PK4, a eukaryotic initiation factor 2α(eIF2α) kinase, is essential for the development of the erythrocytic cycle of Plasmodium

Abstract

In response to environmental stresses, the mammalian serine threonine kinases PERK, GCN2, HRI, and PKR phosphorylate the regulatory serine 51 of the eukaryotic translation initiation factor 2α (eIF2α) to inhibit global protein synthesis. Plasmodium, the protozoan that causes malaria, expresses three eIF2α kinases: IK1, IK2, and PK4. Like GCN2, IK1 regulates stress response to amino acid starvation. IK2 inhibits development of malaria sporozoites present in the mosquito salivary glands. Here we show that the phosphorylation by PK4 of the regulatory serine 59 of Plasmodium eIF2α is essential for the completion of the parasite's erythrocytic cycle that causes disease in humans. PK4 activity leads to the arrest of global protein synthesis in schizonts, where ontogeny of daughter merozoites takes place, and in gametocytes that infect Anopheles mosquitoes. The implication of these findings is that drugs that reduce PK4 activity should alleviate disease and inhibit malaria transmission.

Fig. 1.

PK4 phosphorylates Plasmodium eIF2α in vitro and yeast eIF2α. (A) In vitro kinase assay. The purified GST-PfPK4KD with (lane 2) or without (lane 1) substrate PfeIF2α was incubated with [γ-³²P]ATP at 37 ºC for 30 min. Radiolabeled proteins were separated by SDS/PAGE and visualized by Coomassie staining and autoradiography. As shown, the PK4 kinase autophosphorylates and phosphorylates eIF2α. (B) PK4 phosphorylates eIF2α in yeast. Plasmids expressing GST or GST fused to the intact PK4KD were transformed into yeast strain H2557 expressing wild-type eIF2α. Whole-cell extracts were subjected to SDS/PAGE followed by immunoblot analysis using antibodies to GST, to phosphorylated Ser51 (eIF2α-P), or to total yeast eIF2α. (C) PK4 regulates yeast cell growth. An empty vector or plasmids expressing the positive control human PKR, GST, or GST fused to the PK4KD under the control of a galactose-inducible promoter were transformed into isogenic yeast strains expressing wild-type eIF2α (eIF2α; H2557) or nonphosphorylatable eIF2α-S51A (J223). Patches of two independent transformants for each plasmid were grown to confluence on synthetic dextrose (SD) plates, where the galactose-inducible promoter is repressed, and then replica-plated to synthetic galactose (SGal) medium, where the galactose-regulated promoter is induced. Plates were incubated at 30 ºC for 2 d.

Fig. 2.

PbPK4cKO development into liver stages. (A) Excision of PbPk4 in PbPK4cKO sporozoites was quantitated by qPCR. Shown are the means ± SD of three independent experiments. *P < 0.0001 (two-tailed Student's t test). (B) C57BL/6 mice (six mice per group) were i.v. injected with 10⁴ TRAP/FlpL or PbPK4cKO sporozoites. Liver-stage parasite burden was measured by qPCR, and the means ± SD (**P = 0.8353) are shown. This experiment was performed two times with similar results. (C) HepG2 cells infected with TRAP/FlpL or PbPK4cKO sporozoites were subjected to in situ hybridization followed by immunofluorescence with anti-HSP antibody to reveal the infected hepatocytes. A Cy3-probe was used to label the PK4 mRNA. Infected cells that presented PK4 mRNA were counted in both wild-type and mutant parasites, and results are expressed as a percentage. Only 33 ± 27% exoerythrocytic forms contained PK4 transcripts. (Scale bars, 20 μm.) ***P < 0.0001.

Fig. 3.

Generation of parasites with PbeIF2α Ser59 substitutions. (A) Diagram depicting the PbeIF2α genomic locus, the pBCeIF2α_S/* (A, T, D, or S) allelic replacement vectors, and the PbeIF2α Ser59 mutant parasite locus. Triple lines indicate the replacement vector backbone. The mutated nucleotides are presented in lowercase. The DHFR cassette is under the control of the 5′ UTR of EF1α and the 3′UTR of DHFR-TS. The restriction sites MscI and BstBI (underlined) were introduced with the S/A and S/S mutations, respectively. (B) BstBI digestion analysis of the PbeIF2α PCR product from S/S population gDNA. (C) Replacement-specific PCR analysis of PbeIF2α genomic locus. Lanes 1, 3, and 5: primers P1+P2; lanes 2, 4, and 6: primers P3+P4. Confirmation of integration is shown by primers P1+P2 and P3+P4 that amplify an 820- and a 550-bp fragment, respectively, from the gDNA of PbeIF2α S/S clone C1 and S/T clone E1; note that these bands were absent in the wild-type parasites. (D) MscI digestion analysis of the PbeIF2α PCR product from the pBCeIF2α_S/A-transfected and pyrimethamine-resistant parasite population gDNA. MscI digestion of the PbeIF2α PCR product from wild-type parasites produced two bands of 720 and 560 bp. Although three bands of 450, 270, and 560 bp were expected from digestion of the S/A mutant population, only the wild-type pattern was observed.

Fig. 4.

Phosphorylation of eIF2α in gametocytes and schizonts. (A) P. berghei ANKA-infected mouse blood was separated by 70% percoll gradient centrifugation. (Upper) Old trophozoites (OT), schizonts (S), and gametocytes (G). (Lower) Ring stage (R) and young trophozoites (Y). Upper and Lower bands were collected and uninfected erythrocytes were removed by 0.2% saponin treatment. PbeIF2α-P and total PbeIF2α were revealed by Western blots using specific antibodies (21). Densitometry results are presented below each band. This experiment was repeated with identical results. (B) Kinetic incorporation of [35S] Met/Cys in parasites from percoll separated Upper and Lower bands. (C) Mature schizonts were synchronized in an in vitro culture of blood from a P. berghei ANKA-infected mouse. Gametocytes were collected from the infected mouse treated with pyrimethamine for 2 d. Blood stages, BS. PbeIF2α-P and total PbeIF2α were revealed by Western blots. Densitometry results are presented below each band. (D) Stress granule-associated protein phosphorylated eIF2α (22) was revealed by immunofluorescence in P. falciparum male and female gametocytes and in schizonts. (Scale bar, 10 μm.)