Compositions containing beneficial microorganisms for corals and methods of use thereof

Abstract

Compositions containing beneficial coral microorganisms (BMC), methods for preparing the same and methods of use thereof to confer beneficial traits to coral are provided. The BMC includes consortia of bacteria isolated from red sea coral hosts such as Pocillopora verrucosa, Stylophora pistillata and Galaxea fascicularis. or Acropora.

Description

[0001] COMPOSITIONS CONTAINING BENEFICIAL MICROORGANISMS FOR CORALS AND METHODS OF USE THEREOF CROSS-REFERENCE TO RELATED APPLICATION

[0002] The present application claims priority to U.S. Provisional Application No. 63 / 587,260, filed October 2, 2023, the contents of which are hereby incorporated by reference in their entirety.

[0003] FIELD OF THE INVENTION

[0004] The present invention relates compositions and methods of beneficial microorganisms to confer beneficial traits to corals in their habitats, such as the marine environment.

[0005] BACKGROUND OF THE INVENTION

[0006] Microorganisms are key components of all multi-cellular life due to their crucial roles in nutrient cycling and metabolism (Ainsworth et al., 2010; Garren and Azam, 2012; Thompson et al., 2014). In the past few decades, the important relationship between coral and its microbial symbionts have been documented (reviewed in Bourne et al., 2016). However, coral-microbiome interactions are not yet fully understood due to complex interactions, which include host modulation of the physiology of symbiotic cells (Barott etal., 2015) and the influence of symbiotic cells on their host (Sharp and Ritchie, 2012). The responses of coral-associated microbial communities to shifts in coral health (Harvell et al., 2007; Bourne et al., 2008; Jones et al., 2008; Glasl et al., 2016; Guest et al., 2016) and environmental stressors (Hughes et al., 2003; Doney et al., 2012; Vega Thurber et al., 2014; Garren et al., 2015; Kwiatkowski et al., 2015) have been extensively explored and reported. Changes in environmental conditions may modify coral physiology, leading to changes in the structure, spatial arrangement and abundance of the local population. These environmental disturbances can also directly or indirectly induce shifts in the associated microbial communities, leading to the emergence of disease (Harvell et al., 2007; Mouchka et al., 2010), which is an ongoing threat to coral reefs worldwide (Mumby and Steneck, 2008; Harvell et al., 2007; Vega Thurber et al., 2009; Krediet et al., 2013; Rogers and Miller, 2013).

[0007] The growing threats to corals have increasingly generated reports of whitening which, as a result, weakens the holobiont and facilitates the proliferation of opportunistic pathogens. For this reason, the number of diseases and reef mortality predictions has been increasing considerably. It is possible that in 30 years there will be more than 30% loss in the coverage of coral reefs around the world. BR 102017004458A discloses a composition containing a microbial consortium a plurality of bacteria selected from a group including the genera: Pseudoalter omonas spp, Halomonas spp, Cobetia spp. Rosado, et al., disclose a consortium including five Pseudoalter omonas sp., a Halomonas taeanensis and a Cobetia marina-related species, isolated from Pocillopora damicornis and surrounding seawater. Rosado, et al., (The ISME Journal (2019) 13:921-936). Santoro, et al., Sci. Adv. 2021; 7: eabg3088) discloses a composition containing a microbial consortium a plurality of bacteria isolated from M. hispida containing the following 6 bacterial strains: Bacillus lehensis (M20); Bacillus oshimensis (M24); lehensis (M3), Brachybacterium conglomeration (Ml); Pianococcus rifietoensis (CM29) and Salinivibrio sp. (F2).

[0008] Although promising, in -situ manipulation of the coral’s microbiome using probiotics has remains largely untested, especially in the red sea. One of the main challenges of applying probiotics to corals in natural conditions is the unknown overall effectiveness in the coral host and other organisms living in close association with the coral.

[0009] There is still a need for compositions and methods that are beneficial to coral under adverse stress conditions.

[0010] It is an object of the present invention to provide compositions containing Beneficial Microorganisms for Coral (BMC).

[0011] It is also an object of the present invention to provide methods of protecting coral by providing compositions containing Beneficial Microorganisms for Coral (BMC).

[0012] SUMMARY OF THE INVENTION

[0013] Compositions containing a consortium of bacteria which function as Beneficial Microorganisms for Coral (BMC) and their use on corals in the marine environment are provided. The disclosed consortia are isolated from the red sea. One advantage associated with Red Sea bacteria is that they evolved in environmental conditions that are similar to the future of the rest of the world (warmer) and therefore, may confer more benefits than consortia obtained from other non-red sea environments.

[0014] In one embodiment, the composition (herein BMC Consortia 1 Pocillopora)) includes isolated bacterial consortium of bacterial strains: (Pocillopora) includes 6 strains in 1 / 6 proportion: 30H (Pseudoalter omonas sp ), 33H (Pseudoalter omonas sp ), SAT10 (Halomonas sp ), SAA19 (Bacillus sp.)lSulcliffiella sp., 65H (Cobetia sp ), 81H (Cobetia sp ), preferably, isolated from Stylophora pistillata Clade IV, Galaxea fascicularis and Pocillopora verrucosa. The BMCs can be isolated from coral hosts such as Pocillopora verrucosa, Stylophora pistillata and Galaxea fascicularis and do include Vibrio spp., which is present in the natural host in combination with the isolated species Thus, the selected disclosed BMC is different from the naturally occurring population, as naturally present in the coral host. Additionally the disclosed consortium is obtained following isolation from three different coral hosts and accordingly do not exist in nature together in the same host.

[0015] Accordingly the composition of isolated bacteria, which further exclude pathogenic bacterial such as Vibrio spp., with which they co-exist in nature, is also, functionally different from the individual species as they occur in nature.

[0016] In another embodiment, the composition (herein BMC Consortia 2 (Acropora)) is composed by 4 strains in 1 / 4 proportion: 10 (Pseudoalteromonas sp.), 16 (Pseudoalteromonas sp.), SAT19 (Halomonas sp.), 81H (Cobetia sp.), isolated from the Red Sea from the coral, Acropora. The compositions can be in liquid or solid form and preferably include an effective amount of one or more stabilizers, and do include Vibrio spp., which is present in the natural host in combination with the isolated species. Thus, the selected disclosed BMC is different from the naturally occurring population, as naturally present in the coral host. Additionally the disclosed consortium is obtained following isolation from three different coral hosts and accordingly do not exist in nature together in the same host.

[0017] The disclosed consortium of BMC is in some forms obtained following culture overnight in marine culture media of each bacterial species in the consortium such as marine broth and harvested during the peak of each species exponential growth phase based on colony-forming units counting, to ensure an equal concentration of each strain in the consortium. In some forms, the disclosed BMCs include one or more agents that are effective as a stress antagonist in coral. Examples, include hydrogen, phosphate and ammonium, in effective amounts.

[0018] The disclosed compositions are beneficial to coral and sponges and non-toxic towards non-target species such as sponges.

[0019] Methods of conferring beneficial characteristics to corals and sponges are provided. In one embodiment, the corals are obtained from the red sea. The methods disclosed herein include contacting / inoculating a coral with an effective amount of the disclosed compositions.

[0020] In some forms, the methods include further treating the coral or sponges with one or more agents that are effective as a stress antagonist , such as hydrogen, phosphate and ammonium, in effective amounts. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] FIG. 1 is a schematic showing the Experimental design workflow

[0022] FIG. 2 is a schematic showing the Experimental design covering seasonal variations. The top panel shows sampling (coral fragments, water sediment and fish tissue). The bottom panel shows treatment and control groups.

[0023] FIG. 3 shows the effect of sampling time and treatment on bacterial community structure.

[0024] FIG. 4 shows Coral Bacterial community structure.

[0025] FIG. 5 shows Bacterial community of Pocillopora verrucosa.

[0026] FIG. 6 shows beta-diversity.

[0027] FIG. 7 shows bacterial community composition. Genera with relative abundances greater than 2% are shown.

[0028] FIG. 8 shows data on in-situ PAM and CBASS experiments. FIG. 8A are line graphs showing CBASS profiles by sampling time at different tested temperatures. The X-axis represents temperature (C°) and the Y-axis shows the photochemical efficiency value (PAM). PAM values below 0.6 indicate possibly stressed or damaged photochemical systems of the algae symbionts in corals. FIG. 8B are graphs showing in-situ PAM profiles of the two treatments at the different sampling times (T1-T4). X-axis represents the treatments (BMC= Probiotic; CTRL= Control) and the Y-axis shows the Fv / Fm values.

[0029] FIG. 9: Nonmetric multidimensional scaling (MDS) of Bray-Curtis dissimilarities of microbiomes found in sponges and seawater samples; square = Stylissa carteri; circle = Callyspongia crassa; diamond = S. carteri seawater; triangle = C. crassa seawater; fill shapes = TO; and hollow shapes = T1

[0030] FIG. 10: Strains of beneficial microorganisms for corals (BMC) associated with the Stylissa carteri microbiome at TO and Tl; the same letters correspond to statistical differences in treatments between samples or groups (three-way univariate PERMANOVA, p < 0.05); bars correspond to mean value ± standard error.

[0031] FIG. 11 Strains of beneficial microorganisms for corals (BMC) associated with the Stylissa carteri seawater microbiome at TO and Tl; bars correspond to mean value ± standard error.

[0032] FIG. 12 Strains of beneficial microorganisms for corals (BMC) associated with the Callyspongia crassa seawater microbiome at TO and Tl; bars correspond to mean value ± standard error. FIG. 13. Amplicon sequence variants (ASVs) of the Stylissa carteri microbiome with relative abundance > 0.1% affected by coral probiotics inoculation in comparison with the control; c = class; o = order; f = family; and g = genus.

[0033] FIG. 14: Amplicon sequence variants (ASVs) of the Callyspongia crassa microbiome with relative abundance > 0.1% affected by coral probiotics inoculation in comparison with the control, p = phylum; c = class; o = order; f = family; and g = genus.

[0034] FIG. 15: Seawater temperature during in situ coral probiotics inoculation at Al Fahal reef, Saudi Arabia, Red Sea; CC = Callyspongia crassa, and SC = Stylissa carteri.

[0035] FIGS. 16A-16B are maps showing the Study location at the Coral Probiotic Village, at Al Fahal reef, in the central Red Sea (FIG. 16A), and Spatial distribution of P. verrucosa colonies used for the experiment (FIG. 16B). Colors represent the probiotic (yellow) and control (red) treatments.

[0036] FIG. 17 is a schematic showing the temperature profiles and time settings of Coral Bleaching Automated Stress System (CBASS) experiments.

[0037] FIG. 18A is a canonical analysis of Principal Coordinates (CAP) shows spatial community structure of P. verrucosa associated bacteria according to sampling time and treatments. FIG. 18B shows Shannon diversity index by sampling time and treatments. Statistical difference is denoted with an asterisk (*). FIG. 18C are plots showing multidimensional scaling (MDS) showing the spatial community structure of P. verrucosa associated bacteria according for each sampling time (T1-T3). FIG. 18D are stacked-bar plot showing abundant families for probiotic and control treatments by sampling time (T1-T3). Families with >2% of total abundance are shown. Endozoicomonadaceae, Woeseiaceae, Desulfosarcinaceae, Pirellulaceae, and Flavobacteriaceae, were the most abundant across time and treatments.

[0038] FIG. 19A is a CAP plot shows water bacterial community structure by sampling time and treatment. FIG. 19B shows Shannon diversity index by treatment and sampling time (*=significant difference). FIG. 19C are MDS showing community structure of probiotic vs. controls in T1 and T3. FIG. 19D shows a CAP plot by sampling time and treatment of sediment bacterial communities. FIG. 19E shows Shannon diversity index by sampling time between probiotic and control treatments (ns= not significant differences). FIG. 19F is an MDS showing ordination by treatment for T1 and T3 of sediment bacterial communities. FIG. 19G is a heat map of bacterial community composition at the family level, of water (left panel) and sediment (right panel) (families with > 2% of total abundance are shown). FIGS. 20A-20F are Non-metric Multidimensional Scaling (nMDS) plots of bacterial communities included in the study. FIG. 20A shows bacterial communities of coral, water, and sediment. FIG. 20B shows P. verrucosa bacterial community according to sampling times and treatments. FIGS. 20C-20F show / < verrucosa bacterial community in Tl, T2, T3, and T4, according to treatment, respectively.

[0039] FIG. 21 shows statistical analysis of the alpha-diversity metrics (Kruskal-Wallis test) showed a significant increment in diversity in (Richness (S) (p-value = 1.27e-05***), Shannon (H’) (p-value = 0.0005***), Simpson (p-value = 0.008**), and haol (p-value = 1.043e-05***) in T3 in probiotic compared with control treatment.

[0040] FIG. 22 shows bacterial composition of P. verrucosa at the family level. The top 10 most abundant families are shown. The statistical difference and significance are denoted with an asterisk (*).

[0041] FIG. 23A shows CBASS profiles by sampling time at different tested temperatures. The X-axis represents temperature (C°) and the Y-axis shows the photochemical efficiency value (PAM). PAM values below 0.6 indicate possibly stressed or damaged photochemical systems of the algae symbionts in corals. FIG. 23B shows in-situ photosynthetic efficiency (Fv / Fm) profiles of the two treatments at the different sampling times (T1-T4). X-axis represents the treatments, and the Y-axis shows the Fv / Fm values.

[0042] Figure 24 shows percent change in Chlorophyll a concentration of both hard coral species over the 48h experimental period with n=5 per species and experiment. The median is represented by the black horizontal line and each separate measurement is indicated by a dot in the respective treatment color. Differing letters over boxplots indicate significant differences between treatments.

[0043] Figure 25 shows percent change in Inverted Mean Intensity Grey Values of both hard coral species over the 48h experimental period with n=5 per species and experiment. The median is represented by the black horizontal line and each separate measurement is indicated by a dot in the respective treatment color.

[0044] Figure 26 shows Fv / Fm values of both hard coral species over the 48h experimental period, measured at TO, 12, 24, and 48h with n=5 per species and experiment. Bars indicate the mean value and error bars standard deviation . Differing letters over treatments indicate significant differences in the percent change of TO to T48 between treatments (p < 0.05) .

[0045] Figure 27 shows PgrOss:R ratio of both hard coral species over the 48h experimental period with n=5 per species and experiment. The median is represented by the black horizontal line and each separate measurement is indicated by a dot in the respective treatment color. The dashed line indicates a ratio of 1. Values above the dashed line represent greater energy acquisition over utilization.

[0046] Figure 28 shows a phylogenomic inference of publicly available genomes from 14 Pseudoalter omonas sp. strains and the genomes of our two Pseudoalteromonas galatheae strains (pBMCl and pBMC2; highlighted in bold), totaling 16 genomes. The tree was assembled by comparing 1000 proteins using the codon tree method of the PATRIC platform. This method selects global protein families as homology groups and analyzes aligned proteins and DNA encoding single-copy genes using the RAxML program. Based on RAxML, the Jones-Taylor-Thornton protein model was considered the best model to construct this tree. Gray dots on the branches correspond to a bootstrap value of 100. The isolate source is indicated on the right side of each strain. The blue-shaded area represents the monophyletic clade comprising pBMCl and pBMC2.

[0047] Figure 29shows phylogenomic inference of publicly available genomes from 26 Cobetia sp. strains, the genome of Cobetia sp. BMC6 from Rosado and colleagues26, and the genome of our two Cobetia sp. strains (pBMC3 and pBMC4; highlighted in bold) totaling 29 genomes. The tree was assembled by comparing 1000 proteins using the codon tree method of the PATRIC platform. This method selects global protein families as homology groups and analyzes aligned proteins and DNA encoding single-copy genes using the RAxML program. Based on RAxML, the Jones-Taylor-Thomton protein model was considered the best model to construct this tree. The isolate source is indicated on the right side of each strain. Gray dots on the branches correspond to a bootstrap value of 100. The blue-shaded area represents the monophyletic clade comprising pBMC3 and pBMC4. The Cobetia sp. BMC strain from the study by Rosado and colleagues26is shaded in green.

[0048] Figure 30 shows phylogenomic inference of publicly available genomes from 24 Halomonas sp. strains, the genome of Halomonas sp. BMC7 from Rosado and colleagues26and the genome of our Halomonas sp. strain pBMC5 (highlighted in bold), totaling 26 genomes. The tree was assembled by comparing 1000 proteins using the codon tree method of the PATRIC platform. This method selects global protein families as homology groups and analyzes aligned proteins and DNA encoding single-copy genes using the RAxML program. Based on RAxML, the LG model was considered the best model to construct this tree. Gray dots on the branches correspond to a bootstrap value of 100. The isolate source is indicated on the right side of each strain. The blue-shaded area represents the strains that form a monophyletic clade with pBMC5. The Halomonas sp. BMC strain from Rosado and colleagues26is shaded in green.

[0049] Figure 31 shows phylogenomic inference of publicly available genomes from 1 Sutcliffiella sp. strain and ten Bacillus sp. strains and the genome of our Sutcliffiella sp. strain pBMC6 (highlighted in bold), totaling 12 genomes. The tree was assembled by comparing 1000 proteins using the codon tree method of the PATRIC platform. This method selects global protein families as homology groups and analyzes aligned proteins and DNA encoding single-copy genes using the RAxML program. Based on RAxML, the LG model was considered the best model to construct this tree. Gray dots on the branches correspond to a bootstrap value of 100. The isolate source is indicated on the right side of each strain. The blue-shaded area represents the strains that form a monophyletic clade with pBMC6.

[0050] Figure 32 shows genome maps of the six putative BMC strains. The rings (from outside to inside) indicate the nucleotide position in the genome, predicted genes on the forward strand, predicted genes on the reverse strand, location of putative prophages (as predicted by PHASTEST), and percent GC content (higher GC content = yellow, lower GC content = purple). Genes predicted to encode beneficial proteins are color-coded in the outer rings. The circular genomes were visualized using DNAplotter175.

[0051] DETAILED DESCRIPTION OF THE INVENTION

[0052] I. DEFINITIONS

[0053] “Beneficial Microorganisms for Corals” (BMC) as used herein refers to microorganism that can live symbiotically with corals and confer benefits to the coral host including nutritional (“probiotics”) and protective mechanisms that improve coral fitness and contribute to coral resilience.

[0054] As used herein, “isolated,” and like terms relate to the isolation of an organism in a form that is substantially free (at least 60% free, preferably 75% free, and most preferably 90% free) from other components normally associated with the organism in a native environment.

[0055] Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range, unless otherwise indicated herein, and each separate value is incorporated into the specification as if it were individually recited herein.

[0056] Use of the term “about” is intended to describe values either above or below the stated value in a range of approx. + / - 10%; in other forms the values may range in value either above or below the stated value in a range of approx. + / - 5%; in other forms the values may range in value either above or below the stated value in a range of approx. + / - 2%; in other forms the values may range in value either above or below the stated value in a range of approx. + / - 1%. The preceding ranges are intended to be made clear by context, and no further limitation is implied.

[0057] II. COMPOSITIONS

[0058] The BMC compositions disclosed herein include a consortium of bacteria with beneficial characteristics for corals and sponges. In some forms, the disclosed BMCs include one or more agents that are effective as a stress antagonist in coral. Examples, include hydrogen, phosphate and ammonium, in effective amounts.

[0059] Microbial consortium is a mixture, association, or assemblage of two or more microbial species, which in some instances are in physical contact with one another. The microbes in a consortium may affect one another by direct physical contact or through biochemical interactions, or both. For example, microbes in a consortium may exchange nutrients, metabolites, or gases with one another. Thus, in some examples, at least some of the microbes in a consortium may be metabolically interdependent. Such interdependent interactions may change in character and extent through time and with changing culture conditions. The disclosed consortium of BMC is in some forms is obtained following culture overnight in marine culture media such as marine broth and harvested during the peak of each species exponential growth phase based on colony-forming units counting, to ensure an equal concentration of each strain in the consortium.

[0060] The disclosed BMC or isolates thereof may be described to be in an “isolated” or “biologically pure” form. These terms are intended to mean that the bacteria have been separated from an environment or one or more constituents, cellular or otherwise, which they may be associated with if found in nature or otherwise. The isolates or cultures of the bacteria in the BMC contain a predominance of the bacteria of the disclosed bacterial species.

[0061] In some forms the BMC composition (herein BMC Consortia 1 (Pocillopora) includes a combination of the following bacteria: BMC Consortia 1 (Pocillopora) is composed by 6 strains in 1 / 6 proportion: 30H (Pseudoalter omonas sp.), 33H (Pseudoalter omonas sp. , SAT 10 (Halomonas sp.), SAA19 (Bacillus sp.), 65H (Cobetia sp.), 81H (Cobetia sp ), preferably, isolated from Stylophora pistillata Clade IV, Galaxea fascicularis and Pocillopora verrucosa, and optionally, in combination with an effective amount of a stabilizer and do include Vibrio spp., which is present in the natural host in combination with the isolated species. Accordingly the composition is isolated bacteria is different from the population as they are naturally present in the coral host The BMCs can be isolated from coral hosts such as Pocillopora verrucosa, Stylophora pistillata and Galaxea fascicularis.

[0062] In another embodiment, the composition (herein BMC Consortia 2 (Acropora)) is composed by 4 strains in 1 / 4 proportion: 10 (Pseudoalteromonas sp.), 16 (Pseudoalteromonas sp.), SAT19 Halomonas sp.). 81H (Cobetia sp.), preferably isolated from the red sea coral, Acropora.

[0063] The genomic sequences for each bacterial strain in the consortium are deposited in the European Nucleotide Archive (ENA) under the Project ID PRJEB62849 (https: / / www.ebi.ac.uk / ena / browser / view / PRJEB628490); the accession numbers for each bacterial strain in the consortium is provided in PRJEB62849 as follows in Table A:

[0064] Table A. Deposit information for bacterial strains in PRJEB62849

[0065] The individual species are included in the composition in equal amounts.

[0066] In some embodiments, the disclosed compositions are in liquid form. The liquid compositions include an effective amount of a liquid stabilizer such as 20% glycerol or 3.5% NaCl.

[0067] Another embodiment the disclosed compositions are in solid or gelatinous form.

[0068] In some forms, the BMC composition consists essentially of the disclosed bacterial combinations and the stabilizer. In some forms, the combination of bacteria in the BMC compositions consist of 30H (Pseudoalter omonas sp.), 33H (Pseudoalter opmonas sp.), SAT10 (Halomonas sp.), SAA19 (Bacillus sp.), 65H (Cobetia sp.), 81H (Cobetia sp.) i.e., no additional bacterial specie is included in the BMC composition.

[0069] In some forms, the bacteria combinations in the BMC compositions consist of 10 (Pseudoalter omonas sp , 16 (Pseudoalter omonas sp , SAT19 (Halomonas sp.) and 81H (Cobetia sp.) i.e., no additional bacterial specie is included in the BMC composition.

[0070] The BMC compositions can be obtained from coral, according to methods described in the Examples, reproduced in part and the methods “Production of BMC consortium” as disclosed in the Example, incorporated herein. (See FIG. 1 ((l)-(4))). The methods include: (a) isolating bacteria from the coral of interest, (b) functional screening of the bacteria isolates for traits selected from for reactive oxygen species (ROS) scavengers, phosphate assimilation, an antagonistic effect against Vibrio coralliliticus, and genes related to nitrogen cycling, as nitrogenase subunits (nifH) and denitrification (nirK), as well as the DMSP degradation (dmdA) gene (can be screened through whole genome analysis), and (c) selecting strains on the basis of being positive for one or more of the desired traits; each strain is collected proportionally at the peak of its growth curve, which has been determined to achieve a concentration between 108-l 09cells / ml per strain.

[0071] (a) Isolation of Coral-associated bacteria

[0072] Briefly, coral fragments from the coral of interest, herein, (a) Stylophora pistillata Clade IV, Galaxea fascicularis and Pocillopora verrucosa are collected for example, by snorkeling and Scuba diving at Thala reef (22°15'46.9"N 39°03'05.9"E), Aquarium (22°23T5.6"N 38°55'07.2"E) and Al Fahal reef (22°18'18.4"N 38°57'52.5"E) respectively, in the central Red Sea, at depths between 1 to 10 m, between February and May 2021. Coral fragments can be collected using pliers and transported inside 50 ml falcon tubes on ice to the laboratory. Immediately, the fragments are macerated using for example 50 ml of 3.5 % sterile saline solution with a sterile mortar and pestle. Optionally, serial dilutions up to 10'6are done using the macerate paste with 3.5% saline solution, and 100 pL of each dilution is plated in Marine Agar (MA) (Zobell 2216, HiMedia Laboratories, Mumbai, India) diluted Marine Agar (DMA), and Lenox Broth (LB) (Sigma-Aldrich®). Plates are incubated at 25°C (corresponding to the in-situ water temperature registered at the sampling sites) overnight. In parallel, the macerate is incubated in a 250 mL sterile Erlenmeyer at 25°C overnight with crystal beads at 130 rpm. After this first incubation, triplicate subsamples (100 pL) of 10'4, IO'5, 10'6dilutions are plated into Petri dishes containing approximately 20 mL of Marine Agar media and diluted marine agar medium (Marine Agar Medium 2* diluted with 3.5% NaCl and agar adjusted). Alternatively, 0.5 cm coral fragments are cut and placed on the petri dishes, with both media. All the plates are incubated at 25°C for at least 48 h or until visible bacterial colonies are observed. Approximately > 350 bacterial isolates are obtained, based on colony morphology, and are preserved at -80°C using sterilized glycerol with a final concentration of 20%, for further analysis.

[0073] (b) Characterization and test of bacterial isolates (functional screening of bacterial isolates)

[0074] Bacterial strains identified as potential human or coral pathogens (e.g. Vibrio spp.) are excluded . The remaining bacterial isolates are tested for beneficial functions as follows:

[0075] For reactive oxygen species (ROS) scavengers, 20 pL of pure culture of each bacterial strain is placed on a portable microscopy layer, and immediately, a drop of 3% (v / v) of hydrogen peroxide is added in the middle of it. The criterion of a positive result is evaluated qualitatively, based on the production of bubbles, as a proxy of catalase reaction.

[0076] Phosphate assimilation can be tested according to Nautiyal, 1999, using Pikovskaya’s agar culture media (HIMEDIA®).

[0077] The antagonistic effect against Vibrio coralliliticus (well-known coral pathogen) (Ben-Haim et al., 2003) can be assessed through the diffusion agar method (Marval et al., 1990): First, 20 pl of each bacterial strain is spot-inoculated onto 2.5% NaCl LB agar, placing three spots for each strain (representing replicates). The plates are incubated at 26°C for as long as necessary for the strain to grow. The strains are inactivated by chloroform volatilization, followed by pouring 3 ml of semisolid 2.5% NaCl LB medium (0.7% agar) containing the Vibrio indicators over the inactivated spots. These plates are then incubated at 28°C for 16 hours, and the antagonistic activity is indicated by inhibition halos around or no detection of Vibrio growth over the colony spot.

[0078] Genes related to nitrogen cycling, as nitrogenase subunits (nifH) and denitrification (nirK), as well as the DMSP degradation (dmdA) gene are screened through whole genome analysis.

[0079] (c) Selection and production of BMC

[0080] Six bacterial strains (Halomonas sp., two Pseudoalteromonas spp., two Cobetias spp., and one Sutcliffiella sp.) are chosen for the probiotic consortium, based on the beneficial traits summarized in Table 1. Table 1: Beneficial traits tests performed with the BMC isolated from different coral species. (ntfH) nitrogen fixation; (nirK) denitrification; catalase activity; (-) test with negative result; and (+) positive activity.

[0081] Strain BMC ID Present data Coral-Host catalase

[0082] 30H Pseudoalteromonas sp. ASV1346 Pocillopora verrucosa (-)

[0083] 33H Pseudoalteromonas sp. ASV1346 Pocillopora verrucosa (-)

[0084] 65H Cobetia sp. ASV4551 Pocillopora verrucosa 25% (+)

[0085] 81H Cobetia sp. ASV4551 Pocillopora verrucosa 10% (+) , , , , ASV3046 Stylophora pistillata

[0086] SAT10 Halomonas sp. (z+z) (z+z) 50% (z+z)

[0087] Clade IV

[0088] SAA19 Sutcliffiella sp . ASV8312 Galaxea fascicularis (-) (+) 25% (+)

[0089] As the probiotic consortium is composed of a diverse combination of bacteria, each strain is collected proportionally at the peak of its growth curve, which has been determined to achieve a concentration between 108- 109cells / ml per strain. Fresh bacterial cultures are centrifuged at 6000g for 5 min, and the cell pellets washed three times with saline solution (3.5% NaCl), followed by centrifugation and resuspension in 1 L of saline solution for a final concentration of approximately 108'9cells / ml per strain. The composition can then be stored at 4°C until in-situ inoculation in the corals, or stored in a stabilizer such as glycerol, or further processed with a stabilizer into solid formulations as described herein.

[0090] (d) Coral Stress antagonists

[0091] Corals, for example, Hard corals are highly susceptible to heat stress, often resulting in bleaching and mortality. Coral bleaching, triggered by heat stress, breaks down the vital symbiotic relationship between corals and their zooxanthellae algae, leading to the loss of pigmentation and, eventually, coral mortality. In some forms, the disclosed BMCs or methods of treating coral / sponges include adding one or more agents that are effective as a stress antagonist in coral. Examples, include, but are not limited to, hydrogen, phosphate and ammonium, in effective amounts. H2 can be used at a concentration of about 0.01 to 10 ppm, preferably 0.1 to 5 ppm, more preferably, about.3-0.5 ppm by exchanging 10 % of coral tank water with hydrogen-enriched distilled water, “phosphate” (PO4) can be added at a concentration between about 0.2-10 pM, for example, between about 0.5 to about 5 pM, and preferably about 1, 2, 3 or 4 pM. PO4 can be provided by ISfeHPCUx2 H2O, (Sigma Aldrich, CAS 10028-24-7). “Ammonium” (NH4) can be added at a concentration between about 0.03 to about 15 pM, from about 0.5 to 10 pM, for example about 1, 2, 3 pM, 4, 5, 6, 7 or 8. Ammonium can be provide by NH4CI ((Sigma Aldrich, CAS: 12125-02-9). A. Liquid Compositions

[0092] In some forms, the disclosed compositions are in liquid form. The liquid compositions include an effective amount of a liquid stabilizer such as 20% glycerol or 3.5% NaCl.

[0093] In some forms, the disclosed consortia or compositions of microorganisms are in a liquid medium (such as a culture medium or fermentation medium) or inoculum.

[0094] In some forms, consortia or compositions of microorganisms are present on a solid or gel medium (such as a culture plate) containing or supporting microorganisms, for example, Marine Agar.

[0095] The disclosed consortia of microorganisms or compositions comprising Microorganisms can be maintained at a temperature that promotes the growth of the microorganism (s), for example, at about 25-45 °C (for example, about 30-25 °C. In dome forms, the compositions are stored at temperatures at which the microorganism (s) do not grow or are inactive, such as less than 25 °C (e.g., 4 °C, -20 °C, -40 °C, -70 °C, or below). One skilled in the art can formulate cold storage compositions, for example, by including stabilizers (such as glycerin). Still in additional examples, the compositions are stored at ambient temperature, such as about 0-35 °C (e.g., about 10-30 °C or about 15-25 °C).

[0096] In some forms, the formulation includes the disclosed microorganisms, provided in a cryopreservative. Cryopreservation is used by almost all microbial culture collections in the developed world. Some use mechanical mechanisms to achieve low temperature, but the preferred methodology involves storing cultures in the vapor phase of liquid nitrogen. If used correctly, liquid nitrogen poses little risk. To reduce the risks of cryoinjury, traditional approaches for cry opreservation have involved controlled cooling at -1°C min-1, typically in the presence of a cryoprotectant such as glycerol, trehalose, or DMSO.

[0097] Media for preservation of cells are known in the art, for example, CRYO-GOLD™ (cry opreservation medium) and CROSSTOR® (cry opreservation freeze media) designed to mitigate temperature-induced molecular cell stress responses during freezing and thawing. All CROSSTOR® products are pre-formulated with USP grade DMSO, a permeant solute cryoprotective agent which helps mitigate damage from the formation of intracellular ice. CROSSTOR® is offered in several packages and pre-formulated with DMSO in final concentrations of 2%, 5%, and 10%. A preferred medium for cell preservation includes 5-10 % DMSO, for example, CRYOSTOR® CS10 (a uniquely formulated serum-free, animal component-free, and defined cry opreservation medium containing 10% dimethyl sulfoxide (DMSO)). Additionally, cryoprotectants / cryoprotectant additives which can be include in a cell composition (for cryopreservation) are known in the art and include, ethylene glycol (EG), antioxidants such as taurine, Metformin, gamma amino butyric acid (GABA). Cells may be suspended in a "freeze medium" such as cell culture medium containing 15-20% fetal bovine serum (FBS) and 7-10% DMSO, with or without 5-10% glycerol, at a density, for example, of about 1-10 x 106cells / ml. The cells are dispensed into glass or plastic vials, which are then sealed and transferred to a freezing chamber of a programmable or passive freezer. The optimal rate of freezing may be determined empirically. For example, a freezing program that gives a change in temperature of -1 °C / min through the heat of fusion may be used. Once vials containing the cells have reached -80 °C, they are transferred to a liquid nitrogen storage area.

[0098] Cryopreservation and Resuscitation of natural aquatic prokaryotic communities is described in Rain-Franco, et al., Front. Microbiol., Volume 11 | Article 597653 (2021).

[0099] B. Solid compositions

[0100] In some forms, consortia or compositions of microorganisms (such as a composition of consortium 1 or 2) are present in dry compositions such as dry powder, grains or granules or pellets. Dry compositions can be prepared by adding an Osmo protectant (such as sugar, egg trehalose and / or maltodextrin) to the microbial composition in solution in the desired ratio. This solution is combined with a dry carrier or absorbent such as an organic or inorganic carrier (discussed further below) at the desired concentration of the microorganism composition (such as 2-30%, e.g. 2.5-10%, 5-15%, 7.5-20% or 15-30%). Granules can be created by including clay or polymer binders that serve to hold the granules together or impart specific physical or degradable properties. Granules can be formed using rotary granulation, mixer granulator or extrusion, as several possible methods. In other examples, dry compositions can be prepared by spraying or soaking a liquid microorganism composition on / in a solid carrier. Additional methods for preparing dry compositions comprising one or more microorganism species are known to one of ordinary skill in the art, for example, as described in Formulation of Microbial Biopesticides: Beneficial Microorganisms, Nematodes and Seed Treatments, Burges, ed., Springer Science, 1998; Bashan, Biotechnol. Adv. 16: 729-770,1998; Ratul et al., Int. Res. J.Pharm. 4: 90-95, 2013.

[0101] In some forms, the microbial cells can be immobilized on a solid support to enhance delivery efficiency. Immobilization of cells offers numerous advantages, including maintenance of their metabolic activity owing to better-controlled chemical conditions and physical protection. Also, slow release of cells from a solid support can reduce the need for constant reinoculation. Additionally, immobilized cells will disperse less widely compared with free-living cells, thereby effectively reducing potential issues concerning escaped microorganisms affecting or interacting with nontarget organisms. Microorganisms can be immobilized in carriers made from three types of material: inorganic, organic, or a composite of both. Organic carriers include naturally occurring products (e.g., coconut husk, chitin, alginate, agar, and guar gum) and synthetic polymers (e.g., polypropylene, polyurethane, and acrylamide). Inorganic carriers include charcoal, clay, silica, ceramics, and celite, among others. Whereas inorganic carriers are usually more resistant, less costly, and easier to handle and reuse, organic carriers are more adaptable and typically biocompatible and therefore safer for other organisms Composite carriers, which combine the advantages of both organic and inorganic carriers, provide some level of compromise between these respective (dis)advantages and include substances such as polyvinyl alcohol, sodium alginate, and polyvinyl alcohol-guar gum. Another option is bioencapsulation, where living carriers, such as the live feed organisms Artemia spp., rotifers, or copepods, are used to successfully deliver nutrients, immune stimulants, and probiotics to several aquatic organisms.

[0102] III. METHODS OF USE

[0103] Methods of protecting coral are provided, by providing the coral with microorganisms which confer a combination of beneficial phenotypes required to withstand stress in their environment. Corals are constantly under threat of exposure to toxic compounds, including those produced directly by anthropogenic activity, such as sewage pollution and oil spills and those produced by the coral meta organism itself during episodes of stress

[0104] The disclosed consortia include a combination of microorganisms which can contribute to one or more of coral nutrition and growth, mitigation of stress and toxic compounds, pathogen control, and early life development. . ROS produced by corals and their endosymbiotic algae under high temperatures seem to be capable of inducing coral bleaching. Microbial nitrogen fixation serves as a nitrogen input for the meta organism, alongside heterotrophy, and is an important source of nutrients for both corals and Symbiodiniaceae . The activity of microbial enzymes that participate in other phases of nitrogen cycling can help to control its distribution in the meta organism, limiting nitrogen entry into the Symbiodiniaceae and controlling pathogens and the symbiosis between algae and corals. In parallel, corals living in waters with excess dissolved inorganic nitrogen have shown greater susceptibility to thermal stress, thus, of microorganisms that cycle this element are important for protection against bleaching (mitigation of stress or toxic compounds. (Peixoto, etal., Annu. Rev. Anim. Biosci. 2021. 9:265-88)

[0105] The disclosed methods rely on the synergistic and interdependent relationship between a macroscopic host organism and its associated microorganisms, which forms a meta organism, or holobiont. In this relationship, the host and its microbiome act as a biological unit to constitute the resulting phenotype and maintain the fitness of the meta organism.

[0106] The methods include delivering to coral, an effective amount of the disclosed compositions containing Beneficial Microorganisms for Coral (BMC). All corals have mouths, and different types of corals have different sized mouths and benefit from getting the appropriate sized foods. Large Polyp Stony corals along with colonial polyps typically have larger mouths so they can easily consume pellet or frozen foods. Many soft corals and Small Polyp Stony corals have much smaller mouths and will require liquid or powder foods with smaller particles.

[0107] The disclosed methods can be applied in coral reef aquaculture ex situ (in aquarium) and in situ (at sea / ocean). Delivery options include via water column, live feeds or inert diets, or via slow-release devices.

[0108] Delivery strategies for probiotic bacteria need to target the host micro-niche where their putative beneficial function is required. If colonization of tank surfaces or external host surfaces such as coral mucus is desired, then delivery into the water column by running water over a biofilter or by adding suspended free-living bacterial cells is a viable option in smaller aquaria (Thatcher, et al., Current Opinion in Biotechnology 2022, 73:380-386). Initial investigations have demonstrated that enrichment of rotifers with probiotic cocktails is a viable delivery method for the coral Pocillopora damicornis

[0109] In some forms, the coral is inoculated with an effective amount of the disclosed compositions as demonstrated in the Examples below. Spot feeding with a bulb syringe or feeding device or inoculation with a syringe can be used as a method of delivering the BMC to the coral.

[0110] In some forms, the coral is a red sea coral, for example, acropora, gorgonia, tubastrea, whip coral or fingia. However, the BMC compositions disclosed herein can be beneficial to coral whose natural habitat is not the red sea. EXAMPLES

[0111] EXAMPLE 1: In situ application of a probiotic consortium in the hard coral Pocillopora verrucosa and its response in the coral associated microbiome and surrounding environment.

[0112] Materials and Methods

[0113] Bacterial isolation and probiotic consortium preparation

[0114] Coral fragments from Stylophora pistillata Clade IV, Galaxea fascicularis and Pocillopora verrucosa, were collected by snorkeling and Scuba diving at Thala reef (22°15'46.9"N 39°03'05.9"E), Aquarium (22°23'15.6"N 38°55'07.2"E) and Al Fahal reef (22°18'18.4"N 38°57'52.5"E) respectively, in the central Red Sea, at depths between 1 to 10 m, between February and May 2021. Coral fragments were collected using pliers and transported inside 50 ml falcon tubes on ice to the laboratory. Immediately, the fragments were macerated using 50 ml of 3.5 % sterile saline solution with a sterile mortar and pestle. Serial dilutions up to 10-6 were done using the macerate paste with 3.5% saline solution, and 100 pL of each dilution was plated in Marine Agar (MA) (Zobell 2216, HiMedia Laboratories, Mumbai, India) diluted Marine Agar (DMA), and Lenox Broth (LB) (Sigma- Aldrich®). Plates were incubated at 25°C (corresponding to the in-situ water temperature registered at the sampling sites) overnight. In parallel, the macerate was incubated in a 250 mL sterile Erlenmeyer at 25°C overnight with crystal beads at 130 rpm. After this first incubation, triplicate subsamples (100 pL) of 10'4, 10'5, 10'6dilutions were plated into Petri dishes containing approximately 20 mL of Marine Agar media and diluted marine agar medium (Marine Agar Medium 2* diluted with 3.5% NaCl and agar adjusted). Alternatively, 0.5 cm coral fragments were cut and placed on the petri dishes, with both media. All the plates were incubated at 25°C for at least 48 h or until visible bacterial colonies were observed. Approximately, > 350 bacterial isolates were obtained, based on colony morphology, and were preserved at -80°C using sterilized glycerol with a final concentration of 20%, for further analysis.

[0115] 16S rRNA gene sequencing of bacterial isolates

[0116] Each bacterial isolate from the glycerol stocks was re-growth using 200 pL of the stock and inoculated into 6 mL of Marine broth (HiMedia Laboratories, Mumbai, India), and incubated overnight at 26°C with 140 rpm agitation. For bacterial DNA extraction, 2 ml of bacterial liquid culture were centrifuged for 5 minutes at 10.000 rpm to obtain a pellet and washed twice with 3.5% saline solution to wash the cells from the culture media. Bacterial pellets were stored at -20°C until DNA extraction procedures. The DNA extraction was performed using the Wizard® Genomic DNA purification kit (Promega Corporation, USA), following the protocol for Gram-positive and Gram-negative bacteria. Genomic DNA was purified with the GFX™ PCR DNA and Gel band purification Kit (Cytiva Company, USA) and then quantified using Nanodrop™ 8000 Spectrophotometer (Thermo Scientific™) and Qubit™ dsDNA assay kit (Invitrogen™). To target the full 16S rRNA gene, universal primers 27F and 1492R were used (Frank et al., 2008), using the AmpliTaq Gold® 360 Master Mix (applied Biosystems®, by Life Technologies™) with the following PCR conditions: One cycle of initial denaturation at 95°C for 5 min, 30 cycles of denaturation at 95°C for 1 min, annealing at 50°C for 1 min, extension at 72°C for 1 min, and one cycle of final extension at 72°C for 7 min. PCR products were verified using 1% agarose gels and visualized in a Bio-Rad® transilluminator. PCR products were sent for Sanger sequencing to Macrogen (Korea) for taxonomic identification by Sanger. The forward and reverse sequences (1000-1500 bp) were processed to remove low-quality bases and generate contigs using the ChromasPro software. Ambiguities in the assembled sequences were resolved visually (either by choosing the base from the read with the cleaner signal or changing the consensus base to “N”). Cleaned assembled DNA sequences from each of the BMC isolates were then identified using the EzBioCloud server (Yoon et al., 2017). The top-hit taxon, obtained from average nucleotide identity (ANI), was used to estimate the taxonomy of the BMC isolates.

[0117] Functional screening of bacterial isolates

[0118] Bacterial strains identified as potential human or coral pathogens (e.g. Vibrio spp.) were excluded (n=305). The remaining bacterial isolates were tested for beneficial functions (Rosado et al., 2017 and Santoro et al., 2021), as follows:. For reactive oxygen species (ROS) scavengers, 20 pL of pure culture of each bacterial strain was placed on a portable microscopy layer, and immediately, a drop of 3% (v / v) of hydrogen peroxide was added in the middle of it. The criterion of a positive result was evaluated qualitatively, based on the production of bubbles, as a proxy of catalase reaction. Phosphate assimilation was tested according to Nautiyal, 1999, using Pikovskaya’s agar culture media (HIMEDIA®). The antagonistic effect against Vibrio coralliliticus (well-known coral pathogen) (Ben-Haim et al., 2003) was assessed through the diffusion agar method (Marval et al., 1990): First, 20 pl of each bacterial strain was spot-inoculated onto 2.5% NaCl LB agar, placing three spots for each strain (representing replicates). The plates were incubated at 26°C for as long as necessary for the strain to grow. The strains were inactivated by chloroform volatilization, followed by pouring 3 ml of semisolid 2.5% NaCl LB medium (0.7% agar) containing the Vibrio indicators over the inactivated spots. These plates were then incubated at 28°C for 16 hours, and the antagonistic activity was indicated by inhibition halos around or no detection of Vibrio growth over the colony spot. Genes related to nitrogen cycling, as nitrogenase subunits (nifH) and denitrification (nirK), as well as the DMSP degradation (dmdA) gene were screened through whole genome analysis.

[0119] Selection of BMCs and Probiotic preparation

[0120] Six bacterial strains (one Halomonas sp., two Pseudoalter omonas galatheae, two Cobetia amphilecti and one Sutcliffiella sp.) were chosen for the probiotic consortium, based on their beneficial traits (Table 2). The selected strains were sequenced by PacBio (Bioscience CoreLabs, in KAUST, Saudi Arabia) for screening of their whole genomes (Raimundo et al., In Prepl) The genomic sequences for each bacterial strain in the consortium are deposited in the European Nucleotide Archive (ENA) under the Project ID PRJEB62849 (https: / / www.ebi.ac.uk / ena / browser / view / PRJEB628490); the accession numbers for each bacterial strain in the consortium is provided in PRJEB62849 are provided in Table A, above.

[0121] Thus, the disclosed BMC in some forms, includes Pseudoalteromonas galatheae (BMC1 genome); Pseudoalteromonas galatheae (BMC2 genome); Cobetia amphilecti (BMC2 genome); Cobetia amphilecti (BMC4 genome); Halomonas sp. (BMC6 genome); and Sutcliffiella sp. (BMC8 genome) as shown in the Table above..

[0122] As the probiotic consortium is composed of a diverse combination of bacteria, each strain was collected proportionally at the peak of the exponential growth phase, to achieve a concentration of 109cells / ml per strain. Fresh bacterial cultures were centrifuged at 6000g for 5 min, and the cell pellets were washed three times with saline solution (3.5% NaCl), followed by centrifugation and resuspension in 1 L of saline solution for a final concentration of approximately 10'9cells / ml per strain. The probiotic was stored at 4°C until in-situ inoculation in the corals.

[0123] Table 2. Summary of the BMC traits results for the six selected BMCs for the probiotic consortium. (nifH) nitrogen fixation; (nirK) denitrification; catalase activity; siderophores production; phosphate assimilation. (-) indicates a negative result, (+) indicates positive activity. For the catalase test, results are denoted with: (+), (++) and (+++) from less to high production of bubbles, as a proxy of the catalase reaction.

[0124] Experimental design

[0125] The experimental design workflow is shown in FIG. 1. Pocillopora verrucosa healthy colonies were randomly assigned to the Probiotic and Control treatments (n=15 per treatment).

[0126] The experiment was performed in Al Fahal reef, Thuwal, Saudi Arabia, with coordinates 22°18'18.4"N 38°57'52.5"E, at a site known as the “Coral Probiotic Village” (Garcias-Bonet et al. In Prepl), 15 km off-shore from King Abdullah University of Science and Technology, in the central Red Sea. Thirty visually healthy colonies (no visual signs of bleaching or disease) of P. verrucosa of the brown morphotype were selected for the experiment and were randomly assigned to the probiotic and control treatments (n= 15 colonies per treatment) (FIGS. 16A-16B). Colonies were qualitatively evaluated for their health status throughout the experiment, using a coral health chart to access signs of bleaching. Colonies were tagged with their corresponding treatment and monitored for their health status during the experiment. The experiment was performed covering seasonal variations, from summer to winter 2021, and late spring 2022. Three main sampling points were considered for analysis: Tl, before the inoculations (late August 2021), T2 after one and a half months of inoculations (mid-October 2021), T3 at the end of the inoculations (late November 2021) and T4 after five months from the last inoculations (April 2022) (Table 1, Figure 2). Inoculations were performed three times per week using 50 ml plastic syringes containing 30 ml of the probiotic consortium, released slowly all over the coral colony. The corals may uptake the bacteria from the solution while it is in contact with the coral. As this is a non-invasive delivery method, needles were not used. Instead, inoculations were applied around the coral and were performed regularly to increase the chances for the bacteria to be taken up and / or interact with the coral holobiont. It is believed that corals can incorporate some of the probiotics in a similar manner as other wildlife probiotic applications. The control treatment consisted of a 3.5% NaCl solution and was applied the same way.

[0127] Samples of coral fragments from each colony were collected for coral-associated bacterial community analysis at all sampling times, using pliers and collection bags, and were kept in sterile 5 mL cryovials with DESS buffer (Gaither et al., 2011; Gray et al., 2013) and immediately snap-frozen in liquid nitrogen on the boat. Samples were transported to the laboratory (less than 3 hours after collection) and stored at -80°C until further processing.

[0128] In parallel, sediments and water surrounding the coral colonies were taken at T1 and T3, to monitor their bacterial communities and to assess water nutrients and dissolved organic carbon (DOC). For this, 10 colonies (using a random number generator) distributed in different areas of the experimental study site (five of each treatment) were sampled: sediment samples were collected right at the bottom of each colony, using sterile 50 ml falcon tubes, between 1-5 cm depth. Water samples were collected at approximately 30 centimeters distance to the same colonies, using 2 L dark bottles that were acid-washed prior to the sample collection. Samples were stored on ice on the boat and filtered the same day in the laboratory using a filtration rack with 0.22pm Millipore Sigma membranes attached to a vacuum pump. All equipment and materials used were thoroughly sterilized to avoid contamination. Filter membranes were individually stored at -80°C for less than two weeks until DNA extraction.

[0129] Monitoring of in-situ physicochemical parameters Multiparameter CTDs (Ocean Seven 310 Multiparameter CTD, Idronaut) were deployed in two locations within the experimental area to monitor temperature, salinity, and pH. CTDs were placed on metal frames, pointing north, and replaced every two-three weeks to retrieve the data and replaced with another one, to cover seasonal variations. The average data for seawater temperature are summarized in Table 3.

[0130] Table 3. Date of sampling (dd-mm-yy), summary of procedures, biological components and water temperature across the experiment. Mean, Minimum (Min), and Maximum (Max) values for seawater temperature are shown. Sampling Date of Visual assessment, CBASS, Biological Mean Water time sampling in-situ PAM components temperature

[0131] (Min - Max)(C°)

[0132] T1 24-08-2021 Yes Coral, sediment, 31.97 water (31.64-32.39)

[0133] T2 06-10-2021 Yes Coral 29.84

[0134] (29.58-30.17)

[0135] T3 23-11-2021 Yes Coral, sediment, 28.72 water (28.54-29.01)

[0136] T4 18-04-2022 Yes Coral 28.31

[0137] (28.21-28.46)

[0138] Inorganic nutrient analysis

[0139] Seawater collected during T1 and T3 was used for inorganic nutrient analysis of the surrounding water of each of the randomly selected coral colonies (n= 10 per treatment). Briefly, water was filtered on the boat with 0.22pM Millex®-GV filters (PVDF Membrane, Merck Millipore Ltd., Ireland), into 15mL falcon tubes. Following that, samples were placed on ice on the boat, and then frozen at -20°C until analysis. Inorganic nutrients analyzed were: Silica (Si(OH)4), Nitrite (NO2-), Nitrate (NO3-), and Phosphate (PCU3'). All measurements were performed using a segmented flow analyzer (Model AA3 HR, SEAL Analytical Inc.) with the following detection limits: Silicates 0.08322pmol.L-l; Nitrite 0.0217 pmol.L-1; Nitrate 0.0322 pmol.L-1; Phosphate 0.01052 pmol.L-1.

[0140] Coral Bleaching Automated Stress System (CBASS) experiments.

[0141] To evaluate the behavior of the coral’s heat response during the experiment, shortterm acute heat stress assays were performed to determine coral thermotolerance in genotype (per-colony) level. Here, CBASS was used as a proxy to assess coral health and if the longterm inoculation of coral probiotics had an effect on coral’s thermal tolerance threshold, as well as their natural thermotolerance variation during a seasonal time frame. In Tl, T2, T3, and T4, four fragments of each colony were collected. The fragments were transported in seawater to the wet lab facility of the Coastal and Marine Resources Core Lab (CMRCL, KAUST), where the set-up was ready to receive the corals. Briefly, the system consists of four 10L flow-throughs supplied with raw seawater collected from the site a day before the runs. Each tank runs different regimes temperature independently and the light setting was adjusted to correspond to in-situ irradiance (600 pmol photons m-2 s— 1), which was adjusted using an LI- 193 Spherical Underwater Quantum Sensor (LI-COR) and manual adjustment of dimmable 165W full spectrum LED aquarium lights (Galaxyhydro). The lights followed a 12: 12 hr day / night cycle. The temperature of each tank was controlled using the ITC-310T-B (Inkbird) thermostat connected to an IceProbe Thermoelectric chiller (Nova Tec) and 200W titanium aquarium heaters (Schego). HOBO Pendant Temperature Data Loggers were placed to record every 10 minutes the temperature of each tank during the experiment. One fragment corresponding to each colony was distributed with a different temperature condition. The temperature regime of each tank was 1 control / baseline - 30°C, 1 medium - 33 °C, 1 high - 36°C, and 1 extreme - 39°C. The CBASS assays run for 18 hours, where the temperature of the 30°C tank was maintained at 30°C for the entire experiment, while in the other tanks, the temperature was increased to 33 °C, 36°C, and 39°C, respectively, and then returned to 30°C overnight until the end of the experiment. The detailed temperature profiles are provided in FIG. 17. After 7 hours of the beginning of the experiment (and one hour in darkness), all fragments had their endosymbiotic algae photosynthetic efficiency (Fv / Fm) measured using a pulse-amplitude modulation (PAM) fluorometry (Diving PAM II, Walz). The measurement also matched the temperature ramping down to 30°C. The data were analyzed according to Voolstra and collaborators (2020) (C. R. Voolstra et al., 2020), where Fv / Fm values were used to evaluate the treatment’s ED50, corresponding to its thermal threshold.

[0142] In-situ PAM for coral health monitoring

[0143] The photochemical efficiency of the Symbiodiniaceae was assessed through the maximum quantum yield of PSII photochemistry Fv / Fm. For this, a diving-PAM system (Diving PAM II, Walz) with a red-emitting diode was implemented (LED; peak at 655 nm). PAM data was collected after sunset, at least 30 min after complete darkness, to ensure there was full photochemical dissipation of the reaction centers. The diving PAM was configured as follows: measuring light intensity = 6; gain = 2; and damping = 4. Statistical analyses were performed using a two-way ANOVA in R studio (R Core Team) to test for differences in the thermal tolerance between treatments.

[0144] DNA extraction, library preparation and sequencing of bacterial communities DNA from the coral fragments was extracted using the DNeasy® Blood & Tissue kit (Qiagen), with the Gram-positive bacteria pre-treatment, according to manufacturer instructions, with the following modification: coral fragments of approximately 0.5 g were used directly for the extraction. The lysis incubation step after adding the proteinase K, was carried out overnight for approximately 16 hours, at 56°C, with a constant agitation of 650 rpm in a Thermomixer (ThermoFisher®).

[0145] DNA extraction of water samples was done using the DNeasy® Blood & Tissue kit (Qiagen) by cutting the filter in small pieces with a sterile cutter and tweezers. The protocol was done following manufacturer instructions with the following modifications: At the sample pre-extraction preparation stage, half of the membrane filter was cut into smaller pieces, and then placed in 1.5mL microcentrifuge tubes. The volume of all the following solutions on the kit was adjusted to similar proportional volumes (thus not changing any concentration of compounds) to fully immerse all the membrane filter pieces into solution: ATL buffer added was 540pL, and Proteinase K was 60pL. After adding these solutions, the incubation step at 56°C was conducted for three hours. Then, the volumes of buffer AL and ethanol was 400pL, buffer AW1 and AW2 was 500pL, and the final elution buffer AE was 50pL. DNA samples were stored at -20°C until downstream analyses.

[0146] DNA of sediment samples were extracted using the DNeasy® PowerSoil Kit (Qiagen), with the following modification: 12.5 pL of Proteinase K was added to approximately 0.5 g of sediments, for incubation overnight at 56°C, with a constant agitation at 650 rpm, in a Thermomixer. The downstream steps were performed according to the kit’s protocol. DNA concentration and purity for all samples were quantified using the Qubit™ dsDNA assay kit (Invitrogen™) and Nanodrop™ 8000 Spectrophotometer (Thermo Scientific™). Sequencing of the V3-V4 regions of the 16S rRNA gene using universal primers 34 IF and 785R for the coral, sediments, and water samples, at Novogene Corporation-Inc, in China, in a NovaSeq platform. In brief, PCR mixtures contained 15 pL of Phusion® High-Fidelity PCR Master Mix (New England Biolabs), 0.2 pM of each of the forward and reverse primers, and 10 ng of the samples’ genomic DNA. The thermal cycling conditions were as follows: A first denaturation step at 98°C for 1 minute, followed by 30 cycles at 98°C for 10 seconds, 50°C for 30 seconds, and 72°C for 30 seconds, and a final extension of 5 min at 72°C. PCR products were verified and quantified by mixing an equal volume of IX loading buffer (contained SYB green) with PCR products and performing electrophoresis on 2% agarose gels. For the library preparation, PCR products were purified using Qiagen Gel Extraction Kit (Qiagen, Germany). Sequencing libraries were generated with NEBNext® Ultra™ II DNA Library Prep Kit (Cat No. E7645). The library quality was evaluated on the Qubit@ 2.0 Fluorometer (Thermo Scientific™) and Agilent Bioanalyzer 2100 system. Libraries were sequenced on a NovaSeq platform (Illumina) and 250 bp paired- end reads were generated.

[0147] Bacterial community analyses.

[0148] The DADA2 pipeline was used to infer Amplicon Sequence variants (AS Vs) (Callahan et al., 2016) using the 16S rRNA gene-based amplicon libraries of coral, sediment and water. In short, the raw reads were decontaminated of phiX, and adapter-trimmed using the “BBDuk” tool from the BBMap suite (Bushnell B, http: / / sourceforge.net / projects / bbmap / ). PCR primers were then removed from the reads using the “cutadapt” tool (Martin, 2011). After performing concatenation of the forward and reverse reads via “justConcatenate” option in the “mergePairs” function of DADA2, the sequences were analyzed under the pseudo-pooling mode by following the standard DADA2 (version 1.22) workflow and by using the SILVA database, version 138.1 (Quast et al., 2012). The potential contaminant ASVs that were identified in the negative controls and the study samples were removed from the analysis by the “decontam” tool (Davis et al., 2018) using the prevalence-based method (on the default threshold setting).

[0149] For downstream analysis, the coral, sediment and water ASVs reads were processed separately. Analysis of alpha and beta diversity, diversity plots, ordinations and statistical comparisons were done in R version 4.2.2 (R Core Team, 2018) using the functions in Phyloseq (McMurdie & Holmes, 2013) and Vegan version 2.6-4 (Oksanen, F.J., et al. (2017). All plots were generated using ggplot2 version 3.4.0.

[0150] In brief, reads corresponding to mitochondria, chloroplast and singletons were removed., resulting in 46803 ASVs for the coral dataset, 34099 for the water, and 68354 for the sediment.

[0151] Alpha and beta diversity were analyzed using the rarefied ASV counts to the minimum sample depth (18364 reads for corals, 46210 for sediments, and 99850 for water) with the function “estimate. diversity” from Phyloseq, with the default diversity indices (Observed S, Shannon H’, Simpson, and Chaol). Statistical comparisons for the beta diversity metrics were calculated by implementing the Kruskal-Wallis test, previous testing of the null hypothesis for normal distribution of the data (Shapiro-Wilks p-value < 0.05). Non-metric Multidimensional Scaling (nMDS) analyses were generated from Bray-Curtis distances of Wisconsin-square root transformed ASV raw counts using the “vegdist” and “metaMDS” functions in Vegan. The homogeneity of variances between treatments in each sampling time was tested using the “betadisper” and “permutesf ’ functions in Vegan using Bray-Curtis distances and 999 permutations. Statistical differences were assessed using PERMANOVA with “sampling time” and “treatment” as factors, implementing the “adonis2” function in Vegan, from generated Bray-Curtis distances, with 999 permutations.

[0152] To identify differentially abundant ASVs between “control” and “probiotic” treatments, the Analysis of Composition of Microbiomes with Bias Correction (ANCOM- BC2) function (Lin & Peddada, 2020) was used on ASVs absolute counts, using a conservative approach, with the Benjamini -Hochberg (BH) method to correct for false positives and adjusted p-value < 0.01. An ASV was considered significant in the probiotic treatment, when it was significantly more abundant than the “control” (reference) treatment under the aforementioned parameters.

[0153] Species indicator analysis (De Caceres & Legendre, 2009) was performed using the package “indicspecies” to identify bacterial taxa that were associated with the probiotic treatment, after checking if differences were observed between treatments with the PERMANOVA. Radial plots were used to show the top 5 indicators at the family level, using ggplot2.

[0154] Coral calcification.

[0155] Coral calcification of the inoculated colonies was assessed as a proxy of the coral’s health during T1 and T3. For the determination of the calcium carbonate precipitation rates of the coral colonies, the total alkalinity anomaly method technique was employed (Chisholm & Gattuso, 1991). To do so, 150-160 g of seawater filtrated by 0.22 pM pore membranes (MF- Millipore) was weighed on a precision balance and placed on 35 different 250 mL glass beakers, of which five were used as blank controls and the other 30 were used to incubate one individual coral fragment each, corresponding to the 30 studied coral colonies. Before the incubations, three borosilicate vials were filled with the initial water for comparison of alkalinity values with the water used in the incubations. The beakers were closed with transparent plastic films, and an open cell incubation of the coral fragments was performed inside a I-22VL incubator (Percival-Scientific, USA) over a period of three hours under a temperature of 26°C, a light intensity of 200 PAR and a 100 rpm rotation. After the incubation, the water from the beakers was collected, placed into borosilicate glass vials and subsequently treated with five drops of mercury chloride for long-term storage before the determination of its total alkalinity. For determination of total alkalinity, samples were analyzed at the Integrated Ocean Processes Lab (KAUST). Sampling was done according to Dickson and collaborators (Dickson et al., 2007). Two parameters of the CO2 system were measured from the same sample. The total alkalinity was measured by open-cell titration with a Titrator (Titrino, Metrohm, Herisau, Switzerland). The dissolved inorganic carbon (DIC) was measured using an AS-C6L DIC analyzer (Apollo Scitech, Newark, USA). Instruments were calibrated using certified reference materials from A. Dickson laboratory at SCRIPPS Institution of Oceanography (La Jolla, USA). The CO2 system was calculated from the two variables and the error propagation was estimated using the functions "carb" and "error" of the R package "seacarb".

[0156] Results

[0157] FIG. 5 shows Bacterial community of Pocillopora verrucosa. Significant differences between Probiotic-treated corals vs controls in T3 (Late November).

[0158] FIG. 6 shows beta-diversity. Significant differences between Probiotic-treated corals vs controls in T3 (Late November) FIG. 7 Shows bacterial community composition. Genera with relative abundances greater than 2% are shown. Bacterial composition was dominated by Endozoicomonas across sampling times and treatments.

[0159] Coral ’s thermal tolerance in CBASS experiments

[0160] No significant differences were detected between treatments in the photochemical efficiency at different temperatures in CBASS experiments (Two-way ANOVA, p-value > 0.05) (FIG. 8A).

[0161] In-situ photochemical efficiency of corals (In-situ PAM)

[0162] Although treatment had no statistical significance on the PAM values (Two-way ANOVA, p-adj. > 0.05), the in-situ PAM data increased in T3 (FIG. 8B), most-likely due to the decrease in water temperature during late November, at the beginning of the winter period which could be attributed to seasonal variations (Ansari et al., 2022) characterized by seawater temperature changes, which affect thermal tolerance of corals (Evensen et al., 2021; Osman et al., 2018; Sawall et al., 2014; C. Voolstra et al., 2020).

[0163] Bacterial community of P. verrucosa

[0164] Overall, there was a significant effect of sampling time on the bacterial communities associated to P. verrucosa (Adonis p-adj. 0.001***) (Figure 18 A). Treatment had a significant effect on T3 samples (Adonis p-adj. < 0.05) whereas for the other sampling times (Tl, T2, T4) bacterial communities were not affected by treatment (Adonis test, p-adj. > 0.05) (Figure 18C). Analysis of the Beta-diversity revealed a significantly higher Shannon index in the probiotic treatment vs control (Kruskal-Wallis test, p-adj. < 0.05) (Figure 18B) also in T3. Bacterial taxa associated to P. verrucosa were dominated by the phylum Proteobacteria (42.6%), followed by Bacteroidota (11.7 %), Plancomycetota (8.9%) and Verrucomicrobiota (6.2%). The most abundant families regardless of the treatments were Endozoicomonadaceae (20%), followed by Woesiaceae (4.3%), Desulfosarcinaceae (3.8%), Pirellulaceae (3.4%) and Flavobacteriaceae (3.2%). These families were observed in very similar abundances in both treatments (Endozoicomonas (10.14% & 9.8%), Woesiaceae (2.16% & 2.15%), Desulfosarcinaceae (1.9% & 1.9%) and Pirellulaceae (1.7% & 1.7%) in control and probiotic treatments, respectively (Figure 2 ID).

[0165] Bacterial community response to the probiotic treatment

[0166] Halomonas spp increased significantly in T3 and T4 in probiotic-treated colonies, Cobetia spp abundance was higher in T2 only and it was not detected in T3 and T4. Pseudoalter omonas spp only showed significant increase in T2 (higher in control -treated colonies) and T3 (higher in probiotic-treated colonies). The genus Sutcliffiella sp, corresponding to one of the inoculated BMC, was not detected in any of the sampling times in the coral samples.

[0167] Bacterial communities of water and sediment

[0168] Water bacterial communities showed to be significantly heterogeneous during all sampling times regardless of the treatment (Adonis p-adj. 0.001***), whereas the sediments- associated bacteria remained more stable during the sampling times and was not affected by treatment either (Adonis p-value > 0.05)

[0169] FIG. 19A is a CAP plot shows water bacterial community structure by sampling time and treatment. FIG. 19B shows Shannon diversity index by treatment and sampling time (*=significant difference). FIG. 19C are MDS showing community structure of probiotic vs. controls in T1 and T3. FIG. 19D shows a CAP plot by sampling time and treatment of sediment bacterial communities. FIG. 19E shows Shannon diversity index by sampling time between probiotic and control treatments (ns= not significant differences). FIG. 19F is an MDS showing ordination by treatment for T1 and T3 of sediment bacterial communities. FIG. 19G is a heat map of bacterial community composition at the family level, of water (left panel) and sediment (right panel) (families with > 2% of total abundance are shown).

[0170] Effect of probiotic inoculation and seasonality on coral, sediment, and water bacterial communities

[0171] The bacterial communities of P. verrucosa, marine sediment, and reef seawater were investigated during a long-term BMCs inoculation, from summer to winter 2021, and monitored until spring 2022, in the Red Sea, to assess the microbiome response to the inoculation of a probiotic consortium and the seasonal variations. The results showed that the bacterial community composition in the coral, water, and sediment are significantly different (Adonis-test p-value 0.001***) and have a differential response to seasonality and to the probiotic inoculation (FIGS. 20A-20F). For instance, there was a significant effect of sampling time on the bacterial communities associated with P. verrucosa (Adonis p-adj. 0.001***) and the interaction between sampling time and treatment (Adonis p-value 0.04*) (Figure 23B). Treatment had a significant effect on T3 samples (late November, at the end of the inoculations) (Adonis p-value 0.005**) and not in the other sampling times (Tl, T2, T4) (Adonis p-values > 0.05) (Figure 23C-23F). The analysis of variance showed significant differences in T3 samples (Beta-disperse test p-value 0.018*), and statistical analysis of the alpha-diversity metrics (Kruskal-Wallis test) showed a significant increment in diversity in all evaluated diversity metrics in T3 in probiotic compared with control (Richness (S) (p- value = 1.27e-05***), Shannon (H’) (p-value = 0.0005***), Simpson (p-value = 0.008**), and Chaol (p-value = 1.043e-05***) in T3 (Figure 24).

[0172] The water bacterial community showed to be heterogeneous through time and it was not significantly changed by the probiotic inoculation (Adonis-test sampling time p-value 0.001***; treatment p-value 0.07), whereas the sediments-associated bacteria remained stable during the time and weren’t changed significantly by treatment either (Adonis-test sampling time 0.07)

[0173] Probiotic application elicits coral-microbiome restructuring after a long-term in-situ inoculation

[0174] Significant composition changes occurred in T3 at the end of the probiotic inoculations between control and probiotic-treated corals (Adonis treatment, p-adj. 0.005**). Although there was a significant dispersion among samples (Betadisperse p-value 0.018*), differences were observed in the relative abundance in several groups of the dominant taxa (Figure 25). Overall, some of the top 10 most abundant families were significantly enriched in the probiotic-treated corals, such as Rikenellaceae (p-value 0.00012***), Prevotellaceae (p-value 2.8e-05 ***), Lachnospiraceae (p-value 8.6e-05***), and Rhodobacteraceae (p- value 0.00078). On the contrary, Endozoicomonadaceae ’s relative abundance decreased in probiotic (p-value 0.0078**), and other families like Simkaniaceae (p-value 0.074), Burkholderiaceae (p-value 0.76), Unclassified Alphaproteobacteria (p-value 0.14), Spirochaeta (p-value 0.66), and Phycisphaeraceae (p-value 0.48), did not significantly differ between treatments (Figure 25).

[0175] The thermal tolerance response of P. verrucosa was evaluated in CBASS experiments at different temperatures (30, 32, 36, 38C°) to assess the coral photochemical efficiency (Fv / Fm) as a health proxy of the symbiotic algae. No significant differences were detected between treatments at different temperatures (p-value > 0.05) (Figure 26A and 26B). In parallel, each coral colony was monitored during the duration of the experiment by measuring the photosynthetic efficiency (Fv / Fm) in situ. No significant effect of treatment was observed in any of the sampling times neither (p-value>0.05). EXAMPLE 2- Not only for corals: Beneficial Microorganisms for Corals (BMC) can also be acquired by sponges

[0176] The aim of this study was to assess if sponges can incorporate BMCs into their microbiomes in a 4-week in situ experiment, using Stylissa carteri and Callyspongia crassa as models.

[0177] Materials and Methods

[0178] Isolation and selection of BMCs

[0179] BMCs were isolated from three different coral hosts, namely, Stylophora pistillata, Galaxea fascicularis. and Pocillopora verrucosa, collected at Thala (22°15'46.9"N 39°03'05.9"E), Aquarium (22°23T5.6"N 38°55'07.2"E), and Al Fahal reefs (22°18T8.4"N 38°57'52.5"E) in the Saudi Arabian Red Sea25. Briefly, coral fragments were macerated and plated onto marine agar (Zobell 2216, HiMedia Laboratories, Mumbai, India) and Luria- Bertani broth (Sigma-Aldrich®, St. Louis, MO, USA) at 3.5% salinity. To identify and exclude potential coral or clinical pathogens (eg., Vibrio spp.) from >400 obtained isolates, the genomic DNA of the bacterial isolates was extracted using the Wizard® Genomic DNA purification kit (Promega Corporation, Madison, WI, USA) and the partial 16S rRNA gene was sequenced using the universal primers 27F and 1492R55and the Sanger sequencing approach on an ABI 3730x1 platform (Macrogen, Seoul, South Korea). More than 300 isolates were identified as potential coral or clinical pathogens and excluded from the next steps. The remaining strains were tested for their beneficial functions for corals, including catalase activity, siderophores production, urease activity, phosphate assimilation, and antagonistic effects against the well-known coral pathogen Vibrio coralliilyticus56’5'1, as previously described by Rosado et al. (2O19)20, Santoro et al. (2021)21, and Delgadillo- Ordonez et al. (2023)25. A total of six strains were selected as BMCs as they were positive to at least one of the aforementioned trait tests25. In Raimundo et al. (2024; see Example 3)58, the taxonomic identity of the six BMC strains was updated based on the full-length 16S rRNA gene sequences obtained by genome sequencing and was defined as follows: two strains of Pseudoalter omonas galatheae and Cobetia amphilecti and one strain of Halomonas sp. and Sutcliffiella sp. The selected BMC strains were stored in 20% glycerol at -80°C.

[0180] Production of the BMC consortium

[0181] In total, 1.2 L of the BMC consortium was produced weekly for the experiment, divided into three stages, namely pre-inoculum, inoculum, and consortium assembly25. Briefly, the pre-inoculum stage consisted of the activation of each bacterial strain from the frozen stock overnight in marine broth at 26°C and 140 rpm. For the inoculum stage, 1 mL of the pre-inoculum of each strain was added into separate 250-mL Erlenmeyer flask containing 99 mL of marine broth and incubated overnight at 26°C and 140 rpm. For the consortium assembly stage, each BMC strain was harvested during the peak of its exponential growth phase based on colony -forming units counting, to ensure an equal concentration of each strain in the cocktail25. The content of the Erlenmeyer flasks of each strain was separately centrifuged, washed, and resuspended in 3.5% NaCl solution. Next, 100 mL of each strain was added to a 1-L borosilicate flask (600 mL of BMC consortium in each flask) with a final concentration of 1 x 108'9cells / mL25and stored at 4°C for <1 week until in situ inoculation on the sponges. All steps excluding centrifugation were performed in sterile conditions inside a laminar flow cabinet.

[0182] Experimental design and field sampling

[0183] The 4-week in situ experiment was performed from May to June 2022 at the Al Fahal reef (photos not shown). The two most common sponge species in this reef, S. carteri and C. crassa, were selected for the experiment (photos not shown). In total, 30 sponges were tagged by SCUBA diving in the field and randomly assigned in quintuplicates to the three treatments: (1) control (no inoculation), (2) BMC inoculation once a week (BMC l x), and (3) BMC inoculation thrice a week (BMC 3x; schematic not shown; map not shown). Individuals of C. crassa measured approximately 40 cm in height, while S. carteri measured around 15 cm in height. Because of time limitations for sample processing, the experiment for S. carteri started 1 week after that for C. crassa but followed the same schedule.

[0184] Before initiating the experiment (TO), a square-like piece of sponge (approximately 16 cm2for C. crassa and 4 cm2for S. carteri) was cut with a scalpel and placed underwater into a 50-mL Falcon tube. For C. crassa, the top, middle, and base of the sponges were collected, whereas for S. carteri, only the top and middle ones were collected because of the small size of the individuals. In addition, 1 L of seawater was collected in a plastic bottle (cleaned with 10% bleach, 70% ethanol, and sterile Milli-Q water) next to each sponge replicate. Immediately after sampling, the sponge samples were washed in sterile artificial seawater to remove transient and loosely-attached microorganisms, squeezed into a 5-mL cryovial, and snap-frozen in liquid nitrogen (LN2). The seawater bottles were kept in a box with ice for 1 h until laboratory processing. The seawater samples were filtered onto 0.22-pm filters (47 mm, Millipore, Burlington, MA, USA) in a filtration apparatus (Sartorius [Gottingen, Germany] for C. crassa and Millipore for S. carteri). The filters were carefully folded, placed in 5-mL cryovials, and snap-frozen in LN2. Sponge and seawater samples were transferred from LN2 to a -80°C freezer until DNA extraction.

[0185] Three days after TO sampling, BMC inoculation was performed every Sunday (BMC 3x), Tuesday (BMC 3x) and Thursday (BMC 3x and l x) for 4 weeks for each species. Each plastic syringe (cleaned with 10% bleach and Milli-Q water) was filled with 30 mL of the BMC consortium and stored at 4°C 1 day before each inoculation. For each sponge individual, the content of one syringe (30 mL of the BMC consortium) was slowly and manually inoculated over the sponge body without any enclosing system (movie not provided). Due to the size differences between the sponge species, the probiotics were inoculated over the entire sponge body of S. carteri and only on the front part of the body of C. crassa. No inoculation was performed for control sponges. Photographs were taken during the experiment to document the regeneration / healing process of sponges after TO tissue sampling (photos not shown)

[0186] At the end of the experiment (Tl), the sponges and seawater were sampled three days after the previous BMC inoculation as described for TO, being C. crassa sampled on the front part where the BMC was being inoculated along the experiment. Approximately 1.5 months and two years after the end of the experiment, a nontoxic fluorescent dye (Fluorescein F2456, Sigma-Aldrich) was applied to the sponges to verify that they were pumping properly, which is a common health proxy used for sponges59. The in situ seawater temperature was continuously recorded using a Multiparameter CTD probe (Ocean Seven 310, Idronaut, Brugherio, Italy). The seawater temperature ranged from 27.71°C (lowest daily minimum temperature) to 30.71 °C (highest daily maximum temperature) throughout the experiment. The daily mean temperature was 28.26°C-30.34°C (Figure 15).

[0187] Microbiome analysis

[0188] Total DNA from the sponges and seawater was extracted using the DNeasy PowerSoil kit (Qiagen, Hilden, Germany) following the manufacturer’s protocol and stored at -80°C. DNA was quantified using a Qubit 4 (Invitrogen, Thermo Fisher Scientific Inc., Waltham, MA, USA) and a NanoDrop One (Thermo Fisher Scientific Inc.). The hypervariable region V3-V4 of the 16S rRNA gene was sequenced on the Illumina NovaSeq Platform (2 x 250 bp) by Novogene (Beijing, China) using the universal primers 341F (‘CCTAYGGGRBGCASCAG’60)(SEQ ID NO:1) and 806R (‘GGACTACNNGGGTATCTAAT’61) (SEQ ID NO:2). One replicate for the 5. carteri seawater sample failed Novogene’s quality control and was thus, not available for subsequent analyses.

[0189] Raw sequences were quality -filtered using Trimmomatic (version 0.38) in a sliding window of 4 bp and reads with a quality score below 15 or shorter than 100 bp were discarded62. Paired-end reads were merged using USEARCH (version 11.0.667), and contigs <325 bp or >550 were removed. Low-quality contigs (Q < 15) and contigs with more than one ambiguous base were also removed63. The UNOISE algorithm was used to cluster, denoise, and generate unique amplicon sequence variants (ASVs). Chimeras were removed using the UCHIME2 algorithm (high confidence mode;64). Taxonomy was assigned to the ASV sequences using the BLCA algorithm65with identity and coverage intervals set to 95%- 100% against the genome taxonomy database (GTDB, version 207;66). All non-bacterial and non-archaeal ASV sequences were removed, and reads were mapped to the remaining ASV sequences to estimate relative abundances. Data were normalized for read counts using the DESeq2 package67. Alpha (ASV richness and Shannon diversity H') and beta diversities were calculated based on normalized data using the R package vegan (version 2.6-4;68) in R software (version 4.1.3).

[0190] Data analysis

[0191] To potentially identify BMCs in sponge and seawater samples, their 16S rRNA gene sequences were searched using BLAST+ (version 2.9.0;69) against the ASV dataset. Only hits with a nucleotide identity of 100% against at least one of the 16S rRNA gene copies of BMC isolates were considered. Due to the limited resolution of 16S rRNA gene these identified ASVs are referred to as “BMC-related strains”.

[0192] PERMDISP analysis was performed using the R package vegan (version 2.6-4; function betadisper.^) to check data homogeneity. To compare alpha diversity metrics and detect differences in the relative abundance of the BMC strains in sponge and seawater samples, three-way univariate PERMANOVA based on Euclidean distance matrices was performed using the R package vegan (version 2.6-4; function adonis2 ) with the factors “sponge body part” (levels: top, middle and base), “time” (levels: TO and Tl) and “treatment” (levels: control, BMC 1 x, and BMC 3 x) with 9,999 permutations for the sponges, and two- way univariate PERMANOVA was performed with time and treatment as factors for the seawater samples. The post hoc pairwise test was performed with the R package RVAideMemoire (function pairwise. perm.manova^' once PERMANOVA generated significant results (p < 0.05). For beta diversity, nonmetric multidimensional scaling was performed using the R package viridis (version 0.6.4) to visually compare the Bray-Curtis dissimilarities of microbial communities. One-way and three-way PERMANOVA based on Bray-Curtis dissimilarity using 9,999 permutations was performed with the R package vegan (version 2.6- 4; function adonis2^') to detect significant differences in the microbial communities structure over time, treatment, and sponge body part. To determine which ASVs were affected by the BMC 1 x and BMC 3 x treatments relative to the control, univariate generalized linear models (GLMs) with a negative binomial or Poisson distribution (according to the abundance distribution of each ASV) were used using the R package mvabund. The default family used by mvabund for multivariate GLMs is negative binomial, which assumes a quadratic mean-variance relationship and a log-linear response. This assumption was confirmed, as the mean-variance relationship showed no patterns in the Residuals vs Fitted plot. In addition, the quantile-quantile (QQ) plot showed a straight line, indicating a good model fit. These GLMs accounted for multiple testing, correlations between ASVs, and compositional data71. Univariate GLMs were generated using “sponge body part”, “time,” and “treatment” as fixed factors. Differentially abundant ASVs between BMC treatment and control groups (regardless of the sponge body part) were analyzed in additional univariate GLMs to assess which microbial species were changing between these groups (control x BMC l x and control x BMC 3x). Differences were considered statistically significant at < 0.05. Significant ASVs related only to the factor time were not considered in the results, as they represent the natural variation of the microbiome. An increase or decrease in the relative abundance of the microorganisms for BMC treatments was only considered if the variation between T1 and TO in the treatment samples exceeded the natural variation in the control samples during the same period. Results from univariate GLM analysis for ASVs with relative abundance > 0.1% were plotted using Microsoft Excel, and analyses were conducted using R software (version 4.1.3).

[0193] Results

[0194] Sponges remained healthy during and long after BMC treatment

[0195] All sponge individuals were visually healthy during and after the experiment, with no external damage, necrosis or color changes detected (photos not shown). In addition, all individuals were pumping water 1.5 months after the experiments, excluding one C. crassa individual (BMC 1X_R2) that was not found after the experiment ended. After two years, all S. carteri individuals were found and observed to be actively pumping water (photos not shown). For C. crassa, all individuals found were also pumping, but some individuals from both treatments and control were not found (photos not shown).

[0196] Environmental and species-specific microbiomes

[0197] In total, 7,323 ASVs were obtained for the microbial assemblages of S. carteri, C. crassa, and seawater samples and were assigned to 40 phyla in the domains Bacteria (34; 94.4% ASVs) and Archaea (6; 5.6% ASVs; photos not shown). The average number of reads per sample was 83,744 ± 13,200 (photos not shown). Regardless of the sampling time and treatments, the sponges and seawater had different microbiomes, emphasizing the speciesspecific microbial communities usually found in sponges (Figure 9; one-way PERMANOVA, p = 0.001; Table 4;39).

[0198] Table 4: Multivariate PERMANOVA with Bray-Curtis distance for Beta diversity. CC = Callyspongia crassa: SC = Stylissa carteri. Statistically significant effects in bold (p < 0.05)

[0199] PERMANOVA

[0200] Source F.Model R2 / i-value

[0201] (a) CC vs CC-seawater 195.2807 0.5789857 0.001

[0202] (b) SC vs SC-seawater 179.8231 0.6162058 0.001

[0203] (c) CC vs SC 370.8368 0.7147465 0.001

[0204] BMC-r elated strains are naturally found in sponges

[0205] The two probiotic strains of Pseudoalter omonas were found to match AS VI 346, the two strains of Cobetia were assigned to ASV4551, the Halomonas strain was assigned to ASV3046, and the Sutcliffiella strain was assigned to ASV8312. All BMC -related strains had low relative abundance (<0.01) in seawater at TO (data not shown). Hereafter, all of these ASVs will be preceded by “BMC” to distinguish them from other ASVs present in sponges and seawater samples.

[0206] Table 5: Univariate PERMANOVA with Euclidean distance for BMC strains. CC = Callyspongia crassa SW = seawater. Statistically significant effects in bold (p < 0.05) UNIVARIATE PERMANOVA

[0207] Source Factor df MS F R2p-value

[0208] (a) CC_Pseudoalteromonas Sponge body 2 0.089 0.494 0.011 1

[0209] Time 1 0.178 1.976 0.023 0.500

[0210] Treatment 2 0.044 0.494 0.011 1

[0211] Time:Treatment 2 0.044 0.494 0.011 0.833

[0212] (b) CC_Cobetia Sponge body 2 _ _ _ _

[0213] Time 1 _ _ _ _

[0214] Treatment 2 _ _ _ _

[0215] Time:Treatment 2 _ _ _ _

[0216] (c) CC_Halomonas Sponge body 2 0.089 1 0.022 1

[0217] Time 1 0.044 1 0.011 1

[0218] Treatment 2 0.08910.022 1

[0219] Time:Treatment 2 0.08910.022 1

[0220] (d) CC_Sutcliffiella Sponge body 2 0.089 1 0.022 1

[0221] Time 1 0.044 1 0.011 1

[0222] Treatment 2 0.08910.022 1

[0223] Time:Treatment 2 0.08910.022 1

[0224] (e) CC_S\N_Pseudoalteromonas Time 1 4.033 3.227 0.112 0.097

[0225] Treatment 2 0.067 0.027 0.002 0.953

[0226] Time:Treatment 2 2.067 0.827 0.057 0.453

[0227] (f) CC_S\N_Cobetia Time 1 0.133 2.000 0.071 0.487

[0228] Treatment 2 0.067 0.500 0.036 1

[0229] Time:Treatment2o.O67 0.500 0.036 0.826

[0230] (g) CC_S \N_Halomonas Time 10.033 0.049 0.002 1

[0231] Treatment 20.267 0.195 0.016 0.881

[0232] Time:Treatment2o.267 0.195 0.016 0.849

[0233] (h) CC_S\N_Sutcliffiella* Time 10.300 2.000 0.049 0.121

[0234] Treatment 20.867 2.889 0.141 0.053

[0235] Time:Treatment21 .400 4.667 0.227 0.017

[0236] *PERMDISP p < 0.05

[0237] Most BMC-related strains were naturally associated with the S. carteri microbiome, except BMC-Halomonas (Figure 10). The relative abundance of all BMC-related strains increased over time regardless of the treatment and sponge body part, indicating a natural increase independent of inoculation (three-way univariate PERMANOVA, time, / ? < 0.05; Figure 10; data not shown). Overtime, the relative abundance of BMC-Pseudoalteromonas increased in average from 0.0003% to 0.0027%, BMC-Cobetia from 0.0001% to 0.0014%, BMC-Sutclifftella from 0.0001% to 0.0008%, and BMC-Halomonas changed from no reads detected in TO to 0.0011% in T1 (Figure 10; data not shown). The natural presence of BMC- Pseudoalteromonas in S. carteri was enriched by the BMC 1 x and BMC 3* treatments, as their relative abundance was increased approximately 20* in both treatments (0.0023% in BMC lx, and 0.0021% in BMC 3x) compared with that in the control sponges (0.0001%; three-way univariate PERMANOVA post hoc pairwise, controlxBMC l x: / ? = 0.029; control x BMC 3x; p = 0.026; Figure 10; data not shown). In addition, the relative abundance of BMC-Pseiidoalleromonas at T1 was significantly higher for both treatments than for the control (three-way univariate PERMANOVA post hoc pairwise, T1 - controlxBMC l x: / = 0.032; T1 - controlxBMC 3x; p = 0.025; data not shown). No reads of BMC- Pseudoalteromonas were identified in the control samples at Tl, while the relative abundance in average was 0.0042% for BMC lxand 0.0039% for BMC 3x. The relative abundance of BMC-Pseiidoalleromonas increased approximately 15 x over time in both treatments, ranging from 0.0003% to 0.0042% BMC l x, and from 0.0002% to 0.0039% BMC 3x (three-way univariate PERMANOVA post hoc pairwise, BMC l x - TO x Tl : ? = 0.040; BMC 3x - TO x Tl : p = 0.025; data not shown).

[0238] Moreover, BMC-Cobetia was enriched in S. carteri microbiome only when applied 3 x per week, as the relative abundance of these bacteria was significantly higher in BMC 3 x (0.0016%) than in the control (0.0001%; three-way univariate PERMANOVA post hoc pairwise, control x BMC 3x; / = 0.019; Figure 10; data not shown). Similarly, the relative abundance of BMC-Cobetia at Tl was significantly higher for BMC 3x (0.0028%) than for the control (0.0003%; three-way univariate PERMANOVA post hoc pairwise, Tl - ControlxBMC 3x; / ? = 0.016; data not shown), and it increased approximately 9x over time, ranging from 0.0003% to 0.0028% (three-way univariate PERMANOVA post hoc pairwise, BMC 3x - TO x Tl: / ? = 0.014; data not shown).

[0239] For S. carteri seawater samples, all probiotic strains were found at both sampling times (Figure 11). The BMC-Pseudoalteromonas, BMC-Halomonas, and BMC-Cobetia strains were significantly correlated with the interaction between time and treatment, and the latter two strains were significantly correlated with treatment (two-way univariate PERMANOVA, / ? < 0.05; data not shown). However, no post hoc significance was found.

[0240] By contrast, there is no evidence based on our data that C. crassa can acquire BMC- related strains as no significant changes in the relative abundances of probiotics were observed according to time, treatment, or sponge body part (Table 5a-d). At TO, none of the BMC-related strains was found in the C. crassa individuals, whereas at Tl, 2 of 45 sponge samples contained 'SMC-Sutcliffiella (control_R2), BMC-Halomonas, and BMC- Pseudoalteromonas (BMC 1 *_R3; data not shown). For the C. crassa seawater samples, all BMC-related strains were found at both sampling times, excluding BMC-Cobetia in TO (Figure 12; data not shown; Table 5e-h). No statistical differences were observed regarding time and treatment (Table 5e-h).

[0241] Time and BMC inoculation were correlated with microbiome changes

[0242] The richness (862 ± 207) and Shannon diversity (2.0 ± 0.3) of the S. carteri microbiome did not differ with time, treatment, and sponge body part (Table 6a, Table 6a-b). However, the structure of its microbial communities changed with treatment (three-way PERMANOVA, treatment factor, / ? = 0.036; Table 6c).

[0243] Table 6: Alpha diversity indices of the sponges and seawater microbiomes. SC = Stylissa carterr, CC = Callyspongia crassa

[0244] Source Time Group Shannon (H') ASV Richness

[0245] {a) Stylissa carteri TO control 2.2 ± 0.3 882 ± 140

[0246] TO BMC 1x 2.2 ± 0.4 856 ± 265

[0247] TO BMC 3x 1 .9 ± 0.1 786 ± 221

[0248] T1 control 2.1 ± 0.4 925. ± 286

[0249] T1 BMC 1x 2.0 ± 0.2 873 ± 73

[0250] T1 BMC 3x 1 .9 ± 0.2 851 ± 21 1

[0251] (b) SC_seawater TO control 3.8 ± 0.8 2061 ± 614

[0252] TO BMC 1x 3.7 ± 0.6 2519 ± 904

[0253] TO BMC 3x 5.1 ± 0.6 3780 ± 505

[0254] T1 control 4.9 ± 0.2 3649 ± 491

[0255] T1 BMC 1x 4.7 ± 0.1 3467 ± 261

[0256] T1 BMC 3x 4.5 ± 0.2 31 10 ± 226

[0257] (c) Callyspongia crassa comroi 3.2 ± 0.1 206 ± 43

[0258] TO BMC 1x 3.2 ± 0.2 198 ± 53

[0259] TO BMC 3x 3 2 ± 0.1 197 ± 70

[0260] T1 control 3.2 ± 0.2 361 ± 189T1BMC 1x 3.1 ± 0.3 341 1 163

[0261] T1 BMC 3x 3.2 1 0.2 539 1 183

[0262] (d) CC_seawater TO control 4.7 1 0.3 2530 1669

[0263] TO BMC 1x 4.9 1 0.5 2579 1464

[0264] TO BMC 3x 4 4±o,6 2095 1 586

[0265] T1control 4.7 1 0.3 2394 1490

[0266] T1BMC 1x 4.6 1 0.1 2384 1235

[0267] T1 BMC 3x 4.2 1 0.2 2127 1 157

[0268] Increased BMC inoculation (z.e., thrice a week) was correlated with more changes in the S. carteri microbiome than the BMC l x treatment, as 424 ASVs (approximately 6.8% of all S. carteri ASVs with summed mean relative abundance of 5.4%-10.4% across the different time points and treatments; data not shown) significantly differed between the control and BMC 3 x groups. For the BMC 1 x treatment, 334 ASVs (approximately 5.3% of the microbiome, with a mean relative abundance of 4.7%-8.7% of the microbiome; data not shown) changed significantly. Of these ASVs, 158 were shared between treatments (data not shown). Considering the summed mean relative abundances of ASVs at the phylum level, only changes related to the BMC 3x treatment were detected, with the relative abundances of Pseudomonadota decreasing over time and that of Bacteroidota, Cyanobacteria, and Thermoplasmatota increasing over time. Based on ASVs with relative abundance > 0.1%, Pelagibacter sp. (ASV36) exhibited an increase in relative abundance in the BMC lxand BMC 3x treatments whereas those assigned to the class Gammaproteobacteria (ASV30; ASV37; ASV393; ASV722) were decreased in both treatments (Figure 5). An increase in the abundance of Synechococcus sp. (ASV5), Rhodobacteraceae (ASV18), and Flavobacteriales (ASV32; ASV54) was observed only for the BMC 3x inoculation (Figure 13), whereas a decrease in the abundance of Alphaproteobacteria was observed in the BMC 1xgroup (ASV55; ASV334; Figure 13).

[0269] Table 7: Univariate PERMANOVA with Euclidean distance for Alpha diversity and multivariate PERMANOVA with Bray-Curtis distance for Beta diversity. SC = Stylissa carteri,' CC = Callyspongia crassa. Statistically significant effects in bold (p < 0.05) PERMANOVA

[0270] Source Factor df MS F R2p-value

[0271] (a) SC_richness Sponge body 1 82882 1.867 0.033 0.175

[0272] Time 1 26125 0.589 0.010 0.445

[0273] Treatment 2 72433 0.816 0.029 0.451

[0274] Time:Treatment 2 5573 0.063 0.002 0.939

[0275] (b) SC_shannon Sponge body 1 0.005 0.056 0.001 0.802

[0276] Time 1 0.167 1.760 0.029 0.190

[0277] Treatment 2 0.544 2.872 0.093 0.067

[0278] Time:Treatment 2 0.093 0.492 0.016 0.611

[0279] (c) SC_microbiome Sponge body 1 0.009 0.172 0.003 0.957

[0280] Time 1 0.096 1.775 0.029 0.124

[0281] Treatment 2 0.243 2.235 0.074 0.036

[0282] Time:Treatment 2 0.048 0.439 0.015 0.898

[0283] (d) CC_richness* Sponge body 2 184555 5.871 0.065 0.004

[0284] Time 1 1025280 65.227 0.359 0.0001

[0285] Treatment 2 170602 5.427 0.060 0.005

[0286] Time:Treatment 2 186237 5.924 0.065 0.003

[0287] (e) CC_shannon Sponge body 2 0.053 0.768 0.018 0.469

[0288] Time 1 0.015 0.447 0.005 0.502

[0289] Treatment 2 0.044 0.634 0.015 0.532

[0290] Time:Treatment 2 0.003 0.042 0.001 0.959

[0291] (f) CC_microbiome Sponge body 2 0.589 4.336 0.068 0.0003

[0292] Time 1 1.773 26.092 0.205 0.0001

[0293] Treatment 2 0.333 2.447 0.039 0.0169

[0294] Time:Treatment 2 0.367 2.697 0.042 0.0118

[0295] *PERMDISP p < 0.05

[0296] For C. crassa, the richness (307 ± 179) of its microbiome displayed significant differences with all factors, however as the PERMDISP analysis also produced significant results, these differences were not further assessed (Table 5c and Table 7d). In addition, there was no statistical support for differences in Shannon diversity (3.2 ± 0.2) with time, treatment, and sponge body part (Table 6c and Table 7e). Despite not acquiring any BMC strains, the structures of the microbial communities associated with C. crassa were significantly different with the three factors, including the interaction between time and treatment (three-way PERMANOVA, / ? < 0.05; Table 7f). As observed for S. carteri, the BMC 3 x treatment also caused more microbiome changes for C. crassa than the BMC 1 x treatment. In total, 274 ASVs changed significantly (approximately 7.2% of all C. crassa ASVs with a summed mean relative abundance of 18.9%-25.7% across different time points and treatments; data not shown) in the BMC 3 x treatment compared with the control, whereas only 123 ASVs changed significantly for BMC l x treatment (approximately 3.2% C. crassa ASVs with a summed mean relative abundance of 21.3%-24.4% for different time points and treatments; data not shown), with 54 ASVs shared between treatments. Considering the summed mean relative abundance of ASVs, the bacterial phylum Verrucomicrobiota and the archaeal phylum Thermoproteota displayed increasing abundance in both BMC groups. For the BMC 1xtreatment, the relative abundance of Chlorojlexota, and Nitrospirota increased whereas that of Actinobacteriota decreased. For the BMC 3 x treatment, the relative abundance of the bacterial phyla Bacteroido la. Cyanobacteria, Nitrospirota, and Poribacteria increased, whereas those of Chloroflexota and Desulfobacterota decreased. Regarding ASVs with relative abundance > 0.1%, the abundance of Pedosphaerales (ASV25; ASV59; ASV93; ASV126) and the archaeon Nitrosopumilus sp. (ASV19; ASV73; ASV94; ASV165; ASV171; ASV335; ASV903) increased in both BMC groups, whereas those assigned to the Gammaproteobacteria decreased (ASV51; ASV52). As observed for the S. carteri microbiome after the BMC 3x treatment, the relative abundance of Synechococcus spp. (ASV5; ASV42), Flavobacteriales (ASV35), and Rhodobacteraceae (AS VI 8) increased in the C. crassa microbiome (data not shown). Additionally, the relative abundance of Prochlorococcus sp. (ASV22) and Alphaproteobacteria (ASV14) increased in the C. crassa microbiome following BMC 3x treatment whereas that of Desulfobacterota (ASV45) decreased (data not shown). For the BMC 1 x treatment, an increase in the relative abundance of Nitrospirales (ASV26) was observed whereas that of Acidimicrobiales (ASV65; ASV2027) decreased (data not shown).

[0297] Discussion

[0298] This study provides the first insights into how the in situ inoculation of coral probiotics could affect filter-feeding organisms, specifically sponges. Probiotic-related strains naturally inhabited the S. carteri microbiome at low abundances, and two of them (BMC- Cobetia and ^MC-Pseudoalleromonas) were further enriched in the S. carteri microbiome after BMC inoculation. Conversely, the C. crassa microbiome apparently had no natural presence of BMC-related bacteria, and there is no evidence based on our data that C. crassa can acquire these microorganisms.

[0299] The relative abundance of other AS Vs in the S. carteri microbiome also changed following the BMC treatment. This result highlights the capability of manipulating the S. carteri microbiome by applying coral probiotics, which was unexpected because this sponge species has relatively stable microbiomes across different biogeographic localities45. Among the microbial changes triggered by the BMC inoculation, the increased abundance of Synechococcus sp. (Cyanobacteria) could benefit the sponge host by providing nutritional input if it becomes a dominant member, as has been shown for other sponge species46-48. Similarly, the structure of the C. crassa microbiome significantly changed following the BMC inoculation, including increases in the abundance of the cyanobacteria Synechococcus sp. and Prochlorococcus sp. Moreover, the C. crassa microbiome exhibited an increase in the abundance of the ammonia-oxidizing Nitrosopumilus sp. ( Thermoproleola). which could be beneficial by facilitating the detoxification of ammonia (metabolic waste) in the sponge49,50. In addition, the C. crassa microbiome presented changes in body parts and over time. Natural microbial changes in sponge microbiomes can occur because of seasonal fluctuations in environmental parameters, such as temperature and water quality51. During this experiment, the temperature increased by approximately 3°C in the field (Figure 15), potentially triggering the observed natural microbial changes.

[0300] These studies show that application of BMCs on nearby corals may restructure the microbiome of neighboring sponges, potentially affecting them. Based on the present data, however, these effects are not visually detectable and do not impair filtration. These results support the notion that coral probiotic inoculation has potentially very limited or negligible negative effects on the fitness of nontargeted organisms, which is an important criterion for the safe application of probiotics in the field15,25. Thus BMC disclosed here could be sued for sponges, especially for the treatment of sponge disease epidemics, which have been threatening some sponge species with extinction54.

[0301] References

[0302] 1. IPCC. Climate Change 2022: Impacts, Adaptation and Vulnerability

[0303] 2. McCulloch, M. T., Nat. Clim. Chang. 14, 171-177 (2024).

[0304] 3. Soga, M. & Gaston, K. J. Front. Ecol. Environ. 16, 222-230 (2018).

[0305] 4. Peixoto, R. S. & Voolstra, C. R. Front. Mar. Sci. 10, 1218531 (2023).

[0306] 5. Radecker, N., Nat. Comm. 14, 1-10 (2023). 6. Radecker, N. et al. Proc. Natl. Acad. Sci. USA 118, e2022653118 (2021).

[0307] 7. Hoegh-Guldberg, O. Mar. Freshw. Res. 50, 839-866 (1999).

[0308] 8. Hughes, T. P. et al. Nature 543, 373-377 (2017).

[0309] 9. Eakin, C. M., Coral Reefs 38, 539-545 (2019).

[0310] 10. Duarte, G. A. S. et al. Front. Mar. Sci. 7, 515393 (2020).

[0311] 11. Pratchett, M. S. et al. 265-289 (2021).

[0312] 12. Voolstra, C. R. et al. Nat. Rev. Earth Environ. 2, 747-762 (2021).

[0313] 13. Reshef, L., Environ. Microbiol. 8, 2068-2073 (2006).

[0314] 14. Peixoto, R. S., Front. Microbiol. 8, 341 (2017).

[0315] 15. Peixoto, R. S. et al. Nat. Microbiol. 7, 1726-1735 (2022).

[0316] 16. Peixoto, R. Front. Mar. Sci. 6, 686 10.3389 / fmars.2019.00686 (2019).

[0317] 17. Peixoto, R. S., Annu. Rev. Anim. Biosci. 9, 289-311 (2021).

[0318] 18. Peixoto, R. S. et al. Annu. Rev. Anim. Biosci. 9, 265-288 (2021).

[0319] 19. Fragoso dos Santos, H. et al. Sci. Rep. 5, 1-11 (2015).

[0320] 20. Rosado, P. M. et al. ISME J. 13, 921-936 (2019).

[0321] 21. Santoro, E. P. et al. Sci. Adv. 7, eabg3088 (2021).

[0322] 22. Silva, D. P. et al. Microbiome 9, 1-19 (2021).

[0323] 23. Li, J. et al. Environ. Microbiome 18, 1-14 (2023).

[0324] 24. Ushijima, B. et al. Commun. Biol. 6, 1-13 (2023).

[0325] 25. Delgadillo-Ordonez, N. et al. Commun. Biol. 7, 434 (2024).

[0326] 26. de Goeij, J. M. et al. Science 342, 108-110 (2013).

[0327] 27. Diaz, M. C. Bull. Mar. Sci. 69, 535-546 (2001).

[0328] 28. Wulff, J. Bull. Mar. Sci. 69, 831-846 (2001).

[0329] 29. Bell, J. J. Estuar. Coast. Shelf Sci. 79, 341-353 (2008).

[0330] 30. Padua, A., J. Mar. Biol. Assoc. UK 93, 889-898 (2013).

[0331] 31. Reigel, A. M. et al. Commun. Biol. 7, 1-12 (2024).

[0332] 32. Slattery, M., Mar. Ecol. Prog. Ser. 476, 71-86 (2013).

[0333] 33. Taylor, M. W., Microbiol. Mol. Biol. Rev. 71, 295-347 (2007).

[0334] 34. de Oliveira, Environ. Microbiol. Rep. 12, 619-638 (2020).

[0335] 35. Garritano, A. N. et al. ISME J. 17, 1517-1519 (2023).

[0336] 36. Giles, E. C. et al. FEMS Microbiol. Ecol. 83, 232-241 (2013).

[0337] 37. de Voogd, N. J. et al. World Porifera Database. (2024).

[0338] 38. Cleary, D. R., Microb. Ecol. 80, 103-119 (2020). 39. Thomas, T. et al. Nat. Commun. 7, 1-12 (2016).

[0339] 40. Riesgo, A., Mol. Biol. Evol. 31, 1102-1120 (2014).

[0340] 41. Schmittmann, L., Front. Immunol. 12, 689051 (2021).

[0341] 42. Ryu, T. et al. BMC Genomics 17, 1-11 (2016).

[0342] 43. Mallon, C. A., Trends Microbiol. 23, 719-729 (2015).

[0343] 44. Vinderola, G., Foods 11, 1077 (2022).

[0344] 45. Cleary, D. F. R. et al. Mol. Ecol. 31, 4932-4948 (2022).

[0345] 46. Thacker, R. W. Integr. Comp. Biol. 45, 369-376 (2005).

[0346] 47. Konstantinou, D., PLoS One 13, e0195001 (2018).

[0347] 48. Hudspith, M. et al. ISME J. 16, 2076-2086 (2022).

[0348] 49. Feng, G., Sci. Rep. 6, 1-11 (2016).

[0349] 50. Zhang, S. et al. mSystems 4, 288-307 (2019).

[0350] 51. Glasl, B. et al. ISME J. 14, 1435-1450 (2020).

[0351] 52. Garcias-Bonet, N. et al. Trends Microbiol. 32, 252-269 (2023).

[0352] 53. Daisley, B. A. et al. ISME J. 14, 476-491 (2020).

[0353] 54. Luter, H. M. & Webster, N. S. Invertebrate Pathology 109-122 (2022).

[0354] 55. Frank, J. A. et al. Appl. Environ. Microbiol. 74, 2461-2470 (2008).

[0355] 56. Ben-Haim, Y. et al. Int. J. Syst. Evol. Microbiol. 53, 309-315 (2003).

[0356] 57. Ushijima, B. et al. (2014). Appl. Environ. Microbiol. 80, 2102-2109

[0357] 58. Raimundo, I., Sci. Rep. 14, 14514(2024).

[0358] 59. Bell, J. J. et al. Ecol. Indic. 78, 470-488 (2017).

[0359] 60. Muyzer, G., Appl. Environ. Microbiol. 59, 695-700 (1993).

[0360] 61. Caporaso, J. G. et al. Proc. Natl. Acad. Sci. USA 108, 4516-4522 2011).

[0361] 62. Bolger, A. M., Bioinformatics 30, 2114-2120 (2014).

[0362] 63. Wemheuer, B. & Wemheuer, F. Methods Mol. Biol. 1539, 75-84 (2017).

[0363] 64. Edgar, R. C. bioRxiv 074252 doi: 10.1101 / 074252 (2016).

[0364] 65. Gao, X., BMC Bioinformatics 18, 1-10 (2017).

[0365] 66. Parks, D. H. Nat. Biotech. 38, 1079-1086 (2020).

[0366] 67. Love, M. I., Genome Biol. 15, 1-21 (2014).

[0367] 68. Oksanen, R Package version 2.5-7 (2020).

[0368] 69. Camacho, C. et al. BMC Bioinformatics 10, 1-9 (2009).

[0369] 70. Herve, M. R Package version 0.9-81-2 (2022).

[0370] 71. Wang, Y.. R Package ‘mvabund’. (2022). EXAMPLE 3- Unlocking the Genomic Potential of Red Sea Coral Probiotics

[0371] Coral reefs are one of the most diverse and productive ecosystems on Earth. However, they are currently facing unprecedented challenges due to overfishing, pollution, and climate change 1-4. Rising sea surface temperatures and increased greenhouse gas concentrations leading to global warming 5 and ocean acidification have emerged as key stressors, threatening the survival of coral reefs worldwide 3,6-8. These stressors cause periods of high seawater temperatures, leading to the loss of Symbiodiniaceae, the algal symbionts that reside within the holobiont and fulfill up to 90% of the corals’ nutritional requirements 9 except when they become nutrient competitors and can be expelled 10-12. This loss, especially over a long-term, can lead to devastating bleaching events worldwide, in which corals are deprived of vital nutrients, leading to increased disease susceptibility and reduced reproductive success and skeletal growth 2,13-16. This disruption extends beyond the loss of algal symbionts because it is followed by shifts in the composition and function of the entire coral-associated microbiome 17 . This causes post-heat stress disorder 18 that compromises the coral’s ability to grow, resist pathogens, modulate nutrient cycling, and maintain holobiont homeostasis 18- 27.

[0372] The symbiotic relationship between corals and their bacterial symbionts 28,29 plays a pivotal role in maintaining the health of these organisms and consequently the functioning of reef ecosystems. Although some corals and their microbiomes have shown the capacity to recover from acute stress events 30,31, the coral microbiome may not fully restore its original composition and functionality following a severe disturbance 32.

[0373] Rehabilitation of the coral microbiome 33 has recently emerged as a promising strategy to enhance coral health and resilience 18,24,26,33-42. Coral-associated bacteria possess potential beneficial traits that enhance the fitness and resilience of their coral host by maintaining homeostasis through their involvement in various essential processes, such as nutrient cycling, production of antibiotics and antimicrobial compounds, and mitigation of toxic compounds 33,43. Harnessing the symbiotic interactions between corals and their associated bacteria 28 allows researchers to selectively introduce these beneficial microorganisms for corals (BMC) 33 into the coral holobiont for use as customized probiotics. This can help maintain or even enrich the proportion of native beneficial microorganisms to enhance coral health and resilience under stress 33,35.

[0374] In this study, an in silico investigation of the probiotic potential of six putative BMC (pBMC) strains isolated from corals in the Red Sea was tested with the aim of developing the first Red Sea BMC consortium. This was achieved by analysing the complete genomes of all six strains and screening for the presence of potential beneficial mechanisms for corals. The present study provides a framework for the selection of customized BMC consortia based on the presence of particular BMC characteristics that help the host survive and thrive under stress conditions.

[0375] Materials and Methods pBMC isolation and genome sequencing pBMC strains were isolated from coral fragments (± 5 cm long) of Pocillopora verrucosa, Galaxea fascicularis, and Stylophora pistillata Clade IV. The fragments were collected between February and April 2021 from Al Fahal reef (22°18T8.4"N, 38°57'52.5"E), Aquarium reef (22°23T5.6''N, 38°55'07.2''E), and Tahala reef (22° 15 '46.9 "N, 39°03'05.9''E) in the eastern central Red Sea, Saudi Arabia, as described by Delgadillo- Ordonez and colleagues44.

[0376] Briefly, coral fragments were macerated in 50 mL of 3.5% sterile saline solution using a sterile mortar and pestle. Then, the macerated paste was serially diluted up to 10-6 with 3.5% saline solution, and 100 pL of each dilution was plated on Zobel Marine Agar 2216 (MA; HiMedia Laboratories, Mumbai, India), diluted marine agar (DMA; 2 / dilution of MA with 3.5% NaCl and adjusted agar content) and Lennox Broth (Sigma- Aldrich) and cultured overnight at 25°C. The macerate was also incubated in parallel at 25°C overnight and 130 rpm in a 250-mL sterile Erlenmeyer flask containing crystal beads. After this first incubation, 100 pL triplicate subsamples of 10-4, 10-5, and 10-6dilutions were plated on approximately 20 mL of MA and DMA. Alternatively, 0.5-cm long coral fragments were plated on both media. All plates were incubated at 25°C for at least 48 h or until visible bacterial colonies were observed. Based on colony morphology, >350 isolates were obtained, which were preserved at -80°C in sterilized 20% glycerol. To select pBMC strains, the isolates were screened for various beneficial traits using polymerase chain reaction (PCR) assays targeting nitrogen cycle-related genes involved in nitrogen fixation (nifH) and denitrification (nirK) as well as biochemical analyses of catalase activity, siderophore production45and phosphate assimilation46. Finally, the antagonistic effect of the isolates against the coral pathogen Vibrio coralliilyticus47,48was tested following the methodology described by Rosado and colleagues26and Santoro and colleagues18. Based on the positive results of these tests, six strains were selected as pBMC and stored in glycerol stocks (20%) at -80°C. The genomic DNA (gDNA) were extracted from these six pBMC strains using Wizard Genomic DNA Purification Kit (Promega, Madison, WI, USA) according to the manufacturer’s protocol. The resultant gDNA was purified with AMPure beads (1 :8 ratio; Beckman Coulter) and quantified using Qubit dsDNA Assay Kit (Invitrogen). Then, the quality and integrity of the extracted gDNA were confirmed via spectrometry using a NanoDrop 8000 spectrophotometer (Thermo Fisher Scientific) and via electrophoresis using the 4200 TapeStation System (Agilent Technologies). Next, SMRTbell libraries were prepared using SMRTbell Express Template Prep Kit 2.0 (Pacific Biosciences) and Barcoded Overhang Adapter Kit 8A (Pacific Biosciences) according to the PacBio multiplexed microbial libraries protocol. The quality of the libraries was assessed using Qubit dsDNA High Sensitivity Assay Kit (Thermo Fisher Scientific) and FEMTO Pulse pulsed-field capillary electrophoresis system (Agilent Technologies). For sequencing, the libraries were first annealed using Primer v4 via Sequel Binding Kit 3.0 (Pacific Biosciences) and then sequenced using SMRT Cell IM v3 LR Tray (Pacific Biosciences) via the Sequel system at the Bioscience Core Lab of King Abdullah University of Science and Technology, Saudi Arabia. pBMC genome assembly and taxonomic classification

[0377] The quality of raw reads was assessed using LongQC tool49, and the reads were trimmed according to the default parameters. Next, the reads were assembled using Canu50, Flye51, Raven52, Miniasm / Minipolish53, and Redbean54. The generated assemblies were then evaluated and compared to determine the quality of general genome assembly via Quast55, genomic completeness / contamination via CheckM software56, chimerism via GUNC57, and indel errors / pseudogenes via ideel . The assemblies from Canu were considered the best based on assembly quality, genomic completeness, and lower number of indel s / interrupted open reading frames. FastANI58, and Genome-to-Genome Distance Calculator59were used to assess the phylogenetic relatedness between the study genomes and the publicly available closest reference strain genomes.

[0378] Structural and functional annotation of pBMC genomes

[0379] The six pBMCs genomes were annotated using Prokka60and the PATRIC Bacterial and Viral Bioinformatics Resource Center (BV-BRC)61, which uses the RASTtk online server62, according to the default parameters. The output data were visualized on the BV- BRC interface. We screened for protein-coding genes that might provide beneficial characteristics for host health and resilience33,43by searching the subsystems generated by the RASTtk platform in BV-BRC. Additionally, the six pBMC genomes were searched for putative prophage sequences using the PHASTEST online analysis tool63. This tool analyzes the proportion of phage proteins within a bacterial genomic region and scores the potential phages based on their length and inclusion of several phage-related proteins, such as “capsid,” “head,” “integrase,” “plate,” “tail,” “fiber,” “coat,” “transposase,” “portal,” “terminase,” “protease,” or “lysin,” in the input file. Only phages considered intact based on the abovementioned criteria were used for further analysis. Intact phages were queried using the IMG / VR v4 online BLAST database, and similarity was assessed based on their percent identity with high-quality reference phages64. AntiSMASH v765was also used to identify, annotate, and analyze secondary metabolite biosynthetic gene clusters (BGCs) present in the genomes of all six pBMC strains.

[0380] Phylogenomic trees

[0381] A phylogenomic assemblage was constructed using the codon tree method via the BV-BRC platform. This method randomly selects up to 10 proteins to represent each genuslevel family and merges them to create a single set of representatives, preventing cluster formation, which is based on genus similarity rather than protein similarity (PATRIC global protein families [PGFams])61,66. Amino acids and nucleotide sequences were used for analyzing each selected gene from the PGFams. Protein sequences were aligned using MUSCLE67, and encoding gene sequences were aligned using the Codon align function in BioPython68. The output was a concatenated alignment of all proteins and nucleotides as well as a partitioned file describing the alignment of proteins for analysis using RAxML69. Support values were generated using 100 rounds of the “Rapid” bootstrap option70in RAxML. The phylogenomic trees were constructed by comparing and analyzing proteincoding genes (n = 1000, Figures 31-35; n = 500, data not shown) shared among all genomes using the best protein model method identified by RAxML71. The output, in Newick format, was then visualized via iTOL version 672,73. The trees generated are separated by genus (Pseudoalter omonas, Cobelia. Halomonas, and Siilcliffiella . and the genomes used to construct these trees were selected from the BV-BRC platform database based on various criteria, such as being reference genomes, complete genomes, and isolated from corals and / or aquatic environments, as per their availability.

[0382] Pangenomic analysis and gene categorization

[0383] The Prokka annotations were used to generate a separate pangenome for each pBMC genus (Pseudoalter omonas, Cobetia, Halomonas, and Sutcliffiella') using the Roary tool73, with the default settings. For each pangenome, sequences from the closest genomes to the present pBMC strains, which were identified based on the phylogenomic tree results, were added to the analysis. For the pangenomic analysis of Pseudoalter omonas sp., in addition to the two Pseudoalter omonas galatheae genomes (pBMCl and pBMC2), 19 additional Pseudoalter omonas sp. genomes were included (from Figure 28). For the pangenomic analysis of Cobetia sp., in addition to the two Cobetia amphilecti genomes (pBMC3 and pBMC4), three Cobetia sp. genomes were added (from Figure 29). The pangenome of Halomonas sp. included the single Halomonas sp. genome (pBMC5) and three other Hal omonas sp. genomes (from Figure 30). For Sutcliffiella sp., its pangenome consisted of the genome of Sutcliffiella sp. pBMC6, one other Sutcliffiella sp. genome, and two Bacillus sp. genomes (from Figure 31). All genomes used for comparison with the pBMC genomes were obtained from the BV-BRC genome database61. The pangenomes were created and visualized using the Anvi’o platform74, and the genes were categorized according to these results. Singleton genes were defined as genes that were only present in one or more of the pBMC genomes and not in the other genomes that were part of the same pangenome. These singleton genes were identified for each pBMC and screened for their potential to enhance the health status and resilience of the host using UniProt75.

[0384] Data availability

[0385] Scripts that were used in the assembly and evaluation of the study isolate genomes are available at zenodo.org / records / 10774634. The raw reads and assemblies from this study are publicly available in the European Nucleotide Archive (study accession number: PRJEB62849).

[0386] Results

[0387] Genomic features of pBMC

[0388] The six pBMC genomes were sequenced using PacBio, resulting in several cleaned and trimmed high-quality reads of 137126-345950, with an average read length of 5827 bp. The genomes were assembled using Canu50, resulting in 2-9 contigs per strain. CheckM analysis revealed a high level of genome completeness (>99% for all genomes except Sutcliffiella sp. pBMC6, which was >97.5%) and a low level of contamination (<2.5%) for all genomes (Table / 8).

[0389] Table 8. Genome characteristics of our six putative beneficial microorganisms for corals (pBMC) strains.

[0390] The genomes of P. galatheae pBMCl and pBMC2 had very similar characteristics, with a GC content of 42.94% and 42.98% and a total length of 5528212 and 5474806 bp, respectively. The genomes of Cobetia sp. pBMC3 and pBMC4 also had very similar characteristics, with a GC content of 62.32% and 62.31% and a total length of 4046628 and 4041826 bp, respectively (Table 8).

[0391] Taxonomic attribution pBMCl and pBMC2 represented the same species, sharing an average nucleotide identity (ANI) of >99.5% at both genome and 16S rRNA gene levels. The DNA-DNA hybridization (DDH) and ANI values with the strain Pseudoalter omonas galatheae were 92% and 99.17% for pBMCl and 92.4% and 99.19% for pBMC2, respectively, suggesting that they belonged to this species (Table 9). Notably, the species-level thresholds are 70% for DDH76and 95%-96% for ANI77. Similarly, pBMC3 and pBMC4 showed an ANI of >99.5% at the genome and 16S rRNA gene levels. Their DDH and ANI values with other Cobetia sp. were also high, and pBMC3 and pBMC4 were classified as Cobetia sp. in this study (Table 9). pBMC5 showed 97.94% ANI with the genomes of two strains of Halomonas meridiana (Table 9); however, compared with the type strain of H. meridiana (H. meridiana DSM5425), pBMC5 showed an ANI of <90% and DDH of <70%, suggesting that it is a novel species. pBMC6 also potentially represents a novel species, with the closest species genome being that of Bacillus horikoshii at an ANI of 85.96% (Table 9). The systematic classification of the family Bacillaceae has undergone several modifications due to the implementation of new taxonomic polyphasic techniques78,79, which have led to the creation of new genera from the previously classified genus Bacillus, including the new genus Sutcliffiella19. Considering this, the overall taxonomic classification of the six pBMC strains investigated in our study should be as follows: P. galatheae (n = 2), Cobetia sp. (n = 2), Halomonas sp. (n = 1), and Sutcliffiella sp. (n = 1).

[0392] Table 9. Taxonomic classification of all genomes of our putative beneficial microorganisms for corals (pBMC) according to their phylogenomic trees using two different tools. The Genome-to-Genome Distance Calculator (GGDC) was used to calculate DNA-DNA hybridization (DDH) values, and FastANI was used to calculate the average nucleotide identity (ANI) values.

[0393] Phylogenomic trees

[0394] The phylogenomic relationships between the six pBMC genomes and the closest publicly available genomes were evaluated. 115 genomes were used to assemble the phylogenomic tree of P. galatheae pBMCl and pBMC2 (Figure 8). A well-supported clade, formed by pBMCl, pBMC2, and 14 other Pseudoalteromonas sp. strains, was established in this phylogenomic tree. Consequently, another phylogenomic tree was assembled using the genomes of this clade (Figure 31). All 16 strains included in this new tree were host- associated. For the phylogenomic tree of Cobetia sp. pBMC3 and pBMC4, the genomes of 27 other Cobetia sp. were added, and pBMC3 and pBMC4 formed a clade (Figure 29). The phylogenomic tree of Halomonas sp. pBMC5 included 25 other Halomonas sp. genomes (Figure 30). The Halomonas sp. pBMC5 formed a small clade with one Halomonas sp. strain and one H. meridiana strain, creating a coral-associated clade. In the phylogenomic tree of Sutcliffiella sp. pBMC6, all 11 publicly available genomes were included (Figure 34). There was a small clade between Sutcliffiella sp. pBMC6 and two other Sutcliffiella sp. strains (Sutcliffiella horikoshii and B. horikoshii). one of which was isolated from soil and the other from a clean room.

[0395] Potentially beneficial gene functions

[0396] The BV-BRC platform was used to screen for genes encoding proteins that are potentially beneficial for corals33 43 80anc[ their endosymbiotic algae81. Subsystems that are known to play a role in the health and resilience of corals, as described by Peixoto and colleagues33,43, and Matthews and colleagues81, were selected, and genes associated with oxidative stress, nitrogen cycle, ectoine and B-complex vitamin biosynthesis, iron and sulfur metabolism, and dimethylsulfoniopropionate (DMSP) degradation were identified (Table 10).

[0397] Within the subsystems of interest, all pBMC harbored genes related to protection from reactive oxygen species (ROS) and biosynthesis of vitamins B2, B7, and B9. pBMCl, pBMC2, pBMC5, and pBMC6 shared genes related to catalase and peroxidase activities and vitamin B 12 biosynthesis. All pBMC, except for pBMC6, harbored genes involved in vitamin B6 biosynthesis. pBMC6 was the only pBMC harboring genes involved in iron acquisition. Genes related to ectoine biosynthesis were only detected in pBMC3, pBMC4, and pBMC5, whereas genes related to the nitrogen cycle were found in pBMC3, pBMC4, and pBMC5. DMSP degradation-related genes were only detected in pBMC3 and pBMC4, whereas genes related to sulfur metabolism were only identified in pBMCl and pBMC2 (Table 10). Table 10. Examples of gene functions related to putative beneficial traits for coral found in the genomes of our putative beneficial microorganisms for corals (pBMC).

[0398] Putative prophages

[0399] Two intact prophages were predicted in the genomes of P. galatheae strains pBMCl and pBMC2 (Figure 32). They were most similar to the tailed phages that infect P. galatheae S4498 and P. piscicida 36Y RITHPW. No putative prophages were detected in either of the Cobetia sp. strains (pBMC3 and pBMC4). Four intact prophages were found in Halomonas sp. pBMC5, of which three were most similar to the tailed phages of the bacteria Halomonas meridiana Rlt3, H. axialensis ACH-L-8, and Halomonas sp. The bacterial host of the fourth phage remained unknown because our results were obtained from a pooled assembly. Three prophages were found in Sutcliffiella sp. pBMC6, and they were most similar to the tailed phages infecting Marinisporobacter balticus DSM 102940, S. horikoshii FJAT-14233, and B. cereus HuA2-9.

[0400] Pangenomes

[0401] A pangenomic analysis was conducted for each pBMC clade to screen for genes unique to pBMC strains (data not shown). For the pangenome of pBMCl and pBMC2, 19 Pseudoalter omonas sp. strains were added, including BMC from the study by Rosado et al. (2018) (data not shown). 714 unique genes (i.e., 714 genes were identified to be present in the genomes of pBMCl or pBMC2 but absent in the genomes of other 19 Pseudoalteromonas sp. strains). Of these 714 unique genes (450 of which were shared between pBMCl and pBMC2), 604 were classified as hypothetical proteins, and the remaining 110 had known functions. Of the 110 genes with known functions, 49 were shared by pBMCl and pBMC2. Furthermore, 3 of these 110 unique pBMC genes encoded proteins related to previously described beneficial traits for corals (Table 11).

[0402] There were 32 singleton genes identified in the pangenome of pBMC3 and pBMC4, which included 3 other Cobetia sp. strains (data not shown). Of these 32 singleton genes (with no genes being shared between pBMC3 and pBMC4), 5 had known functions, and the remaining were classified as hypothetical proteins. Of these 5 unique pBMC genes, none encoded proteins related to previously assigned beneficial traits for corals (Table 11).

[0403] Halomonas sp. pBMC5 formed a clade with two other Halomonas sp. strains, and a pangenome involving these three strains and Halomonas sp. BMC7 from the study by Rosado and colleagues26was established (data not shown). A total of 740 genes were identified as singletons, which were present in the pBMC5 genome alone. Among these 740 genes, 536 were classified as hypothetical proteins, and 204 had known functions. Of these 204 unique pBMC5 genes, 4 encoded proteins related to previously assigned and potentially novel beneficial traits for corals related to antibiotic synthesis and transport and sulfur assimilation (Table 11)

[0404] Table 11 : Proteins likely to benefit corals that were only identified in Pseudoalter omonas galatheae pBMCl and pBMC2, Cobetia amphilecti pBMC3 and pBMC4, Halomonas sp. pBMC5, and Sutcliffiella sp. pBMC6 pangenomes. Functions retrieved from UniProt.

[0405] Sutcliffiella sp. pBMC6 formed a clade with another Sutcliffiella sp. and one Bacillus sp. strain (Figure 31). For a more comprehensive pangenome, B. horikoshii strain FJAT- 14233 was added to the above-mentioned strains (data not shown). This pangenome harbored the highest number of unique pBMC genes, with 3088 genes present only in the pBMC6 genome. Of these, 1755 were classified as hypothetical proteins, and the remaining 1333 had known functions. Among these 1333 genes, 27 encoded proteins related to previously assigned and potentially new beneficial traits for corals related to siderophore, carotenoid, vitamin, and antibiotic synthesis, nitrogen, sulfur and iron metabolism, chemotaxis, degradation of toxic compounds and stress resistance (Table 11).

[0406] Biosynthesis of natural product

[0407] Mining of the six pBMC genomes on anti SMASH to screen for secondary metabolite BGCs revealed 49 BGCs across the six pBMC genomes. These included terpene (n = 2), type I and III polyketide synthase (PKS; n = 9 and n = 1, respectively), ectoine (n = 3), nonribosomal peptide synthetases (NRPSs; n = 8), nonribosomal peptide metallophores (n = 4), opine-like zincophores (n = 1), NRPS-independent (NI) lucA / IucC-like siderophores (n = 3), beta-lactone (n = 1), aryl polyene (n = 2), class I and IV lanthipeptides (both, n = 2), and lasso peptide (n = 1) as well as some putative clusters.

[0408] The NI siderophore cluster found in pBMC3 and pBMC4 showed 75% homology with known BGCs encoding desferrioxamine E (BGC0001572, BGC0001478, NCBI: MH015039.1). The ectoine cluster from pBMC5 showed 75% homology with several known ectoine-coding BGCs (BGC0000855, BGC0000856, BGC0000859, NCBI: DQ238213.1). The ectoine cluster of pBMC3 and pBMC4 showed 66% homology with the ectoine BGC000852 (NCBI: AF316874.1). The opine-like metallophore BGC found in pBMC6 showed 100% similarity with a known BGC encoding bacillopaline (BGC0002488, NCBI: CP002869.1), and the lasso peptide cluster showed 80% homology with the known BGC encoding paeninodin (BGC0001356, NCBI: AHKH01000064.1). Two terpenoid clusters were found in pBMC6. The first showed low homology with known clusters, highlighting the importance of continuing to study the biosynthesis of this product in marine organisms. The other terpenoid gene cluster showed higher (66%) homology with a known BGC encoding a carotenoid (BGC0000645, NCBI: FJ040212.1). The pBMC6 genome contained a type III PKS cluster, and all other strains contained a type I-PKS cluster.

[0409] A protein family (Pfam)-based analysis was performed using antiSMASH, revealing squalene and phytoene synthase in Sutcliffiella sp. pBMC6 and a carotenoid biosynthesisencoding gene (Table 12). The genomes of P. galatheae pBMCl and pBMC2 contained lanthionine synthetase and lantibiotic biosynthesis-encoding genes as well as thioesterase synthase, which is necessary for peptide antibiotic production (Table 12). Cobetia sp. pBMC3 and pBMC4 harbored Pfam domains related to siderophore production and vitamin B12 and ectoine biosynthesis (Table 12), and ectoine synthase-encoding genes were present n Halomonas sp. pBMC5. Table 12. Protein families (Pfams) that may be beneficial for the host and are present in the genomes of the six putative beneficial microorganisms for corals (pBMC) strains.

[0410] Discussion

[0411] Halomonas sp. pBMC5 and Sutcliffiella sp. pBMC6 as candidate species

[0412] The systematic classification of the family Bacillaceae has undergone numerous modifications in recent years due to the implementation of new taxonomic polyphasic techniques78,79. This has resulted in the creation of new genera from the previously classified Bacillus genus, such as the new genus Sutcliffiella79, and justifies the presence of B. horikoshii strains in the phylogenomic tree of pBMC6 — likely an older classification of the current S. horikoshii — and the dominance of Bacillus sp. genomes in this tree. The identification of three putative prophages from pBMC6 further corroborates these data because three different bacterial genera were predicted as their hosts, namely, Marinisporobacter, Sutcliffiella, and Bacillus.

[0413] Genome screening reveals previously proposed beneficial traits of pBMC

[0414] Following the classification of each pBMC for identification and phylogenomic analysis, the pBMC genomes were screened for genes encoding proteins that are potentially beneficial for corals. Genes were screened which were related to catalase, urease, and siderophore production; phosphate assimilation; and nitrogen cycle and DMSP degradation through biochemical tests and PCR assays44, which are typically employed for BMC selection. Genes involved in other potential beneficial traits were also detected (Table 10), including those related to oxidative stress, such as superoxide dismutases (all pBMC genomes), which exert an antioxidant effect by catalyzing the dismutation of superoxide (an ROS molecule that causes cell damage)82; catalase KatE (all pBMC genomes) and catalase- peroxidase KatG (pBMCl, pBMC2, pBMC5, and pBMC6 genomes), both of which protect cells from the toxic effects of H2O2 and aerated growth conditions83-87; manganese catalase (pBMC5 and pBMC6 genomes), which is also involved in the protection of cells from H2O288,89; and glutathione synthetase (all pBMC genomes except for pBMC6), which produces glutathione that can subsequently be used by glutathione peroxidase (all pBMC genomes except for pBMC3 and pBMC4) to scavenge ROS, such as H2O290,91. When seawater temperatures rise, the coral holobiont produce ROS, resulting in cell damage in both host and its symbionts92-94. A direct correlation between bleaching and ROS production has been previously reported3, and ROS-scavenging pBMC strains were hypothesized to mitigate coral bleaching33, making this a crucial trait when selecting pBMC.

[0415] Several genes involved in vitamin B-complex biosynthesis were found in our pBMC genomes, such as riboflavin synthase (all pBMC), pyridoxine 5 ’-phosphate synthase (all pBMC except for pBMC6), biotin synthase (all pBMC), dihydrofolate synthase and thymidylate synthase (all pBMC), and cobalamin synthase (all pBMC except for pBMC3 and pBMC4) for the biosynthesis of vitamins B2, B6, B7, B9, and B12, respectively. Vitamin B2 is necessary for glutathione reductase activity, which is involved in stress reduction by increasing antioxidant potential, and B2 deficiency increases lipid peroxidation95. Vitamin B6 catalyzes approximately 2% of all prokaryotic functions96, but it has not been widely studied in the marine environment97,98. It acts as an antioxidant during light exposure and against oxidative stress99 100vitamin B7 is a cofactor in several metabolic pathways, such as fatty acid biosynthesis, amino acid metabolism, and gluconeogenesis101. Vitamin B12 is involved in several metabolic pathways, including the production of the antioxidants glutathione and DMSP102, which are important for neutralizing high concentrations of ROS generated from heat stress events33,43. Bacteria that exist in association with corals possess genes encoding for proteins related to the biosynthesis of essential vitamins, such as Bl, B2, and B7, whereas their coral host does not have the capacity to produce them103. This suggests that the coral holobiont can only take up these essential vitamins through heterotrophic feeding and / or from its bacterial symbionts. Furthermore, coral symbionts from the family Symbiodiniaceae are auxotrophs for various B-complex vitamins, which they acquire from exogenous sources such as bacteria81>1O2,IO4-IO6 This highlights the important role of bacterial symbionts in ensuring coral health. For these reasons and because of the close interaction between several B-complex vitamins, the presence of genes encoding proteins related to B- complex vitamin biosynthesis is suggested as a BMC trait.

[0416] Other genes related to metabolism were also screened. Siderophore synthase was present in pBMC6. This enzyme produces siderophores that can scavenge iron from the environment, a trait that is beneficial for corals33,80and other organisms107. In general, the bioavailability of iron in oceans is extremely low; consequently, the growth and survival of organisms that use iron for essential physiological processes, such as photosynthesis and nitrogen fixation, cannot be guaranteed108 109Thus, bacteria that exist in association with other marine organisms, such as corals and microalgae, produce siderophores to capture and concentrate iron into a bioavailable form that can be used by other organisms110 111Apart from siderosphere production, it was also found that pBMC produced ectoine (pBMC3, pBMC4, and pBMC5) and betaine (pBMCl, pBMC2, pBMC3, pBMC4, and pBMC5), are proposed to be compounds that plays a role in beneficial mechanisms in corals80. Ectoines and betaines are important for osmoregulation and act as protective agents under thermal stress and high irradiance112“114. They also contribute to the nitrogen biomass of corals in reefs115. In marine microalgae, the ectoine content was found to increase in the presence of bacteria, highlighting the crucial role of these microorganisms in host health116. Betaine and ectoine production significantly improves environmental stress tolerance, including pH stress117and heat stress118in aquatic organisms119, such as corals and their symbionts114. Ectoines can help mitigate the harmful effects of heat stress, high salinity, ROS, and radiation120. Pei et al. identified betaine lipids as leading metabolite drivers for differentiating heat-bleached corals from healthy ones, revealing new tools to screen for heat-resistant corals and their symbionts, such as BMC114 121

[0417] Genes involved in the nitrogen cycle were found, including nitrate reductase (pBMC3, pBMC4, and pBMC5) and cyanate hydratase (pBMC5). The presence of these genes was previously proposed as a BMC trait33,43because balancing this nutrient’s availability contributes to maintaining desirable levels of bioavailable nitrogen, limiting algal growth and leading to an accumulation of photosynthates in algal cells that, when released, feed the coral host and promote its growth. Additionally, increased coral catabolic activity due to an environmental stressor leads to host starvation and increased nitrogen availability to Symbiodiniaceae members of the holobiont, potentially causing destabilization of the host’s nutrient cycle and of the Symbiodiniaceae-coral interaction10,122.

[0418] Screening for genes related to DMSP degradation and sulfur metabolism revealed the presence of DMSP CoA transferase / lyase DddD (pBMC3 and pBMC4) and acryloyl-CoA reductase AcuI / YhdH (BMC1 and pBMC2). DMSP is found in several marine organisms, including Symbiodiniaceae105,123, and is a ROS scavenger in marine algae124, an attribute that has been proposed as a BMC trait due to its antioxidant activity33. Mechanisms of DMSP breakdown have also been hypothesized as BMC traits because a high DMSP concentration can lead to dysbiosis and signal the location of more vulnerable coral to pathogens through chemotaxis18,33,125. Discovery of the presence and putative wide distribution of BMC-associated prophages

[0419] Prophages are DNA from bacteriophages (or bacteria-infecting viruses) that are integrated into the genomes of their bacterial host upon infection. They are mainly studied as contributors to the virulence of pathogenic reef bacteria, such as V. coralliilyticus, because they encode virulence factors126. The detection of putative prophages in four of our six pBMC strains revealed the potentially widespread presence of prophages in marine bacteria, suggesting the importance of prophages associated with beneficial bacteria. Prophages also protect the bacterial host against virulent phage infections via superinfection exclusion127. Thus, prophages might both expand the metabolic capabilities and protection of beneficial bacteria.

[0420] Pangenomes exhibit novel pBMC beneficial traits and other applicable functions Pangenomic analyses was conducted that included the pBMC strains and strains closely related to them in the phylogenomic trees to screen for genes unique to our pBMC strains. In the Pseudoalter omonas pangenome, 110 genes with known functions were only present in the pBMC genomes (i.e., they were present in one or both of the P. galatheae pBMC strains). This included a protein involved in chemotaxis and some proteins involved in iron / sulfur metabolism. Chemotaxis may be an important BMC trait because it was previously suggested to play a crucial role in defining patterns of microbial diversity, coral metabolism, coral infection dynamics, and chemical cycling processes, thereby influencing coral holobiont health128. Despite the absence of DMSP-related genes in the genomes of the Pseudoalter omonas sp. pBMC strains compared with the other Pseudoalter omonas sp. strains, the metabolism and production of sulfur compounds has been proposed as a BMC trait because they inhibit the growth of coral pathogens and also play a role in the structure of bacterial communities of the coral holobiont33.

[0421] In the Cobetia pangenome, five genes were found with known functions that were only present in the pBMC genomes, but none of them showed a clear benefit to the host. In the Halomonas pangenome, 204 genes with known functions were found that were only present in the pBMC5 genome, including genes involved in carotenoid biosynthesis, antibiotic biosynthesis, and sulfur metabolism. Although not directly related to host health but more to the environment as a whole, the presence of mercury resistance-related genes in the pBMC5 genome is of crucial interest. Heavy metals are known to be environmental toxins due to their bioaccumulation in the food chain, becoming increasingly hazardous for the higher trophic levels129 130The use of bioremediation for the removal of toxic metals has been studied, but only a few studies were performed in a marine environment131. The trend is similar for mercury-resistant marine bacteria, and the use of mercury-resistant marine bacteria for bioremediation of mercury contamination has received little attention132. However, their use is associated with certain advantages, such as simple process, lower amount of secondary metabolites, and lower cost than commonly used chemical technologies as well as better adaptability and higher resistance to adverse environmental conditions compared with terrestrial bacteria132 133. Some studies have reported that mercury-resistant marine bacteria show a higher capability for mercury bioremediation and can reduce the toxic effects of mercury in contaminated environments132 134.

[0422] The Sutcliffiella genome were found that had the highest amount of singleton genes of all the pangenomes generated in this study, with 1333 genes with known functions present only in the pBMC6 genome, including genes encoding proteins related to the biosynthesis of siderophores, carotenoids, antibiotics, vitamins Bl, B2 and Bl 2; nitrogen, iron, and sulfur metabolism; chemotaxis; and oxidative stress resistance. Other than the abovementioned genes, we identified a gene that was involved in the detoxification of reactive aldehydes, which are highly reactive organic chemical compounds that mostly arise due to oxidative stress135. Proteins involved in the degradation of aromatic compounds were also identified, such as phenols, cresols, catechols, and diphenyl phosphate (DPHP). Phenols and cresols are harmful to the environment136, and catechol bioaccumulation negatively affects the entire ecosystem137. DPHP is used as a chemical additive in numerous industrial products, and because it does not bind with other chemicals, it is easily spread to the environment, where it has been widely detected138-14°. When in the environment, DPHP has a long half-life and is immunotoxic and neurotoxic to other organisms141. Despite not being comprehensively studied in marine organisms, recent studies have reported its negative effects on fish growth, energy metabolism, and reproduction142“145. Lastly, a regulatory system was identified that functions under specific stress conditions, such as hypoxia and starvation, and appears to be beneficial to host health and resilience. pBMC and the importance of their secondary metabolites

[0423] Secondary metabolites are important for defense against microorganisms, toxic compounds, and UV radiation as well as essential for symbiotic relationships146; they are abundantly found within the coral holobiont147. Accordingly, the BGCs and Pfams of the six pBMC strains was assessed using the anti SMASH platform, revealing 11 secondary metabolite production clusters: ectoine, beta-lactone, terpene, aryl polyene, lasso peptide, NRPS, nonribosomal peptide metallophores, NI siderophores, opine-like zincophores, class I and IV lanthipeptides, and type I and III PKS. Ectoines are important because they reduce the effects of heat stress, high salinity, ROS, and radiation. Beta-lactone derivatives are extremely diverse, comprising 30 distinct families, many of which have antimicrobial activity148, while others are important elements for antibiotic production146. Terpenes are diverse organic compounds that play a role in defense mechanisms in plants and fungi149,15°, serving as antioxidants and protecting cells from oxidative stress151, and their presence indicates production of diverse bioactive compounds152. Some terpenes and carotenoids, such as squalenes, are pigments involved in photosynthesis and signaling. However, they also possess antioxidant activity and can neutralize oxidative stress153-155. Aryl polyenes are pigments that are structurally and functionally related to carotenoids and confer protection against photo- oxidative damage and lipid peroxidation156 157Lasso peptides have antimicrobial and antiviral properties and also show thermal and chemical resistance146 158. NRPSs are sources of newly discovered antibacterial agents that have also been widely studied for their antiviral and anti-inflammatory properties146 159.

[0424] Microorganisms scavenge metal ions from the environment via metallophores160. Some of these metallophores, such as siderophores like desferrioxamine E, play important roles in biocontrol and bioremediation161-163, and are considered beneficial when selecting for BMC33. Zinc is also an essential nutrient for several cellular processes and is taken from the surrounding environment by bacteria164. Similar to iron availability, zinc is also present at low concentrations in the environment and is captured by zincophores (also called opine- like zincophores in bacteria), such as bacillopaline, produced by microorganisms165. Lanthipeptides that possess antimicrobial activity are known as lantibiotics166, but their functions are not limited to this; they may also possess antifungal and antiviral properties 167,168 Tepresence ofat ieastonep[<s BGC in each pBMC genome suggests that these bacterial strains have the necessary genomic tools to synthesize various polyketides that are likely to have beneficial properties159.

[0425] Red Sea pBMC and their Indo-Pacific Ocean counterparts

[0426] When examining the pangenomes of each genus, it was noted that the pBMC had a unique set of genes that were absent in the genomes of the BMC from the study by Rosado and colleagues26. Moreover, some of these genes represent potential beneficial traits or mechanisms (data not shown, Table 12)26. The results of the present study revealed potential insights into how bacteria help corals during periods of stress. Notably, all pBMC examined in this study, even those from the same genus, were distinct and contributed differently to the health and resilience of corals, highlighting the advantages of using microbial consortia107 169.

[0427] References

[0428] 1. Hoegh-Guldberg, O. et al. Science 318, 1737-1742 (2007).

[0429] 2. Heron, S. F., Sci Rep 6, (2016).

[0430] 3. Hughes, T. P. et al. Nature 2017546: 7656546, 82-90 (2017).

[0431] 4. Knowlton, N. et al. (2021) doi: 10.53642 / NRKY9386.

[0432] 5. McCulloch, M. T., Nature Climate Change 2024 14:2 14, 171-177 (2024).

[0433] 6. Lough, J. M. J. Geophys. Res 117, 9018 (2012).

[0434] 7. Duarte, G. A. S. et al. Front Mar Sci 7, 515393 (2020).

[0435] 8. Voolstra, EMBO Rep (2023) doi:10.15252 / EMBR.202356826.

[0436] 9. L., M. Coral Reefs : Ecosystems of the World 25, 75-87 (1990).

[0437] 10. Radecker, N. et al. Proc Natl Acad Sci USA 118, e2022653118 (2021).

[0438] 11. Cui, G. et al. Nature Communications 2023 14:1 14, 1-12 (2023).

[0439] 12. Radecker, N., Nature Communications 2023 14:1 14, 1-10 (2023).

[0440] 13. Baker, A. C., Estuar Coast Shelf Sci 80, 435-471 (2008).

[0441] 14. Oliver, J. (2018) doi:10.1007 / 978-3-319-75393-5_3.

[0442] 15. Burge, C. A. et al. Ann Rev Mar Sci 6, 249-277 (2014).

[0443] 16. Cantin, N. E. PLoS One 9, (2014).

[0444] 17. Peixoto, R. S. Front Mar Sci 10, 1218531 (2023).

[0445] 18. Santoro, E. P. et al. Sci. Adv vol. 7 https: / / www.science.org (2021).

[0446] 19. Alagely, A., The ISME Journal 2011 5:10 5, 1609-1620 (2011).

[0447] 20. Sharp, K. H. Biological Bulletin 223, 66-77 (2012).

[0448] 21. Krediet, C. J., Proceedings of the Royal Society B: Biological Sciences 280, (2013).

[0449] 22. Williams, A. D., PLoS One 10, e0144902 (2015).

[0450] 23. Hemandez-Agreda, A., mBio 7, (2016).

[0451] 24. Welsh, R. M. et al. PeerJ2017, e3315 (2017).

[0452] 25. Nair, S., Abraham, 453-478 (2019)

[0453] 26. Rosado, P. M. et al. The ISME Journal 2019 13:4 13, 921-936 (2019).

[0454] 27. Mohamed, FEMS Microbiol Rev 47, (2023).

[0455] 28. Reshef, L., Environ Microbiol 8, 2068-2073 (2006).

[0456] 29. Apprill, A., Appl Environ Microbiol 78, 7467 (2012).

[0457] 30. Bourne, D., The ISME Journal 2008 2:4 2, 350-363 (2007).

[0458] 31. Ziegler, M. et al. Nature Communications 2019 10:1 10, 1-11 (2019).

[0459] 32. Tracy, A. NEnviron Microbiol Rep 7, 471-479 (2015).

[0460] 33. Peixoto, R. S., Frontiers in Microbiology vol. 8 (2017). 34. Fragoso et al. Scientific Reports 2015 5:1 5, 1-11 (2015).

[0461] 35. Peixoto, R. S., Front Mar Sci 6, 487855 (2019).

[0462] 36. Damjanovic, K., Front Microbiol 10, 452611 (2019).

[0463] 37. Morgans, C. A., Restor Ecol 28, 282-288 (2020).

[0464] 38. Doering, T. et al. Microbiome 9, 1-16 (2021).

[0465] 39. Silva, D. P. et al. Microbiome 9, 1-19 (2021).

[0466] 40. Zhang, Y. et al. BMC Microbiol 21, 1-14 (2021).

[0467] 41. Li, J. et al. Environ Microbiome 18, 1-14 (2023).

[0468] 42. Moradi, M. et al. Front Mar Sci 10, 1212690 (2023).

[0469] 43. Peixoto, R. S. et al. 9, 265-288 (2021).

[0470] 44. Peixoto, R. S. et al. (2023).

[0471] 45. Schwyn, B. Anal Biochem 160, 47-56 (1987).

[0472] 46. Nautiyal, C. S. FEMS Microbiol Lett 170, 265-270 (1999).

[0473] 47. Ben-Haim, Y., Appl Environ Microbiol 69, 4236-4242 (2003).

[0474] 48. Giambiagi-Marval, M., J Gen Microbiol 136, 1591-1599 (1990).

[0475] 49. Fukasawa, Y., G3 Genes\Genomes\Genetics 10, 1193-1196 (2020).

[0476] 50. Koren, S. et al. Genome Res 21, 722-736 (2017).

[0477] 51. Kolmogorov, M., Nature Biotechnology 2019 37:5 37, 540-546 (2019).

[0478] 52. Vaser, R. & Sikic, M. Nature Computational Science 2021 1:5 1, 332-336 (2021).

[0479] 53. Wick, R. R. FlOOOResearch 2019 8:2138 8, 2138 (2019).

[0480] 54. Ruan, J. & Li, H. Nat Methods 17, 155-158 (2020).

[0481] 55. Gurevich, A., Saveliev, V., Vyahhi, N. & Tesler, G. QUAST: quality assessment tool for genome assemblies. Bioinformatics 29, 1072-1075 (2013).

[0482] 56. Parks, D. H., Genome Res 25, 1043-1055 (2015).

[0483] 57. Orakov, A. et al. Genome Biol 22, 1-19 (2021).

[0484] 58. Yoon, S. H., Antonie van Leeuwenhoek, International Journal of General and Molecular Microbiology 110, 1281-1286 (2017).

[0485] 59. Meier-Kolthoff, J. P., Nucleic Acids Res 50, D801-D807 (2022).

[0486] 60. Seemann, Bioinformatics 30, 2068-2069 (2014).

[0487] 61. Olson, R. D. et al. Nucleic Acids Res 51, D678-D689 (2023).

[0488] 62. Brettin, T. et al. Scientific Reports 2015 5:1 5, 1-6 (2015).

[0489] 63. Wishart, D. S. et al. Nucleic Acids Res 51, W443-W450 (2023).

[0490] 64. Camargo, A. P. et al. Nucleic Acids Res 51, D733-D743 (2023).

[0491] 65. Blin, K. et al. Nucleic Acids Res 49, W29-W35 (2021).

[0492] 66. Davis, J. J. et al. Front Microbiol 7, 118 (2016).

[0493] 67. Edgar, R. C. MUSCLE: BMC Bioinformatics 5, 1-19 (2004).

[0494] 68. Cock, P. J. A. et al. Bioinformatics 25, 1422-1423 (2009).

[0495] 69. Stamatakis, A. Bioinformatics 30, 1312-1313 (2014).

[0496] 70. Stamatakis, A., SystBiol 57, 758-771 (2008). 71. Le, S. Q. & Gascuel, O. Mol Biol Evol 25, 1307-1320 (2008).

[0497] 72. Letunic, I. Nucleic Acids Res 49, W293-W296 (2021).

[0498] 73. Page, A. J. et al. Bioinformatics 31, 3691-3693 (2015).

[0499] 74. Eren, A. M. et al. Community -led, integrated, reproducible multi-omics with anvi’o. Nature Microbiology 2020 6:1 6, 3-6 (2020).

[0500] 75. Bateman, A. et al. Nucleic Acids Res 51, D523-D531 (2023).

[0501] 76. Meier-Kolthoff, J. P., Arch Microbiol 195, 413-418 (2013).

[0502] 77. Kim, M., Int J Syst Evol Microbiol 64, 346-351 (2014).

[0503] 78. Woo, P. C. Y., Clinical Microbiology and Infection 14, 908-934 (2008).

[0504] 79. Gupta, R. S., Int J Syst Evol Microbiol 70, 5753-5798 (2020).

[0505] 80. Rosado, P. M. et al. mSystems 8, (2023).

[0506] 81. Matthews, J. L. et al. Environmental Microbiology vol. 22 1675-1687 (2020).

[0507] 82. Hayyan, M., Chem Rev 116, 3029-3085 (2016).

[0508] 83. Rochat, T. et al. Microbiology (N Y) 151, 3011-3018 (2005).

[0509] 84. Bohacova, V., Curr Microbiol 53, 430-434 (2006).

[0510] 85. Jittawuttipoka, T., J Bacteriol 191, 7372-7377 (2009).

[0511] 86. Godobikova, J., Arch Microbiol 192, 175-184 (2010).

[0512] 87. Bemroitner, M., J Exp Bot 60, 423-440 (2009).

[0513] 88. Kono, Y. & Fridovich, I. J Bacteriol 155, 742 (1983).

[0514] 89. Whittaker, J. W. Arch Biochem Biophys 525, 111-120 (2012).

[0515] 90. Nja Lsson, R. & Norgren, Acta Paediatr 94, 132-137 (2005).

[0516] 91. Barycki, J. J. et al. Redox Biochemistry 11-47 (2007)

[0517] 92. Weis, V. M., Trends Ecol Evol 23, 369-376 (2008).

[0518] 93. Roberty, S., Plant Cell Environ 39, 2713-2724 (2016).

[0519] 94. Szabo, M., Advances in Photosynthesis and Respiration 45, 459-488 (2020).

[0520] 95. Ashoori, M. doi: 10.1017 / S0007114514000178.

[0521] 96. Percudani, R. & Peracchi, A. EMBO Rep 4, 850-854 (2003).

[0522] 97. Sanudo-Wilhelmy, S. A. et al. Proc Natl Acad Sci USA 109, 14041-14045 (2012).

[0523] 98. Gomez-Consamau, L. et al. Environ Microbiol 20, 2809-2823 (2018).

[0524] 99. Havaux, M. et al. BMC Plant Biol 9, 1-22 (2009).

[0525] 100. Vanderschuren, H. et al. Front Plant Sci 4, 48827 (2013).

[0526] 101. Zempleni, J., Biofactors 35, 36-46 (2009).

[0527] 102. Croft, M. T., Nature 438, 90-93 (2005).

[0528] 103. Hochart, C. et al. Nature Communications 2023 14:1 14, 1-16 (2023).

[0529] 104. Tang, Y. Z., Proc Natl Acad Sci USA 107, 20756-20761 (2010).

[0530] 105. Robbins, S. J. et al. Nat Microbiol 4, 2090-2100 (2019).

[0531] 106. Pogoreutz, C. et al. The ISME Journal 2022 16:8 16, 1883-1895 (2022).

[0532] 107. do Carmo, F. L. et al. Journal of Microbiology 49, 535-543 (2011).

[0533] 108. Behrenfeld, M. J. et al. Biogeosciences Discuss 5, 4235-4270 (2008). 109. Hopkinson, B. M. BioMetals 22, 659-669 (2009).

[0534] 110. Amin, S. A. et al. Proc Natl Acad Sci USA 106, 17071-17076 (2009).

[0535] 111. Lawson, C. A., Environ Microbiol Rep 10, 7-11 (2018).

[0536] 112. Jebbar, M., FEMS Microbiol Lett 154, 325-330 (1997).

[0537] 113. Shivanand, 100, 1516-1521 (2011).

[0538] 114. Pei, J. Y. et al. Anal Bioanal Chem 414, 7635-7646 (2022).

[0539] 115. Ngugi, D. K., iScience 23, 101120 (2020).

[0540] 116. Fenizia, S., Marine Drugs 2020, Vol. 18, Page 42 18, 42 (2020).

[0541] 117. Van-Thuoc, D., Appl Microbiol Biotechnol 97, 6271-6278 (2013).

[0542] 118. Teixido, N. et al. Lett Appl Microbiol 41, 248-252 (2005).

[0543] 119. Bownik, A., J Comp Physiol B 184, 961-976 (2014).

[0544] 120. Pastor, J. M. et al. Biotechnol Adv 28, 782-801 (2010).

[0545] 121. Roach, T. N. F. et al. Nature Ecology & Evolution 2021 5:4 5, 495-503

[0546] 122. Radecker, N. et al. The ISME Journal 2021 16:4 16, 1110-1118 (2021).

[0547] 123. Yost, D. M., Aquatic Toxicology 98, 367-373 (2010).

[0548] 124. Sunda, W., Nature 2002 418:6895 418, 317-320 (2002).

[0549] 125. Garren, M. et al. The ISME Journal 2014 8:5 8, 999-1007 (2013).

[0550] 126. Weynberg, K. D., Scientific Reports 2015 5:1 5, 1-9 (2015).

[0551] 127. Bondy-Denomy, J. et al. The ISME Journal 2016 10:12 10, 2854-2866

[0552] 128. Tout, J. et al. The ISME Journal 2015 9:8 9, 1764-1777 (2015).

[0553] 129. Monperrus, M. et al. Journal of Env Monitoring 7, 693-700 (2005).

[0554] 130. Fang, T. H., Mar Environ Res 61, 224-243 (2006).

[0555] 131. Das, N., Indian J Biotechnol 7, 159-169 (2008).

[0556] 132. Dash, H. R., Env Science and Pollution Research 21, 2642-2653 (2014).

[0557] 133. Singh, S., Curr Opin Biotechnol 19, 437-444 (2008).

[0558] 134. Zhang, W., Appl Microbiol Biotechnol 93, 1305-1314 (2012).

[0559] 135. Fritz, K. S. Free Radio Biol Med 59, 85-91 (2013).

[0560] 136. Duan, W. et al. Ecotoxicol Environ Safl57, 441-456 (2018).

[0561] 137. Paasivirta, J. et al. Chemosphere 14, 469-491 (1985).

[0562] 138. Hoffman, K., Environ Health Perspect 123, 160-165 (2014).

[0563] 139. Zhao, F. et al. (2016) doi:10.1021 / acs.est.6b02474.

[0564] 140. Yan, Z. et al. Environmental Pollution 274, 116586 (2021).

[0565] 141. Ren, R. et al. 3 Biotech 11, 1-9 (2021).

[0566] 142. Zhu, Y. et al. Environ Sci Technol 49, 14579-14587 (2015).

[0567] 143. Li, Y. et al. Environ Sci Technol Lett 5, 649-654 (2018).

[0568] 144. Chen, Q. et al. Environ Sci Technol 55, 13122-13131 (2021).

[0569] 145. Yang, R. et al. Environ Int 158, 106928 (2022).

[0570] 146. Cho, S. H. et al. J. Microbiol. Biotechnol. 30, 505-514 (2020).

[0571] 147. Modolon, F., J Appl Microbiol 129, 1441-1457 (2020). 148. Robinson, S. L., Nat Prod Rep 36, 458-475 (2019).

[0572] 149. Yamada, Y. et al. Proc Natl Acad Sci USA 112, 857-862 (2015).

[0573] 150. Oldfield, E.. Angew Chem IntEdEngl 51, 1124-1137 (2012).

[0574] 151. Baccouri, B. & Rajhi, I. (2021) doi:10.5772 / INTECHOPEN.96638.

[0575] 152. Cheng, A. X. et al. J Integr Plant Biol 49, 179-186 (2007).

[0576] 153. Stahl, W. & Sies, Mol Aspects Med 24, 345-351 (2003).

[0577] 154. Spanova, M. European J of Lipid Sci and Tech 113, 1299-1320 (2011).

[0578] 155. Saini, R. K. et al. Antioxidants 2022, Vol. 11, Page 239 11, 239 (2022).

[0579] 156. Rajagopal, L., FEBS Lett 415, 125-128 (1997).

[0580] 157. Schoner, T. A. et al. ChemBioChem 17, 247-253 (2016).

[0581] 158. Zhu, S. et al. Journal of Biological Chemistry 291, 13662-13678 (2016).

[0582] 159. Raimundo, I., Mar Drugs 16, (2018).

[0583] 160. Reitz, Z. L., bioRxiv 2022.12.14.519525 (2022)

[0584] 161. Tam, T. et al. Curr Med Chem 10, 983-995 (2005).

[0585] 162. Johnstone, T. C. Dalton Transactions 44, 6320-6339 (2015).

[0586] 163. Soares, E. V. Applied Microbiology and Biotechnology 2022 106:11 106, 3985-4004 (2022).

[0587] 164. Andreini, C., JI of Biol Inorganic Chemistry 13, 1205-1218 (2008).

[0588] 165. Morey, J. R. mSystems 5, (2020).

[0589] 166. Willey, J. M. & Van Der Donk, 61, 477-501 (2007).

[0590] 167. Ferir, G. et al. PLoS One 8, e64010 (2013).

[0591] 168. Mohr, K. I. et al. Chemie International Edition 54, 11254-11258 (2015).

[0592] 169. Garcias-Bonet, N. et al. Trends Microbiol 0, (2023).

[0593] 170. Schultz, J. et al. mSystems 7, (2022).

[0594] 171. Sweet, M. et al. mSystems 6, (2021).

[0595] 172. Doering, T. et al. Microbiome 11, (2023).

[0596] 173. Maire, J. et al. Functional potential and evolutionary response to long-term heat selection of bacterial associates of coral photosymbionts. mSystems 8, (2023).

[0597] 174. Peixoto, R. S. et al. Nature Microbiology 2022 7:11 7, 1726-1735 (2022).

[0598] 175. Carver, T., Bioinformatics 25, 119-120 (2009).

[0599] EXAMPLE 4- Testing stress antagonists for their potential to enhance coral recovery after natural heat exposure

[0600] Potential Stress Antagonists

[0601] To address the need for solutions to enhance coral resilience, the present study focuses on testing potential stress antagonists to aid in the rapid recovery of corals following bleaching events. Four candidates were tested: hydrogen, phosphate, ammonium, and probiotics. Hydrogen (H2), exhibits properties that suggest it may serve as an effective antagonist. The goal of the present study was to test the effects of these stress antagonists in aiding coral recovery on the exemplary hard-coral species Acropora spp. and Pocillopora verrucosa, with a focus on their physiological parameters. The results can provide valuable insights for conservation strategies in the face of increasing heat exposure.

[0602] Materials and Methods

[0603] Experimental Design

[0604] To assess the short-term effect of different stress antagonists on the recovery potential of hard corals, two hard coral species, Acropora spp. and Pocillopora verrucosa, commonly found in the Red Sea were collected. Fragments were sampled at the Coral Probiotics Village, Al Fahal Reef (22O18'19.1"N, 38°57'55.0"E), located about 20 km off the King Abdullah University of Science and Technology, Thuwal, in the Saudi Arabian coast of the Red Sea. Sampling was conducted while SCUBA diving, using pliers (Pliers, ACE) to cut coral fragments about 2-4 cm length from 5 coral colonies per species at approximately 10 m depth. All fragments upon collection in the field looked healthy and, except for Pocillopora Colony 1, did not visually appear to be bleached. Colonies were marked to allow resampling of the same colonies throughout the experiment duration. Due to space limitations, each experimental run had to take place separately, to be able to run incubations, yet all experiments were conducted within a time period of two weeks.

[0605] On each sampling day, two fragments per colony were collected and placed in black sampling bags filled with seawater along with 80 L of seawater in plastic buckets. Fragments were then transported to the laboratory in a plastic box (Cooler Rubbermaid 25x36 cm 22 L) filled with seawater from the sampling site, to shield them from high irradiation and temperatures on board during boat transfer. After arrival, all fragments were photographed for visual coloration, then one fragment per colony was frozen and placed in the -80 °C freezer for physiological analyses at the start of the experiment, while the other fragment was placed into a 3D printed holder and then transferred to the experimental tanks.

[0606] Experimental tanks were small glass aquaria filled with 10 L of seawater collected from the reef. In total 5 aquaria were used, each holding one randomly chosen Acropora spp. and P. verrucosa fragment (in addition to one turf algae and Xenia spp. fragment), to avoid tank effects. The experimental tanks were equipped with an aquarium pump (Barst Submersible Pump, JR-250B) and a heater (Schego heater titanium tube IP68, 200 W 250 mm x 25 mm) connected to a thermostat (Inkbird temperature controller ITC-308) while natural lighting was simulated with a LED lamp replicating a 12 h light / dark-cycle (PAR ~ 200 pmol m'2s'1, salinity 39.0 ± 1.0 %o, temperature ambient 30.0 ± 0.5 °C) . Salinity was adjusted three times a day, in the mornings when the light was turned on, middays, and in the evening when the lights were turned off.

[0607] On each sampling day, an experiment with a different treatment was started. Treatments consisted of 1) “control” with no addition of anything, 2) “hydrogen” (H2) at a concentration of 0.3-0.5 ppm by exchanging 10 % of tank water with hydrogen-enriched distilled water, supplied with Red Sea salt using a hydrogen water generator (Hydrogen-rich Water Cup, ABS-FQ-02, Aukewel), 3) “phosphate” (PO4) at a concentration of 2 pM (Sigma Aldrich, Na2HPO4x2 H2O, CAS 10028-24-7), 4) “ammonium” (NH4) at a concentration of 3 pM (Sigma Aldrich, NH4C1, CAS: 12125-02-9), and 4) a “probiotic” treatment using tailor- made BMCs.

[0608] Oxygen Fluxes

[0609] For oxygen flux measurements, fragments were placed into a 3D-printed holder and placed into an incubation jar without being exposed to air. Jars were closed airtight for the duration of both the light and dark incubations and placed in a water bath set to the respective treatment temperature and on magnetic stirrer plates (Hanna instruments, Magnetic stirrer HI 200M) with a magnetic stirrer in each jar to ensure constant water mixing. At the beginning and end of each incubation the oxygen concentration in the water was measured using an Oxygen probe (Mettler Toledo SevenGo Duo Pro with InLab OptiOx). Both light and dark incubations were around 1.5 hours long, with light incubations using an LED Aquarium Light with an intensity of around 200 pmol m'2s'1. After the dark incubation fragments underwent PAM measurements. Oxygen flux data, net photosynthesis (Pnet) and respiration (R) rates, was calculated by subtracting start and end oxygen concentrations of the light and dark incubations, respectively. Rates were then normalized by the surface area of each fragment (SA), total volume of the incubation jar (V), total incubation time (h) and the background oxygen fluxes (blank) as follows:

[0610] Pnet or R = oxy / h — blank / h) / SA * V

[0611] Using Pnet and R, the gross photosynthesis (Pgross) was then calculated according to following formula:

[0612] Pgross = Pnet + | / ?|

[0613] PAM Measurements Photosynthetic efficiency was measured directly at the start of the experiment, after 12hours, 24hours and after the dark incubation after 48h using a Diving PAM (DIVING- PAM-II / R, Walz GmbH) according to Ralph & Gademann (2005) and Ralph et al. (2005), after the corals were dark-adapted for >30 min. For each fragment, the dark level fluorescence yield (F0) and the maximum fluorescence yield (Frri) were measured, and the optimal PSII quantum yield FvIFrri) was calculated for a random point placed on the coral fragment. In addition, a rapid light curve (RLC) was generated for a different point through an exposure of ten light-series with increasing irradiance from 0 to 1455 pmol photons m'2s' f Between the twelve light-series, the fragments were illuminated with the respective actinic light for 10 seconds, each ended by a saturation pulse.

[0614] After the PAM measurements, fragments were photographed for visual coloration and survival assessment, and subsequently stored in the - 80 °C freezer until further analysis.

[0615] Visual Coloration and Survival

[0616] Visual coloration was assessed by taking pictures of each fragment against a white background using a mobile phone camera (iPhone 7) before the start of the experiment and at the end of the 48-h stress antagonist run. Pictures were analyzed following the protocol of (Mclachlan & Grottoli, 2021) using the software ImageJ taking ten separate gray values per fragment, standardizing them against the white background, and calculating the mean intensity gray (MIG) values. MIG values were inverted, so that a lower value, or a reduction indicates paling / bleaching of the fragment, while higher values correlate with darkening of the fragment.

[0617] Survival was determined by two factors: Either a photosynthetic efficiency (FvIFrri) below 0.3, and / or more than 75 % tissue loss of the fragments.

[0618] Airburshing and Surface Area Determination

[0619] Hard coral fragments were defrosted on ice for Ih before airbrushing the tissue off the skeleton using 5ml of micro filtered seawater and an airbrush gun. Tissue slurry was collected in a falcon tubes (VWR 15ml Centrifuge Tube, Cat. No. 525-1084) and then transferred into 3 different 1.5ml Eppendorf tubes (Fisher Scientific, Cat. No.: 05-408-137) and frozen again at -80 °C until further physiological analyses.

[0620] Surface area was determined by building a 3D model using the software Agisoft Metashape (Version 2.1.0, Professional) from 50 pictures taken of each fragment. Models were then analyzed using MeshLab (Version 2023.12). Chlorophyll Content

[0621] To measure total pigment content, slurries were removed from the freezer and placed on ice for thawing. All steps of measurement preparation were conducted in the dark under red light. Slurries were homogenized by vertexing them (VWR vortexer mini, Cat. No. : 58816-123). Afterward, 500 pl of samples were aspirated into a 2 ml Eppendorf tube (Fisher Scientific, Cat. No.: 05-408-146) and centrifuged for 5 minutes (4 °C, 5000 rpm, Eppendorf centrifuge 5424 R). Supernatant was discarded and 2 mL of 90 % acetone (VWR, Acetone for HPLC, Cat. No.: 20067.320) added. Samples were sonicated in an ice water bath for 10 min (Bransonic Ultrasonic Cleaner, model: 5510E-MTH) and then placed inside a cardboard box in a fridge at 4 °C for 48 hours of extraction time. After the extraction, samples were taken from the fridge, vortexed again for 2 seconds and then placed in a benchtop centrifuge (Cole Parmer, Veron Hills, Illinois 60061) for 5 seconds before pipetting 200 pl of supernatant into the respective well plates in duplicates. In each 96-well plate (VWR Tissue Culture plate, European Cat. No.: 734-2327) two wells were filled with 200 pl of 90 % acetone for blanks. Samples were then placed in a spectrophotometer plate reader (Thermo Scientific, spectrophotometer, Ref.: A51119700) and read at wavelengths of 630, 663, 664, and 750 nm wavelength. Sample absorbance values were corrected by subtracting the average absorbance of the blanks. Afterwards, chlorophyll a and c2 concentrations were calculated using the equations of Jeffrey and Humphrey (1975), while accounting for the path length of the well plate by dividing the concentrations by 0.555 (Krueger, 2020). After that values were converted from concentration to the absolute amount of extracted chlorophyll, by multiplying them with the extraction volume, and then multiplied by the dilution factor. Lastly, the absolute amount of chlorophyll a and c2 to the surface area of the respective fragment was quantified.

[0622] Protein Content

[0623] Protein concentrations were determined using a Bradford assay kit (Thermo Scientific, Coomassie Protein Assay Kit, Ref.: 23200). First, a dilution of standards was mixed using an Albumin Standard (Thermo Scientific, Albumin Standard, Ref.: 23209) and MilliQ water for dilution. Tissue slurry was thawed on ice and then homogenized thoroughly by vertexing (VWR vortexer mini, Cat. No.: 58816-123). Depending on the sample the tissue slurry was diluted in a new 1.5 ml Eppendorf cup (Fisher Scientific, Cat. No.: 05-408-137) with MilliQ water to fall into the range of the standard curve. After homogenization, the samples were placed in a small benchtop centrifuge (Cole Parmer, Veron Hills, Illinois 60061) for 5 seconds. Then, 5 pl of sample supernatant was transferred in technical triplicates in a 96-well plate (VWR Tissue Culture plate, European Cat. No.: 734-2327). Using a multichannel pipette 250 pl of Coomassie reagent was transferred into each well and properly mixed with the sample without producing bubbles. The 96-well plate was placed on a shaker (Eppendorf, MixMate) for 30 seconds at 500 rpm, and subsequently incubated at room temperature for 10 min. After the incubation, the plate was placed back on the shaker for 30 seconds at 500 rpm, and then put into a spectrophotometer plate reader (Thermo Scientific, spectrophotometer, Ref.: A51119700) to measure the absorbance of the wells at a wavelength of 595 nm. Sample absorbance was corrected by subtracting the average absorbance of the blank standard, then absorbance values more than 0.03 units away within the triplicates were discarded. Protein concentration was then calculated using the standard curve produced (using a quadratic curve fit, forced through the origin), multiplied by the respective dilution factors of tissue slurry used and additional dilutions, and, finally, standardized to surface area.

[0624] Data Analysis

[0625] To account for possible differences in the starting values of the separate experimental runs the percentage change from start to end point for visual coloration was calculated, FvFm, chlorophyll analyses, and protein content, which was further normalized using the formula:

[0626] ValueNormalized = (ValueOriginal / 0) + 1

[0627] Data was then analyzed using a non-parametric permutational multivariate analysis of variance (PERMANOVA) using PRIMER-E version 6 software with the PERMAONVA+ add-on. To test for differences between antagonist treatments separately for each species a 1- factorial PERMANOVA was performed, based on Euclidean distance resemblance measures. Type III (partial) sum of squares was used with unrestricted permutation of raw data (999 permutations), and PERMANOVA pairwise tests with parallel Monte Carlo tests were carried out when significant differences were found. The results were considered to be significant at a - value lower than 0.05 (p < 0.05). Results

[0628] Control Treatment

[0629] A survivorship of 60% was observed fox Acropora spp. fragments placed in the control treatment. Coloration, estimated as mean intensity gray values (Figure 24), was reduced by 8 %, and also decreasing chlorophyll values of -38 % for chlorophyll a (Figure 25) and -46 % for c2, the highest decrease among all treatments, were observed. Photosynthetic efficiency in terms of Fv / Fm declined by 12 % fox Acropora spp. (Figure 26). Furthermore, Acropora spp. showed the highest Pnet values (data not shown) with 55.3 mg O2 m-2h-1and with a respiration of -107.4 mg O2 m-2h-1having a PgroSs:R ratio of 1.3 (Figure 27), which is in the middle of all other treatments. Protein content fox Acropora spp. increased by 29 % over the 48-h period (data not shown).

[0630] As for P. verrucosa, only 40 % of the fragments in the control tanks survived. Their coloration was reduced by 10%, the highest color loss among all treatments, aligned with the decreased values in both chlorophyll a and c2 values. For P. verrucosa both chlorophyll values decreased the most among all treatments with -40 and -41 %, respectively. In addition, photosynthetic efficiency declined by 24 %, also the sharpest decline among all treatments. This negative trend continues with P. verrucosa in the control treatment also experiencing the lowest Pnet rates of only 2.7 mg O2 m-2h-1, and the lowest PgrOss:R ratio of 1.0, and the biggest decrease in protein content with -37 %.

[0631] Hydrogen Treatment

[0632] The H2 treatment increased the survival for Acropora spp. by 25 %, to a total survival of 75 % of the fragments. Yet, the photo physiology of Acropora spp. was not improved but the worst compared to all other treatments, including the control. Coloration had the biggest decrease of all, dropping 10 %, in line with a decrease of 43 and 35 % in chlorophyll a and c2 concentrations. Vox Acropora spp. hydrogen also led to the largest decline in Fv / Fm with -24 %, the lowest, and negative, Pnet rates of -30.3 mg O2 m-2h-1, highest R rates with -126.8 mg O2 m-2h-1, and the lowest PgroSs:R ratio of 1.0. Yet, protein content increased by 122 % over the experimental period.

[0633] For P. verrucosa, on the other hand, H2 increased survival to 80 %, which is 40 % higher than in the control treatment. All other parameters slightly improved compared to the control treatments, with coloration and Fv / Fm only decreasing by 3 %, and a loss in chlorophyll a and c2 by 36 and 32 %, respectively. The oxygen fluxes under H2 enrichment, fragments showed higher Pnet and R rates compared to the control, also leading to an overall higher PgrOss:R ratio of 1.2. Finally, protein content had the lowest decrease under H2 with only 1 % decline.

[0634] Phosphate Treatment

[0635] For both hard corals, PO4 increased survival by 40 %, to an overall survival of 100 % in Acropora spp. and 80 % in P. verrucosa. Apart from that all other parameters fall in line with the other treatments, without any extreme positive or negative response. Though, notably, for both hard corals respiration under PO4 enrichment was the lowest or Acropora spp. and P. verrucosa with only -79.0 and -74.2 mg O2 m-2h-1, respectively.

[0636] Ammonium Treatment

[0637] Like PO4, also NFL increased survival by 40 %, to an overall survival of 100 % in Acropora spp. Looking at the coloration NFL led to the least severe decline in coloration with -2 %, and chlorophyll a and c2 with -10 and -9 %. Fv / Fm increased significantly by 11 % over the 48-h period compared to the control (PERMANOVA, pMC = 0.04). While Pnet and R rates were not the best in the NFL treatment, the overall PgrOss:R ratio was the highest of all treatments (1.5). Yet, protein content had the largest decrease under NFL enrichment (-15 % overall).

[0638] P. verrucosa showed similar trends to Acropora spp. under NFL enrichment with an increase in survival by 40 %. Besides this, NFL increased the coloration of fragments by 2 %, most likely due to significantly increased levels of chlorophyll a by 27 % (PERMANOVA, pMC = 0.04) compared to the control treatment. NFL also increased Fv / Fm by 13 %, leading to the highest measured Pnet rate of 54.4 mg O2 m-2h-1, and a PgrOss:R ratio of 1.6, the highest among all treatments.

[0639] Probiotics

[0640] The probiotic treatment increased the survival of Acropora spp. fragments by 20 %, compared to the control treatment. While coloration and chlorophyll a and c2 concentrations decreased by 5, 29, and 24 %, respectively, the photosynthetic efficiency (Fv / Fm) still increased by 11 %, being significantly higher than in the control (PERMANOVA, pMC = 0.02). None of the other parameters showed an extreme positive or negative response.

[0641] Applying the BMC1 probiotic mix on P. verrucosa led to an increase of 40 % in survival of fragments. Though chlorophyll a and c2 concentrations declined by 21 and 8 %, respectively, both coloration and Fv / Fm increased by 1 and 13 % over time, respectively. Interestingly, probiotics treatment led to the highest R rate for P. verrucosa among all treatments with -110.5 mg O2 m-2h-1. Tablel2: Survival of hard corals in the different experimental treatments (n = 5 per treatment and species, except n = 4 for Acropora spp. in the hydrogen treatment).

[0642] Discussion

[0643] This study is the first to comprehensively evaluate the effects of different stress antagonists on short-term coral recovery after a marine heat wave in a comparative framework. Hydrogen, phosphate, ammonium and custom-designed probiotics were subjected to comparative analysis, providing insights into their potential to enhance coral recovery following heat stress. This approach emphasizes the comparative evaluation of these stress antagonists, with the aim of elucidating the nuanced responses of hard corals to different interventions and informing targeted conservation strategies. The aim of this study was to understand coral stress responses and recovery mechanisms and provide evidencebased tools for coral reef conservation and management.

[0644] Effect of Hydrogen

[0645] In the control treatment, corals showed clear signs of stress during the 48-h experiment, with the highest overall mortality and bleaching, and impaired photosynthesis with the highest respiration and lowest photosynthesis rates, particularly for P. verrucosa fragments. Under H2 enrichment, an increase in survival for both Acropora spp. and P. verrucosa was observed, with survival rates increasing by 15% for Acropora spp. and 40% for P. verrucosa fragments.

[0646] Effect of Phosphate

[0647] The observed increase in survival rates for both Acropora spp. and P. verrucosa under PO4 enrichment highlights a promising avenue for enhancing coral resilience to heat stress. In particular, the significant improvement in survival, reaching 100 % for Acropora spp. and 80 % for P. verrucosa, demonstrates the effectiveness of PO4 as a stress antagonist in shortterm coral recovery efforts. Effect of Ammonium

[0648] The use of NH4 enrichment as a stress antagonist in coral recovery after heat exposure presented interesting results and provides deeper insights into the dynamics of coral-algal symbiosis and nutrient cycling within the coral holobiont. The present study demonstrated that NH4 enrichment significantly improved short-term survival and P. verrucosa, with a remarkable increase of 40 %, resulting in an overall survival of 100 % of Acropora spp. fragments. Furthermore, fragments exposed to enriched NH4 conditions exhibited the highest Pgross:R ratios, increased Fv / Fm values, chlorophyll concentrations and coloration values compared to other treatments.

[0649] Effect of Custom-made Probiotics

[0650] In the present study, notable improvements were observed in survival rates for both Acropora spp. and P. verrucosa following probiotic treatment, with survival increasing by 20 % for Acropora spp. and 40 % for P. verrucosa compared to control treatments. In addition, a reduction in bleaching and chlorophyll concentrations was observed under probiotic treatment.

[0651] Interestingly, the probiotic treatment led to the highest respiration rates measured in P. verrucosa fragments in the present study. The photosynthetic efficiency (Fv / Fm) values observed in the present study showed a notable increase for both Acropora spp. and P. verrucosa, increasing by 11% and 13% respectively. The present study uniquely demonstrated an actual increase in Fv / Fm over time under probiotic treatment, suggesting a more pronounced positive effect on coral health and photosynthetic performance. Furthermore, a significant correlation was found between Fv / Fm levels and certain physiological parameters. For Acropora spp. Fv / Fm was positively correlated with coloration, while for P. verrucosa it showed a positive correlation with chlorophyll a concentration. This suggests that the observed increase in Fv / Fm under probiotic treatment might be due to mitigation of bleaching and reduction in chlorophyll content, which were less severe in the probiotic treatment compared to the control treatment. These results highlight the potential role of probiotics in maintaining or enhancing key physiological processes essential for coral health and resilience.

[0652] By promoting shifts in the coral microbiome and enhancing metabolic activity, probiotics offer a promising strategy for enhancing coral resilience to environmental stressors. Furthermore, the present results highlight two key stress antagonists, NH4 and probiotics, which had the most positive effects on coral recovery. These interventions not only increased survival rates, but also significantly improved physiological parameters such as Fv / Fm ratio and chlorophyll a content during the short-term recovery experiment.

Claims

Attorney REF: KAUST 2023-049-02 PCTWe claim:

1. A BMC composition comprising a population of isolated bacteria selected species: OH (Pseudoalter omonas sp), 33H (Pseudoalteromonas sp.), SAT10 (Halomonas sp.), SAA19 (Bacillus sp.) / Sutcliffiella sp.), 65H (Cobetia sp.) and 81H (Cobet / a sp) in equal amounts and optionally, in combination with an effective amount of a stabilizer.

2. A BMC composition comprising a population of isolated bacteria selected species: 10 (Pseudoalteromonas sp), 16 (Pseudoalteromonas sp , SAT19 (Halomonas sp.), 81H (Cobetia sp) in equal amounts, and opitionally, in combination with an effective amount of a stabilizer.

3. The composition of claim 1 or 2, wherein the bacterial population is isolated from a coral hosts such selected from the group consisting of Pocillopora verrucosa, Stylophora pistillata and Galaxea fascicularis or from red sea acropora.

4. The composition of any one of claims 1-3, in the form of a liquid, optionally, wherein the liquid comprise a marine culture medium5. The composition of any one of claims 1-4, wherein the further comprising a marine culture medium, optionally, wherein the marine culture medium is agar or Luria-Bertani broth.

6. The composition of any one of claims 1-5, in the form of a solid, optionally, in a an organic carrier selected from the group consisting of coconut husk, chitin, alginate, agar, and guar gum or inorganic carrier selected from the group consisting of . charcoal, clay, silica, ceramics, and celite.

7. The composition of any one of claims 1-5 wherein the liquid is NaCL or glycerol.

8. The composition of claim 7 comprising about 3.5% NaCl.

9. The composition of claim 8, comprising about 20% glycerol .

10. The composition of claim 6, wherein the composition is in the form of a dry powder, grains or granules or pellets.Attorney REF: KAUST 2023-049-02 PCT11. The composition of any one of claims 1-10, wherein the bacterial population is isolated from Pocillopora verrucosa, Stylophora pistillata and Galaxea fascicularis or red sea acropora, by a method comprising:(a) isolating bacteria from Pocillopora verrucosa, Stylophora pistillata and Galaxea fascicularis or red sea acropora,(b) screening the bacteria isolates for traits selected from reactive oxygen species (ROS) scavengers, phosphate assimilation, an antagonistic effect against Vibrio coralliliticus, production of sciderophores, degradation of urease, and genes related to nitrogen cycling, as nitrogenase subunits (nifH) and denitrification (nirK), as well as the DMSP degradation (dmdA) gene (can be screened through whole genome analysis), and(c) selecting strains on the basis of being positive for one or more of the desired traits, culturing the selected strains and collecting the strain at the peak of its growth curve.

12. The composition of claim 11 wherein each strain is collected proportionally at the peak of its growth curve, when it reaches a concentration between about 108and about 109cells / ml per strain.

13. The composition of any one of claims 1-12, therein the composition does not include coral or clinical pathogens, optionally wherein the coral of clinical pathogen is a Vibrio spp.

14. A method of conferring onto coral or spongesone or more traits beneficial to the coral or sponge, the method comprising providing the coral with an effective amount of the composition of any one of claims 1-13.

15. The method of claim 14, wherein the coral is a red sea coral.

16. The method of claim 15, wherein the coral is P. verrucosa or Acropora spp.

17. The method of claim 14, wherein comprising treating the coral or sponge with about 10 mL to about 500 mL. of composition.Attorney REF: KAUST 2023-049-02 PCT18. The method of claim 14 or 17, wherein the sponge is C. crassa. or S. carteri19. The method of any one of claims 14, 17, or 18 wherein the beneficial trait is selected from the group consisting of an increased abundance of Synechococcus sp., Prochlorococcus sp. Moreover, the C. crassa microbiome and the ammonia-oxidizing Nitrosopumilus sp. (Thermoproleola). in the sponge, when compared to an untreated sponge.

Images