Published in Journal of Biogeography 38, issue 4, 640-650, 2010 which should be used for any reference to this work 1 COI gene and ecological data suggest size-dependent high dispersal and low intra-specific diversity in free-living terrestrial protists (Euglyphida: Assulina) Enrique Lara1*, Thierry J. Heger1,2,3 , Rodrigo Scheihing4,5 and Edward A. D. Mitchell1 1 Laboratory of Soil Biology, Institute of Biology, University of Neuchâtel, Rue Emile Argand 11, 2000 Neuchâtel, Switzerland, WSL, Swiss Federal Institute for Forest, Snow and Landscape Research, Ecosystem Boundaries Research Unit, Wetlands Research Group, Station 2, 1015 Lausanne, Switzerland, 3 Ecole Polytechnique Fédérale de Lausanne (EPFL), Laboratory of Ecological Systems, Station 2, 1015 Lausanne, Switzerland, 4 Centro de Estudios Cientı́ficos (CECS), Casilla 1469, Valdivia 5110246, Chile, 5 Escuela de Graduados, Facultad de Ciencias, Universidad Austral de Chile, Valdivia, Chile 2 ABSTRACT Aim Propagule size and ecological requirements are believed to be major factors influencing the passive dispersal of free-living terrestrial protists. We compared the colonization potential of three closely related testate amoeba species (Assulina muscorum, A. seminulum, A. scandinavica, ranging from 40 to 100 lm in length). Location Europe. Methods We collected individual Assulina species cells from Sphagnum peatlands across Europe. We sequenced a 550-bp fragment of the mitochondrial cytochrome c oxidase subunit I gene (COI) to estimate the within-species variability, as a proxy for gene flow. We reviewed existing ecological and palaeoecological data to assess the ecological tolerance of Assulina species and how rapidly they colonized developing peatlands. Results We obtained COI sequences for 30 individuals of A. seminulum from eleven sites, for 39 of A. muscorum from six sites, and for six of A. scandinavica from two sites. We observed three haplotypes for A. seminulum and two for A. muscorum, often co-existing in the same sites. The sequences of A. scandinavica from the two sites were identical. Assulina muscorum and A. seminulum haplotypes varied by only 1–2 nucleotides, resulting in >99.5% similarity. Genetic diversity within A. seminulum was higher than that within A. muscorum. Ecological and palaeoecological records showed that A. muscorum was much more frequent and abundant than A. seminulum, and had a somewhat broader ecological tolerance for pH, moisture and water-table depth. Assulina muscorum always appeared early during the developmental history of peatlands, either before or simultaneously with A. seminulum. Main conclusions The lack of genetic structure observed with a variable marker such as COI suggests high gene flow between the sites and thus rapid transport (at an evolutionary scale) over large distances, in agreement with the palaeoecological records. Thus, geographical distance alone does not seem to prevent the dispersal of testate amoebae, at least within Europe. Nevertheless, genetic diversity was significantly lower within A. muscorum than within A. seminulum, suggesting that its smaller size and abundance and/or broader ecological tolerance influence its effective dispersal capacity. These results are in agreement with the often earlier colonization of peatlands by A. muscorum and its broader ecological tolerance. Keywords Assulina, COI gene, dispersal, Europe, palaeoecological records, peat bog, Sphagnum, terrestrial, testate amoebae, Western Palaearctic. *Correspondence: Enrique Lara, Laboratory of Soil Biology, Institute of Biology, University of Neuchâtel, Rue Emile Argand 11, 2000 Neuchâtel, Switzerland. E-mail: [email protected] Present address: Biodiversity Research Center, University of British Columbia, Vancouver, BC, Canada V6T 1Z4. 2 INTRODUCTION The existence or absence of biogeographical patterns in the distribution of microorganisms is a long-standing yet unresolved debate. Defenders of the ‘ubiquity theory’ claim that all microorganisms can be found anywhere on Earth as long as suitable conditions are met (Finlay & Clarke, 1999; Finlay et al., 1999, 2001; Finlay & Fenchel, 2004), while proponents of the ‘moderate endemicity theory’ claim that at least some species have limits to their distribution ranges (Foissner, 2006; Smith & Wilkinson, 2007; Foissner et al., 2008; Smith et al., 2008; Van de Vijver & Mataloni, 2008; Vanormelingen et al., 2008). Considerable efforts have been invested over the last decade to acquire the new data needed to inform this debate (Caron, 2009). Several terrestrial testate amoeba species are increasingly cited as examples corroborating the moderate endemicity theory, because some morphologically very distinctive species, such as Apodera vas (=Nebela vas), have been shown to present truly geographically limited ranges (Smith & Wilkinson, 2007; Smith et al., 2008). In most other cases, however, morphology alone cannot be used to delimit reliable taxonomic units (Heger et al., 2009), and studies based on intra-morphospecies genetic diversity are urgently needed (Heger et al., 2010a). The main dispersal agent of testate amoebae is thought to be the wind (Wilkinson, 2001), but other vectors such as birds, mammals, insects or human activities have been reported for other soil protists (Revill et al., 1967; Schlichting & Sides, 1969; Charalambidou & Santamarı́a, 2002; Wilkinson, 2010). Several factors have been proposed in the literature to explain the distribution range of testate amoebae: test (shell) size (Wilkinson, 2001; Yang et al., 2010), population size (Finlay & Fenchel, 2004) and the ability to form drought-resistant propagules (cysts) (Corliss & Esser, 1974; Foissner, 1987). The ecological tolerance spectrum (Mitchell & Meisterfeld, 2005) and the spatial distribution of favourable habitats in the landscape are also potential, but less often discussed, factors. The potential habitat of a species with a narrow ecological spectrum can be hypothesized to be patchier than the generalists, and these species will therefore be less likely to reach a suitable habitat by random dispersal. The distribution of species observed today can be considered as a snap-shot of a pattern changing with time (Lara et al., Figure 1 Scanning electron microscope images of (a) Assulina muscorum, (b) A. seminulum and (c) A. scandinavica. The three cells are represented at the same scale. Scale bar = 20 lm. (a) 2008; Smith et al., 2008). Factors controlling this pattern therefore also influence the pace at which new environments are colonized as well as the genetic patterns at different spatial scales. A propagule from a large, rare, drought-sensitive species that is restricted to a very particular environment will theoretically have few chances to be transported to a new suitable habitat and to form a new population. These events will thus happen rarely. Conversely, a small, abundant, cystforming species with a broad ecological spectrum will have many chances to form new populations. Thus, we can expect that the first species will build new populations at a slower rate than the second; in other words, it will disperse more slowly. A second corollary of that example is that, as propagules move more slowly in the first case than in the second, mutations will also spread more slowly, and we can thus expect a higher degree of genetic differentiation among populations than in the second case (assuming that selective pressure is homogeneous in all cases and that generation times are similar). In this study we aimed to bring new elements to the cosmopolitanism versus endemism debate by estimating the colonization potential of three closely related species of euglyphid terrestrial testate amoebae (Lara et al., 2007): (1) Assulina scandinavica Penard, 1890, a large (c. 100 lm long and 1.9 · 105 lm3), rare species restricted to Sphagnum habitat; (2) A. seminulum (Ehrenberg, 1848) Leidy, 1879, a medium-sized (c. 85 lm long and 0.55 · 105 lm3), more frequent but also ecologically restricted species; and (3) A. muscorum Greeff, 1888, a small (c. 45 lm long and 0.12 · 105 lm3), frequent and tolerant species (Fig. 1). We compared the genetic differentiation among populations at the European scale and reviewed ecological and palaeoecological data to determine the colonization pace of these species during peatland development and the breadth of their respective ecological niches. MATERIALS AND METHODS Sampling Sphagnum mosses potentially containing Assulina species were collected from 13 locations in northern and central Europe (Fig. 2; Table 1). Cells were then extracted from wet Sphagnum mosses and isolated using inverted microscopy. Only living (b) (c) 3 Figure 2 Map of the study area showing the locations of our 13 sampling sites of Assulina species in Denmark, Estonia, Poland, Russia, Switzerland (inset), Sweden and the United Kingdom. Table 1 List of sites from which Assulina taxa were sampled for COI gene sequencing. Haplotypes A. muscorum A. seminulum Site Country Latitude Longitude AM1 AM2 AS1 AS2 AS3 A. scandinavica Praz Rodet Sortel Lörmoos Grindelwald Chaux d’Abel Glenn Dee Holmegard Moose Ryggmossen Bory Tucholskie Moszne Verkhozimskie Zuratkul Pikassaare Switzerland Switzerland Switzerland Switzerland Switzerland UK Denmark Sweden Poland Poland Russia Russia Estonia 4634¢ 4644¢ 4659¢ 4637¢ 4759¢ 5659¢ 5516¢ 6002¢ 5336¢ 5116¢ 4304¢ 5453¢ 5925¢ 610¢ E 722¢ E 725¢ E 759¢ E 655¢ E 339¢ W 1152¢ E 1719¢ E 1800¢ E 2260¢ E 4623¢ E 5913¢ E 2551¢ E 3 0 18 0 2 0 0 0 3 0 4 7 0 0 0 0 0 0 0 0 0 2 0 0 0 0 2 0 3 0 0 0 1 0 3 1 5 0 0 1 1 2 2 0 0 0 1 0 1 1 0 4 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 3 3 0 0 0 0 0 0 0 N N N N N N N N N N N N N cells were taken: these could easily be recognized by the presence of a nucleus surrounded by small vesicles of granular appearance. Each cell was isolated with a narrow-diameter pipette under an inverted microscope, washed in distilled water, deposited individually (one cell per tube) inside a polymerase chain reaction (PCR) tube containing 10 lL of PCR reaction buffer with concentrations of 1.5 mm MgCl2 and 10 nmol of each dNTP (deoxynucleotide triphosphate) (Promega), and stored at )20 C or processed immediately for the PCR reaction. Scanning electron microscopy (SEM) of Assulina species was performed as described in Heger et al. (2009). PCR and sequencing Polymerase and primers were added subsequently to the tubes containing the cells to perform the PCR reaction (15 lL containing 1.5 mm MgCl2, 10 nmol of each dNTP, 20 pmol of each primer and 1 U Taq DNA polymerase; GoTaq, Promega Madison, WI, USA). A first PCR was performed using Eucox1F and Euglycox1R, a pair of euglyphid-specific cytochrome c oxidase subunit I (COI) primers designed in a previous study (Heger et al., 2010b). The cycling profile consisted of a 5-min initial denaturation step (95); followed by 40 cycles of 95 C for 15 s, 40 C for 30 s, and 72 C for 90 s; and a final extension at 72 C for 10 min. Assulinaspecific primers were designed for a second PCR (Assucox 1F: 5¢-AAYATGAGRGCYAGRGG-3¢ and Assucox 1R: 5¢-CGTAATGAAARTGWCCYACC-3¢). A nested PCR protocol was applied using these last two primers for both Assulina muscorum and A. seminulum. The cycling profile was a 2-min initial denaturation step (95 C); followed by 40 cycles of 95 C for 15 s, 52 C for 30 s, and 72 C for 90 s; and a final extension at 72 C for 10 min. In the case of A. scandinavica, a combination of Assucox 1F and Euglycox 1R was used for the second step of the nested protocol; the cycling profile was the same, except for the annealing temperature (45 C). The PCR 4 products were purified using either a High Pure PCR Purification Kit (Roche, Basel, Switzerland) or a Wizard SV Gel and PCR Clean-Up System (Promega), and directly sequenced without a cloning step. Sequencing was carried out using a BigDye197 Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Carlsbad, CA, USA) and analysed with an ABI-3130xl DNA Sequencer (Applied Biosystems). COI sequences were deposited in GenBank with the accession numbers HM641248–HM641252. Sequences were manually aligned and compared, and sequence identity values were obtained using the software BioEdit 7.0.9.0 (Hall, 1999). The COI sequences were easily aligned, as no insertions or deletions were detected. Genetic diversity patterns of A. muscorum and A. seminulum were evaluated using the software DnaSP (Rozas et al., 2003). The following standard genetic diversity indices were calculated: number of segregating sites (S); number of haplotypes (h); haplotype diversity (Hd); average number of differences (P) and nucleotide diversity (p). Standard diversity indices are measures of the frequency of each haplotype in the sample (Nei & Jin, 1989). Haplotype diversity Hd is defined as the probability that two randomly chosen haplotypes are different in the sample. The mean number of pairwise differences (P) is defined as the mean number of differences between all pairs of haplotypes in a sample. Nucleotide diversity (p) is defined as the mean number of nucleotide differences per site between two sequences. In addition, in order to test if there is a correlation between the number of haplotypes observed and the number of sampling sites, we performed a Kendall test on the data (i.e. a nonparametric rank test) using the R software package (R Development Core Team, 2008). Review of present-day and palaeoecological data We reviewed 767 published present-day (surface samples) and palaeoecological samples from Europe and Alaska taken from the following publications (original sources for these data): Mitchell et al., 1999, 2001; Lamentowicz & Mitchell, 2005; Payne et al., 2006; Payne & Mitchell, 2007; Lamentowicz et al., 2008, 2010a,b. In addition, we analysed unpublished data from Karelia and Penza regions (kindly provided by Yuri Mazei, Department of Zoology and Ecology, Penza VG Belinsky State Pedagogical University). We used the present-day and palaeoecological data to evaluate: (1) frequency of presence, (2) relative dominance, (3) relative abundance, and (4) density of A. muscorum and A. seminulum. We used the present-day data to evaluate the ecology [optimum and tolerance for depth to water table (DWT), percentage moisture, and pH] of A. muscorum and A. seminulum in peatlands. We used the palaeoecological data to assess: (1) how rapidly Assulina species colonized peatlands during their developmental history, (2) which species appeared first out of A. muscorum and A. seminulum (A. scandinavica is too rarely reported to allow any meaningful interpretation), and (3) in cases where several species co-occurred, how many years separated their respective first records. RESULTS Molecular data Our PCR protocol allowed us to amplify sequences of c. 550 bp. Sequences from A. muscorum showed the existence of two haplotypes (AM1 and AM2), which differed by a single position, representing 0.2% of the sequence divergence. In total, 37 (95%) sequences belonged to haplotype AM1, and two (5%) to haplotype AM2. In the case of A. seminulum, we discovered three haplotypes (AS1, AS2 and AS3), which differed by 1–2 nucleotides (see Fig. 3), and two sites exhibiting single nucleotide polymorphism (SNP); this represents between 0.2 and 0.4% of the sequence divergence. The haplotypes were distributed in the following manner: 15 belonged to AS1 (50%), 13 to AS2 (43%) and 2 to AS3 (7%). For A. scandinavica, we found only one haplotype (i.e. all six sequences were identical). There was no obvious geographical pattern for the distribution of the various haplotypes encountered: AS1 was found in six sites located in four countries, and AS2 in eight sites in five countries. In addition, some sites hosted several haplotypes: for A. seminulum this was the case for Lörmoos (AS1 and AS2), Praz Rodet (AS1 and AS2), Moszne (AS1, AS2 and AS3) and Verkhozimskie (AS1 and AS2). For A. muscorum, AM1 and AM2 were found in Bory Tucholskie (Table 1, Figs 1 & 2). The diversity indices show that A. seminulum was genetically more diverse than 2nt AS2 Figure 3 Representation of the three haplotypes found in Assulina seminulum (AS1, 2 and 3) and the differences observed between sequences. The sampling sites at which they were found are also indicated. nt denotes a single nucleotide difference. Sampling sites Grindelwald (CH) Praz Rodet (CH) Sortel (CH) Lörmoos (CH) Pikasaare (EE) Moszne (PL) Verkhozimskie (RU) Ryggmossen (SE) AS3 1nt 1nt Sampling sites Moszne (PL) Sampling sites Lörmoos (CH) Praz Rodet (CH) Holmegaard (DK) Bory Tucholskie (PL) Moszne (PL) Verkhozimskie (RU) AS1 5 Table 2 Values obtained for the various genetic diversity estimators within Assulina muscorum and A. seminulum using DnaSP (Rozas et al., 2003), with the number of segregating sites (S), the number of haplotypes (h), haplotype diversity (Hd), the average number of differences (P) and nucleotide diversity (p). Species Individuals S h Hd  p Assulina seminulum Assulina muscorum 30 40 2 1 3 2 0.577 0.097 0.637 0.097 0.00115 0.00018 A. muscorum (Table 2). All observed SNPs concerned only silent mutations (i.e. they occurred in the third position of the codon and therefore did not change the amino acid sequence), except for the mutation that differentiated AS1 from AS2, which was responsible for a change between a valine and an alanine (two hydrophobic amino acids). The Kendall test showed that there was no significant correlation between the number of sampling sites and the number of haplotypes in the case of A. muscorum (P = 0.720, n = 40 for the global analysis; if we exclude the Lörmoos site from which we retrieved 18 sequences, P = 1.000). However, this relationship proved to be almost significant for A. seminulum (P = 0.057, n = 30). In contrast to the low level of within-taxon genetic diversity, the three species were clearly distinct from each other. The percentage of similarity of the nucleotide sequences (calculated by direct comparison of the sequences excluding the gaps and non-alignable parts if encountered) was 84% between A. seminulum and A. muscorum, 82% between A. seminulum and A. scandinavica, and 79% between A. muscorum and A. scandinavica. When translated into amino acid sequences, the percentages of similarity are, respectively, 95% between A. seminulum and A. muscorum, 88% between A. seminulum and A. scandinavica, and 87% between A. muscorum and A. scandinavica. Ecological data Assulina species are very common in surface moss samples as well as in subfossil samples from peatlands in Europe and Alaska. Indeed, A. muscorum was the most frequent testate amoeba species listed in a compilation of European and North American data, while A. seminulum ranked fourth (Gilbert & Mitchell, 2006). Of the 767 peat samples analysed in this study, at least one Assulina species was observed in 650 samples (84.7%). Furthermore, A. muscorum and A. seminulum co-occurred in 52.5% of samples, whereas A. muscorum was found alone in 30.2%. Assulina seminulum was found alone in only 2.0% of the samples. Furthermore, A. muscorum was more abundant than A. seminulum in 86.9% of the samples in which both species were observed, whereas A. seminulum was more abundant than A. muscorum in only 10.6% of these cases (Table 3). Assulina scandinavica was not found in these samples and is in fact only rarely recorded in ecological studies. Assulina muscorum was also systematically more abundant than A. seminulum in all samples for which abundance was Table 3 Summary statistics of presence and relative dominance of Assulina muscorum versus A. seminulum in surface and palaeoecological peat samples from Europe and Alaska (n = 767). See text for the list of corresponding references. Samples with: Total n % Assulina spp. A. muscorum only A. seminulum only A. muscorum and A. seminulum A. muscorum > A. seminulum A. seminulum > A. muscorum A. muscorum = A. seminulum 650 223 13 403 565 69 16 84.7 30.2 2.0 52.5* 86.9* 10.6* 2.5* *Percentage of samples containing both Assulina species. recorded. This was the case in terms of both relative abundance (percentage of total number of testate amoebae) and density (cells per gram of dried material; Table 4). When present, A. scandinavica was much less abundant than A. seminulum (data not shown). Present-day ecological studies also reveal that A. muscorum generally occurs in slightly drier and less acidic microsites than A. seminulum and has a broader tolerance with respect to the ecological gradients to which testate amoebae have been shown to respond most strongly: moisture content of mosses, watertable depth and pH. However, the differences in optima were not systematic, and differences in tolerance were mostly small: in some cases A. seminulum had an optimum in drier microsites and a slightly larger tolerance than A. muscorum (Table 5). Palaeoecological data We analysed 13 published palaeoecological studies of peat bogs (Table 3). In 38% of cases, A. muscorum and A. seminulum were observed for the first time in the same sample. They are thus recorded in these cases as having arrived ‘at the same time’, although it is clearly impossible to ascertain this, given the relatively low temporal resolution of the records, especially in the basal part of the peat deposits. In 62% of cases, A. muscorum appeared before A. seminulum, and we did not find a single study in which A. seminulum appeared first. Based on the age–depth chronologies of the studies, we estimated the time lag between the appearance of A. muscorum and A. seminulum to be between 30–50 years and 7500 years. In the latter case, unfavourable ecological conditions over much of the developmental history of the peatland are suspected to be the main explanation for the absence of A. seminulum (Table 6). DISCUSSION The genetic diversity observed within the three species examined in this study can be considered as very low. The 6 Table 4 Relative abundance and density of Assulina muscorum and A. seminulum in surveys of present-day communities. Location A. muscorum Toolik Lake (Alaska, USA), poor fen, control plots of nutrient enrichment experiment Alps of northern Italy, Hylocomium splendens along elevational transects Swiss Jura Mountains, Le Cachot bog North-west Poland, peatlands Swiss Jura Mountains, Le Cachot bog Swiss Jura Mountains, Chaux d’Abel bog Swiss Jura Mountains, Chaux d’Abel bog Swiss Eastern Alps, Subalpine peatlands European & North American peatlands 9337 (1634) 1118 (1118) 4299 (608) 47 (33) 12.9% 19.1% 11.8% 3146 18.7% 14.4% 8.5% (2.3) (6.1) (2.4) (954) (4.6) (2.0) (0.5) Reference A. seminulum 1.2% 2.2% 3.8% 459 1.6% 2.6% 2.3% Mitchell, 2004 Mitchell et al., 2004 (0.5) (1.6) (1.3) (231) (1.0) (0.7) (0.3) Mitchell & Gilbert, 2004 Lamentowicz & Mitchell, 2005 Kishaba & Mitchell, 2005 Laggoun-Défarge et al., 2008 Laggoun-Défarge et al., 2008 Lamentowicz et al., 2010b Gilbert & Mitchell, 2006 Data are either densities (expressed as individuals per gram dry weight of moss) or relative abundances (expressed as percentage of total testate amoeba community; indicated by %); values in brackets are standard errors of the means. Table 5 Optimum and tolerance values from ecological studies of Assulina muscorum and A. seminulum for percentage moisture, depth to water table (DWT) and pH. A. muscorum A. seminulum Location Variable Optimum Tolerance Optimum Tolerance Reference Jura (Switzerland and France) Europe Canada (NE Ontario) Canada (Lake Superior) Jura (Switzerland and France) Finland Europe Canada (Lake Superior) Canada (Newfoundland) Canada (NW Ontario-Minnesota) Jura (Switzerland and France) Finland Canada (Lake Superior) Percentage moisture Percentage moisture Percentage moisture Percentage moisture DWT (cm) DWT (cm) DWT (cm) DWT (cm) DWT (cm) DWT (cm) pH pH pH 34 c. 89 c. 68 c. 72 59 16.6 c. 18 c. 15 17.44 42.53 4.3 4 4.6 19 + ()) + 49 11.6 + + 11 25.59 0.3 0.5 + 38 c. 90.5 c. 70 c. 71 38 10.9 c. 18.5 c. 17 13.95 39.29 4 4 4.2 18 ) (+) ) 33 7.57 ) ) 9.68 17.84 0.1 0.48 ) Mitchell et al., 1999 Charman et al., 2007 Charman & Warner, 1992 Booth, 2001 Mitchell et al., 1999 Tolonen et al.,1992 Charman et al., 2007 Booth, 2001 Charman & Warner, 1997 Charman & Warner, 1997 Mitchell et al., 1999 Tolonen et al., 1992 Booth, 2001 Optima are calculated by the weighted average of species’ relative abundance (see Mitchell et al., 1999 for the formula). Signs ‘+’ and ‘)’ indicate which species had the highest tolerance; when differences are minimal, these signs are in brackets. 0.2 and 0.4% of intraspecific COI sequence divergence observed for Assulina species is far below the threshold of 2–3% generally accepted for discriminating species of Metazoa (Hebert et al., 2003). Our values of intraspecific genetic divergence are comparable to those observed in scleractinian corals, which are well known for the conservation of their mitochondrial genes (Shearer et al., 2002). However, in contrast to these cnidarians, interspecific variation is important in the genus Assulina, reaching about 20% between the three species that constitute this genus. COI is indeed quite variable in euglyphids: a recent study of another euglyphid genus, Cyphoderia, showed that this gene was three times more variable than small subunit ribosomal RNA (SSU rRNA; Heger et al., 2010b). Within the genus Assulina, COI might be relatively even more variable than in Cyphoderia: we observe as much as 97.2% sequence identity between A. muscorum (AJ418791) and A. seminulum (EF456749) at the SSU rRNA gene level and only 84% at the COI level (in other words, COI appears more than five times more variable than SSU rRNA). In addition, it appears that some haplotypes are 100% identical, even though they originated from sites as far from each other as Praz Rodet (Switzerland), Pikassaare (Estonia), Ryggmossen (Central Sweden) and Verkhozimskie (Russia), as in the case of AS2 for A. seminulum, or Glenn Dee (Scotland) and Chaux d’Abel (Switzerland) for A. scandinavica. These locations represent separations of thousands of kilometres, mostly over apparently unsuitable habitat (i.e. not Sphagnum peatlands), especially for A. seminulum and A. scandinavica. Most interestingly, several cells collected from the same site belonged to different haplotypes. Assuming that COI is sensitive enough to reveal the genetic structure of Assulina populations, our data strongly suggest that cells can move Table 6 Palaeoecological records of Assulina muscorum and A. seminulum: position of the first record of A. muscorum from the base of the record, number of samples between the first records of A. muscorum and A. seminulum, and estimated corresponding time lag. A. muscorum present in Xth oldest sample Lag between first records of A. muscorum and A. seminulum (number of samples) Lag between first records of A. muscorum and A. seminulum (years) Country Location Context Belgium USA Scotland Scotland Ipenrooi, Valley of River Mark Upper Michigan Traligill Basin, blanket peat – UAM1 Traligill Basin, peat above cave entrance – UAM4 Männikjärve bog Praz-Rodet Tuchola Temple Hill Moss Tourbière du lac Malbaie – Core Mal-1 Tourbière du lac Malbaie – Core Mal-2 Tourbière du lac Malbaie – Core Mal-3 Glen West, County Fermanagh Eweburn Bog Holocene sequence Holocene sequence Last c. 2700 years Last c. 2200 years 2 1 1 8 4 9 3 0 380–480 300 30–50 0 Last 60 years Holocene sequence Holocene sequence Last c. 7000 years Holocene sequence Holocene sequence Holocene sequence Last c. 2500 years Holocene sequence 1 6 2 1 2 26 9 2 2 0 2 0 1 26 0 10 25 0 0 100–150 0 100–200 7500* 0 4500* 2300* 0 Estonia Switzerland Poland Scotland Québec, Canada Québec, Canada Québec, Canada Northern Ireland New Zealand Reference Beyens, 1985 Booth et al., 2004 Charman et al., 2001 Charman et al., 2001 Charman et al., 2004 Mitchell et al., 2001 Lamentowicz et al., 2008 Langdon et al., 2003 Lavoie & Richard, 2000 Lavoie & Richard, 2000 Lavoie & Richard, 2000 Swindles et al., 2007 Wilmshurst et al., 2003 *Habitat not suitable for A. seminulum over much of this period (probably too minerotrophic). 7 8 quite rapidly across Europe. Another explanation would be recent selective sweeps (Hurst & Jiggins, 2005) that might have occurred in all three Assulina species and eliminated or strongly reduced genetic diversity. Several other protists also present low genetic variability of the COI gene across large geographical distances. In ciliates, it has been observed that isolates of Tetrahymena thermophila from relatively distant geographic origins within North America [Woods Hole (MA) and Allegheny National Forest (PA)] have COI sequences that diverge by less than 1% (Lynn & Strüder-Kypke, 2006). Strains of Paramecium multimicronucleatum originating from Italy, Germany and Australia presented identical COI sequences (Barth et al., 2006), suggesting that these species frequently cross major biogeographical barriers. Carchesium polypinum, another ciliate, did not show patterns of COI genetic diversity associated with geographical distance at the Grand River basin scale, in North America (Gentekaki & Lynn, 2009). The marine prasinophyte Micromonas pusilla also presented cosmopolitan COI haplotypes (Šlapeta et al., 2006), and identical COI sequences of the freshwater diatom Sellaphora capitata have been found in Belgium and Scotland (Evans et al., 2007). Thus, the reduced intraspecific genetic diversity found in other protists is consistent with our observations in Assulina species, whose ability to travel rapidly across large distances hence appears unsurprising. Such results can be seen as corroborating the ‘ubiquity theory’, providing evidence for the more or less instantaneous dispersal of microbial propagules across the globe. The conclusions drawn from all these aquatic species forming small propagules can thus be extended to the larger terrestrial testate amoebae. However, it is possible that genetic markers that are even more sensitive than COI, such as microsatellites, could reveal more about the biogeography of Assulina species, as shown for freshwater diatoms (Evans et al., 2009). In our study, all diversity estimators revealed a higher genetic diversity in A. seminulum than in A. muscorum (Table 2). Assulina scandinavica, the largest, rarest species and the only one whose test length is around the 100-lm threshold proposed for the cosmopolitan distribution of testate amoebae (Wilkinson, 2001), did not present any SNPs in the six sequences we analysed. This might suggest a rapid migration capacity in this species also. However, we were able to retrieve only a limited number of cells, and it would be necessary to sample more sites and retrieve more individuals to assess the genetic diversity in this species at the COI level. This is consistent with our hypothesis that a rarer, larger, more specialized species that forms smaller populations is expected to spread more slowly than a more frequent, smaller, more tolerant species that forms larger populations. These results are partly corroborated by our observations of ecological and palaeoecological data, which show that: (1) A. muscorum is more frequent than A. seminulum in a large data set from European and North American peatlands, and, where it co-occurs with A. seminulum, it is in most cases the more abundant species (see Tables 3 & 4); and (2) A. seminulum became established at the same time as or later than A. muscorum during the developmental history of the studied peatlands (see Table 6). The higher abundance of A. muscorum could be attributed to its smaller size. Within closely related taxa, smaller organisms can be expected to reach higher densities than larger ones. The higher frequency of A. muscorum could, on the other hand, also be explained by its broader ecological tolerance and preference for more common microenvironmental conditions (e.g. drier microsites). Although there is indeed a trend for an optimum (in relative abundance) in drier, less acidic microsites and for a higher tolerance at least with respect to the two most commonly measured variables, pH and water table depth, these differences in ecological preferences were not as clear as for population size and abundance (Table 5). Furthermore, when other ecological variables are compared, the pattern is even more pronounced (e.g. see Tolonen et al., 1992). In some cases (Tourbière du Lac Malbaie cores 1 and 3, Lavoie & Richard, 2000; and Glenn West, Swindles et al., 2007), changes in the ecology of these peatlands could be held responsible for the absence of A. seminulum for a long time. Indeed, other indicators have shown that these sites were relatively minerotrophic. This may have prevented the establishment of this species. However, as existing ecological data did not reveal any marked difference in ecological optima between the two species it is not clear if these palaeoecological patterns can be held responsible for the absence of A. seminulum. In all other records, a lag of between 0 and 480 years has been observed between the appearance of A. muscorum and of A. seminulum. We can thus reasonably consider that A. muscorum spreads more rapidly than A. seminulum (and probably even more rapidly than A. scandinavica, although there are not enough data for this species to draw any firm conclusion). These data suggest that new suitable environments can be colonized within at most a couple of hundred years, depending probably on factors such as distance to the closest population nucleus, competition of migrants with locally established populations (although this effect seems to be marginal, if it exists at all; Wanner & Xylander, 2005; Wanner et al., 2008), direction of dominant winds, migratory patterns of birds, and, of course, ‘chance events’, as attested by the critical importance of the tail-end of propagule dispersal distribution functions for explaining the rapid post-glacial colonization patterns of trees (Clark, 1998). The available data may not allow us to draw definitive conclusions about the causes of the observed patterns. However, (1) the clear differences in size (and hence even more in biovolume and biomass) and abundance (either relative or density) observed between A muscorum and A. seminulum; and (2) the comparatively smaller differences in optima and tolerance for the main ecological gradients suggest that size (and related population size) matters more than the (relatively small and partly inconsistent) differences in ecological requirements in determining the dispersal potential of Assulina species. 9 CONCLUSIONS Molecular and palaeoecological data suggest that the terrestrial euglyphid testate amoebae of the genus Assulina are capable of rapid colonization of new habitats. However, both lines of evidence also suggest that the colonization pace was slower in A. seminulum, a larger, less abundant, and ecologically slightly more specialized species than A. muscorum. Propagule size and to a lesser extent biotic constraints of species therefore influence the dispersal capability of Assulina species. Further studies combining molecular and ecological data are required to determine the degree to which this may also hold true for testate amoebae in general and for other microbial species. ACKNOWLEDGEMENTS We thank Mariusz Lamentowicz, Mariusz Ga˛bka and Yuri Mazei for providing Sphagnum samples from Russia and Poland. Yuri Mazei also provided unpublished data for the analysis of modern testate amoeba communities. The authors also wish to thank Humphrey Smith and an anonymous referee for fruitful comments on the manuscript. E.L. was supported by an Ambizione fellowship from the Swiss National Science Foundation (PZ00P2_122042); T.J.H. by Swiss NSF projects no. 205321-109709/1 and 205321-109709/2; R.S. by a fellowship from CONICYT; and E.A.D.M. by Swiss NSF projects no. 205321-109709/1 and 205321-109709/2. The authors also wish to thank José Fahrni for technical support and helpful discussions. REFERENCES Barth, D., Krenek, S., Fokin, S.I. & Berendonk, T.U. (2006) Intraspecific genetic variation in Paramecium revealed by mitochondrial cytochrome c oxidase I sequences. Journal of Eukaryotic Microbiology, 53, 20–25. Beyens, L. (1985) On the subboreal climate of the Belgian Campine as deduced from diatom and testate amoeba analyses. Review of Palaeobotany and Palynology, 46, 9–31. Booth, R.K. (2001) Ecology of testate amoebae (Protozoa) in two Lake Superior coastal wetlands: implications for paleoecology and environmental monitoring. Wetlands, 21, 564–576. Booth, R.K., Jackson, S.T. & Gray, C.E.D. (2004) Paleoecology and high-resolution paleohydrology of a kettle peatland in Upper Michigan. Quaternary Research, 61, 1–13. Caron, D.A. (2009) Past President’s address: Protistan biogeography: why all the fuss? Journal of Eukaryotic Microbiology, 56, 105–112. Charalambidou, I. & Santamarı́a, L. (2002) Waterbirds as endozoochorous dispersers of aquatic organisms: a review of experimental evidence. Acta Oecologica, 23, 165–176. Charman, D.J. & Warner, B.G. (1992) Relationship between testate amoebae (Protozoa, Rhizopoda) and microenvironmental parameters on a forested peatland in northeastern Ontario. Canadian Journal of Zoology–Revue Canadienne de Zoologie, 70, 2474–2482. Charman, D.J. & Warner, B.G. (1997) The ecology of testate amoebae (Protozoa: Rhizopoda) in oceanic peatlands in Newfoundland, Canada: modelling hydrological relationships for palaeoenvironmental reconstruction. Écoscience, 4, 555–562. Charman, D.J., Caseldine, C., Baker, A., Gearey, B., Hatton, J. & Proctor, C. (2001) Paleohydrological records from peat profiles and speleothems in Sutherland, northwest Scotland. Quaternary Research, 55, 223–234. Charman, D.J., Brown, A.D., Hendon, D. & Karofeld, E. (2004) Testing the relationship between Holocene peatland palaeoclimate reconstructions and instrumental data at two European sites. Quaternary Science Reviews, 23, 137–143. Charman, D.J., Blundell, A. & ACCROTELM Members (2007) A new European testate amoebae transfer function for palaeohydrological reconstruction on ombrotrophic peatlands. Journal of Quaternary Science, 22, 209–221. Clark, J.S. (1998) Why trees migrate so fast: confronting theory with dispersal biology and the paleorecord. The American Naturalist, 152, 204–224. Corliss, J.O. & Esser, S.C. (1974) Comments on the role of the cyst in the life-cycle and survival of free-living protozoa. Transactions of the American Microscopical Society, 93, 578– 593. Evans, K.M., Wortley, A.H. & Mann, D.G. (2007) An assessment of potential diatom ‘‘barcode’’ genes (cox1, rbcL, 18S and ITS rDNA) and their effectiveness in determining relationships in Sellaphora (Bacillariophyta). Protist, 158, 349–364. Evans, K.M., Chepurnov, V.A., Sluiman, H.J., Thomas, S.J., Spears, B.M. & Mann, D.G. (2009) Highly differentiated populations of the freshwater diatom Sellaphora capitata suggest limited dispersal and opportunities for allopatric speciation. Protist, 160, 386–396. Finlay, B.J. & Clarke, K.J. (1999) Ubiquitous dispersal of microbial species. Nature, 400, 828. Finlay, B.J. & Fenchel, T. (2004) Cosmopolitan metapopulations of free-living microbial eukaryotes. Protist, 155, 237–244. Finlay, B.J., Esteban, G.F., Olmo, J.L. & Tyler, P.A. (1999) Global distribution of free-living microbial species. Ecography, 22, 138–144. Finlay, B.J., Esteban, G.F., Clarke, K.J. & Olmo, J.L. (2001) Biodiversity of terrestrial protozoa appears homogenous across local and global spatial scales. Protist, 152, 355–366. Foissner, W. (1987) Soil protozoa: fundamental problems, ecological significance, adaptation in ciliates and testaceans, bioindicators, and guide to the literature. Progress in Protozoology, 2, 69–212. Foissner, W. (2006) Biogeography and dispersal of microorganisms: a review emphasising protists. Acta Protozoologica, 45, 111–136. Foissner, W., Chao, A. & Katz, L.A. (2008) Diversity and geographic distribution of ciliates (Protista: Ciliophora). Biodiversity and Conservation, 17, 329–343. 10 Gentekaki, E. & Lynn, D.H. (2009) High-level genetic diversity but no population structure inferred from nuclear and mitochondrial markers of the peritrichous ciliate Carchesium polypinum in the Grand River basin (North America). Applied and Environmental Microbiology, 75, 3187–3195. Gilbert, D. & Mitchell, E.A.D. (2006) Microbial diversity in Sphagnum peatlands. Peatlands: basin evolution and depository of records on global environmental and climatic changes (ed. by I.P. Martini, A. Martı́nez Cortizas and W. Chesworth), pp. 287–317. Elsevier, Amsterdam. Hall, T.A. (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/ NT. Nucleic Acids Symposium Series, 41, 95–98. Hebert, P.D.N., Ratnasingham, S. & de Waard, J.R. (2003) Barcoding animal life: cytochrome c oxidase subunit I divergences among closely related species. Proceedings of the Royal Society B: Biological Sciences, 270, 96–99. Heger, T.J., Mitchell, E.A.D., Ledeganck, P., Vincke, S., Van de Vijver, B. & Beyens, L. (2009) The curse of taxonomic uncertainty in biogeographical studies of free-living terrestrial protists: a case study of testate amoebae from Amsterdam Island. Journal of Biogeography, 36, 1551–1560. Heger, T.J., Mitchell, E.A.D., Todorov, M., Golemansky, V., Lara, E., Leander, B.S. & Pawlowski, J. (2010a) Molecular phylogeny of euglyphid testate amoebae (Cercozoa: Euglyphida) suggests transitions between marine supralittoral and freshwater/terrestrial environments are infrequent. Molecular Phylogenetics and Evolution, 55, 113–122. Heger, T.J., Pawlowski, J., Lara, E., Leander, B.S., Todorov, M., Golemansky, V. & Mitchell, E.A.D. (2010b) Comparing COI and SSU rDNA barcodes for assessing the diversity and phylogenetic relationships of cyphoderiid testate amoebae (Rhizaria: Euglyphida). Protist (available online, doi:10.1016/j.protis.2010.05.002). Hurst, G.D.D. & Jiggins, F.M. (2005) Problems with mitochondrial DNA as a marker in population, phylogeographic and phylogenetic studies: the effects of inherited symbionts. Proceedings of the Royal Society B: Biological Sciences, 272, 1525–1534. Kishaba, K. & Mitchell, E.A.D. (2005) Changes in testate amoebae (Protists) communities in a small raised bog. A 40-year study. Acta Protozoologica, 44, 1–12. Laggoun-Défarge, F., Mitchell, E.A.D., Gilbert, D., Disnar, J.-R., Comont, L., Warner, B. & Buttler, A. (2008) Cutover peatland regeneration assessment using organic matter and microbial indicators (bacteria and testate amoebae). Journal of Applied Ecology, 45, 716–727. Lamentowicz, M. & Mitchell, E.A.D. (2005) The ecology of testate amoebae (Protists) in Sphagnum in north-western Poland in relation to peatland ecology. Microbial Ecology, 50, 48–63. Lamentowicz, M., Obremska, M. & Mitchell, E.A.D. (2008) Autogenic succession, land-use change, and climatic influences on the Holocene development of a kettle hole mire in Northern Poland. Review of Palaeobotany and Palynology, 151, 21–40. Lamentowicz, M., Lamentowicz, Ł., van der Knaap, W.O., Ga˛bka, M. & Mitchell, E.A.D. (2010a) Contrasting species– environment relationships in communities of testate amoebae, bryophytes and vascular plants along the fen-bog gradient. Microbial Ecology, 59, 499–510. Lamentowicz, M., van der Knaap, W.O., Lamentowicz, Ł., Van Leeuwen, J.F.N., Mitchell, E.A.D., Goslar, T. & Kamenik, C. (2010b) A near-annual palaeohydrological study based on testate amoebae from a sub-alpine mire: surface wetness and the role of climate during the instrumental period. Journal of Quaternary Science, 25, 190–202. Langdon, P.G., Barber, K.E. & Hughes, P.D.M. (2003) A 7500year peat-based palaeoclimatic reconstruction and evidence for an 1100-year cyclicity in bog surface wetness from Temple Hill moss, Pentland Hills, southeast Scotland. Quaternary Science Reviews, 22, 259–274. Lara, E., Heger, T.J., Mitchell, E.A.D. & Ekelund, F. (2007) SSU rRNA reveals a sequential increase in shell complexity among the euglyphid testate amoebae (Rhizaria: Euglyphida). Protist, 158, 229–237. Lara, E., Heger, T.J., Ekelund, F., Lamentowicz, M. & Mitchell, E.A.D. (2008) Ribosomal RNA genes challenge the monophyly of the Hyalospheniidae (Amoebozoa: Arcellinida). Protist, 159, 165–176. Lavoie, M. & Richard, P.J.H. (2000) The paleoecology of a peatland in the Laurentian Highlands (Quebec): identifying the role of climate on peat accumulation. Géographie Physique et Quaternaire, 54, 169–185. Lynn, D.H. & Strüder-Kypke, M.C. (2006) Species of Tetrahymena identical by small subunit rRNA gene sequences are discriminated by mitochondrial cytochrome c oxidase I gene sequences. Journal of Eukaryotic Microbiology, 53, 385–387. Mitchell, E.A.D. (2004) Response of testate amoebae (Protozoa) to N and P fertilization in an Arctic wet sedge tundra. Arctic, Antarctic, and Alpine Research, 36, 77–82. Mitchell, E.A.D. & Gilbert, D. (2004) Vertical micro-distribution and response to nitrogen deposition of testate amoebae in Sphagnum. Journal of Eukaryotic Microbiology, 51, 485–495. Mitchell, E.A.D. & Meisterfeld, R. (2005) Taxonomic confusion blurs the debate on cosmopolitanism versus local endemism of free-living protists. Protist, 156, 263–267. Mitchell, E.A.D., Buttler, A.J., Warner, B.G. & Gobat, J.M. (1999) Ecology of testate amoebae (Protozoa: Rhizopoda) in Sphagnum peatlands in the Jura Mountains, Switzerland and France. Écoscience, 6, 565–576. Mitchell, E.A.D., van der Knaap, W.O., van Leeuwen, J.F.N., Buttler, A., Warner, B.G. & Gobat, J.M. (2001) The palaeoecological history of the Praz-Rodet bog (Swiss Jura) based on pollen, plant macrofossils and testate amoebae (Protozoa). The Holocene, 11, 65–80. Mitchell, E.A.D., Bragazza, L. & Gerdol, R. (2004) Testate amoebae (Protista) communities in Hylocomium splendens (Hedw.) B.S.G. (Bryophyta): relationships with altitude, and moss elemental chemistry. Protist, 155, 423–436. 11 Nei, M. & Jin, L. (1989) Variances of the average numbers of nucleotide substitutions within and between populations. Molecular Biology and Evolution, 6, 290–300. Payne, R. & Mitchell, E.A.D. (2007) Testate amoebae–environment relationships and a hydrological transfer function from mires in the Central Rhodope Mountains, Greece. Protist, 158, 159–171. Payne, R.J., Kishaba, K., Blackford, J.J. & Mitchell, E.A.D. (2006) Ecology of testate amoebae (Protista) in SouthCentral Alaska peatlands: building transfer-function models for palaeoenvironmental studies. The Holocene, 16, 403–414. R Development Core Team (2008) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Revill, D.L., Stewart, K.W. & Schlicht, H. (1967) Passive dispersal of viable algae and protozoa by certain craneflies and midges. Ecology, 48, 1023–1027. Rozas, J., Sánchez-Del Barrio, J.C., Messeguer, X. & Rozas, R. (2003) DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics, 19, 2496–2497. Schlichting, H.E., Jr & Sides, S.L. (1969) The passive transport of aquatic microorganisms by selected Hemiptera. Journal of Ecology, 57, 759–764. Shearer, T.L., Van Oppen, M.J.H., Romano, S.L. & Worheide, G. (2002) Slow mitochondrial DNA sequence evolution in the Anthozoa (Cnidaria). Molecular Ecology, 11, 2475–2487. Šlapeta, J., López-Garcı́a, P. & Moreira, D. (2006) Global dispersal and ancient cryptic species in the smallest marine eukaryotes. Molecular Biology and Evolution, 23, 23–29. Smith, H.G. & Wilkinson, D.M. (2007) Not all free-living microorganisms have cosmopolitan distributions – the case of Nebela (Apodera) vas Certes (Protozoa, Amoebozoa, Arcellinida). Journal of Biogeography, 34, 1822–1831. Smith, H.G., Bobrov, A. & Lara, E. (2008) Diversity and biogeography of testate amoebae. Biodiversity and Conservation, 17, 329–343. Swindles, G.T., Plunkett, G. & Roe, H.M. (2007) A multiproxy climate record from a raised bog in county Fermanagh, Northern Ireland: a critical examination of the link between bog surface wetness and solar variability. Journal of Quaternary Science, 22, 667–679. Tolonen, K., Warner, B.G. & Vasander, H. (1992) Ecology of testaceans (Protozoa, Rhizopoda) in mires in southern Finland. 1. Autecology. Archiv für Protistenkunde, 142, 119–138. Van de Vijver, B. & Mataloni, G. (2008) New and interesting species in the genus Luticola D.G. Mann (Bacillariophyta) from Deception Island (South Shetland Islands). Phycologia, 27, 451–467. Vanormelingen, P., Verleyen, E. & Vyverman, W. (2008) The diversity and distribution of diatoms: from cosmopolitanism to narrow endemism. Biodiversity and Conservation, 17, 393–405. Wanner, M. & Xylander, W.E.R. (2005) Biodiversity development of terrestrial testate amoebae: is there any succession at all? Biology and Fertility of Soils, 41, 428–438. Wanner, M., Elmer, M., Kazda, M. & Xylander, W.E.R. (2008) Community assembly of terrestrial testate amoebae: how is the very first beginning characterized? Microbial Ecology, 56, 43–54. Wilkinson, D.M. (2001) What is the upper size limit for cosmopolitan distribution in free-living microorganisms? Journal of Biogeography, 28, 285–291. Wilkinson, D.M. (2010) Have we underestimated the importance of humans in the biogeography of free-living terrestrial microorganisms? Journal of Biogeography, 37, 393–397. Wilmshurst, J.M., Wiser, S.K. & Charman, D.J. (2003) Reconstructing Holocene water tables in New Zealand using testate amoebae: differential preservation of tests and implications for the use of transfer functions. The Holocene, 13, 61–72. Yang, J., Smith, H.G., Sherratt, T.N. & Wilkinson, D.M. (2010) Is there a size limit for cosmopolitan distribution in free-living microorganisms? A biogeographical analysis of testate amoebae from polar areas. Microbial Ecology, 59, 635–645. BIOSKETCH Enrique Lara is a postdoctoral fellow at the Swiss National Science Foundation at the University of Neuchâtel, Switzerland. His research focuses on the diversity, ecology and evolution of testate amoebae and other eukaryotic microorganisms. Author contributions: E.L. conceived the idea; E.L. and T.J.H. performed the experiments; R.S. analysed the genetic data and E.A.D.M. analysed the ecological and palaeoecological data; E.L., T.J.H. and E.A.D.M. wrote the manuscript. Editor: John Lambshead