INTRODUCTION
Babesia bovis, a tick-transmitted haemoprotozoon, grows asexually within bovine erythrocytes (RBC) and causes a severe disease characterized by fever, anaemia, haematuria, and hypotensive shock syndrome, with occasional death in the infected cattle (Kuttler, Reference Kuttler and Ristic1988; Homer et al. Reference Homer, Aguilar-Delfin, Telford and Persing2000). The disease, called bovine babesiosis, often results in great economic loss in the livestock industry worldwide (Brown and Palmer, Reference Brown and Palmer1999). For the successful development of novel control measures against babesiosis, the biological mechanism whereby B. bovis infects the host RBC and asexually replicates within the infected RBC should be clearly elucidated (Yokoyama et al. Reference Yokoyama, Kumar, Bork, Okamura, Igarashi, Yano, Nam, Anuar and Shen2005, Reference Yokoyama, Okamura and Igarashi2006).
Cholesterol is a major component of the erythrocyte membrane and occupies more than 30% of the total weight of erythrocyte membrane lipids (Yawata, Reference Yawata2003). It plays important roles in the homeostatic stabilities of membrane fluidity and permeability (Yawata, Reference Yawata2003) as well as in raft formation (Simons and Ikonen, Reference Simons and Ikonen1997). The membrane cholesterol can be removed by treatment with methyl-β-cyclodextrin (MCD), a cholesterol-depleting agent (Christian et al. Reference Christian, Haynes, Phillips and Rothblat1997; Niu and Litman, Reference Niu and Litman2002); however, MCD never binds to or permeates the treated RBC (Ohtani et al. Reference Ohtani, Irie, Uekama, Fukunaga and Pitha1989). In Plasmodium falciparum and Toxoplasma gondii, cholesterol reduction in the host cell membrane is known to inhibit their growth in host cells (Coppens et al. Reference Coppens, Sinai and Joiner2000; Samuel et al. Reference Samuel, Mohandas, Harrison, McManus, Rosse, Reid and Haldar2001). For both parasites, host cholesterol is considered to be a crucial component for the formation of their parasitophorous vacuole membrane within the infected cells (Lauer et al. Reference Lauer, Vanwye, Harrison, McManus, Samuel, Hiller, Mohandas and Haldar2000; Coppens and Joiner, Reference Coppens and Joiner2003). However, the biological association of the cholesterol reduction with the asexual growth of Babesia parasites had not been studied yet. In the present study, we describe the effect of cholesterol reduction in host bovine RBC on the in vitro growth of B. bovis.
MATERIALS AND METHODS
In vitro cultivation of B. bovis
The Texas strain of B. bovis was cultured in purified bovine RBC with a serum-free GIT medium (Wako Pure Chemical Industrial, Ltd, Osaka, Japan) as described previously (Bork et al. Reference Bork, Okamura, Matsuo, Kumar, Yokoyama and Igarashi2005).
Cholesterol reduction in the bovine erythrocyte membrane
To reduce the amount of cholesterol in the erythrocyte membrane, bovine RBC were treated with MCD (Sigma, St Louis, MO, USA) as described below. An MCD stock solution of 5 mm was first prepared by dissolving 66 mg of MCD in 10 ml of phosphate-buffered saline (PBS) or a GIT medium and then sterilized by filtrating with a 0·25 μm filter (Toyo Roshi, Tochigi, Japan). Working media containing the indicated concentrations of MCD were prepared by diluting the stock solution with the PBS or GIT medium. For the reduction of membrane cholesterol, 2×109 bovine RBC were washed with PBS and incubated in 1 ml of PBS containing the indicated concentrations of MCD for 30 min at 37°C. The treated RBC were washed with PBS again and subjected to an in vitro growth assay or invasion test as described below. On the other hand, the working MCD suspended in the GIT medium was used for an in vitro replication test.
Water-soluble cholesterol (free cholesterol) (Sigma) was used to prepare the cholesterol-complex form of MCD, which is inactive for the reduction of membrane cholesterol in contrast to that of the non-conjugated form of MCD (Christian et al. Reference Christian, Haynes, Phillips and Rothblat1997). Briefly, 1, 3, and 5 mm MCD were pre-mixed with 37, 112, and 187 μg/ml of the free cholesterol, respectively, and then used for the treatment of bovine RBC. In our preliminary experiment, pre-treatment of host RBC with the indicated concentrations of the prepared MCD-cholesterol complex showed no effect on the original amount of membrane cholesterol (data not shown, Table 1). These complex-treated RBC were also used as a control, together with a PBS-treatment control, for the in vitro growth assay.
Cholesterol concentrations of MCD-treated RBC
† The percentage of each concentration to the concentration of PBS-treated RBC.
* P=0·007, ** P=0·001 (versus control concentration).
The cholesterol concentration of the MCD-, MCD-cholesterol complex-, or PBS-treated RBC was tested on the hypotonically lysed RBC. Briefly, 1×109 cells of the treated RBC were lysed with a 5 mm phosphate buffer, pH 7·4, and washed with the buffer until the haemoglobin pigment disappeared (Okamura et al. Reference Okamura, Yokoyama, Wickramathilaka, Takabatake, Ikehara and Igarashi2005). Finally, the cholesterol concentrations of the erythrocyte ghosts were measured by using a Cholesterol E-test kit (Wako Pure Chemical Industrial, Ltd) according to the manufacturer's protocol.
In vitro growth assay with MCD-treated RBC
The in vitro growth assay with MCD-treated RBC was conducted according to a previously described method (Okamura et al. Reference Okamura, Yokoyama, Wickramathilaka, Takabatake, Ikehara and Igarashi2005) with some modifications. Briefly, B. bovis-infected RBC were diluted with MCD-, MCD-cholesterol complex-, or PBS-treated RBC to obtain 1% parasitaemia in a mixture containing 80% of treated RBC and 20% of normal cells. The 50 μl erythrocyte mixture was subsequently suspended in 450 μl of the GIT medium and incubated in a 48-well tissue-culture plate (Nunc A/S, Roskilde, Denmark) in a humidified multi-gas water-jacket incubator at 37°C. At 24 h post-initiation of the culture, each parasitaemia was determined and translated to percentage growth relative to the control parasitaemia of the culture with PBS-treated RBC. The assay was carried out in triplicate for each concentration of MCD in 3 separate experiments.
In vitro invasion and replication tests using high-voltage pulsing
The effect of reducing the amount of cholesterol in MCD-treated RBC was evaluated on erythrocyte invasion by B. bovis in an in vitro invasion test using high-voltage pulsing as described previously (Franssen et al. Reference Franssen, Gaffar, Yatsuda and De Vries2003; Okubo et al. Reference Okubo, Wilawan, Bork, Okamura, Yokoyama and Igarashi2006) with some modifications. Briefly, after B. bovis-infected RBC were suspended in equal volumes of a GIT medium, the mixture of 400 μl was subjected to 5 intermittent (10 sec, 0°C) high-voltage pulses (1·5 kV, 400 Ω, 25 μF) in the Bio-Rad Gene Pulser II (Hercules, CA, USA) with a 0·2 cm pulser cuvette (Bio-Rad). After washing with the GIT medium, the pellet was resuspended with 300 μl of the GIT medium and transferred into a 96-well tissue-culture plate (Nunc A/S) with the MCD- or PBS-treated RBC at a 10% packed cell volume. After incubation at 37°C for 30 min, the number of infected RBC was counted out of a total of 5000 RBC in Giemsa-stained smears. The invasion activity was calculated as the relative percentage of the number of infected RBC in the culture with MCD-treated RBC to that in the culture with the PBS control. On the other hand, the effect of reducing the amount of membrane cholesterol in infected RBC was also evaluated on the replication of intraerythrocytic B. bovis using the high-voltage pulsing method. After the high-voltage pulsing, the purified parasites were suspended onto normal RBC and incubated at 37°C as described above. At 1 h post-initiation of the culture, the infected RBC were washed with the GIT medium and incubated in the presence of the indicated concentrations of MCD in the GIT medium at 37°C for 5 h. After incubation, Giemsa-stained smears were prepared, and the replication activity was calculated as the ratio of the number of divided parasite-containing RBC to the entire population of infected RBC, among which more than 500 cells of the infected RBC were monitored. These assays were carried out in triplicate for each concentration of MCD in 3 separate experiments.
Statistical analysis
Differences in all percentages were statistically analysed using an independent Student's t-test at P<0·05 and P<0·01 as the values representing a significant difference.
RESULTS
Effect of cholesterol reduction on the in vitro growth of B. bovis
First, the reductions in the cholesterol concentrations were evaluated in the membranes of 1, 3, and 5 mm MCD-treated RBC. As shown in Table 1, the membrane fractions of 1, 3, and 5 mm MCD-treated RBC (1×109 cells) had cholesterol contents of 3·72, 3·41, and 3·05 mg, respectively, indicating that approximately 1, 9, and 19% of the normal membrane cholesterol (3·75 mg/1×109 cells) were depleted by treatment with 1, 3, and 5 mm MCD, respectively. On the other hand, treatment with 5 mm MCD-cholesterol complex showed almost the same concentration of membrane cholesterol as that of the PBS-treatment control (Table 1). Additionally, treatment with 1 and 3 mm MCD-cholesterol complex did not change the original concentration (data not shown) that was shown in the PBS control. Because the treatment with 10 mm MCD resulted in haemolysis of the treated RBC during incubation (data not shown), the 1, 3, and 5 mm MCD-treated RBC were selected for the subsequent experiments.
Next, the effects of a cholesterol reduction were studied on the in vitro growth of B. bovis with the MCD-pre-treated RBC (Fig. 1). The inhibitory effect on their growth was significant in the culture with 5 mm MCD-treated RBC (Parasitaemia (Pa)=2·81%) compared with the control PBS-treated RBC (Pa=3·81%) (P<0·05), while the culture with 1 mm MCD-treated RBC significantly promoted the parasite's growth (Pa=4·57%) (P<0·05) (Fig. 1). On the other hand, pre-treatment of the cultured RBC with 1, 3, and 5 mm MCD-cholesterol complex did not show any significant difference on the growth of parasites as compared to that of the PBS control (5 mm treatment, Fig. 1; 1 and 3 mm treatments, data not shown). These findings indicated that the growth of the parasite was positively influenced by the amount of host membrane cholesterol, but not by other unknown effects of MCD. During the cultivations, the applied RBC did not rupture even in the culture with 5 mm MCD-treated RBC (data not shown).
Effect of MCD-treated RBC on the in vitro growth of Babesia bovis. Relative values are expressed as the percentage of parasitaemia in the culture with MCD pre-treated RBC compared to that in the control culture with PBS-treated RBC in an in vitro growth assay. Each value represents the mean±standard deviation (s.d.) in 3 separate trials. The asterisks indicate significant differences (P<0·05) between the MCD-treated and control cultures. Cho: cholesterol.
Effects of a cholesterol reduction on the invasion and replication activities by B. bovis
In order to study the effects of cholesterol reduction on the growth of B. bovis in more detail, we carried out in vitro invasion and replication tests. In the in vitro invasion test, the invasion activity of the parasite was significantly inhibited in the 3 and 5 mm MCD-treated RBC (Pa=1·24 and 0·94%, respectively) (P<0·01) compared with that in the control PBS-treated RBC (Pa=1·79%) (Fig. 2). In contrast, the 1 mm MCD-treated RBC enhanced the activity of B. bovis invasion (Pa=2·35%) (Fig. 2) in a manner similar to that observed in the in vitro growth assay (Fig. 1).
Effect of MCD-treated RBC on erythrocyte invasion by Babesia bovis. Relative values are expressed as the percentage of parasitaemia in the culture with MCD pre-treated RBC compared to that in the control culture with PBS-treated RBC in an in vitro invasion test. Each value represents the mean±s.d. in 3 separate trials. The double asterisks indicate significant differences (P<0·01) between the MCD-treated and control cultures.
To examine the effects of cholesterol reduction on the intraerythrocytic replication by B. bovis, an in vitro replication test was carried out. When the purified parasites were incubated in normal RBC for 1 h, all parasites within the infected RBC were observed as a ring forming on the stained smears (Fig. 3A). At 1 h post-initiation of the culture, the GIT medium was replaced with an MCD-containing medium, and the cultures were further incubated for 5 h. After incubation, the rates of divided parasite-containing RBC (Fig. 3B) to all infected RBC were determined as shown in Fig. 3C. Supplementation with 5 mm MCD had a significant inhibitory effect on the parasite's replication (P<0·05) compared with the control rate in the normal GIT medium. On the other hand, the media with 1 and 3 mm MCD did not show any significant effect, although the replication activity slightly increased in the 1 mm MCD-supplemented culture.
(A) Intraerythrocytic Babesia bovis after the incubation of purified parasites with normal RBC at 1 h. (B) B. bovis parasites after further 6 h incubation. The arrowheads and arrow indicate the ring-formed and divided parasites within the infected RBC, respectively. (C) Effect of MCD treatment on B. bovis replication. Relative values are expressed as the rates of divided parasite-containing RBC to all infected RBC. Each value represents the mean±s.d. in 3 separate trials. The asterisks indicate significant differences (P<0·05) between the MCD-treated and control RBC.
DISCUSSION
In the present study, the effects of cholesterol reduction in the host erythrocyte membrane were demonstrated on the in vitro growth of B. bovis using MCD-treated bovine RBC. In addition to its biochemical role as a cholesterol scavenger, MCD is also known as a cholesterol donor by forming a complex with cholesterol (Christian et al. Reference Christian, Haynes, Phillips and Rothblat1997). We therefore determined in the preliminary study, the optimal mixture rate of MCD and free cholesterol to prepare the complete MCD-cholesterol complex that does not influence the original amount of normal membrane cholesterol. The treatment with 5 mm MCD removed approximately 19% of the membrane cholesterol, but the 5 mm MCD-cholesterol complex did not change the membrane cholesterol content, indicating that the non-conjugated MCD functioned as a cholesterol scavenger, while the MCD-cholesterol complex prepared in the present study functioned as neither cholesterol scavenger nor donor on the equilibrium balance. Importantly, cholesterol reduction significantly led to growth inhibition of B. bovis. In general, cholesterol is in a dynamic equilibrium between the RBC and the culture medium (Yawata, Reference Yawata2003); however, since the GIT medium contains low concentrations of cholesterol (0·019±0·001 mg/ml), incidental influence of the medium cholesterol may be excluded in the present study. The cholesterol reduction in the host erythrocyte membrane must therefore directly influence the growth effect.
More detailed investigations of in vitro invasion and replication tests revealed that the growth inhibitory effect was mainly due to the direct inhibition of the parasite's erythrocyte invasion activity. Compared with the results of the in vitro growth assay, the in vitro invasion test showed greater effects of MCD on the inhibition of parasites, especially with regard to the 3 mm MCD treatment. This difference would result from the experimental procedures i.e. in the invasion test, all RBC had been treated with MCD and used to evaluate the direct effect on the parasite invasion; in the growth assay, however, 80% of all RBC had been pre-treated with MCD, while the other 20% remained as untreated RBC in the culture with B. bovis. Since the B. bovis parasite preferably invades intact RBC, the MCD effect was weaker in the growth assay than it was in the invasion test. Because a reduction in membrane cholesterol increases the fluidity of the erythrocyte membrane (Yawata, Reference Yawata2003), it could be assumed that the RBC were more pliable. As a result, B. bovis might not be able to invade these pliable RBC efficiently. In addition, the in vitro replication test revealed that the 5 mm MCD treatment of the infected RBC also influenced the intraerythrocytic replication of B. bovis. Because the reduction in membrane cholesterol causes an increase in the permeability to H2O and other hydrophilic micromolecules (Deuticke and Ruska, Reference Deuticke and Ruska1976), the change in an intraerythrocytic environment might influence the replication of B. bovis. In P. falciparum and T. gondii, the cholesterol of the host cell membrane is known to be a crucial component of their parasitophorous vacuole membrane (Coppens and Joiner, Reference Coppens and Joiner2003; Lauer et al. Reference Lauer, Vanwye, Harrison, McManus, Samuel, Hiller, Mohandas and Haldar2000). In Babesia parasites, however, the parasitophorous vacuole exists only when parasites invade the RBC but then disappears relative to their growth (Igarashi et al. Reference Igarashi, Aikawa, Kreier and Ristic1988). Further study is required to understand the exact role of membrane cholesterol in the invasion and replication phases by Babesia parasites.
Interestingly, 1 mm MCD-treated RBC promoted both the invasion activity and growth of B. bovis. In the replication test, 1 mm MCD also had a positive effect on parasite replication although the difference from the control was not significant. One possible explanation for this discrepancy is that there is an ideal amount of membrane cholesterol, especially for B. bovis invasion to host RBC. Reduction of host membrane cholesterol is known to cause an improvement of erythrocyte deformability (Shiga et al. Reference Shiga, Maeda and Kon1990). Possibly, intact RBC without MCD treatment are too resistant from the viewpoint of erythrocyte deformability for efficient invasion by B. bovis. However, after treatment of the RBC with 1 mm MCD, the RBC achieved adequate hardness, which possibly made them suitable for invasion by the parasites. On the other hand, 3 or 5 mm MCD treatment is considered to make the RBC too pliable for efficient invasion. Efficient invasion of B. bovis would have a positive effect on the parasite's growth.
In conclusion, erythrocyte invasion and replication by B. bovis are influenced by the amount of cholesterol in the host erythrocyte membrane. These findings provide useful information to further understand the biological mechanism of the asexual growth of Babesia parasites.
This study was supported by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science, the Industrial Technology Research Grant Program from the New Energy and Industrial Technology Development Organization (NEDO) of Japan, Grants from Promotion of Basic Research Activities for Innovative Biosciences (PROBRAIN), and The 21st Century COE Program (A-1), Ministry of Education, Culture, Sports, Science, and Technology, Japan.
