Deviating the level of proliferating cell nuclear antigen in Trypanosoma brucei elicits distinct mechanisms for inhibiting proliferation and cell cycle progression

Abstract

The DNA replication machinery is spatially and temporally coordinated in all cells to reproduce a single exact copy of the genome per division, but its regulation in the protozoan parasite Trypanosoma brucei is not well characterized. We characterized the effects of altering the levels of proliferating cell nuclear antigen, a key component of the DNA replication machinery, in bloodstream form T. brucei. This study demonstrated that tight regulation of TbPCNA levels was critical for normal proliferation and DNA replication in the parasite. Depleting TbPCNA mRNA reduced proliferation, severely diminished DNA replication, arrested the synthesis of new DNA and caused the parasites to accumulated in G2/M. Attenuating the parasite by downregulating TbPCNA caused it to become hypersensitive to hydroxyurea. Overexpressing TbPCNA in T. brucei arrested proliferation, inhibited DNA replication and prevented the parasite from exiting G2/M. These results indicate that distinct mechanisms of cell cycle arrest are associated with upregulating or downregulating TbPCNA. The findings of this study validate deregulating intra-parasite levels of TbPCNA as a potential strategy for therapeutically exploiting this target in bloodstream form T. brucei.

Abbreviations

Mcm

mini-chromosome maintenance proteins

PCNA

proliferating cell nuclear antigen

Tb

Trypanosoma brucei

Hs

Homo sapiens

Sc

Saccharomyces cerevisiae

Sp

Schizosaccharomyces pombe

CDK

cyclin dependent kinase

Gadd

growth arrest and DNA-damage

MyD

myeloid differentiation primary response gene

Orc

origin recognition complex

Cd

Cluster of differentiation

GINS

Go, Ichi, Nii, complex

RT-PCR

reverse transcriptase-polymerase chain reaction

EdU

5-Ethynyl-2′deoxyuridine

DAPI

4′, 6-diamidino-2-phenylindole

HU

hydroxyurea

H₂O₂

hydrogen peroxide

Introduction

Trypanosoma brucei is the protozoan parasite that causes African trypanosomiasis, also known as sleeping sickness. The bloodstream form of this vector-borne parasite proliferates in human blood and cerebrospinal fluid, where it causes a meningoencephalitic disease during late stage infections that is most often fatal if left untreated.¹ DNA replication in T. brucei, as in all other organisms, is spatially and temporally coordinated to produce a single exact copy of the genome per cell division. In human cells, DNA replication is initiated at origins that become licensed by the components of the origin recognition complex: (Orc1-6), Ctd1, Cdc6, and the mini-chromosome maintenance proteins (Mcm 2–7) that form the pre-replication complex in G1.^2,3 The pre-replication complex in trypanosomatids consists of homologs for Orc1/Cdc6 proteins and a Cd45-Mcm2-7-GINS complex.^4-6 Activation of replication origins occurs at S phase and involves formation of replication forks that become primed by the DNA polymerase α/primase complex. This allows the replication machinery to bind to activated forks and begin synthesis of new DNA.^7,8 At S phase, components of the DNA replication machinery localize into punctate foci in the nucleus that cover several replication forks or replicons.⁹

Proliferating cell nuclear antigen (PCNA) is a central component of the replication machinery originally identified in patients with systemic lupus erythematosus.¹⁰ PCNA acts as a homotrimeric DNA sliding clamp that provides DNA polymerases δ and ε with the processivity necessary to duplicate the entire genome. It functions as a moving platform that complexes with many proteins to process signals important for directing DNA replication and repair pathways, which regulate cell cycle events.^11-13 PCNA was also identified as a cyclin based on its cyclical pattern of expression in mammalian cells.^14,15 Quaternary complexes in mammalian cells including CDKs, p21^WAF1(CIP1 and other cyclins can regulate the cell cycle by interacting with PCNA.^16-18 Upregulation of PCNA occurs in cancerous cells allowing it to serve as a reliable diagnostic biomarker for predicting oncogenesis.^19-21 Ligand-induced apoptosis is prevented by interactions between PCNA and negative regulators of growth such as MyD118 and Gadd45.¹² This indicates that upregulating PCNA in mammalian cells may sequester negative regulators of growth that signal apoptosis or cell cycle check points.²²

A recent study used TbPCNA to establish the temporal and spatial patterns of DNA replication in T. brucei and reported differential regulation of TbPCNA during the cell cycle.²³ We examined the consequences that deregulating intra-parasite TbPCNA levels had on proliferation and DNA replication in bloodstream form T. brucei. This study demonstrates that either depleting or overexpressing TbPCNA severely reduces proliferation and DNA replication in the parasite. The parasites were more sensitive to increased intra-parasite levels of TbPCNA than they were to depleted levels. Overexpressing human PCNA in T. brucei arrested its proliferation with similar efficiency as overexpressing TbPCNA. Finally, down regulation of TbPCNA hypersensitized T. brucei to the replication inhibitor hydroxyurea. These findings reveal the important roles that proper levels of TbPCNA play in maintaining proliferation and DNA replication in T. brucei. They also suggest that overexpressing TbPCNA arrests cell cycle progression by a different mechanism(s) than depleting it. This study also validates deregulating intra-parasite levels of TbPCNA as a viable strategy for therapeutic exploitation of T. brucei.

Results

Endogenous expression of TbPCNA in T. brucei

We subcloned the TbPCNA coding region into a modified version of the pLew111 expression vector ^24,25 that included a hemagglutinin tag (HA). Figure 1A diagrams the strategy for generating clones that express HA-tagged TbPCNA at endogenous levels in the 221strain of wild type T. brucei. Selection with phleomycin identified 6 resistant transfectants. Immunoblot analysis was done to identify phleomycin-resistant transfectants that expressed HA-tagged TbPCNA (TbPCNA_HA). Expression of the endogenously tagged TbPCNA using this strategy was contingent on the plasmid being successfully integrated into one of the TbPCNA chromosomal loci by a homologous recombination event. Positive transfectants identified by immunoblot assays expressed a polypeptide with a predicted molecular mass of 35 kDa (32.3 kDa plus the 3 kDa HA-tag). Figure 1B demonstrates a representative stable transfectant that expressed TbPCNA_HA at endogenous levels.

Figure 1.

TbPCNA localizes to the nucleus and cytoplasm in bloodstream form T. brucei. (A) Scheme for endogenously tagging TbPCNA. (i) TbPCNA-HA vector linearized with NruI, which cuts at a unique site within the TbPCNA-HA cDNA. (ii) Linearized plasmid was stably transfected into T. brucei where a homologous recombination event occurred at the TbPCNA locus. (iii) Representation of the heterozygous TbPCNA locus in clones that express endogenously tagged TbPCNA_HA. (B) Immunoblot with antibodies against hemagglutinin (HA) detected clones that expressed endogenously tagged TbPCNA_HA. VSG 221 loading control was from Stain-Free^TM gel image. Positive clones were fixed in ice-cold methanol and stained with α-HA antibodies. Representative parasites from each stage of the cell cycle: (C) (G1/S) parasites (D) S phase parasites (E) G2 phase parasites and (F) M phase parasites were selected to show nuclear staining patterns of endogenously-tagged TbPCNA. Arrows point to punctate replication foci in the nucleus. K-kinetoplast, eK-elongated kinetoplast, N-nucleus, α-HA-anti-hemagglutinin antibodies, DAPI- 4',6-diamidino-2-phenylindole, dihydrochloride. Scale bars represent 5 μm.

T. brucei cell cycle stages can be estimated by the ratio of nuclei to kinetoplast, which are the DNA containing organelles in the parasite. G0/G1 parasites typically contain 1 nucleus and 1 kinetoplast (1N1K). Parasites in late S phase typically have an elongated kinteoplast and 1 nucleus (1N1eK).^26,27 G2 parasites typically are 1N2K, and M phase parasites are 2N2K.²⁸ Indirect immunofluorescence analysis demonstrated that HA-tagged PCNA localized to distinct spots in 1N1K parasites (Fig. 1C). Distinct punctate nuclear spots were also detected in the nuclei of 1N1eK parasites (Fig. 1D). Punctate spots in 1N2K parasites were detected, but to a lesser extent (Fig. 1E). In 2N2K parasites, the anti-HA antibodies co-localized to the cytoplasm (Fig. 1F). These immunofluorescent studies indicated that TbPCNA was primarily expressed in the nucleus during G1, S, and G2 phases, but entered the cytoplasm during M phase, which was the only observation that differed from the previous study.

PCNA is expressed throughout the cell cycle but is elevated during S phase in mammalian cells.^29,30 Similar expression patterns have been reported for PCNA homologs in yeast.³¹ and in the related kinetoplastid parasite, Leishmania donovani.³² TbPCNA was reported as being upregulated during S phase in procyclic stage T. brucei but not detectible in the G2/M phases.²³ We synchronized the cultures with hydroxyurea as described by Forsythe et al.³³ Nuclear and cytoplasmic fractions from T. brucei synchronized in S, G2 and M phases were prepared and examined by immunoblot analysis. TbPCNA expression was detected in the cytoplasmic and nuclear fractions of each synchronized population of T. brucei. This indicated that TbPCNA like other PCNA homologs was expressed throughout the cell cycle (S1).+

Depleting TbPCNA in T. brucei reduced proliferation and DNA replication

A 710-bp fragment from the TbPCNA cDNA was subcloned into pZJM.³⁴ to examine the effects that depleting its mRNA by RNA interference (RNAi) would have on T. brucei. The pTbPCNA-RNAi plasmid was linearized with restriction endonuclease NotI and transfected into T. brucei.³⁵ Stable transfectants were selected with phleomycin and used to express double stranded RNA under control of the pZJM tetracycline-inducible promoters.³⁴

To examine the effects that RNAi had on the steady-state levels of TbPCNA mRNA in T. brucei, stable transfectants were diluted to 10⁵/ml and cultured in media without or with tetracycline for 48 h. Reverse transcriptase PCR (RT-PCR) analysis was done to examine the steady state levels in these stable transfectants. Semi-quantitative analysis by RT-PCR showed that depletion of TbPCNA mRNA occurred after induction with tetracycline for 48 h (Fig. 2A and S2). Because TbPCNA mRNA was depleted in these stable transfectants upon tetracycline induction, we designated them as TbPCNA_D clones. These clones were counted every 24 h and their cumulative numbers were graphed to analyze their growth patterns for 6 d Continuously culturing TbPCNA_D clones in media containing tetracycline clearly reduced their growth rate in comparison to culturing them in tetracycline-free media (Fig. 2B). This indicated that depleting TbPCNA mRNA had a negative effect on T. brucei proliferation.

Figure 2.

TbPCNA mRNA depletion diminished proliferation and DNA replication. TbPCNA_D clones were diluted to 1 × 10⁵ per/ml and propagated in growth media without (Tet (−)) or with 1.0 μg/ml tetracycline (Tet (+)) for 48 h. The amount of 1.5 μg of total RNA was extracted from the parasites and used for RT-PCR amplification using primers to TbPCNA or α tubulin (α-Tub). (A) 5 μl of RT-PCR reaction (after 23 of 35 cycles) was resolved on an agarose gel and stained with SYBR® Safe. Fluorescence intensity of the bands was quantified using a ChemiDoc^TM MP. (i) RT-PCR product of TbPCNA after 23 cycles. (ii) RT-PCR product of α-Tub after 23 cycles used as a loading control for total RNA. (B) Growth curves representing the mean cumulative count over 6 d for TbPCNA_D clones grown in media without (solid lines) or with (dashed lines) tetracycline. Graphs represent the mean counts with standard error from 3 independent clones repeated in triplicate. (C) Example of non-induced or induced TbPCNA_D clones after labeling with EdU for 2 h. Arrows point to EdU-labeled nuclei and DAPI-stained nuclei in the same clones. (D) Bar graph representing the mean percent with standard error for EdU-labeled nuclei in control (Tet (-)) or induced (Tet (+)) clones. Results were obtained from experiments done in triplicate counting a minimum of 150 parasites per slide per condition. *** The difference between labeled and unlabeled nuclei had a p-value <0 .05. Scale bars in graphs represent 5 μm.

Non-induced TbPCNA_D clones were cultured in tetracycline-free medium that contained the nucleotide analog EdU at a final concentration of 100 μM for 2 h to assess the baseline levels of DNA replication in T. brucei. EdU labeled parasites were fixed with paraformaldehyde and double-stained with Alexa Fluor® 488 azide, which specifically attaches to EdU by click chemistry,³⁶ and with DAPI (Fig. 2C). The percentages of EdU-positive nuclei in non-induced TbPCNA_D clones were calculated by dividing the number of EdU-labeled nuclei by the number of DAPI-stained nuclei. Using this method, we detected labeling in 17% of control TbPCNA_D clones after 24 h when the combined frequencies of S and G2/M of parasites were at their peak.

EdU labeling was performed again in TbPCNA_D clones cultured in media containing tetracycline. After 24 h of TbPCNA mRNA depletion, we calculated that 0.7% of the TbPCNA_D clones had EdU-positive nuclei after being labeled for 2 h (Fig. 2D). Parasite viability was examined by trypan blue assay, which demonstrated that more than 98% of the parasites were still viable after TbPCNA had been depleted (data not shown). These results strongly suggested that depleting TbPCNA mRNA arrested DNA replication in T. brucei without initially killing the parasite.

Depleting TbPCNA mRNA prevented normal cell cycle progression in T. brucei

Cytometric analysis demonstrated that TbPCNA_D clones cultured in tetracycline-free HMI-9 media underwent normal cell cycle progression. Tetracycline induction, which led to the depletion of TbPCNA mRNA in these clones, prevented T. brucei from undergoing normal cell cycle progression. The mean frequency of G0/G1 parasites dropped to 40% at T₄₈ post-induction (Fig. 3A), but S phase (Fig. 3B) and G2/M phase (Fig. 3C) parasites each accumulated to a frequency of 30% at this time point. These data indicate that depleting TbPCNA mRNA delayed or inhibited S and G2/M phase progression in these clones.

Figure 3.

Cell cycle progression ceased in TbPCNA_D clones upon tetracycline induction. Asynchronous TbPCNA_D clones were diluted to 1×10⁵/ml to stimulate proliferation. Clones were cultured in media without (solid lines) or with (dashed lines) tetracycline at T₀ and allowed to grow for 48 h. Aliquots of the induced TbPCNA_D clones were fixed and prepared for analysis by flow cytometry every 8 h, from T₀ to T₄₈ and line graphs representing the mean frequencies of the cell cycle phases: (A) G0/G1, (B) S phase, or (C) G2/M obtained from cytometric histograms were plotted. Line graphs represent the mean frequencies with standard error obtained from histograms of 4 independent experiments. (D) Nuclei and kinetoplasts were counted in the parasites 48 h post induction to estimate the frequency of 1N1K, 1N2K, and 2N2K in TbPCNA_D population. The bar graph represents the mean frequencies with standard error calculated from experiments done in triplicates counting a minimum of 150 parasites from each condition.

We stained the nuclei and kinetoplasts in non-induced or induced TbPCNA_D clones with DAPI at T₄₈ and counted them by fluorescent microscopy to visually estimate their cell cycle distribution. The frequency of 1N1K parasites decreased in tetracycline-induced clones at T₄₈, which correlated with the G0/G1 frequency drop observed by cytometric analysis. At this time point, the frequency of 1N2K and 2N2K parasites increased also (Fig. 3D). These analyses together indicated that depletion of TbPCNA mRNA caused T. brucei to accumulate in G2 and M phases.

Overexpressing TbPCNA in T. brucei arrested proliferation

The TbPCNA-pLew111 expression vector was transfected into T. brucei after being linearized with NotI, which allowed for tetracycline inducible overexpression of TbPCNA in the parasite.^24,25 Eight phleomycin-resistant transfectants were screened by immunoblot analysis to identify ones that expressed TbPCNA_HA upon tetracycline induction. We designated tetracycline-inducible transfectants able to overexpress TbPCNA_HA as TbPCNA_OE clones. A representative immunoblot showed that basal levels of TbPCNA_HA were detected in non-induced TbPCNA_OE clones (Fig. 4A). Such a result indicated that the promoter in this plasmid was not tightly regulated. Detection of high basal levels of TbPCNA_HA in non-induced TbPCNA_OE clones was consistent with the 20% read-through originally reported using the parental pLew82 plasmid.,²⁴ which is the backbone of pLew111. Several independent TbPCNA_OE clones were diluted to 10⁵/ml and cultured in HMI-9 media containing tetracycline at T₀ to test the effects that overproducing TbPCNA had on proliferation in T. brucei. Mean levels of proliferation were arrested in the TbPCNA_OE clones as early as 24 h and lasted through day 6 post induction (Fig. 4B). The luciferase assay provides a linear correlation between intra-parasite ATP concentrations and the amount of proliferation.³⁷ This assay was used to measure proliferation in TbPCNA_OE clones cultured for 48 h in 96-well plates. Read-outs from the luciferase assay showed that proliferation in non-induced TbPCNA_OE clones was more than 5-fold higher than it was in tetracycline-induced clones (Fig. 4C). Parasite viability was examined by trypan blue assay, which demonstrated that, similarly to TbPCNA_D clones, greater than 98% of the parasite population remained viable after 48 h overexpression of TbPCNA (data not shown). These collective results clearly indicated that overexpressing TbPCNA arrested proliferation in T. brucei.

Figure 4.

Overexpression of TbPCNA arrests proliferation in bloodstream form T. brucei. A representative TbPCNA_OE clone was grown in media without (Tet (−)) or tetracycline (Tet (+)). (A) Extracts prepared from about 2 × 10⁶ parasites were resolved on a 10% SDS-PAGE gel and transferred to PVDF membrane for examination by immunoblot analysis using α-HA antibodies. Note basal levels of TbPCNA_HA in control parasites (Tet (−) lane). VSG loading control represents Stain-Free^TM image obtained from gel prior to transfer. (B) Three independent TbPCNA_OE clones were cultured in growth media without (solid lines) or with (dashed lines) tetracycline and the mean count of these parasites was plotted as a line graph. Mean and standard deviation were calculated from these clones tested in 3 independent experiments. (C) TbPCNA_HA transfectants were diluted to 10⁵/ml and plated in 96-well plates. After 48 h of incubation, proliferation of control (Tet (−)) or induced (Tet (+)) clones was checked by luciferase assay, which returned values in relative light units (RLU). Bar graph represents the mean log₁₀ RLUs with standard error for 3 independent clones replicated in triplicate experiments. ** Indicates that the difference in RLU values observed under each condition was significant with a p-value <0 .05.

Overexpressing TbPCNA arrested T. brucei in G2/M

Cell cycle examination showed that the frequency of G0/G1 non-induced TbPCNA_OE clones was about 40% at T₀ but increased to above 50% by T₄₈. Such a low frequency of G0/G1 parasites was consistent with the reduced growth rate observed in non-induced TbPCNA_OE clones. Overexpression of TbPCNA in T. brucei for 48 h reduced the frequency of G0/G1 parasites in the population to about 19% (Fig. 5A). It had little effect on the frequency of S phase T. brucei, which remained about 20% throughout the 48 h period of induction (Fig. 5B). G2/M frequencies in non-induced clones remained about 40% until T₃₂ and then declined, however the population of induced clones began shifting toward a G2/M majority at T₈. By T₁₆ post-induction, the frequency of G2/M parasites was 60% and remained above that level until T₄₈ (Fig. 5C). These results indicate that overexpressing TbPCNA specifically arrested the parasites at G2/M. The representative histograms in S3 demonstrated the rapid onset of G2/M arrest that occurred upon induction of TbPCNA_OE clones.

Figure 5.

Overexpressing TbPCNA arrested T. brucei in G2/M phase. Line graphs representing the cell cycle frequencies: (A) G0/G1; (B) S and (C) G2/M were plotted using data obtained from histograms generated from non-induced (Tet (−) solid lines) or induced (Tet (+) dashed lines) clones in 8 h intervals for 48 h. Each line graph represents the mean with standard error for 4 independent experiments. (D) TbPCNA_OE clones were stained with DAPI to count their nuclei (N) and kinetoplasts (K) by fluorescence microscopy. Graphs were plotted showing changes in the frequencies of 1N1K, 1N2K or 2N2K clones in non-induced (black bars) or induced (gray bars) cultures after 48 h. Bar graphs show the mean frequency with standard error for 1N1K, 1N2K or 2N2K clones. Results were obtained from counting a minimum of 200 parasites per slide per condition tested.

Frequencies for nuclei and kinetoplasts were 84% 1N1K, 9% 1N2K and 7% 2N2K for non-induced TbPCNA_OE clones at T₄₈. These frequencies were typical for T. brucei cultures proliferating as asynchronous populations. Counts for nuclei and kinetoplasts of induced TbPCNA_OE clones at T₄₈ demonstrated that the frequencies of 1N1K: 1N2K: 2N2K parasites were 54%: 26%: 20% respectively (Fig. 5D). These ratios were atypical in comparison to the ratios in control parasites and further demonstrated that overexpressing TbPCNA in T. brucei cultures resulted in the accumulation of G2 and M phase parasites that were nearly 3 times the levels observed in control parasites.

Overexpressing TbPCNA reduced T. brucei DNA replication without damaging DNA

Non-induced or induced TbPCNA_OE clones were diluted at T₀ as described above and briefly labeled with EdU for 2 h during their log phase growth period (Fig. 6A). At T₂₄, about 17% of the non-induced clones had labeled nuclei after short EdU treatments. Inducing these clones for 24 h resulted in only 1% of them having labeled nuclei after the 2 h EdU treatment period (Fig. 6B). We observed that brief EdU incubations labeled less than 8% of nuclei in non-induced clones at T₄₈, as they approached stationary phase. Overproduction of TbPCNA in clones for 48 h led to about 3% of nuclei in the population being EdU-labeled (Fig. 6C). This suggested that overproducing TbPCNA in T. brucei also reduced its ability to synthesize new DNA.

Figure 6.

Overexpressing TbPCNA inhibited DNA replication in T. brucei. TbPCNA_OE clones were cultured in media without or with tetracycline for 24 or 48 h and then labeled for 2 h with EdU. The percent of clones undergoing DNA replication was quantified by fluorescence microscopy. (A) Representative micrograph of control (Tet (−) column) or induced (Tet (+) column) TbPCNA_OE clones labeled with EdU after 24 h. Arrows point to EdU-labeled nuclei and to DAPI-stained nuclei of the same clone. (B) Bar graph representing the mean percent with standard error of EdU-labeled nuclei from control (Tet (−)) or induced (Tet (+)) clones at 24 h time points. (C) Bar graphs showing percent of nuclei labeled with EdU in control (Tet (−)) or induced (Tet (+)) clones at 48 h time points. Results for graphs were obtained from counting a minimum of 250 parasites per slide per condition. Scale bars represent 5 μm.

Phosphorylation of histone γH2A(X) has been used as a reliable marker for detecting DNA damage in mammalian cells.³⁸ Recently, trypanosome histone γH2A(X) was identified and shown to function as a reliable marker for DNA strand breaks in T. brucei.³⁹ We utilized antibodies raised against trypanosome histone γH2A(X) to examine if reducing DNA replication by upregulating TbPCNA caused accumulation of DNA stand breaks in the parasite. Quantitation by indirect immunofluorescence revealed that these antibodies stained about 5% of non-induced parasites, whereas they stained about 10% of TbPCNA_OE clones after 24 h of induction (data not shown). These values were less than or equal to the baseline levels reported in control T. brucei³⁹, suggesting that overexpressing TbPCNA did not cause any appreciable accumulation of DNA strand breaks.

Overexpressing human PCNA arrested proliferation in T. brucei

TbPCNA was 35% identical to human PCNA and 36% identical to PCNAs of yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe based on Clustal W alignments. Alignment of TbPCNA with these homologs demonstrated that TbPCNA contained amino acid insertions. The first insertion was located between residues 186 and 197 and the second was located between residues 206 and 226 (Fig. 7A). Similar amino acid insertions were recognized in PCNA homologs from the related kinetoplastid parasite Leishmania donovani.⁴⁰ As a member of the sliding clamp family, it is predicted that TbPCNA can assume a toroid structure similar to other PCNA homologs. The homology model for TbPCNA, based on monomer structures of yeast and human PCNAs (PDBs 3K4X and 3VKX respectively), predicted that it had 2 domains bridged by an interdomain-connecting loop (Fig. 7B).

Figure 7.

Proliferation in T. brucei was arrested after being induced to overexpress human PCNA. (A) Clustal W alignment of PCNA homologs: Trypanosoma brucei (Tb), Saccharomyces cerevisiae (Sc), Homo sapiens (Hs), and Schizosaccharomyces pombe (Sp) were generated using the utility ESPript 3.0 to show their similarities. Red blocks indicate identical amino acids and white blocks indicate strongly similar amino acids. Black triangles represent sites where amino acid insertions occur in TbPCNA and other kinetoplastid parasites. (B) Homology model of T. brucei, yeast and human PCNA monomers with arrow pointing to unstructured loop in TbPCNA formed by the amino acid inserts. (C) Representative HsPCNA_OE clone cultured in media without or with tetracycline were examined by immunoblot analysis to detect HsPCNA_HA by α-HA antibodies. VSG loading control represents Stain-Free^TM image obtained from gel prior to transfer. (D) Line graph comparing growth rates for 90–13 (dashed line), TbPCNA_OE (black line) and HsPCNA_OE (gray line) cultured in tetracycline-free media. (E) Graph comparing growth rates of HsPCNA_OE clones cultured in non-induced (black line) or induced conditions (dashed line). Growth curves represent the mean value with standard deviations for experiments repeated 3 times using 3 independent clones.

We subcloned the human HsPCNA cDNA into pLew111 for overexpression in T. brucei to test the notion that it would arrest growth in the parasite just as overproducing TbPCNA did. Phleomycin selection and immunoblot analysis were done as mentioned above to screen for stable transfectants that expressed HsPCNA upon tetracycline-induction. Based on the selection and screen, we identified 6 stable transfectants that expressed HsPCNA upon tetracycline induction that we designated as HsPCNA_OE clones. Anti-HA antibodies detected a faint 32 kDa band in extracts from a non-induced representative HsPCNA_OE clone, but a very prominent band was detected from the induced clone (Fig. 7C). These observations were very consistent with those from TbPCNA_OE clones.

The mean growth rates from several independent HsPCNA_OE and TbPCNA_OE clones cultured in tetracycline-free media were compared to that of parental 90-13 clones to determine the consequences of basal levels of expression from either PCNA homolog in the parasite. We observed that the doubling time for either HsPCNA_OE or TbPCNA_OE clones cultured in tetracycline-free media was about 24 h, whereas for 90-13 parasites it was about 7-9 hours (Fig. 7D). The effects of overproducing HsPCNA in T. brucei were examined by comparing the mean growth rates of several independent clones cultured in media without or with tetracycline. This experiment clearly demonstrated that overexpressing HsPCNA in T. brucei arrested its proliferation (Fig. 7E).

Downregulating TbPCNA chemosensitized T. brucei

Luciferase assays were used to determine the IC₅₀ of tetracycline for TbPCNA_D or TbPCNA_OE clones. The tetracycline IC₅₀ was determined to be about 100 ng/ml for TbPCNA_D clones and about 10 ng/ml for TbPCNA_OE clones (data not shown). Inducing parasites with these concentrations of tetracycline was done to attenuate them without abrogating proliferation. Non-attenuated or attenuated clones were treated with serial dilutions of the DNA damaging agent H₂O₂ for 48 h to examine if they developed hypersensitivity. We expected hypersensitivity in attenuated clones to correlate with a decrease in their IC₅₀ for H₂O₂ in comparison to the IC₅₀ observed in non-attenuated clones. The IC₅₀ of H₂O₂ remained constant in TbPCNA_D or TbPCNA_OE clones whether they were attenuated or not, which indicated that attenuating these clones did not make them hypersensitive to H₂O₂ (data not shown).

Treating non-attenuated TbPCNA_OE or TbPCNA_D clones with serial dilutions of the replication inhibitor hydroxyurea (HU) demonstrated that they had a similar IC₅₀ for this agent, which was between 100 and 50 μM (Fig. 8A). The IC₅₀ of HU for attenuated TbPCNA_OE clones also fell between 100 and 50 μM, which indicated that they were not hypersensitive to HU treatment. However in attenuated TbPCNA_D, the IC₅₀ of HU was between 50 and 25 μM (Fig. 8B). Such an IC₅₀ reduction suggested that down-regulating TbPCNA and treating with HU had an adjuvant effect on proliferation in these clones, and indicated that attenuating TbPCNA_D clones made them hypersensitive to HU treatment.

Figure 8.

Down-regulating TbPCNA hypersensitized T. brucei to hydroxyurea. Luciferase assays were used to measure proliferation in TbPCNA_D or TbPCNA_OE clones treated with hydroxyurea (HU) for 48 h. 10⁴ parasites were grown in 96-well plates and treated with serial dilutions of HU followed by incubation at 37°C and 5% CO₂ for 48 h. The luciferase reagent (CellTiter-Glo®) was added to each plate to obtain relative light unit (RLU) readouts by luminometry. (A) Graph showing mean RLU values for non-attenuated clones of TbPCNA_D (open bars) and TbPCNA_OE (gray bars) treated with HU. Arrow is situated between HU concentrations in which the mean RLU values of non-attenuated clones were estimated to be 50% of the values for non-HU treated non-attenuated controls at 0 μM. (B) Graph that shows the mean RLU values for attenuated clones treated with HU. Arrow is situated between concentrations in which the mean RLU values for HU-treated attenuated TbPCNA_D clones (black bars) were 50% of the values for non-HU treated attenuated controls at 0 μM concentration. RLU values represent the mean with standard deviations for 3 independent experiments.

Discussion

Several components of the DNA replication machinery are highly regulated in the cell. The consequences of deleting or deregulating PCNA homologs in other model eukaryotic organism: yeast, mammals, and Drosophila have been very well characterized demonstrating that the PCNA gene and its functional product are essential for cell survival. Gene products such as ERK8 and p21 can arrest cell cycle progression in human cells by reducing intracellular levels of PCNA.^44,45 The p21^Cip1(Waf1 protein arrests DNA replication in human cells by sequestering PCNA, which prevents progression from G1 and G2 phases. Furthermore, several mechanisms of posttranslational modification: phosphorylation, ubiquitylation/sumoylation, and ADP ribosylation are responsible for regulating the function and stability of PCNA in mammalian cells and yeasts.

Much less is known about how components of the DNA replication machinery are regulated or the consequences of deregulating their levels in T. brucei. We demonstrated that either upregulating or downregulating TbPCNA levels in T. brucei severely reduced proliferation by different mechanisms that remain uncharacterized. Abnormal growth phenotypes associated with depleting TbPCNA in this study were consistent with those predicted from the whole genome RNAi targeted screen of T. brucei.⁵⁰ This study demonstrates that depleting TbPCNA by RNAi prevented G2 cell cycle progression in T. brucei. Such a phenotype is consistent with the S and G2/M phase arrest observed after PCNA was depleted in human cell by siRNA ^51,52 and demonstrates an evolutionarily conserved function of PCNA.

Integration into the proper locus was essential for T. brucei clones to express endogenously tagged TbPCNA_HA. Stable clones that expressed endogenously tagged TbPCNA_HA grew at similar rates as wildtype parasites (data not shown). Therefore, we hypothesize that normal levels of TbPCNA were not perturbed in the endogenously tagged TbPCNA_HA clones. It is possible that the increased cytoplasmic staining of TbPCNA observed during M phase with these clones was cause by misdirection from the HA-tag. Immunoblots of synchronized parasites demonstrated that TbPCNA expression was still present during both G2 and M stages of the parasites. This indicates that TbPCNA expression is present during every stage of the cell cycle, which is the case in most other eukaryotic model organisms. The previous study followed TbPCNA expression in a single parasite and reported irregular patterns of TbPCNA expression in M phase parasites.²³ Such irregular expression patterns in M phase parasites might explain why TbPCNA was not detected in the nucleus of a single parasite. We examined multiple parasites per field of view and detected TbPCNA expression in the cytoplasm and nucleus of M phase parasites from endogenous and inducible TbPCNA clones. Further studies are needed to determine if cytoplasmic expression of TbPCNA plays a role in regulating the T. brucei cell cycle.

Upregulation of PCNA in mammalian cells is a hallmark of increased proliferation that serves as a predictable diagnostic marker for a broad range of cancers.^19-21,53,54 In the related species L. donovani, upregulating LdPCNA was associated with clinical isolates of this parasite being resistant to antimonial compounds. Overexpressing LdPCNA laboratory strains protected it from such drugs.⁴⁰ The phenotype observed after upregulating TbPCNA in T. brucei was in stark contrast to that reported for upregulating it in human cells or in L. donovani. Reduced proliferation in non-induced TbPCNA_OE clones was correlated with marginal increases of intra-parasite TbPCNA levels caused by the lax plasmid promoter.

Observations from this study led to the hypothesis that bloodstream form T. brucei has thresholds for its TbPCNA levels. This hypothesis was supported by the arrested proliferation in T. brucei observed after overexpressing either TbPCNA or HsPCNA. A similar outcome was reported after overexpressing pcn1, the PCNA homolog in the fission yeast S. pombe also supports this hypothesis. Upregulating pcn1 triggered a checkpoint that delayed the yeast from exiting G2/M, even when it was upregulated in checkpoint mutants of S. pombe.⁵⁵ Overexpressing TbPCNA in bloodstream form T. brucei allowed the parasite to enter S phase but severely delayed progress beyond the G2/M phase. This is consistent with the notion that S phase in the kinetoplast was unaffected by overexpressing TbPCNA, whereas the nuclear S phase was inhibited or delayed. This represents a plausible explanation because in T. brucei, S phase in the kinetoplast precedes nuclear S phase.²⁸ This would indicate that TbPCNA in the nucleus is regulated differently than it is in the kinetoplast, a matter that needs to be investigated further. We concluded that the phenotype associated with overexpressing TbPCNA resulted from triggering an uncharacterized G2/M checkpoint in the parasite because both G2 and M phase parasites accumulated after overexpressing TbPCNA. Similar abnormal proliferation and cell cycle defects occurred when either human PCNA or TbPCNA were overexpressed in T. brucei (S4). Therefore, we conclude that overproducing either of these homologs may trigger the same mechanism that leads to G2/M arrest in the parasite. Two likely mechanisms for how overexpressing TbPCNA arrested proliferation in bloodstream form T. brucei were that excess TbPCNA triggered a checkpoint kinase or that it sequestered interacting proteins that regulate the cell cycle. The latter mechanism seems more plausible, because the function of PCNA is directed by its ability to form stable interactions with many proteins. Future studies to identify TbPCNA interacting proteins may help determine if upregulating TbPCNA sequesters critical cell cycle regulators in the parasite.

HU acts as a replication inhibitor in eukaryotic cells and depleting TbPCNA associated with arrested DNA replication in T. brucei. We conclude that attenuating TbPCNA_D clones and treating them with HU led to an additive or synergistic effect that made these parasites hypersensitive to HU. Therefore, we interpret these observations to indicate that overexpressing or depleting TbPCNA mRNA in the parasite inhibited proliferation by separate mechanisms.

Currently, mammalian cells and yeast are the predominant model system for characterizing PCNA. This has fundamentally restricted the types of studies for chemotherapeutic interventions of PCNA to human proliferative diseases such as cancer, arthritis and nephritis. ^52,56-59 The results of this study validate TbPCNA as a viable target for therapeutic intervention against African trypanosomiasis and broaden the categories of where PCNA therapeutics may be beneficial to infectious disease research. Future studies will elucidate mechanisms of TbPCNA regulation in T. brucei. Information gained from such studies can also reveal how deregulating TbPCNA triggers G2/M arrest. This information in combination with that obtained from several novel classes of PCNA inhibitors^52,59 may lead to the development of innovative therapies that target components of the T. brucei DNA replication machinery.

Material and Methods

Cell culture

Bloodstream form T. brucei were incubated in 5% CO₂ at 37°C in HMI-9 medium modified to contain 20% fetal bovine serum.⁶⁰ Parasites were cultured in media containing 100 U/ml penicillin and 100 μg/ml streptomycin. Selection medium contained 5 μg/ml hygromycin B, 2.5 μg/ml G418, and 2.5 μg/ml phleomycin. Parasites were induced by adding tetracycline to the medium at a final concentration of 1.0 μg/ml, unless further specified. The TbPCNA coding region was subcloned at the HindIII (http://www.thermoscientificbio.com/restriction-enzymes/hindiii/) and AflII (http://www.thermoscientificbio.com/restriction-enzymes/bspti-aflii/) restriction sites of the C-terminal HA-tagged modified version of pLEW111.²⁵ The restriction endonucleases NruI (https://www.neb.com/products/r0192-nrui) and NotI (https://www.neb.com/products/r0189-noti) were purchased from New England Biolabs.

Introduction of RNA interference (RNAi) transgenes into T. brucei

For electroporation, 10⁷ parasites were pelleted by centrifugation and washed once with 10 ml of phosphate buffered saline (PBS) (4.3 mM Na₂HPO₄ 137 mM NaCl, 2.7 mM KCl, 1.4 mM KH₂PO₄) pH 7.4. 1-to-10 μg of NotI (http://www.thermoscientificbio.com/restriction-enzymes/noti/) linearized plasmid was nucleofected into T. brucei using a Nucleofector™ Kit for Human T Cells (http://www.lonza.com/products-services/bio-research/transfection/nucleofector-kits-for-primary-cells/nucleofector-kits-for-primary-blood-cells/nucleofector-kits-for-human-t-cells.aspx). The parasites were pulsed using program X-001, transferred to 10 ml of modified HMI-9 media, and incubated overnight at 37°C with 5% CO₂. The next day, stable clones were selected and grown in media containing 5.0 μg/ml hygromycin B, 2.5 μg/ml G418, and 2.5 μg/ml of phleomycin.

RNA interference (RNAi) and quantification

RNAi was induced by supplementing selection media with tetracycline to a final concentration of 1 μg/ml. Parasites were counted using a Multisizer™ 3 Coulter Counter® Tetracycline induced or non-induced bloodstream form parasites were pelleted by centrifugation and resuspended in 1 ml of TRIzol® reagent (https://www.lifetechnologies.com/order/catalog/product/15596026). For all samples, 1.5 μg of total RNA was used as a template for amplifying coding regions with the SuperScript® III One-Step RT-PCR kit (http://www.lifetechnologies.com/order/catalog/product/12574018) and gene-specific primers: (TbPCNA forward 5′-AAGCTTATGCTTGAGGCTCAGGTTCT-3′ and reverse 5′-CTTAAGCTCGGCGTCGTCACCTTTG-3′). Primers for Tb-α tubulin (α-Tub forward 5′-ATGCGTGAGGCTATCTGCATCCACA-3′ and α-Tub reverse 5′- AGGTTGCGGCGAGTCAAATCATAAAT-3′) were used as internal controls to check for equal loading and integrity of T. brucei mRNA. Five microliters from cDNA samples were collected at 20, 23, and 25 PCR cycles to ensure that products were still in the linear range of amplification. PCR amplified cDNA products were resolved by agarose gel electrophoresis, stained with SYBR® Safe (https://www.lifetechnologies.com/us/en/home/life-science/dna-rna-purification-analysis/nucleic-acid-gel-electrophoresis/dna-stains/sybr-safe.html), and quantified by ChemiDoc^TM MP.

Immunoblots

Expression of hemagglutinin-tagged TbPCNA expressed in T. brucei was examined by immunoblot as previously described.⁶¹ Stain-Free® gels used to resolve polypeptides for immunoblots were obtained from BioRad (http://www.bio-rad.com/en-es/product/mini-protean-precast-gels/mini-protean-tgx-stain-free-precast-gels). Histone γH2A(X) antibodies used in this study were a generous gift from David Horn.³⁹

Luciferase assay

Parasites were diluted to 10⁵/ml and a total of 10⁴ were incubated in 96-well white opaque tissue culture plates in media with or without tetracycline at 1 μg/ml in biosafety cabinets for 48 h at 37°C with 5% CO₂. After the 48 h incubation period, 50 μl of CellTiter-Glo® reagent (https://www.promega.com/products/cell-health-and-metabolism/cell-viability-assays/celltiter_glo-luminescent-cell-viability-assay/) was added to each well and plates were placed on an orbital shaker at room temperature for 2 min to induce lysis. After a 10 minute incubation to stabilize the signal, the ATP bioluminescence of each well was determined using a SpectraMax® L plate reader. Raw values measured in relative light units (RLU) were converted to log₁₀ and percentage inhibition was calculated relative to the controls. Curve fitting was performed with Prism 4 software or Excel.

5-Ethynyl-2′deoxyuridine (EdU) labeling

At T₀ bloodstream form TbPCNA RNAi clones were subcultured at 1×10⁵/ml in HMI-9 media without or with tetracycline to a final concentration of 1.0 μg/ml. After 48 h, EdU (https://www.lifetechnologies.com/order/catalog/product/A10044?ICID=search-product) was added in the culture media to final a concentration of 100 μM and labeling proceeded for 2 h. The parasites were pelleted by centrifugation, washed with 10 ml phosphate buffered saline (PBS) and fixed with 0.5% paraformaldehyde (https://www.emsdiasum.com/microscopy/products/chemicals/paraformaldehyde.aspx) for 30 min. The fixed parasites were resuspended in 100 μl of PBS, attached to poly-L-lysine coated slides and stained with Click-It® (https://www.lifetechnologies.com/order/catalog/product/C10337) and 4′,6-diamidino-2-phenylindole (DAPI) (https://www.lifetechnologies.com/order/catalog/product/P36931). Visualization and quantification of labeled nuclei was done using indirect fluorescence microscopy and CellProfiler 2.0 (Massachusetts Institute of Technology).

Quantification of nuclei and kinetoplasts

Parasites were visualized using an Axiovert A.1 microscope from Carl Zeiss and photographed using the Axiocam MRc camera system. The DAPI-stained nuclei and kinetoplasts in T. brucei were photographed using a 63× or 100× objective. CellProfiler 2.0 and Zen software were used to count the nuclei and kinetoplasts from T. brucei images obtained by microscopy.

Flow cytometry analysis

T₀ parasites were subcultured to 1 × 10⁵/ml and grown for 48 h in HMI-9 medium with or without tetracycline (https://www.goldbio.com/product/1896/tetracycline-hydrochloride). After 48 h (T₄₈), 1×10⁶ cells were pelleted and washed 3 × 5 ml of PBS and fixed with 0.5 % paraformaldehyde (https://www.emsdiasum.com/microscopy/products/chemicals/paraformaldehyde.aspx) for 30 min, which was then removed by washing with 5 ml of PBS. Washed parasites were RNAse A (http://www.amresco-inc.com/RNASE-A-SOLUTION-10MGML-E866.cmsx) treated and stained with propidium iodide (10μg/ml) (http://www.sigmaaldrich.com/catalog/product/fluka/81845?langDenandregionDUS). Cytometric data was collected with an Accuri™ Flow Cytometer and analyzed with FlowJo.

Indirect immunofluorescence microscopy

Parasites were washed once with PBS before attaching them to poly-L-lysine coated microscope slides. Attached parasites were then fixed with cold methanol for 5 minutes and allowed to air dry. The attached parasites were made permeable with 0.5% Triton X-100 for 5 minutes at room temperature and washed with PBS containing 0.5% BSA. Primary antibodies were diluted 1:200 in PBS containing 0.5% BSA and incubated on the prepared slides for 1 h in a humidified chamber at 37°C. Slides were washed 3 times with 0.5% BSA in PBS. AlexaFluor® 488 or 594-conjugated secondary antibodies (488: http://www.lifetechnologies.com/order/catalog/product/A21206?CID=search-a21206 or 594: https://www.lifetechnologies.com/order/catalog/product/A21207?ICID=search-a21207) were diluted 1:1000 with PBS and incubated on the slides for 1 h at room temperature in a humidified chamber. The slides were washed 3 times with PBS and mounted with ProLong® Gold with DAPI (https://www.lifetechnologies.com/order/catalog/product/P10144?ICID=search-product). Micrographs were generated using an Axiovert A.1 and AxioCam MRc camera.

Genotoxic agents

Hydrogen peroxide (H₂O₂) (http://www.fishersci.com/ecomm/servlet/itemdetail?LBCID=81833512&storeId=10652&langId=-1&LBCID=81833512&catalogId=29104&productId=3017791&distype=0&fromSearch=0&hasPromo=1) was purchased from ThermoFisher. Hydroxyurea crystalline form (http://www.fishersci.com/ecomm/servlet/itemdetail?itemdetail=%27item%27&storeId=10652&productId=16609104&catalogId=29104&matchedCatNo=50525180&fromSearch=1&searchKey=50-525-180&highlightProductsItemsFlag=Y&endecaSearchQuery=%23store%3DRE_SC%23nav%3D0%23rpp%3D25%23offSet%3D0%23keyWord%3D50-525-180%23searchType%3DPROD%23SWKeyList%3D%5B%5D&xrefPartType=From&savings=0.0&xrefEvent=1405969990000_0&searchType=PROD&hasPromo=0) was purchased from ThermoFisher.

Statistical analysis

The Student's t-test was used to compare numerical means obtained from non-induced or induced samples.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Michael Klemba, Janet Webster and Ling Chen at Virginia Tech and Kenneth Kreuzer at Duke for reading and critical comments.

Funding

This work was supported in part by Start-up fund NIFA139696, VT Drug Discovery Center VTCDD #119286. ACR receives support from NSF S-STEM grant: DUE-0850198. ALV receives support from the Costa Rican Ministry of Science, Technology and Telecommunications (MICITT).