Toxoplasma gondii harbors a hypoxia-responsive coproporphyrinogen dehydrogenase-like protein

ABSTRACT

Toxoplasma gondii is an apicomplexan parasite that is the cause of toxoplasmosis, a potentially lethal disease for immunocompromised individuals. During in vivo infection, the parasites encounter various growth environments, such as hypoxia. Therefore, the metabolic enzymes in the parasites must adapt to such changes to fulfill their nutritional requirements. Toxoplasma can de novo biosynthesize some nutrients, such as heme. The parasites heavily rely on their own heme production for intracellular survival. Notably, the antepenultimate step within this pathway is facilitated by coproporphyrinogen III oxidase (CPOX), which employs oxygen to convert coproporphyrinogen III to protoporphyrinogen IX through oxidative decarboxylation. Conversely, some bacteria can accomplish this conversion independently of oxygen through coproporphyrinogen dehydrogenase (CPDH). Genome analysis found a CPDH ortholog in Toxoplasma. The mutant Toxoplasma lacking CPOX displays significantly reduced growth, implying that T. gondii CPDH (TgCPDH) potentially functions as an alternative enzyme to perform the same reaction as CPOX under low-oxygen conditions. In this study, we demonstrated that TgCPDH exhibits CPDH activity by complementing it in a heme synthesis-deficient Salmonella mutant. Additionally, we observed an increase in TgCPDH expression in Toxoplasma when it grew under hypoxic conditions. However, deleting TgCPDH in both wild-type and heme-deficient parasites did not alter their intracellular growth under both ambient and low-oxygen conditions. This research marks the first report of a CPDH-like protein in eukaryotic cells. Although TgCPDH responds to hypoxic conditions and possesses enzymatic activity, our findings revealed that it does not directly affect acute Toxoplasma infections in vitro and in vivo.

IMPORTANCE

Toxoplasma gondii is a ubiquitous parasite capable of infecting a wide range of warm-blooded hosts, including humans. During its life cycle, these parasites must adapt to varying environmental conditions, including situations with low-oxygen levels, such as intestine and spleen tissues. Our research, in conjunction with studies conducted by other laboratories, has revealed that Toxoplasma primarily relies on its own heme production during acute infections. Intriguingly, in addition to this classical heme biosynthetic pathway, the parasites encode a putative oxygen-independent coproporphyrinogen dehydrogenase (CPDH), suggesting its potential contribution to heme production under varying oxygen conditions, a feature typically observed in simpler organisms like bacteria. Notably, so far, CPDH has only been identified in some bacteria for heme biosynthesis. Our study discovered that Toxoplasma harbors a functional enzyme displaying CPDH activity, which alters its expression in the parasites when they face fluctuating oxygen levels in their surroundings.

INTRODUCTION

Infectious pathogens encounter diverse oxygen conditions throughout their life cycles. For example, pathogenic bacteria inhabiting the host’s intestine are exposed to a notably reduced oxygen environment, ranging from 5% to 0.5% (1) compared to 21% ambient conditions. This substantial fluctuation in oxygen levels demands an adaptive response from these pathogens to effectively cope with hypoxic conditions. Similarly, Toxoplasma gondii needs to deal with a substantial reduction in oxygen while inhabiting the host’s organs during in vivo infection, for example, intestine and spleen. To adapt to hypoxic conditions, the parasites activate the host’s hypoxia-inducible factor (HIF) to stabilize the HIF1-α subunit to upregulate the expression and activity of a group of host genes, including hexokinase 2 (2, 3), which further enhances glycolytic activity in the host cells. This is achieved by inhibiting the activity of host prolyl hydroxylase 2 (PHD2), a key negative regulator of HIF1-α. PHD2 is an α-ketoglutarate-dependent dioxygenase whose high K_m toward oxygen allows it to respond to changes in oxygen availability, making it and related prolyl hydroxylases key cellular oxygen-sensing enzymes (4). Toxoplasma expresses two oxygen-sensing prolyl hydroxylases named PHYa and PHYb that are important for growth at low and high O₂, respectively, and appear to act in response to changes in oxygen availability by altering the parasite proteome (5–7). In addition to oxygen, intracellular iron availability can also modulate the activity of prolyl hydroxylase, which further affects HIF activation and subsequent gene transcription regulation (8).

During different infection stages, Toxoplasma faces diverse nutrient environments within its host and must adapt its metabolic network in response to nutritional variations. The identification of host hexokinase 2 upregulation as a requirement for parasite growth indicated alterations in host glucose metabolism as one such mechanism (2). Recent dual metabolomic profiling studies further revealed that Toxoplasma reprograms the host’s metabolism, shifting it from mitochondrial oxidative phosphorylation to glycolysis, and increases in host and parasite pentose phosphate pathway (9). Within the parasite, sedoheptulose bisphosphatase, an enzyme that channels sugar molecules from the glycolytic pathway into the pentose phosphate pathway (9), is also active and provides an alternative pathway to generate ribose. Since most of these assays were performed at ambient oxygen, it remains unclear how Toxoplasma adjusts its metabolome under low-oxygen conditions to fulfill its nutritional needs.

Heme is a vital metabolite in nearly all organisms and is involved in many fundamental metabolic processes, such as cytochrome formation, cellular redox defense, and oxygen sensing and transport. In most eukaryotic cells, the heme biosynthetic pathway comprises eight reactions distributed across two subcellular locations, the cytoplasm and mitochondria (10, 11). Our previous work along with other two research articles has revealed that Toxoplasma parasites possess a complete and functional heme biosynthesis pathway (12–14). However, four reactions occur within the apicoplast, a vestigial plastid that is believed to have originated from endosymbiosis (15). Toxoplasma parasites heavily depend on their own heme production for intracellular growth and acute virulence (12–14). Within the heme biosynthesis pathway of Toxoplasma, only the antepenultimate reaction takes place in the cytoplasm, producing protoporphyrinogen IX (PROTOgen IX) from coproporphyrinogen III (COPROgen III) catalyzed by COPROgen III oxidase (CPOX). CPOX utilizes oxygen as an electron acceptor to decarboxylate two propionic acid groups on the A and B pyrrole rings of COPROgen III, ultimately yielding PROTOgen IX (16). In contrast, certain prokaryotic cells like Escherichia coli and Salmonella encode coproporphyrinogen dehydrogenase (CPDH), which utilizes radical S-adenosylmethionine (SAM) as an electron acceptor to convert COPROgen III to PROTOgen IX (17, 18). The possession of dual enzymes in these bacteria enables flexible adaptation to various growth conditions.

By analyzing the genome of Toxoplasma parasite, an ortholog of bacterial CPDH-like protein, a radical SAM enzyme has been identified and named T. gondii CPDH (TgCPDH) (TGGT1_288640). Deletion of TgCPOX leads to a ∆cpox mutant with significant growth defects (12), suggesting that the parasites possess a bypass pathway for heme production, possibly utilizing TgCPDH as an alternative enzyme. A previous study localized TgCPDH to the parasite’s mitochondrion via endogenous gene tagging, and its deletion did not result in the loss of virulence in Type II Toxoplasma parasites (14).

Here, we conduct a transgenera complementation experiment of expressing exogenous TgCPDH in a heme synthesis-deficient Salmonella mutant (18) to prove that TgCPDH is a functional enzyme in heme biosynthesis. Our findings also revealed that TgCPDH expression increased in response to hypoxia, but its deletion or overexpression in heme-deficient ∆cpox parasites did not alter their growth. In addition, the loss of TgCPDH in ∆cpox did not further reduce heme production. These results suggested that TgCPDH is not directly involved in heme production in Toxoplasma. In summary, our research identified a hypoxia-responsive radical SAM enzyme in Toxoplasma with CPDH activity, marking the first observation of such an enzyme in a eukaryotic organism.

RESULTS

Toxoplasma encodes a functional CPDH ortholog, an oxygen-independent heme biosynthesis enzyme

Some prokaryotes employ two distinct strategies to convert COPROgen III into PROTOgen IX during the heme biosynthetic pathway. They can utilize either the oxygen-dependent coproporphyrinogen oxidase (HemF) (19, 20) or the oxygen-independent CPDH (HemN) (18, 21, 22) to complete the antepenultimate step in this process. HemF orthologs are prevalent in both eukaryotic and prokaryotic cells. In mammalian cells, this ortholog is known as CPOX (10). In contrast, HemN orthologs have not been identified in eukaryotic cells. While exploring the genome of Toxoplasma, we discovered a radical SAM enzyme, TGGT1_288640, which exhibited similarity to E. coli HemN (EcHemN) (Fig. 1A). A pairwise alignment revealed 10.4% sequence identity and 25.4% residue similarity using the BLOSUM 45 matrix (Fig. 1A). Moreover, according to the crystal structure of EcHemN, we found that four crucial residues within the catalytic site and a CXXXCXXC motif were conserved in the Toxoplasma ortholog (Fig. 1A). Therefore, we named this enzyme TgCPDH throughout this paper. TgCPDH possesses a long nonhomologous C-terminal tail relative to EcHemN (Fig. 1A). Furthermore, we superimposed the alphafold-predicted structure of TgCPDH (23) with the solved crystal structure of HemN (21) and observed that both proteins exhibited a highly similar core structure, consisting of a group of α-helices and β-sheets (Fig. 1B), indicating that TgCPDH structurally resembles E. coli CPDH.

Fig 1.

Toxoplasma encoded a CPDH-like protein. (A) Comparison of the primary structures of EcHemN and TgCPDH through pairwise alignment using ClustalW. The catalytic residues, as determined from the crystal structure of EcHemN, were highlighted with red boxes, while the iron-sulfur cluster binding motif, CXXXCXXC, was enclosed in a blue box. The UniProt IDs for EcHemN and TgCPDH proteins are P32131 and S7W4N9, respectively. (B) Molecular docking analysis of EcHemN and TgCPDH. The crystal structure of EcHemN was obtained from the RCSB Protein Data Bank database (Protein Data Bank ID: 1OLT), while the predicted structure of TgCPDH was generated using the AlphaFold algorithm. Structural alignment was performed using PyMol, with EcHemN and TgCPDH depicted in green and red cartoon backbones, respectively. Two SAM molecules were shown in yellow. (C) Transgenera complementation of TgCPDH in the heme-auxotrophic Salmonella TE3006 strain restored the growth of transformed bacteria in the medium lacking exogenous heme. The plasmid pRJ8276, encoding B. japonicum HemN2, served as a positive control. Furthermore, EcHemN and TgCPDH were cloned into pUC19 under the lac promoter and then introduced into Salmonella TE3006. An empty pUC19 plasmid was included as a negative control. The first column was spotted with 2 µL of PBS-diluted bacterial cultures with an OD₆₀₀ of 0.8, and the following three columns were inoculated with bacterial cultures by 10-fold serial dilution.

To assess whether TgCPDH functions as a CPDH, we PCR-amplified the region with high similarity to EcHemN based on the alignment of primary sequences. Then, we cloned this region under a lac promoter in the pUC19 vector and introduced it into a heme-auxotrophic Salmonella strain, TE3006, for transgenera complementation. The Salmonella TE3006 strain lacks both hemF and hemN and cannot grow in a medium without heme supplementation (18). We also introduced an empty pUC19, pRJ8276 plasmid-encoding Bradyrhizobium japonicum hemN2 (24), or EchemN-encoding plasmid individually as negative and positive controls. Our findings demonstrated that the addition of heme into the medium partially improved the growth of this heme-auxotrophic Salmonella strain, while TgCPDH successfully restored the growth of the Salmonella TE3006 strain in a heme-deficient medium (Fig. 1C). This result indicated that TgCPDH indeed exhibits CPDH activity.

The expression of TgCPDH responded to hypoxic conditions, but it did not facilitate parasite growth under low-oxygen levels

Given that TgCPDH exhibited structural similarity to HemN and functionally complemented the growth of heme-auxotrophic Salmonella, it is speculated that TgCPDH is responsive to low-oxygen conditions. To investigate this, we cultured the RH∆ku80::TgCPDH^myc strain under both ambient and low-oxygen conditions (21% and 3%, respectively) and assessed its mRNA and protein levels through real-time quantitative PCR (qPCR) and immunoblotting, respectively. Our analysis revealed that the transcript abundance of TgCPDH was increased by ~2.5-fold under 3% O₂ conditions compared to the ambient level (Fig. 2A). Similarly, the translation level of TgCPDH was increased by ~2.5-fold under hypoxia compared to that under ambient oxygen (Fig. 2B). Subsequently, we hypothesized that TgCPDH can support parasite growth under low-oxygen conditions. To test this, we genetically deleted TgCPDH in RH∆ku80::nLuc, resulting in a ∆cpdh::nLuc strain (Fig. S1). Both strains were inoculated into human foreskin fibroblasts (HFFs) for a plaque assay under 21% and 3% O₂ conditions. The plaques were allowed to grow for 7 days before staining with crystal violet and quantification using optical microscopy. The plaques formed by RH∆ku80::nLuc parasites under 3% O₂ were significantly smaller than those formed under 21% O₂ conditions. However, the plaques of ∆cpdh::nLuc grown under 3% O₂ showed comparable sizes to those formed under ambient O₂ level (Fig. 2C), suggesting that TgCPDH is not involved in parasite growth in hypoxic conditions. Additionally, we investigated the role of TgCPDH in the parasite’s acute virulence. During in vivo dissemination, tissue oxygen concentrations are generally lower than those in in vitro tissue culture (25). We subcutaneously inoculated five mice for each parasite strain with 100 RH∆ku80::nLuc parasites and either 100 or 1,000 ∆cpdh::nLuc parasites individually and monitored mouse mortality daily. In comparison to the parental strain, ∆cpdh::nLuc parasites did not exhibit attenuated virulence. Mice infected with both low and high inoculum doses of ∆cpdh::nLuc parasites succumbed to infection within 12–13 days post-infection, a timeframe similar to that of RH∆ku80::nLuc-infected mice (Fig. 2D). Overall, these results indicated that the expression of TgCPDH was altered in the parasites under different oxygen conditions, but TgCPDH did not play a role in parasite growth under hypoxic conditions, as observed in both in vitro and in vivo infections.

Fig 2.

Toxoplasma increased TgCPDH expression in response to hypoxia but did not require it for parasite growth under low-oxygen conditions. (A) Real-time quantitative PCR analysis showed that the mRNA abundance of TgCPDH was increased by ~2.5-fold when the parasites were grown under 3% O₂ relative to ambient O₂ conditions. Toxoplasma RNA polymerase II served as a housekeeping gene for normalization. Five biological replicates were included in the analysis. (B) Immunoblotting analysis revealed an elevation in TgCPDH expression in Toxoplasma parasites when cultured under 3% O₂ conditions compared to those grown under ambient O₂ levels. Lysates were probed with anti-myc antibody for TgCPDH detection. Toxoplasma actin was also probed as a loading control. The full-length TgCPDH band was indicated by an arrowhead, and a series of smaller bands were denoted with an asterisk, possibly representing degradation products. Densitometry analysis was derived from five biological replicates. (C) TgCPDH-deficient parasites were grown under both ambient and 3% O₂ conditions for 7 days to evaluate the role of TgCPDH in parasite growth via plaque assay. While wild-type parasites (RH∆ku80::nLuc) formed smaller plaques under 3% O₂ compared to 21% O₂, ∆cpdh::nLuc parasites did not exhibit an increased growth reduction under 3% O₂ compared to ambient O₂ conditions. Thirty-four plaques from three biological replicates were measured for statistical significance calculation. Statistical significance for the data in Panels A–C was assessed using a two-tailed unpaired Student’s t-test, with P values indicated in the plot. (D) TgCPDH was dispensable for acute virulence of Type I Toxoplasma parasites. Five CD-1 mice for each strain received subcutaneous injections of 100 RH∆ku80::nLuc and either 100 or 1,000 ∆cpdh::nLuc parasites, and mouse mortality was continuously monitored and recorded daily.

TgCPDH did not facilitate the growth of heme-deficient Toxoplasma tachyzoites

Due to the successful transgenera complementation of TgCPDH in the heme-auxotrophic Salmonella strain, we expected that TgCPDH could potentially compensate for the absence of TgCPOX and aid in completing heme production since they both can convert COPROgen III to PROTOgen IX. To investigate the transcription and translation levels of TgCPDH in heme-deficient parasites, we employed qPCR and immunoblotting to assess the mRNA and protein abundances of TgCPDH in wild-type RH∆ku80, ∆cpox, and ∆cpoxCPOX parasites, respectively. Under standard growth conditions of 21% O₂, Δcpox parasites exhibited approximately a 2.5-fold higher transcript level of TgCPDH compared to wild-type parasites (Fig. 3A). For protein quantification, initially, we attempted to endogenously tag TgCPDH with a C-terminal 3 × HA epitope tag. However, the signals for both immunoblotting and immunofluorescence proved to be very weak. Consequently, we opted to genetically insert a Spaghetti-Monster 10 × myc (smGFP-myc) epitope tag (26, 27) at the C-terminus of TgCPDH using CRISPR-Cas9-mediated genome manipulation. The predicted molecular weight (MW) of TgCPDH is approximately 112 kDa, while that from the smGFP-myc tag is approximately 40 kDa. As anticipated, the apparent MW for TgCPDH-smGFP-myc was around 150 kDa (Fig. 3B). We also probed the lysates with anti-TgSAG1 antibody as a loading control for normalization. The quantitative analysis revealed that the protein level of TgCPDH in ∆cpox was comparable to that in the wild-type and ∆cpoxCPOX strains. We further conducted an immunofluorescence analysis to assess if TgCPDH alters its subcellular localization in ∆cpox. Our data demonstrated that TgCPDH remained within the mitochondria in all three strains (Fig. 3C), and the fluorescence intensities of anti-myc signals were comparable, corroborating our immunoblotting quantification. To investigate whether TgCPDH plays a crucial role in the growth of ∆cpox mutant, we genetically deleted the TgCPDH gene in the ∆cpox::nLuc strain that had been generated in a prior study, resulting in a ∆cpox∆cpdh::nLuc strain (Fig. S1). The double knockout strain remained viable under ambient oxygen conditions. We quantified its growth through a plaque assay alongside ∆cpox::nLuc and ∆cpoxCPOX::nLuc strains under both 21% and 3% O₂ conditions. Remarkably, we did not observe a significant difference in parasite growth between ∆cpox∆cpdh::nLuc and the other strains under regular and hypoxic conditions (Fig. 3D). In addition, we quantified the total heme abundance in the ∆cpox∆cpdh::nLuc parasites as described previously (12, 28). We did not observe different heme levels between ∆cpox and ∆cpox∆cpdh strains (Fig. S2). In conclusion, our findings suggested that although the parasites increase transcript levels of TgCPDH in the ∆cpox mutant, TgCPDH does not significantly contribute to heme biosynthesis within ∆cpox parasites for their intracellular growth.

Fig 3.

Transcript level of TgCPDH was increased in the ∆cpox mutant, but it was not involved in intracellular growth of ∆cpox. (A) Real-time quantitative PCR analysis revealed approximately a 2.5-fold increase in TgCPDH transcript level in ∆cpox compared to RH∆ku80 and ∆cpoxCPOX parasites. Toxoplasma RNA polymerase II was used as a housekeeping gene for normalization. Five biological replicates were included in the figure. (B) The quantitative immunoblotting analysis did not detect the elevated expression of TgCPDH in ∆cpox mutant. TgSAG1 was used as a loading control. Three biological replicates were included in the figure. (C) Immunofluorescence assay demonstrated that TgCPDH remained localized within the mitochondria of ∆cpox parasites. TgF1β was used as a marker of Toxoplasma mitochondrion. Scale bar: 2 µm. (D) Deletion of TgCPDH in the ∆cpox mutant did not result in a reduction in parasite growth under both ambient and 3% O₂ conditions, as determined by plaque assay. The areas of 12–34 plaques per strain from three biological replicates were measured and presented as mean sizes ± standard deviations. Statistical significance for the data in Panels A–D was calculated using a two-tailed unpaired Student’s t-test, with P values marked in the plot.

The overexpression of TgCPDH did not result in an increased parasite growth rate in heme-deficient parasites

The 2.5-fold increase in TgCPDH mRNA levels observed in ∆cpox did not translate into a significant elevation of TgCPDH protein expression. This suggested that heme metabolism may not be substantially improved in ∆cpox, thereby continuing to limit parasite growth. To address this challenge, we conducted an overexpression experiment by placing TgCPDH under the control of a Toxoplasma tubulin promoter and adding a C-terminal 3 × HA epitope for immunodetection. These plasmids were introduced into RH∆ku80::nLuc and ∆cpox::nLuc strains, resulting in the creation of RH∆ku80::nLuc/pTub-TgCPDH^HA and ∆cpox::nLuc/pTub-TgCPDH^HA strains. First, we assessed the mRNA levels of TgCPDH in the TgCPDH overexpression strains using qPCR. The analysis revealed that the mRNA levels were dramatically elevated compared to the strains not overexpressing TgCPDH and the transcription levels were similar in both RH∆ku80::nLuc/pTub-TgCPDH^HA and ∆cpox::nLuc/pTub-TgCPDH^HA (Fig. 4A). Second, we examined the protein levels of the overexpressed TgCPDH in both strains. Similarly, the protein levels of TgCPDH were comparable when overexpressed in both RH∆ku80 and ∆cpox (Fig. 4B). To ensure that overexpressed TgCPDH maintained its subcellular localization, we conducted immunofluorescence staining using anti-HA antibodies along with anti-Toxoplasma mitochondrial ATPase (TgF1β) for Toxoplasma mitochondrion recognition. The analysis demonstrated that overexpressed TgCPDH still remained localized to the mitochondria (Fig. 4C). We monitored parasite growth by measuring luciferase activity in these strains, as they expressed nanoluciferase. The strains were cultured in confluent HFFs, and parasite growth was assessed every 24 hrs over a total period of 96 hrs. We did not observe an enhanced parasite growth in ∆cpox::nLuc/pTub-TgCPDH^HA relative to ∆cpox::nLuc (Fig. 4D). Similarly, overexpression of TgCPDH in RH∆ku80 did not result in improved growth either (Fig. 4D). In summary, although TgCPDH was effectively overexpressed in Toxoplasma parasites, such overexpression did not lead to an increase in the growth of heme-deficient parasites under standard ambient oxygen growth conditions.

Fig 4.

Overexpression of TgCPDH did not enhance the growth rate of Δcpox. TgCPDH overexpression was achieved by placing an exogenous TgCPDH gene under the control of a Toxoplasma tubulin promoter, and a 3 × HA tag was added to its C-terminus for immunodetection. (A) Real-time quantitative PCR analysis demonstrated a significant increase in TgCPDH transcript levels in the strains overexpressing TgCPDH. Toxoplasma RNA polymerase II served as a housekeeping gene for normalization. Four biological replicates were included in the figure. A two-tailed unpaired Student’s t-test was used to calculate statistical significance with P values marked in the plot. (B) Immunoblotting analysis revealed a comparable expression level of exogenously overexpressed TgCPDH in both RH∆ku80 and ∆cpox parasites. TgActin served as a loading control. (C) Immunofluorescence assay confirmed that overexpressed TgCPDH retained its localization within the mitochondrion of Toxoplasma. TgF1β was included as a marker of Toxoplasma mitochondria. Scale bar: 2 µm. (D) A luciferase-based growth analysis did not detect significant improvement in parasite growth in the ∆cpox mutant overexpressing TgCPDH compared to ∆cpox alone. The assay was conducted in triplicate, and the statistical significance of the growth rates at 96 hrs post-infection was determined using a two-tailed unpaired Student’s t-test, with P values indicated in the plot.

DISCUSSION

Over 600 members of the radical SAM protein superfamily have been identified through bioinformatic searches (29). This group of proteins utilizes an unconventional [4Fe-4S]⁺ cluster as a cofactor to catalyze a reductive cleavage of SAM, generating a radical species, typically a 5-deoxyadenosyl radical, which abstracts hydrogen atoms of the substrates they catalyze (30, 31). These radical-based enzymatic reactions are believed to represent an ancient and conserved mechanistic approach to handling challenging chemical reactions (29). Radical SAM proteins catalyze a wide range of reactions, including the anaerobic oxidations of COPROgen III into PROTOgen IX (18, 22, 24), a critical step within the heme biosynthetic pathway. So far, this reaction has only been observed in some bacteria and is catalyzed by oxygen-independent HemN/HemZ proteins (18, 22, 24, 32). However, a recent examination of the Toxoplasma genome has revealed the presence of an ortholog of HemN, therefore, named TgCPDH (14). Notably, a HemZ ortholog was not found in the Toxoplasma genome. Through primary structure alignment, it is evident that TgCPDH possesses all four catalytic residues found in EcHemN, specifically Y⁵⁶, H⁵⁸, G¹¹¹, and G¹¹³, and additionally features a CXXXCXXC motif for [4Fe-4S]⁺ cluster binding (21, 22, 33, 34). Our transgenera complementation of TgCPDH in a Salmonella mutant deficient in converting COPROgen III to PROTOgen IX demonstrated that TgCPDH exhibits CPDH activity. Notably, bacterial CPDH enzymes encode a fourth Cys preceded by the [4Fe-4S]⁺ binding site to form a CXXXCXXCXC motif, with the first three cysteines being responsible for iron-sulfur cluster binding, while the fourth plays a role in catalysis (22, 24, 35). Some proteins previously classified as HemN have been reclassified as HemW due to the absence of this critical cysteine. As anticipated, HemW cannot rescue the growth of heme-auxotrophic Salmonella mutants (35). Instead, it acts as a heme chaperone, facilitating heme trafficking and insertion into hemoproteins (35). In the case of TgCPDH, only the CXXXCXXC motif is observed, but several cysteine residues follow this motif (Fig. 1A). These additional cysteines may be structurally proximate to the [4Fe-4S]⁺-binding motif and potentially mediate enzymatic catalysis. It is also plausible that TgCPDH functions as a heme chaperone, aiding in the transport and conjugation of heme into mitochondrial hemoproteins, given its mitochondrial localization. To investigate this possibility further, a series of biochemical assays assessing mitochondrial respiration and mitochondrial membrane potentials can be conducted in both wild-type and ∆cpdh parasites, which will provide insights into the potential role of TgCPDH in the parasite’s mitochondrial health.

Our study has revealed that TgCPDH expression is responsive to low-oxygen conditions. In mammalian cells, hypoxia can lead to an accumulation of reactive oxygen species (ROS) in the mitochondria, resulting in cellular damage (36). Consequently, cells must regulate mitochondrial activity to control ROS production (37). Interestingly, our ∆cpdh mutant did not exhibit more severe growth defects under hypoxia conditions; instead, it grew at a similar rate to ambient oxygen conditions. Quantifying ROS levels in the ∆cpdh mutant under both ambient and hypoxia conditions may help determine if CPDH expression levels correlate with ROS levels in the parasites. It is possible that a compensation mechanism operates in the ∆cpdh mutant to offset its loss. We currently do not know what regulates the responsiveness of TgCPDH to low O₂. One possibility is that PHYa regulation of the SCF-E3 ubiquitin ligase complex may control the protein levels of an O₂-regulated transcription factor (6, 38). Alternatively, hypoxia may trigger an increase in ROS that impacts parasite gene expression. Future transcriptomic and proteomic comparisons between wild-type and TgCPDH-deficient parasites may shed light on this compensation mechanism. Furthermore, the activation of HIF is mediated by intracellular iron availability, which further impacts a set of genes activated by HIF (8). This regulation occurs via the control of prolyl hydrolase activity, which relies on iron as a cofactor for this enzyme (8). Consequently, it is plausible that maintaining iron homeostasis plays a critical role in the TgCPDH expression in the parasites.

The immunofluorescence assay has confirmed the mitochondrial localization of TgCPDH, but its exact sub-mitochondrial location remains unknown. Notably, TgCPOX catalyzes the conversion of COPROgen III into PROTOgen IX in the cytosol of the parasites (12). If TgCPDH is located in the mitochondrial intermembrane space or matrix, COPROgen III must cross the mitochondrial membrane(s) via a transporter before undergoing catalysis. In mammalian cells, the ABCB6 transporter has been identified as mediating the import of heme precursors across mitochondrial membranes (39), facilitating the final steps of heme biosynthesis from the cytosol to the mitochondria. Toxoplasma contains an ortholog of ABCB6 (TGGT1_269000) localizing at the mitochondrion (40), but its function requires further investigation.

Despite our findings indicating that TgCPDH exhibits CPDH function via transgenera complementation, the deletion of this ortholog in heme-deficient Toxoplasma ∆cpox parasites did not result in arrested growth and reduced heme production. These findings weaken the hypothesis that TgCPDH is directly involved in heme biosynthesis. Since the ∆cpox mutant remains viable in tissue culture, it is possible that alternative enzymes may facilitate endogenous heme production or that the parasite may acquire heme or heme biosynthetic intermediates from the host. Previous research has shown that the deletion of the last enzyme, TgFECH, within the heme biosynthetic pathway, is lethal (12, 14). Additionally, the addition of exogenous heme in the medium did not enhance the growth of heme-deficient mutants (12, 14). These findings suggest that Toxoplasma parasites are not able to acquire heme from host cells or cannot obtain sufficient amounts to support parasite growth. A previous study proposed that parasites can scavenge PROTO IX or PROTOgen IX, which are the products catalyzed by TgPPO or TgCPOX, respectively, from host cells to boost heme production (14). Notably, PROTOgen IX can be autonomously oxidized into PROTO IX, and the primary difference between PROTO IX and intact heme is the absence of a ferrous ion. Therefore, it remains unclear how PROTO IX can be transported into the mitochondria rather than the intact heme molecule. Furthermore, the observation that deletion mutations in TgALAS and TgUROD mutants, which respectively encode the first and fifth enzymes within the pathway, grew poorly or died in standard D10% medium (12, 13) weakens the speculation that Toxoplasma can scavenge PROTO IX from its host. Future investigations involving isotope-labeled heme or heme intermediates may help trace the fate of host heme or its precursors in the parasites. Furthermore, a study has shown that a HemY protein from Bacillus subtilis can catalyze both CPOX and PPO reactions (41). Hence, it is possible that TgPPO can perform the function of TgCPOX, albeit with lower efficacy, potentially explaining the viability of the ∆cpox mutant. Further experiments testing whether TgPPO can rescue the growth of CPOX-deleted Saccharomyces cerevisiae in a medium lacking heme may help address this question.

MATERIALS AND METHODS

Chemicals and reagents

The chemicals used in this study were of analytical grade and were acquired from Avantor, unless specified otherwise. All oligonucleotide primers used in this work, as listed in Table S1, were obtained from Eurofins.

Host cell and parasite culture

HFFs were provided by the American Type Culture Collection (catalog #: SCRC-1041). HFFs were grown in D10% growth medium consisting of Dulbeccos’s Modified Eagle Medium, 4.5 g/L glucose, 10% Cosmic Calf Serum (HyClone, SH30087.03), 10 mM HEPES, 4 mM glutamine, and 10 mM Pen/Strep. Host cells and parasites were incubated at 37°C with 5% CO₂ and either 21% or 3% O₂. Parasites were grown continuously in a low O₂ environment. Low O₂ growth was maintained in an InVivo2 300 hypoxia chamber (Baker, Sanford, ME, USA). All Toxoplasma strains used for this study were maintained by 2-day serial passage in HFF cells supplemented with D10% media before use in all assays.

Complementation of TgCPDH-like ortholog into Salmonella TE3006

Preparation of competent Salmonella TE3006 cells

The Salmonella TE3006 strain, lacking both the oxygen-dependent CPOX (HemF) and the oxygen-independent CPDH (HemN), was generously provided by Dr. Hans-Martin Fischer from ETH Zürich. Competent Salmonella TE3006 cells were prepared using the Zymo Mix and Go Kit (Zymo Research), following the provided instructions. In brief, the competent bacteria were cultured in Luria-Bertani (LB) broth supplemented with 15 µg/mL hemin, tetracycline, and kanamycin until they reached an optical density of 600 nm (OD₆₀₀) of 0.6. Subsequently, the cells were centrifuged at 3,000 × g for 10 min at 4°C and washed twice with a wash buffer to remove residual hemin. The competent cells were then resuspended in a competent buffer and stored at −80°C for later use in the transformation process.

Construction of plasmids expressing EcHemN or Toxoplasma TgCPDH

The EcHemN gene was PCR-amplified from E. coli genomic DNA and inserted into the pUC19 plasmid using BamHI and EcoRI restriction sites. Based on the high homologous region between TgCPDH and EcHemN from primary structure alignment, we isolated the DNA sequence corresponding to a partial region of TgCPDH, spanning from residues 38 to 790, from a Toxoplasma cDNA library, which contains the conserved catalytic residues and motifs. This sequence was subsequently cloned into the pUC19 plasmid through Gibson DNA assembly. Additionally, we incorporated a 3 × HA epitope tag at the C-termini of both EcHemN and the truncated TgCPDH.

Transformation of Salmonella TE3006 cells

Salmonella transformations were carried out by combining 100 µL of competent cells with 2 µL of plasmid DNA (400–700 ng/µL) in a culture tube, followed by incubation on ice for 30 min. The mixture was then supplemented with 1 mL of uper Optimal broth with Catabolite repression (SOC) medium containing 20 mM glucose and shaken at 225 rpm at 37°C for 1 hr. Subsequently, the bacteria were centrifuged at 3,000 × g at room temperature for 10 min, and the resulting pellet was resuspended in 200 µL of SOC medium with glucose. The suspension was then spread onto LB plates containing ampicillin, kanamycin, and tetracycline, with and without 15 µg/mL hemin. These plates were incubated at 30°C overnight to allow for the formation of single colonies.

Assessment of growth in transformed Salmonella TE3006 cells

The individual clones of Salmonella TE3006 strains transformed with plasmids expressing B. japonicum HemN2 generously provided by Dr. Hans-Martin Fischer from ETH Zürich, EcHemN, or Toxoplasma TgCPDH were cultured overnight in the medium supplemented with ampicillin, kanamycin, and tetracycline, both with and without 15 µg/mL hemin. The cultures were then diluted in 1× phosphate-buffered saline (PBS) to achieve an OD₆₀₀ of 0.8, with three additional 10-fold serial dilutions. Two microliters of the diluted bacterial suspensions was spotted onto LB plates supplemented with ampicillin, kanamycin, tetracycline, and 10 mM isopropyl β-D-1-thiogalactopyranoside (IPTG), with and without 15 µg/mL hemin. After drying for 3 hrs, the plates were incubated at 30°C for 3 days before imaging.

Generation of transgenic Toxoplasma strains

Generation of Δcpdh::nLuc and ΔcpoxΔcpdh::nLuc strains

To create the Δcpdh::nLuc and ΔcpoxΔcpdh::nLuc strains, following established protocols (12, 42), we employed CRISPR-based genome modification techniques to delete the TgCPDH gene (TGGT1_288640) in RHΔku80Δhxg::nLuc and Δcpox::nLuc (Table S2). First, we designed a guide RNA targeting the 3′-end of the TgCPDH gene, as previously described (12, 42). Second, we used primers carrying 50-bp homologous regions flanking the start and stop codons of TgCPDH to amplify a pyrimethamine resistance cassette. After electrophoresis and gel extraction, we mixed the repair template with the guide RNA and introduced it into filter-purified Toxoplasma parasites suspended in Cytomix buffer through electroporation, as detailed in previous studies (12, 42). Following pyrimethamine selection, we cloned out the mutant parasites in 96-well plates pre-seeded with HFFs. The correct mutant clones were screened by PCR to confirm the correct integration of 5′- and 3′-homologous regions (ARMs) and the removal of the TgCPDH coding sequence.

Generation of Spaghetti-Monster-myc tag (smGFP-myc)-labeled TgCPDH strains

To label the TgCPDH gene with a smGFP-myc epitope tag, we genetically inserted this tag at the C-terminus of the TgCPDH gene in RHΔku80, Δcpox, and ∆cpoxCPOX parasites using the CRISPR technique. Similar to the previous procedure, 50-bp homologous regions flanking the stop codon of TgCPDH were placed at the 5′- and 3′-ends of the smGFP-myc epitope tag and a chloramphenicol resistance cassette via PCR. We combined this repair template with a guide RNA targeting the 3′-end of the TgCPDH gene and introduced them into Toxoplasma parasites via electroporation, as outlined above. After transfection, we selected the parasites using chloramphenicol and subsequently isolated individual clones. Correct gene tagging was screened through PCR and immunoblotting.

Generation of Toxoplasma strains overexpressing TgCPDH

We PCR-amplified the coding sequence of TgCPDH from a Toxoplasma cDNA library and constructed it into a plasmid containing a pyrimethamine resistance cassette. The TgCPDH gene was driven by a Toxoplasma tubulin promoter (pTub) for its overexpression, and a 3 × HA epitope was added to the C-terminus of TgCPDH for immunodetection. The resulting plasmids were introduced into RH∆ku80::nLuc and ∆cpox::nLuc parasites using standard electroporation parameters. Following drug selection, we cloned the transfected parasites and subsequently screened them via PCR.

SDS-PAGE and immunoblotting

Parasites were grown in confluent HFFs in D10% medium, following a 2-day pass regimen before each experiment. Parasites were harvested by syringing infected host cells with a 25-gauge needle and purifying the parasites through a 3-µm filter. Purified parasites were pelleted at 1,000 × g and then resuspended in 1× SDS-PAGE sample buffer (40 mM Tris, pH 6.8, 1% SDS, 5% glycerol, 0.0003% bromophenol blue, and 50 mM DTT) with the addition of 2% (vol/vol) β-mercaptoethanol. Subsequently, samples were boiled for 10 min before SDS-PAGE. Proteins resolved on SDS-PAGE were semi-dry transferred onto polyvinylidene fluoride (PVDF) membranes. A chemiluminescent immunoblotting strategy was used to detect target proteins. Initially, the blots were blocked with 5% non-fat milk in PBS-T [PBS buffer with 0.1% (vol/vol) Tween-20]. Primary antibodies were prepared in 1% non-fat milk dissolved in PBS-T. Secondary antibodies, either goat α-mouse or α-rabbit IgG conjugated with horseradish peroxidase, were subsequently applied. To develop chemiluminescence signals, the blots were exposed to SuperSignal WestPico chemiluminescent substrate (Thermo Fisher). Finally, images were captured using the Azure C600 Imaging System.

Immunofluorescence microscopy

Filter-purified parasites were introduced into 6-well chamber slides containing confluent monolayer HFFs. Parasites were cultured in D10% medium at 37°C, with an atmosphere of 5% CO₂ and 21% O₂, for 4 hrs. Following this initial incubation, non-invaded parasites were washed away, and fresh D10% medium was replaced. Slides were then incubated for an additional 20 hrs to allow the parasites to form parasitophorous vacuoles, each containing four to eight parasites, before formaldehyde fixation. To stain the intracellular parasites, infected HFFs were permeabilized using 0.1% Triton X-100 in 1× PBS. Staining was performed using mouse anti-TgF1β antibodies to target the Toxoplasma mitochondrion, along with rabbit anti-HA or rabbit anti-myc antibodies to label TgCPDH. Subsequently, the slides were stained with goat anti-mouse or goat anti-rabbit IgGs conjugated with different fluorescent dyes (Alexa 488 or 594, Thermo Fisher). Observations were made using a Leica DMi8-inverted epifluorescence microscope at a magnification of 1,000×, equipped with a CCD camera. The captured images were then processed using Leica LAS X software.

Real-time quantitative PCR

Toxoplasma parasites were cultured in HFFs for a 2-day period prior to the extraction of total RNA. Extracellular parasites were subjected to filter purification and resuspended in ice-cold 1× PBS. Total RNA was then extracted from the parasites using the Direct-zol RNA MiniPrep Plus Kit (Zymo Research). Approximately 100 ng of total RNA was used in the detection and quantification of TgCPDH transcripts using the Luna Universal One-Step RT-PCR Kit (NEB). Data acquisition was collected using the Bio-Rad CFX96 Touch Real-Time PCR detection system. A double delta cycle threshold (ΔΔCT) analysis was applied to determine the abundance of TgCPDH transcripts relative to those in wild-type parasites, following previously reported procedures (42). Toxoplasma RNA Polymerase II was used as a housekeeping gene for normalization.

Luciferase-based growth assay

Parasites were filter-purified in phenol-red free D10% medium and subsequently inoculated into 96-well white solid-bottom plates with pre-seeded confluent HFFs. Typically, each well was inoculated with 1,500 tachyzoites. However, for strains displaying growth defects, a higher inoculum of 7,500 tachyzoites per well was used to ensure the generation of reliable signals for measurement. After allowing the parasites to invade the host cells for a 4-hr duration, the culture medium was replaced with fresh phenol-red free D10% to eliminate any non-invaded parasites. Bioluminescence signals were then recorded at 24-hr intervals over a total period of 96 hrs, following established protocols (12, 43). The average signal readings at each time point were divided by the average readings at 4 hrs to calculate the fold change of intracellular growth for each strain.

Plaque assay

Freshly lysed parasites were filter-purified and subsequently resuspended in room temperature D10% medium, resulting in a parasite concentration of 100 parasites/mL. A half milliliter of this diluted parasite suspension was then inoculated into 24-well plates pre-seeded with confluent HFFs. Plates were placed in incubators at 37°C with 5% CO₂, subjecting them to both ambient or low-oxygen conditions (21% and 3% O₂, respectively) for a continuous period of 7 days without disturbance. Each well was stained with 0.002% crystal violet in 70% ethanol for a duration of 5 min, followed by a thorough rinse with water. To quantify the differences in plaque size, 12–34 plaques were captured at 25× magnification using a Leica DMi8 microscope. These plaque areas were then plotted and analyzed using Prism GraphPad software to assess size variations among different strains.

Fluorescence-based heme quantification

Parasites were cultured in HFFs in D10% medium for 2 days before use. They were prepared by filter purification using cold 1× PBS. Following centrifugation at 1,000× g for 10 min, the parasites were resuspended in 400 µL of cold 1× PBS. Total heme was released from parasites through sonification with a Branson Analog Sonifier equipped with a mini horn. The sonication was carried out at an output intensity of 3 and a duty percentage of 20% for 10-sec intervals, repeated four times with a 30-sec rest period between each cycle to minimize heat generation. For each strain, two samples were prepared by taking 100 µL of the sonicated parasite suspension and mixing it with 900 µL of 2 M oxalic acid in solid black Eppendorf tubes. One sample was vortexed and boiled for 30 min, while another tube was kept at room temperature to serve as background control. A standard curve was generated using different concentrations of hemin in 1× PBS, including 0, 5, 14.7, 44.3, 133, 400, and 1,200 nM. The fluorescence of all samples and standards was measured at a 400-nm excitation wavelength and a 608-nm emission wavelength using a BioTek H1 Hybrid plate reader. To calculate the total heme content, the readings obtained at room temperature were subtracted from the boiled readings. The fluorescence values were then normalized to the number of parasites for each strain. The total heme content from the wild-type parasites was used as the reference (set at 100%) to calculate the relative heme abundance in the other strains.

Molecular docking analysis

The 3D crystal structure of EcHemN (Protein Data Bank ID: 1OLT) was retrieved from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB) (21). Additionally, the predicted 3D structure of TgCPDH was obtained from the AlphaFold protein structure database (23). Subsequently, the alignment of TgCPDH with the EcHemN structure was carried out using Pymol Molecular Graphics 2.0 (Schrödinger LLC in New York, NY, USA).

Statistical analysis

All statistical analyses were calculated using GraphPad Prism software (Version 8). Detailed methods for individual assays were specified in the figure legends.

ETHICS APPROVAL

CD-1 outbred mice were subcutaneously individually injected with 100 RH∆ku80::nLuc, 100 or 1,000 ∆cpdh::nLuc parasites prepared in 1× PBS. Mice were monitored post-infection for adverse symptoms for a course of 30 days. Mice that showed over 20% drop in initial weight were euthanized by CO₂ in compliance with the protocol approved by Clemson University's Institutional Animal Care and Use Committee (Animal Welfare Assurance A3737-01, protocol number 2019-035). The survival rate was plotted, and statistical analysis was calculated using the log-rank (Mantel-Cox) test.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/msphere.00092-24.

Figure S1. msphere.00092-24-s0001.pdf.

Creation of ∆cpdh::nLuc and ∆cpox∆cpdh::nLuc Toxoplasma strains.

Figure S2. msphere.00092-24-s0002.pdf.

Heme quantification in the TgCPDH-deficient Toxoplasma strain.

Supplemental legends. msphere.00092-24-s0003.pdf.

Legends for supplemental figures and tables.

Supplemental tables. msphere.00092-24-s0004.xlsx.

Primers and Toxoplasma parasites used in this study.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.