Editosome RNase III domain interactions are essential for editing and differ between life cycle stages in Trypanosoma brucei

Abstract

Multiprotein editosomes catalyze gRNA-specified insertion and deletion of uridines to create functional mitochondrial mRNAs in Trypanosoma brucei Three functionally distinct editosomes are distinguished by their single KREN1, KREN2, or KREN3 RNase III endonuclease and, respectively, KREPB8, KREPB7, and KREPB6 partner proteins. These endonucleases perform the first catalytic step of editing, cleaving mRNA in diverse mRNA/gRNA heteroduplex substrates. We identified divergent and likely noncatalytic RNase III domains in KREPB4, KREPB5, KREPB6, KREPB7, KREPB8, KREPB9, and KREPB10 editosome proteins. Because known RNase III endonuclease functional domains are dimeric, the editing endonucleases may form heterodimers with one or more of these divergent RNase III proteins. We show here using conditional null cell lines that KREPB6, KREPB7, and KREPB8 are essential in both procyclic form (PF) and bloodstream (BF) cells. Loss of these proteins results in growth defects and loss of editing in vivo, as does mutation of their RNase III domain that is predicted to prevent dimerization. Loss of KREPB6, KREPB7, or KREPB8 also dramatically reduces cognate endonuclease abundance, as does the RNase III mutation, indicating that RNase III interactions with their partner proteins stabilize the endonucleases. The phenotypic consequences of repression are more severe in BF than in PF, indicating differences in endonuclease function between developmental stages that could impact regulation of editing. These results suggest that KREPB6, KREPB7, and KREPB8 form heterodimers with their respective endonucleases to perform mRNA cleavage. We also present a model wherein editosome proteins with divergent RNase III domains function in substrate selection via enzyme-pseudoenzyme interactions.

Keywords: RNA editing; RNase III; developmental regulation; endonuclease; mitochondria; trypanosomes.

FIGURE 1.

RNase III domains have been identified in several editosome proteins. (A) Schematics of KREN1, KREN2, and KREN3 editosomes, with proteins containing an RNase III domain highlighted in color. Nonessential KREPB9 and KREPB10 are shown apart as they variably associate with editosomes. Editosomes contain 12 common core proteins, while KREN1, KREN2, and KREN3 and their partner proteins are mutually exclusive. (B) Alignment of RNase III domain sequences from KREPB4-B10, KREN1-N3, and C. jejuni (accession Q9PM40). A total of 299 sequences from 35 kinetoplastid species and strains were aligned using MUSCLE. Only T. brucei sequences and the RNase III signature motif/C-terminus are shown here for clarity. Colored cylinders denote the α-helices shown colored in the C. jejuni RNase III dimer structure (PDB ID 3O2R). A universally conserved glycine residue in the RNase III signature motif is shown shaded in magenta. Asterisks denote residues that are universally conserved in catalytic RNase IIIs, equivalent to D44 and E116 in C. jejuni. (C) Schematic showing location of crosslinks between RNase III domain containing editosome proteins identified in CXMS (McDermott et al. 2016). Domains are highlighted as indicated based on bioinformatics predictions (Worthey et al. 2003; Carnes et al. 2012b; McDermott et al. 2016). Black dots distributed across proteins indicate the positions of lysine residues available for CXMS crosslinking (McDermott et al. 2016).

FIGURE 2.

KREPB6, KREPB7, and KREPB8 are essential for in vitro growth in BF and PF cells. Cumulative growth of BF (left panels) or PF (right panels) CN cells in which tet-regulatable WT KREPB6, KREPB7, or KREPB8 is conditionally expressed or repressed. Data are representative of two independent experiments.

FIGURE 3.

Real-time PCR analysis shows loss of KREPB6, KREPB7, or KREPB8 expression, and impact on RNA editing in both BF and PF CN cells. Relative RNA abundance is shown for KREPB6, KREPB7, KREPB8, and never-edited mRNAs COI and ND4 (black bars), pre-edited mRNAs (white bars), and edited mRNAs (gray bars). For each target amplicon, the relative change in RNA abundance was determined by using telomerase reverse transcriptase (TERT) mRNA as an internal control, with BF or PF CN cell lines that have either KREPB6, KREPB7, or KREPB8 repressed compared to the same cell line in which KREPB6, KREPB7, or KREPB8 was expressed. Data are shown as means ± SEM from two independent experiments.

FIGURE 4.

Western blot analyses show KREPB6, KREPB7, and KREPB8 are critical for endonuclease abundance in BF and PF. The KREPB6, KREPB7, and KREPB8 CN cell lines were modified by introducing HA-epitope tags into the endogenous loci of each endonuclease. (A) Western analysis of total lysates (equivalent of 1 × 10⁷ BF cells/lane, and 5 × 10⁶ PF cells/lane) from BF and PF CN cells in which the tet-regulatable WT KREPB6, KREPB7, or KREPB8 alleles were expressed or repressed for 2 and 4 d, respectively. The western blots were probed with monoclonal antibodies against the HA-epitope tag on the cognate partner endonucleases, the editosome proteins KREPA1, KREPA2, KREL1, and KREPA3, and quantified by densitometry using mitochondrial Hsp70 for normalization. Asterisk denotes position of the TAP-tagged regulatable KREPB6 protein in PF, which is also detected by our editosome protein monoclonal antibodies. (B) Western quantification of the relative change in abundance of each endonuclease was determined in BF or PF CN cell lines that have either KREPB6, KREPB7, or KREPB8 repressed compared to the same cell line in which KREPB6, KREPB7, or KREPB8 was expressed. Data are shown as means ± SEM from two independent experiments (four experiments for PF KREPB6/KREN2-HA, KREPB7/KREN2-HA, and KREPB8/KREN1-HA).

FIGURE 5.

The highly conserved glycine in the RNase III domains in KREPB6, KREPB7, and KREPB8 is essential for in vitro growth in BF (left panels) and PF (right panels) cells. Cumulative growth of BF and PF CN cells that constitutively express V5-tagged WT, or RNase III Glycine (G) to Arginine (R) or Valine (V) mutant versions of KREPB6, KREPB7, or KREPB8 and in which the tet-regulatable WT alleles were repressed. Data are representative of two independent experiments.

FIGURE 6.

Western blots of editosome IPs from BF and PF CN cells with V5-tagged WT or mutant KREPB6, KREPB7, and KREPB8 and HA-tagged KREN3, KREN2, or KREN1 endonucleases. The cells either had KREPB6, KREPB7, or KREPB8 expression eliminated (CN) or exclusively expressed mutant G-R or WT RNase III V5-tagged KREPB6, KREPB7, or KREPB8. The IPs and input lysates were probed with monoclonal antibodies against the HA-epitope tag on the endonucleases, the V5-tag on the WT or mutant KREPB proteins, and the KREPA1, KREPA2, KREL1, and KREPA3 editosome proteins. For comparisons of G-R mutant relative to WT KREPB protein, cyan arrows highlight loss of cognate endonuclease (HA probing) in KREPB protein IPs, while yellow arrows highlight loss of cognate KREPB protein (V5 probing) in endonuclease IPs. Asterisks indicate heavy chains of IP antibodies.

FIGURE 7.

Proposed model for RNase III interactions within editosomes. Top panel (A) shows RNase III domain structures modeled on C. jejuni 3O2R for KREPB4 (orange), KREPB5 (yellow), KREPB6 (light green), and on 1O0W for KREN3 (dark green) (McDermott et al. 2016) in two configurations: On the left, two KREPB5 domains dimerize while KREPB4 forms dimers with KREPB6 and KREN3; and on the right, two dimers of KREPB4 with KREPB5 form while KREPB6 and KREN3 dimerize. Universally conserved glycine residue critical for function shown in magenta in each domain. Interchange between these configurations could facilitate ES recognition and cleavage by KREN3 using enzyme–pseudoenzyme regulation of catalytic activity. Bottom panel (B) shows crosslinks identified by CXMS between KREPB4, KREPB5, KREPB6, and KREN3 (represented by ovals), all of which are consistent with configurations in the model shown in the top panel (McDermott et al. 2016). Crosslinked Lys residues (K) are indicated by their position in the protein, and crosslinks depicted by black bars. Similar configurations are proposed for KREPB7-KREN2 and KREPB8-KREN1 but are not shown.