Plasmodium falciparum has been the focus of much research due to it being the causative agent of malaria. This article describes some of the recent findings surrounding the unique biology of this organism.

Diagram of the major organelles including the rhoptries, micronemes and polar rings near the apical end.

This article may be too long to read and navigate comfortably. Please consider splitting content into sub-articles and/or condensing it. (December 2012)

It has been suggested that this article be split into multiple articles. (Discuss) Proposed since December 2012.

Morphology

The nucleus, mitochondrion, apicoplast and the microtubules of Plasmodium sporozoites are linked to the parasite pellicle via long tethering proteins. The tethers originate from the inner membrane complex and are arranged in a periodic fashion following a 32 nanometer repeat. The tethers pass through a subpellicular structure that encompasses the entire parasite, probably as a network of membrane associated filaments.^[1]

The pellicle is a structure unique to the Apicomplexa made up of four components: the plasma membrane, the inner membrane complex, the subpellicular network and the subpellicular microtubules.^[2] The subpellicular network consists of a two-dimensional network of intermediate filaments located on the cytoplasmic side of the inner membrane complex and acts as a membrane skeleton. The proteins - the inner membrane complex proteins (IMCs) - that compose this structure are functional homologs of the articulins, the membrane skeleton proteins of free-living protists.

Cell division

Cell division occurs through a process known as schizogony. This is a type of mitotic division in which multiple rounds of nuclear divisions occur before the cytoplasm segments.

DNA synthesis begins in the relatively small trophozoites but nuclear subdivision, which leads to the formation of multinucleate cells, occurs only during schizogony. Whether or not any gap phases exist between each round of DNA synthesis and mitosis is unknown. Eventually, a schizont composed of 8–32 nuclei undergoes segmentation, which culminates with the formation of individual merozoites that burst from the erythrocyte into the blood stream.

Liver stages

Invasion of the hepatocytes appears to involve at least 2 proteins: sporozoite invasion-associated proteins (SIAP)-1 and -2.^[3] These proteins bind heparin sulfate and chondroitin sulfate type membrane receptors on host cells.

Latency of sporozoites is controlled by the eIF2 alpha kinase IK2, a general inhibitor of protein synthesis.^[4] Puf2 participates in the regulation of IK2 and inhibits premature sporozoite transformation. In contrast Puf1 appears to be dispensable.

The RNA binding protein family PUF member Pumilio-2 (Puf2) appears to be involved in transformation of sporozoites into the hepatic stage of the life cycle.^[5] Knock out mutants of this gene exhibit genome wide transcriptional changes resulting in loss of gliding motility, cell traversal ability, reduction in infectivity and trigger metamorphosis typical of early Plasmodium intra-hepatic development.

The division of the liver stages into thousands of merozoites is a complex process. In parallel with nuclear division, the apicoplast and mitochondrion become two extensively branched and intertwining structures.^[6] The organelles subsequently undergo morphological and positional changes prior to cell division. Finally to form merozoites, the parasite undergoes cytokinesis.

Type II fatty acid biosynthesis is vital for this stage in the life cycle.^[7] This pathway may be inhibited by the antibiotic triclosan.

During this stage of development the sporozoite selectively discards organelles unnecessary for growth at this stage of the life cycle. Among these are the micronemes and the inner membrane complex.^[8]

The host iron regulatory hormone hepcidin which is synthesised in the liver and spleen, appears to be able to inhibit growth of the liver stages.^[9] Levels of this hormone are elevated during infection and seem to correlate with the anaemia often found in malaria.^[10] Erythrocytic parasitaemia, above a minimum threshold, impairs the growth of subsequent liver cell sporozoite infection.^[9] The production of hepcidin leads to the redistributes iron away from hepatocytes thus slowing the development of the iron dependent liver stage.

Liver hepcidin expression is upregulated and downregulated during the early and late stages of malaria infection respectively.^[11] Inflammation and erythropoietin, rather than the iron sensing pathway, are involved in the regulation of hepcidin expression. Treatment of malaria infected mice with anti hepcidin neutralizing antibodies increased parasitemia and mortality rates. Overexpression of hepcidin improves the outcome.

Lipocalin 2, a host protein that sequesters iron, is upregulated during infection and appears to be involved in the host response.^[12] This protein increases both host macrophage function and granulocyte recruitment and decreases reticulocytosis.

Expression of the iron sequestering protein ferritin (ferritin H chain in mice) is associated with decreased tissue damage.^[13] The mechanism appears to be via prevention of activation of the proapoptotic c-Jun N-terminal kinase.

Invasion of the hepatocyte seems to require the RON4 protease.^[14]

Within the liver actin reorganization is a dynamic process in part controlled by the actin severing and capping protein - gelsolin.^[15] The hepatocyte cytoskeleton may contribute to parasite elimination.

Within the genome is encoded a homolog of macrophage migration inhibitory factor. This gene appears to be important for parasite development in the liver.^[16]

In Plasmodium bergei a protein - liver specific protein 2 (LISP2) - is expressed 24  hours after infection and rapidly increases during the liver stage schizogony. LISP2 is carried first to the parasitophorous vacuole and subsequently transported to the cytoplasm and nucleus of host hepatocytes. Mutations in this gene result in arrested development of the merozoites.^[17]

Two other proteins (p52 and p36)in Plasmodium bergei appear to be important in the formation of the parasitophorous vacuole membrane in the liver.^[18]

Erythrocyte invasion

This is a complex and poorly understood process. The merozoite initially contacts the erythrocyte and rotates until the rhoptery containing part is adjacent to the erythrocyte membrane. A tight contact is then established and the parasite enters the erythrocyte. This happens within seconds making the invasion process difficult to analyse.

Initial adhesion

Two families of proteins are known to be involved in this process: the reticulocyte binding-like homologues (PfRh or PfRBP) and erythrocyte binding-like (EBL) proteins. The EBL family are principally located in the micronemes and the Reticulocyte binding Homolog (PfRH) family are principally located in the rhopteries. Ligands from the EBL family largely govern the sialic acid dependent pathways of invasion and the RH family ligands (except for RH1) mediate sialic acid independent invasion.^[19]

The PfRh family consists of five proteins and a pseudogene: PfRh1, PfRh2a, PfRh2b, PfRh3, PfRh4 and PfRh5. PfRh3 is a transcribed pseudogene in all strains examined to date.^[20] All the other members of this family bind to erythrocyes and antibodies to them inhibit invasion. PfRh5 is located within the rhoptries and appears to be an essential gene.^[21]

Reticulocyte binding like protein homologue 2a (PfRH2a) is processed both in the schizont as well as during invasion resulting in proteins with different erythrocyte binding properties.^[22] It also moves from the rhoptry neck to the moving junction during merozoite invasion. PfRh2a undergoes a cleavage event in the transmembrane region during invasion consistent with activity of the membrane associated PfROM4 protease.^[23] Both PfRh2a and PfRh2b bind to red blood cells. The erythrocyte-binding domain lies within a 15 kDa region at the N-terminus of each protein.

PfRh4 binds to a second protein P. falciparum Rh5 interacting protein (PfRipr). PfRipr has a molecular weight of 123 kiloDaltons with 10 epidermal growth factor-like domains and 87 cysteine residues distributed along the protein.^[24]

The receptor for the PfRh5 protein appears to be the Ok blood group antigen, basigin.^[25] Blocking access to this protein on the erythrocyte surface appears to inhibit erythrocyte invasion completely. Binding of the Rh5 protein appears to be critically dependent on a single residue within the Rh5 protein.^[3]

The EBL family of proteins includes EBA-175, EBA-181 (also known as JESEBL), EBA-140 (also known as BAEBL) and EBL-1. Whilst these parasite ligands function in merozoite invasion by binding to specific receptors on the erythrocyte, they appear also to have a central role in activation of the invasion process. Binding of EBA-175 to its receptor, glycophorin A, restores the basal cytosolic calcium levels after interaction of the merozoite with the erythrocyte and triggers the release of rhoptry proteins.^[26]

These proteins have several domains. Region II which is responsible for ligand-erythrocyte interaction during invasion, consists of two homologous F1 and F2 domains.^[27]

A member of the EBL family of proteins (maebl) has been shown to be present in Plasmodium gallinaceum.^[28] This protein is now known to be conserved in the primate, rodent and avian infecting species suggesting that it may play an important role in erythrocyte invasion.

The EBL proteins have a Duffy like binding domain (DBL) unique to Plasmodium species. The crystal structure of EBA-140 has been solved.^[29] This protein binds glycophorin C on the host cell membrane. The two domain binding region is present as a monomer. Both domains are required for binding to occur. Its electrostatic surface has a basic patch spanning both DBL domains that is important in the binding mechanism.

The protein EBL-1 appears to bind to glycophorin B.^[30]

There seems to be some overlap between the functions of these proteins. Loss of EBA-175 can be compensated by increased expression of PfRh4.^[31]

Another family of proteins involved in the invasion process are the thrombospondin related anonymous protein (TRAP) family. These proteins are type I cell surface proteins with one or more extracellular thrombospondin type-I repeats (TSR) domains and/or von Willebrand factor like (vWF) A domain(s) and an acidic cytoplasmic tail with a subterminal tryptophan residue. The cytoplasmic tails of TRAP, CTRP, TLP and MTRAP interact with the enzyme aldolase.

The motile forms have their own stage specific cell surface TRAP family member: TRAP and S6 (also known as TREP) occur on the sporozoites; CTRP is found on the ookinetes; MTRAP is expressed in the merozoites; and TLP is present on both sporozoites and merozoites. Other members of this family are the proteins CSP, SPATR, TRSP, WARP and PTRAMP. Roles for several of these proteins has been discovered: TRAP is critical for sporozoite invasion of the mosquito salivary glands, infection of mammalian liver and sporozoite gliding motility; CTRP is required for invasion of the mosquito midgut; and S6 is important for both sporozoite gliding motility and invasion of mosquito salivary glands. TLP has a role in sporozoite cell traversal. The cytoplasmic tail of TRAP is essential for gliding motility and invasion of the mosquito's salivary glands. Both the TSR and A-domains of TRAP are required for the invasion of the mosquito salivary glands. Penetration of the mammalian hepatocytes however requires the TSR, the A-domain and the cytoplasmic tail. In contrast only the A-domains of CTRP are essential for infectivity by the ookinete.

Another protein thought to be involved in the invasion process is the merozoite-specific thrombospondin related anonymous protein homolog (MTRAP). The receptor for this protein has been identified as the GPI-linked protein semaphorin-7A (CD108).^[32] The MTRAP monomers interact via their tandem TSR domains with the Sema domains of a Semaphorin-7A homodimer.

The protein PfTCTP causes the release of histamine in the host.^[33] This protein also activates the basophils.

A GPI-anchored micronemal antigen (GAMA) also appears to be essential in the process of erythrocyte invasion.^[34]

Merozoite proteins 8 and 10 which are thought to be involved in the invasion process appear to be under purifying selection.^[35]

The cysteine rich protective antigen appears to play a role in this process.^[36]

The ribosomal phosphoprotein P0 also seems to be involved in the invasion process.^[37]

A double C2 domain (DOC2) protein appears to be involved in the invasion of the erythrocyte.^[38] DOC2 proteins recruit the membrane fusion machinery an essential part of the Ca²⁺-dependent exocytosis mechanism.^[39] These proteins have a Munc13-interacting domain and tandem C2s (designated C2A and C2B) which are connected by a short polar linker. The C2 domains bind phospholipids in a Ca²⁺-dependent manner. Elucidating their precise role in erythrocyte invasion requires further work.

In Plasmodium vivax a number of tryptophan rich antigens are involved in erythrocyte invasion.^[40] Homologs of these proteins are found in P. falciparum - tryptophan-threonine rich antigen (PfTryThrA) and merozoite associated tryptophan rich antigen (PfMaTrA) and Plasmodium yoelii. These proteins also seem to be involved in the invasion process.

Invasion

The invasion process appears to be ATP dependent^[41] and may involve a purogenic signalling pathway.

The invasion process requires a coupling of the actin-myosin motor to the surface receptors. The myosin molecule involved belongs to the single-headed class XIV myosin. For the thromobospondin related anonymous protein on the sporozoites, aldolase which can bind actin forms this connection.^[42] This connection requires tryptophan and negatively charged amino acids in the ligand's cytoplasmic tail. PfRH2b also binds aldolase with its cytoplasmic tail. This binning requires an aromatic amino acid (phenylalanine or tyrosine) rather than tryptophan again also in the context of negatively charged amino acids. PfLRH2a does not bind aldolase. A second protein glyceraldehyde-3-phosphate dehydrogenase can also bind actin. It is capable of biding the cytoplasmic tails of some of the PfRh and Duffy biding ligands ligands in an aromatic amino acid dependent manner.

The motor behind the invasion process is an actinomyosin motor complex that is assembled below the parasite's plasma membrane.^[43] This complex includes myosin, myosin tail domain interacting protein and glideosome associated proteins 45 and 50. It is anchored on the inner membrane complex which underlies the cell membrane. Myosin, myosin tail domain interacting protein and GAP45 first form a complex that then associates with GAP50. GAP45 is phosphorylated by calcium dependent protein kinase 1 on a number of serine residues. Removal of these residues does not appear to affect the assembly of this complex. This complex may have other function in addition to its role erythrocyte invasion.

GAP45 is phosphorylated in response to Phospholipase C and calcium signaling.^[44] It is phosphorylated by the P. falciparum kinases Protein kinase B and Calcium dependent protein kinase 1, both of which are calcium dependent enzymes, at Serine89, Serine103 and Serine149. Phosphorylation of these sites is differentially regulated during parasite development.

Some details of the invasion process are known.^[45] The rhoptery protein RON2 is inserted into the erythrocyte membrane. The protein AMA1 secreted from microneme then binds to RON2. RON2 forms part of a macromolecular complex which includes RON2, RON 4, RON5 and RON8.

The calcium dependent protein kinase 1 appears to play a role in micronene discharge.^[46] The drug purfalcamine is a specific inhibitor of this kinase and it also inhibits micronene discharge and erythrocyte invasion. These kinases typically have an N terminal kinase domain and C terminal calmodulin like domain with calcium binding EF hands. The N and C terminals are joined by a junction domain. The C terminal appears to interact with the junction domain in the process of binding calcium.

Apical complex

The rhopteries appear to have subcompartments allowing for differential secretion during the life cycle.^[47] Two of these are known as the neck and the bulb.^[48] A number of rhoptry neck proteins are conserved between apicomplexan species and are involved in host cell invasion. Bulb proteins in contrast are less well conserved between the apicomplexa and most likely evolved for a particular lifestyle. In In the majority of species studied to date, rhoptry content is involved in formation and maintenance of the parasitophorous vacuole.

The rhoptery neck proteins (RONs) along with the micronemal AMA1 protein are important in the penetration of the erythrocyte.^[14] These form part of the moving junction which initially binds to the erythrocyte surface and is involved in the entry of the parasite into the erythrocyte cytoplasm. The mechanisms involved in this process are still being elucidated. The protein RON8 appears to be central to the binding of parasite to the erythrocyte surface.^[49] Apical Sushi Protein and Rhoptry Neck protein 2 are released early following the formation of the tight junction between the merozoite and the erythrocyte. The rhoptry protein PFF0645c is released only after invasion is complete.

The rhoptry protein 2 of Plasmodium vivax has been cloned.^[50] The 1,369 amino acid protein is encoded PVX_099930 gene. The gene has nine introns and the protein contains a signal peptide at its N-terminus and 12 cysteines predominantly in its C-terminal half. It is localized in one of the apical organelles of the merozoite, the rhoptry, and the localization pattern is similar to its homolog in P. falciparum.

Apical membrane antigen-1 (AMA-1) - the product of the Pfama1 gene - is a surface exposed protein that plays a role in erythrocyte invasion. It is shed from the parasite surface predominantly via the action of the protease Sub2^[51] Sub2 is released from the micronemes and can also act on the MSP1/6/7 complex and PTRAMP - another micronemal protein. Sub2 appears to be an essential gene.

The residues of the RON2 protein binds to the AMA-1 protein have been identified.^[52] It also appears that the formation of the junction and parasitophorous vacuole are molecularly distinct steps in the invasion process. Positive diversifying selection appears to have acted in the RON2 protein of Plasmodium vivax.^[53]

The receptor binding site of AMA-1 comprises the hydrophobic groove and a region that becomes exposed by displacement of the flexible domain II loop.^[54]

The P. falciparum apical sushi protein is the homolog of the P. vivax RON1 protein.^[55]

Several proteins are involved in the binding of the sporozoite to the various tissues it attaches to. TRAP, S6 and TLP have been implicated in these processes.^[56] Heparin like molecules bound to the surface of the erythrocyte appear to be important in this process which involves the merozoite surface protein 1.^[57]

Maurer's clefts

In 1902 the German physician Georg Maurer discovered an unusual staining pattern in the cytoplasm of erythrocytes infected with P. falciparum. These structures were subsequently named Maurer's clefts. These consist of a convoluted set of membranes that lie within the erythrocyte's cytoplasm and appear to be involved in secrection from the erythrocyte.^[58] They are known to have proteins of parasite origin within them including the Maurer's cleft two transmembrane proteins (PfMC-2TM)^[59] The clefts appear to originate from vacuoles budding off them the parasitophorous vacuole membrane which then diffuse within the erythrocyte cytoplasm before taking up residence at the cell periphery.^[60]

Another protein associated with these structure is skeleton-binding protein 1 (SBP1). This protein is involved in transport of the var gene protein, pfEMP1 (erythrocyte membrane protein 1) to the erythrocyte surface.^[61]

Other proteins associated with these structures include membrane associated histidine rich protein 1 and ring exported protein 1 and 2.^[62][63]

Mutations in the ring exported protein 1 (Rex 1), a protein normally found in Maurer's clefts, reduces transport of the var gene products to the erythrocyte surface.^[64]

The erythrocyte protein ankyrin is found in these structures.^[65]

The parasite generates a host derived actin cytoskeleton within the cytoplasm of the erythrocytes that connects the Maurer's clefts with the host cell membrane and to which transport vesicles are attached.^[66] Hemoglobin oxidation products which are enriched in hemoglobin S and C containing erythrocytes inhibit actin polymerization. This may account for their protective role in malaria.

The protein trophozoite exported protein 1 (PFF0165c) is located within the clefts.^[67] The protein's N-terminal region is intrinsically unstructured but it also has a coiled coil domain. It appears to lack export motifs such as PEXEL, signal sequence/anchor or a transmembrane domain. Transport of this protein to the clefts is sensitive to inhibition by Brefeldin A. This is normally associated with proteins that are have co-translation translocation into the endoplasmic reticulum or posttranslational insertion into the endoplasmic reticulum followed by vesicular transport from the endoplasmic reticulum via Golgi apperatus to the cell surface.

Mutations in REX1 and Pf332 proteins result in distortion of Maurer's clefts morphology suggesting that they play a role in its structure.^[68][69]

Pf332 protein

The 700 kiloDalton protein Pf332 is the largest known exported asexual malaria protein. The protein has three parts: an N-terminal Duffy binding like domain followed by a putative transmembrane region and a large number of negatively charged repeats that are not identical but have the consensus (X)₃-EE-(X)₂-EE-(X)_2–3 where E is glutamic acid and X is a hydrophobic amino acid. The repeat portion of the protein consititue more than 90% of the protein. The protein has a predicted isoelectric point (pI) of 3.8. It is known to associate with the erythrocyte plasma membrane.^[70][71]

The Pf332 protein can first be detected within the parasite at 20–24 hours post invasion, after which it translocates across the parasitotopherous vacuole membrane into the host cell cytosol.^[72] It is initially synthesised in the endoplasmic reticulum and eported to the host cytosol. From there it is trafficked as part of a multimeric protein complex to Maurer's clefts. It may interact with two chaperone proteins - PF14_0700 (a hypothetical protein with a J domain) and PFB0595w (a heat shock protein 40).^[73] It is associated with the cytoplasmic side of Maurer's clefts in a peripheral manner throughout trophozoite maturation and schizogony.^[74] In the clefts both the N and C-termini are localised to the erythocyte cytosol.^[74] Export of Pf332 is sensitive to treatment with Brefeldin A^[75] The export signal appears to be encoded in the N terminal domain^[76]

It interacts with the erythrocyte cytoskeleton and binds actin.^[69][77]

Parasitophorous vacuole

Within a red blood cell, P. falciparum resides inside the parasitophorous vacuole. This is formed during erythrocyte invasion.

The proteins originating in the parasite pass through the membrane of the parasitophorous vacuole and are transported to the cytoplasm or membrane of the erythrocyte.^[78] Although this transport mechanism is largely unknown some details have been elucidated.^[79] Ingestion of the erythrocyte cytoplasm begins in mid-ring-stage parasites. Host cytoplasm is internalised via cytostome-derived invaginations and then concentrated into several acidified peripheral structures. Haemoglobin digestion and haemozoin formation occur within these vesicles. The ring-stage parasites can adopt a deeply invaginated cup shape, but they do not take up haemoglobin via macropinocytosis. As the parasite matures the haemozoin containing compartments coalesce to form a single acidic digestive vacuole (pH 4.5 - 5.5) that is then fed by haemoglobin containing vesicles. Some haemoglobin degradation also occurs in compartments outside the digestive vacuole.

The enzyme phosphatidyl-inositol-3-kinase (PI3K) has been implicated in this process.^[80] PI3K is located in vesicular compartments near the membrane and in the digestive vacuole and is involved in endocytosis from the host and trafficking of hemoglobin in the parasite. Its inhibition with wortmannin or LY294002 results in entrapment of hemoglobin in vesicles within the parasite cytoplasm preventing its transport to the digestive vacuole.

The pH of the digestive vacuole is maintained by a V-type H(+)-ATPase.

A signal sequence at the N terminal of proteins targeted to the arasitophorous vacuole has been identified.^[81] The signal appears to reside in the 55 amino acids of the N terminal of the protein. There may be a retention signal at the C terminal.

The micronemal protease ROM1 appears to be essential for proper parasitophorous vacuole modification to allow parasite development.^[82] This protease is able to cleave the proteins AMA1 and MAEBL.

Apicoplast

Plasmodium falciparum, and most other members of the phylum Apicomplexa, contain an organelle termed an apicoplast.^[78] The apicoplast is an essential plastid, homologous to a chloroplast, although the apicoplast itself lacks any photosynthetic function. Evolutionarily it is thought to have been derived through secondary endosymbiosis. As humans do not harbor apicoplasts, this organelle and its constituents are seen as a possible target for antimalarial drugs.

It contains a 35-kb genome, which encodes for 30 proteins. The genome of this organelle has now been sequenced for several species.^[83] It appears to be conserved and to encode ~30 genes in all species examined.

The plastid genome replicates at the late trophozoite stage of the parasite intraerythocytic cycle. It proceeds predominantly via a D-loop/bi-directional ori mechanism with replication ori localized within the inverted repeat region. The process of replication involves a nuclear-encoded DnaJ homolog that binds to the ori site.^[84]

The DNA polymerase involved in the replication of its genome is Pfprex (Klenow-like polymerase). This enzyme has been cloned, expressed and purified.^[85] The enzymes is relatively error prone and shows a bias toward T->C mutations.

Other nuclear encoded proteins are transported into the apicoplast. Transport into the apicoplast are not well understood. These proteins has a signal in the N terminal but unlike many other organisms this appears to be a disordered chain rather than a conserved sequence.^[86] It was thought that a specific signal peptide was responsible for this targeting ^[78] and it was estimated that 551, or roughly 10%, of the predicted nuclear-encoded proteins are targeted to the apicoplast. This hypothesis now appears to be incorrect. It appears that a relative enrichment within the protein of positively charged amino acid residues (Arginine, Histidine, Lysine) particularly at the N terminal of the protein may be sufficient to target the protein to the apicomplast.^[87]

The biosynthesis of this organelle is not well understood. Phosphatidylinositol 3-monophosphate has been shown to be involved in its biosynthesis in the apicomplexian Toxoplasma gondii.^[88] It seems likely that this enzyme is involved in the formation of this organelle in the Plasmodium species also.

This organelle appears to be essential in the liver stages.^[89]

The functions of this organelle remains to be fully determined but it appears to be involved in the metabolism of fatty acids, isoprenoids and heme.^[78] There are two pathways for protein lipoylation in Plasmodium - one in the mitochondrion and the other in the apicoplast. The apicoplast pathway is not found in the vertebrate host and relies on de novo lipoic acid synthesis.^[90]

The role of the apicoplast in the blood stages has been clarified.^[91] Inhibition of isoprenoid precursor biosynthesis with the antibiotic fosmidomycin (an inhibitor of the enzyme DOXP reductoisomerase) causes delayed death in this parasite. This effect can be overcome with the addition of isopentenyl pyrophosphate (IPP) to the culture medium. Continued culture in the presence of this agent leads to the loss of the apicoplast genome and these mutants fail to process or localize organelle proteins. These auxotrophs can be grown indefinitely in asexual blood stage culture but are entirely dependent on exogenous IPP for survival.

Iron-sulphur prosthetic groups are assembled in this organelle.^[92] One component (SufB) is encoded in the apicoplast genome and a second (SufC) is encoded in the nucleus. SufB also exhibits ATPase activity. Other pathways that have been linked to this organelle include biosynthesis of isoprenoid precursors, fatty acids, heme and lipoic acid.^[93]

A gene ''Plasmodium''-specific Apicoplast protein for Liver Merozoite formation (PALM) has been shown to be important for merozoite formation.^[94] Knock out mutants are unable to release merozoites into the blood from the liver stages. Mutants lacking this gene appear to be able to elicit at least temporary immunity.

Falcilysin a zinc metalloprotease is found in the apicoplast.^[95] It is a member of the M16 protease group and has maximal activity at neutral pH. It appears to be an essential gene. Its function in this organelle is not quite clear but it appears to be involved in the degradation of transit peptides.

The enzyme thioredoxin peroxidase is found in the apicoplast, the mitochondrion and the cytosol.^[96]

Autophagy is membrane-mediated degradation process that involves a series of proteins known as Atg proteins. Atg8 is expressed during development and localises to the apicoplast.^[97]

Two C3 sugar phosphate transporter are present in the membrane of the apicoplast of Plasmodium berghei - triose phosphate transporter and phosphoenolpyruvate transporter.^[98] Knock out mutants of the triose phosphate transporter fail to survive. Phosphoenolpyruvate transporter knock out mutants survive in the blood stages but suffer defects during maturation. These latter mutants also do not survive in the liver or mosquito stages.

An ATP dependent caseinolytic protease (ClpP) is present in the apicomplast. Its function is currently unknown.^[99]

Mitochondrion

Plasmodium lacks mitochondrial pyruvate dehydrogenase^[100][101] and the hydrogen ion translocating NADH dehydrogenase (Complex I, NDH1). The mitochondrion contains a minimal DNA genome (~6 kilobases) and carries out oxidative phosphorylation in the insect vector stages by using 2-oxoglutarate as an alternative means of entry into the tricarboxylic acid cycle and a single-subunit flavoprotein as an alternative NADH dehydrogenase (NDH2). In the blood stages mitochondrial enzymes are down regulated and parasite energy metabolism relies mainly on glycolysis. The enzyme malate quinone oxidoreductase was acquired from an epsilon proteobacteria via lateral gene transfer. This transfer occurred in an ancestor of the Apicomplexa.

The ATP synthase is localised to the mitochondrion, is assembled as a large dimeric complex and appears to be essential for in the blood stages of the life cycle.^[102] Its function in these stages is not yet clear.

The sulfhydryl:cytochrome c oxidoreductase Erv1/ALR/GFER/HSS (Essential for Respiration and Vegatative growth/Augmenter of Liver Regeneration/Growth Factor Erv1-like/Hepatic regenerative Stimulation Substance/hepatopoietin) is an essential sulfhydryl oxidase for required oxidative protein import into the mitochondrial intermembrane space. It is one of several enzymes involved in electron transferase activity. It is encoded by all eukaryotes and cytoplasmic DNA viruses sequenced to date. The enzyme from P. falciparum differs significantly from that found in yeast and humans with altered cysteine motifs and intermolecular disulfide bonds.^[103] Despite successful cloning and expression in yeast, the parasite enzyme fails to function in yeast. A second related enzyme - Mia40 - does not appear to be present in P. falciparum.

Deletion of the gene in the rodent parasite Plasmodium berghei for the flavoprotein subunit of succinate dehydrogenase - part of the complex II - showed impairment of ookinete function and oocyst formation.^[104]

The gene for the flavoprotein subunit of succinate dehydrogenase can be disrupted in the parasite.^[105] Its disruption causes growth retardation of the intraerythrocytic forms. It appears that complex II functions as a quinol-fumarate reductase to form succinate from fumarate in the intraerythrocytic parasite.

The dicarboxylate-tricarboxylate carrier homolog has been cloned from P. falciparum.^[106] This protein may mediate the oxoglutarate-malate exchange across the inner mitochondrial membrane required for the branched pathway of tricarboxylic acid metabolism.

The ClpQ protease and ClpY ATPase have been cloned.^[107] ClpQY function disruption caused hindrance in the parasite growth and maturation of asexual stages of parasites. Features of apoptosis like cell death are also found.

The mitochondrial pathway of protein lipoylation relies on scavenging from the host and can be inhibited with the lipoic acid analog 8-bromo-octanoic acid. Use of this agent inhibits growth and significantly reduces merosome formation. Schizogony is the phase most affected by this inhibition.^[90]

Atovaquone, a 2-hydroxynaphthoquinone, is a competitive inhibitor of the quinol oxidation site of the mitochondrial cytochrome bc1 complex and is used as an antimalaria agent.^[108] Inhibition of this enzyme leads to the collapse of the mitochondrial membrane potential and disruption of pyrimidine biosynthesis. These effects are lethal to the parasite.

The mitochondrial RNA polymerase appears to be an essential gene for the erythrocytic stages.^[109]

Over half the genome of the mitochondrion encodes the genes for three classic mitochondrial proteins: cytochrome oxidase subunits I and III and apocytochrome b.^[110] The remainder encodes 34 RNA genes of which 27 have been assigned to ribosomal RNA (12 to the small subunit and 15 to the large subunit). These genes are fragmented and are encoded on both strands.

The mitochondrial thioredoxin peroxidase-2 does not appear to be essential.^[111]

Digestive vacuole

During growth of the parasite and as part of its digestion of the erythrocyte's haemoglobin, fusion of digestive vesicles occurs and gives rise to a large digestive vacuole.^[112] This vacuole the interior of which is maintained at a low pH (pH 4.5 - 5.5), processes 60-80% of the ingested hemoglobin and provides a pool of amino acids that is crucial for parasite growth and development. The membrane contains ion pumps and transporters that maintain its low pH. During haemoglobin digestion the heme is released from hemoglobin. Haem is toxic to the parasite and is detoxified by biocrystallization to hemozoin within the vacuole. Quinoline drugs, including chloroquine, act by binding to heme and thus prevent its sequestration into hemozoin.

It has been shown that micromolar concentrations of chloroquine partially permeabilized the parasite's digestive vacuole membrane and that this event appears to precede mitochondrial dysfunction.^[113]

Quinine has been shown localise to a non acidic compartment within the digestive vacuole.^[114] It may colocate with haemozoin. It's localisation within the parasite is not altered by the presence or absence of a functional multidrug resistance gene.

The digestive vacuole is able to activate both the alternative complement and the intrinsic clotting pathway.^[115] The digestive vacuole membrane has the capacity to assemble prothrombinase, a key enzyme of the intrinsic clotting pathway.^[115] The capacity of this membrane to activate both complement and coagulation can be suppressed by low molecular weight dextran sulfate. Phagocytosis of these membranes drives the polymorphonucleocytes into a state of functional exhaustion.

Two multi-spanning digestive vacuole membrane proteins are known: the multidrug resistance protein 1 and Chloroquine Resistance Transporter (CRT).^[116] The CRT protein moves from the endoplasmic reticulum to the Golgi apparatus before becoming associated with the digestive vacuole. The digestive vacuole forms in the ring stages of the parasites life cycle. Chloroquine sensitivity is not influenced by the absence of CRT from the digestive vacuole bringing into question its relationship (if any) to chloroquine resistance.^[116] Mutations in the CRT gene have been associated with sensitivity and resistance to quinolines.^[117] In particular a lysine to isoleucine at codon 76 (Adenosine -> Thymine at base 227) mutation and a valine to phenylalanine (Guanine -> Thymine at base 1108) mutation have been associated with changes in drug sensitivity.

The membrane of the digestive vacuole is four nanometers in thickness with patches that may be up to 12 nanometers in thickness.^[118]

The digestion of haemoglobin produces large quantities of ferriprotoporphyrin IX which it unable to digest and is potentially toxic to the parasite. To avoid the toxicity the ferriprotoporphyrin is converted to haemozoin. Chloroquine inhibits this process. The mechanisms behind this process are still unclear. In vitro conversion of ferriprotoporphyrin to haemazoin is enhanced at a temperature of 41C when compared to its conversion at 37C.^[119] It is possible that the rise in temperature that occurs in malaria may be part of a strategy to enhance this reaction at the later stages of growth when the ferriprotoporphyrin concentration is likely to be high.

Nucleolus

A hat-like structure polarized towards one side of the nucleus that stains with nucleolar markers has been described^[120] It seems likely that this unusual structure is the nucleolus.

Endoplasmic reticulum

This forms a set of reticular structures adjacent to the nuclear regions during the trophozoite and schizont stages.^[121] In the late schizont stage it forms globular structures surrounding each budding merozoite.

Ribosomes

Unlike other eukaryotes studied to date Plasmodium species have two or three distinct SSU rRNA (18S rRNA) molecules encoded within the genome.^[122] These have been divided into types A, S and O. Type A is expressed in the asexual stages; type S in the sexual and type O only in the oocyte. Type O is only known to occur in Plasmodium vivax at present. The reason for this gene duplication is not known but presumably reflects an adaption to the different environments the parasite lives within.

The Asian simian Plasmodium species - Plasmodium coatneyi, Plasmodium cynomolgi, Plasmodium fragile, Plasmodium inui, Plasmodium fieldi, Plasmodium hylobati and Plasmodium simiovale - have a single single S-type-like gene and several A-type-like genes. Phylogenetic analyses has shown that gene duplication events giving rise to A- and S-type-like sequences took place independently at least three times in the Plasmodium evolution.

The phosphoprotein P0 occurs as a complex with two other small acidic ribosomal proteins (P1 and P2).^[123] A pentameric complex [(P1–P2) P0 (P1–P2)] form the stalk of the large ribosomal subunit, which seems to play a role in the GTPase elongation centre of the ribosome.

The P2 protein is exported to the infected erythrocyte surface at 30 hrs post merozoite invasion, concomitant with extensive oligomerization. It is largely largely composed of alpha helical and random coil domains.^[123]

Sexual differentiation

Sexual parasite development is controlled by a DEAD box RNA helicase of the DDX6 family, termed DOZI.^[4]

The Puf2 gene, a member of the Puf family of transcriptional regulators, has been shown to be involved in gamete formation.^[124]

FACT (facilitates chromatin transcription) is a dimeric complex of two proteins - SPT16 and SSRP1 - which acts as a histone chaperone in the (dis)assembly of nucleosome (and chromatin) structure during transcription and DNA replication.^[125] It is an essential gene in Plasmodium. Changing its promoter to one expressed only in the blood stages leads to changes in the male gametocytes. The mutant gametocytes have delayed DNA replication and gametocyte formation. Male gamete fertility is strongly reduced. Female gametocytes appear to be normal. When successful fertilization is achieved, the ookinetes generate oocysts that arrest early in development and fail to enter sporogony.

The proteins cell division cycle protein 20 and its homologue, CDC20 homologue 1 are central to the cell cycle activating the anaphase-promoting complex/cyclosome (APC/C) in mitosis and facilitating degradation of mitotic APC/C substrates.^[126] A single homolog of this gene has been identified in Plasmodium berghei. It appears to be essential in male gametogensis but not for asexual reproduction. Blockage occurs at the nuclear spindle/kinetochore stage.

A gametocyte development 1 gene (Gdv1) which encodes a perinuclear protein has been identified.^[127] Its mechanism of action is not known. Homologues of this gene have been found in Plasmodium vivax, Plasmodium knowlesi and Plasmodium gallinaceum.

Egress from the erythrocyte

This is an essential step in the life cycle. The calcium dependent protein kinase PfCDPK5 which is expressed in the merozoite is essential for this process.^[128] Deletion mutations of this gene result in cell arrest in the late schizont stages. Merozoites released from these schizonts are capable of invasion.

Large holes appear in the cytoskeleton ~35 hours post invasion.^[129] This occurs at the same time as the loss of cytoskeletal adaptor proteins that are part of the junctional complex, including α/β-adducin and tropomyosin. This is followed by the proteolysis of many cytoskeletal proteins during egress at ~48 hours post infection. This later proteolysis is mediated by the erythrocyte's own calpain-1.

Along with the release from the erythrocyte of the merozoites, the now functionless digestive vacuole is also released. These are can active complement and are rapidly taken up by the polymorphs.^[130] On ingestion the digestive vacuoles induce a vigorous respiratory burst which drives the cells into a state of functional exhaustion, blunting production of reactive oxygen species and microbicidal activity upon challenge with bacterial pathogens.

The serine repeat antigen (SERA) multigene family encode a series of proteins with a putative papain-like cysteine protease motif. One of these SERA5 (120 kiloDaltons) is produced at the late trophozoite/schizont stage. It is secreted together with other SERAs into the parasitophorous vacuole in an infected erythrocyte where it is cleaved into three fragments: an N-terminal domain (47 kDa), a central domain containing putative papain-like cysteine protease motifs (56 kDa) and a C-terminal domain (18 kDa). This N-terminal fragment is then cleaved in turn into two 25 kDa fragments. These fragments become covalently linked to the C-terminal 18 kDa fragment via disulfide bonding and attach to the merozoite surface. The central fragment is further cleaved to 50 kDa and 6 kDa fragments before being shed to the medium. These proteolytic cleavages are carried out by a subtilisin-like serine protease called PfSUB1 and the inhibition of this processing, likewise, results in blockade of merozoite release.^[131] SERA6 may also be involved in schizont rupture and merozoite release from the erythrocyte. Both SERA5 and SERA6 are essential for blood stage parasite viability.^[132] SERA6 is found in parasitophorous vacuole where it is activated by cleavage by the serine protease PfSUB1 just prior to egress. The release of PfSUB1 may be controlled by a calcium flux within the exomemes of the merozoites.^[133] The release may be under the control of a phopholipase C.

A protein - gamete egress and sporozoite traversal - has been identified that appears to be involved in the egress of male and female gametes from the erythrocyte.^[134] It is also involved in sporozoite migration.

Kinases

Although several kinases are known in P. falciparum (~90^[135]) very little is known about them. Several are cyclin dependent kinase kinase like kinases (CLK): of these two - the Lammer kinase homologue PfCLK-1 and PfCLK-2 have been cloned.^[136] CLKs in other eukaryotes are involved in the regulation of mRNA splicing through phosphorylation of serine/arginine-rich proteins. Both are transcribed throughout the asexual blood stages and in gametocytes. PfCLK-1/Lammer possesses two nuclear localization signal sites while PfCLK-2 possesses one of these signal sites upstream of the C-terminal catalytic domains. The two PfCLKs form complexes with proteins with predicted nuclease, phosphatase or helicase functions.

Although the kinases are primarily localized in the parasite nucleus, PfCLK-2 is also present in the cytoplasm. They are important for completion of the asexual replication cycle. Substrates phosphorylated by the PfCLKs include the Sky1p substrate, splicing factor Npl3p, and the plasmodial alternative splicing factor PfASF-1.

Within the genome is a family of four protein kinases (Pfnek-1 to -4) that are related to the NIMA (never-in-mitosis/Aspergillus) family of kinases. The members of this latter family play important roles in mitosis and meiosis. Pfnek-1 (PFL1370w) is expressed in asexual parasites and male gametocytes.^[137] It is an essential gene for completion of the asexual cycle. The other three - Pfnek-2 (PFE1290w), -3 (PFL0080c) and -4 (MAL7P1.100) - are expressed predominantly in gametocytes.

Pfnek-2 is predominantly expressed in gametocytes and is required for DNA replication during meiosis and ookinete development.^[138]

The plasmodial mitogen-activated protein kinase kinase Pfnek-3 has both serine/threonine and tyrosine kinase activities.^[139]

Pfnek-4 is expressed in stage II to V gametocytes and in a subset of asexual stage parasites undergoing schizogony.^[140] It is also required for the completion of meiosis in the ookinete.^[138]

A mitogen activated protein kinase (MAP kinase) gene is located on is located on chromosome 14.^[141] It is predominantly expressed in gametocytes and gametes/zygotes. The protein has 882 amino acid residues and possesses a TDY dual phosphorylation site upstream of the highly conserved VATRWYRAPE sequence within subdomain VIII. Within the carboxyl-terminal segment the protein contains an unusually large and highly charged domain. This region includes two repetitive sequences of either a tetrapeptide or octapeptide motif.

A subgroup of cyclin-dependent kinases (CDK) including crk-5 have an activation loop that contains a novel Proline-Threonine-x-Cytosine motif which is absent from all known CDKs outside the Apicomplexa.^[142]

The protein PFD0975w appears to be homologous with the right open reading frame 2 kinase RIO-2, a kinase involved in ribosome biogenesis and other cell cycle events.^[143] This enzyme is unique among the kinases in the genome because along with the kinase domain, it also has a highly conserved N-terminal winged helix domain.^[144]

The right open reading frame 2 protein kinase may be a potential drug target.^[145]

The protein kinase CK2, a serine/threonine protein kinase, has one catalytic subunit (PfCK2) and two regulatory ones (PfCK2beta1 and PfCK2beta2).^[146] This enzyme is found both in the cytoplasm and the nucleus. Substrates include the nucleosome assembly proteins (Naps), histones and two members of the Alba family. Both of the two regulatory subunits are required for completion of the asexual erythrocytic cycle.

There are at least three adenylate kinases (AK) encoded in the genome - PfAK1, PfAK2 and a GTP:AMP phosphotransferase (PfGAK).^[147] There are two additional adenylate kinase-like proteins - PfAKLP1 (which is homologous to human AK6) and PfAKLP2. PfAK1, PfAKLP1, and PfAKLP2 are found in the cytosol. PfGAK is located in the mitochondrion. PfAK2 is located at the parasitophorous vacuole membrane and this localization is driven by N-myristoylation.

The calcium dependent protein kinases (CDPK) are part of a superfamily found in plants, ciliates and some apicomplexa. They are not present in fungi or animals. They have three domains: a variable N-terminal region involved in substrate recognition and protein interaction, a kinase catalytic domain and a regulatory domain. The regulatory domain has two subdomains - an autoinhibitory junction domain and a calmodulin like domain. The calmodulin domain has four EF hands. These hands, upon binding calcium, undergo a structural change that moves the junction domain from its autoinhibitory interaction with the substrate binding site of the kinase domain which in turn activates kinase domain catalytic activity.

In P. falciparum CDPK5 controls parasite egress from host cells.^[148] In P. bergei CDPK3 is essential for the ookinete to traverse the mosquito midgut epithelium^[149] and CDPK4 is involved in development of the male gametocyte.^[150]

The homolog of calcium dependent protein kinase 1 (CDPK1) in Toxoplasma gondii is calcium dependent protein kinase 3 (TgCDPK3). This protein in Toxoplasma is localised to the inner membrane and is not an essential gene.^[151] It is involved in Ca(2+) ionophore control and host cell egress. The role of this protein in Plasmodium is not currently known. It is however expressed and localises with proteins at the perifery of the schizonts and merozoites involved in gliding motility^[152] and can can phosphorylate these proteins.^[153] Inhibition of CDPK 1 is associated with a block in development at the schizont level.^[152] In P bergei CDPK1 regulates transcription of stored mRNA during ookinete development in the mosquito midgut.^[154]

The cyclic guanine monophosphate dependent protein kinase is essential for the initiation of gametogenesis and for blood stage schizont rupture and may also be involved in ookinete differentiation and motility and liver stage schizont development.^[155]

Phosphatases

There are 27 putative protein phosphatases in the genome. These can be classed into groups: phosphoprotein phosphatases, metallo-dependent protein phosphatases, protein tyrosine phosphatases and NLI interacting factor-like phosphatases.^[156]

Cysteine proteases

A number of cysteine proteases have been identified this organism including four falcipains, serine repeat antigens (SERA), dipeptidyl aminopeptidase 1, dipeptidyl aminopeptidase 3 and a calpain homolog.^[157] The falcipains belong to the papain family of enzymes (clan CA).

Falcipain-1 appears to be important in the development of the oocysts in the mosquito.

Falcipain-2 is involved in the hydrolysis of haemoglobin and appears to be a non essential gene. It also promotes host cell rupture by cleaving the skeletal proteins of the erythrocyte membrane.

Falcipain-3 appears to be an essential gene but its function has yet to be firmly established.

Dipeptidyl aminopeptidase 1 is found in the digestive vacuole and is also an essential gene.

Dipeptidyl aminopeptidase 3 appears to be involved in the release of the merozoites from the erythrocyte.

Although most SERAs are cysteine proteases some have serine at the active site. SERA-5 and SERA-6 appear to be essential genes: SERA-5 also seems to be involved with egress of the merozoites.

Metallopeptidases

There is at least one M1 family aminopeptidase in the genome (PfA-M1). This is a zinc binding metalopeptidase with optimal activity at pH 7.4, and remains at least 40% active between pH 5.8-8.6. Immunofluorescence studies have shown that in trophozoites that it diffusely found in the parasite cytoplasm with accumulations outside the digestive vacuole while in schizonts it is progressively located to a vesicle like pattern ending as a single location in released merozoites. It exists as two major isoforms, a nuclear 120 kDa species and a processed species consisting of a complex of 68 and 35 kDa fragments.^[158]

There are at least 2 essential metallopeptidases encoded in the genome - PfA-M1 and Pf-LAP.^[159] Specific inhibition of PfA-M1 causes swelling of the parasite digestive vacuole and prevented proteolysis of haemoglobin derived oligopeptides. This inhibition is lethal to the parasite probably by starvation. Inhibition of Pf-LAP is lethal to parasites early in the life cycle, prior to the onset of haemoglobin degradation suggesting a different role for this enzyme.

Falcilysin a zinc metalloprotease found in the apicoplast.^[95] It is a member of the M16 protease group and has maximal activity at neutral pH. It appears to be an essential gene. Its function in this organelle is not quite clear but it appears to be involved in the degradation of transit peptides.

M18 AAP is a metallo-aminopeptidase that has a highly restricted specificity for peptides with an N-terminal glutamine or asparagine residue.^[160]

Aspartly proteases

At least four aspartyl proteases known as plasmepsins are involved in the degradation of hemoglobin by Plasmodium falciparum.^[161]

The histo-aspartic protease (HAP) has been crystallised.^[162] This protein has high sequence similarity to pepsin-like aspartic proteases, but one of the two catalytic aspartates, Asp32, is replaced in this enzyme by a histidine residue. The propeptide interacts with the C-terminal domain of the enzyme, forcing the N- and C- terminal domains apart. This mechanically separates His32 and Asp215 and prevents formation of the mature active site. This mechanism is similar to those of other proplasmepsins. The enzyme has a number of unique features and may be a useful drug target.

There are at least 10 aspartic proteases encoded within the genome. Plasmepsins I, II, IV and histo-aspartic protease are known to be involved in the digestion of haemoglobin.^[163] These four enzymes share 50-79% amino acid sequence identity^[164] and are located on chromosome 14 (gene identifiers PF14_0076, PF14_0077, PF14_0078, and PF14_0075 respectively).^[165] Plasmepsins I and II are present in the food vacuole and make the initial cleavages in the hemoglobin molecule. The proplasmepsins I and II are both type II integral membrane proteins that are transported through the secretory pathway before cleavage to the soluble form. This reaction occurs within the food vacuole and the cleavage occurs immediately after a conserved Leucine-Glycine dipeptidyl motif.^[166] This reaction may be blocked calpain inhibitors. It appears that plasmepsin II and IV are capable of autoactivation as well as activation each other's inactive form.^[167] These two proteins are not glycosylated. Plasmepsin I is synthesized and processed to the mature form soon after the parasite invades the red blood cell, while plasmepsin II synthesis is delayed until later in development.

Plasmepsin V, an integral membrane protein, is located within the endoplasmic reticulum but not in the Golgi apperatus.^[168] The gene is expressed over the course of asexual intraerythrocytic development. The amount of the protein in the parasite is lowest in the ring stage and increases steadily through schizogony. It appears to be involved in the export of proteins to the erythrocyte.

Intramembrane proteases

A presenilin-like signal peptide peptidase is known to be present in the endoplasmic reticulum.^[169]

At least two rhomboid proteases (ROM1 and ROM4) are present in the genome.^[170] ROM4 appears to be involved in the gliding motility of the sporozoite.

Serine proteases

A SUMO specific protease PfSENP1 (PFL1635w) has been identified in the genome but its importance if any is not known^[171]

There are at least 3 subtilisin like proteases encoded in the genome.^[172] These are serine proteases. One of these (PfSUB3) is expressed at late asexual blood stages. In the merozoites SUB2 has been implicated in shedding of adhesins at a juxtamembrane position.^[173]

Erythrocyte proteins taken up

A small number of erythrocytic proteins are taken up by the parasite during the course of its life cycle. The role these play is not clear. Among these proteins is dematin which interacts with the parasite's 14-3-3 protein.^[174]

The parasite is capable of making use of the erythrocyte's own enzymes. The enzymes PAK1 and MEK1 neither of which are encoded in the Plasmodium genome have been shown to be phosphorylated and activated during the course of infection^[175]' In vitro work has shown that inhibition of these enzymes is fatal to the parasite.

Plasmodium ingests kininogen from which its proteases generate vasoactive peptides.^[176] The role this may play in the pathophysiology of malaria is not yet understood.

Transport/secretion

The uninfected erythrocyte lacks a regulated transport system. Vesicular transport within both the parasite and the infected erythrocyte cytoplasm must be provided by the parasite itself.

Both the cytoplasmic pH (7.3) and the inside-negative plasma membrane potential (-95mV) are kept fairly constant during the intra erythrocytic cycle. This is due to the action of a V-type H(+)-ATPase which is also responsible for the pH of the digestive vacuole.

The intracellular concentration of chloride ions has been estimated to be 48 milliMolar.^[177] It appears to actively import using ATP both hydrogen ions and chloride ions in a linked fashion via a DIDS sensitive transporter in the cytoplasmic membrane.

One difficulty the parasite has in acquiring nutrients from the cytoplasm is the presence of phosphate groups on these molecules. It appears to have overcome this by secreting an acid phosphatase (glideosome-associated protein 50 - GAP50 ) into the cytoplasm that is then taken up into the digestive vacuole.^[178]

The parasite has an absolute requirement for isoleucine - an amino acid absent from human haemoglobin. A saturable neutral amino acid (methionine, leucine, isoleucine) transporter appears to be encoded by the parasite and this protein functions in the infected erythrocyte membrane.^[179]

The P. falciparum Na+/H+ exchanger (PfNHE1) is located on chromosome 13 (gene PF13_0019).^[180] This gene may be involved in resistance to quinine.^[181]

Two folate transporters (PfFT1 and PfTF2) have been cloned.^[182] Substrates include folic acid, folinic acid, the folate precursor pABA and the human folate catabolite pABAG(n). 5-methyl tetrahydofolate is not transported by PfFT1 and only poorly by PfFT2. The activity of both transporters may be inhibited by probenecid or methotrexate.

An intracellular purine permease (PfNT2) has been shown to be localised to the endoplasmic reticulum.^[183] This protein is a member of the equilibrative nucleoside transporter family.

Within the genome are encoded 11 Rab GTPases.^[184] These proteins are typically involved in vesicle transport. Casein kinase-1 has been shown to interact with Rab5B and the catalytic subunit of cAMP-dependent protein kinase A interacts with Rab5A and Rab7.

Many proteins in the genome carry a host targeting signal.^[185] This signal sequence is recognised by phosphatidylinositol-3-phosphate in the endoplasmic reticulum.

The thioredoxin 2 protein is part of the multi-protein complex embedded within the parasitophorous vacuolar membrane and is thought to be involved in protein secretion.^[186]

A protein - Armadillo Repeats-Only - has been localised to the cytosolic face of the rhoptries.^[187] A putative signal sequence in the first 20 amino acids has also been identified.

The parasite possesses its own equilibrative nucleoside transporter 1. All members of this protein family have 11 transmembrane segments. The gene product is located in the parasite's plasma membrane and knock out mutants have shown that this is an essential gene at least at physiological concentrations. In the 11th transmembrane segment two mutations have been shown to affect its activity: a phenylalanine (Phe) to leucine (Leu) at residue 394 (F394L) via cytosine (C) or uracil (U) to adenosine (A) or guanine (G) at the third codon position and a cysteine (Cys) to glycine (Gly) mutation at either glycine in a conserved glycine-X-X-glycine motif (where X is any amino acid) via a cytosine to uracil at the second codon position.^[188] Additional work suggests that the 11th transmembrane segment is largely alpha helical. It has been suggested that this transmembrane segment may be the actual purine transport channel.

Within the genome there are encoded four equilibrative nucleoside transporters (ENTs). ENT 1 is the major route of purine nucleoside/nucleobase transport in the erythrocytic stages. Knock out mutants have been generated that can survive. ENT4 has been cloned and expressed.^[189] It does not appear to transport either hypoxanthine or adenine monophosphate but does transport adenine and 2'-deoxyadenosine. It is inhibited by dipyridamole.

The parasite is unable to synthesize purines (including adenosine, hypoxanthine and adenine) and must take these up from the host. Three purine transporters have been studied: the human equilibrative nucleoside transporter (hENT1), the human facilitative nucleobase transporter (hFNT1) and the parasite-induced new permeation pathway (NPP). The bulk of transport is facilitated by host's own transporters rather than through the NPP.^[190] Hypoxanthine and adenine were transported mainly through the hFNT1 pathway whereas adenosine entered predominantly through the hENT1 system. The rate of purine uptake in infected cells was approximately twice that of uninfected erythrocytes. The rate of adenosine uptake was greater than the rate of hypoxanthine uptake in infected human red blood cells. Furosemide inhibits the transport of purine bases through the hFNT1.

The parasite can uptake polyamines from the host. Two of these - putrescine and spermidine - are taken up in a temperature, pH and membrane potential dependent mechanism.^[191]

The clag3 genes on chromosome 3 appear to be involved in anion transport rather than in cell adherence as originally thought.^[192]

The clag3 gene family encode a parasite ion channel known as the plasmodial surface anion channel. Its activation appears to involve an intracellular domain.^[193]

At least two of the clag3 genes appear to be involved in the surface anion channel which functions in nutrient uptake.^[194]

The plasma membrane protein aquaglyceroporin mediates the transport of both glycerol and water.^[195]

An ATP-binding cassette (ABC) transporter encoded by the gene Pf14_0244 (PfABCG2) on chromosome 14 appears to have some role in the asexual stages, gametocyte stages and in the oocyst.^[196]

A copper transport protein (PF14_0369) has been identified^[197] This protein is expressed in early ring stage and translocating from the erythrocyte plasma membrane to a parasite membrane as the parasites developed to schizonts. Inhibition of copper uptake with neocuproine inhibits the ring to trophozoite transition.

Protein export

There is at least one protein export system in the parasite.^[198] Several proteins are known to be involved in this process: HSP101 (a AAA+ ATPase), a protein of no known function termed PTEX150, and the apparent membrane component EXP2 - all of which are located within the dense granules of the merozoites.

For proteins destined for the erythrocyte a motif is known - RxLxE/D/Q Arginine - any amino acid - Leucine - any amino acid - Aspartic acid / Glutamic acid / Glutamine.^[185] Proteins with this motif bind to the lipid phosphatidylinositol 3-phosphate within the endoplasmic reticulum (ER). Cleavage of the motif results in the release of the protein from the ER membrane. Cleavage within the ER is carried out by the aspartyl protease plasmepsin V.^[199] The peptide chain is cleaved between the RxL and xE/Q/D submotifs.^[200] The xE/Q/D submotif then acts as an export signal to the erythrocyte. Mutation of the arginine to alanine results in the loss of binding to phosphatidylinositol 3-phosphate and an approximately threefold reduction in the export rate of the protein to the erythrocyte.

This motif also appears to be present in oomycetes.^[201]

Protein export appears to be complex. It has been thought that protein export depended at least in part on the presence of a Plasmodium export element (PEXEL) within the protein. This does not actually appear to be the case.^[202]

Actin

Within the genome are encoded two forms of the protein actin - I and II. The first form (I) is present in significantly greater quantities. Actin II appears to be essential for the process of exflagellation.^[203] Deletion of this gene results in viable asexual stages. During the formation of the male gametes actin I is found initially in both the nucleus and the cytoplasm. After activation it is found only in the cytoplasm. In actin II deletion mutants actin I remains in both the nucleus and the cytoplasm after activation. Morphologically in the actin II mutants male gametocyte DNA was replicates normally and axonemes are assembled but egress from the host cell is inhibited and axoneme motility is abolished.

Two proteins P. falciparum actin-depolymerizing factor 1 (PfADF1) and P. falciparum actin-depolymerizing factor 2 (PfADF2) are involved in the polymerisation of actin.^[204] PfADF1 has ben crystallised and despite having significant differences from other proteins with similar function it is capable of severing actin filaments. PfADF2, like canonical ADF proteins but unlike ADF1, binds to both globular and filamentous actin, severing the filaments and inducing nucleotide exchange on the actin monomer.^[205] The crystal structure of PfADF1 shows major differences from the ADF consensus, explaining the lack of F-actin binding. PfADF2 structurally resembles the canonical members of the ADF/cofilin family.

The actins found In Plasmodium and in Toxoplasma are divergent both in sequence and function and only form short, unstable filaments in contrast to the stability of conventional actin filaments.^[206] This inherent instability of parasite's actin filaments is a critical adaptation for their gliding motility.

Actin is involved in the expression of the var genes.^[207] The var introns interact with an 18 base pair nuclear protein binding element which recruits actin and repositions the var DNA from a transcriptionally repressive to a transcriptionally active perinuclear compartment.

The presence of actin microfilaments has been demonstrated in the ookinete in the pellicle and in the apices.^[208]

There are two formin genes encoded in the genome.^[209] These associate with and nucleate both mammalian and Plasmodium actin filaments. Another gene profilin - also encoded in the genome but only as a single copy - sequesters actin monomers preventing their polymerisation.

Aldolase, an actin binding protein, is involved in the moving junction that forms during the invasion of the erythrocyte.^[210]

Several hemoglobinopathies that protect carriers from severe malaria may do so by interfering with host actin reorganization.^[66]

The cyclase associated proteins are among the most highly conserved regulators of actin dynamics. They catalyze nucleotide exchange on actin monomers from ADP to ATP and recycle actin monomers from ADF/cofilin for new rounds of filament assembly. The Plasmodium falciparum cyclase associated protein is entirely composed of β-sheet domains and efficiently promotes nucleotide exchange on actin monomers.^[211]

Motility

Thrombospondin Related Anonymous Protein (TRAP) is a type I transmembrane proteins which has several extracellular adhesive domains and a cytoplasmic domain that recruits the glycolytic enzyme aldolase. Normally only small amount of TRAP found on the sporozoite surface. TRAP is involved in cell motility.^[212]

Its' tandem von Willebrand factor A and thrombospondin type I repeat domains connect through the proline rich stalk, transmembrane and cytoplasmic domains to the parasite's actin dependent motility apparatus.^[213] Binding is dependent on the presence of a metal ion. The protein is capable of considerable conformational changes.

The cytoplasmic domain binds to F-actin which connects to myosin A. Within the transmembrane domain it has a canonical rhomboid cleavage site (Ala-Gly-Gly-Ile-Ile-Gly-Gly). Rhomboid proteases are a family of serine proteases that require helical instability in the transmembrane domain and have specific residue requirements in their P1, P4 and P2′ positions. These proteases are responsible for intramembraneous cleavage.

TRAP binds to receptors on the host and is translocated posteriorly by the actomyosin motor. It is then normally cleaved by a calcium independent serine protease. Removal of the cytoplasmic domain abolish the motility of the parasite. Mutations in the rhomboid cleavage site are defective in TRAP shedding and display slow, staccato motility and reduced infectivity.^[214] The reduction in infectivity is particularly marked if the sporozoites are inoculated intradermally rather than intravascularly. Prevention of cleavage of the TRAP protein entirely renders the sporozoites uninfectious and immobile. The rhomboid protease normally involved in TRAP cleavage appears to be the ROM4 protease. This protease is found across the entire sporozoite surface suggesting it has functions in addition to TRAP cleavage.

The circumsporozoite- and thrombospondin-related adhesive protein (CTRP) is a modular multidomain protein containing six tandem von Willebrand factor A like domains and seven tandem thrombospondin type I repeat-like domains.^[215] The A domains of CTRP are critical for ookinete gliding motility and oocyst formation. The thrombospondin domains are fully redundant.

The cell-traversal protein for ookinetes and sporozoites (CelTOS) is a protein involved in the invasion of both vertebrate and insect host cells.^[216]

The C-terminal tail of myosin A (MyoA) and its light chain, myosin A tail domain interacting protein (MTIP) are essential parts of the gliding motility apperatus.^[217]

Dynein light chain 8 is present in P. falciparum as a homodimer.^[218] The dimer is formed by the interaction of the β₀ chains on one molecule with the β₂ chains of the second.

Surface exposed proteins

Enolase is bound to the surface of P. falciparum and several other pathogens.^[219] In this location it binds plasminogen which is thought to function in the degradation of the extracellular matrix surrounding the targeted host cell, thereby facilitating pathogen invasion.

The ETRAMP family is characterized by a predicted signal peptide, a short lysine rich stretch, an internal transmembrane domain and a highly charged C-terminal region of variable length. The highly charged terminal region appears to be involved in protein-protein interactions.^[220] The gene ETRAMP 10.3 has been shown to be expressed in the liver, sporozoites and blood stages.^[221] Within the liver and blood stages it is localized to the parasitophorous vacuole membrane. It is also exported to the erythrocyte during the blood stages. It appears to be an essential gene in the blood stages.^[222]

The receptor for the attachment protein PfRh4 has been identified as complement receptor 1.^[223]

Merozoite surface protein 7 appears to enhance the virulence of the parasite at least in the rodent.^[224]

Merozoite surface protein 2 is one of the most abundant proteins on the surface of merozoites, is intrinsically unstructured and forms amyloid-like fibrils in solution.

The mature parasite-infected erythrocyte surface antigen (MESA) is exported to the erythrocyte cytoplasm where it binds to the N-terminal 30 kiloDalton domain of the erythrocyte protein 4.1R via a 19-residue sequence.^[225] This sequence is also found in a number of other proteins in the parasite. Their role in remodeling of the erythrocyte are still under investigation.

The proteins Pf12, Pf34, Pf92 and Pf38 are associated with detergent resistant membrane microdomains through glycosylphosphatidylinositol anchor sequences.^[226] These microdomains are considered organizing centers for the assembly of molecules implicated in cell signaling.

Positive diversifying selection is present in clag2, clag8 and clag9 but not in clag3.1 and clag3.2.^[227]

A protein PfMSPDBL1 (encoded by PF10_0348 gene) that is a member of the MSP3 family and has both Duffy binding-like (DBL) domain and secreted polymorphic antigen associated with merozoites (SPAM) domain appears to be critical for erythrocyte invasion.^[228]

The cleavage of MSP 1 appears to involve a purinergic signalling pathway.^[229]

The MSP1 protein binds the pro inflammatory protein S100P.^[230] This binding appears to prevent the usual NFκB activation in monocytes and chemotaxis in neutrophils. S100P appears to be able to bind to at least 2 alleles of MSP1 which are separated by at least 27 million years of evolution suggesting that this inhibition mechanism may also be of considerable age.

The merozoite specific thrombospondin related anonymous protein (MTRAP) is thought to be released from the micronemes during merozoite invasion and mediates motility and host cell invasion through an interaction with aldolase.^[210] MTRAP is a highly extended bifunctional protein that binds to an erythrocyte receptor and the merozoite motor. MTRAP specific antibodies fail to inhibit parasite development in vitro.

Thrombospondin related apical membrane protein (PTRAMP) is a surface exposed protein whose function is currently unknown.^[210]

The gene PFE0565w is transcribed in both the erythrocytic and sporozoite stages.^[231] The protein is only expressed in the salivary gland sporozoite stage.

The circumsporozoite protein has been shown to be an inhibitor of the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB).^[232] Its nuclear localization signal alone is sufficient to block NF-κB activation.

A group of proteins known as the 6-cys domain proteins - so called because they contain modules with six characteristic cysteines forming three intra-molecular disulphide bonds between C1 and C2, C3 and C6, and C4 and C5 - are surface exposed proteins.^[233] The first P12 - named after the clone it was isolated from - was described in 1990.^[234] There are at least nine members of the 6-cys family. Most family members contain two 6-cys modules, but up to seven modules can be found in a single protein, in addition to incomplete modules containing fewer cysteine residues. About half of the 6-cys family members characterised to date possess glycosylphosphatidylinositol (GPI) moieties that anchor them to the outer leaflet of the plasma membrane, while those that lack GPI-anchors presumably remain associated with the parasite surface via interactions with other membrane proteins. Of this family P12, P38 and P41 are blood stage antigens. P230 and P48/45 - another two members of this family - are expressed on the surface of gametes.

There are a family of LCCL/lectin adhesive-like protein (LAP) proteins encoded in the genome.^[235] The six members are expressed in gametocytes and form a multi-protein complex.

The 60S stalk ribosomal acidic protein P2 (gene PFC0400w) as well as forming part of the ribosome complex is surface exposed where it forms homo-tetramers.^[123] This protein is exported to the erythrocyte surface 26-28 post invasion and persists there for 6–8 hours. Treatment with antiP2 antibodies causes mitotic arrest at the first nuclear division and disruption of the tubovesicular network which is set up during the trophozoite stages. Removal of the antibodies al lows the reformation of the tubovesicular network and mitotic division to continue.

The phosphoprotein P0 is surface exposed during the asexual erythrocytic stages and antibodies to this protein appear to be protective.^[123] It is also present on the surface of the merozoites.

The Ring-Infected Erythrocyte Surface Antigen (RESA/Pf155) protein appears to affect the mobility of the erythrocyte membrane.^[236]

The circumsporozoite protein forms a dense coat on the sporozoite's surface.^[237] It consists of approximately 400 amino acids organized into three domains: an N-terminal domain containing a conserved pentapeptide (region I), a highly repetitive species specific central domain and a C-terminal domain containing a second conserved sequence (region II). It is involved in invasion of the mosquito's salivary glands and the binding sporozoites to liver cells.

Ubiquitin

The addition of the small protein ubiquitin to other proteins as part of post translational processing is widespread in most eukaryotes. This is also the case with P. falciparum where this process occurs at all stages of the asexual life cycle.^[238] Ubiquitylation involves the covalent attachment of a ubiquitin moiety to lysine residues of protein substrates via the hierarchical intervention of an E1 ubiquitin activating enzyme, an E2 ubiquitin conjugating enzyme, and an E3 ubiquitin ligase that is usually involved in specific substrate recognition

Ubiquitylation is involved in removing misfolded proteins from the endoplasmic reticulum - a process known as Endoplasmic-reticulum-associated protein degradation. This process is a prerequisite for subsequent retro-translocation to the cytosol and destruction by the 26S proteasome. Aberrant proteins are recognized by endoplasmic reticulum luminal chaperone proteins and protein disulfide isomerases to help discriminate properly folded proteins from misfolded proteins. Misfolded proteins are shuttled to the DER1 translocon complex which forms a hydrophobic pore to allow the retro-translocation of proteins through the endoplasmic reticulum membrane. Several components of the system are known to be present in the parasite: HRD1 (E3 ubiquitin ligase), UBC (E2 ubiquitin conjugating enzyme) and UBA1 (E1 ubiquitin activating enzyme).^[121] HRD1 localizes to the endoplasmic reticulum membranes, while UBC and UBA1 localize to the cytosol. HRD1 interacts with membrane bound proteins needed for retro-translocation and helps form the hydrophobic pore complex. Another member of this pathway is the signal peptide peptidase.

The genes PFL1245w is the E1 ubiquitin activating enzyme, PFL0190w is the E2 ubiquitin conjugating enzyme and PF14_0215 is the E3 ubiquitin ligase. PFL1245w (E1) contains a ubiquitin activating enzyme active site, two ubiquitin like activating enzyme catalytic domains, two ThiF repeats and a catalytic cysteine at the N-terminal end. PFL0190w (E2) is 147 amino acid residues in length and contains an ubiquitin conjugating enzyme domain takes up almost its whole length. PF14_0215 (E3) has multiple transmembrane domains, an E3 RING zinc finger (zf-C3HC4) domain on its C-terminal half and a predicted signal peptide consistent with endoplasmic reticulum targeting. The presence of four transmembrane domains is compatible with a pore forming ability and to be able to participate in the recognition and translocation of misfolded proteins across the endoplasmic reticulum membrane.

Autophagy

A number of proteins involved in autophagy are known to be present in the genome. These include Atg8 and Atg3.^[239] The functions of these proteins in the parasite are still being elucidated.

Gene regulation

CCR4-associated factor 1 is involved in the regulation more than 1000 genes during malaria parasite's intraerythrocytic stages.^[240] Mutations in this gene result in mistimed expression, aberrant accumulation and localization of proteins involved in parasite egress and invasion of new host cells. This leads to the premature release of predominantly half-finished merozoites in turn drastically reducing the intraerythrocytic growth rate of the parasite.

A homolog of the DEAD box (Asparagine-Glutamate-Alanine-Asparagine) RNA helicase DDX19 (Dbp5) has been cloned from the P. falciparum genome and has been termed PfD66.^[241] This protein has intrinsic nucleic acid dependent ATPase and RNA binding activities and ATP dependent bipolar DNA and RNA unwinding activity.

SAP1 has been shown to be involved in the post transcriptional control of liver stage genes.^[242]

The protein PfMyb1 is a transcription factor belonging to the tryptophan cluster family.^[243] Inhibition of this gene reduces growth by ~40% with the mortality being concentrated at the trophozoite-schizont interface.

During the life cycle the telomeres and telomere associated repeat elements are transcribed as long non coding RNAs.^[244] They are transcribed by RNA polymerase II as single-stranded molecules. In the ring stage, these transcripts are located in a single perinuclear compartment that does not co-localize with any known nuclear subcompartment. During the schizont stage they are found at several nuclear foci. At least some of these can form stable and repetitive hairpin structure that is able to bind histones. Their function requires further elucidation.

Heat shock proteins

A number of heat shock proteins 40 (hsp40) have been predicted from the sequenced genome. Only one is predicted to be a cytosolic canonical Hsp40 capable of interacting with the major cytosolic Hsp70 an interaction that has been confirmed experimentally.^[245]

PfGECO is a type IV heat shock protein 40 expressed in gametocyte stages I to IV and is exported to the erythrocyte cytoplasm.^[246] This gene appears to be non essential.

Heat shock proteins Hsp70 and Hsp90 are both expressed in P. falciparum. They are linked by an essential adaptor protein known as the Hsp70-Hsp90 organising protein (Hop). This protein co-localises with PfHsp70 and PfHsp90 at the trophozoite stage and forms a complex with them.^[247]

The protein Aha1 interacts with HSP90.^[248]

Heat shock protein 20 has been shown to have a critical role in sporozoite motility.^[249] This role appears to be via substrate adhesion.

A Hsp40 class of chaperone (PFB0090c; PF3D7_0201800; KAHsp40) is located in a chromosomal cluster together with knob components KAHRP and PfEMP3.^[250] This protein has a PEXEL motif required for transport to the erythrocyte compartment. It occurs in punctuate spots in the erythrocyte periphery, distinctly from Maurer's clefts. These structures may be knobs particularly since it is found in a complex the known knob proteins KAHRP, PfEMP3 and Hsp101.

Molecular biology

There are two translation elongation factor G proteins encoded in the genome.^[251] One is located in the mitochondrion and the second in the plastid. Both appear to be inhibitable with fusidic acid

Polynucleotide kinase/phosphatase (PNKP) is a bifunctional enzyme that can phosphorylate the 5'-OH termini and dephosphorylate the 3'-phosphate termini of DNA. It is a DNA repair enzyme involved in the processing of strand break termini, which permits subsequent repair proteins to replace missing nucleotides and rejoin broken strands. A P. falciparum gene encoding a protein with 24% homology to human PNKP has been cloned.^[252] This enzyme dephosphorylates single-stranded substrates or double-stranded substrates with a short 3'-single-stranded overhang, but not double-stranded substrates that mimicked single-strand breaks.

Sir2A is a member of the sirtuin family of nicotinamide adenine dinucleotide dependent deacetylases. In P. falciparum it has been has been shown to regulate the expression of surface antigens to evade the detection by host immune surveillance.^[253] While it is a poor deacetylator of histones it also catalyzes the hydrolysis of medium and long chain fatty acyl groups from lysine residues. Proteins are present in P. falciparum with these modifications and these can be removed by can be removed by PfSir2A in vitro. This suggests that this may be its role rather than the deacetylation of histones.

The telomerase (tert) is a large protein (2518 codons) and has a predicted molecular weight of ~280 kiloDaltons.^[120] It has the usual telomerase specific motifs within the N-terminal half of the protein (GQ/N, CP, QFP and T) and reverse transcriptase (RT) specific motifs in the C-terminal half. The N-terminal half is required for efficient binding of the RNA template, defining the 5′ RNA template boundary, multimerization and interactions with associated proteins. The RT domain is essential for the catalytic activity. The protein contains several nuclear localization signals and is found in the nucleolus.

A histone deacetylase (HDAC1) has been cloned.^[254] The protein has 449 amino acid residues and localises to the nucleus. Its molecular weight is 50 kiloDaltons and it is predominantly expressed in mature asexual blood stages and in gametocytes.

A novel DNA/RNA binding protein PfAlba has been described.^[255] This protein is related to the archaeal protein Alba (Acetylation lowers binding affinity). There are at least four paralogs of the PfAlba gene and these proteins form a complex with the P. falciparum specific TARE6 (Telomere-Associated Repetitive Elements 6) subtelomeric regions. Also associated with the TARE6 regions are PfSir2 a histone deacetylase. In the early blood stages the PfAlba proteins are enriched at the nuclear periphery and associate with the PfSir2 proteins. When the parasite switches from trophozoite to the schizont stage the PfAlba proteins move to the cytoplasm. These proteins will also bind single stranded RNA but the reason for this binding is not known.

A number of novel DNA binding sites have been identified along the genome.^[256] Their function - if any - remains to be determined.

Aminoacyl-tRNA synthetases are required for protein synthesis. Alanine tRNA synthetase, Glycine tRNA synthetase and Threonine tRNA synthetase are dually localised to the cytosol and the apicoplast.^[257] These enzymes do not appear to be present in the mitochondrion.

Tyrosyl tRNA synthetase is secreted by the parasite into the cytoplasm of the infected erythrocyte.^[258] On lysis of the erythrocyte it is released into the blood stream where it is pro inflammatory. It is specifically bound by and taken up by host macrophages and leads to enhanced secretion of the cytokines tumor necrosis factor-alpha and interleukin 6. This interaction also increases the adherence linked host endothelial receptors ICAM-1 and VCAM-1.

The eukaryotic translation initiation factor 2α has a regulatory serine at position 51. This can be phosphorylated by several kinases. Three are known in P falciparum: IK1, IK2 and PK4.^[259] IK1 regulates stress response to amino acid starvation; IK2 inhibits development of malaria sporozoites present in the mosquito salivary glands; and PK4 is essential for the completion of the parasite's erythrocytic cycle.

The centromeres occupy a 4-4.5 kilobase region in each chromosome.^[260] The centromeres cluster to a single nuclear location prior to and during mitosis and cytokinesis but dissociate soon after invasion.

The single stranded DNA binding protein (SSB) plays an important role in all known organisms. A SSB protein is encoded in the genome and localises to the apicoplast.^[261] It forms a homo-tetramer alone and when bound to single stranded DNA. The protein binds 52-65 nucleotides/tetramer.^[262] While similar in its overall structure to that of the SSB of E. coli it differs at the carboxy terminal region. Although it binds single stranded DNA in a similar fashion to the SSB of E. coli it does so with the opposite polarity. There are a number of other functional differences between this protein and that of E. coli. The basis for these differences has yet to be determined.

The RuvB protein belongs to AAA+ family of enzymes which are involved in diverse cellular activities. There are at least 3 copies of this protein in the genome.^[263] RuvB3 possesses the Walker motif A, Walker motif B, sensor I and sensor II conserved motifs similar to yeast and human RuvB like proteins. It has single stranded DNA dependent ATPase activity. The protein is mainly expressed during intraerythrocytic schizont stages and localizes to the nuclear region. In the merozoite the protein relocalizes to the sub nuclear region.

A helicase - PfH45 - of 398 amino acid residues (molecular weight 45 kiloDaltons) is a unique bipolar helicase with both the 3' to 5' and 5' to 3' directional helicase activities.^[264] It is expressed in all the intraerythrocytic developmental stages and has a role in translation.

The transcription factor NF-YB is localised in the nucleus during the erythrocytic stages of the life cycle.^[265] Melatonin and cyclic adenosine monophosphate modulate the expression of NF-YB. NF-YB is also more ubiquitinated in the presence of melatonin.

SET is a conserved nuclear protein involved in chromatin dynamics.^[266] In P falciparum it is expressed in both asexual and sexual blood stages but strongly accumulates in male gametocytes. In P falciparum there are two distinct promoters upstream. he one active in all blood stage parasites while the other active only in gametocytes and in a fraction of schizonts possibly committed to sexual differentiation. In ookinetes both promoters exhibit a basal activity, while in the oocysts the gametocyte-specific promoter is silent and the reporter gene is only transcribed from the constitutive promoter.

A protein (PfSR1) involved in alternative splicing has been described.^[267] It appears that regulation of this gene is essential for the parasite's normal physiology.

A number of the DExD/DExH-box containing pre-mRNA processing proteins (Prps) - PfPrp2p, PfPrp5p, PfPrp16p, PfPrp22p, PfPrp28p, PfPrp43p and PfBrr2p - are present in the genome.^[268] PfPrp16p a helicase and a member of DEAH-box protein family with nine collinear sequence motifs has been cloned. It binds to RNA, hydrolyses ATP and appears to be involved in splicing.

A putative tyrosine site specific recombinase has been isolated.^[269] The N-terminus has the typical alpha helical bundle and potentially a mixed alpha-beta domain resembling that of λ-Int. The C-terminal domain has the putative tyrosine recombinase conserved active site residues Lysine-Histadine-Lysine-(Histadine/Tryptophan)-Tyrosine. The gene is expressed differentially during the erythrocytic stages being maximal in the schizont stage. The open reading frame encodes a ∼57 kiloDalton protein. Knockout mutants are viable and appear normal. DNA binding studies suggest a number of targets include the subtelomeric regions.

Apetala 2 (AP2) family proteins are transcription factors that have DNA-binding domains of ~60 amino acids called AP2 domains. 27 AP2-family genes have been identified in the Plasmodium falciparum genome.^[270] One of these proteins appears to play a critical role in the liver stage development of the parasite.^[271]

Several nucleic acid repair pathways are known. These include the nucleotide excision repair, the mismatch repair, the base excision repair, the double strand break repair and the cross link repair pathways. DNA replication errors - base substitution mismatches and insertion-deletion loops - are primarily corrected by the mismatch repair system.

The DNA helicase II (uvrD) is a superfamily 1A helicase which plays an essential role in the mismatch repair pathway. Homologues of UvrD include the proteins PcrA and Rep. These proteins have a two domain (1 and 2) structure with each domain made of two sub-domains (1A, 1B, 2A and 2B) and a C-terminal extension. They are DNA-dependent ATPases with 3′ to 5′ helicase activity. The helicase activity is located in the N terminal domain.

The UvrD protein of P. falciparum has been cloned.^[263] This gene (PFE0705c) is located on chromosome 5 and contains no introns. It is 4326 bases in length, encodes a protein of 1441 amino acids and has a predicted molecular weight of ~170 kiloDaltons. The two domains and their subdomains are present: The 1A domain is from amino acid 1–722; the 1B domain is from amino acid 150–464; the 2A domain is from amino acid 723–1441; and the 2B domain is from amino acid 896–1359. There is no C-terminal extension. The ATPase and helicase activty are confined to domain 1A and 1B (the N-terminal and first half of the C terminal). It is expressed in the schizont stages of intraerythrocytic development and it colocalizes with PfMLH, a protein involved in mismatch repair. Both PfDH60 - another helicase - and PfMLH are also expressed in schizont stages.

The MutL homolog (MLH) - part of the DNA mismatch repair system - has been cloned.^[272] MLH possess ATPase and endonuclease activities. Its expression is maximal in the schizont stage.

Redox balance

The proteins gamma-glutamylcysteine synthetase and glutathione synthetase which are involved in the synthesis of glutathione appear to be essential genes in P. falciparum.^[273] Inhibition of the glutathione biosynthesis by the parasite is lethal. Its levels appear to be tightly regulated. The enzyme glutathione reductase is highly specific for its substrate glutathione disulfide.^[274]

The thioredoxin system like the glutathione system is responsible for maintaining the redox balance in the cell. The thioredoxin reductase reduces thioredoxin and a number of other low molecular weight compounds.^[274] The other members of this system include five peroxiredoxins differentially located in the cytosol, apicoplast, mitochondria and nucleus with partially overlapping substrate preferences. It also includes members of the thioredoxin superfamily with three thioredoxins, two thioredoxin-like proteins, a dithiol and three monocysteine glutaredoxins and a redox-active plasmoredoxin being encoded in the genome.

The infected host cell is under considerable oxidative stress. Normal erythrocytes have a ratio of reduced (GSH) and oxidized glutathione (GSSG) of 321.6 while the GSH/GSSG ratio in infected cells is 26.7.^[275] The ratio in the parasite is 284.5. Efflux of GSSG from the intact infected cell is more than 60-fold higher than the rate observed in normal erythrocytes. This export process is mediated by permeability pathways that the parasite induces in the erythrocyte's membrane. Exogenous gamma-glutamylcysteine is not converted into GSH in the infected erythrocyte suggesting that the erythrocytes' own GSH synthetase may not be functional. This may be due to the lower levels of magnesium (Mg²⁺) in the infected erythrocyte (0.5 milliMolar) compared to the normal erythrocytes (1.5-3 mM). The lower level of results in cessation of gamma-glutamylcysteine synthesis and of GSH synthesis in the infected erythrocyte. The parasite maintains a level of 4 mM magnesium. The parasite membrane is impermeable to both gamma-glutamylcysteine and GSH.

Glutathione export from parasitized cells is inhibited partially by both the compound MK571 and by furosemide.^[276] These agents are inhibitors of the 'new permeability pathways' induced by the parasite in the host erythrocyte membrane.

Ferriprotoporphyrin IX is released inside the food vacuole of the malaria parasite during the digestion of host cell hemoglobin.^[277] Undegraded ferriprotoporphyrin IX accumulates in the membrane fraction and is degraded by reduced glutathione in a radical mediated mechanism.

Three phosducin-like proteins have been identified in Plasmodium berghei.^[278] Their role in the parasite's metabolism has yet to be clearly established.

The 1-Cys peroxiredoxin enzyme appears to be located in the cytoplasm.^[279]

Within the cytoplasm two peroxiredoxins - T peroxiredoxin-1 and 1-Cys peroxiredoxin - are produced at differing points in the life cycle.^[280] Disruption of the T peroxiredoxin-1 enzymes renders the parasite hypersensitive to heat stress. This does not occur with knock out mutants of 1-Cys peroxiredoxin suggesting that these enzymes have different roles in the life cycle.

Investigation of the liver stages of these enzymes in Plasmodium berghei has shown that both TPx-1 and 1-Cys Prx are present in the cytosol but differ in their expression patterns.^[281] TPx-1 is transcribed shortly after infection of the hepatocyte and expression continues until the schizont stage. Transcription of 1-Cys Prx starts after the parasite has developed into the schizont stage.

Miscellaneous proteins

Polyamine biosynthesis in these parasites is controlled by a unique bifunctional S-adenosylmethionine decarboxylase/ornithine decarboxylase (PfAdoMetDC/ODC).^[282] On the secondary structure level PfAdoMetDC is similar to that of the human protein. This bifunctional enzyme ensure coordination decarboxylated AdoMet and putrescine for the subsequent synthesis of spermidine.

P. falciparum contains both cytosolic and mitochondrial serine hydroxymethyltransferase isoforms.^[283] This is a pyridoxal phosphate dependent enzyme which plays a vital role in the de novo pyrimidine biosynthesis pathway. Both genes are expressed throughout the erythrocytic stages.^[284] Both enzymes appear to be essential.

The first two reactions of the pentose phosphate pathway in P. falciparum are catalysed by a single bifunctional enzyme - glucose 6-phosphate dehydrogenase 6-phosphogluconolactonase.^[285] This is distinct from the case in humans where the enzymes are separate. In animals this pathway is usually found in the cytosol while in plants it is found in the plastids. The location of this reaction is not currently known in P. falciparum.

Fusions between these two enzymes (glucose 6-phosphate dehydrogenase and 6-phosphogluconolactonase) have also been reported in chordates.^[286] The chordate fusion differs in its orientation to that in Plasmodium (in Plasmodium the 6-phosphogluconolactone is found at the N-terminus of the glucose 6-phosphate dehydrogenase protein), indicating that at least two separate fusion events have occurred. The metazoan fusion appears to have occurred near the bases of the metazoan and apicomplexan lineages. This fusion event was not found in any of the three sequenced Cryptosporidium genomes. It was not found in Perkinsus marinus or in either of the ciliate (Paramecium tetraurelia and Tetrahymena thermophila) genomes. More data will be needed to estimate the timing of this fusion event.

Only one of the two metazoan paralogs of glucose 6-phosphate dehydrogenase is fused, indicating that the fusion occurred after a duplication event. This duplication event occurred in an ancestor of the choanoflagellates and metazoans. Another fusion event between these enzymes occurred in an ancestor of the protozoan parasites Trichomonas and Giardia lamblia. In Giardia, the proteins are fused in opposite orientations. A third fusion event occurred between glucose 6-phosphate dehydrogenase with phosphogluconate dehydrogenase in a diatom species (Phaeodactylum tricornutum).

Phosphoinositide-specific phospholipase C (PI-PLC) is a major regulator of calcium-dependent signal transduction usually by liberation of calcium from intracellular stores through the action of its product, inositol-(1,4,5)-trisphosphate. These genes are found in P. falciparum and appear to be essential.^[287] The genes are twice as long as their mammalian counterparts and belong to the delta class of phospholipase C proteins.

The mechanism of action of the triose phosphate isomerase enzyme has been investigated in some detail.^[288] The conserved glutamic acid residue at position 97 is involved in the catalytic proton transfer. Modification of this residue may reduce the rate of catalysis by 9000 fold.

The type II NADH:ubiquinone oxidoreductase has been shown to be redundant in the blood forms but to be essential in the mosquito midgut.^[289]

Chorismate synthase (CS) catalyses the seventh and final step of the shikimate pathway. P. falciparum chorismate synthase (PfCS) is unique in terms of enzymatic behavior, cellular localization and in having two additional amino acid inserts compared to any other CS.^[290]

A phosphatidylserine decarboxylase has been cloned from the parasite Plasmodium knowlesi.^[291] It seems highly probably this enzyme is also found in P. falciparum.

Both the 2-C-methyl-d-erythritol-4-phosphate and shikimate pathways are functional in P. falciparum and vitamin E biosynthesis occurs.^[292]

Membrane biogenesis in this organism involves the enzyme phosphoethanolamine methyltransferase which catalyses the methylation of phosphoethanolamine to phosphocholine. This pathway is found in plants and nematodes but not in humans. The enzymes in P. falciparum is a multi-functional unlike that of plants and nematodes.^[293] The enzyme from P. falciparum has been cloned and its structure solved.

A homolog of the inhibitor of protein phosphatase 1 has been cloned.^[294] This gene is essential for survival and appears to be localised to the nucleus. A conserved 41- Lysine-Valine-Valine-Arginine-Tryptophan- 45 motif is essential for its inhibition activity.

The parasite actively synthesises pyridoxal-phosphate (vitamin B₆). This process involves two sets of reactions: condensation of ribulose 5-phosphate, glyceraldehyde-3-phosphate and ammonia produced from glutamine. These actions are carried out by separate subunits. The synthase domain is known as Pdx1 and the glutaminase domain as Pdx2. In P. falciparum the core Pdx1 is a dodecamer and forms the core of the enzyme. There are up to 12 Pdx2 subunits surrounding the Pdx1 subunit.^[295] The majority of the synthesis is carried out by Pdx1. The pentose substrate is covalently attached through its C1 and forms a Schiff base with the Lysine 84 residue. The ammonia transfer between Pdx2 glutaminase and Pdx1 active sites is regulated by a transient tunnel.

At least one function of B₆ in this parasite is as an antioxidant.^[296]

Orotate phosphoribosyltransferase catalyzes the magnesium dependent condensation of orotic acid with 5-α-D-phosphorylribose 1-diphosphate to yield diphosphate and the nucleotide orotidine 5'-monophosphate. This enzyme has been crystallised.^[297]

The enzyme 1-deoxy-D-xylulose-5-phosphate reductoisomerase can be inhibited by the antibiotic fosmidomycin which has been shown to potentially useful as an antimalarial.^[298]

There are several versions of the enzyme glutamate dehydrogenase (GDH) encoded in the genome. Of these GDH1 and 3 appear to locaise in the cytoplasm and GDH2 to the apicoplast.^[299]

HMB-PP synthase (IspG), an iron-sulphur (4Fe4S) protein involved in isoprenoid biosynthesis, has two domains - a TIM barrel and a 4Fe4S domain - in bacteria. In plants and malaria parasites there is an additional large insert domain.^[300] This is a second TIM barrel that interacts with the other TIM barrel.

Adenylate kinases are phosphotransferases that catalyze the interconversion of adenine nucleotides. There are at least three adenylate kinases (PfAK1, PfAK2 and GTP:AMP phosphotransferase) encoded in the genome.^[301] PfAK1 and PfAK2 both catalyse the conversion of ATP and AMP to two molecules of ADP. PfGAK instead has a preference for GTP and AMP and does not accept ATP as a substrate.

Deoxyhypusine synthase catalyzes the first step in hypusine biosynthesis of eukaryotic initiation factor 5A (EIF5A). Inhibitors of this enzyme may be of use theraputically.^[302]

Within the genome there are two adenylyl cyclases - ACα and ACβ.^[303] ACα contains six predicted transmembrane domains and a single carboxy-terminal catalytic domain homologous to sAC-like ACs. It is a predicted bifunctional protein comprising both a potassium channel and an AC that is conserved in the alveolates. It is expressed in the gametocytes. ACβ has no predicted transmembrane regions and possesses two AC catalytic domains. It has a marked pH dependence and is required for the erythrocytic stages.

Asymmetrical diadenosine 5',5″-P¹,P⁴-tetraphosphate hydrolase (EC 3.6.1.17) catalyses the conversion of diadenosine 5',5″-P¹,P⁴-tetraphosphate (Ap₄A) to ATP and AMP and diadenosine 5',5″-P¹,P⁵-pentaphosphate (Ap₅A) to ATP and ADP. This enzyme from the parasite has been cloned and expressed.^[304]

Palmitoylation - the reversible addition of a lipid moiety to a cysteine residue - appears to be common in this parasite.^[305] Its role in its biology is not yet understood.

A glycerophosphodiesterase has been cloned.^[306] This enzyme is found in the parasitophorous vacuole, food vacuole and cytosol. It appears to be an essential gene but its specific function is currently unclear.

The translationally controlled tumor protein appears to bind artemisinin.^[307] This may contribute to its anti malarial action.

An N-myristoyltransferase is present in the genome.^[308] It is involved in protein trafficking.

Calcium fluctuation

In recently invaded erythrocytes the Ca²⁺ concentration increases about 10 fold.^[309] The Ca²⁺ content increases as the parasite matures.^[310] In infected erythrocytes, Ca²⁺ is almost exclusively localized in the parasite compartment and changes but little in the cytosol of the host cell.

Cytosolic calcium²⁺ increases evoked by extracellular stimuli are may be observed in the form of oscillating Ca²⁺ spikes in eukaryotic cells. Spontaneous spikes in the calcium levels have been observed in Plasmodium falciparum.^[311] The frequency of Ca2+ oscillations are higher in early ring forms than that in early trophozoites. Blockage of this oscillation results in the cessation of intraerythrocytic maturation and death of the parasite. This effect is maximal in the trophozoites.

An inositol phosphate kinase with a role in calcium metabolism has been cloned.^[312]

References