Moss‐cyanobacteria associations: A model for studying symbiotic interactions and evolutionary strategies

Small, evergreen, and omnipresent, the bryophytes—comprising the liverworts, hornworts, and mosses—receive little attention, even with more than 19,000 species distributed across the globe (Brinda and Atwood, 2024). They colonize almost any habitat and can find a place to settle between and on rocks, on other plants, on soil, on walls, on cars, and elsewhere. Having no vascular system and lacking roots makes them the ideal colonizers on any substrate. But they are small. They do not flower. They are difficult to identify. And yet they fulfill crucial ecosystem functions (Eldridge et al., 2023), are used frequently in biotechnology (e.g., Horn et al., 2021), and serve as models in physiological and genetic studies (e.g., Beaulieu et al., 2025). They are not “lower plants” that have led to the evolution of “higher plants.” Rather, both vascular plants and bryophytes are derived from a complex ancestral land plant (Harris et al., 2022). Recent viewpoints have synthesized the important roles that bryophytes play across ecosystems, calling for renewed attention to and inclusion of bryophytes in empirical and theoretical research (Deilmann et al., 2024; Rousk and Villarreal, 2025). One key feature of bryophytes is their ubiquitous associations with microorganisms, including N₂‐fixing prokaryotes (diazotrophs) that can supply ecosystems with readily available nitrogen (N). This association was first described in 1909—in half a sentence—by plant ecologist Eugene Warming in lecture notes at the University of Copenhagen. The field has progressed tremendously since then, and many of the abiotic controls of this key ecosystem function (i.e., N₂ fixation) have been identified in the past decade. It is now time to look ahead.

MOSS‐CYANOBACTERIA ASSOCIATIONS AS A NEW MODEL SYSTEM TO STUDY THE EVOLUTION OF SYMBIOTIC INTERACTIONS

All plants associate with microorganisms, and one key role that microorganisms play is the fixation of atmospheric N₂, converting inert N₂ into plant‐available N. One enigmatic group that performs this ecosystem function is the Cyanobacteria, whose members often associate with vascular and non‐vascular plants. Although liverwort‐cyanobacteria and hornwort‐cyanobacteria associations have been studied intensively, moss‐cyanobacteria interactions remain comparatively understudied—even though all mosses are colonized by N₂‐fixing cyanobacteria. However, the degree of colonization varies widely among moss species, leading to large differences in N₂ fixation rates. Nevertheless, in unpolluted ecosystems, such as arctic tundra and boreal or tropical cloud forests, moss‐cyanobacteria associations can contribute half of total ecosystem N input (Permin et al., 2022). Nitrogen availability and humidity are the key drivers of cyanobacterial N₂ fixation associated with mosses, independent of habitat (Alvarenga and Rousk, 2022). These abiotic controls have been assessed and synthesized elsewhere (Alvarenga and Rousk, 2022; Rousk, 2022), but the biotic controls remain underexplored. In particular, we do not know what type of symbioses mosses and cyanobacteria engage in, even though the answer to this question has implications for ecosystem N input because it is likely the most abundant plant‐bacteria symbiosis, given the globally significant biomass of mosses (Porada et al., 2014).

A beneficial symbiosis via nutrient exchange has been suggested (Bay et al., 2013), but others question this paradigm and instead place the symbioses near the parasitic end of the spectrum, where cyanobacteria exploit a weakened moss (Rousk et al., 2013). In addition, different moss species harbor their N₂‐fixing colonizers in different locations (endophytically inside their cells or epiphytically on their leaf surfaces; Figure 1), which indicates a diversity of types of symbioses and a continuum of interactions with cyanobacteria. Indeed, closely related moss species seem to harbor more closely related prokaryotes, including diazotrophs (Holland‐Moritz et al., 2021). Intriguingly, some mosses secrete compounds that increase N₂ fixation by cyanobacterial associates (Alvarenga et al., 2023), and some reports suggest cyanobacterial host specificity (Holland‐Moritz et al., 2021), while others find the opposite (Alvarenga et al., 2023). Such contradictory findings suggest that these symbioses are strongly influenced by nutrient availability and other environmental conditions, which makes them an ideal model for unraveling the evolution and plasticity of beneficial symbioses. For instance, diazotrophs are inhibited by increased N input but can recover promptly upon removal of the stressor (Wang et al., 2022). Thus, when N is readily available, cyanobacteria may become obsolete, or even costly to the moss host. Although little is known about these interactions at the molecular and metabolic levels, key processes in establishing and maintaining them may parallel those in legume‐rhizobia interactions, thus suggesting comparative research endeavors.

Figure 1.

Examples of moss species across climates with different rates of cyanobacterial colonization and location: (A) a Thuidium species from tropical cloud forests in Costa Rica, several hundred cyanobacterial colonies of which (B) are distributed epiphytically over moss shoots (bright yellow‐red); (C) a Leucodon species from Mediterranean forests in Italy with cyanobacterial colonizers (bright yellow, D) forming spherical colonies, mostly on the moss leaf tips; and (E) a Sphagnum species from boreal forests in Sweden, with cyanobacterial cells (F) within the dead, hyaline cells of the moss. Micrographs were taken with a USB2.0 CMOS Camera (Touptek, Hangzhou, China) mounted on an Olympus BX61 fluorescence microscope (magnification: B, ×40; D, ×100, F, ×200). Photo credits: K. Rousk, L. A. Clasen, D. De Alvarenga.

Taken as a whole, (1) mosses are easily sampled, transported, and stored; (2) they are found in various habitats that differ in environmental conditions; (3) they show large variation in symbiotic associations with cyanobacteria in terms of numbers, identity, and location; and (4) the symbioses are plastic (i.e., they can change when environmental conditions change). Thus, these associations can aid us in answering questions such as what types of interactions plants and diazotrophs engage in and how symbiotic relationships originate.

MOSS‐CYANOBACTERIA ASSOCIATIONS AS A NEW MODEL SYSTEM TO UNRAVEL EVOLUTIONARY STRATEGIES OF THE COLONIZERS

Besides serving as model systems for studying the evolution of symbiotic interactions, these associations can also be used to test evolutionary theories of life strategies. Moss‐cyanobacteria associations are prevalent beyond arctic, boreal, and tropical habitats. They have been discovered in temperate grasslands (Calabria et al., 2020) and in forest ecosystems such as Mediterranean oak forests. Nitrogen fixation by moss‐associated diazotrophs is 10 times higher in Mediterranean habitats than in tropical cloud forests, and three orders of magnitude higher than in temperate forests (Figure 2A). Low N₂ fixation rates in temperate regions are likely due to the relatively high background N input from atmospheric deposition that inhibits N₂ fixation activity (Wang et al., 2022), which is also likely the case for Mediterranean ecosystems. Nonetheless, moss‐associated N₂ fixation is higher in Mediterranean forests than in other investigated ecosystems, particularly after rain events. This hints at other controlling factors beyond N deposition. Mediterranean mosses are dry during most of the year, resulting in negligible N₂ fixation activity. After rain events, though, N₂ fixation rates in mosses can resuscitate quickly (Figure 2B). This is not true for N₂ fixation in other ecosystems, and it may be the result of different life strategies of organisms adapted to drought and to short windows of activity. The r/K selection theory was coined in the 1970s (MacArthur and Wilson, 1967), wherein r‐strategies (lives that are fast and short, often in unstable environments) are contrasted with K‐selected strategies (carrying capacity of the species, often found in stable environments). Having originated in the science of reproductive strategies, this concept has been applied to other fields, including conservation biology and plant dispersal, because it is about maximizing fitness in different environments. Perhaps the moss‐colonizing cyanobacteria in Mediterranean habitats with sporadic rain events are r‐selected strategists, while those in tropical ecosystems with a stable climate are K‐selected strategists. These dynamics make moss‐cyanobacteria symbioses a powerful, accessible system for the application of macroecological and evolutionary concepts to microbial communities associated with plant hosts.

Figure 2.

(A) Nitrogen fixation (as ethylene production in nmol g dw⁻¹ h⁻¹) in mosses from tropical montane cloud forests (two species collected in Malaysia), Mediterranean oak forests (two species collected in southern France), and temperate deciduous forests (two species collected in southern Sweden) as an example of cross‐climate differences in N₂ fixation rates and, likely, cyanobacterial strategies (n = 6 species⁻¹). The tropical mosses were collected in super‐humid montane cloud forests that are constantly immersed in clouds, even in the dry season, remaining moist (relative humidity >80%) throughout the year. Thus, mosses here do not dry out, in contrast to the other two ecosystems (Mediterranean and temperate), which experience clear seasonality and wet and dry periods, and extreme dry‐rewet cycles in Mediterranean habitats. Nitrogen fixation was measured under optimal temperatures (25°C), moisture levels (100% moss‐moisture), and light levels (550 µmol m⁻² s⁻¹). (B) Nitrogen fixation in mosses collected in Mediterranean forests (two species, southern France) and temperate deciduous forests (two species, southern Sweden) (i.e., the same four species in A, under the same light and temperature conditions) when dry (green bars) and 2 h after rewetting (blue bars). Significant differences denoted by lowercase letters.

CONCLUSIONS

Moss‐cyanobacteria associations represent a promising and underutilized model system for exploring the evolution of symbiotic interactions and microbial life‐history strategies. The variability in colonization patterns and responses to changing environmental conditions among moss species and associated cyanobacteria hint at different evolutionary trajectories within the group of mosses. Additionally, the moss leaf can serve as a playground for the study of microbe‐microbe interactions and the fate of fixed N₂ at the microscale—as cyanobacteria are surrounded by a plethora of other microorganisms that potentially take up the fixed N₂ before the moss host has access to it. By integrating physiological, ecological, and evolutionary perspectives, the study of moss‐cyanobacteria systems can deepen our knowledge of plant‐microbe interactions and contribute broader insights to our understanding of the coevolution of life on land. As such, these systems deserve greater attention in both theoretical frameworks and empirical studies.

Rousk, K. 2025. Moss‐cyanobacteria associations: A model for studying symbiotic interactions and evolutionary strategies. American Journal of Botany 112(8): e70086. 10.1002/ajb2.70086