SF3B1 phosphorylation is an evolutionarily conserved step in spliceosome activation carried out by the divergent, OTS964-insensitive kinase CRK9 in trypanosomes

Figures

• 

  Figure 1.. Trypanosome SF3B1 is phosphorylated in CRK9-dependent manner. (A) Schematic depiction of the modified CRK9 and SF3B1 loci in the CRK9WT/AS cell lines expressing HA-SF3B1. Plasmids CRK9WT-PTP-NEO or CRK9AS-PTP-NEO were integrated into one CRK9 allele and the remaining wild-type allele was replaced by the hygromycin phosphotransferase gene (HYG). In each cell line, integration of pPURO-HA-SF3B1 into one SF3B1 allele fused the HA tag N-terminally to the splicing protein. (B) Immunoblot showing the expression of HA-SF3B1 (top panel, anti-HA antibody), CRK9-PTP (middle panel, PAP reagent detecting the tag), and CRK9AP (anti-CRK9AP immune serum) in unmodified cells (427) and in CRK9WT-PTP (WT) and CRK9AS-PTP (AS) expressing cell lines. deg, putative CRK9-PTP degradation product. (C) Anti-HA immunoblot of whole cell lysate (WCL) that was subjected to gel filtration (GF) in the absence or presence of phosphatase inhibitors (p.i.) and incubated for 0 or 20 min. The top band corresponds to phosphorylated (p)HA-SF3B1. (D) 2- and 3-primer PCR assays to monitor cis and trans splicing defects of PAP1 RNA in WT and AS cells that were treated with DMSO or 10 μM 1-NM-PP1 (inh) for 3 h. (E)…

• 

  Figure 2.. CRK9 directly phosphorylates the NTD of SF3B1 in vitro. (A) Schematic of T. brucei SF3B1 and the fusion proteins GST-SF3B1-N and GST-SF3B1-C with N-terminal domain (NTD), HEAT repeat domain, anchor region and GST tag depicted to scale. (B) In vitro kinase assays were carried out with endogenous, tandem affinity-purified CRK9WT-P (WT) or CRK9AS-P (AS) enzyme, substrates GST, GST-SF3B1-N, or GST-SF3B1-C in the presence of ATPγS or its bulky analog 6-cHe-ATPγS. Kinase reactions were analyzed by immunoblotting. Thiophosphorylated proteins after alkylation with PNBM were detected with anti-thiophosphate ester antibody 51–8 (α-TPE), CRK9-P with anti-ProtC antibody, and GST fusion proteins with anti-GST antibody (α-GST). A degraded (degr.) GST-SF3B1 product of ∼40 kDa was consistently detected in our samples. In the α-TPE blot, CRK9-P is detected due to autophosphorylation and the red arrow indicates the expected position of GST-SF3B1-C.

• 

  Figure 3.. CRK9 phosphorylates TP motifs in the NTD of SF3B1 and interacts with SF3B1 in vivo. (A) In vitro kinase reactions were carried out with endogenous, tandem affinity-purified CRK9WT-P and decreasing concentrations of wild-type GST-SF3B1-N(-WT) or of GST-SF3B1-N-APmut. A non-specific band (non-sp.) was detected with α-TPE antibody even when ATPγS was omitted in a control reaction. (B) CRK9-PTP was immunoprecipitated (arrow) and detected in immunoblots in extract (Inp), supernatant (SN) and precipitate (P) with the PAP reagent. Co-precipitated HA-SF3B1 was detected with an anti-HA antibody, and TFIIB and CRK9AP were detected as negative and positive controls, respectively, with previously raised and characterized polyclonal immune sera. x-Values denote relative amounts analyzed.

• 

  Figure 4.. Inhibition of CRK9AS with 1-NM-PP1 reduces the interaction of PRP19 with the active spliceosome. (A) Schematic depiction of the CRK9 and PRP19 loci in the procyclic trypanosome lines CRK9AS-HAee/PRC5-PTP and CRK9AS-HAee/PRP19-PTP which exclusively express analog-sensitive, HA-tagged CRK9 and, from one endogenous allele, C-terminally PTP-tagged PRC5 or PRP19. (B) Schematic of RIP assays. (C) Immunoblot analysis of RIP assays with detection of TFIIB as negative control, showing extract (Inp), supernatant (SN), and precipitate (P). x-Value indicates relative amount loaded. (D) Primer extension of total RNA prepared from extract and precipitates with biotinylated oligonucleotides SL_PE and U2_PE, which are antisense to SL RNA and U2 snRNA, respectively. Their specific products have respective lengths of 126/130 nt, 102 nt, and 91 nt for SL RNA with [hypo-]/methylated cap, U2 snRNA, and the SL intron, which typically is present in the trans splicing intermediate Y-structure. Arrows point to U2 signals in precipitates which are strongly reduced in PRP19-PTP RIPs. Bottom panel is a longer exposure of the U2 and Y signals with asterisks depicting the loss of the SL intron in…

• 

  Figure 5.. The TP motifs in SF3B1’s NTD are of crucial importance to cell viability and pre-mRNA splicing. (A) Schematic outline of the experimental strategy. First, a cell line for conditional SF3B1 silencing was generated in which dox induces the expression of a hairpin RNA that targets the 3′ UTR of endogenous SF3B1 mRNA for degradation by the RNAi pathway. In second steps, complete RNAi-resistant SF3B1 transgenes under the control of a dox-controlled promoter were stably transfected into this cell line to determine their rescue potential. The transgenes were either wild-type (WT) or mutated such that the 13 TP motifs of the NTD were changed to AP motifs (APmut); they carried the HA tag sequence at the 3′ end of their coding regions and a different, RNAi-resistant 3′ UTR. (B) Culture growth of three independently derived clonal lines for each type of cell line in the absence and presence of dox were monitored for four days. The knockdown efficiency was determined by RT-qPCR after 1 day of induction for one representative cell line for each type, using α-tubulin mRNA for normalization. (C) Immunoblot of whole cell lysates of cell lines treated with or without doxycycline for 1 day,…

• 

  Figure 6.. Trypanosomes and CRK9 are insensitive to OTS964. (A) Non-linear regression EC50 analysis of fluorescence measurements using a viability assay after cells were incubated at different OTS964 concentrations for 48 h. (B) Cell lines which exclusively express either CRK9WT-PTP or CRK9AS-PTP were treated with DMSO, 10 or 25 μM OTS964 or 10 μM 1-NM-PP1 for 3 h. Total RNA preparations of these cells were then reverse-transcribed with random hexamers and subjected to 2- and 3-primer PCR assays to monitor PAP1 splicing defects. Arrows indicate minor splicing defects in CRK9WT-PTP-expressing cells at the highest OTS964 concentration. (C) Left, representative immunoblot in which thiophosphorylated CRK9WT-P (autophosphorylation) and GST-SF3B1-N were detected after in vitro kinase assays with ATPγS and increasing concentrations of OTS964. Right, the IC50 of OTS964 on CRK9WT-P autophosphorylation was determined by measuring the immunoblot signal strengths of three biological replicates by densitometry. Please note that it required an OTS964 concentration of 125 μM to reproducibly detect a mild reduction of the GST-SF3B1-N signal.

• 

  Figure 7.. Predicted binding of OTS964 to the CRK9 ATP-binding pocket. (A) Sequence alignment around the ATP-binding pocket of the kinase domains of T. brucei CRK9 (residues 267–685) and human CDK11B (Uniprot ID: P21127-12; human CDK11B isoform 7). Asterisks indicate identical residues, and colons mark conserved positions. The OTS964-interacting residues in CDK11B, identified from the structure of the CDK11B-OTS964 complex (PDB ID: 7UKZ), are highlighted in blue if identical in CRK9 and in red if non-identical. (B) Chai-1-predicted model of the CRK9–OTS964 complex. The N-terminal (residues 1–266) and C-terminal (residues 686–744) domains are omitted for clarity. M438, C518, and OTS964 are shown as licorice sticks. The CRK9-specific insertion (residues 349–430) is colored gray. (C) Superimposition of CRK9–OTS964 and CDK11B–OTS964 complex structures. CRK9 and CDK11B are shown in blue and magenta, respectively. (D) Close-up view of residues surrounding OTS964 in the CRK9 ATP-binding site. (E) Molecular surface representation of the OTS964-binding cavities in CDK11B, CRK9, and CRK9AS. The cavities were defined using the predicted structures of the CRK9– and CRK9AS–OTS964 complexes and the…