2.1. Cell culture and cell cycle synchronization

T. cruzi epimastigotes (Cl Brener strain) were grown in liver infusion tryptose (LIT) medium supplemented with 10% heat-inactivated fetal bovine serum at 28 °C. Different cell cycle stages were obtained after synchronization as previously performed [11,19]. Briefly, parasites in a log-phase were treated with 20 mM of Hydroxyurea (HU) for 24 h. HU was then removed from the medium by centrifugation (1200 x g, 10 min), the cells were washed three times in phosphate-buffered saline (PBS) and resuspended in fresh LIT medium. To obtain G1/S, S and G2/M enriched parasite populations, samples were collected at 0, 6 and 12 h post-HU release respectively. The samples were kept at -80 ºC until the cell cycle synchronization confirmation. A sample of parasites harvested before the incubation with HU was used as a control of asynchronous growth. For each time-point, an aliquot of 10⁷ parasites/mL was washed twice in cold-PBS and fixed overnight at 4˚C with 70% Ethanol-PBS. Propidium iodide (PI) staining was carried out by incubation of the fixed parasites for 30 min at 37˚C in PBS containing 20 μg/mL PI and 20 μg/mL RNAse (Invitrogen) for DNA-specific staining. Four technical replicates per biological sample were analyzed for DNA-content in a Flow cytometer (Attune NXT - Thermo). The proportion of G1/S, S and G2/M cells in the samples was determined by DNA content analyzed in FlowJo software (BD).

2.2. Nuclei Purification

Frozen harvested parasites (1 × 10⁹ cells per replicate; four biological replicates per cell cycle phase and an asynchronous control) were rinsed and resuspended in 1 mL of suspension buffer (10 mM potassium glutamate, 250 mM sucrose, 2.5 mM CaCl₂, 1 mM phenylmethylsulfonyl fluoride - PMSF).. Subsequently, lysis was induced by the addition of 0.1% Triton X100. The resulting lysate underwent centrifugation, followed by washing and resuspension in 1 ml of suspension buffer. The lysate was then carefully layered atop a 50% Percoll solution (Sigma) in a suspension buffer (4 ml) and subjected to centrifugation at 63,000 × g at 4 °C for 2 hours, as previously described [20]. The nuclei, located at the bottom of the self-forming Percoll gradient were isolated, washed thoroughly with a suspension buffer, and storage at -80°C to further analysis.

2.3. TMT labeling and quantitative proteomics analysis

Nuclei protein extracts were prepared with FASP Protein Digestion Kit (Expedeon-Abcam-USA) and digested with trypsin (Promega) by following the manufacturer’s recommendations. Briefly, proteins were solubilized in SDT lysis buffer (4% SDS, 100 mM Tris-HCl pH 7.6, 0.1 M DTT), exchanged into UA buffer (8 M urea in 0.1 M Tris-HCl pH 8.5), and alkylated with 50 mM iodoacetamide (IAA) prepared in UA. Proteins were digested with trypsin (Promega; sequencing grade) and peptides were recovered using 50 mM ammonium bicarbonate (ABC). Four biological replicates were analyzed per condition (G1/S, S, and G2/M) using TMT-based quantitative proteomics. After the digestion of proteins into peptides, the samples were labeled with TMT (Thermo-TMT 16plex Mass Tag Labeling kit cat. 90110). Briefly, 50 μg of peptides were resuspended to 40 μl of 100 mM Tetraethylammonium bicarbonate (TEAB) and 10 μl of 100% acetonitrile. Ten microliters of the TMT label reagent were added to each sample, and the reaction was allowed to proceed for 1 hour at room temperature. For reaction quenching, we added 8 μl of 0.5% hydroxylamine per sample and incubated for 15 minutes. Equal amounts of each sample (126, 127N, 127C, 128N, 128C, 129N, 129C, 130N,130C, 131N, 131C, 132N) were combined and dried in a speed vac. Peptides were separated and analyzed by an UltiMate 3000 RSLC nano system coupled to an Orbitrap Fusion Tribrid mass spectrometer (both from Thermo Scientific). Peptides were firstly loaded and concentrated on to an Acclaim PepMap300 trap column (300 µm x 5 mm) packed with C18 (5 µm, 300 Å) (Thermo) in loading buffer of 1% acetonitrile with 0.1% formic acid then separated in an EASY-Spray column (75 µm x 50 cm) packed with C18 (2 µm, 100 Å, Thermo Scientific) at a flow rate of 300 nL/min. Mobile phase A (0.1% formic acid in water) and mobile phase B (0.08% formic acid in 80% acetonitrile/ 20% H₂O) were used to establish a solvent gradient of 4% B for 1.5 min, 4–60% for 178.5 min, 60–99% for 15 min, and held at 99% for 5 min. A further 10 minutes at initial conditions for column re-equilibration was used before the next injection. The Orbitrap Fusion acquired data continuously high-resolution precursor scan at 120 000 resolving power (over a mass range of m/z 400 – 1600) followed by top speed (2.5s) CID fragmentation (35%) and detection of the top precursor ions from the MS scan in the linear ion trap using turbo scan speed. Triple stage mass spectrometry (MS3) higher-energy collisional dissociation (HCD; 55%) was performed on the top ions from the MS2 CID scan, with up to 10 synchronous precursor selection (SPS) scans isolated with the precursor ion and any TMT loss ions excluded from the selection. Orbitrap detection of the TMT quantitation label from the MS3 fragmentation was acquired at a resolution of 50000 over a mass range of 100–500 m/z.

2.4. TMT data processing

Thermo Scientific Proteome Discoverer (version 2.5) with SEQUEST search engines were used for peptide/protein identification. The first step of the data processing was to check the TMT labeling efficiency (greater than 90%, there is less than a 3-fold difference in an individual channel). Reporter ion intensities in each TMT channel were aggregated to the peptide level. Mean values were then calculated for each identified peptide across all TMT channels, and the values for each channel were then divided by this mean. The resulting ratios were log₂-transformed and used to make distribution plots for each TMT channel within a plex. Normalization was applied based on total peptide amount, with factors ranging from 1.0 to 2.46 for different samples (S2B Fig). These values were used to ensure consistent protein quantification across the TMT channels. The protein searches were performed against a T. cruzi database (Trypanosoma cruzi CL Brener Esmeraldo-like V.54, available on TriTrypDB). Proteins were grouped by applying the maximum parsimony rule (i.e., the protein groups in the final report represent the shortest possible list needed to explain all confidently observed peptides). The quantitation module within Proteome Discoverer software was used to assess the ratios for individually tagged samples. The height of reporter ions detected with mass tolerance ±10 ppm was adjusted, considering the isotopic correction factors provided by the TMT kit manufacturer. The searches were performed with the following parameters: MS/MS accuracy 0.6 Da for CID and 0.1 Da for HCD, trypsin digestion with two missed cleavages allowed, fixed carbamido-methyl modification of cysteine, and variable modification of oxidized methionine and phosphorylation (S, T, Y). TMT Pro was set as a variable modification of the N-terminus. The number of proteins and protein groups and numbers of peptides were estimated using Proteome Discoverer, with a false discovery rate (FDR) of 5%, and peptide rank 1 applied as a cutoff limit. Proteins showing fold changes ≥ 1.5 or ≤ -1.5 and and p-value < 0.05 for each set of comparisons were considered and displayed on volcano plots.

2.5. Gene Ontology (GO) and metabolic pathways enrichment and other T. cruzi proteomes

We conducted the GO enrichment analysis through the TriTrypDB database (https://tritrypdb.org/tritrypdb/app/) to identify significantly enriched GO terms among the selected proteins. We searched for the total proteins found in each cell cycle phase by cellular component category. Next, we examined the differentially expressed proteins (fold changes ≥ 1.5 or ≤ -1.5 and p-value < 0.05) in the ratios of G1/S vs. S and G2/M vs. G1/S to identify significantly enriched GO terms in biological processes and molecular functions. We set the threshold value for GO terms with a p-value cutoff of 0.05.

T. cruzi metabolic enzymes were classified as those belonged to the following GO terms available at TriTrypDB: “cellular amino acid metabolic process” (GO: 0006520), “carbohydrate metabolic process” (GO: 0005975), “pyruvate metabolic process” (GO: 0006090), “ATP metabolic process” (GO: 00046034), “nucleotide metabolic process” (GO: 0009116), “nucleoside metabolic process” (GO: 0009117), “lipid metabolic process” (GO: 0006629), “generation of precursor metabolites and energy” (GO: 0006091) terms. Metabolic pathway enrichment was conducted by KEEG tool though the TriTrypDB DB considering a p-value cutoff of 0.05 (Bonferroni). The list of proteins detected at the nuclear [18], chromatin [21] and cell cycle [17] proteomics was obtained from supplemental material and compared to our dataset.. Subcellular localization predictions were performed for all proteins detected in the nuclear proteome using DeepLoc 2.1 [22]. For each protein, the top predicted localization, additional predicted localizations, predicted sorting signals, membrane type, and compartment probability scores were retrieved. For the subset of metabolic/non-canonical metabolic proteins, DeepLoc outputs were further integrated with protein annotation, genomic location, TrypTag localization of the corresponding T. brucei orthologs when available and overlap with published T. cruzi nuclear/chromatin proteomic datasets.

2.6. Preparation of epimastigotes extracts for enzymatic assays

Enzymatic activity assays were performed with the samples obtained from whole extracts, nuclear and cytoplasmic extracts of epimastigotes forms. Parasites in the exponentially growing phase in LIT medium were collected and washed twice with chilled 1X PBS. Cell lysis was performed according to [23]. Briefly, parasites in hypotonic buffer (10 mm HEPES, pH 7.9; 1.5 mm MgCl₂; 10 mm KCl; 0.5 mm DTT; and protease inhibitors) with 0.1% NP40, and the cytoplasm was separated from the nucleus after the addition of 1 M sucrose and centrifugation at 500 x g for 5 minutes at 4°C. The supernatant containing the cytoplasm was collected and kept at 4°C, while the pellet was washed with a 0.35 M sucrose solution. After centrifugation at 1100 x g for 5 minutes at 4°C, the supernatant was discarded, and the nucleus was solubilized in a buffer containing 0.35 M sucrose, 0.5 mM DTT; 10 mM HEPES pH 7.9; 33 mM MgCl₂; 10 mM KCl; and a cocktail of protease inhibitors (Sigma). The total extract was obtained by lysing the parasites with buffer containing 20 mM Tris-HCl, 0.25 M sucrose, 1 mM EDTA, 0.1% Triton-X 100 (v/v), 1 mM PMSF, and a cocktail of protease inhibitors (Sigma) followed by sonication for five cycles (5 seconds at 20% power and 30 seconds pause on ice). After this process, the extracts were centrifuged at 13,000 x g for 30 minutes at 4°C, and the supernatants were quantified using the Bradford method [24]. Following the activity assays, total, cytoplasmic, and nuclear extracts were analyzed via Western Blotting (WB) to confirm the nucleus and cytoplasm separation, using anti-eIF5a antibodies (cytoplasmic protein) and anti-histone H3 antibodies (nuclear protein).

2.7. Enzymatic assays

Hexokinase (HK) activity assay: HK was assessed indirectly through a coupled enzymatic reaction involving glucose-6-phosphate dehydrogenase (G6PD, Sigma). In this assay, HK catalyzes the conversion of glucose and ATP into glucose-6-phosphate (G6P), which is subsequently oxidized by G6PD, generating NADPH. The formation of NADPH is monitored by measuring absorbance at 340 nm. Thus, in the presence of glucose and ATP, an increase in absorbance over time indicates HK activity, with a linear rise in NADPH production reflecting the enzymatic reaction rate. In this assay, the reaction contained: 50 mM Tris-HCl pH 7.6 buffer; 8 mM D-glucose; 9.6 mM sodium salt ATP; 19 mM MgCl₂; 0.1 mM NADP⁺; 5U/ml G6PD enzyme [25]. The reaction was initiated by adding 100 μg of extract and monitored for 5 minutes at 28°C. A blank was performed with the same preparation, in the absence of glucose.

Pyruvate Kinase (PK) Activity Assay: Pyruvate kinase activity was measured using a coupled enzymatic assay with lactate dehydrogenase (LDH), by monitoring the decrease in NADH absorbance at 340 nm. In this reaction, PK catalyzes the transfer of a phosphate group from phosphoenolpyruvate (PEP) to ADP, forming ATP and pyruvate. The pyruvate produced is then converted to lactate by LDH, which simultaneously oxidizes NADH to NAD ⁺ . As NADH is consumed, its absorbance at 340 nm decreases over time, and this decay is used as a readout of PK enzymatic activity. A blank was performed with the same preparation, in the absence of PEP [26]. The reaction consists of 39 mM potassium phosphate buffer pH 7.6; 0.58 mM PEP; 0.11 mM NADH; 6.8 mM MgSO₄·7H₂O; 1.5 mM ADP; 10 U of LDH; and 250 μg of parasite extract to be measured for activity, in a final volume of 200 μl. The activity was monitored for 5 minutes at 28°C.

Citrate Synthase (CS) Activity Assay: The activity of citrate synthase (CS) was measured based on the decay of Acetyl-CoA in the reaction. Oxaloacetate (OA) was used by CS as a substrate. CS transfers the CoA group to oxaloacetate, generating citrate and Coenzyme A without the acyl group (CoA-SH). A blank was performed with the same preparation, in the absence of OA [27]. The reaction consists of 100 μM DTNB; 7.5 mM Tris-HCl pH 8.1; 0.03 mM Acetyl-CoA; 500 μM oxaloacetate; and 100 μg of parasite extract to be measured for activity, in a final volume of 200 μl. The activity was monitored for 5 minutes at 28°C.

Glycerol Kinase (GK) Activity Assay: Glycerol kinase catalyzes the transfer of a phosphate group from ATP to glycerol, resulting in the formation of glycerol-3-phosphate (G3P). The activity of this enzyme was measured at 340 nm from a coupled reaction, adding PK and LDH enzymes to the reaction. G3P has its phosphate group transferred to ADP by the action of PEP, generating ATP and pyruvate. Pyruvate was then converted by LDH into lactic acid, via oxidation of NADH, which is consumed from the medium, resulting in a decreasing absorbance. The reaction consists of 39 mM potassium phosphate buffer pH 7.6; 0.58 mM PEP; 0.11 mM NADH; 6.8 mM MgSO₄·7H₂O; 1.5 mM ADP; 10 U of PK/LDH; and 100 μg of parasite extract to be measured for activity, in a final volume of 200 μl. The activity was monitored for 5 minutes at 28°C [28].

Pyrroline-5-carboxylate dehydrogenase (P5CDH) Activity Assay. The activity of P5CDH was measured based on the increase in NADH formation, catalyzed by pyrroline-5-carboxylate (P5C) dehydrogenase, which irreversibly converts NAD⁺-dependent P5C into glutamate. The formation of NADH released by the enzyme is measured at 340 nm [28]. The reaction consists of 100 mM KH₂PO₄ buffer pH 7.2; 40 mM NAD⁺; 100 μg parasite extract, in a final volume of 200 μl. The activity was monitored for 5 minutes at 28°C. For all enzymatic assays, a control group was included and consisted of reaction solutions without the enzyme substrates to rule out residual activity.

2.8. Immunofluorescence analysis of metabolic enzymes

Immunofluorescence assays were performed using T. cruzi epimastigotes expressing H2B.V-Myc, generated as previously described [29]. For IFA, the parasites were adhered to poly-lysine-treated 8-well slides and fixed with 4% ultrapure paraformaldehyde. After fixation, slides were incubated with PBS supplemented with 20 mM glycine, permeabilized with 0.2% NP40 in PBS, blocked with 50% SFB in PBS and 3% BSA for 2 hours, then incubated with 8M UREA in PBS, and washed. After washing, slides were incubated for 2 h with antibodies against T. brucei hexokinase, citrate synthase, or glycerol kinase (1:2000), kindly provided by Dr. Frédéric Bringaud. Following primary antibody incubation, slides were washed and incubated for 40 min at room temperature in the dark with species-specific secondary antibodies: anti-mouse Alexa Fluor 555 (Thermo Fisher Scientific, 1:2000) to detect anti-Myc, and anti-rabbit Alexa Fluor 488 (Thermo Fisher Scientific, 1:2000) to detect the metabolic enzyme antibodies. Images were acquired on an Olympus BX51 fluorescence microscope equipped with a 100 × oil-immersion objective and controlled by Olympus cell^R software. Bright-field, DAPI, Alexa Fluor 488, and Alexa Fluor 555 channels were acquired using comparable exposure settings within each experiment. Z-stacks were collected through the parasite body and deconvolved in cell^R before visualization. Fluorescence intensity line profiles were generated from deconvolved images by drawing a line across the parasite and nuclear region; DAPI, Alexa Fluor 488, and Alexa Fluor 555 signal intensities were extracted along the same line using cell^R.

3.1. Cell cycle synchronization and nuclei isolation: Validation and overview of nuclear proteins across the cell cycle

To decipher the nuclear cell cycle proteins signature, we proceeded with a high resolution labelled quantitative proteomics experiment, detailed in the flowchart at S1 Fig. To synchronize the cell cycle in epimastigotes forms, we used HU that inhibits the ribonucleotide reductase enzyme, blocking the conversion of ribonucleotides into deoxyribonucleotides (dNTPs) leading to a depletion of dNTP pools, effectively preventing DNA synthesis [19]. As a result, cells treated with HU are unable to progress through the S phase and are instead halted at the G1/S transition. Thus, cells in S and G2/M were collected upon HU release and cell cycle phase were further confirmed by flow cytometry analysis (Fig 1A).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Characterization of Nuclear Protein Extracts and Histone Abundance During the T. cruzi Cell Cycle.

(A) Representative flow cytometry histograms indicate the distribution of the DNA content in the experimental groups G1/S, S, G2/M and asynchronous control. (B) Immunofluorescence images of epimastigotes (Epi, top row) and isolated nuclear fractions (Nuclei, bottom row). DNA was stained with DAPI (blue), and the flagellum was labeled with the anti-flagellum monoclonal antibody mAb 2F6 (green). Brightfield images are shown on the right. N, nucleus; K, kinetoplast. (C) SDS-PAGE (15%) of epimastigote and nuclear extracts stained with Coomassie blue, showing enrichment of histones (~15 kDa) in the nuclear fraction. (D) Representative western blot assay showing the presence of nuclear marker protein (anti-H3) and the absence of ant-HSP70 (endoplasmatic reticulum), anti-aldolase (glycosomal) in the nuclear fraction and whole cell lysate. (E) Venn diagram showing the overlap of proteins identified in the nuclear proteome across the three cell cycle phases analyzed. (F) Venn diagram comparing proteins identified in our nuclear cell cycle proteome (CC_PN, green) with those previously reported in the T. cruzi nuclear proteome (PN, orange) (1) and chromatin-associated proteome (Chr_P, blue) (2).

https://doi.org/10.1371/journal.pntd.0013299.g001

Upon cell cycle synchronization, we purified the nuclei and evaluated them by fluorescence microscopy demonstrating the presence of intact nuclear structures without kinetoplast contamination (Fig 1B). The nuclear protein extracts were analyzed by Coomassie-stained gel electrophoresis (Fig 1C) and western blot (Fig 1D). We found that the nuclear extracts were enriched in histone H3 and minimally contaminated with endoplasmic reticulum (HSP70) and glycosome (aldolase) proteins (Fig 1D). The results indicated a successful T. cruzi nuclear enrichment.

We quantitatively evaluated TMT-labelled nuclear peptides from cell cycle phases (G1/S, S, and G2/M) by LC-MS/MS. We identified 2,937 proteins belonging to 1,560 protein groups (S1 Table). Principal component analysis (PCA) revealed a distinct separation (explaining 62.3% of the variability) among G1/S, S, and G2/M in the first two components, indicating differences in the nuclear proteomes across cell cycle phases (S2A Fig). Quality assessment of TMT reporter ion distributions after normalization showed comparable abundance profiles across channels, supporting consistent quantitative performance within the TMT plex (S2B Fig). Cellular component analysis based on Gene Ontology (GO) showed that “chromatin” and “chromosome” terms were overrepresented among proteins detected across all cell cycle phases/transitions (S2C Fig), supporting the successful recovery of nuclear-associated proteins.

We identified similar numbers of proteins in each cell cycle phase, with more exclusive proteins in S phase (35 proteins), followed by G1/S (33 proteins) and G2/M (3 proteins) (Fig 1E) (S2 Table and Fig 1E. Among those exclusive proteins of G1/S we found signaling and regulatory proteins such as Mitogen-Activated Protein Kinase 10 (MAPK 10) (TcCLB.506229.10) that can play important roles during cell cycle progression [30]. Previous nuclear proteomics analysis [18] identified 957 nuclear proteins, in which 420 (44%) were found on both analyses. Comparison of previous chromatin proteomics retrieved 864 proteins (55%) (Fig 1F).

3.2. Histones, nucleotide/nucleoside metabolic proteins and kinases profiling during cell cycle

It is well established that canonical histones, including those from T. cruzi, are synthesized in a cell cycle dependent manner, reaching maximal levels in S-phase [31]. In contrast, the timing of synthesis for T. cruzi histone variants remains unclear, as their expression may not be tightly coupled to DNA replication. As expected, we observed a peak of both canonical (Fig 2A) and variant (Fig 2B) protein levels during S-phase. On average, their abundance increases more than threefold compared to G1/S phase, followed by a return to near G1/S levels in G2/M for canonical histones and H2B.V (Fig 2B). This pattern reflects the expected upregulation of histone synthesis during S-phase, when histones become proportionally more abundant when compared to the total protein content, and their proportional levels decline post-replication as the cell progresses into G2/M. Interesting, the levels of variants histones H2A.Z and H3.V remained relatively high in G2/M suggesting a differential regulation of histone variant dynamics, potentially playing a distinct role in chromatin organization or maintenance after DNA replication (Fig 2B). In accordance, we detected two nucleosome assembly proteins (TcCLB.507031.29 and TcCLB.505983.20) that also were increased threefold in S-phase suggesting their involvement in histone deposition (S1 Table).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Cell cycle-associated variation of histones, kinases, and nucleotide metabolism-related proteins in the T. cruzi nuclear proteome.

(A) Relative abundance of canonical histones during G1/S, S, and G2/M phases. (B) Relative abundance of histone variants during G1/S, S, and G2/M phases. (C) Relative abundance profiles of selected kinases detected in the nuclear proteome, including metabolic kinases and signaling-associated kinases. (D) Relative abundance of selected proteins associated with nucleotide metabolism across the cell cycle. Statistical significance was assessed by ANOVA (*p < 0.05, **p < 0.01). (E) Schematic representation of nucleotide metabolism in T. cruzi, including pyrimidine de novo and salvage pathways and the purine salvage pathway. Enzymes are color-coded according to their detection and abundance variation across the cell cycle: orange, identified but not quantified; green, no significant variation; red, increased abundance in G1/S compared with S and G2/M; gray, decreased abundance in G2/M compared with S and G1/S; and white, not detected. Pyrimidine metabolism: carbamoyl-phosphate synthetase II (CPSII, EC:6.3.5.5), aspartate carbamoyltransferase (ATCase, EC:2.1.3.2), dihydroorotase (DHOase, EC:3.5.2.3), dihydroorotate dehydrogenase (DHODH, EC:1.3.98.1), orotate phosphoribosyltransferase (OPRT, EC:2.4.2.10), orotidine-5-phosphate decarboxylase (OMP, EC:4.1.1.23), uracil phosphoribosyltransferase (UPT, EC:2.4.2.9), UMP-CMP kinase 2 (UMPK, EC:2.7.4.22), cytidine triphosphate synthase (CTPS, EC:6.3.4.2), uridine-cytidine kinase (UCK, EC:2.7.1.48), nucleoside diphosphate kinase (NDK, EC:2.7.4.6), ribonucleoside-diphosphate reductase (RNR, EC:1.17.4.1), 5′-deoxynucleotidase (5′DN, EC:3.1.3.89), thymidylate kinase (TK, EC:2.7.4.9), and phosphoribosylpyrophosphate synthetase (PRPS, EC:2.7.6.1). Purine metabolism: inosine-adenosine-guanosine-nucleoside hydrolase (IGNH, EC:3.2.2.1), hypoxanthine-guanine phosphoribosyltransferase (HGPRT, EC:2.4.2.8), inosine-5′-monophosphate dehydrogenase (IMPDH/INMD, EC:1.1.1.205), guanylate kinase (GK, EC:2.7.4.8), adenosine kinase (AK, EC:2.7.1.20), and adenylate kinase (ADK, EC:2.7.4.3). Abbreviations: PRPP, phosphoribosyl pyrophosphate; AMP, adenosine monophosphate; ADP, adenosine diphosphate; ATP, adenosine triphosphate; dADP, deoxyadenosine diphosphate; dATP, deoxyadenosine triphosphate; IMP, inosine monophosphate; GMP, guanosine monophosphate; GDP, guanosine diphosphate; GTP, guanosine triphosphate; dGDP, deoxyguanosine diphosphate; dGTP, deoxyguanosine triphosphate; XMP, xanthosine monophosphate; UMP, uridine monophosphate; UDP, uridine diphosphate; UTP, uridine triphosphate; CMP, cytidine monophosphate; CDP, cytidine diphosphate; CTP, cytidine triphosphate; dCDP, deoxycytidine diphosphate; dCTP, deoxycytidine triphosphate; dTDP, deoxythymidine diphosphate; dTTP, deoxythymidine triphosphate.

https://doi.org/10.1371/journal.pntd.0013299.g002

Nucleotide synthesis is also finely tuned to the cell cycle, ensuring the presence of substrates for DNA synthesis. A multiomics analysis performed in yeast, revealed that nucleotides synthesis peaked in G1 and S phase and proteins associated with nucleobase metabolism are periodically expressed along the cell cycle [32]. Among them, four proteins showed increased abundance in the HU-arrested G1/S group relative to G2/M, with fold changes ranging from 5.7 to 37 (Fig 2D–2E). These proteins included nucleoside diphosphate kinase 1 (NDK), nucleoside phosphorylase, ribonucleoside-diphosphate reductase small chain (RNR), and ribose-5-phosphate isomerase (RPI). Because HU directly affects dNTP metabolism through inhibition of ribonucleotide reductase, we interpret these increases cautiously, as the extent to which they represent primary effects of HU exposure versus secondary consequences of cell-cycle arrest remains unclear. All these enzymes were related to the production of more substrates (nucleotides and their parts) for DNA synthesis. For example, the enrichment of NDK may increase the conversion of diphosphate nucleotides into triphosphates. RNR catalyzes the conversion of ribonucleotides into deoxyribonucleotides, therefore favoring DNA replication. RPI catalyzes the utilization of ribose-5-phosphate for nucleotide synthesis. Additionally, nucleoside phosphorylase plays a key role in the salvage pathways of nucleosides, which are essential in trypanosomes since they lack the de novo purine synthesis pathway and rely entirely on salvage mechanisms to obtain purines [33,34].

In addition to the regulation of nucleotide synthesis, proper coordination of cell cycle events also depends on signaling pathways. Kinases play a pivotal role in the regulation of the cell cycle, orchestrating key events such as DNA replication, mitosis, and cell division [35]. We found 42 protein kinases, which were further classified into “Serine/Threonine kinases”, “Glycolysis/Gluconeogenesis Kinases” and “Phosphorylation/Signaling Kinases” (S3 Table) (Fig 2C). Among the kinases, 24% were up-regulated in G1/S (p-value <0.05 and FC > 1.5) including hexokinase (TcCLB.510121.20), pyruvate phosphate dikinase (TcCLB.510101.140 and TcCLB.506297.190); Phosphoenolpyruvate carboxykinase (TcCLB.507547.90), nucleoside diphosphate kinase 1 (TcCLB.508707.200), arginine kinase (TcCLB.507241.30), phosphoglycerate kinase (TcCLB.505999.90), homoserine kinase (TcCLB.503597.10) and mevalonate kinase (TcCLB.436521.9)(Fig 2C). Notably, we detected upregulation in G2/M vs. G1/S for the receptor for activated C kinase 1 (TcCLB.511211.130), a protein involved in cytokinesis regulation [36], reinforcing the good quality of our results.

3.3. Differentially expressed proteins across the cell cycle

To get further insights into nuclear protein dynamics, we identified 359 proteins exhibiting upregulation and/or downregulation (FDR < 0.05, fold changes ≥ 1.5 or ≤ -1.5 and p-value < 0.05) (Fig 3A-3C) (S4 Table). Comparison between protein expression from G1/S vs. S and G2/M vs. G1/S, retrieved higher number of differentially expressed proteins (DEPs), related to S vs. G2/M. Specifically, in the G1/S vs. S, we detected 100 up-regulated and 78 down-regulated proteins (Fig 3A). Regarding the G2/M vs. G1/S, 84 proteins were down-regulated, while 89 were up-regulated (Fig 3C). Conversely, in the S vs. G2/M, only 8 proteins were differentially expressed (Fig 3B).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Differentially expressed proteins across the cell cycle.

Volcano plots show the differentially abundant nuclear proteins between (A) G1/S vs. S, (B) S vs. G2/M, and (C) G2/M vs. G1/S phases. Proteins with significantly increased (red) (pValue <0.05 and FC > 1.5) or decreased (green) (pValue <0.05 and FC < -1.5) abundance are highlighted, and the total number of upregulated and downregulated proteins in each comparison is indicated. (D–E) Bar plots display enriched Gene Ontology (GO) terms from the Biological Process category for the comparisons G1/S vs. S (D) and G2/M vs. G1/S (E). GO terms were obtained from TriTrypDB tools and analyzed using ReviGO, filtering by p-value < 0.05. Bar lengths represent the log-scaled size of the GO term groups. Upregulated processes are shown in red, while downregulated processes are shown in green.

https://doi.org/10.1371/journal.pntd.0013299.g003

To further investigate the biological roles of DEPs across the T. cruzi cell cycle, we analyzed overrepresented GO terms within the “biological process” category (S5 Table). The top terms are presented in Fig 3D (G1/S vs. S) and Fig 3E (G2/M vs. G1/S). GO enrichment analysis comparing G1/S and S phases revealed that proteins upregulated in G1 were associated with metabolic and biosynthetic processes, including “amino acid metabolic process” (GO:0006520), “protein folding” (GO:0006457), “gluconeogenesis” (GO:0006094), and “tricarboxylic acid metabolic process” (GO:0072350) (Fig 3D). These terms suggest that T. cruzi cells are metabolically active during G1, accumulating energy and building blocks required for the upcoming DNA replication. In contrast, proteins more abundant in the S phase were represented by”chromosome organization” (GO:0051276), “DNA conformation change” (GO:0071103), “protein–DNA complex assembly” (GO:0065004), and “chromatin organization” (GO:0006325) (Fig 3D) terms, indicating that chromatin remodeling and nucleoprotein assembly processes are prominent during active DNA synthesis, as expected.

In the G2/M vs. G1/S comparison, a similar set of metabolic terms remained downregulated in G2/M - “amino acid metabolic process” (GO:0006520), “gluconeogenesis” (GO:0006094), and “protein folding” (GO:0006457)- and thus overrepresented in G1/S (Fig 3E). Conversely, upregulated proteins in G2/M were associated with “gene expression” (GO:0010467), “translation” (GO:0006412), “amide metabolic process” (GO:0043603), and “peptide metabolic process” (GO:0006518) (Fig 3E), suggesting a shift toward increased protein synthesis and gene expression as the cell prepares for mitosis. Together, these results indicate a functional transition across the cell cycle, from a metabolically active G1/S phase to a G2/M phase characterized by upregulation of gene expression and translation-related processes.

3.4. Metabolic compartmentalization

The recurrent enrichment of GO terms related to amino acid metabolism, gluconeogenesis, and the TCA cycle, suggests the presence of metabolic enzymes in the nuclear proteome. This prompted further investigation into their nuclear localization and the potential role of metabolic compartmentalization within the nucleus. The presence of metabolic enzymes within the nuclear compartment was already detected in other organisms [37–39] offering insights about their role within the nuclear environment.

Metabolic enzymes and metabolism-associated proteins represented 5.96% of the nuclear proteome, corresponding to 91 proteins (Fig 4A and S6 Table). KEGG pathway analysis showed overrepresentation of “Fatty acid elongation”, “Citrate cycle (TCA cycle)”, “Pyruvate metabolism”, and “Pyrimidine metabolism” among this subset (p < 0.05, Bonferroni correction) (Fig 4B). To further support their nuclear association, we compared this list with previously published T. cruzi chromatin-associated (Chr_P) and nuclear proteomic (PN) datasets [18,21](Fig 4C). Twenty-eight metabolic proteins were also identified in at least one of these datasets, including hexokinase (TcCLB.508951.20), pyruvate dehydrogenase E1 component alpha subunit (TcCLB.507831.70), pyruvate dehydrogenase E1 beta subunit (TcCLB.510091.80), pyruvate phosphate dikinase (TcCLB.510101.140), citrate synthase (TcCLB.509801.30), and 3-phosphoglycerate kinase (TcCLB.511419.50), which are key enzymes in central carbon metabolism.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Analysis of nuclear metabolic enzymes across the T. cruzi cell cycle.

Proportion of nuclear proteins associated with metabolism (5.96%) identified using GO terms related to key metabolic processes, including “cellular amino acid metabolic process” (GO:0006520), “carbohydrate metabolic process” (GO:0005975), “pyruvate metabolic process” (GO:0006090), “ATP metabolic process” (GO:00046034), “nucleotide metabolic process” (GO:0009116), “nucleoside metabolic process” (GO:0009117), “lipid metabolic process” (GO:0006629), and “generation of precursor metabolites and energy” (GO:0006091). (B) KEGG pathway enrichment analysis of the metabolic enzymes identified in (A), performed using TriTrypDB’s tool with a Bonferroni-corrected p-value cutoff of 0.05. Enriched pathways are shown according to fold enrichment. (C) Venn diagram comparing the nuclear proteins identified in this study (CC_PN) with those present in the nuclear proteome (CC_PT) (De Castro Moreira Dos Santos et al., 2015) and those at chromatin proteome (Chr_P) (De Jesus et al., 2017). (D) Hierarchical clustering analysis of all protein expression profiles across different cell cycle phases. The protein data abundance was normalized before clustering by the Euclidean distance ratio. (E) Venn diagram between the nuclear cell cycle proteins identified in this study (CC_NP) and the Cell cycle proteome (CC_P) (Santos Júnior et al., 2021). (F) Box plots representing the relative abundance (%) of nuclear metabolic proteins found in (A) and ribosomal proteins across cell cycle phases.

https://doi.org/10.1371/journal.pntd.0013299.g004

To add an additional layer of evidence for potential nuclear association, we performed DeepLoc 2.1 analysis [22] of the 91 metabolic/non-canonical metabolic proteins identified in the nuclear proteome (S6 Table). As expected for enzymes with canonical metabolic functions, most proteins retained predicted extranuclear localizations. Nevertheless, a subset displayed nuclear-associated features: five proteins (5.5%) included nucleus among their predicted localizations, six proteins (6.6%) showed a nuclear probability score ≥ 0.5, and 22 proteins (24.2%) contained predicted nuclear localization and/or export signals. These included enzymes associated with glycolysis, nucleotide metabolism, and amino acid metabolism, such as enolase (TcCLB.504105.140), UMP-CMP kinase (TcCLB.506893.100), deoxyribose-phosphate aldolase (TcCLB.507559.80), glutamine-dependent carbamoyl-phosphate synthase (TcCLB.508373.10), and pyruvate kinase 2 (TcCLB.507993.390). In addition, 18 of the 77 proteins with available TrypTag [40] ortholog information (23.3%) had nuclear, nucleoplasmic, nuclear lumen, or nuclear envelope localization reported for their T. brucei orthologs, including adenylate kinase 1 (TcCLB.503479.30), hypoxanthine-guanine phosphoribosyltransferase (TcCLB.506457.40), nucleoside diphosphate kinase 1 (TcCLB.508707.200), and the alpha (TcCLB.507831.70) and beta (TcCLB.510091.80) subunits of pyruvate dehydrogenase (S6 Table). Together, these findings indicate that although most enzymes preserve canonical metabolic localization signatures, a subset shows predicted or comparative evidence compatible with nuclear trafficking or nuclear [40].

Molecular traffic between the nucleus and cytoplasm is mediated by nuclear pore complexes (NPCs) in the nuclear envelope. These complexes allow the passage of ions, metabolites, and small proteins (40–60 kDa), but restrict larger macromolecules unless they are bound by nuclear transport receptors, such as importins and exportins [41]. Notably, ~ 68% (1052) of the nuclear proteins identified in our proteomic analysis were under 60 kDa (S3 Fig), including 74 (82.2%) metabolic proteins, suggesting that their nuclear localization may be facilitated by passive diffusion through NPCs rather than active transport mechanisms.

Temporal compartmentalization of some enzymes can influence cellular function mainly in the dynamic control of cell cycle progression [37]. Then, we investigated whether different cell cycle phase/transition were associated with a distinct profile of metabolic protein expression. We found a hierarchical cluster of highly abundant proteins at G1/S compared with S and G2/M that are enriched in metabolic proteins (Fig 4D and S7 Table). To assess whether these metabolic proteins were specifically localized inside the nucleus during the G1 phase or simply upregulated in G1, we compared our dataset with the T. cruzi cell cycle proteomics analysis published by [17] (Fig 4E). Among the metabolic-associated proteins identified in our nuclear proteome, twelve were only enriched at G1 in our current dataset, including four proteins from carbohydrate metabolism. Together, these results suggest a specific enrichment of nuclear metabolic proteins specifically associated with the G1 phase, suggesting a possible increase in the nuclear import or retention of metabolic proteins during G1.

To assess whether the abundance of metabolic proteins in nuclear extracts is comparable to that of nuclear proteins, we compared their levels to ribosomal proteins, which are typically abundant in the nucleus. Metabolic enzymes abundance is confirmed to be higher at G1/S phase and lower in S and G2/M phase in contrast with ribosomal proteins that display a progressive enrichment across the cell cycle. Interestingly, metabolic enzymes showed nuclear abundance values in the same order of magnitude as ribosomal proteins across all cell cycle phases (Fig 4F). This comparable abundance indicates that metabolic enzymes are consistently detected in nuclear preparations, although the relative partitioning of these proteins between the nucleus and cytoplasm cannot be determined from the present dataset. The higher abundance observed in the HU-arrested G1/S group suggests that this nuclear-associated metabolic protein pool may vary across the cell cycle, supporting a model of dynamic metabolic compartmentalization within the nuclear fraction.

3.5. Validation of nuclear metabolic compartmentalization

Given that nearly 6% of the nuclear proteome corresponded to metabolic enzymes and metabolism-associated proteins, primarily associated with glycolysis, pyruvate metabolism, and the TCA cycle (Fig 4B), we mapped the identified enzymes onto central carbon metabolism pathways (Fig 5A). (highlighted in red, Fig 5A). Our goal was to determine whether these enzymes are confined to a few specific steps or broadly distributed across the pathways. We found that they are indeed widespread, with enzymes present at multiple points throughout both glycolysis and the TCA cycle.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Key enzymes of glycolysis and the TCA cycle exhibit nuclear activity.

(A)Activity of hexokinase (HK), citrate synthase (CS), and pyruvate kinase (PK), glycerol kinase (GK) and pyrroline-5-carboxylate dehydrogenase (P5CDH) were obtained using either nuclear and cytoplasmic extracts. The controls used here were not found in chromatin and nuclear proteomes. Reactions without the enzyme substrates were used to rule out residual activity. Enzymatic activities were normalized with the total activities of parasite extract for each listed enzyme. Means ± SD of triplicate data (n = 3) are graphically represented. *p < 0.01; **p < 0.001; *** p < 0.0001. One-way ANOVA (Bonferroni’s Multiple Comparison Test). (B) KEGG map of the glycolysis and TCA cycle pathways. (C) Immunofluorescence analysis of hexokinase (HK), citrate synthase (CS), and glycerol kinase (GK) in T. cruzi epimastigotes expressing H2B.V-Myc as a nuclear marker. HK, CS, and GK were detected with Alexa Fluor 488-conjugated secondary antibody, H2B.V-Myc was detected with Alexa Fluor 555-conjugated secondary antibody, and DNA was stained with DAPI. Images were acquired using an Olympus BX51 fluorescence microscope controlled by Olympus cell^R software and processed/assembled using ImageJ. Bright-field, individual fluorescence channels, and merged images are shown. White arrows indicate the linear regions of interest drawn across the parasite body and nuclear region.

https://doi.org/10.1371/journal.pntd.0013299.g005

To determine whether they are enzymatically active inside the nucleus, we evaluated the activity of three key metabolic enzymes, hexokinase, pyruvate dehydrogenase E1 component alpha subunit, and citrate synthase,which play essential roles in cellular energy metabolism. Hexokinase is canonically located at T. cruzi glycosomes, while the other two at mitochondrion. We assessed their activity in total, cytoplasmic, and nuclear extracts from asynchronous epimastigotes. Of note, the availability of standardized enzymatic assays for these enzymes was also a key factor for their selection. As negative controls for nuclear activity, we also assayed glycerol kinase (GK) and pyrroline-5-carboxylate dehydrogenase (P5CDH), which were not detected in our nuclear proteome and are primarily associated with glycosomal and mitochondrial metabolism, respectively.

Our analysis confirmed the nuclear activity of hexokinase (TcCLB.508951.20 and TcCLB.510121.20), pyruvate kinase (TcCLB.507993.390), and citrate synthase (TcCLB.509801.30) (Fig 5B). In contrast, P5CDH and GK -both negative controls, as expected, displayed no activity in the nuclear extracts. Since enzymatic activity in nuclear fractions does not define localization, we performed immunofluorescence microscopy as an orthogonal validation approach. We focused on HK and CS, two enzymes detected in the nuclear proteome and enzymatically active in nuclear fractions (Fig 5B). As antibodies specifically raised against the corresponding T. cruzi proteins were not available, we used antibodies raised against the T. brucei orthologs. Prior to immunofluorescence, we confirmed high similarity between the T. brucei antigens and their T. cruzi counterparts by BLAST analysis, and western blotting detected main bands compatible with the expected molecular masses of the target proteins, supporting their use in T. cruzi (S5 Fig). GK was validated in parallel and used as a non-nuclear metabolic control.

To define the nuclear compartment, we performed immunofluorescence assays in parasites expressing tagged H2B.V [29], which was used as a nuclear/chromatin-associated marker together with DAPI staining. HK showed the expected punctate extranuclear distribution, compatible with its canonical glycosomal localization, but also displayed a detectable signal overlapping or closely associated with the H2B.V/DAPI-positive nuclear region (Fig 5C, upper panel). CS showed a strong extranuclear signal, consistent with its expected mitochondrial localization, together with a reproducible nuclear-associated signal overlapping the H2B.V-marked compartment (Fig 5C- middle panel). In contrast, GK showed a predominantly extranuclear distribution and did not display comparable enrichment within the H2B.V/DAPI-positive nuclear region (Fig 5C- low panel). Fluorescence intensity line profiles across individual cells further supported these localization patterns: HK and CS showed signal peaks that partially coincided with the H2B.V/DAPI-positive nuclear region, whereas GK displayed a distinct extranuclear profile with limited overlap with the nuclear marker (S6 Fig). These nuclear-associated signals were observed throughout the Z-stack optical sections, indicating that they were not confined to a single focal plane (S7 Fig) and supporting the conclusion that HK and CS, despite their predominant cytoplasmic localization, are also present in the nucleus.

These results support the presence of selected energy metabolism enzymes associated with the nuclear compartment and demonstrate measurable enzymatic activity in nuclear fractions (S8 Fig). Together with the immunofluorescence data, they argue against a model in which all metabolic enzymes detected in nuclear preparations represent passive contaminants generated during fractionation. Rather, the data suggests that at least a subset of these enzymes may be specifically associated with the nucleus in T. cruzi, although definitive assignment of nuclear localization for individual proteins will require further experimental validation.