Hemoglobin metabolism in the mala...
P1: ARK/mkv P2: RVA/ARK/MBL/plb QC: MBL/abe T1: MBL August 7, 1997 17:33 Annual Reviews AR038-04 PLASMODIUM HEMOGLOBIN DEGRADATION 99 morphologically distinct phases are then observed. The ring stage, lasting approximately 24 h in P. falciparum, accounts for about half of the intraery- throcytic cycle, but it is metabolically nondescript (50). It is followed by the trophozoite stage, a very active period during which most of the red blood cell cytoplasm is consumed. Finally, parasites undergo 4���5 rounds of binary divisions during the schizont stage, producing merozoites that burst from the host cell to invade new erythrocytes, beginning another round of infection. The clinical manifestations of malaria result from schizont rupture and additionally, in the case of P. falciparum, from trophozoite adherence to endothelial cells (90). THE ROLE OF HEMOGLOBIN DEGRADATION Quantitation Hemoglobin comprises 95% of the cytosolic protein of the red blood cell, where it is present at a concentration of 5 mM. During the intraerythrocytic cycle, the host cell cytoplasm is consumed and an estimated 60���80% of the hemoglobin is degraded (10, 92, 98). A number of lower estimates have also been reported these most likely reflect assessment of less mature parasites still in the process of hemoglobin consumption. Hemoglobin proteolysis releases heme and generates amino acids. The heme moiety does not appear to be metabolized or recycled (39), but instead, is stored as an inert polymer known as the malaria pigment hemozoin (123). Utilization Amino acids derived from globin hydrolysis are incorporated into parasite pro- teins (124, 140) and also appear to be available for energy metabolism (123). Hemoglobin proteolysis may be essential for survival, because Plasmodium has a limited capacity for de novo amino acid synthesis (123, 141). However, hemoglobin degradation alone appears insufficient for the parasite���s metabolic needs since it is a poor source of methionine, cysteine, glutamine, and glutamate and contains no isoleucine at all. P. falciparum can be maintained in a red blood cell culture system provided that the five amino acids limiting in hemoglobin are present in the medium (36, 48). Thus, the parasite appears able to rely on both hemoglobin proteolysis and exogenous amino acids for growth. Amino acids are readily taken up by parasites from the culture medium (34, 82, 123). This is true for amino acids that are missing from hemoglobin as well as for those that are abundant, and it suggests that the reliance on hemoglobin degradation may reflect a limited availability of certain amino acids in the host serum. When Plasmodium is cultured in a rich medium containing 20 amino acids, the extent of hemoglobin proteolysis is similar to that seen in vivo. Interestingly, more amino acids are generated from hemoglobin breakdown than are used Annu. Rev. Microbiol. 1997.51:97-123. Downloaded from arjournals.annualreviews.org by Utah State University on 07/27/06. For personal use only.
P1: ARK/mkv P2: RVA/ARK/MBL/plb QC: MBL/abe T1: MBL August 7, 1997 17:33 Annual Reviews AR038-04 100 FRANCIS ET AL for protein synthesis, resulting in the diffusion of some hemoglobin-derived amino acids into the host cell (160). This raises the possibility that hemoglobin catabolism may have additional functions. One hypothesis is that parasites digest host cell cytosol to prevent premature red cell lysis that might occur if parasite growth were not compensated by reduced host cell volume (50, 160). Perhaps the strongest argument that hemoglobin degradation is necessary for survival comes from studies with protease inhibitors. When hemoglobin pro- teolysis is blocked, parasite development is interrupted (48, 76, 107, 109, 113, 115). In addition, when parasites are cultured in a medium lacking most amino acids, they show increased sensitivity to a protease inhibitor known to block hemoglobin proteolysis (48). HEMOGLOBIN INGESTION Cytostome Hemoglobin ingestion during the ring (also known as young trophozoite) stage of development appears to be limited. However, hemozoin can be detected in early stage parasites (4, 98), which suggests that the cellular machinery for ingestion and proteolysis is present. Ultrastructural studies of several species of Plasmodium are consistent with this finding (117, 130, 131). These studies reveal that, early in development, small portions of cytoplasm are taken up by micropinocytosis. Vesicles containing a tiny hemozoin crystal can some- times be seen. As the parasite matures, a larger volume of hemoglobin is ingested by means of a cytostomal system that is formed by invagination of the parasitophorous vacuolar membrane and the parasite plasma membrane. The cytostome is a poorly understood structure spanning these two mem- branes that separate parasite and erythrocyte cytoplasms (Figure 1). In P. fal- ciparum, the cytostome is a large double membrane-enclosed pear-shaped structure (73, 95). In some instances, several cytostomes can be seen in one trophozoite-stage parasite (73, 130). When the cytostomes ingest red cell cy- toplasm, double membrane-delimited vesicles are formed by budding. The hemozoin-containing vesicles appear to fuse, forming one or two large, sin- gle membrane-enclosed digestive vacuoles that contain a cluster of hemozoin crystals (130) (Figure 1). Variation in cytostome morphology and activity oc- curs between species (1, 133, 134). In murine plasmodia, a single branching tubular structure appears to fill with hemoglobin and break down into vesicles. Hemozoin accumulates in the vesicles as the parasite matures (132). Digestive Vacuole Once the digestive vacuole is formed, it appears to be the primary site of hemoglobin degradation (97). Vesicles budding from the cytostome transport Annu. Rev. Microbiol. 1997.51:97-123. Downloaded from arjournals.annualreviews.org by Utah State University on 07/27/06. For personal use only.
P1: ARK/mkv P2: RVA/ARK/MBL/plb QC: MBL/abe T1: MBL August 7, 1997 17:33 Annual Reviews AR038-04 PLASMODIUM HEMOGLOBIN DEGRADATION 101 Figure 1 Transmission electron micrograph of a Plasmodium falciparum trophozoite inside an erythrocyte. c, cytostome v, transport vesicle dv, digestive vacuole. (After Reference 55.) hemoglobin to the vacuole where their outer membrane fuses with the single membrane���delimited vacuole. Fusion releases a single membrane���enclosed vesicle that can occasionally be seen inside the vacuole (95, 157). The vesicle is then lysed and hemoglobin is hydrolyzed. The agent responsible for vesicular lysis has not been identified, but a phospholipase activity has been hypothesized (51, 71). Whatever the identity of the lytic agent, its activity appears to be highly specific: It is able to distinguish between the vesicular and vacuolar membranes. When parasites are cultured in the presence of chloroquine, unlysed transport vesicles are more abundant (95, 157, 158). Digestive vacuoles in P. falciparum are acidic organelles with a pH estimated at 5.0���5.4 (70, 156). An ATPase activity that could be responsible for the proton gradient was detected in partially purified digestive vacuoles (24), and genes for the proton pump subunits VAP-A and VAP-B have been cloned (66, 67). Both subunits are expressed throughout the intraerythrocytic cycle. Immunolo- calization studies of the vacuolar B subunit indicate that it is associated with the digestive vacuole but that it is also distributed over most of the parasite (67). The degradative capacity and acidic pH of digestive vacuoles are characteristic of lysosomes and yeast vacuoles. Acid phosphatase activity typically found in these organelles was associated with hemozoin in the digestive vacuoles of the Annu. Rev. Microbiol. 1997.51:97-123. Downloaded from arjournals.annualreviews.org by Utah State University on 07/27/06. For personal use only.
P1: ARK/mkv P2: RVA/ARK/MBL/plb QC: MBL/abe T1: MBL August 7, 1997 17:33 Annual Reviews AR038-04 102 FRANCIS ET AL lower vertebrate parasites P. berghei and P. gallinaceum (2). In P. falciparum, acid phosphatase was found in endocytic vesicles but not in digestive vac- uoles (132). ��-glucuronidase, ��-galactosidase and acid phosphatase activities were absent from digestive vacuoles that were isolated from P. falciparum (55). The lack of nonproteolytic acid hydrolases in digestive vacuoles suggests that, during their relatively short intraerythrocytic cycle, parasites may not need to degrade and recycle macromolecules other than hemoglobin and that digestive vacuoles may be specialized for that purpose (55). HEMOGLOBIN DEGRADATION Proteases In P. falciparum, some hemoglobin degradation is seen during the ring and early schizont stages of development, but the vast majority of degradation oc- curs during the 6���12 h trophozoite stage (159). Acid proteases present in parasitized red blood cell extracts or partially purified parasite extracts from different species of Plasmodium have been proposed to act as hemoglobinases (3, 9, 58, 59, 74, 75, 119, 125, 144, 145). Some of these were probably derived from the red blood cell or parasite cytoplasm (53, 120, 122). In addition, a trophozoite-stage cysteine protease of 28 kD was observed in cell extracts from P. falciparum (112) (see Falcipain Specificity and Inhibition section, below). Enzymes that are present in the digestive vacuole and that are capable of hemoglobin degradation were identified by subcellular fractionation (55). The dense iron-rich hemozoin crystals that are present in digestive vacuoles allowed differential centrifugation and Percoll density gradient separation for purifi- cation of these organelles. Marker enzyme analysis and electron microscopy indicated that the digestive vacuoles are not contaminated by other membranes and compartments (55). When the digestive vacuole lysate was added to de- natured globin and incubated at acid pH, both aspartic protease and cysteine protease activities were detected. Aspartic proteases account for 60���80%, and cysteine protease(s) account for 20���40% of the globin-degrading activity in pu- rified digestive vacuoles (47, 52, 55). The combination of pepstatin, a specific aspartic protease inhibitor, and E-64, a specific cysteine protease inhibitor, com- pletely blocks this globin digestion, suggesting that aspartic and cysteine pro- teases are the primary enzymes responsible for globin proteolysis in the vacuole (52). Order of Action When native hemoglobin is incubated with digestive vacuole lysate, the disap- pearance of substrate can be analyzed by SDS-PAGE. Hemoglobin digestion is inhibited when pepstatin is added to such incubations, whereas the effect Annu. Rev. Microbiol. 1997.51:97-123. Downloaded from arjournals.annualreviews.org by Utah State University on 07/27/06. For personal use only.
P1: ARK/mkv P2: RVA/ARK/MBL/plb QC: MBL/abe T1: MBL August 7, 1997 17:33 Annual Reviews AR038-04 PLASMODIUM HEMOGLOBIN DEGRADATION 103 of E-64 is minimal (52, 55). The data suggest that the process of hemoglobin breakdown is ordered, and that it requires an aspartic protease-mediated initial cleavage, followed by secondary aspartic protease and cysteine protease cleav- ages. Other enzymes may participate in the breakdown of globin fragments, but this has not been shown. The combined activity of vacuolar hydrolases is expected to produce progressively smaller peptides. Exopeptidases could then generate free amino acids, as is the case in lysosomes and yeast food vacuoles (19, 65, 68). However, in contrast to the activity seen in mammalian and yeast degradative organelles, no exopeptidase activity has been described from P. fal- ciparum digestive vacuoles. Neutral pH-requiring aminopeptidases have been identified in trophozoite extract prepared from P. falciparum (19, 27, 65, 68) and rodent Plasmodium (22, 27). This suggests that either the aminopeptidase has a neutral pH optimum and still functions in the vacuole, that other uniden- tified peptidases exist in the vacuole, or that the final steps in the hemoglobin proteolysis pathway occur in the cytoplasm. Recent data suggest that vacuolar degradation generates small peptides but no free amino acids, which supports this last mechanism (K Kolakovich et al, unpublished). Plasmepsin Specificity and Inhibition Two aspartic proteases, plasmepsins I and II, and one cysteine protease, falci- pain, have been purified from digestive vacuoles of P. falciparum (52, 54). All three proteases have pH optima near 5, the physiological pH of the vacuole. Plasmepsins I and II were shown by mass spectroscopic analysis to cleave native hemoglobin with remarkable specificity between ��33Phe-34Leu (52, 54). This is also the site of initial cleavage by digestive vacuole extracts. The ��33���34 bond is in the hinge region of the hemoglobin molecule, the domain responsible for holding the tetramer together when oxygen is bound (101). Cleavage at this site can be envisioned to unravel hemoglobin and facilitate its proteolysis by other degradative enzymes. This region is highly conserved among vertebrate species (35). No homozygous mutations have been mapped to this position (15). Thus, it appears that the parasite has selected a particularly vulnerable site for its initial attack (54). When the tertiary structure of hemoglobin is relaxed, plasmepsins I and II can cleave at other sites (52, 54). The enzymes have divergent substrate specificities for these secondary cleavages. Plasmepsin I prefers phenylalanine at the P1 position, while plasmepsin II prefers hydrophobic residues on both sides of the scissile bond, especially leucine at the P10 position. It is not entirely clear why the parasite should have two distinct enzymes with apparently similar functions. Interestingly, plasmepsins I and II have completely different patterns of expression during the intraerythrocytic cycle (49). Plasmepsin I is synthesized during the early ring stage, and its synthesis continues through Annu. Rev. Microbiol. 1997.51:97-123. Downloaded from arjournals.annualreviews.org by Utah State University on 07/27/06. For personal use only.