Genomic organization and effects of ivermectin selection on Onchocerca volvulus P-glycoprotein

Molecular & Biochemical Parasitology 143 (2005) 58–66

Genomic organization and effects of ivermectin selection on Onchocerca volvulus P-glycoprotein Bernadette F. Ardelli, Sean B. Guerriero, Roger K. Prichard ∗ Institute of Parasitology, Macdonald Campus, McGill University, 21 111 Lakeshore Road, Ste. Anne de Bellevue, Que., Canada H9X 3V9 Received 7 March 2005; accepted 16 May 2005 Available online 13 June 2005

Abstract Ivermectin (IVM) was first developed for use with livestock. It is now the only drug used for mass treatment of onchocerciasis. It is difficult to prove whether reports of sub-optimal responses to IVM in some Onchocerca volvulus infected patients are a result of drug resistance, as procedures typically used to examine IVM efficacy in livestock can not be performed on humans. To determine the effects of IVM on O. volvulus, one approach is to examine allele frequencies before and after treatment. Allele(s) linked to resistance may increase in frequency after repeated treatment. Mass treatment of large human populations to reduce transmission of O. volvulus will impose selection pressure for resistance. P-glycoprotein has been implicated as a candidate IVM resistance gene in nematodes. In this study, the intron–exon structure of O. volvulus P-glycoprotein (OvPGP) has been defined. The gene spans 10.6 kb, is AT-rich, contains 24 exons and a high proportion of class 0 introns. The genetic diversity of 28 loci spanning the entire OvPGP gene was examined in four O. volvulus populations from the Volta Region of Ghana. Worms collected in 1999 and 2002 from IVM treated patients showed reduced genetic polymorphism and an increase in the number of loci not in Hardy–Weinberg equilibrium. Changes in allelic patterns and a reduction in diversity at many loci in P-glycoprotein in the parasites from IVM treated patients in 1999 and 2002 suggest that IVM is imposing selection on this gene, consistent with a possible development of IVM resistance. © 2005 Elsevier B.V. All rights reserved. Keywords: Onchocerca volvulus; P-glycoprotein; Ivermectin; Genetic polymorphism

1. Introduction Onchocerciasis, or river blindness, is a disease that results from infection with the filarial nematode Onchocerca volvulus. Along with causing serious skin pathologies, onchocerciasis is considered one of the leading causes of blindness in the developing world. Control of onchocerciasis depends almost entirely on yearly treatments with 150 ␮g/kg of ivermectin (IVM). Although very effective at reducing the number of O. volvulus microfilaria (MF) and inhibiting adult reproduction, at this standard dose, IVM does not kill or permanently

Abbreviations: IVM, ivermectin; OvPGP, Onchocerca volvulus Pglycoprotein; MF, microfilariae; ABC, ATP binding cassette; HWE, Hardy– Weinberg equilibrium; ORF, open reading frame ∗ Corresponding author. Tel.: +1 514 398 7729; fax: +1 514 398 7857. E-mail address: [email protected] (R.K. Prichard). 0166-6851/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.molbiopara.2005.05.006

sterilize the majority of adult female O. volvulus; females are capable of regenerating MF within 1 year [1]. Repeated annual treatment is required to suppress microfilarial production and reduce skin and ocular lesions [2]. Some studies suggest IVM has a modest effect on adult worm viability [3,4]. In these studies, where there was decreased fertility displayed by female worms, it is not known if this was a result of an IVM-induced change in the reproductive capacity of female worms or the fertility of male worms, and/or if IVM has a direct effect on female worm survival, or on both fertility and survival. The ATP-binding cassette (ABC) superfamily comprises more than 200 proteins, the majority of which mediate the selective transport of substances across biological membranes [5]. IVM is a substrate of P-glycoprotein [6,7] and accumulates to abnormal levels in the brain of transgenic mice lacking P-glycoprotein, resulting in neurotoxicity [8,9].

B.F. Ardelli et al. / Molecular & Biochemical Parasitology 143 (2005) 58–66

In helminths, P-glycoprotein is the most well-studied member of the ABC-transporter superfamily. In H. contortus and C. elegans, resistance to macrocyclic lactones and other compounds was associated with a P-glycoprotein gene [10–12]. In C. elegans, resistance to various toxic compounds was due to pgp-3 [11], while deletion of pgp-1 conferred hypersensitivity to heavy metals [13]. Within parasitic nematodes, the P-glycoprotein genes of H. contortus are the most extensively studied ABC transporter. It was demonstrated by [12] that the restriction patterns of a P-glycoprotein homologue from H. contortus differed between IVM-sensitive and IVMresistant strains, with increased P-glycoprotein expression in the resistant strain. A P-glycoprotein homologue was cloned from O. volvulus and was found to be differentially expressed between larval and adult stages of O. volvulus [14] and preliminary work on a 356 bp region of O. volvulus Pglycoprotein suggested IVM was imposing selection pressure on P-glycoprotein in this parasite [15]. Despite an extensive amount of data suggesting a possible involvement of P-glycoprotein in resistance to IVM and related drugs in parasitic nematodes (see [16] for review), functional studies have not been conducted on parasitic nematodes that provide a mechanism for linkage between P-glycoprotein and IVM resistance. The absence of a suitable animal model in which to successfully reproduce the life cycle of O. volvulus has slowed efforts toward unravelling the process(es) that might be involved in IVM resistance in this parasite, should it develop. Thus, alternative measures are required to determine the effectiveness or ineffectiveness of IVM against O. volvulus. Should resistance develop, the allele(s) or gene(s) capable of conferring resistance need to be present in O. volvulus populations before treatment. The mass treatment of large human populations to break transmission of O. volvulus, and the tendency to increase the frequency of IVM treatment in onchocerciasis from once per year to up to two or more times per year, will increase selection pressure for resistance [3,17–19]. The early stage of resistance would manifest itself as a change in frequency and diversity of alleles. The alleles or genes that might be involved in IVM resistance in O. volvulus are not known. However, P-glycoprotein is a logical candidate as modifications in this protein could reduce the effective drug concentrations in the parasite, as it has in other organisms [10,20]. We have analysed the whole P-glycoprotein gene from populations of O. volvulus derived from untreated patients and patients that had been repeatedly treated with IVM in order to determine regions of the gene which show altered genetic diversity.

2. Materials and methods 2.1. Origin of O. volvulus samples West Africa is endemic for onchocerciasis and has a well established chemotherapy research centre, located at

59

the Hohoe Hospital in Hohoe, Ghana. All patients used in the study were examined by an attending physician from Hohoe and participants had to show clinical evidence of infection with O. volvulus (i.e. visible evidence of fibrous nodules, ocular or skin lesions or troublesome itching). In 1999, nodules were obtained by Dr. K. Awadzi (Onchocerciasis Chemotherapy Research, Centre, Hohoe Hospital, Ghana) from several villages in the Volta Region of Ghana. In 2002, a second collection was organized in Ghana by Mr. Mike Osei-Atweneboana (Noguchi Memorial Institute for Medical Research, Accra, Ghana) in the Volta Region of Ghana. After obtaining informed consent, visible nodules and skin snips were removed from each subject. Ethical approval for the study was obtained from McGill University, The Noguchi Memorial Institute for Medical Research and the Hohoe Hospital, Ghana. A total of 215 adult O. volvulus were used in the present study; of these 79 (31 IVM treated-adult female, 48 nontreated-adult male) were collected in the Volta Region in 1999 and a further 136 (76 IVM treated-adult female, 60 nontreated-adult female) were collected from the same region and villages in 2002. The patients sampled in 1999 and in 2002 were different. Following examination by the attending physician, superficial nodules were excised from patients under local anaesthetic and aseptic conditions. Excision sites were dressed and excised nodules were placed immediately in liquid nitrogen and shipped to the Institute of Parasitology, McGill University. Worms were removed from nodules using a modified procedure of [21]. Details of the procedure are available in [15]. 2.2. Genetic data collection and statistical analysis Primer pairs for amplification of 28 loci from a Pglycoprotein gene of O. volvulus (GenBank Accession No. AY884214) were designed and used to amplify these regions from the gene as well as to construct the intron–exon boundaries. The introns were analysed and classified as either class 0 (where the splice occurs between the codons), class 1 (where the codon is interrupted between the first and the second nucleotide), or class 2 (where the splice occurs between the second and third nucleotide of the codon. DNA preparation from worms and marker typing followed procedures outlined in [15]. Marker data was collected from 28 loci across the O. volvulus P-glycoprotein gene (10,574 bp) which had a mean of 2.38 sequence variants per locus (range 2–10). Another study [15] using some of the same O. volvulus samples, examined seven polymorphic control genes. In these control genes, there were no differences in genetic polymorphism between worms removed from treated and non-treated patients. For each locus, Hardy–Weinberg equilibrium (HWE) and genetic polymorphism were determined. Genetic polymorphism was determined using either single strand conformation polymorphism analysis (SSCP) or restriction fragment length polymorphism analysis (RFLP) [22]. To test for significant differences in polymorphism, a

60

B.F. Ardelli et al. / Molecular & Biochemical Parasitology 143 (2005) 58–66

Table 1 Exon–intron organization of O. volvulus P-glycoprotein gene 3 Acceptor

Exon No.

Size (bp)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

121 190 96 134 172 242 148 219 117 129 171 103 136 136 249 161 242 196 154 99 96 201 147 233

ATGCAGgtaaca AAAACTgtaagt ATACAAgttcct CAAGGCgtaaat GCTCGAgtatga GTTCTAgtaagt GATCGAgtaagc ACATCGgtaagt CAACAGgttcga AGTGAGgtaaaa AAAAAGgtcatt AAATCGgtgagt ATCCAAgtaaaa GGAAAGgtttgt TCTGCAgtaagt AAAGCGgtagta GGTCAAgtatgc TTTCACgtaaat GTCCGGgtacat TATGCGgtaagt CAATTGgtaagc CCAAAGgtaata GAAAGAgtttgt –

5 Donor

ttttagGGGTG ttgcagTTTGAA tttcagTATAT tttcagGAATAT ttgcagCAAATG tcacagCTATGG tttcagAAACCG cttcagACTGGT attcagGAACCG gtaaagGGTTAC gtttagGCATTT ttccagGCACTC ttatagTACCAG ttgcagAAAAAA ttgcagACATTT tataggGTCTCG caatagCTGGCA tttcagCGCTTC ttcaggCAATAA ttcaagGGACAA ttatagACTTTG atttagGGCTAC tttcagACAGTT –

Intron No.

Phase

Size (bp)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 –

2 0 0 2 0 2 0 0 0 0 0 1 2 0 0 2 0 2 0 0 0 0 0 –

210 215 211 134 127 373 628 775 187 157 128 253 238 684 258 123 237 66 190 76 368 150 317 –

Note: the exon and intron sequences are shown in uppercase and lowercase letters, respectively.

Likelihood Chi-square test was used and the p-values were adjusted using the Bonferroni correction factor.

3. Results and discussion Analysis of the organization of O. volvulus P-glycoprotein indicates that it spans approximately 10.6 kb (not including promoter) and is comprised of 24 exons ranging in size from 96 to 242 bp. The size of the introns and the sequences at the intron–exon boundaries are shown in Table 1. The average size of the exons was 162 bp. The shortest exons, exon 3 and 21; and the longest exon, exon 15 were 96 and 249 bp, respectively. Most introns were small, with the shortest being 76 bp, while the longest was 684 bp, and the majority were class 0. Several regions of the gene are characterized by clusters of introns in which splice junctions are all of the same class (i.e. class 0: introns 2, 3, 5, 7–11, 14,15, 17, 19–23; class 1: intron 12; class 2: introns 1, 4, 6, 13, 16, 18; Table 1). Notable in this respect are the regions encoding the first and second nucleotide binding domains (NBD), in which all of the introns are class 0. This type of organization could increase the probability of alternative splicing events that result in variant mRNAs in which the original open reading frame (ORF) of OvPGP is maintained. The two homologous halves of P-glycoprotein are connected by a linker region, which is highly divergent in MDR1 transporters across many species [23]. In OvPGP, the three exons that encode the linker region are separated by class 1 and 2 introns, while the introns at

each end of the region are class 0. Consequently, alternative splicing of the linker is incompatible with maintenance of the ORF unless the entire region is eliminated. Work by [24] indicated that the integrity of the linker region was not essential for P-glycoprotein function. Thus, alternative splicing of the linker region would probably still result in a functional P-glycoprotein, provided the entire region is eliminated. The majority of the intron–exon boundaries of OvPGP conform to the consensus splice junction sequences for eukaryotic genes [25]. Of the 23 introns, 16 are class 0 (where the splice occurs between the codons), one is class 1 (where the codon is interrupted between the first and the second nucleotide), and six are class 2 (where the splice occurs between the second and third nucleotide of the codon (Table 1 and Fig. 1A). A comparison of the intron–exon organization of OvPGP and OvPGP mRNA is shown in Fig. 1. Similar to other members of the ABC superfamily, there is no apparent correlation between structural or functional domains of PGP and regions encoded by individual exons. The P-glycoprotein gene of O. volvulus is AT-rich and is relatively compact (Fig. 1B). Similar to other sequenced genes of O. volvulus, OvPGP is interrupted quite often by introns which are relatively small in size. For example, the 15 P-glycoprotein genes in C. elegans range in size from 11 to 27 exons, with the size of the introns ranging from 48 to 1592 bp. As in O. volvulus, C. elegans P-glycoprotein genes are interrupted quite often by short introns. In accordance with other genes of O. volvulus that have been sequenced, the majority of the exons are small [26]. The study of [26]

B.F. Ardelli et al. / Molecular & Biochemical Parasitology 143 (2005) 58–66

61

Fig. 1. (A) Genomic organization of O. volvulus P-glycoprotein and alignment of the splice junctions with O. volvulus P-glycoprotein mRNA. The intron–exon organization is illustrated at the bottom of the figure. The coding regions are indicated by the closed boxes and introns are depicted by thin lines. The region of OvPGP mRNA encoded by each exon are shown at the top. Superimposed on the mRNA is a schematic of the protein with regions predicted to be transmembrane helices (numbered 1–12) or nucleotide binding domains (NBD) indicated by solid bars. Within each NBD, the locations of the Walker A and B motifs are indicated by uppercase letters. The transmembrane helices included in each membrane spanning domain and the linker region are indicated by a line. (B) Plot of G + C content for each coding and non-coding region of O. volvulus P-glycoprotein.

showed that the genome of O. volvulus is relatively AT-rich, with the overall AT content being approximately 68% (based upon sequences of genes and partial gene fragments). They also showed that the intron sequences were slightly more ATrich. The results obtained for O. volvulus P-glycoprotein are consistent with the studies of [26]. The two populations of O. volvulus sampled from the Volta Region in 1999 ranged from no IVM treatment to between four and eight IVM treatments. Genetic polymorphism decreased after treatment with IVM (Table 2) whereas the number of loci that deviated from HWE increased after IVM treatment. In non-treated populations, 15 of 28 loci were polymorphic, while in IVM treated samples, 14 of 28 loci were polymorphic (Table 2). The two populations of O. volvulus sampled from the Volta Region in 2002 ranged from no IVM treatment to between four and ten treatments. Similar to the samples collected in 1999, the 2002 samples also showed a loss of polymorphism after IVM treatment. In non-treated populations, 10 of 28 loci were polymorphic, while 8 of 28 loci were polymorphic in IVM treated samples (Table 2). A reduction in polymorphism is expected in treated samples undergoing selection. However, it is interesting to note a loss of polymorphism, from 15 to 10 polymorphic loci, between the non-treated samples collected in 1999 and 2002,

respectively. The majority of the worms used in the present study were adult females. With female worms, it is possible that microfilariae and stored sperm may contribute to the DNA sample. However, the amplification will be predominantly of the DNA derived from the large mass of the parent worm. The treated patients had been under IVM treatment for several years and the skin microfilarial counts were low (Osei-Atweneboana and Prichard, unpublished), indicating that most of the adult female O. volvulus from the treated patients would have had few microfilariae and sperm in utero. Thus, the contribution from sperm and microfilariae to the genotypes determined in the adult worms from treated patients should be minimal. In addition, the RFLP and SSCP patterns were consistent with diploid genotypes, and patterns consistent with three or more haploid genotypes were not detected. Taken together, the results suggest that the use of IVM in Ghana for onchocerciasis control is exerting selection pressure on O. volvulus and as a result is reducing genetic diversity. IVM causes dermal MF to decrease rapidly with a 95% reduction within 1 week of treatment. After 1 year, skin MF densities increase, but the levels are approximately 10–20% of pre-treatment levels. IVM affects adult O. volvulus fecundity and slows the release of MF from the uterus

62

B.F. Ardelli et al. / Molecular & Biochemical Parasitology 143 (2005) 58–66

Table 2 Summary of genetic polymorphism analysis of O. volvulus P-glycoprotein Locusa

Allelic sequences at locus 1999 No IVM

ITD1 ITD2 TM2 ITD4 TM5 Walker A1 Linker ITD9 TM8 TM9 ITD11 TM10 TM11 ITD14 Walker A2 ITD15

3 4 2 2 2 10 3 2 1 2 2 3 3 2 5 9

P-valueb 1999

HWEc

0.0007*d

0.0002*

0.0898 0.5153 0.005* 0.0235* 0.8955 0.403 0.8291 – 0.0028 0.5376 0.4423 0.4068 0.0002* 0.3620 0.2004

0.0008* 0.607 0.0025* 0.0253* 0.125 0.716 0.459 – 0.085* 0.884 0.756 0.584 0.0003* 0.753 0.996

IVM 3 4 2 2 3 8 1 2 1 2 2 3 2 3 2 10

Allelic sequences at locus 2002 No IVM

IVM

1 4 4 4 2 6 1 1 2 2 1 2 1 2 1 4

1 5 3 3 2 3 1 1 1 2 1 2 1 1 1 4

P-value 2002

HWE

– 0.1522 <0.001* 0.0075* <0.001* 0.0096* – – 0.1629 0.9734 – 0.18* – 0.1717 – 0.0449*

– 0.306 0.0008* 0.0045* 0.001* 0.074* – – 0.753 0.607 – 0.842 – 0.756 – 0.0008*

a Regions of O. volvulus P-glycoprotein which were examined for genetic polymorphism. These loci are indicated on Fig. 1. Only those regions which were polymorphic are shown. b P-value associated with Likelihood χ analysis of allele frequencies in treated and non-treated samples. The Bonferonni correction was applied to the p-values. c P-value associated with Hardy–Weinberg equilibrium analysis. d Significant.

[27]. A study by [3] demonstrated that monthly treatment with IVM for 3 years, in comparison to the standard annual treatment, killed more female worms. Thus, if IVM is selectively eliminating more susceptible adult worms this could reduce diversity in genes involved in resistance or genetically linked to other genes involved in resistance. For example, a study on tubulin [15] provided evidence of genetic selection by IVM on this gene. Since IVM has significant effects on O. volvulus fecundity, it is likely to select for females with genotypes that can produce MF in the presence of IVM, or for females that can recover their reproductive capacity much earlier. If IVM kills certain female worms, and affects fecundity in others, this might reduce the number of genotypes of O. volvulus transmitted and be reflected in an overall loss of gene diversity. Several of the loci examined showed differences in the frequencies of different gene sequences (Table 2). In the 1999 samples, the region designated ITD-1 (located at the 5 end before the first transmembrane domain) showed a significant difference in frequencies of allelic sequences (p = 0.0007); variant AA was in higher frequency in non-treated samples, while variant AB was in higher frequency in treated samples (Fig. 2A). Similarly, the region designated ITD-2 also showed a significant difference in frequency of allelic sequences (p = 0.0898); in this region of OvPGP, variant AB and variant BC were in higher frequency in treated worms. Also, the region designated ITD-4 showed a significant difference in frequency (p = 0.005) of allelic sequences; however, variant AB was in higher frequency in non-treated worms. Similar results were obtained for the region designated TM5, with variant AB showing a significantly higher frequency

in non-treated worms. The Walker A motif, located in the first nucleotide binding domain, as well as in the region designated ITD-15, located at the 3 end of the gene, were very polymorphic. Although there was a reduction in polymorphism between the treated and non-treated samples in these regions, this was not significant (p = 0.3620 for Walker A and p = 0.2004 for ITD-15). The region designated ITD14, located within the second nucleotide binding domain between the Walker A and B motifs, showed a significant (p = 0.0002) difference in frequency of allelic sequences. One variant, AB, was only found in worms removed from IVM treated patients. A large region of P-glycoprotein, located between the Walker B motif of the first NBD and TM9, was polymorphic in worms removed from non-treated patients; a reduction in polymorphism was observed in worms removed from patients taking IVM. In these regions of the gene, three sequence variants were detected in non-treated worms, while only one variant was detected in treated worms. There was no significant difference in frequency of allelic variants in this region of P-glycoprotein. The sequence variants detected in the treated worms were the most common variants detected in non-treated worms. As well, the two variants that were not detected in worms removed from treated patients were in low frequency in worms removed from non-treated patients. The same 28 loci of P-glycoprotein were examined from a population of O. volvulus collected in 2002. These samples were from the same villages that were sampled in 1999. Although several regions of this gene were polymorphic in the 2002 samples, only five loci showed a significant selection after IVM treatment (Table 2). In contrast to the samples collected in 1999, samples from 2002 showed four allelic

B.F. Ardelli et al. / Molecular & Biochemical Parasitology 143 (2005) 58–66

sequences at the second transmembrane domain. Variant AA was in significantly higher frequency in treated worms, while variant AB was in higher frequency in non-treated worms (p ≤ 0.001). The region designated ITD-4 was very polymorphic. Four allelic sequences were detected in the 2002 samples, while only two allelic sequences were detected in the 1999 samples. Variant AC was in significantly higher frequency in samples removed treated patients. Similar to the 1999 samples, transmembrane domain five was also polymorphic in the 2002 samples. In contrast to the samples collected in 1999, variant AB was in significantly higher frequency in treated samples removed from patients in 2002. Similar to the 1999 samples, the first Walker A domain and the locus designated ITD-15 were also polymorphic (Table 2).

63

Three other ABC transporter genes cloned from O. volvulus also showed a significant reduction in polymorphism after treatment with IVM [22,28]. These genes, designated OvMDR-1, OvABC-1 and OvABC-3, were examined in the same population of O. volulus used in the present investigation. However, in contrast to the present study, only a small region, ranging between 300 and 350 bp, within the first NBD was examined in the other ABC transporters. Eng and Prichard [15], using restriction fragment length polymorphism analyses, examined a 356 bp region, located within the linker region, of O. volvulus P-glycoprotein. A significant difference in genetic polymorphism was detected between non-treated and IVM treated worms. This study, combined with the results of the previous studies, demonstrate that IVM is reducing polymorphism in four O. volvulus ABC trans-

Fig. 2. (A–J) Single strand conformation polymorphism analysis (SSCP) or restriction fragment length polymorphism analysis (RFLP) of the 1999 and 2002 samples of O. volvulus P-glycoprotein. Regions which show a reduction in polymorphism after treatment with IVM are represented. A–B: SSCP analysis of ITD-1, the region located at the 5 end of the gene, before the first transmembrane domain; C–D: SSCP analysis of the second transmembrane domain; E–F: RFLP analysis of the Walker A motif located within the first nucleotide binding domain; G–H: SSCP analysis of the eleventh transmembrane domain; I–J: RFLP analysis of ITD-15, the region located at the 3 end of the gene, after the second nucleotide binding domain.

64

B.F. Ardelli et al. / Molecular & Biochemical Parasitology 143 (2005) 58–66

Fig. 2. (Continued ).

porter genes. It is not known if IVM resistance is caused by a single gene or is a result of multigene effects. IVM resistance in one gastrointestinal nematode isolate appeared to be mediated by a single gene or gene locus with primarily dominant effects [29]. Many C. elegans genes are arranged in operons, and this includes ABC transporter genes. C. elegans operons contain on average 2.6 genes and the majority of downstream genes in operons were usually trans-spliced by SL2 [30]. Also, functionally related genes in C. elegans were found together in operons. Thus, if O. volvulus genes are also arranged in operons, this might account for the observed reduction in polymorphism in several ABC transporters. For example, if IVM is selecting on one gene, and the others are in close proximity, these genes may also be affected because they are “linked”. In C. elegans, operons appear to be a means to co-regulate functionally related proteins. If the O. volvulus ABC transporter genes are functionally related, it may account for the similar effects observed in the four ABC transporter genes. There was also a loss of genetic polymorphism in the worms removed from non-treated patients between the 1999 and 2002 collection period. Considering the prolonged widespread use of IVM in Ghana and the effects IVM has on reproduction in O. volvulus, it is possible that patients, in the few communities not being treated with IVM, may become selectively infected with larvae that were progeny of worms better able to reproduce, despite IVM treatment. Alternatively, IVM may be selectively eliminating adult worms with specific P-glycoprotein alleles (e.g. susceptible), allowing only relatively IVM-tolerant worms to produce MF. Either

or both of these mechanisms could account for a reduction in gene diversity in non-treated patients. Although the interval between the two collection periods (i.e. 1999 and 2002) is not long, it is sufficient for three generations of O. volvulus [31,32]. If IVM kills certain female worms and affects fecundity in others, this might have an effect on the genotype of O. volvulus transmitted and be reflected in a loss of gene diversity. Four of the 28 loci examined from the 1999 samples (i.e. ITD-1, ITD-4, TM5, ITD-14) and five of 28 loci from the 2002 samples (i.e. TM2, ITD-4, TM5, WA1, ITD-5) showed significant selection following treatment with IVM. The majority of these regions were located in the putative transmembrane helices (i.e. Fig. 2A–J; ITD-1, ITD-2; ITD4; ITD-15; ITD-14), regions that are not normally associated with drug resistance. Several allelic sequences were found in the fifth and ninth transmembrane domains, as well as in the Walker A sites of the NBD (Table 2). There is a great deal of information available on the relationship between SNPs and functional polymorphisms in human MDR1 Pgp. For example, three serine residues located in the linker region of P-glycoprotein are targets for in vivo phosphorylation [33]. It was suggested that phosphorylation of the linker region modulates the interaction of some drugs with MDR1, particularly at low concentrations. In a different study, it was demonstrated that mutations in TM9 affected drug resistance in the DC-3F/AD11 cell line. This cell line displays high levels of resistance to actinomycin D and vincristine. The study demonstrated that TM9 cooperated with TM6 to mediate drug resistance [34]. Detailed kinetic studies by [35] demonstrated

B.F. Ardelli et al. / Molecular & Biochemical Parasitology 143 (2005) 58–66

that only one NBD site is able to hydrolyse ATP at any given time. When one site binds ATP, it induces a conformational change that reduces the ATP affinity in the second ATP site. MDR1 exhibits a drug-stimulated ATP hydrolytic activity, which is closely related to drug transport function. Thus, the alternate catalytic cycling of ATP hydrolysis is coupled to the development of drug resistance. Based on these studies, it is tempting to speculate that mutations found in OvPGP, particularly those found at a significantly higher frequency in drug-treated worms, could be related to IVM tolerance. However, functional expression and characterization of wild and mutant P-glycoproteins must be performed. The changes in allelic patterns and a reduction in diversity at many loci in P-glycoprotein in the parasites removed from IVM treated patients in 1999 and 2002 suggest that IVM is affecting this gene. Since IVM treatment was only implemented as part of the OCP in 1988, O. volvulus populations in Ghana would have been exposed to a maximum of 11 years of IVM treatment for the 1999 samples and 14 years for the 2002 samples, which could include up to 14 generations of O. volvulus. The hallmarks of resistance selection, in particular a reduction in genetic diversity are apparent in these O. volvulus populations from the Volta Region of Ghana. Although the patterns of polymorphism varied for each population, the most dramatic loss of polymorphism was observed for ITD1 (Fig. 2A and B), TM2 (Fig. 2C and D), the Walker A site within the first nucleotide binding domain (Fig. 2E and F), TM11 (Fig. 2G and H) and ITD-15 (Fig. 2I and J). In the 1999 samples, three allelic variants were detected for ITD-1. The 2002 samples show a loss of diversity at this locus, with the most common allelic variant increasing in frequency. An allelic variant, AA, in TM2, which was not detected in the 1999 samples and was presumably rare, became very common in the 2002 treated samples. The ITD-1 (from 3 variants to 1), the Walker A site (from 8 variants to 3), TM11 (from 2 variants to 1) and ITD-15 (from 10 variants to 4) showed the most dramatic reduction in diversity between the 1999 and 2002 samples. So far resistance to IVM in human populations infected with O. volvulus has not been unequivocally shown. However, several recent reports have indicated sub-optimal responses to IVM in some patients in the Pru and Lower Black Volta river basins of Ghana [4,36] and in the Khartoun Region of Sudan [27]. Should IVM resistance develop, it would be difficult to recognize, but it would likely manifest as an enhanced viability of microfilariae, a more rapid return to fecundity in female worms, or greater survival and microfilarial production of some adult worms following repeated IVM treatment. A definitive marker for IVM resistance selection was not detected in OvPGP. However, the results suggest that a sequence under selection by IVM is in close proximity to OvPGP. The results of the present study demonstrate that IVM is causing a significant loss of polymorphism in populations of O. volvulus in the Volta region of Ghana. This is consistent with a possible development of IVM resistance.

65

Acknowledgements This investigation received financial support from the African Programme for Onchocerciasis Control, the Onchocerciasis Control Program in West Africa and the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases. BF Ardelli was supported by a postdoctoral fellowship from the Natural Sciences and Engineering Research Council (NSERC) of Canada and bridging funds from a Fonds Qu´eb´ecois de la recherche sur la Nature et les Technologies (FQRNT) Centre Grant (Centre for Host-Parasite Interactions). SB Guerriero was supported by an NSERC undergraduate research award. We gratefully acknowledge Dr. K. Awadzi for collecting O. volvulus nodules from Ghana in 1999 and Mr. M. Osei-Atweneboana for collecting the nodules in Ghana in 2002.

References [1] Albiez EJ, Walter G, Kaiser A, et al. Histological examination of onchocercomata after therapy with ivermectin. Trop Med Parasitol 1998;39:93–9. [2] Dadzie Y, Neira M, Hopkins D. Final report of the conference on the eradicability of Onchocerciasis. Filaria J 2003;2:2. [3] Gardon J, Boussinesq M, Kamgno J, Gardon-Wendel N, DemangaNgangue, Duke BO. Effects of standard and high doses of ivermectin on adult worms of Onchocerca volvulus: a randomised controlled trial. Lancet 2002;360:203–10. [4] Awadzi K, Boakye DA, Edwards G, et al. An investigation of persistent microfilaridermia despite multiple treatments with ivermectin in two onchocerciasis endemic foci in Ghana. Ann Trop Med Parasitol 2004;98:231–49. [5] Higgins CF. ABC transporters: from microorganisms to man. Annu Rev Cell Biol 1992;8:67–113. [6] Didier A, Loor F. The abamectin derivative ivermectin is a potent P-glycoprotein inhibitor. Anticancer Drugs 1996;7:745–51. [7] Pouliot JF, L’Heureux F, Liu Z, Prichard RK, Georges E. Reversal of P-glycoprotein-associated multidrug resistance by ivermectin. Biochem Pharmacol 1997;53:17–25. [8] Schinkel AH, Smit JJ, van Tellingen O, et al. Disruption of the mouse mdr1a P-glycoprotein gene leads to a deficiency in the blood-brain barrier and to increased sensitivity to drugs. Cell 1994;77:491–502. [9] Schinkel AH, Wagenaar E, van Deemter L, Mol CA, Borst P. Absence of the mdr1a P-glycoprotein in mice affects tissue distribution and pharmacokinetics of dexamethasone, digoxin, and cyclosporin A. J Clin Invest 1995;96:1698–705. [10] Lincke CR, Broeks A, The I, Plasterk RH, Borst P. The expression of two P-glycoprotein (pgp) genes in transgenic Caenorhabditis elegans is confined to intestinal cells. EMBO J 1993;12:1615–20. [11] Broeks A, Janssen HW, Calafat J, Plasterk RH. A P-glycoprotein protects Caenorhabditis elegans against natural toxins. EMBO J 1995;14:1858–66. [12] Xu M, Molento M, Blackhall W, Ribeiro P, Beech R, Prichard R. Ivermectin resistance in nematodes may be caused by alteration of P-glycoprotein homolog. Mol Biochem Parasitol 1998;91:327–35. [13] Broeks A, Gerrard B, Allikmets R, Dean M, Plasterk RH. Homologues of the human multidrug resistance genes MRP and MDR contribute to heavy metal resistance in the soil nematode Caenorhabditis elegans. EMBO J 1996;15:6132–43. [14] Huang YJ, Prichard RK. Identification and stage-specific expression of two putative P-glycoprotein coding genes in Onchocerca volvulus. Mol Biochem Parasitol 1999;102:273–81.

66

B.F. Ardelli et al. / Molecular & Biochemical Parasitology 143 (2005) 58–66

[15] Eng KLJ, Prichard RK. A comparison of genetic polymorphism in populations of Onchocerca volvulus from untreated- and ivermectintreated patients. Mol Biochem Parasitol, in press. [16] Kerboeuf D, Blackhall W, Kaminsky R, von Samson-Himmelstjerna G. P-glycoprotein in helminths: function and perspectives for anthelmintic treatment and reversal of resistance. Int J Antimicrob Agents 2003;22:332–46. [17] Duke BO, Zea-Flores G, Castro J, Cupp EW, Munoz B. Effects of multiple monthly doses of ivermectin on adult Onchocerca volvulus. Am J Trop Med Hyg 1990;43:657–64. [18] Duke BO, Zea-Flores G, Castro J, Cupp EW, Munoz B. Effects of 3 month doses of ivermectin on adult Onchocerca volvulus. Am J Trop Med Hyg 1992;46:189–94. [19] Awadzi K, Attah SK, Addy ET, Opoku NO, Quartey BT. The effects of high-dose ivermectin regimens on Onchocerca volvulus in onchocerciasis patients. Trans R Soc Trop Med Hyg 1999;93:189–94. [20] Juliano RL, Ling V. A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 1976;455:152–62. [21] Schulz-Key H, Albiez EJ, Buttner DW. Isolation of living adult Onchocerca volvulus from nodules. Tropenmed Parasitol 1977;28: 428–30. [22] Ardelli BF, Prichard RK. Identification of variant ABC transporter genes among Onchocerca volvulus collected from ivermectin treated and untreated patients in Ghana, West Africa. Ann Trop Med Parasitol 2004;98:371–84. [23] Germann UA. P-glycoprotein—a mediator of multidrug resistance in tumour cells. Eur J Cancer 1996;32A:927–44. [24] Hrycyna CA, Airan LE, Germann UA, Ambudkar SV, Pastan I, Gottesman MM. Structural flexibility of the linker region of human P-glycoprotein permits ATP hydrolysis and drug transport. Biochemistry 1998;37:13660–73. [25] Shapiro MB, Senapathy P. RNA splice junctions of different classes of eukaryotes: sequence statistics and functional impli-

[26] [27]

[28]

[29]

[30] [31] [32] [33]

[34]

[35]

[36]

cations in gene expression. Nucleic Acids Res 1987;15:7155– 74. Unnasch TR, Williams SA. The genomes of Onchocerca volvulus. Int J Parasitol 2000;30:543–52. Ali MM, Mukhtar MM, Baraka OZ, Homeida MM, Kheir MM, Mackenzie CD. Immunocompetence may be important in the effectiveness of Mectizan (ivermectin) in the treatment of human onchocerciasis. Acta Trop 2002;84:49–53. Ardelli BF, Prichard RK. Reduced genetic variation of an Onchocerca volvulus ABC transporter gene following treatment with ivermectin. Trans Roy Soc Trop Med Hyg, in press. Dobson RJ, LeJambre L, Gill JH. Management of anthelmintic resistance: inheritance of resistance and selection with persistent drugs. Int J Parasitol 1996;26:993–1000. Blumenthol T, Evans D, Link CD, et al. A global analysis of Caenorhabditis elegans operons. Nature 2002;417:851–3. Schulz-Key H. Observations on the reproductive biology of Onchocerca volvulus. Acta Leidensia 1990;59:27–43. Duke BOL. The population dynamics of Onchocerca volvulus in the human host. Trop Med Parasitol 1993;44:62–8. Szab´o K, Bakos E, Welker E, et al. Phosphorylation site mutations in the human multidrug transporter modulate its drug-stimulated ATPase activity. J Biol Chem 1997;272:23165–71. Song J, Melera PW. Transmembrane domain (TM) 9 represents a novel site in P-glycoprotein that affects drug resistance and cooperates with TM6 to mediate [125 I]iodoarylazidoprazosin labeling. Mol Pharm 2001;60:254–61. Sauna ZE, Smith MM, M¨uller M, Kerr KM, Ambudkar SV. The mechanism of action of multidrug-resistance-linked P-glycoprotein. J Bio Biomembr 2001;33:481–91. Awadzi K, Attah SK, Addy ET, et al. Thirty-month follow-up of sub-optimal responders to multiple treatments with ivermectin, in two onchocerciasis-endemic foci in Ghana. Ann Trop Med Parasitol 2004;98:359–70.