is an obligate, intracellular, apicomplexan parasite that is found worldwide. This parasite causes spontaneous abortion in cattle and neural system dysfunction in dogs and results in major economic losses in the breeding industry
. During its life cycle, N. caninum
is exposed to various oxidative stresses, and the parasites develop complex redox networks to maintain redox balance in different stages
. Glutaredoxins (GRXs) are ubiquitous oxidoreductases that maintain a cellular redox balance with the thioredoxin family and catalyse thiol-disulphide exchange reactions by utilizing glutathione (GSH) 
. The number and localization of GRXs differ by species. Humans have four GRXs, which are located in the cytoplasm, nucleus and mitochondria; yeast possesses seven GRXs, located in the cytoplasm, nucleus, mitochondria and endoplasmic reticulum/Golgi 
. Structurally, GRXs are composed of four β-sheets and three α-helices, with the β-sheets surrounded by α-helices 
. GRXs are divided into monothiol (CXXS) GRXs and dithiol (CXXC) GRXs depending on the number of cysteine residues 
. GRXs are involved in DNA/RNA synthesis, Fe–S cluster assembly, cell signal transduction, apoptosis and cell proliferation 
In parasites, GRXs are distributed in different subcellular compartments. For example, Trypanosoma brucei
has five GRXs: two dithiol TbGRXs (TbGRX1 and TbGRX2) and three monothiol GRXs. GRX1 contains the same CPYC active site as human GRX1 but exhibits a greater amino acid identity (39%) with the human mitochondrial GRX2 (CSYC active site). TbGRX1 coordinates iron–sulphur clusters
. TbGrx2 is not essential in vitro or in vivo during the bloodstream stage, but, under fever-like conditions in a mammalian host, TbGrx2 deficiency leads to an increase in thermotolerance. In the procyclical stage, TbGrx2 deficiency significantly affects the morphology of the parasite and leads to irreversible proliferative arrest
. The three groups of monothiol GRXs localize to the mitochondria and cytoplasm and are related to the synthesis of iron–sulphur clusters 
. Additionally, Trypanosoma cruzi
GRX (TcGRX) is linked to apoptosis-like cell death during infection. In the amastigote stage, the overexpression of TcGRX increases its general resistance to oxidative damage and intracellular replication
. Plasmodium falciparum
expresses three monothiol GRX-like proteins (GLP1, GLP2, GLP3), which localize to the cytoplasm and mitochondria. P. falciparum
also has one typical dithiol GRX (PfGRX1), which localizes to the cytoplasm
. To further elucidate the redox-based, parasite–host cell interactions and the mechanisms of antimalarial action, the redox-sensitive, green, fluorescent protein is coupled to human Grx 1 (hGrx1-roGFP2), with pH and glutathione-dependent redox potential in different subcellular compartments detected via the targeted transfer of hGrx1-roGFP2 into the parasite cytoplasm, mitochondria, or apicoplast
2. GRX S14 and GRX C5 Localize to the Apicoplast
To investigate the localization of GRX S14 and GRX C5, we introduced a haemagglutinin (HA) epitope tag at the C-terminus of GRX S14 and GRX C5 in the N. caninum wild-type (WT) strain (Nc1) (Figure 1a). Western blotting verified the expected molecular masses of ~40 kDa for GRX S14-HA and GRX C5-HA (Figure 1b). An immunofluorescence assay (IFA) revealed that GRX S14 and GRX C5 localized to the apicoplast (Figure 1c).
Figure 1. Cellular localization of GRX S14 and GRX C5. (a) Strategy for constructing GRX S14-HA and GRX C5-HA parasites. (b) Western blotting indicated that the HA tag was successfully added. αHA was used to detect GRX S14 and GRX C5; mouse anti-actin was used as a control. (c) IFA indicated GRX S14 and GRX C5 to both be distributed in the apicoplasts of parasites. αHA was used to detect GRX S14 and GRX C5 (green), whereas rabbit anti-SRS2 (red) served as a parasite surface marker. Rabbit anti-ENR (red) was used as an apicoplast marker, and the nuclear DNA was stained with Hoechst (blue) (bar = 5 μm).
3. GRX S14 and GRX C5 Together Affect the Growth of Parasites, and Their Function Depends on the CXXC/CXXS Motif
To investigate the function of the GRX S14 and GRX C5 proteins, it generated two single-gene knockout strains (Δgrx C5 and Δgrx S14) and overexpression strains through CRISPR/Cas9-mediated homologous recombination (Supplementary Figure S1c,d). All strains were validated using PCR. The plaque formation ability of these strains was analysed, and the plaque assays showed no obvious difference among the Nc1, single-gene knockout and overexpression strains (Figure 2a) (Δgrx C5, F(2, 56) = 3.104, p = 0.0527; Δgrx S14, F(2, 30) = 2.500, p = 0.0990). Moreover, the steps of the lytic cycle (invasion, intracellular replication and egress) were not significantly affected in the gene-edited strains (Figure 2b–d).
Figure 2. Lack of GRX S14 and GRX C5 alone did not affect the growth of parasites. (a) Plaque assays comparing growth of wild-type, knockout and overexpression parasites. Each well contained 300 parasites, and plaques were stained for 9 days. Plaque areas were counted by randomly selecting at least 20 plaques and measuring the pixel point with Photoshop C6S software (Adobe, San Jose, CA, USA). Data were compiled from three independent experiments. (b) A total of 1 × 105 parasites were inoculated on human foreskin fibroblast (HFF) cells in 12-well plates and cultured for 24 h. IFA was performed with anti-NcSRS2 antibodies and Hoechst staining. The invasion ratio of wild-type, knockout, overexpression and complementary parasites was based on the number of parasite-infecting cells divided by the number of total cells in one horizon. Data are the mean ± SD (error bars) of three independent experiments. Statistical analysis showed no significant difference (Δgrx C5, F(2, 6) = 0.06720, p = 0.9357; Δgrx S14, F(2, 6) = 2.238, p = 0.1879). (c) Intracellular replication of different parasite strains was compiled from three separate assays, with 100 total PVs of each strain counted in each assay. Statistical analysis showed no change (Δgrx C5, F(8, 15) = 0.6462, p = 0.7287; Δgrx S14, F(8, 15) = 0.6732, p =0.7080). (d) The egress ability of parasites was assessed after treatment with the calcium ionophore A23187. IFA was employed to detect the integrity of the parasitophorous vacuole (PV). The average number of ruptured PVs was determined by counting 100 random vacuoles per slide. Statistical analysis showed no significant difference (Δgrx C5, F(2, 6) = 2.734, p = 0.1432; Δgrx S14, F(2, 6) = 3.622, p = 0.0930). (e) Mouse survival after infection with different strains. BALB/c mice (n = 5) were injected intraperitoneally with 8 × 106 doses of parasites. Statistical analysis was performed using the survival curve of GraphPad Prism (San Diego, CA, USA).
Because the single-gene knockout strains of N. caninum did not exhibit altered growth, we hypothesized that the GRX S14 and GRX C5 proteins had a synergistic effect on the growth of the parasites. To test this hypothesis, we constructed double-gene knockout strains (Δgrx C5Δgrx S14). PCR verified the successful construction of the Δgrx C5Δgrx S14 strain (Figure 3a), and a significant reduction in plaque formation size was observed in the Δgrx C5Δgrx S14 parasites compared with the Nc1 parasites (Figure 3b) (t-test: t(58) = 7.758, p < 0.0001). This entry further evaluated the influence on Δgrx C5Δgrx S14 parasite growth in vivo. The survival rate of mice infected with the Δgrx C5Δgrx S14 parasites was 80%, and that of mice infected with Nc1 was 40%, indicating a significantly reduced pathogenicity for the Δgrx C5Δgrx S14 parasites in mice (Figure 3c). Subsequently, we compared the steps of the lytic cycle (invasion, intracellular replication and egress) between the Nc1 and Δgrx C5Δgrx S14 parasites. Although the intracellular replication of the Δgrx C5Δgrx S14 parasites was significantly reduced (Figure 4d) (F(3, 8) = 5.005, p = 0.0305), the invasion and egress were not affected (Figure 3e,f) (t-test: t(2) = 1.901, p = 0.1977; t(2) = 0.5156, p = 0.6575).
Figure 3. Lack of GRX S14 and GRX C5 together affected the growth of parasites. (a) PCR identification of the Δgrx S14 Δgrx C5 strain. PCR1 and PCR2 suggest successful homologous integration; in PCR3 and PCR4, fragments of grx S14 and grx C5 were amplified. (b) Plaque assay comparing the growth of wild-type and knockout parasites. (c) Mouse survival after infection with different strains. (d) Intracellular replication of Δgrx S14 Δgrx C5 compared with Nc1 parasites. Asterisks indicate statistically significant results. (e) Invasion assay of Δgrx S14 Δgrx C5 and Nc1 parasites. (f) The egress ability of Δgrx S14 Δgrx C5 and Nc1 parasites. Statistical analysis was performed using GraphPad Prism (San Diego, CA, USA). n.s: not significant means, *** p < 0.001, and * p < 0.01.
To assess the importance of the putative CXXS active site of GRX S14 and the CPFC active site of GRX C5, it used ΔΔgrx S14 grx C5 as the base strain and generated a complementary grx S14 sequence with a mutation of the cysteine in the CXXS active site to alanine at the UPRT site (comΔgrx S14AXXSΔgrx C5). In the same way, we constructed the grx C5 mutant strain (Δgrx S14comΔgrx C5AXXA), grx C5 complemented strain (Δgrx S14comΔgrx C5) and grx S14 complemented strain (comΔgrx S14Δgrx C5). The phenotype assays showed that complementing the full sequence of grx S14 or grx C5 could restore the growth ability of the parasites (Figure 4) (F(2, 57) = 21.77, p < 0.0001), but complementing the mutant sequence could not restore growth (F(2, 57) = 1.126, p = 0.3316).
Figure 4. Phenotypes of strains with mutations of key amino acids. Plaque assay comparing growth of the comΔgrx S14AXXSΔgrx C5, Δgrx S14comΔgrx C5AXXA, comΔgrx S14Δgrx C5, Δgrx S14comΔgrx C5, Δgrx S14Δgrx C5 and Nc1 parasites. n.s: not significant means, *** p < 0.001.
4. Redox Homeostasis of Δgrx C5Δgrx S14 Parasites Was Not Affected
As GRXs play crucial roles in redox-dependent signalling pathways by utilizing GSH as a direct electron donor 
, we measured the GSH/GSSG content in Δgrx
S14 parasites. The results showed no significant differences in the GSH/GSSG ratio between Δgrx
S14 parasites and Nc1 parasites (Figure 5
= 0.3968, p
Figure 5. grx C5 and grx S14 double-gene deficiency did not affect the levels of GSH, ROS and OH. (a) Δgrx S14Δgrx C5 and Nc1 parasites were collected and lysed by three cycles of freezing in liquid nitrogen and thawing at 37 °C. The supernatant of each sample was collected for GSH and GSSG measurement. The GSH/GSSG ratio was calculated, as represented by bar charts according to three independent experiments. (b) Reactive oxygen species (ROS) levels of parasites under oxidative stress were determined by FACS analysis using DCFH-DA, whereby the mean fluorescence intensity reflected the ROS level in parasites. (c) The hydroxyl radical (OH) content was detected by FACS analysis, with the mean fluorescence intensity reflecting the hydroxyl radical level. n.s: not significant means.
GRXs play an important role in the reactive oxygen species (ROS) antioxidant system. To examine redox homeostasis, we compared the ROS and hydroxyl radical (OH) levels between Δgrx C5Δgrx S14 and Nc1 parasites and found no significant differences between the two strains (Figure 6b,c) (t-test: t(2) = 0.9853, p = 0.4284; t(2) = 2.371, p = 0.1411). These data indicate that the double-gene deficiency of GRX C5 and GRX S14 does not affect the redox homeostasis of tachyzoites.
5. GRX S14 and GRX C5 Are not Involved in the Control of Protein Trafficking to the Apicoplast
Previous research showed that thioredoxins contributed to the control of protein trafficking to the apicoplast [17
]. The localization of the apicoplast proteins, the enoyl acyl carrier protein reductase (ENR) and acyl carrier protein (ACP), was observed in Nc1, Δgrx
C5 and comΔgrx
C5. As the localization of the apicoplast proteins ENR and ACP in the deletion or mutant strain was normal (Figure 6
), GRX S14 and GRX C5 do not appear to influence protein trafficking to the apicoplast.
Figure 6. Location of apicoplast proteins in grx S14 and grx C5 deletion and mutant strains. Locations of ENR (green) and ACP (red) in comΔgrx S14AXXSΔgrx C5, Δgrx S14comΔgrx C5AXXA, comΔgrx S14Δgrx C5, Δgrx S14comΔgrx C5, Δgrx S14Δgrx C5 and Nc1 strains were detected by IFA. Parasite shapes were visualized with anti-NcSRS2 (red); nuclear DNA was stained with Hoechst (blue). Scale bar = 5 μm.
6. Double-Gene Depletion Affects the Expression of Several Proteins
To explore which pathways of the parasite are affected by double-gene depletion, it performed a comparative proteomic analysis between Δgrx S14Δgrx C5 and the Nc1 parasites. The Δgrx S14Δgrx C5 parasites showed a significant downregulation of 29 proteins (fold change > 2, p-value < 0.05, Figure 7a). According to GO enrichment analysis, the biological functions of the proteins with downregulated expressions were mainly related to the electron transport chain and tricarboxylic acid cycle (Figure 7b,c).
Figure 7. The comparative proteome of Δgrx S14Δgrx C5 and Nc1. (a) Volcano plots showing log2 protein ratios vs. −log2 p values of the global proteome in Δgrx S14Δgrx C5 compared to Nc1 parasites. (b) Gene Ontology (GO) analysis of proteins with downregulated expressions in Δgrx S14Δgrx C5 compared to Nc1 parasites based on biological process and (c) molecular function (see Supplementary Dataset S1).
GRXs are ubiquitous oxidoreductases that maintain a cellular redox equilibrium and catalyse thiol-disulphide exchange reactions by utilizing GSH 
. GRXs are classified as monothiol (CXXS) or dithiol (CXXC) GRXs depending on the number of cysteine residues present in the redox active site
. The biological functions of GRXs include DNA/RNA synthesis, Fe–S cluster assembly, cell signal transduction, apoptosis and cell proliferation 
. Only a few GRXs from parasites have been reported, mainly for trypanosomes and malaria parasites
The entry identified five putative GRXs in N. caninum
. Our previous study showed that the GRX1 (NcGRX1) and GRX3 (NcGRX3) of N. caninum
are located in the cytoplasm 
and NcGRX5 in mitochondria. The present study found that GRX S14 and GRX C5 localize to the apicoplasts. The apicoplast is an essential, nonphotosynthetic plastid found in related apicomplexan pathogens 
. This organelle is the product of a secondary endosymbiosis event and is homologous to the chloroplasts of algae and plants
. Therefore, the functions of GRX S14 and GRX C5 might be similar to those of plant GRXs. GRXs have a crucial role in the developmental process of A. thaliana
. For example, the lack of Class I GRX C1 and GRX C2 proteins in Arabidopsis
can lead to an impaired embryonic development and even cause death 
GRX S14, GRX S15, GRX S16 and GRXS17 are Class II GRXs. GRX S14 is composed of two domains: an N-terminal domain with an endonuclease activity and a C-terminal domain with a GRX motif 
. The silencing of tomato GRXS16 results in an increased sensitivity to osmotic pressure 
. GRX S17 consists of a TRX-like domain and three GRX domains and plays a key role in controlling plant development. The A. thaliana
GRX S17 mutant strain displays a slowed primary root growth and impaired flowering at 28 °C 
. This mutant strain exhibits severe nutritional and reproductive development impairment under a long-day photoperiod 
. GRX S14 and GRX S15 are associated with oxidative stress, high temperature and arsenic exposure 
One study showed that NcGRX1 is important for microneme protein-mediated parasite growth, but that NcGRX3 deficiency does not affect parasite growth 
. Our study showed that the deletion of GRX S14 or GRX C5 alone did not affect the growth of N. caninum
, which was consistent with previous research on Arabidopsis
. GRX C5 has a CPFC active site and is homologous to Arabidopsis
Class I GRX C5. AtGRXC5, which has two forms, is expressed in Escherichia coli
. The monomeric apoprotein of AtGRXC5 exhibits a deglutathionylation activity in mediating the recycling of the plastidial methionine sulfoxide reductase B1 and peroxiredoxin IIE. The dimeric holoprotein of AtGRXC5 incorporates a (2Fe–2S) cluster 
. In our study, single-gene deletion did not affect the growth of parasites, whereas the simultaneous deletion of both apicoplast GRXs reduced their growth. These results revealed that GRX S14 or GRX C5 might have a synergistic effect during parasite growth. In addition, parasites with a mutated cysteine in the CPFC motif of GRX C5 or the CGYS motif of GRX S14 displayed a reduced growth, indicating that these motifs were the key active sites of each GRX.
GRX S14 is a new signalling molecule in plants that regulates the Ca2+
transport activity of CAX1 by interacting with the N-terminal region of CAX1 (cation exchanger)
and protecting against protein oxidative damage 
and poplar GRX S14 are located in the chloroplast and form a bridge with the (2Fe–2S) cluster and two external GSH ligands. GRX S14 is used as a scaffold protein for (2Fe–2S) cluster assembly because it transfers the complete cluster to the receptor protein regulated by GSH 
. Our previous research revealed that N. caninum
GRX1 deficiency decreases the ratio of reduced GSH/GSSG, causing a significant accumulation of hydroxyl radicals in parasites, and increases the number of apoptotic cells under oxidative stress (H2
) conditions 
. In the present study, GRX S14 and GRX C5 double-gene deletion did not affect the GSH/GSSG ratio of parasites, nor did it alter levels of ROS and OH. Thus, GRX S14 and GRX C5 may not have important roles in regulating the redox balance in N. caninum
In addition, glutaredoxin and thioredoxin are oxidoreductases that together maintain a redox balance in cells. Previous studies have found that thioredoxin 1 of T. gondii
is located in the apicoplast and involved in the control of protein trafficking to this organelle 
. However, our results showed that a lack of both GRX proteins or mutations in the key active site did not affect localization of the apicoplast proteins ACP and ENR, indicating that the apicoplast GRX might not be involved in apicoplast protein import.
Apicoplast processes involve multiple metabolic pathways, including the synthesis of haem, type II fatty acids, and isoprenoid precursors, among others 
. After the double deletion of grx
S14 and grx
C5, only 29 proteins were downregulated more twice, and no known apicoplast proteins were identified. Surprisingly, the downregulated proteins are involved in the mitochondrial electron transport chain (ETC) and TCA cycle. This result suggests that GRX S14 and GRX C5 may not be involved in apicoplast function but instead may be related to the ETC and TCA cycle. Regardless, the mechanism by which apicoplast proteins are involved in the mitochondrial ETC or TCA process remains unclear. In summary, we identified two new GRXs localized to the apicoplasts. Double-gene deletion resulted in a significant growth defect and caused the downregulation of the expression of proteins involved in the electron transport chain and TCA cycle.