The Journal of Heredity 2001:92(3)
© 2001 The American Genetic Association 92:248-253
Widespread Heteroplasmy in Schistosomes Makes an mtVNTR Marker "Nearsighted"
From the Department of Biological Sciences, Purdue University, West Lafayette, IN, 47907 (Curtis and Minchella), CENBIOS, UNIVALE, Governador Valadares, Minas Gerais, Brazil (Fraga), and Centro de Pesquisas Renée Rachou, Fundação Oswaldo Cruz, Belo Horizonte, Minas Gerais, Brazil (de Souza and Corrêa-Oliveira).
Address correspondence to Jason Curtis at the address above or e-mail: jcurtis{at}bilbo.bio.purdue.edu..
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mitochondrial markers are often hailed as the preferred DNA elements for analyses of population subdivision. To this end we have employed a mitochondrial repeat element to examine the population structure in Schistosoma mansoni (human blood flukes). Schistosome isolates were collected from each of 21 different patients representing seven different areas of a Brazilian village. These parasite isolates demonstrate substantial genetic polymorphism, with an average of 10 genotypes infecting each patient, which is more readily detected because of high levels of heteroplasmy (i.e., 72.5% of the individual worms exhibit multiple versions of this repeat region with different numbers of repeats). Due to the high number of common haplotypes in the population, this repeat element from S. mansoni has a large proportion (47%) of its genetic variation described by differences among mitochondrial genomes within individual worms. However, when only rare haplotypes are considered, population structure can be detected. It seems that heteroplasmy in the schistosome population of Melquiades is both the source of plentiful genetic variation and a confounding factor in the analysis of that variation. Thus the schistosome population in Melquiades may actually be more strongly subdivided than we are able to detect using this mitochondrial marker.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mitochondrial DNA elements are often regarded as one of the preferred types of markers for studies of population subdivision, parentage, and clonality, due to their rapid nucleotide substitution rate and (typically, at least in most animals) uniparental inheritance (Avise 1994). In particular, recent interest has focused on the control region, or D-loop, of the mitochondrial genome (Stoneking 2000). This region may contain a site of rapid substitution and length heterogeneity caused by variable numbers of tandem repeats, though the mutational processes that underlie this variability are unclear (reviewed by Lunt et al. 1998). Haplotype variation in these mtVNTRs can be measured by either polymerase chain reaction (PCR)- or random fragment length polymorphism (RFLP)-based methods, and methods exist for analyzing these data for studies of population subdivision (Birky et al. 1989).
One such mtVNTR is found in the genome of the digenetic blood fluke, Schistosoma mansoni Sambon, 1907, and the sequence of this 62 bp repeat corresponds to the repetitive element of pSM750 (Spotila et al. 1991). This VNTR was described from the S. mansoni mitochondrial genome by Peña et al. (1995), who noted that its presence would explain the mtDNA length variation observed in this species (Depres et al. 1991). This repeat sequence appears to be localized in the control region of the mitochondrial genome, but this portion of the mtDNA has not been fully sequenced (Le et al. 2000), so the exact location is unknown. The marker has been used to demonstrate overdispersion of the parasite within its intermediate host snails (Biomphalaria glabrata) at two infection foci in Brazil (Minchella et al. 1995). One reason this repeat element was useful in the study of parasite aggregation was that there was a substantial amount of haplotype diversity within a relatively small sample size, making it quite feasible to detect differences between individual parasites within a single host. During that study it also became clear that individual worms contained multiple versions of the element, each with a different number of repeats, which has been taken as evidence of heteroplasmy, as has been described for other animal mtDNA (Rand 1994).
The likelihood of observing population subdivision in natural populations is increased by several features of the schistosome life cycle. Adult S. mansoni worms live in the veins surrounding the intestines and liver of more than 80 million people worldwide. The worms are dioecious, with the female laying about 300 eggs/day. These eggs may become caught up in the tissues of the human host, where the immune response can cause severe organomegaly in chronic cases. Eggs that are released in the patients' feces hatch in freshwater, and in areas of poor sanitation, the parasites can infect freshwater snails in nearby streams and ponds. The parasites reproduce asexually within the snail intermediate host, releasing free-living stages that can infect human hosts that have contact with contaminated water, thus completing the life cycle.
Because of the dependence on aquatic intermediate hosts to complete the life cycle, one might expect schistosome populations to cluster around streams that meet the requirements of both suitable snail habitat and contamination by human feces. The movements of infected humans counteract this localization. In rural villages in Brazil, houses are often clustered around watercourses, with many residents working nearby and rarely venturing far from their homes. By examining schistosomes that infect human hosts living near separate watercourses within a single village, it should be possible to determine the degree to which the parasite population is genetically subdivided. Given that schistosome populations already exhibit polymorphism for traits such as drug resistance and pathology, such subdivision might have significant epidemiologic consequences (Curtis and Minchella 2000).
In this study we sample parasites in human hosts within a single Brazilian village and describe the results of our attempts to characterize population subdivisions for these parasites. First, we examine the number of different parasite genotypes that infect each of 21 human patients from a single village using the hypervariable mtVNTR marker. A single worm's genotype is represented by the collection of haplotypes (i.e., repeats of different sizes) that the worm exhibits, without an estimate of the relative frequencies of those mitochondrial haplotypes within the cells of the individual worm. Next, we analyze the distribution of the haplotypes within and among individual worms to determine the extent of heteroplasmy and haplotype diversity for schistosomes in this village. Finally, we measure the degree of population subdivision for the parasites by investigating the way in which haplotypes are distributed within and between distinct areas of the village. We thus assess whether this mtVNTR element in S. mansoni is well suited to studies of population subdivision.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Field Collection of Schistosomes
Because it is not possible to directly assess the genotypes of the schistosomes within human patients (i.e., they are sequestered deep within the circulatory system), parasite eggs are raised through a single generation of laboratory-reared snails and mice. To this end, fecal samples were collected in October 1997 from 40 individuals who live in the municipality of Melquiades, in the northeast of the state of Minas Gerais (approximately 500 km from Belo Horizonte), Brazil. An earlier survey (July 1997) of this village estimated schistosome prevalence at 75% in human hosts. Patients were chosen at random from a list of people who were positive for schistosome eggs in the July survey. Melquiades is a collection of boroughs along several independent tributaries that flow into a larger stream over a distance of 5 km, such that houses are arranged in clusters separated by agricultural land (Figure 1). Patients were selected so that each of seven major boroughs was represented (Table 1) while maintaining their relative proportions in the whole village.
|
).
|
Each patient deposited their feces over the course of 24 h into 750 ml plastic jars. Feces were shipped on ice to the Laboratory of Malacology, Centro de Pesquisas Renè Rachou, Belo Horizonte, Brazil. For each patient isolate, the feces were suspended in 0.85% saline and washed with dechlorinated water at room temperature. The sediments were exposed to light for at least 1 h to induce eclosion of the S. mansoni eggs. Batches of laboratory-reared snails (Biomphalaria glabrata) were exposed to the larvae (miracidia) from the 22 successfully hatched isolates for 6–15 h. Each batch of snails (representing a patient isolate) was maintained independently and transported to Purdue University, West Lafayette, Indiana.
At 4–6 weeks postinfection, snails from each isolate were exposed to light for at least 4 h to encourage emergence of larval stages (cercariae). Cercariae were collected from the snails of each isolate on at least two separate days. Batches of these cercariae (approximately 125 larvae in 15 ml of water) were used to infect BALB/C mice (optimally four mice per patient isolate given sufficient numbers of infected snails). Infected snails of all but one patient isolate survived to release cercariae. Mice representing the 21 patient isolates were euthanized at 45 days postinfection (i.e., the minimum time for development and sexual maturation of schistosomes in the vertebrate host), and schistosome worms were collected from the hepatic portal vein and mesenteries by dissection. Worms were stored at -80°C for genetic analysis.
Genetic Analysis of Schistosomes
For each patient isolate, we tried to obtain the genotype (defined by the profile of pSM750 haplotypes) for a minimum of 25–30 worms, though this number was sometimes limited by the availability of male worms. Male worms were used exclusively because they yield substantially more DNA for analysis due to the sexual dimorphism in schistosomes. Other experiments in our laboratory have already verified that male and female worms are both heteroplasmic, and that offspring of both sexes exhibit the mitochondrial profile of their mother worm (i.e., the collection of pSM750 haplotypes has high heritability).
DNA was extracted from individual schistosomes as in Sorensen et al. (1998). Briefly, worms were placed in 1.5 ml microcentrifuge tubes and pulverized on dry ice with a chilled pestle. The pestle was rinsed with a buffer containing 100 mM NaCl, 25 mM sucrose, 10 mM EDTA, and 2% (m/v) SDS in 50 mM Tris pH 8.0 (Brindley et al. 1989). This mixture was lysed at 65°C for 30 min, then 8 M KOAc was added to a final concentration of 1 M, and lastly, it was chilled on ice for 30 min. Salts were removed from the mixture by spinning at 12,000 rpm for 10 min. Standard ethanol precipitation practices (Sambrook et al. 1989) were used to pellet the DNA from the supernatant. Pellets were resuspended in 16 ml of water (a 3 ml aliquot of this was further diluted 1:10 with water and stored at -80°C for future studies).
DNA samples were completely digested with the endonuclease RsaI, according to the manufacturer's specifications, at 37°C for 1 h. Digested DNA fragments were electrophoresed through 1% agarose gels at 44 V for 14.5 h (along with a 100 bp ladder as a size standard) and transferred to nylon membranes by Southern blot techniques. Nylon membranes were prehybridized in 6x SSC, 5x Denhardt's, 0.1% SDS, and 6% (w/v) PEG for at least 4 h at 65°C. Purified pSM750 was labeled with [32P]dCTP using an oligolabeling kit and collected by separation on a G-50 Sephadex column. Filters were washed four times with 4x SSC, 0.2% SDS for 20 min each and once with 4x SSC for 20 min, then placed with autoradiography film at -80°C for at least 24 h.
DNA fragments on the autoradiograph were compared to the 100 bp ladder for sizing (all fragments contain a constant amount of flanking DNA and a multiple of the 62 bp repeat). Fragments could be reliably scored between 150 and 1700 bp (i.e., from 2 to 26 repeats). All bands were scored as present/absent without an estimate of frequency. Because the frequency of worms within the original human patient is obscured by stochasticity in the laboratory infections (e.g., efficiency of the parasites' asexual reproduction in infected snails), each distinct worm genotype (haplotype banding profile) was counted only once per patient isolate for calculations of genetic diversity. Thus within each patient isolate, all detectable genotypes were assumed to occur with equal frequency.
Population Genetic Analyses
The numbers of bands corresponding to each repeat size were summed across the unique genotypes within each patient isolate for the entire village. These data were used to plot frequency distributions of repeat sizes for homoplasmic and heteroplasmic worms. A Kolmogorov–Smirnov two-sample test (one-tailed test for large samples) was used to determine if larger repeats are more frequent in heteroplasmic individuals.
The analysis of genetic subdivision in the parasite population was performed at two different spatial levels: within each of the seven boroughs, and within individual human patients. These analyses followed the method of Cesaroni et al. (1997), in that genetic diversity was calculated at each level of the population hierarchy as K = 1 - 
xi2, where xi is the frequency of a haplotype with i repeats, giving values for Kb (within individual worms), Kc (within boroughs or within patients), and Kt (entire village) (Birky et al. 1989). Hierarchical C statistics were calculated to find the proportion of the total genetic variation contained within individuals [Ci = mean Kb/Kt], among individuals within subpopulations [Cip = (mean Kc - mean Kb)/Kt], and among subpopulations [Cpt = (Kt - mean Kc)/Kt] (Lewontin 1972).
Another hierarchical analysis was performed using only rare haplotypes (Slatkin 1985) in order to remove some of the "noise" created by ubiquitous haplotypes. In this case, "rare" means that the haplotype was present in 10 or fewer distinct genotypes in the population. This second analysis again considered subdivision at the level of boroughs and for individual human patients.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Table 1 summarizes the results for the 211 distinct genotypes detected in 460 male schistosomes from 21 patients in the village of Melquiades. Patients were generally infected by a diverse collection of schistosome genotypes (range = 3–19, mean ± SE = 10.04 ± 0.81). A large majority of these schistosomes were heteroplasmic, exhibiting up to five different repeat copies of the pSM750 element (village-wide heteroplasmy = 72.5%, mean haplotypes/genotype = 2.24), so that a total of 465 haplotypes were scored. The cumulative frequency distribution for pSM750 repeat copy number (Figure 2) is skewed toward haplotypes with smaller numbers of repeats (mean = 8.6 repeats, mode = 4 repeats). However, the distributions of repeat copy numbers for haplotypes in heteroplasmic and homoplasmic individuals are not significantly different (Kolmogorov–Smirnov:
2 = 5.42, P > .05), indicating that haplotypes with larger numbers of repeats are not more common in heteroplasmic individuals in this population.
|
Table 1 also shows the values for genetic diversity for the pSM750 marker within each host and the averages across all hosts in the village. The average genetic diversity for individual worms, Kb, is calculated for all of the schistosome genotypes within each patient and also for the entire village (mean Kb = 0.432). If each patient is taken to contain a subpopulation of the total schistosome population in the village, then Kc, the genetic variation represented by each patient's infection, can be calculated and averaged across all patients (mean Kc = 0.867). These two values, Kb and Kc, along with Kt ( = 0.922), the total genetic variation in the village, are used to calculate the C statistics describing the degree of genetic subdivision. Differentiation between haplotypes within a single worm (Ci) describes 46.9% of the total variation present in the village, whereas the differentiation between schistosomes infecting the same patient (Cip) explains a further 47.1% of the variation. This means that only 6.0% of the genetic variation at pSM750 can be attributed to differences between the infection in different patients (Cpt).
Table 2 displays the values for genetic variation if the village is partitioned into seven geographically distinct boroughs (with two to five patients in each borough). These values are used in the calculation of C statistics, as above. Ci remains the same (0.469), while Cip is larger (0.506), meaning that subdivision by geography describes even less of the total genetic variation with this marker (Cpt = 0.025).
|
The hierarchical statistics changed appreciably when only rare haplotypes were used in the analysis (Table 3). Haplotypes meeting this criteria were found in 19 of the 21 patient isolates, and the frequency of heteroplasmy (i.e., the occurrence of two rare haplotypes in the same worm's genotype) dropped to 28.3%, with a mean of 1.35 haplotypes/genotype. Under this analysis, differences between haplotypes within the average worm describe a smaller proportion of the overall variation (Ci = 0.17). When human patients are considered to be the subpopulations, Cip is similar (0.467), but Cpt is much higher (0.365). If the subpopulations are taken to be the seven boroughs, both Cip and Cpt increase when only rare haplotypes are used (0.695 and 0.138, respectively).
|
Regardless of the hierarchical analyses, worms with the same genotype are almost never found in two different patients. Thus the haplotype profiles that describe individual genotypes in a patient's schistosome infection do not appear to be correlated with any epidemiologic parameters. These include household, age, sex, geography, or patient worm burden as estimated from counts of schistosome eggs in the feces (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The most remarkable feature of the pSM750 mtVNTR element in S. mansoni from human patients in this Brazilian village is the amount of genetic diversity that can be resolved at relatively small scales. This feature made the marker useful when it was desirable to detect differentiation between individual parasites in the same snail host (Minchella et al. 1995). Given this ability to distinguish individuals, the element also seemed well suited as a marker for population subdivision. If there is any substantial clustering of the parasites in the village, an mtVNTR marker should detect patterns in the distribution of haplotypes or genotypes (represented as sets of haplotypes) that can be used to differentiate subpopulations. In this study we found such a wealth of genetic variation with this marker that patterns in the variation are potentially becoming obscured by a set of common haplotypes.
Mitochondrial VNTR markers appear to be among the most variable genetic elements used for population studies, but the process by which high levels of variation are generated remains unknown. Attempts have been made to fit different mutational models to observed mtVNTR distributions with mixed success (Lunt et al. 1998). For example, strand slippage during replication might be responsible for small shifts in repeat copy number and a means by which new haplotypes are generated. One particular phenomenon that would seem to support this model of mutation has been observed in studies of fish mtVNTR elements: individuals that are heteroplasmic tend to contain elements with a higher number of repeats (Cesaroni et al. 1997). Thus if mistakes are more likely to occur during the replication of larger repeat regions, new haplotypes (and thus heteroplasmy) would be generated more frequently by larger haplotypes. While this idea is intuitively pleasing, we find that the distribution of repeat numbers within pSM750 elements is not significantly different between homoplasmic and heteroplasmic worms in the Melquiades population (Figure 2). Other mechanisms have been implicated in the generation of variability in VNTR elements, and these could influence the level of heteroplasmy detected with pSM750. For example, unequal crossing over would not traditionally be invoked as a mechanism that could generate variability in mtVNTR, but recent evidence suggests that recombination might actually occur between mitochondrial genomes (Awadalla et al. 1999; Hey 2000). Our laboratory continues to investigate the mechanism of mutation for the pSM750 element through analysis of S. mansoni worms that are clonal siblings produced by asexual reproduction in the snail host.
While the amount of schistosome genetic diversity contained within a single vertebrate host has been difficult to estimate (especially in human patients), the data from Melquiades provide further evidence that the diversity is substantial. The accumulation of many different genotypes within individual vertebrate hosts could have important consequences for the long-term maintenance of genetic variation in schistosome species, since the vertebrate host is the site of sexual reproduction. In fact, patients in Melquiades can probably be thought of as "genetic mixing bowls" for the parasite. Patients are the most likely agents of schistosome dispersal; they move much more than snails, and they potentially sample a variety of different parasite larvae in a short time, bringing together a wide range of parasite genotypes. While this is the first report of schistosome diversity within human patients, there seems to be a similar situation for schistosomes infecting rats, with 78 genotypes infecting 10 hosts on the island of Guadeloupe (Barral et al. 1996). The estimate of population structure from their study (genotypic diversity within a single rat host, HW = 0.251, and between rat hosts, HB = 0.053) is similar to our estimate from the hierarchical analysis of all haplotypes, despite differences in genetic marker (they employed eight polymorphic RAPD markers) and host distribution (all of their rats came from a single locality).
The effect that a high amount of intrahost diversity could have on the epidemiology of schistosomiasis remains unknown. Evolutionary theory dictates that where infection with multiple genotypes is common, the most successful genotypes will be those that reproduce at the highest rate, as long as their eggs are released before the host is severely damaged or killed (Frank 1996). If a schistosome genotype responds to the presence of other genotypes in their host by laying eggs at an increased rate, this could affect acute pathogenicity because most of the symptoms are caused by the response to eggs trapped in host tissues. Of course, the logic behind this prediction becomes confounded if the various genotypes infecting a single host are closely related, since theory would then predict that genotypes should "cooperate" more in exploiting host resources. In this light, the results of the current research underscore the value of tracking individual parasite genotypes and their relatedness within hosts.
In terms of following individual genotypes, the pSM750 mtVNTR marker is excellent, but the analysis of schistosomes in Melquiades indicates that this element might be less well suited for a study of population genetic structure. One particular feature of this repeat element might be responsible for obscuring the population structure in Melquiades: there is a set of haplotypes that is common to many of the genotypes. The effect of this phenomenon is that individual heteroplasmic worms tend to contain one or more of these common haplotypes, thereby representing a significant fraction of the population's genetic variation (i.e., Ci is increased). When these haplotypes are excluded from the hierarchical analysis, the schistosome population appears to be more strongly subdivided. However, haplotypes—whether common or rare—appear to be "packaged together" in genotypes that are passed from generation to generation, and it remains true that there were no correlations between worm genotypes and geographical clusters. This quandary highlights some of the limitations of indirect methods for estimating gene flow and population structure (Bossart and Prowell 1998). It seems that heteroplasmy in the schistosome population of Melquiades is both the source of sufficient genetic variation and a confounding factor in the analysis of that variation.
While heteroplasmy due to mtVNTR variation has now been documented in a number of organismal systems, few studies have considered the utility of mtVNTR markers in measuring population structure, and methods for their application are not fully established, though markers that detect similar levels of heteroplasmy and intraindividual diversity have provided similar results (reviewed by Lunt et al. 1998). In a study of the sea bass (Dicentrarchus labrax), mtVNTR markers gave a Ci value of 0.291 for a population with 52.1% heteroplasmy and a mean Kb of 0.285; Cpt was 0.043 for eight populations around the Mediterranean Sea (Cesaroni et al. 1997). When all of the haplotypes are considered in the hierarchical analysis, the Kb and Ci levels are even higher for schistosomes in Melquiades.
Two important features of populations that might result in subpopulations remaining undifferentiated (or less differentiated) are gene flow and mutation. For schistosomes, it is assumed that the majority of the parasites' gene flow is due to movements of the human patients. One would then be tempted to conclude that human hosts are frequently dispersing schistosomes within the village of Melquiades. However, for pSM750, preliminary evidence shows that mutation could also be playing a significant role, as has been shown for hypervariable elements in the control region of the human mitochondrial genome (Stoneking 2000). For some laboratory populations of S. mansoni, new pSM750 haplotypes are being generated at a rate greater than 10-1 per haplotype per generation (Bieberich AA, personal communication). Even if mutation rates are not nearly as high in natural populations, similar haplotypes that appear in different subpopulations may not be related by descent from a recent ancestor but may, instead, be similar by virtue of coincidental mutation to the same repeat copy number. This is even more likely to be a problem since there is a fairly narrow range of common repeat sizes (Figure 2).
Thus the schistosome population in Melquiades may actually be more strongly subdivided than we are able to detect in this initial investigation using pSM750 as a marker. To resolve this question, and extend our understanding of schistosome epidemiology, it is necessary to assess schistosome genetic variation with more appropriate markers. Our laboratory is developing microsatellite markers to examine schistosome gene flow within rural villages and to explore further the high levels of parasite genetic diversity contained within human and snail hosts.
|
Acknowledgments |
|---|
This work was funded by the National Institutes of Health through an international collaboration in disease research, "Human Immune Responses to Defined Schistosome Antigens" and a research grant (R01-AI 42768; to D.J.M.).
|
Footnotes |
|---|
Corresponding Editor: William S. Modi
Received February 21, 2000
Accepted January 15, 2001
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-
Avise JC, 1994. Molecular markers, natural history and evolution. New York: Chapman & Hall.
Awadalla P, Eyre-Walker A, Smith JM, 1999. Linkage disequilibrium and recombination in hominid mitochondrial DNA. Science 286:2524–2525.
Barral V, Morand S, Pointier JP, and Théron A, 1996. Distribution of schistosome genetic diversity within naturally infected Rattus rattus detected by RAPD markers. Parasitology 113:511–517.
Birky CW Jr, Fuerst P, and Maruyama T, 1989. Organelle gene diversity under migration, mutation, and drift: equilibrium expectations, approach to equilibrium, effects of heteroplasmic cells, and comparison to nuclear genes. Genetics 121:613–627.
Bossart JL and Prowell DP, 1998. Genetic estimates of population structure and gene flow: limitations, lessons, and new directions. Trends Ecol Evol 13:202–206.
Brindley PJ, Lewis FA, McCutchan TF, Bueding E, and Sher A, 1989. A genomic change associated with the development of resistance to hycanthone in Schistosoma mansoni. Mol Biochem Parasitol 36:243–252.[ISI][Medline]
Cesaroni D, Venanzetti F, Allegrucci G, and Sbordoni V, 1997. Mitochondrial DNA length variation and heteroplasmy in natural populations of the European sea bass, Dicentrarchus labrax. Mol Biol Evol 14:560–568.[ISI]
Curtis J and Minchella DJ, 2000. Schistosome population genetic structure: when clumping worms is not just splitting hairs. Parasitol Today 16:70–73.
Depres L, Imbert-Establet D, Combes C, Bonhomme F, and Monnerot M, 1991. Isolation and polymorphism in mitochondrial DNA from Schistosoma mansoni. Mol Biochem Parasitol 47:139–140.[ISI][Medline]
Frank SA, 1996. Models of parasite virulence. Q Rev Biol 71:37–78.[Medline]
Hey J, 2000. Human mitochondrial DNA recombination: can it be true? Trends Ecol Evol 15:181–182.[Medline]
Le TH, Blair D, McManus DP, 2000. Mitochondrial DNA sequences of human schistosomes: the current status. Int J Parasitol 30:283–290.[ISI][Medline]
Lewontin RC, 1972. The apportionment of human diversity. Evol Biol 6:381–398.
Lunt DH, Whipple LE, and Hyman BC, 1998. Mitochondrial DNA variable number tandem repeats (VNTRs): utility and problems in molecular ecology. Mol Ecol 7:1441–1455.[Medline]
Minchella DJ, Sollenberger KM, and de Souza CP, 1995. Distribution of schistosome genetic diversity within molluscan intermediate hosts. Parasitology 111:217–220.
Peña HB, de Souza CP, Simpson AJG, and Peña SDJ, 1995. Intracellular promiscuity in Schistosoma mansoni: nuclear transcribed DNA sequences are part of a mitochondrial minisatellite region. Proc Natl Acad Sci USA 92:915–919.
Rand DM, 1994. Concerted evolution and RAPping in mitochondrial VNTRs and the molecular geography of cricket populations. In: Molecular ecology and evolution: approaches and applications (Schierwater B, Streit B, Wagner GP, and Desalle R, eds). Basel: Birkhäuser; 227–245.
Sambrook J, Fritsch EF, and Maniatis T, 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
Slatkin M, 1985. Rare alleles as indicators of gene flow. Evolution 39:53–65.
Sorensen RE, Curtis J, and Minchella DJ, 1998. Intraspecific variation in the rDNA ITS loci of 37-collar-spined echinostomes from North America: implications for sequence-based diagnoses and phylogenetics. J Parasitol 84:992–997.[Medline]
Spotila LD, Rekosh DM, and LoVerde PT, 1991. Polymorphic repeated DNA element in the genome of Schistosoma mansoni. Mol Biochem Parasitol 48:117–120.[ISI][Medline]
Stoneking M, 2000. Hypervariable sites in the mtDNA control region are mutational hotspots. Am J Hum Genet 74:1029–1032.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||