Profilin is involved in G1 to S phase progression and mitotic spindle orientation during Leishmania donovani cell division cycle

Profilin is a multi-ligand binding protein, which is a key regulator of actin dynamics and involved in regulating several cellular functions. It is present in all eukaryotes, including trypanosomatids such as Leishmania. However, not much is known about its functions in these organisms. Our earlier studies have shown that Leishmania parasites express a single homologue of profilin (LdPfn) that binds actin, phosphoinositides and poly- L- proline motives, and depletion of its intracellular pool to 50%of normal levels affects the cell growth and intracellular trafficking. Here, we show, employing affinity pull-down and mass spectroscopy, that LdPfn interacted with a large number of proteins, including those involved in mRNA processing and protein translation initiation, such as eIF4A1. Further, we reveal, using mRNA Seq analysis, that depletion of LdPfn in Leishmania cells (LdPfn+/-) resulted in significantly reduced expression of genes which encode proteins involved in cell cycle regulation, mRNA translation initiation, nucleosides and amino acids transport. In addition, we show that in LdPfn+/- cells, cellular levels of eIF4A1 protein were significantly decreased, and during their cell division cycle, G1-to-S phase progression was delayed and orientation of mitotic spindle altered. These changes were, however, reversed to normal by episomal expression of GFP-LdPfn in LdPfn+/- cells. Taken together, our results indicate that profilin is involved in regulation of G1-to-S phase progression and mitotic spindle orientation in Leishmania cell cycle, perhaps through its interaction with elF4A1 protein.


Introduction
Profilin is a key regulator of actin dynamics that plays a central role in almost all vital cellular processes, including endocytosis, motility, signal transduction, metabolism, cell division, etc. [1][2][3]. It is a 16kDa actin-binding protein that has been recognized as one of the crucial proteins for cell survival [4]. Besides binding to actin, profilin also binds to a large array of actinbinding membrane proteins that contain poly-L-proline (PLP) motives in their structure, and

Pull-down assay and mass spectrometry
Full-Length profilin was cloned and expressed with GST-tag (GST-LdPfn), as described earlier [11]. For the pull-down assay, purified recombinant GST-LdPfn or GST alone was incubated with 100μL of Glutathione-Sepharose affinity beads for 2 hours. The unbound proteins were removed by centrifugation at 1000 x g for 10 minutes at 4˚C and the recombinant proteins associated with the beads were incubated with 2mg protein of clear lysate of L. donovani promastigotes (clear supernatant obtained after sonication and centrifugation) overnight at 4˚C. The next day, unbound proteins were removed, and the beads were washed five times with 1x PBS. Finally, the bound proteins were eluted by boiling the beads at 96˚C for 5 min in 50ul of SDS-poly acrylamide gel electrophoresis (SDS-PAGE) loading buffer (250mM Tris, 10% SDS, 0.5% bromophenol blue, 50% glycerol, and 500mM 2-mercaptoethanol). Eluates were run on 12% polyacrylamide gel. Three biological replicates of each of GST and GST-LdPfn pull-down samples were stained with silver nitrate, the gel was rinsed thrice with double distilled water; the respective lanes were excised, kept in labelled Eppendorf tubes submerged with double distilled water, and submitted at the mass spectrometry facility (Institute for Stem Cell Regeneration and Medicine, NCBS-TIFR Campus, Bellary Road, Bangalore Karnataka, India) for analysis, using LC-MS (Orbitrap Fusion™ Tribrid™ Mass Spectrometer). Only the protein hits with a score cut off 25, significance threshold p<0.05 were considered. The data thus-obtained were used for searching the www.tritrypdb.org database (version 51), using the Leishmania donovani (BPK282A1) strain as reference. Data has been deposited at PRIDE-ProteomeXchange Consortium [12] and can be accessed with the identifier: PXD026036. Among the three biological replicates, the proteins present in all three replicates or at least two replicates of GST-LdPfn, but not present in any of the replicates of GST control, were taken into consideration.

Total RNA isolation and library construction
Total RNA from three independent biological replicates was isolated from LdPfn +/+ (control) and LdPfn +/promastigotes using TRIzol reagent (Ambion, Life Technologies), according to the manufacturer's instructions. RNA samples were treated with DNase I (2μg) and the RNA concentration was determined, using a spectrophotometer at A260/280 (Nanodrop ND1000, Thermo Scientific, USA). In addition, the RNA integrity was evaluated using TapeStation (Agilent) and Qubit (Invitrogen). Using NEBNextUltra TM II RNA Library Prep Kit for Illumina, Poly (A) mRNA magnetic isolation was performed. Library preparations were carried out, using the NEBNextUltra TM II RNA Library Prep kit (Illumina), according to the manufacturer's instructions.

RNA-Seq and data analysis
Paired-end reads (2 x 150 bp) were obtained using the Illumina HiSeq 2500 platform at the Bio-IT centre at the Institute of Bioinformatics and Applied Biotechnology. Raw data were generated for each of the libraries from 6 samples. The quality of the produced data was analysed using FastQC by Phred quality score. Reads with Phred quality scores lower than 20 were discarded. Reads were aligned to the L, donovani (BPK282A1) genomic data obtained from TriTrypDB version 51 (www.tritrypdb.org), using Bowtie2 (-x option) [13,14]. All analyses were carried out using the Tophat pipeline with the following versions: Tophatv2.1.1, Bowtie2 v2.3.51 [15]. Tophat is a fast splice junction mapper for RNA-Seq reads. It aligns RNA-Seq reads to genomes, using the ultra-high throughput short read aligner Bowtie and then analyses the mapping results to give the transcript counts. The gene expression level values were calculated from the transcript counts. Gffread tool [16] was used to convert gff files to gtf files for read-count calculation using HTSeq. The HTSeq version 0.12.4 (htseq-count -f option) was used to count the number of reads aligned to protein-coding genes. HTSeq [17,18] is a Python package that gives the infrastructure to processed data from high-throughput sequencing assays. HTSeq calculates the number of mapped reads to each gene. DeSeq tool was used for differential gene expression analysis between samples in protein-coding genes. DeSeq [18,19] is an R package to measure variance-mean dependence in count data from high-throughput RNA-Seq assays and test for differential expression depending on a model applying the negative binomial distribution. Differentially expressed (DE) genes were classified as genes with a Benjamini-Hochberg multiple testing p-value of <0.05. Principal component analysis plot (PCA plot), clustered heat map, and volcano plots were created using base R plot PCA function (or base R prcomp function), gplotsheatmap.2 functions and ggplot function of gglplot2 package respectively. Functional annotation was performed using GO (gene ontology) and the Kyoto Encyclopaedia of Genes and Genomes (KEGG) and Metabolic Pathways from all Domains of Life (MetaCyc) using the L.donovani (BPK282A1) GO annotations provided in TriTrypDB.

Real time quantitative polymerase chain reaction (RT-qPCR) validation assays
DNase-treated RNA (2μg) was reverse transcribed with MMLV reverse transcriptase (NEB). Equal amounts of cDNA were assessed in triplicate in a total volume of 10ul containing SYBR Green (Kapa Biosystems) and the primers used were given in S1 Table. The mixture was incubated at 95˚C for 20 sec, 55˚C for 30 sec, and 72˚C for 1 sec for 40 cycles. Negative controls were included in the RT-qPCR assays to detect DNA contamination in RNA samples. The fold-change in expression of up and down regulated genes was determined, using RT-qPCR. Reactions were carried out on a fast RT-qPCR system (Applied Biosystems). The results were quantified by the delta-delta CT method with actin as a reference gene and β-tubulin as an endogenous control to normalize each sample. The specificity of the reaction was verified by melt curve analysis. All the experiments were conducted at least three times and the results were expressed as mean ± SEM of three independent experiments.
permeabilized, using 0.5% (v/v) Triton X-100 for 15 minutes, and washed with PBS-glycine. The washed cells were blocked with 3% bovine serum albumin in PBS for 2 hours at 25˚C and labelled with anti-LdPfn antibodies [11] (1:100) and or anti-tubulin antibodies (Santacruz biotechnologies cat no. sc-5286 and Sigma cat.no. T7816) (1: 500) overnight at 4˚C. The labelled cells were washed with 0.5% bovine serum albumin in PBS to remove non-specifically bound antibodies and again labelled with Alexa Fluor 568 -conjugated goat anti-mouse IgG (Thermo Fisher Scientific, cat no. A21043) or Alexa Fluor 488-conjugated goat anti-rabbit IgG (Thermo Fisher Scientific, cat no. A32731) or Fluor 488-conjugated goat anti-mouse IgG (Thermo Fisher Scientific, cat no.A32723) or Alexa Fluor 568-conjugated goat anti-rabbit IgG (Thermo Fisher Scientific, cat no.A21069) secondary antibodies depending on the experiment. The coverslips were mounted using prolong diamond anti-fade mounting media containing 4,6-diamidino-2-phenylindole (DAPI, Invitrogen). To analyse the nuclear division patterns, the coverslips were treated with RNase (5mg/ml) followed by labelling with propidium iodide (PI). Images were captured on Nikon laser scanning confocal microscope c2 using a 100X1.4 NA (oil) plan apochromatic lens.

Cell cycle analysis
For cell cycle analysis, Leishmania cultures (5 x 10 7 cells) were synchronized by incubating them with 200 mg ml -1 of N-hydroxyurea (HU) (Sigma) overnight (12-14 hours). The cells were washed and then re-suspended in DMEM media containing 10% FCS without HU. Aliquots of cell suspension were drawn at 0 hour to 10 hours at regular time intervals of 2 hours. The cells were washed with cold PBS and resuspended in 50μl of PBS. The cell suspension was mixed with 150μl of fixative solution (1% Triton X-100, 40mM citric acid, 20mM sodium phosphate, 200mM sucrose) and incubated at 25˚C for 5 minutes. After incubation, 350μl of diluent buffer (125mM MgCl 2 in PBS) was added to it and stored at 4˚C until further use. Before the cell cycle analysis, the fixed cell suspension was treated with 50μg RNase (5mg ml -1 in 0.2M sodium phosphate buffer, pH7.0) for 2 hours at 37˚C and then incubated with 50μg PI (5mg ml -1 in 1.12% sodium citrate) for 30 minutes at 25˚C. The samples were analysed in Gallios flow cytometer (Beckman coulter) and proportions of the G1, S, and G2M populations were determined using ModFit LT software (Verity Software House, Topsham, ME, USA).

Statistical analysis
All the experiments were conducted at least three times and the results were expressed as the standard error of the mean (mean ± SEM) of three experiments. The data were statistically analysed by ANOVA test with replication. A p-value of <0.05 was considered significant.

Leishmania profilin interacts with a large number of cellular ligands, including elF4A1
To map the ligand binding profile of LdPfn, we performed a pull-down assay in lysates of midlog phase Leishmania promastigotes, using glutathione S-transferase (GST)-tagged LdPfn [11] and GST protein alone (negative control), as a bait, as described in 'Materials and Methods'. LdPfn bound to its ligands was isolated by using glutathione-Sepharose affinity beads. The beads were washed thoroughly to remove unbound proteins and the bound proteins were eluted by boiling the beads in SDS-polyacrylamide gel electrophoresis (PAGE) sample loading buffer. The eluates from three independent pull-down experiments of both GST-LdPfn and GST-alone were resolved on SDS-PAGE. The gels were silver-stained, excised, and subjected to trypsin digestion followed by liquid chromatography-mass spectrometry (LC-MS) analysis ( Fig 1A). The proteins present in all three replicates (4 proteins) and those present in at least two replicates (34 proteins) of GST-LdPfn pull-down, but not present in any one of the GSTalone pull-down controls are listed in Table 1. A summary of each protein hit is given in S2 Table. The LdPfn ligands thus detected were classified according to their deduced function or cell location, and the percent of proteins representing each group are shown in Fig 1B. Some of the LdPfn ligands, such as actin (LdBPK_041250.1), valosin-containing protein (LDBPK_361420.1), mitochondrial outer membrane protein porin (LDBPK_020430.1), and eIF4A1 (LdBPK_010790.1), have earlier been identified as the potential ligands also of the T. cruzi profilin [8]. Interestingly, unlike T. cruzi [8], we have also identified gamma-tubulin complex component 3-like protein (LdBPK_362370.1), a key component of the microtubule-organizing centre [20], as one of the potential ligands of LdPfn. While valosin-containing protein VCP/p97 has been shown to be essential for the intracellular development of Leishmania [21], T. brucei porin (Tb927.2.2510) has been reported to be the main metabolite channel in the mitochondrial outer membrane and is required to support efficient oxidative phosphorylation [22]. Further, the eukaryotic initiation factor 4A1 (eIF4A1) is abundantly present in Leishmania cytoplasm [23], and this protein has been suggested to be the main translation initiation factor involved in protein synthesis in T. brucei [24]. To further confirm the presence of this protein in the LdPfn interactome, we analysed the pull-down eluates from GST-LdPfn and GST-alone by western blotting, using anti-eIF4A1 antibodies. Results given in Fig 1C and S1 Fig show that eIF4A1 (45.3kDa) was present in the input lysates used in both the GST-LdPfn and GST-alone pull-down assays and also in the GST-LdPfn pull-down eluates and pass-through fraction from the GST-alone pull-down assay, but it was completely absent in the pass-through fraction from GST-LdPfn pull-down and GST-alone eluate. These results strongly indicate that LdPfn interacts with multiple cellular ligands, a majority of which constitute the proteins that may play an important role in regulating mRNA processing, translation initiation, cell metabolism and mitochondrial functions.
As LdPfn appeared to interact with proteins involved in mRNA processing (LDBPK_353150.1 and LdBPK_322350.1) and protein translation initiation (elF4A1), we performed transcriptomic analysis of mRNA isolated from mid-log phase LdPfn +/+ and LdPfn +/cells, employing high throughput RNA sequencing technique (RNA-Seq), to identify differentially expressed genes in LdPfn +/cells.
Transcriptomic data analysis revealed that the genes involved in DNA transcription, mRNA translation, cell cycle regulation, membrane transport, and mitochondrial activity were differentially expressed in LdPfn depleted cells L. donovani (BPK282A1) contains 36 chromosomes and has a haploid genome size of 32.44 Mb, which encodes a total of 8135 genes (8023 protein-coding genes and 112 non-proteincoding genes) [25]. These parasites display a unique way of controlling their gene expression in that the mRNAs are made from polycistronic precursors by SL-trans splicing and polyadenylation [26][27][28]. Many protein-coding genes of unrelated functions are arrayed in long clusters on the same DNA strand. Intergenic regions of polycistronic pre-mRNAs are cotranscriptionally processed by two reactions: polyadenylation of the upstream gene and trans-  splicing of the capped mini exon to the downstream gene [29], thus generating monocistronic units ready for degradation or translation. It is presumed that all polycistronic precursor RNAs are transcribed approximately at the same rate. As a consequence, the regulation of gene expression occurs most entirely post-transcriptionally [30]. To identify differentially expressed genes in profilin depleted Leishmania promastigotes, we analysed the transcriptomes of both the mid-log phase LdPfn +/+ and LdPfn +/-Leishmania promastigotes, using high-throughput RNA sequencing technique (RNA-Seq). Total RNA from three independent biological replicates of each LdPfn +/+ and LdPfn +/promastigotes was extracted and after library preparation, RNA sequencing was carried out on Illumina HiSeq 2500. The RNA-Seq datasets generated in the analysis with each sample have at least 30 million read pairs (S2 Fig and S3 Table). Principal Component Analysis (PCA) was used to analyse the relationship between the samples (S3 Fig), showing a clear separation between LdPfn +/+ and LdPfn +/samples. RNA-Seq data were aligned to the L. donovani genome (L. donovani BPK282A1, NCBI taxon ID: 981087). 8135 transcripts were identified in the data set, which correlated with the data reported earlier [25]. RNA-Seq data revealed that 254 genes were differentially expressed (DE) having an adjusted p-value <0.05. Based on the DE genes, volcano plots were generated (Fig 2A), showing distribution of the transcripts by comparing the fold change in the expression (log 2 ) of each group with the corresponding adjusted p-value (-log 10 ). Among these DE genes, 112 genes (~44%) were hypothetical with unknown functions. The complete list of differentially expressed (DE) genes with known functions either upregulated or downregulated and those with unknown functions along with their respective transcript IDs and log2 fold change are listed in S4 Table. Further, a clustered heat map was generated with the top 30 up and top 30 down-regulated genes to evaluate the reproducibility of the biological replicates ( Fig 2B). The data obtained from RNA-Seq was validated by RT-qPCR of 11 genes, 5 up-regulated and 6 down-regulated including profilin transcript (Fig 2C). Results of the RT-qPCR analysis showed a strong correlation with the RNA-Seq data, thus validating the RNA-Seq results ( Fig  2C).
To evaluate the probable effects of differentially expressed genes in LdPfn +/cells, we performed gene ontology (GO) analysis of 254 DEGs to identify the cellular component, molecular functions, and biological processes enriched due to profilin depletion. Fig 3A shows that the transcripts encoding proteins were mainly localized to the nuclear component, mitochondrial complex, and TORC2 complex. The molecular functions (Fig 3B) enriched by these transcripts were mainly predicted to include the transmembrane transporter activity, phosphatidylinositol binding, kinase activities, and ubiquitin-protein transferase activity. Biological processes that were enriched due to depletion in the LdPfn levels are shown in Fig 3C. From RNA-Seq data, we identified some of the genes enriched in GO biological processes and then validated their mRNA expressions by RT-qPCR. Fig 3D shows   findings from the RNA-Seq data (with log 2-fold change, cut off above +0.8 for upregulated and below -0.8 for down regulated genes) were grouped based on their cellular activities ( Table 2) that could have been affected due to LdPfn depletion. Results given in Table 2 clearly indicate that expression of genes that are involved in regulation of mitochondrial activity  The Hsp60/10 complex is believed to be responsible for accelerating the folding of polypeptides imported into mitochondria, as well as reactivation of denatured proteins, and diminishing aggregation of non-native polypeptides and partially unfolded kinetically trapped intermediates [82] (Continued ) (such as ATOM14, ATOM11), cell division (such as CYC2-like cyclin, cell division cycle protein 20 (CDC20) and kinesin 13.1), DNA transcription (such as histone-lysine N-methyltransferase), mRNA translation (such as eIF4A1), and amino acids and nucleosides transport was significantly reduced in LdPfn depleted cells.

Depletion of intracellular pool of LdPfn results in significantly reduced levels of eIF4A1 protein in LdPfn +/cells
The translation initiation factor 4A1 (eIF4A1), a DEAD-box RNA helicase, is a component of the translation initiation complex eIF4F, which binds to the cap structure of eukaryotic mRNA and helps in recruiting the small ribosomal subunit. As elF4A1 is a component of the LdPfn interactome ( Table 1) and expression of its gene is significantly down-regulated in the LdPfn +/cells (Table 2), we examined whether the intracellular levels of elF4A1 protein have also been affected in these cells. For this, we probed the SDS-electrophoretograms of lysates of mid-log phase LdPfn +/+ , LdPfn gene complemented LdPfn +/cells (LdPfn +/-comp ) and LdPfn +/cells by western blotting, employing anti-LdPfn and anti-eIF4A1 antibodies (Fig 4A and S1  Fig). Results presented in Fig 4B clearly show that depletion of the intracellular pool of LdPfn results in about 50% reduction in cellular levels of eIF4A1 (Fig 4B) in Leishmania promastigotes. That the decreased expression of elF4A1 is caused due to depletion of LdPfn in Leishmania cells, was confirmed by episomal expression of GFP-LdPfn gene in the LdPfn +/cells (Fig 4A), which restored the elF4A1 levels to normal.

PLOS ONE
As cellular levels of key cell cycle proteins are strictly controlled by tightly regulated synthesis of such proteins during the G1-phase of the cell cycle [31][32][33], it may be inferred that the reduced expression of translation initiation factor 4A1 in LdPfn +/cells may adversely affect protein synthesis and consequently the cell division cycle in LdPfn +/cells. To confirm this conclusion, we analysed the cell division cycle in LdPfn +/+ , LdPfn +/-comp and LdPfn +/cells, using flowcytometry and immunofluorescence microscopy.

LdPfn is involved in regulation of the G1-to-S phase progression
As the gene expression in trypanosomatids is largely controlled at the post-transcriptional level, the main control points in Leishmania gene expression should therefore be mRNA degradation and translation [34]. This means that, the poor mRNA processing and translation would reflect poor gene expression in LdPfn +/cells. To test this possibility, we analysed the cell division cycle in LdPfn +/-, LdPfn +/+ and LdPfn +/-comp cells. The cells were synchronized by treating them for about 12 hours with HU. Approximately 70% of the cells were synchronized at the G1/S border by HU treatment. After releasing the HU block, the cells were stained with PI to probe the total DNA content and then the samples were processed for the cell cycle analysis by flow cytometry. The LdPfn +/+ and LdPfn +/-comp cells immediately entered the S-phase attaining the peak at 2 hours, and the G2/M peak was achieved at 4 hours after releasing the HU block. However, the LdPfn +/cells lagged in entering from the G1-to S-phase by at least 1 hour and remained mostly in the S-phase up to 6 hours (Fig 5A, S4A Fig). These results clearly revealed that profilin plays an important role in G1-to-S phase progression in Leishmania cell cycle. To further analyse this finding, we performed quantitative flow cytometric analysis with BrdU incorporation in LdPfn +/+ , LdPfn +/and LdPfn +/-comp cells to determine the number of cells present in different phases of cell cycle at different time points, after removal of the HU block. There were significantly higher number of LdPfn +/cells (71.7%), as compared to LdPfn +/+ (53.1%) or LdPfn +/-comp cells (59.5%), in the G1-phase at 2 hours after releasing the HU block, whereas at the same time point, considerably lesser number of LdPfn +/cells (6.0%), compared to LdPfn +/+ (30.3%) or LdPfn +/-comp (24.1%) cells, were present in the Sphase. Similarly, the number of LdPfn +/cells in the G1-phase was much higher (61.4%) than that of the LdPfn +/+ (31.9%) or LdPfn +/-comp cells (25.6%) at 4 hours after release of the HU block, while only 24.6% LdPfn +/cells, compared to 53.9% LdPfn +/+ and 56.2% LdPfn +/-comp cells, were in the S-phase at the same time period (Fig 5B and 5C).To rule out the possibility of toxic effects of HU on Leishmania promastigotes during synchronization conditions, we performed cytometric assay with BrdU incorporation in asynchronously growing mid-log phase Leishmania cells, without HU treatment, at 2 and 4 hours. In these conditions also, a significantly lesser number of LdPfn +/cells, compared to LdPfn +/+ and LdPfn +/-comp cells, transited from the G1-to-S phase at both the time points (S4B and S4C Fig). These results demonstrate that LdPfn is involved in regulation of G1-to-S phase progression during Leishmania cell division cycle.

LdPfn is involved in regulation of the mitotic spindle orientation
Kinesin 13-1 has been shown to be exclusively an intranuclear protein that regulates spindle assembly during mitosis, and its depletion leads to abnormalities in spindle structure and nuclear division [35,36]. As shown in Table 2, depletion of LdPfn in Leishmania cells resulted in a significant down regulation of expression of the kinesin 13-1 transcript, we considered it of interest to investigate the segregation pattern of the nucleus during mitosis (Fig 6A). For this, the cells collected from the mitotic phase were labelled with anti-tubulin antibodies, anti-LdPfn antibodies, and DAPI or PI and then examined under the fluorescence microscope.

Fig 5. Retardation of cell cycle progression in LdPfn +/cells. (A)
Representative flow cytometry data of LdPfn +/+ , LdPfn +/and LdPfn +/-comp cells. The samples were collected, after releasing hydroxyurea (HU) block, at 2 hours interval for up to 10 hours. 20,000 events were analysed at every time-point. Three independent experiments were performed, and one representative dataset is shown here. G1 (first red peak), S (grey peak) and G2/M (second red peak) phases are indicated in the histogram itself along with percent of cells in each phase. LdPfn +/+ cells entered into S-phase at 2 hours after release of HU block. However, transition of LdPfn +/cells from G1-to S-phase was considerably delayed, compared to LdPfn +/+ and LdPfn +/-comp cells. (B) Representative flow cytometry data with BrdU incorporation in LdPfn +/+ , LdPfn +/and LdPfn +/-comp cells. The cells were collected, after releasing the HU block, at 2 hours interval for up to 10 hours, and then labelled with anti-BrdU antibodies, as described in 'Materials and Methods'. 10,000 events were analysed at every time-point. Three independent experiments were performed, and one representative dataset of 2 hours and 4 hours is shown. G1, S and G2/M phases are indicated in the histogram along with the percent of cells in each phase. In LdPfn +/cells, a significantly lesser number of cells (6.01% and 24.61%, respectively, at 2 hours and 4 hours after releasing HU block) exhibited BrdU incorporation in S-phase, as compared to LdPfn +/+ cells (30.26% and 53.93%, respectively, at 2 hours and 4 hours after releasing HU block), and LdPfn +/-comp cells (24.08% and 56.23%, respectively at 2 hours and 4 hours after releasing HU block). (C) Bar diagram showing considerably lesser number of BrdU labelled LdPfn +/cells (green bar) in S-phase at 2 hours and 4 hours after releasing the HU block, compared to LdPfn +/+ cells (red bar) and LdPfn +/-comp cells (blue bar) at the same time points. p-value ��� <0.001 at both 2 hours and 4hours after releasing the HU block.
https://doi.org/10.1371/journal.pone.0265692.g005 The cells were stained with anti-tubulin (red) and anti-LdPfn (green) antibodies and DAPI (blue). Analysis of dividing cells revealed that the plane of the nuclear division during karyokinesis in the LdPfn +/+ and LdPfn +/-comp cells was positioned parallel to the flagellar base, leading to the formation of laterally arranged spindle between the two dividing nuclei. In contrast, in the LdPfn +/cells the dividing nuclei were arranged nearly perpendicular to the flagellar base, leading to a longitudinally formed mitotic spindle. Scale: 2μm; F, Flagellar base; K, Kinetoplast; N, Nucleus. (C) (a) The cells were alternately labelled with tubulin (green) and the nucleus and the kinetoplast with propidium iodide (red). In this case also, the nuclei division plane orientation in dividing LdPfn +/cells was altered, as compared to dividing LdPfn +/+ cells. Scale: 2μm. Results revealed that the plane of the nuclear division during karyokinesis in the LdPfn +/+ cells was positioned parallel to the flagellar base, leading to the formation of a laterally arranged spindle between the two dividing nuclei. In contrast, in about 70% LdPfn +/cells, the dividing nuclei were arranged nearly perpendicular to the flagellar base, leading to a longitudinally formed mitotic spindle (Fig 6B). However, the division pattern in LdPfn +/-comp cells was similar to that of the control LdPfn +/+ cells, indicating that the defect in the arrangement of dividing nuclei and orientation of spindle in LdPfn +/cells was a specific defect due to depletion in intracellular levels of profilin. To further confirm, we labelled the cells with ant-tubulin antibodies (green) and their nucleus and kinetoplast with PI (red) and then analysed them under a fluorescent microscope (Fig 6C). Quantification of the spindle positioning (lateral or longitudinal) in the dividing cells revealed that about 66% of LdPfn +/cells (n = 140) possessed nearly longitudinally positioned spindle, while only 11% of LdPfn +/+ cells (n = 152) and 10% of LdPfn +/-comp cells (n = 146) showed such aberration (Fig 6C). These results demonstrate that profilin is involved in regulation of the mitotic spindle orientation and nucleus positioning during the mitotic phase in dividing Leishmania cells. However, quantification of the percent of dividing cells with division furrow (Fig 6D) showed no significant difference between the two cell types, suggesting cytokinesis has not been significantly affected.

Discussion
Profilin is a ubiquitous actin-binding protein present in all eukaryotic cells, including unicellular eukaryotic organisms, such as Plasmodium, Toxoplasma, Acanthamoeba Dictyostelium, Tetrahymena, etc. It plays an important role in a number of cellular activities such as cell growth, intracellular trafficking, gene transcription, cell division, cell signalling, etc. [1][2][3][4][5][6]. Profilin has been suggested to be vital for the invasive blood stage Plasmodium falciparum and its complete deletion has been shown to have severe effects on viability of the parasites [37,38]. In related parasite Toxoplasma gondii, this protein is known to be involved in cell motility, host cell invasion, immune evasion, and virulence [39]. Further, Acanthamoeba castellanii contains three profilin isoforms, each one of which exhibits different subcellular localization, diverse ligand binding properties and biological functions [40]. Furthermore, Dictyostelium amoebae lacking profilin isoforms I and II increase in cell size by up to 10 times, their motility is adversely affected, and a wide ring of filamentous actin accumulates beneath the plasma membrane, which blocks their development [41]. Besides this, in Tetrahymena thermophila, loss of profilin affected nuclear positioning, stomatogenesis, and cytokinesis [2]. Results of the present study clearly reveal that profilin is involved in regulation of G1-to-S phase progression and mitotic spindle orientation during Leishmania cell division cycle.
Cell division cycle may be visualised as a developmental process wherein a cell duplicates itself in to two progeny cells. During this process the cell first grows in size, replicates its chromosomes, segregates a full set of chromosomes to each of two new nuclei, and then divides into daughter cells [32]. In eukaryotes, the cell cycle comprises four discrete phases: G1, S, G2, and M. During the G 1 phase, the cell congregates the building blocks of chromosomal DNA and the associated proteins, and also sufficient energy reserves for completing the replication process of each chromosome in the nucleus. Whereas in the S phase, DNA replication occurs to form identical pairs of DNA molecules. The building blocks of proteins and nucleic acids, namely amino acids and nucleobases/ nucleosides, are largely imported by the Leishmania parasite from the host environment through the plasma membrane by specific transporter furrow in LdPfn +/cells (n = 118), as compared to LdPfn +/+ (n = 125) and LdPfn +/-comp (146) cells, indicating that the cytokinesis was not much affected by LdPfn depletion in Leishmania cells. https://doi.org/10.1371/journal.pone.0265692.g006 proteins [42][43][44]. As depletion of LdPfn intracellular pool in Leishmania cells resulted in downregulation of genes that encode amino acids and nucleosides transporters, it may severely limit availability of the DNA and protein building blocks in G1 phase, which, in turn, may result in slowing down the progression of the G1-to-S phase transition. This is strongly supported by our quantitative flow cytometry data, which showed that the number of cells transited from G1-to-S phase at different time periods considerably decreased in LdPfn depleted Leishmania promastigotes, Furthermore, it is consistent with our observation that expression of genes that encode CDC20 and CYC2-like cyclin, which are known to play a critical role in regulation of trypanosomatids cell cycle [45][46][47], is markedly reduced in LdPfn +/cells.
Purines and pyrimidines are basic building blocks of nucleic acids. Distinct from their mammalian and insect hosts, Leishmania parasite lacks the metabolic machinery to produce purine nucleotides de novo and mainly relies on the acquisition of preformed purine nucleobases and nucleosides through its plasma membrane-bound specific permeases from the host [43]. The first such permease, designated LmaNT3, was identified in L. major, which showed about 33% identity to L. donovani nucleoside transporter 1.1 (LdNT1.1), and is, thus, a member of the equilibrative nucleoside transporter (ENT) family [44], which seems to be key elements controlling nucleoside and nucleotide pool for DNA synthesis [48]. As expression levels of the genes that encode nucleoside transporter 1 (LdBPK_151260.1.1; LdBPK_151250.1.1) was considerably reduced in LdPfn +/cells, it may be inferred that the decreased availability of purine nucleobases and nucleotides would significantly affect the DNA and RNA synthesis, and consequently the DNA replication, during the 'S' phase of the cell cycle.
DNA replication in Leishmania is primarily determined by the active transcription [49]. In this group of organisms RNA polymerase II transcription is polycistronic and individual mRNA are excised by trans-splicing and polyadenylation [34]. The lack of individual gene transcription control is mainly compensated by post-transcriptional mechanisms, including tight translational control and regulation of mRNA stability/translatability by RNA-binding proteins [50]. As LdPfn interacts with the proteins that are involved in mRNA processing and translation initiation, these interactions would be adversely affected in LdPfn depleted cells, which in turn could affect the mRNA processing and translatability.
Translation of mRNA into proteins is initiated with the binding of the multimeric translation initiation complex elF4F to the cap structure of mRNA present at its 5' end. The elF4F complex is comprised of the cap binding protein, elF4E, the highly conserved mRNA helicase, elF4A, and the large scaffolding protein having binding sites for both elF4E and elF4A, elF4G is responsible for recruiting ribosomal subunits to the initiation codon of mRNA. Although several potential elF4F homologues have been identified in the L. major database, only three elF4E, two elF4A and one elF4G (elF4G3) have so far been characterized [23]. While elF4A1 and elF4E3 have been reported to be abundantly present, the other proteins are either moderately abundant or not detected in L. major promastigotes [23]. Similar to L. major, elF4A-1 is very abundant and predominantly cytoplasmic protein also in T. brucei, and its depletion to 10% of regular levels dramatically decreases protein synthesis one cell cycle following doublestranded RNA induction and blocks cell proliferation [24]. Based on these findings, it has been suggested that only the elF4A1 protein is involved in protein synthesis [24]. As the expression of elF4A1 was found to depend on the intracellular levels of LdPfn, the decreased availability of elF4A1 in LdPfn +/cells should affect the protein translation process.
The direction in which a cell divides is determined by the orientation of its mitotic spindle at metaphase. The spindle orientation in eukaryotes is controlled by a conserved biological machine that mediates a pulling force on astral microtubules. Constraining the localization of this machine to only certain regions of the cortex can thus determine the orientation of the mitotic spindle [51]. Kinesins and dyneins are microtubules-dependent motor proteins that transport cargo in opposite directions along microtubules. However, the kinesin 13 family of proteins in trypanosomatids do not possess cargo transporting property, but they do depolymerize microtubules at their ends, and thereby control microtubule length [36]. It has been shown that in kinesin 13 family of proteins, kinesin 13-1 is exclusively intranuclear in both T. brucei and L. major, where it mainly localizes to the mitotic spindle and spindle poles and plays a central role in regulating the spindle assembly during mitosis [35,36]. As expression of the gene that encodes kinesin 13-1 was significantly down regulated in the LdPfn depleted Leishmania cells, we speculate that LdPfn might have been involved in regulating the spindle orientation by controlling the expression of kinesin 13-1 protein in these cells.
Based on our proteomic and transcriptomic data analyses, it appears that besides having a role in cell cycle regulation, LdPfn may also be involved in regulating the Leishmania mitochondrial activity. Leishmania has a single large mitochondrion, which is distributed in branches under the subpellicular microtubules and in a specialized region, called kinetoplast, it houses its unusual genome, known as kDNA [52]. Mitochondria in eukaryotic cells are responsible for oxidative phosphorylation, and harness energy from numerous substrates through electron transport chains. They are semi-autonomous cell organelles which have their own DNA and protein synthesizing machinery [52,53]. However, Leishmania mitochondrion is completely devoid of tRNA-encoding genes and hence, it imports nucleus-encoded tRNAs for protein synthesis [54] through a receptor-mediated pathway [55,56]. Several mitochondrial proteins that are involved in the electron transfer mechanisms and also the outer membrane protein, porin, which is the main metabolite channel and is required to support efficient oxidative phosphorylation in T. brucei [22], interact with LdPfn in Leishmania promastigotes, it may be envisaged that activity of these proteins would be adversely affected by depleting intracellular pool of LdPfn. This is consistent with our finding that expression of transcripts that encode for proteins involved in mitochondrial electron transport activities, mitochondrial protein folding, and protein import across the mitochondrial outer membrane was significantly decreased in LdPfn +/cells. Further, protein import in trypanosomatid mitochondrion is mediated by the archaic translocase of the outer membrane (ATOM) complex, consisting of six subunits [57,58]. Among them, expression of transcripts encoding ATOM14 and ATOM11 was significantly reduced in the LdPfn +/cells. As knockdown of ATOM14 in T. brucei has been shown to result in growth arrest and drastically reduced protein and tRNA import into the mitochondrion [59] and ATOM11, which is exclusively present in trypanosomatids and is known to have a direct role in mitochondrial tRNA import [60], These results strongly suggest that LdPfn could play an important role in regulating the protein synthesis and oxidative phosphorylation in Leishmania mitochondrion.
Finally, our earlier studies have shown that LdPfn binds actin, PLP motif-containing proteins and membrane phosphoinositides, especially PI(3,5)P2, PI(4,5)P2 and PI(3,4,5)P3 [11]. As membrane phosphoinositides play a central role in regulation of cell signalling, membrane trafficking, actin remodelling, nuclear events and intracellular transport [61][62][63], depletion of LdPfn intracellular levels should affect LdPfn interactions with these lipids, which in turn should affect phosphoinositide metabolism and consequently their cellular functions, including regulation of actin dynamics, cellular metabolism and membrane signalling. This is well supported by our present and earlier findings that depletion of LdPfn in Leishmania cells not only affected the intracellular transport [11], but it appeared to affect also the expression of proteins that regulate cell signalling, membrane transport, and actin remodelling. It has earlier been suggested that cortical actin remodelling plays a key role during cytokinesis and early mitosis in eukaryotic cells [64]. However, in Leishmania cells, the role of actin dynamics has been shown only in the basal body and kinetoplast separation, cleavage furrow progression and flagellar pocket division [65]. Although no such aberrations were observed during cytokinesis of the LdPfn depleted cells, based on the presently available data, we cannot completely rule out the role of actin remodelling in cell division cycle of these cells.
Supporting information S1 Table. Primer list used for performing quantitative real-time PCR. (XLSX) S2 Table. Potential binding partners of L.donovani profilin. Four proteins were identified as present in all three GST-profilin pulldown assays, while 34 proteins were identified in at least two GST-profilin pulldown assays. Only proteins that were not detected in any of the GSTalone pulldown controls were considered. The protein IDs were given according to L. donovani BPK282A1 strain as downloaded from the TriTrypDB (version 51) database (www. tritrypdb.org). (XLSX) S3 Table. Statistics for RNA-Seq data sets. Reads alignment was done by Bowtie2, using the L. donovani genome (Leishmania donovani BPK282A1, NCBI taxon ID:981087) [17].