Skip to main content
NCBI home page
Springer logoLink to Springer
. 2026 Sep 15;208(12):638. doi: 10.1007/s00203-026-05164-2

The community patterns and assembly mechanism varied in eukaryotic microorganisms with different cell sizes

Wei Hu 1, Xiaofan Zhao 2, Zenghui Zhang 3, Yumei Qian 4, Xinxin Lai 5, Kexin Zhang 6, Pengfei Yang 1, Zongcan Yang 7,✉, Hanghang Han 3,✉
PMCID: PMC13577960  PMID: 42742650

Abstract

Cell size is a fundamental determinant of eukaryotic microorganisms’ ecology, influencing community dynamics, physiological processes, and ecological interactions. However, it is still unclear whether eukaryotic microorganisms of different size have similar structural and assembly mechanisms. In this study, we sorted eukaryotic microorganisms’ cells into two sizes classes (large-sized eukaryotic microorganisms (LE) and small-sized eukaryotic microorganisms (SE)) and compared their structural and assembly mechanisms along the Han River. The results showed that LE fractions had higher diversity compared to SE fractions, the two size classes harbored distinct eukaryotic microorganisms’ communities. And. Assembly mechanisms of the two classes were significantly different. We conclude that cell size is an important factor that determines eukaryotic microorganisms’ structure and assembly mechanisms. These findings underscore the critical role of cell size in determining eukaryotic microorganisms’ community and uncover a direct relationship between microbial morphology and ecological traits.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00203-026-05164-2.

Keywords: Cell size, Eukaryotic microorganisms, Community patterns, Assembly mechanism

Introduction

In river ecosystems, microbial communities (including bacteria, archaea and eukaryotic microorganisms) are key components that drive material cycling, energy flow and ecosystem stability (Wang et al. 2025). Among them, eukaryotic microorganisms (microbial eukaryotes), including protists (such as flagellates, ciliates, amoebae), microalgae (such as diatoms, green algae) and microfungi, play an irreplaceable role in the food web structure, organic matter decomposition and pollutant degradation of river ecosystems (Barbosa et al. 2022; Sleith and Katz, 2022; Wang et al. 2025). Eukaryotic microorganisms occupy multiple trophic levels in river ecosystems and serve as both primary producers (such as photosynthetic microalgae), consumers (such as predatory protists), and decomposers (such as saprophytic fungi), thereby forming complex nutritional interactions within the river food web (An et al. 2025; Xiong et al. 2025). Therefore, studying the community dynamics of eukaryotic microorganisms is crucial for understanding the material cycling and energy flow in river ecosystems.

Cell size is one of the most fundamental and functional pivotal traits of microorganisms, which profoundly shapes microbial physiological metabolism, resource utilization strategies, dispersal capacity, and stress tolerance (He et al. 2026). For microorganisms of different cell sizes, variations in individual volume significantly differentiate their adaptability to environmental filtering, sensitivity to abiotic gradients such as nutrient availability, temperature, salinity, as well as their vulnerability to biotic interactions including predation, competition, and symbiosis (Hu et al., 2023, 2024). Typically, small-sized microorganisms feature high reproductive efficiency and strong dispersal ability, enabling them to adapt to fluctuating and oligotrophic environments (Haro-Moreno et al. 2018). In contrast, large-sized microorganisms possess complex cellular structures and strong metabolic capacity, playing irreplaceable roles in organic matter decomposition and energy transfer (Hu et al. 2022). Such size-dependent trait differences are presumed to fundamentally reshape community distribution patterns and the relative contribution of deterministic and stochastic assembly processes.

To date, research focusing on microbial cell size has predominantly targeted bacteria inhabiting drinking water (Liu et al., 2018), rivers (Proctor et al. 2018; Zhang et al. 2023), oceans (Hu et al. 2023a) and lakes (Santos et al. 2019). A limited body of studies has also explored cell-size patterns of eukaryotic microbes (He et al. 2026; Ma et al., 2022) and revealing substantial compositional divergence between eukaryotic microbial communities partitioned by cell size (0.2–3 and 3–200 μm) (Ma et al., 2022). Nevertheless, an important research gap remains: numerous prior investigations collected eukaryotic microorganisms using 0.45 μm filter membranes (Egge et al. 2021; Gregori et al. 2025; Sleith and Katz, 2022). It remains unclear how community composition differs between eukaryotic microbes smaller than 0.45 μm and those retained on filters with pore sizes greater than 0.45 μm. Relevant empirical evidence addressing this comparison is still scarce.

Given this research gap, this study focuses on the size-dependent differentiation of eukaryotic microbial communities. By partitioning eukaryotic microorganisms into different cell size fractions, we systematically compare the differences in their community diversity, composition structure, and spatial distribution patterns. Furthermore, we quantify the relative contributions of deterministic environmental filtering and stochastic dispersal limitation to the assembly of size-specific eukaryotic microbial communities. This study aims to clarify the community assembly mechanisms in eukaryotic microorganisms with varying cell sizes, enrich the theoretical system of microbial trait-based ecology, and provide a more refined perspective for revealing the maintenance mechanisms and ecological functions of eukaryotic microbial ecosystems.

Materials and methods

Water samples

Water samples were collected in December 2024 from a tributary near the Danjiangkou Reservoir (China). Samples were taken from 10 sites along the river. At each site, 5 L of surface water was collected and transported to the laboratory under low-temperature conditions. The sampling locations are shown in Figure S1.

Environmental factor determination

Routine physical and chemical parameters (temperature (T), pH, dissolved oxygen (DO) were measured on-site using a multifunctional water quality analyzer (ORION 520 M-01 A); Total phosphorus (TP) was measured using a multi-parameter water quality analyzer (DR3900, Hach, Loveland, CO, USA); Total nitrogen (TN), total carbon (TC), inorganic carbon (IC), total dissolved organic carbon content (TOC) were determined using a total organic carbon analyzer (multiN/C3100, Analytikjena, Germany). Table S1 shows the water parameters for each site.

Filtration

The water sample was sequentially passed through membrane filters with pore sizes of 0.45 μm to separate small-sized eukaryotic microorganisms (SE) and large-sized eukaryotic microorganisms (LE). The 0.45 μm Durapore® membrane filter (Merck Millipore, USA) was employed to retain LE, while the 0.1 μm Durapore® membrane filter (Merck Millipore, USA) was used to capture SE from the 0.45 μm filtrate.

DNA extraction, PCR amplification and illumina sequencing

Total environmental DNA was extracted with the Power Water DNA isolation kit (Omega Bio-Tek, USA). The quality and concentration of DNA were determined by 1.0% agarose gel electrophoresis and a NanoDrop2000 spectrophotometer (Thermo Scientific, United States) and kept at -80 ℃ prior to further use. For eukaryotic microorganisms community, the 18S rRNA genes were amplified using the universa primers: SSU0817F (5’-TTAGCATGGAATAATRRAATAGGA-3’) and 1196R (5’-TCTGGACCTGGTGAGTTTCC-3’). Each DNA sample was individually PCR-amplified in triplicated 25 ul reactions. Cycling conditions were an initial denaturation step at 95 °C for 33 min; 27 cycles of 95 °C for 30 s, 55 °C for 30 s and 72 °C for 30 s; and a final 10-min extension at 72 °C. All PCR reactions were carried out with Phusion® High-Fidelity PCR Master Mix (New England Biolabs). After electrophoresis, the PCR products were purified using the AMPure® PB beads (Pacifc Biosciences, CA, USA) and quantified with Qubit 4.0 (Thermo Fisher Scientific, USA). Purified products were pooled in equimolar and DNA library was constructed using the SMRTbell prep kit 3.0 (Pacifc Biosciences, CA, USA) according to PacBio''s instructions. Purified SMRTbell libraries were sequenced on the Pacbio Sequel IIe System (Pacifc Biosciences, CA, USA) by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). High-fidelity (HiFi) reads were obtained from the subreads, generated using circular consensus sequencing via SMRT Link v11.0.

Data processing

Purified amplicons were combined in equimolar concentrations and subjected to paired-end sequencing on the Illumina PE300 platform (Illumina, San Diego, USA) following the standard procedures provided by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The resulting raw FASTQ files were de-multiplexed with an in-house Perl script and subsequently quality-filtered with fastp (v0.19.6) (Chen et al. 2018). Subsequently, operational taxonomic units (OTUs) were clustered based on 97% sequence similarity using UPARSE, and a representative sequence from each OTU was chosen for taxonomic assignments (Edgar 2013). Microeukaryotic OTUs were against the Protist Ribosomal Reference (PR2 4.10) database (Zhang et al. 2024a, 2024b). For each OTU, the most dominant sequence was designated as the representative sequence. Raw data have been deposited to National Center for Biotechnology Information (NCBI) under the BioProject number PRJNA11466480.

Statistical analysis

Statistical analyses were primarily conducted using R software (v4.3.2) (Liu et al. 2024a, 2024b). For network analysis, Spearman’s rank coefficients (q) between all 97%-cutoff OTUs in each sample were calculated pairwise using R package Hmisc (Harrell, 2008). Subsequently, those significant (FDR-adjusted P value < 0.01) and robust (q ≥ 0.6) correlations between OTUs were exported using R package igrap. Network visualization was conducted using Gephi. Non-Metric Multidimensional Scaling analysis (NMDS) was utilized to visualize the community composition of eukaryotic microorganisms. The linear discriminant analysis (LDA) effect size (LEfSe) (http://huttenhower.sph.harvard.edu/LEfSe) was performed to identify the significantly abundant taxa (phylum to genera) of eukaryotic microorganisms among the different groups (LDA score > 2, P < 0.05). Turnover indicates species replacement, whereby species occurring at one site are substituted by distinct species at another site. Nestedness arises from differences in species richness rather than species turnover, where the species pool at one site constitutes a subset of the assemblage at another site (Liu et al. 2024a, 2024b). To disentangle the contributions of species turnover and richness differences to community variation of eukaryotic microorganisms, we partitioned taxonomic beta diversity based on Bray–Curtis dissimilarity. Statistical computations were performed using the “agricolae” and “adespatial” packages, and visualized using the “ggtern” and “vcd” packages (Liu et al. 2024a, 2024). Additional bioinformatic analyses of the water microbiota were performed using the Majorbio Cloud platform (https://cloud.majorbio.com). In the neutral community model, the OTUs were separated into three partitions depending on whether they occurred more frequently than (above partition), less frequently than (below partition), or within (neutral partition) the 95% confidence interval of the neutral community model predictions. The R2 obtained by the neutral community model represents the overall fit to the neutral model, suggesting the contribution of stochastic processes, and the m-value is an estimate of migratory ability for microbes (Liu et al. 2024a, 2024b).

The null model analysis was used to calculate the β nearest taxonomic unit index (βNTI). The deterministic and stochastic processes were classified into five ecological processes based on the Rap-Crick index (RC Bray) values of βNTI and Bray-Curtis. the fraction with βNTI < − 2 is considered as the homogeneous selection, whereas those with βNTI > 2 as the heterogeneous selection. The fraction of pairwise comparisons with RC < − 0.95 and |βNTI|<2 is treated as the homogenizing dispersal, while those with RC > 0.95 and |βNTI|<2 as the dispersal limitation. The remains with |βNTI|<2 and |RC|<0.95 represent the drift (Hu et al., 2023).

Based on the intra-module connectivity (Zi) and inter-module connectivity (Pi) of each node, all nodes were categorized into four categories: network hubs (Zi > 2.5; Pi > 0.62), module hubs (Zi > 2.5; Pi ≤ 0.62), connectors (Zi ≤ 2.5; Pi > 0.62) and peripherals (Zi ≤ 2.5; Pi ≤ 0.62). Among them, network hubs, module hubs, and connectors were considered putative keystone species (Zhang et al. 2024a, 2024b). The neutral model was employed to assess the contribution of neutral processes to community assembly (Woodcock et al., 2006). This model characterizes the relationship between species abundance and frequency, with a neutral-based assembly process predicted within the model and a niche-based assembly process implied outside the model. Based on the detected ratio of OTUs among all samples, microbial OTUs were assigned into three groups: persistent (detected in > 80% of samples), intermittent (detected in 20–80% of samples), and transient (detected in < 20% of samples) (Liu et al. 2024a, 2024b). Then, frequency and abundance distribution of these OTU groups among SE and LE were counted and compared (Liu et al. 2024a, 2024b). The niche breadth of each OTU was calculated based on Levin’s niche breadth index using the “MicroNiche” package (Liu et al. 2024a, 2024b). According to the niche breadth value, microbial OTUs were categorized into environmental specialists (< 1.5) or generalists (> 3) (Liu et al. 2024a, 2024b). All the analysis of the SE/LE groups was carried out using the Majorbio Cloud platform (https://cloud.majorbio.com).

Results

Composition of eukaryotic microbial communities

In this study, we mainly analyzed the community composition of small-sized eukaryotic microorganisms (SE) and large-sized eukaryotic microorganisms (LE). The results showed that a total of 561,608 reads in LE fractions and 530,198 reads in SE fractions were identified. Further analysis showed that there was no significant difference between the number of reads of two fractions (Figure S2). The NMDS plot showed that the pronounced differences in microeukaryotic community composition (at the OTU level) with different cell size (Fig. 1A), which is statistically confirmed by ANOSIM (R2 = 0.602, P = 0.001). The total number of OTUs shared in the two fractions is 192, and the number of unique OTUs is 101 (LE) and 46 (SE), respectively (Fig. 1B). Further analysis showed that as the total number of OTUs increases, the community similarity of the two fractions of eukaryotic microorganisms shows a significant upward trend (Fig. 1C). Additionally, the Shannon index and the community dissimilarity between two fractions were significantly different (Fig. 1D and F). Beta diversity partitioning analysis uncovered that turnover, rather than nestedness, played a more dominant role in community succession in LE and SE communities (Fig. 1E).

Fig. 1.

Fig. 1

Open in a new tab

Principal component analysis showing the distribution pattern of eukaryotic microorganisms’ community (A); The number of unique and common species in LE and SE (B); Trend analysis of similarity in common species communities (C); Analysis of Shannon index differences between different groups (D); Triangular plots showing community ecological processes that contribute to eukaryotic microorganisms. Each point represents a pair of samples, its position is determined by a triplet of values from the similarity (S) = 1 D (dissimilarity), Repl (species replacement), RichDiff (richness difference) matrixes; each triplet sums to 1 (E); Analysis of community dissimilarity differences between different groups (F). * P < 0.05

The community composition of eukaryotic microorganisms with different cell size were compared. A total of 339 OTUs were identified across all fractions corresponding to 25 phyla (LE: 24, SE: 22), 59 classes (LE: 54, SE: 56), 86 orders (LE: 77, SE: 81), 116 families (LE: 101, SE: 108) and 146 genera (LE: 128, SE: 129) (FigureS3). At the class level, the dominant eukaryotic microbial communities were Intramacronucleata, Cryptophyceae, Chytridiomycota, Malasseziomycetes, Eurotiomycetes, Kathablepharidae, Dothideomycetes, Sordariomycetes, respectively (Fig. 2A). Further analysis of the community composition of the two fractions of eukaryotic microorganisms revealed that 21 (76.03%) phyla and 111 (76.03%) genera were shared, 3 (12%) phyla and 17 (11.64%) genera (Paraphelidium, Nuclearia, Paramecium, the diatoms (Ceratium, Cyclotella, Skeletonema)) were unique to the LE fractions, and 1 (4%) phylum and 18 (12.33%) genera (Lembadion, Tetrahymena, the fungal genus (Oxyporus, Sterigmatomyces, Carcinomyces, Saitozyma)) were specific to the SE fractions (FigureS3).

Fig. 2.

Fig. 2

Open in a new tab

The composition, diverse species and persistent patterns of eukaryotic microbial communities; Relative abundance of eukaryotic microbial communities at the class level (A); The bar chart shows the average relative abundance differences of the same species among different groups (B); Persistence pattern of fungal species (abundance and proportion of quantity) (C)

Further analysis showed that the relative abundance of Spirotrichea was significantly higher in LE fractions than that in SE fractions. Whereas the relative abundance of Cryptophyceae, Malasseziales, Eurotiales, Capnodiales, Hypocreales, Saccharomycetales, Dothideales, Agaricales, Pleosporales were significantly higher in SE fractions than that in LE fractions (Fig. 2B).

Subsequently, we partitioned microbial taxa into persistent and transient species based on their prevalence among samples. LE and SE fractions showed prevalence of persistent species, which accounted for 88.38% in LE group and 88.62% in SE group. However, persistent, and intermittent OTUs occupied the most relative abundance of LE fractions (84.64%), SE fractions (89.37%) (Fig. 2C).

The assembly process of eukaryotic microbial communities

In the present study, a significant difference of beta deviation index was found between LE and SE fractions (Fig. 3A). Niche width analysis showed that the SE fractions had a narrower niche width than the LE fractions (Fig. 3B). A significant difference in niche overlap among species was also observed between the LE and SE fractions (Fig. 3C).

Fig. 3.

Fig. 3

Open in a new tab

The beta deviation index (A), habitat breath (B) and niche overlap (C) of the eukaryotic microorganisms community. Dashed lines indicate niche breadth cut-offs (D). Stacked bar graph displaying the relative proportions of generalists, neutralists and specialists within LE and SE assemblages (E). Scatter plot illustrating log-transformed niche breadth versus averaged relative abundance to distinguish generalists (orange), neutralists (grey), and specialists (blue). * P < 0.05, ** P < 0.01, *** P < 0.001

Additionally, all species were subsequently categorized into generalist and specialist groups (Fig. 3D). The proportions of generalist, neutralists and specialist species in the LE fractions were 87.85%, 6.16% and 5.99%, respectively, whereas in the SE fractions, these proportions were 93.37%, 3.52% and 3.11%, respectively (Fig. 3E).

A null model was used to explore the community assembly mechanisms of the LE and SE fractions (Fig. 4A). The results revealed that homogeneous selection (63.08%) was the dominant process, followed by drift (22.65%) and dispersal limitation (9.16%) in the LE fractions. In contrast, drift (66.36%) was the dominant process in the SE group, followed by dispersal limitation (18.11%) and homogeneous selection (12.69%). All the relative importance of different ecological processes in the two groups were significantly different (Fig. 4B). Overall, stochastic processes primarily shaped the LE and SE fractions (Fig. 4C). The neutral community model (NCM) analysis further showed that species dispersal was higher in the LE fractions (m = 0.0755; Fig. 4D) than in the SE fractions (m = 0.0036; Fig. 4E).

Fig. 4.

Fig. 4

Open in a new tab

Community assembly mechanisms of LE and SE communities. A, B Relative importance of five ecological assembly processes. Neutral community model fit for LE and SE assemblages (C, D). Solid blue lines represent the best-fit neutral model; dashed blue lines denote 95% confidence intervals. Points indicate taxa with occurrence frequencies above (cyan), below (red), or within (black) model predictions. * P < 0.05, ** P < 0.01, *** P < 0.001

The co-occurrence pattern of eukaryotic microbial communities

The network analysis was employed to explore microbial interaction patterns. In the present study, LE fractions network contains 149 nodes (species) and 943 edges (associations), among which the proportion of positive correlations is 74% and the proportion of negative correlations is 26%. The SE fractions network consists of 147 nodes and 708 edges, with a positive correlation ratio of 82% and a negative correlation ratio of 18% (Fig. 5A and B).

Fig. 5.

Fig. 5

Open in a new tab

Co-occurrence network analysis of eukaryotic microorganisms community of different cell sizes. Network visualization depicting eukaryotic microorganisms community interactions within different size fractions (A: LE, B: SE). Comparison of the topological properties across eukaryotic microorganisms with different cell size (C, D, E and F). Zi-Pi plots show the distribution of OTUs based on their topological roles in networks (G: LE, H: SE). * P < 0.05, ** P < 0.01, *** P < 0.001

The statistical test results of key topological parameters showed that there were significant differences in the structural characteristics of the two networks (P < 0.05) (Fig. 5): the node connectivity, degree centrality, and closeness centrality of the LE group were significantly higher than those of the SE fractions, indicating that the majority of species in the LE fractions had higher connectivity and more direct connections within the network; while the betweenness centrality of the SE fractions was significantly higher than that of the LE fractions, suggesting that a few species in the SE fractions had stronger “bridging” roles and the core nodes of the network were more concentrated.

Based on Zi-Pi analysis, the key species in the two fractions were identified (Fig. 5C and D). The LE fractions identified 3 network hub species and 1 module hub species, while the majority of the remaining species were peripheral species; The SE fractions identified 7 connector species and 1 network hub species, with no module hub species, and the proportion of peripheral species was higher. This results indicated that there are more global hub species in the LE fractions, which have a stronger role in maintaining the overall network structure; While the key species in the SE fractions are mainly connectors between modules, the network structure is more dependent on weak connections between modules, and the stability of the community may be more susceptible to changes in key nodes.

Drivers of eukaryotic microbial communities

Mantel tests were applied to explore the relationships between environmental variables and community composition of LE and SE fractions (Fig. 6A). The results showed that the community structure of SE fractions exhibited significant correlations with TC, TN and TP. In contrast, most environmental variables displayed non-significant associations with the LE fractions. Pearson correlations between dominant eukaryotic taxa and environmental parameters for LE and SE fractions, respectively (Fig. 6B and C). the results showed that the relative abundance of the taxa such as strombidium in LE fractions and Malassezia in SE fractions were significantly corrected with environmental factors including TC, TN and TP at the genus level. Variation partitioning analysis further quantified the independent explanatory power of each environmental factor for eukaryotic community turnover in LE and SE fractions, showing that the higher contribution of environmental factors to the distribution of LE community were T, pH, TN, TP (Fig. 6D); Whereas higher contribution of environmental factors to the distribution of SE fractions were TOC, DO, pH, TC (Fig. 6E).

Fig. 6.

Fig. 6

Open in a new tab

Influence of environmental factors variables on the community of eukaryotic microorganism with different cell size. Mantel test results illustrate the correlation between environmental factors and the community structure of eukaryotic microorganism (A). Relationships between eukaryotic microorganisms’ taxa and environmental factors (B: LE, C: SE). Contribution of environmental factors to the distribution of eukaryotic microorganisms with different cell size (D: LE, E: SE). * P < 0.05, ** P < 0.01, *** P < 0.001

Discussion

The cell size of eukaryotic microorganisms (such as algae, protists, and fungi) is a core characterization of their fundamental life history strategies (Zhang et al. 2024a, b, c), which is directly related to energy acquisition and metabolic rate (Andersen et al. 2016; Blanchard et al. 2017), environmental adaptability and interspecific interaction (Brose et al. 2017), and thereby shapes community structure through multi-dimensional mechanisms. Previous study have characterized the distribution of eukaryotic microorganisms of different sizes in water bodies (De Vargas et al. 2015). For example, the microbial eukaryotic plankton communities within different sizes (0.2–3 and 3–200 μm) had significant differences in their community composition (Ma et al., 2022), which was found that the small and the large size fractions of the eukaryotic microorganism’s community differed markedly in taxonomic structure. This size-dependent effect is commonly reported in the bacteria in freshwater and marine ecosystems (Hu et al., 2023; Liu et al. 2025, 2016).

It is worth noting that a greater OTU overlap between LE and SE fractions was found, indicating that different groups of eukaryotic microorganisms can also undergo dynamic transformations. Similar phenomena were reported in the previous study that the variability in community dissimilarity between bacterial size fractions suggests a dynamic exchange between the two fractions (Hu et al., 2023). Those results indicated that widespread phenotypic plasticity in cell size within microorganisms. Single species may form both small and large morphotypes and thus be recovered in two size fractions. Plastic adjustment of cell size acts as an adaptive trait responding to ambient niches, community interactions and resource regimes. Smaller morphotypes favor dispersal and survival under nutrient limitation, whereas larger cells support higher metabolic rates and trophic activity. This intraspecific size variation complicates physical size fractionation, and indicates that the 0.45 μm cutoff separates ecological phenotypes, not rigid taxonomic groups. Despite this, we can still observe the size differences in the ecological phenotypes within the eukaryotic microbial community. As we found that the diversity within the SE fractions was significantly higher than that in the LE fractions, consistent with a previous report (Blais et al. 2024). However, this pattern contrasts with observations for bacteria, where larger-cell-size fractions exhibited significantly higher diversity than small-cell-size fractions. Such divergent patterns likely reflect fundamental differences between eukaryotic and prokaryotic microorganisms. For eukaryotes, size-fractionated filtering largely captures taxonomically distinct groups with inherent cell-size variation and divergent environmental preferences. Our results suggest that the small-cell-size fraction harbours higher eukaryotic microbial diversity, which merits greater attention in future investigations.

Based on the ecological niche breadth and distribution, eukaryotes communities were classified into generalists and specialists: generalists occur across diverse habitats and possess broad ecological niches, whereas specialists are restricted to specific habitats due to their narrow environmental tolerance (Li et al. 2023a, b). In the present study, the SE fractions exhibited a higher proportion of generalists and a lower proportion of specialists compared with LE fractions in terms of relative abundance, indicating that the degree of ecological variability in the LE fractions during the succession stage was greater than that of SE fractions. Generalists tolerate wide environmental fluctuations under variable river hydrology, which facilitates their persistence under frequent mass dispersal. In comparison, LE fractions harbour more specialists adapted to narrow niche spaces. These specialists rely on specific resources or microhabitats; they are more sensitive to local environmental selection and contribute substantially to biogeochemical functions within the river.

The community construction mechanism is one of the cores of ecological research (Nemergut et al., 2013). In this study, we found that homogeneous selection (63.08%) was the dominant process in the LE fractions and drift (66.36%) was the dominant process in the SE fractions (Fig. 4B). All the relative importance of different ecological processes in the two groups were significantly different (Fig. 4B). Similar phenomena have also been observed in bacterial community assembly. For example, the assembly processes of small-sized and large-sized bacterial communities differ in marine ecosystems (Hu et al., 2023). A study based on the distance-decay model also confirmed the role of cell size (Astorga et al. 2012), and dispersal limitation may gradually become more powerful in explaining community variation with the increase in cell size (Zhang et al. 2018), which was in line with our study (Fig. 4B). And larger plankton may undergo more environmental filtration and have lower plasticization, while smaller plankton have a stronger diffusion capacity (Li et al. 2023a, b).

These results indicated that cell size exerts a significant impact on microbial community assembly. Further analysis showed that the assembly process of LE and SE communities were dominated by stochastic processes (Fig. 4C). Despite limited research on the assembly of eukaryotic microbial communities with different cell sizes, the results in the previous study have supported the predominant role of stochastic processes in shaping eukaryotic community assembly (Chen et al. 2019; Mo et al. 2021). It is critical to acknowledge several analytical limitations in the present study. Ecological inferences drawn from null model analyses and co-occurrence networks built on sequence-based relative abundance should be interpreted cautiously. Correlations between taxon abundances do not necessarily represent true biological interactions, as synchronous shifts can arise from shared environmental filtering rather than direct microbial crosstalk (Pettersen et al. 2021).

River environmental heterogeneity underpins the high diversity of eukaryotic microbial communities (Gu et al. 2024; Okamoto et al. 2022; Sun et al. 2023). Water‑quality variables and physical conditions exert selective pressure to filter for more well‑adapted taxa. In this study, we found that the SE fractions had a stronger correlation with TC, TN, and TP than other environmental factors, such T, pH, DO and TOC. We suspect that it might be related to these reasons: First, some of these factors (e.g. T, pH, DO and TOC) are not substantially different across the ten sampling sites, leading no significant correlation between the community of two fractions with the environment factors. Second, environmental factors, such as TC, TN, and TP, could directly influence the growth and reproduction of SE communities by regulating their substrate utilization and metabolic processes (Ji et al. 2025). Those results indicated that nitrogen and phosphorus nutrients are the key environmental factors driving the differentiation of microbial community structure in water environments. For instance, low-oxygen habitats can enrich anaerobic protists, whereas high-nutrient regimes favour fast-growing algal groups (Zhang et al. 2012). The results were consistent with the research laws of microbial ecology in most freshwater and landscape water bodies (Chen et al. 2019). Total nitrogen and total phosphorus, as core nutrient indicators of water bodies, directly regulate the available resource levels of microorganisms (Hu et al., 2023). Through environmental screening, they change the growth rates, niche adaptability and competitiveness of different groups, thereby leading to directional succession of community composition. Additionally, biotic interactions including predation (e.g., ciliates grazing on bacteria), competition (e.g., light-resource competition among diverse algae), and symbiosis (e.g., mutualistic associations between fungi and algae) further modulate eukaryotic community structure (Xie et al. 2022).

Furthermore, this study adopts an operational 0.45 μm filter cutoff for size fractionation, which has certain limitations. It should be noted that such filtration cutoffs are somewhat arbitrary and do not strictly align with biological cell-size boundaries. Other pore-size thresholds may have equal or greater ecological relevance for resolving eukaryotic microbial size-based patterns. In addition, cell aggregates and small pico-eukaryotic cells may pass through or be retained non-specifically by 0.45 μm membranes. Sequential multi-pore-size fractionation would be required in future investigations to characterize continuous cell-size-dependent patterns more comprehensively.

Conclusions

This study systematically analyzed the composition differences and construction mechanisms of eukaryotic microbial communities of different particle sizes, and clarified that particle size is a key trait regulating community structure and ecological processes. The research showed that the communities of different cell sizes exhibit significant differentiation in terms of α diversity and community composition. The large-sized eukaryotic microorganisms have lower diversity and broader ecological niches than small-sized group. The assembly process of large-sized and small-sized eukaryotic microorganisms’ communities were dominated by stochastic processes. As the cell size decreases, the contributions of drift and dispersal limitation increase significantly, while the contributions of homogeneous selection and heterogeneous selection decrease significantly. In conclusion, the size differentiation that drives the ecological niche differentiation and construction process of eukaryotic microorganisms is an important dimension for understanding the structure of micro-food webs, biogeochemical cycles, and community stability. The study provides a unified framework for the assembly mechanism of different-sized eukaryotic microbial communities, and also offers theoretical support from the size perspective for resource protection and microbial community regulation.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1 (708KB, docx)

Acknowledgements

This work was supported by the Science and Technology Research Project in Henan Province (252102320216), Key Research Project of Higher Education Institutions in Henan Province (26A180028), Open Project of Key Laboratory of Agro-Forestry Environmental Processes and Ecological Regulation (AFEPER202504) and Doctoral Startup Fund of Zhengzhou University of Light Industry (2025BSJJ027).

Author contributions

W.H.: Conceptualization, experiment operation, data curation, formal analysis, writing-original draft, visualization, funding acquisition. Y.Q.: data validation. Z.Z., X. Z.: figure plotting. X.L., K. Z.: result interpretation. P.Y.: Supervision, Z. Y., H. H.: manuscript revision. All authors have read and agreed to the published version of the manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Conflict of interest

The authors declare no competing interests.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Zongcan Yang, Email: yangzongcan666@163.com.

Hanghang Han, Email: hanhanghang88@163.com.

References

  1. An Y, Zhang L, Li S, Zhang Y (2025) The response of soil eukaryotic microbial communities to afforestation in mountainous area of the Loess Plateau, northern China. PLoS One 20:e0317235. 10.1371/journal.pone.0317235 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Andersen KH, Berge T, Gonçalves RJ, Hartvig M, Heuschele J, Hylander S, Jacobsen NS, Lindemann C, Martens EA, Neuheimer AB, Olsson K, Palacz A, Prowe AEF, Sainmont J, Traving SJ, Visser AW, Wadhwa N, Kiørboe T (2016) Characteristic sizes of life in the oceans, from bacteria to whales. Annu Rev Mar Sci 8:217–241. 10.1146/annurev-marine-122414-034144 [DOI] [PubMed] [Google Scholar]
  3. Astorga A, Oksanen J, Luoto M, Soininen J, Virtanen R, Muotka T (2012) Distance decay of similarity in freshwater communities: do macro- and microorganisms follow the same rules? Glob Ecol Biogeogr 21:365–375. 10.1111/j.1466-8238.2011.00681.x [DOI] [Google Scholar]
  4. Barbosa M, Lefler FW, Berthold DE, Briggs-Gonzalez VS, Mazzotti FJ, Laughinghouse HD (2022) Trophic state drives the diversity of protists in a tropical river (New River, Belize). Microorganisms 10:2425. 10.3390/microorganisms10122425 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Blais M, Matveev A, Lovejoy C, Vincent WF (2024) Size-dependent community patterns differ between microbial eukaryotes and bacteria in a permafrost lake–river–sea continuum. Limnol Oceanogr 69:667–680. 10.1002/lno.12511 [DOI] [Google Scholar]
  6. Blanchard JL, Heneghan RF, Everett JD, Trebilco R, Richardson AJ (2017) From bacteria to whales: Using functional size spectra to model marine ecosystems. Trends Ecol Evol 32:174–186. 10.1016/j.tree.2016.12.003 [DOI] [PubMed] [Google Scholar]
  7. Brose U, Blanchard JL, Eklöf A, Galiana N, Hartvig M, Hirt R, Kalinkat M, Nordström G, O’Gorman MC, Rall EJ, Schneider BC, Thébault FD, Jacob E, U (2017) Predicting the consequences of species loss using size-structured biodiversity approaches. Biol Rev 92:684–697. 10.1111/brv.12250 [DOI] [PubMed] [Google Scholar]
  8. Chen S, Zhou Y, Chen Y, Gu J (2018) fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics 34(17):i884–i890 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chen W, Ren K, Isabwe A, Chen H, Liu M, Yang J (2019) Stochastic processes shape microeukaryotic community assembly in a subtropical river across wet and dry seasons. Microbiome 7:138. 10.1186/s40168-019-0749-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. De Vargas C, Audic S, Henry N, Decelle J, Mahé F, Logares R, Lara E, Berney C, Le Bescot N, Probert I, Carmichael M, Poulain J, Romac S, Colin S, Aury J-M, Bittner L, Chaffron S, Dunthorn M, Engelen S, Flegontova O, Guidi L, Horák A, Jaillon O, Lima-Mendez G, Lukeš J, Malviya S, Morard R, Mulot M, Scalco E, Siano R, Vincent F, Zingone A, Dimier C, Picheral M, Searson S, Kandels-Lewis S, Tara Oceans Coordinators, Acinas SG, Bork P, Bowler C, Gorsky G, Grimsley N, Hingamp P, Iudicone D, Not F, Ogata H, Pesant S, Raes J, Sieracki ME, Speich S, Stemmann L, Sunagawa S, Weissenbach J, Wincker P, Karsenti E, Boss E, Follows M, Karp-Boss L, Krzic U, Reynaud EG, Sardet C, Sullivan MB, Velayoudon D (2015) Eukaryotic plankton diversity in the sunlit ocean. Science 348:1261605. 10.1126/science.1261605 [DOI] [PubMed] [Google Scholar]
  11. Edgar RC (2013) UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat Methods 10:996 [DOI] [PubMed] [Google Scholar]
  12. Egge E, Elferink S, Vaulot D, John U, Bratbak G, Larsen A, Edvardsen B (2021) An 18S V4 rRNA metabarcoding dataset of protist diversity in the atlantic inflow to the arctic ocean, through the year and down to 1000 m depth. Earth Syst Sci Data 13:4913–4928. 10.5194/essd-13-4913-2021 [DOI] [Google Scholar]
  13. Gregori I, Martino F, Zane L, Pallavicini A, Vezzi A (2025) Filtering microbial diversity: evaluation of different pore sizes for <sc>eDNA</sc> community profiling. Environ DNA 7:e70221. 10.1002/edn3.70221 [DOI] [Google Scholar]
  14. Gu Y, Meng D, Liu Z, Zhang M, Yang Z, Yin H, Liang Y, Xiao N (2024) Biotic and abiotic properties mediating sediment microbial diversity and function in a river–lake continuum. Front Microbiol 15:1479670. 10.3389/fmicb.2024.1479670 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Haro-Moreno JM, López-Pérez M, De La Torre JR, Picazo A, Camacho A, Rodriguez-Valera F (2018) Fine metagenomic profile of the Mediterranean stratified and mixed water columns revealed by assembly and recruitment. Microbiome 6:128. 10.1186/s40168-018-0513-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. He Q, Chen B, Deng K-K, Yan P, Lin G-J, Chen Q-W, Chen Y-P, Guo J-S (2026) The relative roles of cell size and phylogeny in driving dissimilarity of algal functional trait. Bioresour Technol 445:134144. 10.1016/j.biortech.2026.134144 [DOI] [PubMed] [Google Scholar]
  17. Hu W, Zhang H, Lin X, Liu R, Bartlam M, Wang Y (2022) Characteristics, biodiversity, and cultivation strategy of low nucleic acid content bacteria. Front Microbiol 13:900669. 10.3389/fmicb.2022.900669 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hu W, Zheng N, Zhang Y, Bartlam M, Wang Y (2023a) Spatiotemporal dynamics of high and low nucleic acid-content bacterial communities in Chinese coastal seawater: Assembly process, co-occurrence relationship and the ecological functions. Front Microbiol 14:1219655. 10.3389/fmicb.2023.1219655 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hu W, Zheng N, Zhang Y, Bartlam M, Wang Y (2023b) Spatiotemporal dynamics of high and low nucleic acid-content bacterial communities in Chinese coastal seawater: assembly process, co-occurrence relationship and the ecological functions. Front Microbiol 14:1–17. 10.3389/fmicb.2023.1219655 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hu W, Zheng N, Zhang Y, Li S, Bartlam M, Wang Y (2024) Metagenomics analysis reveals effects of salinity fluctuation on diversity and ecological functions of high and low nucleic acid content bacteria. Sci Total Environ 933:173186. 10.1016/j.scitotenv.2024.173186 [DOI] [PubMed] [Google Scholar]
  21. Ji D, Zhang J, Li F, Li W, Bi L, Li W, Fu Y, Wang Y (2025) Divergent driving mechanisms shape the temporal dynamics of benthic prokaryotic and eukaryotic microbial communities in coastal subtidal zones. Microorganisms 13:1050. 10.3390/microorganisms13051050 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Li T, Liu G, Yuan H, Chen J, Lin X, Li H, Yu L, Wang C, Li L, Zhuang Y, Senjie L (2023a) Eukaryotic plankton community assembly and influencing factors between continental shelf and slope sites in the northern South China Sea. Environ Res 216:114584. 10.1016/j.envres.2022.114584 [DOI] [PubMed] [Google Scholar]
  23. Li W, Lu Q, Alharbi SA, Soromotin AV, Kuzyakov Y, Lei Y (2023b) Plant–soil–microbial interactions mediate vegetation succession in retreating glacial forefields. Sci Total Environ 873:162393. 10.1016/j.scitotenv.2023.162393 [DOI] [PubMed] [Google Scholar]
  24. Liu J, Hao Z, Ma L, Ji Y, Bartlam M, Wang Y (2016) Spatio-temporal variations of high and low nucleic acid content bacteria in an exorheic river. PLoS One 11:e0153678. 10.1371/journal.pone.0153678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liu H-Q, Li H-J, Li S-C, Xiang Y-Z (2024a) Differentiation strategies for microbial community assembly and their associations with soil functions in lemon farmland. Appl Soil Ecol 201:105493. 10.1016/j.apsoil.2024.105493 [DOI] [Google Scholar]
  26. Liu Q, Jia J, Hu H, Li X, Zhao Y, Wu C (2024b) Nitrogen and phosphorus limitations promoted bacterial nitrate metabolism and propagation of antibiotic resistome in the phycosphere of Auxenochlorella pyrenoidosa. J Hazard Mater 468:133786. 10.1016/j.jhazmat.2024.133786 [DOI] [PubMed] [Google Scholar]
  27. Liu J, Zhang L, Zou H, Shi H, Wang J, Xia F, Yang Y (2025) Cell-size‐dependent responses of bacterial communities to warming in the alpine grasslands of the Qinghai‐Tibet plateau. Glob Change Biol 31:e70285. 10.1111/gcb.70285 [DOI] [PubMed] [Google Scholar]
  28. Ma G, Logares R, Xue Y, Yang J (2022a) Does filter pore size introduce bias in DNA sequence-based plankton community studies? Front Microbiol 13:969799. 10.3389/fmicb.2022.969799 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mo Y, Peng F, Gao X, Xiao P, Logares R, Jeppesen E, Ren K, Xue Y, Yang J (2021) Low shifts in salinity determined assembly processes and network stability of microeukaryotic plankton communities in a subtropical urban reservoir. Microbiome 9:128. 10.1186/s40168-021-01079-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Nemergut DR, Schmidt SK, Fukami T, O’Neill SP, Bilinski TM, Stanish LF, Knelman JE, Darcy JL, Lynch RC, Wickey P, Ferrenberg S (2013) 2013. Patterns and Processes of Microbial Community Assembly. Microbiol Mol Biol Rev 77:342–356. 10.1128/MMBR.00051-12 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Okamoto S, Takenaka M, Tojo K (2022) Seasonal modifications of longitudinal distribution patterns within a stream: interspecific interactions in the niche overlap zones of two ephemera mayflies. Ecol Evol 12:e8766. 10.1002/ece3.8766 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pettersen JP, Gundersen MS, Almaas E (2021) Robust bacterial co-occurence community structures are independent of r- and K-selection history. Sci Rep 11:23497. 10.1038/s41598-021-03018-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Proctor CR, Besmer MD, Langenegger T, Beck K, Walser J-C, Ackermann M, Bürgmann H, Hammes F (2018) Phylogenetic clustering of small low nucleic acid-content bacteria across diverse freshwater ecosystems. ISME J 12:1344–1359. 10.1038/s41396-018-0070-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Santos M, Oliveira H, Pereira JL, Pereira MJ, Gonçalves FJM, Vidal T (2019) Flow cytometry analysis of low/high DNA content (LNA/HNA) bacteria as bioindicator of water quality evaluation. Ecol Indic 103:774–781. 10.1016/j.ecolind.2019.03.033 [DOI] [Google Scholar]
  35. Sleith RS, Katz LA (2022a) Illuminating protist diversity in pitcher plants and bromeliad tanks. PLoS One 17:e0270913. 10.1371/journal.pone.0270913 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Sun L, Wu L, Liu X, Huang W, Zhu D, Wang Z, Guan R, Liu X (2023) Reducing the risk of benthic algae outbreaks by regulating the flow velocity in a simulated south–north water diversion open channel. Int J Environ Res Public Health 20:3564. 10.3390/ijerph20043564 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wang X, Liu J, Chai B, Wu T (2025) Protozoa-driven micro-food webs shaping carbon and nitrogen cycling in reservoir ecosystems. Environ Microbiome 20:109. 10.1186/s40793-025-00772-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Woodcock S, Head IM, Nee S, Curtis TP, Lunn M, Sloan WT (2006) Quantifying the roles of immigration and chance in shaping prokaryote community structure. Environ Microbiol 8(4):732–740 [DOI] [PubMed] [Google Scholar]
  39. Xie N, Wang Z, Hunt DE, Johnson ZI, He Y, Wang G (2022) Niche partitioning of labyrinthulomycete protists across sharp coastal gradients and their putative relationships with bacteria and fungi. Front Microbiol 13:906864. 10.3389/fmicb.2022.906864 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Xiong W, Chen T, Du X, Hou L, Chen Y, Han J-L, Zhan A (2025) Ecological impacts of treated effluent on multitrophic biodiversity and their interactions. Environ Res 277:121585. 10.1016/j.envres.2025.121585 [DOI] [PubMed] [Google Scholar]
  41. Zhang X, Ye N, Liang C, Mou S, Fan X, Xu J, Xu D, Zhuang Z (2012) De novo sequencing and analysis of the ulva linza transcriptome to discover putative mechanisms associated with its successful colonization of coastal ecosystems. BMC Genomics 13:565. 10.1186/1471-2164-13-565 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zhang H, Huang X, Huang L, Bao F, Xiong S, Wang K, Zhang D (2018) Microeukaryotic biogeography in the typical subtropical coastal waters with multiple environmental gradients. Sci Total Environ 635:618–628. 10.1016/j.scitotenv.2018.04.142 [DOI] [PubMed] [Google Scholar]
  43. Zhang H, Hu W, Liu R, Bartlam M, Wang Y (2023) Low and high nucleic acid content bacteria play discrepant roles in response to various carbon supply modes. Environ Microbiol 25:3703–3718. 10.1111/1462-2920.16539 [DOI] [PubMed] [Google Scholar]
  44. Zhang K, Chen C, Long D, Wang S, Li J, Lin X (2024a) Unveiling sedimentary microbial communities: Restored mangrove wetlands show higher heterogeneity and network stability than natural ones. Ecol Indic 166:112274. 10.1016/j.ecolind.2024.112274 [DOI] [Google Scholar]
  45. Zhang G, Guo Z, Ke Y, Li H, Xiao X, Lin D, Lin L, Wang Y, Liu J, Lu H, Hong H, Yan C (2024b) Comparative analysis of size-fractional eukaryotic microbes in subtropical riverine systems inferred from 18S rRNA gene V4 and V9 regions. Sci Total Environ 953:175972. 10.1016/j.scitotenv.2024.175972 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1 (708KB, docx)

Data Availability Statement

No datasets were generated or analysed during the current study.

Articles from Archives of Microbiology are provided here courtesy of Springer

RESOURCES