Plasmodium falciparum chaperonin 60 is an apicoplast-localized protein

Our previous work suggested that the apicoplast employs caseinolytic protease (Clp)-mediated degradation to maintain organellar functions [14,15]. In order to further investigate proteostasis regulation in the apicoplast, we performed an affinity screen, immunoprecipitating several different Clp subunits coupled with mass spectrometry analysis [15]. Notably, we found a conserved apicoplast-resident heat shock protein (HSP60) homolog, termed chaperonin 60 (CPN60, PF3D7_1232100). The putative apicoplast CPN60 in P. falciparum is 718 amino acids in length, while the mitochondrial HSP60 (PF3D7_1015600) is 580 amino acids long, and the two isoforms share 30% sequence identity (46% similarity) (S1 Fig). To test whether and how CPN60 is involved in proteostasis and other apicoplast functions, we employed a genetic approach using the tetR-Dozi system, a Plasmodium-specific method [27] (Fig 1A). Using CRISPR/Cas9, we incorporated a V5 tag followed by the regulatory cassette at the 3′ end of the CPN60 gene, creating the CPN60^V5-apt parasite line. Genotyping of independently isolated clones was performed by PCR analysis, and showed correct integration into the cpn60 locus (Fig 1B). Western blot analysis revealed expression of endogenous V5-tagged CPN60 at the predicted 85 kDa size (Fig 1C). The presence of a double band suggested that CPN60 contains an apicoplast-targeting transit peptide that is cleaved upon its organellar localization. To verify this, we treated CPN60^V5-apt parasites with the antibiotic chloramphenicol (CHL), which specifically inhibits apicoplast protein translation and leads to organelle loss [6]. 96-h post CHL treatment, we observed accumulation of the higher molecular weight band of CPN60 and disappearance of its  lower molecular weight form (Fig 1C). This indicates that CPN60 has a transit peptide that cannot be cleaved off when organelle functions are compromised. Homologs of the HSP60 family are often considered house keeping genes, retaining constant expression levels throughout the cell cycle [23]. However, protein extraction from synchronized parasites revealed that CPN60 levels were relatively low at the first half of the 48-h asexual cycle, and increased only at the later trophozoite stage, 36 hours post-invasion (Fig 1D, see protein levels and corresponding blood smears). This was also manifested with increased CPN60 mRNA levels, as measured by quantitative real-time PCR (Fig 1E). Quantification of CPN60 protein levels shows that expression peaks around the time that schizogony (nuclear replication), and apicoplast development begin [28] (Fig 1F).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Generation and characterization of transgenic CPN60^V5-apt parasites.

A. Schematic representation of CRISPR-mediated homologous recombination through CPN60 homology regions enabling the integration of the V5 tag and the tetR-Dozi system. B. PCR amplification of genomic DNA isolated from CPN60^V5-apt transfected parasites. Primers used are depicted in panel A and in Table 1. Primers are integration specific and thus can amplify 1.6 KB (P7 + P6) and 11 KB (P7 + P4) products only if integration occurs at the desired genomic locus. C. Western blot depicting two clones of CPN60^V5-apt parasites (A4 and F11) following 96-hour incubation with or without 30 μM chloramphenicol (CHL) using an anti-V5 antibody and anti-aldolase as a loading control. D. CPN60^V5-apt parasites were synchronized by incubation with 5% sorbitol followed by Percoll-mediated schizont-enrichment the following day. Western blot depicting expression profile of tagged CPN60 throughout parasite life cycle using an anti-V5 antibody. Right: Blood smears demonstrate parasites’ developmental stage at the time of protein extraction. E. CPN60^V5-apt parasite clones (A4 and F11) were synchronized using 5% sorbitol, and 8ml of culture were collected at the following time points: early ring stage (~12 h post-invasion), early trophozoite stage (~28 hpi), late trophozoite stage (~36 hpi), and schizont stage (~44 hpi). Parasites were liberated from host red blood cells using 0.1% saponin, and total mRNA was extracted and reverse-transcribed to cDNA. Quantitative real-time PCR (qRT-PCR) was performed to quantify transcript levels of the apicoplast-encoded chaperonin CPN60. Primers P14 + P15 were used to amplify CPN60 and two housekeeping genes (aldolase P16 + P17 and arg-tRNA synthetase P18 + P19) for normalization. Gene expression was normalized to the mean ΔCt of both housekeeping genes (aldolase and arg-tRNA synthetase), and expression levels were calculated relative to the early ring stage using the ΔΔCt method. Data are presented as fold change (2^–ΔΔCt). Values represent the mean expression from two parasite lines, and error bars indicate the standard deviation. Statistical analyses were performed using GraphPad Prism. F. Densitometric analysis of protein bands from Western blot in 1D. V5 signal was normalized to aldolase. Figure created by Biorender.

https://doi.org/10.1371/journal.ppat.1013275.g001

We further monitored CPN60 expression and localization in different developmental stages using immunofluorescence assay (IFA). Recently, we showed that apicoplast development is tightly coordinated with nuclear replication and the cell cycle [29]. In accordance with these observations, we saw CPN60 at an early stage in a dot-like structure, in a typical elongated apicoplast morphology during the late trophozoite stage, a branched structure during schizogony and divided daughter organelles post segmentation, in the last hour before egress (Fig 2A). We took advantage of ultrastructure expansion microscopy to verify the details of these structures and observed CPN60 in a consistent and uniform organelle localization (Fig 2B). Lastly, we combined this IFA with a mitotracker staining to demonstrate that the intricate structure is distinct from the parasite’s mitochondrion and therefore marks the apicoplast (Fig 2C).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Organellar Localization and expression of CPN60.

A. Immunofluorescence microscopy of CPN60^V5-apt parasites in different developmental stages. Z stack images processed as Maximum Intensity show from left to right: merged DIC contrast, anti-V5 (CPN60, red), DAPI (parasite nucleus, blue), and merge of fluorescent channels. The imaging was performed using a Nikon Spinning Disk confocal fluorescence microscope equipped with a 1005 x/1.4NA objective. Scale bar is 2.5 μM. B. A representative image of an expanded parasite demonstrates uniform apicoplast localization of CPN60 in a well-developed organelle during mid-schizogony (eight nuclei). Images from left: NHS-Esther (total protein, grey), merge, DNA (red), anti-V5 antibody (CPN60, green), and a 3D visualization. Images were captured as Z-stacks (20 slices of 0.1 μm each) at 63x objective using Airyscan microscopy. The images are displayed as maximum intensity projections. Expansion factor is 5X. Scale bar is 5 μm. C. Immunofluorescence microscopy of CPN60^V5-apt parasites processed as Maximum Intensity show from left to right: merged DIC contrast, DAPI (parasite nucleus, blue), MitoTracker Deep Red FM (mitochondrion, magenta), anti-V5 antibody (CPN60, green), and merge of fluorescentt channels. The imaging was performed using a Nikon Spinning Disk confocal fluorescence microscope equipped with a 100x/1.4NA objective. Scale bar is 2.5 μM.

https://doi.org/10.1371/journal.ppat.1013275.g002

CPN60 is essential for parasite’s survival due to its function in apicoplast biogenesis

To look into CPN60 cellular function, we activated the tet-aptamer system, to induce translational repression of CPN60 (Fig 3A). We induced protein knockdown by removal of the stabilizing ligand anhydrotetracycline (aTC), effectively reducing CPN60 protein levels within 24 hours, dropping below detection after 72 hours (Fig 3B). The consequences of CPN60 knockdown were parasite’s death, indicating that its activity is essential for parasite’s viability (Fig 3C). It was shown that apicoplast-related cytotoxicity can be reversed by addition of isopentenyl pyrophosphate (IPP), an essential apicoplast-producing metabolite [11]. We observed that media supplementation with IPP completely restored mutants’ growth, indicating that CPN60 essential activity is restricted to apicoplast function (Fig 3D). Delayed cell death due to apicoplast-associated damage is observed in parasites treated with doxycycline or chloramphenicol. Such parasites complete their first 48 hours cell cycle, reinvade new cells and die only at the end of the 2^nd cycle after 96 hours of treatment [30]. We therefore synchronized the cultures and removed aTC from early rings stage CPN60^V5-apt parasites and observed their intraerythrocytic development (Fig 3E). During the first cycle, knockdown parasites developed like +aTC control until the trophozoite stage, but then could not process into normal schizogony (Fig 3F). The few persisting parasites in the culture could not egress or reinvade and failed to form new ring-stage parasites in the 2^nd cycle (Fig 3G). For comparison, we induced knockdown in a different apicoplast-resident protein, PfClpS, which we previously showed to be required for organelle biogenesis [15]. Induction of PfClpS knockdown using the same tet-aptamer system under the same experimental settings resulted in a delayed cell death, with parasites death observed 5 days post aTC removal (S2 Fig). This reinforces the observation that, unlike many other apicoplast proteins, CPN60 ablation results in an apicoplast-immediate cell death.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. CPN60 is essential for parasite’s viability due to its apicoplast function.

A. Schematic representation of the tetR-Dozi system, which enables protein knockdown upon removal of the tetracycline analogue aTC. B. Western blots depicting the expression of the tagged CPN60 using an anti-V5 antibody following knockdown induction. A significant reduction in protein levels is already observed after 24 hours, and falls below detection 72 hours after aTC removal. C. Growth curve of CPN60^V5-apt parasites. Independent CPN60^V5-apt clones A4 and F11 were grown with or without 0.5 mM aTC, and parasitemia was monitored every 24 h over two cycles via flow cytometry. The 100% of growth represents the highest value of calculated parasitemia (final parasitemia in the presence of aTC). Normalized data are represented as mean ± SEM for three technical replicates. D. Growth curve of CPN60^V5-apt parasites, clone A4, supplemented with 200 μM IPP. The 100% of growth represents the highest value of calculated parasitemia (final parasitemia in the presence of aTC). Normalized data are represented as mean ± SEM for three technical replicates. E. CPN60^V5-apt parasites were synchronized by incubation with 5% sorbitol followed by Percoll-mediated schizont-enrichment the following day. Synched parasites were washed 8 times and then incubated with or without aTC. Giemsa-stained blood smears were imaged using an upright Eclipse E200 Microscope. F. Growth rates of synchronous CPN60^V5-apt parasites. Clones A4 and F11 were grown with or without 0.5 mM aTC, and parasitemia was monitored every 24 h over two cycles via flow cytometry. Data are represented in columns as mean ± SEM for three technical replicates. G. Blood smears of synched CPN60^V5-apt parasites with or without aTC were analyzed for specific parasites developmental stages. Minus aTC parasites demonstrate a delayed trophozoite stage already on the first replication cycle, fail to egress, and die during schizogony. Dead schizonts are defined as morphologically aberrant mutant cells with more than one nucleus that linger for more than 48 hours but did not complete their first replication cycle. H. Parasites were cultured without aTC and supplemented with IPP for 17 days. Parasites culture was then divided into four groups treated as following: (A) +aTC, + IPP; (B) +aTC, -IPP; (C) -aTC, + IPP; (D) -aTC, -IPP. Parasitemia of these four groups was monitored every 24 hours over two cycles via flow cytometry. Normalized data are represented as mean ± SEM for three technical replicates. I. Immunofluorescence microscopy of untreated and group (A) CPN60^V5-apt parasites. Z stack images processed as Maximum Intensity show from left to right: merged DIC contrast, anti-V5 (CPN60, red), DAPI (parasite nucleus, blue), and merge of fluorescentt channels. The imaging was performed using a Nikon Spinning Disk confocal fluorescence microscope equipped with a 100x/1.4NA objective. Scale bar is 2.5 μM. J. Western blot depicting expression of untreated and group (A) CPN60^V5-apt parasites using an anti-V5 antibody and anti-aldolase as a loading control.

https://doi.org/10.1371/journal.ppat.1013275.g003

We then tested whether essentiality of CPN60 is limited to supporting an apicoplast metabolic pathway, or it is also required for the physical biogenesis of the organelle. If the former is correct, then IPP supplementation during CPN60 knockdown should be sufficient to retain an intact organelle, albeit dysfunctional. However, if the latter is true and CPN60 is required for apicoplast biogenesis, its knockdown will result in organelle loss, regardless of IPP supplementation. To distinguish between these two possibilities, we grew CPN60^V5-apt parasites for 17 days without aTC while supplementing the media with IPP, and then added back aTC to restore CPN60 expression (Fig 3H). We observed that this treatment resulted in parasites growth that is completely dependent on IPP supplementation, regardless of aTC presence (i.e., CPN60 expression) (Fig 3H). Moreover, IFA showed that CPN60 in treated parasites lost its distinct apicoplast elongated structure and appeared instead in vesicle-like cytoplasmic puncta (Fig 3I). Moreover, processing of CPN60 was lost and it migrated on the gel at the size of its cytoplasmic form (Fig 3J). Collectively, these results demonstrate that CPN60 activity is required for apicoplast biogenesis, and during knockdown the organelle is lost and cannot be retrieved even when CPN60 expression is restored.

Plasmodium CPN60 is not involved in heat shock response

In most organisms, HSP60 family members are involved in stress pathways, particularly heat shock (HS) response. During infection, the malaria parasite faces periodic fevers, making HS one of its main stressors. To investigate the function of apicoplast CPN60 during HS, we measured its expression levels following 24 hours incubation at 40^oC. Wildtype parasites subjected to HS, grew slower and exhibited morphological defects (Fig 4A). However, this response was not accompanied by elevated CPN60 protein levels (Fig 4B). We also followed potential changes in transcript levels following HS, and as a positive control used leucine-rich repeat protein (LRR5), a Plasmodium gene that is nonessential under ideal growing conditions, however was found to be associated with HS response in the parasite [31,32]. Here too, we found that while LRR5 transcripts were elevated by two- and three-folds post HS, there was no effect on CPN60 transcription (Fig 4C).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. CPN60 is not involved in Heat Shock response.

A. Workflow of HS experiments and representative blood smears. Parasites are subjected to HS at 40^oC for 24 hours and then allowed to recover at normal temperature (37^oC). B. Parasites were subjected to HS and recovery as depicted in (A). Left: Cellular lysates were extracted every 24 hours over a span of four days and subjected to Western blot analysis using an anti-V5 antibody to follow tagged CPN60 levels. Right: Quantification of CPN60 levels post-heat shock demonstrate no significant increase in protein levels. C. P. falciparum culture was subjected to heat shock at 40°C for 24 hours, followed by return to standard conditions (37°C). Twelve milliliters of parasite culture were collected at 24, 48, and 72 hours post-heat shock for RNA extraction and cDNA synthesis. Quantitative real-time PCR (qRT-PCR) was performed to assess the expression levels of CPN60 (P14 + P15), and the heat shock-responsive gene LRR5 (P20 + P21). Gene expression was normalized to the mean ΔCt of two housekeeping genes; aldolase (P16 + P17) and arg-tRNA synthetase (P18 + P19), and values were calculated relative to the non-heat-shocked control using the ΔΔCt method. The graph displays –ΔΔCt values. Error bars represent the standard deviation from three biological replicates. While LRR5 expression increased post-heat shock, CPN60 levels remained unchanged, suggesting that CPN60 is not transcriptionally regulated in response to heat stress. D. To calculate EC50 of aTC, CPN60^V5-apt parasites were thoroughly washed to remove aTC, and then incubated in serial aTC dilutions in a 96-well plate. Parasitemia was measured after 4 days using flow cytometry, showing an EC50 of 4.3 nM at 37^oc or 4.5 nM following 24 h incubation at 40^oc. Data are fit to a dose-response equation and are represented as mean ± SEM. One representative experiment out of three is shown. E. Parasites were thoroughly washed and then incubated with 8 nM aTC (2X EC50). Left: Parasites lysates were obtained every 24 hours and subjected to western blot analyses by being probed with antibodies against V5 (green) and aldolase (loading control, red). The protein marker sizes that co-migrated with the probed protein are shown on the left. Right: Densitometric analysis of protein bands on the left indicating 50% drop in CPN60 expression. F. CPN60^V5-apt parasites were washed and incubated with different aTC concentrations (8 nM, 4 nM, 16 nM and 500 nM), subjected to HS and then allowed to grow at 37^oC for three days, while being measured daily by flow cytometry. Data are fit to an exponential growth curve and are represented as mean ± SEM for three technical replicates.

https://doi.org/10.1371/journal.ppat.1013275.g004

Next, we sought to analyze the effect of CPN60 depletion on parasite’s survival following HS. Because complete knockdown of CPN60 leads to parasite’s death, we were interested in decreasing CPN60 levels to the point where subtle effects could be tested. To do that, we performed a half-maximal effective concentration (EC50) assay for aTC, and found that 4.5 nM support 50% survival under normal temperature as well as following incubation at 40^oC (Fig 4D). Next, we measured CPN60 protein levels while growing the parasite’s with 2X EC50 dose (8 nM aTC) compared to its levels in non-induced parasites which are cultured at 100X EC50 (500 nM aTC, Fig 4E). Quantification revealed that at this aTC concentration, CPN60 cellular levels were decreased by 50% (Fig 4E). Next, we measured parasite’s growth at different aTC concentrations following 24 hours incubation at 40^oC. While wildtype parasites recover after heat exposure, their growth is slightly attenuated (Fig 4F, blue curve). Importantly, parasites with reduced CPN60 levels (16, 8, 4 nM which correspond to purple, green and red curves, respectively in Fig 4F) exhibited compromised growth, in an aTC dose-dependent manner. However, exposure to heat did not alter their relative growth rate compared to the effect that it had on wildtype parasites (Figs 4F and S2B). These data suggest that unlike canonical CPN60 orthologues, the Plasmodium CPN60 is probably not involved in HS response and possesses an apicoplast-specific function.

Co-expression and co-localization of CPN60 and PfClpP, the active protease of the Clp complex

We previously identified CPN60 in a mass spectrometry screen as a potential interactor of the Clp complex [15]. To look deeper into this potential interaction, we expressed a tagged copy of the Clp protease subunit, PfClpP. PfClpP is a highly potent protease and in order to avoid adverse effects, we used a transgenic PfClpP with a point mutation in its proteolytic active site (S264A), rendering it catalytically inactive [15]. Furthermore, we designed integration into the parasite’s genome under the endogenous hsp110 promoter, which enables constant but moderate levels of expression [15]. This construct was transfected on the background of the CPN60 conditional mutants, generating the CPN60^V5-apt; ClpP^DEAD-Ty parasite line (Fig 5A). Clones were isolated and integration at the hsp110 locus was verified using diagnostic PCR (S3A Fig). Western blot demonstrated the co-expression of the endogenous CPN60^V5 at 85 KDa, as well as the transgenic ClpP^DEAD-Ty (Fig 5B). The expected processing pattern of PfClpP is consistent with a model of a full-length PfClpP, presumably in transit through the secretory pathway (fraction III), an apicoplast-localized PfClpP (fraction II), and a mature PfClpP after proteolytic removal of an inhibitory pro-domain (fraction I) (Fig 5B). We verified that expression of ClpP^DEAD-Ty had no effect on parasite’s growth or CPN60 function, as can be seen by a consistent EC50 at 37^oC or following HS (Fig 5C). Immunofluorescent microscopy, revealed that CPN60 co-localizes with PfClpP in the apicoplast during all cell cycle stages (Fig 5D). Therefore, we concluded that ectopic expression of ClpP^DEAD-Ty does not affect its processing or overall apicoplast morphology. Furthermore, it colocalizes with CPN60 without affecting its activity, indicating that the co-expression could be used to evaluate interaction between the two.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. CPN60 co-localizes, binds and stabilizes PfClpP.

A. A diagram depicting integration of a transgenic inactive PfClpP with a point mutation in the proteolytic active site (S264A) and a C-terminal Ty tag. CRISPR-mediated integration into the parasite’s genome under the endogenous hsp110 promoter enables constant and moderate expression. B. Western blot analysis of lysates from CPN60^V5-apt; ClpP^DEAD-Ty parasite line, probed with antibodies against V5 (red) and Ty (green) showing the co-expression of the two proteins. Bottom: Schematic representation of the PfClpP processing: Full length cytoplasmic PfClpP with the transit peptide (TP, III), apicoplast-localized fraction after TP removal (II) and mature PfClpP after proteolytic cleavage of the pro-domain (I). C. EC50 of aTC in CPN60^V5-apt; ClpP^DEAD-Ty parasite in 37^oC (blue) and following 24 hours incubation at 40^oC (green). Parasites were washed to remove aTC, and then incubated in serial aTC dilutions in a 96-well plate. Parasitemia was measured after 4 days using flow cytometry, showing comparable EC50s of 4.5 nM and 3.9 nM. Data are fit to a dose-response equation and are represented as mean ± SEM. One representative experiment out of three is shown. D. Immunofluorescence microscopy of CPN60^V5-apt; ClpP^DEAD-Ty in different developmental stages. Z stack images processed as Maximum Intensity show from left to right: merged fluorescentt channels and DIC contrast, anti-Ty (ClpP^DEAD-Ty, green), anti-V5 (CPN60, red), and DAPI (parasite nucleus, blue), along with merged fluorescent channels. The imaging was performed using a Nikon Spinning Disk confocal fluorescence microscope equipped with a 100x/1.4NA objective. Scale bar is 2.5 μM. Right: Cytofluorograms depicting the fluorescence of each pixel in green channel (x-axis) against the red channel (y-axis) of maximum projection confocal images on the left. Pearson’s coefficients (r) above 0.5 indicate significant co-localization between two fluorescent channels. E. Co-IP of CPN60^V5-apt and ClpP^DEAD-Ty. Parasites were isolated and sonicated, and extracts were incubated with either anti-V5 antibody-conjugated beads (for CPN60 pulldown, left) or anti-Ty antibody-conjugated beads (for PfClpP pulldown, right). Input and IP samples were loaded on SDS-page and blotted with anti-Ty and anti-V5 antibodies. F. CPN60^V5-apt; ClpP^DEAD-Ty parasites were incubated without aTC for 24 h, and lysates were extracted and subjected to Western blot analysis using antibodies against V5 (CPN60, red), Ty (ClpP^DEAD-Ty parasite, green) and aldolase (red). CPN60 knockdown results in the disappearance of the processed form of PfClpP (I) and accumulation of the inactive zymogen (II). Gel image is representative of three biological replicates. G. Densitometric analysis of PfClpP protein bands from Western blot in (F). Each PfClpP processed form (I, II, III) was calculated as a fraction of total PfClpP expression. Results indicate a significant reduction in PfClpP processing upon CPN60 knockdown, suggesting that the interaction with CPN60 is essential for PfClp complex activity.

https://doi.org/10.1371/journal.ppat.1013275.g005

CPN60 binds and stabilizes the Clp complex

To test for physical interactions between CPN60 and PfClpP, we performed immunoprecipitation using the double-mutant parasites. The pulldown of ClpP^DEAD-Ty resulted in the co-IP of CPN60, in the form of a weak band (Fig 5E, right panel). The reciprocal IP of CPN60 revealed that it interacts with the mature ClpP protease (I), as well as with the zymogen form (II) but not with the full length PfClpP fraction (III) (Fig 5E, left panel). Importantly, pulldown of CPN60^V5 was specific to PfClpP as it failed to co-IP unrelated highly-expressed proteins (S3B Fig), suggesting a specific interaction with ClpP however weak or transient. This indicates that CPN60 and PfClpP interact only upon co-localization to the apicoplast and that the chaperone binds the Clp complex both before and after protease activation. This might be a direct interaction, or indirect through a third unknown partner.

Typically, Clp proteolytic subunits oligomerize into heptameric rings, consisting of the core of the Clp complex [17]. Likewise, CPN60 orthologues are part of large heptameric or tetradecameric chaperonin complexes [23]. Therefore, the interaction observed here is unlikely to occur between individual monomers, rather at an interface between large complexes. To better understand this interaction, we utilized previously solved structures of CPN60 (PDB ID 7K3Z) [33] and PfClpP (PDB ID 2F6I) [22] and set to predict how these two large complexes might interact in their 3D conformations (S3C and S3D Figs). We used AlphaFold3 (AF3) [34], a state-of-the-art model with a diffusion-based architecture capable of handling large protein complexes, offering improved accuracy and speed compared to AlphaFold2 [35]. The AF3 predictions provided a model in which the two rings are stacked on each other, whereby the individual monomers tilt in opposite directions. Each monomer of the PfClpP heptameric ring makes contacts with two monomers in CPN60, and each CPN60 monomer contacts two monomers in PfClpP (S3E Fig). In this configuration it is possible for both CPN60 and ClpP complexes to form tetradecameric barrels, which interact via a stable interface.

We next examined the significance of this interaction, and tested possible effects on PfClpP activity. We previously showed that PfClpP accumulates in the apicoplast as an inactive zymogen, and upon assembly of the PfClpP oligomer, it is auto-cleaved to remove the inhibitory pro-domain and becomes activated [15]. Thus, the ratio between the inactive (band II) and active (band I) PfClpP forms is indicative of its activation state. To test the processing pattern of PfClpP upon CPN60 knockdown, we removed aTC from non-synchronous parasites and observed PfClpP processing 24 hours later, before parasites begin to die. We detected a reduction in the levels of PfClpP active band (I) and accumulation of the inactive zymogen (II), once CPN60 levels drop (Fig 5F). This shift in the ratio of active: inactive PfClpP could be quantified (Fig 5G), suggesting that the interaction between CPN60 and PfClpP is essential for PfClpP activation and complex stability.

Exploiting the apicoplast chaperonin system as an antimalarial target

Inhibition of protein synthesis in the apicoplast is a clinically approved drug target, as indicated by the antimalarial effect of doxycycline. Nevertheless, such treatment directly affects only a fraction of the total apicoplast proteome. Although largely unexplored, post-translational regulation in the organelle is no less critical and likewise can prove druggable. In this regard, the essentiality of CPN60 described here supports its viability as a potential antimalarial target.

To test this, we took advantage of a series of bacterial inhibitors designed to target the prokaryotic CPN60 orthologue, GroEL. As far as we are aware, chaperonin inhibitors have never been explored against Plasmodium. Thus, we conducted a pilot study to test the efficacy of a panel of phenylbenzoxazole (PBZ) analogues. PBZ compounds were previously shown to bind and inhibit the activity of GroEL in E. coli, inducing a separation and a twist of the two rings that decoupled their inter-ring allosteric signaling, thereby blocking the folding cycle [36]. Detailed cryoEM structures showed that the PBZ analogue PBZ1587 binds at the GroEL ring-ring interface [37]. Moreover, it was shown that mutations of key residues within the interface binding site render GroEL resistance both in vitro and in E. coli, demonstrating on-target antibacterial effects and a validated tool for other studies. We therefore hypothesized that PBZ analogues have the potential to bind the Plasmodium chaperonin systems, either the apicoplast CPN60, the mitochondrial HSP60, or both. Sequence alignments indicated that E. coli GroEL, human mtHSP60, and P. falciparum mitochondrial HSP60 shared high conservation of their overall sequences and the PBZ1587 ring-ring interface binding sites (S1 and S4 Figs), but apicoplast CPN60 was much less conserved. However, upon generating homology models of apicoplast CPN60 and mitochondrial HSP60 bound with PBZ1587 at the ring-ring interface as it does with GroEL, we were encouraged to see that, despite the differences in binding site residues, many key ligand-binding site interactions were predicted (S5 and S6 Figs), suggesting the inhibitors could still be effective in P. falciparum. Furthermore, the apicoplast CPN60 isoform contains an extended alpha-helix at the ring-ring interface near the PBZ1587 binding site, a feature we hoped might provide selective targeting as the site is more well defined compared with mitochondrial HSP60.

To test the antimalarial potential of the PBZ inhibitors, we incubated parasites with serial dilutions of each (50 μM to 50 nM) and monitored growth after 72 or 96 hours to calculate the inhibitor EC50 (Figs 6 and 7A). Although some inhibitors showed moderate activity against P. falciparum growth (1134, 1135, 1259), others produced an EC50 at a single-digit micromolar range (1001, 1064, 1133, 1207, 1221) or lower (1587), demonstrating a potent anti-Plasmodium activity (Figs 6 and 7A). We then incubated synchronized ring-stage parasites with a selected panel of the more potent inhibitors and followed their intraerythrocytic replication. During the first cycle, treated parasites developed normally like untreated control until the trophozoite stage, but then could not proceed to normal schizogony, and failed to complete their cycle (Fig 7B). The kinetics of this developmental failure was comparable to the one observed upon CPN60 knockdown (Fig 3G). We therefore tested whether inhibitors cytotoxicity is associated with impaired apicoplast development. We incubated non-synched CPN60^V5-apt parasites with PBZ inhibitors, and fixed the cells after 24 hours, before parasites begin to die. Strikingly, IFA demonstrated that PBZ treatments led to aberrant apicoplast morphologies, including fragmented, shrunk, or undeveloped organelles (Fig 7C). These results suggest at least one apicoplast-associated target, with CPN60 being a likely candidate.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. PBZ compounds used in this study and their EC50 values.

https://doi.org/10.1371/journal.ppat.1013275.g006

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Testing phenylbenzoxazole (PBZ) GroEL inhibitors for anti-Plasmodium activity.

A. Wildtype NF54 parasites were seeded at 0.5% parasitemia and incubated in serial dilutions of PBZ compounds, ranging from 50 μM to 50 nM. Parasitemia was measured after 72 or 96 hours to calculate the drugs’ half-maximal effective concentrations (EC50). Data are fit to a dose-response equation and are represented as mean ± SEM. B. CPN60^V5-apt parasites were synchronized and treated with 12.5 μM of different PBZ inhibitors. Up: Giemsa-stained blood smears were imaged using an upright Eclipse E200 Microscope. Bottom: Blood smears of CPN60^V5-apt parasites with or without PBZ treatment were analyzed for specific parasites developmental stages. PBZ-treated parasites demonstrate a delayed trophozoite stage already on the first replication cycle, fail to egress, and die during schizogony. C. Immunofluorescence microscopy of CPN60^V5-apt parasites following 24-hour incubation with 12.5 μM of different PBZ inhibitors. Z stack images processed as Maximum Intensity show from left to right: merged DIC contrast, anti-V5 antibody (CPN60, red), DAPI (parasite nucleus, blue), and merge of fluorescent channels. The imaging was performed using a Nikon Spinning Disk confocal fluorescence microscope equipped with a 100x/1.4NA objective. Scale bar is 2.5 μM.

https://doi.org/10.1371/journal.ppat.1013275.g007

We tried to test this directly, and performed EC50 assays for the PBZ compounds during CPN60 partial knockdown. This, however, did not lead to a significant EC50 shift for the compounds, hindering definitive conclusions (S7 Fig). We therefore assessed specificity towards broader apicoplast function, by coincubation of PBZ compounds with IPP. We incubated the parasites with a fixed inhibitor concentration of 12.5 μM (roughly 2-3X of most EC50s), and monitored their growth over 96 hours. While IPP did not affect growth of untreated parasites, it fully restores viability of chloramphenicol-treated parasites, which inhibit apicoplast translation (Fig 8A). However, the toxic effect of the PBZ inhibitors could not be restored by IPP, suggesting that (1) the 12.5 μM test concentration may have been too high above inhibitor EC50 values, impeding rescue by IPP supplementation (i.e., PBZ1587), or (2) inhibitors could be functioning against additional targets, the most likely candidate being mitochondrial CPN60. However, the moderately potent inhibitor PBZ1214 (EC50 = 9.6-12.5 μM) presented an intriguing result as IPP treatment sensitized parasites to the inhibitor (Fig 8B and 8C). The reason(s) for IPP sensitization of PBZ1214 is unclear at present and will require further investigation.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. Testing specificity of PBZ compounds against apicoplast activity.

A. Wildtype NF54 parasites were incubated with either DMSO (blue, control), 30μM Chloramphenicol (green, CHL) or 12.5μM of different PBZ compounds (shades of red), with (dashed) or without (solid) 200μM IPP. Parasitemia was monitored every 24 h over at least two cycles via flow cytometry. 100% of growth represents the highest value of calculated parasitemia (final parasitemia with DMSO). Normalized data are represented as mean ± SEM for three technical replicates. B. Wildtype NF54 parasites were incubated with DMSO (blue, control) or 12.5μM of PBZ1214 (red), with (dashed) or without (solid) 200μM IPP. Parasitemia was monitored every 24 h over two cycles via flow cytometry. Data are represented as mean ± SEM for three technical replicates. C. Wildtype NF54 parasites were seeded at 0.5% parasitemia and incubated in serial dilutions of PBZ1214 with (red) or without (blue) 200μM IPP. Parasitemia was measured after 72 hours to calculate the drugs’ half-maximal effective concentrations (EC50). Data are fit to a dose-response equation and are represented as mean ± SEM.

https://doi.org/10.1371/journal.ppat.1013275.g008

Plasmids construction

Genomic DNA was isolated from P. falciparum using NucleoSpin blood kit (Machery-Nagel). For the generation of the CPN60^V5-apt; conditional mutant, 500 bps from the C-term and 3’UTR from the of cpn60 gene (Pf3D7_1232100) were amplified from genomic DNA using primers P1 + P2 and P3 + P4, respectively. The two products were conjugated together using overlap extension PCR and inserted as one piece into pMG74-tetR-Dozi vector [27]. For expression of a C-term CPN60 guide RNA, oligos P8 + P9 were annealed and inserted into pUF1-Cas9-guide. Primer P7 was used in combination with primer P6 or P4 to test for accurate integration in the parasite genome following transfections. Evaluating aptamer repeats number was done using primers P5 + P6.

For the generation of ClpP^DEAD parasites, we transfected parasites with the puc-HSP110-ClpP^DEAD-Ty plasmid which we used previously to study ClpP processing and activation [15]. It encodes the entire Open Reading Frame of ClpP (PF3D7_0307400), mutated in three residues: Ser264Ala, Glu308Arg and Arg285Glu, following by a triple Ty tag. Expression is driven under the Pfhsp110c gene (PF3D7_0708800). Briefly, the HSP110 expression system comprises of a repair plasmid that contains homology sequences from the Pfhsp110c gene. The repair sequences include the last 429 bp (not including the stop codon) from the Pfhsp110c gene, followed by a 2A skip peptide, ClpP^DEAD-Ty sequence and the first 400 bps from Pfhsp110c 3’UTR. For expression of a Pfhsp110c guide RNA, oligos P10 + P11 are expressed from pUF1-Cas9-guide. Primers P12 and P13 were used to test for accurate integration in the parasite following transfections.

All constructs utilized in this study were confirmed by Sanger sequencing. PCR products were inserted into the respective plasmids using In-Fusion Snap assembly Master mix (Takara). All restriction enzymes used in this study were purchased from New England Biolabs. Oligonucleotides used in this study are summarized in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Primers and Oligos used in this study.

https://doi.org/10.1371/journal.ppat.1013275.t001

Cell culture and transfections

Parasites were cultured in RPMI medium supplemented with Albumax I (Gibco) and transfected as described earlier [59–61]. Synchronization of cultures were achieved by intermittent sorbitol and Percoll treatments. To generate CPN60^V5-apt parasites, a mix of two plasmids was transfected into NF54 wildtype parasites in the presence of 0.5 mM anhydrotetracycline (aTC, Cayman Chemicals). The plasmid mix contained 40 μg of pUF1-Cas9-guide expressing the CPN60 guide, and 20 μg of pMG74-based donor plasmid with CPN60 homology regions, linearized by EcoRV. Drug pressure was applied 48 hours post transfection using 2.5 μg/ml Blasticidin (BSD, Sigma). For each transfection, at least 2 clones (A4 and F11) were isolated via limiting dilutions and used in all subsequent experiments.

To generate CPN60^V5-apt; ClpP^DEAD-Ty parasites, a mix of two plasmids was transfected into CPN60^V5-apt parasites. The mix contained 2 plasmids; 40 μg of pUF1-Cas9-Pfhsp110c-guide and 40 μg of the marker-less repair plasmid puc57-HSP110-ClpP^DEAD. Drug pressure was applied 48 hours post transfection, using 1 μM DSM1 (BEI Resources), selecting for Cas9 expression. DSM1 was removed from the culturing media once parasites clones were isolated by limiting dilution. For each transfection, at least 2 clones (D12 and H10) were isolated via limiting dilutions and used in all subsequent experiments.

Growth assays

For CPN60 knockdown assays, asynchronous parasite cultures were washed 8 times and incubated without aTC. Throughout the course of the experiment, parasites were sub-cultured to maintain parasitemia between 1–5% and parasitemia was monitored every 24 hours via flow cytometry. Growth assay during PfClpS knockdown was performed following the same steps. The parasitemia levels were determined in comparison to negative control (uninfected RBCs). Cumulative parasitemia at each time point was back calculated based on actual parasitemia multiplied by the relevant dilution factors. Parasitemia in the presence of aTC at the end of each experiment was set as the highest relative parasitemia and was used to normalize parasites growth. Normalized data from technical replicates were analyzed and fit to an exponential growth curve using Prism (GraphPad Software, Inc.) All experiments shown are representative of at least three biological replicates. For IPP rescue, media was supplemented with 200 μM IPP (Isoprenoids LC). Controls for apicoplast ablation were done using 30 μM chloramphenicol. For apicoplast ablation during CPN60 knockdown, parasites were incubated without aTC for 17 days while being supplemented with IPP. After these 17 days, this culture was divided into two groups, one with aTC (to allow CPN60 re-expression and accumulation) and one without aTC. Both groups were supplemented with IPP. After additional 16 days, each group was further divided into two cultures, with or without IPP. The growth of the four treated groups were then measured by flow cytometry, as well as analysed by IFA and WB.

Flow cytometry

Aliquots of parasite cultures (5 μl) were stained with 8 μM Hoechst (ThermoFisher Scientific) in PBS. The fluorescence profiles of infected erythrocytes were measured and analyzed by flow cytometry on a CytoFlex S (Beckman Coulter). The parasitemia data were analyzed and presented using Prism 10.5.0 (GraphPad Software, Inc.).

Recovery after heat shock

To induce heat shock response, parasites were incubated at 40^oC for 24 hours and then allowed to recover at 37^oC to mimic the characteristic malarial fever episodes. Growth of recovered parasites was monitored over the next following days using flow cytometry and protein samples were isolated as described below. When CPN60 levels were to be manipulated, aTC was first washed away, and parasites were subjected to heat shock 24 hours later to allow protein levels to drop.

Half-maximal effective concentration (EC50)

To generate an EC50 curve for aTC, asynchronous cultures of CPN60^V5-apt or CPN60^V5-apt; ClpP^DEAD-Ty parasites were washed 8 times and then seeded at 0.5% parasitemia in a 96 well plate with varying concentrations of aTC. To measure aTC EC50 after recovery from heat shock, parasites were similarly treated, incubated for 24 hours at 40^oC, and then returned to 37^oC for the rest of the experiment. To generate EC50 curves for HSP60 inhibitors, NF54 wildtype parasites were seeded at 0.5% parasitemia in a 96 well plate with varying concentrations of each compound. In all cases parasitemia was measured after 72- and 96-hours using flow cytometry. Data were fit to a dose-response equation using Prism 10.5.0 (GraphPad Software, Inc.).

In-house anti-Ty tag monoclonal antibody purification

The hybridoma BB2 clone cell line (a kind gift from Boris Striepen) was adapted to serum-free medium. 500 ml spent medium of the hybridoma clone containing mAbs was collected, concentrated and purified by HiTrap Protein G HP antibody purification 1 ml column (Cytiva) operated by peristaltic pump, as per manufacturer protocol. The bound antibody was eluted using 10 ml 0.1 M Glycine-HCl buffer, pH 2.7 for ten minutes, and then neutralized with 1M Tris-HCl, pH 9.2 buffer. The protein content was determined and adjusted to 1 mg/ml.

Immunoprecipitation

Co-Immunoprecipitation (Co-IP) protocols were performed using anti-V5 antibody (CPN60^V5-apt) or anti-Ty antibody (ClpP^DEAD-Ty). Pellets from 4x10⁸ parasites were isolated using cold saponin and were lysed and sonicated in Extraction Buffer (40 mM Tris HCl pH 7.6, 75 mM KCl, and 1 mM EDTA, 15% glycerol) supplemented with HALT protease inhibitor (Thermo). 10% of the sample was kept for later analysis (input sample). Rabbit anti-V5 (Cell Signaling, D3H8Q) or anti-Ty, BB2 (inhouse purification) antibodies were crosslinked to Dynabeads protein G beads (Invitrogen) by incubating with 5 mM BS3 crosslinker (CovaChem) for 30 minutes. Quenching of crosslinker was performed using 1 M Tris HCl (pH 7.5) for 30 minutes and then washing with 1.2M Glycine HCl (pH 2.5) to remove excess unbound antibody. Antibody-conjugated beads were then washed 3 times with PBS and incubated with the supernatant at 4°C. Washes were performed using a magnetic rack (Life Technologies). Samples were run on SDS-page and blotted with anti-V5 or anti-Ty antibodies.

Western blot

Western blots were performed as described previously [15]. Briefly, parasites were collected and host red blood cells were permeabilized selectively by treatment with ice-cold 0.04% saponin in PBS for 15 min, followed by a wash in ice-cold PBS. Cells were lysed using RIPA buffer, sonicated, and cleared by centrifugation at 4oC. The antibodies used in this study were mouse anti-Ty, BB2 (in-house purification), rabbit anti-V5, D3H8Q (Cell Signaling, 1:3000), mouse anti-V5, E9H80 (Cell Signaling, 1:3000), rabbit anti-aldolase ab207494 (abcam, 1:3000). The secondary antibodies that were used are IRDye 680CW goat anti-rabbit IgG and IRDye 800CW goat anti-mouse IgG (LICOR Biosciences, 1:20,000). The Western blot images and quantifications were processed and analyzed using the Odyssey infrared imaging system software (LICOR Biosciences). Three biological replicates were used for quantification of protein expression levels using Prism 10.5.0 (GraphPad Software, Inc.).

Microscopy and image processing

For IFA, cells were fixed using a mix of 4% paraformaldehyde and 0.015% glutaraldehyde and permeabilized using 0.1% Triton-X100. Primary antibodies used are mouse anti-Ty1, BB2 (Invitrogen, 1:100), and rabbit anti-V5, D3H8Q (Cell Signaling, 1:100). Secondary antibodies used are Alexa Fluor 488 and Alexa Fluor 546 (Life Technologies, 1:100). For mitochondrial staining, live cells were incubated with 300 nM Mito-tracker Deep Red (Invitrogen) for 15 minutes, and then fixed and stained as described above. Cells were mounted on Fluoroshield with DAPI (Sigma). The imaging was performed using a Nikon Spinning Disk confocal fluorescence microscope equipped with a 100x/1.4NA objective. Images were collected as Z-stack, and displayed as maximum intensity projection. Image processing, analysis and display were performed using Zeiss ZEN 3.7 Software. Adjustments of brightness and contrast were made for display purposes. Blood smears were imaged using an upright Eclipse E200 Microscope (Nikon), equipped with x100 oil objective Type NVH, and captured using a digital Sight 1000 microscope camera and NIS-Elements Software (Nikon).

Ultrastructure expansion microscopy

Ultrastructure expansion microscopy was performed as previously described [62] with minor modifications. Briefly, parasite cells were adhered to poly-d-lysine coated coverslips, washed and fixated in PFA, and then incubated on gels, followed by transfer to denaturation buffer. Gels were then incubated at 95°C, and then expanded in doubled-distilled water. The gels were then measured for expansion factor (5X in this case). The gel is then blocked (3% BSA) and incubated with antibodies. Dyes and antibodies used in this study: rabbit anti-V5, D3H8Q (Cell Signaling, 1:100), Alexa Fluor 488 goat anti-rabbit (1:200, Cell Signaling), NHS-ester Alexa Fluor 405 (1:250, Thermo Fisher Scientific) and SYTOX Deep Red (1:1000, Thermo Fisher Scientific). The imaging was performed using Airyscan2 LSM980 confocal microscope equipped with 63x/1.4NA objective.

RNA extraction and quantitative real time PCR (qRT-PCR)

Total RNA was extracted using the Macherey-Nagel NucleoSpin RNA Prep Kit. RNA concentration and purity were assessed using a NanoDrop spectrophotometer. For reverse transcription, 10 µL of the total RNA (from a 40 µL eluate) were used to synthesize complementary DNA (cDNA) using the QuantaBio qScript cDNA Synthesis Kit.

Quantitative PCR was performed using the Luna Universal qPCR Master Mix (New England Biolabs) on a QuantStudio 5 Real-Time PCR System. Gene-specific primers were used to amplify CPN60 (P14 + P15), LRR5 (PF3D7_1432400, P20 + P21), and the two housekeeping genes aldolase (P16 + P17) and arg-tRNA synthetase (P18 + P19) for normalization. Each reaction was performed in triplicate. Relative expression levels were calculated using the ΔΔCt method, normalizing to the geometric mean of the housekeeping genes and comparing expression levels in heat-shocked samples to the non-treated control. Statistical analyses were performed using Prism 10.5.0. (GraphPad Software, Inc.)

AF3 structure prediction and Alanine scanning

The inputs are protein and peptide sequences. All different versions were run with random seed 2 or 1. Runs was done on alphafold server from GitHub (https://alphafoldserver.com/). Structural Visualization was done using the PyMOL Molecular Graphics System (Version 2.4.0) and UCSF ChimeraX (Version 1.4). Robetta computational alanine scanning was applied to define the critical interface residues, as implemented in http://robetta.org. Detailed protocol was described by Kortemme et al [59].

Generating the apicoplast and mitochondrial PfCPN60 models with PBZ1587 bound at the ring-ring interface

CryoEM structures were recently reported indicating PBZ1587 binds with 7-fold symmetry between adjacent subunits across the ring-ring interface of E. coli GroEL [37]. To support whether PBZ inhibitors would similarly bind to and inhibit the apicoplast and mitochondrial PfCPN60 isoforms, we generated homology models for the two PfCPN60 isoforms based on the GroEL-1587 cryoEM structure (9C0B) using Schodinger Maestro. We generated two models for each isoform, one with the apical domains in a conformation where the substrate binding sites are oriented inwards towards the central cavity (like in the GroEL-1587 9C0B structure), and the other with the apical domains in an extended conformation and oriented upwards from the central cavity (like in the apicoplast PfCPN60 structure), described as follows.

1. Apicoplast PfCPN60–1587 model with apical domains positioned inwards: We first employed the build homology model, multiple templates (homomultimer) function using default parameters and aligning the apicoplast PfCPN60 sequence onto the GroEL-1587 tetradecamer structure (9C0B). This provided a reasonable initial model for further refinements of the ligand binding sites. As the alpha-helices spanning residues 496–541 present in the apicoplast PfCPN60 structure were not represented as such in the homology model, we then spliced these secondary structures into the model. Next, we independently refined each of the 7 binding sites using the Refine Protein-Ligand function using default parameters, with the exception of refining atoms within 10 Å of the ligands. We then conducted Induced Fit Docking of the PBZ1587 inhibitor independently at each site using default parameters for Extended Sampling, with the exception of allowing refining of residues within 8.0 Å of ligand poses. The docked structure was then subjected to final refinement through the Protein Preparation Workflow using default parameters.
2. Mitochondrial PfCPN60–1587 model with apical domains positioned inwards: We first employed the build homology model, multiple templates (homomultimer) function using default parameters and aligning the mitochondrial PfCPN60 sequence onto the GroEL-1587 tetradecamer structure (9C0B). Next, we independently refined each of the 7 binding sites using the Refine Protein-Ligand function using default parameters, with the exception of refining atoms within 10 Å of the ligands. We then conducted Induced Fit Docking of the PBZ1587 inhibitor independently at each site using default parameters for Extended Sampling, with the exception of allowing refining of residues within 8.0 Å of ligand poses. The docked structure was then subjected to final refinement through the Protein Preparation Workflow using default parameters.
3. Apicoplast PfCPN60–1587 model with apical domains extended upwards: We first aligned the individual apicoplast PfCPN60 rings on those of the GroEL-1587 tetradecamer structure (9C0B) so as to retain the apical domains in the extended, upwards conformation as in the reported PfCPN60 structure. Next, we independently refined each of the 7 binding sites using the Refine Protein-Ligand function using default parameters, with the exception of refining atoms within 10 Å of the ligands. We then conducted Induced Fit Docking of the PBZ1587 inhibitor independently at each site using default parameters for Extended Sampling, with the exception of allowing refining of residues within 8.0 Å of ligand poses. The docked structure was then subjected to final refinement through the Protein Preparation Workflow using default parameters.
4. Mitochondrial PfCPN60–1587 model with apical domains extended upwards: We first employed the build homology model, multiple templates (homomultimer) function using default parameters and aligning the mitochondrial PfCPN60 sequence onto the apicoplast-1587 homology model generated above, with the apical domains extended upwards (model 3). Next, we independently refined each of the 7 binding sites using the Refine Protein-Ligand function using default parameters, with the exception of refining atoms within 10 Å of the ligands. We then conducted Induced Fit Docking of the PBZ1587 inhibitor independently at each site using default parameters for Extended Sampling, with the exception of allowing refining of residues within 8.0 Å of ligand poses. The docked structure was then subjected to final refinement through the Protein Preparation Workflow using default parameters.