The intestinal microbial diversity in mud crab (Scylla paramamosain) as determined by PCR-DGGE and clone library analysis

Authors

Errata

This article is corrected by:

  1. Errata: Corrigendum Volume 114, Issue 6, 1874, Article first published online: 17 May 2013

Correspondence

Shengkang Li, Guangdong provincial key laboratory of marine biology, Shantou University, Shantou 515063, China. E-mail: lisk@stu.edu.cn

Abstract

Aims

To identify the intestinal microbial diversity in mud crab and to investigate the bacterial difference in the intestinal microbiology between wild crabs (WC), pond-raised healthy and diseased crabs (DC).

Methods and Results

The intestinal microbial community of mud crab Scylla paramamosain from WC, pond-raised healthy crabs (HC) and DC were examined by Denaturing Gradient Gel Electrophoresis (DGGE) and clone library analysis of 16S rDNA gene. Eight of 21 representative DGGE bands were affiliated with unidentified or unclassified bacteria. Vibrio, Pseudoalteromonas and Shewanella were also found from the DGGE gel. Analysis of clone libraries revealed that all sequenced clones were grouped into either of the following phyla: Proteobacteria, Firmicutes, Tenericutes, Bacteroidetes, Fusobacteria, Cyanobacteria and unidentified or unclassified bacteria. The phylotypes affiliated with Firmicutes were not found in DC library, yet DC had a little portion of Cyanobacteria which did not exist in both WC and HC library. Real-time PCR showed that the abundance of the total bacterial load in WC were significantly three times higher than that in healthy and DC, the abundance of Bacteriodes in healthy and WC were as much four times, three times as that in DC, respectively.

Conclusions

Statistical analysis showed that the bacterial communities in intestine of the mud crab from these three populations were significantly different. The phylotypes of the Bacteriodes and Tenericutes were the dominant population in the gut of the mud crab.

Significance and Impact of the Study

This study demonstrated significant differences in the intestinal bacterial composition of three crab populations. This knowledge will increase our understanding of the effect of growth conditions on bacterial community composition in the crab gut and provide necessary data for further development of probiotic products for diseases prevention in crab farming.

Introduction

The mud crab Scylla paramamosain (S. paramamosain), widely cultured in brackish and seawater ponds with salinities below 9‰ along the coast of southern China, has suffered from seriously infectious diseases in recent years (Li et al. 2012). An epidemic named as ‘milky disease’, which breaks out mainly in the fall (from September to November) when the crab is near maturity, results in large economic losses in crab farming. The name of ‘Milky disease’ came from the typical symptom of the diseased mud crab, mainly because the body fluid inside the carapace and appendages of the mud crab became milky when the disease spread. The other symptom of milky disease crab showed a frail situation of lethargy, decreasing appetite, reducing ingestion and paralysis of ambulatory legs. The histopathological examination and molecular identification of the pathogens suggested that Hematodinium sp. (Li et al. 2008), together with the mixed infection of Aeromonas hydrophila and Vibrio parahaemolyticus (Xia et al. 2010), were the main pathogens of the diseased crabs (DC).

The use of antibiotics in aquaculture for diseases prevention and treatment may cause development of antibiotic resistance among pathogens-infected animals and humans, affecting the successive development of its aquaculture. In aquaculture industry, the manipulation of the gut microbiota can enhance growth, feed digestion, immunity and disease resistance of the host organism (Burr et al. 2005). The beneficial microbial species (named as ‘probiotics’) and the beneficial compounds (named as ‘prebiotics’) have been introduced in the intestinal tract of the host organism, which harvest significant effectiveness (Li et al. 2007b; Aly et al. 2008). Especially, probiotics have been applied successfully in shrimp and other crusteaceans which have high commercial value (Vaseeharan and Ramasamy 2003; Rengpipat et al. 2000; Wang 2007; Chu et al. 2011). To understand the roles of gut bacteria of the mud crab, it should be done at the first step to investigate the composition of bacterial groups in the gut of the host.

Recently, the intestine microbial diversity has been studied in a wide range of aquatic animals. Shrimp intestinal microbial diversity investigated by molecular-dependent methods have revealed that the predominant bacterial population in the intestine of Chinese shrimp Fenneropenaeus chinensis were Proteobacteria and Vibrio sp. (Liu et al. 2011). In the gut of another freshwater culture animal, the Chinese mitten crab, the investigation illustrated that the Proteobacteria and Bacteroidetes might be the dominant population (Li et al. 2007a). The vibrio group was a dominant component in the gut microflora of abalone (Tanaka et al. 2004). In another report, Proteobacteria belonging to the gamma subclass (mainly Aeromonas and Enterobacteriaceae) dominated the intestinal microbiota of rainbow trout (Huber et al. 2004). However, in the economically important aquaculture species mud crab S. paramamosain, the diversity and dominant population of the intestinal bacteria is still unknown.

Previous researches on intestine microbial diversity in aquatic animals were mostly replying on classical cultivation methods. Yet, these studies have limited the understanding of the complexity of the gut ecosystem and the contribution of different bacteria to processes such as competitive exclusion (Kim et al. 2007). It is unequivocal that <10% of intestinal microbes could be cultured by currently available methods. Molecular methods based on the recovery of microbial diversity by sequence analysis of 16S rDNA gene have proven powerful for studying microbial communities. Denaturing Gradient Gel Electrophoresis (DGGE), a cultured independent method, had been successfully introduced into the research of complex microbial populations of different environments, such as soil fungal communities (Oros-Sichler et al. 2006), lake sediment bacterial diversity (Li et al. 2006), the bacterial endophytes of plant (West et al. 2010), the intestine microbial diversity of terrestrial and aquatic animals (Huber et al. 2004; Janczyk et al. 2007; Liu et al.2011). Meanwhile, the clone library analysis of 16S rDNA gene had proved to be a powerful and useful approach for the analysis of microbial populations although its artifacts and bias (Acinas et al. 2005).

In this study, culture-independent method PCR–DGGE, clone library analysis of 16S rDNA gene were used to investigate the bacterial diversity and dominant population in the intestinal tract of three different mud crab S. paramamosain populations [wild crabs (WC), milky crabs and healthy crabs] from Shantou, China. The variations in bacterial communities among three libraries were compared by statistical methods. The abundance of Bacteriodes and the total bacterial load were performed by the real-time PCR. The data here would lay basis for further development of potential probiotics for diseases prevention in cultivation of mud crab S. paramamosain.

Materials and methods

Sampling of mud crabs

The collected crabs were 1-year-old animals with a weight of approx. 100 g. WC were captured in Shantou bay, healthy and DC were sampled from two ponds in Niutianyang crab farm in Shantou nearby each other. The crabs which showed the typical ‘milky disease’ symptom were collected for further analysis in laboratory. All these three populations with three individuals of each group were euthanized, and the whole intestinal tracts were immediately removed with a sterile tweeze and clamped to prevent loss of samples. The samples were stored at −80°C before DNA extraction.

DNA extraction from the intestine

Nucleic acid extraction of the intestinal tracts was performed essentially as described previously (Huber et al. 2004) with some modifications. Accurately, 100 mg intestine of Scylla paramamosain from each population was separately suspended in a 3 ml phosphate buffer and 80 µl of 20% polyvinylpolypyrrolidone (PVPP) were added to a final concentration of 0·5%, mixed by the vortex and then centrifuged at 400 g for 6 min, collected the supernate and added lysis buffer I (5 mol l−1 NaCl, 0·1 mol l−1 EDTA, pH 8·0) and lysozyme at the final concentration of 2·5 mg ml−1 and RNase A with a final concentration of 10 mg ml−1. The solution was incubated at 37°C for 30 min. Then, lysis buffer II (5 mol l−1 NaCl, 0·5 mol l−1 Tris–HCI, PH 8·0), 50 μl of 20% sodium dodecyl sulphate (SDS) with a final concentration of 1% and 50 µl of 20% PVPP were added, and then incubated on ice for 5 min. The lysates were purified twice by extraction with an equal volume of phenol–chloroform–isoamyl alcohol [25 : 24 : 1 (v/v/v)], and the residual phenol was removed by extraction with an equal volume of chloroform–isoamyl alcohol (24 : 1). Finally, nucleic acids were precipitated with isopropyl alcohol, rinsed with 70% frozen ethanol and suspended in 30 ml of TE buffer. The quality and quantity of the total DNA was checked by 0·8% agarose gel electrophoresis and measured by a UV/Visible Spectrophotometer (Amersham Biosciences, Piscataway, NJ) at 260 and 280 nm. Nucleic acid extracts were stored at −20°C until further analysis.

PCR-DGGE analysis

The extracted DNA samples were further analysed using PCR-DGGE fingerprinting technology. A 180-bp fragment (V3 region) of the 16S rRNA gene was amplified using the bacteria universal primers GC-338F (5′-GCclamp-ACTCCTACGGGAGGCAGCAG-3′) and 518R (5′-ATTACCGCGGCTGCTGG-3′) (Ovreás et al. 1997), the GC clamp sequence is 5′-CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGGGG-3′, as described previously (Hovda et al. 2007). Amplification was carried out in 50 μl reaction contained 5 μl of 10× PCR buffer, 4 μl dNTP mixture (0·2 µmol each), 1U Taq DNA polymerase (TaKaRa, Dalian, China), 0·4 μmol of each primer, and 2 μl of genomic DNA. The touchdown PCRs were cycled in a thermal cycler with an initial denaturation at 95°C for 5 min followed by 20 cycles of denaturation for 45 s at 94°C, primer annealing for 45 s from 65 to 55°C (at each temperature for two cycles) and primer extension for 1 min at 72°C. For another 20 cycles at the annealing temperature of 55°C for 45 s, followed by additional elongation at 72°C for 10min. Sterile distilled water was used instead of genomic DNA isolated from the microbial community for negative control in PCR. Band was excised from the gel, and the DNA was purified with DNA purification kit (Promega, Madison, WI, USA).

DGGE was performed with a Bio-Rad DCode mutation detection system (Bio-Rad, USA) according to the manufacturer's instructions. Approximately, 1·5–2 μg of PCR product was deposited in each well of 8% (weight in volume, w/v) polyacrylamide gels containing a linear 30–50% denaturant gradient. Electrophoresis was performed for 30 min at a constant voltage of 30 V at 60°C and then with an increase in the voltage to 130 V for another 6 h. The gels were stained with dissolved EB in TAE solution and photographed. DGGE band recovery and clustering were carried out as described previously (Wang et al. 2005).

Clone library construction

The intestinal contents of three populations of crab individuals were pooled, respectively, and total DNA was extracted and purified as described previously. 16S rRNA genes were amplified with a Bacteria-specific primer pair B27F&B1492R (Danovaro et al. 2006) and were cloned by using the pMD-19T vector cloning system and introduced into competent Escherichia coli DH5α cells. Inserts of the expected size (approx. 1500 bp) were amplified by PCR with the M13 forward and reverse primer. Clones were grouped on the basis of the restriction patterns obtained after digestion with Afa I/Msp I (Takara). For restriction analysis, 10 μl of PCR products, 2 μl of 10× TE buffer, 2 μl 0·1% BSA and 1 μl (1 U) of an enzyme were added to a sterile tube, the final volume was made up to 20 μl with autoclaved MilliQ water and incubated at 37°C for 3–4 h. Digested product was visualized by electrophoresis in 1·5% agarose gel along with DL2000 marker (Takara). Different OTUs are detected by the Quantity One software and eye identification as a supplement. Three clones were used for sequencing in Beijing Genomics Institute, Shenzhen Branch (Shenzhen, China) in each OUT (one clone was sequenced if the OTU has only one). To determine the closest known relatives of the 16S rRNA sequences obtained, searches were performed with the Blastn program (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The 16S rRNA gene sequences obtained from the intestinal bacteria in mud crabs have been deposited to GenBank with the accession numbers from HE610317 to HE610337 and HE611094 to HE611123.

Phylogenetic tree construction

The richness, diversity and dominance indexes within the microbial populations as well as the similarities between the microfloras of different intestinal products were calculated from clone library analysis using the Ward method (Ampe and Miambi 2000), with the helping of Quantity One software package following the strategy proposed by Eichner et al. (1999). Phylogenetic tree of the sequence analysis of the three clone libraries were also constructed from a matrix of pairwise genetic distances by neighbor-joining (N-J) method. The bootstrap analysis of 1000 replicates was performed (Boon et al. 2002). The construction of N-J tree of different libraries from the isolated colonies with the bootstrap of 1000 resamplings was performed using Mega 4.1 software (Arizona State University, AZ, USA).

Real-time quantitative PCR

The representative clone from the phylotype H-B6 was used to establish the standard curve in each RTQ-PCR experiment. As H-B6 assigned to Bacteroidetes is one of the most dominant phylotypes common in three crab populations. Recombinant plasmid DNA of H-B6 was isolated, purified and serially diluted with a ratio of 1 : 10 in distilled water. One-microlitre aliquots of each dilution were used for RTQ-PCR to generate the standard curve, and used as quantification standards for mud crab intestine samples. The DNA isolated from three samples in each group was mixed with an equal volume and was run using an ABI 7300 Real-time Detection System (Applied Biosystems, Foster City, CA, USA). The data were recorded and analysed with the corresponding Monitor software.

Primers BfrF and BfrR were used to quantify the Bacteroidetes group of organisms, whereas a set of universal primers (the forward primer, 5′-TCCTACGGGAGGCAGCAGT-3′ and the reverse primer, 5′-GGACTACCAGGGTATCTAATCCTGTT-3′) (Li et al. 2007a) were used for the amplification of the 16S rRNA gene to estimate the total bacterial load of the intestine of three populations of mud crabs with the same conditions of DNA extraction at one experiment. The amplification was performed in a total volume of 20 μl, containing 10 μl of 2 × SYBR Green I real-time PCR Master Mix (TaKaRa), 1 μl of the diluted genomic DNA and 0·8 μl of each primer (primers specific for Bacteroidetes and bacteria universal primers, respectively). The real-time PCR program was 95°C for 1 min, followed by 40 cycles of 95°C for 15 s and 60°C for 60 s. Dissociation analysis of amplification products was performed at the end of each PCR to confirm that only one PCR product was amplified and detected. During the PCR program, the signal was measured and then the data were analysed by the related software.

Statistical analysis

The mean and SD value of peak intensities in the DGGE profile was calculated by the WS software (Microsoft, Redmond, WA, USA). The figures were pictured by Origin 7.0 if necessary. Coverage index was calculated by

display math

(n1 is the number of the OUT and N is the number of all clones). Simpson's index was calculated according to the equation:

display math

(ni is the number clones in the OUT and N is the number of all clones).

Results

DGGE analysis of V3 regions

The total genomic DNA were extracted from the intestine of mud crabs with a range of 400 ng to 2 μg from each sample, which is similar to that in Chinese mitten crab Eriocheir sinensis (Li et al. 2007a). A 180bp of V3 region of the 16S rRNA gene was amplified by PCR from the purified genomic DNA and DGGE profiles were performed in different sources of mud crabs (WC: sample 1,2,3; healthy crabs: sample 4,5,6; DC: sample 7,8,9) (Fig. 1). Twenty DGGE bands on average were observed from each sample, and each sample profile displayed a unique banding pattern theoretically. All bands appearing in the DGGE gel were initially collected for sequencing. Forty-three representative bands from the DGGE gel were eluted and sequenced, 21 unique sequences were obtained eventually, while eight of which were affiliated with unidentified or unclassified bacteria. In addition, Vibrio, Pseudoalteromonas and Shewanella were also found (Table 1). The cluster analysis of these band-types showed that sample 1 and sample 8 treed in different subbranch within their own group, while sample 4 was even not classified in the same branch of the healthy crab group (Fig. 2), indicating that obvious intersubject variation existed in intestinal bacteria in all three populations. Generally, the crabs from different sources were significantly clustered (Fig. 2), indicating that intergroup variation was principle characteristic among the intestinal communities. All sequences derived from DGGE bands and compared by a Blast program in GenBank website were deposited in the EMBL database.

Figure 1.

The Denaturing Gradient Gel Electrophoresis profile of the selected samples from the intestine of different mud crabs 1–3 are samples from the gut of the wild crabs, 4–6 from that of the healthy crabs (HC), 7–9 from that of the diseased crabs which suffered from ‘milky disease’ as described in the materials and methods. The 21 high intensive bands information were listed in Table 1.

Figure 2.

Cluster analysis showing the similarity of the bacterial communities in intestine of three mud crab populations.

Table 1. The cloned sequences and their distribution by Denaturing Gradient Gel Electrophoresis method
CloneClosest relative (accession no.)Percentage similaritySourceBand distributiona
  1. WC, wild crabs; DC, diseased crabs; HC, healthy crabs.

  2. a

    The distribution of cloned bands in the intestine were shown.

1Clostridia bacterium (GU136592.1)97Marine sedimentHC, DC
2Pedobacter sp. LMG 10343 (AF329963)98SoilDC
3Pseudoalteromonas sp. GSO3TIBB4 (DQ005207.1)95 Paralichthys dentatus WC, HC, DC
4Cetobacterium somerae WAL 14325 (AJ438155)96GutWC, HC, DC
5Uncultured bacterium clone BP1393 Neanthes glandicincta WC,DC
6Vibrio harveyi (JN183120.1)99 Portunus trituberculatus WC, HC, DC
7Shewanella sp. (BSi20505)96Arctic seaWC, HC, DC
8Uncultured bacterium clone SC57 (GU293190.1)98 Pelteobagrus fulvidraco WC, HC, DC
9Shewanella sp. 1-Aa (AY770009.1)94Arctic seaWC, HC, DC
10Anaerorhabdus furcosa (HM038002.1)97Marine sedimentWC, HC, DC
11Uncultured bacterium clone SC57 (GU293190.1)93 Pelteobagrus fulvidraco WC, HC, DC
12Uncultured bacterium clone SAug3996 Indo-Pacific oyster HC
13Prevotella sp. oral clone DO039 (AF385513)92Environmental sampleWC, HC, DC
14Mycoplasma hyopneumoniae (GU227406.1)95SwineWC, HC, DC
15Uncultured Mycoplasmataceae bacterium (EU646198.1)97HepatopancreasHC
16Uncultured Holophaga sp. (AM072424)94Mining wasteWC, HC
17Uncultured bacterium GQ866067.197GutDC
18Psychrilyobacter sp. STAB 704 (JF825448.1)95St. Andrews BayWC, DC
19Uncultured Prevotella PUS9.18096Environmental sampleWC, HC, DC
20Prolixibacter bellariivorans (AY918928.1)95Marine sedimentWC, HC, DC
21Photobacterium damselae (FR687001.1)97 Miichthys miiuy WC, HC, DC

Phylogenetic analysis of 16S clone library

Three clone libraries of 16S rDNA gene from wild, pond-raised healthy and DC were constructed to investigate the microbial diversity of intestine. In total, 153 clones from WC, 127 ones from HC, 95 from DC were retrieved in the colony libraries. The percentage coverage of WC, HC, DC libraries was 95·78, 90·32, 91·04%, respectively (Table 2), indicating that each clone library covered most of the micro-organisms in intestine of the mud crab.

Table 2. Comparison of coverage and diversity indices of 16S rRNA gene clone libraries
Clone libraryNumber of clonesNumber of phylotypesCoverage (C) index (%)Simpson's index (D) (%)Shannon–Wiener index (H)
  1. WC, wild crabs; DC, diseased crabs; HC, healthy crabs.

WC1532595·7893·052·83
HC1272690·3289·782·68
DC951891·0484·112·15

The clone analysis showed that the sequence population was the least diverse in the DC and the most diverse in the WC. This was further confirmed by calculating Shannon–Wiener index (H) and Simpson's indices (D) (2·15–2·83, 0·8411–0·9305, respectively) (Table 2). The data indicated that the species richness of WC was highest, followed by that of HC, while DC had the lowest species richness. This result was consistent with the DGGE profiles (Fig. 1). Thus, the bacterial diversity was higher in the guts of healthy crabs than in those of sick crabs.

The sequences of a length of approx. 1500 bp for clones representing individual phylotypes were obtained from clone libraries and used to construct the phylogenetic tree (Fig. 3). A total of 178 sequences of bacterial 16S rDNA gene were obtained from clone library analysis, and 129 unique sequences were eventually obtained after a filtering procedure removing highly identical sequences with a criteria of 97% cut-off. Phylogenetic analysis showed that seven clones were clustered into Proteobacteria. These clones were common in various libraries and distributed three clones, two clones, two clones in the WC, HC, DC library, respectively. The phyla Fusobacteria was also common in three crab populations and three clones in WC, seven clones in HC, five clones in DC were retrieved from the clone libraries, respectively. The other two phyla Bacteroidetes and Tenericutes were listed as some of the most dominant components of the intestinal bacteria in mud crab. In phylogenetic tree, 29 clones were assigned to be Bacteroidetes (12 clones in WC, 9 clones in HC, 8 clones in DC, respectively) and 32 clones to be Tenericutes (10 clones in WC, 13 clones in HC, 9 clones in DC, respectively). Of particular note in this survey, no clones assigned to Firmicutes were found in DC library (9 Firmicutes clones in WC and 7 ones in HC, respectively), whereas a little portion of Cyanobacteria (3/34) existed. It is noteworthy that a substantial part of unidentified or unclassified bacteria was found in all of three libraries, with the details of 11 clones in WC, nine ones in HC and seven clones in DC library.

Figure 3.

Dendrogram of 16S rRNA gene sequences showing the phylogenetic affiliation of intestinal bacteria in mud crabs. The neighbor-joining tree was constructed from 16S rRNA gene sequences of representative clones of each phylotype and sequences retrieved from the GenBank database. The scale bar represents five nucleotide substitutions per 100 nucleotides. Bootstrap values (100 replicates) are shown at branch nodes. W represents wild crab, while H and D represent healthy and diseased pond-raised crabs from two nearby farms.

The identified taxa for each clone library mainly grouped into seven different phyla (Fig. 4). All sequenced clones were grouped into either of the following phyla: Proteobacteria, Firmicutes, Tenericutes, Bacteroidetes, Fusobacteria, Cyanobacteria and unidentified or unclassified bacteria. About 64·3% of the phylotypes of the clone libraries in the WC were affiliated with Bacteroidetes, Tenericutes and Firmicutes. Nearly, 61·6% of the phylotypes of the clone libraries in the healthy crabs were affiliated with Bacteroidetes, Tenericutes and Fusobacteria, while 64·7% of the phylotypes in the clone library of diseased ones were classified into Bacteroidetes, Tenericutes and Fusobacteria. Similarly, a significant part of unidentified or unclassified bacteria were found in three libraries. The proportion of the phylotypes of uncultured bacteria in WC, HC and DC library was 23·20%, 19·24% and 20·58%, respectively (Fig. 4). Bacteroidetes and Tenericutes dominated the intestine microbial community in three crab populations. The data indicated that the diversity and dominant components in the intestine of mud crabs of three populations were significantly different.

Figure 4.

Percentage abundance diagram of 16S rRNA gene library clones from wild crab (WC), pond-raised healthy crab (HC) and diseased crab. Percentage distribution was calculated on the basis of relative abundance in the total analysed clones. (image) Unclassified bacteria; (image) Bacteroidetes; (image) Proteobacteria; (image) Firmicutes; (image) Tenericutes; (image) Fusobacteria and (image) Cyanobacteria.

Enumeration of the intestinal microflora by Real-time PCR

Absolute expression provides the exact copy number following transformation of the data via a standard curve. Standard curve methods have become widely used for the purpose of calibrating real-time PCRs against known concentrations of nucleic acids. RTQ-PCR results indicated that the corresponding copy numbers of total bacterial load were 10·63 ± 1·24 × 106, 2·44 ± 0·87 × 106 and 2·72 ± 1·34× 106 copies per microlitre of DNA from WC to DC library, whereas the copy numbers of the genera Bacteroidetes were 4·90 ± 1·80 × 103, 6·37 ± 1·53 × 103 and 1·67 ± 1·03× 103 copies per microlitre of DNA in WC, HC and DC library, respectively (Fig. 5). Therefore, the abundance of the total bacterial load in wide crabs were significantly more than three times higher than that in healthy and DC, the abundance of Bacteriodes in healthy and WC were as much four times, three times as that in DC, respectively (< 0·05). The pond-raised healthy and DC are equivalent abundance in term of total bacterial load but the 16S rRNA gene copy number of the genera Bacteroidetes in healthy crabs was much higher than that in DC (< 0·05), indicating that the genera Bacteroidetes might play an important role on the disease in the intestine of mud crab. The data are presented as the means of triplicate determinations. The standard deviation of means varied by ≤15% (Fig. 5).

Figure 5.

The 16S rDNA gene copy number of the total bacteria and Bacteriodes load by real-time quantitative PCR in three crab populations. WT, total bacteria of wild crabs (WC); HT, total bacteria of healthy crabs; DT, total bacteria of diseased crabs (DC); WB, Bacteriodes of WC; HB, Bacteriodes of healthy crabs; DB, Bacteriodes of DC.

Discussion

An epidemic in mud crab farms named as ‘milky disease’, which breaks out mainly in the fall (from September to November) when the crab is near maturity, results in large economic losses in crab farming in southern China. The name of ‘Milky disease’ came from the typical symptom of the diseased mud crab, mainly because the body fluid inside the carapace and appendages of the mud crab became milky when the disease spread. Until now, there is no such report that this milky disease is related to the intestinal pathogens or the change of intestinal microbial composition. The purpose of this study was to describe the diversity of intestinal bacteria from wild crab (WC), pond-raised healthy crab (HC) and DC populations and provide necessary data for further development of probiotic products for diseases prevention in crab farming.

It is unequivocal that conventional culture-based techniques could not present a comprehensive picture of the intestine microbial diversity, the culture-independent methods PCR-DGGE and clone library analysis of 16S rDNA were used in the present study to compare the diversity of microbes associated with the intestine of mud crabs. In this study, although intersubject variation existed individually in the DGGE profiles of every population, we focused mainly on intergroup variations. Cluster analysis of DGGE profiles showed that intergroup variations were obvious (Fig. 2). Each group of crab still shares similarity in gut community structure so that the samples from each group could be pooled to get an overall description of each group's gut bacterial structure. Variations in the gut bacteria community among WC were more obvious than pond-raised crabs in present study. Our results are a little bit different from that done in Chinese mitten crab (Li et al. 2007a), where pond-raised crabs had higher bacterial diversity. The higher total diversity and abundance of the intestinal bacteria in WC than that in the pond-raised crabs (healthy crabs and DC) are probably because that WC have more diverse food resources compared with the pond-raised crabs.

The values of the 16S rDNA gene copy number obtained here by RTQ-PCR ranged from approx. 10·6 × 106 to 2·4 × 106 CFU g−1 from WC to DC. The 16S rDNA gene copy number can converted into cell counts as it is assumed that the average number of copies of 16S rDNA gene per cell was the same for three crab populations (Liu et al. 2005), although the number of rDNA gene operons per bacterial genome could varied (Klappenbach et al. 2001). The abundance of the total bacterial load in WC were significantly more than three times higher than that in healthy and DC, the abundance of Bacteriodes in healthy and WC were as much four times, three times as that in DC. This is similar to the results that are obtained in rainbow trout (Kim et al. 2007). However, typical viable counts in mud crab intestine (106–108) are significantly lower than those reported for humans and terrestrial animals (approx. 1011 CFU g−1) (Mead 1997). These were partly because that the anaerobic microbial communities of the intestines of the mud crab were underestimated (Austin and Austin 1999).

In our study, it is showed that only the intestine of DC had a little portion of Cyanobacteria. It could be presumed that the phylotypes affiliated with Cyanobacteria, maybe in part, resulted in the crab disease. Cyanobacterial lipopolysaccharides are frequently reported as toxins responsible for a variety of health effects in humans or other animals, from skin rashes to gastrointestinal, respiratory and allergic reactions (Stewart et al. 2006). It is showed that isolated intestinal enterocytes from chicks are both deformed and killed by exposure to Microcystis toxins, in a dose-dependent and time-dependent manner (Falconer et al. 1992). Blooms of blue-green algae (Cyanobacteria) in many eutrophic to hypertrophic ponds and lakes in tropical zones could multiply so prolifically and produce a wide range of potent toxins as secondary metabolites. The health of mud crab would be affected when these Cyanobacteria could enter into the intestine of the host.

The majority of phylotypes in three crab populations detected by clone library analysis were assigned to Bacteriodes, Proteobacteria, Fusobacteria and Tenericutes. Proteobacteria and Fusobacteria were just common but not dominant in intestine of three crab populations by clone analysis. Actually, Bacteriodes and Tenericutes dominate the gut microflora of mud crabs in our study. Some Bacteroides species might be opportunistic pathogens owing to their association with a variety of soft tissue and other infections (Liu et al. 2003). Besides Bacteroides, some genera which were affiliated with Proteobacteria were found to be common in different populations. Shewanella algae is of great interest to people because of its ability to reduce the amount of radioactive waste in groundwater by making it less soluble. Photobacterium damselae subsp. piscicida (previously known as Pasteurella piscicida) is a Gram-negative rod-shaped bacterium that causes disease in fish. It was first isolated in mortalities occurring in natural populations of white perch Morone americanus and striped bass Morone saxatilis in 1963. From 1990, it has caused economic losses in different European countries (Barber and Swygert 2000).

A big portion of Gram-positive bacteria clones assigned to Firmicutes were retrieved from WC and HC library (pond-raised health crabs), respectively, whereas no Firmicutes clone was found in DC library (pond-raised DC). Some genera belong to Firmicutes such as Clostridium sp., Clostridia bacterium as well as Eubacterium moniliforme were identified as unique in WC and HC libraries. In some reports, Clostridium isolates from pig intestinal tract were found to be multiple enzymatic activities (Varel and Pond 1992). The significance of Firmicutes to the balance of gut bacteria community in mud crab needs further investigation.

Strictly anaerobic bacteria were not found in our study, which have previously been isolated from other aquatic animals. It has been reported that the predominant bacteria isolated from the salmonid gut are aerobes or facultative anaerobes (Ringø et al. 1995). Meanwhile, our results showed that a few phylotypes affiliated with Gram-positive bacteria existed in the gut of mud crab, which is a little different from the study of Chinese mitten crab (Li et al. 2007a,b). Actually, we have isolated a Gram-positive Bacillus by traditional cultivation in LB medium, the preliminary experiment shows that the isolate could inhibit some of the pathogens from mud crab (unpublished results), which might be candidate for probiotics development in the future. The Gram-positive bacteria also existed in the intestine of shrimp by some other reports (Liu et al. 2011). The difference of intestinal microbial diversity between mud crab and other crustaceans needs further investigation.

Acknowledgements

This manuscript has been financially supported by China Natural Science Foundation (31172424), Guangdong Provincial Scientific and Technological Project (2012B020308007) and Shantou Municipal Scientific &Technological Project (2010-164).

Ancillary