Salinity is one of the most important factors influencing the distribution of the coastal green seaweed Ulva ^[1][122]. It is also probably the most important environmental factor to consider for the use of Ulva spp. in the bioremediation of coastal fresh/brackish wastewaters or in any other aquaculture application outside the sea. Ulva belongs to the intertidal zone and is a euryhaline genus that tolerates a wide range of salinities, from the hypohaline to hyperhaline zones ^[2][123]. Generally, intertidal seaweeds have a higher capacity to withstand changes in salinity than subtidal species. For example, the seaweeds present in the pools formed during the retreat of the sea can experience large variations in salinity during the day, ranging from 0.1 to 3.5 times that of seawater ^[3][124]. Importantly, Ulva spp. have the remarkable characteristic to include marine, brackish and freshwater species ^[2][123]. This makes this genus ideal for the investigation of the mechanisms involved in salinity tolerance and adaptations, as well as a source of diverse species which can be deployed for bioremediation, according to the salinity observed in the wastewater to be treated and the availability of water sources.

Variations in salinity can cause osmotic, ionic and oxidative stresses, which have a strong effect on the cellular functions of photosynthetic organisms ^[4][125]. Under salinity stress, variations in osmolarity disturb cell turgor pressure, ion distribution and metabolic reactions, and often lead to an accumulation of reactive oxygen species (ROS). This accumulation of ROS is responsible for damage to protein complexes, membranes and other cellular components, thus affecting metabolism and growth, leading in extreme cases to cell death ^[1][5][6][7][122,126,127,128]. Such damage can result from either hypo or hyper-salinity treatments ^[8][129]. A number of publications document the impact of salinity on the growth rate and nitrate and phosphate uptake of Ulva species ^[9][10][11][114,130,131]. Disturbances in carbon and nitrogen metabolism due to changes in hypersaline conditions have also been described in the species U. pertusa, with an increase in the compatible solute proline ^[12][132]. In response to hyposalinity stresses, growth and physiological impacts of salinity have been largely documented in a number of Ulva species, such as U. intestinalis, U. prolifera, U. linza, U. limnetica, U. lactuca and U. australis ^{[8][12][13][14][15][16][17][18]}[129,132,133,134,135,136,137,138]. A six-day exposure of U. prolifera to hyposaline conditions, from 30 ppt to 10 ppt, has a significant impact on growth rate and photosynthetic performance, decreasing growth rate by 65% ^[19][139]. Lu et al. (2006) ^[8][129] also found that after only 4 days of exposure of Ulva fasciata to hyposalinity (10 ppt), there was a reduction in maximum photosynthetic quantum efficiency (Fv/Fm) of 10%, which was proposed to be due to oxidative damage in chloroplasts ^[8][129].

Salinity is also known to affect the morphology of Ulva species ^{[20][21][22][23]}[140,141,142,143]. Some Ulva species, such as U. compressa and U. intestinalis can be found with two distinct morphotypes, tubular and foliose thalli. Indeed, U. compressa is found as a monostromatic tubular morphotype in a saline/hypersaline environment and a distromatic foliose form in a low salinity environment, such as estuarine sites ^[21][141]. It is not clear whether these differences can be attributed to a direct effect of salinity or are an indirect effect of salinity associated variation in the microbiome ^[24][144]. No obligate foliose species have been recorded in freshwater ecosystems and tubular morphotypes are found in a broader range of salinities ^[25][26][27][3,145,146]. For example, U. flexuosa is the only Ulva species known to date which is able to grow from ultra-oligohaline to hyperhaline zones where salinity exceeds 50 PSU ^{[22][26][28][29]}[4,142,145,147]. U. torta also shows a very wide range of salinity tolerance, from 1 to 36 PSU ^[2][123]. Valiela et al. (1997) ^[30][148] have hypothesised that those tubular cells have a better survival potential under low salinity conditions, e.g., a higher surface-volume ratio allowing for more rapid nutrient uptake compared to the foliose morphotype. In addition, an increase in the number of branches associated with a thallus in an aggregated form under low salinity has been reported for the species U. prolifera ^[31][6]. It was hypothesised that this may allow for better protection against increased turgor caused by lower salinity, as this new morphology would allow for the establishment of a more stable microenvironment around Ulva thalli. Contradictory observations were made in the distribution of U. compressa, which more frequently presents the tubular morphotype at high salinity and the foliose morphotype at low salinity ^[21][24][141,144]. Going further, Rybak et al. (2018) ^[2][123] hypothesised that an ancestral tubular morphotype carried tolerance and rapid adaptation mechanisms that are independent of morphotype, with these being lost among more recently diverged foliose and/or tubular species, but experimental evidence is unfortunately lacking.

Recent gene expression studies identified candidate genes for involvement in tolerance to short-term low salinity conditions ^[4][125]. In one study, genes involved in photosynthesis and glycolysis were typically shown to be up-regulated in response to hypo-salinity stress ^[4][125]. An earlier study demonstrated the downregulation of many genes related to lipid metabolism, membrane and cell adhesion (51–93 genes) when U. prolifera and U. linza were cultured in fresh/brackish water compared to seawater ^[32][149]. The same study also identified some upregulated genes, encoding an ion transporter, a hydrolase and multiple heat shock proteins. Despite the insight that such comparative transcriptomics can offer, a more thorough understanding of the mechanisms of acclimatisation and tolerance to salinity variations is likely to require targeted strategies to identify genes involved in the process, such as via genome-wide association studies or QTL mapping linked with reverse genetic validation. Indeed, gene expression studies cannot identify with certainty causal genes.

Ulva spp. depend on mutualistic bacteria for proper development and growth ^{[33][34][35][36]}[1,150,151,152]. This dependence is not related to the presence of a single, defined bacterium, rather, it can be achieved by redundant partnerships and the details of these requirements are poorly described ^[37][153]. A useful study system for this dependency has been developed and termed “tripartite symbiosis”, where Roseovarius sp. MS2 and Maribacter sp. MS6 are sufficient to restore normal development in the Ulva host ^[38][154]. Ulva does not survive or grows at a very low rate with an undeveloped cell wall when deprived of its microbiome ^[39][120]. Ulva-associated bacteria also provide nutrient cycling and disease resistance for their host ^[40][41][155,156].

The change in environmental conditions during establishment in aquaculture settings often causes stress in the seaweed and changes in the associated microbiome ^[40][42][155,157]. Ulva adaptation to new environmental conditions can be considered to occur via changes in the metabolism of the seaweed depending on its genetic characteristics and changes in the bacterial community associated with Ulva that provides support through the production of algal growth and morphogenesis-promoting factors (AGMPFs) ^[43][158]. The composition of the microbiome associated with Ulva spp. is influenced by the geographical location as well as abiotic factors, such as temperature, salinity and nutrient concentrations ^[44][45][159,160]. Even if a core microbiome with the essential bacteria exists in macroalgae, Burke et al., 2010 ^[44][159] and Tujula et al. (2010) ^[46][161] have demonstrated that the composition of the microbiome changes both seasonally and geographically. Understanding this microbiome-Ulva complex is, therefore, essential given its importance for the adaptation of Ulva spp. to its environment, which will vary between aquaculture systems.

Many studies have examined the impact of growing conditions on the epiphytic microbiomes of seaweed ^[47][48][162,163]. In Fucus vesiculosus, an increase in salinity can cause a significant loss in bacterial community diversity ^[49][164]. Saha et al. (2020) ^[47][162] have shown that the epibacterial communities of an invasive red seaweed (Agarophyton vermicullophylum) changed significantly in terms of species richness and diversity according to the salinity. Concerning Ulva species, Tujula et al. (2010) ^[46][161] have shown that the microbiome associated with the species Ulva australis can vary considerably among the individuals collected from the same area and between different seasons. Califano et al. (2020) ^[42][157] have investigated the impact of wild Ulva transfer in a controlled environment (IMTA) on the composition of its microbiome and showed that the implementation of IMTA results in detectable changes in the epiphytic bacterial community. Another more recent study, focusing on the impact of one environmental factor, salinity, on the Ulva bacterial community has shown that the Ulva-associated microbiome is strongly structured by salinity ^[24][144]. Interestingly, the differences in bacterial communities at low and high salinity were quantitative rather than qualitative. These studies highlight that changes in bacterial communities are strongly environment dependent, which is an important consideration for the establishment of a new Ulva aquaculture farm ^[50][51][52][91,165,166].

To date, studies on associations between microbiota and conditions remain correlative, and only hypotheses can be made regarding the ability of bacteria to facilitate host adaptation to environmental factors. While studies have identified the bacteria required for Ulva development ^[33][34][35][1,150,151], studies identifying specific bacteria influencing the growth of mature thallus and the biochemical composition of the biomass are still lacking. To date, a limited number of studies have attempted to demonstrate that certain bacteria can promote Ulva growth ^[53][54][167,168] and can affect the biochemical composition of Ulva ^[55][116]. Further, examination of the molecular mechanisms driving Ulva: microbial interactions is still limited. For example, are there certain bacteria adapted to a specific environment that may be better than others for promoting Ulva growth? If they exist, such bacteria could be of critical importance to the optimisation of Ulva yields and biomass composition in aquaculture conditions. Thus, the use of different “cocktails” of bacteria could directly impact the biochemical content and the growth of Ulva ^[43][56][158,169]. Future studies should investigate the effect on Ulva phenotype of the microbiome:host genotype interactions, and the impact of environmental conditions on these interactions. For example, the exchange of resources and chemical signals from both host seaweed and epiphytic bacteria, and the impact of environmental conditions on these exchanges, should be documented.

Natural variation refers to changes in phenotype between individuals from the same species, which are explained by genetic differences. As a result, to assess the extent of natural variation within a species, individuals must be grown in the same environmental conditions in order to exclude changes in phenotypes due to the environment. Natural variation within Ulva species has been studied both for foliose and tubular species ^{[57][58][59][60]}[23,170,171,172]. Lawton et al. (2013) ^[59][171] reported high levels of variation in the specific growth rate of the foliose thallus of U. ohnoi, with strains cultivated in the same location showing > two-fold variation in growth rates. Fort et al. (2019) ^[57][23] also reported extensive variation within U. lacinulata species, >four-fold, from 0.092 to 0.371 mg·mg⁻¹·d⁻¹. This variation was in fact as high as that observed between six different Ulva foliose species. Moreover, the authors reported a similar extent of natural variation for a large range of biochemical traits, e.g., starch content and protein content. Interestingly, Fort et al. (2020) ^[61][103] subsequently reported that for a given species, the Ulva strains originating from green tide areas have higher protein, pigments, lower starch content and higher growth rates than other samples, making green tide areas suitable places for the collection of strains for aquaculture if the biomass produced is destined to feed/food applications. Huo et al. (2013) ^[62][173] also identified several strains from a same species in greentides.

Although natural variation has been identified as being very high within Ulva species, the associated genes are still unknown. A recent study has demonstrated the importance of intraspecific variation in mitochondrial genomes within the species Ulva compressa ^[63][174]. However, many previous studies of Ulva organellar genomes have shown very few differences within Ulva species, and high variation between species, suggesting that a large part of the natural variation within Ulva species is explained by nuclear encoded genes ^[60][64][43,172]. A recent study written by Fort et al. 2022 ^[65][11], details the genomic resources available in Ulva. Hence, nuclear DNA marker association studies, such as genome wide association studies or quantitative trait loci analyses, should be considered with growth and metabolite profiles to engineer, select or breed for improved yield and biomass characteristics in aquaculture.

However, before undertaking such targeted improvement strategies, significant productivity gains can already be achieved by simply screening this existing natural genetic variation to identify and isolate fast-growing strains with desirable characteristics. An important aspect of strain selection is to select strains in the environment they will be cultivated in afterward. The phenotype is dependent on the genotype (G), as well as the environment (E), and their interaction (G*E) is often an important contributing factor; hence, strain selection must be performed under environmental conditions as close as possible to those the strains will be cultivated in.