genetic variation and species identification among selected leeches (hirudinea) revealed by rapd...

Biologia 67/4: 721—730, 2012Section ZoologyDOI: 10.2478/s11756-012-0063-4

Genetic variation and species identification among selected leeches(Hirudinea) revealed by RAPD markers

Aleksander Bielecki1 & Kornelia Polok2

1Department of Zoology, University of Warmia and Mazury in Olsztyn, Oczapowskiego 5, 10–718, Olsztyn; Poland; e-mail:[email protected] of Genetics, University of Warmia and Mazury in Olsztyn, Plac Lodzki 3, 10–967 Olsztyn, Poland; e-mail:[email protected]

Abstract: Leech taxonomy is based on unstable morphological characters. The overall level of genetic variability andspecies differentiation is unknown. Using the RAPD assays genetic diversities and genetic similarities were estimated intwelve species collected in North-Eastern Poland and representing three families and five subfamilies. Ten primers revealed204 reproducible bands. Genetic diversities varied from 0.099 to 0.219 classifying studied species among variable inverte-brates. Total 45 markers comprised 22% of all amplified bands were unique for species thus enabling their identification.Genetic similarities among species (0.528–0.811) evidenced several stages of differentiation, which is mirrored in the currenttaxonomy. The UPGMA and multidimensional scaling (nMDS) based on our RAPDs are congruent and reflected traditionaldivision into “Rhynchobdellida” and Arhynchobdellida. The RAPD approach proved to be an effective tool in populationand evolutionary studies of leeches. For the first time, genetic parameters were estimated enabling to compare leeches withoutcomes from other animals.

Key words: RAPD; leeches; gene diversity; species specific markers; Nei’s identity

Introduction

Leeches (Hirudinea) are highly specialized herma-phroditic annelids widely distributed on all continentsbut Antarctica. The Holarctic realm with a half of 91genera and the highest number of species is the rich-est and the most diverse region. Out of 676 species,71% are important components of freshwater ecosys-tems, 15% are marine and slightly less, 14% are terres-trial organisms (Sket & Trontelj 2008). Leeches play im-portant role as invertebrate predators like members ofthe family Erpobdellidae (Pfeiffer et al. 2005). However,most are famous as ectoparasitic bloodsuckers with thebest-known “medicinal leeches”, which have attractedtherapeutic applications for centuries. Several speciesfrom the Glossiphoniidae family are also of commer-cial interests owing to anticoagulative properties of thesalivary secretions, antistasin, ghilanten, lefaxin, thero-statin, for instance (Siddall et al. 2011). Many speciesfeed on fishes and birds, frequently transmitting viruses,bacteria and intracellular parasites, thus triggering dis-eases. Unfortunately, a recent renaissance in leech appli-cations and the knowledge about their genetic variationand evolutionary relationships are largely uncoupled.Leeches are believed to be primarily freshwater an-

imals derived from oligochaetes, Acanthobdella. Theyare divided into two orders, “Rhynchobdellida” andArhynchobdellida (Sket & Trontelj 2008). Hundreds ofspecies have been described based on shapes, color pat-

terns, eyes and genital openings. Nevertheless, the factremains that these features are unstable and the level ofvariability has not been assessed what makes personalexperience the most important in delineating species.Such uncertainties have prompted molecular studiesthat yielded surprising results on different taxonomicallevels. A striking example involves “medicinal leeches”.For years, the prevailing view has been that all varietiesin Europe represent the same taxon, Hirudo medicinalis(L., 1758). However, RAPD markers revealed a sec-ond European taxon, Hirudo verbana (Carena, 1820)(Trontelj et al. 2004). In another study, a battery ofRAPDs suggested a reproductive isolation within thesnail leech,Glossiphonia complanata (L., 1758) complex(Verovnik et al. 1999), whereas AFLPs distinguishedmorphologically similar erpobdellid leeches (Govedichet al. 1999). These data imply that leeches diversityis underestimated and a vast of cryptic species waitsto be brought to light. One notion is that assessing ge-netic diversity by means of molecular markers is limitedto several, well-known species. Furthermore, compar-isons focus on individuals rather than on populations.Consequently, the overall level of genetic variability islargely unknown. Whether observed diversity is typicalof conspecific populations or it is a sign of speciationis indeed difficult to decide. In majority of plants andanimals, similar doubts have been clarified by means ofisoenzymes. Unfortunately, it is not true for leeches andthis has necessitated frequent reinterpretation of mor-

c©2012 Institute of Zoology, Slovak Academy of Sciences

722 A. Bielecki & K. Polok

Table 1. Sampled leech species and their systematics.

Systematics1 ChromosomeSpecies Locality number2

Order Family Subfamily [2n]

Glossiphonia complanata (L., 1758) 1 Czarna Hancza3 Rhynchobdellida Glossiphoniidae Glossiphoniinae 26Glossiphonia complanata (L., 1758) 2 TylkowkoGlossiphonia complanata (L., 1758) 3 Ukiel LakeGlossiphonia concolor (Apáthy, 1888) Czarna Hancza 28Hemiclepis marginata (O.F. Muller, 1774) 1 Czarna Hancza 32Hemiclepis marginata (O.F. Muller, 1774) 2 TylkowkoPlacobdella costata (F. Muller, 1846) Czarna Hancza UnknownAlboglossiphonia papillosa (Braun, 1805) Ukiel Lake Haementeriinae UnknownHelobdella stagnalis (L., 1758) Tylkowko UnknownTheromyzon maculosum (Rathke, 1862) Czarna Hancza Theromizinae UnknownTheromyzon tessulatum (O.F. Muller, 1774) Czarna Hancza 16Piscicola geometra (L., 1761) Tylkowko Piscicolidae Piscicolinae 32Erpobdella nigricollis (Brandes, 1900) Tylkowko Arhynchobdellida Erpobdellidae Erpobdellinae UnknownErpobdella octoculata (L., 1758) Tylkowko 16Erpobdella testacea (Savigny, 1822) 1 Czarna Hancza 22Erpobdella testacea (Savigny, 1822) 2 Tylkowko

Explanations: 1Based on Integrated Taxonomic Information System; 2Based on Cichocka and Bielecki 2008. 3Each location (popula-tion) was represented by 20 individuals.

phologically based wisdom about their taxonomy andevolution.Despite a diverse array of molecular approaches

available, most leech studies employ nuclear rDNA,mitochondrial cytochrome c oxidase and NADH dehy-drogenase subunits to infer phylogenetic relationships(Siddall 2002; Williams & Burreson 2006). One advan-tage of sequence analyses is that they give first in-sights into molecular diversity of leeches; a disadvan-tage is that they do not represent intraspecific and in-trapopulational variation. Furthermore, they tag only atiny portion of genomes passing over prevailing, highlyvariable repetitive and transposon sequences. Molecularmethods only in combination provide powerful charac-ters for delineating clades, thus each estimation fromany single-gene sequence needs corroboration from ad-ditional loci. The technically straightforward randomamplified length polymorphism (RAPD) has frequentlybeen used to detect variation and discriminate mor-phologically similar, closely related taxa (Polok et al.2005b; Szczecinska et al. 2006; Baczkiewicz et al. 2008)including leeches (Verovnik et al. 1999; Trontelj et al.2004). If properly done, RAPDs offer a quick way ofscreening potential molecular markers from many lociin many individuals. Estimated genetic variation pa-rameters are comparable to these derived from enzy-matic data (Nybom 2004). In another survey, a bat-tery on nearly 3000 DNA markers was used to as-sess genetic similarities and resolve evolutionary re-lationships among species from the genus Lolium aswell as between Lolium and representatives of Poeae,of which the taxonomic position was well grounded(Polok 2007). Both Nei’s genetic similarities and phy-logenetic trees based on RAPDs did not differ fromthese revealed by ISJ (Intron Splice Junctions marker),AFLP (Amplified Fragment Length Polymorphism), B-SAP (Bacteria Specific Amplification Polymorphism),DNA transposons, retrotransposons as well as organelle

DNA hence providing usefulness of RAPDs in resolv-ing relationships even at higher taxonomic levels. TheRAPDs are of great importance in less studied taxa(Liu et al. 2007; Boulila et al. 2010; Krebs et al. 2010),in which sophisticated methods are barely trouble-some. The present studies are aimed at assessing theeffectiveness of RAPDs in population and evolution-ary studies of leeches. Parameters of genetic variationwere estimated whenever possible. Genetic divergenceof selected leech species was analysed in terms of ge-netic similarity and species-specific markers. The for-mer gives the crude overview of genetic relatedness; thelatter provides tools helpful in species identification.

Material and methods

Surveyed species and samplingTwelve species were collected in three localities in North-Eastern Poland, the Czarna River Hancza/Turtula ponds(54◦15′ N, 22◦50′ E), a fish pond in Tylkowko (53◦39′ N,20◦46′ E) and Ukiel Lake in Olsztyn (53◦47′ N, 20◦29′ E)in 2007 and 2008. They belong to two orders, three families,five subfamilies and eight genera (Table 1). Samples werecollected from objects at bottoms of water bodies, fishes andfishing baskets. They were kept in beakers with water whiletransporting to the laboratory, where they were identifiedbased on morphology (Sample photos are available as sup-plementary data.). The majority of species were identifiedonly in one locality. For each of them 20 individuals werecollected for the analysis. However, three species, namelyG. complanata, Hemiclepsis marginata (O.F. Muller, 1774)and Erpobdella testacea (Savigny, 1822) were found in 2–3 locations, so that 2–3 populations represented them. Intotal, 20 individuals per each population represented thesespecies. Leeches were starved for several weeks to avoid con-tamination from host/prey DNA found in the gastric re-gions. Finally, they were frozen at –70◦C until use.

DNA extractionTotal genomic DNA was extracted from the caudal suckerusing the modified CTAB method. Briefly, tissues were

Genetic variation of leeches 723

ground in a mortar with 300 µl of the pre-warm CTABbuffer (2% CTAB, 100 mM Tris-HCl, pH 8.0, 20 mM EDTA,1.4 M NaCl, 2% β-mercaptoethanol), then transferred intoa 1.5 ml Eppendorf tube, to which Proteinase K (final con-centration 500 µg ml−1) was added to remove proteins.After 1 h of incubation at 65◦C, another portion of Pro-teinase K was added (final concentration 250 µg ml−1)and the mixture was further incubated at 65◦C for 12 h.Afterwards nucleic acids were extracted three times usingchloroform-isoamyl alcohol (24:1), precipitated with 96%ethanol, washed and dissolved in sterile, deionized H2O asdescribed by Polok (2007). RNA was removed with RNA-ase at a final concentration 200 µg ml−1. The quality ofDNA was verified on 1% agarose while the purity was as-sessed spectrophotometrically and it ranged between 89%and 98%. The average DNA content in the samples was42 µg and ranged from 3.25 µg to 89.7 µg depending on thebody size.

Primer screening and RAPD-PCR conditionsTwenty ten-nucleotide primers from Operon Technologies(from A, B and D kits) were initially tested in fourspecies, Helobdella stagnalis (L., 1758), Theromyzon tessula-tum (O.F. Muller, 1774), Theromyzon maculosum (Rathke,1862) and E. testacea representing different degrees of tax-onomic relationships. Each reaction was done twice usingdifferent DNA isolates of the same sample and only primersgiving identical results were selected. In total ten primers,giving reproducible and reliable banding patterns were fi-nally selected to analyse 12 species (Table 2). These primersgave identical fingerprints for two PCR runs of the samesample.

Each PCR reaction was performed in 20 µl volumecontaining 20 mM (NH4)2SO4, 50 mM Tris-HCl, pH 9.0at 25◦C, 2.0 mM MgCl2, 2× concentrated Enhancer withbetaine (Epicentre Technology), 200 µM dNTPs, 0.3 µMprimer, 1 U of Tfl polymerase (Epicentre Technology) and80 ng of the template DNA. To ensure repeatability, all sam-ples were amplified in the same MJ Mini Personal Ther-mal Cycler (BioRad). After 3 min of initial denaturation at94◦C, samples were subjected to 45 cycles of 1 min at 94◦C,1 min at 36◦C, 2.5 min at 72◦C and 5 min of the final exten-sion at 72◦C. PCR products were loaded into a 1.5% (w/v)agarose gel containing 0.5 µg ml−1 ethidium bromide andseparated in 1× TBE buffer (Tris-Borate-EDTA) at 100 Vconstant power, visualized under UV light (312 nm), pho-tographed with Olympus Camera and stored as .jpg files.The 100–2000 bp ladder was used as a standard. The bandswere scored in a computer using the Technical Designer soft-ware to identify bands of the same mobility.

Statistical analysisAll bands that could be reliable read were treated as inde-pendent loci, and scored either present (1) or absent (0).We assumed that possible, non-homology errors from co-migrating bands are random. Such random errors althoughmay reduce the absolute similarities, have not effect onrelative similarities nor the relationships among the taxa(Adams & Rieseberg 1998). The polymorphic products werereferred according to the Operon recommendations as fol-low, A03–1 when referring to a band revealed by the A03primer and the first band from the anode. Basic geneticparameters were estimated for species collected in morethan one locality, namely G. complanata, H. marginata andE. testacea, so that data from multiple populations wereavailable (Table 1). The mean number of alleles per locus

(Na), effective number of alleles (Ne), percentage of poly-morphic loci (P) and the Nei’s gene diversity (H) werecalculated using the POPGENE 1.32 software (Yeh et al.2000). The gene diversity H is equal to the average propor-tion of heterozygotes in a randomly mating population ofdiploids and therefore, the parameter is frequently called as“heterozygosity” (Nei 1987). Diversity was also measuredin categorical data using Shannon’s diversity (Hpop). Thisquantity differs from the Nei’s gene diversity by calculatingthe logarithm of allele frequencies instead of their squares.The output is the weighted geometric mean of the averageabundance of different alleles. Each parameter was averagedover all both loci and populations to give the total aver-age value for a species. Statistically significant differencesamong species means of all genetic parameters were testedby the analysis of variance (MANOVA) with the LSD test inthe STATISTICA 9.0 software. For a relatively large num-ber of alleles the distribution of parameters is approximatelynormal, and the ordinary statistics can be used.

Nei’s (1987) pairwise genetic identities and distance forall 12 species were determined using the POPGENE 1.32.A general point for evolutionary conclusions involves thedegree of congruence among clustering derived from dif-ferent algorithms (Avise 2004). Accordingly, relationshipsamong leech species were evaluated using UPGMA and non-parametric multidimensional scaling (nMDS) with 3-D visu-alization. All phylogenetic conclusions were rooted in com-parisons between groupings revealed by both methods. TheUPGMA amalgamation was based on Euclidean distancesderived from the matrix of Nei’s similarities. The major as-sumption underlying UPGMA clustering is that evolution-ary rates are equal along all dendrogram branches. Notwith-standing this limitation, UPGMA is often superior to otherdistance matrix methods in recovering the true tree basedon genetic distance/similarity. This is because a distancemeasure has a smaller coefficient of variation and an av-eraging feature of UPGMA tends to reduce large stochas-tic errors (Nei 1987; Nei & Kumar 2000). Non-parametricmultidimensional scaling (nMDS) was based on the matrixof Nei’s genetic distances. This is a heuristic method ex-plaining distances in terms of underlying dimensions. Non-metric MDS based on the Lingoes-Guttman test finds bothnon-parametric relationships and metric distances betweenitems. In other words, it tests the relationship each vari-able has to every over variable across all cases. Therefore,it does not depend on a population or evolutionary modelfitting any distribution. Also, it produces the 3-D trees thatcan be viewed from any angle each furnishing better under-standing of the relative position and evolutionary distanceof each species. nMDS in an iterative process, in which con-figurations of the data points are produced sequentially. Astress value is calculated for each iteration by comparing theranked distances between observations in the original dataset with the ranked distances between nMDS scores from theordination. The lower a stress value, the better fit betweenthe ordination and the original data. The procedure runs un-til a minimum stress value is reached. The algorithm provedto be a powerful method in resolving evolution of grassesfrom the genus Lolium (Polok 2007) and repositioning ofrodents, chiroptera and primates in phylogenetic trees (Mil-ner et al. 2004). This is a useful approach for a large set ofdata and with a growing number of molecular data MDS isseen a modern means for resolving phylogenetic trees, whichwas emphasized by using MDS in monitoring evolution ofthe largest family of transmembrane receptors (GPCRs) inhumans (Pélé et al. 2011). To visualize evolutionary relation-


Table 2. RAPD primer sequences and amplification products revealed in 12 leech species.

Primer Sequence Fragment size Total n◦ Mean n◦ of bands Mean n◦ (%) of polymorphic(5’→3’) range [bp] of bands per species bands per species

A01 CAGGCCCTTC 540–1580 21 3.7 2.0 (49.0%)A02 TGCCGAGCTG 780–1900 21 3.7 3.7 (48.7%)B01 GTTTCGCTCC 300–912 17 2.1 0.0 (0.0%)B02 TGATCCCTGG 800–2080 22 5.1 1.3 (21.0%)B04 GGACTGGAGT 420–2200 23 5.8 2.7 (41.3%)B05 TGCGCCCTTG 440–1600 26 9.4 3.0 (26.0%)B08 GTCCACACGG 600–1640 17 2.0 0.0 (0.0%)D01 ACCGCGAAGG 360–1640 24 9.1 6.7 (54.3%)D05 TGAGCGGACA 620–1600 11 1.1 0.0 (0.0%)D16 AGGGCGTAAG 400–1600 22 8.3 2.7 (34.0%)

Total over loci 204 50.2 22.0 (43.8%)Mean per primer 20.4 5.0 2.2 (27.4%)

Table 3. Species specific bands in 12 leech species.

Species Number of identified bands Number of bands specific per species % of bands specific per species

G. complanata 67 5 7.5G. concolor 38 4 11.1H. marginata 54 5 9.3P. costata 38 1 2.6A. papillosa 31 1 3.2H. stagnalis 53 2 3.8T. maculosum 73 8 10.9T. tessulatum 58 8 13.7P. geometra 47 4 8.5E. nigricollis 36 0 0.0E. octoculata 44 2 4.5E. testacea 63 5 7.9

Mean per species 50.2 3.8 6.9

ships among studied leeches a number of dimensions werereduced from nine to three using the Guttman-Lingoes test.At least 100 iterations were used in each step. In the finalrun, the best configuration was received after 18 iterations(stress value, 0.009) and this was used for plotting a 3Ddiagram. The goodness of fit was assessed by the Shepardplot. Both the UPGMA and the nMDS were carried in theSTATISTICA 9.0 software.

Results

RAPD variationRAPD assays revealed 204 reproducible bands rangingfrom 300 bp to 2200 bp. The number of scored bandsper primer was high and varied from 11 (D05) to 26(B05) with an average of 20.4 (Table 2). All amplifiedbands were polymorphic. Certainly as many as 12 dif-ferent species contributed to both such a huge num-ber of bands and 100% of polymorphic bands acrossall samples. Values were significantly lower when cal-culated per species – on average 50 bands were am-plified and 22 (44%) were polymorphic. The highestpolymorphism per species revealed the D01 primer (6.7bands, 54% polymorphic) while the B02 (1.3 bands,21% polymorphic) the lowest. Bands amplified by threeprimers, namely B01, B08 and D05, although polymor-

phic among species, were always monomorphic withina species.A number of amplified bands in studied species

ranged from 31 in Alboglossiphonia papillosa (Braun,1805) to 73 in T. maculosum (Table 3). Surprisingly,populations of selected species collected from the samewater body appeared to produce identical DNA pro-files. A striking example involves monomorphic sam-ples of T. maculosum (Fig. 1) although they were col-lected independently in 2007 and 2008 and DNA wasseparately isolated. By accidence, this observation wasfortuitous confirming high reproducibility of RAPDs inthe present studies. However, it was a curse disablingestimation of genetic diversities except of species foundin different localities, i.e., G. complanata, H. marginataand E. testacea. Among these three taxa G. complanataexhibited the highest and relatively large genetic vari-ability as emphasised by high polymorphism (P =58%), number of alleles (Na = 1.58 and Ne = 1.37)and gene diversity (H = 0.219, Hpop = 0.326). Thisspecies was twice as variable as H. marginata and E.testacea (Table 4). By contrary, the latter two speciesdisplayed similar values of all genetic variation param-eters as demonstrated by polymorphic loci (24% and30%, respectively) and gene diversities (H = 0.099 andH = 0.125).


Table 4. Genetic variation revealed by RAPD markers in three leech species.

Species L P Na Ne H Hpop I

G. complanata 39 58.2 1.58a 1.37a 0.219a 0.326a 0.836H. marginata 13 24.1 1.24b 1.17b 0.099b 0.146b 0.918a

E. testacea 19 30.2 1.30b 1.21b 0.125b 0.182b 0.881ab

Explanations: Variation parameters: L – the number of polymorphic loci; P – the percentage of polymorphic loci; Na – average observednumber of alleles; Ne – effective number of alleles; H – average Nei’s gene diversity (heterozygosity) in a population (Nei 1987), Hpop– the Shannon’s index over loci. I – average Nei’s genetic identity among populations (Nei 1987). Different letters (a, b, c) indicatesignificant differences among species at P = 0.05 in LSD test.

Table 5. Nei’s unbiased measures of genetic identity (above diagonal) and distance (below diagonal) among 12 leech species.

G.complanata1

G.complanata2

G.complanata3

G.concolor

H.marginata1

H.marginata2

P.costata

A.papillosa

H.stagnalis

T.maculosum

T.tessulatum

P.geometra

E.nigricollis

E.octoculata

E.testacea1

E.testacea2

G. complanata 1 **** 0.862 0.830 0.748 0.673 0.704 0.698 0.654 0.619 0.583 0.648 0.566 0.648 0.610 0.616 0.623G. complanata 2 0.149 **** 0.818 0.723 0.598 0.642 0.660 0.604 0.597 0.562 0.585 0.528 0.572 0.585 0.604 0.585G. complanata 3 0.186 0.201 **** 0.717 0.591 0.635 0.642 0.610 0.589 0.543 0.579 0.547 0.554 0.591 0.535 0.591G. concolor 0.290 0.324 0.333 **** 0.610 0.679 0.660 0.654 0.631 0.583 0.635 0.604 0.635 0.635 0.629 0.610H. marginata 1 0.396 0.515 0.526 0.494 **** 0.918 0.635 0.654 0.644 0.600 0.635 0.591 0.560 0.585 0.616 0.598H. marginata 2 0.350 0.444 0.454 0.387 0.085 **** 0.642 0.673 0.701 0.593 0.654 0.648 0.616 0.604 0.649 0.642P. costata 0.359 0.415 0.444 0.415 0.454 0.444 **** 0.692 0.673 0.574 0.698 0.591 0.635 0.660 0.642 0.649A. papillosa 0.425 0.505 0.494 0.425 0.425 0.396 0.368 **** 0.722 0.665 0.616 0.660 0.704 0.642 0.723 0.679H. stagnalis 0.481 0.516 0.533 0.460 0.440 0.356 0.397 0.326 **** 0.680 0.707 0.650 0.678 0.673 0.688 0.660T. maculosum 0.539 0.577 0.611 0.539 0.512 0.522 0.555 0.408 0.385 **** 0.659 0.602 0.609 0.599 0.653 0.625T. tessulatum 0.434 0.536 0.547 0.454 0.454 0.425 0.3359 0.484 0.3469 0.417 **** 0.604 0.635 0.649 0.667 0.635P. geometra 0.569 0.638 0.603 0.505 0.526 0.434 0.526 0.415 0.431 0.507 0.505 **** 0.642 0.667 0.673 0.642E. nigricollis 0.434 0.558 0.592 0.454 0.580 0.484 0.454 0.350 0.389 0.497 0.454 0.444 **** 0.723 0.806 0.811E. octoculata 0.494 0.536 0.526 0.454 0.536 0.505 0.415 0.444 0.397 0.512 0.434 0.406 0.324 **** 0.767 0.736E. testacea 1 0.484 0.505 0.626 0.464 0.484 0.434 0.444 0.324 0.374 0.427 0.406 0.396 0.217 0.265 **** 0.881E. testacea 2 0.474 0.536 0.526 0.494 0.515 0.444 0.434 0.387 0.415 0.470 0.454 0.444 0.209 0.307 0.127 ****

Fig. 1. Summarized PCR patterns of 12 leech species amplified with the RAPD primer A01: 1, 2 – H. stagnalis; 3 – P. costata; 4 – T.tessulatum; 5, 6 – T. maculosum; 7 – E. testacea 1; 8 – E. testacea 2; 9 – E. octoculata; 10 – E. nigricollis; 11 – G. complanata 1;12 – G. complanata 2; 13 – G. complanata 3; 14 – G. concolor; 15 – H. marginata 1; 16 – H. marginata 2; 17 – A. papillosa; 18 – P.geometra.

Species specific markersA follow-up of low band sharing (Fig. 1) provided tal-lies of markers specific to one species yet absent in oth-ers that can help in their identification (Table 3). To-tal 45 unique markers comprised 22% of all amplifiedproducts. On average 3.8 bands (6.9%) were unique perspecies. The highest number of unique markers was intwo Theromyzon species (8 bands in each). The oth-ers possessed from no unique bands as in E. nigricollis

to five bands in G. complanata, H. marginata andE. testacea. To pick one setting as an example, a bandat the B04-4 and B04-22 loci distinguished two species,P. costata and A. papillosa, respectively. Apart fromspecies specific bands three markers were typical of allmembers of higher taxa. They involved a band at theA01-13 locus in the subfamily Glossiphoniinae, a bandat B02-13 in all representatives of „Rhynchobdellida”and a band at A02-20 typical of all studied Arhyn-


Fig. 2. Dendrogram of 12 leech species based on genetic similarities: 1 – Glossiphoniinae; 2 – Haementeriinae; 3 – Theromizinae; 4 –Piscicolinae; 5 – Erpobdellinae.

chobdellida. However, a caution should be made as thelater was represented only by Erpobdella therefore, themarker can be a genus specific as well.

Genetic similarities and cluster analysisAs expected from distinctive fingerprints, pairwise Nei’sgenetic similarities were low (0.528–0.811), well withinthe range of values associated with biological species(Table 5). Unsurprisingly, lower Nei’s coefficients werenoted for comparisons across different subfamilies thanwithin them, hence confirming the utility of RAPDs forsuch analyses. Overall, the Glossiphoniinae and Pisci-colinae were the least similar as demonstrated by theNei’s coefficient (I = 0.582), statistically lower thanbetween other subfamily pairs. Mean similarities be-tween the other subfamilies (0.603–0.681) did not differstatistically (Table 6). Divergence of the Glossiphoni-inae and Piscicolinae is also demonstrated by individualcomparisons. For instance, I = 0.528 between G. com-planata 2 and P. geometra was the lowest among allspecies pairs. By contrary, similarities within a sub-family were generally higher and for congeners closeto values typical of the subspecies level as for pairs E.nigricollis – E. testacea (I = 0.806–0.811), E. octocu-lata – E. testacea (0.736–0.767) and G. complanata – G.concolor (0.717–0.748). Two Theromyzon species withthe Nei’s coefficient closer to the level of non-siblingspecies are exceptions (I = 0.659). With regard to con-specific populations, similarities from I = 0.818 to I =0.918, much below levels associated with populations,indicated early stages of evolutionary divergence. TheG. complanata 3 from Ukiel Lake was the most diverged(Table 5).

Table 6. Mean Nei’s similarities among species within a subfamilyand between subfamilies.

Glo

ssip

hon

iinae

Hae

men

teri

inae

Ther

omiz

inae

Pis

cico

linae

Erp

obdel

linae

Glossiphoniinae 0.699bc

Haementeriinae 0.642bc 0.722ab

Theromizinae 0.605cd 0.667bc 0.659bc

Piscicolinae 0.582d 0.655bc 0.603cd 1.000a

Erpobdellinae 0.609cd 0.681bc 0.641cd 0.642bcd 0.787a

Explanations: The statistically significant values at P = 0.05 inLSD test are indicated by different letters.

Both the UPGMA and multidimensional scaling(MDS) are congruent and grouped species in generalagreement with their taxonomic status. On the UP-GMA tree three clades were distinguished (Fig. 2). Firstconsisted of all Glossiphoniinae but P. costata, sec-ond comprised Haementeriinae and Theromizinae andthird, Erpobdellinae. Grouping of P. geometra with rep-resentatives of Arhynchobdellida on the UPGMA treewas unexpected, however on the MDS plot its posi-tion was somewhere in the middle between Glossiphoni-inae and Erpobdellinae (Fig. 3). Likewise, T. maculo-sum and T. tessulatum were dispersed among differentgroups on the UPGMA dendrogram while the MDSplaced them together. On the other hand, P. costatawas always separated from its subfamily Glossiphoni-inae.


Fig. 3. Multidimensional scaling of 12 leech species based on RAPD data. Solid circles – Glossiphoniinae, open circles – Haementeriinae,bracket circles – Theromizinae, triangles – Piscicolinae, squares – Erpobdellinae.

Discussion

The knowledge about genetic diversity and evolutionof leeches has both theoretical and practical implica-tions. Firstly, this can serve as the basis for taxonomy.Secondly, it implies management recommendations forboth predaceous and ectoparasitic species. The pointat issue is that leeches generally suffer from the lackof molecular analyses of populations including isoen-zymatic studies. With this respect, the present RAPDdata are the first estimations of genetic diversity pa-rameters and divergence levels of a wealth of speciesrepresenting different taxonomic relationships. Likely,the technical reasons are behind scarce population ge-netic data. They include difficulties in finding a suit-able number of populations and material contamina-tion from gastric regions. The former were also en-countered in our studies, in which only three speciesinhabited at least two collection sites thus, only forthem genetic diversity could be assessed. The latter canbe overcome by starvation for a period. However, thiscan influence enzymatic activities but not DNA. RecentRAPD analyses providing first data about microevolu-tionary processes in leeches confirmed their usefulness.Thus, simple and cost effective nuclear DNA mark-ers can be a driving horse of leech population genetic

studies as isoenzymes did for many plants and ani-mals.One common finding is that RAPD markers reveal

various levels of polymorphism in three taxa, G. com-planata, H. marginata and E. testacea. In all species thepercentage of polymorphic loci (24%–58%) and genediversity (0.099–0.219) fall within ranges recorded inabout a thousand of animal and plant populations bymeans of isoenzymes. The polymorphism is from 10%to 50% whilst average diversities vary from 0.020 to0.150 (Sole-Cava & Thorpe 1991). Leech genetic varia-tion is also the same magnitude as in invertebrates, inwhich the mean coefficient of gene differentiation (HT–HS/HT) is 0.171 with a range of 0.060–0.263 (Avise2004). In contrast to enzymatic data, there are no suchsummaries for DNA markers but comparative analy-ses in grasses demonstrate that protein and DNA-basedmethods describe the same pie of a population history(Polok 2007).Among three studied species, G. complanata with

P = 58% and H = 0.219 classifies to most variableorganisms. Indeed, it evidences as high variation as44 highly variable marine invertebrates with the meanpolymorphism, P = 54% and H = 0.206 (Sole-Cava& Thorpe 1991). Likely, high RAPD polymorphism inG. complanata mirrors morphological diversity empha-


sized by colour variation observed in the populationfrom Lake Ukiel (photos – supplementary data). Theother two species, H. marginata and E. testacea aretwofold less variable. Although these data can be lim-ited to studied populations, a common belief is thatgenetic polymorphisms are products of evolutionaryforces, life history and ecological factors. If so, thesedifferences would be because of various ectoparasiticproprieties. Among three species, only H. marginata isan obligate parasite of fishes and amphibians and thisspecies is the least variable. Another explanation in-volves the nature of water bodies. Both H. marginataand E. testacea were not found in Lake Ukiel, fromwhere the most diverse population of G. complanataoriginated. Postglacial Lake Ukiel, lying nearby a cityof Olsztyn, is polluted, littered with bottles, cans andplastics. They are perfect sites to refuge and prey pro-viding especially rich environments for leeches. A finalpoint to consider is that populations inhabiting one wa-ter body were monomorphic. From one side leeches arehermaphrodites that may explain the phenomenon, butfrom the other, the cross-fertilization is predominant.Whatever is the reason; these data corroborate confir-mation in more extensive sampling.Significant challenges when working on leeches

both morphologically variable and poorly studied arerepeated field collections followed by species identifica-tion. This means that different researchers may workon different species still classified under common taxo-nomic names. In an illustrative example, mtDNA read-ily separated five distinct species in laboratory strainsof Helobdella commonly used in developmental research(Bely & Weisblat 2006). Therefore, a battery of 45RAPD markers discriminating studied taxa can findapplications in confirming diagnoses based on morphol-ogy. This is not to say that DNA fingerprints shouldreplace traditional taxonomy but they rather shouldcomplement the data from anatomy, body size and lifestyles. Notwithstanding such a vast number of uniquebands in our study certainly result from sampling di-verse taxonomic species, when coupling with low bandsharing, they inevitable entail more important conclu-sions. All studied taxa delineated based on morphologyare biological species.Genetic similarities in studied leeches (0.528–

0.811) evidence apparently several stages of speciationprocess, which is mirrored in the current taxonomy. Inprincipal, the closest taxonomic relationships the high-est Nei’s similarities were noted. The lowest values ininterfamiliar comparisons are a considerable support forreproductive barriers between most species analysed,thus confirming the rationale for their delineation. Un-fortunately, the only available leech’s estimations arefrom the G. complanata aggregate, in which RAPDbased similarities of 0.57–0.59 led the authors to dis-tinguish the morphs as biological species (Verovnik etal. 1999). The data are scarce as well for other an-nelids. Exotic earthworms (Apporectodea) proved to ex-hibit RAPD similarities from 0.466 to 0.590 (Dyer et al.1998). A classic survey in Drosophila willistoni complex

demonstrates that Nei’s similarities from 0.38 to 0.56are typical of sibling and non-sibling species with fullpostzygotic isolation barriers (Avise 2004). In plants,the mean Nei’s identity of 0.66 is associated with ob-viously reproductively isolated grasses from the genusLolium and cereals (Polok 2007). Molecular appraisalsemploying several categories of DNA markers (RAPD,ISJ, B-SAP) revealed that some liverworts consist ofreproductively isolated cryptic species with similari-ties ranging from 0.182 to 0.690 (Polok et al. 2005a;Baczkiewicz et al. 2008).If we apply the Drosophila model, similarity mag-

nitudes in congeners (0.717–0.811) may imply only apartial isolation. It can not be excluded, however thatit is a case of a sudden speciation associated with lit-tle changes at the allelic level. Different chromosomenumbers in G. complanata and G. concolor as well asin E. octoculata and E. testacea support the conclusion(Table 1). Chromosome aberrations are most commonmechanisms of rapid speciation. Different chromosomenumbers inevitable entail at least a partial sterility ofputative hybrids, so do reproductive barriers. ThreeHirudo species differ in only one chromosome in haploidgenomes (n = 12, 13 or 14) yet the hybrids have a verypoor reproductive potential (Utevsky et al. 2009). Oneintriguing observation is a relatively high divergence ofconspecific populations especially well pronounced inG. complanata but also noticeable in H. marginata andE. testacea. The results are in line of evidences fromhighly diverged G. complanata in Slovenia (Verovniket al. 1999), E. punctata in northern Arizona (Govedichet al. 1999) in addition to high morphological diversi-ties (Sket & Trontelj 2008). The emerging message isthat leech species are evolving and further studies em-ploying more populations and markers are necessary todocument a wide spectrum of outcomes.Studied species grouped much in agreement with

traditional taxonomy. Placements of “Rhynchobdell-ida” and Arhynchobdellida as well as Glossiphoni-inae, Haementeriinae and Erpobdellinae are well sup-ported. Nonetheless, conflicting relationships are ob-served. First, P. geometra, consequently groups withArhynchobdellida despite the proboscis stated for in-clusion into “Rhynchobdellida”. Likewise, 18S rDNAand mtDNA, join Piscicolidae with Arhynchobdellida(Apakupakul et al. 1999) suggesting independent evo-lution of the proboscis in different groups, possible dueto several point mutations. Second, P. costata groupswith Haementeriinae and Theromizinae instead of Glos-siphoniinae. Similar relationships on the rDNA andmorphological tree (Siddall et al. 2005) may indicatethe necessity for regrouping. Third, conflicting posi-tions of T. maculosum and T. tessulatum are difficultto explain. Both species differ in colour (supplemen-tary data), genital openings and distribution. Low ge-netic identity confirms their split. However, the formerhas never been studied molecularly and more extensivesampling is necessary for general conclusions.Finally, the RAPDs provide powerful characters

to describe the leech’s diversity and genetic similari-


ties. The approach enables, for the first time, to lookat leeches in the light of outcomes from dozens of livingcreatures. Despite RAPD usefulness is frequently ques-tioned due to possible co-migration of non-homologousfragments they often perform surprisingly well in recov-ering proper phylogenetic trees even in distant taxa asexemplified in plants (Polok 2007; Wong et al. 2007),fungi (Obornik et al. 2000) and animals (Spiridonovaet al. 2008). An intriguing point is why RAPDs areso effective. One explanation is randomness of non-homologous, co-migrating bands that have little effecton relative similarities (Adams & Rieseberg 1998). Sec-ondly, recent data from Pinus genome suggest transpo-son origin of at least some RAPD products (Morse et al.2009). It is well documented that transposons are majordiversifying forces in eukaryotic genomes (Polok 2007;Oliver & Greene 2009; Polok & Zielinski 2011) but theirrole in leech evolution remains to be disclosed. Keepingin mind that a comprehensive picture can only be drawnon various data, simple DNA markers such as RAPDscan successfully complement morphological and singlegene approaches. Last but not least, DNA assays usu-ally support morphologically based hypotheses. Theyare the most useful when they address controversial ar-eas. As demonstrated in our studies, leeches are not anexception.

Acknowledgements

The molecular part of this work was supported by the Euro-pean Union, the Marie Curie Host Fellowship for the Trans-fer of Knowledge programme, under the project N◦ MTKD-CT-2004-509834. We thank Sylwia Ciaglo-Androsiuk forhelp in DNA isolation and Dominika Bogdanska for hervaluable technical support.

References

Adams R.P. & Rieseberg L.H. 1998. The effects of non-homologyin RAPD bands on similarity and multivariate statistical or-dination in Brassica and Helianthus. Theor. Appl. Genet. 97(2): 323–326. DOI: 10.1007/s001220050902

Apakupakul K., Siddall M.E. & Burreson E.M. 1999. Higher levelrelationships of leeches (Annelida: Clitellata: Euhirudinea)based on morphology and gene sequences. Mol. Phylogenet.Evol. 12: 350–359. DOI: 10.1006/mpev.1999.0639

Avise J.C. 2004. Molecular Markers, Natural History, and Evo-lution. 2nd Edition. Sinauer Associates, Inc. Publishers, Sun-derland, Massachusetts, 684 pp. ISBN-10: 0878930418, ISBN-13: 978-0878930418

Baczkiewicz A., Sawicki J., Buczkowska K., Polok K. & ZielinskiR. 2008. Application of different DNA markers in studies oncryptic species of Aneura pinguis (Jungermannipsida, Met-zgeriales). Cryptogamie Bryol. 29: 3–21.

Bely A.E. & Weisblat D.A. 2006. Lessons from leeches: a call forDNA barcoding in the lab. Evol. Dev. 8 (6): 491–501. PMID:17073933

Boulila A., Bejaoui A., Messaoud C. & Boussaid M. 2010. Ge-netic diversity and population structure of Teucrium polium(Lamiaceae) in Tunisia. Biochem. Genet. 48 (1–2): 57–70.DOI: 10.1007/s10528-009-9295-6

Cichocka J. & Bielecki A. 2008. Biological diversity of leeches(Clitellata: Hirudinida) based on the characteristics of thekaryotypes. Wiad. Parazytol. 54 (4): 309–314. PMID:19338221

Dyer A.R., Fowler J.C.S. & Baker G.H. 1998. Detecting ge-netic variation in exotic eartworms, Aporrectodea spp. (Lum-bricidae), in Australian soils using RAPD markers. SoilBiol. Biochem. 30 (2): 159–165. DOI: 10.1016/S0038-0717(97)00098-9

Govedich F.R., Blinn D.W., Hevly R.H. & Keim P.S. 1999. Cryp-tic radiation in erpobdellid leeches in xeric landscapes: amolecular analysis of population differentiation. Can. J. Zool.77 (1): 52–57. DOI: 10.1139/z98-178

Integrated Taxonomic Information System [database on the In-ternet]. [modified 2009 Dec 24: cited 2010 May 28]. Availablefrom: http://www.itis.gov

Krebs C., Mahy G., Matthies D., Schaffner U., Tiebre M.S.& Bizoux J.P. 2010. Taxa distribution and RAPD mark-ers indicated different origin and regional differentiation ofhybrids in the invasive Fallopia complex in central-westernEurope. Plant Biol. 12 (1): 215–223. DOI: 10.1111/j.1438-8677.2009.00219.x

Liu Y.G., Chen S.L. & Li B.F. 2007. Genetic differentiationamong common and selected hatchery populations of floun-der: evidence from RAPD. Biochem. Syst. Ecol. 35 (10):689–695. DOI: 10.1016/j.bse.2007.04.012

Milner M., Bansode A.G., Lawrence A.L., Nevagi S.A., Patward-han V. & Modak S.P. 2004. Molecular phylogeny in 3-D. Curr.Issues Mol. Biol. 6 (2): 189–200. PMID: 15119827

Morse A.M., Peterson D.G., Islam-Faridi M.N., Smith K.E., Mag-banua Z., Garcia S., Kubisiak T.L., Amerson H.V. et al. 2009.Evolution of genome size and complexity in Pinus. PLoS ONE4 (2): e4332. DOI: 10.1371.journal.pone.0004332

Nei M. 1987. Molecular Evolutionary Genetics. Columbia Uni-versity Press, New York, Guildford, Surrey, 512 pp. ISBN:0231063210, 9780231063210

Nei M. & Kumar S. 2000. Molecular Evolution and Phylogenet-ics. Oxford University Press, New York, 333 pp. ISBN-10:0195135857, ISBN-13: 978-0195135855

Nybom H. 2004. Comparison of different nuclear DNA markersfor estimating intraspecific genetic diversity in plants. Mol.Ecol. 13 (5): 1143–1155. DOI: 10.1111/j.1365-294X.2004.02141.x

Obornik M., Klic M. & Zozka L. 2000. Genetic variability andphylogeny inferred from random amplified polymorphic DNAdata reflect life strategy of enthopathogenic fungi. Can J. Bot.78 (9): 1150–1155. DOI: 10.1139/cjb-78-9-1150

Oliver K.R. & Greene W.K. 2009. Transposable elements: power-ful facilitators of evolution. BioEssays 31 (7): 703–714. DOI:10.1002/bies.200800219

Pélé J., Abdi H., Moreau M., Thybert D. & Chabbert M. 2011.Multidimensional scaling reveals the main evolutionary path-ways of class A G-protein-coupled receptors. PLoS ONE 6(4): e19094. DOI: 10.1371/journal.pone.0019094

Pfeiffer I., Brening B. & Kutschera U. 2005. Molecular phylogenyof selected predaceous leeches with reference to the evolutionof body size and terrestrialism. Theor. Biosci. 124 (1): 55–64.DOI: 10.1016/j.thbio.2005.05.002

Polok K. 2007. Molecular Evolution of the Genus Lolium L. Stu-dio Poligrafii Komputerowej “SQL”, Olsztyn, 317 pp. ISBN:8388125524, 9788388125522

Polok K., Sawicki J., Kubiak K., Szczecinska M., Korzekwa K.,Szandar K. & Zielinski R. 2005a. Evolutionary divergencewithin Pellia endiviifolia (Dicks.) Dum. from Poland, pp.241–252. In: Prus-G�lowacki W. & Pawlaczyk M.E. (eds),Variability and Evolution: New Perspectives, WydawnictwoNaukowe UAM, Poznan, 562 pp. ISBN: 83-232-1599-5

Polok K., Urbaniak L., Korzekwa K., Androsiuk P., Ciag�lo S.,Kubiak K. & Zielinski R. 2005b. Genetic similarity of Pinussylvestris populations on the base of DNA markers, pp. 253–267. In: Prus-Glowacki W. & Pawlaczyk E. (eds), Variabil-ity and Evolution: New Perspectives, Wydawnictwo NaukoweUAM, Poznan, 562 pp. ISBN: 83-232-1599-5

Polok K. & Zielinski R. 2011. Mutagenic treatment induces hightransposon variation in barley (Hordeum vulgare L.). ActaAgr. Slov. 97 (3): 179–188. DOI: 10.2478/v10014-011-0012-x

Siddall M.E. 2002. Phylogeny of the leech family Erpobdelli-dae (Hirudinida: Oligochaeta). Invertebr. Syst. 16: 1–6. DOI:10.1071/IT01011


Siddall M.E, Budinoff R.B. & Borda E. 2005. Phylogenetic eval-uation of systematics and biogeography of the leech fam-ily Glossiphoniidae. Invertebr. Syst. 19 (2): 105–112. DOI:10.1071/IS04034

Siddall M.E., Min G.-S., Fontanella F.M., Phillips A.J. & WatsonS.C. 2011. Bacterial symbiont and salivary peptide evolutionin the context of leech phylogeny. Parasitology 138 (13):1815–1827. DOI: 10.1017/S0031182011000539

Sket B. & Trontelj P. 2008. Global diversity of leeches (Hirudinea)in freshwater. Hydrobiologia 595 (1): 129–137. DOI: 10.1007/s10750-007-9010-8

Sole-Cava A.M. & Thorpe J.P. 1991. High levels of genetic vari-ation in natural populations of marine lower invertebrates.Biol. J. Linn. Soc. 44 (1): 65–80. DOI: 10.1111/j.1095-8312.1991.tb00607.x

Spiridonova L.N., Korobitsyna K.V., Yakimenko L.V. & Bog-danov A.S. 2008. Genetic differentiation of subspecies of thehouse mouseMus musculus and their taxonomic relationshipsinferred from RAPD-PCR data. Russ. J. Genet. 44 (6): 732–739. DOI: 10.1134/S1022795408060148

Szczecinska M., Sawicki J., Polok K., Holdynski Cz. & Zielinski R.2006. Comparison of three Polygonatum species from Polandbased on DNA markers. Ann. Bot. Fenn. 45: 379–388.

Trontelj P., Sotler M. & Verovnik R. 2004. Genetic differen-tiation between two species of the medicinal leech, Hirudomedicinalis and the neglected H. verbena, based on random-amplified polymorphic DNA. Parasitol. Res. 94 (2): 118–124.DOI: 10.1007/s00436-004-1181-x

Utevsky S., Kovalenko N., Doroshenko K., Petrauskiene L. &Klymenko V. 2009. Chromosome numbers for three speciesof medicinal leeches (Hirudo spp.). Syst. Parasitol. 74 (2):95–102. DOI: 10.1007/s11230-009-9198-2

Verovnik R., Trontelj P. & Sket B. 1999. Genetic differentiationand species status within the snail leech Glossiphonia com-planata aggregate (Hirudinea: Glossiphoniidae) revealed byRAPD analysis. Arch. Hydrobiol. 144 (3): 327–338.

Williams J.I. & Burreson E.M. 2006. Phylogeny of the fish leeches(Oligochaeta, Hirudinida, Piscicolidae) based on nuclear andmitochondrial genes and morphology. Zool. Scr. 35 (6): 627–639. DOI: 10.1111/j.1463-6409.2006.00246.x

Wong C.L., Ng S.M. & Phang S.M. 2007. Use of RAPD indifferentiation of selected species of Sargassum (Sargas-saceae, Phaeophyta). J. Appl. Phycol. 19 (6): 771–781. DOI:10.1007/s10811-007-9236-x

Yeh F.C., Yang R.C., Boyle T.B.J., Ye Z.H. & Mao J.X. 2000.POPGENE, the user-friendly shareware for population ge-netic analysis. Molecular Biology and Biotechnology Cen-tre, University of Alberta, Canada. http://www.ualberta.ca/∼fyeh/index.htm http://www.ualberta.ca/∼fyeh/popgene.html

Received October 18, 2011Accepted February 10, 2012

genetic variation and species identification among selected leeches (hirudinea) revealed by rapd...

Documents