The Vibrio fortis strain S10-1 (MCCC 1H00104) was isolated from sea cucumber from the coral reef of Lembeh Strait, North Sulawesi, Indonesia, and kindly provided by the Third Institute of Oceanography, MNR (Xiamen, China). The corals S. guttatus and P. damicornis were artificially bred and kindly provided by the coral conservation laboratory of the Third Institute of Oceanography. The YtnP enzyme was prepared by this lab in a previous study (Sun et al., 2021).

The S10-1 strain was inoculated in a 100-ml of Marine Broth 2216 (BD, USA) media and incubated at 30°C with shaking at 180 rpm to obtain the bacterial growth curve. The overnight culture of S10-1 at exponential growth phase, with an absorbance of about 0.50 at the optical density (OD) at 595 nm (approximate 1.0 × 10⁸ CFU/ml), was resuspended using the same volume of seawater to prepare a bacterial solution for the further experiment of bacterial infection.

The aquaria with the size of 60 × 45 × 45 cm (length × wide × high) were prepared for coral culture, with 20 L pump-recycling seawater at a salinity of 33.3‰, a protein skimmer to remove dissolved organic compounds, and a dimmable LED light to simulate the diurnal rhythm (light on between 6 a.m. and 6 p.m.). The live stone was moved from the coral conservation laboratory, randomly separated into pieces, and pre-cultured individually in the aquaria for 7 days at 25°C to maintain clean and stable water quality for coral growth. The corals were cut into branches with a length of about 80 mm, randomly separated, and placed in the aquaria for the subsequent experiment, where each aquarium contains at least four coral branches for biological repetition. The overview of the bacterial infection experiment is shown in Figure 1 and described later.

The condition of coral health was recorded by a digital camera and judged by the marking of the color scale (Rosado et al., 2019), according to the PANTONE Color Manager software V2.2.0. The images of corals in each group were taken using a digital camera (Canon EOS 600D with 50 mm lens, Japan), using the parameter of aperture at F/3.5, the exposure time for 1/50 s, and the sensitivity at ISO 200.

The Chlorophyll Fluorescence Quantum Yield (Qy, Fv’/Fm) of coral was determined by the mean value of the measurement in light at the tip and middle of coral branches, using pulse amplitude modulated (PAM) fluorimetry (FluorPen FP100, Czech) in triplicate. The measurement was taken at a fixed time every day before light off for the first Vibrio infection and performed every 4 h during the experiment in repeated trials. Statistical differences of Qy analyses were determined using one-way ANOVA.

The coral branches, about 5 g of live stone, and 500 ml of water from each group were sampled after bacterial infection. The sampled coral branches and live stones were immediately frozen by liquid nitrogen and crushed into powder using a pre-sterilized mortar and pestle, and the water was filtered by the 0.22 μm membrane (Millipore, USA) to collect the microorganisms on the filter membrane. The treated coral, stone, and water samples containing the symbiotic microorganism were then stocked at −80°C for further analysis.

Total genomic DNA was extracted from samples using the MP FastDNA Spin kit for Soil (MP Biomedicals, USA). The DNA extract was checked on 1% agarose gel, and DNA concentration and purity were determined with NanoDrop 2000 UV-vis spectrophotometer (Thermo Scientific, USA) and used as a template DNA to amplify the hypervariable region V3–V4 of the bacterial 16S rRNA and Symbiodiniaceae ITS2 sequence with the primer pairs (listed in Table 1) by PCR using Phusion^® High-Fidelity PCR Master Mix kit (NEB, UK). The PCR amplification of 16S rRNA was performed as follows: initial denaturation at 95°C for 3 min, followed by 30 cycles of denaturing at 98°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 15 s, and single extension at 72°C for 5 min, and end at 4°C. The PCR products containing bands of 400–450 bp were operated using electrophoresis on 2% agarose gel for detection, and the samples were extracted from the 2% agarose gel and purified using the DNA Gel Extraction Kit (Omega, China) according to the manufacturer’s instructions and quantified using a NanoDrop 2000 spectrophotometer (Thermo Scientific, USA).

The purified amplicons were pooled in equimolar and paired-end sequenced on an Illumina MiSeq PE300 platform/NovaSeqPE250 platform (Illumina, San Diego, USA) according to the standard protocols by Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China).

The raw bacterial 16S rRNA and Symbiodiniaceae ITS2 sequencing reads were demultiplexed, quality filtered by fastp version 0.20.0 (Chen et al., 2018), and merged by FLASH version 1.2.7 (Magoè and Salzberg, 2011) with the following criteria: (i) the 300 bp reads were truncated at any site receiving an average quality score of <20 over a 50 bp sliding window, and the truncated reads shorter than 50 bp and reads containing ambiguous characters were discarded; (ii) only overlapping sequences longer than 10 bp were assembled according to their overlapped sequence. The maximum mismatch ratio of the overlap region is 0.2. The reads that could not be assembled were discarded; (iii) samples were distinguished according to the barcode and primers, and the sequence direction was adjusted with exact barcode matching or a maximum of two nucleotide mismatch in primer matching.

Operational taxonomic units (OTUs) with a 97% similarity cutoff were clustered using the UPARSE version 7.1 (Edgar, 2013), and the chimeric sequences were identified and removed. The taxonomy of each OTU representative sequence was analyzed by RDP Classifier version 2.2 (Wang et al., 2007) according to the SILVA ribosomal RNA database and algal symbionts Genomic resource database (Shi et al., 2021), using a confidence threshold of 0.7. Alpha diversity index, including observed OTUs, and richness estimators, such as Ace, Chao, Shannon, and Simpson indices, were calculated based on the frequency of OTUs and genera in the assigned sequence collections after rare sequences were removed. The OTUs were analyzed by Student’s t-test to compare the difference of the abundance on the genus level.

The condition of coral health is shown in Figure 2. After the inoculation of S10-1 for about 30 h, the corals in blank groups as the control were still observed in good condition, while the infected ones repeated in Groups 3 and 4 were observed with less tentacles extending for predation (Day 3). The second inoculation of Vibrio on Day 5 caused the symptoms to be more obvious on the second day (Day 6), and the images showed that Group 3 infected with Vibrio was observed with bleaching from C7 to C5, and Group 4 observed more serious bleaching from C5 to C1, accompanied with peeling of coral tissue. The damages mainly occurred starting from the middle of coral branches to the tip ends, while the coral polyps still seemed to be in good condition at the tip ends.

The results from the visual response of the coral holobiont health in each treatment were confirmed by the chlorophyll fluorescence quantum yield (Fv’/Fm) measured by PAM fluorometry to indicate the algal symbionts’ photosynthetic function. As shown in Figure 3, the Qy values from duplicated blank groups were similar, ranging between 0.50 and 0.40 in the first 6 days and remaining above 0.35 on Day 7 (P-value was 0.96857), while the Qy values from duplicated Vibrio infection groups also did not present a significant difference (P-value was 0.37669), with coral 3 being observed with a decreasing ratio of 13.17% after 6 days and further decreasing to about 0.21 on Day 7, while coral 4 present more sensitive to the Vibrio infection, decreased rapidly from healthy status at about 0.45 to an obvious bleaching status at Qy valve of 0.208 on Day 6 than 0.014 on Day 7. According to the results of statistical analysis, the Vibrio infection significantly affected the photosynthetic function of coral holobiont with a P-value of 0.00491.

Later on, the S10-1 strain was isolated from the infected S. guttatus branches. According to Koch’s rule, to verify the actual causative agent of V. fortis, subsequent infection experiments were carried out and inoculated the isolated S10-1 on both S. guttatus and P. damicornis; in addition, the QQ enzyme was used to inhibit the pathogenicity of S10-1 as control. As shown in Figure 4, the two species of coral were observed with a significant decrease in Qy value accompanied by obvious bleaching, after the inoculation of re-isolated S10-1 for 30 h. Interestingly, the corals in QQ-treated groups were observed with a reduction tendency but far better than that of infected groups, while the corals in blank groups remained in good condition. The results demonstrated that the infection of V. fortis again caused coral bleaching after bacterial inoculation, confirming the hypothesis that V. fortis was highly responsive to coral bleaching and was a potential pathogen related to coral health. In addition, the positive effects of QQ also indicated the probability of quorum sensing (QS) inhibition as a potential strategy for coral disease control.

The diversity and richness of symbiotic bacteria from S. guttatus were further analyzed by the Alpha index, as shown in Table 2. The coverage of OTU from each sample was in the range of 98.79 to 99.95%, which indicated good recovery and reliable reflection of the symbiotic bacterial community of the tested samples.

The community of coral symbiotic bacteria was observed with a remarkable difference between the healthy corals and the Vibrio-infected ones; the infection of V. fortis increased the richness as indicated by the higher indices of Ace and Chao and enriched the diversity as indicated by the higher Shannon index and lower Simpson index. In contrast, the planktonic bacteria in the water sample of healthy corals present higher richness than that of diseased corals and showed similar diversity between groups. In addition, the richness and diversity of the bacterial community in live stones were rarely affected by the additional V. fortis and showed no significant difference.

The microbial diversity in S. guttatus of each treatment group was analyzed. As shown in the community species abundance clustering in Figure 5, the upper clustering tree analyzed using the Bray–Curtis algorithm demonstrated a similarity and consistent tendency on the species abundance of the bacterial community in the duplicates, which indicated good reliability of the experiments. The results also indicated the relevance of bacterial community in the coral culture system, indicating that the Vibrio infection changed the abundance of the bacterial community in the symbiotic status of coral holobiont and planktonic status in seawater and that there was no significant difference among stones exposed under immersion of Vibrio, indicating the ability of self-regulation and stronger colonization resistance, which was probably acted by the abundant microorganisms adhered on the stones.

The community of algal symbionts was also analyzed to assess the effects of Vibrio infection as shown in Figure 9. The community of algal symbionts was observed with the most components of Cladocopium sp. in coral holobionts and very few Breviolum sp. and Fugacium sp. in live stone and seawater. No significant difference in community abundance and diversity was observed among the groups in coral holobiont, which indicated the rare effect of Vibrio on the algal symbionts, and therefore verified that the phenomenon of bleaching might be caused by the bacterial infection reasoned by Vibrio and secondary disorder of bacterial community.

The role of Vibrios in coral diseases was still limitedly known (Munn, 2015), though there are some Vibrio species had been identified as opportunistic pathogens to cause coral disease, there is disagreement about whether they should be regarded as primary causing agents, as some Vibrios appeared to contribute to nitrification (Thurber et al., 2009) or involved in defense against pathogens (Ritchie, 2006).

Unlike the better-studied species, such as V. parahaemolyticus and V. coralliformis, V. fortis is still rarely reported. To the best of our knowledge, it was reported for its pathogenicity to marine animals such as shrimps (Thompson et al., 2003), sea horses (Wang et al., 2016), and sea urchins (Ding et al., 2014), as well as one of the lists of abnormal abundant coral symbiotic microbes (Ou et al., 2018). Predictably, the activity of sea cucumber carrying V. fortis in coral reefs area would increase the risk of bacterial infection to the coral holobiont. However, the disease development and pathogenicity mechanism of coral bleaching caused by V. fortis are not yet clear.

It is currently known that the pathogenicity of Vibrio sp. is parallelly regulated by at least three QS pathways, LuxM/LuxN-related AHLs, CqsA/CqsS-related CAI-1, and LuxS/LuxP-related AI-2; all lead to the regulation of core regulatory protein LuxO (Herzog et al., 2019) and then to the synergistic action of key regulatory proteins AphA (Lu et al., 2018) and OpaR (Zhang et al., 2016), which regulate downstream exopolysaccharide synthesis genes to affect biofilm formation and virulence relative genes to release virulence factors including hemolysin. The findings of QS signals produced by bacteria in the coral mucus layer (Li et al., 2017), the signaling molecules and active hydrolytic enzymes released by V. shilonii AK1 in outer membrane vesicles (Li et al., 2016), and the interference of QS regulation against the AHL-mediated opportunistic bacteria of resident microbes in bleaching’s initiation and progression (Zhou et al., 2020), in recent years, all supported the hypothesis of bacterial infection under QS regulation, which is consistent with the positive inhibition of V. fortis pathogenicity under QQ enzyme treatment in this study (Figure 4), and this phenomenon indicates the potential strategy of QQ on coral disease control.

In this article, the infection of V. fortis was designed as the only variable factor in coral health. The symptom development of coral bleaching and tissue lysis verified and strongly supported the hypothesis of “bacterial bleaching.” The findings in this study demonstrated that its infection also caused significant changes in the coral-associated microsystem, especially the increased abundance of opportunistic pathogenic bacteria, which might synergistically affect coral health.

Among the microbial community, the decreasing abundance of Bacillus belonging to phylum Firmicutes seems to be the most serious side effect caused by Vibrio infection (Figure 5). Bacillus had been widely accepted as the main antagonistic bacteria and used for biological control in agriculture (Jiang et al., 2018) and aquaculture (Wang et al., 2019), and recently reported as marine probiotics to increase coral resistance to bleaching (Rosado et al., 2019). Bacillus can produce secondary metabolites, including lipopeptides and polyketones with antibacterial, antiviral, or antitumor activities (Wu et al., 2019), and the quorum quenching (QQ) enzymes, such as AiiA (Dong et al., 2000) and YtnP (Sun et al., 2021), to degrade AHLs, inhibiting the biofilm formation and toxin release of AHLs/Lux-mediated pathogens including V. fortis that was known to contain an AHLs intermediated LuxM/LuxN QS pathway (Ding et al., 2014). The subsequent abnormal abundant Bacillus, B. circulans for instance, resulted from the simulated outbreaking of V. fortis when the inoculation of 10⁵ CFU/ml was far beyond the threshold of QS; therefore, it might further weaken the resistance of coral holobiont to the stress from pathogens (Satpute et al., 2010; Niu et al., 2014).

The increased abundance of Rhodococcus in phylum Actinobacteria might also be related to the decreasing abundance of Bacillus, because Rhodococcus sp. was reported with antibacterial activity against Bacillus substilis (Mahmoud and Kalendar, 2016). Among these, the most composition species, R. erythropolis, produces different types of QQ enzymes (Ryu et al., 2020) to recognize and degrade AHLs and could be used for biofouling control (Ergön-Can et al., 2017). The increasing abundance might be speculated to result from the self-regulation of the coral symbiotic microbiota after the sense of abnormally increased concentration of AHLs generated by the inoculated V. fortis.

The increasing abundance of the genera Ralstonia and Burkholderia–Caballeronia–Paraburkholderia in phylum Proteobacteria was probably eligible for the causal relationship of the declined abundance of Bacillus because of the less inhibition from the Bacillus-produced surfactin (Grady et al., 2019). However, the correlation between Ralstonia and Burkholderia–Caballeronia–Paraburkholderia of coral health is still unclear (Bernasconi et al., 2019). Reports are claiming that Ralstonia sp. is a known gram-negative phytopathogenic bacteria (Hayashi et al., 2019) and dental opportunistic pathogen (Tuttlebee et al., 2002), while the Burkholderia–Caballeronia–Paraburkholderia, which is mostly constituted by Paraburkholderia fungorum, is reported as a plant probiotic bacteria (Rahman et al., 2018) but still risky to human health (Tan et al., 2020).

In addition, considering that the planktonic bacteria in the water could flow into the gastrovascular cavity of coral along with the filter feeding, the significantly increased abundance of Thalassobius (also detected with higher abundance in bleaching coral but with no significant difference) and Tenacibaculum in water might also affect the coral health (Figure 7). Thalassobius belonging to the family Rhodobacteraceae has been reported as a microbial bioindicator enriched in the Stony Coral Tissue Loss Disease accompanied by Vibrio (Becker et al., 2021), and has been associated with invertebrate diseases (Roder et al., 2014), while the increased abundance of Tenacibaculum was also consistent with the findings from microbial community shift in the White syndrome-affected Echinopora lamellosa in aquaria (Smith et al., 2015).

Another unclassified genus belonging to the family Rhodobacteraceae and the genus Shimia was also detected in the planktonic bacterial community in water, though there was no significant difference observed yet. The indicator species in the coral hosts, family Rhodobacteraceae and genus Shimia, were observed to increase their relative abundance when corals are under stress (Casey et al., 2015; Pootakham et al., 2019) or with the emergence of Porites white patch syndrome accompanied with Vibrio (Séré et al., 2013). This phenomenon was verified in the later coral bleaching in the blank group caused by the deterioration of water quality 2 days after the experiments stopped (data not shown).

The virulence from the shift bacterial community of abundant opportunistic pathogens might not be the only reason for coral bleaching. The nutrients sources and waste products for coral holobiont are also a concern, which mainly come from the metabolism of symbiotic microbes such as carbon and nitrogen fixation (Gibbin et al., 2019) and the metabolic integration from algal symbionts (Sun et al., 2020), in such closed aquaria without extra exogenous nutrient sources. The shift in the bacterial community no doubt leads to an imbalance in the nutrient food chain in the coral holobiont (Raina et al., 2009); however, in this study, there is no direct relevance between the infection of V. fortis and the status of algal symbionts. The dominant population in the tested S. guttatus was Cladocopium sp., though with few relative abundances of other Symbiodiniaceae according to the reported ITS2 type (Yu et al., 2020), and no observation of Durusdinium sp., which was reported with stronger stress resistance (Sun et al., 2020); therefore, the findings indicated that the bleaching in this study was caused by Vibrio infection and the following shift of bacterial community, but not by the dissociation of algal symbionts.

To sum up, this study expands the cognition of the correlation between coral symbiotic Vibrio and coral health. According to the experience of other reported Vibrio pathogens, the outbreak and pathogenicity enhancement of Vibrio may be related to environmental changes (Kimes et al., 2012). The symbiotic changes caused by the outbreak of V. fortis, especially the increased abundance of other pathogenic bacteria, may also cause more serious damage to the coral holobiont. In addition, the reduced abundance of potential probiotics Bacillus under the competition of microbiota might also provide an experimental experience for probiotic strategy (Zhao et al., 2019) or QS regulation (Zhou et al., 2020) on pathogen control or resistance enhancement (Rosado et al., 2019).