The central problem for Theileria parva during merogony is how to form numerous individual, uninucleate merozoites from a syncytial schizont so that each merozoite contains a single nucleus and a prescribed assortment of organelles. The way T. parva packages all the requisite organelles into free merozoites is by binding these organelles to the nuclear envelope, which in turn becomes associated, both directly and through the rhoptry complex, with the schizont plasma membrane. Formation of the merozoites occurs In a synchronous manner by a budding process. The merozoites develop with the rhoptry complex at the apical end by the progressive, outward évagination of the schizont plasma membrane. This évagination of the plasma membrane is associated with, and presumably induced by, the development of an orderly array of tubules that originate from the apical end and progressively form a longitudinal basket enclosing first the rhoptry complex, then the mitochondria and ribosomes, and finally the nucleus.

The process of merogony is compared to sporogony within the tick salivary gland and with the differentiation of the intra-erythrocytic piroplasm stage. Because all three processes occur by a morphologically similar mechanism, the possibility that the parasite uses a single cassette of genes to perform each of these similar processes is discussed.

In general eukaryotic cells cannot form cell organelles de novo and require the presence of pre-existing copies within the cells either to act as templates or from which new organelles can develop by growth and fission. In the case of specialized organelles (e.g. secretory granules or specialized invasive organelles such as rhoptries, microspheres etc.) that appear to be made de novo, the cell must, however, contain the requisite organelles from which to synthesize these specialized cellular inclusions. Thus, the mechanism(s) whereby an organism distributes its cell organelles into daughter cells during cell division is a fundamental problem in cell biology. In higher eukaryotic cells there appear to be enough copies of most cell organelles so that random distribution alone should ensure that at least one copy of each organelle would be present in each daughter cell after cytokinesis. In the case of organelles occurring in low copy numbers such as the Golgi complex and the endoplasmic reticulum, these organelles fragment, prior to cell division, into numerous vesicular structures, which then become randomly distributed between the daughter cells. Even though there may be no special mechanism(s) required to ensure that each daughter cell acquires at least one copy of each organelle, there are examples of unequal cell division (e.g. during early embryonic differentiation) or cell divisions in which certain cellular components become selectively segregated into only one daughter cell (e.g. the preferential localization of P granules into only one daughter cell during early embryogenesis in Caenorhab-ditis elegans, see Strome and Wood, 1983). It is thought that this asymmetrical segregation of cell organelles involves the cytoskeleton.

There is a third category, which is the antithesis of the unequal distribution of cellular components and is of particular relevance to the present study. This is the problem of how organisms are able to produce cells which contain either a fixed number of cell organelles or a selected range of organelles (e.g. sperm of many organisms). A particularly good example is the sperm of Caenorhabditis elegans where ribosomes are excluded from the mature sperm by a particular gene product (see Ward, 1986). Of particular interest to us is the problem of how syncytia become cellularized and produce large numbers of individual cells containing a precise and identical set of organelles.

Theileria parva is a tick-borne protozoan parasite that causes East Coast fever, an acute and often fatal disease in cattle, which is endemic in large areas of East and Central Africa (see Norval et al. 1991). The life cycle of T. parva is shown diagrammatically in Fig. 1. The parasite is transmitted to cattle when the tick, Rhipi-cephalus appendiculatus, feeds on a susceptible host. Within the life cycle of T. parva there are three stages at which the parasite undergoes a period of extensive growth involving nuclear division without cytokinesis, prior to sporogony, merogony and to a lesser extent, at the piroplasm stage (Fig. 1). This results in all three stages in the formation of a multinucleate syncytium (Fig. 1). From these syncytia the parasite, in order to continue its life cycle, needs to form uninucleate cells (the sporozoites and merozoites) each of which contain, as a minimum, a nucleus and nuclear envelope, at least one mitochondrion, ribosomes and the appropriate secretory organelles required by the parasite to invade the next host cell.

Fig. 1.

A highly simplified drawing of the life cycle of Theileria parva. There are three stages (at sporogony, merogony and at the piroplasm stage) in which the parasite undergoes karyokinesis without cytokinesis resulting in the formation of a multinucleate syncytium. To continue its life cycle the parasite has to form uninucleate cells from these syncytia. For a detailed account of the life cycle of T. parva, including the sexual stages within the tick gut, see Norval et al. (1991).

Fig. 1.

A highly simplified drawing of the life cycle of Theileria parva. There are three stages (at sporogony, merogony and at the piroplasm stage) in which the parasite undergoes karyokinesis without cytokinesis resulting in the formation of a multinucleate syncytium. To continue its life cycle the parasite has to form uninucleate cells from these syncytia. For a detailed account of the life cycle of T. parva, including the sexual stages within the tick gut, see Norval et al. (1991).

Until now individual investigators have studied separately either sporogony (Fawcett et al. 1982,1985), merogony (Schein et al. 1978) or the piroplasm stage (Conrad et al. 1985, 1986; Fawcett et al. 1987) in Theileria, and no one has taken an overview of all three stages. As T. parva undergoes the same or a very similar process three times during its life cycle it would seem reasonable to predict that the parasite might use the same mechanism to develop uninucleated cells containing a similar set of organelles from a syncytium. The only major difference between these life cycle stages must be the appearance of different molecules on the cell surface, related to the specificity of the subsequent host cell.

In the present study we have described in detail the process of merogony in T. parva to determine how uninucleate cells containing a fixed complement of organelles are formed from a syncytium. We will then relate these observations to what is known about the similar and related processes of sporogony within the tick salivary gland (Fawcett et al. 1982, 1985) and the generally more limited development of the intra-erythrocytic piroplasm stage (Conrad et al. 1985, 1986; Fawcett et al. 1987). Until now there is only one brief and rather incomplete report describing the process of merogony in Theileria (see Schein et al. 1978).

Parasites and cells

Theileria parva (Muguga stock) and the tick, Rhipicephalus appendiculatus maintained in the Tick Unit at ILRAD, were used. Ticks were infected as nymphs by feeding them on cattle infected with ground up tick stabilate (3087), after which the engorged nymphs were maintained at 23–25°C and 80% relative humidity and allowed to moult to the adult stage.

Most of our observations were made on samples of peripheral blood and lymph node cells collected from stabilate-infected cattle. These cattle had very high piroplasm parasitemia (up to 70%) and low (<20%) packed cell volumes. Small samples (e.g. 5–10;μl) of peripheral blood collected into heparin were fixed immediately for electron microsocpy as described below. Lymph node cells were aspirated from the prescapular lymph node into either RPMI 1640 culture medium or directly into one of the fixatives described below.

Observations were also made on in vitro cultures of schizont-infected lymphoblastoid cells. Bovine peripheral blood lymphocytes (PBL) were obtained from the defibri-nated blood of uninfected cattle (Lalor et al. 1986), isolated on Ficoll-Hypaque (Pharmacia, Sweden) gradients, washed three times in Alsever’s solution and resuspended in RPMI 1640 culture medium with 10 mM Hepes buffer supplemented with 10% heat-inactivated foetal bovine serum (Gibco, Paisley, Scotland, UK). Bovine PBLs were infected in vitro with T. parva by incubation with sporozoites derived from salivary glands of infected adult ticks fed on a rabbit for 4 days (Brown, 1987). The infected cells were maintained in RPMI 1640 culture medium with 10 mM Hepes buffer supplemented with 10% heat-inactivated foetal bovine serum, 5×10−5 M 2-mercaptoethanol, 2 mM L-glutamine and 50 μg ml-1 gentamycin.

Merogony was observed in in vitro infected PBLs as early as day 10 post-infection and low levels (10–20%) of merogony were observed throughout the period of observation to day 48.

Electron microscopy

Samples were fixed in suspension by the addition of an equal volume of either (Method 1) a mixture of 2.5% glutaraldehyde, 2% formaldehyde and 0.01% picric acid in 0.1 M phosphate or cacodylate buffer (pH 7.2), or (Method 2) in a freshly prepared solution containing 1% glutaraldehyde (from an 8% stock solution supplied by Electron Microscope Sciences, Fort Washington, PA, USA), 1% OsO4 and 0.05% M phosphate buffer (pH 6.2). In the case of the first fixative, samples were fixed at room temperature (22°C) for 2 h, pelleted, post-fixed in buffered, 1% OsC>4 for 60 min, washed, and “en bloc” stained with aqueous 0.5% uranyl acetate for 6–16 h. In the case of the low ionic strength glutaraldehydeosmium tetroxide combined fixative (Method 2), fixation was carried out on ice (0-4°C) for 40-45 min after which the samples were pelleted, washed several times in distilled water to remove excess phosphate and then “en bloc” stained with uranyl acetate as described above.

The samples were dehydrated with either ethanol or acetone and embedded in an Epon-Araldite mixture.

Ultrathin sections (50–70 nm thick) were collected on uncoated copper grids, double stained with aqueous uranyl acetate and lead citrate, and examined in a Zeiss EM 10A electron microscope.

The purpose of merogony is the production, from a multinucleate syncytial schizont, of a large number of individual merozoites, each of which contains a precise collection of organelles. The following description of merogony in T. parva is based on the examination of a large number of sections of both in vitro cultured material and material obtained from the lymph nodes of cattle having very high piroplasm counts. Material fixed with a hypotonic fixative (Method 2) did not provide optimal preservation of the material, but proved most revealing because much of the cytoplasmic matrix was extracted revealing many details which are not observable following more conventional fixation (Method 1). From a wide range of images of both conventionally and hypotonically fixed material we have reconstructed the stages of merogony that culminate in free merozoites.

The schizont prior to the onset of merogony

The schizont lies free within the host cell cytoplasm and is surrounded by a plasma membrane that has no obvious outer surface coat (Fig. 2A,B). Micropores, thought by earlier investigators to be involved in the uptake of host cell components, are present in the outer plasma membrane at frequent intervals around the schizont surface. The micropores measure about 50–80 nm in diameter and up to 150 nm in depth.

Fig. 2.

(A) Low power electron micrograph of part of a T. parva-infected lymphocyte showing the presence of a multinucleate schizont (Sc). HN, host cell nucleus. (B) High magnification micrograph of the schizont plasma membrane showing the absence of any obvious surface coat. Sc, schizont cytoplasm. HC, host cell cytoplasm. (C) Part of a T. parva schizont showing a schizont nucleus (N) and a mitochondrion (M). The schizont nucleus is surrounded by a typical nuclear envelope with ribosomes attached to the cytoplasmic face of the outer membrane (arrows). Note that the nuclear contents have a uniform, granular appearance, and that the mitochondrion (M) consists of an outer and inner membrane without any cristae.

Fig. 2.

(A) Low power electron micrograph of part of a T. parva-infected lymphocyte showing the presence of a multinucleate schizont (Sc). HN, host cell nucleus. (B) High magnification micrograph of the schizont plasma membrane showing the absence of any obvious surface coat. Sc, schizont cytoplasm. HC, host cell cytoplasm. (C) Part of a T. parva schizont showing a schizont nucleus (N) and a mitochondrion (M). The schizont nucleus is surrounded by a typical nuclear envelope with ribosomes attached to the cytoplasmic face of the outer membrane (arrows). Note that the nuclear contents have a uniform, granular appearance, and that the mitochondrion (M) consists of an outer and inner membrane without any cristae.

The schizont nuclei are scattered randomly throughout the cytoplasm, have a uniform, granular appearance and exhibit no differentiation into areas of dispersed and condensed chromatin. The nuclei are surrounded by a typical nuclear envelope (Fig. 2C) with nuclear pores. The nuclear pores measure between 60 and 70 nm in diameter and are therefore slightly smaller in diameter than the nuclear pores present in the host cell nuclear envelope (85–95 nm in diameter). Occasionally ribosomes were found attached to the cytoplasmic face of the outer nuclear membrane.

The schizont mitochondria are usually found in pairs and occur randomly throughout the cytoplasm. They consist of an outer and inner membrane but lack cristae. In the normal schizont the nuclei and mitochondria are not associated.

The schizont cytoplasm contains conspicuous numbers of membrane-free ribosomes and some small clusters of polysomes. These ribosomes measure between 20 and 25 nm in diameter and are, therefore, somewhat smaller than those of the host cell. In the early stage of merogony, ribosomes of comparable size to host cell ribosomes become attached to the newly formed, rough endoplasmic reticulum. Thus the difference in size of the schizont and host cell ribosomes may be due to the schizont ribosomes existing as ribosomal subunits and not fully assembled ribosomes. If this is the case, then it seems reasonable to conclude that the parasite is engaged in only a minimal amount of protein synthesis, possibly involved in the production of more ribosomal subunits, and is relying on the host cell for the majority of its metabolic and synthetic requirements. Apart from numerous ribosomes and occasionally some small membrane-bounded vesicles, the schizont cytoplasm contains very few other organelles. In particular, no smooth or rough endoplasmic reticula or anything resembling a Golgi apparatus were observed in the schizont cytoplasm.

Early events in merogony

The initial events in merogony are characterized by a series of structural and organizational changes within the schizont that apparently occur either simultaneously or in rapid succession. These changes include the extensive elaboration of the nuclear envelope and the appearance of endoplasmic reticulum, the formation of an obvious external coat on the outer surface of the schizont plasma membrane, changes in the schizont nucleus, the appearance of rhoptries and the association of mitochondria with the schizont nuclei.

Appearance and formation of endoplasmic reticulum

A major change in the schizont cytoplasm is the development and elaboration of both rough (RER) and smooth (SER) endoplasmic reticulum (Fig. 3A). In particular, the appearance of stacks or parallel arrays of RER was commonly observed in schizonts in the initial phase(s) of merogony. Because in the same section we often see extensive elaboration of the outer membrane of the schizont nuclear envelope (Fig. 3B), it seems reasonable to assume that these elaborations give rise to the endoplasmic reticulum. At this stage significantly more ribosomes are attached to the outer nuclear membrane than were seen prior to the onset of merogony.

Fig. 3.

(A) Low magnification micrograph of a section through a T. parva schizont at an early stage in merogony. The early stages of merogony are characterized by: (i) the appearance of extensive arrays of rough and smooth endoplasmic reticulum (RER and SER respectively), (ii) the condensation of the nuclear chromatin and the localization of nuclei (N) at the periphery of the schizont, (iii) the appearance within the schizont cytoplasm of electron dense rhoptry-like inclusions (arrows). (B) Schizont nucleus showing the apparent formation of smooth endoplasmic reticulum (SER) from the outer nuclear membrane. (C) A pair of opposed schizont plasma membranes showing the presence of a 20-25 nm thick surface coat. Sc, schizont cytoplasm.

Fig. 3.

(A) Low magnification micrograph of a section through a T. parva schizont at an early stage in merogony. The early stages of merogony are characterized by: (i) the appearance of extensive arrays of rough and smooth endoplasmic reticulum (RER and SER respectively), (ii) the condensation of the nuclear chromatin and the localization of nuclei (N) at the periphery of the schizont, (iii) the appearance within the schizont cytoplasm of electron dense rhoptry-like inclusions (arrows). (B) Schizont nucleus showing the apparent formation of smooth endoplasmic reticulum (SER) from the outer nuclear membrane. (C) A pair of opposed schizont plasma membranes showing the presence of a 20-25 nm thick surface coat. Sc, schizont cytoplasm.

The appearance of both RER and SER within the schizont at this early stage of merogony is indicative of an increased synthetic activity by the schizont that is in keeping with the formation of a schizont surface coat and accessory secretory organelles.

Surface coat

The outer surface of the schizont becomes covered by a prominent, 20–25 nm thick, coat (Fig. 3C). This coat covers the entire surface of the schizont, persists throughout merogony and is present on the surface of the mature merozoites. Although the exact appearance varies depending upon the fixatives used, this coat material often has a distinct “peg-like” substructure (cf. Bannister et al. 1986).

Nuclear changes

The initial changes in the schizont nuclei are a condensation of the chromatin and the close apposition of the membranes of the nuclear envelope around part of the nucleus. In many schizont nuclei the condensed chromatin also became localized along the inner nuclear membrane in the region where the nuclear membranes were closely apposed (Fig. 4A). This close apposition of the nuclear envelope membranes was most apparent in hypotonically fixed material (Method 2). It is almost certainly not an artifact because even in sections of material showing extensive swelling of the remaining nuclear envelope and looking extremely “poorly fixed”, the two membranes at the point of apposition still remained closely apposed.

Fig. 4.

(A) Schizont nucleus showing the condensation and marginalization of chromatin and the close apposition of the two nuclear envelope membranes around part of the nucleus. In this micrograph the nucleus is now located close to the outer schizont plasma membrane with the specialized portion of the nuclear envelope adjacent to the schizont membrane. Note also the presence of two schizont mitochondria (M) closely associated with the non-specialized part of the nuclear envelope. (B) High magnification micrograph of the specialized portion of the schizont nuclear envelope. In this region the apposed leaflets of the individual nuclear envelope membranes fuse to produce a pentalaminate structure. This specialized portion of the nuclear envelope is separated from the schizont plasma membrane by a 15–20 nm gap. Note also the presence of the surface coat on the outer face of the schizont plasma membrane (arrows).

Fig. 4.

(A) Schizont nucleus showing the condensation and marginalization of chromatin and the close apposition of the two nuclear envelope membranes around part of the nucleus. In this micrograph the nucleus is now located close to the outer schizont plasma membrane with the specialized portion of the nuclear envelope adjacent to the schizont membrane. Note also the presence of two schizont mitochondria (M) closely associated with the non-specialized part of the nuclear envelope. (B) High magnification micrograph of the specialized portion of the schizont nuclear envelope. In this region the apposed leaflets of the individual nuclear envelope membranes fuse to produce a pentalaminate structure. This specialized portion of the nuclear envelope is separated from the schizont plasma membrane by a 15–20 nm gap. Note also the presence of the surface coat on the outer face of the schizont plasma membrane (arrows).

We initially observed nuclei with both condensed nuclear material and a modified nuclear envelope within the central regions of the schizonts. However, it is apparent that concomitant with, or immediately following, these initial changes some nuclei become relocated to contact the periphery of the schizont (Fig. 4A). Here the nuclei are invariably orientated with the modified portion of the nuclear envelope closely associated with the schizont plasma membrane (Fig. 4B). Although the nuclear envelope does not fuse with the outer plasma membrane, the association of the nucleus with the outer membrane is intimate, with the membranes being separated by a uniform gap of 15–20 nm. Around the remaining regions of the schizont nuclear envelope the two component membranes are separated by a variable and much wider spacing (Fig. 4A).

The process whereby the schizont nuclei migrate to the periphery of the schizont and become preferentially orientated with respect to the outer plasma membrane is unclear. We have been unable to find any obvious structural entities that could account for the movement and there appears to be no morphological specialization) associated with the outer plasma membrane in the regions where the nuclei attach.

Development of the rhoptry complex

Concurrent with the above changes, electron dense rhoptry-like structures appear within the schizont cytoplasm. The site of origin of the rhoptries is unclear. In a small number of schizonts, profiles of SER containing material of similar density to the contents of mature rhoptries were observed (Fig. 5A). Arising from the SER were numerous smaller, flask-shaped structures that resembled the rhoptries seen in mature merozoites. Thus the rhoptries appear to arise from the SER.

Fig. 5.

(A) Part of a conventionally fixed (Method 1) T. parva schizont showing profiles of smooth endoplasmic reticulum (SER) containing material of a similar electron density and consistency to the mature merozoite rhoptries. (B) A schizont nucleus located close to the schizont plasma membrane showing the presence of rhoptries (R) associated with both the nuclear envelope and with fibrous material that projects towards the schizont membrane. Note that the specialized portion of the schizont nuclear envelope is adjacent to the schizont plasma membrane (arrowheads). (C) High magnification micrograph of a portion of the schizont nucleus showing the attachment of the rhoptry complex with the spindle pole body (SP) and through a fibrous structure with a dense structural specialization associated with the schizont plasma membrane (arrows).

Fig. 5.

(A) Part of a conventionally fixed (Method 1) T. parva schizont showing profiles of smooth endoplasmic reticulum (SER) containing material of a similar electron density and consistency to the mature merozoite rhoptries. (B) A schizont nucleus located close to the schizont plasma membrane showing the presence of rhoptries (R) associated with both the nuclear envelope and with fibrous material that projects towards the schizont membrane. Note that the specialized portion of the schizont nuclear envelope is adjacent to the schizont plasma membrane (arrowheads). (C) High magnification micrograph of a portion of the schizont nucleus showing the attachment of the rhoptry complex with the spindle pole body (SP) and through a fibrous structure with a dense structural specialization associated with the schizont plasma membrane (arrows).

Whereas newly formed individual rhoptries could be found throughout the schizont cytoplasm, they were most frequently observed associated in small clusters with one margin of the schizont nuclei in close proximity to a dense plaque structure present in the nuclear envelope (Fig. 5B). These dense plaques, which span the nuclear envelope, closely resemble the spindle pole bodies described by Schein et al. (1978) and Fawcett et al. (1985). These bodies are thought to play a role in nuclear division. Prior to the onset of merogony the spindle pole bodies often have microtubules associated with them.

When the nuclei are close to, or have formed their tight association with, the schizont plasma membrane, we can find fine fibrous structures that join the nuclear dense plaques to structural specializations which develop beneath and are usually associated with the plasma membrane (Fig. 5C). The rhoptries are usually associated with these fibrous structures. In a single section there can be two or even three of these rhoptry-associated fibrous connections between the nuclei and the plasma membrane.

The structural specializations, that form the modified apical complex of the merozoite, develop beneath the outer plasma membrane and consist of a plate-like structure composed of a central, inwardly projecting peg, and a number of concentric rings of either denser material or membranous tubules lying parallel to the schizont plasma membrane (Fig. 6). These membranous tubules become flattened to form a slightly concave disc or annulus which, in favourable images, can be seen to be composed of two closely apposed membranes and subsequently forms the very much reduced inner membrane complex in the mature merozoite. The fine fibrous material connecting the rhoptries and the nuclear envelope with the schizont plasma membrane attaches to the inwardly projecting peg in the central region of this subplasmalammel specializations. The whole of this structural specialization is separated from the schizont plasma membrane by a uniform 10–20 nm gap.

Fig. 6.

High magnification micrographs, in three different planes of section, of the structural specialization associated with the schizont plasma membrane to which the rhoptry complex ultimately attaches. (A) is a section cut perpendicular to the attachment complex and shows the central dense peg-like structure (arrow) which projects from the plasma membrane and to which the individual rhoptries attach, and a number of membranous tubules. (B) is an oblique, en face section through the complex showing that the structure is composed of a central dense peg and a series of concentric dense rings. (C) is a slightly oblique section showing a group of six rhoptries and the two closely apposed membranes which form part of the much reduced inner membrane complex (arrows).

Fig. 6.

High magnification micrographs, in three different planes of section, of the structural specialization associated with the schizont plasma membrane to which the rhoptry complex ultimately attaches. (A) is a section cut perpendicular to the attachment complex and shows the central dense peg-like structure (arrow) which projects from the plasma membrane and to which the individual rhoptries attach, and a number of membranous tubules. (B) is an oblique, en face section through the complex showing that the structure is composed of a central dense peg and a series of concentric dense rings. (C) is a slightly oblique section showing a group of six rhoptries and the two closely apposed membranes which form part of the much reduced inner membrane complex (arrows).

Mitochondria

At some point during the early stages in merogony the mitochondria become closely associated with the outer membrane of the nuclear envelope, at a point where the membranes of the nuclear envelope, are not closely apposed (Fig. 4A). The adjacent leaflets of the mitochondrial and nuclear membranes do not fuse, however, and are separated by a small, 10–25 nm wide gap.

Thus, during the early events in merogony prior to the formation of the free mature merozoites, the rhoptries, nucleus, mitochondria and outer plasma membrane have become interconnected in an orderly and reproducible manner.

Later events in merogony: the “budding” of merozoites from the syncytial schizont

Up to this point in merogony the schizont is still a multinucleate syncytium, although the constituent organelles present in a mature merozoite are intimately associated with both themselves and with the schizont plasma membrane. Thus, the next step in the process is the formation of individual unicellular merozoites from the syncytial schizont.

In conventionally fixed material the cytoplasm of the schizonts and developing merozoites is so electron dense (Fig. 7A) that information relating to the intermediate stages in merogony was extremely difficult to discern. Accordingly, we have concentrated on examining material fixed at low ionic strength (Method 2) which “washes out” much of the soluble, electron dense cytoplasmic constituents (Fig. 7B). By examining large numbers of sections we have been able to reconstruct the probable sequence of events occurring during the formation of merozoites from the syncytial schizont (Fig. 8).

Fig. 7.

(A) Micrograph of a conventionally fixed (Method 1) schizont showing merozoites in the process of budding from the main syncytial body. Note that the parasite cytoplasm is extremely electron dense and that it is therefore difficult to discern many of the structural details of the process of budding. (B) Micrograph of a schizont in the process of merozoite budding fixed with low ionic strength glutaraldehyde-osmium tetroxide combined fixative (Method 2).

Fig. 7.

(A) Micrograph of a conventionally fixed (Method 1) schizont showing merozoites in the process of budding from the main syncytial body. Note that the parasite cytoplasm is extremely electron dense and that it is therefore difficult to discern many of the structural details of the process of budding. (B) Micrograph of a schizont in the process of merozoite budding fixed with low ionic strength glutaraldehyde-osmium tetroxide combined fixative (Method 2).

Fig. 8.

(A and B) High magnification electron micrographs of the earliest stage in merozoite budding. (A) Beneath the initial protrusion, which consists of the peg to which the rhoptries attach and the membranous cisternal structure (arrowheads) that forms the very much reduced inner membrane complex of the mature merozoite, are a series of tubular structures (arrows) that are closely associated with the schizont plasma membrane and extend from the apex around the lateral margins of the nucleus (B). (C and D) As merozoite budding proceeds, the rhoptry complex (R) is always located at the apical end of the developing merozoite. Arising from the apical region of the bud the tubular structures (arrows) extend basally forming a sublamellar basket progressively enclosing the rhoptry complex, mitochondria and nucleus. The nucleus (N) is always situated at the base of the bud with the specialized portion of the nuclear envelope (arrowheads) associated with the most basal portion of the budding merozoite. (EF) Final stages of merozoite budding. The tubular structures continue around to the basal end of the nucleus (arrows in E) and terminate in a ring structure at the most basal constriction of the bud (arrowheads in F). (G) Grazing section through the basal region of a budding merozoite showing the ends of the tubules terminating in an electron dense ring structure (arrows).

Fig. 8.

(A and B) High magnification electron micrographs of the earliest stage in merozoite budding. (A) Beneath the initial protrusion, which consists of the peg to which the rhoptries attach and the membranous cisternal structure (arrowheads) that forms the very much reduced inner membrane complex of the mature merozoite, are a series of tubular structures (arrows) that are closely associated with the schizont plasma membrane and extend from the apex around the lateral margins of the nucleus (B). (C and D) As merozoite budding proceeds, the rhoptry complex (R) is always located at the apical end of the developing merozoite. Arising from the apical region of the bud the tubular structures (arrows) extend basally forming a sublamellar basket progressively enclosing the rhoptry complex, mitochondria and nucleus. The nucleus (N) is always situated at the base of the bud with the specialized portion of the nuclear envelope (arrowheads) associated with the most basal portion of the budding merozoite. (EF) Final stages of merozoite budding. The tubular structures continue around to the basal end of the nucleus (arrows in E) and terminate in a ring structure at the most basal constriction of the bud (arrowheads in F). (G) Grazing section through the basal region of a budding merozoite showing the ends of the tubules terminating in an electron dense ring structure (arrows).

The earliest identifiable protrusion from the surface of the schizont occurs at the point where the rhoptry complex attaches to the inwardly projecting peg structure and ring of flattened membranous tubules closely associated with the schizont plasma membrane. Just lateral to the tip of the protrusion, which contains the rhoptry complex and the flattened membranous tubules of the much reduced inner membrane complex, appear some tubular structures (Fig. 8A,B), which form a regular arrangement radiating out from the point at which the rhoptry complex attaches to the plasma membrane. These tubules are sometimes flattened in cross section and measure between 30-65 nm in diameter, and are, therefore, somewhat larger in size than similar sub-plasmalammel microtubules described in other Apicomplexa. As the tip of the developing merozoite protrudes further from the surface of the schizont, these radiating tubules form a basket enclosing the rhoptry complex. The lateral walls of the tubules are closely applied to the membrane forming the protrusion. As the developing merozoite buds further from the surface of the schizont the tubules enclose the nucleus, passing between the nuclear envelope and the outer plasma membrane. High resolution images of cross sections through these tubules show that they make a direct contact with the outer leaflet of the nuclear envelope but are separated from the schizont plasma membrane by a small 10–20 nm gap (Fig. 8B).

Budding of the merozoite proceeds in an orderly manner with the modified portion of the nuclear envelope invariably being orientated towards the basal portion of the developing merozoite (Fig. 8C,D). A possible reason for the close association of the tubule with the modified portion of the nuclear envelope is that the basket of tubules, by following the contour of the nucleus, ensures the formation of a rounded merozoite. The tubules continue around the basal portion of the nucleus. At the last stage the ends of the tubules terminate in some electron dense material which in grazing section appears as a ring of dense material (Fig. 8E-G). It seems reasonable to suspect that this ring structure is involved in the final separation of the merozoite from the syncytial schizont. Thus the basket of tubules that extends from the rhoptry complex to the basal ring encloses a cytoplasmic volume beneath the rhoptry complex that contains both the mitochondria still attached to the nuclear envelope and some free ribosomes (Fig. 9).

Fig. 9.

Three-dimensional drawing of a budding merozoite just prior to separation from the syncytium showing the basket of tubules enclosing the rhoptry complex (R), a mitochondrion (Mito), ribosomes and the nucleus (N).

Fig. 9.

Three-dimensional drawing of a budding merozoite just prior to separation from the syncytium showing the basket of tubules enclosing the rhoptry complex (R), a mitochondrion (Mito), ribosomes and the nucleus (N).

While merozoite budding from the syncytial schizont is a highly synchronous process it is our impression that two or more waves of budding may occur. We have observed frequently free merozoites within a host cell in which newly budding merozoites are forming from the schizont. At the end of merogony the host cell contains large numbers of mature merozoites and usually a residual schizont body which contains a number of nuclei, rhoptries, mitochondria and various amounts of endoplasmic reticulum. Eventually the mature merozoites are liberated by the breakdown of the host cell plasma membrane. Whether this is due to parasite-induced lysis or to the physical presence of large numbers of parasites within the cell is not known.

The free merozoite, which is covered by a distinct 2025 nm thick surface coat, measures between 1 and 2 gm in diameter and contains a single eccentric nucleus, between 3 and 6 rhoptries, one or two mitochondria, microspheres and some cytoplasm containing free ribosomes. Unlike other genera of Apicomplexa, the merozoites of T. parva have no clearly defined apical complex and a conoid or similar apical structure is absent. The rhoptries attach to an inwardly projected peg at the opposite pole to the nucleus, and are associated with a very much reduced inner membrane complex composed of two tightly apposed membranes. No cytostome or micropore was found, even though numerous cytostomes are present in the schizont. We did not observe the tubular basket in the free, mature merozoites but because of the density of the merozoite cytoplasm it is difficult to be certain that this is true in all mature merozoites. As we did not see any microspheres or obvious microsphere precursors in the schizont cytoplasm or in the earlier stages of merozoite budding these organelles would seem to be formed at a very late stage in merozoite formation. Furthermore, the number of microspheres present in the mature merozoites was significantly less than observed in T. parva sporozoites.

Whilst the above description was the most commonly encountered method of merozoite budding, in some schizonts merozoite budding had started prior to the final karyokinesis resulting in a number of merozoites budding from a central nuclear mass, a similar situation to that described by Schein et al. (1978). The stages in merozoite budding are, however, essentially the same as described above. The only difference is that as the merozoite finally buds away the nucleus has to separate from the nuclear mass, which remains within the schizont.

Occasionally large, highly branched schizonts undergoing merogony were observed in the in vitro cultures. In these cases the merozoites, that develop in an identical manner as described above, bud from the ends of the branches in a situation similar to the formation of sporozoites from the highly branched labyrinthine sporoblast (see Fawcett et al. 1982, 1985).

Cellularization from a syncytium: the nucleus plays a pivotal role

The central problem for T. parva during merogony is how to form many individual uninucleate cells from a syncytium. Moreover, because eukaryotic cells cannot form organelles de novo, each resultant individual cell must contain all the organelles or organelle precursors that the parasite requires, not only for the next stage, but for all subsequent stages of its life cycle. For example, although ribosomes will assemble spontaneously from their component parts they cannot be made without the presence of pre-existing ribosomes to synthesize the required proteins. Thus, the mature merozoite, which has a distinctive surface coat, must contain at a minimum a nucleus and nuclear envelope, a mitochondrion, ribosomes and the rhoptry complex to allow for entry and establishment in the next host cell.

We have shown that the way T. parva controls what organelles are destined to be included in the merozoites is by coupling these organelles to the nuclear envelope, that in turn associates both directly and indirectly to the schizont plasma membrane. Thus, as shown diagram-matically in Fig. 10, a portion of the two membranes of the nuclear envelope become tightly apposed and in turn become associated with the schizont plasma membrane. The rhoptry complex associates with the nuclear spindle pole bodies and subsequently with a newly formed specialization situated immediately beneath the plasma membrane. The mitochondria bind to an unspecialized portion of the nuclear envelope, to which ribosomes also attach. All that is now required is a mechanism to bud off individual nuclei with their associated array of organelles from the syncytial schizont. This appears to be accomplished by the progressive évagination of the plasma membrane in association with an array of tubules that originate from the apically situated rhoptry complex. These tubules elongate and enclose the nucleus with its attached mitochondria in a manner resembling an array of ribs.

Fig. 10.

Diagrammatic summary of the process of merozoite formation in Theileria parva. 1. Schizont prior to the onset of merogony. 2. The earliest events in merogony include the appearance of both RER and SER, a prominent surface coat, rhoptries and changes in the appearance of the nuclei. 3. The schizont nuclear envelope membranes become closely apposed around part of the nucleus and this specialized portion becomes preferentially associated with the schizont plasma membrane (see insert). The bodies become associated with the nuclear spindle pole bodies and are attached to a sublamellar specialization beneath the schizont plasma membrane. 4–6. The merozoites bud from the surface in a synchronous manner with the rhoptry complex at the apical end. An array of tubular structures arising from the apical end forms a sublamellar basket enclosing the rhoptry complex, mitochondria and the basally located nucleus.

Fig. 10.

Diagrammatic summary of the process of merozoite formation in Theileria parva. 1. Schizont prior to the onset of merogony. 2. The earliest events in merogony include the appearance of both RER and SER, a prominent surface coat, rhoptries and changes in the appearance of the nuclei. 3. The schizont nuclear envelope membranes become closely apposed around part of the nucleus and this specialized portion becomes preferentially associated with the schizont plasma membrane (see insert). The bodies become associated with the nuclear spindle pole bodies and are attached to a sublamellar specialization beneath the schizont plasma membrane. 4–6. The merozoites bud from the surface in a synchronous manner with the rhoptry complex at the apical end. An array of tubular structures arising from the apical end forms a sublamellar basket enclosing the rhoptry complex, mitochondria and the basally located nucleus.

In retrospect, it seems obvious that in order to have a precise assortment of organelles sequestered into each merozoite, the organelles must self-associate in some way. Furthermore, since the nucleus dominates in size and is limited by a membrane that has inserted in it at least one spindle pole body to which other objects could be attached, it would be a reasonable candidate to begin this self association. It is likely that a similar mechanism may exist in other cases where cellularization from a syncytium occurs, as will be discussed in the next section.

How cellularization proceeds from syncytia in other organisms: again the nucleus appears to play a pivotal role

Although cellularization from a syncytium is not a common phenomenon in biological systems, it nevertheless occurs in, for example, insect embryogenesis, endosperm cellularization in plants and in spermatogenesis. In all cases, when cellularization occurs, a common feature is the central role played by the nucleus in orchestrating the non-random distribution of organelles and other inclusions as it does in merogony in T. parva and in other apicomplexan parasites. For example, during insect embryogenesis there is a series of rapid nuclear divisions without cytokinesis giving rise to the syncytial blastoderm in which a proportion of the nuclei migrate outwards in a stepwise manner to become aligned beneath the syncytial surface membrane. Each peripherally located nucleus is associated with a domain of structured cytoplasm. By the subsequent invagination of the outer plasma membrane these nuclei with their associated cytoplasm form a cellular layer (see Foe and Alberts, 1983). The whole of this process of cellularization involves a complex network of interacting cytoskeletal proteins (see Miller et al. 1989). A similar process involving the cytoskeleton occurs in the process of cellularization of the endosperm of higher plants (Fineran et al. 1982). During spermatid differentiation from a syncytium, the nucleus with its limited but highly selective array of associated organelles buds away from the mass of cytoplasm leaving an anucleate residual mass (Baccetti and Afzelius, 1976). Again the nucleus plays a pivotal role in determining what organelles and inclusions will be present in the mature spermatid by binding to the requisite organelles in a manner similar to that described in merogony (Tilney, 1976).

Merogony, sporogony ‘and piroplasm differentiation in T. parva all occur by a similar mechanism

It is our contention that merogony, sporogony and piroplasm differentiation, all processes involving cellularization from a syncytium, have or undergo similar if not identical intermediate steps resulting in mature merozoites, sporozoites and piroplasms all of which contain a similar set of organelles packaged in an identical manner. In particular, the end products of merogony, sporogony and piroplasm differentiation are all approximately 1–2 μm in diameter. In each case, the zoite nucleus is eccentrically located with a modified portion of the nuclear envelope closely apposed to the outer plasma membrane. The mitochondria are associated with the non-specialized portion of the nuclear envelope, to which ribosomes also attach. At the pole opposite to the nucleus, the rhoptries are attached to a submembrane specialization, although in T. parva, unlike other Apicomplexa (e.g. Toxoplasma gondii, see Nichols and Chiappino, 1987), the merozoites and sporozoites do not possess a clearly demarcated apical complex. Because of these structural similarities between the different developmental stages in T. parva, it seems reasonable to suspect that the mechanisms of cellularization from the respective syncytia will be similar to the process of merogony as described above. Although the information on sporogony and piroplasm differentiation in the literature is less than complete, what is known is consistent with the notion that during sporogony the process whereby the rhoptry complex initially associates with the spindle pole bodies and subsequently with a sub-plasmalemmal specialization (see Fig. 26, Fawcett et al. 1985) is identical to what we have found in merogony. Similarly, in both sporogony and merogony the micronemes or microspheres appear to be formed either just before or immediately after the completion of zoite formation (Fawcett et al. 1985).

The only major difference between merogony and sporogony in T. parva is the apparent absence in sporogony of a basket of tubules similar to the ones formed during merozoite budding. As we have been unable to see this basket of tubules in schizonts undergoing merogony fixed in the same manner as Fawcett et al. (1982, 1985) it is likely that this is not a real difference in the two processes.

How the life cycle of Theileria may have evolved and what controls it

Up until now investigators have concentrated their attention on specific stages in the life cycle of a single parasitic protozoan. There are several reasons for this, not the least of which is the problem of obtaining infected material from more than one or two stages of the life cycle that invariably involves two separate hosts. For example, T. parva is a tick-borne protozoan parasite of cattle and in each host different cells are infected at different times (i.e. in the tick intestinal epithelial cells and then selective cells in the salivary gland, in the cow a subpopulation of lymphocytes and then red blood cells). Thus to compare what is happening in different hosts and in different host cells requires many animals (both ticks and cattle) that have been infected at different times. In fact some of these intermediate stages have not been well described, due in part to the difficulties of obtaining suitable material, either because of the low densities of parasites obtained from experimental animals or due to the lack of alternative tissue culture systems.

This report not only describes the process of merogony in T. parva in detail, but it also attempts for the first time to relate ultrastructural events that occur at merogony, sporogony, and at the piroplasm stage; three stages that each involve cellularization from a syncytium. As already mentioned, there are remarkable similarities in the structure of the merozoites, sporozoites and piroplasms and in the process of their formation from syncytia. Because there are these repeating patterns in the life cycle of Theileria, it is unlikely that the three processes evolved independently. What would appear more likely is that the developmental pattern of cellularization has been evolved once, and that the same process is used repeatedly by the parasite throughout its life cycle. The most logical way of repeatedly undergoing the same developmental process is for the organism to use the same cassette of genes with only the replacement of a few genes each time to allow for minor differences in, for example, the composition of the surface coat. Furthermore, the sporozoites and merozoites of T. parva enter and establish within their respective host cells in a morphologically similar manner (see Shaw et al. 1991; Shaw, unpublished observations), indicating that the parasite probably uses an identical or a very similar method of entry into the different host cells. A logical conclusion from these observations would be that T. parva and by extension many, if not all, members of the Apicomplexa which undergo similar if not identical developmental processes a number of times during their life cycles, use a single cassette of genes to perform each of these similar and repeated processes. Thus the parasite may be able to undergo a number of seemingly disparate developmental processes numerous times throughout its life cycle without the requirement of an extensive genome.

Similarities between Theileria and related protozoan parasites, and how by knowing their comparative cell biology one can progress more rapidly in understanding any one of the protozoa

The protozoan phylum, Apicomplexa, includes a large number of parasitic organisms of both medical and veterinary importance. The processes of sporogony, merogony and gametogenesis are a common feature of the life cycle and all involve the formation of individual cells from a multinucleate syncytial body. Furthermore, each newly formed cell contains a single nucleus and a precise complement of organelles. Therefore, it is not too surprising that the processes of sporogony, merogony and gametogenesis appear to proceed in a similar manner. From the various descriptions of the individual processes in a wide range of Apicomplexa (e.g. Varghese, 1977; Pacherco and Fayer, 1977; Schein et al. 1978; Wong and Desser, 1978; Dubremetz and Elsner, 1979; Entzeroth, 1983; Fawcett et al. 1982, 1985; Mehlhom and Schein, 1984; Meis et al. 1985; Barta et al. 1987; Klein et al. 1988) sporogony, merogony and gametogenesis all invariably proceed in a synchronized manner and involve a co-ordinated sequence of events resulting in a large number of cells being formed simultaneously from a syncytium. Characteristically, the co-ordinated sequence of events involves extensive mitotic nuclear division followed by some degree of condensation of the nuclear material, the formation, at an early stage, of a distinct extracellular coat (e.g. Fawcett et al. 1982, 1985; Glascodine et al. 1990; Clark et al. 1989; Hamilton et al. 1988; Posthuma et al. 1988), multiplication of the mitochondria, an increase in endoplasmic reticulum and ribosomes, and the production of various secretory organelles (e.g. rhoptries, micronemes and microspheres). The formation of the individual cells from the syncytium occurs by a progressive budding process with the apical complex (a characteristic feature of the Apicomplexa consisting, in the majority of cases, of the polar ring and conoid complex) forming the leading pole of the developing cell. In the merozoites and sporozoites of T. parva, the conoid is absent, and the polar ring-inner membrane complex very much reduced as compared with the zoites of other Apicomplexa. The budding of the individual cells is associated with, and presumably induced by, the development of an orderly array of tubules arising from the apical structure. These tubules form the subpellicular cytoskeleton of the developing cell. In some organisms such as T. parva, the mature merozoites have secondarily lost this cytoskeletal complex while in many others (e.g. Toxoplasma, Eimeria) it is a characteristic feature of the free zoites and is involved in parasite motility and in the process of host cell invasion (e.g. Russell and Sinden, 1981; Russell, 1983). Thus, what is now apparent is that by being aware of how related protozoan parasites function we can in many cases predict how a particular parasite may function.

We are grateful to Patrick Theuri and Daniel Ngugi for technical assistance, Francis Mwakima and colleagues in the ILRAD Tick Unit for provision of sporozoites, Dr Tom Dolan for providing the lymph node biopsy samples and other material from infected cattle and for helpful comments on the manuscript, and Bob Golder for the excellent drawings. L. G. Tilney was supported by grant HD 14474 from the National Institutes of Health.

Baccetti
,
B.
and
Afzelius
,
B. A.
(
1976
).
The Biology of the Sperm Cell
.
Karger, Basel
.
Bannister
,
L. H.
,
Mitchell
,
G. H.
,
Butcher
,
G. A.
,
Dennis
,
E. D.
and
Cohen
,
S.
(
1986
).
Structure and development of the surface coat of erythrocytic merozoites of Plasmodium knowlesi
.
Cell Tissue Res
.
245
,
281
290
.
Barta
,
J. R.
,
Boulard
,
Y.
and
Desser
,
S. S.
(
1987
).
Ultrastructural observations on secondary merogony and gametogony of Dactylosoma ranarum Labbe, 1894 (Eucoccidiida; Apicomplexa)
.
J. Parasitai
.
73
,
1019
1029
.
Brown
,
C. G. D.
(
1987
).
Theileriidae
.
In: In vitro Methods of Parasite Cultivation
, (ed.
A. E. R.
Taylor
and
J. R.
Baker
), pp.
230
253
.
Academic Press
,
London
.
Clark
,
J. T.
,
Donachie
.
S.
,
Anand
,
R.
,
Wilson
,
C. F.
,
Heidrich
,
H-G.
and
McBride
,
J. S.
(
1989
).
46-53 kilodalton glycoprotein from the surface of Plasmodium falciparum merozoites
.
Mol. Biochem. Parasitai
.
32
,
15
24
.
Conrad
,
P. A.
,
Denhatn
,
D.
and
Brown
,
C. G. D.
(
1986
).
Intraerythrocytic multiplication of Theileria parva in vitro: an ultrastructural study
.
Int. J. Parasitai
.
16
,
223
229
.
Conrad
,
P. A.
,
KeUy
,
B. G.
and
Brown
,
C. G. D.
(
1985
).
Intraerythrocytic schizogony of Theileria annulata
.
Parasitology
91
,
67
82
.
Dubremetz
,
J. F.
and
Elsner
,
Y. Y.
(
1979
).
Ultrastructural study of schizogony of Eimeria bovis in cell cultures
.
J. Protozool
.
26
,
367376
.
Entzeroth
,
R.
(
1983
).
Electron microscope study of merogony preceding cyst formation of Sarcocystis sp. in roe deer (Capreolus capreolus)
.
Zeit. Parasitenk
.
69
,
447
456
.
Fawcett
,
D. W.
,
Buscher
,
G.
and
Doxsey
,
S.
(
1982
).
Salivary gland of the tick vector of East Coast fever. III. The ultrastructure of sporogony in Theileria parva
.
Tissue and Cell
14
,
183
206
.
Fawcett
,
D. W.
,
Conrad
,
P. A.
,
Grootenhuis
,
J. G.
and
Morzaria
,
S. P.
(
1987
).
Ultrastructure of the intraerythrocytic stage of Theileria species from cattle and waterbuck
.
Tissue and Cell
19
,
643
655
.
Fawcett
,
D. W.
,
Young
,
A. S.
and
Leitch
,
B. L.
(
1985
).
Sporogony in Theileria (Apicomplexa: Piroplasmida)
.
J. Submicrosc. Cytol
.
17
,
299
314
.
Fineran
,
B. A.
,
Wild
,
D. J. C.
and
Ingerfeld
,
M.
(
1982
).
Initial wall formation in endosperm of wheat, Triticum aestivum: a réévaluation
.
Can. J. Botany
60
,
1776
1795
.
Foe
,
V. E.
and
Alberts
,
B. M.
(
1983
).
Studies of nuclear and cytoplasmic behaviour during the five mitotic cycles that precede gastrulation in Drosophila embryogenesis
.
J. Cell Sci
.
61
,
31
70
.
Glascodine
,
J.
,
Tetley
,
L.
,
Tait
,
A.
,
Brown
,
D.
and
Shiels
,
B.
(
1990
).
Developmental expression of a Theileria annulata merozoite surface antigen
.
Mol. Biochem. Parasitai
.
40
,
105
112
.
Hamilton
,
A. J.
,
Davies
,
C. S.
and
Sinden
,
R. E.
(
1988
).
Expression of circumsporozoite protein revealed in situ in the mosquito stages of Plasmodium berghei by the Lowicryl-immunogold technique
.
Parasitology
96
,
237
280
.
Klein
,
T. A.
,
Akin
,
D. C.
,
Young
,
D. G.
,
Telford
,
S. R.
and
Butler
,
J. F.
(
1988
).
Sporogony, development and ultrastructure of extrinsic stages of Plasmodium mexicanum
.
Int. J. Parasitai
.
18
,
463
476
.
Lalor
,
P. A.
,
Morrison
,
W. L
,
Goddeeris
,
B. M.
,
Jack
,
R. J.
and
Black
,
S. J.
(
1986
).
Monoclonal antibodies identify phenotypically and functionally distinct cell types in the bovine lymphoid system
.
Vet. Immunol. Immunopathol
.
13
,
121
140
.
Mehlhorn
,
H.
and
Schein
,
E.
(
1984
).
The Piroplasms: Life cycle and sexual stages
.
Advances in Parasitology
23
,
37
103
.
Meis
,
J. F. G. M.
,
Verhave
,
J. P.
,
Jap
,
P. H. K.
and
Meuwlssen
,
J. H. E. T.
(
1985
).
Fine structure of exoerythrocytic merozoite formation of Plasmodium berghei in rat liver
.
J. Protozoology
32
,
694699
.
Miller
,
K. G.
,
Field
,
C. M.
and
Alberts
,
B. M.
(
1989
).
Actin binding proteins from Drosophila embryos: a complex network of interacting proteins detected by F-actin affinity chromatography
.
J. Cell Biol
.
109
,
2963
2975
.
Nichols
,
B. A.
and
Chiappino
,
M. L.
(
1987
).
Cytoskeleton of Toxoplasma gondii
.
J. Protozool
.
34
,
217
226
.
Norval
,
R. A. I.
,
Perry
,
B. D.
and
Young
,
A. S.
(
1991
).
The Epidemiology of Theileriosis in Africa
.
Academic Press
,
London
..
Pacherco
,
N. D.
and
Fayer
,
R.
(
1977
).
Fine structure of Sarcocystis cruzi schizonts
.
J. Protozoology
24
,
382
388
.
Posthnma
,
G.
,
Meis
,
J. F. G. M.
,
Verhave
,
J. P.
,
Holllngdale
,
M. R.
,
Ponnudurai
,
T.
,
Meuwissen
,
J. H. E. T.
and
Geuze
,
H. J.
(
1988
).
Immunogold localization of circumsporozoite protein of the malaria parasite Plasmodium falciparum during sporogony in Anopheles stephensi midguts
.
Eur. J. Cell Biol
.
46
,
18
24
.
RusseH
,
D. G.
(
1983
).
Host cell invasion by Apicomplexa: An expression of the parasite’s contractile system?
Parasitology
50
,
199
209
.
Russell
,
D. G.
and
Sinden
,
R. E.
(
1981
).
The role of the cytoskeleton in the motility of coccidian sporozoites
.
J. Cell Sci
.
50
,
345
359
.
Schein
,
E.
,
Mehlhorn
,
H.
and
Warnecke
,
M.
(
1978
).
Electron microscopic studies on the schizogony of four Theileria species of cattle (T. parva, T. lawrencei, T. annulata and T. mutons)
.
Protistologica
14
,
337
348
.
Shaw
,
M. K.
,
Tilney
,
L. G.
and
Musoke
,
A. J.
(
1991
).
The entry of Theileria parva sporozoites into bovine lymphocytes: Evidence for MHC class I involvement
.
J. Cell Biol
.
113
,
87
101
.
Strome
,
S.
and
Wood
,
W. B.
(
1983
).
Generation of asymmetry and segregation of germ-line granules in early CaenorhabdiUs elegans embryos
.
Cell
35
,
15
25
.
Tilney
,
L. G.
(
1976
).
The polymerization of actin. II. How non-filamentous actin becomes non-randomly distributed in sperm - evidence for the association of actin with membranes
.
J. Cell Biol
.
69
,
51
72
.
Varghese
,
T.
(
1977
).
Fine structure of the endogenous stages of Eimeria labbeana. 5. Schizonts and merogony with special reference to the rhoptry-micrineme system
.
J. Protozool
.
24
,
376382
.
Ward
,
S.
(
1986
).
Asymmetric localization of gene products during development of Caenorhabditis elegans spermatozoa
.
In Society for Developmental Biology 44th Annual Symposium
(ed.
J. G.
Gall
) pp.
55
75
.
Alan R. Liss
,
New York
.
Wong
,
S. T.
and
Desser
,
S. S.
(
1978
).
Ultrastructural observations on renal schizogony of Leucocytozoon dubreuili in the American robin
.
J. Protozool
.
25
,
302
314
.