Unravelling Diatom–Microbiome Dynamics in the Red Alga Gelidium Corneum (Florideophyceae, Rhodophyta)
Begoña Sánchez-Astráin, María Borrego-Ramos, Raquel Viso, Camino F. de la Hoz, Saúl Blanco, José A. Juanes
TL;DR
This study explores how the diatom communities on the red alga Gelidium corneum change with seasons and locations, showing that environmental factors strongly influence their diversity and structure.
Contribution
The study is the first to examine diatom–microbiome dynamics in Gelidium corneum, revealing seasonal and environmental influences on community variation.
Findings
Diatom community structure was more strongly shaped by seasonal shifts than by location-specific factors.
Diatom diversity was higher in autumn, while spring samples showed lower diversity but more distinct community structures across populations.
Environmental variables, except irradiance, significantly shaped diatom distribution.
Abstract
The microbiome plays a crucial role in host health, a recognition that has grown in recent years across many organisms. These host–microbiome associations may not be uniform at the host-species level; instead, they could differ among geographically separated populations, potentially reflecting adaptation to local conditions. In the Cantabrian Sea, rising seawater temperatures are shifting the distribution of many macroalgae species. Among them, Gelidium corneum, a key habitat-forming red alga, is showing a reduction in biomass, yet remains present in this coastal region. Given the wide geographic separation of its populations, it is plausible that G. corneum harbours distinct diatom-dominated microbial communities signalling local adaptation. In this study, we characterized the epibiotic diatom communities associated with G. corneum across three populations spanning 340 km along the…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5- —Universidad de Cantabria
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMarine and coastal plant biology · Microbial Community Ecology and Physiology · Algal biology and biofuel production
Introduction
Marine macroalgae are major contributors to primary production in inshore ecosystems and play a fundamental role in supporting marine life and ecological interactions. In temperate to cold-water regions, large seaweeds underpin the formation of marine forests, creating structurally complex habitats that provide numerous niches for diverse marine organisms [1]. Their thalli also offer living substrata that facilitate attachment, access to light, and exposure to nutrient-rich water currents, promoting interactions between hosts and epibionts [2]. Consequently, epiphytism is a common colonization strategy in benthic communities, especially on rocky substrates where competition for space is intense.
Among the most abundant and diverse epiphytes on macroalgae are algae themselves, ranging from unicellular diatoms and dinoflagellates to larger macrophytes. Filamentous algae often dominate due to their short life cycles, simple morphology, and high photosynthetic rates [3]. Within the microbiota, diatoms typically form the most diverse and abundant surface-associated communities. Epiphyte distribution shows marked temporal and spatial variation, with peak development from spring to late summer [4]. Among microbiomes, seaweed–bacteria relationships are the best studied and are usually mutualistic, with certain bacterial species enhancing macroalgal growth [5], reducing biofouling [6], and supporting reproduction [7].
Diatom–macroalgal associations are widespread in marine ecosystems. Pelagic diatoms account for roughly 40% of oceanic primary production [8], while benthic forms contribute up to 60% of primary production in habitats such as seagrass beds [9]. On macroalgal hosts, diatom assemblages are shaped by host morphology, surface texture, tissue age, seasonality, and local environmental conditions [10, 11]. Although these relationships are generally neutral or competitive due to resource overlap [12], macroalgae can supply dissolved organic carbon and phosphorus to diatoms in nutrient-poor environments [13].
Gelidium corneum (Hudson) J.V.Lamouroux, 1813, is a perennial red alga that usually forms dense, canopy-like meadows in sublittoral communities along the northeastern Atlantic coast [14], from France to Morocco. Growth peaks in spring and summer due to increased light, favouring epiphytic colonization, including macroalgae such as Dictyota dichotoma and bryozoans like Electra pilosa [15, 16]. In autumn and winter, storm-induced water movement causes branch loss, resulting in a decrease in both the standing stock of G. corneum and its epiphytic load [17]. Although its macrobiota is well studied, its microbiomes remain largely unexplored, with diatom associations serving as a key example of this gap.
Along the northern Iberian coast, a thermal gradient has historically shaped the distribution of seaweed species in both intertidal and subtidal zones [18]. In the easternmost regions, where waters are warmer during the summer, G. corneum populations have recently shown greater thermal tolerance compared to those in cooler areas [19]. Such environmental and phenotypic variability may also influence the structure of associated microbiomes. For instance, a study conducted in the Mediterranean Sea on the seagrass Posidonia oceanica revealed differences in epiphytic diatom assemblages among genetically distinct host populations [20]. This finding suggests the existence of host-specific interactions between different species within the diatom genus Cocconeis and P. oceanica genotypes, although the underlying ecological mechanisms remain to be elucidated.
Despite the recognized importance of microorganisms in marine ecosystems, our understanding of how microbial communities interact with macroalgae and influence their adaptability to climate change remains limited. To address this gap, current research efforts are increasingly focused on elucidating how small-scale interactions between macroalgae and their associated microbiota may contribute to host resilience. In line with this, we address the following questions: (1) Is there variation in epibenthic diatom community composition among different G. corneum populations? If not, (2) do seasonal, geographical, or local factors shape the composition of these communities across different populations? Finally, (3) to what extent do environmental variables explain observed microbiome patterns?
By addressing these questions, we aim to characterize the microbiome composition associated with thermally adapted G. corneum populations along the northern Iberian coast [19] across multiple environmental gradients, including seasonal, depth-related, and regional variation. This study provides the first report on the distribution of epiphytic diatoms on G. corneum and examines how geographical distance influences diatom community structure.
Methods
Sampling Site and Plant Material
The northern coastline of the Iberian Peninsula, along the southern Bay of Biscay, extends for more than 600 km from the French border (1°W) to the Galician coast (9°W) (Fig. 1). Sea surface temperature (SST) exhibits a marked west–east thermal gradient and notable seasonal variations. During winter, SST are relatively uniform, averaging 12–13 °C. In summer, the western coast records mean temperatures of about 19.5 °C, although upwelling events can temporarily lower them to around 16 °C. The eastern coast shows greater variability, with mean summer SST reaching up to 21.5 °C (OCLE; [21]).Fig. 1. Map of the northern coast of Spain showing the locations where Gelidium corneum samples were collected: Asturias (A), Cantabria (C), and Basque Country (BC)
Thalli of well-developed Gelidium corneum were collected in autumn 2022 and spring 2023 at two depths (5 and 12 m) in three locations: Asturias (A; 43°33′43″ N, 6°07′29″ W), Cantabria (C; 43°23′50″ N, 4°23′17″ W), and the Basque Country (BC; 43°20′49″ N, 1°54′20″ W). Autumn sampling was performed in November in Asturias and in December in Cantabria and the Basque Country, with the latter delayed by adverse meteorological conditions limiting site accessibility. Spring sampling was conducted uniformly across all locations in April. These sites represent the typical summer temperature gradient, and depth differences mainly reflect variation in light. Three replicate samples were collected by scuba diving at each depth, spaced 2–3 m apart. Healthy individuals were trimmed from the base of the main frond, placed in 1-L containers, and, once at the surface, seawater was removed and samples were immersed in 70% ethanol and kept refrigerated in the dark. In the laboratory, samples were stored at 4 °C until analysis.
Laboratory Processing and Diatom Characterization
The methodology for diatom identification followed the protocol of Borrego-Ramos et al., [22]. Then, samples were shaken for two minutes to detach diatoms from G. corneum, after which the macroalgal material was removed. Each sample was then divided for separate molecular and morphological analyses. This dual approach takes advantage of the complementary strengths of morphological and molecular methods. Morphological analysis enables the identification of visually distinguishable taxa, while molecular techniques reveal cryptic or low-abundance species that might not be detected microscopically.
For the morphological analysis, sediments obtained after shaking were allowed to settle for 24 h and centrifuged (5000 rpm, 5 min). The pellet was oxidized with hydrogen peroxide (5 mL, 12 h in a sand bath) to remove organic matter, followed by washing and treatment with hydrochloric acid to eliminate carbonates. After two rinses with distilled water, frustules were mounted on permanent slides using Naphrax^®^. Observations were carried out under bright-field light microscopy (Olympus BX60), with digital micrographs taken of representative frustules. A total of 400 diatom valves were counted per sample to assess the species composition. For the molecular analysis, at least 200 mL of sample was centrifuged at 4000 rpm for 5 min at 4 °C. Total DNA was extracted using the PowerSoil^®^ DNA Isolation Kit (Qiagen, formerly MoBio). The plastid rbcL marker was amplified by PCR using an equimolar cocktail of five diatom-specific primers [23, 24]: three forward (Diat_rbcL_708F_1, _2, _3) and two reverse (Diat_rbcL_R3_1, _2), which included Illumina adapters. For each sample, three PCR replicates were performed and subsequently pooled. Libraries were purified with Mag-Bind^®^ RXNPure Plus magnetic beads, quantified with a Qubit™ dsDNA HS Assay, and sequenced on an Illumina MiSeq PE300 platform by AllGenetics (A Coruña, Spain). Raw sequencing reads were processed using the DADA2 pipeline implemented in R, following the diat.barcode package [25], which is specifically designed for diatom metabarcoding. Quality filtering included the removal of low-quality reads, dereplication, error model learning, and inference of amplicon sequence variants (ASVs). Chimeric sequences were identified and removed using the consensus method implemented in the diat.barcode package. All steps were performed using the default parameters of the package. Only ASVs with confident taxonomic assignment against the Diat.barcode v7 reference database were retained. Sequences identified only at family level and singletons were excluded from subsequent analyses.
Molecular quantification was initially based on ASV read abundances. To reduce quantification bias related to differences in diatom cell size and gene copy number, read counts were corrected using a biovolume-based correction factor following Vasselon et al., [24]. Corrected read counts were then converted into relative abundances (%) for each taxon within each sample.
For the integrative analysis, relative abundances obtained from the molecular dataset and from the morphological counts were calculated independently and subsequently averaged, assigning equal weight (50:50) to both approaches. This strategy aims to minimize the limitations inherent to each method when used alone, combining the taxonomic resolution of metabarcoding with the quantitative robustness of light microscopy.
Environmental Data
Environmental data were obtained from satellite databases. Data corresponding to the month preceding the sampling date were downloaded, and 30-day time-weighted averages (TWA) were calculated.
Seawater temperature (ST) and nutrient concentrations (phosphate, PO₄; nitrate, NO₃; and silicate, Si) were obtained from the Global Ocean Biogeochemistry Analysis and Forecast products, using daily averages from the 5.07 m and 11.40 m vertical layers (spatial resolution: ~27.8 km for nutrients and ~ 9.3 km for ST) [26].
Depth-specific light availability was derived from MODIS-Aqua satellite data, using daily averaged products (spatial resolution 4 km) (NASA, [27]). Specifically, we used ocean surface photosynthetically active radiation (PAR) and the diffuse attenuation coefficient at 490 nm (Kd490) for downwelling irradiance. Light at each depth (z) was computed following the approach described by Assis et al., [28]. Their approach estimates bottom light using a standard depth-dependent exponential function based on PAR and Kd490:
\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:Light\:at\:bottom\:=\:PAR\:x\:exp(-Kd490\:x\:z)$$\end{document}Where,
PAR: Photosynthetically Active Radiation in µmol photons m^− 2^ s^− 1^.
Kd490: diffuse attenuation coefficient at 490 nm in m^− 1^.
z: depth in m.
Wave exposure was characterized using significant wave height (Hs), derived from a global wave dataset generated by the WaveWatch III model and based on the National Centers for Environmental Prediction (NCEP) and the National Center for Atmospheric Research (NCAR) reanalysis of winds and ice fields (Global Ocean Waves, GOW, [29]) (spatial resolution ~ 13.9 km).
Statistical Analysis
To evaluate differences in diatom composition among G. corneum populations across spatial and temporal scales, we applied a suite of multivariate approaches. Non-metric multidimensional scaling (nMDS) was used to represent the similarity structure of diatom communities in a reduced-dimensional space, with environmental variables incorporated as extrinsic factors to aid interpretation. Hierarchical clustering and the resulting groups (cut at a 0.4 threshold) were superimposed on the nMDS plots.
Analysis of similarities (ANOSIM) tested for significant differences in diatom community among the three factors tested: season, depth, and region. For factors showing significant effects in ANOSIM, similarity percentage analysis (SIMPER) identified the species contributing most to the observed dissimilarities. All the above analyses were carried out using the vegan package [30]. Indicator species analysis (ISA) was used to determine which diatom species were strongly associated with specific factors, employing the indicspecies package [31].
A Mantel test assessed correlations between community composition (abundance-based dissimilarity matrix) and both environmental and geographic distance matrices. Pairwise correlation analyses (Spearman method) explored interrelationships among environmental variables. Both analyses were conducted with the vegan package [30]. All statistical tests were performed at a significance level of 0.05, and analyses and visualizations were carried out in R (v4.3.3; R Core Team, [32]).
Results
Characterization of Diatom Communities
A higher number of taxa at the family, genus, and species levels were identified using the molecular approach compared to the morphological identification based on light microscopy (LM). The raw data obtained from both methods are presented in Fig. 2.Fig. 2. Venn diagrams showing the percentage of overlap between molecular and morphological raw data for epibiotic diatoms of Gelidium corneum at the family, genus, and species levels
After merging both methodological datasets and removing singletons, a total of 268 taxa identified to the species level and 94 to the genus level were recorded. In autumn 2022, 84 genera were identified, 29 of which were exclusive to this season, while in spring 2023, 65 genera were recorded, including 10 unique to spring. Overall, 55 genera were shared between seasons, representing 58.5% of the total genera observed. The dominant genera in autumn were Achnanthidium, Navicula, Staurosira, Pteroncola, and Nitzschia, whereas in spring, the assemblage was dominated by Arcocellulus, Licmophora, Navicula, Pteroncola, and Minidiscus (Supplementary 1).
The Shannon Index (H’) and species Richness (S) varied primarily with season, both peaking in autumn (Fig. 3). During this period, diversity was consistently high across all sites and depths, with H’ values ranging from approximately 2.8 to 3.3 and S reaching its maximum at the Basque Country 12 m site. In spring, both metrics declined markedly, especially in the Basque Country population, where H’ dropped to approximately 1.2–1.4 and S fell below 35 taxa. Images of species identified using the morphological approach via LM are shown in the Supplementary 2.Fig. 3. Shannon Index (H’) and Richness values (S) (mean ± SE; n = 3) of diatom assemblages identified in the three Gelidium corneum populations at different depths (5 and 12 m) and seasons (autumn and spring). Asturias (A; western-population), Cantabria (C; middle-population), and Basque Country (BC; eastern-population)
Diatom Community Structure and Ordination
Non-metric multidimensional scaling (nMDS) ordination (R² = 0.97; stress = 0.084) showed clear clustering of diatom communities across sites (Fig. 4). Nutrients (phosphate, PO₄; nitrate, NO₃; and silicate, Si) and photosynthetically active radiation (PAR) had the strongest influence in spring, whereas significant wave height (Hs) and seawater temperature (ST) were more important in autumn. Autumn samples clustered tightly, indicating greater similarity among populations, while spring samples were more dispersed, reflecting higher spatial variability in community composition.Fig. 4. Non-metric multidimensional scaling (nMDS) of diatom communities based on relative abundances and associated environmental variables in three Gelidium corneum populations at different depths (5 and 12 m) and seasons (autumn and spring). Asturias (A; western-population), Cantabria (C; middle-population), and Basque Country (BC; eastern-population). Circles indicate hierarchical clusters defined at a dissimilarity threshold of 0.4
Analysis of similarities (ANOSIM) results indicated a strong and significant effect of season on diatom community composition (R = 0.7315, p < 0.01; Table 1). Depth showed a negative R value (R = −0.1537, p = 0.9262), reflecting high variability within depth groups, while region, although not significant (R = 0.2106, p = 0.1075), exhibited a low positive R, showing some weak spatial structuring.Table 1. Results of the analysis of similarities (ANOSIM; Bray-Curtis dissimilarity) testing the effects of season, depth, and regionFactorR**p-valueSeason0.73150.0017 **Depth−0.15370.9262Region0.21060.1075R ANOSIM statistic, R; * Indicates p-value < 0.05; ** indicates p-value < 0.01; *** indicates p-value < 0.001
Following ANOSIM results, the similarity percentage analysis (SIMPER) identified the taxa that contributed most to the differences between seasons (Table 2). In autumn, the dissimilarity was primarily driven by three taxa resulting in a cumulative contribution of 16.3%., while in spring was five taxa reached a cumulative contribution of 47.7%. These results show that spring differences were driven by a few dominant taxa, whereas in autumn contributions were more evenly distributed.Table 2. List of taxa ordered by their contribution to similarity into seasons and the cumulative contribution (Cum. Contrib.), according to results of the similarity percentage analysis (SIMPER). Taxa whose contribution is higher than 1.75% are shownAutumnSpringAchnanthidium minutissimum 12.07Arcocellulus mammifer 13.97Hyalosira cf. delicatula 2.42Licmosoma squamosum 12.13Staurosira sp. 1.78Navicula sp. 9.08Pteroncola inane 6.94Minidiscus trioculatus 5.58Cum. Contrib.: 16.28Cum. Contrib.: 47.71
Indicator species analysis (ISA) results showed that diatom distributions were mainly shaped by season and region, with no significant depth effects. In total, 74 species were associated with seasons, 61 with autumn and 13 with spring. Regionally, 13 species showed significant associations: 4 with Asturias, 5 with Cantabria, 3 with the Basque Country, and 1 shared by Asturias and Cantabria (Table 3). A full list of seasonal ISA results is provided in Supplementary 3.Table 3. Results of the indicator species analysis (ISA) by region. The table lists the species identified as significant indicators along with their indicator value statistic (Stat)RegionSpeciesStatp-valueAsturiasMastogloia sp.0.8230.0059Opephora sp.0.8130.0196Navicula* sp.0.7850.0178Navicula normaloides*0.7530.0059Basque CountryMinidiscus trioculatus0.7790.0119Neosynedra provincialis*0.6330.0129Thalassiosira profunda0.4980.0058**CantabriaTryblionella coarctata0.8270.0116**Nitzschia fontifuga0.8240.0062Nitzschia lanceolata var. minima0.7120.0295Thalassiosira* sp.0.6950.0320Cocconeis scutellum* var. posidoniae0.6690.0496Asturias + CantabriaRhoicosphenia* sp.0.7440.0225 Indicates p-value < 0.05; ** indicates p-value < 0.01; *** indicates p-value < 0.001
Environmental Drivers of Diatom Community Distribution
The Mantel test revealed that environmental variables significantly influenced diatom community composition (r = 0.689, p < 0.001; Table 4). The strongest associations were with ST (r = 0.713), NO₃ (r = 0.778), PO₄ (r = 0.636), and Si (r = 0.702), all p < 0.001. Hs had a weaker but significant effect (r = 0.26, p < 0.05), while PAR showed no significant influence (r = 0.08, p = 0.206). Geographic distance did not correlate with community differences (r = −0.007, p = 0.408).Table 4. Results of the Mantel test (Bray-Curtis dissimilarity) assessing correlations between diatom community composition and environmental variables. ST: Seawater temperature, Hs: Significant wave height, PAR: Photosynthetically active radiation, NO₃: Nitrate, PO₄: Phosphate, Si: SilicateVariabler**p-valueST0.713< 0.001 ***Hs0.2590.0231 *PAR0.0860.2066NO_3_0.778< 0.001 ***PO_4_0.636< 0.001 ***Si0.702< 0.001 ***All variables combined0.689< 0.001 *Geographical distance−0.0070.4089r Mantel statistic, r; * Indicates p-value < 0.05; ** indicates p-value < 0.01; *** indicates p-value < 0.001
The correlation matrix illustrating significant relationships among the environmental parameters is presented in Supplementary 4, and the corresponding environmental data plots are provided in Supplementary 5.
Discussion
This study provides the first characterization of epibiotic diatom communities associated with distinct populations of Gelidium corneum along the northern coast of Spain. At the spatial scale examined, geographically distant populations hosted relatively homogeneous diatom assemblages, while seasonal influxes clearly influenced algal microbiome composition through environmental filtering. During spring, when environmental conditions were most favourable for diatom growth, diversity and richness declined, and community composition exhibited greater divergence among host populations.
Is there Variation in Epibenthic Diatom Community Composition Among Different Gelidium Corneum Populations?
Based on our findings, epibenthic diatom communities colonizing different G. corneum populations show limited differentiation. Indicator species analysis (ISA) identified 13 taxa associated with specific regions, but a larger number of taxa were linked to seasonal conditions (Supplementary 3), suggesting that environmental factors exert a stronger overall influence than the identity of geographically separated and phenotypically distinct G. corneum populations.
Previous studies mainly compared epiphytic diatom communities across macroalgal species [10, 11, 33] rather than among populations of the same host (e.g., Majewska et al., [20]. Nevertheless, considering intraspecific variation may provide additional ecological insight, particularly in Rhodophyta, which support high diatom richness [11, 33]. Their suitability for colonization is attributed to complex surface architectures, including articulated and diverse microstructures such as edges and grooves [11]. In contrast, smooth-surfaced Laminariales generally support low diatom abundance [10], while rougher surfaces offer better conditions for attachment [11].
Patterns of diatom colonization likely result from interactions among biotic and abiotic factors. Among biotic factors, host–microbiome interactions are shaped by morphology, life cycle stage, grazing pressure [11], and defensive strategies that regulate colonization, including periodic shedding of outer cell layers, mucilage production, and secretion of bioactive or antifouling metabolites [34]. These metabolites may induce acclimation responses in diatoms and potentially promote co-specialization with hosts. Such dynamics suggest that some diatom lineages could adapt to the structural and chemical traits of their macroalgal hosts, fostering more specific associations.
Moreover, reproductive strategies may reinforce this process: G. corneum relies heavily on clonal propagation [35], while epiphytic diatoms show infrequent sexual reproduction [36]. Similarly, Majewska et al., [20] proposed that the association between Posidonia oceanica and Cocconeis species could arise from restricted dispersal due to predominant asexual reproduction and hydrodynamic patterns limiting contact among benthic populations. Such constraints could promote parallel evolutionary trajectories and convergent biogeographic patterns between hosts and their associated diatoms. Similarly, phenotypically distinct populations of G. corneum [19] may show similar co-specialization with their diatom communities. Future studies aiming to investigate these fine-scale associations could focus on broader geographic regions where Gelidium occurs while minimizing seasonal variation, in order to better disentangle host–microbiome interactions.
Do Seasonal, Geographical, or Local Factors Shape the Composition of these Communities Across Different Populations?
Differences in diatom taxa across G. corneum populations were primarily driven by seasonal variation, reflecting temporal shifts in environmental conditions that regulate diatom growth and community composition [37]. This pattern is consistent with the relatively narrow ecological preferences and tolerances exhibited by different diatom species, which make their assemblages highly responsive to changes in abiotic conditions throughout the year. Consequently, diatom analysis has long been recognized as one of the most reliable micropaleontological methods for reconstructing past environmental and climatic conditions [38].
Geographic separation at the spatial scale examined did not appear to impose strong ecological filters, likely reflecting the connectivity maintained by prevailing marine currents in the region. This pattern contrasts with broader biogeographic studies: for example, Majewska et al., [39] reported pronounced differences in epiphytic diatom communities on the red alga Plocamium cartilagineum between two Antarctic regions, and Burfeid-Castellanos et al., [33] documented region-specific assemblages across multiple macroalgal hosts. However, in our study geographic differentiation in diatom community composition was more pronounced in spring than in autumn, suggesting that summer, when environmental gradients between regions are strongest, may provide greater insight into regionally differentiated diatom communities.
Benthic diatoms exhibit adaptations to fluctuating and attenuated light, with many species achieving optimal growth at low photon flux densities (< 50 µE m⁻² s⁻¹), enabling persistence in shaded environments [40]. Consistent with this tolerance, diatom distribution in our study was largely unaffected by depth, likely due to the narrow range sampled (5–12 m), where light remained sufficient for photosynthesis. As a result, assemblage structure was relatively uniform, and depth-related patterns were absent. In contrast, studies spanning broader depth gradients (e.g., Burfeid-Castellanos et al., [33], Majewska, [39] reported differences such as a greater prevalence of erect forms in deeper zones, attributed to reduced irradiance and the need for elevated positioning [41]. Evidence from lake ecosystems suggests that depth often serves as a proxy for other gradients, including light, substrate, and hydrodynamics [42]. In our system, the narrow depths and homogeneous conditions likely minimized depth effects, reinforcing environmental factors as the primary drivers of epiphytic diatom community structure.
To what Extent do Environmental Variables Explain Observed Microbiome Patterns?
Several physical factors, including light availability, hydrodynamics, salinity, and sediment characteristics, regulate benthic microalgal abundance, composition, and productivity [37]. Consistent with this, most of the measured environmental variables significantly influenced the benthic diatom communities associated with distinct G. corneum populations. Seawater temperature (ST), nutrient availability, and significant wave height (Hs) were the main drivers, whereas photosynthetically active radiation (PAR) did not significantly structure the communities. ST and nutrient levels were negatively correlated, indicating interrelated effects that likely reflect seasonal dynamics.
Rising spring ST enhance metabolic activity, nutrient assimilation [43], and thermal stratification [44]. Nitrogen, often the primary limiting factor, strongly controls diatom composition, diversity, and biomass in coastal ecosystems [45]. Phosphorus interacts with nitrogen, with their N: P ratio influencing plankton dynamics [46], consistent with the nearly perfect positive correlation observed. Silicon, essential for frustule formation, can likewise constrain growth and influence species composition [47]. These nutrient-mediated processes regulate diatom abundance and functional traits, shaping food-web structure and biogeochemical cycling. In this study, higher macronutrient concentrations (PO₄, NO₃, Si) were associated with strong dominance by specific diatom taxa. For example, Licmosoma squamosum represented nearly 50% of relative abundance in spring samples from the easternmost population (Supplementary 6), likely reflecting competitive exclusion under nutrient-rich conditions. Large diatom species may gain advantage through greater storage capacity and “luxury uptake” [48], whereas smaller cells prevail under oligotrophic conditions [49]. Seasonal host dynamics may interact with these effects: after autumn–winter hydrodynamic stress reduces G. corneum biomass, spring regrowth provides new substrate favouring dominant taxa. Consequently, size-based community structure reflects resource availability and habitat turnover, with larger species maintaining growth through pulsed-nutrient periods and smaller species responding rapidly but transiently [50].
In the Basque Country population, dominance of L. squamosum may indicate potential co-specialization or mutualism with G. corneum. Dense aggregations of this large diatom could attenuate light, buffering the host against high irradiance and thermal stress during specific periods of the year (i.e., summer). Similar interactions have been reported, such as bryozoans reducing light exposure on Gelidium thalli [51] and altering nutrient dynamics in other algae [52]. Because G. corneum populations in this region have been strongly affected by climate change, manifested as reductions in cover and biomass, determining whether microbiome interactions, in conjunction with thermal plasticity, contribute to the resilience of remaining populations may elucidate their capacity to persist under warming conditions. As eastern G. corneum populations have been strongly affected by climate change, understanding whether microbiome interactions, together with thermal plasticity, contribute to their resilience may clarify their capacity to persist under warming conditions.
Therefore, this study presents the first comprehensive characterization of epiphytic diatom communities associated with G. corneum, revealing a dynamic and diverse assemblage largely shaped by seasonal environmental changes in the Cantabrian Sea. While seasonality emerged as the main driver of diatom assemblage structure, we were also able to detect some differences regarding microbiome composition among phenotypically distinct populations of G. corneum. These patterns suggest the potential for environment-driven co-associations between host and diatom species, which may have relevance under the holobiont framework, as both host and microbiota could adapt together to local conditions. Given that our analysis combined morphological and molecular data, we recognize that methodological differences across studies may limit direct comparisons of community composition. Standardization of sampling and identification protocols, as previously proposed [33], will be essential to improving cross-study comparability. Then, future research should continue to investigate microbiome-Gelidium assemblages to assess whether this red alga could co-evolve with specific microorganisms, enhancing its ability to adapt to climate change impacts.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1 (DOCX 13.3 MB)
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1NASA (2025) Moderate-resolution imaging spectroradiometer (MODIS) aqua. https://oceancolor.gsfc.nasa.gov/
- 2Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’hara RB, Simpson GL, Solymos P, Stevens MHH, Wagner H (2013) & others. Vegan: Community Ecology Package. In Community ecology package, version (Vol. 2, Issue 9, pp. 1–295). https://CRAN.R-project.org/package=vegan
- 3R Core Team (2024) R: A Language and Environment for Statistical Computing (R version 4.3.3). R Foundation for Statistical Computing. https://www.R-project.org/
- 4Mann D (1988) Why didn’t Lund see sex in Asterionella? A discussion of the diatom life cycle in nature. In: Round FE (ed) Algae and the Aquatic Environment. Biopress Ltd., pp 383–412