Download - Wiley Online Library

FEMS Immunology and Medical Microbiology 26 (1999) 109^113
A comparison of the e¡ects of two dinitroanilines against
Cryptosporidium parvum in vitro and in vivo in neonatal mice
and rats
A. Armson *, K. Sargent, L.M. MacDonald, M.P. Finn, R.C.A. Thompson,
J.A. Reynoldson
Division of Veterinary and Biomedical Sciences, Murdoch University, Murdoch, W.A. 6150, Australia
Received 5 July 1999 ; accepted 14 July 1999
Abstract
The effects of two dinitroanilines, oryzalin and trifluralin, were compared against Cryptosporidium parvum, in vitro using
HCT-8 cells and in vivo using neonatal Swiss ARC mice and Wistar neonatal rats. In vitro, oryzalin and trifluralin exhibited
IC50 values (concentration necessary to cause a 50% inhibition) of 750 and 800 nM, respectively. A viability assay showed that
neither compound produced a cytotoxic effect on the host cells at concentrations as high as 1 WM. The in vivo component of
this study consisted of inoculation of neonatal mice and neonatal rats with 105 viable oocysts of C. parvum per animal and the
subsequent treatment of this infection with trifluralin and oryzalin administered via gastric intubation. At doses of 100 mg kg31
body weight administered twice daily for 3 consecutive days, trifluralin had no statistically significant effect on the number of
oocysts recovered from the gut of either rats or mice compared with controls, whereas at the same concentration, oryzalin
caused 90 and 79% inhibition of oocysts recovered from mice and rats, respectively. ß 1999 Federation of European
Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
1. Introduction
Cryptosporidium parvum is a widespread pathogen
infecting domestic and wild animals as well as humans [1]. The infection causes diarrhoeal disease
which is self limiting in immunocompetent hosts
but can be life threatening in immunocompromised
individuals such as those infected with human immunode¢ciency virus [2]. Cryptosporidiosis has been
recognised as a signi¢cant public health threat with
* Corresponding author. Fax: +61 (08) 9310 4144;
E-mail: [email protected]
the potential for waterborne outbreaks involving
large numbers of people [3] and is also of considerable signi¢cance to the cattle industry, where it is the
most common cause of neonatal calf diarrhoea and
associated calf morbidity [1,4]. At present, there is no
e¡ective treatment for cryptosporidiosis despite the
fact that many compounds have been tested for e¤cacy against this parasite either prophylactically or
as a means of therapy [5,6].
The dinitroanilines are tubulin-binding agents that
were originally recognised for their herbicidal properties. More recently, a number of in vitro studies
have compared the anti-cryptosporidial e¡ects of
several dinitroanilines as well as dinitroaniline ana-
0928-8244 / 99 / $20.00 ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 8 - 8 2 4 4 ( 9 9 ) 0 0 1 2 7 - 3
FEMSIM 1112 8-10-99
110
A. Armson et al. / FEMS Immunology and Medical Microbiology 26 (1999) 109^113
logues [7^9]. However, as far as we are aware, no
previous studies have examined the in vivo e¤cacy
of these compounds against C. parvum. In this study,
we compared the in vitro and in vivo anti-cryptosporidial e¡ects of two dinitroanilines, tri£uralin
and oryzalin, using neonatal mice and rats.
Brie£y, this method involves counting 30 ¢elds of
view which are randomly determined. The parasites
counted in the 30 ¢elds of view were summed and
the mean of duplicates was established. The percentage inhibition (Inh %) was determined by
Inh % ˆ
…total parasites in control3total parasites in drug-treated well†
total parasites in control
2. Materials and methods
2.1. Puri¢cation, in vivo ampli¢cation and isolation of
infective Cryptosporidium oocysts
A bovine Cryptosporidium isolate was collected
from calf faecal samples using phosphate-bu¡ered
saline (PBS)-ether sedimentation and was washed
as described by Morgan et al. [10].
In order to induce infectivity of the oocysts to cell
culture it is necessary to passage them in mice.
Therefore, the resuspended oocysts were inoculated
into 7^8-day old Swiss ARC mice via gastric intubation of 10^12U104 oocysts per mouse using the
method of Meloni and Thompson [11]. The mice
were killed on day 7 post-infection and the oocysts
processed as previously described [11]. Oocysts were
further puri¢ed using a Ficoll gradient and subsequently counted on a haemocytometer. Oocysts
were stored at 4³C in PBS containing ampicillin
and streptomycin.
2.2. Infection of HCT-8 cells with infective oocysts
and in vitro drug assay
HCT-8 cells were grown to con£uence in 24-well
plates in RPMI medium containing 10% foetal bovine serum. The cells were then rinsed in PBS, infected and then treated as described by Meloni and
Thompson [11]. The appropriate drugs were then
delivered in 5-Wl aliquots, in duplicate, dissolved in
dimethyl sulfoxide (DMSO) to a ¢nal concentration
of DMSO of no more than 0.5%. A range of drug
concentrations was used in order to determine the
concentration necessary to cause a 50% inhibition
(IC50 ) of infecting oocysts. The cells were incubated
for 72^120 h with constant exposure to the drugs or,
in the case of the controls, vehicle alone. Cryptosporidium oocysts were then counted using a modi¢cation of the method of Meloni and Thompson [11].
U100
2.3. Measurement of cytotoxicity of drugs
Cytotoxicity to host cells was determined using the
CytoTox 960 kit (Promega) which measures lactate
dehydrogenase as a guide to cell integrity. The assay
was performed according to the manufacturer's instructions and both positive and negative controls
were used. The assay was performed in quadruplicate.
2.4. In vivo analysis of drug activity against
Cryptosporidium
Stock solutions of drugs were made in 5-ml aliquots. Drugs were dissolved in 0.5 ml DMSO which
was then added to 4.5 ml peanut oil using a modi¢cation of the method described by Fayer [12]. Neonatal mice and rats were infected with the oocysts in
the manner described above. On day 4 post-infection, the animals were weighed and the mean weight
for each litter was calculated. Using the mean
weight, the dinitroaniline dose rate was determined
such that each mouse or rat received 100 mg kg31 of
drug in 100 Wl vehicle twice daily for 3 consecutive
days. Controls received vehicle alone. Both treated
and control mice and rats were killed on day 7 postinfection and processed in the manner described
above [10]. For both drug treatments or control,
three litters of either mice or rats were used and
each litter was divided into three replicate groups
giving a total of nine replicates. Mice replicate
groups contained 2^3 animals whereas rat replicate
groups contained two animals. Gut contents were
pooled for each replicate group and the oocyst
load was determined separately and average oocysts
FEMSIM 1112 8-10-99
A. Armson et al. / FEMS Immunology and Medical Microbiology 26 (1999) 109^113
111
Table 1
E¡ect of dinitroanilines in vitro and in vivo on neonatal mice and rats infected with C. parvum compared with untreated controls
Rat/mouse
Rat
Rat
Mouse
Mouse
Drug
Tri£uralin
Oryzalin
Tri£uralin
Oryzalin
Dose (mg kg31 )
100
100
100
100
mg
mg
mg
mg
31
kg
kg31
kg31
kg31
% Inhibition
IC50 value in vitro (nM)
Nil
79
Nil
90
800
750
Neonatal mice and rats were treated 4 days post-infection with either vehicle, oryzalin or tri£uralin at 100 mg kg31 , administered at 12-h
intervals (six doses). Oocysts recovered from treated and untreated animals were counted in order to determine the percentage inhibition.
E¤cacies of the compounds were determined in vitro using HCT-8 cells infected with C. parvum and the oocyst yield was compared between treated and untreated controls.
per mouse were calculated. The assignment of an
animal to a replicate group within the treated litter
was randomised using random numbers. Assays were
performed three times for each drug and each concentration used. Statistical analysis was performed
using a two-tailed paired t-test.
3. Results
3.1. In vitro studies
The mean IC50 values for tri£uralin and oryzalin
against Cryptosporidium in vitro were 800 and
750 nM, respectively (Table 1). At 800 nM, neither
oryzalin nor tri£uralin caused a detectable increase
in lactate dehydrogenase concentrations in the cell
media using the CytoTox 960 kit.
3.2. In vivo studies
Mean oocyst counts determined on treated and
untreated animals demonstrated no signi¢cant e¡ect
following tri£uralin treatment (six doses of 100 mg
kg31 ) for either mice or rats (P v 0.05). In contrast,
oryzalin caused 90% inhibition of oocyst numbers at
the same dose rate for mice (P 9 0.001) and 79%
inhibition for rats (Table 1) (P 9 0.005).
4. Discussion
A safe and e¡ective anti-cryptosporidial agent has
yet to be found. Previous studies have examined a
number of compounds [6,8]. Yet, the most promising
class of compounds remains the tubulin-binding di-
nitroanilines. Like the benzimidazoles, the selective
toxicity of the dinitroanilines is thought to be due to
amino acid di¡erences in tubulin structure between
the lower eukaryotes and mammals. These di¡erences are thought to account for the low mammalian
toxicity (LD50 of 0.5 g kg31 for tri£uralin orally in
1^2-day old rats increasing to 36.5 g kg31 in 56-day
old rats [13] and over 10 g kg31 orally for oryzalin in
adult rats) [14], making this class of compounds a
suitable starting point for the development of a treatment for cryptosporidiosis.
The IC50 values of less than 1 WM achieved in the
current in vitro study are similar to those previously
reported [7] and represent achievable, albeit transient, concentrations within the gastro-intestinal tract.
In fact, for an average neonatal mouse weighing 7 g,
a prepared dose of 100 mg oryzalin or tri£uralin per
kg body weight dissolved in 100 Wl solvent equals a
concentration of approximately 50 WM. The concentration of dinitroaniline will naturally become more
dilute in the gastro-intestinal tract.
The comparable e¤cacies of tri£uralin and oryzalin in vitro are very di¡erent to their e¡ects in vivo in
the strain of mice and rats used in this study. The
total absence of any in vivo anti-cryptosporidial effect by tri£uralin at 100 mg kg31 body weight is
disappointing and di¤cult to understand in view of
the excellent e¤cacy achieved with an identical concentration of oryzalin. However, previous studies revealed that tri£uralin was rapidly cleared in rats by
excretion in both faeces and urine [15]. Of the tri£uralin recovered from the faeces and urine in that
study, only 10% comprised unchanged tri£uralin,
suggesting that its metabolism may be the reason
for the di¡erences in the in vivo e¡ects of the two
drugs [15]. The rate of excretion was dose dependent
FEMSIM 1112 8-10-99
112
A. Armson et al. / FEMS Immunology and Medical Microbiology 26 (1999) 109^113
with 80% clearance in 48 h following an oral dose of
1 mg kg31 , whereas the rate decreased to 60% clearance over the same interval at 10 mg kg31 . It could
be reasonably postulated that the 100-mg kg31 dose
used in the current study would slow the rate of
clearance to an even greater extent. Oryzalin administered intraperitoneally to mice at 200 mg kg31
showed a peak plasma concentration of 25 Wg ml31
which decreased by half after 2.1 h and half again at
14.3 h. In contrast, when 300 mg kg31 was administered orally, the mean peak plasma level achieved
was 4 Wg ml31 , suggesting poor absorption or ¢rst
pass metabolism [14]. As with other gastro-intestinal
parasites con¢ned to the mucosa, poor absorption
may enhance rather than diminish e¤cacy. However,
caution must be exercised when comparing the metabolism of di¡erent drugs in rats and mice where
such di¡erences could be related more to animal
than structural di¡erences of the compounds [16].
The partition coe¤cient (log P) and water solubility
(ppm) values for oryzalin are 2.12 and 39 ppm, respectively, and for tri£uralin, they are 4.81 and 0.33
ppm, respectively. Taking into account the similarity
of their structures coupled with the greater water
solubility of oryzalin and its log P value of between
2^3, it is likely that oryzalin is more capable than
tri£uralin of free movement through eukaryotic cells.
Additionally, the milk diet of the neonatal mice and
rats used in the current study and/or the peanut oil
used as a vehicle for the administration of the compounds may act as a depot for tri£uralin, the less
polar compound, thereby hindering its activity. The
greater water solubility of oryzalin may prevent this
scenario from occurring. Nonetheless, the unusual
nature of Cryptosporidium in the host, being neither
extracellular nor entirely intracellular, does add a
further complication when predicting the e¡ects of
solubility on the e¤cacy of a compound. Williams
and Feil [17] examined the metabolism of tri£uralin
in a rumen microbial culture and found that polar
and non-polar metabolites were produced. Interestingly, only the polar metabolites were resistant to
acid and base hydrolysis and acetylation, but were
susceptible to reduction. Whether or not this resistance to hydrolysis and acetylation played a role in
the sustained e¤cacy of oryzalin in the present study
compared to tri£uralin is unknown and open to
speculation. At present, we are using HPLC to ex-
amine the metabolic changes that occur with these
drugs in the murine model. The di¡erence in e¤cacy
in vivo suggests that oryzalin rather than tri£uralin
has the greater potential and would therefore be a
better lead compound for the design of an e¡ective
anti-cryptosporidial agent.
Acknowledgements
The authors would like to acknowledge the technical assistance and advice given by Dr. Wayne Best
of the Chemistry Centre (WA). We would also like
to acknowledge the continuing ¢nancial support we
receive from SmithKline Beecham Pharmaceuticals
and the Australian Research Council (ARC) in the
form of a SPIRT research grant.
References
[1] Dubey, J.P., Speer, C.A. and Fayer, R. (1990) General biology of Cryptosporidium. In: Cryptosporidiosis of Man and
Animals (Dubey, J.P., Speer, C.A. and Fayer, R., Eds.), pp.
1^29. CRC Press, Boston, MA.
[2] Watzl, B., Huang, D.S., Alek, J., Darban, H., Jenkins, E.M.
and Watson, R.R. (1993) Enhancement of resistance to C
parvum by pooled bovine colostrum during murine retroviral
infection. Am. J. Trop. Med. Hyg. 48, 519^523.
[3] MacKenzie, W.R., Hoxie, N.J., Proctor, M.E., Gradus, S.,
Blair, K.A., Peterson, D.E., Kazmierczak, J.J., Addiss, D.J.,
Fox, K.R., Rose, J.P. and Davis, J.P. (1994) A massive outbreak in Milwaukee of Cryptosporidium infection transmitted
through the public water supply. New Engl. J. Med. 331, 161^
167.
[4] Xiao, L., Herd, R.P. and Rings, D.M. (1993) Concurrent infections of Giardia and Cryptosporidium in on two Ohio farms
with calf diarrhoea. Vet. Parasitol. 51, 41^48.
[5] Naciri, M., Mancassola, R., Rëpërant, J.M., Canivez, O.,
Quinque, B. and Yvorë, P. (1994) Treatment of experimental
ovine cryptosporidiosis with ovine or bovine hyperimmune
colostrum. Vet. Parasitol. 53, 173^190.
[6] Woods, K.M., Nesterenko, M.V. and Upton, S.J. (1996) E¤cacy of 101 antimicrobials and other agents on the development of C. parvum in vitro. Ann. Trop. Med. Parasitol. 90,
603^615.
[7] Arrowood, M.J., Mead, J.R., Xie, L.T. and You, X.D. (1996)
In vitro anticryptosporidial activity of dinitroaniline herbicides. FEMS. Microbiol. Lett. 136, 245^249.
[8] Armson, A., Meloni, B.P., Reynoldson, J.A. and Thompson,
R.C.A. (1999) Assessment of drugs against Cryptosporidium
parvum using a simple in vitro screening method. FEMS (in
press).
FEMSIM 1112 8-10-99
A. Armson et al. / FEMS Immunology and Medical Microbiology 26 (1999) 109^113
[9] Benbow, J.W., Bernberg, E.L., Korda, A. and Mead, J.R.
(1998) Synthesis and evaluation of dinitroanilines for treatment of cryptosporidiosis. Antimicrob. Agents Chemother.
42, 339^343.
[10] Morgan, U.M., Reynoldson, J.A. and Thompson, R.C.A.
(1993) Activities of several benzimidazoles and tubulin inhibitors against Giardia spp. in vitro. Antimicrob. Agents Chemother. 37, 328^331.
[11] Meloni, B.P. and Thompson, R.C.A. (1996) Simpli¢ed methods for obtaining puri¢ed oocysts from mice and for growing
Cryptosporidium parvum in vitro. J. Parasitol. 82, 757^762.
[12] Fayer, R. and Fetterer, R. (1995) Activity of benzimidazoles
against cryptosporidiosis in neonatal BALB/c mice. J. Parasitol. 81, 794^795.
[13] Budavari, S., O'Neil, M.J., Smith, A. and Heckelman, P.E.
(1989) The Merck Index, 11th edn. Merck, Rathway, NJ.
113
[14] Dvorakova, K., Dorr, R.T., Gallegos, A., McClure, T. and
Powis, G. (1997) Pharmacokinetic studies of the herbicide and
anti-tumor compound oryzalin in mice. J. Chromatogr. B.
696, 275^281.
[15] Erkog, F.U. and Menzer, R.E. (1985) Metabolism of tri£uralin in rats. J. Agric. Food Chem. 33, 1061^1070.
[16] Reynoldson, J.A., Thompson, R.C.A. and Meloni, B.P. (1991)
In vivo e¤cacy of albendazole against Giadia duodenalis in
mice. Parasitol. Res. 77, 325^328.
[17] Williams, P.P. and Feil, V.J. (1971) Identi¢cation of tri£uralin
metabolites from rumen microbial cultures. E¡ect of tri£uralin
on bacteria and protozoa. J. Agric. Food Chem. 19, 1198^
1204.
FEMSIM 1112 8-10-99