An open question regarding mixoplankton diversity is how their unique evolutionary histories have shaped cell metabolism and physiology of extant protist lineages (Kim and Maruyama, 2014; Mansour and Anestis, 2021). We are gaining a better mechanistic understanding of why and how phago-mixotrophy is maintained in contemporary aquatic systems through insights into the genetic and cellular integration of photosynthetic symbionts/organelles and phago-heterotrophic hosts, and by deciphering the cryptic ancestral genetic signatures of phagotrophy. Such insights have facilitated a better understanding of both non-constitutive (NCM) and constitutive (CM) cultures through use of transcriptomics and proteomics (Hattenrath-Lehmann et al., 2021; Li et al., 2021a; Koppelle et al., 2022; Johnson et al., 2023). Efforts focusing on genetic commonalities among diverse mixoplankton may also reveal conserved traits retained from early phagotrophic ancestors, while instances of gene loss/acquisition can shed light on traits that have enabled the diversification of younger lineages. Ongoing culture work suggests a mixotrophy continuum, ranging from mostly photosynthetic groups (e.g., prasinophytes, haptophytes) to mostly phagotrophic (e.g., dictyochophytes) (Li et al., 2022; Edwards et al., 2023a), but with a high degree of variability within lineages and even among species (Calbet et al., 2011; Wilken et al., 2020).

The various omics approaches serve as powerful tools for investigating the evolutionary trajectories of protists, the biochemical underpinnings of mixotrophic behavior, and potentially even resource allocations between phototrophic and phagotrophic states. Gene-based models using reference databases of protistan genomes or transcriptomes can be leveraged to identify core metabolic features of phago-mixotrophy in lab cultures (Burns et al., 2018; Bock et al., 2021; Koppelle et al., 2022) and in field populations (Lambert et al., 2022; Alexander et al., 2023), enabling explorations into trophic modes of previously unexamined lineages. Comparative genomics can be particularly valuable in examining patterns of plastid gene loss or retention in lineages of mixoplankton (Dorrell et al., 2019), and transcriptomics and proteomics can furthermore shed light on the regulation of genes among closely related mixotrophs differing in trophic abilities (Lie et al., 2018). However, only a small subset of mixoplankton have been cultured in the lab, sequenced, and available in reference annotation databases which models are trained on, limiting our ability to infer phago-mixotrophy in the field based on sequences. Single-cell sorting approaches, in which mixotrophic status can be empirically ascertained and combined with single-cell amplicon, genomic or transcriptomic sequencing, offers a promising way to address this issue (Wilken et al., 2019; Needham et al., 2022). Furthermore, strengthening reference sequence databases with the inclusion of more cultured representatives and supplementing with metagenome assembled genomes (MAGs) derived from the natural environment (Delmont et al., 2022; Alexander et al., 2023) could expand our ability to identify mixoplankton taxa and metabolism. Once robust phylogenies are constructed, phylogenetic modeling can help us to infer likely gains and losses of function and understand how mixotrophic life histories affected rates of diversification (reviewed in Morlon, 2014).

The richest body of historical research on mixoplankton has been focused on characterizing traits related to phagotrophy and photosynthesis and in identifying trophic interactions. The vast majority of these studies have used rate-based measurements to document photosynthesis and ingestion, and have helped to provide a foundational framework for understanding the occurrence of these traits among mixoplankton. Less well understood, however, are how mixotrophic behavior differs among diverse eukaryotic lineages, both as a function of the environment and due to physiological adaptations. Microevolution is a valuable lens through which to explore phago-mixotrophy, including epigenetics and other regulatory processes. Such approaches could deepen our understanding of how external conditions shape phago-mixotrophy phenotype and may help to predict mixoplankton functional biology and trophic status. For instance, how is the ecophysiology of different species related to epigenetic processes that modulate mixotrophic traits, and how do these relationships vary with ecosystem type? Such networks between resource limitation and trait expression are likely highly diverse and involve some combination of the above scenarios and would need to be broadly surveyed across phylogenetic lineages, types of mixoplankton, and across diverse ecosystems.

While the application of microevolutionary approaches such as epigenetic mechanisms or documenting the distribution and frequency of alleles across populations remain in their infancy for marine microeukaryotes, it may hold promise for understanding interactions between short-term acclimation and long-term adaptation strategies of mixoplankton. For example, DNA modifications and small non-coding RNAs may function as epigenetic mechanisms promoting diversification and driving phenotypic plasticity in microbial eukaryotes (Weiner and Katz, 2021), and could facilitate adaptation in response to environmental pressures. Phenotypic demonstrations of microevolutionary processes in mixoplankton have recently been documented with observations of cultures losing their capacity to feed due to long-term maintenance without prey (Blossom and Hansen, 2021), and in documenting metabolic and behavioral changes in response to warming temperatures (Wilken et al., 2013; Ferreira et al., 2022; Lepori-Bui et al., 2022). Understanding genetic processes that underlie such regulatory changes are needed, and studies are underway by using “experimental evolution” approaches for deciphering molecular dynamics of long-term environmental change.

In recent years an increase in the use of omics, experimental, and modeling approaches for characterizing mixotrophic metabolism and trophic modes across resource gradients has provided a wealth of new insights into mixoplankton physiology and biogeography. This information will also facilitate progress in deciphering evolutionary processes that have shaped lineage diversity of protistan phago-mixotrophs and can be used to validate existing theoretical approaches. Historically, adaptive dynamics models have been used to understand the circumstances under which mixotrophy is evolutionarily viable. These models simulate evolution as a series of small mutations in an organism’s (functional) trait(s) and are used to identify evolutionarily stable strategies (ESSs) that are robust to invasion by slightly different strategies (Diekmann, 2003). Such models have been used to identify environmental conditions under which mixotrophs should specialize (Troost et al., 2005b, 2005a), to evaluate a mixotroph’s optimal balance between phagotrophy and phototrophy as a function of environmental temperature (Gonzalez et al., 2022), and to understand the adaptiveness of kleptoplasty (Brown et al., 2023). These theoretical predictions can be tested with new data on the distribution (and plasticity) of mixotrophs in nature, and better constrained by data on the phylogeny of mixotrophs and their traits.

The fundamental constraints on mixotrophic metabolisms will likely be revealed by investigating key traits related to the three routes of resource acquisition: photosynthesis, nutrient uptake, and phagotrophy (Andersen et al., 2015). Although seemingly straightforward, trade-offs have been hard to characterize (Dolan and Pérez, 2000; Mitra et al., 2023b). The diverse metabolic strategies of mixoplankton complicate the interpretation of trade-offs, for example, not all mixoplankton can take up inorganic nutrients or survive in darkness (Schoener and McManus, 2017; Lie et al., 2018). Methodological challenges also hamper our ability to quantify these traits. For example, while it is relatively easy to quantify chlorophyll a content or photochemical yield of photosystem II (Fv/Fm) as a proxy for investment in light harvesting capacity, it is less straightforward to quantify the relative investments in nutrient uptake and phagotrophy. Furthermore, attempting to tease apart the investments into different nutrient acquisition modes in the field adds another layer of difficulty. Here we propose the application of single-cell techniques in situ (Beisner et al., 2019; Wilken et al., 2019) and traditional physiological measurements obtained in the lab to start estimating these traits across environmental gradients both within and across species. While we call for in situ approaches throughout this paper, we also acknowledge the importance of laboratory-based work with cultures to advance mixoplankton research. These can then be used to better inform trait-based models and predict the large-scale impact of mixotrophy ( Figure 1 ).

Mixoplankton have the metabolic flexibility to modulate the different resource acquisition modes and thus must trade between investing in different cellular and metabolic “machinery”. Flow cytometric methods can potentially provide cell-level detail on an organism’s investment in photoautotrophy and phagotrophy, while distinguishing among co-occurring populations in a natural community (González-Gil et al., 1998; Anderson et al., 2017; Wilken et al., 2019). Specifically, the amount of autofluorescence from plastids can be compared with the fluorescence of acidic food vacuoles stained with acidotropic dyes, such as LysoTracker, in both laboratory and field samples (e.g., Anderson et al., 2017). As mentioned in Box 1 , the latter approach is not free of caveats, but the promise of effective trait quantification at the single cell level makes it worth understanding potential biases better (Wilken et al., 2019). If applied successfully, quantification of traits related to photoautotrophy and phagotrophy, can be complemented by those related to nutrient uptake. For example, alkaline phosphatase activity can be assessed by flow cytometry and might serve as a proxy for investments into phosphorus acquisition (González-Gil et al., 1998).

Accurate volume measurements of organelles are another way to quantify traits. In order to switch investments between strategies, mixoplankton must allocate cellular space for different structures. Novel techniques such as 3D subcellular imaging can provide detailed information on the space occupied by plastids and feeding vacuoles, with the potential to determine the metabolic flexibility of different processes (Colin et al., 2017; Uwizeye et al., 2021). As an example, these tools can be used to investigate trade-offs in CMs related to maximizing inorganic nutrient uptake through larger surface area to volume (SA/V) ratios (to accommodate nutrient transporters) versus maximizing ingestion through larger cellular biovolume (smaller SA/V ratios) occupied by feeding vacuoles. By integrating these measurements with models based on resource allocation theory, in which the mixoplankton need to build different structures to optimize the investment of carbon and nitrogen to different functions (Berge et al., 2017), it is possible to validate model predictions and generate new hypotheses about mixotrophy trade-offs ( Figure 1 ).

Uptake kinetics have long been used to characterize trade-offs, especially across species, for both strict autotrophic and heterotrophic protists (Hansen et al., 1997; Edwards et al., 2012). In mixoplankton, the matter is complicated by potential interactions between different resource acquisition pathways. For example, do mixoplankton with higher photosynthetic rates have lower affinity for prey uptake? Recent studies provide initial insights into how these trade-offs play out among nanoplankton (Edwards et al., 2023a, 2023b; Barbaglia et al., 2024). When performed both across and within different mixotrophic species, these analyses can help to reveal the relative roles of intraspecific phenotypic plasticity on setting mixotrophy trade-offs. So far generating these types of data relies on experimentation with cultured isolates. However, stable isotope approaches with high phylogenetic resolution such as RNA based stable isotope probing (RNA-SIP) allow probing resource acquisition by mixoplankton directly within natural communities (Wilken et al., 2023). While RNA-SIP can at best be semi-quantitative, more targeted approaches such as nanoSIMS or Chip-SIP (Mayali et al., 2012) are also more quantitative. If employing isotope sources for both auto- and heterotrophic metabolism, these might also provide information on trade-offs. Such datasets can then be targeted to test modeling predictions, such as the potential correlation between mixotrophic metabolism and cell size (Chakraborty et al., 2017). On the other hand, trait correlations derived from empirical data can be used to inform mechanistic models; they can be used to define mixotrophic trade-offs in models that resolve community and ecosystem dynamics at different scales (Ward and Follows, 2016; Serra-Pompei et al., 2019). Specifically, if two traits are correlated, it is possible to derive parameters that describe this relationship and use it to constrain trait-based models ( Figure 1 ), as it is commonly done for the allometric relationships resolving resource uptake kinetics for autotrophs and heterotrophs (Ward and Follows, 2016).

Quantification of mixoplankton traits in situ is a first step toward understanding trade-offs. Observed trait correlations might not be true trade-offs (a correlation does not equal causation) or the lack of empirically observed correlations does not exclude the possibility of a real trade-off. Therefore, if trait correlations are observed that likely reflect trade-offs, the mechanistic basis of such trade-offs needs to be understood. Such underlying mechanisms can be of different kinds. For instance, resource limitation might cause trade-offs as resources allocated to one function cannot be used for another. However, even in the presence of plentiful resources, trade-offs might arise through biophysical constraints. This includes limited cell volume or cell surface to accommodate all cellular machineries required for photosynthesis, phagotrophy and nutrient uptake, and all other cellular functions. Finally, trade-offs can arise in the interaction with other species, such as competitors, predators, or parasites. Trait correlations identified with the methods described in section 3.1 can generate hypotheses about underlying mechanisms, which will in turn determine the choice of methods required to test them.

To understand trade-offs associated with limited resource availability, the energetic costs and elemental investments involved in building cellular structures such as plastids versus digestive vacuoles should be quantified. A combination of traditional measurements of macromolecular composition, stoichiometry, and physiological rates (e.g. growth, photosynthesis, and ingestion) with the omics toolbox can be powerful to quantify the biomolecular structures and associated costs invested into each strategy. For example, proteomic approaches applied to cultured isolates in the lab can be used to link changes in strategy following an environmental perturbation to the respective costs of phenotypic adjustments (Zhang et al., 2022). Proteome models work hand-in-hand with the methods mentioned above allowing for assessment of different proteins, incorporating energetic, space, and stoichiometric constraints, and optimizing resource allocation to maximize growth across different environmental conditions (Molenaar et al., 2009; McCain et al., 2021; Leles and Levine, 2023). Such modeling frameworks can reveal new trade-offs as well as the mechanisms underpinning trait trade-offs ( Figure 1 ). In contrast, one might choose a bottom-up approach in systems biology to simulate whole systems from (incomplete) genomes. Genome-scale metabolic models are already applied in the model microalga Chlorella to predict mixotrophic metabolism during utilization of dissolved organic carbon (Zuñiga et al., 2016) but would require further development to capture utilization of chemically much more complex prey organisms.

The optimization of the nutritional balance in mixoplankton might also be determined by constraints other than resource availability. These may be specific to certain taxonomic groups and only become apparent by comparison across species or broader taxonomic units differing in morphology. An example is the relationship between flagellar arrangements found in nano-sized protists and the flow fields they generate, which in turn affect rates of prey encounter and ingestion (Nielsen and Kiørboe, 2021). Such relationships can be resolved by video-microscopy and particle flow field analysis. A previous study suggested that flow fields created by some haptophytes do not support sufficiently high rates of prey ingestion for growth through feeding alone (Dölger et al., 2017), making the mixotrophs reliant on photosynthesis. Various techniques of live microscopy will remain important for observing the behavior of individual cells and integrating these observations with models accounting for hydrodynamics at the relevant scale.

Finally, a trade-off can arise due to direct interference of a trait with species interactions (e.g. predation defense or competition ability). Potential examples include prey searching behavior or the generation of a feeding current that can also attract predators and thus creates a trade-off between optimizing prey capture and avoiding predation (Kiørboe and Thomas, 2020). There could also be synergies for instance in mixoplankton that produce toxins to defend themselves from predators, immobilize their prey and/or inhibit competitors (Berge et al., 2012; Traboni et al., 2020). Assessing such trade-offs requires variation in the presence of interacting species, so they can best be investigated experimentally by manipulating i) the availability of prey, ii) the trophic landscape including pressure from parasites, viral infection, or predators, or iii) the presence of competitors for the same prey particles or dissolved nutrients. Gradually adding complexity to controlled experiments with forced species interactions should allow for insight into in situ community conditions and the external factors that might dictate costs and benefits for a range of traits.

Ocean basin-scale transects can be used to examine the presence and activity of mixoplankton across large gradients of light, temperature and nutrients. Mixoplankton are able to respond to changing environmental conditions by modulating the photosynthetic and phagotrophic modes of nutrition (González-Olalla et al., 2019). When active mixoplankton are a prominent part of the protist community, a relevant number of organisms are simultaneously being producers and consumers at the bottom of the food web. This has the potential to fundamentally alter food web dynamics by increasing trophic efficiency and the size of organisms (Sanders, 1991; Ward and Follows, 2016). Latitudinal transects can thus reveal the ecological niches of mixoplankton, providing new insights into when the mixoplankton community is more likely to be functioning as producers or consumers.

There are a few on-going decadal transect studies, such as the International Global Ship-based Hydrographic Investigations Program (GO-SHIP; www.go-ship.org) where biological assessments are now included on some US ship based transects (Bio-GO-SHIP; www.biogoship.org). Along these basin-scale transects there are opportunities to monitor and examine the presence and abundance of mixoplankton using tools, such as flow and imaging cytometry at the surface using flowthrough seawater systems as well as at discrete depths. In addition, the pilot Bio-GO-SHIP program includes both eukaryotic metagenomic and transcriptomic data collection and may be analyzed for both mixotrophic presence and activity through comparative genomics. By using these transects, insight to the variability of mixoplankton abundance across different light, temperature and nutrient regimes can be assessed. Similarly, there are long research expeditions that collect plankton and molecular data that may be used to assess mixotrophy, for example, the TARA Oceans Expedition (Faure et al., 2019; Lambert et al., 2022), the Malaspina Expedition in 2010 as well as other cruises that traverse oceanic basins (e.g. Atlantic Meridional Transects).

As a part of basin-scale oceanic transects, a combination of targeted ship-board incubations and observations would be ideal. The development of routine field-based assays for assessing mixotrophic activity would be valuable tools in incorporating oceanic transits using ships involved in both oceanographic research or commercial shipping. These routine measurements could involve either periodic sampling or continuous auto-sampling for incubations coupled with FLP grazing incubations or isotope probing. Such sampling could be preserved for later analysis and could include assessment of genetic diversity. When feasible, incorporating observations that require more immediate processing, such as cell imaging, flow cytometry, or omics approaches, would greatly enhance the potential inference and scope of routine large scale sampling efforts. To conduct basin-scale transects, several suggestions can be considered. First, it is beneficial to sample at different depths within a particular location, as environmental gradients are also associated with depth. Additionally, it is essential to target traditionally under-sampled regions across the globe, such as the Indian Ocean, South Atlantic, South Pacific, and the Southern Ocean, as most data comes from easily accessible coastal marine or freshwater systems. By focusing on these areas, which have received relatively less scientific attention, we can fill critical knowledge gaps into the unique dynamics and biodiversity of mixoplankton in these ecosystems.

Modeling approaches can also be applied to investigate the ecological niches of mixoplankton. For example, Edwards (2019) combined data from studies that targeted in situ mixotrophic nanoflagellates across the global ocean with a competition model that described mixotrophs and their specialized autotroph and heterotroph competitors. However, the dataset compiled was limited in terms of spatial and temporal coverage and similar datasets are largely absent for other mixoplankton functional types. As an alternative, theoretical models that describe different mixoplankton functional types can be used to run simulations under idealized scenarios of environmental stress to evaluate competitive abilities (Leles et al., 2018; Anschütz and Flynn, 2020), with the drawback of a large number of unconstrained parameters. Large-scale models that resolve physics such as 1D (vertical resolution; water column models) and 3D (vertical and horizontal resolution) models can capture variation in, for example, temperature and nutrient concentrations that more closely reflect the real environment. Such models can then be applied to investigate the biogeography of mixotrophy at regional (Lin et al., 2018; Li et al., 2022) and global (Ward and Follows, 2016) scales, and help guide future empirical research. In addition, they can be applied to investigate species shifts and their biogeochemical roles in future climate scenarios (Dutkiewicz et al., 2021). Alternatively, Species Distribution Models (SDMs) can be powerful tools when long-term time series are available. However, their predictive skill to investigate spatial shifts of microbial taxa under future climate change scenarios was found to be low when applied to data from the Continuous Plankton Recorder program (Brun et al., 2016), which emphasizes the need for more targeted surveys.4.2. How will the seasonal cycle of mixoplankton presence and activity respond to a changing climate?

The seasonal cycle of phytoplankton and zooplankton has been well studied across aquatic systems and climate regimes for decades, particularly through long-term monitoring stations that collect plankton data alongside other environmental factors. However, mixoplankton have largely been an overlooked part of the plankton community, limiting our ability to understand drivers of plankton temporal patterns and further predict community response to climate change (Schneider et al., 2021). One way to address this research gap is through reanalyzing long-term monitoring of plankton microscopy and imaging data. Species/genera within these datasets can be classified as mixoplankton (including different types such as CMs and NCMs) based on previous experimental evidence for mixotrophy in peer-reviewed literature (Mitra et al., 2023a). An ideal dataset to analyze the seasonal cycle of mixoplankton is one with enough data to account for intra- and inter-annual variability in a given location to discern the average annual cycle and allow for statistical analysis of environmental conditions associated with variability. While long-term monitoring datasets with plankton taxa data are not common, many exist that would allow for reanalysis of mixoplankton abundance. Examples of these long-term monitoring locations include: Martha’s Vineyard Coastal Observation (Hunter-Cevera et al., 2016), the L4 station in the Western English Channel (Widdicombe et al., 2010), the Narragansett Bay Long-Term Plankton Time Series in Rhode Island (Smayda, 1998), Helgoland Roads in the North Sea (Wiltshire et al., 2010), German Long-Term Ecosystem Research Network site Lake Müggelsee (Gsell et al., 2016), and multiple locations through the French national phytoplankton monitoring network (Hernández Fariñas et al., 2015). Furthermore, NSF Long Term Ecological Research (LTER) sites such as Northeast U.S. Shelf (Fowler et al., 2020) and Northern Gulf of Alaska (Batten et al., 2018) collect phytoplankton data that could be recategorized as mixoplankton. One major drawback to simply reclassifying taxa in datasets as mixoplankton is that there is no confirmation that a specific species or genus is a mixoplankton at a location or that it was actively engaging in mixotrophy at the time of collection. This could lead to an overestimate of mixoplankton abundance and proportion.

Despite these drawbacks, historic records contain a plethora of data that can be used to rapidly expand our understanding of mixoplankton spatiotemporal distribution and associated environmental determinants. Finding a way to constrain what species/genera in a given dataset are mixoplankton and under what conditions or time of year they are most likely utilizing mixotrophy (vs. strict photoautotrophy or strict phagotrophy) would likely improve the accuracy of any abundance estimations. One approach that might be used to help identify mixoplankton within previously collected samples is BrdU labeled prey experiments. Repeated BrdU labeled bacteria experiments can help to develop a list of known mixoplankton taxa in a region where a long-term dataset exists and help identify at what time of the year or under what conditions these taxa are actively engaging in mixotrophy (Dobbertin da Costa et al., 2024). On its own, this data could expand our understanding of mixoplankton’s biogeography, however, when combined with historical datasets, this data can exponentially expand our estimations of active mixoplankton’s temporal and spatial variability. Another possible approach is a combination of flow and imaging cytometry in combination with FLP grazing incubations or isotope probing, as described in section 4.1.

Despite being able to revisit long-term time series and identify mixoplankton species, we cannot retrieve information on the relative contributions of photosynthesis and phagotrophy to their metabolism; presence of a known mixoplankton does not indicate the importance of both nutrient acquisition modes. Mathematical models that resolve carbon and nutrient fluxes and mechanistically represent plankton growth based on environmental conditions can provide insights into the potential metabolic strategy of mixoplankton over the seasonal cycle (Berge et al., 2017; Ghyoot et al., 2017; Leles et al., 2021). These models can be compared against mixoplankton biomass and other environmental variables derived from long-term time series datasets and, in turn, we can analyze the emergent metabolic strategy obtained from the simulations. For example, phagotrophy was found to be important for mixoplankton within the nanoplankton size spectrum during summer due to nutrient limitation but also during winter due to carbon limitation (Leles et al., 2021). It is noteworthy that, although a powerful tool, models are limited by our current understanding of mixoplankton trade-offs (see section 3). It is thus imperative that new datasets are generated to quantify modern in situ seasonal cycles of mixotrophs simultaneously engaging in phototrophy and phagotrophy and their associated environmental conditions. This would be a challenging endeavor that requires some combination of methods, such as uptake of fluorescent particles, LysoTracker and flow cytometry, stable isotopes, and molecular techniques (Anschutz et al., 2024). However, if activity cycles can be quantified at the same locations as long-term monitoring datasets, then it could be possible to hindcast activity using environmental and presence data.

Once the annual cycle of mixoplankton presence and/or activity are better understood, mathematical or statistical modeling can be used to investigate how mixoplankton will respond to climate induced shifts in the environment. Species distribution models (SDMs) can be used in combination with outputs from Earth System Models (ESMs; e.g., sea surface temperature, mixed layer depth, pH, chlorophyll, PAR, etc.) to evaluate shifts in the environmental envelope of mixotrophic taxa (sensu Barton et al., 2016). While SDMs are powerful for evaluating changes to the suitable habitat/environment for individual taxa, they are unable to simulate feedbacks between plankton groups and ocean biogeochemistry. In this case, mathematical models that simulate nutrient fluxes and the mechanisms driving plankton growth can be applied instead (see section 5.2).

A key parameter that needs further investigation is the extent to which phagocytized prey are assimilated by CMs into lipid, nucleic acid, protein, or carbohydrate biomass for growth versus egested, expelled, or respired. Quantifying the amount of carbon fixed through photoautotrophy compared to the carbon assimilated from prey would improve the accuracy of carbon flux predictions in modeling (Ward and Follows, 2016). Similarly, for NCMs, what percentage of assimilated carbon originates from acquired plastid photosynthetic activity versus ingested particles? Figure 2 illustrates possible intracellular and extracellular fates of consumed prey, and the potential impacts these fluxes might have on surrounding organisms. For example, variability in the mixoplankton exometabolome across different trophic conditions may influence surrounding bacteria and archaea dependent on dissolved organic matter, and other nearby phytoplankton that are competing for nutrients. Furthermore, mixoplankton likely present different intracellular biochemistry compared to phytoplankton which may impact fitness of zooplankton feeding on them (Traboni et al., 2020, 2021), see section 5.2. Importantly, such biochemical parameters are likely different for CMs versus NCMs and across different environmental conditions such as light regimes (Fischer et al., 2022).

New and emerging methods including high-energy imaging, labeling, and cell-sorting, will allow scientists to track the fate of ingested particles in order to understand the impact of mixotrophy on small-scale chemical changes to large-scale carbon and nutrient cycling. To illuminate the small-scale cellular impacts of mixotrophy, methods that quantify and locate intracellular elemental composition (including C, N, P), such as single-cell stable isotope probing using nanoSIMS (Mayali, 2020), energy dispersive X-ray microanalysis (Norland et al., 1995; Segura-Noguera et al., 2012, 2016) and synchrotron–based X-ray fluorescence microprobe (Twining et al., 2003) could provide insight about the effect of mixotrophic metabolism on mixoplankton stoichiometry. Meanwhile, quantifying the mass balance and stoichiometry of extracellular protist egesta is challenging (e.g., Nagata, 1996) but is needed to track the fate of ingested prey. In culture, stable and radioisotope tracer methods can track elements from labeled prey to cellular and extracellular material. In the field, the contributions of mixoplankton to the complex extracellular dissolved matter pool is much more challenging and would require isolation of protist species by flow cytometry or microscopy for both intracellular and extracellular analyses. To identify mixotrophic cells, the operator would have to target cells based on their size, chlorophyll fluorescence, and the presence of a vacuole (via vacuole staining). While the presence of ingested prey could also be used to identify mixotrophs (based on FLP also analyzed for their elemental composition), the contribution of prey to the mixotroph’s elemental composition would be difficult to tease apart in field samples.

Stable isotope probing of biomolecules or single cells is an emerging tool that can help track the activity of specific microbes in complex environments (Alcolombri et al., 2022). For example, Terrado et al. (2017) quantified mixotrophic carbon and nitrogen acquisition in Ochromonas sp. strain BG-1 by single-cell stable isotope probing using nanoSIMS. Applied to natural environments, such approaches can help estimate carbon and nutrient fluxes in food webs. Future applications would improve our understanding of the role of mixotrophy in elemental fluxes along the continuum of mixotrophic strategies. However, it is important to acknowledge that these methods are time intensive, expensive and require specialized instrumentation and trained personnel making them unsuitable for studies conducted over global scales. Therefore, future advancements in methodology should prioritize enhancing quantification accuracy at the community level across larger scales.

Method advancement is needed to overcome challenges associated with moving from controlled laboratory culture studies to in situ assessments of the intra- and extracellular biochemistry of mixed communities that engage in a variety of trophic modes. Complications could arise in natural abundance isotope measurements and stoichiometry assessments during attempts to attribute heterotrophic metabolism to phagotrophy if a mixoplankton is simultaneously incorporating significant amounts of dissolved carbon through osmotrophy (Godrijan et al., 2020, 2022), which can also alter the biochemistry (Cecchin et al., 2018) and isotope composition (Estep and Hoering, 1981; Zhang et al., 2009; Cormier et al., 2022). Additionally, it is important to note that the methods discussed here require a sample concentration step, whereby microorganisms are typically collected on a filter membrane, or retained by it to prepare concentrated liquid samples for flow cytometry cell sorting of low-abundance samples (Berthelot et al., 2019, 2021; Duhamel et al., 2019). A major issue is that an unknown, but likely large, fraction of protists are fragile, and cells can rupture even when using gentle filtration with either vacuum, syringe or peristaltic pumps. This creates a significant bias when trying to determine important parameters in the likely majority of mixotrophs in both laboratory and field studies. Methodological improvements are therefore needed to overcome this issue, which would open exciting new applications, especially in field studies where cell abundances are relatively low.

It is hypothesized that the combined use of photosynthesis and phagotrophy will enhance trophic transfer efficiency of carbon and nutrients and bolster secondary production (Ward and Follows, 2016; Moorthi et al., 2017). However, the value (or detriment) of mixoplankton to higher trophic levels in the aquatic food web remains to be extensively tested (but see Traboni et al., 2020, 2021), and the nutritional quality of mixoplankton is debated (Traboni et al., 2021; Vad et al., 2021). Mixoplankton are theorized to be of superior food quality because they maintain more stable intracellular C:N:P ratios compared to phytoplankton despite environmental variability (Katechakis et al., 2005; Moorthi et al., 2017) and synthesize essential fatty acids and sterols at a higher concentration than strict heterotrophs (Boëchat et al., 2007). The stability of mixoplankton as a food source is complicated by variability in trophic plasticity (see section 3) and innate ability to produce chloroplasts (Mitra et al., 2016). Therefore, the biochemical composition of mixoplankton taxa, and thus viability as a food source, likely differs across the various mixoplankton functional groups and environmental gradients (see section 4). Though it remains difficult to estimate the relative contribution of photosynthesis versus phagotrophy to intracellular budgets, it is likely that trophic transitions between nutritional modes are accompanied by changes in the fatty acid and sterol signature of mixoplankton (Calderini et al., 2022). This is likely to be especially complex in NCMs that have varying capacities for acquired photoautotrophy (Johnson, 2011; Leles et al., 2017).

While protist nutritional value has been considered in relation to grazer health and productivity (Jónasdóttir, 1994; Jónasdóttir et al., 2002), baseline biochemical and stoichiometric profiles for mixoplankton taxa in relation to trophic strategy are lacking (Ma et al., 2022). It is necessary to start categorizing the C:N:P ratios, fatty acid, and sterol compositions of mixoplankton functional groups under different environmental conditions so we can begin to properly understand their quality as prey items and nutrient stability under abiotic and biotic stress (e.g., nutrient, light, and prey limitation). Traditional approaches to understanding predator-prey relationships, such as trophic markers (fatty acid composition, natural abundance amino acid ¹⁵N/¹⁴N ratios) and stable isotope probing can be employed to explore how mixotrophic ingestion processes can transform prey nutrients, but interpretation of results will also require basal profiles of said prey items (e.g., heterotrophic bacteria, phytoplankton) across varying environmental conditions. The nutritional quality of mixoplankton can be deduced by elemental analysis and stable isotope probing has been used to assess the nutritional profile of phytoplankton (Grosse et al., 2017). However, as described in 5.1, applying this approach in situ will require isolation and concentration of mixoplankton.

A more complete understanding of the nutritional quality of mixoplankton requires in situ studies that assess a combination of life history traits of micro- and meso-zooplankton grazers in response to varying prey items. Most studies offer mixotrophic prey in comparison to strict nutritional specialists (Jónasdóttir, 1994; Jónasdóttir et al., 2002; Jones et al., 2002). We suggest future work that quantifies fecundity (e.g. egg production, hatching success) and growth efficiency of zooplankton predators in conjunction with a single prey species in monoculture reared in conditions along the mixotrophic spectrum or varying proportions of phagotrophy and photoautotrophy ratios (Traboni et al., 2021). Given the potential for toxicity (Hiltunen et al., 2012; Tang et al., 2020) or inedibility (Vad et al., 2020) among mixoplankton, we also see a need for prey-preference assays (as in Castellani et al., 2008; Parrish et al., 2012), but focused on mixoplankton and their varying nutritional value as prey. Advances in flow cytometric sorting and imaging allow for discrimination between potential mixotrophic prey items in situ, which could allow for studies of prey selection in mixed assemblages, though the biochemical composition of mixoplankton as prey is largely understudied.

As new data become available on mixoplankton palatability, biochemistry, and trophic transfer efficiency, they can be used to parameterize ecosystem models. These mathematical models simulate abiotic and biotic components of marine ecosystems, linking nutrient availability to primary production and grazing (e.g., Darwin; Follows et al., 2007), and, in some cases, higher trophic levels (e.g., COBALT; Stock et al., 2014). To do so in a computationally tractable manner, these models typically collapse plankton diversity into fewer dimensions (i.e. functional groups and size classes) in order to represent the feedbacks between different trophic groups and carbon/nutrient cycles (Ward and Follows, 2016; Dutkiewicz et al., 2020; Leles et al., 2021; Serra-Pompei et al., 2022). Models vary in the degree of functional complexity they represent, and in the geographic scale modeled, from regional (Leles et al., 2021) to global (Ward and Follows, 2016). Although not all ecosystem models include representations of mixotrophy, those that do have already demonstrated the importance of mixotrophy to carbon cycling (Ward and Follows, 2016). However, these models’ accuracy hinges on the quality of input data. Thus, information on mixotroph palatability can directly affect trophic linkages built into models, mixoplankton stoichiometry can be used to estimate trophic transfer efficiency, and so on.