Characterization of two putative potassium channels in Plasmodium falciparum
© Waller et al. 2008
Received: 26 October 2007
Accepted: 24 January 2008
Published: 24 January 2008
Skip to main content
© Waller et al. 2008
Received: 26 October 2007
Accepted: 24 January 2008
Published: 24 January 2008
Potassium channels are essential for cell survival and participate in the regulation of cell membrane potential and electrochemical gradients. During its lifecycle, Plasmodium falciparum parasites must successfully traverse widely diverse environmental milieus, in which K+ channel function is likely to be critical. Dramatically differing conditions will be presented to the parasite in the mosquito mid-gut, red blood cell (RBC) cytosol and the human circulatory system.
In silico sequence analyses identified two open-reading frames in the P. falciparum genome that are predicted to encode for proteins with high homology to K+ channels. To further analyse these putative channels, specific antisera were generated and used in immunoblot and immunofluorescence analyses of P. falciparum -infected RBCs. Recombinant genome methods in cultured P. falciparum were used to create genetic knock outs of each K+ channel gene to asses the importance of their expression.
Immunoblot and IFA analyses confirmed the expression of the two putative P. falciparum K+ channels (PfK1 and PfK2). PfK1 is expressed in all asexual stage parasites, predominantly in late stages and localizes to the RBC membrane. Conversely, PfK2 is predominantly expressed in late schizont and merozoite stage parasites and remains primarily localized to the parasite. Repeated attempts to knockout PfK1 and PfK2 expression by targeted gene disruption proved unsuccessful despite evidence of recombinant gene integration, indicating that pfk1 and pfk2 are apparently refractory to genetic disruption.
Putative K+ channel proteins PfK1 and PfK2 are expressed in cultured P. falciparum parasites with differing spatial and temporal patterns. Eventual functional characterization of these channels may reveal future pharmacological targets.
In the course of a persistent human infection, Plasmodium falciparum parasites are exposed to dramatic changes in the extracellular milieu in terms of pH, osmolarity and ionic constituents. To survive such conditions transporter pumps, exchangers and ion channels operate to maintain the intra-parasite environment in the face of varying external conditions. Many classes of putative P. falciparum channels and transporters have been identified . These are thought to be critical for parasite viability [2, 3] and offer potential new molecular targets for the development of novel anti-malarials.
Here, a bioinformatic approach was used to identify two putative K+ channel genes (pfk1 and pfk2). This report details the cellular investigation of these putative K+ channel proteins, showing their differential expression and cellular localization in asexual blood stage parasites. The identification of only two putative K+ channels encoded within the P. falciparum genome, in conjunction with their differing expression and localization profiles suggests that these proteins may play unique and potentially critical roles in the maintenance of parasite viability.
Plasmodium falciparum 3D7 parasites were maintained by in vitro culture using standard culture conditions  and supplemented with 0.5% Albumax II (Invitrogen). Parasitaemias were determined by microscopic examination of Giemsa stained thin blood smears.
Sequence (5'→ 3') a
cgg gaattc ATGAATAATGATAATATTGGGAG
pfk1 full length (+)
ccc aagctt TCAGACTTGGTCATGCGGTTC
pfk1 full length (-)
cgg gaattc GTTGATAAGAAAGGAAAAATAATAGAT
pfk1 F2 (+)
pfk1 F2 (-)
ccgg gaattc atTTTAATAATCATGATCATCAGAATC
pfk1 F3 (+)
ccc aagctt TAAACATTCTTGATCATCTTTTTCC
pfk1 F3 (-)
ccgg gaattc tcAAAAAAGGACAAAACCATCAACTT
pfk1 F4 (+)
ccc aagctt CATATTTTTTTCCATATCTTTAATG
pfk1 F4 (-)
cgc ggatcc gATGAAAAGCGGATTATTTTCTATG
pfk2 full length (+)
ccg gaattc TCACAATATATAAACTATATCATCGAATC
pfk2 full length (-)
ccgg gaattc ATAATGATAATGATAATAATAATATTATA
pfk2 F3 (+)
ccc aagctt tTTTAATTCCTTTTTATTATCAATTGT
pfk2 F3 (-)
ccgg gaattc GAATATCAAGAATATTTACCAAAAAC
pfk2 F4 (+)
ccc aagctt ATTATGAATATGAATATTTGTATTTGA
pfk2 F4 (-)
aaa ctgcag TGAAAACAGAATGCGATTAGTTCG
pfk1 KO (+)
cgc ggatcc GATGATGCAAATGTTAATGGTAC
pfk1 KO (-)
aaa ctgcag TTCTTATCTTATTGAACGGATCTC
pfk2 KO (+)
cgc ggatcc TTGTATGGTATATATAGTAGATGC
pfk2 KO (-)
hrp2 3'UTR (+)
aaggaaaaaa gcggccgc AATGAATAATGATAATATTGGGAG
ccc aagctt AGATATAATACCATCCGTTTGG
ccg gaattc AATGAAAAGCGGATTATTTTCTATG
cgc ggatcc TTGTATGGTATATATAGTAGATGC
ccc aagctt ATTACAAAAATAATAATTGAAATTACT
Approximately 1.0 kb regions of pfk1 and pfk2 (downstream of, but not including the endogenous ATG start site) were amplified by PCR (Table 1) and cloned into the transfection plasmid pminiBSD , generating the plasmids pminiBSD/pfk1 -KO and pminiBSD/pfk2 -KO. Transfection of P. falciparum 3D7 parasites was performed using previously described methods [8, 9]. Transfected cultures were screened by PCR for plasmid integration  using specific primers (Table 1) approximately three months after transfection. Integration positive cultures were then cloned by limiting dilution .
Recombinant Glutathione S -transferase (GST) fusion proteins of PfK1 and PfK2 were generated. Regions of pfk1 and pfk2 (Figure 1B) were cloned into pGEX-KG and the encoded fusion proteins expressed and purified using standard affinity chromatography . Rabbit polyclonal antibodies were generated (Covance), and affinity purification was performed using the GST-fusion proteins immobilized on CN-Br Sepharose (Amersham Biosciences). The paired pre-immune control sera were obtained prior to immunization of the rabbits, and were used in simultaneous immunochemical methodologies (see above) as controls. No cross-reactivity was observed.
Parasite lysates were prepared from cultures synchronized by sorbitol lysis . Cultures were sampled at regular intervals and the cells lysed using 0.15% saponin in phosphate-buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.4 mM KH2PO4, pH7.4). Parasites were washed in STE (100 mM NaCl, 20 mM Tris-HCl pH8.0, 50 mM EDTA), before resuspending in reducing SDS-PAGE loading buffer and incubation at 37°C for 30 minutes. Samples were not boiled to facilitate solubility and prevent aggregation of transmembrane proteins [13, 14]. Proteins were resolved in 7.5% or 4–15% gradient polyacrylamide gels (Bio-Rad Laboratories) before transfer to nitrocellulose membranes. Western blotting was performed using affinity purified anti-PfK1 or anti-PfK2 antibodies (1/500) and detected using Western Lightning Chemiluminescence Reagent Plus (PerkinElmer Life Sciences). Control western blots were performed using polyclonal anti-KAHRP (provided by R. L. Coppel) and anti-GRP(BiP) antisera (; MRA20, MR4 ATCC).
Cellular localization of PfK1 and PfK2 by IFAs was performed on sorbitol-synchronized parasites. Thin blood smears were prepared from cultures at approximately 12-hour intervals before fixation with methanol/acetone (9:1). Slides were incubated with affinity purified antibodies in PBS containing 0.5% Bovine Serum Albumin (BSA) for 30 minutes, washed extensively with PBS and stained with goat anti-rabbit AlexaFluor®488 conjugate antibody (1/1000; Molecular Probes). For double-labelled IFAs, thin blood smears were prepared in the same manner and sequentially incubated with primary antibodies before staining with goat anti-rabbit AlexaFluor®488 and goat anti-mouse AlexaFluor®594 conjugate antibodies (both 1/1000; Molecular Probes). Parasite nuclei were visualized with 4',6-Diamidino-2-phenylinodole dihydrochloride (DAPI; 10 μg/ml). Slides were washed and mounted in Fluoromount-G (Southern Biotech). Cells were visualized using an Olympus IX81 electronically motorized microscope. Images were captured using a Sensicam QE cooled CCD camera (The Cooke Corporation) and IP Lab Spectrum Scientific Imaging Processing Software (Scanalytics Inc.). Images were then deconvolved using Vaytek Image software (Vaytek Inc.). The primary antibodies were: affinity purified anti-PfK1 F4 (1/250) and anti-PfK2 F4 (1/100), polyclonal rabbit anti-KAHRP (1/250) and anti-MESA (1/250 ; provided by R. L. Coppel) and mouse mAb5.2 anti-MSP1 (1/100; MRA94, MR4 ATCC).
Searches of the P. falciparum 3D7 genome database using the conserved K+ channel selectivity filter GYG motif yielded two gene sequences, designated here pfk1 (PFL1315w, or pfkch1 ) and pfk2 (PF14_0622). Pfk1 is approximately 6.1 kb and is located on chromosome 12. The pfk2 gene is approximately 4.4 kb and located on chromosome 14. Both genes possess a single exon and are predicted to encode highly charged proteins; PfK1 is rich in asparagine, lysine, arginine, glutamic and aspartic acid, whereas PfK2 is rich in asparagine, lysine, leucine, isoleucine and tyrosine. Both protein sequences are predicted to have six membrane-spanning regions and a pore region (Fig. 1B). Alignments of the pore region sequence from PfK1 and PfK2 with other well characterized K+ channels demonstrates overall similarity between these K+ channel proteins (Figure 1C), with closest similarity observed within the pore region of Ca2+-activated K+channels.
Additional blots were performed in parallel using antibodies against the endoplasmic reticulum (ER) marker protein Glucose-regulated protein GRP(BiP)  and the Knob Associated Histidine Rich Protein (KAHRP; Fig. 2). GRP(BiP) mRNA has been shown by microarray to be constitutively expressed across the 48 hr asexual life cycle [17–19]. Conversely, KAHRP is expressed by intracellular parasites in a stage-specific manner (see Cooke et al for review ). GRP(BiP) was readily detected in all samples at approximately equal levels, whereas maximal KAHRP expression was detected in mature trophozoites and schizonts.
Control IFAs were performed with KAHRP and Mature parasite-infected Erythrocyte Surface Antigen (MESA) antisera. KAHRP is exported into the red blood cell (RBC) cytosol where it associates with the membrane skeleton, forming electron-dense, knob-like structures at the RBC membrane. MESA is exported from the intracellular parasite into the RBC cytosol where it associates with the membrane skeleton, but has no specific association with knobs (see Cooke et al for review ). Increasing intensity of punctate KAHRP labeling was observed with increasing parasite maturity, with maximal labeling detected on the RBC membrane of trophozoites and schizonts (Figure 3C). Increased MESA labeling was observed at the RBC membrane of the mature trophozoites and schizonts with a nearly confluent appearance (Figure 3D). The pattern of PfK1 labeling more closely resembled the punctate appearance of the RBC membrane-associated protein KAHRP, rather than that of MESA.
Asexual P. falciparum parasites must rapidly respond to the widely variant ionic conditions, pH and osmolarities of their external milieus. In order to adapt to these environments, parasites most likely utilize ion channels and transporters to facilitate their regulation of intracellular ion concentrations. Allen and Kirk have shown that the transmembrane potential across the Plasmodium parasite plasma membrane is primarily generated by an electrogenic proton pump . The proton efflux in this study was balanced by influx of K+, presumably through such surface channels. This K+ transport may be carried by PfK1 and/or PfK2. During blood-stage development, various parasite proteins are synthesized and exported into the RBC resulting in modification of normal RBC structure and function . New permeation pathways (NPP) are hypothesized to facilitate the transport of low molecular mass solutes [2, 3] from the RBC membrane to the parasite and are likely to include many of the predicted parasite-encoded channels and transporters encoded in the parasite genome .
K+ channels are essential for the viability of all cells and are likely to be critical for malaria viability. The P. falciparum 3D7 genome encodes only two putative K+ channel genes, pfk1 and pfk2. The search criteria employed in this study specified that the selectivity filter motif be flanked by at least two predicted membrane-spanning segments. Although the number of transmembrane domains can vary with K+ channel type, in all cases, a single ion channel is created by the assembly of a combination of subunits that contribute four inner helices and four selectivity filter domains. Voltage-gated and Ca2+-activated K+ channels contain six (or seven, in the case of big conductance (BK) K+ channel) membrane spanning segments with the ion selectivity filter between the 5th and 6th membrane segment. Both PfK1 and PfK2 are predicted to have six membrane-spanning domains, and a pore region between the 5th and 6th segments based on hydropathy analysis and sequence alignments (Figure 1). The predicted pore-forming structures of PfK1 and PfK2 most resemble those of Ca2+-activated K+ channels. PfK1 also has a cluster of evenly spaced arginines in the 4th membrane-spanning segment that suggests the possibility of voltage-sensitivity. PfK2 lacks these charged residues in the 4th transmembrane segment and is thus more akin to intermediate conductance (IK) or small conductance (SK) channels. Full protein sequence alignments of PfK1 and PfK2 with other characterized K+ channels show less sequence similarity outside of the conserved pore-forming regions (data not shown), however their overall topologies are consistent with other voltage-gated and Ca2+-activated K+ channels.
A possible third P. falciparum K+ channel gene, PF14_0342, has been suggested from searches detailed by Martin et al . This sequence is unlikely to encode a K+ channel due its lack of several key features that are common to many K+ channels. The predicted selectivity filter contains a GKG motif instead of GYG (Fig. 1C) and lacks appropriate flanking transmembrane regions.
Here, the differential expression and localization of two putative P. falciparum potassium channels have been demonstrated. Recent global gene expression studies show pfk1 mRNA detectable throughout the entire 48-hour asexual cycle [17–19, 22]. In support of these data, PfK1 was detected by immunoblot in all RBC stages, though more protein was detected in mature forms. By IFA, PfK1 expression was detected from early trophozoites through schizont stage parasites. In ring stage parasites, the density of PfK1 may be too low, such that its detection was beyond the sensitivity of IFA. More intense labeling in mature-iRBCs (Figure 3) and also in multiply-iRBCs was consistently observed.
Like KARHP and MESA, PfK1 localizes to the RBC membrane. It is hypothesized that PfK1 is expressed by the parasite and exported into the RBC, where it is inserted into the RBC plasma membrane. Recent bioinformatic analyses of trafficked protein sequences have identified a conserved PEXEL/VTS motif that targets proteins for trafficking into the RBC [23, 24], in addition to the SS motif that directs proteins via the ER into the secretory pathway. In silico analysis [24, 25] of the PfK1 protein sequence for motifs indicated as potential export/transport signals failed to identify a VTS-like sequence, however manual searches yielded two potential VTS sequences, FFYRKLKNTFM and NFRRFLSSYKS (located 99 and 169 residues from the N-terminus, respectively). PfK1 also contains a predicted SS sequence (YDKRF SS RIKPK), located upstream of the VTS motifs and 20 residues from the N terminus of PfK1 (K. Haldar, personal communication).
PfK2 is primarily expressed in late schizonts and merozoites, in agreement with previous gene expression studies [18, 19, 23]. By IFA, PfK2 was detected in some, but not all, young ring stage parasites. Manual analysis of the PfK2 sequence revealed the presence of a possible SS cleavage motif and a low value VTS motif, but their configuration and location within the protein make it unclear whether these could function in the export of PfK2 (K. Haldar, personal communication).
From its localization, it is hypothesized that PfK1 encodes a parasite-induced ion channel that results in modification of the normal RBC to facilitate intracellular survival and maturation of the parasite by virtue of regulating K+ flux across the iRBC membrane, thereby effecting maintenance of the membrane potential and electrochemical gradient across the RBC plasma membrane and composition of the RBC cytosol.
PfK2, which is expressed predominantly in the merozoite, may function in the maintenance of membrane potential and the electrochemical gradient across merozoite membranes that are exposed to the human circulatory system. It is possible that PfK2 may act in concert with PfK1 during this time to adapt to environment changes the parasites are exposed to when released from the rupturing RBC. Invasion studies using the Apicomplexan parasite Toxoplasma gondii have shown that host cell invasion is accompanied by fluctuations of intracellular Ca2+ [26, 27], similar to those described during antigen-stimulated activation and proliferation of T cells. In those cells, effective Ca2+ signaling and ultimately downstream gene expression are modified by K+ channel activity to maintain the cell membrane potential . By analogy, it is hypothesized that PfK2 is a merozoite-specific K+channel that functions during the process of merozoite invasion of RBCs, perhaps functioning in ways to similarly effect intracellular Ca2+ signaling and gene transcription in the invading merozoite.
Heterologous expression and patch-clamp studies are required to definitively assign PfK1 and PfK2 to a class of K+ channels. Functional expression of both native and codon-optimized PfK1 and PfK2 cDNAs in a variety of heterologous systems has so far proven unsuccessful (McBride and McDonald, unpublished data). Expression of active parasite-encoded K+ channels in heterologous systems would illuminate the potential function of PfK1 and PfK2. Such a system could be adapted for the screening of novel specific blockers in a high through-put format for identification of possible new anti-malarials.
To examine the importance of these K+ channels for parasite survival, attempts were made to knockout pfk1 and pfk2 expression. The ability to detect 'snap shots' of integration by PCR and inability to isolate parasite clones despite repeated attempts are consistent with the hypothesis that both pfk1 and pfk2 are critical for parasite viability. Failed attempts to disrupt various parasite genes thought to be necessary for parasite invasion, viability or propagation have been reported [10, 29–31]. Further experimentation using either specific gene expression knockdown  or tetracycline-dependant conditional expression systems  may help unravel the precise functional roles of PfK1 and PfK2 in the regulation of K+ physiology in malaria-iRBCs.
Red Blood Cell
Indirect Immunofluorescence Assay
Plasmodium falciparum Potassium channel
polymerase chain reaction
Glutathione S -transferase
Phosphate Buffered Saline
Knob-Associated Histidine Rich Protein
Glucose-Regulated Protein (Binding protein)
Bovine Serum Albumin
Mature parasite-infected Erythrocyte Surface Antigen
infected Red Blood Cells
Uninfected Red Blood Cells
New Permeation Pathways
Small, Intermediate and Big Conductance K+ channels, respectively
Plasmodium Export Element
Vaculolar Transport Signal
We thank Ji-Mee Hwang and Danmei Qin for assistance during this investigation. We thank MR4 (Malaria Research and Reference Reagent Resource, ATCC) for providing the anti-GRP(BiP) and anti-MSP1 antibodies (contributed by John H. Adams and David C. Kaslow, respectively), Ross L. Coppel for providing the anti-KAHRP and anti-MESA antibodies and the Analytical Imaging Facility at Albert Einstein College of Medicine for their expert technical assistance and use of their imaging facilities. KLW is supported by an Australian NHMRC Howard Florey Centenary Research Fellowship. This research was funded by the USAMRMC grant #DAMD17-02-1-0290.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.