus; arrowheads: new MAC (NM).

Figure 3. Effects of PDSG1 and PDSG2 knockdowns on developmental-specific small RNA.

(A–C) Histograms of sRNAs size for MAC genome-matching (yellow) and IES- matching (green) reads of the early and late developmental stages for control and PDSG1 and PDSG2 knockdowns. (D, E) Representative sequence logos of 25 nt and 27 nt sRNAs for control and knockdown cells from the late developmental stage. Note that due to the low abundance of iesRNAs in the knockdowns in (E) a single sRNA is relatively abundant in the 27 nt sRNAs.

Figure 4. Effect of PDSG1 depletion on the retention of injected IES (51A2591).

Lanes 1 and 2, control non-silenced cells. Lanes 3 and 4, control silencing with empty vector (EV). Lanes 5 and 6, PDSG1 depleted cells. Lanes 7 and 8, DCL2 and DCL3 depleted cells. Lanes 2, 4, 6 and 8, IES 51A2591 injected cells. IES-mtA promoter (control mcIES).

Figure 5. Effects of Pdsg1 and Pdsg2 depletion on H3K27me3 histone modification.

(A, D, G) Maternal MAC in vegetative cells of control, PDSG1-KD and PDSG2-KD. (B, E, H) Middle stage of development with the fragments decorated with H3K27me3. (C, F, I) Late stage of development with histone modification present in the new MAC. J Peptide competition assay. Magenta: DAPI; yellow: H3K27me3; white arrow: MAC; arrowhead: new MAC. Scale bar 5 µm.

Figure 6. Proposed model of Pdsg1 and Pdsg2 roles in elimination of mcIESs during MAC development.

(A) Early development, MIC germline genome is transcribed and transcripts are processed by Dcl2/3 into scnRNAs. (B) scnRNAs are transported from the MIC to parental MAC by Ptiwi01/09. Once in the parental MAC, scanning takes place filtering out the MAC genome-matching scnRNAs. We propose that the matching of scnRNAs to complementary sequences may be driven by a multiprotein complex that may include Pdsg1. (C) scnRNAs without matching sequences are transported to the new MAC where they target the excision of complementary sequences. This study suggests that Pdsg2 may be part of the DNA excision machinery. (D) After the excision, IESs are used as templates for iesRNA production to ensure reinforce the signal for an efficient targeting of IES excision.