Regulation of Trypanosoma cruzi invasion of nonphagocytic cells by the endocytically active GTPases dynamin, Rab5, and Rab7.

During invasion of nonphagocytic cells by Trypanosoma cruzi (T. cruzi), host cell lysosomes are recruited to the plasma membrane attachment site followed by lysosomal enzyme secretion. The membrane trafficking events involved in invasion have not been delineated. We demonstrate here that T. cruzi invasion of nonphagocytic cells was completely abolished by overexpression of a dominant negative mutant of dynamin. Likewise, overexpression of a dominant negative mutant of Rab5, the rate-limiting GTPase for endocytosis, resulted in reduced infection rates compared with cells expressing Rab5 wild-type. Moreover, cells expressing the activated mutant of Rab5 experienced higher infection rates. A similar pattern was also observed when Rab7-transfected cells were examined. Confocal microscopy experiments showed that parasites colocalized with green fluorescent protein-Rab5-positive early endosomes after 5 min of invasion. These data clearly indicate that newly forming T. cruzi phagosomes first interact with an early endosomal compartment and subsequently with other late component markers before lysosomal interaction occurs.

Rab22a affects the morphology and function of the endocytic pathway.

Soon after endocytosis, internalized material is sorted along different pathways in a process that requires the coordinated activity of several Rab proteins. Although abundant information is available about the subcellular distribution and function of some of the endocytosis-specific Rabs (e.g. Rab5 and Rab4), very little is known about some other members of this family of proteins. To unveil some of the properties of Rab22a, one of the less studied endosome-associated small GTPases, we have expressed the protein tagged with the green fluorescent protein in CHO cells. The results indicate that Rab22a associates with early and late endosomes (labeled by a 5 minute rhodamine-transferrin uptake and the cation-independent mannose 6-phosphate receptor, respectively) but not with lysosomes (labeled by 1 hour rhodamine horseradish peroxidase uptake followed by 1 hour chase). Overexpression of the protein causes a prominent morphological enlargement of the early and late endosomes. Two mutants were generated by site-directed mutagenesis, a negative mutant (Rab22aS19N, with reduced affinity for GTP) and a constitutively active mutant (Rab22aQ64L, with reduced endogenous GTPase activity). The distribution of the negative mutant was mostly cytosolic, whereas the positive mutant associated with early and late endosomes and, interestingly also with lysosomes and autophagosomes (labeled with monodansylcadaverine). Cells expressing Rab22a wild type and Rab22aS19N displayed decreased endocytosis of a fluid phase marker. Conversely, overexpression of Rab22aQ64L, which strongly affects the morphology of endosomes, did not inhibit bulk endocytosis. Our results show that Rab22a has a unique distribution along the endocytic pathway that is not shared by any other Rab protein, and that it strongly affects the morphology and function of endosomes.

Recruitment of coat-protein-complex proteins on to phagosomal membranes is regulated by a brefeldin A-sensitive ADP-ribosylation factor.

Particle internalization in macrophages is followed by a complex maturation process. We have previously observed that proteins bound to phagocytosed particles are sorted from phagosomes into a heterogeneous population of vesicles that fuse with endosomes. However, the mechanism and the protein machinery involved in the formation of these phagosome-derived vesicles are largely unknown. It has been shown that vesicles coated with coat protein complex type I (COPI) have a role in both secretion and endocytosis. To address the possibility that COPI proteins might participate in the formation of phagosome-derived vesicles we studied the recruitment of beta-COP to highly purified phagosomes. The binding of beta-COP to phagosomal membranes was regulated by nucleotides and inhibited by brefeldin A (BFA). An ADP-ribosylation factor 1 (ARF1) mutant defective in GTP hydrolysis supported the binding of beta-COP to phagosomes independently of added nucleotide. AlF(4) and Gbetagamma subunits, agents known to modulate heterotrimeric G-protein activity, were tested in the beta-COP binding assay. AlF(4) increased beta-COP association, whereas binding was inhibited by the addition of Gbetagamma subunits. Our results suggest that COP proteins are recruited to phagosomal membranes by a mechanism that involves heterotrimeric GTP-binding proteins and a BFA-sensitive ARF. In addition, our findings indicate that COPI proteins are involved in the recycling of components from phagosomes to the cell surface.

Trypanosoma cruzi: phosphatidylinositol 3-kinase and protein kinase B activation is associated with parasite invasion.

Multiple signal transduction events are triggered in the host cell during invasion by the protozoan parasite Trypanosoma cruzi. Here, we report the regulation of host cell phosphatydilinositol 3-kinase (PI3K) and protein kinase B (PKB/Akt) activities by T. cruzi during parasite-host cell interaction. Treatment of nonphagocytic cells (Vero, L(6)E(9), and NIH 3T3) and phagocytic cells (human and J774 murine macrophages) with the selective PI3K inhibitors Wortmannin and LY294002 significantly impaired parasite invasion in a dose-dependent fashion. A strong activation of PI3K and PKB/Akt activities in Vero cells was detected when these cells were incubated with trypomastigotes or their isolated membranes. Consistently, we were unable to detect activation of PI3K or PKB/Akt activities in host cells during epimastigote (noninfective) membrane-host cell interaction. Infection of transiently transfected cells containing an inactive mutant PKB showed a significant inhibition of invasion compared with the active mutant-transfected cells. T. cruzi PI3K-like activity was also required in host cell invasion since treatment of trypomastigotes with PI3K inhibitors prior to infection reduced parasite entry. Taken together, these results indicate that PI3K and PKB/Akt activation in parasites, as in host cells induced by T. cruzi, is an early invasion signal required for successful trypomastigote internalization.

A phorbol ester-binding protein is required downstream of Rab5 in endosome fusion.

Previous observations indicate that a zinc and phorbol ester binding factor is necessary for endosome fusion. To further characterize the role of this factor in the process, we used an in vitro endosome fusion assay supplemented with recombinant Rab5 proteins. Both zinc depletion and addition of calphostin C, an inhibitor of protein kinase C, inhibited endosome fusion in the presence of active Rab5. Addition of the phorbol ester PMA (phorbol 12-myristate 13-acetate) reversed the inhibition of endosome fusion caused by a Rab5 negative mutant. Moreover, PMA stimulated fusion in the presence of Rab5 immunodepleted cytosol. These results suggest that the phorbol ester binding protein is acting downstream of Rab5 in endosome fusion.

Zn2+ depletion blocks endosome fusion.

Fusion among endosomes is an important step for transport and sorting of internalized macromolecules. Working in a cell-free system, we previously reported that endosome fusion requires cytosol and ATP, and is sensitive to N-ethylmaleimide. Fusion is regulated by monomeric and heterotrimeric GTP-binding proteins. We now report that fusion can proceed at very low Ca2+ concentrations, i.e. < 30 nM. Moreover, fusion is not affected when intravesicular Ca2+ is depleted by preincubation of vesicles with calcium ionophores (5 microM ionomycin or A23187) in the presence of calcium chelators (5 mM EGTA or 60 mM EDTA). The results indicate that fusion can proceed at extremely low concentrations of intravesicular and extravesicular Ca2+. However, BAPTA [1,2-bis-(o-aminophenoxy)ethane-N,N,N',N'-tetra-acetic acid], a relatively specific Ca2+ chelator, inhibits fusion. BAPTA binds other metals besides Ca2+. We present evidence that BAPTA inhibition is due not to Ca2+ chelation but to Zn2+ depletion. TPEN [N,N,N',N'-tetrakis-(2-pyridylmethyl) ethylenediamine], another metal-ion chelator with low affinity for Ca2+, also inhibited fusion. TPEN- and BAPTA-inhibited fusions were restored by addition of Zn2+. Zn(2+)-dependent fusion presents the same characteristics as control fusion. In intact cells, TPEN inhibited transport along the endocytic pathway. The results indicate that Zn2+ depletion blocks endosome fusion, suggesting that this ion is necessary for the function of one or more factors involved in the fusion process.

Calcium-dependent fusion among endosomes.

Fusion among endosomes is an important step for transport and sorting of internalized macromolecules. Working in a cell-free system, we have previously reported that, in the absence of externally added calcium, endosome fusion requires cytosol, ATP, and is sensitive to N-ethylmaleimide (NEM) and to anti-NEM-sensitive factor (NSF) antibody. This cytosol-dependent fusion is regulated by monomeric and heterotrimeric GTP-binding proteins. Further studies have revealed, however, that in the presence of micromolar concentrations of free calcium, fusion is observed even in the absence of cytosol and ATP. At the electron microscope level, Ca(2+)-dependent endosome aggregation and fusion were similar to that observed for cytosol-dependent fusion. Calcium-dependent fusion was not affected by non-hydrolyzable analogs of GTP or GDP nor by NEM or anti-NSF antibody. However, Ca(2+)-dependent fusion was abrogated by trypsin treatment of the vesicles or by a membrane wash with 60 mM EDTA indicating that peripheral proteins are required. An anti-annexin II antibody and an annexin II peptide blocked Ca(2+)-dependent fusion by 50%. After the EDTA wash, Ca(2+)-dependent fusion was reconstituted by addition of purified annexin II and arachidonic acid. We conclude that endosomes can fuse by two mechanisms, one that has an absolute requirement for calcium and is probably mediated by annexins, and another that does not require calcium.

Fusion of newly formed phagosomes with endosomes in intact cells and in a cell-free system.

Phagosomes are membrane-bound vesicles, formed by the receptor-mediated internalization of particulate ligands, which exchange soluble and membrane proteins with other endocytic compartments as a part of their maturation process. This exchange of material is undoubtedly mediated by fusion of phagosomes with other membrane-bound compartments of the endocytic pathway. By using a particulate probe (fixed Staphylococcus aureus coated with mouse anti-dinitrophenol monoclonal antibody) localized in phagosomes and a soluble probe (dinitrophenol-derivitized beta-glucuronidase) internalized by receptor-mediated endocytosis, we have studied phagosome-endosome and phagosome-lysosome fusion in intact cells and in a cell-free system. Vesicle fusion was assessed by measuring beta-glucuronidase activity associated with S. aureus particles after lysis of the membranes. In intact macrophages, newly formed phagosomes fused with early endosomes and with lysosomes. Fusion with lysosomes was observed to commence after a short lag period of about 5 min. In broken-cell preparations, phagosomes were able to fuse with early endosomes. It was not possible to reconstitute phagosome-lysosome fusion in vitro. In vitro phagosome-endosome fusion required energy and cytosolic- and membrane-associated proteins. A nonhydrolyzable analog of GTP stimulated fusion at low cytosol concentrations and inhibited fusion at high cytosol concentrations. These observations indicate that the mechanisms mediating phagosome-endosome fusion are similar to those described for endosome-endosome fusion. Our results suggest that exchange of material with endosomes is an important step in the process of phagosome maturation.

Receptor-mediated entry of beta-glucuronidase into the parasitophorous vacuoles of macrophages infected with Leishmania mexicana amazonensis.

125I-labeled rat preputial gland beta-glucuronidase was shown by light and electron microscopic radioautography to accumulate within the parasitophorous vacuoles of in vitro derived bone marrow macrophages infected with Leishmania mexicana amazonensis. beta-glucuronidase uptake was mediated by the mannose receptor, since the penetration of the ligand was inhibited by mannan. Uptake was detected as soon as 4 h after incubation of infected cells with the ligand, and increased at 24 and 48 h. The label persisted in the vacuoles for at least 24 h after a 24-h pulse with the ligand, a finding compatible with the relatively long half-life of labeled beta-glucuronidase in normal macrophages. Parasitophorous vacuoles were also labeled in macrophages exposed to the ligand only before infection, indicating that secondary lysosomes containing the ligand fused with the parasitophorous vacuoles. Another mannosylated ligand, mannose-BSA, which, in contrast to beta-glucuronidase, is rapidly degraded in macrophage lysosomes, did not detectably accumulate in the vacuoles. The results support and extend information previously obtained with electron opaque tracers that emphasizes the phagolysosomal nature of Leishmania parasitophorous vacuoles. In addition, the results suggest that appropriate mannosylated molecules may be used as carriers for targeting of leishmanicidal drugs to the parasitophorous vacuoles of infected macrophages.