Phosphoethanolamine-N-methyltransferase is a potential biomarker for the diagnosis of P. knowlesi and P. falciparum malaria

Abstract

Background: Plasmodium knowlesi is recognised as the main cause of human malaria in Southeast Asia. The disease is often misdiagnosed as P. falciparum or P. malariae infections by microscopy, and the disease is difficult to eliminate due to its presence in both humans and monkeys. P. knowlesi infections can rapidly cause severe disease and require prompt diagnosis and treatment. No protein biomarker exists for the rapid diagnostic test (RDT) detection of P. knowlesi infections. Plasmodium knowlesi infections can be diagnosed by PCR.

Methods and principal findings: Phosphoethanolamine-N-methyltransferase (PMT) is involved in malaria lipid biosynthesis and is not found in the human host. The P. falciparum, P. vivax and P. knowlesi PMT proteins were recombinantly expressed in BL21(DE3) Escherichia coli host cells, affinity purified and used to raise antibodies in chickens. Antibodies against each recombinant PMT protein all detected all three recombinant proteins and the native 29 kDa P. falciparum PMT protein on western blots and in ELISA. Antibodies against a PMT epitope (PLENNQYTDEGVKC) common to all three PMT orthologues detected all three proteins. Antibodies against unique peptides from each orthologue of PMT, PfCEVEHKYLHENKE, PvVYSIKEYNSLKDC, PkLYPTDEYNSLKDC detected only the parent protein in western blots and P. falciparum infected red blood cell lysates or blood lysates spiked with the respective proteins. Similar concentrations of PfPMT and the control, PfLDH, were detected in the same parasite lysate. The recombinant PfPMT protein was detected by a human anti-malaria antibody pool.

Conclusion: PMT, like the pan-specific LDH biomarker used in RDT tests, is both soluble, present at comparable concentrations in the parasite and constitutes a promising antimalarial drug target. PMT is absent from the human proteome. PMT has the potential as a biomarker for human malaria and in particular as the first P. knowlesi specific protein with diagnostic potential for the identification of a P. knowlesi infection.

Fig 1. Purification and molecular exclusion chromatography of recombinant P. falciparum, P. vivax and P. knowlesi PMT.

Samples (A, D and G) from the respective purification steps of three recombinantly expressed PMT orthologues, purified with an anti-His₆ affinity resin, were resolved on 12.5% reducing SDS-PAGE gels. Western blots (B, E and H) of the gels were probed with the appropriate antibodies. (M) Molecular weight marker; (lane 1) untransformed E. coli lysate; (lane 2) induced E. coli lysate; (lane 3) final wash and (lanes 4 to 8) eluents 1 to 5 from the Talon affinity matrix. The elution profiles (280 nm) were overlaid on the respective blots (B, E and H). Elution profiles (C, F and I) in milli absorbance units (mAu) of each PMT orthologue from a Sephacryl S-200 chromatography matrix. The calibration standards were plotted on the primary axis (dashed line), with the respective sizes indicated above each peak. Each PMT profile (black line) was plotted on the secondary axis and the estimated protein size indicated above the respective peaks in bold.

Fig 2. Detection of the recombinant PMT orthologues with the respective anti-recombinant PMT IgY.

(A) Reducing 12.5% SDS-PAGE reference gel for (B to D) western blots. The reference gel was the same as shown in Fig 4(A) as all experiments were performed at the same time on the same batch of samples. (M) Molecular weight marker, (lane 1) uninfected red blood cell lysate (25 μg), (lane 2) untransformed E. coli lysate (25 μg), (lanes 3 to 5) rPfPMT, rPvPMT and rPkPMT respectively. (B) Western blots probed with anti-rPfPMT or (C) anti-rPvPMT or (D) anti-rPkPMT IgY (10 μg) and detected with a rabbit anti-chicken-HRPO secondary antibody. (E to G) IgY against each of the whole recombinant PMT proteins (100 ng) was used to detect a range (0.8 to 100 ng) of concentrations of (E) rPfPMT, (F) rPvPMT and (G) rPkPMT in an ELISA. (H to J) IgY was diluted (100 to 1 ng) and used to detect a single concentration (100 ng) of rPfPMT, rPvPMT and rPkPMT respectively. Antibodies against a protein were denoted as “α” and the protein name. ELISA results present averages of triplicate values with standard deviations. Student’s t-test with p ≤ 0.05 and ≤ 0.001 are indicated with “*” or “**” respectively.

Fig 3. Sequence alignment of PMT orthologues and location of peptide epitopes on the PfPMT crystal structure.

(A) Alignment of P. vivax (XP_001614208.1), P. knowlesi (XP_002259925.1) and P. falciparum (Pf3D7_1343000.1) PMT protein orthologues. Potential epitopes selected by Predict7^™ analysis are indicated on the sequences as the common (boxed “common”) and species specific epitopes (boxed “Pf” or “Pv” or “Pk”). (B) The surface location of each selected peptide epitope is indicated on the 3D crystal structure of PfPMT (3uj6).

Fig 4. Detection of the recombinant PMT orthologues with the respective anti-PMT peptide IgY.

(A) Reducing 12.5% SDS-PAGE reference gel for (B to E) western blots. The reference gel was the same as shown in Fig 2(A). All experiments were performed at the same time on the same samples. (M) Molecular weight marker, (lane 1) uninfected red blood cell lysate (25 μg), (lane 2) untransformed E. coli lysate (25 μg) and (lanes 3 to 5) rPfPMT, rPvPMT and rPkPMT respectively. (B) Western blots probed with anti-PLENNQYTDEGVKC, (C) anti-PfCEVEHKYLHENKE, (D) anti-PvVYSIKEYNSLKDC, or (E) anti-PkLYPTDEYNSLKDC IgY (10 μg) and detected with a rabbit anti-chicken-HRPO secondary antibody using 4-chloro-1-naphthol and H₂O₂. (F to I) Detection of rPfPMT, rPvPMT and rPkPMT respectively (0.8 to 100 ng) in an ELISA with anti-peptide IgY (100 ng). (J to M) ELISA plates were coated with the PMT orthologues at 100 ng and detected with different dilutions of anti-PMT peptide IgY, with 100 to 1 ng (J to L) and 500 to 10 ng (M). All ELISAs were done in triplicate and the standard deviations included. Antibodies against a peptide were denoted as “α” and the peptide sequence. Student’s t-test with p ≤ 0.05 and ≤ 0.001 are indicated with “*” or “**” respectively.

Fig 5. PMT peptide detection with the anti-rPMT antibodies and DAS-ELISA based capture of rPMT orthologues.

(A) PMT peptides coated at 100 ng were detected with the anti-recombinant PMT IgY (100 ng). (B to D) The recombinant PMT proteins were captured with the anti-peptide antibodies and detected with the anti-PvPMT-HRPO conjugate. All results represent triplicate values with standard deviations. Antibodies against a peptide were denoted as “α” and the peptide sequence. Student’s t-test with p ≤ 0.05 and ≤ 0.001 are indicated with “*” or “**” respectively. Note the different scale in (B) compared to (C and D).

Fig 6. Spiked blood DAS-ELISAs and detection of native PfPMT in a Pf(D10) parasite culture lysate.

(A) Uninfected, A-positive human whole blood lysates were spiked with recombinant PMT orthologues (100 ng) and captured with anti-peptide IgY (all coated at 500 ng, except anti-PfCEVEHKYLHENKE at 1 μg). Captured PMT proteins were detected with anti-rPvPMT-HRPO coupled IgY (500 ng). rPfLDH (100 ng) spiked blood lysates served as a negative control. (B) Native P. falciparum PMT was detected on western blots, with the Coomassie stained reference gel on the left (SDS-PAGE). (M) Molecular weight marker, (lane 1) uninfected A-positive human whole blood lysate (100 μg) and (lane 2) Pf(D10) culture lysate (100 μg). Western blots were probed with: anti-rPfPMT (10 μg); anti-PLENNQYTDEGVKC or anti-PfCEVEHKYLHENKE or anti-rPfLDH IgY (100 μg). (C) Detection of native P. falciparum PMT in a Pf(D10) culture lysate with the anti-PfCEVEHKYLHENKE as capture and anti-rPvPMT-HRPO as the detection antibody. TMB + H₂O₂ was used as the substrate in all ELISAs, each performed in triplicate with standard deviations shown. Antibodies were denoted as “α” and the peptide sequence or protein name. Student’s t-test with p ≤ 0.05 and ≤ 0.001 are indicated with “*” or “**” respectively.

Fig 7. Detection of anti-rPfPMT antibodies in a human anti-malaria hyperimmune antibody pool.

A human anti-malaria hyperimmune antibody pool was passed over a rPfPMT affinity column. The affinity purified anti-rPfPMT human antibodies (100 ng) were used to detect each of the recombinant PMT orthologues or each of the PMT peptide epitopes coated directly onto ELISA plates at 100 ng per well. The pool before (antibody pool) and the pool after (depleted antibody pool) passing the human antibodies over the rPfPMT affinity resin were used at 10 μg. All readings were done in triplicate with standard deviations shown. A cut-off value for positive reactions of three times the standard deviation of the background control was included as a horizontal dashed line. Student’s t-test with p ≤ 0.05 and ≤ 0.001 are indicated with “*” or “**” respectively.