|
|
In this work, we demonstrated that the acquisition of the benzimidazole (BZ) resistance in the small ruminant parasite Teladorsagia circumcincta is linked to the selection of individuals that are characterized by a tyrosine (Tyr) at the amino acid 200 of their isotype 1 ß-tubulin gene. This mutation is recessive since only homozygous mutant (Tyr/Tyr) individuals are BZ resistant. In the BZ resistant populations, a decrease of the restriction polymorphism (RFLP) of the isotype 1 b-tubulin gene was observed. This decrease of ß-tubulin polymorphism results only from the selection of homozygous mutant individuals. Analysis of the sequences of a 550 bp fragment of the isotype 1 b-tubulin gene in different natural BZ resistant or susceptible populations confirmed this observation and showed that the alleles which carries the mutation of the residue 200, are not the same in all the BZ resistant studied populations. This result suggests that resistance alleles were present in populations prior to selection by BZ treatment.
The appearance of benzimidazole (BZ) resistance in trichostrongylid nematode communities has stimulated efforts to determine how these parasites become resistant. Knowledge of the molecular changes that confer resistance may lead to the development of rapid, reliable diagnostic protocols, and hence limit the spread of this phenomenon on farms. The molecular basis of resistance to BZ involves alterations of the b-tubulin gene, b-tubulin being the target of BZ. Two main types of change occur in the trichostrongylid nematodes. One is a mutation at the amino acid 200 (Phe-Tyr) of the isotype 1 b-tubulin gene, which leads to BZ resistance in the three dominant species in temperate zone, Haemonchus contortus, Trichostrongylus colubriformis and Teladorsagia circumcincta (Kwa et al., 1993b; Kwa et al., 1994; Lehrer et al., 1995; Elard et al., 1996). This mutation is functional in the expression of the BZ susceptibility or resistance and is not simply a marker linked to the emergence of resistance (Kwa et al., 1995). The second recorded change in b-tubulin gene is a loss of allelic diversity of this gene in H. contortus and T. colubriformis (Roos et al., 1990; Kwa et al., 1993a; Beech et al., 1994; Lubega et al., 1994; Grant & Mascord, 1996). This reduction in b-tubulin polymorphism has been interpreted as the result of selection, and is considered to be a major method of acquiring BZ resistance. But several of these studies were carried out using laboratory populations that have been generally isolated from the field for many generations. There is good evidence that the genetic variability of these strains can decrease in the laboratory due to genetic drift or selection linked to artificial breeding conditions (Gasnier et al., 1992; Nadler, 1990). Another problem is that the detected restriction polymorphism can result from mutations in coding or non-coding regions. The high conservation of the coding sequences of the b-tubulin gene, demonstrated by the great similarities between the amino acid sequences in different species (Elard et al., 1996), suggests that most of the detected polymorphism in this gene arise from nucleotide substitutions located in the introns or from synonymous substitutions in the coding regions. But these mutations are selectively neutral and so cannot play a role in the acquisition of resistance. The sole mutation which generates the change of an amino acid in the isotype 1 b-tubulin gene is the mutation of the residue 200. To evaluate the respective role of the 2 described mechanisms of BZ resistance (mutation and selection of a resistant allele) in T. circumcincta, we have developed 2 approaches using natural populations. In a first time, the importance of the mutation of the residue 200 of the isotype 1 b-tubulin gene in the acquisition of the BZ-resistance, was tested by typing worms from BZ treated or non-treated isolates of the same strains. A PCR method recently perfected by us (Humbert & Elard, 1997) allows us to determine if the worms are homozygous (Tyr/Tyr or Phe/Phe) or heterozygous (Tyr/Phe) for the mutation which seems to be decisive in the acquisition of the resistance. In a second time, the genetic variability of the isotype 1 b-tubulin gene was evaluated in natural BZ susceptible and resistant populations of T. circumcincta using 2 approaches : restriction polymorphism of the complete b-tubulin sequence was estimated by PCR-RFLP, nucleotide mutations were then identified by sequencing a 550 bp fragment from the central part of this gene. This fragment contains 2 non-coding regions and 3 coding regions, including the site responsible for the mutation of the amino acid 200 which is involved in the resistance to BZ.
Parasite
collection and DNA isolation.
Five BZ susceptible (SuBOU, SuLEL, SuLM, SuPRO, SuTOU) and 4 BZ resistant (ReCAS, ReGAU, ReGP, ReECH) isolates of Teladorsagia circumcincta were collected from goat farms (except for SuBOU, SuPRO and SuLM, which were isolated from sheep) located in central France. The resistance was estimated by the method of Coles et al. (1992) (SuLEL, LD50 = 0.03 µg ml-1; ReCAS, LD50 = 0.85 µg ml-1; ReECH, LD50 = 0.31 µg ml-1; ReGAU, LD50 = 0.33 µg ml-1; ReGP, LD50 = 0.15 µg ml-1) or in the field, by checking the treatment efficacy (100 % reduction of the egg output after BZ treatment of the host in SuBOU, SuPRO, SuTOU, SuLM populations). The genomic DNA from at least 20 adult male worms in each isolate was prepared as previously described (Humbert & Cabaret, 1995).
Detection
of the mutation of the amino acid 200 of the isotype 1 b-tubulin gene in worms
recovered from BZ treated or non-treated isolates.
Females of T. circumcincta were isolated from 2 resistant populations (ReCAS and ReGP). Eggs from these females were cultured to obtain infective larvae (L3). For each population, 1 worm-free lamb was infected with these L3. After 21 days, all faeces were collected from these lambs (from day 21 to day 36) and cultured to yield infective larvae. For each isolate, 2 worm-free lambs were then infected with 4000 of these larvae. After 28 days, one of the 2 lambs was treated with Panacur (Fenbendazole) at the commercial dose (5mg/kg body weight). All lambs were slaughtered at the day 35 and the parasites recovered. The presence of a phenylalanine (TTC) or a tyrosine (TAC) at the residue 200 was detected in these worms according to the method described by Humbert and Elard (1997).
Amplification
of the isotype 1 b-tubulin gene and restriction enzyme digestion.
The isotype 1 b-tubulin gene was amplified in 2 steps. A 3.6 kb fragment was first amplified with the ExpandTM Long Template PCR System (Boehringer, Mannheim), according to the supplier's instructions, with the primers TUB3 (5'-GAG GAG CCC CAT GCC GAG AAG ACG TGG AAG-3') and TUB5 (5'-ATG CGT GAA ATC GTT CAT GTA CAA GCC GGT-3'). The second step was a nested PCR which generates a 3.5 kb fragment using the primers N5 (5'-TGT ACA AGC CGG TCA ATG CGG-3') and N3 (5'-AGC TCT AGC GGG TAT AGC AG-3'). The procedure was : 50 µl reaction mixture, 5.0 µl 10X Taq buffer (Promega), 1.5 U Taq polymerase (Promega), 80 µmoles of each dNTP (Promega); 30 cycles of 92 °C for 50 sec, 58 °C for 60 sec, 72 °C for 90 sec; final extension 72 °C for 10 min. After control by agarose gel electrophoresis, only the best amplifications (1 band with a good intensity) were kept for the restriction study. These PCR products were digested using 5 enzymes (EcoR1, EcoRV, Hinf1, Dra1 and Rsa1; Eurogentec) and then separated on a 1.5 % agarose gel (GIBCO-BRL). The electrophoresis gels were analyzed with a BioImage system. Restriction fragments were scored as 1 (present) or 0 (absent) in a data table. Then, a correspondence analysis was performed using the ADE-4 software package (Thioulouse et al., 1997).
Sequencing
and cloning of the central part of the isotype 1 b-tubulin gene.
An aliquot of amplified product from the isotype 1 b-tubulin gene was used as template in a new PCR (25 µl reaction mixture, 2.5 µl 10X Taq buffer (Promega), 1.0 U Taq polymerase (Promega), 80 µmoles of each dNTP) with two primers (25 pmoles of each; RES1 : 5' CCAACTGACGCATTCTTTGG 3’; M2 : 5' GATCAGCATTCAGCTGTCCA 3'). This permits the amplification of a 550 bp fragment from the central part of the gene. PCR was carried out in a MJ Research thermal cycler using the following conditions : one cycle of 94 °C for 90 sec, 58 °C for 60 sec, 72 °C for 90 sec; 39 cycles of 92 °C for 50 sec, 58 °C for 60 sec, 72 °C for 90 sec; final extension 72 °C for 10 min. The PCR products were purified on QIAquicks columns (Qiagen). The PCR products were directly sequenced using the RES1 primer, but some of them were cloned in the pGEM-T vector (Promega) according to the supplier's instructions and sequenced using M13 Forward and Reverse sequencing primers on an Applied Biosystems 377 automated DNA sequencer (Applied Biosystems, Perkin Elmer). The GCG package (Genetics Computer Group, Inc. Madison) was used for analysis. A Wagner parsimony analysis was conducted with PHYLIP 3.5c (Phylogeny Inference Package, Felsenstein (1993)).The bootstrap option was used to run 100 replicates.
Typing
of worms from BZ treated or non-treated isolates from resistant populations.
The results of the BZ treatment experiment on 2 resistant populations (Table 1) show that the most resistant population (ReCAS, LD50 = 0.85 µg ml-1) was characterized without treatment by the high prevalence of homozygous resistant individuals (Tyr/Tyr for the residue 200). In the treated lambs, this prevalence increases to 100 % (LD50 = 1.05 µg ml-1). The other resistant population (ReGP, LD50 = 0.15 µg ml-1) was characterized without treatment by a great proportion of heterozygous (Phe/Tyr for the residue 200) and homozygous resistant (Tyr/Tyr) individuals. After BZ treatment, only homozygous resistant individuals were observed in the worm population and the LD50 was estimated to 0.97 µg ml-1. For the ReGP population, the restriction polymorphism of the worm populations collected from BZ treated or non treated lambs was compared using Rsa1 and Dra1 restriction enzymes. For Dra1, 8 restriction profiles (in 18 studied individuals) were observed in the worm population of the non treated lamb. In the worm population of the BZ treated lamb, only 3 profiles were observed in 19 individuals. The same type of results was observed with the Rsa1 enzyme : Eleven profiles were present in the non treated population and only 3 profiles in the treated population. The restriction profiles observed in the worm population after BZ treatment were the same as those observed in the resistant individuals of the non treated population.
Restriction
polymorphism of the isotype 1 b-tubulin gene in natural BZ-susceptible and
resistant populations.
Digestion of the amplified fragment (3.5 Kb) gave 2-12 bands in restriction profiles that varied with the populations and the restriction enzymes used. The multi-enzyme patterns were determined for each individual by addition of the results obtained with each enzyme (Dra1, EcoR1, EcoRV, Hinf1,Rsa1). Results (Table 2) show that the number of distinct restriction patterns is greater (P < 0.05, Mann-Whitney U-Test) in BZ susceptible populations than in the resistant. In addition, the most resistant population (ReCAS, LD50 = 0.85 µg ml-1) was characterized by the presence of only 2 different restriction patterns in the 18 studied individuals. Conversely, the ReGP population which was characterized by the lower LD50 value (LD50 = 0.15 µg ml-1) in the resistant populations, presented 10 different restriction patterns (for 17 studied individuals). The multi-enzyme patterns were not the same in all of the BZ resistant populations but several of these patterns were observed in different BZ resistant populations. The correspondence analysis on these restriction data (Fig. 3) confirm these observations. The subsets of individuals (populations) are summarized by ellipses that contain 90 % of the individuals according to a Gaussian distribution. The comparison of these ellipses clearly shows that the genetic variability of the isotype 1 b-tubulin gene estimated by restriction polymorphism, is lower in the BZ-resistant populations than in the susceptible.
A 550 bp fragment in the central region of the gene from 85 individual worms was first amplified and then directly sequenced. These individuals were isolated from 8 populations (4 resistant populations : ReCAS, ReGP, ReECH, ReGAU; 4 susceptible : SuBOU, SuLM, SuPRO, SuLEL). Only 55 of the 85 sequences were clearly readable in full and 30 sequences were only readable with some ambiguous sites on the first 152 bases. The 55 complete sequences were aligned and treated by parsimony analysis. The resulting consensus tree is shown in figure 2. There were 2 groups of sequences highly supported by the bootstrap proportions (Al. 1 and Al. 2, Fig. 3). The first group (Al. 1) contained 42 individuals and the second 13 (Al. 2). The existence of these 2 groups of sequences was confirmed and the differences between the sequences verified by cloning 25 individuals from the 55. The results (Fig. 3) showed that the 2 groups differed by 14 synonymous substitutions in the coding regions (348 bp), and at 52 sites in the non-coding regions (202 bp). These 2 types of sequences seem to represent 2 alleles of the isotype 1 b-tubulin gene. Hence, there must be heterozygous individuals in the populations. The 30 sequences that were only partially readable come from these putative heterozygous individuals. The beginning of these sequences (base 0 to base 53) was clearly readable; in this portion, the sequences of the 2 alleles are identical. From the base 54 to the base 152, 4 ambiguous sites (positions 54, 60, 117, 120) which correspond to 4 differences between the 2 alleles were found. After the base 152, the presence of a deletion at the beginning of the first intron, which generates a gap between the 2 sequences, made reading impossible the subsequent sequence of the heterozygous individuals. The evidence for presence of the 2 alleles in these individuals was obtained by cloning PCR fragments from 4 of these putative heterozygous worms and sequencing 4-8 clones for each of them. The 2 alleles were found in each individual, which were thus heterozygotes. Except for the population ReCas, in which only 6 individuals were sequenced, all the other populations had the 2 alleles. The first allele was the most common allele in 6 populations (Al.1 frequencies > 63%), while allele 2 was only dominant in the resistant population ReGP (Al.2 frequency = 65%). The parsimony analysis (Fig. 2) was only performed on homozygous individuals (Al.1/Al.1 or Al.2/Al.2). In the first group of sequences (Al. 1, Fig. 2), cluster I contains individuals belonging to 3 resistant populations (ReGP, ReCAS and ReGAU). These individuals had 4 mutations in common in the 2 introns and were characterized by a tyrosine (TAC) at residue 200 of the b-tubulin (position 326 in Fig. 3). Cluster III contained individuals from the resistant population ReECH, and cluster II individuals from the susceptible population SuPRO. These 2 clusters were not supported using the bootstrap resampling. They were not defined by specific sites, but by combinations of nucleotides at different positions. In addition, all the individuals of the ReECH populations had a tyrosine at residue 200. In the second group of sequences (Al. 2, Fig. 2), cluster IV was supported by a high bootstrap proportion (BP=83) and contained the individuals of the ReGP population. These individuals had 5 mutations in common in the 2 introns and a tyrosine at residue 200 of the b-tubulin. The individuals of the susceptible populations SuLEL, SuLM and SuBOU were not arranged in cluster. Only 2 individuals from resistant populations, ReGP6 and ReGAU1, were not well classified. Analysis of their sequences revealed that ReGP6 had a TTC codon (Phe) at residue 200 (position 326 in Fig. 3), while ReGAU1 was heterozygous (TTC/TAC) at this position. This individual was also heterozygous at the 4 positions (in introns 1 and 2) which define cluster I.
The first goal of this work was to evaluate the importance of the mutation of the residue 200 of the isotype 1 b-tubulin gene in the BZ resistance. The typing of worms (Table 1) recovered from BZ treated or non treated resistant strains (ReGP, ReCAS) shows that only homozygous individuals (Tyr/Tyr) for the residue 200 were resistant and that homozygous (Phe/Phe) and heterozygous (Phe/Tyr) were eliminated from the worm population after BZ treatment. The LD50 value estimated on the egg hatching (Coles et al., 1992), was correlated with the proportion of homozygous resistant individuals in the worm populations. So, the mutation (Phe to Tyr) at the residue 200 of the isotype 1 b-tubulin gene confers BZ resistance and appears to be recessive. This result is in full agreement with observation of Kwa et al. (1995) which had proposed this hypothesis after experiments of transformation of a Caenorhabditis elegans susceptible strain with a BZ resistant Haemonchus contortus allele. The comparative study of the b-tubulin sequences in different organisms supports the hypothesis that the nature of the amino acid at position 200 is fundamental for the BZ susceptibility or resistance. Numerous studies showed that BZ altered microtubules of the parasites but not those of the vertebrate host. An examination of the b-tubulin sequences of vertebrates indicated that a tyrosine is present at the position 200 in five (classes I to V) of the six isotype classes present in mammals (Sullivan, 1988). Conversely, Katiyar et al. (1994) have observed that the Phe-200 is a strong predictor of benzimidazole susceptibility for protozoan parasites. The second goal of this study was to determine if a decrease of polymorphism of the isotype 1 b-tubulin gene was present in natural BZ resistant populations of T. circumcincta. Even if the number of individuals in some populations was low due to difficulties to obtain a good amplification of the complete isotype 1 b-tubulin sequence, all the results obtained in this RFLP study were very consistent. The BZ-resistant populations were characterized by a low diversity of the restriction profiles in comparison with the susceptible populations. This reduction is more important in the resistant population (ReCAS) which presented the higher LD50 than in the other. Results from the sequencing of the central part of the isotype 1 b-tubulin gene confirm the decrease of polymorphism of this gene in resistant populations. That’s we observed in the parsimony analysis that individuals of these populations were grouped in cluster when individuals of susceptible populations were not arranged in such clusters. So, the genetic variability of the isotype 1 b-tubulin gene in the natural resistant populations of T. circumcincta is lower than in the susceptible populations. This result confirms those reported on laboratory strains with 2 other trichostrongyle nematodes, Haemonchus contortus (Beech et al., 1994; Kwa et al., 1993a, b; Lubega et al., 1994; Roos et al., 1990) and Trichostrongylus colubriformis (Grant & Mascord, 1996). If this mutation of the residue 200 is the main factor of acquiring BZ, the decrease of the polymorphism of the isotype 1 b-tubulin gene observed by RFLP and sequencing in the resistant populations, must then be interpreted as the selection of the homozygous mutant individuals (rr). This hypothesis is supported by the RFLP results obtained on the BZ treated or non treated strains of the ReGP population in which we observed that most of the allelic diversity present in the non-treated populations was carried by the susceptible worms (rS and SS). In addition, this hypothesis allows to understand why the same allele was not observed in all resistant populations. Excepted the mutation of the residue 200, all other mutations which define these different alleles, were selectively neutral (Fig. 2) and cannot play a role in the acquisition of the BZ resistance. The selection of these different alleles in resistant strains is only linked to the presence of the mutation at the residue 200. The sequencing of the central part of the isotype 1 b-tubulin gene give some indications on the origin of the BZ-resistant alleles which carry the mutation of the residue 200 in the BZ resistant populations. All individuals of the 4 resistant populations are not contained in a monophyletic group which suggests that the resistant alleles have not the same origin for all resistant populations. So, it’s very probable that these alleles were present in the worm populations prior to selection by BZ treatments and that the emergence of the resistance in a flock is not linked to BZ-resistant worm’s migration. This agree with observations of Cabaret and Gasnier (1994) which demonstrate that helminthic populations in dairy-goat farms are isolated (no introduction of parasited animals after the constitution of the herds). The small genetic variability of the isotype 1 b-tubulin gene observed in the most resistant population (ReCAS) suggests that bottlenecks must occur in the resistant populations when a strong selection pressure is exercised by BZ treatments. To confirm this assertion, a comparative study of the heterozygosity in BZ susceptible or resistant worm populations is necessary using neutral markers as microsatellites for example. One other question linked to this work concerns the possible consequences of this mutation on the fitness of the parasites. The b-tubulin gene is one of the most conserved genes which shows that mutations in coding regions of this gene are counter selected. The comparison of the fitness of Tyr-200 or Phe-200 parasites in absence of BZ selection pressures is very interesting because it has many implications in the management of BZ resistance in the field.
Acknowledgements
L. Elard was supported by a grant from the Région Centre and the INRA. We thank M. Peloille from the sequencing service of INRA (Station P.A.P., Tours), C. Sauvé for technical assistance, J. Cabaret and N. Gasnier for providing parasite populations and J. Cabaret for stimulating discussions. The English text was checked by Dr. Owen Parkes.
Beech RN, Pritchard RK, Scott ME (1994) Genetic variability of the b-tubulin genes in Benzimidazole-susceptible and -resistant strains of Haemonchus contortus. Genetics, 138, 103-110.
Cabaret J., Gasnier N. 1994. Farm history and breeding history influences on the intensity and specific diversity of nematode infection of dairy-goats. Veterinary Parasitology, 53,219-232.
Coles GC, Bauer C, Borgsteede FHM, Geerts S, Klei TR, Taylor MA, Waller PW (1992) World Association for the Advancement of Veterinary Parasitology (W.A.A.V.P.) methods for the detection of anthelmintic resistance in nematodes of veterinary importance. Veterinary Parasitology, 44, 35-44.
Elard L, Comes AM, Humbert JF (1996) Sequences of b-tubulin cDNA from benzimidazole-susceptible and -resistant strains of Teladorsagia circumcincta, a nematode parasite of small ruminants. Molecular and Biochemical Parasitology, 79, 249-253.
Felsenstein J (1993) Phylogenetic Inference Programs (PHYLIP), Manual 3.5c. University of Washington, Seattle.
Gasnier N, Cabaret J, Moulia C (1992) Allozyme variations between laboratory reared and wild populations of Teladorsagia circumcincta. International Journal for Parasitology, 22, 581-587.
Grant WN, Mascord LJ (1996) Beta-tubulin polymorphism and benzimidazole resistance in Trichostrongylus colubriformis. International Journal for Parasitology, 26, 71-77.
Humbert JF, Cabaret J (1995) Use of the Random Amplified Polymorphic DNA (RAPD-PCR) for identification of Trichostrongylids Nematodes. Parasitology Research, 81, 1-5.
Humbert JF, Elard L (1997) A simple PCR method for rapidly detecting point mutations. Technical Tips Online, http://tto.trends.com, T40076.
Katiyar SK, Gordon VR, McLaughlin GL, Edlin TD (1994). Antiprotozoal activities of benzimidazoles and correlations with b-tubulin sequence. Antimicrobial Agents Chemotherapy, 38, 2086-2090.
Kwa MS, Kooyman FNJ, Boersema JH, Roos MH (1993a) Effect of selection for benzimidazole resistance in Haemonchus contortus in b-tubulin isotype 1 and isotype 2 genes. Biochemical and Biophysical Research and Communications, 191, 413-419.
Kwa MS, Veenstra JG, Dijk MD, Roos MH (1995) b-tubulin genes from the parasitic nematode Haemonchus contortus modulate drug resistance in Caenorhabditis elegans. Journal of Molecular Biology, 246, 500-510.
Kwa MS, Veenstra JG, Roos MH (1993b) Molecular characterisation of b-tubulin genes present in benzimidazole-resistant populations of Haemonchus contortus. Molecular and Biochemical Parasitology, 60, 133-144.
Kwa MS, Veenstra JG, Roos MH (1994) Benzimidazole resistance in Haemonchus contortus is correlated with a conserved mutation at amino acid 200 in b-tubulin isotype 1. Molecular and Biochemical Parasitology, 63, 299-303.
Lehrer S, Davey H, Watson T, Wilkins RJ (1995) Sensitive PCR for detecting Benzimidazole resistant sub populations of ovine nematodes in the Waikito. Proceedings of the New Zealand Society of Animal Production, 55, 209-210.
Lubega GW, Klein RD, Geary TG, Pritchard RK (1994) Haemonchus contortus: The role of two b-tubulin gene subfamilies in the resistance to benzimidazole anthelminthics. Biochemical Pharmacology, 47, 1705-1715.
Nadler S (1990) Molecular approaches to studying helminth population genetics and phylogeny. International Journal for Parasitology, 20, 11-29.
Roos MH, Boersema JH, Borgsteede FHM, Cornelissen J, Taylor M, Ruitenberg EJ (1990) Molecular analysis of selection benzimidazole resistance in the sheep parasite Haemonchus contortus. Molecular and Biochemical Parasitology, 43, 77-88.
Sullivan KF (1988) Structure and utilization of tubulin isotypes. Annual Review of Cellular Biology, 4, 687-716.
Thioulouse J, Chessel D, Doledec S, Olivier JM (1997) ADE-4: A multivariate analysis and graphical display software. Statistics and Computing, 7, 75-83.
QUATRIEME
ARTICLE