In a grain of sand: An overlooked over‐summering habitat of macroalgae

Seaweeds, or macroalgae, are primary producers on coasts whose biomass and species composition of their communities fluctuates seasonally. These fluctuations, which arise from phenological variations of many species (van den Hoek et al., 1995), partly drive spatiotemporal fluctuation in food webs and element cycling in coastal ecosystems (Hurd et al., 2014). Many of these species' life cycle has alternation of generations between the diploid body (sporophyte generation) and the haploid body (gametophyte generation). Sporophytes are developed from fertilized zygotes and meiotically produce haploid meiospores, which develop into gametophytes that produce gametes to form zygotes through fertilization. In many species, one generation is macroscopic and the other is microscopic, and these two generations alternate in response to seasonal environmental conditions. For example, some brown algae (e.g., Desmarestia ligulata) have macroscopic sporophytes, which form seaweed forests from spring to fall that then disappear in winter, when microscopic gametophytes form (Edwards, 2000). Red algae of the Bangiales, like the economically important alga Pyropia yezoensis, flourish during the winter–spring as macroscopic plants (gametophytes) and develop microscopic filamentous sporophytes (called “conchocelis”) in summer that bore into bivalve shells (Wang et al., 2020). This phenology is an adaptive strategy to avoid seasonal environmental stressors (Hurd et al., 2014, chap. 2). However, the ecology, including specific habitats and the timing of reproduction of microscopic stages in the field that are essential to understand the mechanism and evolution of macroalgal phenology, has been poorly studied. In Monostroma angicava, an ulvophycean green macroalga inhabiting mid-to-low intertidal zones of northern Japan, including Muroran, macroscopic foliose gametophytes flourish from winter to spring, release biflagellate gametes to reproduce sexually, and disappear in early summer (Tatewaki, 1969; Togashi & Cox, 2001), similar to what is seen in the Bangiales (Figure 1a,b). Fertilized quadriflagellate zygotes, with negative phototaxis (Figure 1c), swim toward and settle on benthic substrata. Settled zygotes, which lose flagella, develop into microscopic unicellular cyst-like sporophytes called the “codiolum” stage with their volume increasing (Horinouchi et al., 2019). Codiolum sporophytes are usually spherical without complex rhizoids in plastic or glass culture dishes (Horinouchi et al., 2019; Tatewaki, 1969). In related species (e.g., Gomontia polyrhiza), settled zygotes and codiolum sporophytes can bore into bivalve shell probably by dissolving the shell's calcareous matrix while they grow (Kornmann, 1959; O'Kelly et al., 2004). When the sporophytes of these species grow in shells, the main cell body sinks into the substrate and elongates multiple rhizoids irregularly toward the surface of the shell, resulting in a characteristic complex shape of boring codiolum sporophytes (Kornmann, 1959). As boring Bangiales sporophytes likely utilize extracellular carbonic anhydrase to dissolve shell calcium carbonate (CaCO3) (Wang et al., 2020), boring codiolum sporophytes might have a similar mechanism. However, it is unclear whether M. angicava has this boring ability. Additionally, in many populations, such as those around Muroran, bivalves are infrequent in M. angicava habitats (Figure 1a,b), it is therefore unknown what sporophytes inhabit in summer. Here we report that zygotes and microscopic codiolum sporophytes of the green macroalga M. angicava have an endolithic (within-solid-substrate) lifestyle boring into calcareous substrates and mainly inhabiting inside of calcareous sand grains, a niche completely different from that of its epilithic (on-rock) or epiphytic (on-other-macroalga/plant) macroscopic gametophytes. In the summer of 2022, we noticed a thin deposit of sand in M. angicava habitats (Figure 1d) (sand depth 8.8 ± 11.9 mm, n = 75). Through microscopic observation of this sand that was mainly composed of magnetite, quartz, and calcareous grains (e.g., limestones and shell fragments), we found many codiolum sporophytes within the calcareous grains (Figure 1f). These sporophytes had complex shapes, being ellipsoidal and with rhizoids elongating irregularly toward the surface of sand grains (Figure 1g,h). These cells often grew densely and varied in size. Small cells of several micrometers in size (see Appendix S1 for details) were likely seen soon after settlement (Figure 1h) since zygotes were 5–7.3 μm in diameter after settlement (Tatewaki, 1969). We also found that sporophytes bored into tiny shell fragments (Figure 1i). Morphological and genetic analysis using gametophytes that developed from isolated calcareous grains confirmed that nearly all were M. angicava (93.0%, n = 14 grains) (Appendix S1: Figure S1). Bangiales sporophytes, as well as other microscopic ulvophyceans (Phaeophila sp., Acrochaete sp., and Gomontia sp.), also grew in these calcareous grains (Figure 1j). Additionally, spherical sporophytes grew sparsely in the pockets of the surface of noncalcareous sand grains (Appendix S1: Figure S2). We did not find codiolum sporophytes in other areas, such as rocks, calcareous algae, or shells of living snail. Through observation of field sporophytes, we detailed the life cycle phenology of M. angicava (Figure 2) (see Horinouchi & Togashi, 2023; Tatewaki, 1969 for life cycle variations). M. angicava sporophytes collected in early winter (November and December, n = 297) were significantly but only slightly larger than those in summer (July and August, n = 811): 23.4 ± 12.7 × 9.5 ± 6.3 μm (length × width, mean ± SD) in summer and 25.8 ± 9.6 × 12.0 ± 5.3 μm in winter (length: W = 100,726, p = 3.0 × 10−5; width: W = 79,106, p < 2.2 × 10−16; Mann–Whitney U test). This suggests that young individuals settled in spring and grew slowly from summer to winter. Additionally, sizes of sporophytes varied within a single calcareous grain probably due to differences in the timing of settlement, and some grains had small sporophytes even in winter (Figure 1k, Appendix S1: Figure S3). Culture experiments confirmed that they were M. angicava. Sporophytes began to mature, by converting their cytoplasm into zoospores, in November (Appendix S1: Figure S2). Many empty sporophytes, that is, after zoospore release, were found in December (Figure 1l). An isolated sand grain collected in December released zoospores that developed into young filamentous gametophytes within 7 days after collection (Figure 1m), suggesting that M. angicava sporophytes may reproduce around December in the field. Given that M. angicava zoospores take a few months to grow to form observable-sized gametophytes under colder and shorter photoperiod conditions, such as in December (Tatewaki, 1969), this inference is consistent with the observation that observable-sized gametophytes were not found in the field in December but were found in February. To assess the abundance of M. angicava sporophytes after over-summering and the potential of sand as a sporophyte habitat, we investigated calcareous grains (n = 680) in 50 μL sand samples collected from 10 sampling points at Botofurinai, Muroran, in early winter of 2022 (see Appendix S1 for details). Calcareous grains were 0.86 ± 0.34 mm × 0.59 ± 0.24 mm (length × width, mean ± SD), with an estimated surface area of 1.63 ± 1.43 mm2, and estimated volume of 0.23 ± 0.38 mm3. The estimated surface area of calcareous grains per 1 m2 was 6.5 ± 3.3 m2/m2. The proportion of calcareous grains in sand was 10.5 ± 6.8% in volume. Codiolum sporophytes grew in 4.0% of calcareous grains, and 3.0 ± 34.6 (range: 0–667.8) individuals bored into a single grain. The estimated density of sporophytes in this habitat was 8.3 ± 9.1 × 106 individuals/m2. Our understanding of the natural history and ecology of endoliths (organisms living within solid substrates) remains poor, although bioerosion studies focusing on mainly microalgae and fungi have highlighted its ecological and geological significance (Golubic et al., 1981; Tribollet, 2008). Specifically, tiny sand grains have been overlooked as a seasonal habitat of macroalgal life cycle stages. Although a few studies have reported that Bangiales sporophytes inhabit grains of coral skeletons in tropical ecosystems (Perry, 1998), little attention has been paid to its importance for macroalgal phenology and demography in general. This study indicates that sand grains are a main over-summering habitat for the microscopic sporophyte stage of M. angicava. The field density of these endolithic stages was high, similar to the geometric median (6.5 × 106 individuals/m2) between the maximum density of zygotes per year (3.3 × 1011 individuals/m2) and the density of mature gametophytes, which survived from zygotes of the previous year (1.3 × 102 individuals/m2) (Togashi et al., 1997), suggesting that many zygotes bore into sand grains and grow into sporophytes to survive (see also Appendix S2). Although the proportion of calcareous grains was not high, their total surface area, where M. angicava zygotes can settle, bore, and grow, was large, demonstrating a high potential of calcareous grains as endolith algal habitat. Although sand grains may seem an unstable habitat, the seasonal abundance of M. angicava suggests that sand grains are a more stable habitat than our current perception and can play a central role in macroalgal phenology. Although dormant or long-lived macroalgal microscopic stages are expected, evidence from the field remains poor (Edwards, 2000). In laboratory culture, sporophytes of M. angicava stop growth under dim light, survive for several years, and then restart growth and reproduce after transfer to appropriate light conditions (Horinouchi, unpublished data). In the field, stirring by wave or benthos activities causes burial and resurfacing of sand grains, entailing changes in light conditions, which can stop growth. Sand grains having small sporophytes in winter suggest that some sporophytes hardly grew, probably due to their surrounding environments. Together with larger sporophytes observed in summer, some sporophytes appear not to reproduce in the winter of their settlement and may survive for several years. Since stopping growth entails reduced metabolism, this can be a type of dormancy. Dormancy can be an adaptation (e.g., bet-hedging strategy) to unexpected spatiotemporal environmental changes (Cohen, 1966; Gremer, 2023). The degree of sand stirring will be affected by irregular events such as typhoons, varying the number of dormant individuals and the gametophyte population size the following year. Such dormant abilities can contribute to robustness and succession of macroalgal communities as a seed bank (Schoenrock et al., 2021). Notably, this is similar to the buried seed strategy of many land plant species in forest ecosystems (Nakagoshi, 1985), which may be an interesting parallel between these evolutionarily distinct groups in aquatic and terrestrial ecosystems. Further investigation on ecophysiological mechanisms and fitness benefits of dormancy is needed. Microscopic stages inhabiting sand grains have implications for macroalgal ecology. First, seasonal escape into calcareous substrates is likely an adaptive strategy to avoid both intertidal physical stresses increased in summer (e.g., desiccation) and biotic stresses (e.g., grazing by snails or microscopic ciliates) (Horinouchi & Togashi, 2022; Lubchenco & Cubit, 1980). Second, the sympatric growth of codiolum sporophytes and Bangiales sporophytes in the same grain suggests an overlooked interaction (e.g., competitions at micro-niches) among macroalgal microscopic stages. Third, inhabiting sand grains may be a novel overlooked dispersal mode. Macroalgae, such as M. angicava, may have a low dispersal ability and consequent low gene flow among neighboring populations, since they have tiny sporophytes (Bell, 1997) and their gametes and zoospores are not long-lived (Santelices, 1990). Sand grains could be dispersal agents, moved by water currents and attaching to animals, increasing the dispersal abilities of M. angicava more than previously thought. Even if bivalve populations coexist, inhabiting sand grains can facilitate dispersal and/or be a risk-spreading strategy by utilizing a niche different from bivalve shell. More detailed culture experiments and phylogeographic analyses of neighboring populations are needed to investigate these hypotheses further. Sand grains as a seasonal habitat for macroalgal microscopic stages may be a hitherto unrecognized but important substrate, as seen in M. angicava. As seen in microboring green algae of coral skeletons (Verbruggen & Tribollet, 2011) as well as Bangiales sporophytes (Wang et al., 2020), boring codiolum sporophytes must dissolve CaCO3. M. angicava and related green macroalgae might be important bioeroders that affect carbon and calcium cycles in temperate ecosystems. Future studies on microscopic stages inhabiting sand grains can contribute to our understanding of macroalgal phenology, demography, and their trophic effects in coastal ecosystems and their changes under recent global climate changes. Yusuke Horinouchi conceived the study, performed the field survey, and collected the specimens. Yusuke Horinouchi and Kosei Mochizuki collected the data. All authors analyzed the data. Yusuke Horinouchi drafted the original manuscript, and Yusuke Horinouchi, Kensuke Ichihara, and Tatsuya Togashi revised the manuscript. All authors approved the final version of the manuscript. We thank Joe Zuccarello for helpful comments. This study is funded by Japan Society for the Promotion of Science (nos. 22K20664 and 23K14260 to Yusuke Horinouchi) and Fujiwara Natural History Public Interest Incorporated Foundation to Tatsuya Togashi. The authors declare no conflicts of interest. Data (Horinouchi et al., 2024) are available in Figshare at https://doi.org/10.6084/m9.figshare.25018187. Appendix S1. Appendix S2. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.