Volume 165, Issue 4 pp. 795-808
Full Access

Phylogenetic relationships and taxonomy of Altiplano populations of Biomphalaria (Gastropoda: Planorbidae): inference from a multilocus approach

GONZALO A. COLLADO

GONZALO A. COLLADO

Laboratorio de Genética y Evolución, Departamento de Ciencias Ecológicas, Facultad de Ciencias, Universidad de Chile, Las Palmeras 3425, Ñuñoa, Santiago, Chile

Search for more papers by this author
MARCO A. MENDEZ

Corresponding Author

MARCO A. MENDEZ

Laboratorio de Genética y Evolución, Departamento de Ciencias Ecológicas, Facultad de Ciencias, Universidad de Chile, Las Palmeras 3425, Ñuñoa, Santiago, Chile

E-mail: [email protected]Search for more papers by this author
First published: 25 July 2012
Citations: 1

Abstract

Isolated closed basins provide a natural laboratory to study the differentiation among wild populations. Here we examined the phylogenetic relationships of the Southern Altiplano populations of Biomphalaria, a genus with medical importance, using nuclear (ITS1, ITS2) and mitochondrial (16S) ribosomal gene markers and a species of Helisoma as outgroup. Phylogenetic trees based on separate and combined analyses show that these populations form a particular lineage within Biomphalaria along with Biomphalaria peregrina (d'Orbigny, 1835) and Biomphalaria oligoza Paraense, 1974. The origin of this clade was estimated to have occurred in the middle Pleistocene. Molecular analyses showed that the nominal species Biomphalaria crequii (Courty, 1907) from the Salar de Ascotán and Biomphalaria costata (Biese, 1951) from the Salar de Carcote, previously synonymized with Biomphalaria andecola (d'Orbigny, 1835) and B. peregrina, respectively, are distinct taxa. Molecular data did not resolve the relationship of Biomphalaria aymara Valdovinos & Stuardo, 1991 from the Isluga swamps to other Biomphalaria species, but confirm that the populations from the Lauca and Huasco basins may represent a distinct undescribed species of Biomphalaria from the Southern Altiplano. Snails examined for trematodes were found to be positive in some Altiplano localities.

© 2012 The Linnean Society of London, Zoological Journal of the Linnean Society, 2012, 165, 795–808.

INTRODUCTION

The freshwater snails of the genus Biomphalaria Preston, 1910 are widely distributed in the Neotropical and African regions where some species are intermediate hosts of Schistosoma mansoni Sambon, 1907, a digenetic trematode which causes schistosomiasis (WHO, 1993, 1998). The taxonomy of Biomphalaria has been principally based on shell morphology and the anatomy of the reproductive system (Paraense, Fauran & Courmes, 1964; Paraense, 1966, 1981, 1984, 2003; Valdovinos & Stuardo, 1991), but the similarity of these characters among taxa has made classification difficult (Paraense, 1988; Estrada et al., 2006). In fact, convergent shell evolution among African Biomphalaria species has been detected, complicating species identification using conchological characters (Plam et al., 2008). The incorporation of molecular techniques, based mainly on restriction fragment length polymorphism (RFLP) analyses, has helped in the identification and recognition of species, although most of the studies have concentrated on Neotropical species (Caldeira et al., 1998, 2000; Vidigal et al., 1998, 2000a, b, 2001, 2002; Spatz et al., 1999, 2000; Velásquez et al., 2002). At the time of writing, 30–37 species are considered valid within Biomphalaria (DeJong et al., 2001; Jørgensen, Kristensen & Stothard, 2007; Caldeira, Teodoro & Carvalho, 2008).

In the south-western Altiplano of South America there is a great diversity of hydrological systems inhabited by populations of Biomphalaria. The physical structure of the area has been shaped primarily by intense volcanic activity from the Miocene to the Holocene and a series of palaeolakes that have been explained by climatic changes (Wörner et al., 2000; Fornari, Risacher & Féraud, 2001; Risacher, Alonso & Salazar, 2003; Strecker et al., 2007). At present, the landscape comprises isolated closed basins generally delineated by high chains of volcanoes (Risacher et al., 2003). The hydrological systems restricted to the basins constitute enclosed evaporitic environments. Thus, this Altiplano region constitutes an ideal natural scenario to study the evolutionary diversification of the aquatic biota. From this area have been described Biomphalaria costata (Biese, 1951) from the Salar de Carcote, Biomphalaria aymaraValdovinos and Stuardo, 1991 from the Isluga swamps, and Biomphalaria crequii (Courty, 1907) [= Biomphalaria grangei (Courty, 1907), G. A. Collado & M. A. Mendez, unpubl. data] from the Salar de Ascotán. However, the taxonomy of these nominal species and the identity of other populations of the genus recently discovered in the region are not completely resolved. For example, Malek (1985) using morphological characters argued that the species from the Salar de Ascotán [considered by him to be the nominal species Biomphalaria thermala (Biese, 1951)] correspond to Biomphalaria andecola (d'Orbigny, 1835), a species from Peru and Bolivia, while for Paraense (1966) B. costata is a synonym of Biomphalaria peregrina (d'Orbigny, 1835), a species widely distributed in the Neotropical region. In a recent phylogenetic study, we utilized DNA sequences of the mitochondrial gene cytochrome oxidase subunit I (COI) to study the relationship among Altiplano populations of Biomphalaria (Collado, Vila & Méndez, 2011). In this study, sequences of B. peregrina from La Plata and Mendoza were recovered as sister to the Altiplano taxa, while individuals from the Salar de Ascotán, assigned to B. crequii, were recovered in a strongly supported clade. Biomphalaria aymara and B. costata formed a monophyletic group and could not be differentiated phylogenetically. This study also showed that the populations restricted to the Lauca and Huasco basins possibly comprise a distinct subgroup of Biomphalaria, as well as the populations restricted to the Caquena basin.

In the present study we obtained DNA sequences of the mitochondrial 16S rRNA gene and the ITS1 and ITS2 nuclear internal transcribed spacers regions of the rRNA gene complex from Altiplano populations and other species of Biomphalaria to examine its monophyly and systematic position among other species of Biomphalaria through an inclusive, multilocus analysis of the group. We also assess whether B. crequii is a synonym of B. andecola. Additionally, considering that Collado et al. (2011) assessed the synonymy of B. costata and B. peregrina postulated by Paraense (1966) using COI data, which was not supported, in the present study we further evaluate this hypothesis incorporating multilocus sequences (16S, ITS1, ITS2) of B. peregrina from a widespread range of the distribution of the species in South America and B. costata from Salar de Carcote for comparative purposes. Available data from GenBank for B. peregrina include sequences from Uruguay, Brazil (Minas Gerais, Rio Grande Do Sul, Ourinhos) and Argentina (Mendoza) (Table 1). We additionally obtained original sequences of this species from La Plata, Argentina, thanks to samples generously provided by Dr Alejandra Rumi. Finally, we investigate the possible existence of putative species among Altiplano populations of Biomphalaria and briefly discuss the taxonomy of the Neotropical Biomphalaria schrammi (Crosse, 1864).

Table 1. Geographical origin of the samples of Biomphalaria and GenBank accession numbers of the DNA sequences used in the phylogenetic analyses
Species/population Accession number Origin
16S ITS1 ITS2
Biomphalaria alexandrina (Ehrenberg, 1831)a AY030203 AY030371 AY030371 Egypt (1)
AY030204 AY030372 AY030372 Egypt (2)
Biomphalaria amazonica Paraense, 1966a AY030216 AY030384 AY030384 Brazil (1)
AY030217 AY030385 AY030385 Brazil (2)
Biomphalaria angulosa Mandahl-Barth, 1957b DQ084848 DQ084889 Tanzania
Biomphalaria camerunensis (Boettger, 1941)a AY030198 AY030366 AY030366 Cameroon (1)
AY030199 AY030367 AY030367 Cameroon (2)
Biomphalaria choanomphala (Martens, 1879)a AY030202 AY030370 AY030370 Tanzania
Biomphalaria edisoni Estrada et al. (2006)c AY364446 AY364450 Colombia (1)
AY364448 AY364451 Colombia (2)
AY364449 AY364452 Colombia (3)
Biomphalaria glabrata (Say, 1818)a AY030206 AY030374 AY030374 Dominican Republic
AY030207 AY030375 AY030375 Puerto Rico
AY030208 AY030376 AY030376 Brazil
AY030209 AY030377 AY030377 Venezuela
Biomphalaria helophila (D'Orbigny, 1835)a AY030230 AY030399 AY030399 Cuba
Biomphalaria intermedia (Paraense & Deslandes, 1962)a AY030215 AY030383 AY030383 Paraguay
Biomphalaria kuhniana (Clessin, 1883)a AY030210 AY030378 AY030378 Dominican Republic
AY030211 AY030379 AY030379 Colombia
AY030212 AY030380 AY030380 Venezuela
Biomphalaria obstructa (Morelet, 1849)a AY030228 AY030397 AY030397 USA
AY030229 AY030398 AY030398 Mexico
Biomphalaria occidentalis Paraense, 1981a AY030221 AY030389 AY030389 Brazil
Biomphalaria oligoza Paraense, 1974d AF198679 Brazil (1)
AF198680 Brazil (2)
Biomphalaria peregrina (d’Orbigny, 1835)a AY030231 AY030400 AY030400 Uruguay
AY030232 AY030401 AY030401 Brazil (Minas Geraís1)
Biomphalaria peregrina (d’Orbigny, 1835)d AF198676 Brazil (Minas Geraís2)
AF198677 Brazil (RS)
Biomphalaria peregrina (d’Orbigny, 1835)e GU168591 Argentina (Mendoza1)
GU168592 Argentina (Mendoza2)
Biomphalaria peregrina (d’Orbigny, 1835)f DQ143961 Brazil (Ourinhos1)
DQ143962 Brazil (Ourinhos2)
Biomphalaria peregrina (d’Orbigny, 1835)g JF309030 Argentina (La Plata1)
Biomphalaria peregrina (d’Orbigny, 1835)g JF309031 JF308974 JF309001 Argentina (La Plata2)
Biomphalaria peregrina (d’Orbigny, 1835)g JF309032 Argentina (La Plata3)
Biomphalaria pfeifferi (Krauss, 1848)a AY030193 AY030361 AY030361 Senegal
AY030194 AY030362 AY030362 Cameroon
AY030195 AY030363 AY030363 Sudan
AY030196 AY030364 AY030364 Madagascar
Biomphalaria prona (Martens, 1873)a AY030222 AY030390 AY030390 Venezuela (1)
AY030223 AY030391 AY030391 Venezuela (2)
Biomphalaria schrammi (Crosse, 1864)a AY030233 AY030402 AY030402 Brazil (Minas Geraís1)
Biomphalaria schrammi (Crosse, 1864)d AY425743 AF198681 Brazil (Minas Geraís2)
AY425744 AF198682 Brazil (Minas Geraís3)
Biomphalaria smithi Preston, 1910a AY030205 AY030373 AY030373 Uganda
Biomphalaria sp. Boliviaa AY030218 AY030386 AY030386 Bolivia
Biomphalaria sp. Cubaa AY030224 AY030393 AY030393 Cuba (1)
Biomphalaria sp. Cubaa AY030225 AY030394 AY030394 Cuba (2)
Biomphalaria sp. Puerto Ricoa AY030226 AY030226 AY030226 Puerto Rico
Biomphalaria stanleyi (Smith, 1888)a AY030197 AY030365 AY030365 Uganda
Biomphalaria straminea (Dunker, 1848)a AY030213 AY030381 AY030381 Brazil (1)
AY030214 AY030382 AY030382 Brazil (2)
Biomphalaria sudanica (Martens, 1873)a AY030200 AY030368 AY030368 Kenya
AY030201 AY030369 AY030369 Tanzania
Biomphalaria temascalensis Rangel-Ruiz, 1987a AY030227 AY030396 AY030396 Mexico
Biomphalaria tenagophila (d’Orbigny,1835)a AY030219 AY030387 AY030387 Paraguay
AY030220 AY030388 AY030388 Brazil
Biomphalaria sp. Umaquig GU168043 JF308977 JF309004 Umaqui (1)*
Biomphalaria sp. Viocog GU168044 JF309005 Vioco (1)*
Biomphalaria sp. Visvirig GU168045 JF308978 JF309006 Visviri (1)*
Biomphalaria sp. Lauca Surg GU168046 JF308984 Lauca Sur (1)*
Biomphalaria sp. Piacotag GU168049 JF308987 JF309015 Piacota (1)*
JF309036 JF308988 JF309016 Piacota (3)*
Biomphalaria sp. Laucag GU168050 JF308983 JF309012 Lauca (1)*
Biomphalaria sp. Caquenag GU168052 JF308975 JF309002 Caquena (1)*
Biomphalaria sp. Chungarág GU168053 JF309007 Chungará (3)*
Biomphalaria sp. Cotacotanig GU168054 JF308981 JF309008 Cotacotani (2)*
Biomphalaria sp. Chuvirig GU168055 JF308980 JF309010 Chuviri (1)*
Biomphalaria sp. Colpag JF309033 JF308976 JF309003 Colpa (2)*
Biomphalaria sp. Parinacotag JF309035 JF308985 JF309013 Parinacota (1)*
GU168058 JF308986 JF309014 Parinacota (3)*
Biomphalaria sp. Salar de Huascog GU168059 JF308982 JF309011 Salar de Huasco (1)*
Biomphalaria sp. Chibatambog JF309034 JF308979 JF309009 Chibatambo (1)*
Biomphalaria aymara Valdovinos & Stuardo, 1991 and Stuardo, 1991g GU168056 JF308989 Isluga (1)*
GU168057 JF308990 JF309017 Isluga (2–6)*
JF309041 JF309018 Isluga (3)*
JF309042 JF309019 Isluga (4)*
Biomphalaria costata (Biese, 1951)g JF309037 JF308991 JF309020 Salar de Carcote (1V1)*
GU168048 JF308992 JF309021 Salar de Carcote (6V1)*
JF309038 JF308993 JF309022 Salar de Carcote (11V1)*
JF309039 JF308994 JF309023 Salar de Carcote (14V1)*
JF309040 JF308995 JF309024 Salar de Carcote (16V1)*
Biomphalaria crequii (Courty, 1907)g GU168047 JF308996 JF309025 Salar de Ascotán (1V2)*
GU168051 JF308997 JF309026 Salar de Ascotán (6V2)*
JF309043 JF308998 JF309027 Salar de Ascotán (18V2)*
JF309044 JF308999 JF309028 JF309025 Salar de Ascotán (21V2)*
JF309045 JF309000 JF309029 Salar de Ascotán (23V2)*

MATERIAL AND METHODS

Dataset

We compiled DNA sequences of the 16S rRNA gene and ITS1 and ITS2 region available in GenBank from several sources (Table 1) and obtained original sequences of these three molecular loci from Altiplano populations of Biomphalaria and other species of the genus (see below). The sequences used in this study were sourced from across a large fraction of the generic range and thus we can estimate the systematic position of the populations of Biomphalaria restricted to the Neotropical Southern Altiplano. We used Helisoma trivolvis as the outgroup for molecular analyses based on previous phylogenetic studies (Bandoni, Mulvey & Loker, 1995; DeJong et al., 2001; Morgan et al., 2002); its sequences were obtained by DeJong et al. (2001).

Sampling and DNA extraction

We sampled snails of populations of Biomphalaria distributed in the southern Altiplano (18–22°S), which were fixed in absolute alcohol for DNA extraction. Some living specimens were dissected using a stereoscopic microscope to investigate the presence of trematodes. Snails were sampled using a small sieve from macrophytes in localities belonging to six closed basins located at more than 3500 m elevation (Fig. 1). The Caquena basin included five localities (Visviri, 4048 m altitude; Umaqui, 4132 m; Vioco, 4383 m; Colpa, 4384 m; Caquena, 4398 m), and the Lauca basin seven (Parinacota, 4399 m altitude; Chuviri, 4374 m; Piacota, 4427 m; Chibatambo, 4232; Lauca, 4395 m; Lauca Sur, 4232 m; Chungará lake, 4583 m). The basin of the Salar de Huasco (3805 m altitude), Isluga swamps (3776 m), Salar de Carcote (3688 m; spring 1) and Salar de Ascotán (3716 m; spring 2) had one locality each.

Details are in the caption following the image

Sampling localities of populations of Biomphalaria of the southern Altiplano. The number of localities sampled by basin is shown in parentheses. Asterisks indicate basins where Biomphalaria snails were not found.

Genomic DNA was extracted using the CTAB method (cetyltrimethyl ammonium bromide) (Winnepennickx, Backeljau & De Wachter, 1993). We separated the shell of the animals from the soft tissues, and a small part of the foot muscle tissue was dissected and placed in a tube with 600 µL of 100 mm Tris, 1.4 m NaCl2, 20 mm EDTA, 2% CTAB, 0.2% β-mercaptoethanol and 0.01 mg proteinase K. Samples were macerated and incubated at 65 °C for 40 min with agitation every 15 min. Proteins and lipids were removed using 600 µL chloroform by centrifuging at 16 500 g for 10 min. DNA was precipitated in cold isopropanol for 30 min at 4 °C and then centrifuged at 16 500 g for 15 min. Finally, DNA was washed in 70% ethanol, dried and resuspended in 50 µL distilled water.

DNA amplification

We obtained partial sequences from 1–5 specimens per basin and locus. To amplify the 16S mitochondrial gene we used the primers 16Sar (5′-CGCCTGTTTATCAAAAACAT-3′) and 16Sbr (5′-CCGGTCTGAACTCAGATCACGT-3′) (Palumbi, 1996). We amplified the ITS1 nuclear region using primers ETTS2 (5-TAACAAGGTTTCCGTAGGTGAA-3) (Kane & Rollinson, 1994) and ITS1R (5-ACGAGCGAGTGATCCACCGC-3) (Vidigal et al., 2004). The ITS2 nuclear region was amplified using primers ETTS1 (5′-TGCTTAAGTTCAGCGGGT-3′) (Kane & Rollinson, 1994) and ITS2F (5-CGTCCGTCTGAGGGTCGGTTTGC-3) (Vidigal et al., 2000c). PCR amplifications were performed in a total reaction volume of 50 µL with 0.8 U Taq DNA polymerase, 15 mm MgCl2, 200 µm dNTPs and 1 µm of each primer. The PCR programme for 16S included an initial denaturation at 94 °C for 60 s, 35 cycles of 94 °C for 30 s, 55 °C for 50 s and 72 °C for 60 s, and a final extension at 72° for 10 min. The PCR programme for the nuclear regions was an initial denaturation at 95 °C for 60 s, 34 cycles of 94 °C for 45 s, 54 °C for 60 s and 72 °C for 2 min, and a final extension at 72° for 5 min. Amplified products were sequenced by Macrogen Company, South Korea. We incorporated 33 original sequences for 16S (GenBank accession numbers GU168043–GU168059 and JF309030–JF309045), 27 for ITS1 (GenBank accession numbers JF308974–JF309000), and 29 for ITS2 (GenBank accession numbers JF309001–JF309029) (see Table 1).

Sequence edition and phylogenetic analysis

DNA sequences were edited with the BioEdit program (Hall, 1999) and aligned with Clustal X (Thompson et al., 1997), with a final adjustment by visual inspection. For phylogenetic reconstructions we used the maximum-parsimony (MP), maximum-likelihood (ML) and Bayesian analysis (BA) methods. MP analysis was performed with PAUP* (Swofford, 2003), using a heuristic search with the TBR algorithm with the addition of random sequences and retaining 100 trees in each analysis. Characters not informative for parsimony were excluded from the analyses. The statistical confidence of the nodes was evaluated using 100 bootstrap pseudoreplicates (Felsenstein, 1985). ML analysis was also performed with PAUP*, selecting the best evolutionary model with the jModelTest 0.1.1 program (Posada, 2008) under the Akaike Information Criterion. BA was performed using the program MrBayes v. 3.1.2 (Ronquist & Huelsenbeck, 2003), selecting the best evolutionary model with jModelTest under the Bayesian Information Criterion. This analysis was only applied to the groups of the combined data set and was run three times for 5 million generations. Trees were sampled every 1000 generations and posterior probabilities were based on 4501 trees obtained after the burn-in period (10%). With the original sequences determined in this study and those obtained from GenBank we constructed data matrices with 78 sequences for the 16S gene, 75 sequences for ITS1 and 83 sequences for ITS2, which were analysed separately and using total evidence, which allows data to be combined (Kluge, 1989). The combined analysis was performed using two strategies. One approach was to construct a data matrix which included all the taxa for which sequences were available for the three loci we used. The second included taxa for which we did not have sequences for some loci; in this case, entries for the incomplete taxa included ‘missing data’. This has been shown to increase the phylogenetic resolution without producing ambiguous results (Wiens & Reeder, 1995; Wiens, 1998). Using this procedure we obtained a matrix of 67 concatenated sequences in the first case and a matrix of 90 sequences with the second strategy.

Divergence times

To estimate divergence times of the clades and lineages we constructed a tree based on Bayesian inference using the program BEAST (version 1.5.3; Drummond & Rambaut, 2007), which employs a Bayesian Markov chain Monte Carlo (MCMC) algorithm. With this program the distance of the extant terminal taxa and the hypothetical ancestors in the node may be converted to divergence times if there is a reference date, such as a fossil record or geological event, or a known mutation rate. We used a nucleotide substitution rate of 1.6–2.2% per million years for the 16S gene, as estimated for invertebrates (Cunningham, Blackstone & Buss, 1992; Patarnello et al., 1996). In this analysis we use only one sequence by species and population of Biomphalaria. BEAST analysis was performed using a strict molecular clock, the general time-reversible (GTR) substitution model and the Yule process of speciation. We combine output files from independent runs of BEAST using LogCombiner (version 1.5.3). Convergence of posterior distributions for parameter estimates was analysed in Tracer (version 1.5; Rambaut & Drummond, 2007), and the Bayesian tree was obtained after removing burn-in (10%) with TreeAnnotator (version 1.5.3).

RESULTS

Separate analyses of the three datasets (Table 2) yielded trees with well-supported clades that in general were congruent among data partitions. Under the MP analysis we constructed a majority-consensus tree (not shown) of 100 equally parsimonious trees for each dataset (see Table 2 for statistics). The sequences of B. schrammi were located outside of Biomphalaria with 100% bootstrap support in all separate analyses performed. The MP 16S analysis recovered a diverse clade (79% bootstrap support) composed of several subclades including the North American and Caribbean species, the majority of the Neotropical species and all the African species, and its sister group composed of B. peregrina and the Altiplano populations of Biomphalaria (100%). Within this clade were recovered four subclades corresponding to sequences from Lauca and Salar de Huasco basins (85%), Salar de Ascotán (54%), Salar de Carcote (69%), and B. peregrina from Argentina (67%). The remainder of the Altiplano sequences and those of B. peregrina from Uruguay and Brazil were not resolved by this gene. The ML 16S analysis (GTR + G; −lnL = 2284.154) produced three trees which recovered the same main clades and subclades inferred by MP analysis. The MP ITS1 analysis resolved three main clades, the largest composed by the African species and its sister B. glabrata (100%), and a subclade rich in Neotropical species (70%). A second clade included the species from Mexico and the Caribbean (100%); the third clade contained B. peregrina and the Altiplano populations of Biomphalaria (100%). These clades did not show resolution among themselves or with respect to B. helophila, which formed an independent branch in the tree. Within the B. peregrina/Altipano clade we found resolution of the sequences from the Salar de Ascotán (72%), Lauca and Salar de Huasco (52%), and B. peregrina (71%). The ML ITS1 analysis (GTR + G; −lnL = 3090.3933) produced a tree which recovered the same clades inferred by MP, but with the North American and the Caribbean species and B. helophila as consecutive sister groups of the rest of the species. Within the B. peregrina/Altipano clade the sequences from the Lauca and Salar de Huasco and B. peregrina were rseolved. The MP ITS2 analysis recovered five clades, including the B. glabrata and the African species clade (100%) and the clade comprising B. peregrina and the Altiplano taxa, which also included B. oligoza (100%) whose sequences were available for this locus. Within this clade were resolved the sequences from the Salar de Carcote (56%) and B. peregrina together with B. oligoza (71%). The remainder of the Neotropical species formed three well-supported but not resolved clades. As in the 16S and ITS1 analyses, the position of B. helophila was not resolved. The ML ITS2 analysis (TVM + G; −lnL = 3877.094) produced a topology similar to MP analysis, including a cluster comprising B. peregrina, B. oligoza, and the Altiplano taxa.

Table 2. Summary of the amplified sequences of the three locus and tree statistics obtained under maximum parsimony (MP) analysis
Locus Nucleotide composition N F L MP index
A C G T I TL CI RI
16S 36.4 10.5 15.7 37.4 33 412–439 457 108 307 0.49 0.88
ITS1 22.4 25.8 29.3 22.4 27 385–395 494 222 460 0.73 0.97
ITS2 16.8 26.8 27.7 28.7 29 418–471 528 183 594 0.58 0.89
  • CI, consistency index; F, length of the fragment amplified; I, number of parsimony-informative characters; L, final length of the alignment; N, number of individuals sequenced; RI, retention index; TL, tree length.

The combined analysis with the matrix of 90 sequences recovered the same main clades and produced nodal support values similar to those obtained in the combined analysis of 67 sequences and for this reason we focus on the first analysis. The analysis with the matrix of 90 sequences included 1479 characters, of which 513 were parsimony-informative. Excluding B. schrammi, which was located outside Biomphalaria, the majority consensus tree (Fig. 2) of 100 equally parsimonious trees (length = 1402, consistency index = 0.59, retention index = 0.92) recovered a main clade formed by B. helophila, species from North America and the Caribbean, a conglomerate of mainly South American species, and a subclade that includes B. glabrata and the African species (100% bootstrap support). Recovered as sister to this group was a clade comprising B. peregrina, B. oligoza, and the Altiplano populations of Biomphalaria. Within this group, sequences of the first two taxa were recovered as a monophyletic group (65%), as well as the sequences from the Salar de Carcote (93%) and the Salar de Ascotán (96%). The ML analysis with the matrix of 90 sequences recovered the North American and Caribbean species being the sister group of the rest of the species of Biomphalaria (100%) (excluding B. schrammi) (Fig. 3), and the same two groups recovered by the MP analysis. BA recovered a consensus tree (Fig. 4) similar to that of the ML analysis. The North American and the Caribbean species also appear as the sister group of the rest of the species of Biomphalaria (excluding B. schrammi), but the conglomerate of mainly South American species was recovered as the sister group of the clade composed of B. glabrata and the African species. The clade comprising B. peregrina, B. oligoza, and the Altiplano taxa was recovered as the sister group of this last main clade in the ML analysis, but without bootstrap support.

Details are in the caption following the image

Majority consensus tree obtained from the maximum-parsimony analysis using the combined matrix of 90 sequences. Numbers at nodes indicate bootstrap support values (only those above 50% are given). The origin of the sequences is shown using numbers or initials (see Table 1).

Details are in the caption following the image

Tree obtained from the maximum-likelihood analysis using the combined matrix of 90 sequences. Numbers at nodes indicate bootstrap support values (only those above 50% are given). The origin of the sequences is shown using numbers or initials (see Table 1).

Details are in the caption following the image

Consensus tree obtained from Bayesian analysis using the matrix of 90 sequences. Numbers at nodes indicate posterior probability values (only those above 0.94 are given). The origin of the sequences is shown using numbers or initials (see Table 1).

Divergence times

The divergence times between the Biomphalaria clade and B. schrammi and the outgroup clade was estimated at 7.76–6.77 Mya. The origin of the Biomphalaria clade (excluding B. schrammi) occurred in the Early Pliocene (≈ 5.33–4.59 Mya). Divergence between B. glabrata and the African species was estimated to about 2.55–2.20 Mya, in agreement with previous estimates (Woodruff & Mulvey, 1997; Campbell et al., 2000; DeJong et al., 2001). Divergence within the new Neotropical clade occurred in the middle and late Pleistocene (Fig. 5).

Details are in the caption following the image

Reconstruction of the divergence times of lineages of Biomphalaria estimated using a substitution rate of 1.6–2.2% per million years for the 16S locus. We used one sequence per population/species to generate a Bayesian tree. Numbers at nodes represent millions of years. The origin of the sequences is shown using numbers or initials (see Table 1).

DISCUSSION

The majority consensus tree produced by the combined MP analysis revealed that the genus Biomphalaria (excluding the clade formed by the sequences of B. schrammi) is divided phylogenetically into four subclades. One subclade includes the majority of the Neotropical species, a second includes B. glabrata and all the African species, and other the North American and Caribbean species of the genus. Biomphalaria helophila was recovered as sister group to these three main subclades. A fourth subclade comprised B. peregrina, B. oligoza, and the Altiplano taxa, which are sister to the remaining species. The ML tree and the consensus tree produced by BA using the combined matrix of 90 sequences show some incongruence with the MP analysis in the sense that the first two analyses recovered the species from North American and Caribbean as the sister group of the rest of the species of Biomphalaria (excluding B. schrammi), although both analyses recovered the clade composed of B. peregrina, B. oligoza, and the Altiplano taxa with 100% bootstrap support and a posterior probability of 1.00, respectively. This monophyletic group had 100% bootstrap support in the MP separate analyses of the three dataset. The shortest fragment of ITS1 reported by DeJong et al. (2001) in B. peregrina compared with other species of the genus, probably indicative of deletion events according to these authors, was also found for the Altiplano populations, suggesting a common ancestor. Molecular analyses also revealed that the clade B. peregrina/B. oligoza/Altiplano taxa is separated from the other clades by a long branch, probably indicative of the differences in evolutionary rates among species and supporting the designation of a subgenus within Biomphalaria for this group. In fact, Collado et al. (2011) postulated a ‘Southern Altiplano species complex’ of Biomphalaria for the Altiplano taxa. In general, within the B. peregrina/B. oligoza/Altiplano taxa the sequences of B. costata and B. crequii were well resolved with high node support values; they and others in this clade are closely related taxa. Biomphalaria peregrina and B. oligoza were also recovered as closely related species in previous phylogenetic analyses using the ITS2 locus (Vidigal et al., 2000c) and RFLP profiles (Spatz et al., 2000).

The phylogenetic hypothesis postulated by DeJong et al. (2001), one of 234 equally parsimonious trees, topologically congruent with the ML tree, recovered B. peregrina as the sister group to the rest of the species of the genus (DeJong et al., 2001). The MP analysis confirms this inference, but now with a clade of B. peregrina and its close relatives that is widespread across South America. ML and BA recovered the North American and Caribbean species of Biomphalaria as the sister group of the remaining species of the genus (excluding B. schrammi). It has been suggested that Biomphalaria could have originated in North America (DeJong et al., 2001; Morgan et al., 2002), and both topologies agree with this hypothesis.

Taxonomic implications

The molecular phylogenetic analyses showed that the synonymy of the nominal species described for the Chilean Altiplano, i.e. B. crequii and B. costata with B. andecola and B. peregrina, respectively, is not appropriate. In the first case we could only compare the ITS2 region as only this sequence is available for B. andecola; however, we also included this taxon in the combined analysis of 67 and 90 sequences. All these analyses showed that B. andecola forms part of the species-rich Neotropical clade (Dejong et al., 2001) within which it forms a monophyletic group with B. prona. The sequences of B. crequii from the Salar de Ascotán, its type locality (see Courty, 1907), in general were recovered as a strongly supported group within the clade B. peregrina and its close relatives. Thus, both species belong to different clades and represent distinct lineages. For B. peregrina and B. costata we were able to compare sequences of the three loci. The separate and combined analysis indicated that the sequences of B. costata from the Salar de Carcote (spring 1), its type locality (see Biese, 1951), never conform to a monophyletic group compared with those of B. peregrina, which groups with B. oligoza. In the 16S ML analysis the three sequences of B. peregrina from La Plata amplified in the present study formed a monophyletic group with two sequences of Mendoza and two from Brazil and Uruguay reported for this species (DeJong et al., 2001; Estrada et al., 2006); the molecular evidence does not support the synonymy established for these taxa. The COI data obtained by Collado et al. (2011) also show that B. costata from the Salar de Carcote is a distinct lineage of B. peregrina. Radular morphology (G. A. Collado & M. A. Mendez, unpubl. data) confirms the separation of the two taxa. We cannot be conclusive with respect to the Altiplano species B. aymara due to the scarce resolution obtained within its clade, although its sequences are never sister to those of B. costata and B. crequii. The populations from the Lauca and Huasco basins were recovered as a strongly supported group in the 16S MP analysis (85% bootstrap support), confirming previous studies using COI sequence data that suggested that these populations could represent a candidate species of Biomphalaria from the Southern Altiplano (Collado et al., 2011). The ML analysis also recovered this monophyletic group. The populations restricted to the Caquena basin were not resolved in the present study, as compared with previous results (Collado et al., 2011).

Although a number of phylogenetic studies of Biomphalaria have included B. schrammi, its systematic position within the genus has been placed in doubt. Vidigal et al. (2000c) suggested that B. schrammi is a taxon ‘intermediate’ between H. trivolvis and other species of Biomphalaria. DeJong et al. (2001) suggested that B. schrammi is equally distant from other Biomphalaria and from Helisoma from the genetic point of view, and may represent either a separate or an intermediate lineage. The 5.8S sequences obtained by DeJong et al. (2001) were identical for all the Biomphalaria species studied and for H. trivolvis, with the exception of B. schrammi, which had two nucleotide differences. The phylogenetic study of Morgan et al. (2002) using ribosomal 28S RNA and actin exon placed B. schrammi within a clade formed by eight species of Biomphalaria, which supports the inclusion of this species within the genus. However, this result may be due to the small number of species considered and/or the more evolutionarily conserved nature of this molecular locus. The present study, the most inclusive in terms of number of Biomphalaria taxa utilized, in all analyses showed B. schrammi forming a lineage independent of the rest of the species of the genus.

Trematode infection

Within the clade B. peregrina/B. oligoza/Altiplano taxa the species B. peregrina is considered a potential intermediate host of S. mansoni (Paraense & Corrêa, 1973) while B. oligoza is resistant to this trematode (see Spatz et al., 2000). We have found some snails of the species B. aymara collected from the Isluga swamps infected with adult trematodes (Fig. 6). Although there are no records of local cases of schistosomiasis in Chile (OMS, 2008), Paré & Black (1999) reported schistosomiasis in the Chilean flamingo Phoenicopterus chilensis Molina, 1782 caused by trematodes tentatively assigned to the genus Dendritobilharzia Sktjabin & Zakharow (1920). Given that several species of trematodes infect Biomphalaria, the poorly characterized digenean fauna in South America (DeJong et al., 2001), and that the ability of these snails to act as an intermediate host for trematodes appears to have evolved following several separate acquisition events during its evolutionary history (Vidigal et al., 2000c; DeJong et al., 2001), the species and the definitive hosts of the helminthes discovered in the present study need to be elucidated.

Details are in the caption following the image

Trematodes parasitizing Biomphalaria aymara from Isluga, Chilean Altiplano. S, shell; T, trematodes in the soft body.

CONCLUSIONS

Molecular phylogenetic analyses support the monophyly of the clade formed by the Altiplano taxa, B. peregrina, and B. oligoza with strong branch support. MP analysis recovered this clade as sister to the remaining species of Biomphalaria (excluding B. schrammi). Given that ML and BA recovered the clade formed by the North American and Caribbean species as sister to remaining species of Biomphalaria (excluding B. schrammi), further work is needed to clarify the relationships in basal nodes. Taxonomically the present study shows (1) that the nominal species B. crequii is phylogenetically differentiated from B. andecola; they fall in different clades. (2) The synonymy of B. costata and B. peregrina was not supported; the results suggest that they represent distinct lineages within the same clade. (3) The specific status of B. aymara could not be resolved in this study. (4) Snails from some Altiplano localities are infected with trematodes. (5) Biomphalaria schrammi in all the analyses performed appears outside of Biomphalaria, suggesting that it does not belong to this lineage.

Finally, our results suggest that species diversity is possibly underestimated at the Chilean Altiplano and additional sampling is required.

ACKNOWLEDGEMENTS

We thank Dr Sergio Letelier, curator of the Malacology section at the Museo Nacional de Historia Natural de Santiago, Chile, for his collaboration during the development of this study. We also thank Dr Alejandra Rumi for providing samples of B. peregrina from Argentina. This study was partially financed by project MULT 05/04-2 and Domeyko: Biodiversidad Iniciativa Transversal 1 of the Universidad de Chile; FONDECYT 1061256, 1080390 and 1110188. G. Collado thanks CONICYT: Programa Nacional de Becas de Postgrado- Doctorado 2005 No. 21050661 and FONDECYT 3110072.

      The full text of this article hosted at iucr.org is unavailable due to technical difficulties.