Mechanism of Entry and Development of Trypanosoma Dionisii in Non-Phagocytic Cells

J. Cell Sci. 56, 371-387 (1982) 371 Printed in Great Britain © Company of Biologists Limited 1982

MECHANISM OF ENTRY AND DEVELOPMENT OF TRYPANOSOMA DIONISII IN NON-PHAGOCYTIC CELLS

AUDREY M. GLAUERT Strangeways Research Laboratory, Worts' Causeway, Cambridge, CBi +RN, England J. R. BAKER* AND LINDA F. SELDEN Medical Research Council Biochemical Parasitology Unit, Molteno Institute, Downing Street, Cambridge, CBi 3EE, England

SUMMARY Cultures of 'buffalo' (bison) lung (BL) cells were infected with epimastigotes or trypomastigotes of Trypanosoma (Schizotrypanum) dionisii derived from cultures in vitro, fixed after various periods of incubation at 37 °C and examined by light or electron microscopy. Few if any epimastigotes entered the BL cells, but many trypomastigotes did so; they adhered to the cell surface within 2 h and then appeared to sink into furrows on the cell surface until engulfed in parasitophorous vacuoles. Cytochalasin D (5-10 fig ml"1) completely, but rever- sibly, inhibited entry of trypomastigotes without affecting parasite motility. It was concluded that entry depended on the interaction of stage-specific components on the trypomastigote'a surface with receptors on the BL cells, and that this interaction induced active uptake of the protozoa by a phagocytic process not involving pseudopod formation. Soon after entry of the trypomastigotes into BL cells, the membranes of the parasitophorous vacuoles disintegrated and the parasites, which were now lying free in the cytoplasm of the host cell, transformed into amastigotes (micromastigotes) during the next 24-48 h. Replication then occurred, followed by transformation, beginning after 3 days, through a transitional promastigote phase to small intracellular trypomastigotes at 7 days. The promastigotes had a characteristic curved protrusion extending from the lip of the flagellar pocket (or reservoir) into the host cell's cytoplasm. Trypomastigotes, released into the supernatant medium by rupture of the plasma membranes of the BL cells after 8 days, could re-invade other cells.

INTRODUCTION Trypanosoma (Schizotrypanum) dionisii, a parasite of Chiroptera (bats) in Europe (Baker, Green, Chaloner & Gaborak, 1972), is related to Trypanosoma cruzi, the etiological agent of Chagas's disease (Baker, 1976). There is no evidence that T. dionisii is pathogenic for man, and consequently it is a useful organism in studies relevant to Chagas's disease (Baker, 1976; Baker & Selden, 1981). The entry of T. dionisii into mouse peritoneal macrophages, and its subsequent intracellular development, parallel those of T. cruzi. It replicates as amastigotes after entry by a cytochalasin-inhibited process (presumably phagocytosis), following

• Present address: Culture Centre of Algae and Protozoa, (Natural Environment Research Council, Institute of Terrestrial Ecology), 36 Storey's Way, Cambridge CB3 oDT, England. 372 A. M. Glauert, J. R. Baker and L. F. Selden attachment to non-specific trypsin-sensitive receptors on the macrophage membrane (Baker & Liston, 1978). After about 1 week, intracellular differentiation to trypomastigotes occurs and these are subsequently liberated into the medium (Baker et al. 1972; Baker & Liston, 1978). As not all the natural host cells of T. cruzi in vivo are phagocytes, it seemed worthwhile to study comparatively the interaction of T. dionisii with cells that are not normally phagocytic. The availability and use of 'buffalo' (bison) lung (BL) cells, which are not 'professional' phagocytes, as a 'feeder layer' in cultures of haematozoic trypomastigotes of T. brucei (Hill, Shimer, Caughey & Sauer, 1978) prompted us to use them in a comparative study, the results of which are reported in this paper. A brief account has already been published (Baker, Glauert & Selden, 1980).

MATERIALS AND METHODS Trypanosomes Stock P2 of T. dionisii (primary isolate no. LONDON/71/BPUC/2; Baker, 1980) was maintained serially in vitro in L4NCS liquid medium (Baker & Selden, 1081) at 28 °C. Trypomastigotes, from cultures about 1-2 weeks old, were separated (> 97% pure) from other forms by passage through diethylaminoethyl cellulose (Selden & Baker, 1980). Epi- mastigotes (> 90% pure) were harvested from 3-day-old cultures before differentiation to trypomastigotes had occurred.

Cells 'Buffalo' lung (BL) cells (Bu(IMR-3i), Bison bison, ATCC no. CCL40; Flow Laboratories, Irvine, KA128NB, Scotland; catalogue no. 05541) were grown at 37 °C in tissue-culture medium 199 (Wellcome Reagents Ltd, Beckenham, Kent, BR3 3BS, England; catalogue no. TC45) P'118 2O % (v/v) foetal bovine serum (FBS; Flow Laboratories; catalogue no. 29—1 o 1), inactivated at 56 CC for 30 min (hereinafter called 'complete medium').

Light microscopy Cultures of io4 BL cells and 5 x 10* trypomastigotes or epimastigotes in 1 ml complete medium were set up in bottomless miniature Carrel flasks (Baker et al. 1972) cemented to microscope slides with polyvinyl acetate S7 (Cairn Chemicals Ltd, 5-9 Pembroke Road, Ruislip, Middlesex, HA4 8NQ, England) dissolved in acetone (approx. 30%, w/v). After incubation at 37 CC, cultures were washed with medium 199 without serum, fixed for 30 min in Bouin's fluid, washed with 70 % ethanol and stained by the Giemsa-colophonium method (Bray & Garnham, 1962). Daily counts were made on an improved Neubauer haemocytometer of the numbers of extracellular trypanosomes in the supernatant medium from a 10 ml culture of the type used for electron microscopy (see below). Observations were also made on living cultures of BL cells and trypomastigotes in Wild ' Selecta' tissue culture observation chambers (Leitz, Luton, LUi 3HP, England), kept at 37 °C, examined by phase-contrast microscopy and photographed by electronic flash (Liston & Baker, 1978).

Electron microscopy Cultures were grown in 25 cm* plastic tissue-culture flasks (Nunclon Delta, catalogue no. 163371, obtained from Gibco Europe Ltd, P.O. Box 35, Paisley, PA3 4EP, Scotland; or Corning, catalogue no. 25100) containing 10 ml complete medium. When cell growth was about 50 % confluent or more (usually after 5—9 days at 37 °C), between 4-4 x 10' and I-I x io8 washed trypomastigotes were added in 1 ml complete medium. After various periods of Trypanosoma dionisii in non-phagocytic cells 373 incubation at 37 °C, the medium was poured off and replaced with a fixative consisting of 2'5 % (w/v) glutaraldehyde in 0-09 M-cacodylate buffer (pH 7-2) containing 3 mM-calcium chloride. Samples were fixed for 30 min at room temperature and then washed with o-i M- cacodylate buffer (pH 7-2) containing 3 mM-calcium chloride. Fixed cultures were stored in buffer at 4 °C. To remove the cells the necks of the culture flasks were cut off and the monolayer scraped out with a rubber ' policeman'. The cells were centrifuged into a pellet, post- fixed in cacodylate-buffered 1 % (w/v) osmium tetroxide for 1 h at room temperature, washed briefly in distilled water, stained with 0-5 % (w/v) uranyl acetate in water for ih at room temperature, dehydrated in ethanol and embedded in Araldite. Thin sections were cut on a Cambridge Huxley Ultramicrotome, Mark 2, and stained with lead citrate. Specimens were examined in an AEI EM6B or a Philips 201C electron microscope operating at 60 kV.

Cytochalasin treatment Cytochalasin D was used in preference to cytochalasin B since it affects microfilaments without also inhibiting hexose transport (Miranda, Godman, Deitch & Tanenbaum, 1974 a; Miranda, Godman & Tanenbaum, 19746). A stock solution of 1 mg ml"1 cytochalasin D (Sigma, C 0263) in dimethyl sulphoxide (DMSO; BDH, catalogue no. 28216) was added to complete medium to give a final concentration of 1, 5 or 10 /(g ml"1. The diluted solution replaced complete medium in some of the miniature Carrel flask cell cultures (described above) for 1-1-5 h, after which it was removed and replaced with 1 ml of a suspension of trypomastigotes (1 x 10* to 2 x io* ml"1), either in complete medium alone or in fresh cytochalasin solution. Control cultures received solutions of DMSO (0-5 or 1 %, v/v) instead of cytochalasin. After a further 1-2 h at 37 °C, cultures were fixed and stained as described for light microscopy. The various treatments are listed in Table 2 (see Results).

Heat treatment Suspensions of trypomastigotes (5 x io* to 6 x io' ml"1) were immersed in a water bath at 60 °C for 5 min, before addition to Carrel flask cell cultures. After 2 h at 37 °C, the cultures were fixed and stained.

Enumeration of slides Slides were coded so that their treatment was unknown to the enumerator. The proportions of infected cells and of those containing more than 10 parasites were determined in 10 microscope fields (magnification x 500) at predetermined sites distributed evenly over the surface of each stained monolayer (a circle about 1 cm in diameter). Triplicate cultures were examined and means with their standard errors (S.E.M.) were calculated (Baker & Liston, 1978).

RESULTS Entry of trypanosomes into ' buffalo' lung cells Many more parasites entered BL cells when the inoculum consisted largely of trypomastigotes, than when it contained mainly epimastigotes (Table 1). Prior heating of trypomastigotes, which immobilized them, totally prevented their entry. Cytochalasin D (5 or io/jgml"1 in 0-5 or 1% DMSO) also abolished entry into cells treated for 1-1-5 h before, and 1 or 2 h after, addition of trypomastigotes (Table 2), although it had no observable effect on the motility of the protozoa. Cytochalasin D at 1 jig ml"1 reduced, but did not completely prevent, entry. When the drug was removed before the addition of trypomastigotes, entry was similar to that with DMSO alone. Observations by light and electron microscopy showed that trypomastigotes Table I. Infection rates in cultures of BL ceUF given various i& b

Cells with > 10parasites Q Infected cells (%) (%) a" L r -, - r Parasite Number m1-l n Time -x 8.~. X 8.E.

Trypornastigotea : Normal 6 x 10' 4 zh 40'50 4'23 Heated 6 x 108 4 zh o - Trypornastigotea : Normal I x roe 3 zh 13.60 2.1 I Heated 0.5 x 10' 3 zh o - .. Trypomaatigotes 5 x 10' 3 3 d 93'74 0.43 63-68 I '78 Epirnastigotea 14-54 1.81 1'23 123 3 5x roa 3 3 tn f, mean; s.E., standard error of mean; n, number of replicate cultura (all at 37 OC); heated, immersed in water bath at 60 OC for 5 min. ii Table 2. Eflect of cytochalasin D on infection rates in cultures of BL ceUr and T.dionisii ttypomattigot~~

Cells Infected cells with > 10parasites ( %) (%I A A r 1 r \ & & Comp. Concn Time (h) Comp. Concn Time (h) 2 s.E. f 8.E. - Cyt D 10yg ml-' 1-1.5 CY~D 10yg ml-' I 0 - 0 DMSO 1 % 1-1-5 DMSO 1 % I 0.9.5 0.29 I '00 o Medium - 1-1'5 Medium - I 1 '49 0.23 I .04 0.04

- Cfi Il 5 PI3 ml-l 1-1'5 Cyt l3 5 pg- ml'l 2 0 0 Cyt D 5 PB ml-' 1-1.5 Medium 2 18.91 4'76 1'54 0.19 Cfi El I pg d-' 1-1'5 Cfl r, I pg ml-' 2 2'33 0.80 1.78 0.22 DMSO 0.5 % 1-1.5 DMSO 0.5 % 2 I 3.60 2.11 1.41 0.15

Pre- and post-treatment refer to periods before and after addition of I x 10' to 2 x xo6 trypomastigotes; Comp., com- pound; concn, concentration; %, mean; S.E., standard error of mean; Cyt D, cytochalasin D in DMSO; DMSO, dimethyl sulphoxide. Number of cultures per treatment = 3 (at 37 "C). 376 A. M. Glauert, J. R. Baker and L. F. Selden

Figs, i, 2. Phase-contrast photomicrographs of cultures of BL cells and T. dionisii trypomastigotes incubated at 37 °C. Bars, 10 fim. Fig. 1. Living culture, 100 min after addition of trypomastigotes; trypomastigotes (arrows) are attached to the cell surface; electronic flash. Fig. 2. Culture fixed with glutaraldehyde 2 h after addition of trypomastigotes; 2 trypomastigotes (arrows) are apparently lying in furrows in the cell surface, or are in parasitophorous vacuoles.

adhered to the surfaces of the BL cells within 2 h (Figs. 1-3). The parasites were aligned parallel to the surface of the cell (Fig. 3). During entry the trypomastigotes appeared to sink into the cell so that they were seen in furrows in the cell surface (Figs. 2, 4), which subsequently closed around them (Fig. 5). Some of the small vesicles associated with the cell surface, which are a feature of BL cells, had fused with the plasma membrane in regions of contact with parasites (Fig. 4, arrow) and surrounded developing parasitophorous vacuoles (Fig. 5, v). The trypomastigotes appeared to enter the BL cells with some force, since parasites were observed that had caused an indentation of the nucleus (Fig. 6).

Figs. 3-5. Electron micrographs of thin sections illustrating the entry of trypomastigotes of T. dionisii into BL cells. Bars, 1 fim. Fig. 3. A trypanosome with a kinetoplast (ft) characteristic of a trypomastigote is adhering parallel to the surface of a BL cell. The cell cytoplasm contains mito- chondria (m), cisternae of the extensive Golgi apparatus (g) and microfilaments, some of which are organized in bundles (mf) near the cell surface. Small vesicles (v) are associated with the cell surface of the cell in contact with the trypanosome. Incubation, 2 h. Fig. 4. A trypanosome, seen in transverse section, lies within a dip or furrow in the surface of a BL cell. One of the small vesicles (v) near the cell surface has fused with the plasma membrane (arrow). Incubation, 6 h. Fig. 5. A group of trypanosomes in a BL cell is almost completely enclosed within a vacuole, which is surrounded by small vesicles (v). g, Golgi apparatus; n, nucleus of BL cell. Incubation, 2 h. Trypanosoma dionisii in non-phagocytic cells 377

Development of trypanosomes within BL cells In cultures receiving trypomastigotes, most of the intracellular parasites remained in this form for 24 h (Fig. 7). They then were transformed into amastigotes, within obvious vacuoles, and replicated by binary fission (Fig. 8). The kinetics of infection are illustrated in Fig. 9. The proportion of cells containing parasites increased during the first 3 days of culture (Fig. 9 A), as trypomastigotes continued to enter the cells. This was followed by a decline during the following 2 days, probably largely or entirely as a result of multiplication of uninfected BL cells. Replication of amastigotes, 13 CEL56 378 A. M. Glauert, J. R. Baker and L. F. Selden 6

Fig. 6. Electron micrograph of a recently entered trypomastigote, with kinetoplast (k) and flagellum (/). The parasite has indented the nucleus (n) of the BL cell. Incubation, 6 h. Bar, i /tm. indicated by the proportion of cells containing more than io parasites, reached a peak at day 6 (Fig. 9B). Amastigotes transformed into small trypomastigotes from day 7 onwards (Fig. 10). The trypomastigotes were released into the medium, presumably by rupture of the host cells, and some at least re-invaded BL cells. The proportion of infected cells consequently increased between days 7 and 8 (Fig. 9A). In 10 experiments the grand mean percentage of BL cells containing trypanosomes after 5 to 7 days was 76-7 (s.E.M. = 42). The grand mean percentage of these cells containing more than 10 parasites was 778 (S.E.M. = 3-2). The number of extracellular trypomastigotes declined steeply during the first 6 days of culture, but then increased as trypomastigotes emerged from host cells on days 7 and 8 (Fig. 11). In order to confirm re-entry of released trypomastigotes into uninfected BL cells Trypanosoma dionisii in non-phagocytic cells 379

: *

Figs. 7, 8, io. Light micrographs of Giemsa-stained monolayers of BL celb infected with T. dionisii and incubated at 37 °C. Bars, 10 fim. Fig. 7. Cell fixed 24 h after addition of trypomastigotes; intracellular parasites are still mostly in this form. Fig. 8. Cell fixed after 6 days; amastigotes (some dividing) are present in con- spicuous vacuoles. Fig. 10. Cell fixed after 8 days; intracellular parasites are predominantly trypomastigotes. the supernatant medium of one infected cell culture was changed on day 6, to remove as many as possible of the extracellular trypanosomes persisting from the inoculum. Trypanosomes (mainly trypomastigotes that had emerged from cells) were harvested from the supernatant of this culture on day 10, washed and added (at yx io'ml"1) to six uninfected cell cultures; three were fixed 3 days later, and three were fixed after 6 days. About one-third of the recipient cells became infected: on day 3, mean = 33-5% (S.E.M. = 70) and on day 6, mean = 30-3% (S.E.M. = 1*7%). About one-half of the infected cells contained more than 10 parasites: on day 3, mean = 44-9% (S.E.M. = 4-8) and on day 6, mean = 53-5% (S.E.M. = 7-6). 13-2 38o A. M. Glauert, J. R. Baker and L. F. Selden

100 90 80 70 g 60 f 50 o 40 30 20 10 1 2 3 4 5 6 7 8 9 12 3 4 5 6 7 8 9 Days Fig. 9. Percentages of infected cells (A), and of infected cells containing more than 10 parasites (B), in cultures of BL cells at 37 °C infected with 5 x io* trypomastigotes of T. dionisii on day o. Means ± standard errors (vertical bars) of duplicate counts.

106

E S

Io c CO

105

104 1 12 3 4 5 6 7 8 9 10 Days Fig. 11. Numbers of extracellular trypanosomes (ml"1) in one 10 ml culture of BL cells infected with T. dionisii trypomastigotes on day o and incubated at 37 °C. Trypanosoma dionisii in non-phagocytic cells

Figs. 12-17. Electron micrographs of thin sections of BL cells infected with T. dionisii, illustrating development of the parasite. Bars, 1 fim. Fig. 12. A trypomastigote, with kinetoplast (ft), in the cytoplasm of a BL cell after 24 h incubation is enclosed within a membrane, which is beginning to break down (arrow). Fig. 13. The membrane surrounding an intracellular trypomastigote, with kinetoplast (k), has broken (arrow) and the parasite is in direct contact with the cytoplasm of the BL cell. This cytoplasm is vacuolated and shows signs of lysis in regions adjacent to the trypomastigote. Incubation, 24 h. A. M. Glauert, J. R. Baker and L. F. Selden

Fig. 14. After incubation for 2 days, the trypanoaomes are rounded amastigotes (or micromastigotes) with an internal flagellum (/) only and a condensed kinetoplast (k). They lie free in a vacuolated region of the cytoplasm of a BL cell. The breaks in the plasma membrane of the host cell (arrows) were probably caused during removal of the monolayer of cells from the culture flask.

Fine-structural changes during development of T. dionisii within BL cells After incubation for 24 h, the intracellular parasites were still trypomastigotes (Figs. 12, 13) and were enclosed in membranes that were presumably derived from the plasma membrane of the BL cell during entry. The membranes of the parasitophorous vacuoles were in various stages of dissolution and some of the trypomastigotes were in direct contact with the cytoplasm of the BL cell (Fig. 13). The cytoplasm surrounding these escaping parasites was vacuolated and showed signs of localized lysis. During the following 24 h most of the intracellular parasites had developed into rounded amastigotes (or micromastigotes), with no external flagellum and with a characteristic condensed kinetoplast (Fig. 14). The parasites lay within the host cell's cytoplasm in vacuoles lacking limiting membranes. From 3 days onwards elongated parasites with promastigote morphology were seen (Fig. 15). The flagella of many of these forms appeared to emerge from an eccentrically placed flagellar pocket (or 'reservoir'), and from thicker portions of the pocket's rim a broad protrusion had developed (Fig. 15). This protrusion extended, often at an acute angle, into the cytoplasm of the host cell. In some of these elongated forms a number of curved, apparently tubular, structures were present (Fig. 16, arrows). Trypanosoma dionisii in non-phagocytic cells

Fig. 15. An elongated promastigote, with a condensed kinetoplast (k) and an external flagcllum (/), is in a vacuole in a BL cell and has a broad protrusion (/>) extending into the host cell's cytoplasm. Incubation, 4 days. Fig. 16. Curved tubular structures (arrows) are present in the cytoplasm of an elongated promastigote within a BL cell. Incubation, 6 days.

Seven days after the addition of trypomastigotes to cultures of BL cells, many intracellular parasites with typical trypomastigote-type kinetoplasts were seen (Fig. 17), still lying free in the cytoplasm of the host cell. By 8 days, morphologically similar extracellular forms were seen, together with intracellular parasites of pro- A. M. Glaturt, J. R. Baker and L. F. Selden

Fig. 17. After incubation for 7 days, an intracellular parasite has a kinetoplast (k) of trypomastigote-type and lies free in the cytoplasm of a BL cell. mastigote morphology. At this stage many of the BL cells showed signs of lysis and had breaks in their plasma membranes.

DISCUSSION The infection rate of BL cells in cultures receiving epimastigotes of T. dionisii was much lower than that in those inoculated with trypomastigotes, but was not zero (Table 1). However, cultures were not fixed for enumeration until 3 days after addition of the epimastigotes, and probably some epimastigotes had transformed into trypomastigotes by this time and had then entered cells. Therefore, it is likely that epimastigotes cannot enter BL cells, and certainly none was seen by electron microscopy within cells early in the infection. Similarly, it has usually been reported that only trypomastigotes of T. cruzi penetrate into non-phagocytic cells (e.g. see Nogueira & Cohn, 1976), although some assert that epimastigotes can also do so (e.g. see Tanowitz, Winner, Kress & Bloom, 1975). The difference in the ability of trypomastigotes and epimastigotes of T. dionisii to enter BL cells suggests the presence of stage-specific membrane components on trypomastigotes, which interact with the host cell's surface to induce uptake, as proposed for T. mm and mouse peritoneal macrophages (Alcantara & Brener, 1978). The fact that heating trypomastigotes of T. dionisii prevented entry suggests that the active membrane components are heat-labile. Similar induced phagocytic uptake by normally non-phagocytic cells is known to occur with other protozoa. One example is the entry of Plasmodium merozoites into erythrocytes via an invagination of the erythrocyte's surface membrane, which may be induced by a protein secreted from the apical organelles of the merozoite (Bannister, 1977). A protein with similar activity has been found in Toxoplasma gondii and named the ' penetration-enhancing factor' (Lycke, Carlberg & Norrby, 1975). The mechanism of entry of trypomastigotes of T. dionisii into BL cells was unusual Trypanosoma dionisii in non-phagocytic cells 385 in that it did not involve the formation of enveloping pseudopodia by the host cell. Such pseudopodia were a characteristic feature of the ingestion of T. cruzi by phagocytic cells, such as mouse peritoneal macrophages (Tanowitz et al. 1975; Nogueira & Cohn, 1976; Milder & Kloetzel, 1980); they were also observed during uptake of T. dionisii by human monocytes and neutrophils (A. M. Glauert & K. J. I. Thome, unpublished observations; see Thorne, Glauert, Swennsen & Franks, 1979; Thorne et al. 1981). In contrast, the uptake of parasites into BL cells resulted from the induction of dips or furrows in the cell surface by the trypomastigotes, which appeared to sink into the host cell by a mechanism similar to that described for the interiorization of C3-opsonized sheep red blood cells into rat Kuppfer cells (Munthe-Kaas, Kaplan & Sejelid, 1976). Surface furrows have also been observed in a light microscopic study of the entry of trypomastigotes of T. cruzi into non- phagocytic, bovine embryo skeletal muscle cells (Dvorak & Hyde, 1973). In the many studies of the entry of T. cruzi into phagocytic and non-phagocytic cells a full range of mechanisms has been proposed, from a phagocytic process in which the parasite makes little contribution, to active penetration by the parasites in which the surface of the host cell is thought to ' dissolve'. The electron microscopic evidence for membrane dissolution is, however, unconvincing (e.g. see Tanowitz et al. 1975) and it is possible to classify all uptake mechanisms under the general heading of 'phagocytosis', with the parasite playing a role of varying importance. With ' professional' phagocytes, particularly activated or elicited macrophages and neutrophils, ingestion is very rapid for all types of particle, while for other cells considerable stimulation of the host cell by the parasite is necessary. Factors enhancing uptake include the presence of specific parasite membrane components, such as those thought to be present on trypomastigotes but not epimastigotes of T. dionisii, or on some strains of T. cruzi and not others (Alcantara & Brener, 1978), and the opsonization of parasites by antibody or complement (Alcantara & Brener, 1978; Kipnis, Calich & Dias da Silva, 1979). Conflicting evidence has resulted from experiments with T. cruzi and cytochalasins. In general, uptake of trypanosomes into host cells followed the pattern of ingestion of other particles and was inhibited by 5-10 /ig ml"1 cytochalasin B (e.g. see Nogueira & Cohn, 1976), but Alexander (1975) reported that uptake of epimastigotes of T. cruzi into mouse peritoneal macrophages was only partially inhibited. In contrast, Kipnis et al. (1979) found that, although uptake of epimastigotes was blocked, 10 fig ml"1 cytochalasin B had no effect on the entry of bloodstream forms of T. cruzi. These discrepancies may perhaps be explained with reference to our earlier conclusion that successful entry depends both on the parasite and on the host cell. Some combinations of the two may result in a phagocytic system so effective that it can operate, either partially or wholly, even in the presence of cytochalasins. The fact that uptake of trypomastigotes of T. dionisii into BL cells was completely inhibited by cytochalasin D, which had no visible effect on the motility of the parasites, indicated that here the host cell played an active role in the process of ingestion. It appears that motile parasites are necessary to initiate uptake, but that a normally functioning microfilament system in the BL cell is required for the 386 A. M. Glauert, J. R. Baker and L. F. Selden formation of surface furrows and the subsequent entry of the trypomastigotes into the cell. We conclude that entry of T. dionisii is a procedure of 'induced' phagocytosis and not merely active penetration by the parasites, although the parasites do play an essential role in the process. After entry into BL cells, trypomastigotes of T. dionisii soon became free of the enveloping phagosome membrane, as has previously been observed in macrophages infected with T. cruzi (Nogueira & Cohn, 1976; Milder, Kloetzel & Deane, 1976, 1977) or T. dionisii (Liston & Baker, 1978). Our observations agree with those of Milder et al. (1977) that the phagosome membrane appeared to disintegrate rather than that the parasite actively escaped from the vacuole. The subsequent development of trypomastigotes of T. dionisii in BL cells closely paralleled that of T. cruzi in macrophages (e.g. see Behbehani, 1973; Alcantara & Brener, 1978) or muscle cells (Dvorak & Hyde, 1973). The free intracellular parasites changed rapidly into rounded forms (amastigotes or micromastigotes) without passing through any distinguishable intermediate state. In the subsequent reverse transformation, an elongated flagellate form developed within the host cells before the fully differentiated trypomastigote, which could be distinguished by its kinetoplast structure. We have earlier referred to this stage as an epimastigote (Baker et al. 1980), but its shortness and, more important, its apparent lack of an undulating membrane and the near-terminal opening of the flagellar pocket, suggest that it is more correctly regarded as a transitional promastigote. The broad protrusion, which developed from the rim of the flagellar pocket, was a characteristic and previously un- reported feature of these promastigotes, and appeared to be a device for anchoring the parasite in the host cell during the process of elongation. The curved tubular structures seen near the nuclei of these promastigotes resembled the cytostomes described in epimastigotes and developing forms of T. cruzi and T. conorhini (Milder & Deane, 1969). This study of the mechanism of entry of T. dionisii into cells and its subsequent intracellular development has provided further evidence of the close similarity between T. dionisii and T. cruzi, and the consequent usefulness of T. dionisii as a non-pathogenic model for the study of Chagas's disease. We acknowledge the financial support of the Sir Halley Stewart Trust (to A.M.G.). We thank Mr R. A. Parker, Mrs Janet Atherton, Mrs G. E. Coulson and Mr K. J. Clarke for skilled technical assistance. We are grateful to the Wellcome Trust for the provision of the AEI EM6B and Philips 201C electron microscopes.

REFERENCES ALCANTARA, A. & BRENER, Z. (1978). The in vitro interaction of Trypanotoma cruzi bloodstream forms and mouse peritoneal macrophages. Acta Trop. 35, 209-219. ALEXANDER, J. (1975). Effect of the antiphagocytic agent cytochalasin B on macrophage invasion by Leithmania mexicana promastigotes and Trypanotoma cruzi epimastigotes. J. Protozool. aa, 237-240. BAKER, J. R. (1976). Species of the subgenus Schizotrypanum other than Trypanotoma cruzi and their potential usefulness in the laboratory. Tram. R. Soc. trop. Med. Hyg. 70, 126-127. BAKER, J. R. (1980). Primary isolate numbers of stocks of Trypanotoma (Schizotrypanum) species from Chiroptera in England. Sytt. Paratitol. 1, 153-154. Trypanosoma dionisii in non-phagocytic cells 387 BAKER, J. R., GLAUERT, A. M. & SELDEN, L. F. (1980). Interactions between Trypanotoma (Schizotrypanum) dionisii and 'buffalo' lung cells in vitro. J. Protozool. 27, 24A (abstract). BAKER, J. R., GREEN, S. M., CHALONER, L. A. & GABORAK, M. (1972). Trypanosoma (Schiso- trypanum) dionisii of Pipistrellus pipittrellus (Chiroptera): intra- and extracellular development in vitro. Parasitology 65, 251-263. BAKER, ]. R. & LISTON, A. J. (1978). Trypanosoma (Schizotrypanum) dionisii: effect of various agents on attachment and entry to macrophages m vitro and on morphogenesis. J. gen. Microbiol. 104, 79-89. BAKER, J. R. & SELDEN, L. F. (1981). Trypanosoma (Schizotrypanum) dionisii as a model for chemo- therapeutic studies related to Chagas's disease. Trans. R. Soc. trop. Med. Hyg. 75, 80-85. BANNISTER, L. H. (1977). The invasion of red cells by Plasmodium. Symp. Br. Soc. Parasit. »5> 27-55- BEHBEHANI, K. (1973). Developmental cycles of Trypanosoma (Schizotrypanum) cruzi (Chagas, 1909) in mouse peritoneal macrophages in vitro. Parasitology 66, 343-353. BRAY, R. S. & GARNHAM, P. C. C. (1962). The Giemsa-colophonium method for staining protozoa in tissue sections. Indian J. Malariol. 16, 153-155. DVORAK,J. A. & HYDE, T. P. (1973). Trypanosoma cruzi: interaction with vertebrate cells in vitro. I. Individual interactions at the cellular and subcellular levels. Expl Parasit. 34, 268-283. HILL, G. C, SHIMER, S., CAUGHEY, B. & SAUER, S. (1978). Growth of infective forms of Trypanosoma (T.) brucei on buffalo lung and Chinese hamster lung tissue culture cells. Acta Trop. 35, 201-207. KIPNIS, T. L., CALICH, V. L. G. & DIAS DA SILVA, W. (1979). Active entry of bloodstream forms of Trypanosoma cruzi into macrophages. Parasitology 78, 89—98. LISTON, A. J. & BAKER, J. R. (1978). Entry of Trypanosoma (Schizotrypanum) dionisii into macrophages in vitro and its subsequent fate therein. J. gen. Microbiol. 107, 253-262. LYCKE,E.,CARLBERG, K.&NORRBY,R. (1975). Interactions between ToxopZajwiatfondii and its host cells: function of the penetration-enhancing factor of Toxoplasma. Infect. Intmun. 11, 853-861. MILDER, R. & DEANE, M. P. (1969). The cytostome of Trypanosoma crust and T. conorhbri. J. Protozool. 16, 730-737. MILDER, R. & KLOETZEL, J. (1980). The development of Trypanosoma cruzi in macrophagea in vitro. Interaction with lysosomes and host cell fate. Parasitology 80, 139—145. MILDER, R., KLOETZEL, J. & DEANE, M. P. (1976). Interaction of peritoneal macrophages and Trypanosoma cruzi. Rev. Inst. Med. Trop., Sdo Paulo 18, 139. MILDER, R., KLOETZEL, J. & DEANE, M. P. (1977)- Observation on the interaction of peritoneal macrophages with Trypanosoma cruzi. II. Intracellular fate of bloodstream forms. Rev. Inst. Med. Trop., Sdo Paulo 19, 313-322. MIRANDA, A. F., GODMAN, G. C., DEITCH, A. D. & TANENBAUM, S. E. (1974a). Action of cytochalasin D on cells of established lines. I. Early events. J. Cell Biol. 61, 481-500. MIRANDA, A. F., GODMAN, G. C. & TANENBAUM, S. W. (19746). Action of cytochalasin D on cells of established lines. II. Cortex and microfilaments. J. Cell Biol. 6a, 406-423. MUNTHE-KAAS, A. C, KAPLAN, G. & SELJELID, R. (1976). On the mechanism of internalization of opsonised particles by rat Kupffer cells in vitro. Expl Cell Res. 103, 201-212. NOGUEIRA, N. & COHN, Z. (1976). Trypanosoma cruzi: mechanism of entry and intracellular fate in mammalian cells. J. exp. Med. 143, 1402-1420. SELDEN, L. F. & BAKER, J. R. (1980). Aseptic separation of cultivated trypomastigotes from epimastigotes of Trypanosoma (Schizotrypanum) dionisii, using DEAE cellulose. Trans. R. Soc. trop. Med. Hyg. 74, 406-407. TANOWITZ, H., WITTNER, M., KRESS, Y. & BLOOM, B. (1975). Studies of in vitro infection by Trypanosoma cruzi. I. Ultrastructural studies on the invasion of macrophages and L-cells. Am.jf. trop. Med. Hyg. 24, 25-33. THORNE, K. J. I., GLAUERT, A. M., SWENNSEN, R. J. & FRANKS, D. (1979). Phagocytosis and killing of Trypanosoma dionisii by human neutrophils, eosinophils and monocytes. Para- sitology 79, 367-379- THORNE, K. J. I., GLAUERT, A. M., SWENNSEN, R. J., THOMAS, H., MORRIS, J. & FRANKS, D. (1981). Evasion of the oxidative microbicidal activity of human monocytes by trypomastigotes of Trypanosoma dionisii. Parasitology 83, 115-123. {Received 8 December 1981)