Intra-population genetic diversity and its effects on outlining genetic diversity of ciliate populations: Using Paramecium multimicronucleatum as an example

Available online at www.sciencedirect.com

ScienceDirect European Journal of Protistology 67 (2019) 142–150

Intra-population genetic diversity and its effects on outlining genetic diversity of ciliate populations: Using Paramecium multimicronucleatum as an example Xuefen Lua,1 , Eleni Gentekakib,1 , Yiwei Xua,1 , Lijuan Huanga , Yunyi Lia , Xiaotong Lua , Yan Zhaoc , Xiaofeng Lina , Zhenzhen Yia,∗ a Laboratory of Protozoology, Guangzhou Key Laboratory of Subtropical Biodiversity and Biomonitoring, School of Life Science, South China Normal University, Guangzhou 510631, China b School of Science, Mae Fah Luang University, Chiang Rai 57100, Thailand c Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, China

Received 21 July 2018; received in revised form 16 December 2018; accepted 21 December 2018 Available online 27 December 2018

Abstract Questions regarding ciliate distribution (endemism vs. cosmopolitanism) and degree of genetic diversity (high vs. low) remain unsettled, even when the same organism is under investigation. Presence of genes with high copy number and amplification of non-dominant haplotypes might account for the observed discordance in these studies. Herein, we used direct PCR and cloning sequencing to examine intra-population sequence diversity and its effect on assessments of phylogeography of Paramecium multimicronucleatum. Totally, 381 ITS1-5.8S rDNA-ITS2-28S rDNA and 304 mitochondrial cytochrome oxidase subunit I (COI) gene sequences were generated for 18 populations of P. multimicronucleatum. The following results were obtained: (1) Direct sequencing of PCR products captured the dominant ITS and LSU haplotypes, indicating that it is an appropriate strategy for constructing phylogeography of large-scale spatial populations. (2) Deep cloning was deemed more appropriate for the COI gene for population level studies, as direct sequencing could not easily capture the dominant haplotypes. (3) No endemic populations of P. multinucleatum were noted, indicating origin from a single founder population. (4) Nuclear genetic diversity within temporal populations was high, but only the dominant haplotypes seemed to be passed on to subsequent generations. © 2018 Elsevier GmbH. All rights reserved.

Keywords: Eukaryotic microbes; Genetic polymorphisms; Geography; Sequencing strategy; Spatial populations; Temporal populations

Introduction In recent years, discussions on genetic diversity of eukaryotic microbes and geographic distribution of this diversity have resurfaced (e.g. Finlay and Fenchel 1999; Foissner

∗ Corresponding

author. E-mail address: [email protected] (Z. Yi). 1 These authors contributed equally to this work. https://doi.org/10.1016/j.ejop.2018.12.005 0932-4739/© 2018 Elsevier GmbH. All rights reserved.

1999; Katz et al. 2005; Lachance 2004; Smith and Wilkinson 2007; Weisse 2007; Zufall et al. 2013). The discussion has mainly focused on the geographic distribution of microbes, their genetic diversity, and whether there is gene flow or not. In that vein, some researchers advocate wide distribution of microbes and unrestricted gene flow (Finlay 1996), while others propose a large number of geo-specific species and limited gene flow (Foissner 1999). Ciliated protozoa have been used to address this question due to several favorable characteristics including large

X. Lu et al. / European Journal of Protistology 67 (2019) 142–150

size, abundance and high reproduction rate (Grolière et al. 1990). A commonly used method to do this is by assessing intraspecific genetic diversity and/or distribution of a particular species. For instance, endemic distribution was revealed in Tetrahymena thermophila (Zufall et al. 2013), and Stentor coeruleus (Kusch 1998), while Paramecium multimicronucleatum and P. caudatum had cosmopolitan distribution (Barth et al. 2006). Additionally, Katz et al. (2005) found both high levels of diversity and evidence of high gene flow in oligotrich ciliates. In some cases, conflicting results have been obtained for populations of the same species sampled at different geographic areas. Distinct population structure was detected in Chinese populations of Carchesium polypinum, while no such structure was evident in Canadian populations of the same species (Gentekaki and Lynn 2009; Miao et al. 2004). Currently, discussions remain inconclusive, due to the limited number of studies focusing on the topic and the sometimes-conflicting results even within these few studies (e.g. Barth et al. 2008; Gentekaki and Lynn 2009; Katz et al. 2005). Both nuclear and mitochondrial gene markers have been used to examine molecular diversity and its distribution in ciliates. These studies have revealed a high degree of genetic diversity among temporal and/or fine-scale spatial populations (Barth et al. 2008; Gentekaki and Lynn 2009; Miao et al. 2004; Zhao et al. 2013). Notably, in some cases, the level of intra-individual or intra-population genetic polymorphism was greater than that among populations. Moreover, genetic diversity among fine-scale populations was higher than that among large-scale populations (Miao et al. 2004; Tarcz et al. 2012; Zhao et al. 2013). These results have led to variable interpretations regarding gene flow and founder populations in ciliate species. Recent studies have revealed that ciliates possess extremely high copy numbers of rDNA within a single cell and that these copies might not evolve in concert (Gong et al. 2013; Wang et al. 2017). Moreover, numerous genetic polymorphisms have also been identified in the mitochondrial COI, suggesting multiple copies of this gene within a single ciliate cell (Zhao et al. 2013). The level of genetic heterogeneity within populations varies greatly depending on the species (Gong et al. 2013; Wang et al. 2017; Zhao et al. 2013). Though the effect of multiple gene copies on estimates of genetic diversity has been examined in ciliates (e.g. Gong et al. 2013; Zhao et al. 2013), its impact on their phylogeography remains largely unexplored. At the same time, there has been no systematic effort of the degree to which sequencing strategy, namely cloning or direct sequencing, provides similar or disparate results in such studies. Herein, we used Paramecium multimicronucleatum, a cosmopolitan and broadly distributed ciliate to address these questions. The intraspecific genetic diversity of this species has been previously examined using nuclear rDNA and mitochondrial COI gene markers (Barth et al. 2006; Catania et al. 2009; Tarcz et al. 2012). We collected P. multimicronucleatum from 10 locations around Guangzhou, China, at different

143

time points. We used both direct sequencing of PCR products and cloning to obtain internal transcribed spacer 1(ITS1)-5.8 small ribosomal (5.8S) rDNA-ITS2-5 LSU rDNA (ITS15.8S rDNA-ITS2-LSU rDNA) and COI sequences, in order to: (1) examine genetic polymorphisms of these markers within populations of P. multimicronucleatum, (2) determine whether the two sequencing strategies produced similar results, and (3) examine the effect of detected polymorphisms on the temporal and spatial genetic diversity among P. multimicronucleatum populations.

Material and Methods Sample collection information and species identification Samples were collected from bodies of freshwater in Guangzhou, China, at various time points (Table 1). In total, we collected 18 Paramecium multimicronucleatum populations distributed over 10 sampling sites. Herein, spatial population is used to define individuals collected from the same locality and at the same time, while temporal population is defined as individuals collected from the same location at different time points. We collected two temporal populations from each of locations b, c, e, i and k, and four from location d (Table 1). Specimens were picked under a dissecting microscope. Cells were cultured separately at room temperature in freshwater and a rice grain was used to enrich bacterial growth for feeding (Yi et al. 2016). Identification of cells was according to morphological features using both in vivo observations and staining methods including Protargol impregnation and Methyl Green-Pyronin. Moreover, sequences of SSU rDNA were used to further confirm species identification.

DNA extraction, PCR amplification and sequencing One to three pure cultures were established from each population. Then, at least 15 cells from each pure culture were isolated, washed three times in sterile water, and subsequently transferred to 1.5 ml microtubes with a minimum volume of water. Genomic DNA was extracted from multiple clean cells derived from one population using the Qiagen Dneasy Blood & Tissue Kit (QIAGEN, China) according to manufacturer’s specifications with the following modification: only 1/4 of the suggested volume was used for each step (Yi et al. 2016). Primers 5.8s-F (5 -GTAGGTGAACCTGCGGAAGGATC ATTA-3 ) and 28s-R (5 -CATTCGGCAGGTGAGTTG TTACACACTCC-3 ) (Gong et al. 2007) were used to amplify a 1700 bp fragment containing ITS1-5.8S-ITS2-LSU rDNA. Amplification conditions were as follows: 3 min at 98 ◦ C, followed by 30 cycles at 98 ◦ C for 15 s, 54 ◦ C for 15 s, and extension at 72 ◦ C for 1.5 min, with a final extension at 72 ◦ C for 10 min.

144

X. Lu et al. / European Journal of Protistology 67 (2019) 142–150

Table 1. Sampling information of Paramecium multimicronucleatum populations in this study. Population names

Sampling sites

Sampling time

Pop.a1 Pop.b1 Pop.b2 Pop.c1 Pop.c2 Pop.d1 Pop.d2 Pop.d3 Pop.d4 Pop.e1 Pop.e2 Pop.f2 Pop.h3 Pop.i3 Pop.i4 Pop.k1 Pop.k5 Pop.n1

a: Lake in South China Normal University, Guangzhou, China b: Wuyan bridge, Guangzhou, China b: Wuyan bridge, Guangzhou, China c: Dongshan lake, Guangzhou, China c: Dongshan lake, Guangzhou, China d: Donghao river, Guangzhou, China d: Donghao river, Guangzhou, China d: Donghao river, Guangzhou, China d: Donghao river, Guangzhou, China e: Nanzhou bridge, Guangzhou, China e: Nanzhou bridge, Guangzhou, China f: Shangchong river, Guangzhou, China h: Dasha river, Guangzhou, China i: Liuhua lake, Guangzhou, China i: Liuhua lake, Guangzhou, China k: Small pond near South China Normal University, Guangzhou, China k: Small pond near South China Normal University, Guangzhou, China n: Chebei river, Guangzhou, China

2013.06.13 2013.07.24 2013.09.20 2013.07.24 2013.09.20 2013.07.24 2013.09.20 2013.11.24 2014.01.17 2013.08.07 2013.10.10 2013.10.10 2014.06.10 2014.03.16 2014.05.13 2014.01.16 2014.09.20 2014.06.17

Primers COIF (5 -GGWKCBAAAGATGTWGC-3 ) (Strüder-Kypke and Lynn 2010) and COIR (5 TADACYTCAGGGTGACCRAAAAATCA-3 ) (Folmer et al. 1994) were used to amplify a 760 bp fragment of mitochondrial cytochrome c oxidase subunit I (COI). Amplification cycle conditions were as described above. Polymerase chain reaction with 50 ␮l reaction volume was performed using Thermo ScientificTM Phusion highfidelity DNA polymerase (2 U/␮l), 10 ␮l 5X Phusion HF buffer, 1 ␮l of 2.5 mmol/L dNTP mix, 2.5 ␮l of template, and 34.5 ␮l of ddH2 O. Phusion high-fidelity DNA polymerase was chosen, for its extremely low error rates and superior fidelity, which far surpasses that of regular Taq polymerase (∼ 52x). PCR products were purified using Universal DNA Purification Kit (TIANGEN) and were either directly sequenced or cloned, in order to compare nucleotide diversities between sequencing strategy. Five to ten clones per DNA sample were selected and their ITS1-5.8S-ITS2-LSU rDNA and COI genes were sequenced. Cloning was performed using pEASY-Blunt Cloning Vector and Trans 1-T1 Phage Resistant Chemically Competent Cells (TransGen, Beijing, China). All plasmids containing inserts were purified using a QIAGEN plasmid mini kit (Qiagen, Hilden, Germany). All samples were sequenced at the Beijing Genomics Institute (BGI).

Sequence and genetic structure analysis In this study, a total of 381 ITS1-5.8S-ITS2-LSU rDNA, and 304 COI sequences were newly generated. Sequences were imported into BioEdit v7.0.9.1 (Hall 1999) and checked by eye. Primer-binding regions and ambiguous bases at the 5 and 3 ends of the sequences were removed. For downstream analyses, the ITS1-5.8S-ITS2-LSU rDNA fragment

was split into ITS1-5.8S-ITS2 (ITS) and LSU rDNA. And 13, 24 and 16 distinct sequences were detected for ITS (GenBank No. MH544148-MH544160), LSU rDNA (GenBank No. MH544161-MH544184), and COI genes (GenBank No. MH544185- MH544200), respectively. The rest of the sequences are duplicates of the ones reported. In total, six datasets were created. Three datasets contained all the newly generated sequences: ITS-new (n = 381), LSU-new (n = 381), COI-new (n = 304). Three datasets, ITS-all (n = 52), LSUall (n = 51) and COI-all (n = 58), contained only the distinct sequences from this study (duplicates were excluded) along with those available in GenBank (Supplementary document A Table A2). We excluded duplicate sequences to avoid flooding the dataset containing the published sequences in GenBank. Nucleotide sequences of ITS and LSU rDNA were aligned with BioEdit v7.0.9.1 (Hall 1999), while those of COI gene were aligned according to the amino acids sequences using DAMBE 6.0 (Xia 2017). The median-joining method as implemented in the program Network 4.6.1.0 (Bandelt et al. 1999) was used to derive networks depicting the distribution and relationships among ITS, LSU rDNA and COI haplotypes of Paramecium multimicronucleatum. A minimum of one nucleotide site difference was used to define a new haplotype. In total, six networks were constructed, one for each dataset. Networks ITS-new (n = 381), LSU-new (n = 381), and COI-new (n = 304), contained all unique and duplicate sequences generated in this study. Networks ITS-all (n = 52), LSU-all (n = 51) and COIall (n = 58) contained only the distinct sequences generated herein (duplicates were excluded) along with those available in GenBank (Supplementary document A Table A2). Genetic diversity of haplotypes was calculated using the program DnaSP version 5 (Librado and Rozas 2009).

X. Lu et al. / European Journal of Protistology 67 (2019) 142–150

Results Genetic diversity analysis In total, ITS-new had 15 nucleotide polymorphisms, while LSU-new and COI-new had 29 and 49, respectively (Fig. 1). Regarding COI, 39 out of 49 polymorphic sites were in the third codon position, whilst eight and three polymorphisms corresponded to the first and second codon positions, respectively (Supplementary document A Table A3). Two, eight and 18 parsimony informative sites were present in ITS-new, LSU-new and COI-new, respectively (Fig. 1).

Haplotype network analyses and their geographic distribution A total of 17 haplotypes, with 6.5% genetic divergence, were identified within the dataset ITS-all (Fig. 2A). Thirteen of these haplotypes were generated in this study (I1–I13). Of these, only the dominant haplotype was recovered by both cloning and direct sequencing. The rest were recovered by cloning only. The ITS-all network (Fig. 2A) formed two major clades with all newly generated haplotypes being present in Clade 1. Within Clade 1, the dominant type sequence was I1, which contained most of the newly generated sequences along with sequences from Europe (Russia, Italy, Moldova, and Germany), the Americas (USA, Brazil) and Asia (China). The rest of the haplotypes generated by cloning (n = 15) spread into 12 satellites (I2-I13), directly descending from I1 (Fig. 2A; Supplementary document A Table A1). Clade 2 contained three haplotypes (I15-I17); I15 contained sequences from European populations (Italy, Poland, Germany and Cyprus), I16 contained sequences originating from USA and unknown localities, and I17 contained sequences from Australia, Japan and Russia (Fig. 2A). In the Dataset LSU-all, 24 haplotypes with 4.7% genetic divergence were identified, 15 of which were newly generated (Fig. 2B). Of these, two were recovered by both cloning and direct sequencing, 19 were amplified by cloning only and two were found by direct sequencing only. Within the LSU-all network (Fig. 2B), the dominant haplotype was S1, and contained sequences from Europe (Russia and Moldova), America (USA) and Asia (China). The rest of the cloning sequences (n = 20) spread into 14 satellites (Fig. 2B; Supplementary document A Table A1). Variation sites among haplotypes ranged from one to six (Fig. 2B). Haplotypes S16, S19 and S17 contained sequences originating from Europe. Specifically, S16 contained two sequences from Italy, S19 contained one sequence from Poland, and S17 contained three sequences from Cyprus and one from Russia (Fig. 2B). Finally, within Dataset COI-all, 24 haplotypes were identified with 18.2% genetic divergence, 12 of which were generated in this study (C1-C12) (Fig. 2C). The COI-all network was divided into three major clades (Fig. 2C). Clade

145

1 included three haplotypes. Haplotype C20 consisted of three sequences from USA, C23 contained two sequences from unknown locality, and C24 included sequences from Europe (Russia, Italy, Poland, Germany and Cyprus), and Oceania (Australia) (Fig. 2C). Clade 2 included six haplotypes. C12 was the dominant haplotype containing sequences from Europe (Russia, Italy, Moldova, Germany) and Asia (China). Haplotype C14 contained sequences from USA, C16 contained sequences from Europe (Russia, Moldova), while haplotypes C17, C18, and C19 contained sequences from Europe (Russia) (Fig. 2C). Clade 3 was the most diverse consisting of 15 haplotypes. C1 was the most abundant haplotype and contained sequences from Europe (Russia, Italy) and Asia (China). In total, 16 of the newly generated haplotypes grouped within Clade 3: four were derived from direct sequencing, and 12 were derived from cloning. Most sequences derived from direct PCR and cloning belonged to the dominant haplotype (C1) (Fig. 2C; Supplementary document A Table A1). The remaining sequences spread into nine satellites (Fig. 2C). One to five sequences were included within each satellite haplotype (Fig. 2C). Variable sites among haplotypes containing our new sequences ranged from one to three (Fig. 2C; Supplementary document A Table A1). C21 and C22 each contained a single sequence from Italy, C15 contained three sequences from Russia, and C13 contained one sequence from Brazil (Fig. 2C). In the ITS-new network, at least eight sequences from each of our 18 Guangzhou populations belonged to the dominant haplotype (Fig. 3A; Supplementary document A Table A1). And sequences in satellite haplotypes seemed to be from random populations (Fig. 3A; Supplementary document A Table A1). In the LSU-new network (Fig. 3B), the S1 haplotype of LSU-all network (Fig. 2B) was further divided into nine haplotypes [S1(1)–S1(9)]. Similarly, as shown in ITS-new network, sequences from all our 18 Guangzhou populations collected at different time points and sites also fell under the dominant haplotype [S1(2); 268 sequences] (Fig. 3B; Supplementary document A Table A1). Haplotype S1(1) contained 11 sequences derived from direct sequencing and 72 out of 333 sequences from cloning. Haplotypes S1(3)–S1(9) were represented by only one to three sequences. Additionally, satellite haplotype S6 of LSU-all network (Fig. 2B) was further divided into two haplotypes [S6(1), S6(2)] in LSU-new network (Fig. 3B; Supplementary document A Table A1). The number of variable sites among haplotypes ranged from one to eight (Fig. 3B). The nucleotide sequences of haplotypes S3 and S14 differed from the dominant haplotype by 0.69% and 0.69%, respectively (Figs. 1 and 3 B). In the COI-new network, sequences from 16 out of the 18 Guangzhou populations belonged to the dominant haplotype C1(1), and 10 out of these 18 belonged in C2(1) (Fig. 3C; Supplementary document A Table A1). By contrast, sequences in other haplotypes seemed to be from random populations (Fig. 3C; Supplementary document A Table A1). Haplotype

146

X. Lu et al. / European Journal of Protistology 67 (2019) 142–150

Fig. 1. Variable sites and parsimony informative sites in ITS-new, LSU-new and COI-new haplotypes of Paramecium multimicronucleatum. Parsimony informative sites are shaded with purple. Numbers in brackets are same as in Fig. 3.

C12 differed from the dominant haplotype by 32 nucleotides (Figs. 1 and 3C). Sequences in the datasets ITS-new (Fig. 3A), LSU-new (Fig. 3B), and COI-new (Fig. 3C) were longer thus, they had more haplotypes resulting in more diverse networks. Datasets containing all sequences (including those from GenBank) were trimmed according to the shortest sequence, thus some portion of the markers with polymorphisms were lost resulting in less diverse networks.

Discussion Suggested strategies for sequencing and gene marker selection There has been considerable discordance among studies assessing intra-specific diversity in ciliates. For instance, genetic diversity within a population was higher than that among populations. And genetic diversity among fine-scale populations was higher than that among large-scale populations (Miao et al. 2004; Tarcz et al. 2012; Zhao et al. 2013). We supposed that the observed contradictions among studies were partly due to inadequate sequencing depth and choice of sequencing strategy. Given the high copy number of macronuclear rDNA and number of mitochondria within a single ciliate cell (Prescott 1994; Zhao et al. 2013), the dominant haplotype of a population might not be captured, if only a few clones are picked for sequencing. Thus, we chose to test two sequencing strategies (direct sequencing and cloning) on two nuclear and one mitochondrial marker that are commonly used to address these questions. Regarding the two nuclear markers (ITS and LSU rDNA), all sequences derived directly from PCR – except for two PCR sequences of LSU rDNA – and most sequences derived from cloning (ranging from 66.7%–100.0% depending on

population), belonged to the dominant haplotype (Fig. 3A and B). This illustrates that, in the case of rDNA, direct sequencing of PCR products, the preferred and inexpensive option, is likely to capture the dominant haplotypes of each population, though some genetic diversity might still be missed (Fig. 2A–C). A possible explanation for the presence of dominant rDNA haplotype in a population is that the sequence might represent the one amplified from the micronuclear template during macronuclear development following conjugation (Juranek and Lipps 2007; McGrath et al. 2006). Additionally, nucleotide site mutations of macronuclear rDNA have been known to accumulate during vegetative growth after conjugation (Prescott 1994), which would lead to several rDNA haplotypes within a population. This explanation would account for contradictory results (higher intra-population rather than inter-population genetic diversity; higher genetic diversity among fine-scale populations rather than among large-scale). In the case of the COI gene, for each population, direct sequences as well as 37.5–100% cloning sequences belonged to the same haplotype (Fig. 3C). Genetic diversity of COI gene was much higher than the other two rDNA markers, in accordance with previous studies (Zhao et al. 2013). Thus, the “dominant” COI haplotypes could not be easily detected by direct sequencing. A possible explanation is that COI is a mitochondrial gene, thus during conjugation, conjugating cells will retain their copies since mitochondria are not exchanged (Maynardsmith and Szathmáry 1999). Thus, deep cloning sequencing rather than direct sequencing might be more appropriate to use for the COI gene. In the present study, we used ITS, LSU and COI to assess the level of genetic polymorphisms among and within temporal and fine-scale geographical populations. We found that estimates of genetic heterogeneity differed within a given population depending on the marker used to perform that estimate. For instance, when using the ITS region, population

X. Lu et al. / European Journal of Protistology 67 (2019) 142–150

147

Fig. 2. Haplotype networks of Paramecium multimicronucleatum. (A) ITS-all network containing 13 distinct sequences from this study and 39 sequences from GenBank. (B) LSU-all network containing 24 distinct sequences from this study and 27 sequences from GenBank. (C) COI-all network containing 16 distinct sequences from this study and 38 sequences from GenBank. Big black circles indicate intermediate or unsampled haplotypes and short lines between points represent nucleotide substitutions. Wherever there are more than seven substitutions, these are indicated by numbers. Circle size is proportional to haplotype frequency. Different colors represent different sampling locations.

E2 had the highest genetic diversity among all populations, while this was not the case when using LSU and COI genes (Supplementary document A Table A3). These observations confirm previous studies (Tarcz et al. 2012), and reinforce the suggestion that inference of genetic diversity should be based on more than one marker. A possible limitation of this study is that even though most of the sequences had duplicates, some were unique. Though we took utmost care to minimize errors, the possibility that some of the sequences represent cloning and/or sequencing errors cannot be excluded.

Genetic diversity and geographic distribution In all three markers, within each clade, the dominant haplotype contained sequences from several distinct geographic

regions (Fig. 2A–C), suggesting origin from a single founder population. In accordance with previous studies, there was no clear connection between geographical origin and Paramecium multimicronucleatum populations suggesting gene flow among localities (Barth et al. 2006; Tarcz et al. 2012). It is possible that insects or larger migratory animals, whose habitats are associated with water, introduced members of the founder population in bodies of water worldwide. To our knowledge, though a few studies examining finescale diversity of ciliate populations exist, they do not account for intra-population variation (e.g. Gentekaki and Lynn 2009; Przybo´s et al. 2011). Our current investigation of P. multimicronucleatum showed that intra-population genetic divergence at the fine scale level is equal to or higher than that among large-scale spatial populations (Fig. 2A–C). This is in agreement with previous studies of Paramecium bursaria

148

X. Lu et al. / European Journal of Protistology 67 (2019) 142–150

Fig. 3. Haplotype networks of Paramecium multimicronucleatum containing all newly obtained sequences. (A) ITS-new network, (B) LSUnew network, and (C) COI-new network. Big black circles indicate intermediate or unsampled haplotypes, short lines between points represent nucleotide substitutions. Wherever there are more than seven substitutions, these are indicated by numbers. Circle size is proportional to haplotype frequency. Different colors and styles represent different sampling locations and time points, respectively. Numbers in brackets represent subtypes of haplotypes in Fig. 2.

(Zhao et al. 2013). Both P. bursaria and P. multimicronucleatum are outbreeding ciliates, which have been hypothesized to have lower level intra-specific variation than inbreeding species (Nyberg 1988; Tarcz et al. 2012). Currently, we propose that geographic patterns of ciliates, especially those of fine-scale spatial populations of outbreeding paramecia, might be biased, if only one sequence from the cloning library is picked to represent the whole population. Moving forward, further deep sequencing studies are needed to test, whether the same pattern is present for inbreeding ciliates.

Genetic diversity among temporal populations In the present study, we collected Paramecium multimicronucleatum from six localities at various time points (Table 1). Though the ciliate was collected throughout the year in at least one sampling site, it was not found each time in each locality despite intense sampling efforts (Table 1) suggesting seasonal cycling. Cysts of P. multimicronuclea-

tum have never been observed (Wichterman 1986), thus the possibility of cells moving from one water body into another quickly and frequently, is not very likely. Similarly to this study, Coleps, Paramecium and spirotrichs, could not be collected consistently from one sampling location over a certain period of time (Barth et al. 2008; Przybo´s et al. 2011). The temporal variations observed in this and previous works could be caused by changes in the environment including physicochemical variables (Weisse 2006, 2007). Both Barth et al. (2008) and this study (Fig. 3A–C), found that several haplotypes were present within a single temporal population, and some temporal populations usually shared haplotypes. Moreover, network patterns of temporal populations usually differed according to which gene marker was used in the present investigation. For example, temporal samples D1–D4 shared a single haplotype (I1) when using ITS. However, when using LSU rDNA, no haplotype sharing was noted between D1 and D2–D4 (Supplementary document A Table A1). After performing deep sequencing, it became obvious

X. Lu et al. / European Journal of Protistology 67 (2019) 142–150

that for nuclear gene markers, the dominant haplotype of a gene was representative of all temporal samples from a sampling site (Fig. 3A–B), and that only a few temporal populations could be detected in specific satellite haplotypes. This suggests that despite the high nuclear sequence diversity observed within a temporal population, only dominant sequences are passed on to the next generation. This might also account for some of the observed discrepancies in previous studies (e.g. Gentekaki and Lynn 2009; Miao et al. 2004).

Author contributions Z. Yi designed the experiment. L. Huang, Y. Li & X.T. Lu collected samples and performed lab experiment. X.F. Lu, Y. Xu & Z. Yi did data analyses. X.F. Lu, E. Gentekaki, & Z. Yi drafted the manuscript. All authors reviewed the manuscript.

Acknowledgements This work was supported by the Natural Science Foundation of China (grant numbers 31772440, 31430077, 41576148, 31501850, 41576124), Pearl River Science and Technology Nova Program of Guangzhou (grant number 201610010162), and Special Support Program of Guangdong Province to YZ. Many thanks are due to Ms. Ye Tang and Dr. Jun Gong, former students in SCNU, for their help on sequencing and sampling. We also thank the two anonymous reviewers for their constructive suggestions.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/ 10.1016/j.ejop.2018.12.005.

References Bandelt, H.J., Forster, P., Röhl, A., 1999. Median-joining networks for inferring intraspecific phylogenies. Mol. Phylogenet. Evol. 16, 37–48. Barth, D., Krenek, S., Sergei, F., Berendonk, T.U., 2006. Intraspecific genetic variation in Paramecium, revealed by mitochondrial cytochrome c oxidase I sequences. J. Eukaryot. Microbiol. 53, 20. Barth, D., Tischer, K., Berger, H., Schlegel, M., Berendonk, T.U., 2008. High mitochondrial haplotype diversity of Coleps sp. (Ciliophora Prostomatida). Environ. Microbiol. 10, 626–634. Catania, F., W¨urmser, F., Potekhin, A.A., Przybo´s, E., Lynch, M., 2009. Genetic diversity in the Paramecium aurelia species complex. Mol. Biol. Evol. 26, 421.

149

Finlay, B.J., Fenchel, T., 1999. Divergent perspectives on protist species richness. Protist 150, 229. Finlay, B.J., 1996. Biodiversity at the microbial level: the number of free-living ciliates in the biosphere. Quart. Rev. Biol. 71, 221–237. Foissner, W., 1999. Protist diversity: estimates of the nearimpoderable. Protist 150, 363–368. Folmer, O., Black, M., Hoeh, W., Lutz, R., Vrijenhoek, R., 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol. Mar. Biol. Biotechnol. 3, 294–299. 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). Appl. Environ. Microb. 75, 3187–3195. Gong, J., Kim, S.J., Kim, S.Y., Min, G.S., Roberts, D.M., Warren, A., Choi, J.K., 2007. Taxonomic redescriptions of two ciliates, Protogastrostyla pulchra n. g. n. comb. and Hemigastrostyla enigmatica (Ciliophora: Spirotrichea Stichotrichia), with phylogenetic analyses based on 18s and 28s rRNA gene sequences. J. Eukaryot. Microbiol. 54, 468. Gong, J., Dong, J., Liu, X., Massana, R., 2013. Extremely high copy numbers and polymorphisms of the rDNA operon estimated from single cell analysis of oligotrich and peritrich ciliates. Protist 164, 369–379. Grolière, C.A., Chakli, R., Sparagano, O., Pepin, D., 1990. Application de la colonisation d’un substrat artificiel par les ciliés à l’Étude de la qualité des eaux d’une rivière. Eur. J. Protistol. 25, 381–390. Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids Symp. Ser. 41, 95–98. Juranek, S., Lipps, H., 2007. New insights into the macronuclear development in ciliates. Int. Rev. Cytol. 262, 219–251. Katz, L.A., Mcmanus, G.B., Snoeyenbos-West, O.L., Griffin, A., Pirog, K., Costas, B., Foissner, W., 2005. Reframing the microbial ‘everything is everywhere’ debate: evidence for high gene flow and diversity in ciliate morphospecies. Aquat. Microb. Ecol. 41, 55–65. Kusch, J., 1998. Local and temporal distribution of different genotypes of pond-dwelling Stentor coeruleus. Protist 149, 147–154. Lachance, M., 2004. Here and there or everywhere? Bioscience 54, 884. Librado, P., Rozas, J., 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25, 1451. Maynardsmith, J., Szathmáry, E., 1999. The Origins of Life: from the Birth of Life to the Origin of Language. Oxford University Press, New York. McGrath, C., Zufall, R., Katz, L., 2006. Genome evolution in ciliates. In: Katz, L., Bhattacharya, D. (Eds.), Genomics and Evolution of Eukaryotic Microbes. Oxford University Press, New York, pp. 64–77. Miao, W., Yu, Y., Shen, Y., Zhang, X., 2004. Intraspecific phylogeography of Carchesium Polypinum (Peritrichia, Ciliophora) from China, inferred from 18S-ITS1-5.8S ribosomal DNA. Sci. China Ser. C 47, 11–17.

150

X. Lu et al. / European Journal of Protistology 67 (2019) 142–150

Nyberg, D., 1988. Paramecium. In: Görtz, H.D. (Ed.), The Species Concept and Breeding Systems. Springer Press, Berlin, pp. 41–58. Prescott, D., 1994. The DNA of ciliated protozoa. Microbiol. Rev. 58, 233–267. Przybo´s, E., Tarcz, S., Greczek-stachura, M., Surmacz, M., 2011. Seasonal and spatial variability of species occurrence of the Paramecium aurelia, complex in a single natural pond (Protista, Ciliophora). Hydrobiologia 663, 233–244. 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). J. Biogeogr. 34, 1822–1831. Strüder-Kypke, M.C., Lynn, D.H., 2010. Comparative analysis of the mitochondrial cytochrome c oxidase subunit I (COI) gene in ciliates (Alveolata, Ciliophora) and evaluation of its suitability as a biodiversity marker. Syst. Biodivers. 8, 131–148. Tarcz, S., Potekhin, A., Rautian, M., Przybo´s, E., 2012. Variation in ribosomal and mitochondrial DNA sequences demonstrates the existence of intraspecific groups in Paramecium multimicronucleatum (Ciliophora, Oligohymenophorea). Mol. Phylogenet. Evol. 63, 500–509. Wang, C., Zhang, T., Wang, Y., Katz, L.A., Gao, F., Song, W., 2017. Disentangling sources of variation in SSU rDNA sequences from

single cell analyses of ciliates: impact of copy number variation and experimental error. P. Roy. Soc. B-Biol. Sci. 284. Weisse, T., 2006. Freshwater ciliates as ecophysiological model organisms – lessons from daphnia, major achievements, and future perspectives. Arch. Hydrobiol. 167, 371–402. Weisse, T., 2007. Distribution and diversity of aquatic protists: an evolutionary and ecological perspective. Biodivers. Conserv. 17, 243–259. Wichterman, R., 1986. The Biology of Paramecium, 2nd ed. Plenum Press, New York. Xia, X., 2017. Dambe6: new tools for microbial genomics, phylogenetics, and molecular evolution. J. Hered. 108, 431–437. Yi, Z., Huang, L., Yang, R., Lin, X., Song, W., 2016. Actin evolution in ciliates (Protist, Alveolata) is characterized by high diversity and three duplication events. Mol. Phylogenet. Evol. 96, 45–54. Zhao, Y., Gentekaki, E., Yi, Z., Lin, X., 2013. Genetic differentiation of the mitochondrial c oxidase subunit I gene in genus Paramecium (Protista, Ciliophora). PloS One 8. Zufall, R.A., Dimond, K.L., Doerder, F.P., 2013. Restricted distribution and limited gene flow in the model ciliate Tetrahymena thermophila. Mol. Ecol. 22, 1081–1091.