The mechanism by which trophozoites of the protozoan parasite, Entamoeba histolytica destroy mammalian cells is not clear. A variety of amoeba-associated properties have been correlated with virulence (reviewed in 1 and 2). In vitro studies (3-6) have shown that contact of the trophozoite with a target cell is required. Contact killing has been described (6) as occurring in the sequence: adherence, cytolysis, then phagocytosis (possibly) of the target or its remnants. Dependence of in vivo virulence upon the phagocytic competence of the amoeba has also been reported (7). All steps of the sequence are inhibited by cytochalasins (6,8,9), indicating that dynamic participation of the amoeba actin cytoskeleton is required. However, the role of actin in the attack process has not been defined.By fluorescence microscopy of glutaraldehyde-fixed and rhodamine-phalloidin-stained (10) trophozoites, we have revealed the organization and distribution of polymerized actin in Entamoeba involved in a variety of motility-related activities (11, 12), including target cell interactions. Furthermore, by methanol extraction and spectrofluorometric measurement of the bound fluorescence, we have been able to quantitate polymerized actin in trophozoites. In this report, we describe this latter procedure and studies of the organization and quantity of polymerized actin in E. histolytica trophozoites before and immediately after challenge with human red blood cells (RBC) 1. Within 5 s after challenge, polymerized actin appeared at the contact interface with many adherent RBC. A net increase in amoeba polymerized actin content was detectable 1 rnin after the challenge, and reached a maximum of approximately twice the value in unchallenged cells within 4 min. Latex beads, which were phagocytized by E. histolytica trophozoites, neither stimulated an actin response nor were able to diminish the interaction with RBC. RBC, on the other hand, inhibited uptake of latex beads.These results indicate that the initial interaction of E. histolytica trophozoites with target cells is a recognition-specific process that triggers rapid polymerization of amoeba actin at the site of target contact. This actin appears to be involved in phagocytosis of the target cells. The procedure developed to quanThis work was supported in part by grants AI 19023 and RR 08006 from the National Institutes of Health, Bethesda, MD.i Abbreviations used in this paper: BSA, bovine serum albumin; PBS, phosphate-buffered saline; RBC, red blood cell. 546J. ExP. M~D.

To investigate enhanced disease associated with a formalin-inactivated (FI) respiratory syncytial virus (RSV) vaccine, we studied the pulmonary inflammatory response to RSV in BALB/c mice immunized with live RSV, FI-RSV, or combinations of the two. After RSV challenge, the number of granular cells, the ratio of CD4 ؉ /CD8 ؉ lymphocytes, and the level of Th2-like cytokine mRNAs in the bronchoalveolar lavage specimens in mice immunized first with live RSV and then with FI-RSV were lower than that in FI-RSV-immunized mice and close to that in live RSV-immunized mice. These data suggest that prior live RSV infection prevents most of the enhanced inflammatory response seen in FI-RSV-immunized mice and might explain lack of enhanced disease in older FI-RSV-immunized children. A live RSV vaccine might similarly decrease the risk of enhanced disease with non-live RSV vaccines.

A micropore membrane procedure to assay taxis by Entamoeba histolytica is described and the results of studies of responses to a variety of soluble substances, bacteria, an rat colon washings using this procedure are reported. Trophozoites migrated in blind well chambers through 8-micron pore size polycarbonate membranes but not nitrocellulose membranes up to 12 micron pore size. Amoebae were attracted toward fresh axenic culture medium (TYI-S), an enzymatic hydrolysate of casein (Trypticase), and a partially purified preparation of N-acetylneuraminic acid from egg mucin, but not purified N-acetylneuraminate or a variety of other low molecular weight metabolites. The response was verified as chemotaxis by checkerboard analysis. Amoebae migrated most dramatically toward suspensions of all of seven bacterial species tested, including motile and non-motile, gram-negative and gram-positive rods and cocci. This response was diminished when the bacteria concentration gradient was eliminated. The response to bacteria culture filtrates was less than 10% of that to bacterial suspensions. A response to clarified washings from the rat colon was detected; this was diminished but not eliminated by filter sterilization of the washings. We concluded that some soluble molecules, possibly of intermediate molecular size, whole bacteria, and both soluble and particulate components of the rat colon provide tactic stimuli for E. histolytica. Scanning electron micrographs of trophozoites migrating towards attractants through membranes showed narrow, extended pseudopodia entering the membrane pores, and enlarging spheres exiting as the cells proceeded through.

The identity of molecules of mammalian target cells that stimulate contact-dependent attack by Entamoeba histolytica was sought using human erythrocytes (RBC) as a model. Protein-free liposomes prepared from RBC membrane lipids stimulated the same rapid E. histolytica actin polymerization and phagocytosis as did whole target cells. Liposomes constructed from the major phospholipids of RBC stimulated these responses but only if a negatively charged phospholipid was included. The addition to these liposomes of digalactosyl diglyceride significantly enhanced their stimulatory activity. The results demonstrate that ligands that trigger attackrelated responses by E. histolytica reside in the target cell membrane lipid fraction and suggest roles for both glycolipid and phospholipid components.

The dynamics of cytoplasmic streaming, retrograde translocation of externally bound particles and locomotion by Entamoeba invadens were compared. Locomoting amoebae were monopodial, exhibited fountain flow cytoplasmic streaming and translocated externally bound erythrocytes to the rear of cells. The rates of rearward flow of peripheral cytoplasmic vacuoles and of the externally bound particles were equal to the rate of cell forward locomotion. Rhodamine-phalloidin staining revealed a distinct cortical polymerized actin cytoskelton. This was least evident about the periphery of the advancing pseudopod, increased in density toward the rear of the cell and was most concentrated in the uroid. A monoclonal anti-eucaryotic actin antibody, which recognized monomeric Entamoeba actin on immunoblots, stained trophozoites by indirect immunofluorescence throughout the cytoplasm, but not in the cortical regions stained by rhodamine-phalloidin. This and other evidence implied that the antibody recognized only unpolymerized actin in Entamoeba. We propose that locomotion, cytoplasmic streaming and translocation of externally bound particles are driven by a common actin-based mechanism in Entamoeba, possibly involving retrograde cortical actin flow and recycling.