Scanning and transmission electron microscopy of host cell pathology associated with penetration byEimeria papillata sporozoites

Scanning and electron microscopy was used to study the pathogenesis that occurred in mouse epithelial cells that had been penetrated byEimeria papillata sporozoites. Optimal penetration of parasites injected into nonligated and ligated mouse intestine was found to occur at 4–15 min post-inoculation. During initial penetration, the parasite caused disruption of the microvilli of the intestinal cells, which led to detachment of the microvilli from the plasma membrane of the penetrated cell. Host cells penetrated by the parasite showed extensive destruction of the internal cellular organization together with blebbing of host-cell cytoplasm and release of internal organelles such as mitochondria. Ultimately, the penetrated cells completely broke down, leaving vacuolated areas next to ultrastructurally normal epithelial cells.     

Ultrastructural observations of host-cell invasion by sporozoites ofEimeria papillata in vivo

Scanning and transmission electron microscopy were used to study the invasion of mouse small-intestinal epithelium by sporozoites ofEimeria papillata. Some mice received oocysts by gavage and others received either sporocysts or sporozoites by direct injection into the small intestine. The highest concentration of invaded cells were found in ligated intestinal tissues studied at 5–45 min after the inoculation of sporozoites. Sporozoites actively invaded anterior end first, which resulted in extensive damage to the host cell. Such cells showed disrupted microvilli; protuberances of cytoplasm into the lumen, apparently the result of a disrupted plasma membrane; vacuolization of the cytoplasm; and damage to the mitochondria. These damaged cells were rapidly vacated as the sporozoite moved laterally into one or more adjacent intact host cells without entering the lumen. It is suggested that the host cell initially entered from the lumen becomes so severely traumatized that the parasite of necessity enters an adjacent cell as a prelude to further development. Various aspects of host-cell invasion by coccidia and malarial parasites are reviewed.     

Electron microscopic studies on the interaction of rat Kupffer cells andPlasmodium berghei sporozoites

The interactions betweenPlasmodium berghei sporozoites and Kupffer cells in rat liver were studied by transmission electron microscopy. Between 10 and 45 min after inoculation, sporozoites were found in the process of entering Kupffer cells and inside phagolysosomes. The sporozoites entered the Kupffer cells by phagocytosis as determined by the presence of pseudopods and local accumulations of aggregated microfilaments and the resulting exclusion of other organelles in the phagocyte cytoplasm beneath the attached parasite. Sporozoites were taken up either with their anterior end first, or backwards. Scanning electron microscopy of in vitro sporozoite Kupffer cell interaction confirmed these observations. It was concluded that sporozoites are taken up in a normal phagocytic way by the Kupffer cells, regardless of their initial place of contact or position. Thirty min after inoculation sporozoites found in phagolysosomes were still morphologically intact but after 45 min we could encounter completely digested sporozoites.     

Apical organelle discharge by Cryptosporidium parvum is temperature, cytoskeleton, and intracellular calcium dependent and required for host cell invasion

The apical organelles in apicomplexan parasites are characteristic secretory vesicles containing complex mixtures of molecules. While apical organelle discharge has been demonstrated to be involved in the cellular invasion of some apicomplexan parasites, including Toxoplasma gondii and Plasmodium spp., the mechanisms of apical organelle discharge by Cryptosporidium parvum sporozoites and its role in host cell invasion are unclear. Here we show that the discharge of C. parvum apical organelles occurs in a temperature-dependent fashion. The inhibition of parasite actin and tubulin polymerization by cytochalasin D and colchicines, respectively, inhibited parasite apical organelle discharge. Chelation of the parasite's intracellular calcium also inhibited apical organelle discharge, and this process was partially reversed by raising the intracellular calcium concentration by use of the ionophore A23187. The inhibition of parasite cytoskeleton polymerization by cytochalasin D and colchicine and the depletion of intracellular calcium also decreased the gliding motility of C. parvum sporozoites. Importantly, the inhibition of apical organelle discharge by C. parvum sporozoites blocked parasite invasion of, but not attachment to, host cells (i.e., cultured human cholangiocytes). Moreover, the translocation of a parasite protein, CP2, to the host cell membrane at the region of the host cell-parasite interface was detected; an antibody to CP2 decreased the C. parvum invasion of cholangiocytes. These data demonstrate that the discharge of C. parvum sporozoite apical organelle contents occurs and that it is temperature, intracellular calcium, and cytoskeleton dependent and required for host cell invasion, confirming that apical organelles play a central role in C. parvum entry into host cells.     

Flavivirus entry into cultured mosquito cells and human peripheral blood monocytes

Summary The entry modes of Japanese encephalitis (JE) and dengue-2 (DEN-2) viruses into C6/36 mosquito cells and of DEN-2 virus into human peripheral blood monocytes in vitro were studied. Inoculation of either JE or DEN-2 virions into C6/36 cells resulted in direct penetration of the virions into the cytoplasm at the cell surface in 3 stages. At stage 1, virions attached to the plasma membrane of host cells by their envelope spikes; at stage 2, the virion envelopes approximated to and eventually overlapped the host plasma membrane, and in the process the plasma membrane at the attachment sites dissolved; and, at stage 3, virions penetrated into the cytoplasm through the plasma-membrane disruptions created at the adsorption sites. Virions themselves apparently disintegrated at or near the penetration sites, for no virions were seen in the deeper cytoplasm. Coated pits did not form at the virion attachment sites, and virion-containing vesicles were not found in the cytoplasm. In the entry of DEN-2 virus into human peripheral blood monocytes, virions were found, adsorbed onto the external surface of the plasma membrane and attached to the luminal surface of macropinocytic vacuolar membranes. The latter apparently occurred as the result of ruffling and macropinocytic activities of the cells. At both sites virions penetrated into the cytoplasm through the plasma or vacuolar membrane in the same manner as they did through the plasma membrane of C6/36 cells. No evidence of viral entry by receptor-mediated endocytosis was observed. Implications of the entry mode of the mosquito cell-generated DEN-2 virus into human peripheral blood monocytes to an early process of natural, mosquito-transmitted infection is discussed.     

Induction of secretion and surface capping of microneme proteins in Eimeria tenella

Micronemes are secretory organelles of the invasive stages of apicomplexan parasites and contain proteins that are important for parasite motility and host cell invasion. We have examined the induction of microneme secretion in the coccidian Eimeria tenella. When sporozoites were added to MDBK cells in culture, microneme proteins were secreted, capped backwards over the parasite surface and deposited onto underlying host cells from the posterior end of gliding parasites. Induction of secretion was also achieved by the addition of foetal calf serum, or purified albumin, to extracellular sporozoites. Microneme secretion per se was not dependent on parasites being able to move or to invade host cells. However, in the presence of cytochalasin D, which disrupts actin polymerisation and prevents parasite movement, microneme proteins were secreted from the apical tip but were not capped backwards over the sporozoite surface. These observations support the hypothesis that microneme proteins function as ligands which, when secreted out onto the parasite surface, form a link, either directly or indirectly, between the sub-pellicular actin–myosin cytoskeletal motor of the parasite and the surface of target host cells.     

Induction of secretion and surface capping of microneme proteins in Eimeria tenella

Micronemes are secretory organelles of the invasive stages of apicomplexan parasites and contain proteins that are important for parasite motility and host cell invasion. We have examined the induction of microneme secretion in the coccidian Eimeria tenella. When sporozoites were added to MDBK cells in culture, microneme proteins were secreted, capped backwards over the parasite surface and deposited onto underlying host cells from the posterior end of gliding parasites. Induction of secretion was also achieved by the addition of foetal calf serum, or purified albumin, to extracellular sporozoites. Microneme secretion per se was not dependent on parasites being able to move or to invade host cells. However, in the presence of cytochalasin D, which disrupts actin polymerisation and prevents parasite movement, microneme proteins were secreted from the apical tip but were not capped backwards over the sporozoite surface. These observations support the hypothesis that microneme proteins function as ligands which, when secreted out onto the parasite surface, form a link, either directly or indirectly, between the sub-pellicular actin–myosin cytoskeletal motor of the parasite and the surface of target host cells.     

Cryptosporidium parvum glycoprotein gp40 localizes to the sporozoite surface by association with gp15

Cryptosporidium spp. are waterborne apicomplexan parasites responsible for outbreaks of diarrheal disease worldwide. Antigens involved in zoite invasion into host cells have been the focus of many investigations as these may prove to be good vaccine candidates. gp40/15 is a zoite antigen synthesized as a precursor protein and proteolytically cleaved into the mature glycoproteins, gp40 and gp15. gp15 is anchored in the sporozoite membrane by a glycosylphosphatidyl inositol moiety, while gp40 is predicted to be soluble. However, gp40 bears epitopes that recognize a host cell receptor. If this interaction is important for zoite invasion, then gp40 must have some mechanism of associating with the parasite membrane. In these studies we demonstrate that gp40 and gp15 co-localize to the surface membrane of sporozoites and merozoites, and co-immunoprecipitate, suggesting that these antigens associate after proteolytic cleavage to generate a protein complex capable of linking zoite and host cell surfaces.     

In vitro association of leptospires with host cells.

Interactions of Leptospira interrogans with cultured endothelial and kidney epithelial cells were assayed by examining (i) cytoadherence of intrinsically radiolabeled leptospires to eucaryotic cell monolayers and (ii) penetration of leptospires through cell monolayers grown on polycarbonate filters in chemotaxis chambers. L. interrogans serovars attached to cultured cells in a dose- and time-dependent manner. Adherence was diminished following pretreatment of organisms with proteases, rabbit immune serum, or heat. When observed by scanning electron microscopy, most leptospires attached by both ends, rather than just one tip like Treponema pallidum. In penetration assays, 9.7% of added L. interrogans migrated through the monolayer-filter barrier, while only 0.3% of L. biflexa penetrated in the same time interval. Transmission electron microscopy revealed that organisms entered the host cell cytoplasm. These in vitro results indicate that leptospires have an invasive capacity that may be related to pathogenicity in vivo and suggest that further investigation of interactions with host cells may enhance knowledge of leptospiral virulence.     

Antibodies to Plasmodium circumsporozoite protein (CSP) inhibit sporozoite's cell traversal activity

Plasmodium sporozoites are deposited in the skin of the mammalian host by Anopheles mosquitoes. To continue the life cycle, the sporozoites have to invade the host's hepatocytes, where they transform into exoerythrocytic forms (EEFs) inside a parasitophorous vacuole. During their route from the skin to the liver, the parasites traverse the capillary epithelium in the dermis to enter the blood circulation, and cross the endothelium of liver sinusoids to enter the parenchyma. Cell traversal by sporozoites is usually measured by quantifying dyes that enter or are released from cells during incubation with salivary gland sporozoites. These methods do not distinguish cell traversal from cell wounding. Here we validate an assay that quantifies cell traversal of sporozoites through monolayers of MDCK cells that form tight junctions. We compared cell traversal of wt sporozoites and of parasites lacking the Type I membrane protein TLP (TRAP-like protein) previously implicated in cell traversal. We provide direct evidence that TLP ko sporozoites are defective in cell traversal and that they are retained inside the MDCK cytoplasm. We then used the MDCK assay to study the effect of a monoclonal antibody (3D11) to the circumsporozoite protein (CSP) on the parasite's cell traversal. We show that 3D11 inhibits cell traversal at nanomolar concentrations. We conclude that antibodies elicited by CSP-based vaccines are likely to inhibit the migration of sporozoites from the skin to the liver.     

Distribution of cytoskeletal structures and organelles of the host cell during evolution of the intracellular parasitism by Trypanosoma cruzi

The distribution of microtubules, microfilaments, mitochondria, Golgi complex and endosomes/lysosomes was analyzed in Vero cells allowed to interact for different periods of time with the pathogenic protozoan Trypanosoma cruzi and observed by confocal laser scanning microscopy. Microtubules were revealed using a mouse monoclonal anti-alpha-tubulin antibody. Actin filaments were revealed using phalloidin-rhodamine. To identify mitochondria, endosomes/lysosomes and the Golgi complex the cells were labelled with Rhodamine 123, Lucifer yellow and C6-NBD-ceramide, respectively. During cell invasion actin filaments concentrate at the site of parasite penetration in some, but not in all cells, probably depending upon the mechanism used by the trypomastigote form to penetrate into the host cells. Following internalization the trypomastigote form gradually changes into the amastigote form, disruption of the parasitophorous vacuole membrane takes place and the amastigote form enters in direct contact with host cell structures and organelles, and starts to divide. The presence of the parasite in the cytoplasm of the host cell did not induce significant changes in the distribution of actin filaments, microtubules, the Golgi complex, mitochondria and endosomes/lysosomes during the first 48 h of infection. Amastigote forms were seen close to the microtubules. After 72 h of interaction, the number of microtubules and microfilaments around the parasites was reduced and lysosomes and mitochondria were seen in between the parasites.     

Evidence for vesicle-mediated trafficking of parasite proteins to the host cell cytosol and erythrocyte surface membrane in Plasmodium falciparum infected erythrocytes

Plasmodium falciparum malaria parasites actively remodel the host cell cytosol and plasma membrane during the erythrocytic cycle. The focus of this investigation was to characterize intra-parasitic and -erythrocytic secretory pathways. Electron-dense vesicles, similar in appearance to mammalian secretory vesicles were detected in proximity to smooth tubo-vesicular elements at the periphery of the parasite cytoplasm in mature parasites by transmission electron microscopy. Vesicles (60-100 nm diameter), which appeared to be coated, were visualized on the erythrocytic side of the parasite vacuolar membrane and in the erythrocyte cytosol. The vesicles seemed to bind to and fuse with the erythrocyte membrane, giving rise to cup-shaped electron-dense structures, which might be intermediates in knob structure formation. Treatment of mature parasites with aluminum tetrafluoride, an activator of GTP-binding proteins, resulted in the accumulation of the vesicles with an electron-dense limiting membrane in the erythrocyte cytosol into multiple vesicle strings. These vesicle complexes were often associated with and closely abutted the erythrocyte membrane, but were apparently prevented from fusing by the aluminum fluoride treatment. The parasite proteins PfEMP1 and PfEMP3 were found by immunoelectron microscopy to be associated with these vesicles, suggesting they are responsible for transporting these proteins to the erythrocyte membrane.     

Plasmodium falciparum antigens associated with membrane structures in the host erythrocyte cytoplasm

Hybridomas were made from mice immunized with plasma membranes from erythrocytes infected with Plasmodium falciparum. Among the monoclonal antibodies produced, a series reacted with antigens in the host cell cytoplasm. Immunoelectron microscopy, along with indirect fluorescent antibody double labeling experiments, were used to further localize the antigens to membrane structures (presumably Maurer's clefts) in the erythrocyte cytoplasm. The epitopes thus localized are found on three parasite proteins (20 kDa, 29 kDa, and 45 kDa) and one parasite glycoprotein (45 kDa). They are likely to be part of a transport system for the parasite.     

Redistribution of plasma-membrane surface molecules during formation of the Leishmania amazonensis-containing parasitophorous vacuole

Leishmania amazonensis presents two developmental stages that gain access to the host macrophage through phagocytosis. The protozoan resides in a membrane-bound compartment, the parasitophorous vacuole (PV), which can fuse with the endocytic system. For evaluation of the parasite/host-cell interaction process and of PV biogenesis, the two parasite forms or host-cell membrane whose surface had previously been labeled with specific probes for lipids, proteins, and sialoglycoconjugates were allowed to interact for periods varying from 5 to 15 min for adhesion and from 30 to 60 min for PV formation. The fate of fluorescent probes was followed by confocal laser scanning microscopy. In host cells previously labeled with PKH26, DTAF and FITC-thiosemicarbazide, which label membrane lipids, proteins, and sialoglycoconjugates, respectively, interaction with both protozoan forms revealed that adhesion to the macrophage was sufficient for labeling of the parasite surface. In addition, recently formed PVs displayed strongly labeled intravacuolar parasites, except for amastigote-macrophage interaction in a DTAF-labeled macrophage that displayed slight labeling of intravacuolar parasites, with the membrane lining the PV evidently being stained. Therefore, the vacuole modulation presents some particularities such that different host-cell membrane components may be selected, depending on the protozoan form involved. Thereafter, amastigotes labeled with the probes mentioned above displayed a diffuse labeling pattern after interaction with unlabeled macrophages, suggesting the spreading of Leishmania surface molecules during the initial parasite-invasion stages. In particular, intravacuolar DTAF-labeled amastigotes showed a delineating halo around the PV, with the intravacuolar parasite being partially labeled. Promastigotes could not be labeled with 5-(4,6-dichlorotriazinyl)aminofluorescein (DTAF) or with fluorescein-5-thiosemicarbazide, but promastigotes labeled with PKH26 lost the fluorescent probe during the invasion process such that slightly labeled promastigotes were seen inside the PV. These observations indicate the existence of a dynamic process of exchange of membrane-associated glycoproteins and lipids between the parasite and the host cell. Received: 15 May 1999 / Accepted: 10 September 1999     

Survival and antigenic profile of irradiated malarial sporozoites in infected liver cells.

Exoerythrocytic (EE) stages of Plasmodium berghei derived from irradiated sporozoites were cultured in vitro in HepG2 cells. They synthesized several antigens, predominantly but not exclusively those expressed by normal early erythrocytic schizonts. After invasion, over half the intracellular sporozoites, both normal and irradiated, appeared to die. After 24 h, in marked contrast to the normal parasites, EE parasites derived from irradiated sporozoites continued to break open, shedding their antigens into the cytoplasm of the infected host cells. Increasing radiation dosage, which has previously been shown to reduce the ability of irradiated sporozoites to protect animals, correlated with reduced de novo antigen synthesis by EE parasites derived from irradiated sporozoites.     

Ultrastructure of the attachment of Cryptosporidium sporozoites to tissue culture cells

The attachment of Cryptosporidium sporozoites to Madin-Darby canine kidney (MDCK) cells was examined using transmission electron microscopy. As the anterior end of the sporozoite came into close proximity to the MDCK cell, the host cell membrane evaginated around the sporozoite, forming a parasitophorous vacuole. A dense band formed below the host cell membrane at the site nearest to the conoid. Variably electron-dense material was apparently released from the conoid and a large membrane-bound vacuole was formed in the anterior end of the sporozoite, displacing the typical anterior electron-dense organelles (rhoptries and micronemes). The outer membrane of the sporozoite pellicle then fused with the host cell membrane immediately adjacent to the conoid. The membrane surrounding the anterior vacuole was also fused with the common host-parasite membrane, forming Y-shaped membrane junctions where each limb was a unit membrane. A direct link was thereby established between the anterior vacuole of the sporozoite and the host cell cytoplasm. The anterior vacuole membrane separating the sporozoite and the host cell cytoplasm was the precursor of the feeder organelle.     

Specific inflammatory cell infiltration of hepatic schizonts in BALB/c mice immunized with attenuated Plasmodium yoelii sporozoites.

We compared immunization of BALB/c mice with radiation-attenuated versus killed sporozoites of the rodent malaria parasite, Plasmodium yoelii. We employed a suboptimal schedule of only two immunizations, in expectation that some parasites might break through the resultant low level immunity and that it might thus be possible to study the response of the host against these 'breakthrough' schizonts. As a measure of protective immunity, we used histological means to determine the percentages of challenge sporozoites prevented from completing development into hepatic schizonts within the liver. Immunization with attenuated sporozoites led to almost complete protection, whereas immunization with similar dosages of killed sporozoites led to approximately a 75% protection. Fluorescent antibody titers against sporozoites were similar in both sets of immunized animals. However, serum from mice immunized with attenuated sporozoites had a protective effect upon passive transfer into immunologically naive mice subsequently challenged with normal sporozoites; serum from mice immunized with killed sporozoites had no such effect. When mice suboptimally immunized with attenuated sporozoites were challenged, we observed breakthrough schizonts being infiltrated with inflammatory cells, primarily mononuclear cells, and neutrophils; partial depletion of CD4+ or CD8+ cells within these mice prior to challenge prevented the infiltration of breakthrough schizonts. Thus, cellular infiltration of schizonts was apparently secondary to earlier action by lymphocytes. This infiltration was also not observed in mice immunized with killed sporozoites. The more effective protective immunity induced by attenuated sporozoites could be due to their ability to release antigen into the cytoplasm of hepatocytes that they invade or their ability to continue differentiating, thereby presenting new antigens that are not seen after immunization with killed sporozoites.     

A member of a conserved Plasmodium protein family with membrane-attack complex/perforin (MACPF)-like domains localizes to the micronemes of sporozoites

Pore-forming proteins are employed by many pathogens to achieve successful host colonization. Intracellular pathogens use pore-forming proteins to invade host cells, survive within and productively interact with host cells, and finally egress from host cells to infect new ones. The malaria-causing parasites of the genus Plasmodium evolved a number of life cycle stages that enter and replicate in distinct cell types within the mosquito vector and vertebrate host. Despite the fact that interaction with host-cell membranes is a central theme in the Plasmodium life cycle, little is known about parasite proteins that mediate such interactions. We identified a family of five related genes in the genome of the rodent malaria parasite Plasmodium yoelii encoding secreted proteins all bearing a single membrane-attack complex/perforin (MACPF)-like domain. Each protein is highly conserved among Plasmodium species. Gene expression analysis in P. yoelii and the human malaria parasite Plasmodium falciparum indicated that the family is not expressed in the parasites blood stages. However, one of the genes was significantly expressed in P. yoelii sporozoites, the stage transmitted by mosquito bite. The protein localized to the micronemes of sporozoites, organelles of the secretory invasion apparatus intimately involved in host-cell infection. MACPF-like proteins may play important roles in parasite interactions with the mosquito vector and transmission to the vertebrate host.     

The host-parasite relationship ofToxoplasma gondii in the brains of chronically infected mice

Summary The host parasite relationship in the brains of asymptomatic mice chronically infected withToxoplasma gondii was examined at 3, 6 and 12 months post-infection (PI) using electron microscopy. The parasites were located in large numbers within tissue cysts which ranged in size from 10–50 µm in diameter. The cysts were predominantly found in the grey matter. The toxoplasms were enclosed by a cyst wall consisting of a membrane, with irregular invaginations, and an underlying layer of homogeneous osmiophilic material. A detailed examination of 50 cysts revealed that all the cysts were present within intact host cells irrespective of their size or the period PI. The majority of host cells could be positively identified as neurons by the presence of synapses. No extracellular cysts were observed. It is probable that the intracellular location of the cysts protects them from recognition and attack by the host immune system.