Cell Biology of Trypanosoma cruzi

INTERNATIONAL REVIEW OF CYTOLOGY. VOL. 86

Cell Biology of Trypanosoma cruzi WANDERLEYDE SOLJZA Laboratdrio de Ultra-estrutura Celular, Departamento de Biofisica Celular, Instituto de Biofisica, Universidade Federal do Rio de Janeiro, Ilha do Fundao, Rio de Janeiro, B r a d I. 11. 111. IV.

V. VI. VII.

VIII.

IX . X. XI. XII. XIII.

Introduction . . . . . . . . .................. ............................. Developmental Forms ........................ Life Cycle of Trypano ...................... The Cell Surface . . . . . . . . . . . . A. The Plasma Membrane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Surface Charge.. . . . . . . . . . . . . . . . . . . C. The Surface Coat .................................. D. Freeze-Fracture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Mobility of Membrane Components. . . F. Biochemical Analysis of the Cell Membrane.. . . . . . . . . . . . . . . G. Isolation of the Plasma Membrane . . . . . . . . . . . . . . . . . . . . . . . . H. Lipids on the Membrane.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. Lysis of Trypanosoma cruzi . . . . . . . . . . The Subpellicular Microtubules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Microfilaments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Flagellum . . . . . . . . ....................... A. Flagellar Movement.. . . . . . . . . B. Isolation of the Flagellum . . . . . C. Basal Body ................................... Kinetoplast-Mi non . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Kinetoplast . . . . ....................... B. Ultrastructure of the Kinetoplast. ......................... C. Kinetoplast DNA . . . . . . ............ D. Replication of the Kinetoplast.. .......................... The Cytostome, Pinocytotic Vesicles, Lysosomes, and Multivesicular Structures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electron-Dense Granules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .................. Peroxisome-like Organell Endoplasmic Reticulum, The Nucleus.. . . . . . . . . . . . . . . References .........................................

197 198 202 207 207 213 222 226 229 232 238 248 249 252 254 255 258 258 262 264 273

I. Introduction Among the protozoa of the Trypanosomatidae family a large number of species represent agents of diseases such as Chagas’ disease (American tryI97 Copyright 6 1984 by Academic Press, Inc. All rights of reproduction in any form reserved.

ISBN 0-12-3614864

198

WANDERLEY DE SOUZA

panosomiais), sleepness disease (African trypanosomiasis), Leishmaniasis, etc. and therefore they are of considerable medical and veterinary importance. For this reason this group of protozoa has been subject of several studies in the last years. Special programs of investigation in this area have also been supported by the World Health Organization. Trypanosoma cruzi is the agent of Chagas’ disease. It has a complex biological life cycle which is characterized by the presence of several developmental stages which can be observed in the vertebrate and in the invertebrate hosts. The aim of this article is to review some aspects of the cell biology of T. cruzi, giving emphasis specially to those aspects related to the ultrastructure of this pathogenic protozoa. However, at some points references will be made to biochemical and immunological aspects of the parasite as well as to results obtained with other members of the Trypanosomatidae family. 11. Developmental Forms

Protozoa of the Trypanosomatidae family show, during their,life cycle, several forms which can be easily identified by light microscopy in Giemsa-stained preparations. The definition of these forms is based on (1) the general form of the cell, ( 2 ) the position of the kinetoplast in relation to the nucleus, and (3) the region where the flagellum emerges from the flagellar pocket (Hoare and Wallace, 1966). According to these criteria the following forms have been identified: trypomastigote, epimastigote, spheromastigote (amastigote, micromastigote), promastigote, paramastigote, opisthomastigote, and choanomastigote. In the case of T. cruzi the following forms can be identified. (1) Amastigote, also known as spheromastigote, micromastigote, aflagellate form, or leishmania1 form. According to the nomenclature proposed by Hoare and Wallace (1966) the intracellular multiplicative form of T. cruzi and Leishmania is designed as amastigote. However, observations made with phase contrast and electron microscopy show that these forms possess a short flagellum (Figs. l and 2 ) . Therefore, we prefer to call them spheromastigotes (Meyer and De Souza, 1976); Brack ( 1968) designated as spheromastigotes rounded forms observed in the stomach of the invertebrate host. However, further ultrastructural studies are necessary to exclude the possibility that they represent trypomastigotes which became rounded as a consequence of environmental conditions existing in the invertebrate host. More recently Piras et al. (1982) designated as spheromastigotes some rounded forms obtained when tissue culture derived trypomastigotes are incubated for 18 hours at 37°C in a cell-free medium. Rounded forms which multiply outside cells from a established cell line of Triatoma infesrans forming large clusters have also described and designated as staphylomastigotes (Lanar, 1979). (2) Epimastigotes, which are spindle-shaped

FIG. 1 . General aspect of a neuron infected in vitro with Trypanosoma cruzi. Most of the cytoplasm of the cell is occupied by intracellular parasites. X2300. (Courtesy of H . Meyer.)

200

WANDERLEY DE SOUZA

FIG. 2. General aspect of the intracellular multiplicative form of T . cruzi which possesses a short flagellum (F). Part of the mitochondrion (*) is observed. The kinetoplast (K) is seen within the mitochondria1 matrix. N, Nucleus. X 20,000. (After De Souza and Souto-Padron, 1980.)

organisms, 20-40 pm long. The kinetoplast is located anterior to the nucleus. In T. cruzi these forms are observed at the logarithimic phase of growth of T. cruzi maintained in acellular cultures and in the intestine of the invertebrate host. Epimastigotes are also found within vertebrate cells during the end of the intracellular cycle when the spheromastigotes transform into trypomastigotes or vice versa when in the beginning of a new cycle the trypomastigotes transform into spheromastigotes (Meyer and De Oliveira, 1948). They are able to divide. (3) Trypomastigotes. These forms have a length of about 25 pm and a diameter of about 2 pm. The kinetoplast is located posterior to the nucleus (Fig. 3). They can be observed (1) in the tissue cells and in the blood of the vertebrate host, (2) in the posterior intestine, in the feces, and in the urine of the invertebrate host, (3)

CELL BIOLOGY OF TRYPANOSOMA CRUZI

20 1

FIG. 3. General aspect of the trypomastigote form of T . cruzi as seen with the high voltage electron microscope. K, Kinetoplast. X 12,000.

202

WANDERLEY DE SOUZA

at the stationary phase of growth in axenic cultures, and (4) in the liquid phase of cell cultures. These forms are not able to divide. 111. Life Cycle of Trypanosoma cruzi

In the life cycle of T. cruzi there are forms which are able to divide and one form considered to be highly differentiated and responsible for the infectivity of this protozoa, which does not divide. The biological cycle starts when the invertebrate host feeds on the vertebrate host by sucking blood. The invertebrate hosts are Hemiptera and Reduvidae such as Rhodinus prolixus, Triatoma infestans, Panstrongylus megistus, and others. During feeding, the trypomastigote forms in the blood of the infected vertebrate host are ingested by the insect. It has been assumed that in the stomach of the insect most of the bloodstream trypomastigotes transform into epimastigotes and some rounded forms. In the intestine the epimastigotes divide repeatedly by a process of binary fission and can attach to the intestinal cells by hemidesmosomes (Zeledon et al., 1977), and in the rectum a certain proportion of the epimastigotes transform into metacyclic trypomastigotes which are eliminated with the feces and are able to infect the vertebrate host (Brener and Alvarenga, 1976; Brener, 1973; Brack, 1968; Dias, 1934; Zeledon e t a / . , 1977). However, more recently some new data have been obtained which show that the forms of the parasite which pass through the digestive tube of the invertebrate host are more varied than described above. It was shown that the trypomastigote form can transform into a rounded form which possesses a free flagellum. This form, which appears in the stomach, is able to transform into either short epimastigotes which start a process of multiplication in the intestinum or into long epimastigotes which move to the more posterior region of the digestive tract of the bug. Apparently these long epimastigotes are unable to divide. Some days after the ingestion of blood, spheromastigotes which are able to transform into trypornastigotes are found in the rectum. Relying on the information available up to the present moment it seems that both, epimastigotes and the spheromastigotes, are able to transform into trypomastigotes (Brack, 1968; Brener and Alvarenga, 1976). Little is known about the influence of substances in the stomach and in the intestine of the invertebrate host on the rate of division and differentiation of T . cruzi. The trypomastigote forms eliminated in the feces and the urine of the invertebrate host are able to penetrate into vertebrate cells where they transform into spheromastigotes. These divide repeatedly by binary fission and give rise to trypomastigotes which are liberated into the intercellular space, may reach the bloodstream, and can penetrate into other cells initiating a new cycle. Contrary to what is known from the African trypanosomes, the trypomastigote form of T .

CELL BIOLOGY OF TRYPANOSOMA CRUZI

203

cruzi does not divide in the bloodstream. Two morphological types of bloodstream trypomastigotes have been observed: one is slender, with an elongated nucleus, a subterminal kinetoplast, and a short free flagellum; the other is broad, with an oval nucleus, an almost terminal kinetoplast, and a long free flagellum. The predominance of one form over the other is dependent on the strain of T. cruzi and the time of infection. It has been suggested that the slender forms are mainly responsable for the infection of the vertebrate cells while the broad forms are more able to infect the invertebrate host (for a review see Brener, 1973). A better understanding of the life cycle of T. cruzi in vertebrate cells was possible thanks to the pioneering work of Kofoid et al. (1937) and Meyer and collaborators (Romafia and Meyer, 1942; Meyer and Xavier de Oliveira, 1948) who showed that it was possible to reproduce the life cycle of T. cruzi in v i m in tissue cultures. This experimental system was also used in the last years by many others, among them Dvorak and co-workers (Dvorak, 1976; Dvorak and Hyde, 1973; Dvorak and Poore, 1974; Hyde and Dvorak, 1973). The results obtained by all these authors show that once the trypomastigotes enter the cells they maintain their motility for a certain period and collide with the organelles of the host cell, and then progressively assume a rounded to oval shape. Examination of parasites recently penetrated show that although presenting a rounded shape they may have a long flagellum. However, after some hours this flagellum has shortened, reaching a length of about 1 pm. During the trypomastigote- spheromastigote transformation there is an intermediary phase which is designated as epimastigote, although this form is somewhat different from the epimastigote observed in axenic cultures. The spheromastigote form remains unchanged increasing in size, however, for about 35 hours after which time it starts a process of binary division. The significance of this long lag period in the intracellular lyfe cycle of T . cruzi is not yet completely understood. It is influenced by the growth temperature of the vertebrate cells. At higher temperature (38°C) it is shorter than at lower temperature (29°C) (Dvorak, 1976). Recent autoradiographic studies indicates that DNA synthesis occurs during the lag period (Crane and Dvorak, 1979). Using a special cell line of fibroblasts which, due to the loss of hypoxanthine-guanine phosphoribosyltransferase, are unable to incorporate guanine into nucleic acids it was shown by autoradiography that parasite RNA synthesis occurs only 2 hours after the penetration of the parasite inside the vertebrate cell (Crane and Dvorak, 1980). Therefore the synthesis of new RNA does not appear to be required for the process of trypomastigote- spheromastigote transformation (Crane and Dvorak, 1980). These studies also showed that extracellular trypomastigotes actively synthesize RNA. Before division there is a gradual increase in the size of the spheromastigotes. The kinetoplast and the nucleus divide, a new flagellum is formed, and clefts appear in the anterior and posterior regions of the cell which subsequently divides the parasite. The doubling time of the parasite is about 14 hours, although

204

WANDERLEY DE SOUZA

it can be influenced by the temperature of cell growth and is dependent on the strain of the parasite used. The process of cytokinesis requires about 25 minutes for completion (Dvorak, 1976). After the first division other divisions occur so that after a few days many parasites are found inside the host cell. The parasites then start a process of transformation into trypomastigotes via epimastigotes. According to Meyer and Xavier de Oliveira (1948) the number of trypomastigotes produced in a host cell is basically dependent on its volume. Large cells offer conditions for many parasite divisions whereas small cells after few parasite divisions are completely filled. Dvorak (1976) suggests that “the number of parasites produced within a host cell may be under parasite gene control, i.e., differentiation to trypomastigotes is dependent upon the completion of a programmed number (about 9) of divisions. Further studies are certainly necessary to clarify this point. The spheromastigote-trypomastigote transformation takes several hours. During the transformation, changes occur in the general organization of the cell, in the structure of the kinetoplast, and in the flagellum (Meyer, 1968). This latter structure increases in lenghth from 1 to 20 bm. The population of intracellular parasites does not differentiate completely synchronously. Therefore, at a certain time all transitional stages between spheromastigotes and trypomastigotes can be found in one cell. At the end of the intracellular cycle the movement of the parasites, which have now long flagella, is intense and helps in the rupture of the host cell and the liberation of the parasites which are then able to infect new vertebrate cells. When bloodstream trypomastigotes or parasites obtained from the invertebrate host are inocculated into axenic culture media maintained at 28°C they transform into epimastigotes which are able to multiply. About 6x107 cells/ml can be obtained at the stationary phase of growth (Camargo, 1964; Chiari, 1974). Such a system, which is easily reproducible, has been used to obtain large amounts of T . cruzi cells for biochemical and immunological studies. Although no detailed comparative studies have been undertaken between these epimastigotes and those observed in the intestine of the invertebrate host it has been assumed that they are equivalent. A major criticism to the use of these forms as representative of T . cruzi is the fact that they are not infective for vertebrates. When inoculated into mice or incubated in vitro in the presence of normal mice macrophages, they are phagocytised and digested (Nogueira and Cohn, 1976). Therefore, it would be more interesting to work with either bloodstream trypomastigotes or with intracellular spheromastigotes or both, which are the forms found in the vertebrate host and directly responsible for Chagas’ disease. However, it is difficult to obtain large enough amounts of these forms for biochemical studies, although some progress has been achieved in the last years (Carvalho et al., 1981; Gutteridge et al., 1978; Leon et al., 1979; Loures et al., 1980; Sanderson et al., 1980; Schmatz and Murray, 1982; Villalta and Leon, 1979; Murray et al., 1981). However, few laboratories are equipped to work with large quantities of these ”

CELL BIOLOGY OF TRYPANOSOMA C R U Z

205

pathogenic forms. Therefore, it is inevitable to continue with the use of epimastigotes for biochemical and immunological studies. It is important to point out here that no major differences have been detected in the few comparative biochemical and cytochemical studies carried out with epimastigotes and trypomastigotes (Gutteridge and Rogerson, 1979; Meirelles et al., 1982). It has been observed that when vertebrate cells maintained in culture are heavily infected with T. cruzi some of the trypomastigotes liberated into the culture medium transformed into epimastigotes which are able to multiply at 37°C (El-On and Greenblatt, 1977; H. Meyer, personal communication). Unfortunately no further data exist about these epimastigotes. It would be important to characterize them in detail. In acellular culture medium maintained at 28°C some of the epimastigotes transform into trypomastigotes mainly at the end of the logarithimic phase of growth, as have been shown by Camargo (1964) and by Chiari (1974). The percentage of trypomastigotes varies from one strain of T. cruzi to the other and depends also on the composition of the growth medium. Although several papers have been published dealing with factors which could induce the epimastigotetrypomastigote transformation in acellular culture media, difficulties have been found in reproducing the results satisfactorily. More recently, however, some clones obtained from the Y and CL strains show a high rate of differentiation which reaches about 95% when the parasites are cultivated in a special medium (Chiari, 1981). The trypomastigotes obtained under these conditions are able to reproduce the intracellular cycle in normal mouse macrophages maintained in culture (Meirelles et al., 1982). It is possible that this experimental system will open new possibilities for biochemical and physiological studies of trypomastigotes of T . cruzi. It has been reported that, under special conditions of cultivation, spheromastigotes (amastigotes) can be obtained in axenic cultures. It was shown that these forms are able to reproduce the intracellular cycle of T. cruzi in skin muscle cells maintained in tissue cultures (Pan, 1978a). These studies, however, will need to be confirmed. When bloodstream forms of some strains of T. cruzi are inoculated into axenic medium, clumps of rounded forms, which are able to multiply, are observed (Brener and Chiari, 1965; Baker and Price, 1973). These forms are, by morphological criteria, similar to spheromastigotes found inside vertebrate cells. However, further studies are necessary in order to determine if these rounded forms are indeed spheromastigotes, able to reproduce within vertebrate cells. It has been shown that when bloodstream trypomastigotes were placed in the presence of an established cell line of Triatoma infestans they transformate into amastigote-like cells which multiply to form large clusters of cells. The term staphylomastigote was proposed to designate this form. After 10 days of cultivation the staphylomastigotes transform into trypomastigotes (Lanar, 1979).

206

WANDERLEY DE SOUZA

CELL BIOLOGY OF TRYPANOSOMA CRUZI

207

IV. The Cell Surface The cell surface of trypanosomatids can be considered as composed by two components: the plasma membrane and a layer formed by the subpellicular microtubules (Figs. 4 and 5). Although it is now well established that in most of the eukaryotic cells the microtubules and the microfilaments are associated with the plasma membrane (for a review see Weatherbee, 1981), in no other cell type is such association as strong as in trypanosomatids where these two components of the cell surface remain associated even after lysis of the protozoan and isolation of a membrane-enriched fraction. A. THE PLASMAMEMBRANE As in other cell types, the plasma membrane of T . cruzi is composed of proteins, lipids, and carbohydrates which form the glycocalix. This structure is formed by a filamentous material localized on the outer face of the plasma membrane. Cytochemical studies, together with electron microscopy, show that the glycocalix is present in all stages of the life cycle of T . cruzi, even in the intracellular forms (De Souza and Meyer, 1976). It is about 15 nm thick in bloodstream trypomastigotes and 5 nm thick in intracellular spheromastigotes and in epimastigotes (De Souza et al., 1978b). The material which forms the glycocalix gives a positive reaction, which appears as deposition of electrondense reaction products (Figs. 6 and 8) when the protozoan is submitted to the following cytochemical techniques: ( 1) periodic acid-thiosemicarbazide-silver proteinate (De Souza and Meyer, 1976); (2) ruthenium red (De Souza ef al., 1978b); (3) concanavalin A-horseradish peroxidase (Chiari et al., 1978); (4) colloidal iron hydroxide (Martinex-Palomo et al., 1976); and (5) cationized ferritin (De Souza et al., 1977a). The carbohydrate-containing sties on the cell surface of the three developmental stages of T . cruzi have been studied by the use of lectins. It was first reported that epimastigotes obtained from acellular cultures agglutinated intensely with the lectin concanavalin A while trypomastigotes did not agglutinate (Alves and Colli, 1974). However, further studies using trypomastigotes obtained either from axenic cultures or from the blood of infected mice showed that both epimastigote and trypomastigote forms of T. cruzi possess concanavalin A (Con A) receptors on their cell surface and are agglutinated with relatively low concentraFIG. 4. Tangential section through the surface of Lepromonas samueli showing the parallel array of the subpellicular microtubules. X75,OOO. (Courtesy of T. Souto-Padron.) Fic. 5 . Promastigote form of Leishmania mexicuna amazonensis showing the plasma membrane and the layer of subpellicular microtubules. A profile of the endoplasmic reticulum (ER) is seen below the sub-pellicular microtubules. At some points (arrow) it approximates to the plasma membrane. X 135,000. (Courtesy of P. F. Pimenta.)

208

WANDERLEY DE SOUZA

CELL BIOLOGY OF TRYPANOSOMA CRUZI

209

tions of the lectin (Chiari et al., 1978). The presence of Con A receptors on the cell surface of T. cruzi was also detected by light microscopy, using fluoresceinlabeled Con A (Araujo et a l . , 1980; Szarfman et al., 1980), and by electron microscopy (Fig. 7) using Con A conjugated to horseradish peroxidase (Chiari et al., 1978) or to ferritin (De Souza, unpublished observations). Binding of Con A was not observed to trypomastigotes of the MR strain (Araujo et a l . , 1980). Recently it was shown that epimastigotes of T. cruzi appear to have at least two types of concanavalin A receptors: one of low capacity and high affinity and another of high capacity and low affinity. One of these receptors has a five-fold larger affinity constant than the receptor for concanavalin A found in trypomastigotes (Katzin and Colli, 1983). A detailed study of the cell surface of amastigotes, epimastigotes, and trypomastigotes of T. cruzi was recently undertaken by Pereira et al. (1980). These authors tested highly purified lectins with specificities for binding sites containing N-acetylgalactosamine, N-acetylglucosarnine, D-galactose, D-mannose, and sialic acid. Results obtained with wheat germ agglutinin (WGA) showed that only epimastigotes were agglutinated by low concentrations of this lectin. Bloodstream and culture-derived trypomastigotes as well as isolated intracellular forms only agglutinated at higher concentrations of the lectin. Katzin and Colli (1983), however, reported that trypomastigotes have two receptors with different affinities and capacities for WGA. The nature of the WGA receptors in epimastigotes and in trypomastigotes was studied in more detail using labeled WGA (Pereira et al., 1980; Katzin and Colli, 1983). In the case of epimastigotes it was observed that both the WGAinduced agglutination and the binding of WGA to the cell surface of T. cruzi was inhibited by N-acetylglucosamine or by a 1-acid glycoprotein, a sialic acidcontaining glycoprotein. Treatment of the epimastigotes with neuraminidase impeded both the binding of WGA to the cell surface and the agglutination of the cells. Such treatment also released sialic acid in the supernatant. Agglutination of epimastigotes was also observed when the cells were incubated in the presence of a sialic acid-binding lectin isolated from Limulus polyphemus (Pereira et al., 1980). These results suggest that WGA interact with sialic acid of epimastigotes of T. cruzi. However, this does not mean that trypomastigotes do not have sialic acid in their surface. In the case of trypomastigotes, affinity chromatography with WGA-sepharose yielded a glycoprotein of M, 85,000, specific to the trypomastigote stage. The WGA binding properties of these glycoproteins are not FIG.6. lntracellular form of T. cruzi in which the glycocalix can be seen as a dense reactive layer. Cell submitted to the periodic acid-thiosemicarbazide-silver proteinate technique for detection of glycoproteins. X45,OOO. (After De Souza and Meyer, 1976.) FIG. 7. Bloodstream trypomastigote of T. cruzi showing the distribution of concanavalin A binding sites as detected using the concanavalin A-peroxidase technique. Reaction product (arrow) is seen over the whole cell surface but not on the membrane portion which lines the flagellar pocket. K, Kinetoplast. X22,500. (After Chiari et al.. 1978.)

210

WANDERLEY DE SOUZA

CELL BIOLOGY OF TRYPANOSOMA CRUZI

21 1

altered by treatment with neuraminidase of either intact trypomastigotes or the isolated glycoproteins (Katzin and Colli, 1983). Based on the results obtained it was also suggested that intracellular forms do not have sialic acid on the cell surface, as judged by lectin-induced agglutination (Pereira e f al., 1980). Wistaria JZoribunda hemagglutinin, a lectin with specificity for N-acetylgalactosamine, agglutinated both epimastigotes and trypomastigotes obtained from acellular cultures but not agglutinated intracellular forms and bloodstream trypomastigotes. In contrast, the lectin from Phaseolus vulgaris, which also reacts with N-acetylgalactosamine, agglutinated intracellular forms and bloodstream trypomastigotes but did not agglutinate epimastigotes and trypomastigotes from acellular cultures. Peanut agglutinin (PNA from Arachys hypogae), which is specific for the disaccharide DGalP- 1-3~GalNAc and DGalP- l-+4~GlcNAc, only agglutinated the intracellular forms of T . cruzi (Pereira er al., 1980). The two structures mentioned above are commonly found in complex carbohydrates localized on the cell surface, but are available for interaction with lectins only in poorly sialylated cells which are more imature. In mature cells these structures are masked by sialic acid and therefore are not available for interaction. The study of the cell surface of T . cruzi with lectins provides data which show that qualitative and quantitative differences exist in the carbohydrates exposed on the cell surface of the parasite in the various stages of its life cycle. Contradictory results have been described for some lectins (see Pereira et al., 1980; Araujo e f al., 1980), probably as a consequence of the purity of the lectin used, the strain of the parasite, and the method used to assay the cell agglutination. Considering that some lectins showed a specific interaction with determinated forms of the cycle the results obtained open the possibility of the use of affinity chromatography for the purification of the different forms. More recently the agglutination induced by various lectins was studied in various T . cruzi and T . cruzi-like strains. Two different agglutination types were observed. One type (strains Y,CL and FL) was agglutinated by WGA but not by PNA. The second type (Sgo Felipe and OPS strains) was agglutinated by PNA but not with WGA (Schottelius, 1982). B. SURFACE CHARGE By using the colloidal iron hydroxide and cationized ferritin particles it was shown that T . cruzi possesses anionic sites on its cell surface (Martinez-Palomo FIG. 8. Transitional form between epimastigote and trypomastigote, incubated in the presence of cationized ferritin for detection of cell surface anionic sites. Cationized ferritin particles are seen throughout the cell surface (short arrow) and on the membrane which lines the flagellar pocket (long arrow) and the flagellum (F). The kinetoplast (K) shows a morphology typical of transition forms in which part of the DNA is arranged as in epimastigotes (*) and part as in trypomastigotes (**). N, Nucleus. X72,OOO. (After De Souze er al., 1977a.)

212

WANDERLEY DE SOUZA

et al., 1976; De Souza et a l . , 1978b). With colloidal iron, incubation was performed at pH 1.8, a condition which facilitates the detection of sialic acid. Indeed the binding of colloidal iron particles to the cell surface of T . cruzi can be almost completely blocked by previous treatment of the cells with neuraminidase (Souto-Padrh and De Souza, unpublished observations). Biochemical data as well as agglutination of T. cruzi by the lectin from Lirnulus polyphemus and wheat germ agglutinin also indicate the presence of sialic acid on the cell surface of T. cruzi (Pereira et a l . , 1980). The data obtained by using the colloidal iron hydroxide particles indicate that the intensity of the reaction, reflected by the number of particles attached to the cell surface, is higher in bloodstream and tissue culture-derived trypomastigotes than in trypomastigotes derived from acellular cultures or in epimastigotes. It is possible that host cell components or components of the serum of the host may account for the large number of colloidal iron particles on the cell surface of the trypomastigotes. Data obtained by using cationized ferritin particles also show differences between the cell surface of epimastigote and trypomastigote forms of T . cruzi. The incubation of the protozoa with cationized ferritin was performed at physiologic conditions, e.g., at pH 7.2. It was observed that the binding of positive particles was more intense in the bloodstream trypomastigotes than in epimastigotes thus suggesting that the cell surface of the first form had a more negative surface charge than the latter (De Souza et a l . , 1977a). Examination of the net surface charge determined by electrophoretic mobility (EPM) of epimastigotes and trypomastigotes obtained from acellular cultures as well as from bloodstream trypomastigotes indicated that all forms have a negative surface charge which varies in degree according to the developmental stage. Based on the determination of the EPM of mixed populations containing epimastigotes and trypomastigotes from axenic cultures it was observed that during the development of T . cruzi there is a gradual increase in the net surface so that the highly differentiated trypomastigotes are those with the highest values. The EPM indicated a surface charge of -0.52 pm sec- V - cm for epimastigotes and - 1.14 for bloodstream trypomastigotes. Previous qualitative studies (Broom et al., 1936) had shown that bloodstream trypomastigotes of T . cruzi had a negative surface charge. Lanham and Godfrey (1970) also studied the separation of bloodstream forms of several trypanosomes by using a DEAE-cellulose column. They suggested, on the basis of the degree of adsorption of the trypanosomes to the columm, the following order of negative surface charge: T . cruzi > T . lewisi > T . vivax, T . congolense > T . brucei, T. evansi, T. simiae. The authors were able to separate most of the trypanosomes from the red blood cells by using the above mentioned anion exchanger columm. Indeed this technique is used today as a routine method for purification of bloodstream trypomastigotes of T . brucei for biochemical and immunological studies. Bloodstream forms of T . cruzi, however, could not be separated from the red blood cell thus suggesting the existence of

'

CELL BIOLOGY OF TRYPANOSOMA CRUZI

213

a very slight difference in the surface charge between the two cells. Indeed, while the surface charge of T. cruzi, as determined by EPM, is - 1.14, that of erythrocytes is - 1.07 pm sec- V - cm. The EPM of bloodstream forms of T. rhodesiense was -0.1 pm sec- V- cm (Hollingshead et al., 1963) while that of culture form forms was -0.91. Some data obtained indicate that the passage of trypomastigotes of T. cruzi through DEAE-cellulose column may change some of the surface properties of the population of isolated parasites (Goldberg et al., 1976; De Souza et al., 1977a; Villalta and Leon, 1979). At present it is not yet clear if the column indeed alters the surface of the cells or selects an already differentiated population (Souza, 1982). In any case the data available indicate that care might have to be taken in the interpretation of results obtained with parasites isolated in this way. It was shown that the passage of culture forms of T. cruzi through DEAEcellulose changes the kinetics of transport of arginine and lysine by the plasma membrane (Goldberg et al., 1976) and decreases the electrophoretic mobility (De Souza et al., 1977a). It was also shown that the infectivity of bloodstream forms of T. cruzi isolated by such columns is lower than that observed for trypomastigotes isolated by other means (Villalta and Leon, 1979). For these reasons the author of this article prefers to work with parasites isolated by centrifugation on a gradient of metrizamide (Loures et al., 1980). However, recent studies indicate that it is possible to use DEAE-cellulose columns to purify bloodstream trypomastigotes of T. cruzi without interference with their infectivity to mice as well as with the binding of anti-T. cruzi antibodies to the parasite’s surface (Souza, 1982).

’

C. THE SURFACE COAT Electron microscopy has shown a layer of electron-dense material on the unit membrane of the Trypanosomatidae family. This structure, which has been designated as surface coat, is more evident in bloodstream trypomastigotes which belong to the subgenera Trypanosoon and Nanomonas where it reaches a thickness of 15 nm (Vickerman, 1969). The surface coat is lost in acellular cultures and after the parasite enters the vector for a cyclical development. However, it reappears when the parasites transform into metacyclic forms in the salivary gland of the invertebrate host. Vickerman (1969, 1974) suggested that “the surface coat of T. brucei is an endogenous glycocalyx which contains the variant antigens demonstrable by agglutination and neutralizing reactins, and that both, antigenic variation and surface protection, could have a basis in successive replacements of this glycocalix with another containing a different antigen.’ ’ This assumption is supported by various facts: ( I ) loss of both the variant antigen and the surface coat occurs when bloodstream forms of T. brucei are put into acellular cultures; (2) ferritin-conjugated antibodies to a specific variant will bind

214

WANDERLEY DE SOUZA

to the cell surface of the parasite only when the surface coat is present. Remotion of the surface coat impedes the binding of the antibody (Vickerman and Luckins, 1969); (3) metacyclic trypomastigotes isolated from the salivary gland of the invertebrate host carry the variant antigen probably because they possess a surface coat (Vickerman, 1969, 1974; Steiger, 1973). It has been shown that the surface coat of T. brucei indeed contains the variant antigen, which is a glycoprotein which accounts for about 10% of all trypanosome protein (Cross, 1977, 1979; Cross et al., 1980). The purified variant glycoproteins contain a single polypeptide chain of 65,000 daltons. Antigenic specificity of the variant glycoproteins appears to be primarily due to extensive diversity in the amino acid sequence. The amino acid sequence of a variant surface glycoprotein of T. brucei was determined. It consists of 470 amino acid residues with two carbohydrate chains attached at Asn420 and Asp470. The authors reported that the carboxy-terminal region of the glycoprotein, which is close to the membrane, is remarkably hydrophilic. Besides no pronounced hydrophobic regions were found. These results suggest that the variant surface glycoprotein is not an integral protein of the plasma membrane and is probably associated with the membrane through electrostatic interaction of its charged residues with bther membrane components (Allen et al., 1982). This suggestion is supported by the fact that the variant surface coat antigen from T. brucei is released if intact cells are incubated in the presence of calcium and the calcium ionophore A-23187 (Voorheis et al., 1982). More recently, however, it was reported that the variant surface glycoproteins are altered by the procedures employed during their isolation. Using a new method it was suggested, based on the use of charge-shift electrophoresis, that the variant surface glycoproteins have amphiphilic properties and behave as integral membrane proteins (De Almeida and Turner, 1983). Variant antigen glycoproteins have been purified from a number of clones of T. brucei and found to be very different by isoelectric focusing mapping, amino acid composition (Cross, 1975, 1977), and N-terminal amino acid sequence (Bridgen et al., 1976). The total carbohydrate composition of each variant also showed considerable variation (Johnson and Cross, 1977). Selective cleavage of the isolated proteins by mild tryptic hydrolisis indicated that all the carbohydrate was attached in the C-terminal region of certain variant glycoproteins, possibly at the membrane-protein interface (Cross and Johnson, 1976). For two variant proteins that terminate in an aspartic acid and a serine residue, respectively, the sugar side chain is linked through ethanolamine to the a-carboxy group of the amino acid (Holder, 1983).

D. FREEZE-FRACTURE Most of the ultrastructural studies performed with T. cruzi have been done in ultrathin sections. Although this technique gives important information concerning the organization of cellular components it does not reveal details of the

CELL BIOLOGY OF TRYPANOSOMA CRUZI

215

structure of cell membranes. In replicas of freeze-fractured specimens, however, the inner part of the cell membranes is exposed allowing the examination of either the inner or the outer membrane halves. Freeze-fracture replication may cleave through the cytoplasm and/or through the hydrophobic region of cell membranes exposing large surfaces of the P (which represents the outer aspect of the inner membrane half) and the E (which corresponds to the inner aspect of the outer membrane half) faces (Figs. 9 and 10). Both the P and the E faces showed a large number of structures which have been designated as intramembranous particles. Since the fracture plane passes through the center of the lipid bilayer of the membrane, the intramembranous particles represent membrane components that extend at least halfway through the lipid core (Branton et al., 1975; Pinto da Silva and Branton, 1970.) Evidence exists that the intramembranous particles represent integral proteins located in the hydrophobic region of the membrane lipid bilayer, as predicted by the fluid mosaic model for membrane structure (Singer and Nicolson, 1972). The density of the intramembranous particles, e.g., the number of such particles per square micrometer of membrane, reflects the richness in proteins of the membrane. A special pattern in the arrangement of the particles is usually interpreted as a functionally specialized region of the cell membrane. Freeze-fracture studies on T . cruzi show that the structure of the flagellar membrane differs from that of the membrane which encloses the cell body. Both the P and the E faces of the flagellar membrane show only very few intramembranous particles throughout the whole flagellum (Martinez-Palomo et a l . , 1976; De Souza et a l . , 1978a). This difference is not visible in ultrathin sections. It is interesting to note that the low density of particles in the flagellar membrane seems to be a characteristic feature of the membrane of cilia and flagella from other cell types. It is possible that this fact is related to the function of the flagellum. Many proteins involved in the transport of nutrients, ions, regulation of cell growth, etc. do not need to be located on the flagellar membrane. However, in some trypanosomatids an asymmetric distribution of particles has been observed in the flagellar membrane. In Leptomonas collosoma (Linder and Stachelin, 1977), Herpetomonas samuelpessoai (De Souza et al., 1979a), and Leptomonas samueli (Souto-Padrh et a l . , 1980a) the E face has a density of intramembranous particles almost similar to that seen in the membrane which enclose the cell boy. However, the particles found on the E face can be visualized only in favorably shadowed regions where they appear as flat particles. Since they are observed only on the E face it is possible that they represent some kind of receptor proteins localized in the flagellar membrane. It is well known that the flagellum of trypanosomatids is involved, besides the cell locomotion, in the establishment of contacts between the protozoan and other surface in general such as the host cell membrane, the wall of the digestive tube of the invertebrate host, the substratum in which the cells are grown in v i m , etc. (for a review see Vickerman and Preston, 1976).

216

WANDERLEY DE SOUZA

FIG. 9. General aspect of an epimastigote form of T. cruzi as seen in freeze-fracture replicas. The nucleus (N)can be seen with its membrane provided with pores (arrow). The Golgi complex (G) and cytoplasmic vesicles are observed. X37,OOO. (After De Souza er al., 1978b.)

FIG. 10. Aspect of the protoplasmic faces of the membranes which enclose the cell body (B) and the flagellum (F) of T. cruzi. The membrane enclosing the cell body has a high density of intramembranous particles while few particles are seen in the flagellar membrane. A specialized area is seen near the flagellar pocket. It corresponds to the cytostome (C), which is a particle-poor region delimited by a linear array of particles (arrows). X42,OOO. (After De Souza e t a / . , 1978b.)

CELL BIOLOGY OF TRYPANOSOMA CRUZI

217

218

WANDERLEY DE SOUZA

The flagellar membrane also presents another specialized region localized at the base of the flagellum. It can be observed in thin sections of cells fixed in the presence of ruthenium red (Fig. 11) and appears as hair-like projections, perpendicularly oriented in relation to the main axis of the flagellum (Linder and Stachelin, 1977; De Souza et al., 1978a). By freeze-fracture it was possible to visualize at the base of the flagellum, an aggregation of irregular particles (Fig. 12). Such a structure has been designated as a ciliary necklace and is possibly involved in the control of the flagellar motility. Another area of specialization of the flagellar membrane is observed at the region of attachment of the flagellum to the cell body. Such specialization will be described latter. The plasma membrane of epimastigotes of T . cruzi has a higher density of intramembranous particles than that in bloodstream trypomastigotes (De Souza et al., 1978b). Determination of particle density shows that epimastigotes have 1836 and 1450 particles/km2 while bloodstream trypomastigotes have 122 and 125 particles/km2 for the P and E faces, respectively. These data indicate that the two developmental stages hhve a different protein-lipid ratio in their plasma membrane. It is interesting to note that in almost all eukaryotic cells studied up to now with the freeze-fracture technique, the P faces have a higher density of particles than the E faces. In the case of epimastigotes and trypomastigotes of T . cruzi studied as well as in other members of the Trypanosomatidae family, the two faces of the plasma membrane have a similar particle density. In T . brucei, however, the P face has a density of membrane particles higher than on the E face (Smith et al., 1974; Vickerman and Tetley, 1977). In L. samueli the density is higher on the E face than on the P face (Souto-Padr6n et al., 1980a). It is possible that the conditions in which the parasites are cultivated as well as the parasite strain may be responsable for the difference in the structure of the plasma membrane. Futher comparative studies in this area are necessary. The intramembranous particles localized on both faces of the plasma membrane of T . cruzi are homogenously distributed. Only at the region of the cytostome found in epimastigotes and in spheromastigotes a special array of particles was observed and will be described later. The plasma membrane of epimastigotes of T . cruzi was used to analyze the relationship between intramembranous particles and surface coat components detected by cytochemical techniques. It was shown, both in the cytostome and in the flagellar membrane, the existence of an independence of concanavalin A receptors, colloidal iron hydroxyde binding sites, and ruthenium red-stainable surface coat components from the intramembranous particles (Martinez-Palomo et al., 1976). E. MOBILITY OF MEMBRANE COMPONENTS

The physical state of the lipid of a membrane is best described by its fluidity. As with other substances, lipids can exist in a solid or in a liquid phase of varying

CELL BIOLOGY OF TRYPANOSOMA CRUZI

219

FIG. 11. Thin section of an epimastigote form of T . cruzi fixed in the presence of ruthenium red. Hair-like projections are evident at the base of the flagellum (F) (arrows). K, Kinetoplast. X55,OOO. FIG. 12. Freeze-fracture image of the base of the flagellum (F) of epimastigoes of T . cruzi. Groups of particles (*) are seen. X 115,000. (After De Souza et a / . , 1978b.)

220

WANDERLEY DE SOUZA

viscosity, depending on the temperature. Since it is the lipid that provides the matrix in which the membrane proteins are embedded the physical state of the lipid will be an important determinant of the mobility of the proteins and glycoproteins (Shinitzky and Henkart, 1980). The actual demonstration that the proteins of the membrane are in a dynamic state within the plane of the membrane has been one of the main findings in favor of the fluid mosaic model (Singer and Nicolson, 1972). Unfortunately there are no detailed studies of the role of lipids in relation to the fluidity of the plasma membrane of T . cruzi or any other member of the Trypanosomatidae family. However, there are certain indications that at least some components of the plasma membrane of T . cruzi may undergo lateral difusion in the membrane plane. These include the aggregation of intramembranous particles, as seen in freeze-fracture replicas, and the induction of patching and capping of membrane antigens and membrane receptors for concanavalin A. It was observed that when glutaraldehyde-fixed epimastigotes of T . cruzi are incubated in the presence of fluorescein- or peroxidase-labeled concanavalin A, there is a uniform binding of the lectin throughout the cell surface of the parasite, e.g., the membrane which encloses the cell body, the flagellar pocket region, and the flagellum. However, as mentioned previously, there is a more intense binding of the lectin to the region of the cytostome (Szarfman et al., 1980). Similar results are obtained when living, instead of glutaraldehyde-fixed, epimastigotes are incubated at 4°C for 30 minutes or at 4°C for 30 minutes followed by an incubation for 15 to 60 minutes at 28 or 37"C, or when the epimastigotes are incubated in the presence of specific antibodies, with the only difference that the antigenic sites do not show a preferential localization in the region of the cytostome (Szarfman et al., 1980). These observations suggest that the fluidity of the membrane of epimastigotes of T . cruzi is low at least for some membrane components such as the antigenic and the concanavalin A-binding sites. It is well known that cytoplasmic components associated with the plasma membrane, such as microfilaments and microtubules, have a marked influence on the mobility of membrane components. In many cells evidence exists indicating that the microtubules are responsable for anchoring the membrane proteins in place while the microfilaments would cause them to be moved. In such cells, incubation in the presence of drugs which interfere with microtubules, such as colchicine and vinblastine, increases the mobility of membrane components. Otherwise, incubation of the cells with cytochalasin B, which interfers with microfilaments, decreases the mobility of the same membrane components (for a review see Nicolson, 1976). Treatment of epimastigotes of T. cruzi with either colchicine, vinblastine, or cytochalasin B did not induce the mobility of antigenic sites and concanavalin A-binding sites. When living bloodstream trypomastigotes of T. cruzi were incubated for 30 minutes at 4°C in the presence of sera from chronically infected humans or mice

CELL BIOLOGY OF TRYPANOSOMA CRUZI

22 1

containing antibodies against T. cruzi, 50 to 75% of the parasites showed a strong diffuse staining (fluorescein or peroxidase) on the surface. Ten percent of the parasites showed a patch staining and less than 5% exhibited capping. However, when after incubation at 4°C the temperature was raised to 37"C, a large number of parasites form patches and caps. In some cases about 60% of the parasites had antigen-antibody complexes, which were revealed by immunofluorescence or immunoperoxidase, localized in the posterior and in the anterior tips of the cell. Such localization was confirmed by electron microscopy. The induction of capping of antigen-antibody complexes required cellular energy since it could be blocked by treatment of the parasites with sodium azide or iodoacetamide. Cytochalasin B or colchicine did not interfere with the process of capping (Schmunis et al., 1978, 1980). Capping of concanavalin A receptors was also induced in bloodstream trypomastigoes of T . cruzi, contrary to what we observed with epimastigotes. Experiments in which the trypomastigotes were simultaneously labeled with antibodies and concanavalin A showed that the two receptors are independent from each other, e.g., it is possible to induce capping of Con A receptors without interference in the distribution on antigenic sites and vice versa (Szarfman et al., 1980). Recent results show that the intracellular spheromastigotes of T. cruzi, isolated from the spleen macrophages of infected mice, also present mobility of antigenic sites and concanavalin A receptors (Leon et al., 1979). Capping seems to be a general phenomenon in many members of the Trypanosomatidae family as has been demonstrated in Trypanosoma brucei (Barry, 1975, 1979; Barry and Vickerrnan, 1977), Trypanosoma lewisi (Giannini and D'Alesandro, 1978; Cherian and Dusanic, 1977), and Leishmania (Doyle et al., 1974; Dwyer, 1976). In most of the eukaryotic cells the process of capping is induced only when the cells are incubated at temperatures higher than 18°C. In the case of T . cruzi as well as in T . lewisi (Giannini and D'Alesandro, 1978) capping can be induced at 4"C, at least for a certain number of parasites. It is interesting to note that the number of capped trypornastigotes of T . cruzi varied according to the strain of the parasite used. For instance, while capping was easily obtained with parasites from the Y strain it was difficult to induce capping in trypomastigotes of the CL strain (Schrnunis et al., 1980). These two strains also present other immunological differences, probably related to their surface properties, as will be discussed latter. What is the functional role of the process of capping in T . cruzi? Clearly, it indicates that the membrane of bloodstream trypomastigotes and intracellular spheromastigotes, which are the two developmental forms found in the vertebrate host, are fluid for at least some of their components. It has been suggested that there is a process of shedding of the capped antigen-antibody complex

222

WANDERLEY DE SOUZA

(Schmunis et al., 1980). If such shedding occurs in vivo, in the infected vertebrate host, it may partially explain the deposits of antibodies detected in the glomeruli of infected animals (Szarfman et a l . , 1975; Castro and Ribeiro dos Santos, 1977). It is also possible that the so-called exoantigens detected in sera from infected animals and man (Siqueira et al., 1966; Gotlieb, 1977) may have originated from immunocomplexes sloughed from the parasite surface. It is also possible that antibody-induced mobility of surface antigens of T . cruzi or any other parasite is one of the mechanisms which render the parasite resistant to the immune defense of the host. At least two alternatives exist in relation to this point: (1) it has been shown that opsonized parasites are easier ingested by macrophages than normal parasites. At least for some strains of T . C ~ K the Z ~ opsonized trypomastigotes are destroyed by the macrophages (Alcantdra and Brener, 1978). Since the immunological phagocytosis mediated through the Fc receptors localized on the macrophage membrane requires that the parasites have antibodies homogeneously distributed throughout the parasite surface, according to the Zipper hypothesis (Silverstein et al., 1981) capped parasites would not be ingested. Therefore the process of capping would impede the ingestion of opsonized trypomastigotes by macrophages. Since nonopsonized trypomastigotes of T . C ~ K are Z ~ ingested but not destroyed by macrophages, the process of capping would be relevant for a escape of T . cruzi from the microbicidal action of the macrophage. It is interesting to note here that opsonized trypomastigotes of T. C ~ K from Z ~ the CL strain do not cap easily (Schmunis et al., 1980) and are destroyed by macrophages (Alcantdra and Brener, 1978) while those from the Y strain cap easily and survive inside the macrophages. Further studies are necessary in order to establish if there is, or not, a relationship between the ability of T. U K Z ~ to undergo lateral mobility of membrane components, and its ability to survive macrophages; (2) it has been suggested that the process of capping would be relevant as a means for the parasite to free itself of antibodies on its surface. The process of capping in T . brucei is detected only when the indirect, rather than the direct, immunofluorescence method is used. A single layer of antibody is inefective (Barry, 1979). In the case of T . C ~ K capping Z ~ can be observed using a single layer of antibodies (Schmunis et a f . , 1978, 1980). The data obtained in T . brucei indicate that the process of capping does not play an important role in the phenomenon of antigenic variation since capping of a variable antigen of one type is followed by the appearance of antigen of the same type (Barry, 1979). F. BIOCHEMICAL ANALYSIS OF THE CELLMEMBRANE It was shown by ultrastructural cytochemistry that T . C ~ K Zdoes ~ not have a reserve of polysaccharides. Reaction product, indicative of the presence of carbohydrates, was associated mainly with the plasma membrane of the protozoan

CELL BIOLOGY OF TRYPANOSOMA CRUZI

223

although a slight reaction was observed in intracellular membranes which form the Golgi complex, the endoplasmic reticulum, and some cytoplasmic vesicles (De Souza and Meyer, 1976). Therefore, biochemical data on glycoproteins or polysaccharides of T . cruzi may be considered as associated with cellular membranes. A sugar-containing macromolecular complex was isolated from epimastigotes of T. cruzi using extraction with phenol and precipitation with ethanol (Alves and Colli, 1975). Polyacrylamide gel electrophoresis of the isolated glycoprotein fractions showed the presence of four Schiff-positive bands which were designated as bands A, B, C, and D. The gel pattern of bands A, B, and C was deeply altered if the complex was treated with proteolytic enzymes thus suggesting their glycoprotein nature. Band D was resistant to proteolytic enzymes (Lederkremer et al., 1976). It was composed of neutral sugars, glucosamine, phosphorus, amino acids, amide and ester-linked fatty acids, inositol, and long chain bases of the sphingosine family (Lenderkremer et al., 1977, 1978). This complex substance was designated as lipopeptidophosphoglycan (LPPG). It may represent about 15% of the plasma membrane dry weight, precipitates concanavalin A, and inhibits the agglutination of epimastigotes by this lectin (Colli et al., 1981; Franco da Silveira and Colli, 1981). These data strongly suggest that LPPG is a plasma membrane-associated component of epimastigotes of T . cruzi and it may be part of the concanavalin A receptor found on the cell surface of the parasite. Polysaccharides have been isolated from epimastigotes of T . cruzi (Gogalves and Yamaha, 1969; Gottlieb, 1977) and showed to be antigenically active. Galactose, glucose, mannose, xylose, and glucosamine were found as components of the polysaccharide. More recently a D-galacto-D-mannan was isolated from epimastigotes. After methylation and gas-liquid chromatography and mass spectroscopy analysis a number of methylated fragments were identified which indicated a-D-mannopyranosyl units that were nonreducing ends, 2-0-, 2,3-di0-,and 3,4-di-O-substituted structures and D-galactofuranosyl nonreducing endunits. 13C-nuclear magnetic resonance spectroscopy showed the presence of single units, nonreducing ends of a-D-galactofuranose (Gorin er al., 1981). Marcipar et al. (1982) using diidosalycilic acid and lithium salts and phenolwater biphasic partition isolated a glycoprotein fraction which showed, by SDS-polyacrylamide gel electrophoresis, the presence of five glycoprotein bands. The glycoprotein fraction was immunogenic and antibodies obtained in goats against the fraction reacted specifically with T . cruzi but not with other member of the Trypanosomatidae family. A galactose-terminal glycoprotein, with a molecular weight of about 10,000, was isolated from the glycoprotein fraction by affinity chromatography using peanut agglutinin (Marcipar et al., 1982). More recently a glycoprotein, with a molecular weight of 25,000, was isolated from epimastigotes of T . cruzi. It was also shown that it is recognized by serum antibodies of Chagas’ disease patients and that it is located on the cell

224

WANDERLEY DE SOUZA

surface of amastigote, epimastigote, and trypomastigote forms (Mendonqa-Previatto et al., 1982; Scharfstein et al., 1982). Another approach to analyze cell surface associated glycoproteins is the use of enzymatic labeling procedures. Two procedures have been frequently used: (1) the enzyme lactoperoxidase is used to catalyze a reaction in the presence of peroxide so that a radioactive iodine atom can be attached to the tyrosine residues of a protein; (2) the enzyme galactose oxidase is used. It oxidizes galactose and galactosamine residues in carbohydrate chains. When these oxidase sugars are incubated with a radioactive reducing agent (tritiated borohydride) they are returned to the reduced state by incorporating radioactive hydrogen atoms. Since both lactoperoxidase and galactose oxidase do not penetrate the plasma membrane only those proteins and glycoproteins that are acessible to the enzymes will be radioactively labeled. The labeled proteins are identified later in autoradiographs of the polyacrylamide gel electrophoresis. Both procedures, as well as the Iodo-Gen technique, have been used to analyze the membrane components of T . cruzi. Snary and Hudson (1979), by using the lactoperoxidase procedure, found labeled proteins in epimastigotes, amastigotes, and trypomastigotes of T . cruzi. Each developmental form had a characteristic pattern of labeled proteins althought several bands were common to all forms. Proteins with a molecular weight of about 130,000 and 180,000 were suggested to be specific for amastigotes and trypomastigotes, respectively (Snary and Hudson, 1979). Using the lectin from Lens culinaris they detected in the three developmental forms studied a glycoprotein with a molecular weight of 90,000. This glycoprotein contains 18% of carbohydrate of the mannose chain type linked through glucosamine to asparagine in the polypeptide chain (Scott and Snary, 1982). By using the same methodology, Zingales et al. (1979) found two 1251-lactoperoxidase-labeledglycoproteins from epimastigotes of T. cruzi. One had a molecular weight of 100,000 and the other between 80,000 and 90,000, Using the glucose oxidase-lactoperoxidase technique a major component, with a molecular weight of 90,000, was detected only in bloodstream trypomastigotes of T . cruzi (Nogueira et al., 1980, 1981). Such a component was not found in epimastigotes and in acellular culture-derived trypomastigotes. The later has a component with a molecular weight of 75,000. It was suggested that the major band found in the bloodstream trypomastigotes may represent the trypsin-sensitive component of the cell surface of T . cruzi which makes difficult the ingestion of the parasite by macrophages (Nogueira et al., 1980, 1981). It was shown that such a major protein is an acidic protein with a pf of 5.0 while the pf of the protein from epimastigotes and acellular culture-derived trypomastigotes was 7.2. According to the results reported by Nogueira et al. the cell surface of bloodstream trypomastigotes of T . cruzi has only one major component. However, results recently reported by Zingales et al. (1982a), using the Iodo-Gen technique, indicate that trypomastigotes have a larger number of bands

CELL BIOLOGY OF TRYPANOSOMA CRUZI

225

than epimastigotes. More recently Nogueira et al. (1982), using the lactoperoxidase/glucose oxidase method, showed that the glycoprotein with a molecular weight of 75,000 is specific for insect stages of T . cruzi since it was found in epimastigotes and in axenic culture-derived trypomastigotes. On the other hand the glycoprotein with a molecular weight of 90,000 is specific for mammalian stages since it was observed in amastigotes, tissue culture-derived trypomastigotes, and bloodstream trypomastigotes. Tryptic and chymotryptic digestion showed little homology between the peptide pattern of the two glycoproteins, a result which was interpreted as indicative that they are coded by independent genes. The glycoproteins were found in several strains of T. cruzi and thpy lack immunologic cross-reactivity. A more complex pattern of bands was observed by Araujo and Remington (1981). They found (1) two majors bands, of 100,000 and 90,000, in tissue culture-derived trypomastigotes, (2) a group of three to four bands between 68,000 and 94,000, and a single band of 45,000 in amastigotes, and epimastigotes. A band appears on the top of each lane which probably represents larger molecular weight components. This band was intense in epimastigotes, intermediate in amastigotes, and weak in trypomastigotes. The top band of trypomastigotes and amastigotes stained jointly with PAS while strong staining was observed in the top band of epimastigotes. The pattern of bands of epimastigotes and amastigotes was the same, independent of the strain of T. cruzi. In the case of trypomastigotes, however, those from the CL and MR strains have some bands of low molecular weight which were not seen in trypomastigotes of the Y strain. A band of 30,000 appeared only in trypomastigotes of the MR strain but not in those of the Y and CL strains. This fact is interesting since trypomastigotes of the MR strain differed from those of the Y and CL strains in regard to their capacity for binding lectins from Glycine max and Arachis hypogaea (Araujo er al., 1980). More recently Zingales et al. (198213) found, using the Iodo-Gen technique as well as immunoprecipitation, that epimastigotes have bands of 62,000, 72,000, 80,000, 95,000, and 150,000. Tissue-culture derived trypomastigotes have bands of 95,000, 85,000, and 80,000 and high-molecular-weight antigens up to 180,000. Therefore both epimastigotes and trypomastigotes share two main antigens (95,000 and 80,000) while antigens of 85,000 and 72,000 seem to be specific of trypomastigotes and epimastigotes, respectively. A glycoprotein with a molecular weight of 85,000 was recently purified, using affinity chromatography with WGA-sephorose, from trypomastigotes. This glycoprotein was not detected in amastigotes and epimastigotes (Katzin et a f . , 1982). It is important to point out that some differences found by different authors in the molecular weight of the glycoproteins or proteins may be consequence of proteolysis which occurs during the experiment. Other factors such as the strain and the origin of the parasite, and the methodology used may also influence the results. In vivo labeling of epimastigotes of T. cruzi by using the galactose oxidase

226

WANDERLEY DE SOUZA

procedure indicated that the glycoproteins which form the bands A, B, and C cited above are localized on the cell surface. Band D was not detected on the surface of amastigotes and trypomastigotes (Colli et al., 1981; Zingales et al, 1980, 1982a), thus suggesting that either they do not exist or they are masked in these two developmental forms of the T . cruzi life cycle. However, this macromolecule was recently located on the cell surface of all developmental stages of the T. cruzi life’s cycle, using antibodies developed in rabbits against the isolated LPPG (Previato et al., 1982). A monoclonal antibody was obtained from mice immunized with epimastigotes of T . cruzi. (Snary et a f . , 1981). The antibody reacted with epimastigotes from three different strains tested but did not react with amastigotes and trypomastigotes. The antigen recognized by the monoclonal antibody shows two major polypeptide bands of 72,000 and 64,000 daltons. It is possible, however, that the peptide of 64,000 is the result of proteolysis. The antigen, which represents 0.04% of the whole cell protein, is a glycoprotein composed of 46% carbohydrate, 42% protein, and 12% phosphate. All phosphate is linked to the carbohydrate. The carbohydrate is present as two classes of side chains, one linked through glucosamine to asparargine in the polypeptide chain and containing mannose, glucosamine, galactose, and glucose. The second type of chain is linked through fucose and xylose probably to serine or threonine and contains xylose, ribose, galactofuranose, and phosphate. No sialic acid or galactosamine was detected (Scott and Snary, 1982). It was recently suggested that this glycoprotein may function as a receptor involved in the process of transformation of epimastigotes into trypomastigotes since this transformation can be inhibited by a monoclonal antibody against the 72,000 dalton glycoprotein (Sher et al., 1982). Monoclonal antibodies which recognize only amastigotes or espimastigotes of T . cruzi were recently obtained (Araujo et al., 1982). Other antibodies reacted with both forms of T . cruzi. Immunofluorescence showed that the antigenic sites detected by these monoclonal antibodies are not uniformly distributed throughout the parasite surface (Araujo et al., 1982). It is expected that other monoclonal antibodies, which are being developed by several groups, will open new possibilities for the isolation of specific components of the plasma membrane of T . cruzi. G. ISOLATION OF THE PLASMA MEMBRANE The study of purified cell fractions can tell us much about the immunological properties, the enzyme content, and the functional role of a certain cellular structure. In the case of the plasma membrane of a pathogenic parasite it is important to know its composition in order to understand certain aspects of the host cell-parasite interaction. Recently several attempts have been made to obtain a pure plasma membrane

227

CELL BIOLOGY OF TRYPANOSOMA CRUZl

fraction from trypanosomatids including T . cruzi. In all studies, the authors make reference to the fact that trypanosomatids are resistant to the conventional cell breakage methods in consequence of the presence of a continuous layer of microtubules localized below the plasma membrane. Hunt and Ellar (1974) were able to isolate a plasma fraction from Leptomonas collosoma. They found that the subpellicular microtubules remained attached to the plasma membrane, which was then used as a morphological criterium to assess the purity of the fraction. This association of membrane and microtubules has been confirmed at several conditions for L . collosoma (Linder and Staehelin, 1977), T. cruzi (De Souza, 1976a), T . brucei (Voorheis et al., 1979). L. mexicana (Timm et al., 1980), L . donovani (Dwyer, 1980), and T . lewisi (Dwyer and D'Alesandro, 1980). Also in the reviewer's opinion, the association of plasma membrane and microtubule is, when present, the best criterion for assessing the purity of a plasma membrane fraction from trypanosomatids. However, in some cases it has been observed that, for unknown reasons, the microtubules did not remain associated to the plasma membrane of T . brucei (Rovis and Baekkeskov, 1980) and T . cruzi (Pereira et al., 1978; Zingales et al., 1979; Franco da Silveira et al., 1979). In these cases the only way to assess the purity of the fraction is through the assay of enzyme markers. Since different methods have been used to rupture the cells, it is possible that the fact that in some cases the microtubules remain attached to the membrane while in others they do not, depends on differences in the method used. However, even the same method used for isolation of the membrane form T . cruzi (Pereira et al., 1978; Timm et al., 1980) and L . mexicana gives different results. In the case of T . cruzi three procedures have been used to obtain a purified plasma membrane fraction from epimastigotes: ( 1 ) treatment of previously swollen cells with Triton X- 100, followed by disruption using a Dounce-type homogenizer. Through sucessive differential ultracentrifugation the membrane fraction was obtained. It was characterized by electron microscopy (ultrathin sections and freeze-fracture) and by using enzyme markers (acid phosphatase, succinate dehydrogenase, 5'-nucleotidase,Mg2 -ATPase, Na -K -ATPase, and adenylyl cyclase). With electron microscopy it was observed that the fraction had a slight contamination by ribosomes. The enzyme activities of 5'-nucleotidase and Na -K -ATPase, the two classical markers of the plasma membrane, were not detected. The specific activities of Mg2+-dependent ATPase and adenylyl cyclase in the membrane fraction showed a twofold and fivefold increase with respect to the whole homogenate, respectively; (2) rupture of the cells by sonication and isolation of the membrane fraction by differential centrifugation and equilibrium centrifugation in sucrose gradients. The fractions were characterized by electron microscopy, enzyme markers (adenylyl cyclase, succinate cytochrorne c reductase, acid phosphatase, and glucose-6-phosphatase), and distribution of I3II activity (in this case the cells were previously +

+

+

+

+

228

WANDERLEY DE SOUZA

labeled using the lactoperoxidase procedure). A 10-fold increase in the adenylyl cyclase activity was detected with this method. Na+ -K -ATPase and 5 I - n ~ cleotidase activities were not found; (3) induction of vesiculation of the plasma membrane by incubation of the cells with either cross-linking reagents (aldehydes, N-ethylmaleimide, p-chloromercuribenzoate) or acid buffers, followed by isolation of the vesicles by sucrose density centrifugation. The fractions were characterized by distribution of I3’I activity as in (2) and by electron microscopy. Unfortunately the authors did not assay the enzyme markers. Examination of the electron micrographs of T. cruzi treated with vesiculating agents showed that the formation of vesicles occurs at certain points of the membrane which encloses the cell body and the flagellum. Therefore, the plasma membrane fraction isolated by this procedure certainly is not representative of the whole plasma membrane of the cell. However, it is interesting since it opens the possibility for examination of specific regions of the cell surface of T. cruzi which, for unknown reasons, are susceptible to form vesicles. A 30- to 50-fold enrichment of bands A, B, and C was detected in the membrane fraction isolated from the vesicles. As suggested by Colli (1979) this fact may reflect “either a topographical differentiation of the plasma membrane or a lateral migration of these structures during cell vesiculation.” It is important to point out that no microtubules were seen in the region in which the vesicles appeared. An important conclusion from the studies discussed above is that the plasma membrane of epimastigotes of T. cruzi does not have two typical enzymes of the plasma membrane of eukaryotic cells: Na+ -K -ATPase and 5’-nucleotidase. These two enzymes are classic markers to determine the purity of a plasma membrane fraction. On the other hand, it seems evident that adenylyl cyclase is the best enzyme marker to determine the purity of a plasma membrane fraction of T. cruzi. It is interesting to note that also 5’-nucleotidase was not detected in trypomastigotes of T. brucei (Rovis and Baekkeskov, 1980). The low activity reported in L . collosoma (Hunt and Ellar, 1974) and in T . brucei (Voorheis et al., 1979) may represent the activity of nonspecific phosphatases, such as an alkaline phosphatase. In T. brucei, Na -K -ATPase was found and used to establish the purity of the plasma membrane fraction isolated from that protozoan. A 26-fold increase of specific activity was found for that enzyme (Rovis and Backkeskov, 1980). 3’-Nucleotidase has been recently used as a marker for the plasma membrane of Leishmania donovani (Gottlieb and Dwyer, 1981b) and T . brucei (McLaughin, 1982). Further studies need to be done in order to determine if this enzyme is found in the plasma membrane of other trypanosomatids. In epimastigotes of 7‘. cruzi an ATPase activity was biochemically detected (Pereira et af., 1978; Frasch et al., 1978). Such activity was not inhibited by ouabain. By using ultrastructural cytochemistry it was shown that an ATPase activity is localized in the plasma membrane (both in the cell body and flagellum) of amastigotes, epimastigotes, and trypomastigotes of T. cruzi (Meirelles et al., +

+

+

+

229

CELL BIOLOGY OF TRYPANOSOMA CRUZI

submitted). Such ATPase activity was not sensitive to ouabain and K , but was sensitive to Mg2 . Further studies are necessary in order to determine its role in the plasma membrane of T . cruzi. +

+

H. LIPIDSON

THE

MEMBRANE

The only detailed analysis of the lipids of the plasma membrane of T. cruzi was reported by Franco da Silveira (1979) and Franco da Silveira et al. (1981). The determination of the lipids from whole cells was performed by Oliveira et al. (1977). In whole cells it was found that from the total lipids, 35% are phospholipids and 65 are neutral lipids. Among the phospholipids, phosphatidylcholine is the most abundant (40%) followed by phosphatidylethanolamine (28%), phosphatidylinositol (1 2%), sphingomyelin (4%), and smaller amounts of cardiolipin, phosphatidic acid, lysolecitin, phosphatidylserine, and an unidentified phospholipid which accounts for 3%. Analysis of the membrane fractions isolated from epimastigotes of T. cruzi indicate that their composition varies according to the method used for the isolation of the fraction. While the protein:lipid rate for whole epimastigotes was 2.4, for membrane fractions isolated by mechanical rupture of the whole cell (MF) and isolated from vesicles induced by some substances (VF) the values found were 1 and 0.3, respectively (Franco da Silveira, 1979; Franco da Silveira et al., 1981). These data indicate that the VF is very poor in proteins. Among the phospholipids the MF was more abundant in phosphatidylethanolamine (45%) followed by phosphatidylcholine ( 13%), lysophosphatidylcholine ( 1 1%), lysophosphatidylethanolamine (6%), phosphatidylinositol (4%), cardiolipin ( l l % ) , and a unidentified phospholipid (1 1%) similar to that observed by Oliveira et al. (1977) in whole cells. In the VF phosphatidylcholine was the most abundant (33%) followed by phosphatidylethanolamine (31%), the unidentified phospholipid (13%), phosphatidylinositol (lo%),lysophosphatidylcholine (9%), and lysophosphatidylethanolamine(6%). Cardiolipin was not detected in the VF. Since cardiolipin is a phospholipid characteristic of the inner mitochondria1 membrane it is possible that the MF is partially contaminated with mitochondrial membranes of T . cruzi. Phosphatidylcholine and phosphatidyletanolamine were also the major phospholipids found in membrane fractions isolated form L. collosoma (Hunt and Ellar, 1974) and T. brucei (Voorheis et al., 1979). It was observed that the membrane total lipid fraction of T. cruzi contained neutral carbohydrates and long chain amino alcohols thus suggesting the existence of glycolipids in the plasma membrane of T . cruzi (Franco da Silveira et al., 1981). Unfortunately all the available data on lipid composition of T . cruzi have been obtained only in epimastigote forms. Further comparative studies would be important because other studies showed that the membrane fluidity of amastigotes and trypomastigotes is different from that observed in epimastigotes (Szarf-

230

WANDERLEY DE SOUZA

man et al., 1980; Leon et al., 1979). It also would be important to analyze the lipid composition of both the membrane which envelops the cell body and that of the flagellum. Previous freeze-fracture studies suggest that the two membrane may possess a different protein:lipid rate (De Souza et al., 1978b).

I. LYSISOF Trypanosoma cruzi It is well known that epimastigotes of T. cruzi are lysed when incubated in the presence of mammalian serum while trypomastigotes (both from axenic cultures, from tissue cultures, and from the bloodstream) are resistant (Muniz and Borrielo, 1945; Nogueira et al., 1975). It was shown that such lysis is mediated by the alternative pathway of the complement (Nogueira et al., 1975). Taking in consideration that complement mediated lysis involves mainly the participation of the cell surface of the parasite it is possible that the differences found in the susceptibility of epimastigotes and trypomastigotes of T. cruzi to this type of lysis reflect differences in their surface properties. These differences have even been used as the basis of a methodology for obtaining axenic culture-derived trypomastigotes. In this case, axenic cultures at the stationary phase of growth, a condition in which a certain number of epimastigotes transform into trypomastigotes, are incubated in the presence of fresh mammalian serum. The epimastigotes are then readly lysed while the trypomastigotes are not altered and can be easily isolated by centrifugation in a gradient of either albumin (Nogueira et al., 1975) or metrizamide (Nogueira et al., 1980). More recently it was shown that trypomastigotes of T. cruzi treated with trypsin or neuraminidase become susceptible to mammalian complement-mediated lysis (Kipnis et al., 1981). This result suggests that the membrane structure which interacts with the complement is localized on the cell surface of both epimastigote and trypomastigote forms. However, in trypomastigotes it may be masked by sialic-containing components of the plasma membrane. Experiments in which the trypomastigotes were treated with trypsin or neuraminidase and then incubated for a few hours in culture medium showed that they are able to synthesize the membrane component which protects the trypomastigote from lysis. The synthesis of this component could be blocked by inhibitors of protein synthesis thus suggesting its glycoprotein nature (Kipnis et al., 1981). Unfortunately nothing is known about this subject for the intracellular forms of T. cruzi. The only existing data were reported by Muniz and Borrielo (1946) who showed that amastigotes had a behavior similar to that found in trypomastigotes. A more detailed study of this problem would be of interest in view of the fact that intracellular forms apparently have little or no sialic acid on their surface (Pereira et al., 1980). Studies made by incubating bloodstream forms of T. cruzi with serum containing antibodies against the parasite indicate that the cell surface of the parasite may vary from one strain to the other. This aspect has been more extensively

CELL BIOLOGY OF TRYPANOSOMA CRUZI

23 1

analyzed by Kretly (1978) and Kretly and Brener (1976). They showed that while bloodstream forms of the Y strain were agglutinated with serum those from the CL strain were not. Three hypotheses were proposed to explain this result: (1) CL trypomastigotes do not have in their surface antigenic sites to interact with antibodies; ( 2 ) the antigenic sites localized on the surface of CL trypomastigotes would have a more disperse localization, and (3) a process of capping and shedding of antigen-antibody complexes would avoid the agglutination of CL trypomastigotes. With the data available at present we can exclude all three possibilities for the following reasons: (1) antibodies do bind to the surface of CL trypomastigotes, as observed by immunofluorescein and immunoperoxidase labeling; ( 2 ) by immunoelectron microscopy no differences were observed in the distribution of antigenic sites in the surface of Y and CL trypomastigotes (De Souza, unpublished observations). However, we cannot exclude the possibility that in the two strains differences exist in the grade of fluidity of their plasma membrane; (3) the process of capping is less evident in CL than in Y trypomastigotes (Schmunis et al., 1980). It has been shown that bloodstream trypomastigotes (Budzko et al., 1975; Kierszenbaum et al., 1976) are lysed when incubated in the presence of specific antibodies and complement. Krettly (1978) showed that trypomastigotes from the Y strain, isolated from mice at the peak of parasitaemia, are lysed when incubated in the presence of fresh, uninactivated, normal human serum. It was shown that such lysis is dependent on the activation, mainly through the alternative pathway, of the complement by immunoglobulins adhered to the parasite’s surface. Trypomastigotes of the CL strain although they also have immunoglobulins adhering to their cell surface are not able to activate the complement. These data show clearly that differences may exist between some cell surface properties of bloodstream trypomastigotes of the Y and CL strains of T . cruzi. They may explain the differences found in the interaction of the two strains with macrophages (Alcantara and Brener, 1978; Kipnis et al., 1979; Meirelles et al., 1980). It has been shown that trypomastigotes of T . cruzi from the blood of infected mice are coated with immunoglobulins (Kloetzel and Deane, 1977; Krettli et al., 1979). Mice immunized with different T. cruzi antigens present antibodies which can be detected by immunofluorescence using formalin-fixed, but not living, parasites obtained from irradiated mice. Parasites incubated in the presence of these antibodies and complement are not lysed. However, mice chronically infected with T . cruzi harbor both antibodies which can be detected by immunofluorescence which leads to lysis of the living parasites when complement is added (Krettli and Brener, 1982). These antibodies were not detected in mice treated and considered to be cured as evaluated by negative fresh blood examinations, subinoculation, and hemoculture. These results are of great interest since they open the possibility for the use of antibodies against living

232

WANDERLEY DE SOUZA

bloodstream forms of T. cruzi, detected by complement-mediated lysis, as a criterium of cure in Chagas’ disease. Further studies are necessary in order to determine the nature of the antigenic sites which react with the two types of antibodies as well as if the antibodies belong to different immunoglobulin classes. More recently Krettly and Eisen (1980) studied the possibility of the occurrence of a process of fabulation in bloodstream trypomastigotes such as previously shown for Tetrahymena (Eisen and Talan, 1977). In this process the cell is able to cleave molecules of immunoglobulins attached to its surface so that only the Fab fragment remains associated with the parasite membrane. By using immunofluorescent antiserum against mouse Fab and Fc portions of the immunoglobulin it was observed that CL trypomastigotes cleave the immunoglobulin molecule down to Fab which remains attached to the parasite’s membrane. In contrast, the Y trypomastigotes had both Fab and Fc fragments attached to their plasma membrane. These data suggest that the resistance of CL trypomastigotes to complement-mediated lysis results from their ability to fragment the immunoglobulin molecule bound to their surface. V. The Subpellicular Microtubules One of the characteristic features of protozoa Trypanosomatidae is the presence of a layer of microtubules localized below the plasma membrane and designated as subpellicular microtubules (Figs. 4 and 5). It has been observed that the microtubules are connected to each other and to the plasma membrane by short filaments, of still unknown nature (Hunt and Ellar, 1974; De Souza, 1976a; Linder and Staehelin, 1977; Dwyer, 1980). This association is probably responsible for the rigidity of the cell and the difficulties found in the disruption of the cell by mechanical means. Bridges between adjacent microtubules have been reported in a variety of systems such as in the flagellar axoneme, in the axostyle, etc. It has been shown that dynein, a microtubule-bound ATPase, can bind to and induce cross-bridging between adjacent in vitro polymerized microtubules (Haimo et al., 1979). Dwyer (1980) suggested that the arms connecting the subpellicular microtubules to the plasma membrane of Leishmania donovani might represent dynein, although no concrete data were presented so far to support this suggestion. His results indicate that the microtubule-plasma membrane linkage is not mediated by either divalent cations or disulfide-type bonds since the complex was not altered when it was treated with EDTA, 2-mercaptoethanol, or dithiothreitol. By using surface tension disruption of various trypanosomatids, followed by critical point drying, Angelopoulos (1970) showed that the microtubules follow a longitudinal helical pathway in the protozoan cortex. Such array can also be seen

CELL BIOLOGY OF TRYPANOSOMA CRUZI

233

in whole cells adhering to a polylysine-Formvar-coated grid, treated slightly with Triton X- 100, fixed in glutaraldehyde, dehydrated, critical point dried, and observed in the high voltage electron microscope (De Souza and Benchimol, 1983). The nucleating center of the subpellicular microtubules has not yet been identified. Transversal sections through different regions of the trypomastigote form of T . cruzi show that the microtubules are regularly spaced, with a distance equalling 44 nm (center to center). Occasionally one or more microtubules may be lacking, specially in the region of attachment of the flagellum to the protozoan body (Meyer and De Souza, 1976). It was observed that the number of microtubules is related to the diameter of the cell. At the posterior and anterior regions of the protozoan the number of microtubules is smaller while the largest number is found between the nucleus and the kinetoplast, at the region where the Golgi complex is located. Observation of tangential sections of trypomastigotes showed no branching of microtubules, which could lead to an increase in number, or confluence which would result in reduction of this number. Observation of transversal sections through spheromastigotes localized inside the host cells indicates that the number of subpellicular microtubules varies according to the stage of the cell cycle. The maximum number of microtubules in trypomastigotes (120) is virtually the same as the minimum number of these structures in spheromastigotes. In transverse section of dividing forms, however, it was possible to count up to 220 subpellicular microtubules (Meyer and De Souza, 1976). Cross-sections of the subpellicular microtubules of T . cruzi show that they are hollow structures. Their wall has a thickness of about 5 nm and the central portion shows a diameter of about 20 nm. Based on the use of the image reinforcement technique, it was reported that the wall of the microtubules of T . brucei was composed of 12 subunits, called protofilaments (Fuge, 1968). With the use of an association of glutaraldehyde with tannic acid (Mizuhira and Futeasaku, 1972) for fixation of T . cruzi and H.samuelpessoai it was shown that each subpellicular microtubule is composed by 13 protofilaments (Baeta Soares and De Souza, 1977), as in most of the microtubules found in other cells. Little is known about the chemical composition and the behavior of the subpellicular microtubules of trypanosomatids. Based on studies in which T . cruzi was incubated at 0°C or in the presence of colchicine, treatments which depolymerize microtubules in other cells, it has been assumed that the subpellicular microtubules have a behavior similar to those of the flagellum. Any of the two mentioned treatments induced changes in the structure of the microtubules (De Souza, unpublished observations). In other cells the microtubules are made up of a major protein, called tubulin, and other protein components which are known as microtubule-associated proteins. Tubulin is an acid protein, with an isoelectric point of about pH 5.3, which is usually isolated as 115,000 dalton dimer, also known as 6.S tubulin. By

234

WANDERLEY DE SOUZA

treatment with denaturating agents as sodium dodecyl sulfate, urea, etc., it is possible to dissociate the dimer into two different 3 S subunits called as a-and ptubulins of nearly identical molecular weight (about 54,000). They have related, but somewhat different, amino acid sequences and are assumed to derive from a common ancestral protein in an earlier evolutionary period. The main difference between the two subunits is not due to posttranslational modifications since separate messenger RNAs, which code for each subunit, have been isolated from chick embryo brains. The tubulin molecule has binding sites for nucleotides, phosphate, amino acids (Bryan et a f . , 1978; Olmsteld and Borisy, 1975), Mg2+, and alcaloid drugs such as colchicine (Wilson and Bryan, 1974) and vinblastine (Wilson et al., 1975). The examination of tubulin isolated from different sources shows that it maintained most of its physicochemical characteristics during the evolution. Experiments in vitro have shown that it is possible to polymerize tubulin isolated from Chfamydomonas with tubulin isolated from Dicryostelium or from the mammalian brain (Piperno and Luck, 1977; Cappuccinelli et al., 1978). Besides tubulin, the microtubules also have other proteins which are necessary for the formation and function of the structure. Usually these proteins are involved in the generation of motility, in the creation of connections between microtubules and other structures, or in controlling the assembly and disassembly of the microtubules. Some of the microtubule-associated proteins are known, such as dynein (Gibbons, 1963) and nexin (Stephens, 1970) which are associated with the flagellar microtubules. Associated with cytoplasmic microtubules of mammalian cells, two groups of proteins have been detected: (1) some highmolecular-weight proteins, which have been designated as HMW and have a molecular weight of about 300,000 (Borisy et a f . , 1974), and (2) some proteins with a molecular weight around 70,000 which have been designated as tau. (Weingarten et al., 1975). These microtubule-associated proteins, which are known as MAPs, are structural components of microtubules which form thin projections attached to the tubules at regular intervals along their length and which are visualized by electron microscopy (Murphy and Borisy, 1975). Although no biochemical studies exist on the presence of MAPs in the subpellicular observations indicated that these microtubules seem to be similar to the other well characterized microtubules mentioned above. It has been shown that, provided certain conditions are fulfilled, tubulin can assembly into microtubules and vice versa. The assembly solution must have GTP, Mg2+, tubulin, MAPs, and a slightly acid pH. The reaction is reversible and temperature dependent. At physiological temperatures tubulin is assembled into microtubules, the reversing occurring at 0°C. Ca2+ is a potent inhibitor of microtubule assembly. Several experiments have indicated that the concentration of free Ca in the cytoplasm of a cell can regulate the process of assembly and disassembly of microtubules. The cytoplasmic concentration of Ca is regulated through a process of transport and liberation of the ions by the smooth endo-

CELL BIOLOGY OF TRYPANOSOMA CRUZI

235

plasmic reticulum and the mitochondrion (Carafoli and Crompton, 1978). Based on the data obtained in experiments made in vitro it was suggested that the polymerization starts with the presence of a few rings of polymeric tubulin with a sedimentation coefficient of 36 S. When the polymerization starts, first helical ribbons are formed. The tubulin dimers continue to attach themselves to the ribbons until a stage of 13 protofilaments is reached, when the ribbons fold up into a cylindrical shape to form a segment of a microtubule (Penningroth et al., 1976). In vivo, the cells would need a control to determine the places of assembly of tubulin into microtubules, their elongation, and their spatial orientation. The centers which would be able to organize the assembly of microtubules have been designated as microtubule-organizing centers (MTOC), some of which have been identified as the basal body, centriole, kinetochore, etc. At present we do not have precise knowledge about the localization of the organizing center for the subpellicular microtubules of trypanosomatids. Angelopoulos ( 1970) showed that after the disruption of the body of the protozoan the remaining microtubules were always connected to those of the flagellum in the region of the basal body. Further studies are necessary to clarify this point. The analysis of the various microtubules found in different cell types and their response to treatment of the cells with low temperature and with some alcaloids indicate that there are two main groups of microtubules: those which are in constant equilibrium with the tubulin pool through a dynamic process of polymerization and depolymerization and those which are permanent structures and usually do not depolymerize. The first microtubules are labile while the second are called stabile. Usually the cytoplasmic microtubules are labile while those which form the flagellum are stabile. Based on experiments with various physical and chemical agents we may consider the subpellicular microtubules of trypanosomatids as of the stabile group (Messier, 1971; De Souza, unpublished observations). The observation that tubulin has a binding site for the alcaloid drug colchicine opened the possibility of performing several biochemical studies. Using [3H]colchicine the amount of tubulin can be easily assayed. Experiments made on C . fusciculafa show that colchicine did not penetrate the cells but bound in a nonsaturable fashion to the membrane (Rembold and Langenbach, 1978) perhaps by a general hydrophobic interaction with membrane lipids. Taking in consideration that stabile microtubules do not bind colchicine, the understanding of the molecular structure of the subpellicular microtubules of trypanosomatids will progress slowly. However, the interesting association existing between these microtubules and the plasma membrane invites further studies. More recently some data have been published on the microtubules of trypanosomatids. It was found that the major protein from the subpellicular microtubules of Leishmania tropica is composed of two polypeptides which cross-react immunologically

236

WANDERLEY DE SOUZA

with the a and P subunits of pig brain tubulin. Digestion of the subunits with protease indicates that the P subunits of L . tropica and pig brain have similar primary structures. Significant differences were observed for the a subunits (Bordier et al., 1982). Genomic clones and cDNA of T . brucei have been recently isolated. Using different techniques it was suggested that trypanosome a-and P-tubulin genes are “physically linked and clustered in tandem repeats of approximately 13- 17 copies per haploid genome of alternating a-and P-tubulin sequences” (Thomashow et al., 1982). In other organisms, however, the tubulin genes are dispersed. In mammalian cells microtubules are believed to have two basic functions: (1) providing the structural support acting as a sort of cellular skeleton and (2) as part of the machinery recquired for certain types of movements of the whole cell, of intramembranous macromolecules and some intracellular structures. As to the role as a cytoskeleton which gives a certain form to the protozoan, it has been shown that when a trypanosomatid changes its form as a consequence of environmental factors or during the process of differentiation or dedifferentiation, the microtubules also change their spatial orientation (Meyer and De Souza, 1976). It is possible therefore that such changes are related to the grade of spiralization of the microtubules. A detailed study was performed to analyze the changes in shape which can be induced in H . samuelpessoai. A model was then presented to account for the relationship between microtubules arrangement, changes in the cell shape, and changes in the cell volume (De Andrade and De Almeida, 1980). Its central feature consists of an asymmetric departure from the regularly helicoidal distribution of the microtubules upon induction of shape changes. While some microtubules become more linear, other assumes a compensatory overspiralized course. Observations of the profile of the subpellicular microtubules made on both longitudinal and transversal sections of different forms of the protozoan support the model. Such studies should be extended to analyze systems in which the change in shape is more evident such as occurs during spheromastigote-promastigote and spheromastigote-trypomastigote transformation in Leishmania and in T . cruzi, respectively. It is possible that the short filaments which connect the microtubules to each other and to the plasma membrane are involved, through a mechanism of sliding, with the changes in shape which occur in the life cycle of trypanosomatids, causing dramatic changes in the relative positions of several intracellular structures such as the kinetoplast, the Golgi complex and the nucleus. Indeed, preliminary observations indicate that when bloodstream trypomastigotes of T. cruzi are incubated in the presence of cytochalasin B, a change occurs in the form of the protozoan. Ultrastructural analysis of such cells shows that some subpellicular microtubules are displaced, appearing below the normal layer of microtubules (unpublished observations). The available data do not indicate a participation of the subpellicular microtubules in the control of the mobility of intramembranous components of the

CELL BIOLOGY OF TRYPANOSOMA C R U Z l

237

plasma membrane of trypanosomatids. As mentioned previously, treatment of T . cruzi and L . donovani with microtubule-disrupting drugs did not influence the capping of antigenic sites and concanavalin A receptors (Schmunis et ul., 1980; Dwyer, 1976). If we remember that the microtubules of trypanosomatids are not altered by such drugs, these results are explained. The subpellicular microtubules of T. cruzi and other trypanosomatids are observed running throughout the whole protozoan body, with the exception of the region of attachment of the flagellum to the cell body and in the region of the flagellar pocket. This fact is of importance when one considers that endocytosis and/or exocytosis do not occur in any region of the protozoan which contains microtubules (De Souza et al., 1978a). It occurs only at the flagellar pocket region.

VI. Microfilaments Most of the eukaryotic cells have elongated structures with a diameter varying between 4 and 12 nm which have been designated as microfilaments. The thin filaments, with a diameter of 4 to 7 nm, are formed by actin. They can be observed free in the cytoplasm, mainly at the cell cortex or in bundles forming the so-called stress fibers. Other microfilaments, with a diameter of 8-9 nm, are known as intermediate microfilaments. Their composition varies from one cell to the other. Other microfilaments are still thicker (10 to 12 nm) and are composed by myosin. All these microfilaments are found not only in muscle cells but in almost all eukaryotic cells studied up to now (Schliwa and Van Blerkan, 1981). The study of the distribution of these filaments in different cell types has received great attention in the last few years, since it is possible to visualize them by immunofluorescence using specific antibodies labeled with fluorescein or rhodamine (Groeschel-Stewart, 1980). These experiments indicated that the microfilaments play important roles in several cellular processes such as the maintaining of the cell shape, control of the movement of intramembranous macromolecules, movement of intracellular structures, and movement of the cell. Few data are available about the presence of microfilaments in trypanosomatids. They are observed in the flagellum of almost all species of the Trypanosomatidae family forming the paraxial structure (see Fuge, 1969; Vickerman and Preston, 1976; De Souza and Souto-Padrh, 1980). As mentioned previously, short filaments 6 nm thick connect the subpellicular microtubules with each other and with the plasma membrane of all trypanosomatids. In some trypanosomatids a microfibrilar structure has been observed in the region of attachment of the flagellum to the cell body or to the wall of the intestine tube of the invertebrate host (Vickerman and Preston, 1976). Filamentous tracts were also observed in the cytoplasm of C.fusciculuta

238

WANDERLEY DE SOUZA

(Brooker, 1971a) and T. brucei (Vickerman and Preston, 1976). However, in any of the above mentioned sites the nature of the microfilaments was established. In the case of T. cruzi the presence of microfilaments was never observed in the cytoplasm. However, cytochalasin B, a drug which interferes with actin microfilaments, induces marked changes in the general morphology of bloodstream trypomastigotes and inhibits partially the parasite’s movements. The cytochalasin B effect is readily reversed by washing the cells (unpublished observations). However, treatment of the parasites with this drug does not interfere with the process of capping of antigenic sites (Schmunis et al., 1980). Further studies using fluorescein-labeled antibodies against actin, myosin, vimentin, etc. would help in a better understanding of the possible participation of microtubules in the structural organization of trypanosomatids. Using anti-actin antibodies, actin was recently observed by immunofluorescence in the cell body and mainly in the flagellum of T. cruzi. However, with the resolution provided by light microscopy it was not possible to identify the structures which reacted with the antibodies (De Souza et al., 1983). Further studies at the ultrastructural level need to be carried out.

VII. The Flagellum All members of the Trypanosomatidae family have a flagellum that emerges from an invagination called the flagellar pocket. In developmental stages such as promastigote, paramastigote, opisthomastigote, and choanomastigote, the flagellum emerges at the anterior tip whereas in epimastigote and trypomastigote forms it emerges somewhere along the side. The proportion of the total flagellar length to that inside the flagellar pocket region varies according to the developmental stage. The flagellum of T. cruzi has a basic structure which is similar to other flagella, showing a 9 + 2 pattern of axonemal microtubules. In trypanosomatids the length of the flagellum varies according to their developmental stage. For example, spheromastigotes of T. cruzi have a short flagellum, 1 p,m in length (Fig. 2). However, at the end of the intracellular cycle the parasite elongates and the flagellum also grows up to about 20 pm (Meyer and De Oliveira, 1948). This fact makes the flagellum of T. cruzi a suitable model for the study of the natural growth and shortening of this structure. Considering the results obtained with the assembly in vitro and in vivo of flagella of other cell types, in which growth occurs by addition of tubulin at the flagellar tip, it may be possible that also the growth of the flagellum of T. cruzi during spheromastigote-trypomastigote transformation takes place by the same process. Further studies are necessary in order to clarify the process of shortening of the flagellum. It would be important to find out about the destiny of the tubulin during this process. The flagellum is

CELL BIOLOGY OF TRYPANOSOMA CRUZI

239

present in all developmental stages of the life cycle of T . cruzi. Therefore, it is erroneous to call the intracellular stage of the parasite aflagellate or amastigote. Observations made by phase contrast microscopy show that the intracellular multiplicative form of T . cruzi is motile and has a short flagellum. The existence of a flagellum is also confirmed by electron microscopy (Meyer and De Souza, 1976; De Souza and Souto-Padr6n, 1980). At the side of the axoneme of the flagellum of trypanosomatids there is a filamentous, lattice-like structure which has been called paraflagellar or paraxial rod (Figs. 13 and 14). The nature, distribution, structure, and function of it has not yet been studied in detail. It appears as a network of microfilaments laterally localized to the major axis of the axoneme (Fuge, 1969; De Souza and SoutoPadrbn, 1980). Observations made in several species of trypanosomatids show that the paraxial structure is more developed in some species than in others (De Souza and Souto-Padrh, 1980). For example, it appears to be highly developed in Herpetomonas megaseliae. As a consequence of the presence of the paraxial structure, the flagellum of trypanosomatids is wider than that of other cells. Cross-sections have been used to determine the relative position of the paraxial structure to the main axis of the axoneme. For these studies the peripheral doublet microtubules of the axoneme were numbered in clockwise rotation as suggested by Afzelius (1959). Doublet 1 corresponds to that which is equidistant from the two central microtubules of the axoneme. In all trypanosomatids studied the paraxial structure is localized in a region comprised between the peripheral doublets 4 and 7, thus indicating that it maintains a fixed position relative to the axonemal microtubules (Vickerman and Preston, 1976). In epimastigote and trypomastigote forms the flagellum, after its emergence from the flagellar pocket, is attached to the cell body along the entire length up to its free anterior portion. Examination of cross sections at different levels of the protozoan body, which were recognized by the determination of the number of subpellicular microtubules (Meyer and De Souza, 1976), shows that the relative position of the paraxial structure to the axoneme remains constant. However, its relative position to the cell body varies considerably. Sometimes it is in touch with the cell body, sometimes at its side (De Souza and Souto-Padrh, 1980). In the bodonine flagellates it was observed that the paraxial structure also maintains a fixed position relative to the axoneme of the anterior flagella. However, in the recurrent flagellum it is comprised between doublets 2 and 5 or 3 and 6 (Vickerman and Preston, 1976). The paraxial structure has been found in almost all trypanosomatids studied up to now, with two exceptions: (1) it is absent from the flagellum of trypanosomatids which possess an endosymbiont in the cytoplasm as occurs in Crithidia oncopelti (Burnasheva et al., 1968), Crithidia deanei, Blastocrithidia culicis, Blastocrithidia sp., and Crithidia sp. (Freymuller and Camargo, 1981; Lk Souza, unpublished observations); (2) it was not found in intracellular, multi-

FIGS. 13. AND 14. Thin section and negative staining of the flagellum of Herpetomonas megaseliae and T . cruzi, respectively, showing the axoneme (A) and the array of filaments which form the paraxial structure (PS). X60,ooO. (Fig. 13, Courtesy of N. L. Cunha.)

CELL BIOLOGY OF TRYPANOSOMA CRUZI

24 1

plicative forms of T . cruzi (De Souza and Souto-Padrh, 1980) and, apparently, in Leishmania. In the spheromastigote form of T . cruzi the flagellum is short and almost completely located inside the flagellar pocket. It is well known that at the end of the intracellular cycle the spheromastigotes start a process of differentiation to epi- and trypomastigote forms, during which an elongation of the body and the flagellum occurs. It was observed that the paraxial structure appears just at the beginning of this process, suggesting that its components are synthesized and/or assembled as soon as the elongation of the flagellum begins. In T . cruzi the paraxial structure is seen only in the flagellar portion localized outside the flagellar pocket, which is short in this parasite. In stages such as promastigote, paramastigote, and opisthomastigote, that have a deep flagellar pocket, the paraxial structure can also be observed in the flagellar portion located inside the pocket (De Souza and Souto-Padrh, 1980). The paraxial structure is formed by microfilaments 6 nm thick, longitudinally oriented to the main axis of the axoneme and spaced at intervals of 27 nm. They are crossed in two directions by oblique filaments spaced at intervals of 27 nm and at an angle of 45" to the longitudinal filaments (De Souza and Souto-Padrh, 1980). It is possible that the analysis of the paraxial structure with optical diffraction techniques may give more detail about its structural organization. The nature of the microfilaments which form the paraxial structure is not known. They are formed by proteins, since the whole structure disappears when thin sections are treated with proteolytic enzymes (unpublished observations). These proteins are not basic, since the structure does not show reaction product when treated with ammoniacal silver or ethanolic phosphotungstic acid (SoutoPadr6n and De Souza, 1979). A strong fluorescence of the flagellum of T . cruzi was observed when Triton X- 100 permealized cells were incubated with fluorescein-labeled anti-actin antibodies. This staining was not observed in the flagella of Tritrichomonas foetus which does not possess the paraxial structure (De Souza et a l . , 1983). However, actin was not observed by polyacrylamide gel electrophoresis of a highly purified flagellar fraction of Herpetomonas megaseliae. However, an enrichment of two bands with molecular weights of 73,000 and 78,000 was observed (Cunha et al., 1982). It was observed that when flagella of T . cruzi were isolated with the detergent lubrol, most of them do not have the flagellar membrane. However, the paraxial structure remained associated with the axoneme. Observation of such preparations by negative staining with phosphotungstic acid showed the presence of projections 12 nm thick, regularly spaced at intervals of 22 nm, which connect the peripheral doublet microtubules of the axoneme to the paraxial structure (De Souza and Souto-Padrh, 1980). Nothing is known about the functional role of the paraxial structure in trypanosomatids. In euglenoids (Piccini et al., 1975) ATPase activity was found to be associated with it, indicating a possible role in the process of flagellar movement. Probably it is not involved in any step fundamental for the movement of

242

WANDERLEY DE SOUZA

the cell since trypanosomatids which do not possess such structure, as endosymbiont-harboring species such as Crithidia oncopelti, appear to have normal flagellar movements (Holwill, 1965; Holwill and McGregor, 1974, 1976). The absence of the paraxial structure is maintained in such parasites when they are transformed in endosymbiont-free species by treatment of the cells with antibiotics (Freymuller and Camargo, 1981; De Souza, unpublished observations). More detailed comparative physiological and biochemical studies of the flagellum of trypanosomatids may lead to a better understanding of the function of the paraxial structure. Recently (Hyams, 1982) it was shown that the paraflagellar rod of Euglena gracilis is composed of two proteins with molecular weights of 80,000 and 69,000. Trypanosoma cruzi as well as other members of the Trypanosomatidae family do not have mastigonemes, e.g., flagellar hairs or lateral projections from the flagellar surface. The flagellum of trypanosomatids usually is attached to the cell body. This attachment occurs only in a determined region of the flagellum of trypanosomatids in which the flagellum emerges from the central portion. However, it is extensive in developmental stages in which it emerges laterally as occurs with epimastigote and trypomastigote forms (Fig. 15). Such attachment is made by a junction which has been considered to be of the desmosome type (Vickerman and Preston, 1976). When the flagellar movement starts, the wave propagates throughout the flagellum and when reaching the region in which it is attached to the body of the parasite, induces an apparent movement of the body, giving the visual impression of an undulating membrane, and propagates toward the free extremity of the flagellum or vice versa. An undulating membrane of trypanosomatids does not appear in thin sections as an actual structure, as it does in Trichomonadidae (Honigberg et a l . , 1971). Observations of thin sections of the attachment zone by electron microscopy show that the junctional complex is formed by a linear series of apposed macular densities, each measuring 25 nm in diameter and formed by an amorphous material spaced at intervals of 90 nm. Such structures are clearly visualized in T . brucei (Vickerman, 1969; Smith et a l . , 1974; Hogan and Patton, 1976) but are less evident in T . cruzi (De Souza et a l . , 1978b). The presence of a dense material localized in the space between the cell body and the flagellum has not been detected in T . cruzi nor the presence of filaments associated with the submembranous density. In other trypanosomatids, such components are found (Vickerman and Preston, 1976). Therefore, in T . cruzi the junction above mentioned cannot be considered as a typical desmosome, similar to that found between epithelial mammalian cells. Freeze-fracture studies indicate that there is a specialization of the flagellar membrane at the region of attachment of the flagellum to the cell body of epimastigote and trypomastigote forms of T . cruzi (Martinez-Palomo et a l . , 1976; De Souza et al., 1978b). It appears as clusters of intramembranous parti-

CELL BIOLOGY OF TRYPANOSOMA CRUZI

243

FIG. 15. Thin section of an epimastigote form of T . cruzi showing the kinetoplast (K), two basal bodies (b), and the region of the attachment of the flagellum (F) to the cell body (arrows). An oblique section through the cytostome (C) is also seen. X60.000.

244

WANDERLEY DE SOUZA

245

CELL BIOLOGY OF TRYPANOSOMA CRUZI

cles spaced at more or less regular intervals (Figs. 16-19). The clusters are formed by 6 to 14 particles and are seen both on the P and E faces of the flagellar membrane. In the case of epimastigotes of T . cruzi such clusters were not observed in any of the faces of the cell body plasma membrane (Fig. 16). In bloodstream trypomastigotes, however, they were observed also in the cell body plasma membrane (Figs. 18 and 19). This fact was more evident in replicas when the fracture jumped from the P face of the flagellar membrane to the E face of the cell body membrane, and vice versa, the case in which the rows of particles on the adjacent membranes were found to be in register (Fig. 18). Freeze-fracture studies made on T . brucei had revealed the presence of the clusters of membrane particles in the flagellar membrane (Smith et al., 1974; Hogan and Patton, 1976; Vickerman and Tetley, 1977). Such clusters were interpreted as representing “the only obvious specialization of the fracture surfaces of the cell body or flagellum that can be related to the junctional complexes” (Smith et al., 1974). Based on the fact that the clusters were observed only on the E face of the flagellar membrane of T . brucei it was suggested (Smith et al., 1974) that they could be a flagellar membrane specialization involved in the connection of the paraxial structure to the flagellar membrane. However, the observation of these clusters in both faces of the flagellar and cell body membranes in the regions of attachment in T . cruzi shows that the particles represent fractured protein components involved in the cell body-flagellar adhesion. In epimastigotes of T. cruzi, a second specialization was found in both P and E faces of the flagellar membrane (Fig. 17). It appeared as a linear array of particles, longitudinally oriented in relation to the main axis of the flagellum. The functional role of this structure was not yet determined (De Souza et al., 1978b). The nature of the connection between the flagellum and the cell body of trypanosomatids is still uncertain. Vickerman (1 969) observed that “trypanosomes dividing in citrated blood produce a completely free daughter flagellum while the parent flagellum remains attached, suggesting that Ca+ ions chelated by citrate are necessary for adhesion of the developing flagellum, but not for maintenance of attachement. Further studies in this area are necessary. In other developmental stages where the flagellar membrane makes contact with the cell body plasma membrane only at the regions of its emergence from +

”

FIGS. 16- 19. Freeze fracture images showing aspects of the adhesion of the flagellum (F) to the cell body (b) of epimastigote (Figs. 16 and 17) and trypomastigote (Figs. 18 and 19) forms of T . cruzi. In epimastigotes, particles arranged in clusters spaced at regular intervals (Fig. 16) or as a linear array of closely adjacent particles (Fig. 17) were seen on both faces of the flagellar membrane. In bloodstream trypomastigotes the presence of particles in clusters was seen either on the flagellar or the cell body membranes at the region of cell body-flagellar attachment. This is clearly seen in replicas where the fracture jumps from the flagellar membrane to the cell body membrane (Figs. 18 and 19). Fig. 16, x 150,000; Fig. 17, X95,OOO; Fig. 18, X64,OOO; Fig. 19, X I10,OOO. (After De Souza et al., 1978b.)

246

WANDERLEY DE SOUZA

the flagellar pocket, clumps of particles that stand out on the P face of the flagellar membrane have been observed in Lepfomonas collosoma (Linder and Staehelin, 1977), H . samuelpessoai (De Souza et a l . , 1979a), Lepfomonas samueli (Souto-Padrh e f al., 1980a), and Leishmania mexicana (Benchimol and De Souza, 1980). In some cells these clumps appear to be present on all sides of the flagellum while in others, they are observed only at one side. It is interesting to note that the clumps of particles, which resemble the desmosomes seen by freeze-fracture of epithelial cells, were found neither on the P nor on the E faces of the cell body membrane. In thin sections Brooker (1970) found that such junctions possess, in choanomastigotes of C. fasciculafa, all elements which characterize a desmosome, including the tonofilaments. Cytochemical observations made on trypanosomatids treated with ethanolic phosphotungstic acid showed that the aggregates of electrodense material localized below the cell body and flagellar membranes at the junction region are formed by basic proteins. Each dense plate measured about 3.4 nm and was separated from the others by intervals of about 1.7 nm, the same dimensions found for the clumps of particles seen in freeze-fracture replicas (Souto-Padrh and De Souza, 1979). Since these dense structures are seen to be associated both with the cell body and the flagellar membranes, it is difficult to explain the absence of clumps of particles on the P and E faces of the cell body membrane. Based on all these data it is possible to classify the junction mentioned above as a rudimentary desmosome, following the classification of Connel (1978). When T . cruzi is fixed in the presence of substances such as ruthenium red and alcian blue-lanthanum nitrate (De Souza e f a l . , 1977a, 1978b), or when incubated in the presence of colloidal iron hydroxyde and cationized ferritin particles, it has been observed that these substances can penetrate inside the flagellar pocket of epimastigotes but only in few trypomastigotes (Figs. 7 and 8). This observation suggests that the attachment of the flagellum to the cell body of trypomastigotes may act as an occluding junction which can control the nature of the substances which penetrate inside the pocket. Similar observations have been made in lower trypanosomatids. Since in trypomastigotes of T . cruzi, as well as in the lower trypanosomatids, the incorporation of nutritive material occurs at the flagellar pocket region we can understand why the cell needs to have mechanisms to control the substances which will be ingested. It is possible that the junction is a dynamic structure which can be assembled and disassembled according to the necessities of the cell. The flagellum of trypanosomatids can establish contact with all type of surfaces with which it interacts. Contacts have been observed between the flagellum of one cell with the flagellum of another cell, between the flagellum and the chitinous intima of the mosquito hindgut, with debris found in cultures, with millipore filters, etc. (for a review see Vickerman and Preston, 1976). The attachment occurs mainly through the flagellar tip and seems to be nonspecific

CELL BIOLOGY OF TRYPANOSOMA CRUU

247

since it occurs with all substrates tested. The flagellum of T. cruzi and Leishmania may play some role in the process of interaction of these parasites with the host cell. In many situations it has been reported that these parasites touch the host cell surface first with the flagellum and, then, start the process of entrance. A. FLAGELLAR MOVEMENT The sliding filament model (Satir, 1968) is the one generally accepted to explain the movement of flagella and cilia of eukaryotic cells. According to this model the flagellar movement is generated by the action of the dynein arms of the microtubular doublet in a process dependent on ATP. The experiments made on the lateral cilia of the molluscan gills, in which the cells were fixed at different stages of the beat cycle, are responsable for most of the knowledge on the role of nexin links and radial spokes on ciliary movement (Warner and Satir, 1974). High-resolution electron micrographs showed that in unbent regions of the cilium, the radial spokes are not linked to the central sheath and maintain a perpendicular orientation relative to the subfiber from which they originate. However, the radial spokes are attached to the central sheath and show an angular displacement which is proportional to the curvature of the cilium in the bent regions. Therefore, the radial spokes undergo an attachment- detachment cycle with respect to the central sheath according to the curvature of the cilium. The nexin links which connect adjacent doublets would prevent excessive sliding. It has been shown that the initiation of bending in flagella may occur at any point along the axoneme (Brokaw and Gibbons, 1973). In natural conditions, however, the bend is initiated at the base of the flagellum and the wave propagates from base to tip. The flagellum of trypanosomatids, however, is capable of propagating waves from both its base and tip (Holwill, 1965; Holwill and McGregor, 1974, 1976). In normal situations, freely swimming C. oncopelti maintains tip to base waves. However, when stimulated mechanically or electrically, a transient reversal of the wave direction occurs. When the cell is extracted in glycerol or in detergent solution and subsequently reactivated with ATP, it was observed that the flagella have lost the capacity for wave reversal and propagate waves only from the basal end of the flagellum. Holwill and Mcgregor (1974, 1976) showed that calcium ions control the direction, form, and the frequency of flagellar waves in C. oncopelti. Based on data obtained in normal and extracted flagella it was suggested that the flagellar membrane is involved in the control of the direction of wave propagation. The membrane would act as a receptor for extracellular events and regulates the intraflagellar Ca concentration which in turn controls the reaction to determine the direction in which waves will be propagated. The Ca would act in modifying the interaction

248

WANDERLEY DE SOUZA

between the spoke head and the central sheath of the flagellum (Holwill and Mcgregor, 1976). As previously stated, the flagellar membrane of trypanosomatids possesses a special array of intramembranous particles at the base of the flagellum. It has been suggested that such structure may be involved in the control of local membrane permeability and that it could represent an energy-transducing zone of the flagellum. Cytochemical studies have shown that this region also contains Ca2 -binding sites (Bloodgood, 1982). In a liquid medium, the movement of a flagellum produces a force able to propel the cell in the direction opposite to that of the wave propagation. In the more common case, where the wave is from the base to the tip, the cell is pushed by the flagellum. In observations made on trypomastigotes of T. cruzi, maintained in tissue cultures, it is easily observed that the parasite can move in both directions. +

B. ISOLATION OF THE FLAGELLUM Few attempts have been made for the isolation of the flagellum of trypanosomatids (Segura et al., 1977; Pereira et al., 1977; Piras et al., 1981). It was observed that the association of the flagellum with the cell body is stronger in trypanosomatids than in other cells which possess cilia and flagella. Treatment of T. cruzi with low pH, EGTA, and dibucaine did not liberate the flagellum. Pereira et al. (1977) were able to isolate a flagellar fraction which showed, by electron microscopy, a high degree of purity. The inconvenience of the method used was that, as a consequence of the use of lubrol, most of the flagella did not have the flagellar membrane. The isolated flagella gave electrophoretic patterns qualitatively identical to those flagella isolated from other sources. Tubulin, as expected, was the most conspicuous component. The authors did not indicate the possible relationship of one of the bands found with the paraxial structure. Piras et al. (1981) disrupting epimastigotes of T. cruzi by sonication in a medium containing sucrose, albumin, and calcium as stabilizers, obtained by differential centrifugation followed by isopycnic centrifugation a pure fraction containing flagella surrounded by their membrane. After treatment of the fraction with 0.2% Triton X- 100 they also obtained fractions containing either the axoneme or the flagellar membrane. Using the same methodology, with slight modifications, a highly purified fraction containing flagella from H. megaseliae was recently obtained (Cunha et al., 1982).

C. BASALBODY As in other cells, the flagellum of trypanosomatids is connected to the basal body. Although the trypanosomatids have only one flagellum, they have two

CELL BIOLOGY OF TRYPANOSOMA CRUZI

249

basal bodies closely located. Cross-sections through the basal body show that it is composed of nine triplets of microtubules which have been designated as microtubules A, B, and C. The A microtubule is the one more closely located to the axis of the cylinder. The basal body is separated from the extracellular portion of the flagellum by the axosomal plate. One of the central microtubules of the flagellum originates at the axosome. The second one originates at a region situated above the axosome. About 90 nm below the rim of the axosomal plate two structures are seen which have been designated as intermediate and terminal plates. The terminal plate makes the boundary between the transition zone and the basal body proper. Above the plate the peripheral tubules are doublets while below it they appear in triplets. The C microtubule of the basal body terminates near the terminal plate. All these components are clearly observed in the basal body of T. cruzi. The components of the basal body are assembled during division of the cell. It is common to find dividing cells with four basal bodies (De Souza and Meyer, 1974) thus indicating that both the basal bodies connected to the flagellum and the adjacent one are able to replicate. The basal bodies of T. cruzi and other trypanosomatids do not migrate to the nuclear region during division.

VIII. Kinetoplast-Mitochondrion T r y p a n ~ ~ ~cruzi r n a as well as other members of the Trypanosomatidae family possesses only one mitochondrion which extends throughout the cell body. At a certain portion of the mitochondrion, localized near the basal body, there is a complex array of DNA fibrils in the mitochondrial matrix which forms the structure known as kinetoplast (Figs. 2, 3, 8, 15, and 20-23). This structure is diagnostic of the order kinetoplastida to which the Trypanosomatidae family belongs. The presence of only one mitochondrion per trypanosomatid has been established by observation of serial thick sections (Paulin, 1975, 1977) or whole cells (De Souza and Benchimol, 1983) with high-voltage electron microscopy. In all developmental stages of the life cycle of T. cruzi cristae are present in the mitochondrion. Although no morphometric study has been carried out in T. cruzi there are apparently no significant changes in the mitochondrion during the life cycle of this trypanosomatid. Biochemical (for a review see Gutteridge and Rogerson, 1979) and cytochemical studies (Meirelles and De Souza, 1982) carried out in intracellular forms, in epimastigotes and trypomastigotes, also indicate that there are no clear differences in the respiratory metabolism between the three forms. The following mitochondrial enzymes were detected by ultrastructural cytochemistry: cytochrome oxidase, succinate dehydrogenase, isocitrate dehydrogenase, NADPH diaphorase, a-glycerophosphate dehydrogenase, and

250

WANDERLEY DE SOUZA

CELL BIOLOGY OF TRYPANOSOMA CRUZI

25 1

P-hydroxybutyrate dehydrogenase (Meirelles and De Souza, 1982). These enzymes were found to be located, as occurs with mammalian cells, in the inner mitochondrial membrane. Based on studies of the oxygen uptake of T. cruzi cultures in the presence of azide and cyanide it was suggested the T. cruzi has (1) one respiratory terminal sensitive to both inhibitors, (2) another cyanide sensitive but azide insensitive, and (3) a third insensitive to both inhibitors which may correspond to the a-glycerophosphate system. As discussed by Carneiro and Caldas (1982) the “reason for the existance of different respiratory terminal in T. cruzi could be explained based on the need of the parasite to adapt to different environmental conditions.’ ’ Important changes occur in the mitochondrion during the life cycle of Trypanosoma brucei and related species. The mitochondrion of the slender bloodstream forms lack cristae. These forms break down glucose to pyruvate which is not degradated. They oxidize NADH via a glycerophosphate oxidase system. However, in the stumpy bloodstream form tubular cristae appear. These forms are able to oxidise 2-oxoglutaric acid and proline thus indicating the participation of mitochondrial components. However, succinate is not respired (Brown er a l . , 1973; Vickerman and Preston, 1976). Morphometrical analysis indicate that the relative volume of the mitochondrion to the whole cell is three times greater in the stumpy than in the slender form (Hecker et al., 1973; Hecker, 1980). When bloodstream forms of T. brucei are placed in acellular culture medium dramatic changes occur in both the morphology and the physiology of the protozoan. After 24 hours in culture the mitochondrion becomes a reticulum rich in plate-like cristae. After some days in culture the protozoan synthesize many mitochondrial enzymes (succinate dehydrogenase, cytochrome oxidase, etc.) and the respiration becomes sensitive to cyanide (for a review see Vickerman and Preston, 1976). In other trypanosomatids the changes which occur in the mitochondrion are not so dramatic as in T. brucei. In Herpetomonas samuelpessoai it was shown that the development of the mitochondrion varies according to the growth conditions. In a defined medium rich in glucose the mitochondrion shows few cristae and has low cytochrome oxidase activity. In contrast, the number of cristae as well as the cytochrome oxidase activity increase when the cells are grown in a medium without glucose (De Souza et a l . , 1977b). Until now no satisfactory method exists to obtain a mitochondrial fraction from T. cruzi, which shows a satisfactory grade of purity and at the same time metabolic activity. Since the mitochondrion of trypanosomatids is branched and FIGS.20-23. Various aspects of the structural organization of the kinetoplast DNA of T. cruzi. In epimastigotes (Fig. 20) the kinetoplast appears as a rod-like structure. The kinetoplast of an intermediate form is showed in Fig. 22. Figure 21 shows a aspect of the kinetoplast which is seen in most of the trypomastigotes. The basket-like aspect of the kinetoplast, found in few trypomastigotes, is shown in Fig. 23. Figs. 20-22, X60,OOO; Fig. 23, X20,OOO.

252

WANDERLEY DE SOUZA

distributed throughout the whole cell body, it almost certainly will not be possible to isolate the whole organelle intact. Some attempts have been made to isolated mitochondria1 fractions from some trypanosomatids (Simpson, 1972; Braly et a l . , 1974). An important role played by the mitochondrion in eukaryotic cells is the control, in conjunction with the endoplasmic reticulum, of the concentration of free Ca2+ in the cytoplasm. It is well known that the concentration of Ca2+ regulates many cellular processes such as the sliding of ctyoplasmic filaments, the assembly and disassembly of microtubules, the processes of cellular fusion, etc. However, there are no studies on the presence of Ca2 -binding sites or sites of accumulation of Ca in trypanosomatids. +

A. KINETOPLAST The kinetoplast is a specialized region of the unitary mitochondrion of protozoa Trypanosomatidae and Bodonodidae. It is a basophilic structure which was first observed by Ziemann (1898) and by Rabinovitsch and Kempner (1899) in Trypanosoma rotatorium and Trypanosoma lewisi, respectively. Several terms, such as nucleolus, parabasal body, centrosome, micronucleus, blepharoplast, and kinetonucleus, have been used to designate the structure seen, by light microscopy, at the base of the flagellum of trypanosomatids. The term kinetoplast, introduced by Alexeieff (1917), is the one which is used by all protozoologists. The association of the kinetoplast with the mitochondrion of trypanosomatids was suggested based on staining of the mitochondrion with the Janus green B stain (Shipley, 1916). However, only the observation of thin sections of trypanosomatids with the electron microscope clearly demonstrated the association of the kinetoplast with the mitochondrion (Meyer et a l . , 1958; Steinert, 1960; Clarke and Wallace, 1960). There is some confusion about what is considered to be the kinetoplast. Some authors consider the kinetoplast to be the portion of the mitochondrion which contains the fibrous DNA network (Simpson, 1972; Kallinikova, 1981). Others, among which this reviewer is included, consider as kinetoplast only the fibrous network which undoubtedly represents the basophilic body seen by light microscopy (Vickerman and Preston, 1976). Therefore the term kinetoplast will be used in this article to refer to the filamentous material observed inside the mitochondrion of the trypanosomatids. During the normal life cycle of trypanosomatids there are changes in the position of the kinetoplast relative to the nucleus. In the spheromastigote form of T . cruzi the kinetoplast is localized in the anterior portion of the cell. In the trypomastigote form it is localized in the posterior region. In both situations the basal body, from which the flagellum originates, is always localized close to the

CELL BIOLOGY OF TRYPANOSOMA CRWU

253

kinetoplast. Ultrastructural studies did not show any structures which would connect the basal body to the kinetoplast. However, we cannot exclude the possibility that filaments, which would not be preserved by the usual procedures of preparation of biological specimens for electron microscopy, may exist. Further immunocytochemical studies may clarify this important point. It is interesting to point out that in some preparations in which a kinetoplast-enriched fraction was obtained the basal body was found connected to the kinetoplast (Braly et a l . , 1974; Simpson, 1972). It is possible that divalent cations may be involved in the preservation of this association since these two structures could be dissociated by 2 mM EDTA and mild shearing force (Simpson, 1972). It has been suggested, based on ultrastructural observations, that during the changes which occur in the relative position of the kinetoplast to the nucleus the membranes of these two organelles could rupture and that exchange of their contents could occur (Chacraborty and Sanyal, 1962, cited by Kallinikova, 1981; Muhlpfordt, 1963). However, an exchange of nucleus-kinetoplast material has not been shown by other authors. It is not clear which factors determine the localization of the kinetoplast near the basal body. Although such association is observed in most of the protozoa kinetoplastidae, there are some exceptions: (1) in some strains of certain species ( T . equiperdum, T . evansi) the filamentous material can be seen distributed throughout the whole mitochondrion. The same occurs when some trypanosomatids are treated with intercalating drugs such as ethidium bromide and acriflavine (for a review see Hadjuk, 1978); (2) ultrastructural studies have shown that in some bodonidae (Ichyobodo necator, Cryptobia vaginalis, and Dimastigella frypanformis) it is not localized near the basal body but dispersed throughout the whole mitochondrion (Joyon and Lom, 1969; Schubert, 1966; Vickerman, 1974). In some cases although the filamentous material is dispersed, it is concentrated enough in some regions of the mitochondrion to be observed by light microscopy after Feulgen staining. These protozoa have been properly designated as pankinetoplastics (Vickerman, 1974). The fact that such forms exist with filamentous material localized at several regions of the mitochondrion is a good reason to use the term kinetoplast to designate only the filamentous material independently of the region in which it is located. Evidence had been obtained which indicated that the kinetoplast is composed by DNA. It was Feulgen-positive (Baker, 1961) and incorporated tritiated thymidine (Cosgrove and Anderson, 1954). These two properties could be abolished by DNase digestion. By electron microscopy it was then shown that the kinetoplast is formed by fibrils similar to those found in the nucleoid of bacteria (Ris, 1962). These fibrils as the incorporation of [3H]thymidine was eliminated by DNase digestion (Anderson and Hill, 1969; Burton and Dusanic, 1968; Ozeki et al., 1971). Later, the filamentous material was isolated and, as discussed below, showed to be composed of DNA.

254

WANDERLEY DE SOUZA

B. ULTRASTRUCTURE OF THE KINETOPLAST The kinetoplast of epimastigote and spheromastigote forms of T. cruzi present a similar morphology. The filamentous material is arranged in a tightly packed row of fibers oriented parallel to the longitudinal axis of the protozoan (Fig. 20). The whole structure appears as a slightly concave disk of 1 pm length, and a depth of 0.1 pm. Such parameters, altough constant for each species of trypanosomatids, vary from species to species. The values determined for several trypanosomatids were summarized by Simpson (1972). There is a space between the kinetoplast and the inner mitochondrial membrane in which mitochondrial cristae can be seen. This type of kinetoplast, which has been designated as type A, is observed in trypanosomes and leishmanias and in C. fusciculutu (Anderson and Hill, 1969; Brooker, 1971b; Kusel et ul., 1967). In other trypanosomatids, such as C. luciliue and Blustocrithidiu sp. (Steinert and Van Assel, 1967; Muhlpfordt, 1963), the kinetoplast is in more direct contact with the inner mitochondrial membrane, and is called kinetoplast type B. At least some fibrils of the kinetoplast DNA make contact with the inner mitochondrial membrane. Such contact is observed mainly in the portion of the mitochondrion which faces the basal body. In T. cruzi (both epimastigotes and spheromastigotes) the kinetoplast may have the appearance of a double layer. The fibrils appear to loop back upon themselves at the edge of the disk and to cross over one another along the central band. A granular material is sometimes observed at both sides of the kinetoplast (Sookrsi et ul., 1972; Brack, 1968) and it has been suggested that it could represent a site of RNA synthesis (Brack, 1968). The kinetoplast of the trypomastigote form of T. cruzi has an approximately rounded form and may present two aspects (Figs. 21-23). In one, it consists of three or four double-layered rows of fibers lying parallel to each other (Meyer, 1968; Brack, 1968; De Souza and Chiari, 1977). This structure has been designated as basket-like kinetoplast (Meyer, 1968). In the second type the filaments are in a more dispersed state, losing the ordered array typical of the kinetoplast of most of trypanosomatids. Preliminary observations made on trypomastigotes of T. cruzi obtained from different sources suggest that the aspect of the kinetoplast may be related with the grade of differentiation of the parasite. The basket-like kinetoplast is not frequently observed. Until now it was not found in bloodstream trypomastigotes. In axenic cultures forms showing basket-like kinetoplast were seen in division. Trypomastigotes with such kinetoplast are more frequently seen in axenic cultures, in the insect vector, and in tissue cultures. In all these situations the population of T. cruzi is a mixture of different forms in different stages of differentiation. Assuming that what occurs in tissue culture may also occur in vivo we can not exclude the possibility that some host cells may rupture before finishing the intracellular cycle of the parasite releasing imature try-

CELL BIOLOGY OF TRYPANOSOMA CRUZI

255

pomastigotes in the bloodstream. Therefore, there is a possibility of finding basket-like kinetoplasts in the bloodstream. An unusual kinetoplast structure has also been reported for Herpetomonas muscarum ingenoplastis (Wallace el al., 1973). Other trypanosomatids, which harbor endosymbionts in their cytoplasm, such as C.deanei and Blastocrithidia. sp. (Freymuller and Camargo, 1981) also have a different kinetoplast. C. KINETOPLASTDNA It was first reported by Schildkraut et al. (1962) that density gradient centrifugation of total cell DNA extracted from Crithidia showed two DNA satellite bands. One of the bands found in C. oncopelti results from the presence of an endosymbiont in the cytoplasm of this trypanosomatid (Marmur et al., 1963). Clear evidence that the satellite band found in Leishmania enrietti was not of nuclear origin was provided by Dubuy et al. (1965) after treatment of lysed cells with DNase I. Such treatment removed all nuclear DNA but did not interfere with the satellite DNA. In the last 10 years it has been conclusively shown in T. cruzi and in other trypanosomatids that the kinetoplast is formed by a DNA which has a special configuration. Since this topic has been the subject of three excellent reviews, one by Simpson in 1972 and the two others more recently by Borst and Hoeijmakers in 1979 and by Kallinikova in 1981, I will not discuss this topic in detail. The kinetoplast of T. cruzi has been studied in detail by Riou and co-workers (Riou and Delain, 1969; Riou and Pautrizel, 1969; Riou and Yot 1977; Riou and Paoletti, 1967; Brack et al., 1972). It is now established that the kinetoplast DNA represents between 20 and 25% of the total DNA of epimastigotes of T. cruzi. Electron microscopy showed that the kinetoplast consists of a network of 20,000 to 30,000 minicircle molecules associated with each other and with long linear molecules. Each minicircle has a length of about 0.45 km, which corresponds to about 1440 base pairs and a molecular weight of 0.94 X lo6 (Riou and Delain, 1969). Such an arrangement of the kinetoplast DNA was described also for other trypanosomatids (for reviewers see Simpson, 1972; Borst and Hoeijmakers, 1979; Kallinikova, 1981). It has been shown that the diameter of the minicircles from the K-DNA isolated from epimastigotes of T. cruzi corresponds to the width of the kinetoplast, as seen in thin sections. Similar data were obtained for many trypanosomatids (Simpson, 1972). It was suggested that the minicircles, figure eights, and loops of the long molecules are aligned in situ in package fashion to form the K-DNA quaternary structure. It is not well clarified which factors determine the stability of the well-structured association of DNA molecules in the kinetoplast. Basic proteins, which may, at least in part, neutralize the negatively charged DNA molecules in close contact, have been detected, by using

256

WANDERLEY DE SOUZA

cytochemistry (Heroin-Delaeney, 1965; Warton and Kallinikova, 1974) in association with the kinetoplast of several trypanosomatids. In the case of T. cruzi it was observed that changes occur in the pattern of appearance of reaction products, indicative of the localization of the basic proteins during the epimastigote-trypomastigote transformation (Souto-Padron and De Souza, 1978). Biochemical studies also indicate the existance of proteins associated with the kinetoplast of T. cruzi (Saucier et al., 1981). Further studies are necessary to determine the nature of these proteins. Restriction enzyme analysis has shown that the minicircles are heterogeneous in sequence (for a review see Borst and Hoeijmakers, 1979; Simpson and Simpson, 1980). Recently the digestion of kinetoplast DNA of T. cruzi with restriction endonucleases, followed by the electrophoretic analysis of the fragments on polyacrylamide gradient gels, has been used for the characterization of stocks, strains, and clones. It is possible that such an approach can be useful in the classification of trypanosomatids (Morel and Simpson, 1980; Morel et al., 1980). Since the first observations of highly purified K-DNA the presence of another component in the kinetoplast besides the minicircles has been suggested. Such a component was found then in the form of long DNA in several trypanosomatids. Recently a large circle, which has been designated as a maxicircle, with a diameter varying from 6 to 11 ym was observed in the K-DNA of C. luciliae (Borst and Hoeijmakers, 1979), T. brucei, and several other trypanosomatids. Such a structure, measuring about 10 pm and with a molecular weight of about 26X lo6 was recently detected in epimastigotes of T. cruzi (Leon et al., 1980). The information contained in the DNA of a minicircle is small. If transcription and translation occur, a small protein would result. Therefore, it is difficult to understand what is the functional role played by the minicircle in the physiology of the trypanosomatids. The maxicircles, however, have a size comparable to that found in the mitochondrial DNA of other eukaryotic cells. Therefore, it is possible that the maxicircles may contain genetic information. In other eukaryotic cells the mitochondrial DNA produces the RNAs of mitochondrial ribosomes, a set of tRNAs containing information for the synthesis of subunits of enzymes associated with the inner mitochondrial membrane such as cytochrome bc and ATPase (Borst, 1977). Evidence was recently given that a fragment of the maxicircle of T. brucei contains a DNA sequence that encodes cytochrome oxidase subunit I1 (Johnson et al., 1982). Zaitseva and co-workers (cited by Kallinikova, 198 1) isolated a subcellular fraction from C. oncopelti which was rich in kinetoplasts and was able to synthesize protein in v i m . They found polysomes, rapidly labeled messenger RNA, and RNA polymerase. The incorporation of amino acids was inhibited by chloramphenicol and resistant to cycloheximide. Ribosomes have also been isolated from the kinetoplast of C. luciliae (Laub-Kuperstejn and Thirion, 1974) and T. brucei (Hanas et al., 1975). Such studies have not yet been carried out in T. cruzi. Granules containing

CELL BIOLOGY OF TRYPANOSOMA CRUZI

257

ribonucleoproteins, as detected by ultrastructural cytochemistry, were recently observed in the kinetoplast of T . cruzi (Esponda er al., 1983). However, further studies are necessary to establish their nature and function. Two major RNAs, sedimenting at 12 and 9 S, and several other low-molecular-weight RNAs were detected in a kinetoplast-enriched fraction isolated from L . rarenrolue and other trypanosomatids (Simpson and Simpson, 1976, 1980; Cheng and Simpson, 1978). The 12 and 9 S RNAs lack a poly(A) tail, hybridize to the maxicircle fraction of K-DNA, and their synthesis is blocked by ethidium. It was also observed that both RNAs stimulate leucine incorporation into protein by a wheat germ system (Simpson and Lasky, 1975). Both RNAs did not hybridize to minicircle fragments. It was suggested that these RNAs are the mitochondrial ribosomal RNAs. Hybridization studies carried on T . brucei indicate that “the maxicircle is the genetically active component in the K DNA network” (Hoeijmakers er al., 1981). These authors point out several facts which indicate that the two major RNA species found initially in L . turenrolae (Simpson and Simpson, 1976) are mitochondria1 ribosomal RNAs. They would be the smallest rRNAs known in nature, a fact which could explain the difficulties found in isolating mitochondria1 ribosomes from trypanosomatids. Even in the other trypanosomatids there is no evidence that the kinetoplastic ribosomes contain RNAs complementary to either the minicircles or the maxicircles. Autoradiographic studies indicate that [3H]uridine is incorporated by the kinetoplast of trypanosomatids. The label could be removed by RNase digestion thus indicating the ability of the kinetoplast to synthesize RNA. The incorporation of uridine by the kinetoplast accounted for about 3% of the total precursor incorporated by the cell (Ozeki er al., 197 1). For many years it is known that some trypanosomatids do not have a typical kinetoplast neither by light nor by electron microscopic observations (Vickerman and Preston, 1976). Some of these parasites can be considered as mutants which have lost the ability to make a functional mitochondrion. These mutants are found either in nature, as occurs with some strains of T. evansi, T . equinum, and T . equiperdum or can be induced by treatment of some species of trypanosomatids with DNA-intercalating drugs. Several attempts have been made to obtain such mutants with T . cruzi. The mutants, which have been obtained mainly with T . equiperdum and T . brucei, show either a normal kinetoplast or do not show any. In the first case it is considered K + and in the second case, K - or dyskinetoplastic. The analysis of some K + mutants showed that some of them do not present maxicircles, as seen by restriction enzyme digestions and electron microscopy. The network formed by minicircles appears normal. In others, however, maxicircles were observed (Borst and Hoeijmakers, 1979; Fairlamb er al., 1978; Hadjuk, 1978). The analysis of K - or diskinetoplastic mutants showed that although they do not present a typical kinetoplast, the same relative amount of K-DNA is found as amorphous

258

WANDERLEY DE SOUZA

clumps of DNA not limited to the usual regions in which they are located, but dispersed throughout the whole mitochondrion (Hadjuk, 1976, 1978). Five micrometer circles, which resemble the maxicircles, are visualized in these mutants although minicircles are not observed. These results were recently discussed by Borst and Hoeijkmakers (1979) which suggest that some of these mutants “may represent deletion mutants of various parts of the trypanosome K-DNA sequence.” D. REPLICATIONOF THE KINETOPLAST The division of the kinetoplast is coordinated with the process of cell division. Division in T. cruzi begins with the replication of the basal body and flagellum followed by the division of the kinetoplast (De Souza and Meyer, 1974). As previously explained the kinetoplast of interphasic cells has a length and a width which is characteristic for a given species. In dividing forms the length of the kinetoplast increases until it reaches a certain value when partition occurs in the middle. In a certain percentage of epimastigotes of T. cruzi and T. conorhini the kinetoplast doubles transversally and the new band shifts sideways until it lies next to the original band, after which contriction occurs. In some forms parallel disks are seen looking like an amorphous material. By using electron microscope autoradiography Burton and Dusanic (1968) as well as Anderson and Hill (1969) showed that the incorporation of [3H]thymidine apparently occurred mainly along the periphery of the kinetoplast of dividing culture forms of T. fewisi and C .fusciculutu. However, incorporation of material was occasionally found in other regions of the kinetoplast. Unfortunatelly similar studies in which either a statistical analysis of the frequency of labeling patterns had been carried out, or synchronized cultures had been use, have not been carried out yet. Simpson and Simpson (1976) showed by autoradiography of isolated networks of K-DNA, after short pulses, that labeling occurred predominantly at two sites of the network rim separated by 180”. After longer pulses labeling of the entire rim was observed. Based on a large number of experiments carried out in several trypanosomatids the picture which emerges is that the minicircle replication takes place at two opposite positions of the network and leads to an increase in the size of the network. Concomitant with the minicircle replication the newly synthesized DNA is distributed through the network by a process of recombination (see Borst and Hoeijmakers, 1979).

IX. The Cytostome, Pinocytotic Vesicles, Lysosomes, and Multivesicular Structures All the Trypanosomatidae have an invagination of the plasma membrane where the flagellum emerges from the cell forming the flagellar pocket (Fig. 15).

CELL BIOLOGY OF TRYPANOSOMA CRUZI

259

This region is characterized by the absence of the subpellicular microtubules and by the presence of cytoplasmic vesicles. Incorporation of extracellular substances by pinocytosis is one of the most important mechanisms by which cells can ingest macromolecules. In protozoa this process is especially important when they are cultivated in axenic media containing large quantities of proteins. However, this mechanism may also be important in the uptake of macromolecules from the cytoplasm of the host cell by the intracellular stages of trypanosomatids. Steinert and Novikoff (1960) showed that Trypanosoma mega was able to incorporate femtin particles through a specialized region of its surface which was designated as the cytostome. This structure appears as an invagination of the plasma membrane which can penetrate deeply into the cytoplasm of the cell (Figs. 15, 24, and 25). Cross-sections through the cytostome show that the subpellicular microtubules remain associated with the plasma membrane in this region. In the case of T. cruzi the cytostome is found only in spheromastigotes and epimastigotes (Milder and Deane, 1969; Meyer and De Souza, 1973). It disappears during the transformation of these forms into trypomastigotes. In the case of African trypanosomes no cytostome is found (Vickerman and Preston, 1976). Pinocytotic vesicles, localized in the cytoplasm of trypanosomatids, are always seen in association with the cytostome and the flagellar pocket region (Fig. 25). Various studies, in which tracers such as ferritin and horseradish peroxidase were used (Bretaiia and O’Dally, 1976; Brown et al., 1965; Langreth and Balber, 1975; De Souza et d., 1978a), indicate that trypanosomatids incorporate these tracers by a process of pinocytosis. This process occurs at the region of the cytostome in those parasites which possess this structure. In those which do not have a cytostome the incorporation of material occurs through the flagellar pocket. In the case of intracellular forms of T. cruzi it has been shown that even large structures such as melanin granules, which measure 0.7 pm, can be incorporated through the cytostome region by a process of intracellular phagotophy (Meyer and De Souza, 1973) similar to that described in malaria parasites (Rudzinska et al., 1965). Evidence has been obtained indicating that the plasma membrane which covers the cytostome region differs in structure and composition from the plasma membrane of the other regions of the cell body of T. cruzi (Fig. 10). It has been shown by ultrastructural cytochemistry that the cytostome region binds more ruthenium red and concanavalin A than the membranes of the other regions of the protozoan. Freeze-fracture studies carried out on epimastigotes of T. cruzi show that the cytostome appears as a particle-poor region delimited by a pallisade-like row of adjacent intramembranous particles localized close to the flagellar pocket (Martinez-Palomo et al., 1976). During epimastigote-trypomastigote transformation the pallisade-like row of particles is gradually disintegrated, so that it is no longer observed in trypomastigotes (De Souza et al., 1978b). In the case of T. cruzi it has been shown that the pinocytotic vesicles contain-

260

WANDERLEY DE SOUZA

CELL BIOLOGY OF TRYPANOSOMA CRUZI

26 1

ing the peroxidase ingested by the parasite fuse with each other and possibly with vesicles originated from the Golgi complex, forming large multivesicular structures (De Souza et a l . , 1978a). These structures (Fig. 26) have different forms. Some of them appear to contain vesicles in which peroxidase activity was not detected. Others appear uniformly electron dense. It was observed that even short incubation times (about 10 minutes) of epimastigotes with horseradish peroxidase produced large multivesicular structures containing reaction product. Apparently the number of the multivesicular structures is much larger in epimastigotes than in trypomastigotes. Freeze-fracture studies show that the membrane of the vesicles and multivesicular structures formed by ingestion of proteins by T . cruzi have much less intramembranous particles than the plasma membrane. Since they have originated from the plasma membrane, it must have lost many intramembranous particles during this process (De Souza et a l . , 1978a). In the case of T . brucei it has been shown that ferritin uptake occurs through coated vesicles (Langreth and Balber, 1975) which empty into straight tubular extensions of a collecting system formed by smooth membranes. Small lysosomes, originating from the Golgi complex, fuse with the terminal portions of the tubules which contain ferritin, forming digestive vacuoles. However, such vacuoles do not resemble the multivesicular structures found in T . cruzi. In some lower trypanosomatids pinocytotic vesicles can be seen oriented along microtubules which probably are part of a rudimentar cytostome (Souto-Padron et al., 1980a). In these cells a large and unique membrane-bounded cavity containing many vesicles is observed. Such structure is clearly seen in C . fasciculata (Brooker, 1971a), Herpetomonas samuelpessoai (De Souza et a l . , 1976), and Leptomonas samueli (Souto-Padron et a l . , 1980b). Acid hydrolasis, which in other eukaryotic cells is associated with lysosomes, has been detected in homogenates of T . cruzi (Pereira et a l . , 1978) and other trypanosomatids (see Vickerman and Preston, 1976). Acid phosphatase, which can be cytochemically detected using the method of Gomori, has been used as an enzyme marker to characterize structures related with lysosomes. Such enzyme activity is usually detected in certain components of the Golgi complex, GERL, and in the lysosomes. In the case of trypanosomatids acid phosphatase activity has been detected, at the light microscopic level, by several authors (see VickerFIG. 24. Cytostome of epirnastigotes of T . cruzi. This region binds strongly concanavalin A (arrow). X50.000.

FIG. 25. Small vesicles (arrow) are seen near the cytostorne (C). X45,OOO. (After De Souza et

al., 1978a.)

FIG. 26. Epimastigotes incubated in the presence of horseradish peroxidase before glutaraldehyde fixation. Reaction product is seen in the interior of multivesicular structures. X45,OOO. (After De Souza er al., 1978a.)

262

WANDERLEY DE SOUZA

man and Preston, 1976). By electron microscopy acid phosphatase activity has been found in some cytoplasmic vesicles of T . brucei (Seed et al., 1967) and T . raiae (Preston, 1969), in the flagellar pocket of T . cruzi (Creemers and Jadin, 1967), C. fasciculata (Brooker, 1971), and T . brucei (Langreth and Balber, 1975), and in the Golgi complex of T . cruzi (Creemers and Jadin, 1967), C. fasciculata (Brooker, 1971b), and T . brucei (Langreth and Balber, 1975). More recently acid phosphatase activity was detected on the plasma membrane of L . donovani (Gottlieb and Dwyer, 1981a).

X. Electron-Dense Granules Electron-dense granules, often designated as “reservoirs” of metabolic products, pigment bodies, osmiophilic granules, polyphosphate vacuoles, etc. (Anderson and Ellis, 1965; De Souza et a l . , 1975; Heywood et al., 1974; Janovy et a l . , 1974; Maria et a l . , 1972; Molyneux and Robertson, 1974; O’Daly and Bretafia, 1976; Vickerman and Tetley, 1977), are a constant feature in trypanosomatids (Figs. 27 and 28). Usually the dense granules are located within membrane-bound cavities of circular profiles measuring from 0.2 to 0.4 pm in T. cruzi to 3 pm in Trypanosoma cyclopis (Vickerman and Tetley, 1977). They are distributed all over the cell, although in some trypanosomatids they are found more frequently around the nucleus and the kinetoplast. Some vacuoles are completely filled by the electron-dense material while others are only partially occupied, leaving electron-lucent areas. In T . cyclopis (Heywood et a l . , 1974) this aspect was interpreted as a consequence of “shrinkage of the pigment during processing for electron microscopy.” However, based on results obtained in H. samuelpessoai it was suggested that such an aspect may correspond to different stages of formation or degradation of the granules (De Carvalho et al., 1979). At least two types of dense granules can be found in T . cruzi and other trypanosomatids. One is homogeneous, is not membrane-bound, does not present electron-lucent areas, and is possibly constituted of lipids (Fig. 28) since it is not found in cells not fixed with osmium tetroxide. The second one is more electron dense and maintains its electron density also in cells which were not postfixed with osmium. Therefore, its density is due to its composition. It was shown, by using energy-dispersive X-ray microanalysis associated with transmission electron microscopy, that the dense granule of H. samuelpessoai is rich in iron (De Carvalho and De Souza, 1977). T . cyclopis, a trypanosomatid isolated from Malaysian primates, is also rich in pigments visible by light microscopy, which were suggested to correspond to the electron-dense granules seen by electron microscopy. It was reported that this trypanosomatid possesses the dense granule only when the protozoan was cultivated in the presence of hemoglobin (Heywood el al., 1974). However, similar studies by Vickerman and

CELL BIOLOGY OF TRYPANOSOMA CRUZI

263

FIG. 27. Electron-dense granule (arrow) which contains iron is shown in the cytoplasm of Herpetomonas samuelpessoai. X 25 ,OOO. FIG. 28. Bloodstream trypomastigote of T . cruzi showing portion of the mitochondrion (m), lipidic granule (L), and the peroxisome-like organelle (P). X40,OOO. (Courtesy of M. N. L. Meirelles.)

FIGS.29 AND 30. Aspects of the peroxisome-like organelle (P) found in Herpetomons samuelpessoai and Leptomonas samueli, respectively. A large multivesicular structure is also seen in L. samueli (*). X22,OOO. (Courtesy of T. Souto-Padron.)

264

WANDERLEY DE SOUZA

Tetley (1977) suggested that “the inclusion vacuoles and the brown pigment are different entities and that the latter occurs in separate post-nuclear vacuoles. ” Energy dispersive X-ray microanalysis showed that the dense granules contained no iron but have a high phosphate content, an appreciable calcium content, and possibly some Zinc. The postnuclear vacuoles, however, presented the K a and KP peaks of iron. The vacuoles found in H. samuelpessoai and in T . cruzi differ from the postnuclear vacuole of T. cyclopis. Both in T. cruzi (Meirelles and De Souza, 1980) and in H. sarnuelpessoai (De Carvalho et af., 1979) the dense granules contain a peroxidase activity which can be detected by the diaminobenzidine medium at pH 7.6 (Graham and Karnovsky, 1966). Such peroxidase activity is cyanide sensitive and slightly inhibited by aminotriazole. Based on these data, associated with the observation that H. samuelpessoai grows poorly and has fewer dense granules when transferred into a hemin-deprived culture medium, it was suggested that the dense granules may represent a storage form of tetrapyrol derivatives which can be used by the protozoan for synthesis of cytochrome oxidase (De Carvalho et al., 1979). It is well known that the trypanosomatids require preformed tetrapyrroles, usually in the form of hemin or hemoglobin, as in vitro growth factor (Gangham and Karssner, 1971). A peak indicative of phosphorus was also observed in the X-ray spectra obtained from the dense granules of H . samuelpessoai (De Carvalho and De Souza, 1977). As pointed out by Vickerman and Tetley (1977) it is possible that the presence of polyphosphates in trypanosomatids may explain their survival in conditions where little or no respirable substrates are available. In organisms which lack guanidine phosphate, as in the case of C. fascicufata (Janakidewi et af., 1965), enzymes may exist which would catalyze the transfer of phosphate from polyphosphate to ADP. Polyphosphate granules may also represent a storage of phosphate to be used for the synthesis of nucleic acids and phospholipids (Vickerman and Tetley, 1977) from polyphosphate to ADP.

XI. Peroxisome-like Organelle The presence of single membrane-bound cytoplasmic structures which resemble the microbodies or peroxisomes found in many eukaryotic cells has been described in several trypanosomatids (Vickerman and Preston, 1976; McGhee and Cosgrove, 1980). They are usually spherical (Figs. 28 and 29) and have a single 6-nm-thick limiting membrane and a dense and granular matrix (Muse and Roberts, 1973; Brun, 1974; De Souza et al., 1976; Vickerman, 1974; Vickerman and Preston, 1976; Steiger, 1973). In Leptomonas samueli (Fig. 30), however, the microbodies are long structures which can reach 2.8 pm (Souto-Padron and De Souza, 1982).

CELL BIOLOGY OF TRYPANOSOMA CRUZI

265

The microbodies of some cells have a cristalloid core or dense plate in their matrix. The most typical example is the peroxisome of hepatocytes of mouse and rat. Such cristalloids, which are composed of urate oxidase, have been also observed in the microbodies of T . brucei (Vickerman, 1974) and L. mexicana (J. Alexander, cited by Vickerman and Preston, 1976). Cristalloid structures were not found in the microbodies of T . cruzi. Peroxisomes have been defined as organelles that are bounded by a single membrane and that contain catalase and H,O,-producing oxidases (De Duve and Beaudhuin, 1966). Catalase has been used as an enzyme marker which identifies an organelle as peroxisome. Its activity can be cytochemically detected by using the alcaline diaminobenzidine medium (Novikoff and Goldfischer, 1969). Light catalase and/or peroxidase activity, as detected by the DAB technique, has been detected in organelles found in epimastigotes (Docampo et al., 1976) and trypomastigotes (Meirelles and De Souza, 1980) of T . cruzi, in choanomastigotes of C . fasciculata (Muse and Roberts, 1973), and promastigotes of L . samueli (Souto-Padron and De Souza, 1981). In these studies it was shown that the enzyme activity was inhibited by aminotriazole which is quite specific for catalase. Catalase activity was also biochemically detected in microbody-enriched subcellular fractions of C . luciliae (Opperdoes et a l . , 1977). In most of the eukaryotic cells the peroxidatic activity found in the peroxisomes can be easily detected by using short incubation times in the presence of the DAB medium. The same occurs with C . fasciculata (Muse and Roberts, 1973). However, when this procedure was used for some trypanosomatids (L. samueli, H . samuelpessoai, and T . cruzi) very little peroxisomal staining was obtained. Similar results have been previously described for Tetrahymena pyriformis, and it was assumed that microbodies of this protozoan did not contain catalase. However, when the incubation time in the DAB medium was prolonged, reaction product could be seen in the microbodies of T . pyriformis (Fok and Allen, 1975). Similar results have been found for T . cruzi (Meirelles and De Souza, 1980) and L . samueli (Souto-Padron and De Souza, 1981). It is not yet understood which are the factors responsible for the difficulties found in the detection of catalase activity in trypanosomatids. It is possible that either the catalase concentration or its peroxidatic activity or both are lower in some protozoa than in mammalian cells. The role played by the peroxisomes in the cell physiology of trypanosomatids is not yet clear. In other eukaryotic cells it has been shown that they are involved in the oxidation of amino acids and in the formation and breakdown of hydrogen peroxide; they are also involved in the conversion of stored fatty acids into carbohydrates and are the sites where the enzymes which participate in the glyoxylate bypass are located (De Duve and Beaudhuin, 1966; Muller, 1975). Morphometric studies show that the microbodies contribute to the total cell volume in about 3-4% in the case of T . brucei (Bohringer and Hecker, 1975) and

266

WANDERLEY DE SOUZA

5% in L. samueli (Souto-Padron and De Souza, 1980). They are also rich in basic proteins, as was cytochemically detected by using the ethanolic phosphotungstic acid and the ammoniacal silver techniques (Souto-Padron and De Souza, 1978, 1979). As discussed previously, T. brucei and related species have one well-developed mitochondrion only in some stages of the life cycle while in other stages the mitochondrion lacks mitochondria1cristae. The microbodies, however, are present in all stages of the life cycle of the protozoan. It was first suggested that the microbodies of T. brucei contained glycerol-3-phosphateoxidase. However, the results obtained by Opperdoes et al. (1977) showed clearly that this enzyme is located in the mitochondrion only while the particle-bound NAD-linked glycerol-3-phosphatedehydrogenase of T. brucei and C . luciliae is located within the microbodies. It was also shown that the glycolytic enzymes involved in the conversion of glucose to 3-phosphoglycerateare located in the microbodies of T. brucei (Opperdoes and Borst, 1977; Visser et al., 1981), T. cruzi, and C . fasciculata (Taylor et al., 1980). Based on these results the term glycosome was suggested to designate the microbodies of trypanosomatids. In the author's opinion all these data do not exclude the possibility that the microbodies of trypanosomatids are peroxisomes for the following reasons: (1) as previously discussed the presence of catalase, which is an enzyme marker of peroxisomes, has been detected in several trypanosomatids. Although catalase activity was not detected in T. brucei (Steiger, 1973) we can not exclude the possibility that this protozoan has a low enzyme acitvity which can be detected only using special conditions of incubation in the DAB medium. Even the absence of catalase in peroxisomes may occur in certain conditions. Indeed it has been shown that microbody proliferation may occur in rat liver in the complete absence of cytochemically demonstrable peroxidase activity. (2) D-amino acid oxidases (Muse and Roberts,' 1973) as well as isocitrate lyase and malate synthase (Frugulhetti and Rebello, 1977), three enzymes which have been found in peroxisomes of eukaryotic cells, were detected in C. fasciculata and H . samuelpessoai, respectively; (3) NAD-linked glycerol-3-phosphate dehydrogenase, which is found in the microbodies of T. brucei and C . luciliae, was found in the same subcellular fraction of C. luciliae which also contained catalase (Opperdoes et al., 1977). More recently some data have been obtained indicating that the glycosomes of trypanosomatids are involved in carbon dioxide fixation (Opperdoes and Cottem, 1982) and pyrimidine biosynthesis de now (Hammond and Gutteridge, 1982; 1983). Taking all these results together it seems to us that Trypanosomatidaedo have a structure which can be considered as a peroxisome. However, certainly it may have been adapted in trypanosomadis to play other functional roles in addition to those described for the peroxisomes of mammalian cells.

CELL BIOLOGY OF TRYPANOSOMA CRUZI

267

XII. Endoplasmic Reticulum, Ribosomes, and Golgi Complex Rough as well as smooth endoplasmic reticulum is observed in T . cruzi as well as in other trypanosomatids. These structures are especially evident in the perinuclear region and in the region between the nucleus and the flagellar pocket. It is common to find profiles of the endoplasmic reticulum below the plasma membrane and sometimes between the subpellicular microtubules (Fig. 5). No data exist about the composition and the functional role of the endoplasmic reticulum in T . cruzi. It is assumed that it plays the same role demonstrated for other eukaryotic cells. Ribosomes are distributed throughout the cytoplasm of T . cruzi. However, few data exist about their structural organization and composition. They are composed of about 45% protein and 55% RNA (Tanowitz et al., 1975; Grynberg, 1976; Morales and Roberts, 1978; Castro et al., 1981a). Analysis of the ribosomal RNA by centrifugation on sucrose gradients shows the presence of the 5, 21, and 26 S characteristic peaks. The Golgi complex is always present in T . cruzi and is located between the nculeus and the kinetoplast in the region near the flagellar pocket. In the trypanosomatids which have a cytostome the Golgi complex is located near this structure. It is formed by 3-10 cisternae with a structure similar to that found in other eukaryotic cells. Few data exist on the physiology of the endoplasmic reticulum- ribosome-Golgi complex system of T . cruzi. Using the periodic acid-thiosemicarbazide-silver proteinate technique reaction product, indicative of the presence of carbohydrates, was seen in the Golgi complex and in part of the endoplasmic reticulum of T . cruzi (De Souza, 1976b). Similar observations were also reported in T . brucei (Steiger, 1973). These observations suggest that in T . cruzi the ER and the Golgi complex are, as occurs in other eukaryotic cells, involved in the synthesis of glycoproteins. Studies carried out in several other cell types indicate that in some proteins (those containing N-glycosidically linked oligosaccharide chains) the process of their glycosylation is a cotranslational event which starts in the endoplasmic reticulum. However, 0-glycosylation as well as later steps in the processing of N-linked oligosaccharide chains are posttranslational events which take place in the Golgi complex (for a review see Hanover and Lennarz, 1981). The assembly of N-linked oligosaccharide chains involves dolicholphosphate as the anchor for the sugars. A series of glycosyltransferases catalyze sequential addition of sugar residues to the growing saccharide chain. After completion of the oligosaccharide chain, which may be signaled by the addition of three glucose residues, the chain is transferred from the lipid to an asparagine residue on a polypeptide chain after which the three glucose residues are removed and other sugars (usu-

268

WANDERLEY DE SOUZA

ally N-acetylglucosamine, galactose, and sialic acid) are added. Usually this oligosaccharide chain contains two N-acetylglucosamines, nine mannoses, and three glucoses. In the case of C . fasciculata it was shown that the transferred oligosaccharide chain contains two N-acetylglucosamines, seven mannoses, and no glucose (Parodi et al., 1981). In the case of T . cruzi recent results whow that the oligosaccharide chain which is transferred to the protein contains two Nacetylglucosamines, nine mannoses, and no glucose. However, during processing there is removal of one to three mannose residues and addition of glucose (Parodi and Quesada-Allue, 1982; Parodi and Cazzulo, 1982). Assembly of O-linked oligosaccharide chains involves direct transfer of Nacetylgalactosamine from UDP-GalNAc to a serine or threonine residue in the polypeptide. This assembly occurs in the Golgi complex (for review see Hanover and Lannarz, 1981). In the case of T . cruzi results obtained recently with the use of tunicamicin indicate that both N- and O-linked oligosaccharide chains are incorporated into proteins to form glycoproteins located in the plasma membrane of the parasite (Zingales et al., 1982a).

XIII. The Nucleus The nucleus of T . cruzi and other trypanosomatids has a structural organization which appears to be similar to that found in other eukaryotic cells. It is small, measuring about 2.5 pm. Therefore, few data can be obtained with the light microscope about its organization. Although the nucleus contains all information important for the life of the trypanosomatids and the control of their differentiation processes very few studies have been udnertaken so far on its structure. Most of the molecular biologists who work with trypanosomatids have centered their observations on the kinetoplast. However, in view of the importance of the nucleus we expect that it will be studied in more detail in the next few years. In trypomastigotes of T . cruzi the nucleus is elongated and localized in the central portion of the cell. In spheromastigotes and epimastigotes it has a rounded shape. It has a typical nuclear membrane provided with pores. Continuity between the outer nuclear membrane and the endoplasmic reticulum is evident. In favorable freeze-fracture preparations large areas of the inner and the outer nuclear membranes are exposed (Fig. 9). In the convex as well as in the concave faces of both membranes intramembranous particles are randomly distributed. The nuclear pores, as seen in freeze-fracture replicas, have a mean diameter of 80 nm; in L . samueli 25 pores per square micrometer of nuclear membrane were found (Souto-Padr6n et al., 1980a). In interphasic T . cruzi cells the chromatic material agglomerates into masses at

CELL BIOLOGY OF TRYPMOSOMA CRUZI

269

the periphery of the nucleus, below its inner membrane. Occasionally these masses are also found in the more central region. In the center of the nucleus, or situated slightly eccentrically, the nucleolus may be found. During division, changes occur in the organization of the nuclear material. The fate of the nucleolus during the division of trypanosomatids varies from species to species. In T. rhodesiense it persists in the form of small fragments which are easily identified as nucleolar material (Vickerman and Preston, 1970). In T. raiae (Vickerman and Preston, 1970) as well as in T. cruzi (De Souza and Meyer, 1974) it is dispersed completely, reappearing at the final phases of the cell division. In the beginning of the division process of T. cruzi, when the basal body is replicating, the first signs of division can be observed in the nucleus. The chromatin material localized below the inner nuclear membrane and the nucleolus disappear. Both are dispersed over the whole nucleus giving it a homogeneous aspect. Immediately after replication of the basal body, when the kinetoplast shows no morphological signs of division, microtubules appear inside the nucleus of trypanosomatids (Bianchi et al., 1969; Vickerman and Preston, 1970; De Souza and Meyer, 1974; Heywood and Weinman, 1978; Solari, 1980). The nucleus which has a spherical form changes into a more oval one with the major axis perpendicular to the direction of the flagellum. During the continuation of the process the kinetoplast divides, the two newly formed structures move to the sides, and the nucleus becomes progressively elongated. In all studies made on the nucleus in division small electron-dense plaques were observed among the intranuclear microtubules. In the case of T. cruzi the spindle microtubules were seen in connection with the plaques (De Souza and Meyer, 1974; Solari, 1980). A more detailed study of the nuclear division of epimastigotes of T. cruzi, using serial sections and a three-dimensional reconstruction of each divisional stage, was recently reported by Solari (1980). He found that the equatorial spindle is formed by about 120 microtubules arranged in two sets of about 60 microtubules running from each pole to the dense plaques (Figs. 3 1 and 32) and divided into discret bundles which reach a single plaque. Solari (1980) identified 10 plaques which were not located in one plane but were distributed within a region which covers about 0.4 pm from both sides of the actual equatorial plane. The average dimensions of each plaque was of 200 nm length, of 120 nm width, and of 70 nm of thickness. Each plaque had a symmetrical structure formed by transverse bands. Before nuclear elongation occurs the dense plaques split in two halves and begin to migrate to the polar regions. At this time no microtubules were seen between the two halves of each plaque. All microtubules were localized between the plaques and the poles of the nucleus. In the elongated nucleus it was possible to see 10 half-plaques on each side of the dividing nucleus. The nature and functional role of the dense plaques are not yet clear. It has been suggested that they could represent specialized parts of noncondensed chromosomes (Solari, 1980). However, there are no data which support

FIGS.31 AND 32. Nucleus of an epimastigote form of T. cruzi in division. The nuclear membrane remains intact. Spindle microtubules (long arrow) and the dense plates (short arrows) are indicated. Fig. 31, x50,OOO; Fig. 32, X100,OOO. (Courtesy of A. Solari.)

CELL BIOLOGY OF TRYPANOSOMA CRUZI

27 1

this idea. They could also represent kinetochore-like structures which would play an important role in the process of separation of the nuclear material into the two new cells (Solari, 1980). Further cytochemical studies may shed some light on the composition of the electron-dense plaques thus helping to understand their function. As suggested by Solari (1980) “if the relationship to chromatin is proved by histochemical or structural evidence, T. cruzi would most probably have ten chromosomes. Similar studies should be carried out with dividing intracellular forms of T. cruzi and also with other members of the Trypanosomatidae family to confirm and extend these data. More recently a similar study was carried out in Blastocrithidia triatomae where three dense plaques were observed (Solari, 1983). When the nucleus is elongated it divides by constriction in the middle. When the division is completed the chromatin material and the nucleolus reorganize and assume the position seen in interphase cells and the microtubules disappear. During the whole process of division the nuclear membrane remains intact showing, however, a more irregular folded. No centrioles have been observed in connection with the microtubules nor have other structures been found which would suggest participation in their formation. The results obtained until now do not permit a precise statement of the functional role of the intranuclear microtubules in the dividing nucleus. It is feasible, however, to attribute to them the same function assumed for metazoan cells, i.e., helping to separate and move the genetic material into the two newly forming nuclei. They might also be engaged in elongating the dividing nucleus. Basic proteins are important components of the nucleus of eukaryotic cells. However, few investigations have been carried out concerning the presence of basic proteins in protozoa of the Trypanosomatidae family. Stainert (1965) used the acid dyes fast green and bromophenol blue to locate histones in bloodstream forms of T. lewisi and culture forms of T. cruzi and B . culicis. He observed staining of the nucleus of all trypanosomatids studied. Other authors (Beck and Walker, 1964; Stewart and Beck, 1967) used human serum containing an antibody to DNA-linked histone to detect histones on 67 species of the subphylum Sarcomastigophora. They did not find DNA-linked histone antigen in any of 28 species of trypanosomatids examined. Similar results were also obtained with T. gambiense (Thivolet et al., 1965). However, biochemical analysis performed on C. oncopelti showed the presence of histones associated with the DNA of this trypanosomatid. The histones included components electrophoreticallysimilar to those of calf thymus and two additional components rich in lysine and arginine (Leaver and Ramponi, 1971a). Reaction product indicative of the presence of basic proteins was also cytochemically detected in the nucleus of T. cruzi by the use of the ethanolic phosphotungstic acid and the postformalin ammoniacal silver techniques (Souto-Padr6n and De Souza, 1978). Recently, techniques have been developed for the purification of nuclei of trypanosomatids (Pereira, 1978). The isolated nuclei were able to incorporate ”

212

WANDERLEY DE SOUZA

[3H]UTP in a cell-free system. This incorporation could be inhibited by DNase, RNase, acriflavin, and actinomicin D, known inhibitors of the process of transcription. With the isolated nuclei some studies were carried out on the organization of the chromatin of T . cruzi. Chromatin is a term which is used to denote the complex of macromolecules (mainly DNA and associated proteins) which makes up the chromosomal material of the nonmitotic cells. In eukariotic cells there are five classes of histones which are distinguished from each other on the basis of their content of arginine and lysine. The arginine-rich classes are termed H3 and H4, the slightly lysine-rich classes are termed H2A and H2B, and the lysine-rich class is termed H1 (Kornberg, 1977). It has been proposed that the basic structure of chromatin consisted of a repeating subunit composed of DNA and histones. Every subunit is composed of a stretch of DNA approximately 200 base pairs in length together with eight molecules of histone, two of H2A, two of H2B, two of H3, and two of H4. The histone-DNA subunits are connected to each other by stretches of DNA so that the overall structure is essentially that of a beaded string. The histone-DNA subunits have been termed Nu bodies or nucleosomes. Electron microscopy of chromatin material that had been gently prepared from lysed nuclei indicates that the chromatin strands are made of beads with a diameter of approximately 10 nm connected to each other by strands of DNA 1.5 to 2.5 nm in diameter. Whereas the H2A, H2B, H3, and H4 species are present within the nucleosome particles the H1 protein appears to be associated with links of DNA conecting the nucleosomes. The chromatin of purified nuclei of T . cruzi has been studied recently by three different groups (Astolfi Filho et af., 1980; Rubio et al., 1980; Belnat et af., 1981). The results obtained indicate that nucleosomes, apparently similar to those found in other eukaryotic cells, are present in T . cruzi. The DNA repeat in length obtained after micrococcal nuclease digestion of T . cruzi nuclei was found to be 212 (Rubio et al., 1980), 200 (Astolfi Filho et af., 1980), or 185 (Belnat et al., 1981) base pairs. After further digestion a relatively stable core particle containing 140 base pairs of DNA was observed by Belnat et al. (198 1). However, Rubio er af. (1980) did not detect a fragment of 140 base pairs. Even with longer digestion periods or higher enzyme concentrations a core of 140 base pairs was not found. Electrophoresis of T . cruzi acid-soluble proteins of the isolated nuclei on both SDS-polyacrylamide and acid-urea-polyacrylamide gels showed four major bands which were believed to be the H4, H3, H2A, and H2B histones (Astolfi Filho et al., 1980; Rubio et al., 1980). However, a band comparable to the HI was not found. Astolfi Filho et af. (1980) found a fast-moving band, however, which was also seen in several other specie of trypanosomatids. They consider it as a subspecies of histone H1. Since the histone HI seems to be involved in chromatin condensation its absence, or the presence of one subspecies, may be responsible for the instability in the structure of the nucleosome and the fragility of the chromatin of 7'. cruzi as well as for the fact that the

CELL BIOLOGY OF TRYPANOSOMA CRUZI

273

chromatin of trypanosomatids does not condense during the division of these parasites (De Souza and Meyer, 1974; Solari, 1980). Based on analysis of nuclear DNA reassociation studies and microspectrofluorometric measurements of laser-induced fluorescence of cellular DNA, T. cruzi was found to be a diploid organism with a nuclear DNA content of 2.5 X lo8 nucleotide pairs (Lanar et al., 1981; Castro et al., 1981b). A similar conclusion was reached for T . brucei based on electrophoretic variation in 19 enzymes from a series of isolates (Tait, 1980). More recently, based on determination of the nuclear content of T . brucei by quantitative absorption and fluorescence cytophotometry of individual Feulgen-pararosaniline stained cells, Borst et al. (1982) reached the same conclusion. If we assume that the dense plaques seen in the dividing nucleus of trypanosomatids are kinetochores and that the number of plaques is identical to the number of chromosomes T . cruzi could be diploid since it would have 10 chromosomes but B . triatominae could not since it would have only 3 chromosomes (Solari, 1983). Some authors have suggested the possibility that a process of mating could occur in trypanosomatids although no direct proof has been obtained (Tait, 1980). More recently it has been reported that some mutants of C . fasciculata adhere to each other by the flagellum in a way which resemble the mating process which occurs with Chlamydomonas (Hughes el al., 1982).

ACKNOWLEDGMENTS

It is a pleasure to dedicate this review to Dr. H. Meyer. I am most gratiful to Drs. T. U . de Carvalho, N. L. Cunha, T. C. de Ara6jo Jorge, M. N. L. Meirelles, P. Pimenta, A. Solari, and T. Souto-Padron for providing me with illustrations, unpublished results, or discussion. I also thank Mrs. Elizabeth da Fonseca Menda e Sandra Maria de Brito for secretarial assistance. The author's work has been supported by UNDPiWorld BankiWHO Special Programme for Research and Training in Tropical Diseases, Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq), Conselho de Ensino para Graduados da UFRJ (CEPG), and Financiadora de Estudos e Projetos (FINEP).

REFERENCES Afzelius, B. (1959). J. Eiophys. Eiochern. Cyrol. 5 , 269-278. Alcantara, A , , and Brener, Z. (1978). Acru Trop. 35, 209-219. Alexeiff, A. (1917). C . R . Soc. Eiol. 80, 358-361. Allen, G . , Gurnett, L. P., and Cross, G. A. M. (1982). J. Mol. B i d . 157, 527-546. Alves, M. J . M., and Colli, W. (1974). J. Protozool. 21, 575-578. Alves. M. J. M., and Colli, W. (1975). FEES Lett. 52, 188-190. Anderson, W. A,, and Ellis, R. A. (1965). J. Protozool. 12, 483-489. Anderson, W. A,, and Hill, G. C. (1969). J. Cell Sci. 4, 61 1-620.

274

WANDERLEY DE SOUZA

Angelopoulos, E. (1970). J . Prorozool. 17, 39-51. Anziano, D. F., Dalmasso, A. P., Lelchuk, R., and Vasques, C. (1972). Infecr. Immun. 6,860-864. Araujo, D. G.,and Remington, J. S. (1981). J. Imrnunol. 127, 855-859. Araujo, F. G.,Handman, E., and Remington, J. S. (1980). J. Protozool. 27, 397-400. Araujo, F. G.,Sharma, S. D., Tsai, V.,Cox, P.,and Remington, J. S. (1982). Infecr. Immun. 37, 344-349. Astolfi Filho, S., De Almeida, E. R. P., and Gander, E. S. (1978). Acra Trop. 35, 229-237. Astolfi Filho, S., Martins De Sa, C., and Gander, E. (1980). Mol. Biochem. Parasirol. 1, 45-53. Baeta Soares, T. C., and De Souza, W. (1977). 2.Parasirenkd. 53, 149-154. Baker, J. R. (1961). Trans. Rv. SOC. Trop. Med. Hyg. 55, 518-524. Baker, J. R., and Price, J. (1973). Int. J. Parasirol. 3, 549-551. Barry, J. D. (1975). J. Prorozool. 22, 490. Barry, J. D. (1979). J. Cell Sci. 37, 287-302. Barry, J. D., and Vickerman, K. (1977). Parasitology 75. Beck, J. S., and Walker, P. J. (1964). Nature (London)204, 194-195. Belnat, P., Paoletti, J., and Riou, G.(1981). Mol. Biochem. Parasifol. 2, 167-176. Benchimol, M., and De Souza, W. (1980). J. Parasirol. 66, 941-947. Benchimol, M., Elias, C. A,, and De Souza, W. (1981). J. Parasirol. 67, 174-178. Bianchi, L., Rondanelli, E. G.,Carosi, G.,and Gema, G.(1969). J . Parasirol. 55, 1091-1092. Bloodgood, R. A. (1982). In “Prokaryotic and Eukaryotic Flagella” (W. B. Amos and J. G. Ducket, eds.), pp. 353-380. Cambridge Univ. Press, London and New York. Bohringer, S., and Hecker, H. (1975). J . Prorozool. 22, 463-467. Bordier, C., Garavito, R. M., and Armbruster, B. (1982). J. Protozool. 29, 560-565. Borisy, G.G.,Olmsted, J . B., Marcum, J. M., and Allen, C. (1974). Fed. Proc. Fed. Am. SOC. Exp. Biol. 33, 167-174. Borst, P. (1977). Inr. Cell Biol. Pap. Int. Congr. I s r pp. 237-244. Borst, P., and Hoeijmakers, J. H. (1979). Plasmid 2, 20-40. Borst, P., Van Des Ploeg, M., Van Hoek, J. F. M., Tas, J., and James, J . (1982). Mol. Biochem. Parasifol. 6, 13-23. Brack, C. (1968). Acra Trop. 25, 289-356. Brack, C. Delain, E., Riou, G.,and Festy, B. (1972). J . Vltrastruct. Res. 39, 568-579. Brack, C., Bickle, T. A,, Yvan, R., Barker, D. C., Foulkes, M., Newton, B. A,, and Jenni, L. (1976). In “Biochemistry of Parasites and Host-Parasite Relationships” (H. Van Der Bossche, ed.) pp. 21 1-218. North-Holland Publ., Amsterdam. Braly, P., Simpson, L., and Kretzen, F. (1974). J. Prorozool. 21, 782-790. Branton, D. S., Bullivant, S., Gilula, N. B., Karnovsky, M. J., Moor, H., Muhlethaller, K., Northcote, D. H., Packer, L., Satir, P., Satir, B., Speth, V.,Staehelin, L. A,, Steere, R. L., and Weinstein, R. S. (1975). Science 190, 54-56. Brener, Z. (1973). Annu. Rev. Yicrobiol. 27, 347-382. Brener, Z., and Alvarenga, N. (1976). In “New Approaches in American Trypanosomiasis,” pp. 83-86. Pan American Health Organization, Washington, D.C. Brener, Z., and Chiari, E. (1965). J. Parasitol. 6, 922-926. Bretana, A,, and O’Daly, J . A. (1976). Inr. J. Parasitol. 6, 379-386. Bridgen, P. J., Cross, G.A. M., and Bridgen, J. (1976). Nature (London) 263, 613-614. Brinkley, B. R., Cox, S. M., Pepper, D. A,, Wible, L., Brenner, S. L., and Pardue, R. L. (1981). J. Cell Biol. 90, 554-562. Brokaw, C. J., and Gibbons, I. K. (1973). J. Cell Sci. 13, 1-18. Brooker, B. E. (1970). Z . Zellforsch. 105, 155-166. Brooker, B. E. (1971a). Z . Zellforsch. 116, 532-563. Brooker, B. E. (1971b). Protoplasma 72, 19-25.

CELL BIOLOGY OF TRYPANOSOMA CRUZI

275

Brooker, B. E. (1971~).Proroplasma 73, 191-202. Brooker, B. E., and Preston, T. M. (1967). 1. Prorozool. 14, 41. (Suppl.) Broom, J. C., Brown, H. C., and Hoare, C. A. (1936). Trans. R. SOC.Trop. Med. H y g . 30,87- 100. Brown, K. N., Armstrong, J. A., and Valentine, R. C. (1965). Exp. Cell Res. 39, 129-135. Brown, K. N., Evans, D. A., and Vickerman, K. (1973). Int. J. Parasirol. 3, 691-704. Brun, R. (1974). Acra Trop. 31, 219-290. Bryan, R. N., Cutter, G. A,, and Hayashi, M. (1978). Nature (London) 272, 81-83. Budzko, D. B., Pizzimenti, M. C., and Kierszenbaum, F. (1975). Infecr. Immuno. 11, 86-91. Burnasheva, S. A., Ostravskaja, M. V., and Yurzina, G. A. (1968). Acra Prorozool. 6, 357-362. Burton, P., and Dusanic, D. G. (1968). J. Cell Biol. 39, 318-331. Camargo, E. P. (1964). Rev. Inst. Med. Trop. Scio Paulo 6, 93-100. Cappuccinelli, P., Martinotti, G . , and Hames, B. D. (1978). FEBS Len. 91, 153-157. Carafoli, E., and Crompton, M. (1978). Ann. N.Y. Acad. Sci. 307, 269-284. Carneiro, M., and Caldas, R. A. (1982). Acta Trop. 39, 41-49. Carvalho, R. M. G., Meirelles, M. N. L., De Souza, W., and Leon, W. (1981). Infect. Immun. 33, 546-554. Castro, A. C. L. C., and Ribeiro dos Santos, R. (1977). Rev. Goiana Med. 23, 1-13. Castro, C., Hernandez, R., and Castaneda, M. (1981a). Mol. Biochem. Parasirol. 2, 219-233. Castro, C., Craig, S. P., and Castaneda, M. (1981b). Mol. Biochem. Parasirol. 4, 273-282. Cheng, D., and Simpson, L. (1978). Plasmid 1, 297-315. Cherian, P. V., and Dusanic, D. G. (1977). Exp. Parasirol. 44, 14-25. Chiari, E. (1974). Rev. Inst. Med. Trop. Scio Paulo 16, 81-82. Chiari, E. (1981). Ph.D. thesis, University of Minas Gerais, Brazil. Chiari, E., De Souza, W., Romanha, A. I . , Chiari, C. A,, and Brener, 2. (1978). Acra Trop. 35, 113-121. Clarek, T. B., and Wallace, F. G. (1960). J. Prorozool. 7, 115-124. Colli, W. (1979). In “The Membrane Pathobiology of Tropical Diseases,” pp. 131-153. Schwabe, Basel. Colli, W., Andrews, N. W., and Zingales, B. (1981). In. Cell Biol., 1980-1981 pp. 401-410. Connel, C. J . (1978). 1.Cell Biol. 76, 57-75. Cosgrove, W. B., and Anderson, E. (1954). Anar. Rec. 120, 813-814. Crane, M. J., and Dvorak, 1. A. (1979). J. Prorozool. 26, 599-604. Crane, M. J., and Dvorak, J. A. (1980). J. Prorozool. 27, 336-338. Creemers, J., and Jadin, J. M. (1967). Bull. SOC. Pathol. Exor. 60, 33-58. Cross, G. A. M. (1975). Parasirology 71, 393-417. Cross, G. A. M. (1977). Ann. SOC. Belge Med. Trop. 57, 389-399. Cross, G. A. M. (1979). In “Pathogenicity of Trypanosomes” (G. Losos and A. Chovinard, eds.), pp. 32-37. International Development Research Centre, Ottawa. Cross, G. A. M., and Johnson, J. G. (1976). In “Biochemistry of Parasites and Host-Parasite Relationships” (H. Van den Bossche, ed.), pp. 413-420. Elsevier, Amsterdam. Cross, G. A. M., Holder, A. A,, Allen, G., and Boothroyol, J. C. (1980). Am. J. Trop. Med. Hyg. 29, 1027-1032. Cunha, N. L., De Souza, W., and Voloch, A. H. (1982). Proc. Annu. Meet. Basic Res. Chagas’ Dis. Caxambu, Brazil p. 90. De Almeida, M. L. C., and Turner, M. J . (1983). Nature (London) 302, 349-352. De Andrade, P. P.,and De Almeida, D. F. (1980). Enp. Parasirol. 50, 57-66. De Carvalho, T. U.,and De Souza, W. (1977). J . Parasitol. 63, 1116-1117. De Carvalho, T. U., Souto-Padron, T., and De Souza, W. (1979). Exp. Parasirol. 47, 297-304. De Duve, C., and Beaudhuin, P. (1966). Physiol. Rev. 46, 323-357. Delain, E., and Riou, G. (1969a). C.R. Acad. Sci. (Paris) 268, 1225-1227.

276

WANDERLEY DE SOUZA

Delain, E., and Riou. G. (1969b). C . R . Acad. Sci. (Paris) 268, 1327-1330. De Souza, W. (1976a). J. Microsc. B i d . Cell. 25, 189-190. De Souza, W. (1976b). Z. Parasitenkd. 48, 221-226. De Souza, W.. and Benchimol, M. (1983). J. Sftbmicrosp. Cytol. (in press). De Souza, W., and Chiari, E. ( 1977). Rev. Bras. Biol. 37, 67 1-675. De Souza, W., and Meyer, H. (1974). J. Protoioul. 21, 48-52. De Souza, W., and Meyer, H. (1976). 2. Parusitenkd. 46, 179-187. De Souza, W., and Souto-Padron, T. (1980). J. Parasitol. 66, 229-235. De Souzd, W., Grynberg, H., and Nery-Guimarles, F. (1975). Rev. Soc. Bras. Med. Trop. 9, 143-156. De Souza, W., Rossi, M. A,, Kitajima, E. W., Santos, R., and Roitman, I. (1976). Can. J. Microbiol. 22, 197-203. De Souza, W., Arguello, C., Martinez-Pdlomo, A,, Trissl, D., Gonzales-Robles, A., and Chiari, E. (1977a). J. Protozool. 24, 41 1-415. De Souza, W., Bunn, M., and Angluster, J. (1977b). J . Submicrosc. Cytol. 9, 355-361. De Souza, W., De Carvalho, T. U., Benchimol, M., and Chiari, E. (1978a). Exp. Parasitol. 45, 101-1 15.

De Souza, W., Martinez-Palomo, A , , and Gonzales-Robbles, A. ( I 978b). J. Cell Sci. 33, 285-299. De Souza, W., Chaves, B., and Martinez-Palomo, A. (1979a). J. Parasitol. 65, 109-1 16. De Souza, W., Souto-Padron, T., Docampo, R . , Chaves, L., and Leon, W. ( 1 979b). Acta Trop. 36, 253-255. De Souza, W., Meza, I., Martinez-Palomo, A,, Sabanero, M., Souto-Padrh, T., and Meirelles, M. N. L. ( 1983). J . Purasitol. 69, 138- 142. Dias, E. (1934). Mem. Inst. Oswaldo Cruz 28, 1-1 10. Docampo, R., Boiso, J . F., Boveris, A., and Stoppani, A. 0. M. (1976). Experientia 32,972-975. Doyle, J . J., Behin, R., M a d , J . , and Rowe, D. S. (1974). J. Exp. Med. 39, 1061-1069. Dubuy, H.G . , Mattern, C. F., and Riley, F. L. (1965). Science 147, 754-756. Dusanic, D. G . (1968). J . Protozool. 15, 328-332. Dute, R., and Kung, C. (1978). J. Cell B i d . 78, 451-464. Dvorak, J. A. (1976). In “American Trypanosomiasis Research,” pp. 109-120. PAHO Scientific Publication No. 318. Dvorak, I . A , , and Hyde, T. P. (1973). Exp. Parasitol. 34, 268-283. Dvorak, J. A,, and Poore, C. M. (1974). Exp. Parasitol. 36, 150-157. Dwyer, D. M. (1976). J. Immunol. 117, 2081-2091. Dwyer, D. M. (1980). J . Protozool. 27, 176-182. Dwyer, D. M., and D’Alesandro, P. A. (1980). J. Parasitol. 66, 377-389. Easterbrook, K. B. (1971). Can. J. Microbiol. 17, 277-289. Eisen, H., and Talan, M. (1977). Nature (London) 270, 514-515. El-On, J . , and Greenblatt, C. L. (1977). Exp. Parasitol. 41, 31-42. Esponda, P.,Souto-Padron, T., and De Souza, W. (1983). J. Proroiool. 30, 105-110. Fairlamb, A. H., Weislogel, P. 0.. Hoeijmakers, J . H. J., and Borst, P. (1978). J . Cell Biol. 76, 293-309. Fok, A. K., and Allen, R. D. (1975). J. Histochem. Cytochem. 23, 599-606. Franco Da Silveira, J . (1979). Ph.D. thesis, University of Slo Paulo. Franco Da Silveira, J . , and Colli, W. (1981). Biochim. Biophys. Acta 644,341-350. Franco Da Silveira, J., Abrahamson, P. A., and Colli, W. (1979). Biochim.Biophys. Acta 550, 222-232. Frasch, A. C. C., Segura, E. L., Cazzulo, J. J., and Stoppani, A. 0. M. (1978). Comp. Biochem. Phvsiol. 60B, 271-275.

CELL BIOLOGY OF TRYPANOSOMA CRUZI

277

Freymuller, E., and Camareo. E. P. (1981). J . Protozool. 28, 175-182. Frugulhetti, 1. C. P., and Rebello, M. A. (1977). Proc. Annu. Meet. Basic Res. Chagas’ Dis., Caxambu, Brazil p. 50. Fuge, H. (1968). Z. Zell. Mikrosk. Anat. 89, 201-211. Fuge, H.(1969). J. Prorozool. 16, 460-466. Fuge, H. (1977). Int. Rev. Cytol. 6, 1-58. Futeasaku, Y.,Mizuhira, V., and Nakamura, H. (1972). Proc. Int. Congr. Hisrochem. 4, 155-156. Gangham, P. L. Z., and Krassner, S. M. (1971). Comp. Biochem. Physiol. 39B, 5-18. Giannini, M. S., and D’Alesandro, P. A. (1978). Science 201, 916-918. Gibbons, I. R. (1963). Arch. Biol. 76, 317-324. Goldberg, S. S., Silva Pereira, A. A,, Chiari, E., Mares-Guia, M., and Gazzinelli, G. (1976). J . Protozool. 23, 179-186. GonGalves, J. M., and Yamaha, T. (1969). Am. J. Trop. Med. Hyg. 72, 39-44. Gorin, P. A. J . , Barreto-Bergter. E. M.. and CNZ, F. S. (1981). Carbohydr. Res. 88, 177-188. Gottlieb, M. (1977). J . Immunol. 119, 465-470. Gottlieb, M., and Dwyer, D. M. (1981a). Science 212, 939-941. Gottlieb, M. and Dwyer, D. M. (1981b). In “The Biochemistry of Parasites’’ (0.M. Shutzky, ed.), pp. 29-46. Pergamon, Oxford. Graham, R. C., and Karnovsky, M. J. (1966). J . Hisrochem. Cytochem. 14, 291-302. Oroeschel-Stewart, U. (1980). Int. Rev. Cyrol. 65, 193-254. Grynberg, N. F. (1976). MS. thesis, University Federal of Rio de Janeiro. Gutteridge, W. E., and Rogerson, G. W. (1979). In “Biology of the Kinetoplastid” (W. H. R. Lumsden and D. A. Evans. eds.), pp. 619-652. Academic Press, New York. Gutteridge, W. E., Cover, B., and Gaborak, M. (1978). Parasitology 76, 159-176. Hadjuk, S. L. (1976). Science 191, 858-859. Hadjuk, S . L. (1978). Prog. Mol. Subcell. Biol. 6, 158-200. Haimo, L. T., Telzer, B. R., and Rosenbaum, J . L. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 5759-5763. Hammond, D. J., and Gutteridge, W. E. (1982). Biochim. Biophys. Acra 718, 1-10, Hammond, D. J., and Gutteridge, W. E. (1983). Mol. Biochem. Parasitol. 7, 319-330. Hanas, J., Linden, G., and Stuart, K. (1975). J . Cell. B i d . 65, 103-1 11. Hanover, J. A., and Lennarz, W. J. (1981). Arch. Biochem. Biophys. 211, 1-19. Hecker, H. (1980). Pathol. Res. Prac?. 166, 203-217. Hecker, H., Burri, P. H., and Boeheringer, S. (1973). Experientia 29, 901-903. Heroin-DeLaeney, J. (1965). Ann. Histochim. 10, 23-34. Heywood, P., and Weinman, D. (1978). J. Prorozool. 25, 287-292. Heywood, P., Weinman, D., and Lipman, M. (1974). J . Prorozool. 21, 232-238. Hoeijmakers, J. H. J., Snidjers, A., Janssen, J. W. G., and Borst, P. (1981). Plasmid 5 , 81-93. Hoare, C. A., and Wallace, F. G. (1966). Narure (London) 212, 1385-1386. Hogan, J. C., and Patton, C. L. C. (1976). J. Prorozool. 23, 205-215. Holder. A. (1983). Biochem. J. 209, 261-262. Hollingshead, S., Pethica, B. A., and Ryley, J. F. (1963). Biochem. J . 89, 123-127. Holwill, M. E. J. (1965). J . Exp. Biol. 42, 125-137. Holwill, M. E. J., and McGregor, J. L. (1974). J . Exp. Biol. 60, 437-444. Holwill M. E. J. and McGregor, J . L. (1976). J. Exp. B i d . 65, 229-242. Honigberg, B. M., Mattern, C. F. T., and Daniel, W. A. (1971). J . Prorozool. 18, 183-198. Hughes, D. E., Schneider, C. A., and Simpson, L. (1982). Mol. Biochem. Parasitol. Suppl. Inr. Congr. Parasirol., 5th p. 109. (Abstr.) Hunt, R. C., and Ellar, D. J. (1974). Biochim. Biophys. Acra 339, 173-189.

278

WANDERLEY DE SOUZA

Hyams, J. S. (1982). J . Cell Sci. 55, 199-210. Hyde, T. P., and Dvorak, J. A. (1973). Exp. Parasitol. 34, 284-294. Janakidevi, K., Dewey, V. C . , and Kidder, G. W. (1965). J . Biol. Chem. 240, 1754-1757. Janovy, J., Jr., Lee, K. W . , and Brumbangh, J. A. (1974). J. Prorozool. 21, 53-59. Johnson, B. J . B., and Cross, G. A. M. (1977). J. Protozool. 24, 587-591. Johnson, B. J. B., Hill, G. C., Fox, T. D., Stuart, K. (1982). Mol. Biochem. Parasirol. 5,381-390. Joyon, L . , and Lorn, J. (1969). J. Protozool. 16, 705-719. Kallinikova, V. D. (1981). Int. Rev. Cytol. 69, 105-156. Karp, G. (1979). ‘‘Cell Biology.” McGraw-Hill, New York. Katzin, A. M.,and Colli, W. (1983). Biochem. Biophys. Acra 727, 403-411. Katzin, A. M., Armda, M. V., and Colli, W. (1982). Proc. Annu. Meet. Basic Res. Chagus’ Dis.. Caxambu, Brazil p. 94. Kierszenbaum, F., Ivany, J., and Budzko, D. B. (1976). Immunology 30, 1-6. Killick-Kendrick, R., Molyneux, D. H., and Ashford, R. W. (1974). Proc. R. SOC. Ser. B 187, 409-5 19. Kipnis, T. L., Calich, V. L. G., and Dias Da Silva, W. (1979). Parasitology 78, 89-98. Kipnis, T. L., David, J . R., Alper, C. A , , Sher, A., and Dias Da Silva, W. (1981). Proc. Natl. Acad. Sci. U.S.A. 78, 602-605. Kleisen, C. M., and Borst, P. (1975). Biochim. Biophys. Acra 407, 473-478. Kleisen, C. M., Werslogel, P. O., Funck, K., and Borst, P. (1976). Eur. J . Biochem. 64, 153-160. Klcetzel, J., and Deane, M. P. (1977). Rev. Inst. Med. Trop. Srio Paulo 19, 397-402. Kofoid, C. A., McNail, E., and Wood, F. D. (1937). Univ. Calg. Publ. Zool. 4, 23-24. Komberg, R. D. (1977). Annu. Rev. Biochem. 46, 931-954. Krettli, A. U. (1978). Ph.D. thesis, University of Minas Gerais, Brasil. Krettli, A. V., and Brener, Z. (1976). J . Immunol. 116, 755-760. Krettli, A. U.,and Brener, Z. (1982). J. Immunol. 128, 2009-2012. Krettli, A. V., and Eisen, H. (1980). Proc. Annu. Meet. BusicRes. Chugas’Dis., Caxumbu, Brazil, I, 58. Krettli, A. U., Weiz-Carpington, P., and Nussenzweig, R. (1979). Clin. Exp. Immunol. 37, 416-421. Kusel, J . P., Moore, K. E., and Weber, M. M. (1967). J. Prorozool. 14, 283-296. Lanar, D. E. (1979). J. Protozool. 26, 457-462. Lanar, D. E., Levy, L. S., and Manning, J . E. (1981). Mol. Biochem. Parusitol. 3, 327-341. Langreth, S . M., and Balber (1975). J. Prorozool. 22, 40-53. Langreth, S. M., and Godfrey, D. G. V. (1970). J . Prorozool. 22, 40-53. Lanham, S. M., and Godfrey, D. G. V. (1970). Exp. Parusirol. 28, 521-534. Laub-Kuperstejn, R., and Thirion, J. (1974). Biochim. Biophys. Acru 340, 314-322. Leaver, J. L., and Ramponi, G. (1971). Biochem. J . 125, 44. Lederkremer, R. M., Alves, M. J. M., Fonseca, G. C., and Colli, W. (1976). Biochim. Biophys. Acru 44, 85-96. Lederkremer, R. M., Tanaka, C. T., Alves, M .J. M., and Colli, W. (1977). Eur. J . Biochem. 74, 263-267. Lederkremer, R. M., Casal, 0. L., Tamaka, C. T., and Colli, W. (1978). Biochim. Biophys. Acta 85, 1268-1274. Leon, W., Villalta, F., Queiroz, T., and Szarfman, A. (1979). Infect. Immun. 26, 1218-1220. Leon, W., Frasch, A. C. C., Hoeijmakers, J . H. J., Fase-Fowler, F., Borst, P., Brunnel, F., and Davison, J. (1980). Biochim. Biophys. Actu 607, 221-231. Linder, J . C., and Staehelin, L. A. (1977). J. Ulrrusrrucr. Res. 60, 246-263. Loures, M. A., Pimenta, P. F. P., and De Souza, W. (1980). J. Parasirol. 66, 1058-1059. McGhee, R . B., and Cosgrove, W. B. (1980). Microbiol. Rev. 44, 140-173.

CELL BIOLOGY OF TRYPANOSOMA CRUZI

279

McLaughlin, J. (1982). J. Immunol. 128,2656-2663. McNutt, N. S. (1977). In “Dynamics Aspects of Cell Surface Organization” (G. Poste and G. L. Nicolson, eds.), pp. 75- 126. Elsevier, Amsterdam. Mancini, P. E., Strickler, J. E., and Patton, C. L. (1982). Biochim. Biophys. Acta 688, 399-410. Marcipar, A. J . , Lentwojt, E., Segard, E., Afchain, D., Fruit, J., and Capron, A. (1982). Parasite Immunol. 4, 109- 1 15. Maria, T.A., Tafuri, W., and Brener, Z. (1972). Ann. Trop. Med. Parasitol. 66, 423-431. Marmur, J., Cahoon, M. E., Shimura, Y., and Vogel, H. J. (1963). Nature (London) 197, 1228- 1229. Martinez-Palomo, A., De Souza, W., and Gonzales-Robles, A. (1976). J. Cell Biol. 69, 507-513. Meirelles, M. N. L., and De Souza, W. (1980). Proc. Annu. Meet. Basic Res. Chagas’ Dis., Carambu, Brasil. p. BI. Meirelles, M. N. L., and De Souza, W. (1982). Exp. Parasitol. 53, 341-354. Meirelles, M. N. L., De Araujo Jorge, T. C., and De Souza, W. (1980). Parasitology81, 373-381. Meirelles, M. N. L., Chiari, E., and De Souza, W . (1982). Acta Trop. 39, 195-203. Mendonqa-Previato, L.,Gorin, P. A. J., and Previato, J . 0. (1982). Proc. Annu. Meet. Basic Res. Chagas’ Dis., Carambu, Brasil p. 97. Messier, P. E. (1971). J. Prorozool. 18,223-231. Meyer, H.(1968). J. Protozool. 15,614-621. Meyer, H.,and De Oliveira, M. (1948). Parasitology 39, 91-94. Meyer, H.,and De Souza, W. (1973). J. Protozool. 20, 590-593. Meyer, H.,and De Souza, W. (1976). J. Protozool. 23, 385-390. Meyer, H.,Oliveira Musacchio, M., and Andrade Mendonqa, I. (1958). Parasitology 48, 1-8. Milder, R., and Deane, M. P. (1967). J. Protozool. 14,65-72. Milder, R.,and Deane, M. P. (1969). J. Protozool. 16, 730-737. Mizuhira, V., and Futeasaku, Y . (1972). Acta Histochem. Cytochem. 5 , 233-236. Molyneux, D. H. (1969). Parasitology 59, 55-66. Molyneux, D. H., and Robertson, E. (1974). Ann. Trop. Med. Parasitol. 8, 369-377. Morales, N. M., and Roberts, J. F. (1978). Comp. Biochem. Physiol. 59B, 1-4. Morel, C.,and Simpson, L. (1980). Am. J. Trop. Med. Hyg. 29, 1070-1074. Morel, C., Chiari, E., Camargo, E. P., Mattei, D. M., Romanha, A. J., and Simpson, L. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 6810-6814. Muhlpfordt, H. (1963). 2. Tropenmed. Parasitol. 14, 357-398. Muller, M. (1975). Annu. Rev. Microbiol. 29, 467-483. Muniz, J., and Borrielo, A. (1945). Rev. Bras. Biol. 5, 563-576. Murphy, D. B., and Borisy, G. C. (1975). Proc. Natl. Acad. Sci. U.S.A. 72, 2696-2700. Murray, P. K., Boltz, R. C., and Schmutz, D. M. (1982). J. Protozool. 29, 109-113. Muse, K. E., and Roberts, J. F. (1973). Protoplusma 78, 343-348. Nicolson, G. L. (1976). Biochim. Biophys. Acta 457, 57-108. Nogueira, N., and Cohn, Z. (1976). J. Exp. Med. 33, 546-554. Nogueira, N., Bianco, C., and Cohn, Z . (1975). J. Exp. Med. 146, 157-171. Nogueira, N., Chaplan, S., and Cohn, Z. (1980). J. Exp. Med. 132, 447-451. Nogueira, N., Chaplan, S., Tydings, J. D., Unkeless, J., and Cohn, Z. (1981). J. Exp. Med. 153, 629-639. Nogueira, N., Unkeless, J., and Cohn, Z. (1982). Proc. Natl. Acad. Sci. U.S.A. 79, 1259-1263. Novikoff, A. B., and Goldfischer, S. (1969). J. Histochem. Cytochem. 17, 675-680. O’Daly, J. A., and BretaAa, A. (1976). Int. J. Parasitol. 6 , 271-278. Oliveira, M. M., Timm, S. L., and Costa, S. C. G. (1977). Comp. Biochem. Physiol. 58B, 195- 199. Olmsted, 1. B., and Borisy, G. G. (1975). Biochemistry 14, 2996-3005.

280

WANDERLEY DE SOUZA

Opperdoes, F. R., and Borst, P. (1977). FEES Lett. 80, 360-364. Opperdoes, F. R., and Cottem, D. (1982). FEES Lett. 143, 60-64. Opperdoes, F. R., Borst, P., Bakker, S., and Leene, W. (1977). Eur. J. Biochem. 76, 29-39. Ozeki, Y., Sooksri, V., and Ono, T. (1971). Eiken J. 14, 97-118. Pan, S. C. (1978a). Exp. Parasitol. 45, 215-224. Pan, S. C. (1978b). Exp. Parasitol. 45, 274-286. Parodi, A. J., and Cazzulo, J. J. (1982). J . Eiol. Chem. 257, 764-7645. Parodi, A. J., and Quesada-Allue, L. A. (1982). J. Biol. Chem. 257, 7637-7640. Parodi, A. J., Quesada-Allue, L. A,, and Cazzulo, I. J. (1981). Proc. Natl. Acad. Sci. U.S.A. 78, 620 1-6205. Paulin, J. J. (1969). Trans. Am. Microsc. SOC. 88, 400-410. Paulin, J. P. (1975). J . Cell Eiof. 66, 404-413. Paulin, J. P. (1977). Exp. Parasifol. 41, 283-289. Penningroth, S. M., Cleveland, D. M., and Kirschner, M. W. (1976). In “Cell Motility” ( R . Goldman, T. Pollard, and J. Rosenbaum, eds), pp. 1233-1255. Cold. Spring Harbor Lab., Cold Spring Harbor, New York. Pereira, M. E. A,, Loures, M. A. Villalta, F., and Andrade, A. F. P. (1980). J. Exp. Med. 152, 1375- 1392. Pereira, N. (1978). M. S. Thesis, University Federal of Rio de Janeiro. Pereira, N. M., De Souza, W., Machado, R. D., and De Castro, F. T. (1977). J . Protozool. 24, 511-514. Pereira, N. M., Timm, S. L., Costa, S. C. G.,Rebello, M. A , , and De Souza, W. (1978). Exp. Parasitol. 46, 225-234. Piccini, E. V., Albergoni, V., and Cappellotti, 0. (1975). J. Protozool. 22, 331-332. Pinto Da Silva, P., and Branton, D. (1970). J . Cell Eiol. 45, 598-605. Piperno, G., and Luck, D. J. (1977). J. Biol. Chem. 252, 383-391. Piras, M. M., De Rodrigues, 0. O., and Piras, R. (1981). Exp. Parasirof. 51, 59-73. Piras, M. M., Piras, R., and Henriquez, D. (1982). Mol. Eiochem. Parasitol. 6, 67-81. Preston, T. M. (1969). J. Protozool. 16, 320-333. Previato, J. O., Andrade, A. F. B., Gorin, P. A. J., and Mendonsa-Previato, L. (1982). Proc. Annu. Meet. Basic Res. Chagas’ Dis., Caxambu. Brazil p. 77. Rabinovitch, L., and Kempner, W. (1899). Z. Hyg. Infektionskr. 30, 251. Rembold, H., and Langenbach, T. (1978). J. Protozool. 25, 404-408. Renger, H. C., and Wolstenholme, D. R. (1970). J. Cell Biol. 47, 689-702. Renger, H. C., and Wolstenholme, D. R. (1972). J. Cell Eiol. 54, 346-364. Riou, G . , and Delain, E. (1969). Proc. Narl. Acad. Sci. U.S.A. 62, 210-217. Riou, G.,and Pantrizel, R. (1969). J. Prorozool. 16, 509-513. Riou, G.,and Paoletti, C. (1967). J. Mol. Eiol. 28, 377-382. Riou, G., and Yot, P. (1977). Biochemistry 16, 2390-2396. Riou, G.,Yot, P., and Truhant, M. R. (1975). C. R. Acad. Sci. (Paris) 280, 2701-2704. Ris, H. (1962). Proc. Inr. Congr. Electron. Microsc., 5th 2, Abstr. 21. Romaiia, C., and Meyer, H. (1942). Mem. Inst. Oswaldo Cruz 37, 19-27. Rovis, L., and Baekkeskov, S. (1980). Parasitology 80, 507-524. Rubio, J., Rosado, Y., and CastaAeda, M. (1980). Can. J. Biochem. 58, 1247-1251. Rubio, M. (1956). Bol. Chil. Parasitol. 11, 62-69. Rudzinska, M. A., Trager, W., and Bray, R. S. (1965). J. Protozool. 12, 563-576. Sanderson, C. J., Thomas, J. A., and Thomey, C. E. (1980). Parasitology 80, 153-162. Satir, P. (1968). J. Cell Eiol. 39, 77-94. Saucier, J. M., Benard, J., Da Silva, J., and Riou, G . (1981). Eiochem. Eiophys. Res. Commun. 101, 988-994.

CELL BIOLOGY OF TRYPANOSOMA CRUZI

28 1

Scharfstein, J., Rodrigues, M. M., Alves, C. A . , De Souza, W., and Mendonga-Prcviato. L. (1982). Proc. Annu. Meet. Basic Res. Chagas’ Dis.. Caxambu, Brazil p. 161. Schildkraut, C. L., Mandel, M., Smith, J. E., Levisohn, S . , and Marmur, J. (1962). Nature (London) 196, 795-796. Schliwa, M., and Van Blerkam, J. (1981). J. Cell Biol. 90, 222-235. Schmatz, D. M., and Murray, P. K. (1981). J. Parasitol. 67, 517-521. Schmunis, G. A., Szarfman, A., Langenbach, T., and De Souza, W. (1978). Infect. Immun. 20, 567-569. Schmunis, G . A., Szarfman, A., Dc Souza, W., and Langenbach, T. (1980). Exp. Parasitol. 50, 90-102. Schotelius, J . (1982). Z. Parasitenk. 68, 147-154. Schubert, 0 . (1966). Z. Parasitenkd. 27, 271-286. Scott, M. T., and Snary, D. (1982). Proc. Annu. Meet. Basic Res. Chaps’ Dis.. Caxumbu, Brazil p. 161. Seed, J. R., Byram, J . , and Cam, A. A. (1967). J . Protozool. 14, 117-125. Seed, T. M., Gnan, J., Kreier, J., and Pfister, R. (1972). Exp. Parasitol. 31, 399-406. Seed, T. M., Kreier, J . , Al-Abbassy, S. N., and Pfistcr, R. (1973). Z. Tropenmed. Parasitol. 24, 146- 160. SegUrd, E. L., Vasqucz, C., Bronzina, A , , Campos, S. M., Cerizola, J. A,, and Gonzales-Cappa. S. M . (1977). J . Protozool. 24, 540-543. Sher, A., Snary, D., Kirchoff, L. V., and Crane, M. (1982). Proc. Annu. Meet. Basic Res. Chagas’ Dis., Caxamhu, Brazil, p. 163. Shinitzky, M., and Henkart, P. (1980). Int. Rev. Cytol. 60, 121-147. Shipley, P. C. (1916). Anat. Rec. 10, 439-445. Silvcrstein, S. C., Michl, J . , and Loike, J. D. (1981). I n t . Cell B i d . pp. 604-612. Simpson, A. M., and Simpson, L. (1976). J. Protozool. 23, 583-587. Simpson, A. M., and Simpson, L. (1980). Mol. Biochem. Parasitol. 2, 93-108. Simpson, L. (1972). Int. Rev. Cytol. 32, 139-207. Simpson, L., and Da Silva, A. (1971). J. Mol. Biol. 56, 443-473. Simpson, L., and Larsky, L. (1975). J. Cell B i d . 67, 402A. Simpson, L., Simpson, A,, and Larsky, L. (1976). In “Biochemistry of Parasites and Host-Parasite Relationships” (H. Van Den Bossche, cd.), pp. 225-228. North-Holland Publ., Amsterdam. Singer and Nicolson (1972). Science 175, 720-731. Siqueira, A. F., Ferrioli, F., and Cavalheiro, J. R. (1966). Rev. Inst. Med. Trop. Sao Paulo 8, 148- 154. Smith, D. S., Njogu, A. R., Cayer, M., and Jarlfors, U. (1974). Tissue Cell 8, 563-569. Snary, D . , and Hudson, L. (1979). FEBS Lett. 100, 166-170. Snary, D., Ferguson, Maj., Scott, M. T., and Allen, A. K . (1981). Mol. Biochem. Parasitol. 8, 563-569. Soares, T. C. B., and De Souza, W. (1977). Z. Parasitenkd. 53, 149-154. Solari, A. J. (1980). Chromosoma 78, 239-255. Solari, A. J. (1983). Z. Parasitenkd. 69, 3-15. Sookrsi, V. Kudo, N., Ozeki, I . , and Inoki, S . (1972). Bikens J . 15, 23-29. Souto-Padron, T . , and De Souza, W. (1978). J. Histochem. C.ytochem. 26, 349-358. Souto-Padrh, T., and Dc Souza, W. (1979). J. Protozool. 26, 551-557. Souto-Padrbn, T . , and De Souza, W. (1982). Z. Parasitenkd. 222, 153-158. Souto-Padron, T . , Lima, V. M. Q., Roitman, I., and De Souza, W. (1980a). Z. Parasitenkd. 62, 145-157. Souto-Padrbn, T., De Lima, V. M. Q. G., Roitman, I., and De Souza, W. (1980b). Z. Parasitenkd. 62, 127-143.

282

WANDERLEY DE SOUZA

Souza, W. (1980b). 2 Parasirenhd. 63, 127-143. Souza, M. A. (1982). M.S. thesis, FundaGHo Oswaldo Cruz, Rio de Janeiro, Brasil. Steiger, R. F. (1973). Acfa Trop. 30, 64-168. Steinert, M. (1960). J. Biophys. Biochem. Cyfol. 8, 542-546. Steinert, M. (1965). Exp. Cell Res. 39, 69-72. Steinert, M., and Novikoff, A. B. (1960). J. Biophys. Biochem. Cyrol. 8, 563-569. Steinert, M., and Van Assel, S. (1967). J. Cell Biol. 34,489-503. Steinert, M., and Van Assel, S. (1976). Exp. Cell Res. 96, 406-409. Steinert, M., Van Assel, S., and Steinert, G. (1976). In “Biochemistry of Parasites and HostParasite Relationships” (H. Van Den Bossche, ed.), pp. 21 1-218. North-Holland Publ., Amsterdam. Stephens, R. E. (1970). Biol. Bull. 139, 438a. Stewart, J. M., and Beck, S. J. (1967). J . Prorozool. 14, 225. Stoppani, A. 0. M., Docampo, R., De Boiso, J. F., and Frasch, A. C. C. (1980). Mol. Biochem. Parasirol. 2, 3-21. Szarfman, A,, Cossio, P. M., Laguens, R. P., Segal, A,, De La Vega, M. T., Arana, R. M., and Schmunis, G. A. (1975). Biomedicine 22, 489-495. Szarfman, A,, Queiroz, T., and De Souza, W. (1980). J. Parasifol. 66, 1055-1057. Tait, A. (1980). Nature (London) 287, 536-538. Tanowitz, H., Wittner, M., Sveda, M., and Soeiro, R. (1975). J. Parasifol. 61, 1065-1069. Taylor, M. B., Berhausen, H., Heyworth, P., Messenger, N., Rees, L. J . , and Gutteridge, W. E. (1980). Inr. J . Biochem. 11, 117-120. Thivolet, J., Monier, J. C., Lalain, F., and Richard, M. M. (1965). Ann. Insf. Pasreur 109, 817. Thomashow, L. S., Milhausen, M., Rutter, W. J., and Agabian, N. (1983). Cell 32, 35-43. Timm, S. L., Leon, L. L., Pereira, N. M., De Souza, W., Queirbz-cruz, M., Mbrascher, H. M., and Oliveira Lima, A. (1980). Mem. Inst. Oswaldo Cruz. 75, 145-155. Vickerman, K. (1969). J . Cell Sci. 5 , 163-193. Vickerman, K. (1973). J. Prorozool. 20, 394-404. Vickerman, K. (1974). Ciba Found. Symp. Ser. 20, 171-190. Vickerman, K., and Luckins, A. G. (1969). Nature (London) 224, 1125-1126. Vickerman, K., and Preston, T. M. (1970). J . Cell Sci. 6, 365-383. Vickerman, K., and Preston, T. M. (1976). In “Biology of Kinetoplastida” (W. H. R. Lumsden and D. E. Evans, eds.), pp. 35-130. Academic Press, New York. Vickerman, K., and Tetley, L. (1977). Ann. SOC. Belge Med. Trop. 57, 441-455. Villalta, F. V., and Leon, W. (1979). J. Parasirol. 65, 188-189. Villalta, F. V., Katzin, A. M., Leon, W., and Gonzales-Cappa, S. (1980). J . Parasirol. 66, 1053-1055. Visser, N., Opperdoes, F. R.,and Borst, P. (1981). Eur. J. Biochem. 118, 521-526. Voorheis, P. H.,Gale, J. S., Owen, M. J., and Edwards, W. (1979). Biochem. J. 180, 11-24. Voorheis, H. P., Bowles, D. J., and Smith, G. A. (1982). J . Biol. Chem. 257, 2300-2304. Wallace, F. G., Wagner, M., and Rogers, W. E. (1973). J. Prorozool. 20, 218-222. Warner, F. D., and Satir, P. (1974). J. Cell Biol. 63, 35-63. Warton, A,, and Kallinikova, V. D. (1974). Acfu Prorozool. 12, 355-362. Weatherbee, J. A. (1981). Inf. Rev. Cyrol. 12, 113. Weingarten, M. B., Lockweed, A. H.,Hwo, S . Y., and Kirschnar, M. W. (1975). Proc. Narl. Acad. Sci. U.S.A. 72, 1858-1862. Weislogel, P. O., Hoeijmakers, J. H.J., Fairlamb, A. H.,Kleisen, L. M., and Borst, P. (1977). Biochim. Biophys. Acra 478, 167-179. Wilson, L., and Bryan, J. (1974). Adv. Cell. Mol. Biol. 3, 21-72. Wilson, L. Creswell, M., and Chin, D. (1975). Biochemistry 14, 5586-5592.

CELL BIOLOGY OF TRYPANOSOMA CRUZI

283

Witman, G. B., Carlson, K., Berliner, J., and Rosenbaum, J. L. (1972). J . CellBiol. 54,507-539. Zartesava, G. N . , Mett, N. L., and Kdesrukoo, A. A. (1976). Biokhimiya 41, 1145-1149. Zartesava, G. N . , Kolesnikov, A. A., and Shirshov, A. T. (1977). Mol. Cell. Biochem. 14,47-54. Zeledon, R., Alvarenga, N. I., and Schosinsky, K. (1977). In “Chagas Disease,” pp. 59-70. Pan American Health Organization. h b l . No. 347. Ziemann, H. ( 1 898). Zentralbl. Bakreriol. Parasitenkd. Infektionkr. 24, 945. Zingales, B., Carniol, C., Abrahamson, P. A., and Colli, W. (1979).Biochim. Biophys. A m 550, 233-244. Zingales, B., Andrews, N. W., Kuwajima, V., and Colli, W. (1981).Proc. Annu. Meet. BasicRes. Chagas’ Dis., Cuxambu, Brasil p. 69. Zingales, B., Andrews, N. W., Kuwajima, V. Y., and Colli, W. (1982a).Mol. Biochem. Parasitol. 6 , 111-124. Zingales, B., Martin, N. F., De Lederkremer, R., and Colli, W. (1982b). FEBS Lett. 142,238-242.