- Top of page
Calves undergoing initial infection with a virulent strain of the haemoprotozoan parasite Babesia bovis were treated with aminoguanidine (AG), an inhibitor of the inducible form of nitric oxide synthase (iNOS). The mean maximum parasitaemia of the AG treated calves was significantly lower than that of the control cattle. In addition, the febrile response and decrease in packed cell volume (PCV) observed during acute infection were significantly ameliorated in the AG treated cattle relative to the controls. However, AG had no effect on the multiplication of B. bovis in the microaerophilous stationary-phase (MASP) in-vitro culture system. These results provide evidence of a role for nitric oxide (NO) produced in response to acute infection in the pathology of bovine babesiosis.
- Top of page
The haemoprotozoan parasites Babesia bovis and Babesia bigemina, together with the rickettsia Anaplasma marginale, comprise the three causative agents of the tick fever disease complex in Australia. Of these three, B. bovis is considered to be the most pathogenic and causes significant economic losses in the cattle industries of tropical and sub-tropical regions ( McCosker 1981). Cattle acutely infected with B. bovis frequently exhibit severe pathology, including pyrexia, anaemia, organ failure, haemoglobinuria, respiratory and cerebral dysfunction (possibly due to sequestration of infected erythrocytes in capillaries) and hypotensive shock syndrome ( Wright & Kerr 1977).
Nitric oxide (NO) is produced by inducible nitric oxide synthase (iNOS) in macrophages during acute infection and has been shown to mediate resistance against several species of pathogen ( Liew et al. 1990 , Boockvar et al. 1994 , Chan et al. 1995 , Gosselin et al. 1995 , Petray et al. 1995 , Stevenson et al. 1995 ). NO has also been proposed to mediate infectious inflammation and to be immunosuppressive ( Eisenstein et al. 1994 , Mabbott et al. 1995 ). NO has been proposed as a mediator of human cerebral malaria and malaria-induced hypotension and immunosuppression ( Clark et al. 1991 , Rockett et al. 1994 ). The pathology of Plasmodium falciparum infection in humans is similar to that of B. bovis infection in cattle, including similar cerebral involvement and hypotension ( Wright et al. 1988 ). We have shown previously that administration of aminoguanidine (AG), a specific inhibitor of iNOS ( Griffiths et al. 1993 , Wolf & Lubeskie 1995) to calves undergoing acute infection with Anaplasma marginale resulted in decreased severity of infection ( Gale et al. 1997 ). Here we report the results of an investigation into the effect of AG on B. bovis infection in calves.
Mixed British breed dairy calves 6–8 months of age were obtained from a tick free area and were confirmed free of previous exposure to haemoparasites as described previously ( Gale et al. 1997 ). Calves were maintained under conditions which precluded accidental infection with tick-borne pathogens.
On day 0 of the experiment, six calves were infected with a virulent (Lismore) strain of B. bovis by the intravenous inoculation of 7.5 × 105 infected erythrocytes from a splenectomized calf which had been infected with B. bovis by tick challenge. At the same time as infection, AG treatment of three of the calves was commenced by the administration of the compound at a dose of 0.25% of the bicarbonate salt (Sigma Chemical Co.) ad libitum in drinking water. Fresh AG solution was administered daily from days 0 to 3 inclusive, after which calves received water only. The three remaining calves served as untreated controls. Blood samples were obtained daily from all calves. Parasite levels were determined by examination of Giemsa stained thick blood films ( Mahoney & Saal 1961). Plasma levels of AG were determined using a colorimetric assay as described previously ( Gale et al. 1997 ). Packed cell volume (PCV) and temperature values were recorded daily for all calves. Percent change from pre-challenge values were calculated for PCV and temperature absolute values. Data were analysed for significant differences between groups using Students t-test.
The mean daily plasma AG level ( Figure 1a) of the AG treated calves increased from day 1 to a daily mean maximum of 18.8 μg/ml on day 4 and decreased to zero by day 6 following withdrawal of AG on day 4. We have shown previously that plasma nitrate plus nitrite (the stable end-products of NO production) were not detectable using standard methods (nitrate reductase followed by Griess reagent assay) in normal bovine plasma (results not shown). No measurement of the effect of AG treatment on plasma nitrate plus nitrite was therefore attempted.
Figure 1. Effect of administration of AG on B. bovis infection in calves (n = 3). 1(a): Mean plasma AG concentration of treated calves versus days p.i. (b) (c) and (d): Mean daily parasite level, % PCV change and percent temperature change respectively, of AG treated (▪) and control (•) calves. Error bars represent ± 1 SD. For statistical significance of differences see text.
The mean daily parasitaemia of the AG treated calves was reduced relative to that of the controls ( Figure 1b) and this difference was statistically significant on day 7 (P = <0.05). The daily mean maximum parasite level of the AG treated calves was 26 per microlitre compared with 59 for the controls. The decrease in PCV observed in the AG treated calves was significantly less than that of the control calves for the period from day 3 to day 8 post infection (p.i.) ( Figure 1c, P = <0.05 on days 3, 4, 5, 7 & 8, P = <0.01 on day 6). The daily mean maximum PCV decrease in the AG calves was −29% compared with −42% in the controls. The most marked effect of AG treatment was evident in the reduced pyrexia of the AG treated calves from day 5 to day 8 compared to the control calves ( Figure 1d, P = <0.005 on days 5 to 8 inclusive). The daily mean maximum temperature change of the AG treated calves was +1.5% compared with +4.5% for the controls.
The microaerophilous stationary-phase (MASP) culture system was used to investigate whether AG had a direct effect on B. bovis multiplication in-vitro. The culture conditions used were essentially as described by Jorgensen et al. (1992 ) for Babesia bigemina. M199 medium was used with TES buffer (10 m M), 100 U/ml penicillin, 100 μg/ml streptomycin, 250 μg/ml Amphotericin B, 40% bovine serum and bovine erythrocytes to a final PCV of 4%. The B. bovis used was the ‘Anderson’ strain. Eight 1.2 ml culture replicates were set up in 24 well plates using 0, 2, 20 and 200 μg/ml of the free base of AG bicarbonate in the medium. Nine hundred μl of medium was replaced on days 1, 3, 5, 7 and 9. Thin films were prepared from the sedimented erythrocytes for determination of percentage infection rates on days 2, 4, 6, 8 and 10 prior to splitting of cultures (dilution with an equal volume of fresh uninfected erythrocytes in medium) on these days. The mean percent infection rates for the eight replicate cultures at the four AG concentrations are shown in Table 1. These results show that high parasite levels were maintained throughout the 10 day culture period at all AG concentrations and that there was no significant difference in the level of infected erythrocytes between groups after 10 days. We conclude that even at an AG concentration of 200 μg/ml (more than ten times the mean maximum plasma level attained in vivo), AG has no significant toxicity for B. bovis in-vitro. Similar results were obtained using the ‘Cunningham’ strain of B. bovis (results not shown).
Table 1. Effect of AG on in-vitro cultivation of B. bovis Parasitized erythrocytes were cultured for ten days in medium containing 0, 2, 20 or 200 μg/ml of the free base of AG bicarbonate. Percent infected erythrocytes in cultures (eight replicates) were determined by examination of Giemsa stained thick films on days 2, 4, 6, 8 and 10 prior to dilution of cultures 1 in 2 with fresh uninfected erythrocytes.
In summary, the effect of AG treatment on B. bovis infection in calves was to facilitate increased control of parasitaemia to some extent and to markedly ameliorate two pathological symptoms of babesiosis (pyrexia and anaemia). These effects occurred 3–8 days after infection of cattle, although plasma AG levels were low or undetectable from day 6 p.i. onwards due to cessation of treatment on day 4. No direct in-vivo anti-parasite toxicity of AG has been reported to our knowledge and we were unable to demonstrate toxicity of AG for B. bovis in vitro.
The maintenance of higher PCV values in the AG-treated calves may be due at least in part, to the removal of the inhibitory effect of NO on erythropoiesis ( Tsukahara et al. 1997 ). An additional factor may be a reduction in the hypotension-induced ‘pooling or stasis’ of erythrocytes in the organs and subsequent fall in PCV in the circulation ( Wright 1979). None of the calves in either group displayed evidence of haemolytic anaemia (e.g. jaundice or haemoglobinuria). The effect of AG in the reduction of B. bovis-induced pyrexia in the treated calves was consistent with the observations that inhibition of NO production results in hypothermia and depresses lipopolysaccharide-induced fever ( Scammell et al. 1996 ). None of the calves in this experiment showed overt symptoms of neurological dysfunction which are more often observed in older cattle infected with B. bovis.
As discussed above, although nitric oxide plays an important role in the control of many intracellular pathogens, especially those inhabiting cells of monocyte lineage, overproduction of NO via iNOS in these cells is also likely to be involved in host inflammatory disease, hypotension and infection-related immunosuppression. Our observations on the effect of AG on babesiosis are consistent with the hypothesis that NO is not protective against B. bovis infection and that inhibition of NO overproduction leads to decreased inflammation, immunosuppression and severity of disease. Inhibition of NO synthesis was also shown to lead to a reduced parasitaemia during Trypanosoma infection in mice ( Sternberg et al. 1994 ). These authors have postulated that NO may not be toxic for some haemoparasites due to the neutralisation of NO by oxyhaemoglobin ( Mabbott et al. 1994 ). It is interesting to note however, that in a murine Babesia model, treatment of mice with the immunomodulator AS101 ( Sredni et al. 1987 ), resulted in the mitigation of the course of Babesia rodhaini infection and increased survival via the proposed mechanism of increased NO production ( Rosenblatt-Bin et al. 1996 ). Clearly, the protective or detrimental effects of NO produced during acute infection vary according to many parameters, including the parasite and host species under investigation. It should also be noted that Johnson et al. (1996 ) reported that at least in-vitro, NO (chemically derived from sodium nitroprusside or S-nitroso-N-acetyl-penicillamine) was toxic for B. bovis. It is not known at this time how the levels of NO which were toxic in vitro compare with physiological levels produced in-vivo in cattle.
Prevention of losses due to B. bovis infection is currently achieved largely via the use of a vaccine strain of B. bovis, significantly attenuated by passage in splenectomized calves ( Callow et al. 1979 ). Vaccination using live tick fever parasites results in severe clinical disease in a proportion (e.g. 2%) of vaccinated cattle and is not recommended for highly susceptible cattle (i.e. mature or pregnant animals). Long-acting oxytetracycline has been used experimentally to moderate clinical reactions to vaccination with live B. bovis ( Pipano et al. 1987 , Jorgensen et al. 1993 ) and prior vaccination with Babesia culture supernatant has been reported to give some protection against anaemia due to subsequent Babesia infection ( Jorgensen et al. 1993 ). Prior treatment with long-acting oxytretracycline is contraindicated however when A. centrale is included in the live vaccine (to protect against bovine anaplasmosis) because such treatment inhibits the establishment of Anaplasma infection ( Jorgensen et al. 1993 ). The AG treatment described here is the first report of anti-babesial activity using this class of compound, by the novel proposed mechanism of host immune response modification rather than direct anti-parasite effect. Administration of AG also moderates clinical anaplasmosis without blocking the establishment of infection ( Gale et al. 1997 ) and may therefore have practical applications in the moderation of clinical reactions to live tick fever vaccination of highly susceptible cattle.