Composition and species richness of flagellate protozoa from environments associated to the Baía river (Mato Grosso do Sul State, Brazil): influence of the hydrological period and the connectivity

Despite flagellate protozoa are important for the metabolism of aquatic ecosystems, researches about the ecology of this community is scarce. Therefore, this study investigated the influence of flood regime and hydrological connectivity on the patterns of composition and species richness of protozoan flagellates from aquatic environments in the upper Paraná river floodplain. To this end, samplings were undertaken in two periods of the hydrological cycle (March 2008 and September 2008), at six lentic environments associated with the Baía river (three are connected to the river, and three are isolated). 120 taxa were identified in the studied area, distributed in 10 orders and one residual group. This high species richness was sustained by the order Euglenida, which contributed with 70% of total species. The species composition was significantly different between periods and environments, as well as species richness, which was higher in connected environments during the high water period. The results evidenced that the flood pulse and hydrological connectivity are relevant factors driving the structure and dynamic of plankton flagellate protozoa community in the upper Paraná river floodplain.


Introduction
The hydrological regime is considered the driving force both for the ecological functioning and for the maintenance of biodiversity patterns in floodplains, since the periodical oscillations of the water flow induce deep changes in the structure and dynamic of the biota (NEIFF, 1990;JUNK et al., 1989).
Among the aquatic communities in floodplains, flagellate protozoa play important role in the energy flow and nutrient cycling (AZAM et al., 1983), because they transfer to the upper trophic levels the energy obtained through the consumption of bacteria, organic detritus, viruses and pico and nanoplanktonic algae (XU et al., 2005) and, contribute significantly for the primary production in the ecosystems (SAFI; HALL, 1997).
Although it is unquestionable the importance of this community to aquatic ecosystems, studies about are still scarce.Only the autotrophic and Acta Scientiarum.Biological Sciences Maringá, v. 32, n. 4, p. 349-356, 2010 mixotrophic fractions have received a little more attention in studies on phytoplankton.Thus, holistic studies incorporating all the fractions are almost nonexistent, especially in floodplains.Consequently, the effect of the flood regime and the hydrological connectivity on the composition and species richness of flagellate protozoa is not known.Furthermore, few studies have investigated the species composition of flagellate protozoa and most of them did not examine all the fractions of this community (AUER; ARNDT, 2001;COMTE et al., 2006;DOMAIZON et al., 2003;KISS et al., 2009;LAVRENTYEV et al., 2004;TIKHONENKOV, 2007; TIKHONENKOV; MAZEI, 2006MAZEI, , 2008;;WEITERE;ARNDT, 2003).In the same way, there are no records of studies specifically approaching the species richness patterns of these organisms.In Brazil, the studies on this community are still scarce (ARAÚJO; COSTA, 2007; ARAÚJO; GODINHO, 2008;PEREIRA et al., 2005) and most of them does not present taxonomical focus and includes the other components of the protozooplankton.
In this way, the present study aimed to accomplish a species faunistic inventory and investigate the influence of the flood regime and hydrological connectivity on the patterns of composition and species richness of flagellate protozoa from six lentic environments associated to the Baía river in the upper Paraná river floodplain.
To this end, we predict that the patterns of composition and species richness are different between the environments (with and without connection) and between the periods of the hydrological cycle (high and low water), and we expect that: i) the species composition is more similar between the environments with and without connection during the high water period due to the homogenizing effect of the floods (THOMAZ et al., 2007) and ii) the higher species richness occur in the environments connected to the Baía river during the high water period due to the lotic influence.

Material and methods
The present study was developed in the upper Paraná river floodplain that presents high environmental heterogeneity, due to presence of lakes, backwaters, channels and rivers.Among these environments, Baía river (22 o 43'23.16"S;53 o 17' 25.5"W) is situated in the right bank of Paraná river, in the Mato Grosso do Sul State.It is a meandering river that presents a great number of concatenated lakes along its course.Moreover, due to the low current flow and small slope of its bed, this river is characterized as semilotic (SOUZA FILHO; STEVAUX, 1997).
We accomplished two semestrial samplings in six lakes associated to the Baía system (Figure 1), three without connection to the river (Aurélio, Fechada and Pousada das Garças) and other three connected to it (Guaraná, Maria Luiza and Porcos).The first sampling was performed in March 2008, during the high water period, and the second in September 2008, during the low water period.All the samplings, in replica, were performed in the littoral and pelagic region from each (n = 48), using a plankton net (20 μm) and also through passage of a flask at subsurface.The individuals were analyzed in vivo, under optical microscope, to avoid alterations in the cell shape and dimension arisen from the preservation process.The aliquots were obtained using a monochannel micropipette and placed on glass slides, which were examined until no new species arose during this analysis.The individuals were identified, whenever possible, at species level, based on morphological traits and specialized literature.The results were presented according to the classification system developed by Lee et al. (2000).
In order to test the differences in the species richness between the environments (with and without connection with the river) and between the periods (high and low water) we employed a factorial analysis of variance (Two-way ANOVA).The spatial and temporal patterns in the species composition were evaluated using a detrended correspondence analysis (DCA), since according to Gauch (1982), this analysis is the most indicated for group of ecological data because it describes well the distribution of the sampled population.Afterwards, we applied further analyses of variance (Two-way ANOVA) to the scores from each axis to test differences in the species composition between the environments (with and without connection with the river) and periods (high and low water).The assumptions of normality and homoscedasticity were previously tested and the significance level adopted was of p < 0.05.
The detrended correspondence analysis indicated a clear distinction among the environments and mainly between the periods (Figure 3).This pattern was corroborated by the analysis of variance applied to the DCA scores, which evidenced a significant interaction between the factors for the DCA 1 and significant difference between the connected and isolated environments for the DCA 2 (Table 2).The highest species richness were verified in the environments connected to Baía river during the high water period (Figures 4a and b) and these differences were validated by the analysis of variance that evidenced significant differences between the connected and isolated environments (F = 26.915;p < 0.01) and between the hydrological periods (F = 15.561;p < 0.01).

Discussion
The faunistic inventory revealed that the community of flagellate protozoa in the studied environments is quite rich, probably due to the relevant contribution of Euglenida order, which contributed with 67% from the total of identified species.The individuals from this order, specially the genera Euglena, Phacus and Trachelomonas, which were the most species-rich in the present study, are commonly found in aquatic environments with high biochemical demand of oxygen, rich in organic matter and ammonia (REYNOLDS et al., 2002).These preferences might have favored the success of Acta Scientiarum.Biological Sciences Maringá, v. 32, n. 4, p. 349-356, 2010 these organisms during the high water period due to the great accumulation of allochthonous organic matter (JUNK et al., 1989;LEWIS JR. et al., 2000;NEIFF, 1990).
The second most important order, especially during the low water period, was Cryptomonadida that contributed with 6% from the total of identified species.The success of these organisms in this period may be associated to the opportunistic traits that provide them competitive advantage to tolerate adverse conditions in the low water period (BOVO-SCOMPARIN; TRAIN, 2008;JONES, 2000).The other orders contributed far less expressive, i.e. all together contributed with 27% from the total of identified species.These orders were only a little more representative during the low water period, probably because many of these organisms (especially the orders Peridiniales, Gymnodiniales and Chromulinales) present adaptations that permit them to tolerate the unfavorable conditions (JONES, 2000;REYNOLDS et al., 2002).
The pattern of species composition recorded in the present study is quite different from those found by other authors in several temperate rivers and lakes (AUER; ARNDT, 2001;COMTE et al., 2006;DOMAIZON et al., 2003;KISS et al., 2009;LAVRENTYEV et al., 2004;TIKHONENKOV, 2007;TIKHONENKOV;MAZEI, 2006MAZEI, , 2008;;WEITERE;ARNDT, 2003).The predominance of Chrysomonadida is frequently observed in these environments and in general Spumella spp. is the most common species.Similar result to the registered in the present study was verified by Araújo and Costa (2007) that observed a great contribution of Euglenida order for the species richness in reservoirs from Brazilian semiarid region due to the enrichment of the environments by organic matter.
The species composition may be a useful tool to understand the functional processes from an ecosystem since it reflects the environmental characteristics, as observed in the present study.Nevertheless, the first hypothesis of this study cannot be totally accepted, since despite the influence of the flood regime and hydrological connectivity, the species composition from the environments associated and isolated from the Baía river was not similar during the high water period, as previously predicted.This probably occurred because the local factors (such as morphometry, wind action, rainfall etc) were more important than the regional ones.
On the other hand, the patterns of species richness corresponded to the expected, with higher values in the environments connected to the Baía river during the high water period, thus, corroborating our second hypothesis.The greater values of species richness during the high water period may be related to the increase in the connectivity that promotes the exchange of water, sediment, nutrients and organisms between the different habitats (BONECKER et al., 2005;NEIFF et al., 2001;WARD;STANFORD, 1995).Also, to the contribution of periphytic community, since in shallow environments, as wetlands, the increase in the water flow may cause the detachment of periphytic organisms, and then preventing the establishment of limits among the populations from the different compartments of the aquatic ecosystem.
Otherwise, the lower values of richness in the low water period are associated to the absence of lotic activity and to the action of local factors that may limit the flagellate development during this period, such as reduced depth, high turbidity, resuspension of sediment, which cause changes in the biotic interactions (CARVALHO et al., 2001;LEWIS JR. et al., 2000).
In summary, the flood regime and the hydrological connectivity influenced the patterns of composition and species richness of flagellate protozoa as already observed in other planktonic communities from the upper Paraná river floodplain (ALVES et al., 2005;AOYAGUI;BONECKER, 2004;BINI et al., 2003;DOMITROVIC, 2002).However the non corroboration of the homogenizing effect of the floods on the species composition during the high water period may be associated to the temporal scale of the sampling, which was quite limited, since this theory is widely proved with other communities.Therefore it is necessary to accomplish further studies with a longer temporal scale for a better understanding of the processes that regulate the structure and dynamic of this community.In the same way, it is necessary to perform holistic researches evaluating integrally the composition, species richness, abundance and biomass of this community in these systems.

Figure 1 .
Figure 1.Location of the environments associated to the Baía river in the upper Paraná river floodplain.

Figure 3 .
Figure 3. Distribution of the scores along the first two DCA axes, based on presence and absence data of flagellate protozoa species from six environments associated to Baía river, during the high and low water periods (FA: Isolated, high water; FB: Isolated, low water; AA: Connected, high water; AB: Connected, low water).

Figure 4 .
Figure 4. Mean number of registered taxa: a) in the environments (connected and isolated) associated to Baía river; b) during the high and low water periods.

Table 2 .
Eigenvalue of the axis 1 and 2 from DCA and results from the analyses of variance (Two-way Anova) applied to the scores from each axis.*Significant values (p < 0.05).