Diversity and Geographic Distribution of Free-Living Heterotrophic flagellates - Analysis by PRIMER

Summary Introduction Materials and methods Results Discussion Literature Sampling sites Excel spreadsheet

Summary: This account deals with the geographic distribution of approximately 350 species of heterotrophic flagellates (protozoa) which we have reported in over 30 inventories from sites around the world. The sites include marine and freshwater habitats, water column and benthic habitats, and ‘extreme’ (i.e. anoxic and hypersaline) habitats. The communities have been compared and analysed using the Cluster algorithm in the PRIMER package. We sought to minimise the impact of factors extrinsic to the actual distribution of organisms which may distort our perception of the geographic patchiness of organisms. In our analyses, communities from areas which are geographically close to each other do not cluster together. Communities from similar habitats cluster together (i.e., water column communities group together, as do benthic communities, or freshwater and marine communities), and communities from more unusual habitats segregate from those of less ‘extreme’ habitats. We argue that our capacity to draw robust conclusions on the distribution of these organisms is hampered by under-reporting. We favour a model in which there are relatively few species of free-living heterotrophic flagellates, and that most of these species have a cosmopolitan distribution.

Introduction. The degree of undescribed diversity among the protozoa is in dispute (Allsopp et al. 1995; Corliss 1982; May 1990). Corliss (1982) estimated that there are about 115,000 species of extant protists. More generally, May (1990) argued that for every 10-fold decrease in linear dimensions, there are 100-fold more taxa. He reported a fall-off of several orders of magnitude in the number of known species in the size range of protozoa. He was unable to determine if the paucity of known species reflected a lack of taxonomic attention - as is suggested by the apparent revelation of considerable undescribed microbial diversity by molecular means (DeLong et al. 1993; Embley and Stackebrandt 1994; Giovannini et al. 1990; Ward et al. 1990) or if there are few protozoan species - as has also been argued (Fenchel 1993; Finlay 1997, 1998). The taxonomic foundations for the views of Finlay and other protozoan ecologists (Cairns 1993) - that there are a small number of species with a cosmopolitan distribution - have not been established because uninterpreted records of taxa that are held to be conspecific are rarely presented.

Heterotrophic flagellates are significant members of the microbial communities which are important and possibly dominating elements of the biota of natural aquatic ecosystems (Azam et al. 1983; Sherr and Sherr 1988, 1994). Heterotrophic flagellates typically occur at concentrations of about 100 ~ 100,000 cells ml-1 (Berninger et al. 1991). They are consumers of bacteria, are prey for metazoa, and catalyse the recycling and remineralization of nutrients (Caron et al. 1988; Jürgen and Güde 1990; Kirchman 1994; Pace and Vaque 1994; Sherr and Sherr 1988).

The flagellates are one of the four types of protozoa recognized from the earliest taxonomic syntheses. This categorisation was based on the light microscopical appearance. It has become clear that groupings of protists based on light microscopy are often polyphyletic (Patterson 1994). An alternative view is that there are over 70 lineages of protists and these can be differentiated by reference to ultrastructural characteristics (Patterson 1994). From this perspective, there are over 40 lineages of heterotrophic flagellates. Heterotrophic flagellates therefore dominate the high level diversity of eukaryotes.

Despite their abundance in aquatic ecosystems and the dominance of high level diversity of eukaryotes by flagellates, the species-level diversity of this group is largely unknown. However, the recent review of all genera (Patterson and Larsen 1991) makes a full catalogue of known species achievable within the near future.

Here, we report on the degree of cosmopolitanism of heterotrophic flagellates, an issue which bears on the global biodiversity of these organisms.

The geographic distribution of organisms is determined by their evolutionary history and by their physiological preferences - both of which limit the distribution of organisms - and by the forces of dispersal will tend to obscure the patchiness arising from evolutionary history (Zeitzschel 1990). We refer to these three factors as intrinsic (Ekebom et al. 1996) Our capacity to perceive the geographic distribution can be distorted by factors that are not inherent to the biology of the organisms. The principal ‘extrinsic’ factors relate to sampling, taxonomic practices and species concepts. These factors can distort our perceptions if they suggest that communities which are dissimilar are similar, or if they suggest that communities which are similar are dissimilar. Extrinsic factors which might have these kinds of effects include the following.
(1) Under-sampling and under-reporting: These apply when the number of species recorded from a site is (significantly) less than the number of species present. It can be caused by a failure to completely sample sub-habitats, by not taking account of temporal patchiness or by a failure to include sparsely distributed species. Even if diversity is well sampled, only a small number of species present may be reported (Finlay 1997, 1998). The exclusion of rare species may lead to an erroneous generalization of cosmopolitanism, but under-reporting which excludes species more randomly is likely to lead to an erroneous sense of endemism. Sampling protocols structured specifically to address the issue of under-reporting and undersampling are required.
(2) Taxonomic issues: Individual taxonomists usually favour certain taxa and do not report on all members of a community - an example being a preference for choanoflagellates or scale-bearing taxa in some studies of heterotrophic flagellates(e.g., Thomsen et al. 1997; Tong 1997b). Species lists generated by different authors may therefore not be comparable and may lead to an erroneous sense of endemism.
(3) Lumpers and splitters: Different taxonomists take different approaches about whether small differences in appearance should form the basis for new species or not. When lists are compared, those produced by lumpers will contain fewer species and will appear similar. This will lead to a conclusion that the communities are relatively cosmopolitan. Lists made by splitters from the same communities will include more species and there will be less overlap. Comparison of these species lists may favour the conclusion that the dissimilarity of the communities is as a result of endemism.
(4) Inconsistent nomenclature and identification: A species of kinetoplastid which attaches to the substrate by the tip of its posterior flagellum, has a recurrent anterior flagellum and feeds on suspended bacteria is called Bodo saltans in the recent Western literature and Pleuromonas jaculans in the Russian literature (Hänel 1979). The widespread lack of type material adds to the difficulties of ensuring consistent nomenclatures by workers isolated from each other. This can be resolved in part by the use of uninterpreted records (photography, video, etc.). Differences in nomenclature or identification exacerbate differences among species lists and will favour a sense of endemism.
(5) Taxonomic instability: With the passage of time, taxonomists may divide species, merge them, or move them to different genera. These are normal changes in taxonomy. They make comparisons among lists of species more difficult and will usually create an inappropriate sense of endemism.
(6) Nomenclatural instability. The names of taxa change when synonymies are established. Nominal taxa with a short nomenclatural life are reported occasionally. Such taxa will appear to have a restricted distribution even if they have a more widespread distribution. This problem can be rectified when a full taxonomic history of each species is developed.
(7) Species concepts: Broad species concepts will lead to a sense of cosmopolitanism, narrow concepts may lead to a sense of endemism. At this time, the alpha taxonomy of heterotrophic flagellates is based on a morphospecies approach. In the case of species bearing scales or other excrescences (choanoflagellates), the concept may rely on electron-microscopical (i.e. more discriminatory) characteristics. Genera with such traits tend to be more speciose - suggesting that traditional morphospecies offer a facade behind which hide a much greater number of physiological or molecular species (Cairns 1993).
(8) Subjective analysis of data: If the mechanism by which statements about endemism are developed from the data on distribution is not specified, then the reliability of the conclusions must be questioned. We have carried out a series of studies in which we have sought to minimise or standardise the effects of extrinsic factors. We carry out surveys of living material. We work on site where possible to minimise problems of loss of numbers of individuals and of the loss of identity of species associated with fixing, storing or transporting samples (Kemp et al. 1993). We rely heavily on photographic and video records to provide uninterpreted and archivable reference material. We work with data from a small group of workers to minimise factors 2, 3 and 4. In this account, we address the issue of the subjective analysis of data, and report on the analysis of data by PRIMER. Preliminary results were reported by Ekebom et al. (1996).

Results

The data used in this analysis are available as an excel spreadsheet. In a few cases we have updated information published previously - see the following Table - because we are aware that some species were observed in earlier studies but were not reported because of insufficient information; or that identifications were incorrect. We have included these changes in the spreadsheet. We have also updated some previously published lists to accommodate subsequent synonymies.

Below, sites included in this analysis. All sites are marine (inclusive of brackish and hypersaline) unless otherwise specified. Australian site, MB: marine plankton, FB: freshwater benthos, MI: marine ice, MAB: marine anoxic benthos.

Figure 1 (left) shows the number of sites in which species (above) or genera (below) have been encountered in our surveys. Most species and genera have only been reported from one site.

Figure 1 (left) shows the number of sites in which species (above) or genera (below) have been reported in the literature. Most species and genera have only been reported from one site.

The similarities of communities indicated by Cluster in PRIMER and based on the analysis of data on genera is shown in Fig. 3.

Communities from the same geographic region (such as Australia) do not cluster together. The first and second major separations (at about 13% and 20% similarity) separate communities from ‘extreme’ habitats (anoxic and hypersaline) from more ‘normal’ habitats. Communities from anoxic habitats cluster together, whether or not the original sites were marine or freshwater. The most remote community within the normal cluster is that from the water column of freshwater lakes in Antarctica - but we note that the species list is derived from two sampling occasions and from material that was shipped from Antarctica to Sydney before analysis. At a level of about 37% similarity, communities from comparable habitats (all water column communities; all benthic communities) cluster together. The benthic communities divide into freshwater and marine clusters at a level of about 42% similarity.

Groupings at higher levels of similarity usually cannot be rationalised against geography, habitat type, or climate etc. Similar insights are obtained when the analysis is carried on species data (Fig. 5). Species with unclear identity (i.e. species reported as aff. or cfr.) have been excluded from this analysis. When compared with the results on genera, there is minor repositioning of some branches relating to communities from the water column and marine sediments.

Figure 4 Dendrogram showing the (Bray-Curtis) similarities (%) among communities; taxonomic information based on species only.

Communities from the same geographic regions do not cluster together. Communities from ‘extreme’ habitats (firstly hypersaline then anoxic) are the most remote in the analysis. In contrast to the analysis of genera, communities from freshwater sediments cluster external to marine communities. Communities from the water column form a discrete cluster (we regard deep water sediments are unusual for reasons given below); and communities from marine sediments also cluster together.

We carried out one analysis which embraced species lists from other authors (Fig. 5) and the described communities failed to cluster with our own analyses from similar geographical sites (Denmark, Finland, Greenland).

Figure 5. Dendrogram showing the (Bray-Curtis) similarities (%) between the communities from 31 surveys-being those included in figure 4 plus the data of Ilavalko and Thomsen (1997).

Discussion

The data presented here fail to allow an unambiguous picture to emerge. Most species have been reported from one or a few sites (Fig. 1). This may be interpreted as indicating endemism. The same result is obtained from a survey of the literature on the geographic distribution of marine heterotrophic flagellates (Patterson & Lee 2000). The same literature survey reveals that the majority of species and the majority of publications relate to Europe. The high frequency with which species are reported from a single location may not be due to endemism, but to an extrinsic factor - an imbalance in the reporting of communities which has favoured those from European sites.

The apparent endemism in Fig. 1 may also be caused if species are sparsely distributed and are therefore being under-reported. In support of this, the data indicate that over 150 species have been reported from marine sediments, yet frequently fewer than 50% of this number are reported from any one site. Our records suggest that the evidence of occurrence of many species is based on information from one cell or very few cells. Again, this suggests that many species are sparsely distributed and therefore rarely encountered, or are being under-reported for other reasons. About 54 species were described from Fiji in 1985 (Larsen and Patterson 1990). Yet, despite numerous taxonomic surveys, some species have not been described again. While this could be due to endemism, we believe that the study of Botany Bay supports the argument that many species are rare. The Botany Bay community appears distinctive because the community includes some genera (e.g. Clautriavia, Metanema, Heterochomonas, Salpingoeca, Sphenomonas and Stephanoeca) not reported from other benthic sites. Botany Bay is the most exhaustively surveyed benthic site and we attribute its distinctiveness to an increased number of records of rare taxa arising from the more intense nature of the survey. We therefore feel that the results given in Figure 1 are caused not by endemism but by under-reporting. As unique records may reflect rarity as well as endemism, another test for endemism must be generated. In our view this is the occurrence of species in b iogeographically proximate communities but not in distant communities from comparable habitats. In its crudest form, we would have evidence for endemism if we were to find the same species in several Australian sites but not encounter these species from, say, northern hemisphere, Atlantic locations.

Two data sets deserve comment. The remote location of the community from freshwater plankton in Antarctica may be attributable to the unusual processing of this sample (see ‘Materials and Methods’). The data for the mid-North Atlantic site (Patterson et al. 1993) includes surveys of a variety of locations from 20º 28’N to 62º 50’N and of various depths and is therefore a more bundled list than any others in the analysis.

Otherwise, communities do not group by geography but segregate on habitat type - and this is evident in the analyses of genera and species. These groupings are consistent with the view (Fenchel 1987; Patterson et al. 1989) that the distribution of protozoa is largely determined by ecophysiological factors such as oxygen, redox potential, illumination, temperature, and so on. Salinity appears to be a determining factor in community composition, as communities from freshwaters and hypersaline habitats segregate from communities from habitats with a normal salinity. Communities from the water column separate broadly on the basis of temperature and/or latitude. One warm water element (Belize) clustered with communities from cold water - but the species list from that site may have been distorted because this material was shipped from Belize to Copenhagen for analysis.

We included one survey involving species lists from another group of workers (Ikävalko and Thomsen 1997 from the Baltic Sea). The reported communities adopted a peripheral location in our analysis (Fig. 5) and did not cluster with our own surveys from similar habitats from the same geographic region (i.e. Denmark, Finland, Greenland). This has confirmed our suggestion above that the involvement of different taxonomists has the capacity to lead to spurious conclusions about endemism.

We attribute the conflict between the apparent endemism suggested by taxon-based information (Fig. 1) and the lack of endemism suggested by comparisons of communities (Figs 3, 4) to the incompleteness of the data. In the absence of more exhaustive studies, we retain our model (Ekebom et al. 1996; Patterson and Simpson 1996; Tong et al. 1997) that the constituents of these communities are not determined to any detectable extent by geographic location, but they are determined by other factors (habitat type). On the basis of this model, on our taxonomic studies, and on a synthesis of the relevant literature (Patterson & Lee, 2000), we concur with the view (Finlay 1997, 1998) that there is not a great diversity of morpho-species of heterotrophic flagellates - perhaps no more than 3000 in total (dinoflagellates excluded).

Two general insights have therefore emerged. The first is the lack of strong evidence for endemism, and this is in conflict with usual assumptions about metazoa, metaphyta and indeed many algae (Tyler 1996). Secondly, the number of species in this group is considerably fewer than might be expected from the diversity of bacteria in natural ecosystems or from the abundance of individuals of flagellates in aquatic ecosystems or from the fact that over half of the types of eukaryotic organisation include or are exclusively the adaptive group called flagellates.

Methods

The sites studied are in the Table above and shown in the associated figure. Communities were examined on site with the exception of the materials from Antarctic sites which was collected and shipped for examination in Sydney and from the Belize site which was transported to Copenhagen for analysis.
Samples were obtained and processed as indicated elsewhere (Ekebom et al. 1996; Tong et al. 1997). Analysis of similarities among genus and species were conducted using the statistical package PRIMER version 4.0 beta (Clarke 1993). A matrix comprising data for about 350 individual species at 31 different habitats has been assembled (Appendix 1), and the species lists were extracted from published and unpublished data (Table 1). We used PRIMER-CLUSTER, a hierarchical classification technique based on the Bray-Curtis Similarity Coefficient calculated on presence / absence-transformed data employing group average sorting.

• Al Qassab, S., Lee, W. J., Murray, S., and Patterson,. D. J. (2002). Flagellates from stromatolites and surrounding sediments in Shark Bay, Western Australia. Acta Protozoologica, 41: 91-144
• Allsopp D, Colwell RR, Hawksworth DL, eds (1995) Microbial diversity and ecosystem function. CAB International, Oxford, UK.
• Azam F, Fenchel T, Field JG, Gray JS, Meyer-Reil LA, Thingstad F (1983) The ecological role of water-column microbes in the sea. Mar Ecol Prog Ser 10: 257-263.
• Berninger U-G, Caron DA, Sanders RW, Finlay BJ (1991) Heterotrophic flagellates of planktonic communities, their characteristics and methods of study. In Patterson DJ, Larsen J (eds.) The biology of free-living heterotrophic flagellates. Syst. Ass. special vol. No. 45, pp. 39-56, Oxford.
• Cairns JJ (1993) Can microbial species with a cosmopolitan distribution become extinct? Speculations in Science and Technology 16: 69-73.
• Caron DA, Goldman JC Dennett MR (1988) Experimental demonstration of the roles of bacteria and bacterivorous protozoa in plankton nutrient cycles. Hydrobiologica 159: 27-40.
• Clarke KR (1993) Non-parametric multivariate analyses of changes in community structure. Australian J Ecol 18: 117-143.
• Corliss JO (1982) Numbers of species comprising the phyletic groups assignable to the kingdom protista. J Protozool 29: 499.
• Daugbjerg N, Vørs N (1992) Preliminary results from a small scale survey of marine protists from Northern Foxe basin in the vicinity of Igloolik Island June 1992. In Søeberg B, Jensen D, Schurmann H, Steffensen JF, Curtis MA, Vørs N, Daugbjerg N, Bushnell P (eds.) Research on Arctic biology, Igloolik, Northwest territories, Canada. ISBN 87-7834-022-5
• DeLong EF, Franks DG, Alldredge AL (1993) Phylogenetic diversity of aggregate attached vs. free-living marine bacterial assemblages. Limnol Oceanogr 38: 924-934.
• Ekebom J, Patterson DJ, Vørs N (1996) Heterotrphic flagellates from coral reef sediments (Great Barrier Reef, Australia). Arch Protistenkd 146: 251-272.
• Embley TM, Stackebrandt E (1994) The use of 16s ribosomal RNA sequences in microbial ecology. In Pickup RW, Saunders JR, Codd GA (eds.) Molecular appreoaces to environmental microbiology. Chapmann and Hall, Oxford, pp 39-62.
• Fenchel T (1987) The ecology of protozoa. Science Tech Inc, Madison.
• Fenchel T (1993) There are more small than large species? Oikos 68: 375-378.
• Finlay BJ (1997) The global diversity of protozoa and other small species. Australian biologist 10: 131-151.
• Finlay BJ (1998) The global diversity of protozoa and other small species. Int J Parasitol 28: 29-48.
• Foissner W (1987) Soil protozoa: fundamental problems, ecological significance, adaptations in ciliates and testaceans, bioindicators, and guide to the literature. Progress in protistology 2: 69-212.
• Giovannoni SJ, Britschgi TB, Moyer CL, Field KG (1990) Genetic diversity in Sargasso Sea bacterioplankton. Nature 345: 60-62.
• Hänel K (1979) Systematik und Ökologie der farblosen Flagellaten des Abwassers. Arch Protistenkd 121: 73-137.
• Ikävalko J, Thomsen HA (1997) The Baltic Sea ice biota (March 1994): a study of the protistan community. Europ J Protistol 33: 229-243.
• Jürgens K, Güde H (1990) Incorporation and release of phosphorus by planktonic bacteria and phagotrophic flagellates. Mar Ecol Prog Ser 58: 271-284.
• Kemp PF, Sherr BF, Sherr EB, Cole JJ (1993) Handbook of methods in aquatic microbial ecology. Lewis Publishers, Boca Raton, USA.
• Kirchman DL (1994) The uptake of inorganic nutrients by heterotrophic bacteria. Microb Ecol 28: 255-271.
• Larsen J, Patterson DJ (1990) Some flagellates (Protista) from tropical marine sediments. J Nat Hist 24: 801-937.
• May RM (1990) How many species ? Phil Trans R Soc Lond B 330: 293-304.
• Pace ML, Vaque D (1994) The importance of Daphnia in determining mortality rates in protozoans and rotifers in lakes. Limnol Oceanogr 39: 985-966.
• Patterson DJ (1994) Protozoa, evolution and systematics. In Hausmann K, Hülsmann N (eds.) Progress in protozoology, proceedings of the IX international congress of protozoology. Gustav Fischer, Stuttgart, pp 1-14.
• Patterson DJ, Larsen J, eds (1991) The biology of free-living heterotrophic flagellates. Clarendon Press, Oxford.
• Patterson DJ, Larsen J, Corliss JO (1989) The ecology of heterotrophic flagellates and ciliates living in marine sediments. Progress in Protistology 3: 185-277.
• Patterson, D.J. and Lee, W.J. 2000. Geographic distribution and diversity of free-living heterotrophic flagellates. In: 'The Flagellates" Eds Green, J. and Leadbeater, B.S.C. Taylor and Francis, 269-287.
• Patterson DJ, Nygaard K, Steinberg G, Turley C (1993) Heterotrophic flagellates and other protists associated with oceanic detritus throughout the water column in the mid North Atlantic. J mar biol Ass UK 73: 67-95.
• Patterson DJ, Simpson AGB (1996) Heterotrophic flagellates from coastal marine and hypersaline sediments in Western Australia. Europ J Protistol 32: 423-448.
• Sherr EB, Sherr BF (1988) Role of microbes in pelagic food webs: a revised concept. Limnol Oceanogr 33: 1225-1226.
• Sherr EB, Sherr BF (1994) Bacterivory and herbivory: key roles of phagotrophic protists in pelagic food webs. Microb Ecol 28: 223-235.
• Simpson AGB, Hoff J, Bernard C, Burton HR, Patterson DJ (1997) The ultrastructure and systematic position of the euglenozoan Postgaardi mariagerensis, Fenchel et al. Arch Protistenkd 147: 213-225.
• Thomsen HA, Garrison DL, Kosman C (1997) choanoflagellates (Acanthoecidae, Choanoflagellida) from the Weddell Sea, Antarctica, taxonomy and community structure with particular emphasis on the ice biota; with preliminary remarks on choanoflagellates from Arctic Sea ice (Northeast water polynya, Greenland). Arch Protistenkd 148: 77-114.
• Tong SM (1997a) Heterotrophic flagellates from the water column in Shark Bay, Western Australia. Mar Biol 128: 517-536.
• Tong SM (1997b) Choanoflagellates in Southampton Water, including the description of three new species. J mar biol Ass UK 77: 929-958.
• Tong SM (1997c) Heterotrophic flagellates and other protists from Southampton Water, U.K. Ophelia 47: 71-131.
• Tong SM Vørs N, Patterson DJ (1997) Heterotrophic flagellates, centrohelid heliozoa and filose amoebae from marine and freshwater sites in the Antarctic. Polar Biol 18: 91-106.
• Tyler PA (1996) Endemism in freshwater algae. Hydrobiologia 336: 127-135.
• Vørs N (1992a) Heterotrophic amoebae, flagellates and heliozoa from the Tvärminne area, Gulf of Finland, in 1988-1990. Ophelia 36: 1-109.
• Vørs N (1992b) Heterotrophic protists (excl. dinoflagellates, loricate choanoflagellates, and ciliates). In Thomsen HA (ed.) Plankton from inner Danish Waters. An analysis of the autotrophic and heterotrophic plankton in Kattegat. HAV 90 Report. Danish national agency for environmental protection (in Danish). Havsforskning fra Miljøstyrelsen 11: 195-250.
• Vørs N (1993a) Heterotrophic amoebae, flagellates and heliozoa from arctic marine waters (North West Territories Canada and West Greenland). Polar Biol 13: 113-126.
• Vørs N (1993b) Marine heterotrophic amoebae, flagellates and heliozoa from Belize (Central America) and Tenerife (Canary Islands), with descriptions of new species, Luffisphaera bulbochaete n. sp., L. longihastis n. sp., L. turriformis n. sp. and Paulinella intermedia n. sp. J Euk Microbiol 40: 272-287.
• Vørs N, Buck KR, Chavez FP, Eikrem W, Hansen LE, Østergaard JB, Thomsen HA (1995) Nanoplankton of the equatorial Pacific with emphasis on the heterotrophic protists. Deep Sea Res 42: 585-602.
• Ward DM, Weller R, Bateson MM (1990) 16S rRNA sequences reveal numerous uncultured microorganisms in a natural community. Nature 345: 63-65.
• Zeitzschel B (1990) Zoogeography of marine protozoa: an overview emphasizing distribution of planktonic forms. In Capriulo GM (ed.) Ecology of marine protozoa. Oxford University Press, Oxford, pp 139-185