The cell lines Tb HD82-dKO, Tb HD82-cdKO, Tb HD82-OE and Tb HD82-myc were generated for both the T. brucei single-marker bloodstream form (BF) (Wirtz et al., 1999) and the T. brucei 449 procyclic form (PF) (Estevez et al., 2001). Bloodstream trypanosomes were cultured at 37°C and 5% CO₂ in HMI-9 with 10% (v/v) fetal bovine serum (FBS) and procyclic parasites were cultured at 28°C in SDM-79 supplemented with 10% FBS and 7.5 µg/ml hemin.

Parasites were transfected by electroporation in cytomix buffer for bloodstream cells and Zimmerman buffer for procyclic cells, as previously described (Castillo-Acosta et al., 2008). Clones were selected with appropriate selection drugs at the following concentrations: puromycin (Sigma): 0.1 µg/ml (BF) and 1 µg/ml (PF); hygromycin (Sigma): 5 µg/ml (BF) and 50 µg/ml (PF); phleomycin (Sigma): 2.5 µg/ml (BF) and blasticidin (Invitrogen): 5 µg/ml (BF) and 10 µg/ml (PF). Expression of the ectopic copy in TbHD82-OE and TbHD82-myc was induced using 1 µg/ml of doxycycline (DOX) (Sigma).

The TbHD82 alleles replacement (Tb HD82-dKO) was performed by transfection of two gene cassettes containing the blasticidin S deaminase (BSD) or hygromycin phosphotransferase (HYG) resistance markers.

Overexpression of both the native TbHD82 and a c-myc tagged fusion protein (TbHD82-myc) were achieved by transfection with pGRV190 and pGRV191 plasmids, respectively, linearized with NotI.

Replacement of the two alleles of both BF and PF was attained by transfection with two gene cassettes containing the BSD or HYG resistance markers. These cassettes were generated by PCR amplification of a 5’-UTR fragment of 502 bp (position 330-831) and a 3’-UTR fragment of 340 bp (position 281-620) of the regions flanking the open reading frame of the TbHD82 gene using wild type T. brucei genomic DNA as template. Four specific primers were designed taking into account the sequence of Tb427.06.2900 present in the Gene DB database: 5’-CGC GGC CGC AAC TGA GAG GGT GAT TGG CG-3’ (NotI restriction site underlined) and 5’- GCC TCG AGT CGT GTG ATT TGC TAA CGC-3’ (XhoI) for the 5’-UTR region, and 5’-GCA GGC CTG TGC GCA TGT ACG TTG TAG C-3’ (StuI) and 5’-GGC TAG CGC GGC CGC CTA TCG TCC AAG TGT TTC TA-3’ (NheI and NotI) for the 3’-UTR region.

The 5’-UTR and 3’-UTR regions of the TbHD82 were introduced as flanking sequences of the BSD resistance gene in the pHD887 plasmid kindly provided by Christine Clayton (ZMBH, Heidelberg, Germany), yielding the construct pGRV188. The second cassette was generated by replacement of the BSD gene in the pGRV188 plasmid with the HYG selectable marker coming from SnaBI-digested pGR19 (Clayton et al., 2005), giving the construct pGRV189.

Studies on the intracellular localization of TbHD82 localization were performed in BF and PF cell lines that overexpress the TbHD82 native protein (Tb HD82-OE) or a TbHD82-c-myc fusion protein (Tb HD82-myc). The TbHD82 coding sequence (Tb427.06.2900) was amplified by PCR using wild type T. brucei genomic DNA as template and the specific primers: 5’-GCA TTA ATA TGG AAG GGG AAC TTG CCT TC-3’ (AseI restriction site underlined) and 5’-GCG GAT CCT TAG CGG CTG CGT TTA TTT CC-3’ (BamHI) for the native form and 5’-GCA TTA ATA TGG AAG GGG AAC TTG CCT TC-3’ (AseI) and 5’-GCG ATA TCG CGG CTG CGT TTA TTT CCC GC-3’(EcoRV) lacking a stop codon for expression of the c-myc-fusion protein. The fragment corresponding to the native TbHD82 was subcloned in pGRV23b (Castillo-Acosta et al., 2012b) after NdeI and BamHI digestion yielding pGRV190 plasmid. In the case of the c-myc fusion protein, the PCR fragment was subcloned in pGRV33 (Castillo-Acosta et al., 2012b) previously digested with NdeI and HpaI resulting in the pGRV191 plasmid containing the carboxy-terminal additional amino acid sequence SKGKVNEEQKLISEEDL* (c-myc sequence is underlined, the asterisk indicates a stop codon). Both the pGRV190 and pGRV191 plasmids allow for regulated expression by a tetracycline-inducible PARP promoter and puromycin (PAC resistance gene) was used as drug for selection.

For western blot analysis, 5 x 10⁶ cells were harvested, centrifuged and washed in 1 × PBS (137 mM NaCl, 4 mM Na₂HPO₄, 1.7 mM KH₂PO₄ and 2.7 mM KCl). Pellets were resuspended in urea buffer (6 M urea, 10 mM Na₂HPO₄, 1% β-mercaptoethanol, pH 7) and loading sample buffer (67.5 mM Tris-HCl, pH 6.8, 3% SDS, 10% glycerol, 5% β-mercaptoethanol) in a 3:1 ratio, and boiled for 5 min at 99°C. Samples were subjected to electrophoresis in SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes, which were incubated with a rabbit polyclonal anti-TbHD82 (1:10,000) antibody generated against recombinant TbHD82 (Yague-Capilla et al., 2021). The mouse monoclonal anti-Tbβ-tubulin (1:10,000 [Sigma]) antibody was used as loading marker. Goat anti-rabbit IgG (1:5,000) or goat anti-mouse IgG (1:3,000) HPRT-conjugated antibodies (Promega) were used as secondary antibodies and films were developed using the ECL enhanced chemiluminescence detection kit (Amersham Pharmacia Biotech).

Immunofluorescence (IF) studies were performed for intracellular localization as well as the quantification of γH2A phosphorylation. Colocalization analysis was performed in BF and PF cell lines. Briefly, 5 x 10⁶ parasites were harvested and incubated in 1 ml of culture medium without FBS supplemented with 10 µM (BF) and 100 µM (PF) MitoTrackerVR Red CMXRos for 15 min at 37°C (BF) or 28°C (PF), followed by incubation in fresh culture medium without FBS and washing twice with 1 ml of PBS. In the case of γH2A phosphorylation IF, cells were directly washed and then the parasites were fixed in 4% p-formaldehyde (PFA) in wash solution (PBS, 0.2% Tween 20) on a poly-l-lysine coated slide for 20 min. Fixed parasites were washed once in wash solution and permeabilized and blocked with 1% of IGEPAL in blocking solution (wash solution containing 1% blocking reagent [Roche]) during 75 min. Samples were incubated with the corresponding primary antibody; anti-TbHD82 polyclonal antibody (1:100 diluted in blocking solution), monoclonal anti-c-myc (Sigma, 1:100 diluted in blocking solution), or the anti-TbγH2A polyclonal antibody (Life Technologies, 1:100 diluted in blocking solution). Samples were washed six times for 5 min and then labeled with secondary antibodies for 1 h at room temperature: goat Alexa Fluor 488 anti-rabbit (Sigma, 1:40 diluted in blocking solution) or goat Alexa Fluor 488 anti-mouse (Sigma, 1:40 diluted in blocking solution). Preparations were dehydrated in methanol for 1 min, mounted and stained with ProLong Gold Antifade Reagent with DAPI (Life Technologies). Vertical stacks of up to 40 slices (0.2 µm steps) were captured using an inverted Leica DMi8 microscope, 100x objective, and LASX software. Images were deconvolved and pseudo-colored with Huygens Essential software (version 3.3; Scientific Volume Imaging). Colocalization analysis was carried out using the JACoP plugin of the Fiji software (version 1.53q; ImageJ). For 3D images and videos, vertical stacks were captured using a sp8 confocal spectral Leica microscope, 63x objective, with hyvolution technology. Images were processed, rendered and clipped with LASX software (version 3.5.6).

The four canonical dNTPs (dTTP, dCTP, dATP and dGTP) were quantified in Tb BF, Tb BF HD82-dKO and Tb BF HD82-conditional knockout (HD82-cdKO; see Figure S3 ) cells in the absence or presence of DOX using a modified DNA polymerase assay (Horowitz et al., 1997). The oligonucleotides 5’-TTT ATT TAT TTA TTT ATT TAG GCG GTG GAG GCG G-3’, 5’-AAA CAA ACA AAC AAA CAA ACG GCG GAG GAG GCG G-3’, 5’-AAA GAA AGA AAG AAA GAA AGG GCG GAG GAG GCG G-3’, 5′-AAA TAA ATA AAT AAA TAA ATG GCG GTG GAG GCG G-3′ were employed as template for dTTP, dGTP, dCTP and dATP measurements respectively, and the oligonucleotide 5’-CCGCCTCCACCGCC-3’ as primer. Parasites (2 x 10⁶ cells per sample) were harvested by centrifugation and washed once in PBS. Cells were incubated with 250 µl of cold 60% methanol in PBS overnight at -20°C. The suspension was heated for 5 min in boiling water bath, followed by centrifugation for 20 min at 17,000 × g. The supernatant was moved to a fresh tube and dried in a Speed-Vac (Thermo Fischer Scientific). Residues obtained this way were then dissolved in 100 µL of reaction mixture containing 32 nM DNA template, 32 nM DNA primer, NEBuffer 2 (New England BioLabs), 0.3 units of DNA polymerase I Klenow fragment (New England Biolabs) and 0.0032 µCi/µl of the corresponding [³H]deoxynucleotide ([³H]dATP (PerkinElmer) for dTTP measurements and [³H]dTTP (Moravek) for dCTP, dATP and dGTP measurements). Samples were incubated for 15 min at 25°C. The polymerization was stopped by adding 10 mM EDTA and heating at 75°C for 20 min. DNA was then precipitated with 10% (v/v) trichloroacetic acid for 30 min at 4°C. The solution was filtered with Glass Microfibre Filters GF/C (Whatman) and the filters were washed under a vacuum with 30 ml of 5% trichloroacetic acid, rinsed with 3 ml of ethanol and dried. Radioactivity was measured in a LS 6500 Multi-Purpose Scintillation counter (Beckman Coulter). Data were interpolated in standard curves obtained using different concentrations of the corresponding dNTP.

For FACS analysis, 1.5 x 10⁶ parasites were harvested by centrifugation, washed twice with TDB-g buffer (5 mM KCl, 80 mM NaCl, 1 mM MgSO₄, 20 mM Na₂HPO₄, 20 mM glucose, pH 7.4) and permeabilized with 50 µl saponin diluted in TDB-g (0.05% saponin for BF and 0.1% saponin for PF parasites). Cells were incubated at 37°C (BF) and 28°C (PF) for 3 min and the reaction was stopped by adding 450 µl of 10 µg/ml RNase A diluted in TDB-g. Samples were incubated at room temperature for 30 minutes and then propidium iodide was added to a final concentration of 20 µg/ml. The content of the tubes was transferred to special cytometer tubes and analyzed in a FACSymphony A3 SE Cell Analyzer using BDFacsDiva version 8 (Becton Dickinson). Percentages of each phase of the cell cycle were analyzed with FlowJo version 10.7.1.

The rate of reversion to the TK^- phenotype in PF cells transfected with the viral TK gene was measured by the Luria-Delbrück fluctuation test. Stock cultures were prepared at a density of 10³ cells/ml (log phase). The cultures were split in 96-well plates by adding 200 parasites per well. Cultures were incubated for 6 days until the density reached 5-10 x 10⁶ parasites/ml, then cultures were transferred to 24-well plates with 2.5 ml containing 100 µM of ganciclovir (Sigma) and the plates were sealed. Cultures were left growing during 8 days and then cell density was determined. The mutation rate was calculated using the equation described by Valdes et al. (1996) :

where P₀ is the proportion of wells without cell growth and N is the final number of parasites per culture upon addition of ganciclovir.

In order to establish the mutation spectra, genomic DNA of independent TK^- revertants was isolated using DNAzol (Invitrogen) and the TK gene amplified by PCR. The oligonucleotides VCA19 (5’-GAG CCG ATG CTT TTG ACA TGT TAG-3’) and VCA20 (5’-CTT GTG CGC TGT ACG TAA ATG TGT TGC-3’) were used to sequence the TK gene of the different revertant clones.

Levels of DNA fragmentation were measured by TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling) using the In situ Cell Death Detection Kit, Fluorescein (Roche). For each sample, 1.5 x 10⁷ parasites were harvested by centrifugation and washed twice with ice cold PBS. Cells were fixed by incubating in 2% p-formaldehyde for 45 min at room temperature, washed twice with PBS, and permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate for 2 min at 4°C. Samples were then stained with the TUNEL reaction mix (with fluorescein-dUTP) at 37°C for 1 h. FL1-H fluorescence was measured on 10,000 events in a FACSymphony A3 SE Cell Analyzer (Becton Dickinson). Percentages of TUNEL positive cells were analyzed with BDFacsDiva version 8 (Becton Dickinson).

Alignment of the amino acid sequences was performed with the online tool ClustalOmega (EMBL’s European Bioinformatics Institute). The results were transferred to JalView version 2.11 and coincident amino acids were colorized applying Clustal style.

Data were represented as the mean ± SD for each group. SPSS Statistics and GraphPad Software were used for the comparison of data sets. The student’s t test was used in the analysis when two sets of data were compared. Either the two-way ANOVA or the Dunnet’s post hoc tests were used to calculate variances of data sets with normal distribution. Kolmogorov−Smirnov and Shapiro−Wilk tests were applied to verify data normality, and the equality of variances was analyzed with the Levene’s test. A p value ≤ 0.05 was considered statistically significant. To identify differences between populations with categorical data with n < 300 cells, a Fisher’s exact test was applied, whilst a Chi-square test was used when n > 300.

We previously identified Tb927.6.2900 as a member of the HD-domain/phosphodiesterase (PDEase)-like superfamily of proteins which has a total of 13 members in T. brucei (5 members belong to the HD family, and 8 are classified as PDEases) and designated it as TbHD82 taking into account its molecular mass (Yague-Capilla et al., 2021). PANTHER (Protein ANalysis Through Evolutionary Relationships; http://www.pantherdb.org) suggests that Tb927.6.2900 is a deoxynucleoside triphosphate triphosphohydrolase. TbHD82 has a theoretical Mw of 81,734 Da consisting of 735 amino acids with a calculated pI of 7.53 encoded by a single-copy gene located on chromosome 6. According to the Alphafold2 prediction (Jumper et al., 2021; Varadi et al., 2022), TbHD82 secondary structure, showed in Figure S1 , contains two disordered regions (amino acids 641-675 in purple and 713-735 in yellow) with no predicted function and a central region (amino acids 216-366 in red) containing the HD domain conserved within the protein superfamily.

Immunofluorescence analysis was performed in different cell lines in order to identify the intracellular localization of TbHD82. For this purpose, we generated BF and PF cell lines with a tetracycline-controlled inducible expression system that allowed for overexpression of the protein, named as Tb BF HD82-OE and Tb PF HD82-OE. We generated an affinity-purified rabbit polyclonal anti-TbHD82 antibody which was used for determination of enzyme levels upon induction (Yague-Capilla et al., 2021). This antibody specifically recognizes TbHD82 in western blots of parasite extracts ( Figure S2E ). Additionally, we transfected BF and PF cells with a construct encoding a TbHD82-myc fusion protein in order to further evaluate the localization of the protein by using a monoclonal antibody directed against the c-myc epitope.

Immunoblots of T. brucei cell lysates after 2 days of induction showed that anti-TbHD82 recognized the wild-type (WT) TbHD82 protein as well as the endogenous C-terminally myc-tagged version of TbHD82 in both BF and PF ( Figure S2 ).

The anti-TbHD82 antibody labeling revealed a nuclear localization in both BF and PF parental parasites ( Figure 1A , lines 1 and 3). Cell lines which overexpress the untagged protein, Tb BF HD82-OE and Tb PF HD82-OE ( Figure 1A , lines 2 and 4) also exhibit colocalization with nuclear DAPI staining. Mitotracker red staining was additionally performed to discard presence of the protein in the mitochondrion. Figure 1B and Supplementary Videos show the 3D rendering of nucleus and kinetoplast of parasites from the cell lines described above, extracted from the confocal stack.

A colocalization analysis with the Alexa Fluor 488 and DAPI signals was performed for n = 10 cells in each cell line. Pearson’s coefficients of 0.72 ± 0.06, 0.72 ± 0.07, 0.61 ± 0.08 and 0.73 ± 0.05 were obtained for the Tb BF, Tb BF HD82-OE, Tb PF and Tb PF HD82-OE and cell lines, respectively, hence supporting the nuclear localization.

It is known that certain proteins involved in dNTP homeostasis are regulated throughout the cell cycle (Chang et al., 1998; Valente et al., 2016). In the case of human SAMHD1, its function is regulated mainly by two factors: phosphorylation/dephosphorylation and changes in expression (Cribier et al., 2013; Schott et al., 2018). We sought to analyze if expression of TbHD82 is also regulated. For this purpose, levels of TbHD82 in more than 200 cells in both BF and PF of parental and overexpressing the TbHD82-myc fusion protein were measured by immunofluorescence. Cells were classified according to the number of nuclei and kinetoplasts (1N1K, 1N1K*, 1N2K and 2N2K). The quantification of the signal was performed with Fiji software (version 1.53q; ImageJ) (Schindelin et al., 2012), evaluating the intensity of the nucleus of the parasites. In all cases TbHD82 is present throughout the cell cycle yet increased expression was evidenced in 1N1K* cells, which correspond mainly to early S phase cells (Woodward and Gull, 1990) ( Figure 2 ). Colocalization analysis with the Alexa Fluor 488 and DAPI signals was also performed for parasites expressing the c-myc fusion protein. Pearson’s coefficients obtained were 0.71 ± 0.03 and 0.76 ± 0.04 for Tb BF HD82-myc and Tb PF HD82-myc, respectively.

In order to decipher the role of TbHD82 in proliferation and nucleotide homeostasis in T. brucei, we generated TbHD82 knockout mutant (Tb HD82-dKO) cell lines in both BF and PF parasites. TbHD82 null mutants were obtained through the replacement of two alleles with cassettes containing the BSD or HYG resistance markers ( Figure S3A ). Transfectants were verified by PCR, using primers of the selectable markers and flanking genes upstream and downstream the HD82 gene in addition to specific primers of the HD82 gene open reading frame, which confirmed the absence of both alleles coding for TbHD82 ( Figures S3B, C ). The determination of protein levels in knockout cells was performed by western blotting using the anti-TbHD82 rabbit polyclonal antibody ( Figures 3B, D ). Immunoblots of T. brucei cell lysates showed that the anti-TbHD82 antibody recognized the wild-type (WT) protein (81.73 kDa) in BF and PF cells, while no staining was detected in knockout cells. Tb HD82-dKO cells do not exhibit modifications in the levels of TbHD52 thus discarding participation of the mitochondrial orthologue in possible compensatory mechanisms involved in modulation of dNTP pools in null mutants ( Figure 3G ).

In order to evaluate parasite proliferation, cultures were initiated at 3 x 10³ (BF) or 2 x 10⁵ (PF) parasites/ml and monitored during 18 days by counting in a Z1 Coulter counter. Cultures were diluted every 2 days to the initial density and maintained in the absence of the antibiotics previously used to generate the KO cell lines. Both the BF and PF HD82 null mutant cell lines showed a cell density that was slightly reduced compared to that of the parental strains ( Figures 3A, C ).

To validate observations obtained with the knockout lines, we generated a conditional knockout cell line containing a doxycycline-inducible ectopic copy of the HD82 gene (HD82-cdKO) with puromycin (PAC) as a resistance marker for both BF and PF ( Figure S3A ). The conditional knockout allows for doxycycline-mediated control of the expression of HD82 in the null mutant. While the increased levels of protein observed upon induction of the ectopic copy (5-fold and 6-fold in Tb BF HD82-cdKO and Tb PF HD82-cdKO cells, respectively) do not result in pronounced cytotoxicity ( Figure S4 ), the growth defects observed in the HD82-dKO cell line were not efficiently reverted upon overexpression. Thus, conditional knockout cell lines ± DOX exhibit growth profiles comparable to HD82-dKO cells.

Since members of the deoxynucleoside triphosphate triphosphohydrolase family have a role in maintaining dNTP homeostasis by hydrolyzing all four canonical dNTPs (Franzolin et al., 2013), we aimed to elucidate whether HD82 shares this function with its human homologue. For this purpose, we performed polymerase-based dNTP quantification as previously described (Castillo-Acosta et al., 2008; Requena et al., 2014). In the absence of TbHD82, a significant expansion of the dATP, dCTP and dTTP pools was observed while the levels of dGTP remained unchanged. These alterations in null mutants were observed in a similar fashion in both BFs and PFs and were mostly reverted when the expression of the ectopic copy of the gene was induced ( Figures 3E, F ). Thus, dATP, dCTP and dTTP were increased 3.2, 1.8 and 1.3 -fold, respectively, in BF null mutants and 7.8, 1.6, and 2.5-fold, respectively, in Tb PF HD82-dKO cells. Furthermore, DOX-induced expression of an ectopic copy of TbHD82 resulted in certain instances in lower dNTP levels ( Figures 3E, F ), and this dysregulation might be responsible for the defects in proliferation observed. Hence, the data suggests that TbHD82 is a major player in controlling dATP, dCTP and dTTP levels.

Correct dNTP homeostasis is essential for adequate DNA replication and integrity (Kohnken et al., 2015). Nucleotide requirements differ during cell cycle progression and impaired dNTP levels at the onset of S phase can activate replication stress signaling (Forey et al., 2020). To investigate whether the loss or overexpression of TbHD82 can influence cell cycle progression, we performed a FACS analysis and measured N/K patterns in order to establish the percentage of cells in the G1, S phase, late S/G2 phase and mitosis as previously described (Marin et al., 2018). The data summarized in Figure 4A show that FACS analysis evidenced a small yet significant reduction in S phase and an increase in G1 cells in Tb BF HD82-dKO mutants. Furthermore, DAPI staining ( Figure 4B ) clearly illustrates an accumulation of 1N1K (G1 phase cells) and a decrease in 1N1K* (S phase cells) and 2N2K (G2/M phase) cells suggesting defects in the progression through S phase as a consequence of the absence of HD82 and the subsequent imbalance in dNTP pools.

It is well documented that establishing dNTP levels within a range that is optimal for adequate chromosomal replication is a crucial factor in maintaining genome stability (Pai and Kearsey, 2017). Since TbHD82 is important for dNTP homeostasis, we aimed to elucidate its potential role in genomic integrity. First, we performed a mutation rate assay based on the Luria-Delbrück test as described in Valdes et al. (1996) (Valdes et al., 1996; Castillo-Acosta et al., 2012b). For this assay we used Tb PF and Tb PF HD82-dKO cell lines expressing the thymidine kinase (TK) gene from Herpes simplex type 1 virus (HSV-1). As shown in Table 1 , a 3.8-fold increase in the mutation rate was obtained for Tb PF HD82-dKO compared to Tb PF. The mutation spectra ( Table 2 , Figure 5 ) evidenced a notable increase in all types of mutations although GC to CG transversions (G→C and C→G), GC to AT transitions (G→A and C→T) and deletions were the most significantly increased (8.9, 4.4 and 7.9-fold, respectively).

Subsequently, we investigated the presence of DNA repair foci in both parental and knockout BF cells by immunofluorescence ( Figure 4C ). The Tb BF HD82-cdKO cell line was also included in the analysis. For this purpose, we quantified the phosphorylation of Thr130 of trypanosomal histone H2A (known as γH2A), an early event that takes place in the DNA damage response involved in double strand break (DSB) repair (Glover and Horn, 2012). The percentage of γH2A positive cells was determined by fluorescence microscopy, evidencing a significantly increased staining for Tb BF HD82-dKO cells compared to the parental cell line (37% vs 24%, respectively), while induction of enzyme expression in the Tb BF HD82-cdKO mutant for 2 days reduced the number of positive cells from 42% in uninduced to 32% in induced cells ( Figure 4D ). The differential expression pattern of γH2A among cells (one focus, two or more foci or whole nucleus) was also determined and revealed that cells with whole nuclear staining were more abundant when TbHD82 was absent ( Figure 4E ). Likewise, when the cell cycle is considered (by DAPI staining), positive cells were mainly in the G1 and S phases which agrees with the concept that major DSB occurs during DNA replication and that repair probably occurs predominantly during the late S or G2 phases through the homologous recombination (HR) pathway (Marin et al., 2018). In agreement with the postulated defects in progression through the S phase, the null mutant exhibits a significant increase in G1 phase positive cells; 27% and 22% for Tb BF HD82-dKO and Tb BF HD82-cdKO -DOX vs 10% and 9% for Tb BF and Tb BF HD82-cdKO +DOX cells, respectively ( Figure 4F ). Furthermore, Figure 4G shows the percentage of cells in each phase with discrete foci or whole nucleus pattern, clearly indicating that whole nucleus marked cells are predominantly found in late S phase. On the other hand, positive cells in early S cells mainly show discrete foci. In summary, defects in cell cycle progression and increased damage are found upon elimination of TbHD82 in BF cells.

Previous studies have suggested that dNTP triphosphohydrolases such as human SAMHD1 could be involved in repair mechanisms and recruiting repair factors to DSB sites (Daddacha et al., 2017; Coquel et al., 2018). Moreover, in human cells, SAMHD1 has been shown to have an important role in the response and recovery from stress when exposed to ROS (Kapoor-Vazirani et al., 2022). Hence, we aimed to evaluate the behavior of TbHD82 upon genotoxic insult. Once inside the mammalian host, trypanosomes are in contact with oxidative and genotoxic stress as a consequence of the immune response (Mabbott and Sternberg, 1995; Figarella et al., 2018). We first evaluated the proliferation of TbHD82 null mutants upon exposure to the genotoxic agents doxorubicin (DXR), diethylenetriamine-nitric oxide (DETA-NO, an exogenous nitric oxide donor), and the oxidizing agent hydrogen peroxide (H₂O₂). Subtle yet significant increases in EC₅₀ ( Table 3 ) were denoted in knockout compared to parental cells after treatment with DXR ( Figure S5A ) and H₂O₂ ( Figure S5C ) while no significant changes in EC₅₀ values were observed with DETA-NO ( Figure S5B ). DXR is a DNA intercalating agent that causes breakage of DNA strands and inhibition of both DNA and RNA synthesis. DXR inhibits the enzyme topoisomerase II, causing DNA damage and induction of apoptosis (Wassermann et al., 1990). Despite the minor changes in EC50 values, further TUNEL analysis revealed that Tb BF HD82-dKO cells treated with 0.03 µM DXR exhibit significantly increased damage (2.3-fold) compared to parental cells ( Figure 6D ). We subsequently performed treatment and recovery assays in Tb BFs; cells were treated for 48 h with either 0.03 µM DXR or 25 µM DETA-NO and then cultures were washed to remove the genotoxic agent and allowed to recover for 72 h. Interestingly increased TbHD82 expression was observed during recovery for both DXR (5.2-fold) ( Figures 6A, C ) and DETA-NO (9.0-fold) ( Figures 6B, C ).

This observation was analyzed in detail in the case of DXR exposure. First, when DNA repair was quantified by determining γH2A, we observed that parental cells 48 h after wash-out, present a similar number of γH2A positive cells than prior to treatment indicating effective recovery ( Figure 6E ). On the other hand, in null mutants recovery is less efficient suggesting that the absence of HD82 is detrimental to effective repair after genotoxic insult. It must be noted that during recovery of parental cells (16 h) there is a significant increase in the number of cells in S phase (observed by DAPI staining) undergoing active replication which might partially explain the increased levels of TbHD82 whose expression, as previously indicated, is maximum in S phase ( Figures 6F , S6A ). When dNTP levels were measured in Tb BF cells treated with DXR and upon wash-out, surprisingly, increased TbHD82 does not result in diminished dNTPs; rather dCTP is somewhat increased upon DXR treatment. In null mutants, dNTP levels remain increased during treatment and recovery ( Figures 6G , S6B ).