The multinucleated macroschizont stage of the proto-zoon Theileria annulata is an intracellular parasite of bovine leukocytes. The parasite induces the host cell to proliferate, and divides in synchrony with the immortalised host cell. Differentiation to the next stage occurs within the host cell culminating in the release of merozoites and destruction of the leukocyte. In this study clones of Theileria annulata macroschizont-infected cell lines were isolated by limiting dilution and tested for differentiation to the merozoite stage (mero-gony). Two cloned cell lines underwent differentiation with enhanced efficiency, while two others were of lower efficiency. Quantification was carried out using monoclonal antibodies, which showed that over 90% of the cells in an enhanced cloned cell line could be induced to differentiate. By carrying out induction at 41°C for limited periods of time followed by culture at 37°C evidence was obtained that differentiation to the merozoite is a two-step process: a preliminary reversible phase, followed by a second irreversible phase of differentiation. Analysis of the nuclear number of the macroschizont and the growth rate of the cloned cell lines showed that the ability to differentiate was associated with an increase in nuclear number (size) of the macroschizont, generated by a disruption in the synchrony between parasite growth and host cell division. We believe that these results reveal a relationship between a reduction in parasite division and differentiation, and that there are similarities between stage differentiation in parasites and cellular differentiation in higher eukaryotes.

Differentiation of a cell frequently occurs after changes in the cellular environment. In higher eukaryotes these changes can be caused by an alteration in the level of polypeptide growth factors, which regulate the proliferative activity of cellular populations (Cross and Dexter, 1991). Differentiation of protozoan parasites from one life-cycle stage to another can also occur in response to environmental changes, particularly during the transition from the invertebrate vector to the vertebrate host. The molecular mechanisms that regulate parasite stage differentiation have not been fully elucidated, although it has been postulated that an increase in heat shock gene expression is involved (Van Der Ploeg et al., 1985).

Theileria annulata is a protozoon parasite of cattle and is the causative agent of tropical theileriosis, a disease of livestock in areas of Europe, North Africa and Asia (Purnell, 1978). The bovine phase of the life cycle (Fig. 1) is initiated by inoculation of the sporozoite stage with a feeding tick. The sporozoite invades mononuclear leukocytes and differentiates through the trophozoite to the macroschizont stage (Jura et al., 1983). Macroschizont-infected host cells become immortalised, and division of parasite and host cell occurs in synchrony (Hulliger, 1965), the parasite associating closely with the microtubules of the host cell spindle to achieve distribution of the schizont to the daughter leukocytes. Differentiation to the merozoite stage (merogony) takes place within the host cell with the generation of merozoite nuclei, rhoptry and microneme organelles (Melhorn and Schein, 1984). Differentiation culminates in the destruction of the leukocyte and the release of merozoites into the extracellular environment. Free merozoites subsequently invade erythrocytes, where they mature into piroplasms (Conrad et al., 1985). Completion of the cycle occurs with the uptake of infected erythrocytes by the tick vector.

Fig. 1.

Schematic outline of T. annulata stage differentiation in the bovine host.

Fig. 1.

Schematic outline of T. annulata stage differentiation in the bovine host.

In a previous study differentiation of the macroschizont to the merozoite was induced by raising the temperature of the culture to 41°C. Reactivity of macroschizont-infected cells and merozoites with a panel of monoclonal antibodies demonstrated that this process results in significant changes in the antigenic profile of the parasite, indicating that merogony is a major point of differentiation in the mammalian phase of the parasite life cycle and that both positive and negative regulation of gene expression occur during this process (Glascodine et al., 1990). These experiments were carried out with a macroschizont-infected cell line derived by infection of peripheral blood mononuclear cells with sporozoites from a parasite stock. It has been shown that parasite stocks, and the macroschizont cell fines derived from them, consist of more than one type of parasite, which can be distinguished by monoclonal antibody reactivity and restriction fragment length polymorphisms on Southern blots (Shiels et al., 1986; Conrad et al., 1989; Toye et al., 1991). In the present study we have isolated clones of macroschizont-infected cell lines by limiting dilution, and demonstrated that certain cloned cell lines undergo differentiation with a high degree of efficiency. Using these clones we have obtained evidence that merogony is a two-step process, with a preliminary reversible phase leading to a second irreversible phase of differentiation. In addition, we present data that demonstrate that a disruption in the synchrony between parasite growth and host cell division is a determinant of differentiation to the merozoite, and that similarities exist between differentiation in parasites and higher eukaryotic cells.

Cell culture and cloning of the TaA2 macroschizont-infected cell line

The T. annulata (Ankara) macroschizont-infected cell line (TaA2) was obtained by in vitro infection of peripheral blood mononuclear cells with sporozoites from a mixed parasite stock. Maintenance of the cell line in culture was carried out at 37°C in RPMI-1640 (Gibco) supplemented with 20% heat-inactivated foetal calf serum, 8 μg ml-1 streptomycin, 8 units ml-1 penicillin, 0.6 μg ml-1 amphotericin B and 0.05% NaHCO3. Cloning of a low passage number of this cell line was carried out by limiting dilution as previously described (Shiels et al., 1986). To determine whether the cell lines had isoenzyme patterns characterstic of clones, analysis was performed for glucose 6-phosphate isomerase as outlined by Melrose et al. (1984).

Induction of differentiation in vitro

Induction of macroschizont-infected cell lines to differentiate was carried out by increasing the temperature of culture from 37°C to 41°C. The cell number of each culture was estimated by counting, using a haemocytometer, and was adjusted to 1.4 ×105 cells ml-1 by dilution with fresh medium, before and every second day after transfer to 41°C.

Giemsa staining and indirect immunofluorescence assay

Morphological examination of induced cultures was routinely carried out by light microscopy of Giemsa-stained cytospin preparations. A 50 μl sample of culture was spun at 1,500 revs min-1 (240 g) for 5 min using a Shandon cytospin 2. The preparations were then air dried for 10 min and fixed in methanol for 30 min. Staining was performed with Gurr’s improved R66 Giemsas stain (BDH) at 4% in distilled water.

Indirect immunofluorescence assay (IFA) on preparations of differentiating cultures with monoclonal antibodies raised against the piroplasm stage of the parasite (Glascodine et al., 1990) was carried out as described previously (Shiels et al., 1986). Differentiating cultures were centrifuged at 500 g for 5 min and washed three times with sterile phosphate-buffered saline (PBS). Fixation was carried out on ice in 1.8% formaldehyde (BDH) for 10 min and the cells were then washed three times in PBS. Cells were resuspended in PBS (approx. 5×106 ml-1), spotted onto FIFE Multispot slides (C.A. Hendley, Essex) and air dried. Analysis and photography of immunofluorescence was carried out with a Leitz Ortholux II fluorescent microscope and an Orthomat-W camera attachment.

Southern blotting

The isolation of DNA from the different cell lines, gel electrophoresis and Southern blotting were performed using standard methods (Maniatis et al., 1982). The L16 DNA used as a probe was isolated from a genomic library of T. annulata (Jedaida 2), cloned in pUC 18. The probe was radiolabelled by the random priming method (Feinberg and Vogelstein, 1983). Hybridisation and washing were carried out at 65°C as described by Church and Gilbert (1984).

Counts and statistical analysis

To quantify the number of cells that reacted with the monoclonal antibodies by IFA the numbers of positive and negative cells were counted in random microscope fields. Between 500 and 1000 cells were counted for each time point and cell line. Statistical analysis of the values obtained for the different cell lines was carried out using the chi2 distribution and 2×2 contingency tables. Estimation of the macroschizont nuclei number was carried out by counting the nuclei of over 100 infected cells in random fields of Giemsa-stained cytospin preparations. This was done for both cloned cell lines at each time point. The standard error of the mean (S.E.M.) was estimated, and the difference between means tested, using standard statistical formulae (Harper, 1971). In order to analyse the growth rate of the different cell lines the cell number per ml was estimated by counting samples in a haemocytometer, after the culture had been diluted to approximately 1.4 ×105 cells ml-1 at the end of a period of growth. After a further 48 h of culture the cell number was again estimated, and the increase in cell number was calculated by dividing the second count by the first. This procedure was carried out at 37°C and at 2-day intervals at 41°C, until day 6. Counting was carried out in duplicate for duplicate cultures of each cloned cell line. The S.E.M. for each time point was estimated, and the difference between means tested by using the t distribution.

Isolation of cloned macroschizont-infected cell lines

Four cell lines were successfully isolated by limiting dilution of the TaA2-infected macroschizont cell line and showed clonal phenotypes by isoenzyme analysis (data not shown). These cloned cell lines were then tested for their ability to differentiate at 41°C. Two of these clones (C9 and D7) exhibited morphology characteristic of differentiating cultures after 10 days at 41°C (Fig. 2B and C). In approximately 50% of these cells the number of nuclei was greatly increased and the parasite was enlarged, filling most of the host cell cytoplasm. In just over one third of these parasites the nuclei were bulky and stained densely. The macroschizont was so large in some cases that it completely filled the host cell, which also appeared to be enlarged. In addition to the enlarged macroschizonts, parasites containing multiple small densely staining particles were observed within approximately 22% of the host cells. This morphology is indicative of merozoite formation, and from previous studies these forms were identified as microschizonts. A high level of extracellular particles (between 300 and 400 per microscope field) with small densely staining nuclei were also observed, and these were identified as merozoites (Glascodine et al., 1990).

Fig. 2.

Giemsa-stained cytocentrifuge preparation of T. annulata (Ankara) cloned cell lines C9 and E3 following culture at 41°C. (A-D) Cloned cell line C9. (A) At 37°C; (B) day 6 at 4TC; (C) day 10 at 41°C; and (D) day 16 at 41°C. (E-H) Cloned cell line E3. (E) At 37°C; (F) day 6 at 41°C; (G) day 10 at 41°C; and (H) day 44 at 41°. Host cell nucleus (hn), macroschizont (ms), macroschizont nuclei (msn), enlarged macroschizont (ems) and merozoites (me). Bar, 10 μm.

Fig. 2.

Giemsa-stained cytocentrifuge preparation of T. annulata (Ankara) cloned cell lines C9 and E3 following culture at 41°C. (A-D) Cloned cell line C9. (A) At 37°C; (B) day 6 at 4TC; (C) day 10 at 41°C; and (D) day 16 at 41°C. (E-H) Cloned cell line E3. (E) At 37°C; (F) day 6 at 41°C; (G) day 10 at 41°C; and (H) day 44 at 41°. Host cell nucleus (hn), macroschizont (ms), macroschizont nuclei (msn), enlarged macroschizont (ems) and merozoites (me). Bar, 10 μm.

After 10 days at 41°C the level of differentiation of the other two cloned cell lines (E3 and D3) did not appear as extensive (Fig. 2F). Only 8% of the infected host cells contained enlarged macroschizonts (volume of parasite greater than 50% of the host cell cytoplasm), with approximately 1% showing the microschizont morphology consistent with merozoite formation. The small extracellular densely staining merozoites were observed at low levels (20–30 per field). Continued culture over 14 days at 41°C resulted in the loss of the cultures of cloned cell lines C9 and D7. In contrast, after a period of poor growth (from day 4 to 14) the cloned cell lines D3 and E3 recovered and could be kept at 41°C indefinitely. These cells showed the morphology of macroschizont-infected cells (see Fig. 2H), with an occasional infected cell displaying morphology consistent with merozoite formation. From these initial observations we designated the cloned cell lines as having an enhanced (C9 and D7) or a diminished (E3 and D3) differentiation phenotype.

Reactivity of differentiating parasites with monoclonal antibodies

Monoclonal antibodies have been raised that react against the merozoite and piroplasm, but not against the macroschizont of T. annulata (Glascodine et al., 1990). Using IFA, three of these monoclonals (5E1, 1D11 and 1C2) were tested against fixed slide preparations of the cloned cell lines after 8 days of culture at 41°C. Fig. 3A shows that within the represented field the majority (but not all) of cells of the enhanced cloned cell line (C9) reacted with monoclonal antibody 5E1; with most of the fluorescence being located at the periphery of the infected cell. None of the cells of the diminished cloned lines (D3 or E3), however, showed reactivity with antibody 5E1 (Table 1).

Table 1.

Reactivity of monoclonal antibodies against cloned cell lines cultured at 41 °C for 8 days

Reactivity of monoclonal antibodies against cloned cell lines cultured at 41 °C for 8 days
Reactivity of monoclonal antibodies against cloned cell lines cultured at 41 °C for 8 days
Fig. 3.

Immunofluorescence reactivity of monoclonal antibodies 5E1, 1C2 and 1D11 against fixed slide preparations of cloned cell lines C9 and E3 after 8 days of culture at 41°C. (A) Reactivity of cloned cell line C9 with monoclonal 5E1; (B) reactivity of cloned cell line E3 with monoclonal 1C2; and (C) reactivity of cloned cell line E3 with monoclonal 1D11. Note the peripheral labelling with monoclonal antibody 5E1, and the distinct stippled fluorescence with monoclonal antibodies 1C2 and 1D11. Bar, 10 μm.

Fig. 3.

Immunofluorescence reactivity of monoclonal antibodies 5E1, 1C2 and 1D11 against fixed slide preparations of cloned cell lines C9 and E3 after 8 days of culture at 41°C. (A) Reactivity of cloned cell line C9 with monoclonal 5E1; (B) reactivity of cloned cell line E3 with monoclonal 1C2; and (C) reactivity of cloned cell line E3 with monoclonal 1D11. Note the peripheral labelling with monoclonal antibody 5E1, and the distinct stippled fluorescence with monoclonal antibodies 1C2 and 1D11. Bar, 10 μm.

In contrast to 5E1, antibody 1C2 clearly reacted against a small proportion of cells of the diminished (D3 and E3) cloned cell fines but showed no reactivity against cells of the enhanced (D7 or C9) cloned cell lines. The staining pattern differed from that of monoclonal 5E1 in that the fluorescence obtained was stippled and distinct (see Fig. 3 B). The third monoclonal (1D11) gave the same staining pattern as monoclonal 1C2 (Fig. 3 C) but, unlike the other two antibodies, reacted against cells of all the cloned cell lines (Table 1). From these results we concluded that in addition to reacting with the liberated merozoite the three antibodies also detected parasites that were in the process of merogony, and could therefore be used as markers for differentiation.

Genotypic analysis of cloned cell lines by Southern blotting

In order to determine whether any differences between the enhanced and the diminished cloned cell lines existed at the genomic level, Southern blot analysis was carried out. A restriction fragment length polymorphism was observed when the T. annulata DNA probe L16 (Ben Miled, unpublished data) was hybridized to EcoRI-digested cell line DNA. This was most evident in the major bands of 7.1 kb and 1.3 kb, which were specific to either the enhanced (7.1 kb; Fig. 4, tracks 2 and 3) or the diminished (1.3 kb; Fig. 4, tracks 4 and 5) cloned cell lines. Both of these bands were present in DNA isolated from the T. annulata (Ankara) cell line TaA46 (Fig. 4, track 1).

Fig. 4.

Southern blot analysis of EcoRI restriction-digested DNA isolated from different cell lines. The blot was hybridised with the T. annulata L16 probe as described in Materials and methods. Track 1, DNA isolated from cell line TaA46; tracks 2 and 3, DNA isolated from cloned cell lines C9 and D7; tracks 4 and 5, DNA isolated from cloned cell lines E3 and D3.

Fig. 4.

Southern blot analysis of EcoRI restriction-digested DNA isolated from different cell lines. The blot was hybridised with the T. annulata L16 probe as described in Materials and methods. Track 1, DNA isolated from cell line TaA46; tracks 2 and 3, DNA isolated from cloned cell lines C9 and D7; tracks 4 and 5, DNA isolated from cloned cell lines E3 and D3.

Quantitification of differentiation using monoclonal antibodies 1D11 and 5El

To compare the ability of enhanced (cloned cell line C9), diminished (cloned cell line E3) and parental (TaA2) cell lines to differentiate, fixed preparations of cells after different time points of culture at 41°C were tested by IFA with monoclonal antibodies 1D11 and 5E1, and the percentage of positive cells was estimated. The analysis with monoclonal 1D11 (Table 2) showed that at all time points tested the percentage of cells that reacted positively to antibody 1DI1 was highest with the enhanced cloned cell line (C9). At day 12 the percentage of positive cells for this cell line was nine times greater than that estimated for the diminished cloned cell line (E3) and over seven times greater than the parental cell line (TaA2). The number of positive cells in the diminished cloned cell line (E3) increased to 8% at day 14 (data not shown), but after culture for 44 days at 4°C this level was reduced to 0.2%.

Table 2.

Percentage of cells positive with monoclonal 5El or 1D11 before and after induction at 41°C

Percentage of cells positive with monoclonal 5El or 1D11 before and after induction at 41°C
Percentage of cells positive with monoclonal 5El or 1D11 before and after induction at 41°C

Owing to the inability of monoclonal antibody 5E1 to detect differentiating cells of the diminished cloned cell lines (see Table 1) the analysis with this antibody was confined to the enhanced cloned cell line (C9) and the parental line (TaA2). At day 10 over 80% of the cells of clone C9 reacted with monoclonal 5E1, and this high level increased to 92% at day 12. These values compared with a maximum (at day 10) of 18% for the parental cell line (TaA2).

In addition to comparing the abilities of the three cell lines to differentiate, the quantitative analysis showed a difference in the number of cells of the enhanced cloned cell line (C9) that reacted with antibody 5E1 compared with antibody 1D11. Thus, after 10 days of culture at 41°C three times more cells were positive with antibody 5E1 than with 1D11. Interestingly, by day 12 the difference was reduced to one and a half times.

Limited incubation of cultures at 41°C results in a reduction in the level of differentiation

To test whether placing the cultures of macroschizont-infected cells at 41°C resulted in a rapid induction of differentiation, the enhanced cloned cell line (C9) was placed in culture at 41°C. Following incubation for different time periods at the increased temperature, cultures were returned to 37°C. The cultures were then maintained at 37°C up to the time at which visible differentiation, as assessed by Giemsa-stained morphology, had occurred in the control culture at 41°C. Fixed slide preparations were then made and the level of differentiation was assessed by IFA with monoclonal 5E1. Table 3 shows that 24 h at 41°C followed by 6 days at 37°C produced the same result as the negative control (37°C) culture in that no cells were induced to differentiate by this procedure. A proportion of the cells (14%) were positive for differentiation after 48 h in culture at 41°C followed by 5 days at 37°C. This proportion increased the longer the cells were maintained at the higher temperature, so that after 4 days at 41°C followed by 3 days at 37°C 50% of the cells were estimated to be differentiating. In agreement with the above observation, it was found that the highest level of differentiation (73%) was achieved by continuous culture at 41°C for 7 days (see Table 3).

Table 3.

Percentage cells positive with monoclonal 5E1 after different times of culture at 41°C, followed by culture at 3TC % Cells

Percentage cells positive with monoclonal 5E1 after different times of culture at 41°C, followed by culture at 3TC % Cells
Percentage cells positive with monoclonal 5E1 after different times of culture at 41°C, followed by culture at 3TC % Cells

Estimation of number of macroschizont nuclei during culture at 41°C

Morphological analysis of macroschizont-infected cells cultured at 41°C has shown an increase in the number of nuclear particles per macroschizont, which is associated with a larger parasite size, during differentiation to the merozoite (Hulliger et al., 1966; Glascodine et al., 1990; see Fig. 2B). To compare the extent of these changes in the enhanced (C9) with those in the diminished (E3) cloned cell lines, the mean number of nuclei per macroschizont was estimated at successive 2day time points at 41°C, and the data are presented in Fig. 5. For the enhanced cloned cell Une (C9) at 37°C the average number of nuclei per schizont was 20. This number increased by 40% (8 nuclei) after 2 days of culture at 41°C, and over the next 48 h between day 2 and day 4 the mean number of nuclei rose from 27 to 57, an increase of almost 300% from day 0. An accurate estimation of the number of nuclei was not possible at day 6, because many of the cells had macroschizonts with a very large number of nuclei (300-500) closely packed together (see Fig. 2 B).

Fig. 5.

Mean number of nuclei per macroschizont after culture of cloned cell lines C9 and E3 at 37°C and at progressive 2-day time points after induction at 41°C. Filled bar, cloned cell line E3; hatched bar, cloned cell line E3. A dot above the column represents a significant difference (P<0.01) between the means estimated at this and the preceding time point.

Fig. 5.

Mean number of nuclei per macroschizont after culture of cloned cell lines C9 and E3 at 37°C and at progressive 2-day time points after induction at 41°C. Filled bar, cloned cell line E3; hatched bar, cloned cell line E3. A dot above the column represents a significant difference (P<0.01) between the means estimated at this and the preceding time point.

The increase in the mean number of macroschizont nuclei for the diminished cloned cell line E3 over the same period of time was found to be much smaller. Thus, from day 0 (37°C) to day 4 the mean number of nuclei increased by only 30%, from 13 to 17, and it took until day 12 for the mean number of nuclei to rise to a level of 32. Following prolonged (44 days) culture of these cells at 41°C, the mean number of nuclei fell to 11, a level not significantly different (P<0.01) from the macroschizont-infected cells cultured at 37°C.

Comparison of infected cell growth at 41°C and 37°C

In order to analyse the growth rate of the cloned cell lines C9 and E3 during the first 6 days of differentiation at 41°C, cell counts were taken every 48 h. Fig. 6 shows that over the first 2 days of growth at increased temperature the amplification in cell number (7.3-fold) of the enhanced cloned cell line (C9) culture was greater than that estimated for the culture over 48 h at 37°C (4.2-fold). This level of growth at 41°C was not maintained, however, and as the time course progressed the increase in cell number of the culture steadily declined to 6-fold from day 2 to 4 and 3.1-fold from day 4 to 6. In general, the fluctuation in growth observed over the time course was similar for the culture of the diminished cloned cell line (E3). However, at each time point the estimated increase in cell number for the diminished cloned cell line was always greater than for the enhanced (see Fig. 6). For example, after 48 h at 37°C the increase in cell number of the diminished cloned cell line (E3) was estimated at 6.7-fold, compared to 4.2-fold for the enhanced cloned cell line (C9).

Fig. 6.

Increase in cell number of cloned cell line C9 (filled bar) and cloned cell line E3 (hatched bar) over 48 h periods of culture. Days shown represent the value estimated at 37°C (day 0) and at 41°C (days 2–6). A dot above the column represents a significant difference (P<0.01) between the means estimated at this and the preceding time point.

Fig. 6.

Increase in cell number of cloned cell line C9 (filled bar) and cloned cell line E3 (hatched bar) over 48 h periods of culture. Days shown represent the value estimated at 37°C (day 0) and at 41°C (days 2–6). A dot above the column represents a significant difference (P<0.01) between the means estimated at this and the preceding time point.

We have isolated clones of Theileria macroschizont-infected cell lines that showed an enhanced or diminished ability to differentiate to the next stage in the parasite life cycle, the merozoite. Thus, we have established cell lines in which the majority of the parasites (>90%) differentiate over approximately the same time period. These cloned cell lines are a considerable improvement on cultures that have been used in previous studies of merozoite differentiation (Hulliger et al., 1966; Danskin and Wilde, 1976; Fritsch et al., 1988; Glascodine et al., 1990), and are essential for analysis of the molecular events that occur during merogony.

The percentage of cells of the enhanced cloned cell Une (C9) that reacted with monoclonal ID 11 was found to be smaller than the number positive with monoclonal 5E1. In addition to this quantitative difference, monoclonal 5E1 produced a halo type pattern of immunofluorescence, whereas parasites that reacted with monoclonal 1D11 had a stippled pattern (see Fig. 3). Immunoelectron microscopic studies have shown that monoclonal antibody 5E1 binds to the surface of the differentiating macroschizont, whereas monoclonal 1D11 reacts with the rhoptry located within the differentiating parasite (L. Tetley and B. Shiels, unpublished data). We conclude that during the differentiation process the molecule detected by monoclonal 5E1 is expressed on the surface of the differentiating parasite before formation of the rhoptry is completed.

The reactivity of two of the monoclonal antibodies was found to be specific for either the enhanced (monoclonal 5E1) or the diminished (monoclonal 1C2) cloned cell line. Therefore, the epitopes recogised by these monoclonals are not conserved between the parasites of the different cloned cell lines, and the differentiating macroschizonts and merozoites generated from the two types of cloned cell fine must be antigenically distinct. From these results it was predicted that the parasite genotypes represented by the different type of cloned cell line would be distinct, and this was confirmed by the result of the Southern analysis which showed a restriction fragment polymorphism between the enhanced and the diminished cloned cell fines (Fig. 4). Thus, like other stocks of Theileria (Allsop and Allsop, 1988; Conrad et al., 1989) our parental (TaA2) cell line contained at least two different parasite genotypes. These genotypes correlated with the enhanced or diminished differentiation phenotypes of the infected cell lines. This suggests that parasites of different genotype can have differing abilities to differentiate.

The results of the pulse experiment (see Table 3) showed that differentiation was not an immediate response to the increased temperature, as the cells had to be in culture for more than 24 hours at the higher temperature before a proportion were induced to differentiate. Furthermore, it was evident that returning the cultures to 37°C reduced the level of differentiating cells from that obtained by continuing culture at 41°C for 7 days. From these results it appears that in order to differentiate the cells must go through a preliminary phase at 41°C, and that this early phase is reversible if the cells are returned to 37°C. However, returning the culture to the lower temperature after 4 days at 41°C did not result in a reduction in differentiation, when compared with the level observed immediately after 4 days of culture at 41°C (data not shown). Thus, we believe that, as in a number of other systems, including protozoon parasites (Watson et al., 1987; Bruce et al., 1990), the macroschizont once triggered becomes committed to an irreversible phase of differentiation that can proceed to completion at 37°C.

A characteristic that has been observed in the early stages of differentiation to the merozoite is an enlargement of the macroschizont, with a concomitant increase in its nuclear number (Hulliger et al., 1966; Glascodine et al., 1990). Our results show that the number of nuclei per macroschizont and the parasite size increased greatly during the first 6 days of culture of the enhanced cloned cell line C9 (see Fig. 5 and Fig. 2B), but with the diminished cloned cell line (E3) the increase was significantly smaller and took longer to occur. Furthermore, after 44 days of culture of the diminished line at 41°C the level of differentiation was very low (Table 2) and the mean number of nuclei per macroschizont was not significantly different from the number estimated at 37°C (Fig. 5). Thus, there is an association between the increase in the number of nuclei per macroschizont (and parasite size) and the ability of the cell lines to differentiate. We propose that this increase in parasite growth takes place during the initial reversible phase of the differentiation process until a condition is reached that triggers the second irreversible phase of differentiation to the merozoite.

In the growth analysis of the enhanced cell line (C9) we found, as postulated by Hulliger et al. (1966), that the growth of the macroschizont becomes asynchronous with the division of the host cell at 41°C. For the first 2 days the infected cell grew faster at the higher temperature (see Fig. 6). Therefore, in order to account for the increase in the number of parasite nuclei and size observed over this time (see Fig 5; cell line C9), the rate of parasite nuclear division must have been even greater than the increased host cell division. This increase in parasite nuclear division was then coupled with a decline in host cell division from two to six days, and it is likely that at a certain time point host cell division was inhibited completely. Owing to the association of the parasite with host cell spindle for separation of the macroschizont, as the rate of host cell division is reduced the division of parasites also becomes slower. Thus, an increase in parasite nuclear division/growth is accompanied by a decrease in parasite cell division, and consequently the macroschizont is enlarged to the point where the host cell cytoplasm is completely filled (see Fig. 2B). The diminished cell line (E3) divided faster than the enhanced cell line at all time points tested, and the disruption of synchrony between parasite and host cell division was less at 41°C. This and the smaller size of the macroschizonts at 37°C probably caused the diminished phenotype of the E3 cell fine, as it would take longer for the parasite to reach a predetermined size or condition which triggers differentiation.

Studies on the kinetics of replication of T. parva in vivo showed that an increase in the nuclear number of the macroschizont preceded the appearance of pirop-lasm infected erythrocytes. It was argued that microschizont formation (i.e. differentiation to the merozoite) was time-dependent and was preceded by a fixed number of macroschizont-infected cell multiplications (Jarrett et al., 1969). The hypothesis of a mitotic clock regulating differentiation has been proposed for higher eukaryotic cells (Temple and Raff, 1986). Recent studies have demonstrated that the timing of such a clock is determined by control over proliferation, as in oligodendrocyte progenitor cells differentiation occurs prematurely when their proliferation potential is reduced by removal of growth factor (Raff et al., 1988). Moreover, differentiation can be inhibited by continued stimulation of the cells to proliferate by the addition of growth factors (Bogler et al., 1990). We believe the situation to be similar for differentiation to the merozoite, and that the timing of the initial reversible phase is regulated by the rate of increase in parasite size, with a concomitant decrease in the proliferation potential of both host cell and parasite.

In addition to studies carried out on higher eukaryotes, a correlation between a reduction in proliferation and differentiation has also been found for other protozoon parasites. For example, during its life cycle Trypanosoma brucei alternates between proliferative and non-proliferative phases (Vickerman, 1985), and repeated passage of the trypomastigote (bloodstream) stage can result in a loss of differentiation (Hajduk and Vickerman, 1981). In Leishmania differentiation from the insect promastigote stage to the mammalian infective stage (metacyclic) involves a change from logarithmic-to stationary-phase growth (Sacks and Perkins, 1984, 1985), and the inability of Leishmania major promastigotes to differentiate after transfer of an in vitro culture from 25°C to 35°C has been associated with proliferation at the higher temperature (Shapira et al., 1988). It is possible, therefore, that one of the parameters that leads to the triggering of stage differentiation in protozoan parasites is a reduction in the rate of parasite division, and that there are fundamental similarities between the mechanisms that regulate differentiation in parasites and higher eukaryotic cells.

Thanks to Alan May for help with photomicroscopy and photographic printing, to Frank Wright for advice on statistical analysis, and to the Wellcome Trust for continued support.

Allsop
,
B. A.
and
Allsop
,
M. T. E. P.
(
1988
).
Theileria parva: genomic studies reveal intra-specific sequence diversity
.
Mol. Biochem. Parasit
.
28
,
77
84
.
Bogler
,
O.
,
Wren
,
D.
,
Barnett
,
S. C.
,
Land
,
H.
and
Noble
,
M.
(
1990
).
Cooperation between two growth factors promotes extended selfrenewal, and inhibits differentiation, of O-2A progenitor cells
.
Proc. Nat. Acad. Sci. U.S.A
.
87
,
6368
6372
.
Bruce
,
M. C.
,
Alano
,
P.
,
Duthie
,
S.
and
Carter
,
R.
(
1990
).
Commitment of the malaria parasite Plasmodium falciparum to sexual and asexual development
.
Parasitology
100
,
191
200
.
Church
,
G. M.
and
Gilbert
,
W.
(
1984
).
Genomic sequencing
.
Proc. Nat. Acad. Sci. U.S.A
.
81
,
1991
1995
.
Conrad
,
P. A.
,
Baldwin
,
C. L.
,
Brown
,
W. C.
,
Sohan pal
,
B.
,
Dolan
,
T. T.
,
Goddeeris
,
B. M.
,
De Martini
,
J. C.
and
Ole-Moi Yoi
,
O. K.
(
1989
).
Infection of Bovine T-cell clones with genotypically distinct Theileria parva parasites and analysis of their cell surface phenotype
.
Parasitology
99
,
205
213
.
Conrad
,
P. A.
,
Kelly
,
B. G.
and
Brown
,
C. G. D.
(
1985
).
Intraerythrocyte schizogony of Theileria annulata
.
Parasitology
91
,
6782
.
Cross
,
M.
and
Dexter
,
T. M.
(
1991
).
Growth factors in development, transformation, and tumorigenesis
.
Cell
64
,
271
280
.
Danskin
,
D.
and
Wilde
,
J. K. H.
(
1976
).
Simulation in vitro of bovine host cycle of Theileria parva
.
Nature
261
,
311
312
.
Feinberg
,
A. P.
and
Vogelstein
,
B.
(
1983
).
A technique for radiolabelling DNA restriction fragments to high specific activity
.
Anal. Biochem
.
132
,
6
13
.
Fritsch
,
F. M.
,
Melhorn
,
H.
,
Schein
,
E.
and
Hauser
,
M.
(
1988
).
The effects of drugs on the formation of Theileria annulata merozoites in vitro
.
Parasit. Res
.
74
,
340
343
.
Glascodine
,
J.
,
Tetley
,
L.
,
Tait
,
A.
,
Brown
,
D.
and
Shiels
,
B. R.
(
1990
).
Developmental expression of a Theileria annulata merozoite surface antigen
.
Mol. Biochem. Parasit
.
40
,
105
112
.
Hajduk
,
S. L.
and
Vickerman
,
K.
(
1981
).
Absence of detectable alteration in the kinetoplast DNA of a Trypanosoma brucei clone following loss of ability to infect the insect vector (Glossina morsitans)
.
Mol. Biochem. Parasit
.
4
,
17
28
.
Harper
,
W. M.
(
1971
).
Statistics
.
Macdonald and Evans
,
Handbooks London
.
Hulliger
,
L.
(
1965
).
Cultivation of three species of Theileria in lymphoid cell in vitro
.
J. Protozool
.
12
,
649
655
.
Hulliger
,
L.
,
Brown
,
C. G. D.
and
Wilde
,
J. K. H.
(
1966
).
Transition of developmental stages of Theileria parva in vitro at high temperature
.
Nature
211
,
328
329
.
Jarret
,
W. F. H.
,
Crighton
,
C. W.
and
Pirie
,
H. M.
(
1969
).
Theileria parva: Kinetics of regulation
.
Exp. Parasit
.
24
,
9
25
.
Jura
,
W. G. Z. O.
,
Brown
,
C. G. D.
and
Kelly
,
B.
(
1983
).
Fine structure of the early developmental stages of Theileria annulata in vitro
.
Vet. Parasit
.
12
,
31
44
.
Manlatis
,
T.
,
Fritsch
,
E. F.
and
Sambrook
,
J.
(
1982
).
Molecular Cloning: A Laboratory Manual
.
Cold Spring Harbor Laboratory
,
Cold Spring Harbor, NY
.
Melhorn
,
H.
and
Schein
,
E.
(
1984
).
The piroplasm: life cycle and sexual changes
.
Adv. Parasit
.
23
,
37
103
.
Melrose
,
T. R.
,
Brown
,
C. G. D.
,
Morzarla
,
S. P.
,
Ocama
,
J. G. R.
and
Irvin
,
A. D.
(
1984
).
Glucose phosphate isomerase polymorphism in Theileria annulata and Theileria parva
.
Trop. Animal Health Prod
.
16
,
239
245
.
Purnell
,
R. E.
(
1978
).
Theileria annulata as a hazard to cattle in countries on the Northern Mediterranean littoral
.
Vet. Sci. Commun
.
2
,
3
10
.
Raff
,
M. C.
,
Lillen
,
L. E.
,
Richardson
,
W. D.
,
Burne
,
J. F.
and
Noble
,
M. D.
(
1988
).
Platlet-denved growth factor from astrocytes drives the clock that times oligodendrocyte development in culture
.
Nature
333
,
562
565
.
Sacks
,
D. L.
and
Perkins
,
P. V.
(
1984
).
Identification of an infective stage of Leishmania promastigotes
.
Science
223
,
1417
1419
.
Sacks
,
D. L.
and
Perkins
,
P. V.
(
1985
).
Development of infective stage Leishmania promastigotes within phlebotomine sandflies
.
Am. J. Trop. Med. Hyg
.
34
,
456
459
.
Shapira
,
M.
,
McEwen
,
J. G.
and
Jaffe
,
C. L.
(
1988
).
Temperature effects on molecular processes which lead to stage differentiation in Leishmania
.
EMBO
7
(
9
),
2895
2901
.
Shiels
,
B. R.
,
McDougall
,
C.
,
Tait
,
A.
and
Brown
,
C. G. D.
(
1986
).
Antigenic diversity of Theileria annulata macroschizonts
.
Vet. Parasit
.
21
,
1
10
.
Temple
,
S.
and
Raff
,
M. C.
(
1986
).
Clonal analysis of oligodendrocyte development in culture: evidence for a developmental clock that counts cell divisions
.
Cell
44
,
773
779
.
Toye
,
P. G.
,
Goddeeris
,
B. M.
,
lams
,
K.
,
Musoke
,
A. J.
and
Morrison
,
W. I.
(
1991
).
Characterisation of a polymorphic Host cell division and differentiation in Theileria 107 immunodominant molecule in sporozoites and schizonts of Theileria parva
.
Parasit. Immun
.
13
,
49
62
.
Van Der Ploeg
,
L. H. T.
,
Gianni
,
S. H.
and
Cantor
,
C. R.
(
1985
).
Heat shock genes: regulatory role for differentation in parasitic protozoa
.
Science
228
,
1443
1446
.
Vickerman
,
K.
(
1985
).
Developmental cycles and biology of pathogenic trypanosomes
.
Br. Med. Bull
.
41
,
105
114
.
Watson
,
J. D.
,
Hopkins
,
N. H.
,
Roberts
,
J. W.
,
Steitz
,
J. A.
and
Weiner
,
A. M.
(
1987
).
The molecular biology of development
.
In The Molecular Biology of the Gene (4th edn)
, pp.
747
831
.
Benjamin/Cummings Melmo Park, CA
.