Transparent exopolymer particles (TEPs), a class of organic aggregates that can be stained with Alcian Blue (Alldredge et al., 1993; Passow, 2002a), are produced mainly by phytoplankton and bacteria in the upper ocean (Passow, 2002a). They play important roles in the regulation of material cycling, food webs, and the transport of pollutants, including spilled oils and microplastics (Passow, 2002a; Mari et al., 2017; Passow et al., 2017; Galgani et al., 2019; Nagata et al., 2021). Owing to their sticky nature (Rochelle-Newll et al., 2010), TEPs may act as a glue to facilitate the aggregation of living and non-living particles (Quigg et al., 2016; Mari et al., 2017), contributing to the formation of large, sinking aggregates that mediate the vertical transport of organic carbon in the ocean (Engel et al., 2004; Passow and Carlson, 2012). Transparent exopolymer particles may also accumulate in the sea-surface microlayer and influence air–sea gas exchange (Wurl and Holmes, 2008; Galgani et al., 2016; Jennings et al., 2017). Clarifying the dynamics of TEPs and its regulation in marine environments is fundamentally important for a better understanding of oceanic biogeochemical cycles.

The TEP dynamics in marine environments have been studied using colorimetric and microscopic methods (Harlay et al., 2009). Colorimetry is used to examine the distribution of bulk TEPs and its relationships with environmental variables. Such studies have revealed that bulk TEP distributions are affected by several factors, including primary production (Mari and Burd, 1998), chlorophyll a (Chl a) concentration (Ortega-Retuerta et al., 2017), phytoplankton community composition (Taylor et al., 2014), bacterial abundance (Busch et al., 2017), bacterial production (Ortega-Retuerta et al., 2017), and other physical and chemical conditions (e.g., temperature, pH, and salinity) (Piontek et al., 2009; Thornton, 2009; Najdek et al., 2011). Although these data are useful, colorimetry cannot resolve the distinctive dynamics of different types of TEPs. The TEPs in natural seawater are a complex mixture of various polymeric particles with different physicochemical characteristics (Gogou and Repeta, 2010), origins (Passow, 2002b; Bhaskar et al., 2005; Li et al., 2015), and development stages (Passow et al., 1994). In fact, early microscopy studies revealed that there are morphologically distinctive types of TEPs in seawater, including filaments, sheets, clumps, and amorphous blobs (Alldredge et al., 1993; Passow and Alldredge, 1994; Passow, 2002a; Berman, 2010). “Fresh” TEPs, assembled from freshly released TEP precursors, are barely associated with any other types of particles, such as living cells (Passow et al., 1994; Passow, 2002a), whereas “mature” TEPs are associated with various other types of particles (Kiørboe, 2003). Other studies have revealed that morphologically distinctive types of TEPs are produced during the incubation of natural bacterial assemblages (Sugimoto et al., 2007). In Sagami Bay, based on observations of TEP formation during bacterial growth in seawater, Sugimoto et al. (2007) found that the bacteria formed tightly packed TEPs composed of bacterial clusters and capsule-like polymers (Linley and Field, 1982; Radić et al., 2006). Different types of TEPs may play different roles in biogeochemical cycles and food webs. Fresh TEPs are C-rich with a high C:N ratio (Engel and Passow, 2001) and might contribute to the upward transport of organic carbon due to their low densities (Azetsu-Scott and Passow, 2004). In contrast, “matured” TEPs associated with high-density ballast particles may contribute to the downward transport of organic carbon (Nagata et al., 2021). Bacterial cluster-like TEPs can be a N-rich food for grazers due to low bacterial C:N ratio being around 6 (Fukuda et al., 1998). Thus, the microscopic examination of different types of TEPs may help to elucidate the complex behavior and internal dynamics of TEPs in marine environments. However, because microscopic analysis is labor intensive, data concerning the seasonal and vertical variations of different types of TEPs in marine water columns are limited (Sugimoto et al., 2007; Bar-Zeev et al., 2011; Busch et al., 2017; Engel et al., 2020).

Here, we examine the seasonal and vertical variability of the abundance and size distributions of different types of TEPs in Sagami Bay, Japan, to investigate the possible environmental factors controlling the dynamics of each type of TEP. Using the microscopy, we distinguished three types of TEPs. Type I: Polysaccharide matrices made of fibrillar TEP precursors without attached or embedded particles. Type II: Complex aggregates associated with biotic (algae and bacteria) and abiotic (detritus and minerals) particles. Type III: Bacterial clump-type TEPs characterized by densely packed bacterial assemblages embedded in polysaccharide matrices (Sugimoto et al., 2007). Our data demonstrate that the abundance and distribution of particle sizes among TEP subgroups are dynamic parameters that can help evaluate processes involved in regulating TEP dynamics in coastal environments.

Transparent exopolymer particle (TEP) number has been reported to be in the order of 10⁵–10⁷ particles L^–1 in various marine environments, including in Monterey Bay, United States (Passow and Alldredge, 1994), off the coast of Peru (Engel and Galgani, 2016), in the Indian Ocean (Kumar et al., 1998), in Kattegat, Denmark (Mari and Kiørboe, 1996; Mari and Burd, 1998), and in Otsuchi Bay, Japan (Ramaiah et al., 2001). The TEP numbers in Sagami Bay (4.0 × 10⁴–3.0 × 10⁶ particles L^–1) were generally within this range, except for some low values (10⁴ particles L^–1) found in winter (February and March). Because we used 2.0-μm pore-size filters instead of 0.2- or 0.4-μm pore-size filters, which have often been used in previous studies (Passow, 2002a), our estimates of TEP numbers are conservative.

Our results reveal the seasonal dynamics of TEPs in Sagami Bay. The high TEP abundance in April indicates that phytoplankton released a large amount of TEPs during a spring bloom. This result is consistent with previous observations in other coastal environments (Mari and Kiørboe, 1996; Ramaiah et al., 2001). Transparent exopolymer particle size (mean diameter) displayed seasonal variability, with the largest sizes being observed in April, especially in the subsurface layer (20 m) below the pycnocline. At the time of the April sampling, the low transparency (high turbidity) was associated with two consecutive days of precipitation preceding the sampling date (precipitation on April 11 and 12 was 42 and 66 mm, respectively; Odawara Weather Station, Japan Meteorological Agency, 2021). Conceivably, the loading of land-derived inorganic particles mediated by river runoff facilitated the coagulation and subsequent transfer of TEPs to the subsurface layers (Nagata et al., 2021). Our data also reveal that TEPs were abundantly produced during the stratification period in Sagami Bay. High TEP abundances were observed in June and August when nutrients were depleted in the mixing layer. This indicates that phytoplankton and bacteria under nutrient limitation actively produce TEPs and TEP precursors (Mari et al., 2005; Bar-Zeev et al., 2011; Deng et al., 2016).

The size distribution slopes (β values), which reflect the geometric features (fractal dimension, FD) of TEPs at steady state (Mari and Kiørboe, 1996), for individual TEP types and total TEPs were within the range of 2.6–4.2. Similar β ranges have been found in coastal environments in Denmark (Kattegat), where β values varied from 2.6–3.4 over a year (Mari and Burd, 1998). β did not display systematic variations among months and TEP types, except that there was a positive correlation between β and N_d10μm for Type I TEPs, the latter being an indicator of the extent of particle load (Buonassissi and Dierssen, 2010). The positive correlation partly reflects the fact that both β and N_d10μm were relatively low in winter (February and March) and high in summer (June and July) (Supplementary Figure 2 and Supplementary Table 2). Assuming a steady-state particle size distribution, the FD of Type I TEPs in winter (February and March) and summer (June and July) can be estimated to be 0.3–0.8 and 2.3–2.4, respectively, using an equation derived from dimensional analysis (FD = 2β− 5; for sheer coagulation; Jiang and Logan, 1991). Fractal dimension describes a geometric feature (the spatial structure) of the aggregates; a higher FD indicates more tightly packed aggregates (Logan and Wilkinson, 1990). Therefore, our results suggest that Type I TEPs were more tightly packed in summer (higher particle load) than in winter (lower particle load). Alternatively, the variability in β may reflect the non-steady-state conditions of the TEPs, involving processes other than coagulation (e.g., grazing), or both (Mari and Kiørboe, 1996).

In seawater culture experiments using samples collected in Sagami Bay, Sugimoto et al. (2007) found that TEPs densely colonized by bacteria, which correspond to the Type III TEPs in this study, accumulated abundantly during bacterial growth. Because the seawater was prefiltered through a 0.8-μm pore-size filter to eliminate phytoplankton and grazers, this accumulation of TEPs was ascribed to bacterial TEP production, bacterial enhancement of TEP precursor coagulation, or both. The authors suggested that bacteria-driven TEP production was potentially substantial in Sagami Bay, renewing 9–37% of TEP standing stocks per day. Despite the potentially high production rates of Type III TEPs in Sagami Bay, the results of the present study reveal that Type III TEPs generally only accounted for a small fraction of the total TEPs in terms of area, although their contribution in terms of number reached 19–34% of the total TEP number throughout the year. These results suggest that, in the natural environment of Sagami Bay, which differs from the conditions in the seawater cultures used by Sugimoto et al. (2007), Type III TEPs rapidly coagulate with other particles to form Type II TEPs. Similarly, Type I TEPs may also rapidly coagulate with other particles to form Type II TEPs. Thus, Type III and Type I TEPs likely represent transient phases of TEP development, characterized by high turnover and relatively low abundance in terms of area. This model is partly consistent with the results showing a decrease in the relative abundance of Type II TEPs (area) in February, when the Chl a concentration was low. It is likely that the particles, such as algae and detritus, available for coagulation with Type I and Type III TEPs were less abundant and that the frequency of collisions among all particles, including TEPs, decreased in February, leading to a decrease in the formation of Type II TEPs (i.e., the coagulation of Type I and Type III TEPs with other particles).

Our analysis of relationships between TEPs and environmental parameters revealed that the highest correlations were found between TEPs (number and area) and bacterial abundance for all types of TEPs, as well as for total TEPs. Of note, regardless of TEP type, the correlation coefficients between TEPs and bacteria were always higher than those between TEPs and Chl a (Table 3). This, combined with previous results showing active TEP production by bacteria in Sagami Bay (Sugimoto et al., 2007), suggests that our correlation results reflect the important role that bacteria play in the regulation of TEP dynamics in Sagami Bay. Consistent with our results, previous studies have found that TEP concentrations increase with increasing bacterial abundance and production in marine environments (Bar-Zeev et al., 2009; Ortega-Retuerta et al., 2009; Ortega-Retuerta et al., 2010; Yamada et al., 2017). Our results add to the previous proposition by demonstrating that such relationships hold for different types of TEPs. A conceptual model to explain bacterial roles as a driver of TEP dynamics is presented in Figure 6. Both bacteria and phytoplankton can produce TEP precursors (Passow, 2002a), which can drive the formation of Type I TEPs. This process is likely mediated by the spontaneous formation of nanogels (Chin et al., 1998; Verdugo et al., 2004) and their subsequent aggregation (Burd and Jackson, 2009). In addition to the role as a producer of TEP precursors, bacteria can also contribute to the formation of Type III TEPs (Sugimoto et al., 2007). Type I and Type III TEPs can aggregate with various particles suspended in seawater (phytoplankton, bacteria, and organic and inorganic detritus), leading to the formation of Type II TEPs. Although the largest pool of TEPs in Sagami Bay was usually Type II TEPs, Type I and Type III TEPs are considered to play pivotal roles as intermediate pools of TEP transitions. The low relative abundance of Type III TEPs, despite the fact that Type III TEPs were abundantly produced in seawater cultures at rates potentially able to replace the TEP standing stock in <10 days (Sugimoto et al., 2007), suggests that Type III TEPs are a transient component with a high turnover time. Obviously, our model does not capture the complexity of TEP dynamic regulation in marine environments. Bacterial production is fueled by dissolved organic matter produced by phytoplankton, mediated by grazer and virus activities (Cole et al., 1988; Nagata, 2000), implying that bacterial TEP production is ultimately limited by phytoplankton production. In fact, an experimental study revealed that TEP formation is intimately controlled by bacteria–phytoplankton interactions (Van Oostende et al., 2013). An additional level of complexity arises from the fact that bacteria might act not only as TEP producers but also as decomposers of TEPs (Ortega-Retuerta et al., 2017; Nagata et al., 2021). Furthermore, activities of grazers and viruses might affect TEP dynamics in a complex manner (Bhaskar and Bhosle, 2005; Yamada et al., 2018). Future studies are required to better understand the role of bacteria, phytoplankton, other food web components, as well as interactions among them, in the regulation of the internal dynamics of TEPs.

We examined seasonal and vertical variations in the abundance of different types of TEPs in Sagami Bay. The TEP assemblages were generally dominated by Type II TEPs in terms of area, except in February, when the contribution of Type I TEPs was high. Type III TEPs contributed significantly to the total TEPs in terms of number, although their contributions in terms of area were low because of their small particle sizes. Apparently, Type I and Type III TEPs are rapidly incorporated into the Type II TEPs during the productive period in the bay. The results of our TEP size distribution analyses revealed that the size distribution slopes, which reflect the geometric features of TEPs, of each type of TEP changed dynamically with season. The abundance of each type of TEP was significantly positively correlated with bacterial abundance, suggesting that bacteria play an important role in the regulation of TEP dynamics in Sagami Bay. We proposed a conceptual model to explain the major role of bacteria as a driver of the internal dynamics of TEPs.

Important areas for future research include elucidation of the biogeochemical role of each type of TEP in marine systems. Currently, there are divergent views on the role of TEPs in the transfer of carbon and pollutants, such as microplastics, in marine environments. One view emphasizes that TEPs facilitate particle aggregation and enhance downward transport of materials, especially when TEPs are aggregated with dense particles (ballasts) (Passow, 2002a; Mari et al., 2017). The other perspective stresses that TEPs are porous and less dense and enhance the retention and ascending flux of TEPs in the upper water (Azetsu-Scott and Passow, 2004; Jennings et al., 2017; Mari et al., 2017). Although the current model generally assumes TEPs to be a homogeneous pool (Passow, 2002a; Wurl et al., 2011; Mari et al., 2017; Nagata et al., 2021), TEPs in fact consist of distinctive and interacting pools of TEP subgroups, as revealed by the present study (Figure 6). Models assuming TEPs to be a homogeneous pool might be compromised if different TEP subgroups behave differently in transfer processes. Therefore, our results highlight the importance of investigating TEP internal dynamics and controls to improve our understanding of the role of TEPs in carbon and pollutant transfer in marine environments.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

KS, HF, and IK designed the study plan. HF collected samples. KS conducted TEP analyses. KS, HF, and TN analyzed the data. KS, HF, IK, and TN wrote the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.