Accepted Manuscript

The microbiota profile and transcriptome analysis of immune response during

metamorphosis stages in orange spotted grouper (Epinephelus coioides)

Joan Tang Xiao Joe, Pinwen Peter Chiou, Chia-Yu Kuo, Jacqueline Ho Jia Lin, Jen-
Leih Wu, Ming-Wei Lu

PII: S1050-4648(19)30212-8
DOI: https://doi.org/10.1016/j.fsi.2019.03.063
Reference: YFSIM 6031

To appear in: Fish and Shellfish Immunology

Received Date: 4 January 2019

Revised Date: 26 March 2019
Accepted Date: 26 March 2019

Please cite this article as: Xiao Joe JT, Chiou PP, Kuo C-Y, Jia Lin JH, Wu J-L, Lu M-W, The microbiota
profile and transcriptome analysis of immune response during metamorphosis stages in orange spotted
grouper (Epinephelus coioides), Fish and Shellfish Immunology (2019), doi: https://doi.org/10.1016/
j.fsi.2019.03.063.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
1 TITLE:
2 The microbiota profile and transcriptome analysis of immune response during
3 metamorphosis stages in orange spotted grouper (Epinephelus coioides)

5 Joan Tang Xiao Joea,b, Pinwen Peter Chiouc, Chia-Yu Kuoa,b, Jacqueline Ho Jia Linc ,
6 Jen-Leih Wud, Ming-Wei Luc,e*

PT
7
8 a. Doctoral Degree Program in Marine Biotechnology, The College of Life Sciences,

RI
9 National Taiwan Ocean University, Keelung, Taiwan
10 b. Doctoral Degree Program in Marine Biotechnology, Academia Sinica, Taipei,
11 Taiwan

SC
12 c. Department of Aquaculture, National Taiwan Ocean University, Keelung, Taiwan
13 d. Laboratory of Marine Molecular Biology and Biotechnology, Institute of Cellular

U
14 and Organismic Biology,Academia Sinica, Nankang, Taipei, Taiwan
15 e. Center of Excellence for the Oceans, National Taiwan Ocean University,
AN
16 Keelung,Taiwan
17
18
M

19
20
D

21
22
TE

23
24
EP

25
26 CORRESPONDENCE:
27 Dr. Ming-Wei Lu
C

28 mingwei@mail.ntou.edu.tw
AC

29
 ACCEPTED MANUSCRIPT
30 The microbiota profile and transcriptome analysis of metabolism and immune
31 response during metamorphosis stages in orange spotted grouper (Epinephelus
32 coioides)
33
34 Abstract
35 Metamorphosis is a transformation process in larval development associated with
36 changes in morphological and physiological features, including the immune system.

PT
37 The gastrointestinal tract harbors a plethora of bacteria, which might affect the
38 digestion and absorption of nutrients, immunity, and gut-brain crosstalk in the host.

RI
39 In this study, we have performed metagenomic and transcriptomic analyses on the
40 intestines of grouper at the pre-, mid- and post-metamorphosis stages. The sequencing
41 data of 16S rRNA gene showed drastic changes in the microbial communities at

SC
42 different developmental stages. The transcriptomic data revealed that the leukocyte
43 transendothelial migration and the phagosome pathways might play important roles in

U
44 mediating immunity in grouper at the three developmental stages. This information
45 will increase our understanding of the metamorphosis process in grouper larvae, and
AN
46 shed light on the development of antimicrobial strategy during larval development.
47
48 Keywords: Epinephelus coioides, metamorphosis, 16s rRNA metagenomics ，
M

49 intestinal microbiota, immune system, transcriptome analysis

50
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
51 1. Introduction
52 The orange-spotted grouper, Epinephelus coioides, is a commercially important
53 fish in many Asian countries [1]. In grouper larval development, metamorphosis is an
54 essential and critical stage involved in the morphological and physiological changes
55 [2]. During metamorphosis, the digestive system goes through significant changes,
56 including the development of a functional stomach and increase in digestive capacity
57 [3]. The metamorphosis process begins with the appearance of prolonged second

PT
58 dorsal section at 12 days post hatching (dph), and is complete by 50 dph. From 13 to
59 18 dph, the second dorsal section and pelvic fin spines become clearly visible and

RI
60 reach to their maximal lengths (Fig. 1). During the early developmental stage, the
61 adaptive immunity is not yet fully established and functional. Hence, the innate
62 immunity plays an important role in protecting the host against pathogens at the early

SC
63 developmental stage [4].
64 The intestinal mucosal immune system encounters more antigens than any

U
65 other parts of the body do. The gastrointestinal tract is richly colonized by bacteria,
66 archaea and eukaryotes, all together known as the gut microbiota [5]. The quantity of
AN
67 microorganisms living in the gut may reach over 95% of the total population in the
68 whole body [6, 7]. The gastrointestinal microbiota is known to produce beneficial
69 components related to, for example, enhancing the immune system, protecting against
M

70 pathogen, nutrition, and modulating host metabolism [8, 9]. The gut microbiota
71 produce short-chain fatty acids (SCFAs), such as propionate and butyrate, by
D

72 fermentation of the non-digestible substrates like dietary fibers and endogenous

73 intestinal mucus [10]. SCFAs can induce apoptosis of cancer cells, and regulate the
TE

74 intestinal gluconeogenesis; they also play an important role to maintain colon healthy
75 [11]. Numerous evidences have demonstrated a network called “gut-brain axis”, in
EP

76 which bidirectional interactions between the gut microbiota and the central nervous
77 system exist. Not only may the brain influence the function of the gut microbiota [12,
78 13], but also alterations of microbiota composition may be associated with several
C

79 disorders of the nervous system, including neuropsychiatric, neurodegenerative, and

AC

80 neuroinflammatory disorders [14]. As a result, gut microbiota ultimately have a

81 significant impact on the immunity, metabolism, diseases, brain and behaviors [15,
82 16].
83 In recent years, metagenomics has become the major technique to study the
84 intestinal microbiota in host [17]. To investigate the changes in intestinal microbiota
85 and the development of immune system associated with metamorphosis in E. coioides,
86 we have performed metagenomic and transcriptomic analyses on the intestines of fish
87 at the pre-, mid and post-metamorphosis stages. The metagenomics data showed
88 drastic changes in the microbial communities at different developmental stages. The
 ACCEPTED MANUSCRIPT
89 transcriptomic data revealed pathways and molecules that might play important roles
90 in mediating immunity in grouper at the three developmental stages. The new
91 information will increase our understanding of the development of grouper larvae into
92 juveniles, and further assist in the development of antimicrobial strategy in larvae
93 going through the metamorphosis process.

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
94 2. Materials and Methods
95 Experimental animals
96 In this study, fertilized eggs of orange-spotted grouper were collected from
97 hatchery at Pingtung, Taiwan. The animal experiment was conducted at National
98 Taiwan Ocean University (Keelung, Taiwan) by following the institutional IACUC
99 guideline (Approve number: 106009). The tanks were stocked with Chlorella spp.
100 prior to the installment of fertilized eggs. One gram of eggs was placed into a 100L

PT
101 tank with slightly aeration. The temperature was around room temperature with little
102 photoperiod. Before 3dph, the first live feed provided to the larvae was ss-type live

RI
103 rotifers (Brachionus plicatilis), which were fed with Chlorella spp. From 3dph to
104 25dph, the larvae were fed with s-type rotifer; the L-type rotifers were fed to 25dph
105 throughout 35dph juveniles. Subsequently, the juveniles were fed with copepod

SC
106 nauplii from 35dph to 50dph.
107

U
108 DNA extraction and 16S rRNA gene sequencing
109 The samples (12dph: n=5group (150 fish pool together), 18dph: n=4group (120
AN
110 fish pool together), 50dph: n=5group(4 fish pool together)) were collected and
111 washed with sterile phosphate-buffered saline. The genomic DNA was extracted by
112 using Genomic DNA Mini Kit, Geneaid. DNA concentration and purity were
M

113 monitored on 1% agarose gel, and the concentration of DNA was diluted to 1ng/uL
114 using sterile water. Distinct regions in 16S rRNA gene (16S V3-V4) were amplified
D

115 using specific primers (16S V4: 515F-806R) with barcode. All PCR reactions were
116 carried out with Phusion® High-Fidelity PCR Master Mix (New England Biolabs).
TE

117 PCR products were used to undergo quantification and qualification, mixed with
118 equal volume of 1X loading buffer (contained SYBR green), and analyzed by
EP

119 electrophoresis on 2% agarose gel. Samples with bright main strip between
120 400-450bp were chosen for further experiments. The PCR products were mixed at
121 equidensity ratios, and were purified with Qiagen Gel Extraction Kit (Qiagen,
C

122 Germany). Sequencing libraries were generated using NEBNext Ultra DNA Library
AC

123 Pre ® Kit for Illumina by following manufacturer's recommendations, and the index
124 codes were added. The library quality was assessed on the Qubit@ 2.0 Fluorometer
125 (Thermo Scientific) and Agilent Bioanalyzer 2100 system (Agilent Technologies, Inc).
126 At last, the library was sequenced on an Illumina platform and 250 bp paired-end
127 reads were generated.
128 Data split paired-end reads was assigned to samples based on their unique
129 barcode and truncated by cutting off the barcode and primer sequence. Sequence
130 assembly paired-end reads were processed using FLASH (V1.2.7,
131 http://ccb.jhu.edu/software/FLASH/) (Caporaso et al., 2010) to merge paired-end
 ACCEPTED MANUSCRIPT
132 reads when at least some of the reads overlap the read generated from the opposite
133 end of the same DNA fragment. The splicing sequences were afterwards called raw
134 tags. Data Filtration Quality filtering on the raw tags were performed under specific
135 filtering conditions to obtain the high-quality clean tags [18] according to the QIIME
136 (V1.7.0, http://qiime.org/index.html) [19] quality-controlled process. Chimera
137 removal the tags were compared with the reference database (Gold database,
138 http://drive5.com/uchime/uchime_download.html) using UCHIME algorithm

PT
139 (UCHIME Algorithm, http://www.drive5.com/usearch/manual/uchime_algo.html) [20]
140 to detect chimera sequences, and then the chimera sequences were removed [21] to

RI
141 obtain Effective Tags.
142 OTU Production Sequences analysis were performed by Uparse software
143 (Uparse v7.0.1001, http://drive5.com/uparse/) [22]. Sequences with ≥97% similarity

SC
144 were assigned to the same OTUs. Representative sequence for each OTU was
145 screened for further annotation. Species annotation for each representative sequence,

U
146 the GreenGene Database (http://greengenes.lbl.gov/cgi-bin/nph-index.cgi) [23] was
147 used based on RDP 3 classifier (Version 2.2,
AN
148 http://sourceforge.net/projects/rdp-classifier/) [24] algorithm to annotate taxonomic
149 information. Data Normalization OTUs abundance information was normalized using
150 a standard of sequence number corresponding to the sample with the least sequences.
M

151 Subsequent analysis of alpha diversity and beta diversity were all performed based on
152 this output normalized data. The KEGG pathway used an OTU table that had already
D

153 been generated for use with PICRUSt. PICRUSt currently can only use an OTU table
154 with GreenGene OTU identifiers which is the output from closed-reference picking or
TE

155 by filtering out de-novo OTUs after open-reference picking. Statistical significance
156 was accepted at p <0.05, and high significance was accepted at p <0.01.
EP

157 Alpha diversity is applied in analyzing complexity of species diversity for a

158 sample through indices, including Observed-species, Chao1, Shannon, Simpson, ACE,
159 and Good-coverage. All these indices in our samples were calculated with QIIME
C

160 (Version 1.7.0) and displayed with R software (Version 2.15.3). Two indices were
AC

161 selected to identify the community richness. First is the Chao1 estimator (Chao)
162 (http://www.mothur.org/wiki/Chao); second, Abundance-based Coverage Estimator
163 (ACE) (http://www.mothur.org/wiki/Ace). The ACE incorporated data from all
164 species with fewer than 10 individuals, rather than just singletons and doubletons
165 (Anne Chao, 1992). Two indices were used to identify the community diversity. First,
166 Shannon - the Shannon index (http://www.mothur.org/wiki/Shannon); second is
167 Simpson - the Simpson index (http://www.mothur.org/wiki/Simpson). The Shannon
168 index is an information statistic index, which means it assumes all species are
169 represented in a sample and that they are randomly sampled. The Simpson index is a
 ACCEPTED MANUSCRIPT
170 dominance index for it gives more weight to common or dominant species. In this
171 case, a few rare species with only a few representatives will not affect the diversity.
172 Cluster analysis was preceded by principal component analysis (PCoA), which was
173 applied to reduce the dimension of the original variables using the FactoMineR
174 package and ggplot2 package in R software (Version 2.15.3).
175
176 RNA isolation, transcriptome assembly and pathway review

PT
177 Total RNA from the grouper (12dph:n=150, 18:n=120, 50dph:n=4) was
178 extracted by using TRIzol (life technologies), according to the manufacturer’s

RI
179 protocol. RNase-free DNase I (GMbiolab, Taiwan) was used to remove the residual
180 genomic DNA. Then, RNA concentration and quality were verified by using Agilent
181 2100 Bioanalyzer. The extracted RNA was stored at -80°C until processed for library

SC
182 preparation. After RNA extraction, RNA was quantified to 150ng/ul and total RNA
183 was fixed to be 20ug. Later, transcriptome sequencing was carried out by the
company Genomics using Illumina HiseqTM 2000. The data formed was proceeded

U
184
185 using software CLC Genomics Worbench while reads were assembled and the
AN
186 expressions of genes were analyzed. Log2 fold change and FDR p-value were
187 obtained from the result.
188 Using FDR p-value ≤ 0.001 and Log2 fold change abs value > 2, data was
M

189 filtered and output as a csv file. R language (R v3.3.0) was used in order to form the
190 data analysis graph while gene annotation was done using ncbi-blast + v2.3.0.
D

191 Bioconductor v3.3 worked as a tool to find out the gene function and pathway related
192 through Gene Ontology (GO) and KEGG database.
TE

193
194 RNA isolation and gene expression
EP

195 After RNA extraction, 1ug of total RNA was used for cDNA synthesis by using
196 HiScript I Reverse Transcriptase (BIONOVAS, Canada). Reverse transcription was
197 conducted, according to the manufacturer’s protocol with an Oligo (dT)18 primer. The
C

198 synthesis condition of cDNA was set at: 65°C for 5 min, 42°C for 60 min and 70°C
AC

199 for 15 min. Quantitative real-time PCR (qPCR) was performed using the Applied
200 BiosystemTM 7500 Real-Time PCR System (Applied Biosystems, USA) on a
201 TOptical Thermocycler ® (Analytik Jena AG, Germany). The qPCR reaction contains
202 1µl of the cDNA template, 10µl of the 2X qPCRBIO syGreen Master Mix, 0.8µl each
203 of the forward and reverse primer (10pmol/uL) shown in Table.3. The amplification
204 condition was initial denaturation at 95℃ for 5 mins, followed by 40 cycles of 95℃
205 for 5sec, 65℃ for 30sec. The melting curve and cooling were performed at the last
206 step of qPCR. The primers used in this study were listed in Table 1. The expression
207 levels of the target gene were normalized to beta-actin, a housekeeping gene. Fold
 ACCEPTED MANUSCRIPT
208 change in the relative gene expression with control group was determined by the
standard 2- Ct method. The changes were analyzed by unpaired sample t-test.
△△
209
210 Statistical significance was accepted at p <0.05, and high significance was accepted at
211 p <0.01. All data were expressed as mean± standard deviation (mean±SD).
212

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
213 3. Results
214 The intestinal bacterial diversity and richness in grouper at different
215 metamorphosis stages
216 The species accumulation boxplot was used to evaluate the adequacy of sample
217 size and the species richness. As shown in Fig. 2, with the increase in sample size the
218 boxplot showed first a sharp rise that later approached to a plateau at the sample size
219 of 28, indicating adequate sample numbers in the metagenomic analysis. The alpha

PT
220 diversity indices (Chao1, ACE, Shannon and Simpson) were further analyzed to
221 estimate the microbiome richness and diversity in the intestine flora of grouper at the

RI
222 three different metamorphosis stages (Table.1). The varieties in the intestinal bacterial
223 communities at 50dph were significantly decreased as compared to that at 12dph and
224 18dph. The microbiomes diversity and richness at 18dph were higher than those at

SC
225 12dph and 50dph.
226

U
227 Comparison of the intestinal bacterial communities among grouper at different
228 metamorphosis stages
AN
229 The principal component analysis (PCA) was conducted to observe the
230 composition of intestinal flora communities. PCA is a statistical procedure to extract
231 principle components and structures in data by using orthogonal transformation and
M

232 reducing dimensionalities of data[25]. The more similar the compositions among the
233 sample sets are, the closer the distance of their corresponding data points on the PCA
D

234 graph are. Fig.2 shows that the individuals collected at 12, 18 and 50dph were
235 clustered in three distinct areas, indicating significant variation in the intestinal
TE

236 bacterial composition among the three groups (Fig. 4A). The 16s rRNA gene
237 sequencing data further revealed that at the phylum level Proteobacteria is the most
EP

238 abundant species in the midgut, with an abundancy of 86%, 70% and 96% in the fish
239 at 12dph, 18dph and 50dph, respectively. Firmicutes was the second most abundant
240 microbe at 12dph (4%) and 18dph (6%). Actinobacteria, Bacteroidetes and
C

241 Chloroflexi were also abundant in the microbiota at 18dph. The rest of the
AC

242 microbiome at 50dph consisted of minor amount of Firmicutes, Acidobacteria,

243 Actinobacteria and Bacteroidetes. We further analyzed the relative abundance of
244 microbes in the intestine at the genus level. The top 35 dominant genera among all
245 samples was displayed in a species abundance heat map (Fig. 4B). Red color indicates
246 high abundance genera and blue color low abundance. At 12dph, the Exiguabacterium
247 and Acinetobacter were two most dominant bacteria. There were 7 abundant genera at
248 18dph, that is, Propionibacterium, Marivita, unidentified-Holosporacaea,
249 Idiomarinas, Halomonas, Caldisericum, and Nitrosomonas. At 50 dph, Desulfovibrio
250 and Photobacterium were the two most dominant genera. The comparison of
 ACCEPTED MANUSCRIPT
251 intestinal bacterial communities among grouper at different developmental stages was
252 further conducted by the Venn diagram. As shown in Fig. 5, the OTU number at 12,
253 18 and 50dph was 682, 1285 and 388, respectively. The number of OTUs shared by
254 all three groups was 701. In addition, the number of unique OTUs between 12dph and
255 18dph, 18dph and 50dph, 12dph and 50dph, was 929, 339 and 96, respectively.
256
257 Biological pathways affected in the intestines of grouper at different

PT
258 metamorphosis stages
259 The NGS data was further subjected to analysis against KEGG to reveal the

RI
260 biological pathways that were affected in the intestines of grouper at different
261 metamorphosis stages. The results showed that pathways relevant to leukocyte,
262 lysosome and phagosome were the most affected pathways between grouper at 12dph

SC
263 and 50dph (Fig.6A). Further analysis of the transcriptomes showed that the leukocyte
264 transendothelial migration pathway and the phagosome pathway were the two most

U
265 significantly activated pathways. In the leukocyte transendothelial migration pathway,
266 four genes (Rap1, β-catenin, JAM2 and Actin) were significantly up-regulated while
AN
267 two genes (p120ctn and CAMs) down-regulated (Fig.6B). In the phagosome pathway,
268 significant up-regulation of F-actin, ATPase, TUBA and NOS was observed (Fig.6C).
269
M

270 The gene expression of immune system at the metamorphosis stages

271 Quantitative real-time RT-PCR was conducted to assay the expression of
D

272 representative immune genes in whole larvae at 12 and 18dph, and in the spleen, brain
273 and head kidney in juveniles at 50dph. The results showed the expression level of
TE

274 NF-κβ was slightly higher than that of other immune genes (TLR-3, IL-1β, Type I
275 IFN, Type II IFN and Mx gene) at 12dph, 18dph (Fig. 7A) and 50dph (Fig. 7B).
EP

276 These results indicate the fish were cultured in a healthy environment and did not
277 contract infections.
278
C
AC
 ACCEPTED MANUSCRIPT
279 4. Discussion
280 Assessment of the intestinal bacterial composition of healthy E. coioides can
281 help to establish intestinal microbiota database and to identify potential gut probiotics.
282 However, no such information is currently available. Here, we have performed assays
283 on the intestinal microbiota and immune system associated with E. coioides at the pre-,
284 mid-, and post-metamorphosis stages.
285 When the early development stage of grouper, the adaptive immunity is still not

PT
286 fully established and functioning; therefore the innate immunity has a crucial role in
287 protect or fights to get rid of the pathogens, such as viruses, bacteria, and parasites [4,

RI
288 26, 27]. The key cells of the innate immune system addressed here are NF-κB, TLR-3,
289 IL-1β, Type I IFN, Type II IFN and Mx gene. NF-κB is an important transcription
290 factor that responds to various stimuli. NFκB is typically inactive and remains

SC
291 sequestered in the cytoplasm under nonstress-related conditions [28]. When NF-κB is
292 activated, it will regulate the suppression of cellular apoptosis. On the other hands,

U
293 toll-like receptors (TLRs) family are a key pathogen recognition receptors (PRRs),
294 starting up the antivirus signaling pathway [29]. Virus-infected cell secretes interferon,
AN
295 which induce nearby uninfected cells to produce substances that inhibit viral
296 reproduction and transported by the blood stream to alarm other cells in the body [30].
297 IFNs are a family of structurally related cytokines with a hallmark function of
M

298 antiviral activity [31]. Interleukin-1β (IL-1β) is an endogenous pyrogen produced and
299 released at the early stage response following infections, and subsequently considered
D

300 as initiator of the pro-inflammatory response in macrophages, activator of

301 lymphocytes and a synthesis promoter of other cytokines and prostaglandins [32, 33].
TE

302 In this study, the alpha diversity analysis indicates the intestinal microbiota
303 communities in grouper larvae (12 and 18 dph) is richer than those in the juveniles
EP

304 (50dph). Similar result has been observed in Arapaima gigas and Gadus
305 morhuathat[34, 35]. Interestingly, the intestinal microbiota in blunt snout bream
306 (Megalobrama amblycephala) is altered during feeding habit transition [36]. All
C

307 together, these results show that the richness and diversity of intestinal bacterial
AC

308 communities are subjected to the influence of development, feeding habit, and
309 environments. Importantly, we have found profound differences in the intestinal
310 microbial compositions among the grouper at pre-, mid-, and post-metamorphosis
311 stages. The observation raises an interesting question about the impact of the different
312 microbial compositions on the fish going through this critical transformation stage.
313 In this study, we have shown significant variations in the abundance and
314 diversity of intestinal microbiota in grouper at different developmental stages. The
315 most prevalent phylum is the gram-negative Proteobacteria, which often are the most
316 dominant phylum in the intestine of numerous marine fish, including Plectropomus
 ACCEPTED MANUSCRIPT
317 leopardus, Seriola lalandi Ctenopharyngodon idellus, Megalobrama amblycephala,
318 Carassius auratus, and Hypophthalmichthys nobilis [37-39]. In Oncorhynchus mykiss,
319 Proteobacteria can make up over 70% of the total population[40, 41].. Next to
320 Proteobacteria was the phylum of Firmicutes, which took up 4% and 6% of the total
321 population in the larvae at 12dph and 18dph, respectively. Similar observation has
322 been reported previously in farm and aquarium rainbow trout. Both Firmicutes and
323 Proteobacteria are proficient in fermentation of carbohydrates and proteins, and thus

PT
324 promote nutrients uptake by host from the diets[42].
325 A number of intestinal bacteria identified in the grouper guts may be worth

RI
326 further investigation for their beneficial effects on the fish. At genus level,
327 Exiguobacterium was most abundant in the intestine at 12dph. Dominance of
328 Exiguobacterium in fish at early life stage has also been reported in African turquoise

SC
329 killifish (6 weeks). Exiguobacterium has been reported to be anti-inflammatory
330 mediator [43], and has been applied by the industry on bioremediation and

U
331 degradation of toxic substances released into the environments[44]. Another dominant
332 species, Acinetobacter, is known for therapeutic functions, including antibacterium,
AN
333 antivirus, anticancer, insecticide, and enzyme inhibition [45, 46]. Hence, the intestinal
334 Acinetobacter might play a major role in protecting hosts from microbial infection at
335 their early life stage. The gut content of 18dph-grouper was dominated by seven
M

336 major genera, that is, Propionibacterium, Marivita, unidentified-Holosporacaea,

337 Idiomarinas, Halomonas, Caldisericum and Nitrosomonas. Among these species,
D

338 Caldisericum is known to be important in the oxidation of reduced sulfur species

339 because the growth of Caldisericum occurs with anaerobic respiration using sulfur
TE

340 compounds [47]. In aquaculture, Nitrosomonas is well-known as ammonia oxidizer

341 that converts ammonia (NH3) into nitrite (NO2−), and can be applied to reduce the
EP

342 toxicity in water by turning NH3 to NO2−[48]. It will be interesting to further

343 explore these bacteria mentioned above for potential applications on improvements on
344 water quality or host health.
C

345 The KEGG pathway analysis has revealed several pathways and molecules that
AC

346 might play important roles in mediating immunity in grouper at pre-, mid-, and
347 post-metamorphosis stage. In particular, the expression of several molecules in the
348 leukocyte transendothelial migration pathway was significantly regulated in the
349 grouper at different stages (Fig. 6B). CDH5 is a classical cadherin important to the
350 formation of cell junctions [49]. Meanwhile, p120ctn, which was down-regulated, is a
351 component in cadherin-catenin complex that helps to stabilize cadherin on cell
352 membrane and regulates cytoskeletal re-organization leading to junction maturation
353 [50, 51]. p120ctn and CDH5 forms a complex which could affect the expression of
354 β-catenin. In Figure 6B, β-catenin was found to be upregulated and so was
 ACCEPTED MANUSCRIPT
355 actin. β-catenin regulates cell growth and helps to form adherens junction, while
356 actin builds up filament and cytoskeleton. β-catenin has been shown to affect actin
357 through α-caterin, which binds directly to the actin filaments [52-54]. A small GTPase,
358 Rap1, was also up-regulated. Rap1 can be activated by cytokines and
359 neurotransmitters, and is involved in the regulation of cell-cell contact and the
360 formation of adherens junctions especially in the contact between T cells and
361 antigen-presenting cells (APC) [54-56]. JAM2, which was also up-regulated, is a

PT
362 vascular molecule in interendothelial junctional complexes and can also mediate
363 cell-cell contact especially among immune cells [57, 58]. In addition, JAM2 is

RI
364 involved in the process of lymphocyte homing to the secondary lymphoid organs and
365 the regulation of transendothelial migration [58]. These data indicate that leukocyte
366 transendothelial migration might be critical to the development of immune system in

SC
367 grouper going through the metamorphosis process. Further investigation will shed
368 light on our understanding of the role of leukocyte transendothelial migration in the

U
369 development of immune system in grouper.
370 Several genes in the phagosome pathway were also evidently regulated (Fig. 6C).
AN
371 F-actin, which was up-regulated, forms actin filaments to build cytoskeleton that
372 helps in maintaining cell junction and cell shape by remodeling cell cytoskeletal
373 structures to form phagocytic cup [59]. TUBA, which was also upregulated, binds
M

374 microtubules and builds cytoskeletal to provide structural support for cell during the
375 remodeling of cell cytoskeletal. TUBA was found to be linked to actin through
D

376 specific independent mechanism and may regulate actin in endocytosis [60].The
377 upregulation of vATPase on phagosome membrane might induce hydrogen ion to
TE

378 flow into the endosome, leading to acidosis in phagosome, a condition allowing
379 lysosome to decompose. This result was similar to that of 16s rRNA metagenomic
EP

380 analysis (Fig.6A). Future study is required to further illustrate the importance of
381 phagosome pathway to the immunity in grouper throughout the metamorphosis
382 process.
C

383 In summary, we have performed metagenomic and transcriptomic analyses to

AC

384 investigate the microbiota and immune system development in the intestines of
385 grouper at three developmental stages: pre, mid-, and post-metamorphosis. The
386 obtained information will not only increase our knowledge of the development of
387 grouper, but also assist in developing strategy to enhance health in fish going through
388 the critical metamorphosis stage. For example, manipulating the gut microbiomes to
389 resemble a community found in the healthy grouper might enhance the immunity
390 against infections.
391 5. Acknowledgements
392
 ACCEPTED MANUSCRIPT
393 This study was financially supported by Ministry of Science and Technology (grant
394 number:MOST-107-2321-B-019-002). We would like to acknowledge Biotools, Co.,
395 Ltd for assistance with analyzing the 16s rRNA metagenomics.

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
396 References
397 [1] L.-S. Lim, Y. Mukai, Morphogenesis of sense organs and behavioural changes in
398 larvae of the brown-marbled grouperEpinephelus fuscoguttatus(Forsskål), Marine
399 and Freshwater Behaviour and Physiology 47(5) (2014) 313-327.
400 [2] Y.Y. Chen, A.C. Cheng, S.A. Cheng, J.C. Chen, Orange-spotted grouper Epinephelus
401 coioides that have encountered low salinity stress have decreased cellular and
402 humoral immune reactions and increased susceptibility to Vibrio alginolyticus, Fish

PT
403 Shellfish Immunol 80 (2018) 392-396.
404 [3] A.T. Gonçalves, C. Gallardo-Escárate, Microbiome dynamic modulation through

RI
405 functional diets based on pre- and probiotics (mannan-oligosaccharides
406 andSaccharomyces cerevisiae) in juvenile rainbow trout (Oncorhynchus mykiss),
407 Journal of Applied Microbiology, 2017, pp. 1333-1347.

SC
408 [4] A.E. Ellis, Innate host defense mechanisms of fish against viruses and bacteria,
409 Dev Comp Immunol 25(8-9) (2001) 827-39.

U
410 [5] E. Thursby, N. Juge, Introduction to the human gut microbiota, Biochemical
411 Journal 474(11) (2017) 1823-1836.
AN
412 [6] M. Rosenthal, D. Goldberg, A. Aiello, E. Larson, B. Foxman, Skin microbiota:
413 microbial community structure and its potential association with health and disease,
414 Infection, genetics and evolution : journal of molecular epidemiology and
M

415 evolutionary genetics in infectious diseases, 2011, pp. 839-48.

416 [7] S. Virtanen, I. Kalliala, P. Nieminen, A. Salonen, Comparative analysis of vaginal
D

417 microbiota sampling using 16S rRNA gene analysis, PLoS One, 2017, p. e0181477.
418 [8] A.M. O'Hara, F. Shanahan, The gut flora as a forgotten organ, EMBO reports 7(7)
TE

419 (2006) 688-693.

420 [9] I. Sekirov, S.L. Russell, L.C. Antunes, B.B. Finlay, Gut microbiota in health and
EP

421 disease, Physiological reviews 90(3) (2010) 859-904.

422 [10] J.M. Wong, R. de Souza, C.W. Kendall, A. Emam, D.J. Jenkins, Colonic health:
423 fermentation and short chain fatty acids, J Clin Gastroenterol 40(3) (2006) 235-43.
C

424 [11] F. De Vadder, P. Kovatcheva-Datchary, D. Goncalves, J. Vinera, C. Zitoun, A.

AC

425 Duchampt, F. Backhed, G. Mithieux, Microbiota-generated metabolites promote

426 metabolic benefits via gut-brain neural circuits, Cell 156(1-2) (2014) 84-96.
427 [12] M. Carabotti, A. Scirocco, M.A. Maselli, C. Severi, The gut-brain axis: interactions
428 between enteric microbiota, central and enteric nervous systems, Annals of
429 Gastroenterology : Quarterly Publication of the Hellenic Society of Gastroenterology
430 28(2) (2015) 203-209.
431 [13] J.A. Foster, K.A. McVey Neufeld, Gut-brain axis: how the microbiome influences
432 anxiety and depression, Trends in neurosciences 36(5) (2013) 305-12.
433 [14] A.K. Fleck, D. Schuppan, H. Wiendl, L. Klotz, Gut-CNS-Axis as Possibility to
 ACCEPTED MANUSCRIPT
434 Modulate Inflammatory Disease Activity-Implications for Multiple Sclerosis,
435 International journal of molecular sciences 18(7) (2017).
436 [15] R. Cianci, D. Pagliari, C.A. Piccirillo, J.H. Fritz, G. Gambassi, The Microbiota and
437 Immune System Crosstalk in Health and Disease, Mediators of Inflammation 2018
438 (2018) 1-3.
439 [16] S. Rautava, R. Luoto, S. Salminen, E. Isolauri, Microbial contact during pregnancy,
440 intestinal colonization and human disease, Nature reviews. Gastroenterology &

PT
441 hepatology 9(10) (2012) 565-76.
442 [17] S. Mazumdar-Leighton, V.K. Choudhary, Metagenomics at grass roots, Resonance

RI
443 22(3) (2017) 291-301.
444 [18] N. Youssef, C.S. Sheik, L.R. Krumholz, F.Z. Najar, B.A. Roe, M.S. Elshahed,
445 Comparison of Species Richness Estimates Obtained Using Nearly Complete

SC
446 Fragments and Simulated Pyrosequencing-Generated Fragments in 16S rRNA
447 Gene-Based Environmental Surveys, Applied and Environmental Microbiology 75(16)

U
448 (2009) 5227-5236.
449 [19] M. Hess, A. Sczyrba, R. Egan, T.W. Kim, H. Chokhawala, G. Schroth, S. Luo, D.S.
AN
450 Clark, F. Chen, T. Zhang, R.I. Mackie, L.A. Pennacchio, S.G. Tringe, A. Visel, T. Woyke, Z.
451 Wang, E.M. Rubin, Metagenomic Discovery of Biomass-Degrading Genes and
452 Genomes from Cow Rumen, Science 331(6016) (2011) 463-467.
M

453 [20] F. Asnicar, G. Weingart, T.L. Tickle, C. Huttenhower, N. Segata, Compact graphical
454 representation of phylogenetic data and metadata with GraPhlAn, PeerJ 3 (2015).
D

455 [21] T.Z. DeSantis, P. Hugenholtz, K. Keller, E.L. Brodie, N. Larsen, Y.M. Piceno, R. Phan,
456 G.L. Andersen, NAST: a multiple sequence alignment server for comparative analysis
TE

457 of 16S rRNA genes, Nucleic Acids Research 34(Web Server) (2006) W394-W399.
458 [22] B.D. Ondov, N.H. Bergman, A.M. Phillippy, Interactive metagenomic visualization
EP

459 in a Web browser, BMC Bioinformatics 12(1) (2011).

460 [23] D. Bulgarelli, R. Garrido-Oter, Philipp C. Münch, A. Weiman, J. Dröge, Y. Pan,
461 Alice C. McHardy, P. Schulze-Lefert, Structure and Function of the Bacterial Root
C

462 Microbiota in Wild and Domesticated Barley, Cell Host & Microbe 17(3) (2015)
AC

463 392-403.
464 [24] B. Li, X. Zhang, F. Guo, W. Wu, T. Zhang, Characterization of tetracycline resistant
465 bacterial community in saline activated sludge using batch stress incubation with
466 high-throughput sequencing analysis, Water Research 47(13) (2013) 4207-4216.
467 [25] E. Avershina, T. Frisli, K. Rudi, De novo Semi-alignment of 16S rRNA Gene
468 Sequences for Deep Phylogenetic Characterization of Next Generation Sequencing
469 Data, Microbes and Environments 28(2) (2013) 211-216.
470 [26] P.K. Sahoo, P.K. Meher, K.D. Mahapatra, J.N. Saha, R.K. Jana, P.V.G.K. Reddy,
471 Immune responses in different fullsib families of Indian major carp, Labeo rohita,
 ACCEPTED MANUSCRIPT
472 exhibiting differential resistance to Aeromonas hydrophila infection, Aquaculture
473 238(1-4) (2004) 115-125.
474 [27] Y.-C. Yeh, Y.-J. Hsu, Y.-M. Chen, H.-Y. Lin, H.-L. Yang, T.-Y. Chen, H.-C. Wang, EcVig,
475 a novel grouper immune-gene associated with antiviral activity against NNV infection,
476 Developmental & Comparative Immunology 43(1) (2014) 68-75.
477 [28] K.-M. Debatin, Apoptosis pathways in cancer and cancer therapy, Cancer
478 Immunology, Immunotherapy 53(3) (2004) 153-159.

PT
479 [29] R. Huo, X. Zhao, J. Han, T. Xu, Genomic organization, evolution and functional
480 characterization of soluble toll-like receptor 5 (TLR5S) in miiuy croaker ( Miichthys

RI
481 miiuy ), Fish Shellfish Immun 80 (2018) 109-114.
482 [30] R.E. Randall, S. Goodbourn, Interferons and viruses: an interplay between
483 induction, signalling, antiviral responses and virus countermeasures, Journal of

SC
484 General Virology 89(1) (2008) 1-47.
485 [31] M.-W.W. Lu, Jen-Leih, Innate Immune Response of NNV Infection in Fish and Its

U
486 Disease Prevention, Journal of Marine Bioscience and Biotechnology Volume 2( Issue
487 3 ) ( 2007) pp.127-132.
AN
488 [32] P. Alvarez-Pellitero, Fish immunity and parasite infections: from innate immunity
489 to immunoprophylactic prospects, Veterinary Immunology and Immunopathology
490 126(3-4) (2008) 171-198.
M

491 [33] C.A. Dinarello, Immunological and Inflammatory Functions of the Interleukin-1
492 Family, Annual Review of Immunology 27(1) (2009) 519-550.
D

493 [34] G. do Vale Pereira, D.G. da Cunha, J.L. Pedreira Mourino, A. Rodiles, A.
494 Jaramillo-Torres, D.L. Merrifield, Characterization of microbiota in Arapaima gigas
TE

495 intestine and isolation of potential probiotic bacteria, Journal of Applied

496 Microbiology 123(5) (2017) 1298-1311.
EP

497 [35] T. Forberg, E.B. Sjulstad, I. Bakke, Y. Olsen, A. Hagiwara, Y. Sakakura, O. Vadstein,
498 Correlation between microbiota and growth in Mangrove Killifish (Kryptolebias
499 marmoratus) and Atlantic cod (Gadus morhua), Scientific Reports 6(1) (2016).
C

500 [36] J. Wei, X. Guo, H. Liu, Y. Chen, W. Wang, The variation profile of intestinal
AC

501 microbiota in blunt snout bream (Megalobrama amblycephala) during feeding habit
502 transition, BMC Microbiology 18(1) (2018) 99.
503 [37] G.D. Alpini, M. Mekuchi, T. Asakura, K. Sakata, T. Yamaguchi, K. Teruya, J. Kikuchi,
504 Intestinal microbiota composition is altered according to nutritional biorhythms in
505 the leopard coral grouper (Plectropomus leopardus), Plos One 13(6) (2018).
506 [38] T.P.R.A. Legrand, S.R. Catalano, M.L. Wos-Oxley, F. Stephens, M. Landos, M.S.
507 Bansemer, D.A.J. Stone, J.G. Qin, A.P.A. Oxley, The Inner Workings of the Outer
508 Surface: Skin and Gill Microbiota as Indicators of Changing Gut Health in Yellowtail
509 Kingfish, Frontiers in Microbiology 8 (2018).
 ACCEPTED MANUSCRIPT
510 [39] I. Ortega Blázquez, M.J. Grande Burgos, R. Pérez-Pulido, A. Gálvez, R. Lucas,
511 Treatment With High-Hydrostatic Pressure, Activated Film Packaging With Thymol
512 Plus Enterocin AS-48, and Its Combination Modify the Bacterial Communities of
513 Refrigerated Sea Bream (Sparus aurata) Fillets, Frontiers in Microbiology 9 (2018).
514 [40] C.J. Krediet, K.B. Ritchie, A. Alagely, M. Teplitski, Members of native coral
515 microbiota inhibit glycosidases and thwart colonization of coral mucus by an
516 opportunistic pathogen, The ISME Journal 7(5) (2012) 980-990.

PT
517 [41] O. Schildgen, Y. Zuo, W. Xie, Y. Pang, T. Li, Q. Li, Y. Li, Bacterial community
518 composition in the gut content of Lampetra japonica revealed by 16S rRNA gene

RI
519 pyrosequencing, Plos One 12(12) (2017).
520 [42] P. Fan, P. Liu, P. Song, X. Chen, X. Ma, Moderate dietary protein restriction alters
521 the composition of gut microbiota and improves ileal barrier function in adult pig

SC
522 model, Scientific Reports 7 (2017) 43412.
523 [43] P. Smith, D. Willemsen, M. Popkes, F. Metge, E. Gandiwa, M. Reichard, D.R.

U
524 Valenzano, Regulation of life span by the gut microbiota in the short-lived African
525 turquoise killifish, eLife 6 (2017).
AN
526 [44] R.C. Kasana, C.B. Pandey, Exiguobacterium: an overview of a versatile genus with
527 potential in industry and agriculture, Critical reviews in biotechnology 38(1) (2018)
528 141-156.
M

529 [45] J.-S. Lee, Acinetobacter antiviralis sp. nov., from Tobacco Plant Roots, Journal of
530 Microbiology and Biotechnology 19(3) (2009).
D

531 [46] S.A. Wadhwani, M. Gorain, P. Banerjee, U.U. Shedbalkar, R. Singh, G.C. Kundu,
532 B.A. Chopade, Green synthesis of selenium nanoparticles using Acinetobacter sp.
TE

533 SW30: optimization, characterization and its anticancer activity in breast cancer cells,
534 International journal of nanomedicine 12 (2017) 6841-6855.
EP

535 [47] K. Mori, K. Yamaguchi, Y. Sakiyama, T. Urabe, K. Suzuki, Caldisericum exile gen.
536 nov., sp. nov., an anaerobic, thermophilic, filamentous bacterium of a novel bacterial
537 phylum, Caldiserica phyl. nov., originally called the candidate phylum OP5, and
C

538 description of Caldisericaceae fam. nov., Caldisericales ord. nov. and Caldisericia
AC

539 classis nov, International journal of systematic and evolutionary microbiology 59(Pt
540 11) (2009) 2894-8.
541 [48] M.A. van Kessel, R.J. Mesman, A. Arshad, J.R. Metz, F.A. Tom Spanings, S.C. van
542 Dalen, L. van Niftrik, G. Flik, S.E. Wendelaar Bonga, M.S. Jetten, P.H. Klaren, H.J. Op
543 den Camp, Branchial nitrogen cycle symbionts can remove ammonia in fish gills,
544 Environmental microbiology reports (2016).
545 [49] S. Alimperti, S.T. Andreadis, CDH2 and CDH11 act as regulators of stem cell fate
546 decisions, Stem cell research 14(3) (2015) 270-82.
547 [50] A. Kourtidis, S.P. Ngok, P.Z. Anastasiadis, p120 catenin: an essential regulator of
 ACCEPTED MANUSCRIPT
548 cadherin stability, adhesion-induced signaling, and cancer progression, Progress in
549 molecular biology and translational science 116 (2013) 409-32.
550 [51] T. Pieters, F. van Roy, J. van Hengel, Functions of p120ctn isoforms in cell-cell
551 adhesion and intracellular signaling, Frontiers in bioscience (Landmark edition) 17
552 (2012) 1669-94.
553 [52] F. Drees, S. Pokutta, S. Yamada, W.J. Nelson, W.I. Weis, Alpha-catenin is a
554 molecular switch that binds E-cadherin-beta-catenin and regulates actin-filament

PT
555 assembly, Cell 123(5) (2005) 903-15.
556 [53] L.P. Elia, M. Yamamoto, K. Zang, L.F. Reichardt, p120 catenin regulates dendritic

RI
557 spine and synapse development through Rho-family GTPases and cadherins, Neuron
558 51(1) (2006) 43-56.
559 [54] W.J. Nelson, Regulation of cell-cell adhesion by the cadherin-catenin complex,

SC
560 Biochemical Society transactions 36(Pt 2) (2008) 149-55.
561 [55] K. Katagiri, M. Hattori, N. Minato, T. Kinashi, Rap1 functions as a key regulator of

U
562 T-cell and antigen-presenting cell interactions and modulates T-cell responses,
563 Molecular and cellular biology 22(4) (2002) 1001-15.
AN
564 [56] A.L. Knox, N.H. Brown, Rap1 GTPase regulation of adherens junction positioning
565 and cell adhesion, Science 295(5558) (2002) 1285-8.
566 [57] M. Aurrand-Lions, L. Duncan, C. Ballestrem, B.A. Imhof, JAM-2, a novel
M

567 immunoglobulin superfamily molecule, expressed by endothelial and lymphatic cells,

568 The Journal of biological chemistry 276(4) (2001) 2733-41.
D

569 [58] C.A. Johnson-Leger, M. Aurrand-Lions, N. Beltraminelli, N. Fasel, B.A. Imhof,

570 Junctional adhesion molecule-2 (JAM-2) promotes lymphocyte transendothelial
TE

571 migration, Blood 100(7) (2002) 2479-86.

572 [59] P. Rougerie, V. Miskolci, D. Cox, Generation of membrane structures during
EP

573 phagocytosis and chemotaxis of macrophages: role and regulation of the actin
574 cytoskeleton, Immunological reviews 256(1) (2013) 222-39.
575 [60] M.A. Salazar, A.V. Kwiatkowski, L. Pellegrini, G. Cestra, M.H. Butler, K.L. Rossman,
C

576 D.M. Serna, J. Sondek, F.B. Gertler, P. De Camilli, Tuba, a novel protein containing
AC

577 bin/amphiphysin/Rvs and Dbl homology domains, links dynamin to regulation of the
578 actin cytoskeleton, The Journal of biological chemistry 278(49) (2003) 49031-43.
579
 ACCEPTED MANUSCRIPT
580 Tables
581 Table 1.Sequences of primers used in the present study
582
Primer Name Orientation Nucleotide Sequences(5'-3') Primer Usage
341F sense CCTAYGGGRBGCASCAG amplify V3-V4 region
806R antisense GGACTACNNGGGTATCTAAT amplify V3-V5 region
beta-actin sense TCCACCgCAAATgCTTCTAA real-time PCR

PT
beta-actin antisense TgCgCCTgAgTgTgTATgA real-time PCR
NF-kB sense CAGGACGGCAACGGAGA real-time PCR

RI
NF-kB antisense TGCTGCTGACTGCTGAG real-time PCR
TLR3 sense CTggCTTACTACAACCACCCC real-time PCR
TLR3 antisense CAAACTCCCTgCCCTCTTCA real-time PCR

SC
IL-1B sense CCAGCGTTGAGGGCAGAA real-time PCR
IL-1B antisense ATCGTCTCCAGATGTAAGGTT real-time PCR

U
INF-1 sense CTgTgTCCTTCCCgAATCAT real-time PCR
INF-1 antisense gTgCAgCTTCTTgCTCTCCT real-time PCR
AN
IFN-2 sense TggAggCgTACgAAAAgCTg real-time PCR
IFN-2 antisense gCTCgTTgTACCggTgTTTC real-time PCR
Mx gene sense ATTgTCAggTgCAgAggTCA
M

real-time PCR
Mx gene antisense TCTTTCAgCCAgCTCAggAA real-time PCR
D

583
TE

584
585 Table 2. The microbial diversity index including Shannon, Simpson, Chao1 and ACE
586 of 16S rRNA sequence library. Shannon and Simpson index is used to identify
EP

587 bacterial community diversity; besides, Chao1 and ACE is used to identify bacterial
588 richness.
C
AC

589
590
 ACCEPTED MANUSCRIPT
591 Figure legends
592 Figure 1: The morphology of E.coioides during metamorphosis. (A) 12 days after
593 hatching, the second dorsal started to lengthen. (B) 18dph, the second dorsal and
594 pelvic fin spines were clearly visible and reach to the maximal length. (C) 50dph,
595 transformation of larvae into juveniles occurred
596
597 Figure2: The diversity and richness of the intestinal flora in E.coioides. (A) Species

PT
598 accumulation boxplot: The plot demonstrates the adequacy of sample size and the
599 species richness. Plotted by the "Sample number" on the X-axis and "OTU number"

RI
600 on the Y-axis.
601
602 Figure3. Principal component analysis (PCA): each point represents a sample, plotted

SC
603 by the second principal component on the Y-axis and the first principal component on
604 the X-axis, which was colored by group.

U
605
606 Figure4. The species and genera level of intestinal microbiomes in different
AN
607 metamorphosis stages. (A) The top 10 species in the different taxonomic ranks were
608 selected to form the distribution histogram of relative abundance. Plotted by the
609 "Relative Abundance" on the Y-axis and "Samples Name" on the X-axis. "Others"
M

610 represents a total relative abundance of the rest phylum besides the top 10 phylum. (B)
611 The abundance distribution of dominant 35 genera among all samples was displayed
D

612 in the species abundance heatmap. Plotted by sample name on the X-axis and the
613 Y-axis represents the genus. Red color represented most abundance and blue color
TE

614 represented less abundance.

615
EP

616 Figure5. Venn diagram: the OTUs in each metamorphosis stages (12dph, 18dph and
617 50sph) are 682, 1285 and 50dph, respectively. According to the analysis results of
618 OTUs clustering, the common information of the grouper in three different
C

619 development stages is 701 OTUs.

AC

620
621 Figure 6. The functional pathway observed by 16s rRNA gene sequencing and
622 transcriptome analysis. (A) 16s rRNA gene sequencing: the functional changes in the
623 microbiome of 12dph and 50dph. KEGG pathway (level 3) indicates statistically
624 significant differences between the two data sets (p-value<0.05). (B) The analysis of
625 gene expression in leukocyte transendothelial migration pathway through KEGG.
626 This pathway shows the migration of leukocyte from blood into tissue by going
627 through the cell junction between endothelium cells. The up-regulated genes were
628 labelled with red, while the down-regulated genes were labelled with green. (C) The
 ACCEPTED MANUSCRIPT
629 analysis of gene expression in phagosome pathway through KEGG. The up-regulated
630 genes were labelled with red, while the down-regulated genes were labelled with
631 green.
632
633 Figure 7. The expression of immune genes (NF-κB, TLR-3, IL-1β, Type I IFN, Type
634 II IFN and Mx gene) was determined by real-time PCR. NF-κB was slightly higher
635 than the other immune genes (TLR-3, IL-1β, Type I IFN, Type II IFN and Mx gene) at

PT
636 12dph, 18dph (A) and 50dph (B).
637

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

638
639 Figure 1: The morphology of E.coioides during metamorphosis. (A) 12 days after
640 hatching, the second dorsal started to lengthen. (B) 18dph, the second dorsal and

PT
641 pelvic fin spines were clearly visible and reach to the maximal length. (C) 50dph,
642 transformation of larvae into juveniles occurred

RI
643

U SC
AN
M
D
TE
C EP
AC

644
645 Figure2: The diversity and richness of the intestinal flora in E.coioides. (A) Species
646 accumulation boxplot: The plot demonstrates the adequacy of sample size and the
647 species richness. Plotted by the "Sample number" on the X-axis and "OTU number"
648 on the Y-axis.
649
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
650
651 Figure3. Principal component analysis (PCA): each point represents a sample, plotted
M

652 by the second principal component on the Y-axis and the first principal component on
653 the X-axis, which was colored by group.
D

654
655
TE

656
657
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
658
M

A.
D
TE
CEP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D

659 B.
660 Figure4. The species and genera level of intestinal microbiomes in different
TE

661 metamorphosis stages. (A) The top 10 species in the different taxonomic ranks were
662 selected to form the distribution histogram of relative abundance. Plotted by the
EP

663 "Relative Abundance" on the Y-axis and "Samples Name" on the X-axis. "Others"
664 represents a total relative abundance of the rest phylum besides the top 10 phylum. (B)
665 The abundance distribution of dominant 35 genera among all samples was displayed
C

666 in the species abundance heatmap. Plotted by sample name on the X-axis and the
AC

667 Y-axis represents the genus. Red color represented most abundance and blue color
668 represented less abundance.
 ACCEPTED MANUSCRIPT

PT
RI
SC
669

U
670 Figure5. Venn diagram: the OTUs in each metamorphosis stages (12dph, 18dph and
671 50sph) are 682, 1285 and 50dph, respectively. According to the analysis results of
AN
672 OTUs clustering, the common information of the grouper in three different
673 development stages is 704 OTUs.
674
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

675 A.
676

PT
RI
U SC
AN
677 B.
678
M
D
TE
C EP

679 C.
AC

680 Figure 6. The functional pathway observed by 16s rRNA gene sequencing and
681 transcriptome analysis. (A) 16s rRNA gene sequencing: the functional changes in the
682 microbiome of 12dph and 50dph. KEGG pathway (level 3) indicates statistically
683 significant differences between the two data sets (p-value<0.05). (B) The analysis of
684 gene expression in leukocyte transendothelial migration pathway through KEGG.
685 This pathway shows the migration of leukocyte from blood into tissue by going
686 through the cell junction between endothelium cells. The up-regulated genes were
687 labelled with red, while the down-regulated genes were labelled with green. (C) The
 ACCEPTED MANUSCRIPT
688 analysis of gene expression in phagosome pathway through KEGG. The up-regulated
689 genes were labelled with red, while the down-regulated genes were labelled with
690 green.
691

PT
RI
U SC
692 A.
AN
M
D
TE
C EP

693 B.
AC

694 Figure 7. The expression of immune genes (NF-κB, TLR-3, IL-1β, Type I IFN, Type
695 II IFN and Mx gene) was determined by real-time PCR. NF-κB was slightly higher
696 than the other immune genes (TLR-3, IL-1β, Type I IFN, Type II IFN and Mx gene) at
697 12dph, 18dph (A) and 50dph (B).
698
 ACCEPTED MANUSCRIPT

Research highlights

1. We investigate the changes in intestinal microbiota of grouper at the pre-, mid-,

and post-metamorphosis stages.

2. The drastic changes in the microbial communities at different developmental

PT
stages.

3. Some gene in biological pathways was up-regulated at pre, mid-, and

RI
post-metamorphosis.

SC
4. We performed metagenomic and transcriptomic analyses database between the
relationship of microbiota and immune system.

U
AN
M
D
TE
C EP
AC