Structure–function analysis of cysteine residues in the plasmodium falciparum chitinase, PfCHT1

Abstract

The Plasmodium ookinete uses chitinase activity to penetrate the acellular, chitin‐containing peritrophic matrix to invade the mosquito vector. Plasmodium ookinetes from different parasite clades secrete two structurally distinct forms of chitinase, one, a short form lacking a C‐terminal putative chitin‐binding domain (CBD), the other, a long form with both proenzyme and C‐terminal putative chitin‐binding domains. Here, we structurally and functionally characterize the three cysteines in the short chitinase of the human‐infecting malaria parasite, P. falciparum testing the hypothesis that one unpaired cysteine would not contribute to chitinase‐specific enzymatic activity which would identify this residue as potentially involved in intermolecular disulfide bonding and heteromultimeric invasion complex formation as previously described. To test this hypothesis, we produced and characterized recombinant wild‐type and cysteine‐mutation PfCHT1 proteins in E. coli and used biophysical and enzymatic approaches to examine their enzymatic activities and chitin‐binding affinities. The cysteine‐203 PfCHT1 mutation had no effect on chitinolytic and chitin‐binding functions. The cysteine‐220 and cysteine‐230 mutants were enzymatically inactive and did not bind to chitin. The artificial intelligence‐based protein prediction algorithm, AlphaFold, correctly identified the involvement of cys‐220 and cys‐230 in the intramolecular disulfide linkages key to maintaining properly folded chitinase structural integrity. AlphaFold predicted that cys‐203 cysteine is surface exposed and thus involved in intermolecular protein–protein interaction. Production of the cys‐to‐ser 203 PfCHT1 mutant facilitated recombinant protein production. Future cellular and biochemical studies are needed to further understand details of Plasmodium ookinete mosquito midgut invasion.

1. INTRODUCTION

Plasmodium ookinete‐secreted endochitinases (EC.3.2.1.14) belong to family 18 glycohydrolase, which function to hydrolyze β‐1,4‐linkages between the N‐acetylglucosamine units of the solid‐phase chitin polymer. ¹ , ² , ³ The Plasmodium chitinases along with aspartic proteases play an essential role in traversing the chitin‐containing acellular peritrophic matrix (PM) lining the mosquito midgut. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ Within the mosquito midgut, ingested sexual‐stage Plasmodium gametocytes begin their sporogonic development. Within minutes inside the midgut, the gametes emerge, haploid male and female forms merge into diploid zygotes and elongate into a crescent‐shaped motile ookinete, which escape the digestive proteolytic milieu by traversing the midgut wall, all within 12–20 hrs. ⁴ , ⁸

Plasmodium ookinete‐secreted chitinases have been classified into two classes, namely short forms (PfCHT1, PoCHT1, PgCHT2, and PrCHT1) and long forms (PgCHT1, PvCHT1, and PbCHT1). The classification is based on the presence or absence of the proenzyme domain at amino terminal and a cysteine‐rich putative chitin‐binding domain (CBD) at the carboxyl‐terminal of the protein, whereas the short form lacks proenzyme and putative chitin‐binding domains. ⁹ In escaping the mosquito midgut, the role played by the Plasmodium ookinete‐secreted chitinases varies among Plasmodium species. ¹⁰ However, both long and short chitinase forms bind to chitin have been well demonstrated irrespective of the structural differences of the Plasmodium chitinase protein. ¹¹ Structurally, Plasmodium species with long chitinases have four conserved cysteine residues in the chitin‐binding domain at the C‐terminal compared to the three conserved cysteine residues located in the catalytic domain of the short chitinases. ¹⁰ The avian‐infecting P. gallinaceum parasite, whose ookinetes secrete both long and short forms of chitinases, is an excellent model to study the role of cysteines in the Plasmodium transmission biology, ⁴ and more generally, a model for understanding the evolutionary biology of host‐pathogen interactions.

The putative role of conserved cysteines in the CBD of PgCHT1 (long chitinase) has been experimentally demonstrated, using a novel chitin pull‐down assay, to be involved in protein secondary formation that enables chitin substrate binding. ¹¹ E. coli‐expressed rPgCHT1 (long form) mutant proteins (cysteines to serine) were examined for their chitin‐binding affinities, which showed their importance in enabling high affinity, specific and detergent‐resistant interactions of PgCHT1with chitin. ¹¹ Similar chitin‐specific interactions of rPfCHT1, a short chitinase of the human‐infecting parasite Plasmodium falciparum, were observed with both E. coli lysate and purified rPfCHT1 protein. ¹¹ The PfCHT1, lacking proenzyme and chitin‐binding domains, is secreted as an active enzyme covalently linked to other ookinete‐secreted proteins and binds to solid‐phase chitin.

Recently, highly accurate high prediction score and descriptive 3D structures for Plasmodium chitinases have been established using the DeepMind AlphaFold algorithms. ¹⁰ , ¹² , ¹³ , ¹⁴ , ¹⁵ The close insets into the 3D structure of the short chitinase are suggestive of the role played by the cysteines in the intramolecular structural and functional stability. ¹⁰ These structure‐based observations suggest the possible role of the PfCHT1 cysteine residues in maintaining the structure and function; however, this further needs experimental investigation.

In the present study, we heterologously expressed purified and biophysically characterized recombinant WT and mutant PfCHT1 proteins. Additionally, we have predicted the structure of WT‐PfCHT1 protein and experimentally demonstrated the critical role of position −220, −230 cysteine residues in rPfCHT1 protein enzymatic and binding functions, and the lack of a role for the −203 in enzymatic function. This study highlights functional insights into the role played by the short chitinases during mosquito midgut invasion.

2. RESULTS

2.1. AlphaFold 3D‐PfCHT1 protein structure prediction

The 3D PfCHT1‐WT protein structure predicted using the DeepMind's AlphaFold algorithm confirmed the presence of the triosephosphate isomerase (TIM)‐barrel fold consisting of eight beta strands (β1‐8) tethered with the eight alpha helices (α1‐8) within the conserved catalytic domain ⁷ (Figure 1a,b). Further, the PfCHT1‐WT 3D structure shows the involvement of the two cysteines (PfCHT1: −220, −230) in the formation of the intramolecular disulfide linkages. These linkages are probably essential for maintaining the properly folded PfCHT1 protein structure. ¹⁰ On the other hand, the third residue (PfCHT1: −203) was found to be in an unbound state and surface exposed compared to −220 and −230 (Figure 1c). The −203 residues might have a role to play in the experimentally demonstrated heteromultimeric, reduction‐sensitive short chitinase‐containing protein complexes, which further needs experimental evidence. ¹¹ Also, the −203 residue might be independent of maintaining the protein's structure and functional stability.

FIGURE 1.

PfCHT1‐3D protein structure prediction using Deepmind's AlphaFold algorithm and pET‐32b (+) construct design for production of recombinant PfCHT1‐WT and mutant proteins (a) Schematic representation of the primary structure of PfCHT1. (b) Predicted PfCHT1 3D structure with signal peptide. The three cysteines of interest are highlighted in yellow. The predicted disulfide linkage between the cysteine at 220 and 230 is shown in panel (c), magnified view of the predicted disulfide linkage between cysteines located at 220 and 230 amino acids and the free unbound cysteine at 203 position. (d) Schematic representation of cloning vector construct design for the production of recombinant PfCHT1 proteins; (1) PfCHT1‐WT and all mutants (−Triple, −203, −220, and −230) in customized pET32b vector obtained from Gene Universal (Newark DE). (e) Confirmation of positive PfCHT1‐WT and PfCHT1‐mutant transformants by BamHI and XhoI double digestions. All of the positive transformants show the right insert size of 1,063 bp, which is the size of the gene of interest. M‐ 1.5 Kb Molecular Marker; TrxA: Thioredoxin; 6X His tag, a: Thrombin recognition and cleavage site; (b) S‐tag; (c) Enterokinase recognition and cleavage site; (d) Multiple cloning site

2.2. Recombinant PfCHT1‐WT and mutant (−triple, −203, −220, and −230) protein expression

The full‐length minus signal peptide rPfCHT1‐WT and four PfCHT1 mutants generated by replacement of the cysteine with serine (−203, −220, −230, and ‐Triple) were efficiently cloned in pET‐32b (+) vector and double restriction digestion with BamHI and XhoI confirms the right size insert (Figure 1d,e). The rPfCHT1‐WT and ‐mutant proteins were successfully expressed on induction with IPTG in E. coli SHuffle cells except for the −220 mutant at 16°C. The western immunoblot performed with the solubilized and filtered fractions further confirms the presence of the PfCHT1 protein band at 58 kDa when probed with the monoclonal anti‐PfCHT1 (1C3) antibody (Figure 2a,b). Both whole lysate and clarified supernatants of PfCHT1‐WT and −203 mutant protein fractions showed the detectable enzymatic (chitinase) activity measured by using the 4‐MU based microfluorometric assay (Figure 2c). However, PfCHT1‐203 supernatants showed comparable higher activities compared to the ‐WT protein. The results also demonstrated the complete inhibition of the chitinolytic activity of ‐Triple, −220, and −230 PfCHT1 mutant proteins compared to the −203 mutant where the activity was seen comparable to the WT protein.

FIGURE 2.

E. coli expressing PfCHT1 cell pellet solubilization and recombinant PfCHT1‐WT and −203 protein purification using affinity chromatography. The PfCHT1‐ pET‐32b (+) expression plasmid was transformed into SHuffle T7 Express Competent E. coli cells for WT and mutant protein expression. The E. coli cell pellets were solubilized using optimized buffer conditions. (a) Soluble and insoluble fractions were subjected to SDS‐PAGE, followed by (b) western immunoblot analysis with 1C3 antibody to detect rPfCHT1, at 58 kDa. Except PfCHT1‐220 protein, PfCHT1 was detected in all the protein preparations. (c) Comparison of chitin hydrolytic activity using fluorescent 4‐methyl‐umbelliferyl‐N, N′, N″‐β‐D triacetyl‐chitotrioside (4MU) substrate in a microflurometric assay using fetal bovine serum (FBS) as a positive control. Only PfCHT1‐WT and 203 lysates and supernatants were found to be enzymatically active. Solubilized rPfCHT1‐WT (d) and PfCHT1‐203 (e) fractions were then subjected to Ni‐NTA purification, and purification chromatograms showed the presence of eluted protein peaks and western immunoblot confirms the presence of PfCHT1‐WT and −203 in eluted protein peak fractions when probed with 1C3 antibody. (f) The chitinase hydrolytic activity assay using fluorescent ‐4MU substrate further confirmed the presence of active affinity‐purified eluates for both the proteins

2.3. Effect of triton X‐100 on rPfCHT1 protein solubilization and enzymatic activity

In the present study, we have used various buffer mixtures (Figure S1) for the solubilization of the protein expressed in the E. coli system, where most of them resulted in the complete inhibition of the WT protein's chitinolytic activity. However, the addition of the denaturing reagent urea resulted in regain of the chitinase activity suggestive of the presence of misfolded and aggregated rPfCHT1 protein upon solubilization (Figure S1). Further steps involved in the optimization of PfCHT1 protein solubilization included the addition of 1% Triton X‐100 with 20 mM NaCl which resulted in the solubilization of PfCHT1 and allowed to retain the enzymatic activity of the protein. The use of Triton X‐100 enabled the complete solubilization of the E. coli secreted PfCHT1‐WT and ‐mutant proteins.

2.4. PfCHT1‐203 cysteine is highly reactive and leads to the formation of oligomers in solution in WT‐expressed protein

The rPfCHT1‐WT and all mutant except −220 mutant proteins containing the 6xHis tag were expressed in the Shuffle E. coli cells and were further subjected for purification using Ni‐NTA column chromatography using the Akta FPLC system. The PfCHT1‐WT and −203 mutant proteins demonstrated the purified protein elution peak and corresponding bands of ~58 kDa on the western blot when probed with monoclonal anti‐PfCHT1 (1C3) Ab (Figure 2d,e). Both purified eluates (E4) of the WT and −203‐mutant protein were enzymatically active, as measured by using the 4MU substrate (Figure 2f). Fetal bovine serum (FBS) was used as positive control, as it is known to have demonstrated chitinase activity. ¹¹ The purified rPfCHT1‐220, −230, and ‐Triple mutant did not show any measurable chitinase enzymatic activity. However, the elution peaks in the affinity purification chromatograms were observed for ‐Triple, and −230 mutant proteins (Figure S2).

Further, the affinity‐purified eluates of rPfCHT1‐WT and −203 from E4 peak fraction were subjected for size exclusion chromatography using Superdex‐75 Increase column to obtain monomeric fraction of the affinity‐purified proteins (Figure 3a,b). A small high‐molecular aggregate peak fraction at 8.11 ml (E2) was seen in rPfCHT1‐WT SEC elution profile and was enzymatically inactive. The second monomeric peak fraction of rPfCHT1‐WT at 9.91 ml (E3) showed the chitinolytic activity using the 4MU substrate (Figure 3a,c). However, both the peak fractions E2 and E3 showed reactivity with the 1C3 Ab on the western immunoblot. The rPfCHT1‐203 mutant protein SEC purification elution profile demonstrated the presence of only monomeric protein peak at 9.65 ml (E3), which was both enzymatically active and showed reactivity with the monoclonal anti‐PfCHT1 Ab (Figure 3b,c). Both monomeric protein peak fractions (E3) of rPfCHT1‐WT (9.9 ml) and −203 (9.6 ml) were analyzed for homogeneity and molecular weight determination using SEC‐MALS. The Light Scattering (LS), UV Absorbance at 280 nm (UV) and Refractive Index (RI), and calculated Molar mass (g/mol) were plotted as a function of elution time. The rPfCHT1‐WT in comparison to rPfCHT1‐203 showed the presence of a high‐molecular aggregate peak in addition to the single homogenous peak eluted at 69 kDa (Figure 3d). On the other hand, rPfCHT1‐203 protein eluted with a single homogenous peak calculated by LS to be 55 kDa (Figure 3e), without the presence of higher molecular aggregates as observed in WT. These results emphasize the role played by the 203‐cysteine residue in PfCHT1 protein in being reactive and possessing the tendency to form the disulfide bridges when in solution. This residue when replaced with serine in −203 mutant resisted the formation of higher molecular weight oligomers, as evident from −203 protein. Also, the increased chitinolytic potential of affinity‐purified PfCHT1‐203 protein is suggestive of the presence of the functionally folded, monomeric and homogenous protein compared to ‐WT where the presence of aggregates leads to decline in the enzymatic potential of the protein (Figure 2c).

FIGURE 3.

Purification and characterization of PfCHT1‐WT and PfCHT1‐203 proteins using size exclusion chromatography and multi‐angle light scattering (MALS). (a) Affinity‐purified rPfCHT1‐WT protein fraction was loaded onto a gel filtration column (Superdex Increase 75 10/300; Cytiva) for molecular size‐based purification and western immunoblot analysis were performed with 1C3 antibody to detect rPfCHT1. The rPfCHT1‐WT SEC chromatogram shows the presence of an aggregate peak at 8.1 ml elution volume, followed by a monomeric elution volume peak at 9.9 ml. (b) Affinity‐purified rPfCHT1‐203 protein fraction was purified similar to ‐WT protein and western immunoblot performed with 1C3 antibody further confirms the rPfCHT1 presence. The rPfCHT1‐203 SEC chromatogram shows the presence of only monomeric elution volume peak at 9.6 ml. (c) The chitin hydrolytic activity estimated for the SEC‐purified eluates was comparable for both the proteins. (d, e) The SEC‐purified rPfCHT1‐WT and −203 protein fraction from the 9.9‐ and 9.6‐ml elution volume peak were further subjected to SEC‐MALS for homogeneity and molecular weight determination. Light Scattering (LS), UV Absorbance at 280 nm (UV) and Refractive Index (RI), and calculated Molar mass (g/mol) are plotted as a function of elution time. PfCHT1‐WT (d) elutes with a single homogenous peak calculated by LS to be 69 kDa with a minor aggregate peak, PfCHT1‐203 (e) also elutes with a single homogenous peak calculated by LS to be 55 kDa, without the presence of higher molecular aggregates observed in WT

2.5. Mapping rPfCHT1 cysteine residues that interact with solid substrate chitin

Earlier studies conducted on recombinantly produced PgCHT1 have well demonstrated the functional importance of the cysteine residues in the chitin binding. ¹¹ Here, we have demonstrated the functional importance of the three conserved cysteine residues in the recombinantly produced PfCHT1. To assess the chitin‐binding affinity of the WT and mutant rPfCHT1 proteins, they were pulled down from the total E. coli lysate expressing rPfCHT1. Interestingly, the binding affinity of the ‐WT and −203 rPfCHT1 proteins toward solid substrate chitin was found to be highly specific and therefore observed as a single‐enriched band on silver‐stained SDS‐PAGE gels (Figure 4a). The rPfCHT1‐WT and −203 mutant proteins showed high affinity binding for the chitin beads and resisted high concentration detergent washes. Further, the western blots with 1C3 antibody confirm the presence of the rPfCHT1 band at 58 kDa in both WT and −203 mutant protein pull downs (Figure 4a). However, western blots with ‐Triple mutant pull down have shown a very less binding toward chitin beads compared to WT and −203 mutant. As expected, the −230 mutants remain incapable of binding to chitin substrate and therefore, rPfCHT1 is not detected in their corresponding western blots. The fold change was calculated for the chitin‐binding ability of the ‐WT, −Triple, and −203 mutant proteins, which revealed that the WT still most effectively binds to chitin compared to ‐Triple mutant protein where the minimal binding was observed (Figure 4c). Results showed 1.5‐2‐fold increase in the chitin binding with −203 and ‐WT proteins compared to ‐Triple mutant protein.

FIGURE 4.

Estimation of the chitin‐binding affinity of E. coli expressed WT and mutant PfCHT1 protein. (a) For estimation of chitin‐binding affinity, 200 μl of washed chitin beads was added to 50 μg of WT and mutant proteins, which were then washed with the detergent containing Phosphate‐buffered saline (PBST; 1% Tween‐20). Samples of solubilized protein (input) and washed chitin beads after protein incubation were run under reducing conditions on 4–12% bis‐tris SDS‐PAGE gels and western immunoblots were performed using 1C3. PfCHT1‐WT, and −203 proteins showed positive enrichments in pull downs compared to −220 and −230 proteins with minimal binding seen with Triple mutants. (b) The samples were processed similar to chitin bead pull‐down assays, where 200 μl of washed nickel sepharose beads were added to 50 μg of WT and mutant proteins. Western immunoblots confirmed the positive pull downs for ‐WT, −Triple, −203, and −230 proteins compared to −220 protein. (c) The fold change for rPfCHT1‐WT and mutant proteins‐chitin binding was estimated using Image J software. Results showed 1.5–2 fold increase in the chitin binding with −203 and ‐WT proteins compared to ‐Triple mutant protein (Experiments were performed in triplicates)

The control bead (Nickel) binding assay has shown the specific Ni‐bindings and enrichment of rPfCHT1 in E. coli lysates of WT and as well as of the ‐Triple, −203 and −230 mutant proteins (Figure 4b). The −220 mutant does not show any binding with the Ni sepharose beads, whereas −230 mutant shows a positive binding with Ni sepharose beads. The western blots with 1C3 antibody further confirm the presence of the rPfCHT1 at 58 kDa position in ‐WT, −Triple, −203, and −230 mutant Ni pull downs (Figure 4b).

Due to the lack of expression of −220 mutant protein in the Shuffle cells, no binding was observed either with control Ni or chitin solid substrate. Further to understand the functional significance of the −220‐cysteine residue, the −220 mutant PfCHT1 protein was expressed in the Arctic cells where E. coli origin chaperonins facilitated the proper folding of the expressed protein. Arctic cell‐produced −220 mutant protein was solubilized similar to the WT protein and was found to be enzymatically inactive and further failed to show any binding affinity toward chitin. However, this protein showed the positive binding toward the nickel Sepharose control bead (Figure S3). This further confirms the importance of the −220 and −230 cysteine residue in chitin binding, where the presence of serine completely diminishes the chitin binding by abolishing the formation of intramolecular disulfide linkages, thus making protein nonfunctional.

2.6. Effect of temperature and reducing agents on the PfCHT1 stability and chitinolytic function

Effect of temperature on the enzymatic activity of PfCHT1 WT and −203 mutant proteins was investigated at temperature range of 4–55°C. Both PfCHT1‐WT and −203 protein demonstrated optimum enzymatic activity at 30°C with complete inhibition at 55°C (Figure 5a). Further, the effect of reducing agents on chitinolytic activity was assessed to demonstrate the dependency of PfCHT1 protein structure and functional stability on the −220 and −230 mediated intramolecular disulfide linkages. The purified PfCHT1‐WT and −203 proteins were incubated with increasing concentrations of two reducing agents, 2‐mercaptoethanol (β‐ME) and dithiothreitol (DTT). With β‐ME, both purified proteins showed decrease in the chitinolytic activity in concentration‐dependent manner with a substantial decrease at 0.625 mM and complete inhibition of the activity at 2.5 mM (Figure 5c). These results further confirm that the intramolecular disulfide linkages between −220 and −230 cysteines maintain the functional stability of the PfCHT1 protein, as −203 mutant protein had no effect on the functions of the protein. On the other hand, DTT had no significant effect on the chitinolytic activity of both purified WT and −203 mutant protein (Figure 5b). The ineffectiveness of DTT could be due to its incapability to reduce the buried intramolecular disulfide bonds (−220, −230) that are not solvent accessible.

FIGURE 5.

Analysis of intramolecular protein stability under varying thermal and reducing conditions. Chitinase hydrolytic activity was measured using the fluorescent 4‐MU substrate in a microfluorometric assay. (a) Hydrolytic activity at increasing temperatures was used to understand protein stability of active PfCHT1 protein and the PfCHT1‐203. Both ‐WT and −203 are inactive at 55°C. (b) The same activity assay was repeated with increasing concentrations of DTT. Both the ‐WT and −203 remained active through 5 mM concentration, no concentration‐dependent variations were observed (c) Varying conditions of β‐ME were also used to understand protein stability. At 2.5 mM β‐ME, both WT and mutant become inactive, and unable to hydrolyze the solid substrate

3. DISCUSSION

Here, we report the first study demonstrating the essential role of the three conserved cysteine residues of short chitinase, PfCHT1 in maintaining the protein structural stability, chitinolytic, and solid substrate‐binding functions. We experimentally demonstrated that specifically PfCHT1‐220 and −230 cysteine residues play an essential role in maintaining the properly folded protein structure and function through intramolecular disulfide linkages, whereas the PfCHT1‐203 cysteine residue is independent of providing any functional element to the chitinase protein and remained surface exposed in an unbound‐free state for the possible formation of disulfide linkages.

Ortholog of the P. gallinaceum ookinete‐secreted short chitinase (PgCHT2) has been found in human‐infecting species P. ovale (PoCHT1), P. falciparum (PfCHT1), and related species found in the Laverania subgenus. ⁴ Earlier studies have reported the functional significance of these chitinases, and other proteases in the Plasmodium's sporogonic lifecycle inside the mosquito midgut and in further continuing the malaria transmission cycle. ¹ , ² , ⁵ , ⁷ , ⁹ , ¹⁶ Recent studies with recombinantly expressed mutated PgCHT1 (long form of chitinase) protein resulted in diminished and high percent detergent wash‐sensitive binding affinities toward solid substrate chitin compared to wild‐type rPgCHT1 protein. This study confirms the significance of chitin‐binding domain (CBD) in the high affinity chitin binding in long chitinases. ¹¹ However, despite the lack of the CBD in the short chitinases (PfCHT1), the recombinantly expressed rPfCHT1 has shown the highly specific and detergent‐resistant binding affinities toward chitin like rPgCHT1 wild‐type protein. ¹¹ This confirms the strong binding affinities of the Plasmodium ookinete secreted chitinases toward the solid substrate despite their strikingly different structural organization.

However, the difficulties associated with the genetic manipulating the parasites for the biochemical and functional studies of P. falciparum have been well known. ¹⁷ , ¹⁸ Specifically, culturing the P. falciparum ookinetes in an in vitro system in presence of serum limits further biochemical studies involving chitinases. ¹⁹ , ²⁰ , ²¹ Therefore, in the present study, the recombinant WT and mutant PfCHT1 proteins were heterologously expressed in an E. coli system. To examine the solid substrate‐binding functions, three conserved cysteine residues of PfCHT1 were mutated to serine (all together: Triple, and one at a time: −203, −220, −230) and proteins were recombinantly expressed in an E. coli (sHuffle and arctic cells) system. As a result of cysteine mutation, the rPfCHT1‐220 (expressed in Arctic cells), and −230 mutant proteins were unable to bind to the solid substrate chitin in a chitin pull‐down assay compared to nickel pull downs (control pull downs) where positive enrichment of the mutant proteins were seen. The expressed proteins were also incapable of enzymatically degrading the fluorogenic substrate in a chitinase activity assay compared to the rPfCHT1‐WT protein. This is suggestive of the essentiality of these PfCHT1 cysteine residues in maintaining their chitin‐binding functions by stabilizing the expressed protein structure through intramolecular disulfide linkages. These −220 and −230 intramolecular disulfide linkages mediated protein stability, which have been further demonstrated by estimating the chitinolytic potential of WT and −203 proteins in presence of the reducing agents, where both proteins lose the function due to disruption of disulfide linkages. With the critical role played by these cysteine residues, previous studies conducted on the transgenic PfCHT1 with truncated C‐terminal (13 aa upstream of stop codon) demonstrated the essential role of the full‐length protein in the Anopheles freeborni mosquito midgut invasion. ²² This is further suggestive of the regulatory nature of the primary structure of the PfCHT1's C‐terminal, thus mediating multi‐functional roles, whether in hydrolyzing chitin or in the secretory mechanism of the protein. ⁹

Further, the short chitinases (PfCHT1 and PgCHT2) have been recently demonstrated to be secreted as a component of the multimeric, high molecular weight complex comprised of von Willebrand factor A domain‐related protein (WARP) and secreted ookinete adhesive protein (SOAP) proteins. ¹¹ , ²³ , ²⁴ This heteromultimeric HMW complex organization has been postulated to be dependent on the intermolecular disulfide linkages, as evident from the reduction‐sensitive nature of the secreted complex. The AlphaFold structural predictions of PfCHT1 have shown the localization of the −203‐cysteine residue to be surface exposed and in an unbound state compared to the −220 and −230 cysteine residues. Here, functional studies conducted with −203 mutant PfCHT1 protein showed the chitin‐binding function to be entirely independent of the −203‐cysteine residue with no effect on chitin‐binding affinities and was also enzymatically active. The main challenge during the recombinant expression of the Plasmodial origin protein in the E. coli expression system remains with the production of the soluble and functional proteins. ¹⁸ , ²⁵ Here, in the present study, the heterologous expression of −203 mutant protein resulted in homogenous preparations of the protein compared to ‐WT PfCHT1 protein where aggregates constituted a major fraction of the purified protein. This finding is further suggestive of the reactive nature of the −203‐cysteine residue in PfCHT1 which have mediated the intramolecular cross linkages of the monomeric proteins in the absence of other binding partner proteins of the HMW protein complex. ¹¹ This −203‐cysteine residue in PfCHT1 and other short chitinases (PgCHT2) inside the mosquito midgut might be involved in interacting with other micronemal proteins (SOAP and WARP) through the intermolecular linkages at the same time in an HMW complex which needs further experimental demonstration.

4. CONCLUSION

Here, in the present study, we showed the critical functional role played by the −220, and −230 cysteine residues in the short chitinase, rPfCHT1. This chitin‐binding and enzymatic function is entirely independent of the −203‐cysteine residue which might play an essential role in mediating the biogenesis of the secreted HMW complex by mature ookinetes during midgut invasion. Therefore, the −220 and −230 cysteine residues in association with other proteins in an HMW complex bind to chitin with high affinity and thus mediate invasion of the mosquito midgut through enzymatic degradation of the PM layer. However, studying these protein–protein interactions at the molecular and biochemical level will further insight into the Plasmodium ookinete's mechanism of invasion. These studies will provide an important tool to develop novel interventions to block early mosquito midgut invasion by the matured Plasmodium ookinete.

5. MATERIALS AND METHODS

5.1. 3D protein structure prediction using AlphaFold

The primary protein sequence was used to predict the 3D coordinates of the heavy atoms using the AlphaFold novel neural networks. ¹² The predicted PfCHT1‐WT protein structure was visualized using the molecular modeling software UCSF Chimera. ²⁶

5.2. Cloning and expression of PfCHT1‐WT and ‐mutant (−triple, −203, −220, and −230) recombinant proteins

The PfCHT1 nucleotide sequence (PlasmoDB gene ID: PF3D7_1252200)) was obtained using the PlasmoDB database (https://plasmodb.org/plasmo/app/record/gene/PF3D7_1252200#MitoprotForm). The E. coli codon‐optimized sequence encoding the amino acid 29–378 (full‐length protein minus the signal peptide) was cloned in pET‐32b (+) vector (Gene Universal) having a N‐terminal Thioredoxin (Trx A), N‐ and C‐Polyhistidine (His) and N‐terminal S tag. The pET‐32b (+) plasmid constructs for all the four mutant PfCHT1 proteins had cysteine replaced with serine. The triple mutant had all three conserved cysteines (−203, −220, and −230) mutated to serine, whereas other single mutants (−203, −220, and −230) had one cysteine mutated to serine at a time. The recombinant plasmids were then transformed into E. coli DH10B cells (Invitrogen) by electroporation as per the manufacturer's instructions (Bio‐Rad). The PfCHT1‐WT and ‐mutant pET‐32b (+) plasmids were further isolated and double digested with BamHI and XhoI using Monarch Plasmid Miniprep kit (New England Biolabs).

For protein expression, PfCHT1‐WT and ‐mutant pET‐32b (+) constructs were transformed in SHuffle T7 Express Competent E. coli cells (New England Biolabs) certifying the proper intramolecular disulfide bond formation and protein folding. In addition to these, the PfCHT1‐220 pET‐32b (+) mutant plasmid was transformed and expressed in ArcticExpress DE3 Competent E. coli cells (Agilent Technologies). Primary cultures were initiated using single E. coli colony for each protein in Luria broth (LB) media with Ampicillin (100 μg/ml) and grown at 37°C overnight in an incubator shaker. The secondary PfCHT1 cultures were further initiated and grown at 37°C and induced with 1 mM isopropyl‐β‐D‐thiogalactopyranoside (IPTG) (Sigma‐Aldrich). The cultures were then further grown overnight at colder temperatures (16°C for Shuffle T7 Express and 11°C for ArcticExpress DE3 E. coli cells) for protein expression.

5.3. Optimization of solubilization conditions for E. coli cells expressing PfCHT1‐WT and ‐mutant protein

The conditions for solubilizing E. coli cells expressing rPfCHT1 WT protein were optimized using an array of mild and commercially available buffers (Figure S1). The cell pellets of both WT and mutant protein were resuspended in 30 ml of optimized lysis buffer (20 mM Tris–HCl, pH 7.4, 20 mM NaCl) containing 1% Triton‐X 100 and PMSF (1 mM). The cells were then incubated at RT for ~30 min with continuous shaking. The lysates were centrifuged at 18,000g for 10 min at 4°C and further passed through 0.22 μm filters (50 ml Steriflip, EMD Millipore, MT) to obtain the solubilized clear protein fraction. The protein fractions for rPfCHT1‐WT and mutants were further analyzed on the 4–12% bis‐tris sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE) gels. Western immunoblotting was performed using the monoclonal 1C3 (Anti‐PfCHT1; mAb; 2.5 μg/ml) to confirm the presence of the protein in the solubilized protein fractions.

5.4. Affinity purification of rPfCHT1 and chitinase activity analysis

The His‐tagged rPfCHT1 (WT, −Triple, −203, −220, and −230) proteins were purified using the Akta Pure chromatography system (Cytiva). The FPLC system was operated at a flow rate of 0.8 ml/min, and the column was equilibrated using binding buffer (100 mM NaH2PO4, 10 mM Tris HCl, 25 mM Imidazole, and 0.01% Triton‐X‐100, pH 8.0). After equilibration, the clear supernatants of the E. coli cell lysates were passed through the prepacked Ni‐Sepharose HisTrap High Performance 1 ml columns (Cytiva) connected to the Akta system. Purified proteins were then eluted under a constant 500 mM Imidazole (100 mM NaH2PO4, 10 mM Tris HCl, 500 mM Imidazole+0.01% Triton‐X‐100, pH 8.0) concentration in an elution buffer. The purified protein fractions of PfCHT1‐WT and PfCHT1‐mutants were checked on 4–12% SDS‐PAGE to identify the rPfCHT1 positive fractions. Chitinase activity of the E. coli lysate, supernatants, and purified rPfCHT1‐WT and rPfCHT1‐mutant proteins were determined using a microfluorometric assay by adding chitinase substrate (4‐methylumbelliferyl‐N, N′, N″‐β‐D‐triacetylchitotrioside; Sigma). Normalized 30 μg of protein for each of the lysate and supernatant samples and 10 μg of purified protein was used for assessing the activity. The linear increase in the enzymatic activity was determined by kinetic fluorescence detection (Gemini EM, Molecular Devices LLC; excitation, 365 nm; emission, 450 nm) as described. ⁴

5.5. Biophysical characterization of rPfCHT1‐WT and rPfCHT1‐203 using size exclusion chromatography‐multiangle light‐scattering (SEC‐MALS)

Size exclusion chromatography was performed to further characterize the enzymatically active PfCHT1‐WT and −203 mutant proteins on the basis of their molecular size.

The affinity‐purified eluates were then passed through a prepacked Superdex 75 Increase 10/300 GL column (Cytiva) at a linear flow rate of 0.8 ml/min in buffer containing 20 mM Tris/HCl (pH 7.4) and 20 mM NaCl +0.01% Triton‐X‐100. The purified protein fractions were analyzed on SDS‐PAGE, and western immunoblots were performed to confirm the PfCHT1‐WT and −203 using 1C3 antibody. Size exclusion chromatography coupled with in‐line multi‐angle light scattering (SEC‐MALS) studies of purified and enzymatically active PfCHT1‐WT and −203 samples was performed on a HELEOS II multi‐angle light scattering detector (Wyatt Technology, Santa Barbara) with an Optilab T‐rEX refractometer (Wyatt Technology). MALS was measured following in‐line separation with a size exclusion chromatography analytical column (Sepax SRT SEC‐300, Sepax Technologies). Prior to injection, PfCHT1 samples were concentrated to ~1 mg/ml and 100 μl was injected per experiment, the mobile phase was 20 mM Tris pH 7.4, 20 mM NaCl and 0.01% TritonX‐100 with a flow rate of 0.4 ml/min. An Agilent 1,260 HPLC system (Agilent Technologies) was used for the mobile phase flow, sample injection, and measurement of UV absorbance at 280 nm. Data collection and analysis was done using ASTRA 6.1.7.17 software (Wyatt Technology).

5.6. Chitin‐binding affinity analysis of PfCHT1‐WT and mutant protein using chitin pull‐down assay

The nickel binding was analyzed using immobilized Ni Sepharose beads (HisTrap, GE Lifesciences), as a positive control for the binding assays. The E. coli lysates of WT and mutant proteins were centrifuged, and supernatants were filtered through 0.22 μm filters. Ni was washed three times with PBS (pH = 7.4) and made into a 50% slurry with PBS. Fifty microliters of Ni Sepharose bead slurry was added to 1 ml of filtered supernatants and incubated at RT for 30 min on a slow rocker. After incubation, protein‐bound Ni Sepharose beads were washed three times with PBS or PBST (0.05% Tween 20). PBST‐washed beads were washed one more time with PBS and SDS‐sample buffer added to the beads. SDS‐PAGE analysis and western immunoblots were performed according to standard procedure. Coomassie stain was used for gel visualization.

For the chitin‐binding analysis, Chitin beads (New England Biolabs, MA) were used. The E. coli lysates of rPfCHT1‐WT and all the mutants were centrifuged, and supernatants were filtered through 0.22 μm filters. Briefly, chitin beads were washed three times with PBS (pH 7.4) and made into a 50% slurry with PBS and 200 μl of chitin beads was added to 50 μg of filtered supernatants and incubated at RT for 30 min on a slow rocker. The protein‐bound chitin beads were washed three times with PBS or PBST (0.05% Tween 20) and SDS‐sample buffer was added directly to the beads and subjected to SDS‐PAGE analysis and silver stain was used for gel visualization. For western immunoblots, all the primary and secondary antibody incubations were done with 5% nonfat dry milk in PBS containing 0.1% Tween 20 (PBST) overnight at 4°C. For the rPfCHT1‐WT and mutants, the membranes were incubated with 1C3 (Anti‐PfCHT1; mAb; 2.5 μg/ml) as the primary and alkaline phosphatase‐conjugated goat anti‐mouse IgG (H + L) antibody (1:2500) (KPL) was used as secondary antibody to detect rPfCHT1. The reactions were carried under the similar conditions for rPfCHT1‐WT and mutant proteins. The results of western blots were further evaluated to quantitate the protein binding to the chitin and Ni beads using Image J software. ²⁷

5.7. Determination of the PfCHT1‐WT and PfCHT1‐203 mutant protein stability

The thermostability of the PfCHT1‐WT and −203 proteins were assessed with a temperature gradient from 4 to 55°C. Normalized protein concentration of 10 μg was used for analyzing the ‐WT and −203 chitinase activity (as mentioned above) in a temperature gradient. In order to determine the effect of the reducing agents on the protein folding, stability, and functional efficiency of the PfCHT1‐WT and −203 mutant proteins, two commonly used reducing agents β‐mercaptoethanol (BME) and dithiothreitol (DTT) were used. A doubling dilution of β‐ME (0.6‐10 mM) and DTT (0.6‐5 mM) were prepared to analyze the stability of the normalized 5 μg of protein concentrations for both proteins. Both reducing agents were incubated individually in the protein buffer (20 mM tris buffer) cocktail for 10 min. After incubation, the linear increase in the enzymatic activity was determined by using a microfluorometric assay as mentioned above.

AUTHOR CONTRIBUTIONS

Hargobinder Kaur: Conceptualization (lead); formal analysis (lead); investigation (lead); methodology (lead); visualization (lead); writing – original draft (lead); writing – review and editing (lead). Laine Garber: Conceptualization (supporting); investigation (supporting); methodology (supporting); writing – original draft (supporting); writing – review and editing (supporting). James W. Murphy: Conceptualization (equal); formal analysis (supporting); investigation (supporting); methodology (supporting); writing – original draft (supporting); writing – review and editing (supporting). Joseph Vinetz: Conceptualization (equal); formal analysis (equal); funding acquisition (equal); investigation (equal); methodology (equal); project administration (equal); resources (equal); validation (equal); visualization (equal); writing – original draft (equal); writing – review and editing (equal).

Supporting information

Figure S1

Figure S2

Figure S3

Kaur H, Garber L, Murphy JW, Vinetz JM. Structure–function analysis of cysteine residues in the plasmodium falciparum chitinase, PfCHT1 . Protein Science. 2022;31(5):e4289. 10.1002/pro.4289