Unveiling the partners of the DRBD2-mRNP complex, an RBP in Trypanosoma cruzi and ortholog to the yeast SR-protein Gbp2

DRBD2 is an ortholog of the SR-protein Gbp2 in yeast

The results from BLAST together with those from the multiple sequence analysis provided convincing evidence that the T. cruzi DRBD2 (TcCLB.510755.120) protein is an ortholog of Gbp2 proteins in other organisms, such as T. brucei and S. cerevisiae, with approximately 40% similarity between the sequences. However, domain conservation and the protein sizes show that these proteins acquired - or lost - different functions during evolution (Additional file 1: Figure S1).

DRBD2 is an RBP with two RRM domains Fig. 1. Its ortholog in yeast, Gbp2, is an SR-rich protein with three RRM domains (Fig. 1) [28]. The SR-protein Gbp2 forms a mRNP complex that acts in the quality control and export of recently spliced mRNAs [24]; after mRNA transport, the protein remains associated with the transcripts, suggesting a role in translation regulation [29].

Although DRBD2 does not have SR-rich motifs as in Gbp2, DRBD2 contains several predicted phosphorylation sites in its sequence (Additional file 2: Figure S2). Phosphorylation is the main posttranslational modification responsible for nuclear transport and other cellular processes [30]. The phosphoproteome analysis of T. cruzi identified phosphorylated cysteine residues in DRBD2 from exponentially growing epimastigotes (Fig. 1a) [31]. Based on this result, we hypothesize that DRBD2 is involved in (i) mRNA transport, such as the Gbp2 modus operandi [29], and/or (ii) binding transcripts being exported from the nucleus, as DRBD2 is an RBP with two RRM domains [8].

TAP-tagged DRBD2 is detected throughout metacyclogenesis

TAP-tagged DRBD2 is detected throughout the metacyclogenesis process (Fig. 2a and b). It is a cytoplasmic protein, slightly concentrated around the nucleus in unstressed epimastigotes and metacyclic trypomastigotes (Fig. 2a). This pattern of cellular localization would corroborate the hypothesis raised above, that DRBD2 could have a role in the nuclear transport of its target transcripts. In stressed parasites, however, DRBD2 altered its localization, displaying a granular formation scattered in the cytoplasm (Fig. 2a); this shift in the protein localization suggests an alteration of DRBD2 function during stress. In addition, RNA-seq data available at tritryp database, showed that the drbd2 gene is expressed in all developmental forms of the parasite (Additional file 3: Figure S3 A and B) [32–34].

DRBD2 is present in translational complexes in unstressed epimastigotes

We investigated if DRBD2 was present in cytoplasmic complexes involved in translation by performing a polysome fractionation of epimastigotes in exponential growth (Fig. 3a) and epimastigotes subjected to a nutritional stress (Fig. 3b). Nutritional stress is one of the key steps that triggers T. cruzi metacyclogenesis [35]. DRBD2 was detected in polysome-enriched and polysome-independent fractions in both conditions analyzed - unstressed and stressed epimastigotes (Fig. 3a and b, respectively).

The sedimentation profile of DRBD2 on the sucrose density gradients of unstressed epimastigotes showed that its migration was similar to the S7 ribosomal protein - used as a control of polysomal fractions - after treatment with both cycloheximide and puromycin (Fig. 3a). These results indicate that DRBD2 is, in fact, associated with the translation machinery in unstressed epimastigotes. In stressed parasites, however, DRBD2 was associated with polysome-independent heavy complexes after puromycin treatment, differently from the control S7 (Fig. 3b), indicating that DRBD2 is not associated with the translation machinery in epimastigotes subjected to nutritional stress.

In yeast, Gbp2 is also present in cytoplasmic mRNP complexes formed by Gbp2, Npl3, and Hrbd1 [29]; this complex represses the translation of the regulated transcripts. During glucose deprivation, Gbp2, Npl3, and Hrbd1 were found in P-bodies and stress granules in yeast [36, 37], suggesting a translation arrest of the associated transcripts in this condition.

DRBD2 associates with proteins involved in RNA metabolism in epimastigotes

The association of DRBD2 with translation complexes in unstressed epimastigotes motivated us to investigate the proteins partners of its mRNP complex. For this, we used immunoprecipitation (IP) assays and analyzed them by shotgun proteomics. Initially, we performed an SDS-PAGE of the IP assay eluates to compare DRBD2 immunoprecipitated complexes with the control, i.e., parasites expressing only TAP-tagging empty vector (Fig. 4). Figure 4a shows the detection of differentially abundant proteins in the DRBD2 immunoprecipitated complex compared to the control IP assay (lanes 2 and 1, respectively).

Assessment of transcripts regulated by the DRBD2-mRNP complex

We performed RNA-seq analysis of the immunoprecipitated DRBD2-mRNP complex to assess the regulated transcripts. The analysis pinpointed 86 transcripts as more abundant in the DRBD2-mRNP in epimastigotes when compared to the TAP-tag strain control, with an absolute fold-change of at least two (Additional file 5: Table S2 and Additional file 7: Table S4); however, most transcripts mapped to hypothetical proteins. To obtain further information regarding the transcripts, we performed gene ontology analysis (Fig. 5). Most of the mRNAs code for proteins related to biosynthetic process, DNA metabolic process, protein modification, and response to stress (Fig. 5). The complete information of the RNA-seq analysis is found in Additional file 5: Table S2, Additional file 6: Table S3, Additional file 7: Table S4. Table 3 lists the most abundant transcripts identified in this work.

Gene ID	Product description	Max group mean	Log2 fold change	Fold change	FDR p-value
TCDM_06351	ATP-dependent DEAD/H DNA helicase recQ	16,788,12	12,33	5145,66	0%
TCDM_14363	hypothetical protein	1629,85	7,63	198,67	0%
TCDM_09386	hypothetical protein	929,52	7,41	169,66	0%
TCDM_01670	hypothetical protein	605,60	7,25	152,51	0%
TCDM_14159	hypothetical protein	192,69	6,88	118,06	0%
TCDM_01210	hypothetical protein	516,81	6,88	117,47	0%
TCDM_13811	hypothetical protein	386,65	6,72	105,38	0%
TCDM_09098	hypothetical protein	1242,50	6,53	92,72	1%
TCDM_02352	RNA-binding protein	695,00	6,18	72,45	0%
TCDM_06973	hypothetical protein	502,23	5,86	57,96	0%
TCDM_12750	trans-sialidase	47,34	5,86	57,94	0%
TCDM_12372	hypothetical protein	30,65	5,63	49,67	0%
TCDM_13701	unspecified product	347,66	5,54	46,46	1%
TCDM_05675	unspecified product	284,03	5,33	40,18	1%
TCDM_03204	hypothetical protein	323,38	5,21	36,91	1%
TCDM_02335	UDP-galactose transporter	190,16	4,84	28,71	1%
TCDM_05198	DNA polymerase delta catalytic subunit	53,35	4,80	27,90	0%
TCDM_12807	hypothetical protein	58,72	4,75	26,88	1%
TCDM_00712	hypothetical protein	101,03	4,59	24,14	0%
TCDM_09723	hypothetical protein	94,04	4,53	23,03	1%
TCDM_03727	hypothetical protein	34,10	4,50	22,55	0%
TCDM_13534	hypothetical protein	9,56	4,35	20,41	0%
TCDM_06859	hypothetical protein	167,14	4,24	18,86	1%
TCDM_08048	hypothetical protein	13,59	4,01	16,15	0%
TCDM_06310	quiescin sulfhydryl oxidase	7,16	3,95	15,47	0%
TCDM_13669	hypothetical protein	7,50	3,95	15,45	0%
TCDM_14079	hypothetical protein	27,07	3,93	15,28	0%
TCDM_13037	hypothetical protein	79,99	3,90	14,93	0%
TCDM_03172	small GTP-binding protein RAB6	53,57	3,82	14,17	1%
TCDM_04048	hypothetical protein	49,91	3,75	13,50	0%
TCDM_11018	P27 protein	117,38	3,72	13,22	0%
TCDM_00934	hypothetical protein	37,36	3,72	13,17	1%
TCDM_02364	hypothetical protein	87,49	3,71	13,11	0%
TCDM_03710	nuclear lim interactor-interacting factor	19,47	3,68	12,84	0%
TCDM_03688	ATP-dependent DEAD/H RNA helicase - DHH1	39,78	3,68	12,77	0%
TCDM_10796	ATPase	34,11	3,66	12,65	1%
TCDM_12205	complement regulatory protein	12,93	3,62	12,27	1%
TCDM_07935	hypothetical protein	6,67	3,61	12,22	1%
TCDM_13524	hypothetical protein	55,72	3,54	11,66	1%
TCDM_14171	hypothetical protein	79,57	3,53	11,55	1%
TCDM_06468	hypothetical protein	11,22	3,50	11,35	1%
TCDM_01286	hypothetical protein	50,27	3,48	11,17	1%
TCDM_05765	DNA-directed RNA polymerase III subunit	34,71	3,47	11,06	1%
TCDM_03285	small GTP-binding protein Rab11	123,34	3,42	10,74	1%
TCDM_09209	fructosamine kinase	41,17	3,38	10,42	1%
TCDM_13613	retrotransposon hot spot (RHS) protein	10,02	3,36	10,26	0%
TCDM_10239	hypothetical protein	60,17	3,35	10,18	1%
TCDM_14224	hypothetical protein	172,44	3,32	10,00	1%

We then used the results from the T. cruzi ribosome profiling by Smircich et al. [34] to verify if the transcripts identified in this work are associated with polysomes (Additional file 5: Table S2 and Additional file 7: Table S4). Interestingly, some of the most abundant transcripts associated with the DRBD2-mRNP complex showed high levels of expression in the total RNA fraction rather than in the polysomes, in both epimastigotes and metacyclic trypomastigotes (Additional file 3: Figure S3C), indicating a possible role of DRBD2 in negatively regulating the translation of its target transcripts, as described in T. brucei [25].

In silico analysis

Similarities between the DRBD2 (TcCLB.510755.120) sequence from T. cruzi and the Gbp2 sequences from several other organisms from NCBI Refseq and UniProt SwissProt (accessed in September, 2018), including Trypanosoma brucei, Saccharomyces cerevisiae, and Homo sapiens, were assessed using the BLAST algorithm [44]. This analysis was achieved by querying the TcDRBD2 sequence against other organism genomes, followed by inverse alignment using the best candidates from the first search, using the best reciprocal hit (BRH) method [45]. Plus, we used Constraint-based Multiple Alignment Tool (COBALT), contained in NCBI platform, to perform a multiple protein sequence alignment using conserved domain and local sequence similarity information. To predict phosphorylation sites in the DRBD2 sequence, we used the NetPhos 2.0 Server [46] (http://www.cbs.dtu.dk/services/NetPhos/). Secondary structure mapping was obtained with PSIPRED [47].

Parasite cultures and drug treatments

T. cruzi Dm28c strain was originally isolated from Didelphis marsupialis [35, 48] and epimastigotes forms were maintained in axenic culture in liver infusion tryptose (LIT) medium at 28 °C [48]. To prepare 1.0 L of LIT medium, we used 9.0 g of liver infusion broth, 5.0 g of tryptose, 1.0 g of NaCl, 8.0 g of Na₂HPO₄, 0.4 g of KCl, 1.0 g of glucose, 100 ml of fetal bovine serum (heat-inactivated), 10 mg of hemin, and 1.0 L of distilled water. For nutritional stress induction, epimastigotes in the late logarithmic growth phase were harvested by centrifugation at 7000 x g for 5 min at 10 °C and incubated for 2 h at 28 °C in TAU medium (190 mM NaCl, 17 mM KCl, 2 mM MgCl₂, 2 mM CaCl₂, 8 mM phosphate buffer pH 6.0) at a density of 5.0 × 10⁸ parasites/ml. Metacyclic trypomastigotes were prepared as previously described [48].

Recombinant protein expression

The T. cruzi Dm28c strain was used as the DNA template. The DRBD2 (TriTrypDB ID TcCLB.510755.120) open reading frame polymerase chain reaction (PCR) product was cloned into the pDONR™221 vector from Gateway Technology (Invitrogen) and was then recombined into a pTcTAPN vector that adds the Tap-tag in the N-terminus portion of the protein [49]. T. cruzi Dm28c epimastigotes were transfected with these plasmids as described by Lu et al. [50]. Stable lines were selected by adding 500 mg/ml G418 to the culture medium. It is episomally expressed, not recombining into a genomic locus (DRBD2 molecular weight: 31 kDa; TAP-tagging: 21 kDa).

Immunoblotting and immunofluorescence

Immunoblotting and immunofluorescence were performed as previously described [51]. A rabbit anti-protein A antibody (Sigma-Aldrich) was used to detect TAP-tagging (1:40,000 dilution). TcDHH1 (47 kDa) was used as a control of the metacyclogenesis western blotting assay. Anti-TcDHH1 (1:100 dilution) was kindly provided by Jimena Ferreira da Costa. Bound antibodies were detected by the alkaline phosphatase method or by the Odyssey® imager (LI-COR) using the secondary antibodies IRDye 800CW or IRDye 680RD (1:15,000 dilution) [52]. Immunofluorescence slides were analyzed by inverted microscopy (Leica DMI6000B) associated with deconvolution software (Leica AF6000, microscope facility RPT07C PDTIS/ Carlos Chagas Institute - Fiocruz Paraná).

Polysome sedimentation profiles

Polysomes were separated on a sucrose density gradient as previously described, performed once for each condition [53]. The drugs used here were puromycin, which causes premature chain termination during translation, thus disassembling polysomes [54]; and cycloheximide, that interferes with the translocation step in protein synthesis, thus blocking eukaryotic translational elongation and stabilizing polysomes [54]. The resulting fractions were analyzed by western blotting using rabbit anti-protein A (1:40,000 dilution) for DRBD2 detection and mouse anti-S7 (1:1000 dilution), kindly provided by Dr. Stênio P. Fragoso. S7 (25 kDa) is a small ribosomal protein commonly used as a control. The fractionation was performed twice for each treatment and condition.

Immunoprecipitation assays

Immunoprecipitation (IP) assays were performed with cytoplasmic extracts of exponentially growing epimastigotes. Primary rabbit anti-protein A antibodies were coupled to 30 mg of Dynabeads® M-270 Epoxy (Thermo Fisher), following the manufacturer’s instructions. The equivalent of 5.0 × 10⁹ parasites expressing TAP-tagged DRBD2 and control parasites expressing only the TAP-tag were cryo-milled using the Mixer Mill MM 400 (Retsch), resulting in 1 g of cell powder. Approximately 5.0 × 10⁸ parasites were resuspended in 1 ml of solubilization buffer [25 mM and 50 mM sodium citrate; 20 mM Tris-HCl pH 7.5; 5 mM MgCl₂; 0.1% CHAPS; protease inhibitor (Complete Mini Protease inhibitor cocktail tablet, Roche, 1:100) and 40 U/ml RNase OUT (Invitrogen)], followed by centrifugation at 20,000 x g for 10 min at 4 °C. The supernatant was incubated with 3 μl of the magnetic beads under constant stirring at 4 °C for 2 h. The magnetic beads were washed 3 times with the solubilization buffer, and the proteins linked to the beads were eluted with 30 μl of elution buffer (2% SDS and 20 mM Tris-HCl pH 8.0) at 72 °C for 20 min. The eluted fractions were visualized using 13% SDS-PAGE with silver nitrate staining, as previously described [55], and then, TAP-tagged DRBD2 was detected by immunoblotting.

Shotgun proteomic data acquisition and analysis

Digested extracts of the immunoprecipitated lysates were fractionated by a NanoLC Easy 1000 system (Thermo Fisher Scientific). Ionized peptides were analyzed by an LTQ Orbitrap XL ETD (Thermo Fisher Scientific). Sequences of T. cruzi were downloaded from TriTrypDB on March 16, 2016. PatternLab for proteomics 4.0 was used for data analysis, following the provided instructions [56]. Briefly, the search was performed using PatternLab’s integrated Comet [57] search engine. The validity of the Peptide Sequence Matches (PSM) was assessed using PatternLab’s Search Engine Processor module (SEPro) [58]. A cutoff score was established to accept a false discovery rate (FDR) of 1% based on the number of decoys [59]. The results were postprocessed to accept PSM with less than 5 ppm from the average ppm of the pre-validated results. Relative quantitation was performed according to PatternLab’s Normalized Ion Abundance Factors (NIAF) as a relative quantitation strategy [56]. For proteins present in both conditions (DRBD2 assay and control), we used PatternLab’s TFold module [60], with an FDR of 0.05. Only proteins having an absolute fold change ≥ 2 were considered for the TFold analysis.

Ribonomic data acquisition and analysis

The eluted fractions from the immunoprecipitation assays were purified with the miRCURY™ RNA Isolation Kit (Exiqon), with the “Cell & Plant” protocol from the manufacturer’s manual, initiating with the “Cell Lysis” step in the manual. The purified RNAs were quantified and analyzed using the Agilent DNA 1000 kit on an Agilent Technologies 2100 Bioanalyzer and then subjected to sequencing on the Illumina® platform using paired-end configuration on the MiSeq Desktop Sequencer (Illumina®). The sequencing data obtained from 3 biological replicates were analyzed using CLC Genomics Workbench© 7.5.5. The T. cruzi Dm28c reference genome used for mapping was obtained from the NCBI database (AYLP01), and the alignment was performed as follows: additional upstream and downstream sequences of 75 bases, minimum number of reads: 10, maximum number of mismatches: 2, and nonspecific match limit: - 2. We selected possible targets of the DRBD2-mRNP with the β binomial statistical test (Baggerly’s test [61]) with a q-value ≤1%, a minimum RPKM value of 50 and a fold change > 5 when compared to the control (parasites expressing TAP-tag) for significance. The RNA-seq data were deposited in the NCBI Sequence Read Archive (SRA) database under the accession number SRX4560510.