- Email: [email protected]
Comparison of three in vitro techniques to estimate benzimidazole resistance in Haemonchus contortus of sheep
Veterinary Parasitology, 34 (1989) 213-221 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands
213
Comparison of Three In Vitro Techniques to Estimate B e n z i m i d a z o l e Resistance in H a e m o n c h u s c o n t o r t u s of Sheep MARIA V. JOHANSEN 1 and P E T E R J. WALLER 2
1Institute of Hygiene and Microbiology, Royal Veterinary and Agricultural University, Bi~lowsvej 13, DK-1870 Frederiksberg C (Denmark) 2CSIRO Division of Animal Health, McMaster Laboratory, Private Bag No. 1, Glebe, N.S. W. 2037 (Australia) (Accepted for publication 30 March 1989)
ABSTRACT Johansen, M.V. and Waller, P.J., 1989. Comparison of three in vitro techniques to estimate benzimidazole resistance in Haemonchus contortus of sheep. Vet. Parasitol., 34: 213-221. Three in vitro assays to detect benzimidazole resistance, namely, the egg-hatch assay, tubulinbinding assay, and a larval-development assay, were evaluated by estimating the level of benzimidazole resistance in three field isolates of Haemonchus contortus compared with a susceptible reference strain. Comparisons were also made with estimates of benzimidazole resistance of the three field strains obtained from an in vivo controlled anthelmintic efficacy test. All three in vitro tests showed similar, consistent results which also suggested greater sensitivity than the in vivo assay. These results indicate that selection of an in vitro technique to determine benzimidazole resistance should therefore be based on considerations other than precision, such as technical expertise, availability of equipment, cost and speed in which diagnosis is required.
INTRODUCTION
The greatest problem with anthelmintic resistance in nematode parasites of livestock is associated with the benzimidazole group of anthelmintics. Reports of resistance to this class of compounds have been made in parasites of sheep and goats virtually in every country where it has been specifically investigated (for review see Waller, 1986). The problem has reached alarming proportions in the high summer rainfall regions of the world, where the abomasal parasite Haemonchus contortus is endemic. However, other important nematode species of sheep and goats, namely Ostertagia, Trichostrongylus and Nematodirus spp. now show substantial degrees of resistance to the benzimidazoles (Waller, 1986). In addition, the prevalence of benzimidazole resistance in cyathostome parasites of horses is widespread (for review, see Bauer et al., 1986) and to add 0304-4017/89/$03.50
© 1989 Elsevier Science Publishers B.V.
214
M.V. JOHANSEN AND P.J. WALLER
to this concern, benzimidazole resistance has also recently been reported in nematode parasites of cattle (Eagleson and Bowie, 1986; Jackson et al., 1987 ). To aid in the diagnosis of anthelmintic resistance, a range of in vivo and in vitro techniques have been developed. These have varying attributes and limitations depending on whether they are used for field diagnosis or specific research purposes (for review, see Johansen, 1989). The aim of this study was to compare two of the most commonly used in vitro techniques to detect benzimidazole resistance, namely the egg hatch and the tubulin-binding assay, together with a recently developed larval-development assay, in the detection of benzimidazole resistance in H. contortus of sheep. Comparisons were made between three field strains and a standard reference strain of H. contortus, and the results of the in vitro assays were compared with an in vivo anthelmintic efficacy study. MATERIALSAND METHODS Parasite strains
Four different strains of H. contortus, one reference and three field strains, were used in all assays. The reference strain (McM) was isolated some 30 years ago, prior to the commercial release of the benzimidazole anthelmintics. This strain has since been maintained in the laboratory by serial passage in wormfree sheep. The three field strains were derived from a sheep-grazing experiment conducted at the CSIRO McMaster Field Station, Badgery's Creek, N.S.W. For detailed background history and methods of isolation of these strains, see Waller et al. (1989). In summary, the experiment commenced in January 1982 on pasture contaminated with H. contortus having a high level of resistance to thiabendazole derived from treatment with this drug 9 years previously. During the 5 years of the experiment, there were separate groups of sheep treated eight times per year with levamisole (8 LEV), eight times per year with ivermectin (8 IVM) and three times per year with thiabendazole and moved to clean pasture (3 TBZ/move). Pure strains of H. contortus were isolated from each of these treatments in December 1986. Techniques
An in vivo test to estimate TBZ resistance was carried out at the time of isolation of the field strains and reported in detail elsewhere (Waller et al., 1989). All in vitro tests were conducted according to the methods described in the literature. Assays of the three techniques were carried out on a minimum of two occasions, either on the same day or consecutive days, to estimate the variation between assays. For the larval-development and egg-hatch assays, H.
BENZIMIDAZOLE RESISTANCE IN HAEMONCHUS CONTORTUS
215
contortus eggs were recovered from faeces of sheep infected with each strain using the technique described by Dobson et al. (1986). Larval-development assay (E. Lacey, J.M. Redwin and P.J. Waller, personal communication, 1988) Approximately 100 eggs were added to each well of a flat-bottomed microtitration plate containing serial dilutions of TBZ incorporated into agar. The plate was incubated for 7 days at 26 °C and after this time the assay was terminated by the addition of a few drops of iodine to each well. The number of eggs, first-, second- and third-stage larvae were counted, and the percentage of eggs that failed to hatch or reach the third larval stage was estimated. Development through to the infective larval stage is much more sensitive to drug concentration than the hatching of eggs (Waller and Lacey, 1985 ). Estimated ECso concentrations are ~ 10-fold greater for egg-hatch compared with larvaldevelopment within individual assays. In this study, the assay was repeated on two occasions.
Egg-hatch assay (Donald et al., 1980) The procedures used were a modification of the technique described by Le Jambre (1976). Eggs were incubated for 48 h at 27 ° C in serial dilutions of TBZ solubilised in dimethylsulphoxide (DMSO) and contained in wells of a microtitration plate. The number of hatched first-stage larvae were then estimated. The assay was replicated twice at each drug concentration and was repeated on three occasions.
Tubulin-binding assay (Lacey and Snowdon, 1988) Approximately 100 000 infective larvae, derived from faecal cultures from each sheep were used in this assay. Crude extracts of tubulin were prepared by homogenisation of the larvae and incubated with tritiated mebendazole MBZ until equilibrium was reached. The amount of bound drug was then estimated by liquid scintillation spectrophotometry and expressed as pmol mg-1 of worm protein. Tubulin extracts from resistant strains bind substantially less drug than those from susceptible strains. The assay was replicated three times and conducted on two occasions.
Statistical procedures For both the larval-development and egg-hatch assays, the results were subject to logit analysis (Waller et al., 1985) to estimate the TBZ concentration which, on average, would prevent 50% of eggs hatching (ECso). Also, in the case of the larval-development assay, estimates were made of the drug concentration which would prevent 50% of eggs reaching the third larval stage. The variance (s 2) of the ECso for each assay and strain was estimated from the fit
216
M.V. JOHANSEN AND P.J. WALLER
of the logit model to the data, according to the procedures of Dobson et al. (1987). Resistance ratios (RR) for the three test lines were calculated as differences on the loge scale and then transformed to resistance ratios as follows:
RRT = exp (ln EC~o~- In EC~oM) where EC~oT is the test strain and ECsoM is the McM reference strain, EC~os. The 95% confidence interval (CI) for resistance ratios was determined using the following formula: 95% CI=exp [In (RRT) +_2 (X/~M +S~)] where s ~ and s~ are the variances of the ECsos on the log scale for the McM and test strain, respectively. For the tubulin-binding assay, the bound radioactivity was standardised to pmol mg- 1 of protein for each isolate. The extent of resistance for a test strain was expressed as a susceptibility factor (SFT), calculated as the ratio of the test isolate binding (BT) over the susceptible binding (BM), using the McM strain as a standard for susceptibility. The variance (vat) of SFT was determined by the formula of Kendall and Stuart (1977):
2~ var BT var SM ) var SFT = (SET) \ S ~ t B ~ A pooled estimate for the variance of the amount of bound drug over strains and times was determined from the analysis of variance of the binding data. The 95% confidence interval for SFT was estimated as: 95% CI=SFT +2 x//var SFw RESULTS Table I shows the results of an in vivo dose-and-slaughter assay of TBZ at the recommended (44 mg kg -1) and twice the recommended (88 mg kg -~) dose rate for the three field lines tested at the time of isolation by Waller et al. (1989). Both the 8 LEV and 8 IVM isolates showed high levels of TBZ resistance with only 61% efficacy at 44 mg kg -1 TBZ, despite the fact that this drug had not been used for 14 years. TBZ showed only a 5% efficacy at the recommended dose rate against the 3 TBZ per move isolate indicating that a much higher level of resistance had been selected for in this treatment. Results of the in vitro assays are shown in Tables 2, 3 and 4 for the larvaldevelopment, egg-hatch and tubulin-binding assays, respectively. Except for the results of egg hatch on Day i of the larval-development assay, all three assays showed consistency, both in comparison with each other and
BENZIMIDAZOLERESISTANCEIN HAEMONCHUS CONTORTUS
217
TABLE 1 Mean worm numbers of three field strains of H. contortus {with percentage reduction in parentheses ) from an in vivo assay with thiabendazole (Waller et al., 1989 ) TBZ dose rate (mg kg- 1)
Strains
8 LEV 0 44 88
1210 (61) (81)
3 TBZ per move
8 IVM
525 (5) (77)
950 (61) (79)
~Five sheep per group. TABLE2 Estimates from larval-development assays of thiabendazole concentrations which, on average, would prevent 50% of eggs hatching or larvae developing to the third stage for a benzimidazolesusceptible isolate (McM) of H. contortus, and the resistance ratios of three field isolates of this species Stage
Eggs
Assay No. 1 2
Larvae
1 2
ECho1 (95% limits) McM
Resistance ratios e (95% limits) 8 LEV
3 TBZ per move
8 IVM
0.499 (0.456-0.546) 0.629 (0.595-0.666)
4.9 (4.3-5.6) 2.4 (2.0-2.8)
3.3 (2.9-3.8) 5.6 (5.1-6.2)
2.4 (2.1-2.7) 1.8 (1.6-2.0)
0.0468 (0.0404-0.0543) 0.0744 (0.0719-0.0770)
5.7 (4.7-7.0) 4.3 (3.8-4.9)
10.5 (8.9-12.4) 9.3 (8.5-10.1)
3.4 (2.9-4.2) 2.3 (2.0-2.6)
1ECso=ECsothiabendazole concentration (/IM ml-1). 2Resistance ratios--ECso test strain/ECs0 McM.
compared with the in vivo assay. Estimates of resistance (resistance ratios and susceptibility factors) showed that the 3 TBZ per move was the most resistant isolate followed by the 8 LEV which was more resistant than the 8 IVM isolate. For the egg-hatch assay, the standard errors of ECsos between lines were very small and particularly consistent within each assay, with a standard error estimated at 0.0786 by analysis of variance. For the larval-development assay, no estimates of standard errors within assays were determined because no replication was carried out. However, this assay was shown to be similar to the egg-hatch assay in being highly accurate, with very low standard errors for the fit of the data to the model, and repeatable. The range of the 95% confidence limits for both of these procedures was sufficiently narrow for all tests to show
218
M.V. JOHANSENAND P.J. WALLER
TABLE3 Estimates from egg-hatch assays of thiabendazole concentrations which, on average, would prevent 50% of eggs from hatching for a benzimidazole-susceptible isolate (McM) of H. contortus, and the resistance ratios of three field isolates of this species Assay No. 1 2 3
ECso 1 (95% limits) McM
8 LEV
3 TBZ per move
8 IVM
0.245 (0.235-0.255) 0.069 (0.062 0.077) 0.076 (0.070-0.081)
2.8 (2.6 3,0) 5.2 (4.3-6,1) 5.6 (4.9 6.4)
3.3 (3.1 3.5) 9.3 (7.8-11.0) 9.9 (9.0 11.0)
1.4 (1.3-1.5) 3.4 (2.9-4.1) 2.4 (2.1-2.7)
Resistance ratios 2 {95% limits)
~ECso = EC~0 thiabendazole concentration (#M m l - 1). 2Resistance ratio = ECho test strain/ECho McM. TABLE4 Calculated binding and susceptibility factors derived from the incubation of [3H] mebendazole to crude supernatants of a benzimidazole-susceptible isolate (McM) of H. contortus and three field isolates of this species Assay No.
1 2 Waller et al. (1989)
Bound TBZ ~ (95% limits) McM
Susceptibility factors 2 (95% limits) 8 LEV
3 TBZ per move
8 IVM
114 (108 120) 92 (86-96)
0.53 (0.42-0.63) 0.52 (0.40-0.65)
0.37 (0.27 0.47) 0.48 (0.36-0.61)
0.66 (0.55-0.77) 0.83 (0.69-0.98)
100 (94-108)
0.68 (0.61-0.75)
0.33 (0.26 0.40)
0.74 (0.67-0.81)
1Bound TBZ = Bound mebendazole pmoles m g - 1 protein. 2Susceptibility factors = Bound MBZ, test strain/Bound MBZ, McM.
the level of benzimidazole resistance in the 3 TBZ per move isolate to be significantly higher than both the 8 LEV and 8 IVM isolates and the 8 LEV also to be significantly higher than 8 IVM. Although the tubulin-binding assay discriminated between the isolates, the width of the 95 % confidence interval were greater than the other two techniques, so there was overlap in the susceptibility factor estimates especially between 8 LEV and 8 IVM isolates. DISCUSSION
All three in vitro techniques gave consistent and comparable results. They discriminated between the field isolates of H. contortus, with the TBZ selected
BENZIMIDAZOLE RESISTANCE IN HAEMONCHUS CONTORTUS
219
strain (3 T B Z / m o v e ) showing the greatest degree ofbenzimidazole resistance, followed by the 8 LEV isolate and then the 8 IVM strain. These differences were shown to be significant in both the egg-hatch and larval-development assay. Although the in vivo assay showed that all of these isolates had a high level of resistance to TBZ, this assay was unable to differentiate between the isolates as observed for the in vitro procedures. This suggests that the in vitro techniques are more sensitive than the in vivo controlled anthelmintic efficiency test which requires considerable resources such as sheep (in this case, 45 lambs), pen accommodation and feed for at least 4 weeks, and considerable technical labour to conduct worm counts. All three in vitro assays produced highly satisfactory results, with sensitive comparisons between strains and within assays. However, all showed relatively large between-day variation, possibly attributable to variations in ambient or incubator temperature, drug preparation or egg recovery indicating the desirability to incorporate a reference control strain each time an assay is conducted. Consistency of the results is also shown by the independent estimates of benzimidazole resistance of the three field isolates by the tubulin-binding assay (Waller et al., 1989 ), which were in accord with our observations (Table 4). The only aberrant result was in the number of eggs that hatched between isolates for the first larval-developmeot assay. This may be attributed to the fact that for this assay, compared with!the egg-hatch assay, no special precautions are required to obtain eggs at a uniform early stage of development because the assay was specifically developed to discriminate between test isolates by estimating the number of eggs that succeed in developing to the third larval stage. In fact, this assay was shown to provide the correct result when development was allowed to proceed and estimates of third-stage larvae were made. It is of interest to note the much lower EC~0 determined by the egg-hatch assay compared with the ECso of the eggs in the larval-development assay. This is in accord with the observations by Lacey et al. (1987) who showed that the action of benzimidazoles in the egg-hatch assay is principally ovistatic, whereby the drugs exert an inhibitory effect on development resulting in more eggs hatching with increasing time of incubation. The egg-hatch assay is a relatively cheap, simple and fast procedure, with results obtained within 3 days. Although the tubulin-binding assay can be used on all stages of development, from eggs to adult parasites, it has been developed using mainly infective larvae (Lacey and Snowdon, 1988 ). Therefore to obtain a sufficient quantity of parasite material for this assay (100 000 L3 were used in this investigation), a minimum culture time of 7 days is required. Tubulinbinding assays also require access to expensive laboratory apparatus for high performance liquid chromatography estimations, and a source of radiolabelled drug. The larval-development assay has a great advantage, not available with the former techniques, of being able to be used to test simultaneously a c o m -
220
M.V.JOHANSENANDP.J. WALLER
plete range of both broad- and narrow-spectrum anthelmintics. However, the technique requires possibly greater technical expertise than the other in vitro procedures and more time in counting the assay, particularly if both ovicidal and larvacidal effects are to be estimated. The decision on which in vitro procedure to use to detect benzimidazole resistance is one of personal preference and is influenced by such considerations as the availability of equipment, speed with which results are required, and available technical skills rather than precision which is similar for the three techniques investigated here. ACKNOWLEDGEMENT
We are grateful to R.J. Dobson for statistical advice.
REFERENCES Bauer, C., Merkt, J.C., Janke-Grimm, G. and Btirger, H.-J., 1986. Prevalence and control of benzimidazole-resistant small strongyles on German Thoroughbred studs. Vet. Parasitol., 21:189203. Dobson, R.J., Donald, A.D., Waller, P.J. and Snowdon, K.L., 1986. An egg-hatch assay for resistance to levamisole in trichostrongylid nematode parasites. Vet. Parasitol., 19: 77-84. Dobson, R.J., Griffiths, D.A., Donald, A.D. and Waller, P.J., 1987. A genetic model describing the evolution of levamisole resistance in Trichostrongylus colubriformis, a nematode parasite of sheep. IMA J. Math. Appl. Med. Biol., 4: 279-293. Donald, A.D., Waller, P.J., Dobson, R.J. and Axelsen, A., 1980. The effect of selection with levamisole on benzimidazole resistance in Ostertagia spp. of sheep. Int. J. Parasitol., 10:381 389. Eagleson, J.S. and Bowie, J.Y., 1986. Oxfendazole resistance in Trichostrongylus axei in cattle in Australia. Vet. Rec., 119: 604. Jackson, R.A., Townsend, K.G., Pyke, C. and Lance, D.M., 1987. Isolation of oxfendazole resistant Cooperia oncophora in cattle. N.Z. Vet. J., 35: 187-189. Johansen, M.V., 1989. Review of techniques used for the detection of anthelmintic resistance in nematode parasites of domestic livestock. Vet. Res. Commun., in press. Kendall, M.K. and Stuart, A., 1977. The Advanced Theory of Statistics. Vol. 1. Charles Griffin, London, p. 247. Lacey, E. and Snowdon, K.L., 1988. A routine diagnostic assay for the detection of benzimidazole resistance in parasitic nematodes using tritiated benzimidazole carbamates. Vet. Parasitol., 27: 309-324. Lacey, E., Brady, R.L., Prichard, R.K. and Watson, T.R., 1987. Comparison of inhibition of polymerisation of mammalian tubulin and helminth ovicidal activity by benzimidazole carbamates. Vet. Parasitol., 23: 105-119. Le Jambre, L,F., 1976. Egg-hatch as an in vitro assay of thiabendazole resistance in nematodes. Vet. Parasitol., 2: 385-391. Waller, P.J., 1986. Anthelmintic resistance in nematode parasites of sheep. Agric. Zool. Rev., 1: 333 373. Waller, P.J. and Lacey, E., 1985. Nematode growth regulators. In: N. Anderson and P.J. Waller (Editors), Resistance to Anthelmintic Drugs. CSIRO, Melbourne, pp. 137 148. Waller, P.J., Dobson, R.J., Donald, A.D., Griffiths, D.A. and Smith, E.F., 1985. Selection studies
BENZIMIDAZOLE RESISTANCE IN HAEMONCHUS CONTORTUS
221
on anthelmintic resistant and susceptible populations of Trichostrongylus colubriformis of sheep. Int. J. Parasitol., 15: 669-676. Waller, P.J., Donald, A.D., Dobson, R.J., Lacey, E., Hennessy, D.R. and Prichard, R.K., 1989. Changes in anthelmintic resistance status of Haemonchus contortus and Trichostrongylus colubriformis exposed to different anthelmintic selection pressures in grazing sheep. Int. J. Parsitol., in press.
213
Comparison of Three In Vitro Techniques to Estimate B e n z i m i d a z o l e Resistance in H a e m o n c h u s c o n t o r t u s of Sheep MARIA V. JOHANSEN 1 and P E T E R J. WALLER 2
1Institute of Hygiene and Microbiology, Royal Veterinary and Agricultural University, Bi~lowsvej 13, DK-1870 Frederiksberg C (Denmark) 2CSIRO Division of Animal Health, McMaster Laboratory, Private Bag No. 1, Glebe, N.S. W. 2037 (Australia) (Accepted for publication 30 March 1989)
ABSTRACT Johansen, M.V. and Waller, P.J., 1989. Comparison of three in vitro techniques to estimate benzimidazole resistance in Haemonchus contortus of sheep. Vet. Parasitol., 34: 213-221. Three in vitro assays to detect benzimidazole resistance, namely, the egg-hatch assay, tubulinbinding assay, and a larval-development assay, were evaluated by estimating the level of benzimidazole resistance in three field isolates of Haemonchus contortus compared with a susceptible reference strain. Comparisons were also made with estimates of benzimidazole resistance of the three field strains obtained from an in vivo controlled anthelmintic efficacy test. All three in vitro tests showed similar, consistent results which also suggested greater sensitivity than the in vivo assay. These results indicate that selection of an in vitro technique to determine benzimidazole resistance should therefore be based on considerations other than precision, such as technical expertise, availability of equipment, cost and speed in which diagnosis is required.
INTRODUCTION
The greatest problem with anthelmintic resistance in nematode parasites of livestock is associated with the benzimidazole group of anthelmintics. Reports of resistance to this class of compounds have been made in parasites of sheep and goats virtually in every country where it has been specifically investigated (for review see Waller, 1986). The problem has reached alarming proportions in the high summer rainfall regions of the world, where the abomasal parasite Haemonchus contortus is endemic. However, other important nematode species of sheep and goats, namely Ostertagia, Trichostrongylus and Nematodirus spp. now show substantial degrees of resistance to the benzimidazoles (Waller, 1986). In addition, the prevalence of benzimidazole resistance in cyathostome parasites of horses is widespread (for review, see Bauer et al., 1986) and to add 0304-4017/89/$03.50
© 1989 Elsevier Science Publishers B.V.
214
M.V. JOHANSEN AND P.J. WALLER
to this concern, benzimidazole resistance has also recently been reported in nematode parasites of cattle (Eagleson and Bowie, 1986; Jackson et al., 1987 ). To aid in the diagnosis of anthelmintic resistance, a range of in vivo and in vitro techniques have been developed. These have varying attributes and limitations depending on whether they are used for field diagnosis or specific research purposes (for review, see Johansen, 1989). The aim of this study was to compare two of the most commonly used in vitro techniques to detect benzimidazole resistance, namely the egg hatch and the tubulin-binding assay, together with a recently developed larval-development assay, in the detection of benzimidazole resistance in H. contortus of sheep. Comparisons were made between three field strains and a standard reference strain of H. contortus, and the results of the in vitro assays were compared with an in vivo anthelmintic efficacy study. MATERIALSAND METHODS Parasite strains
Four different strains of H. contortus, one reference and three field strains, were used in all assays. The reference strain (McM) was isolated some 30 years ago, prior to the commercial release of the benzimidazole anthelmintics. This strain has since been maintained in the laboratory by serial passage in wormfree sheep. The three field strains were derived from a sheep-grazing experiment conducted at the CSIRO McMaster Field Station, Badgery's Creek, N.S.W. For detailed background history and methods of isolation of these strains, see Waller et al. (1989). In summary, the experiment commenced in January 1982 on pasture contaminated with H. contortus having a high level of resistance to thiabendazole derived from treatment with this drug 9 years previously. During the 5 years of the experiment, there were separate groups of sheep treated eight times per year with levamisole (8 LEV), eight times per year with ivermectin (8 IVM) and three times per year with thiabendazole and moved to clean pasture (3 TBZ/move). Pure strains of H. contortus were isolated from each of these treatments in December 1986. Techniques
An in vivo test to estimate TBZ resistance was carried out at the time of isolation of the field strains and reported in detail elsewhere (Waller et al., 1989). All in vitro tests were conducted according to the methods described in the literature. Assays of the three techniques were carried out on a minimum of two occasions, either on the same day or consecutive days, to estimate the variation between assays. For the larval-development and egg-hatch assays, H.
BENZIMIDAZOLE RESISTANCE IN HAEMONCHUS CONTORTUS
215
contortus eggs were recovered from faeces of sheep infected with each strain using the technique described by Dobson et al. (1986). Larval-development assay (E. Lacey, J.M. Redwin and P.J. Waller, personal communication, 1988) Approximately 100 eggs were added to each well of a flat-bottomed microtitration plate containing serial dilutions of TBZ incorporated into agar. The plate was incubated for 7 days at 26 °C and after this time the assay was terminated by the addition of a few drops of iodine to each well. The number of eggs, first-, second- and third-stage larvae were counted, and the percentage of eggs that failed to hatch or reach the third larval stage was estimated. Development through to the infective larval stage is much more sensitive to drug concentration than the hatching of eggs (Waller and Lacey, 1985 ). Estimated ECso concentrations are ~ 10-fold greater for egg-hatch compared with larvaldevelopment within individual assays. In this study, the assay was repeated on two occasions.
Egg-hatch assay (Donald et al., 1980) The procedures used were a modification of the technique described by Le Jambre (1976). Eggs were incubated for 48 h at 27 ° C in serial dilutions of TBZ solubilised in dimethylsulphoxide (DMSO) and contained in wells of a microtitration plate. The number of hatched first-stage larvae were then estimated. The assay was replicated twice at each drug concentration and was repeated on three occasions.
Tubulin-binding assay (Lacey and Snowdon, 1988) Approximately 100 000 infective larvae, derived from faecal cultures from each sheep were used in this assay. Crude extracts of tubulin were prepared by homogenisation of the larvae and incubated with tritiated mebendazole MBZ until equilibrium was reached. The amount of bound drug was then estimated by liquid scintillation spectrophotometry and expressed as pmol mg-1 of worm protein. Tubulin extracts from resistant strains bind substantially less drug than those from susceptible strains. The assay was replicated three times and conducted on two occasions.
Statistical procedures For both the larval-development and egg-hatch assays, the results were subject to logit analysis (Waller et al., 1985) to estimate the TBZ concentration which, on average, would prevent 50% of eggs hatching (ECso). Also, in the case of the larval-development assay, estimates were made of the drug concentration which would prevent 50% of eggs reaching the third larval stage. The variance (s 2) of the ECso for each assay and strain was estimated from the fit
216
M.V. JOHANSEN AND P.J. WALLER
of the logit model to the data, according to the procedures of Dobson et al. (1987). Resistance ratios (RR) for the three test lines were calculated as differences on the loge scale and then transformed to resistance ratios as follows:
RRT = exp (ln EC~o~- In EC~oM) where EC~oT is the test strain and ECsoM is the McM reference strain, EC~os. The 95% confidence interval (CI) for resistance ratios was determined using the following formula: 95% CI=exp [In (RRT) +_2 (X/~M +S~)] where s ~ and s~ are the variances of the ECsos on the log scale for the McM and test strain, respectively. For the tubulin-binding assay, the bound radioactivity was standardised to pmol mg- 1 of protein for each isolate. The extent of resistance for a test strain was expressed as a susceptibility factor (SFT), calculated as the ratio of the test isolate binding (BT) over the susceptible binding (BM), using the McM strain as a standard for susceptibility. The variance (vat) of SFT was determined by the formula of Kendall and Stuart (1977):
2~ var BT var SM ) var SFT = (SET) \ S ~ t B ~ A pooled estimate for the variance of the amount of bound drug over strains and times was determined from the analysis of variance of the binding data. The 95% confidence interval for SFT was estimated as: 95% CI=SFT +2 x//var SFw RESULTS Table I shows the results of an in vivo dose-and-slaughter assay of TBZ at the recommended (44 mg kg -1) and twice the recommended (88 mg kg -~) dose rate for the three field lines tested at the time of isolation by Waller et al. (1989). Both the 8 LEV and 8 IVM isolates showed high levels of TBZ resistance with only 61% efficacy at 44 mg kg -1 TBZ, despite the fact that this drug had not been used for 14 years. TBZ showed only a 5% efficacy at the recommended dose rate against the 3 TBZ per move isolate indicating that a much higher level of resistance had been selected for in this treatment. Results of the in vitro assays are shown in Tables 2, 3 and 4 for the larvaldevelopment, egg-hatch and tubulin-binding assays, respectively. Except for the results of egg hatch on Day i of the larval-development assay, all three assays showed consistency, both in comparison with each other and
BENZIMIDAZOLERESISTANCEIN HAEMONCHUS CONTORTUS
217
TABLE 1 Mean worm numbers of three field strains of H. contortus {with percentage reduction in parentheses ) from an in vivo assay with thiabendazole (Waller et al., 1989 ) TBZ dose rate (mg kg- 1)
Strains
8 LEV 0 44 88
1210 (61) (81)
3 TBZ per move
8 IVM
525 (5) (77)
950 (61) (79)
~Five sheep per group. TABLE2 Estimates from larval-development assays of thiabendazole concentrations which, on average, would prevent 50% of eggs hatching or larvae developing to the third stage for a benzimidazolesusceptible isolate (McM) of H. contortus, and the resistance ratios of three field isolates of this species Stage
Eggs
Assay No. 1 2
Larvae
1 2
ECho1 (95% limits) McM
Resistance ratios e (95% limits) 8 LEV
3 TBZ per move
8 IVM
0.499 (0.456-0.546) 0.629 (0.595-0.666)
4.9 (4.3-5.6) 2.4 (2.0-2.8)
3.3 (2.9-3.8) 5.6 (5.1-6.2)
2.4 (2.1-2.7) 1.8 (1.6-2.0)
0.0468 (0.0404-0.0543) 0.0744 (0.0719-0.0770)
5.7 (4.7-7.0) 4.3 (3.8-4.9)
10.5 (8.9-12.4) 9.3 (8.5-10.1)
3.4 (2.9-4.2) 2.3 (2.0-2.6)
1ECso=ECsothiabendazole concentration (/IM ml-1). 2Resistance ratios--ECso test strain/ECs0 McM.
compared with the in vivo assay. Estimates of resistance (resistance ratios and susceptibility factors) showed that the 3 TBZ per move was the most resistant isolate followed by the 8 LEV which was more resistant than the 8 IVM isolate. For the egg-hatch assay, the standard errors of ECsos between lines were very small and particularly consistent within each assay, with a standard error estimated at 0.0786 by analysis of variance. For the larval-development assay, no estimates of standard errors within assays were determined because no replication was carried out. However, this assay was shown to be similar to the egg-hatch assay in being highly accurate, with very low standard errors for the fit of the data to the model, and repeatable. The range of the 95% confidence limits for both of these procedures was sufficiently narrow for all tests to show
218
M.V. JOHANSENAND P.J. WALLER
TABLE3 Estimates from egg-hatch assays of thiabendazole concentrations which, on average, would prevent 50% of eggs from hatching for a benzimidazole-susceptible isolate (McM) of H. contortus, and the resistance ratios of three field isolates of this species Assay No. 1 2 3
ECso 1 (95% limits) McM
8 LEV
3 TBZ per move
8 IVM
0.245 (0.235-0.255) 0.069 (0.062 0.077) 0.076 (0.070-0.081)
2.8 (2.6 3,0) 5.2 (4.3-6,1) 5.6 (4.9 6.4)
3.3 (3.1 3.5) 9.3 (7.8-11.0) 9.9 (9.0 11.0)
1.4 (1.3-1.5) 3.4 (2.9-4.1) 2.4 (2.1-2.7)
Resistance ratios 2 {95% limits)
~ECso = EC~0 thiabendazole concentration (#M m l - 1). 2Resistance ratio = ECho test strain/ECho McM. TABLE4 Calculated binding and susceptibility factors derived from the incubation of [3H] mebendazole to crude supernatants of a benzimidazole-susceptible isolate (McM) of H. contortus and three field isolates of this species Assay No.
1 2 Waller et al. (1989)
Bound TBZ ~ (95% limits) McM
Susceptibility factors 2 (95% limits) 8 LEV
3 TBZ per move
8 IVM
114 (108 120) 92 (86-96)
0.53 (0.42-0.63) 0.52 (0.40-0.65)
0.37 (0.27 0.47) 0.48 (0.36-0.61)
0.66 (0.55-0.77) 0.83 (0.69-0.98)
100 (94-108)
0.68 (0.61-0.75)
0.33 (0.26 0.40)
0.74 (0.67-0.81)
1Bound TBZ = Bound mebendazole pmoles m g - 1 protein. 2Susceptibility factors = Bound MBZ, test strain/Bound MBZ, McM.
the level of benzimidazole resistance in the 3 TBZ per move isolate to be significantly higher than both the 8 LEV and 8 IVM isolates and the 8 LEV also to be significantly higher than 8 IVM. Although the tubulin-binding assay discriminated between the isolates, the width of the 95 % confidence interval were greater than the other two techniques, so there was overlap in the susceptibility factor estimates especially between 8 LEV and 8 IVM isolates. DISCUSSION
All three in vitro techniques gave consistent and comparable results. They discriminated between the field isolates of H. contortus, with the TBZ selected
BENZIMIDAZOLE RESISTANCE IN HAEMONCHUS CONTORTUS
219
strain (3 T B Z / m o v e ) showing the greatest degree ofbenzimidazole resistance, followed by the 8 LEV isolate and then the 8 IVM strain. These differences were shown to be significant in both the egg-hatch and larval-development assay. Although the in vivo assay showed that all of these isolates had a high level of resistance to TBZ, this assay was unable to differentiate between the isolates as observed for the in vitro procedures. This suggests that the in vitro techniques are more sensitive than the in vivo controlled anthelmintic efficiency test which requires considerable resources such as sheep (in this case, 45 lambs), pen accommodation and feed for at least 4 weeks, and considerable technical labour to conduct worm counts. All three in vitro assays produced highly satisfactory results, with sensitive comparisons between strains and within assays. However, all showed relatively large between-day variation, possibly attributable to variations in ambient or incubator temperature, drug preparation or egg recovery indicating the desirability to incorporate a reference control strain each time an assay is conducted. Consistency of the results is also shown by the independent estimates of benzimidazole resistance of the three field isolates by the tubulin-binding assay (Waller et al., 1989 ), which were in accord with our observations (Table 4). The only aberrant result was in the number of eggs that hatched between isolates for the first larval-developmeot assay. This may be attributed to the fact that for this assay, compared with!the egg-hatch assay, no special precautions are required to obtain eggs at a uniform early stage of development because the assay was specifically developed to discriminate between test isolates by estimating the number of eggs that succeed in developing to the third larval stage. In fact, this assay was shown to provide the correct result when development was allowed to proceed and estimates of third-stage larvae were made. It is of interest to note the much lower EC~0 determined by the egg-hatch assay compared with the ECso of the eggs in the larval-development assay. This is in accord with the observations by Lacey et al. (1987) who showed that the action of benzimidazoles in the egg-hatch assay is principally ovistatic, whereby the drugs exert an inhibitory effect on development resulting in more eggs hatching with increasing time of incubation. The egg-hatch assay is a relatively cheap, simple and fast procedure, with results obtained within 3 days. Although the tubulin-binding assay can be used on all stages of development, from eggs to adult parasites, it has been developed using mainly infective larvae (Lacey and Snowdon, 1988 ). Therefore to obtain a sufficient quantity of parasite material for this assay (100 000 L3 were used in this investigation), a minimum culture time of 7 days is required. Tubulinbinding assays also require access to expensive laboratory apparatus for high performance liquid chromatography estimations, and a source of radiolabelled drug. The larval-development assay has a great advantage, not available with the former techniques, of being able to be used to test simultaneously a c o m -
220
M.V.JOHANSENANDP.J. WALLER
plete range of both broad- and narrow-spectrum anthelmintics. However, the technique requires possibly greater technical expertise than the other in vitro procedures and more time in counting the assay, particularly if both ovicidal and larvacidal effects are to be estimated. The decision on which in vitro procedure to use to detect benzimidazole resistance is one of personal preference and is influenced by such considerations as the availability of equipment, speed with which results are required, and available technical skills rather than precision which is similar for the three techniques investigated here. ACKNOWLEDGEMENT
We are grateful to R.J. Dobson for statistical advice.
REFERENCES Bauer, C., Merkt, J.C., Janke-Grimm, G. and Btirger, H.-J., 1986. Prevalence and control of benzimidazole-resistant small strongyles on German Thoroughbred studs. Vet. Parasitol., 21:189203. Dobson, R.J., Donald, A.D., Waller, P.J. and Snowdon, K.L., 1986. An egg-hatch assay for resistance to levamisole in trichostrongylid nematode parasites. Vet. Parasitol., 19: 77-84. Dobson, R.J., Griffiths, D.A., Donald, A.D. and Waller, P.J., 1987. A genetic model describing the evolution of levamisole resistance in Trichostrongylus colubriformis, a nematode parasite of sheep. IMA J. Math. Appl. Med. Biol., 4: 279-293. Donald, A.D., Waller, P.J., Dobson, R.J. and Axelsen, A., 1980. The effect of selection with levamisole on benzimidazole resistance in Ostertagia spp. of sheep. Int. J. Parasitol., 10:381 389. Eagleson, J.S. and Bowie, J.Y., 1986. Oxfendazole resistance in Trichostrongylus axei in cattle in Australia. Vet. Rec., 119: 604. Jackson, R.A., Townsend, K.G., Pyke, C. and Lance, D.M., 1987. Isolation of oxfendazole resistant Cooperia oncophora in cattle. N.Z. Vet. J., 35: 187-189. Johansen, M.V., 1989. Review of techniques used for the detection of anthelmintic resistance in nematode parasites of domestic livestock. Vet. Res. Commun., in press. Kendall, M.K. and Stuart, A., 1977. The Advanced Theory of Statistics. Vol. 1. Charles Griffin, London, p. 247. Lacey, E. and Snowdon, K.L., 1988. A routine diagnostic assay for the detection of benzimidazole resistance in parasitic nematodes using tritiated benzimidazole carbamates. Vet. Parasitol., 27: 309-324. Lacey, E., Brady, R.L., Prichard, R.K. and Watson, T.R., 1987. Comparison of inhibition of polymerisation of mammalian tubulin and helminth ovicidal activity by benzimidazole carbamates. Vet. Parasitol., 23: 105-119. Le Jambre, L,F., 1976. Egg-hatch as an in vitro assay of thiabendazole resistance in nematodes. Vet. Parasitol., 2: 385-391. Waller, P.J., 1986. Anthelmintic resistance in nematode parasites of sheep. Agric. Zool. Rev., 1: 333 373. Waller, P.J. and Lacey, E., 1985. Nematode growth regulators. In: N. Anderson and P.J. Waller (Editors), Resistance to Anthelmintic Drugs. CSIRO, Melbourne, pp. 137 148. Waller, P.J., Dobson, R.J., Donald, A.D., Griffiths, D.A. and Smith, E.F., 1985. Selection studies
BENZIMIDAZOLE RESISTANCE IN HAEMONCHUS CONTORTUS
221
on anthelmintic resistant and susceptible populations of Trichostrongylus colubriformis of sheep. Int. J. Parasitol., 15: 669-676. Waller, P.J., Donald, A.D., Dobson, R.J., Lacey, E., Hennessy, D.R. and Prichard, R.K., 1989. Changes in anthelmintic resistance status of Haemonchus contortus and Trichostrongylus colubriformis exposed to different anthelmintic selection pressures in grazing sheep. Int. J. Parsitol., in press.