Copper-transporting ATPase is important for malaria parasite fertility

Homeostasis of the trace element copper is essential to all eukaryotic life. Copper serves as a cofactor in metalloenzymes and catalyses electron transfer reactions as well as the generation of potentially toxic reactive oxygen species. Here, we describe the functional characterization of an evolutionarily highly conserved, predicted copper-transporting P-type ATPase (CuTP) in the murine malaria model parasite Plasmodium berghei. Live imaging of a parasite line expressing a fluorescently tagged CuTP demonstrated that CuTP is predominantly located in vesicular bodies of the parasite. A P. berghei loss-of-function mutant line was readily obtained and showed no apparent defect in in vivo blood stage growth. Parasite transmission through the mosquito vector was severely affected, but not entirely abolished. We show that male and female gametocytes are abundant in cutp− parasites, but activation of male microgametes and exflagellation were strongly impaired. This specific defect could be mimicked by addition of the copper chelator neocuproine to wild-type gametocytes. A cross-fertilization assay demonstrated that female fertility was also severely abrogated. In conclusion, we provide experimental genetic and pharmacological evidence that a healthy copper homeostasis is critical to malaria parasite fertility of both genders of gametocyte and, hence, to transmission to the mosquito vector.

Alignment of the two regions containing the three most carboxy-terminal transmembrane domains (H6-8) of twelve representative P 1B1 -type ATPases, which selectively transport monovalent copper or silver (Argüello, 2003). Highlighted with green is an aspartic acid residue conserved in all P-type ATPases. Conserved residues with potential metal-binding side chains are highlighted with cyan. The combined presence of the "CPC" and "MXXSS" motifs in H6 and H8, respectively, is a determinant of Cu + -selectivity. Both motifs were found in all apicomplexan CuTP sequences. Table S1 lists all protein sequence IDs, the sources, species, and strains.  Most parsimonious phylogenetic tree of copper-transporting P-type ATPases from 26 eukaryotic organisms, including many common model and pathogenic species. The tree was built using Escherichia coli K12 CopA as the outgroup and nodes with bootstrap values <70 were collapsed. Dots indicate nodes with bootstrap values ≥85 (light gray), ≥95 (dark gray), or of 100 (black). All sequences clustered in kingdom specific clades, though both plantae and alveolata formed two distinct clades. All apicomplexan parasites formed a single clade with a highly confident subclade including all Plasmodium spp. Table S1 lists all protein sequence IDs, the sources, species, and strains. The sequences are grouped and shaded according to the major clades as they appear in the phylogenetic tree. Sequences from human pathogens are indicated in bold. The tree was constructed using the PHYLIP package (Felsenstein, 1996). A. Replacement strategy to generate cutp::tag parasites. The carboxy-terminal and 3'UTR sequences of CuTP were cloned into the pBAT-SIL6 P. berghei transfection vector (pCuTP-tag). Upon an double cross-over homologous recombination event, CuTP is predicted to fuse in-frame to the mCherry-3xMyc tag and introduce the GFP and drug-selectable cassettes. Replacement (5'INT and 3'INT, black) and wild type (WT, gray)-specific test primer combinations and expected fragments are indicated. B. Diagnostic PCR confirms successful integration of the CuTP-tag and absence of WT parasites after FACS of the isogenic cutp::tag parasite line. C. Western blot analysis using anti-mCherry antibody shows absence of processed mCherry-3xMyc tag (33 kDa) and presence of a distinct band (black arrow) above 170 kDa indicative of the unprocessed CuTP::tag fusion protein (259 kDa).      (Kenthirapalan et al., 2012) b Reference: (Haussig et al., 2011) c Sequences encoding the two c-myc epitopes are in bold S10

Cloning of plasmids pCuTP-KO and pCuTP-tag
First, a 486 bp fragment of the 3'UTR was amplified from gDNA using the primer combination CuTP-F5-AvrII and CuTP-R4-KpnI (see Table S2 for all primer sequences) and cloned into the pBAT-SIL6 vector  using AvrII and KpnI to generate the intermediate construct pCuTP-IM. For the generation of the CuTP disruption construct (pCuTP-KO), a 561 bp fragment of the 5'UTR was amplified from gDNA using the primer combination CuTP-F2-SacII and CuTP-R1-EcoRI. This fragment was subcloned using EcoRI and SacII into an intermediate pBAT-derived vector. Subsequently, the CuTP targeting fragment was released by cleavage with HpaI and SacII and cloned into pCuTP-IM using PvuII and SacII.
For the tagging construct, termed pCuTP-tag, a 498 bp fragment of the carboxyterminal coding region of CuTP was amplified from gDNA using the gene-specific primers CuTP-F4-SacII and CuTP-R3-PshAI and fused in frame to the mCherry-3xMyc tag of pCuTP-IM using SacII and HpaI.

Western blot analysis
Whole protein extracts of mixed blood stage WT and cutp::tag parasites were isolated and separated on a 8 % SDS-polyacrylamide gel. The separated proteins were blotted on a PVDF membrane, incubated with rat monoclonal anti-mCherry antibodies (1:1000; ChromoTek) and detected with horseradish peroxidase coupled goat anti-rat antibodies (1:5000; Jackson ImmunoResearch).

Cloning of the pTgCuTP-myc plasmid
A 2.3 kb fragment of the carboxy-terminal region of TgCuTP was amplified from genomic DNA using the primer combination TgCuTP-F2-ApaI and TgCuTP-R1-2xmyc-PacI. The resulting fragment was cloned into the p5RT70mycGFP-DD/HX vector (Herm-Götz et al., 2007) using restriction digestion with ApaI and PacI, thus replacing the 5'TUB-Myc-GFP-DD sequence. Prior to transfection, the resulting pTgCuTP-myc plasmid was linearized overnight by restriction digestion with EcoRV and purified via ethanol precipitation. Integration of pTgCuTP-myc is thus predicted to result in the carboxy-terminal tagging of the endogenous TgCuTP fused in-frame to a double c-Myc epitope tag followed by the 3' untranslated region of TgSAG1 along with insertion of the HXGPRT drug-selectable cassette.

In vitro drug inhibition assay
To determine the distribution of asexual blood stages in vitro, mice were infected with purified schizonts and bled after 2 h. The resulting synchronized ring stages were cultured in the absence or presence of the intracellular copper chelator neocuproine (Sigma-Aldrich) for 20 h, as indicated. Thereafter, DNA was stained with Hoechst 33342 (Invitrogen) and the intensity of the nuclear stain analyzed using a MACSQuant Analyzer (Milteny Biotec; Fig. S6). The red blood cell population was gated to exclude small debris and large cells using the forward scatter versus side scatter plot. This subpopulation was further gated for GFP-positive events. The Hoechst-positive cells of the GFP-positive subpopulation were divided into two subpopulations, according to their DNA content, i.e. 1n and multi n (≥8n).