TFIIIB Subunit Bdp1 Participates in RNA Polymerase III Transcription in the Protozoan Parasite Leishmania major

Leishmania major, a protozoan parasite that diverged early from the main eukaryotic lineage, exhibits unusual mechanisms of gene expression. Little is known in this organism about the transcription factors involved in the synthesis of tRNA, 5S rRNA, and snRNAs, transcribed by RNA Polymerase III (Pol III). Here we identify and characterize the TFIIIB subunit Bdp1 in L. major (LmBdp1). Bdp1 plays key roles in Pol III transcription initiation in other organisms, as it participates in Pol III recruitment and promoter opening. In silico analysis showed that LmBdp1 contains the typical extended SANT domain as well as other Bdp1 conserved regions. Nevertheless, LmBdp1 also displays distinctive features, including the presence of only one aromatic residue in the N-linker region. We were not able to produce null mutants of LmBdp1 by homologous recombination, as the obtained double replacement cell line contained an extra copy of LmBdp1, indicating that LmBdp1 is essential for the viability of L. major promastigotes. Notably, the mutant cell line showed reduced levels of the LmBdp1 protein, and its growth was significantly decreased in relation to wild-type cells. Nuclear run-on assays demonstrated that Pol III transcription was affected in the mutant cell line, and ChIP experiments showed that LmBdp1 binds to 5S rRNA, tRNA, and snRNA genes. Thus, our results indicate that LmBdp1 is an essential protein required for Pol III transcription in L. major.


Introduction
In eukaryotic cells, transcription of nuclear DNA is carried out by three different classes of RNA polymerases (Pol): Pol I, II, and III. Each of these enzymes transcribes a specific group of genes, and each depends on transcription factors to bind its promoter regions. Pol I produces 18S, 5.8S, and 28S rRNAs [1], while Pol II generates mRNAs and most microRNAs and snRNAs [2]. Pol III transcribes genes encoding 5S rRNA, tRNAs, U6 snRNA, 7SL RNA, and other small essential RNA molecules [3,4]. Pol III promoters have been divided into three main types: 5S rRNA genes possess a type 1 promoter, composed of three gene-internal elements (boxes A and C and intermediate element); type 2 promoters are present on tRNA genes, and they consist of two gene-internal sequences (boxes A and B); and U6 snRNA genes contain type 3 2 BioMed Research International promoters, which are constituted by three small sequence elements located upstream of the genes (TATA box, proximal sequence element, and distal sequence element) [5].
General transcription factors TFIIIA, TFIIIB, and TFIIIC are required for accurate initiation of Pol III transcription [6]. TFIIIB is needed to form a transcriptionally active preinitiation complex (PIC) at all three types of promoter regions, as it participates in Pol III recruitment and promoter opening [7]. TFIIIB consists of three subunits: the TATAbinding protein (TBP), the TFIIB-related factor 1 (Brf1), and B double prime 1 (Bdp1) [8]. Notably, Bdp1 is specific to Pol III transcription, since it does not show homology with any of the general transcription factors of other RNA polymerases. Early studies showed that Bdp1 contains a SANT (Swi3, Ada2, N-Cor, and TFIIIB) domain, which is a highly conserved ∼50-amino acid motif present in several proteins involved in transcriptional regulation [9]. However, recent analysis established that the SANT domain present in Bdp1 from human and yeast is actually larger and possesses other distinguished features, and accordingly it was renamed extended SANT domain [10][11][12]. In yeast, the Bdp1 extended SANT domain is essential for cell viability, as it is needed for Pol III PIC assembly and stability [13]. The extended SANT domain interacts directly with DNA and with subunits Brf1 and TBP [14,15]. Other regions of Bdp1 associate with subunits C128, C37, and C34 of Pol III [11,13,16] and subunit Tfc4 of TFIIIC [17,18]. Thus, Bdp1 establishes an intricate network of interactions within the Pol III PIC.
The trypanosomatid parasite Leishmania major is the etiological agent of cutaneous leishmaniasis in Asia and Africa [19]. The parasite is transmitted to humans through the bite of infected sandflies of the genus Phlebotomus. In addition to their medical importance, L. major and other trypanosomatid protozoa, such as Trypanosoma brucei and Trypanosoma cruzi, are relevant because they express their genes by uncommon processes, including Pol II polycistronic transcription and trans-splicing [20][21][22].
Little is known in trypanosomatids about either the DNA sequences or the proteins that participate in Pol III transcription initiation. Unlike other organisms, all snRNA genes are transcribed by Pol III in these parasites [23]. Notably, snRNA expression in T. brucei and L. major is controlled by extragenic boxes A and B located within tRNA or tRNAlike sequences [24,25]. Although 5S rRNA genes contain the characteristic gene-internal elements, their promoter activity has not been proven [26]. Neither TFIIIA nor TFIIIC has been identified in trypanosomatids [27]. However, orthologues of TFIIIB subunits TBP and Brf1 have been studied in T. brucei. TBP, which participates in transcription by all three RNA polymerases, has been primarily investigated in the context of Pol II transcription of the spliced-leader (SL) RNA genes [28,29]. On the other hand, knockdown analysis by RNA interference demonstrated that Brf1 is indeed involved in Pol III transcription and that it is indispensable for cell survival of procyclic forms of T. brucei [30]. Another protein that participates in the regulation of Pol III transcription in T. brucei is Maf1, which inhibits tRNA and snRNA transcription by associating with their promoter regions [31].
Here we identified and characterized the TFIIIB subunit Bdp1 in L. major (LmBdp1). Bioinformatic analysis shows that LmBdp1 possesses the characteristic extended SANT domain and other conserved regions that flank this domain. Attempts to generate null mutants produced a cell line with an additional copy of LmBdp1. Nevertheless, Western blot analysis showed that the levels of LmBdp1 were considerably diminished in the mutant cell line. Notably, the growth of this cell line was significantly reduced in relation to wildtype cells. Nuclear run-on assays demonstrated that Pol III transcription was affected in the mutant cell line, and ChIP experiments showed that LmBdp1 binds to 5S rRNA, tRNA, and snRNA genes. pymol .org/2/) using the crystallographic structure of human extended SANT domain (PDB id: 5n9g) and yeast Bdp1 (PDB id: 6F41) as templates. Phosphorylated residues were predicted with PhosTryp (http://phostryp.bio.uniroma2.it/), which was developed for the specific identification of phosphorylated residues in Trypanosomatid proteins, as the method has been trained using phosphoproteomic data from Leishmania, T. brucei, and T. cruzi [32].

Indirect Immunofluorescence Assays.
The cellular localization of LmBdp1 labeled with a PTP tag was determined by indirect immunofluorescence, as previously described [26,34]. For these assays, 1.5 × 10 7 mid-log promastigotes were incubated with a rabbit anti-Prot C polyclonal antibody (Delta Biolabs) and a secondary goat anti-rabbit antibody conjugated with Alexa-Fluor 488 (Life Technologies). Images were obtained with a Zeiss Axio Vert.A1 epifluorescence microscope and analyzed with the ZEN 2012 software (Blue edition).

Southern Blot and PCR Analyses.
For Southern blot experiments, 5 g of genomic DNA was digested with XhoI and SacI (for analysis of single-knockout parasites), or Pst1 (for examination of double-knockout cells). The DNA was separated by electrophoresis on 0.8% agarose gels and transferred to Hybond-NC membranes (GE Healthcare) by capillary action. Blots were hybridized with the 5 targeting region from LmBdp1 (566 bp) labeled with [ -32 P]dCTP using the High Prime labeling system (Roche). Hybridizations were performed in 50% formamide, 5× SSC, 0.2% SDS, and 4× Denhardt's reagent at 42 ∘ C. Filters were washed at 68 ∘ C in 0.2× SSC and 0.1% SDS. To verify the replacement of the LmBdp1 gene with the pac gene, PCR analysis was carried out with oligonucleotides PAC-LOC3 -REV (5 -GTGGGCTTGTACTCGGTCATGG) and Bdp1for-upstream (5 -TGTTGGCAACTTGCCACCGT) which recognizes sequences located upstream of the 5 targeting region; and primers PAC-DHFR-3 -REV (5 -GGAGGG-AGGAATGAGGTGAGCT) and Bdp1-for-upstream. The correct integration of the hyg gene was confirmed by PCR with primers HYG-rev (5 -GTCGGAGACGCTGTCGAA-CT) and Bdp1-for-upstream.

Generation of LmBdp1 Polyclonal Antibody. Competent cells of
Escherichia coli BL21 (DE3) were transformed with plasmid pCold-LmBdp1. Expression of LmBdp1 recombinant protein (LmBdp1r) was induced with 1 mM isopropyl -D-1-thiogalactopyranoside (IPTG) at 37 ∘ C for 18 h. Affinity chromatography with Ni-Sepharose 6 Fast Flow matrix (GE Healthcare) was carried out to purify the LmBdp1r protein, according to the manufacturer's specifications. Around 50 g/animal of purified LmBdp1r was employed to inoculate subcutaneously six-week-old male BALB-C mice in TiterMax Gold adjuvant (Sigma) at a 1:1 ratio. Pre-immune normal mouse serum was collected before inoculation. Serum was collected six weeks after antigen immunization. Western blot analyses against LmBdp1r and protein extracts from promastigotes were performed to confirm the specificity of the anti-LmBdp1 polyclonal antibody.
2.9. Chromatin Immunoprecipitation Assays. The ChIP procedures were performed as previously described [31]. Briefly, 2 × 10 8 promastigotes were cross-linked with formaldehyde (final concentration of 1%) for 5 min at 37 ∘ C. A Vibra-Cell VCX130 ultrasonic processor (Sonics) was employed to lyse the cells (15 s on/off, 40% amplitude, for 5 min). Nuclei were pelleted and resuspended in sonication buffer (1% SDS, 10 mM EDTA, 50 mM Tris-HCl pH 8.1). Chromatin was sonicated to an average DNA size of about 200 to 500 bp with a BioRuptor UCD-200 (Diagenode) (30 s on/30 s off, high intensity) for 30 cycles. The sonicated material was precleared by adding protein A/G plus-agarose beads (Santa Cruz Biotechnology) and mixing for 1 h at 4 ∘ C. Chromatin samples were incubated overnight at 4 ∘ C with rabbit anti-Prot A antibody (Sigma) or nonspecific rabbit immune serum (negative control). Protein-DNA complexes were incubated for 1 h with protein A/G plus-agarose beads and 200 ng of sonicated salmon sperm DNA and washed as previously described [37]. Cross-links were reversed with 200 mM NaCl at 65 ∘ C overnight. Samples were treated with RNase A and proteinase K. DNA was precipitated with sodium acetate and ethanol and quantified. Each ChIP experiment was performed at least three times.

LmBdp1 Possesses the Extended SANT Domain and Other
Conserved Regions. Based on the presence of the SANT domain, gene LmjF36.6530 was identified as a potential orthologue of Bdp1 in L. major (LmBdp1) [20]. However, as SANT domains are also present in other proteins, including the subunits of many chromatin-remodeling complexes [9], we further analyzed the sequence and structure of LmBdp1 to verify its identity. Sequence alignments with Bdp1 orthologues from several species showed that LmBdp1 indeed contains the typical extended SANT domain, characterized by the presence of five -helices (Figure 1(a)). Moreover, two sequences that flank the extended SANT domain, the N-linker and the long arm [11,12], are conserved or semiconserved in LmBdp1. The N-linker, involved in the binding to the minor groove of the DNA, is distinguished by the occurrence of aromatic residues, including the invariably conserved tryptophan (W) residue that is present in LmBdp1 (Figure 1(a)). Nevertheless, LmBdp1 does not contain other conserved aromatic residues (Y291, F294, and Y299) that are important for the interaction Bdp1-DNA (Figure 1(a)) [10]. The long arm [11], also known as helix C [16] or Bdp1 stem [12], which participates in interactions with Brf1 and subunit C34 of Pol III, is also present in LmBdp1 (Supplementary Figure 1(a)). The tether region, located N-terminally to the N-linker (Supplementary Figure 1), interacts with Pol III subunits C128, C37, and C34 and in most Bdp1 orthologues contains several -sheet structures [11]. However, unlike most species, the tether region in LmBdp1 is not predicted to fold into -sheet structures (Supplementary Figure 1). Other amino acids that are important for the function of human Bdp1 are conserved in LmBdp1. These include R334 (R211 in LmBdp1), K338 (K215 in LmBdp1) and R343 (R220 in LmBdp1) (Figure 1(a)). R334 interacts with TBP, while K338 and R343 contact the major groove of the DNA [10]. The extended SANT domain is also conserved in Bdp1 orthologues in T. brucei and T. cruzi (Figure 1(a)), as well as in other species of Leishmania (Supplementary Figure 2). To further study the structure of the extended SANT domain present in LmBdp1, its predicted three-dimensional structure was generated by homology modeling, using as templates the crystal structures of Bdp1 from human (Figure 1(b)) and yeast (Supplementary Figure 1(b)). The architectures of the obtained three-dimensional structures for the extended SANT domain of LmBdp1 are very similar to the ones reported for human and yeast [10][11][12], showing a conserved distribution of the five -helices and the long arm. Thus, these results and the functional data showed below demonstrate that LmjF36.6530 is indeed the orthologue of Bdp1 in L. major.

LmBdp1 Is a Nuclear Protein.
In order to determine the cellular distribution of LmBdp1 in L. major, we performed indirect immunofluorescence experiments. To that end, a cell line where LmBdp1 was labeled with a C-terminal PTP tag was generated. The PTP tag is constituted by Protein A (Prot A) and Protein C (Prot C) epitopes separated by a tobacco etch virus (TEV) protease cleavage site [38]. The expression of the recombinant LmBdp1-PTP protein was confirmed by Western blot analysis with an anti-Prot C antibody (Figure 2(a)). Immunofluorescence experiments were carried out on fixed and permeabilized promastigotes with the same antibody. The images obtained demonstrated that LmBdp1 localizes to the nucleus, as anticipated for a protein that regulates transcription (Figure 2(b)). The observed dotted pattern is similar to that obtained with Brf1 in T. brucei [30].

Targeted Gene Replacement of LmBdp1.
To analyze the effects of the elimination of LmBdp1 on cell growth and transcription in promastigotes of L. major, we intended to generate LmBdp1 null mutant parasites by targeted gene replacement. Although L. major Friedlin is a diploid organism, some chromosomes are aneuploid [39]. However, it has been shown that this strain possesses two copies of chromosome 36 [40], where the LmBdp1 gene is located. Consequently, two sequential rounds of targeted gene disruption were planned to obtain null mutants of LmBdp1, using plasmids that contain puromycin (pac) and hygromycin (hyg) resistance genes. In these plasmids, the selectable marker genes are bounded by 5 and 3 flanking regions of LmBdp1.
The single-knockout cell line for LmBdp1 was generated by transfecting wild-type promastigotes with the targeting cassette from pΔLmBdp1-pac, and clones were selected with puromycin. Southern blot analysis using the 5 flanking region as a probe showed a 6.2 kb band, expected from targeted gene replacement, with genomic DNA from a singleknockout clone digested with XhoI, in addition to the 2.6 kb band that corresponds to the undisrupted gene ( Figure 3). The expected bands of 2.5 kb (gene replacement) and 1.7 kb (undisrupted gene) were also observed with SacI-digested genomic DNA. Moreover, the pac gene was amplified by PCR from the single-knockout clone (Supplementary Figure 3). Thus, these results demonstrate that one copy of LmBdp1 was replaced with the pac gene.
To obtain the double-knockout cell line for LmBdp1, the single-knockout clone was transfected with the targeting cassette from vector pΔLmBdp1-hyg, and cells were selected with puromycin and hygromycin. The obtained transfected clones showed growth defects (see below). Southern blots of PstIdigested genomic DNA from the double-knockout cell line showed that the second allelic copy of LmBdp1 was replaced with the hyg gene (2.8 kb band) (Figure 3). The expected band of 5.1 kb (pac gene) was also observed. However, a band of 4.6 kb that corresponds to the undisrupted LmBdp1 gene was also detected (Figure 3). PCR analysis also showed the presence of coding sequences for hyg and pac in the cell line (Supplementary Figure 3). These results indicated that an additional copy of LmBdp1 was present in the mutant cell line,  which was named double-knockout +1 (DKO+1). Thus, these data strongly suggest that LmBdp1 is an essential gene.

The Amount of the LmBdp1 Protein Was Reduced in the Mutant Parasites.
To further analyze the growth defect in the mutant cells, growth curves of the DKO+1 cell line were obtained and compared to growth curves of LmBdp1 single-knockout and wild-type parasites (Figure 4(a)). The data showed that the DKO+1 cell line is strongly impaired in growth, as it multiplies more slowly and to a lower stationary-phase cell density than single-knockout and wildtype promastigotes (Figure 4(a)). Growth curves of the single-knockout clone and wild-type parasites did not reveal significant differences (Figure 4(a)).
To determine the level of the LmBdp1 protein in the DKO+1 cell line, recombinant LmBdp1 (LmBdp1r) protein was obtained to produce antibodies against it. To that end, the complete LmBdp1 gene was amplified by PCR and cloned into the pColdI vector, where it was fused to a 6×His tag. The LmBdp1r protein was expressed in E. coli, purified by nickel affinity chromatography, and used as antigen for polyclonal The bands shown here and from two more independent assays were quantified and plotted (lower graph). Protein levels of LmBdp1 were normalized to the amount of -tubulin, which was used as loading control. Standard deviation bars are shown. antibody production in mice (data not shown). Western blot analysis with the LmBdp1 polyclonal antiserum showed that, comparing to wild-type cells, the amount of LmBdp1 protein was decreased by ∼70% in the DKO+1 cell line (Figure 4(b)). Thus, in spite of the presence of an additional copy of the LmBdp1 gene in the DKO+1 parasites, the expression of the LmBdp1 protein is considerably diminished.

Pol III Transcription Is Affected in the Mutant Cell Line.
To determine if the reduced levels of LmBdp1 have an effect on Pol III transcription, run-on experiments were performed with isolated nuclei from the DKO+1 cell line and wild-type promastigotes ( Figure 5). The genes analyzed were tRNA-Phe, tRNA-Tyr, 5S rRNA, and U2 snRNA (transcribed by Pol III); tRNA-Sec (transcribed by Pol II and Pol III);tubulin, LmjF.06.0200, LmjF.06.0210, and LmjF.06.0370 (transcribed by Pol II); and the 18S rRNA (transcribed by Pol I). The hybridization signals observed in Figure 5(a) and two more independent experiments were quantitated, setting to 100% the transcription signal obtained with wild-type cells ( Figure 5(b)). Normalization was performed with -tubulin. As anticipated, Pol III transcription was decreased in the DKO+1 cell line, since signal from U2 snRNA, 5S rRNA, tRNA-Phe, and tRNA-Tyr was reduced to 33, 47, 49, and 58% of the control value, respectively (Figures 5(a) and 5(b)). Thus, the nuclear run-on data corroborate the involvement of LmBdp1 in Pol III transcription. Signal of the tRNA-Sec, transcribed by both Pol II and Pol III, was diminished to 78% of the control value. Pol II transcription of -tubulin, LmjF.06.0200, LmjF.06.0210, and LmjF.06.0370 was slightly affected. Interestingly, 18S rRNA signal was reproducibly reduced to ∼55% of the control (Figures 5(a) and 5(b)). To further analyze this unexpected result, the 24S rRNA gene was included in new nuclear run-on experiments. As shown in Figures 5(c) and 5(d), similarly to the 18S rRNA, transcription of the 24S rRNA gene was reduced to 39% in the DKO+1 cell line. Transcription of the spliced-leader (SL) RNA was not affected (Figures 5(c) and 5(d)), supporting the previous results that indicate that LmBdp1 does not participate in Pol II transcription.

LmBdp1 Binds to Pol III Promoters.
To determine whether LmBdp1 associates to Pol III promoter regions in vivo, chromatin immunoprecipitation (ChIP) experiments were carried out, using the L. major cell line that expresses LmBdp1 fused to the PTP tag. Immunoprecipitations were conducted with a ChIP-grade anti-Prot A antibody, which recognizes the two Prot A domains present in the PTP tag [41], and with a nonspecific mouse immune serum as negative control. To assess the binding of LmBdp1-PTP to the L. major genome, qPCR assays were performed with the purified DNA. Data obtained from three independent experiments showed that LmBdp1 is enriched in the 5S rRNA, tRNA-Ala, tRNA-Met, and the promoter regions from the U2 and U4 snRNAs (tRNA-like sequences) ( Figure 6). Enrichment was also detected within the U2 snRNA gene, which contains an internal promoter element [25]. Thus, these results confirm the binding of LmBdp1 to Pol III promoters. As anticipated, enrichment was not observed with the -tubulin gene and the SL RNA promoter region ( Figure 6). Notably, we did not find association of LmBdp1 with the promoter region of the rRNA transcription unit, which suggests that the reduction in 18S and 24S rRNA transcription that was observed in the LmBdp1 DKO+1 cell line ( Figure 5) is an indirect effect.

Discussion
In this work, we characterized the orthologue of the Bdp1 subunit of transcription factor TFIIIB in L. major.  long arm, and the tether region. LmBdp1 possesses the characteristic extended SANT domain, predicted to fold into five -helices (Figure 1). Similarly to Bdp1 from yeast, a sixth -helix that comprises the long arm is present in LmBdp1 (Supplementary Figure 1). Interestingly, the long arm is predicted to be present in all the Bdp1 orthologues that we analyzed, with the exception of the human protein (Supplementary Figure 1) [10]. In yeast, the long arm forms a coiled-coil structure with Brf1 homology domain II and ends next to winged helix domains 2 and 3 from Pol III subunit C34 [11]. The N-linker region is characterized by the presence of 3-6 aromatic amino acids (W, F, and Y) (Figure 1(a)). However, only the conserved W that is located at the end of the N-linker is present in LmBdp1 (Figure 1(a)). Since the missing aromatic residues anchor the N-linker to the major groove of the DNA through aromatic-sugar interactions with the DNA backbone [11], a different type of contacts should occur between the LmBdp1 N-linker and the promoter DNA. The Bdp1 tether, involved in interactions with some Pol III subunits, is folded into several -sheet structures in S. cerevisiae, H. sapiens, Schizosaccharomyces pombe, and D. melanogaster (Supplementary Figure 1) [11,12]. Nonetheless, in L. major and other trypanosomatid parasites this region is not predicted to fold into -sheets (Supplementary Figure 1). Thus, LmBdp1 shares several characteristics with other Bdp1 orthologues, but it also shows some distinctive features.
Our data demonstrate that LmBdp1 participates in Pol III transcription, as the DKO+1 cell line showed reduced levels of tRNAs, 5S rRNA, and U2 snRNA in nuclear runon experiments ( Figure 5). The association of LmBdp1 with genes transcribed by Pol III was demonstrated by ChIP analysis ( Figure 6). While nuclear run-on assays exhibited a reduction in the transcription levels of 18S and 24S rRNAs, ChIP analysis did not reveal the binding of LmBdp1 to the promoter region of the ribosomal transcription unit. Thus, the observed decrease in 18S and 24S rRNA transcription is most likely a secondary effect caused by the reduction of 5S rRNA, as cells coordinate the expression levels of all rRNA species to regulate ribosome biogenesis [42]. The target of rapamycin (TOR) signal-transduction pathway is a key regulator of this process [43]. Interestingly, while most eukaryotes possess one or two TOR kinases, L. major and other trypanosomatids have three different TOR kinases [44] that might coordinate transcription by all nuclear RNA polymerases.
We were unable to generate null mutants of LmBdp1, as the DKO+1 cell line contained an extra copy of LmBdp1 ( Figure 3). It is possible that this additional copy was generated while trying to delete the second endogenous gene of LmBdp1, as attempts to knockout essential genes in Leishmania commonly produce alteration in gene copy number, either by gene amplification or changes in ploidy [33,[45][46][47]. Thus, this result strongly suggests that LmBdp1 is essential for the growth of L. major promastigotes, as has been reported in yeast [17]. Regardless of the moment when the extra LmBdp1 copy was produced, the growth of the DKO+1 cell line is strongly impaired and the expression of LmBdp1 is reduced by 70% (Figure 4). We tried to restore the expression levels of LmBdp1 in the DKO+1 cell line by transfecting it with the pLmBdp1-PTP vector (data not shown). However, several attempts to obtain transfected parasites were unsuccessful due to the inability of the DKO+1 cell line to reach the optimal cell densities for electroporation and to tolerate the drug selection process. As an alternative approach, we transfected the LmBdp1 single-knockout parasites with the pLmBdp1-PTP vector and then tried to knockout the second LmBdp1 allele (data not shown). Nevertheless, we were not capable of deleting the second endogenous copy of LmBdp1, which suggests that the expression of the recombinant LmBdp1-PTP protein in the resultant cell line was not high enough to allow the elimination of the second allelic copy of LmBdp1.
The function of Bdp1 is regulated by phosphorylation of specific amino acids. In logarithmically growing yeast cells, four Bdp1 residues (S49, S164, S178, and S586) are phosphorylated by PKA, Sch9, and CK2 kinases; and Bdp1 is dephosphorylated under conditions that reduce Pol III transcription [48]. Human Bdp1, by contrast, is phosphorylated by CK2 during mitosis to inhibit U6 snRNA transcription [49]. An in silico search allowed us to identify several potential phosphorylation sites in LmBdp1, including T308 (PhosTryp score of 0.933), S129 (score of 0.905), and S327 (score of 0.841) (Supplementary Figure 4). Notably, PKA [50] and CK2 [51] are present in Leishmania. Thus, similarly to yeast and human, the activity of LmBdp1 could be controlled by phosphorylation. This is supported by the fact that Western blot analysis of LmBdp1 often showed two or more bands (Figure 4(b) and Supplementary Figure 5) that might correspond to different phosphorylation states of the protein.
Besides its role in Pol III transcription initiation, Bdp1 is also involved in maturation of tRNAs by interacting with RNAse P, the enzyme required for the site-specific cleavage of the 5 leader sequence of precursor tRNAs [17]. Also, Bdp1 participates in the integration of Ty1, a long terminal repeat retrotransposon, upstream of tRNA genes in S. cerevisiae [52]. Moreover, Bdp1 also interacts with Hmt1, a protein arginine methyltransferase from yeast that inhibits tRNA transcription by methylating an unknown factor of the Pol III complex [53]. Bdp1 seems to also participate in Pol II transcription termination of noncoding RNAs in the vicinity of tRNA genes [54]. It remains to be determined whether Bdp1 also performs multiple functions in L. major and other trypanosomatids.

Conclusions
In the present study, we show that LmBdp1 shares several sequence and structural features with Bdp1 orthologues from other species, such as the presence of the extended SANT domain and the long arm. But some differences were also observed, including the occurrence of only one aromatic residue in the N-linker and the lack of predicted -sheet structures in the tether region of LmBdp1. Our data also demonstrate that LmBdp1 localizes to the nucleus, where it regulates the Pol III transcription of tRNAs, 5S rRNA, and U2 snRNA by associating with their promoter regions. Targeted gene replacement analysis strongly suggests that LmBdp1 is essential for the growth of promastigotes of L. major. Thus, LmBdp1 could be considered a suitable candidate for drug development against Leishmania.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
The authors have no conflicts of interest to declare.