- Email: [email protected]
Pathogenesis of acute hepatopancreatic necrosis disease (AHPND) in shrimp
Fish & Shellfish Immunology 47 (2015) 1006e1014
Contents lists available at ScienceDirect
Fish & Shellfish Immunology journal homepage: www.elsevier.com/locate/fsi
Full length article
Pathogenesis of acute hepatopancreatic necrosis disease (AHPND) in shrimp Hung-Chiao Lai a, Tze Hann Ng a, Masahiro Ando b, Chung-Te Lee c, I-Tung Chen c, Jie-Cheng Chuang d, Rapeepat Mavichak e, Sheng-Hsiung Chang a, Mi-De Yeh a, Yi-An Chiang a, Haruko Takeyama f, Hiro-o Hamaguchi g, Chu-Fang Lo c, Takashi Aoki a, **, Han-Ching Wang a, * a
Institute of Biotechnology, National Cheng Kung University, 701, Taiwan, ROC Consolidated Research Institute for Advanced Science and Medical Care, Waseda University, Tokyo 162-0041, Japan c Institute of Bioinformatics and Biosignal Transduction, National Cheng Kung University, Tainan 701, Taiwan, ROC d Sheng Long Bio-tech International Co., LTD, Long An, Viet Nam e Aquatic Animal Health Research Center, Charoen Pokphand Foods, Bangkok, Thailand f Department of Life Science and Medical Bioscience, Waseda University, 2-2 Wakamatsu-cho, Shinjuku-ku, Tokyo, Japan g College of Science, National Ciao Tung University, 300, Taiwan, ROC b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 July 2015 Received in revised form 26 October 2015 Accepted 2 November 2015 Available online 6 November 2015
Acute hepatopancreatic necrosis disease (AHPND), also called early mortality syndrome (EMS), is a recently emergent shrimp bacterial disease that has resulted in substantial economic losses since 2009. AHPND is known to be caused by strains of Vibrio parahaemolyticus that contain a unique virulence plasmid, but the pathology of the disease is still unclear. In this study, we show that AHPND-causing strains of V. parahaemolyticus secrete the plasmideencoded binary toxin PirABvp into the culture medium. We further determined that, after shrimp were challenged with AHPND-causing bacteria, the bacteria initially colonized the stomach, where they started to produce PirABvp toxin. At the same early time point (6 hpi), PirBvp toxin, but not PirAvp toxin, was detected in the hepatopancreas, and the characteristic histopathological signs of AHPND, including sloughing of the epithelial cells of the hepatopancreatic tubules, were also seen. Although some previous studies have found that both components of the binary PirABvp toxin are necessary to induce a toxic effect, our present results are consistent with other studies which have suggested that PirBvp alone may be sufficient to cause cellular damage. At later time points, the bacteria and PirAvp and PirBvp toxins were all detected in the hepatopancreas. We also show that Raman spectroscopy “Whole organism fingerprints” were unable to distinguish between AHPND-causing and non-AHPND causing strains. Lastly, by using minimum inhibitory concentrations, we found that both virulent and non-virulent V. parahaemolyticus strains were resistant to several antibiotics, suggesting that the use of antibiotics in shrimp culture should be more strictly regulated. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Acute hepatopancreatic necrosis disease (AHPND) Early mortality syndrome (EMS) Vibrio parahaemolyticus Shrimp
1. Introduction Acute hepatopancreatic necrosis disease (AHPND), or early mortality syndrome (EMS), is a bacterial disease which has
* Corresponding author. Institute of Biotechnology, College of Bioscience and Biotechnology, National Cheng Kung University, Tainan 701, Taiwan, ROC. ** Corresponding author. Institute of Biotechnology, College of Bioscience and Biotechnology, National Cheng Kung University, Tainan 701, Taiwan, ROC. E-mail addresses: [email protected] (T. Aoki), [email protected]. tw (H.-C. Wang). http://dx.doi.org/10.1016/j.fsi.2015.11.008 1050-4648/© 2015 Elsevier Ltd. All rights reserved.
seriously impacted the shrimp aquaculture industry, not just in Asia but also in central America (Mexico) [1e4]. AHPND was first described in China in 2009, and since then, this newly emergent disease has become widespread. It affects both Penaeus monodon and Litopenaeus vannamei [1,5]. Within the first 20e30 days after a pond is stocked with postlarvae, AHPND can cause up to 100% mortality [6]. An atrophied pale hepatopancreas is the most commonly observed sign of AHPND in diseased shrimp. Under the microscope, histological analysis reveals sloughing of the hepatopancreas tubule epithelial cells in the early stage of AHPND, while necrosis of the
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
hepatopancreas tubule epithelial cells and massive hemocytic infiltration are observed in the later stages of infection [4,5,7]. In 2013, the causative agent of AHPND, Vibrio parahaemolyticus, was first identified by Lightner and his team [5]. In addition to the two bacterial chromosomes in V. parahaemolyticus, a large extrachromosomal plasmid (~63e69 kbp) was found in all of the tested AHPND-causing V. parahaemolyticus strains, but was absent in the non-AHPND strains [8e10]. The AHPND-associated plasmid studied by our group was named pVA1 [11]. Like all the other instances of the AHPND-causing plasmid, pVA1 contained the PirAvp and PirBvp toxin genes, which express homologs to the Photorhabdus insect-related (Pir) binary toxin. A PCR diagnostic system for AHPND was developed based on the sequences of the AP1, AP2 and Pir (PirBvp) regions in pVA1 [11]. However, the pathogenesis of AHPND still needs to be clarified. In the present study, we characterized several AHPND-causing strains from Thailand and Vietnam. The in vitro expression of PirABvp and single-cell Raman spectral profiles of V. parahaemolyticus strains were also investigated. Finally, by using animal trials, we investigated the accumulation of AHPND-causing strains and the expression of PirAvp and PirBvp in shrimp target tissues after challenge. 2. Materials and methods 2.1. Bacterial strains used in this study A total of seven AHPND-causing and non-AHPND causing bacterial strains were used in this study. Three of the AHPND-causing strains (ThV-1, ThV-16 and 5HP) and two non-AHPND causing strains (ThN-2 and ThN-7) were isolated from Thailand. The AHPND-causing strain M1-1 and the non-AHPND causing strain M2-36 were both isolated from Vietnam. On arrival in the laboratory, all of the above strains were growing in tryptic soy broth (TSB) medium containing 2% NaCl. Subsequently, to maintain each of these strains, colonies were transferred to thiosulfate citrate bile salts sucrose (TCBS) agar and cultured at 30 C. To prepare the bacterial inoculum for the challenge experiments, the strains were once again cultured in TSB medium containing 2% NaCl as described below. 2.2. Experimental animals L. vannamei shrimp (1e2 g mean body weight) were purchased from the Aquatic Animal Center, National Taiwan Ocean University (NTOU). To assess the health of these shrimp, they were maintained in sterilized seawater (30 ppt) for 1e3 days at 25e27 C before being challenged by immersion. Before challenge, some shrimp were also tested for the presence of the AP1, AP2 and Pir sequences. In addition, in some of the experiments, batches of ~20 shrimp were maintained without treatment until the end of the experiment in order to monitor their survival. 2.3. Analysis of bacterial strains by Raman spectroscopy For each strain to be tested, three to four individual colonies were picked up from the culture plate and spread onto a glass slide. A 400-element grid was superimposed on the selected area of this bacterial smear, and Raman spectroscopy measurements were performed with a HeliumeNeon Laser (632.8 nm; HRP350-ECHeNe Laser; Thorlab, Inc.), an inverted microscope (ECLIPSE Ti, Nikon Corporation), a spectrometer (MS3504i, SOL instruments, Ltd., Minsk, Republic of Belarus, 1200 lines/mm) and a charge coupled device (CCD) camera (Newton DU920-M, Andor Technology plc., U.K.). A 100/NA 1.4 objective lens (Plan Apo VC, Nikon
1007
Corporation, Tokyo, Japan) was used to focus the laser beam on the bacterial smear. For each successive element of the 400-element grid, the backscattered Raman light was routed back through the same objective lens and directed into the cross slit of the spectrometer. Raman signals were successfully collected from approximately 200 of the 400 elements. These signals were averaged by IGOR Pro (WaveMetrics) and used to construct the Raman profiles. 2.4. Immersion challenge with AHPND-causing strains To determine the bacterial density that would produce a final CFU/ml of approximately 104, colonies of bacteria were transferred from the TCBS plates to TSB medium containing 2% NaCl, and incubated for 12e16 h at 30 C. The resulting bacterial cultures were then serially diluted, and for each dilution of the bacterial strain, the optical density (OD) at 600 nm was measured using a spectrophotometer. The CFU/ml of all of these cultures was then determined by serial dilution, plating onto TCBS agar, and subsequently counting the number of colonies on the plate. It was found that for all strains, the desired final value of ~104 CFU/ml was produced by a bacterial culture with an OD600 0.1. The challenge tests were conducted according to a slightly modified version of the bacteria immersion protocols described by Tran et al. [5]. Briefly, after each bacteria strain was cultured overnight in TSB medium containing 2% NaCl at 30 C, aliquots of the cultures were transferred into ~100 ml of fresh TSB medium plus 2% NaCl to a bacterial density of OD600 0.1. Each bacterial suspension was then mixed with 900 ml sea water to give an intermediate bacterial density of approximately 106 CFU/ml. The experimental shrimp (30e40 shrimp per group) were then immersed in this sea water for 15 min. Control groups were immersed in 900 ml sea water with 100 ml TSB medium containing 2% NaCl. The shrimp and 300 ml of the bacterial inoculum were then transferred into another tank containing 30 l of sea water (salinity: 30 ppt) to give a final bacterial density of approximately 104 CFU/ml. At 0, 6, 12, 18, 24, 48, 72 and 96 h post the initial bacteria immersion, stomach, hepatopancreas and hemolymph samples were taken from four of the shrimp in each group. 2.5. Preparation of total DNAs for AHPND diagnostic PCR Total DNA was isolated from shrimp stomach and hepatopancreas by using a DTAB/CTAB DNA extraction kit (GeneReach Biotechnology Corp.). For the bacteria, an individual colony was picked from each culture plate and subjected directly to AHPND diagnostic PCR without DNA extraction. 2.6. AHPND diagnostic PCR Diagnostic PCR for virulent AHPND-causing bacteria was carried out using the primer sets AP1F/AP1R and AP2F/AP2R as described by Lo and her team [11], and the primer set PirF/PirR, which was designed based on the PirBvp gene as it appears in the published genome of the pVA1 plasmid [9]. The primer set Vpara1F/Vpara2R was used to amplify the gyrB gene in V. parahaemolyticus [12]. Sequences of all the specific primer sets used in this study are listed in Table 1. PCR amplification with the bacterial colonies as templates followed a protocol of: denaturing at 94 C for 5 min; 30 cycles of 94 C for 30 s, 60 C for 30 s, 70 C for 1 min; plus a final 10 min extension at 72 C. For the shrimp tissue samples, after quantification of the extracted total DNA, 100 ng of each DNA was used as the PCR template. Except for the primer set PirF/PirR, PCR amplification with the other 3 primer sets was performed as described above. For the primer set PirF/PirR, since 30 cycles failed to produce a signal (data not shown), 40 cycles of amplification were used
1008
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
Table 1 PCR primers used in this study. Gene/sequence
Primer name
Primer sequence (50 e30 )
AP1F AP1R
50 - CCTTGGGTGTGCTTAGAGGATG -30 50 - GCAACCTATCGCGCAGAACACC -30
AP2F AP2R
50 - TCACCCGAATGCTGCTTGTGG -30 50 - CGTCGCTACTGTCTAGCTGAAG -30
PirF PirR
50 - GAGCCAGATATTGAAAACATTTGG -30 50 - CCACGCAGCGAGTTCTGTAATGTA -30
AP1
into each well and incubated for 1 h at room temperature. After developing the enzyme activity signal by using 100 ml/well TMB (3,30 ,5,50 -Tetramethylbenzidine) substrate (Sigma) for 15 min in the dark, the developing was stopped by adding the stop solution (1 N HCl) and the absorbance of each well was detected at a wavelength of 450 nm.
AP2
Pir
gyrB Vpara1F Vpara2R
0
0
5 - CGGCGTGGGTGTTTCGGTAGT -3 50 - TCCGCTTCGCGCTCATCAATA -30
instead. 2.7. Detection of the toxin proteins PirAvp and PirBvp in bacteria using western blotting Individual colonies of each bacterial strain were picked from the TCBS plate, added to 3 ml of TSB medium containing 2% NaCl, and cultured at 30 C overnight. The resulting bacterial broth was adjusted to OD600 1.5, diluted 200 with TSB medium containing 2% NaCl, and incubated again at 30 C for another 12 h. After centrifuging (8000 g) for 5 min at 4 C, the supernatant and bacteria cell pellet were collected. The pellet was dissolved in 200 ml PBS before Western blotting. Samples were mixed with SDS sample buffer, separated on 15% SDS-PAGE and transferred onto a PVDF membrane. Each membrane was then cut into two pieces so that top half contained the heavier bands (including PirBvp) and the bottom half contained the lighter bands (including PirAvp). After blocking with blocking buffer (2% BSA in TBST [50 mM TriseHCl, 500 mM NaCl, 0.2% Tween 20, pH 7.5]) for 1 h at room temperature, the lower and upper halves of the membranes were respectively incubated with primary antibodies against PirAvp and PirBvp [11] in blocking buffer for 1 h at room temperature. After washing twice with TBST, the membranes were incubated with secondary antibody conjugated with horseradish peroxidase (HRP) in TBST for 1 h at room temperature. After the signals were developed using a Western Lightning Plus-ECL reagent kit (Perkin Elmer), immunoblot signals were detected using an ImageQuant LAS 4000 mini (GE Healthcare Life Science). 2.8. Detection of the toxin proteins PirAvp and PirBvp in shrimp tissues and hemolymph using ELISA (enzyme-linked immunoassay) For this assay, the samples of shrimp stomach and hepatopancreatic tissue were homogenized in lysis-PBS buffer (3 diluted PBS) and centrifuged (10,000 g), after which the supernatants (protein lysates) were collected. The hemolymph samples were subjected to centrifugation only (8000 g), and again the supernatants were collected. A Bradford protein assay kit (Bio-Rad) was used to quantify the protein concentrations in the supernatants, and samples (0.75 mg total protein/well) were mixed with 100 ml 0.1 M carbonate/bicarbonate buffer, pH 9.6, and coated onto the microtiter wells of a 96-well ELISA plate for 12 h at 4 C. After removing uncoated proteins by three washes with PBST (0.2% Tween 20 in PBS), blocking buffer (2% BSA in PBST) was added into each well and the mixture was incubated for 1 h at room temperature. The wells were washed again and then incubated with the primary antibodies against PirAvp and PirBvp in blocking buffer for 1 h at room temperature. After washing with PBST, the secondary antibody conjugated with HRP in blocking buffer was then added
2.9. Histological observation of the shrimp hepatopancreas using H&E staining Shrimp used in this experiment were challenged by the same immersion procedure as described above. At 0, 6, 12, 18, 24, 48, 72 and 96 h after the initial bacterial immersion, three shrimp were taken from each group. The cephalothorax of these randomly selected samples were fixed with Davidson's fixative for 48 h and then sectioned and stained with hematoxylin and eosin using methods described by Tran et al. [5]. Microphotographs of the hepatopancreatic structures and AHPND lesions were taken by light microscopy. 2.10. Determination of minimal inhibitory concentration (MIC) In this study, determination of the MIC was performed by the agar plate dilution method modified from Andrews [13]. The following antibiotics were selected for this assay: ampicillin (ABPC), chloramphenicol (CP), kanamycin (KM), streptomycin (SM), tetracycline (TC), nalidixic acid (NA), sulfamethoxazole (SMZ), trimethoprim (TMP), enrofloxacin (ERFX), fosfomycin (FOM), ofloxacin (OFLX) and bicozamycin (BCM). All of the above antibiotics were originally effective either against all bacteria or against Gram-negative bacteria only [14]. However, the inappropriate use of antibiotics can result in the development of resistant bacterial strains, and we note that, among the antibiotics tested here, at least CP, SM, TC, SMZ, TMP, ERFX and OFLX are widely used in shrimp farming [15,16]. After the individual antibiotics were dissolved in their corresponding diluents, they were subjected to 2 serial dilution in ddH2O and added to molten Mueller-Hinton agar with final concentrations that ranged from 0.006 to 12.8 mg/ml. To prepare the bacterial inocula, 3e5 individual colonies of each bacterial strain were taken from TCBS plates and subcultured overnight at 30 C in TSB medium containing 2% NaCl. The resulting bacterial suspension was diluted 500 in TSB and 10 ml of the final bacterial inoculum was placed onto the surface of antibiotic agar plates. The bacterial inoculum was completely absorbed within 10e20 min, and the plates were then cultured overnight at 30 C. The MIC of each bacteria was defined as the concentration of antibiotic which completely inhibited bacterial growth. 3. Results 3.1. Characterization of the bacterial strains used in this study To confirm the presence of the AP1, AP2 and Pir sequences that are associated with the AHPND-causing strains of V. parahaemolyticus [11], PCR was performed to pre-screen the seven bacterial strains used in this study. The three virulent Thailand strains, ThV-1, ThV-16 and 5HP, and one of the Vietnam strains, M1-1, were positive for all three sequences (Fig. 1A). For the two non-AHPND-causing strains from Thailand, ThN-2 and ThN-7, PCR results were negative for AP1, AP2 and Pir. The non-virulent strain M2-36, which was isolated from a pond in Vietnam after an outbreak of AHPND, was positive for AP1 and AP2 but showed a negative result for the Pir amplicon. The Pir amplicon includes the toxin gene pirBvp, and as reported in a previous study [11], the
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
1009
Fig. 1. Characteristics of the AHPND and non-AHPND strains used in this study. (A) PCR detection of the AP1, AP2 and Pir sequences was performed on the indicated V. parahaemolyticus strains. Strains ThN-2 and ThN-7 do not contain the virulence plasmid and were used as a negative controls. Vpara is a primer set that is specific for the V. parahaemolyticus gene gyrB. (B) Western blotting analysis was used to detect PirAvp and PirBvp proteins in the culture medium supernatant (S) and the cell pellet (P) of the indicated strains after overnight culturing at 30 C. The white lines that separate each PVDF membrane into two halves indicate where the membranes were cut prior to incubation with the anti- PirBvp (upper half) and anti- PirAvp (lower half) antibodies.
absence of this gene from M2-36 explains why this strain fails to cause AHPND in shrimp. vp
3.2. Detection of PirAB blotting
toxin in culture medium using western
To investigate whether PirABvp toxin is released into the extracellular environment, the bacteria were incubated overnight at 30 C, and the culture medium and bacteria cell pellets were subjected to Western blotting with antibodies against PirAvp and PirBvp. As shown in Fig. 1B, PirAvp (~12 kDa) and PirBvp (~50 kDa) were detected in both the bacteria cell pellets and the culture medium of strains ThV-1, ThV-16, 5HP and M1-1. Neither PirAvp or PirBvp were detected in the bacteria cell pellets or culture medium of strains ThN-2, ThN-7 and M2-36. All of these results were consistent with the PCR result in Fig. 1A, and confirm that ThV-1, ThV-16, 5HP and M1-1 are virulent strains. In the infection pathway experiments described below, M1-1 and 5HP were used as the AHPND-causing strains.
3.3. Antimicrobial resistance profiles of the seven V. parahaemolyticus strains Evaluation of the antibiotic resistance was performed by detecting the minimal inhibitory concentrations (MICs) of 12 antimicrobial drugs (Table 2). While some antibiotics, such as CP, ERFX and OFLX, still seem to be universally effective, there also seems to be universal resistance to ABPC, SM, SMZ, FOM and BCM. The remaining antibiotics are effective against some strains but not against others.
3.4. The Raman spectral profiles of AHPND-causing strains Raman spectral analysis generates a profile of the organic molecules in a cell, and this method has recently been used to detect and identify bacterial strains via their unique “fingerprints” [17,18]. The Raman spectra of five of the bacterial strains in this study all
Table 2 Minimal inhibitory concentrations of selected antibiotics on V. parahaemolytics. Bacteria
ThV-1 ThV-16 ThN-2 ThN-7 5HP M1-1 M2-36
Antibioticsa (mg/ml) ABPC
CP
KM
SM
TC
NA
12.8 12.8 6.4 12.8 12.8 12.8 12.8
3.2 3.2 0.4 1.6 0.8 1.6 1.6
6.4 3.2 6.4 3.2 12.8 6.4 6.4
12.8 12.8 12.8 12.8 S12.8 12.8 6.4
0.4 3.2 0.4 12.8 0.4 12.8 12.8
12.8 1.6 0.8 1.6 0.8 0.8 0.8
Bacteria
Antibiotics (mg/ml) SMZ
TMP
ERFX
FOM
OFLX
BCM
ThV-1 ThV-16 ThN-2 ThN-7 5HP M1-1 M2-36
S12.8 12.8 6.4 S12.8 S12.8 12.8 12.8
3.2 6.4 1.6 6.4 12.8 3.2 3.2
0.4 0.1 0.1 0.2 0.1 0.1 0.1
S12.8 S12.8 S12.8 S12.8 S12.8 S12.8 S12.8
0.2 0.2 0.2 0.4 0.2 0.4 0.2
S12.8 S12.8 S12.8 S12.8 S12.8 S12.8 S12.8
a Antibiotics: ABPC (Ampicillin), CP (Chloramphenicol), KM (Kanamycin), SM (Streptomycin), TC (Tetracycline), NA (Nalidixic acid), SMZ (Sulfamethoxazole), TMP (Trimethoprim), ERFX (Enrofloxacin), FOM (Fosfomycin), OFLX (Ofloxacin), BCM (Bicozamycin).
1010
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
show a similar pattern (Fig. 2), which indicates a general similarity in the biochemical composition of the tested strains, and suggests that, without more sophisticated mathematical analysis, this method is not suitable for distinguishing between AHPND and nonAHPND causing strains of V. parahaemolyticus. 3.5. Location of AHPND-causing bacteria in shrimp challenged by immersion Shrimps were challenged by immersion in the bacterial strains M1-1 or 5HP or in the TSB control, and for each group, at each time point, the stomach and hepatopancreas were collected. Total DNA was extracted from these tissues, and PCR analysis was performed as described above with the primer sets for AP1, AP2 and Pir in order to detect the presence of AHPND-causing bacteria. In stomach, DNAs extracted from shrimp infected with 5HP gave positive PCR results with the AP1, AP2 and Pir primer sets from at 6e24 hpi through to the death of the challenged shrimp at 24 hpi (Fig. 3A). In general, weaker PCR signals were produced by the M11 strain, and shrimp challenged with this strain survived beyond the end of the experiment (96 hpi). We also note that at 96 hpi, only very weak signals were detected from the AP1, AP2 and Pir sequences. In the hepatopancreas (Fig. 3B), shrimp infected with 5HP gave positive PCR results after 12 hpi, while M1-1 results were almost entirely negative for the entire 96 h (Fig. 3B).
only at 24 hpi in the M1-1 treated group. Thereafter, even when the differences were statically significant, they still represented only small increases over the baseline control values (Fig. 4). In the hepatopancreas of shrimp challenged with 5HP, although the levels of PirAvp started to be elevated only at 18 hpi, PirBvp levels were significantly increased throughout the entire experiment from 6 to 24 hpi (Fig. 5). By contrast, in the M1-1 treated group, the statistically significant increases in the levels of PirAvp and PirBvp were generally smaller and did not appear until later (Fig. 5). Lastly, although significantly elevated levels of PirAvp and PirBvp were observed in the hemolymph at 24 hpi and 48 hpi, respectively (Fig. 6), in general there was little evidence that the toxin or its components were distributed via the hemolymph. 3.7. Histopathology of the hepatopancreas of shrimp challenged with the M1-1 and 5HP strains of V. parahaemolyticus Typical AHPND histopathology was observed in the hepatopancreas of the 5HP-treated group from 6 hpi to 24 hpi (Fig. 7). Almost all of the examined hepatopancreas sections of the 5HPinfected shrimp showed sloughing of the hepatopancreas tubule epithelial cells into the lumen. Hemocyte infiltration was also seen in the areas surrounding the necrotic hepatopancreas tubules. Some bacteria colonization was observed at 24 hpi. By contrast, none of these pathological effects were observed in the M1-1 group (Fig. 7).
3.6. Production and location of the PirABvp toxin
4. Discussion
Based on the absence of large numbers of bacteria in the hepatopancreas even when the characteristic AHPND lesions are clearly seen, it has been proposed [11] that the damage to the hepatopancreatic cells is caused by PirABvp toxin that has been released by AHPND-causing bacteria that are still in the stomach. To further explore this proposed mechanism, we investigated whether PirABvp toxin could be detected in the total protein lysates prepared from stomach, hepatopancreas and hemolymph. In stomach, ELISA assays with antibodies against PirAvp and PirBvp detected the presence of elevated levels of both PirAvp and PirBvp in the 5HP treated group starting at 6 hpi, and continuing through to the death of the shrimp at 24 hpi (Fig. 4). Meanwhile, significant increases in the levels of PirAvp and PirBvp first occurred
V. parahaemolyticus is ubiquitous in the marine environment, and although it is usually only an opportunistic pathogen, the AHPND-causing strains of V. parahaemolyticus are uniquely virulent to shrimp because they contain a large plasmid that encodes the binary insecticidal toxin homologs PirAvp and PirBvp [9,10]. The origin of this plasmid and the toxin genes is still an open question. One intriguing possibility is suggested by the inconsistent results for the TSB group in Fig. 3A. Unexpectedly, in the TSB-treated control group, a number of positive PCR results were found with the AP1 and AP2 primer sets and also with the Pir primer set. The reason for these surprising results is unclear, but it is possible that some unidentified bacteria e probably not V. parahaemolyticus because our initial PCR analysis with the Vpara primer set was negative (data not shown) e may have been present in the stomachs of some of the putatively healthy experimental shrimp. Since we already know that in some bacterial species, during recombination events, entire plasmids or mobile genetic elements within the chromosomal or plasmidic genome significantly contribute to changes in genes, including toxin genes [19,20], it seems likely that a series of recombination events involving plasmids and/or mobile genetic elements may also have created the AHPND virulence plasmid that is now found in V. parahaemolyticus. It also seems possible that these events might have taken place in the stomach of a shrimp. If so, then the positive AP1, AP2 and Pir signals in the stomachs of the TSB-only shrimp might simply be a reflection of these potential events. The transfer of R (resistance) plasmids and mobile genetic elements during recombination events is also used by bacteria to achieve antibiotic resistance [21,22]. Unfortunately, the increased selective pressure caused by the widespread use of antibiotics in shrimp aquaculture further serves to drive and enhance this bacterial antibiotic resistance [23,24]. In this study, the antibiotic resistance of AHPND-causing and non-AHPND causing strains was evaluated for the first time (Table 2). It is alarming that all of the tested bacterial strains were resistant to some of the antibiotics (Table 2). This suggests that increased resistance to antibiotics is
Fig. 2. Raman spectra of single bacterial cells of the indicated strains recorded with 635.8 nm excitation. Each spectrum represents the mean “fingerprint” of approximately 200 bacterial cells.
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
1011
Fig. 3. PCR detection of the indicated AHPND-causing strains in shrimp (A) stomach (Stm) and (B) hepatopancreas (Hep). Shrimp were challenged with the M1-1 and 5HP strains by immersion, and tissues were collected from 4 individual animals at the indicated time points. (Only two animals were collected at 24 hpi in the 5HP group because only two animals were still alive.) After total DNAs were extracted, PCR detection of the AP1, AP2 and Pir sequences was performed. Shrimp treated with TSB medium were used as controls.
already becoming a serious problem. Instead of the unregulated use of antibiotics, we therefore recommend the establishment and maintenance of properly managed biosecure systems as the way forward for an eco-friendly, sustainable shrimp aquaculture industry. The Raman spectroscopy data that was used in this study to generate molecular “whole-organism fingerprints” [25e27] also provided information on the molecular content of AHPND-causing bacteria cells. The Raman spectra in Fig. 2 show bands representing the major cell components, including proteins (1003 cm1 and 1659.9 cm1), nucleic acids (782.9 cm1), lipids (1253.7 cm1) and carbohydrates/lipids (1449.5 cm1) [27,28]. We observed a slight increase in the intensity of the Raman band of the AHPND-causing strain 5HP located between 850 and 950 cm1, although the
significance of this feature, if any, still needs to be clarified. Our main result, however, is negative: although all the AHPND-causing strains contain a large unique plasmid, the presence of this plasmid failed to produce any correspondingly unique Raman spectra features that could be used to distinguish between AHPND-causing strains versus non-AHPND-causing strains (Fig. 2). It will be interesting to investigate whether or not useful differences between the AHPND-causing and non-AHPND-causing strains will be revealed by other techniques with higher sensitivity, such as proteomics and metabolomics. The results obtained in the present study suggest that the pathology of the AHPND-causing strain 5HP is consistent with our previous proposal that the characteristic sloughing of epithelial cells into the hepatopancreatic tubules is due to secreted toxins
1012
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014 6 hpi
***
PirAvp
Relative expression
3.5 3 2.5
***
3 2.5
***
***
3 2.5
**
***
3 2.5
***
***
72 hpi
48 hpi
24 hpi 3.5
3.5
**
*
3 2.5
96 hpi 3.5
3.5 3
3
2.5
2.5
2
2
2
2
2
2
2
1.5
1.5
1.5
1.5
1.5
1.5
1
1
1
1
1
1
1
0.5
0.5
0.5
0.5
0.5
0.5
0.5
TSB
M1-1
***
3.5
PirBvp
***
18 hpi 3.5
1.5
0
Relative expression
12 hpi 3.5
3 2.5
5HP
0
TSB
***
3.5
***
M1-1
3 2.5
5HP
0
M1-1
TSB
**
3.5
***
**
3 2.5
0
5HP
M1-1
TSB
3.5
***
3
**
2.5
5HP
***
0
TSB
M1-1
3.5
*
0
M1-1
TSB
0
3
3
2.5
2.5
2
2
2
2
2
2
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1
1
1
1
1
1
1
0.5
0.5
0.5
0.5
0.5
0.5
0.5
TSB
M1-1
5HP
0
TSB
M1-1
5HP
0
TSB
M1-1
0
5HP
TSB
M1-1
5HP
0
TSB
M1-1
0
TSB
M1-1
3.5
3.5
3 2.5
2
0
*
TSB
M1-1
0
**
TSB
M1-1
Tissue: Stm
Fig. 4. Detection of toxins PirAVP and PirBVP in stomach by ELISA assay. The samples used in this experiment were taken from the same individual shrimps as those used in Fig. 3. Each bar represents the mean ± SD. Asterisks indicate statistically significant differences between the respective groups (Student's t test, *p < 0.05, **p < 0.01, ***p < 0.001).
6 hpi
12 hpi
PirAvp
Relative expression
*
3 2.5
3
3
2.5
2.5
***
***
48 hpi
24 hpi
**
3.5
**
3
3.5
**
2.5
72 hpi
*
3
3
2.5
2.5
96 hpi 3.5
3.5
***
3 2.5
2
2
2
2
2
2
2
1.5
1.5
1.5
1.5
1.5
1.5
1
1
1
1
1
1
1
0.5
0.5 TSB
M1-1
**
7
PirBvp
3.5
1.5
0
Relative expression
18 hpi
3.5
3.5
6
5HP
0
***
7
**
M1-1
5HP
0
7
*
TSB
**
***
0
5HP
M1-1
***
TSB
M1-1
7
*
**
5HP
0
M1-1
0
TSB
M1-1
7
7
***
0.5
0.5 TSB
*
6
0
6
5
5
4
4
4
4
4
3
3
3
3
3
3
3
2
2
2
2
2
2
2
5
4
1
1 0
TSB
M1-1
5HP
0
1 TSB
M1-1
5HP
0
0
5HP
5
1
1 M1-1
TSB
5
TSB
M1-1
5HP
0
M1-1
0
M1-1
6 5
***
4
1 TSB
TSB
7
*
6
6
5
6
0.5
0.5
0.5 TSB
***
1 TSB
M1-1
0
TSB
M1-1
Tissue: Hep
Fig. 5. Detection of toxins PirAVP and PirBVP in hepatopancreas (Hep) by ELISA assay. The samples used in this experiment were taken from the same individual shrimps as those used in Fig. 3. Each bar represents the mean ± SD. Asterisks indicate statistically significant differences between the respective groups (Student's t test, *p < 0.05, **p < 0.01, ***p < 0.001).
6 hpi
*
PirAvp
Relative expression
***
24 hpi
*
*
48 hpi
72 hpi
96 hpi
** 1.75
1.75
1.5
1.5
1.5
1.5
1.5
1.5
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.75
1.75
1.75
1.75
1
1
1
1
1
1
1
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.25
0.25 0
PirBvp
18 hpi
1.5
1.75
Relative expression
12 hpi
TSB
M1-1
5HP
0
M1-1
5HP
*
0
1.75
1.75
1.5
1.5
1.5
1.25
1.25
1.25
0.25
0.25
0.25 TSB
TSB
1.75
M1-1
*
***
5HP
***
0
TSB
1.75
M1-1
0
0
TSB
M1-1
1.25
0
1.5
1.5
1.25
1.25
1
1
1
1
1
1
1
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0
TSB
M1-1
5HP
0
TSB
M1-1
5HP
0
TSB
M1-1
5HP
0
TSB
M1-1
5HP
0
TSB
M1-1
0
TSB
M1-1
1.75
1.75
*
1.5
1.25
0.25
0.25 M1-1
TSB
1.75
*
*
1.5
5HP
TSB
M1-1
0
TSB
M1-1
Tissue: Hlm
Fig. 6. Detection of toxins PirAVP and PirBVP in hemolymph (Hlm) by ELISA assay. The samples used in this experiment were taken from the same individual shrimps as those used in Fig. 3. Each bar represents the mean ± SD. Asterisks indicate statistically significant differences between the respective groups (Student's t test, *p < 0.05, **p < 0.01, ***p < 0.001).
rather than the presence of the bacteria itself [11]. Thus, our PCR results show that 5HP initially colonizes the stomach at 6 hpi and only later reaches the hepatopancreas (12 hpi; Fig. 3). Meanwhile,
the PirAvp and PirBvp toxins were also detected in the stomach of the 5HP-challenged shrimp at a similarly early time point (6 hpi; Fig. 4). If the PirABvp toxin functions as a pore-forming toxin (as we
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
1013
Fig. 7. H&E stained photomicrographs of hepatopancreas collected from TSB-treated, M1-1 infected, and 5HP-infected groups at 0e96 hpi. Normal hepatopancreas tubules are seen in the TSB-treated and M1-1-infected groups, while typical AHPND lesions with necrotic and sloughed epithelial cells (black arrows) are seen from 6 hpi in the 5HP-infected group. From 6 to 18 hpi in the 5HP group, the small dark dots are hemocytes that have infiltrated between the cells of the hepatopancreas tubules. At 24 hpi, significant bacterial colonization (blue arrows) was also observed in the lumen of the hepatopancreas tubules. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
proposed in a previous publication [11]), it may damage the shrimp stomach epithelium in the same way that the original Pir (Photorhabdus insect related) binary toxin, PirAB, causes swelling and shedding of the epithelial cells in the digestive organs of a range of insects [29e31]. This putative damage to the stomach would then allow both the toxin proteins and, subsequently, the bacteria themselves to spread from the stomach to the hepatopancreas. Sure enough, in the 5HP-challenged shrimp, Fig. 5 shows elevated levels of PirBvp in the hepatopancreas as early as 6 hpi, and at this time there was already sloughing of the hepatopancreatic tubule epithelial cells (Fig. 7). Interestingly, preliminary experiments suggest that infection with 5HP induces significant upregulation of
several immune-related genes (Toll, My88, ALF, PEN2, PEN3 and PEN4) in the hepatopancreas at this time, whereas by contrast, expression of these same genes is suppressed in the stomach and hemocytes (data not shown). The bacteria themselves, on the other hand, were only detected in the hepatopancreas at 12 hpi (Fig. 3). Presumably at this time, the PirABvp toxin is being produced by the 5HP bacteria in both the stomach and the hepatopancreas and released into both of these organs directly. We note, however, that surprisingly, increased levels of PirAvp were not seen in the hepatopancreas until 18 hpi (Fig. 5). The reason for this is unclear, but previous studies on PirB from other (insecticidal) bacteria as well as our own work on PirBvp suggest that under some circumstances
1014
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
recombinant PirB toxin alone might be sufficient to induce cell damage and mortality in the host [11,32]. It is interesting to note that, although both 5HP and M1-1 were able to secrete the PirAvp and PirBvp toxins in vitro (Fig. 1), neither toxin was detected in the hemolymph except between 24 and 48 hpi (Fig. 6). This suggests that these toxins are not primarily being spread via the shrimp circulatory system, and that they are therefore probably unlikely to cause systemic damage or damage to other remote tissues or organs. In contrast to the 5HP strain, shrimp that were challenged with the M1-1 strain survived through to the end of the experiment (Figs. 3e6) and showed none of the characteristic symptoms of AHPND (Fig. 7). Nevertheless, Fig. 1 suggests that M1-1 was able to produce similar amounts of toxin to 5HP, and we also found that M1-1 caused the hepatopancreas to become pale (which is another clinical sign of AHPND) in postlarvae (data not shown). The reason for M1-1's relatively weak pathogenicity in the 3 g shrimp used here is presently unclear, However, the weaker M1-1 PCR signals from stomach DNA and the almost complete absence of signals from the hepatopancreas (Fig. 3) leads us to tentatively conclude that a relatively slow replication of strain M1-1 in stomach may be the explanation for its weaker pathogenicity. A slower replication rate would also explain why M1-1 generally manages to produce less PirAvp and PirBvp than 5HP in the stomach and hepatopancreas (Figs. 4 and 5). However, we continue to work on this question, and we are presently using a next generation sequencing platform to sequence the genomic DNA and plasmid DNAs of strains M1-1 and 5HP. In the near future, by searching for specific genes/sequences that exist only in M1-1 or 5HP, we hope to better understand on a molecular level the underlying mechanisms that give rise to these differences. Acknowledgments This study was supported financially by the Ministry of Science and Technology (MOST 103-2633-B-006-004 and MOST 104-2622B-006-009 -CC1). One of the Vibrio parahaemolyticus strains, 5HP, was originally provided by Dr. Timothy W. Flegel, Dr. Kallaya Sritunyalucksana, and Dr. Siripong Thitamadee. We would also like to thank Paul Barlow for his helpful criticism of the manuscript. References [1] D.V. Lightner, R.M. Redman, C.R. Pantoja, B.L. Noble, L.H. Tran, Early mortality syndrome affects shrimp in Asia, Glob. Aquac. Advocate 40 (2012). [2] T.W. Flegel, Historic emergence, impact and current status of shrimp pathogens in Asia, J. Invertebr. Pathol. 110 (2012) 166e173. ~ o, C.V. Mohan, Early mortality syndrome threatens Asia's shrimp [3] E.M. Lean farms, Glob. Aquac. Advocate (2012) 38e39. [4] S.A. Soto-Rodriguez, B. Gomez-Gil, R. Lozano-Olvera, M. Betancourt-Lozano, M.S. Morales-Covarrubias, Field and experimental evidence of Vibrio parahaemolyticus as the causative agent of acute hepatopancreatic necrosis disease (AHPND) of cultured shrimp (Litopenaeus vannamei) in northwestern Mexico, Appl. Environ. Microbiol. 81 (2015) 1689e1699. [5] L. Tran, L. Nunan, R.M. Redman, L.L. Mohney, C.R. Pantoja, K. Fitzsimmons, D.V. Lightner, Determination of the infectious nature of the agent of acute hepatopancreatic necrosis syndrome affecting penaeid shrimp, Dis. Aquat. Organ. 105 (2013) 45e55. [6] P. De Schryver, T. Defoirdt, P. Sorgeloos, Early mortality syndrome outbreaks: a microbial management issue in shrimp farming? PLoS Pathog. 10 (2014) e1003919. [7] L. Nunan, D. Lightner, C. Pantoja, S. Gomez-Jimenez, Detection of acute hepatopancreatic necrosis disease (AHPND) in Mexico, Dis. Aquat. Organ 111 (2014) 81e86. [8] H. Kondo, S. Tinwongger, P. Proespraiwong, R. Mavichak, S. Unajak, R. Nozaki, I. Hirono, Draft genome sequences of six strains of Vibrio parahaemolyticus isolated from early mortality syndrome/acute hepatopancreatic necrosis
disease shrimp in Thailand, Genome Announc. 2 (2014) e00221e14. [9] Y.T. Yang, I.T. Chen, C.T. Lee, C.Y. Chen, S.S. Lin, L.I. Hor, T.C. Tseng, Y.T. Huang, K. Sritunyalucksana, S. Thitamadee, H.C. Wang, C.F. Lo, Draft genome sequences of four strains of Vibrio parahaemolyticus, three of which cause early mortality syndrome/acute hepatopancreatic necrosis disease in shrimp in China and Thailand, Genome Announc. 2 (2014) e00816e14. [10] J.E. Han, K.F. Tang, L.H. Tran, D.V. Lightner, Photorhabdus insect-related (Pir) toxin-like genes in a plasmid of Vibrio parahaemolyticus, the causative agent of acute hepatopancreatic necrosis disease (AHPND) of shrimp, Dis. Aquat. Organ. 113 (2015) 33e40. [11] C.T. Lee, I.T. Chen, Y.T. Yang, T.P. Ko, Y.T. Huang, J.Y. Huang, M.F. Huang, S.J. Lin, C.Y. Chen, S.S. Lin, D. Lightner, H.C. Wang, H.J. Wang, H.C. Wang, L.I. Hor, C.F. Lo, Vibrio parahaemolyticus: an opportunistic marine pathogen becomes virulent by acquiring a plasmid that expresses a deadly toxin, Proc. Natl. Acad. Sci. U. S. A. 112 (2015) (2015) 10798e10803. [12] X.Y. Luan, J.X. Chen, X.H. Zhang, J.T. Jia, F.R. Sun, Y. Li, Comparison of different primers for rapid detection of Vibrio parahaemolyticus using the polymerase chain reaction, Lett. Appl. Microbiol. 44 (2007) 242e247. [13] J.M. Andrews, Determination of minimum inhibitory concentrations, J. Antimicrob. Chemother. 48 (2001) 5e16. [14] M.A. Kohanski, D.J. Dwyer, J.J. Collins, How antibiotics kill bacteria: from targets to networks, Nat. Rev. Microbiol. 8 (2010) 423e435. [15] K. Kümmerer, Antibiotics in the aquatic environmentea reviewepart I, Chemosphere 75 (2009) 417e434. € m, S. Gr€ € m, S. Poungshompoo, B.E. Bengtsson, [16] K. Holmstro aslund, A. Wahlstro N. Kautsky, Antibiotic use in shrimp farming and implications for environmental impacts and human health, IJFST 38 (2003) 255e266. [17] J.C. Baritaux, E. Schultz, A.C. Simon, A.G. Bourdat, I. Espagnon, P. Laurent, J.M. Dinten, Fast and robust identification of single bacteria in environmental matrices by Raman spectroscopy, in: SPIE BiOS. International Society for Optics and Photonics, 2015, 93281Ie93281I. [18] X. Wu, Y.W. Huang, B. Park, R.A. Tripp, Y. Zhao, Differentiation and classification of bacteria using vancomycin functionalized silver nanorods array based surface-enhanced Raman spectroscopy and chemometric analysis, Talanta 139 (2015) 96e103. [19] L.S. Frost, R. Leplae, A.O. Summers, A. Toussaint, Mobile genetic elements: the agents of open source evolution, Nat. Rev. Microbiol. 3 (2005) 722e732. [20] K.K. Hill, G. Xie, B.T. Foley, T.J. Smith, A.C. Munk, D. Bruce, L.A. Smith, T.S. Brettin, J.C. Detter, Recombination and insertion events involving the botulinum neurotoxin complex genes in Clostridium botulinum types A, B, E and F and Clostridium butyricum type E strains, BMC Biol. 7 (2009) 66. ~ a, Antibiotic resistance of bacteria from shrimp [21] E.A. Tendencia, L.D. de la Pen ponds, Aquaculture 195 (2001) 193e204. [22] P.M. Bennett, Plasmid encoded antibiotic resistance: acquisition and transfer of antibiotic resistance genes in bacteria, Br. J. Pharmacol. 153 (2008) S347eS357. [23] B.H. Normark, S. Normark, Evolution and spread of antibiotic resistance, J. Intern. Med. 252 (2002) 91e106. n-Aguin ~ aga, P.R. Jensen, Multilocus sequence typing reveals [24] K.C. Freel, N. Milla evidence of homologous recombination linked to antibiotic resistance in the genus Salinispora, Appl. Environ. Microbiol. 79 (2013) 5997e6005. [25] C. Kirschner, K. Maquelin, P. Pina, N.A. Ngo Thi, L.P. Choo-Smith, G.D. Sockalingum, C. Sandt, D. Ami, F. Orsini, S.M. Doglia, P. Allouch, M. Mainfait, G.J. Puppels, D. Naumann, Classification and identification of enterococci: a comparative phenotypic, genotypic, and vibrational spectroscopic study, J. Clin. Microbiol. 39 (2001) 1763e1770. [26] J.F. Almarashi, N. Kapel, T.S. Wilkinson, H.H. Telle, Raman spectroscopy of bacterial species and strains cultivated under reproducible conditions, J. Spectrosc. 27 (2012) 361e365. [27] K. Hamasha, Q.I. Mohaidat, R.A. Putnam, R.C. Woodman, S. Palchaudhuri, S.J. Rehse, Sensitive and specific discrimination of pathogenic and nonpathogenic Escherichia coli using Raman spectroscopy-a comparison of two multivariate analysis techniques, Biomed. Opt. Express 4 (2013) 481e489. [28] E. Staniszewska-Slezak, K. Malek, M. Baranska, Complementary analysis of tissue homogenates composition obtained by Vis and NIR laser excitations and Raman spectroscopy, Spectrochim. Acta A Mol. Biomol. Spectrosc. 147 (2015) 245e256. [29] N. Waterfield, S.G. Kamita, B.D. Hammock, R. Ffrench-Constant, The photorhabdus Pir toxins are similar to a developmentally regulated insect protein but show no juvenile hormone esterase activity, FEMS Microbiol. Lett. 245 (2005) 47e52. [30] M.B. Blackburn, R.R. Farrar, N.G. Novak, S.D. Lawrence, Remarkable susceptibility of the diamondback moth (Plutella xylostella) to ingestion of Pir toxins from Photorhabdus luminescens, Entomol. Exp. Appl. 121 (2006) 31e37. [31] Y. Li, X. Hu, X. Zhang, Z. Liu, X. Ding, L. Xia, S. Hu, Photorhabdus luminescens PirAB-fusion protein exhibits both cytotoxicity and insecticidal activity, FEMS Microbiol. Lett. 356 (2014) 23e31. [32] A. Ahantarig, N. Chantawat, N.R. Waterfield, R. ffrench-Constant, P. Kittayapong, PirAB toxin from Photorhabdus asymbiotica as a larvicide against dengue vectors, Appl. Environ. Microbiol. 75 (2009) 4627e4629.
Contents lists available at ScienceDirect
Fish & Shellfish Immunology journal homepage: www.elsevier.com/locate/fsi
Full length article
Pathogenesis of acute hepatopancreatic necrosis disease (AHPND) in shrimp Hung-Chiao Lai a, Tze Hann Ng a, Masahiro Ando b, Chung-Te Lee c, I-Tung Chen c, Jie-Cheng Chuang d, Rapeepat Mavichak e, Sheng-Hsiung Chang a, Mi-De Yeh a, Yi-An Chiang a, Haruko Takeyama f, Hiro-o Hamaguchi g, Chu-Fang Lo c, Takashi Aoki a, **, Han-Ching Wang a, * a
Institute of Biotechnology, National Cheng Kung University, 701, Taiwan, ROC Consolidated Research Institute for Advanced Science and Medical Care, Waseda University, Tokyo 162-0041, Japan c Institute of Bioinformatics and Biosignal Transduction, National Cheng Kung University, Tainan 701, Taiwan, ROC d Sheng Long Bio-tech International Co., LTD, Long An, Viet Nam e Aquatic Animal Health Research Center, Charoen Pokphand Foods, Bangkok, Thailand f Department of Life Science and Medical Bioscience, Waseda University, 2-2 Wakamatsu-cho, Shinjuku-ku, Tokyo, Japan g College of Science, National Ciao Tung University, 300, Taiwan, ROC b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 July 2015 Received in revised form 26 October 2015 Accepted 2 November 2015 Available online 6 November 2015
Acute hepatopancreatic necrosis disease (AHPND), also called early mortality syndrome (EMS), is a recently emergent shrimp bacterial disease that has resulted in substantial economic losses since 2009. AHPND is known to be caused by strains of Vibrio parahaemolyticus that contain a unique virulence plasmid, but the pathology of the disease is still unclear. In this study, we show that AHPND-causing strains of V. parahaemolyticus secrete the plasmideencoded binary toxin PirABvp into the culture medium. We further determined that, after shrimp were challenged with AHPND-causing bacteria, the bacteria initially colonized the stomach, where they started to produce PirABvp toxin. At the same early time point (6 hpi), PirBvp toxin, but not PirAvp toxin, was detected in the hepatopancreas, and the characteristic histopathological signs of AHPND, including sloughing of the epithelial cells of the hepatopancreatic tubules, were also seen. Although some previous studies have found that both components of the binary PirABvp toxin are necessary to induce a toxic effect, our present results are consistent with other studies which have suggested that PirBvp alone may be sufficient to cause cellular damage. At later time points, the bacteria and PirAvp and PirBvp toxins were all detected in the hepatopancreas. We also show that Raman spectroscopy “Whole organism fingerprints” were unable to distinguish between AHPND-causing and non-AHPND causing strains. Lastly, by using minimum inhibitory concentrations, we found that both virulent and non-virulent V. parahaemolyticus strains were resistant to several antibiotics, suggesting that the use of antibiotics in shrimp culture should be more strictly regulated. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Acute hepatopancreatic necrosis disease (AHPND) Early mortality syndrome (EMS) Vibrio parahaemolyticus Shrimp
1. Introduction Acute hepatopancreatic necrosis disease (AHPND), or early mortality syndrome (EMS), is a bacterial disease which has
* Corresponding author. Institute of Biotechnology, College of Bioscience and Biotechnology, National Cheng Kung University, Tainan 701, Taiwan, ROC. ** Corresponding author. Institute of Biotechnology, College of Bioscience and Biotechnology, National Cheng Kung University, Tainan 701, Taiwan, ROC. E-mail addresses: [email protected] (T. Aoki), [email protected]. tw (H.-C. Wang). http://dx.doi.org/10.1016/j.fsi.2015.11.008 1050-4648/© 2015 Elsevier Ltd. All rights reserved.
seriously impacted the shrimp aquaculture industry, not just in Asia but also in central America (Mexico) [1e4]. AHPND was first described in China in 2009, and since then, this newly emergent disease has become widespread. It affects both Penaeus monodon and Litopenaeus vannamei [1,5]. Within the first 20e30 days after a pond is stocked with postlarvae, AHPND can cause up to 100% mortality [6]. An atrophied pale hepatopancreas is the most commonly observed sign of AHPND in diseased shrimp. Under the microscope, histological analysis reveals sloughing of the hepatopancreas tubule epithelial cells in the early stage of AHPND, while necrosis of the
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
hepatopancreas tubule epithelial cells and massive hemocytic infiltration are observed in the later stages of infection [4,5,7]. In 2013, the causative agent of AHPND, Vibrio parahaemolyticus, was first identified by Lightner and his team [5]. In addition to the two bacterial chromosomes in V. parahaemolyticus, a large extrachromosomal plasmid (~63e69 kbp) was found in all of the tested AHPND-causing V. parahaemolyticus strains, but was absent in the non-AHPND strains [8e10]. The AHPND-associated plasmid studied by our group was named pVA1 [11]. Like all the other instances of the AHPND-causing plasmid, pVA1 contained the PirAvp and PirBvp toxin genes, which express homologs to the Photorhabdus insect-related (Pir) binary toxin. A PCR diagnostic system for AHPND was developed based on the sequences of the AP1, AP2 and Pir (PirBvp) regions in pVA1 [11]. However, the pathogenesis of AHPND still needs to be clarified. In the present study, we characterized several AHPND-causing strains from Thailand and Vietnam. The in vitro expression of PirABvp and single-cell Raman spectral profiles of V. parahaemolyticus strains were also investigated. Finally, by using animal trials, we investigated the accumulation of AHPND-causing strains and the expression of PirAvp and PirBvp in shrimp target tissues after challenge. 2. Materials and methods 2.1. Bacterial strains used in this study A total of seven AHPND-causing and non-AHPND causing bacterial strains were used in this study. Three of the AHPND-causing strains (ThV-1, ThV-16 and 5HP) and two non-AHPND causing strains (ThN-2 and ThN-7) were isolated from Thailand. The AHPND-causing strain M1-1 and the non-AHPND causing strain M2-36 were both isolated from Vietnam. On arrival in the laboratory, all of the above strains were growing in tryptic soy broth (TSB) medium containing 2% NaCl. Subsequently, to maintain each of these strains, colonies were transferred to thiosulfate citrate bile salts sucrose (TCBS) agar and cultured at 30 C. To prepare the bacterial inoculum for the challenge experiments, the strains were once again cultured in TSB medium containing 2% NaCl as described below. 2.2. Experimental animals L. vannamei shrimp (1e2 g mean body weight) were purchased from the Aquatic Animal Center, National Taiwan Ocean University (NTOU). To assess the health of these shrimp, they were maintained in sterilized seawater (30 ppt) for 1e3 days at 25e27 C before being challenged by immersion. Before challenge, some shrimp were also tested for the presence of the AP1, AP2 and Pir sequences. In addition, in some of the experiments, batches of ~20 shrimp were maintained without treatment until the end of the experiment in order to monitor their survival. 2.3. Analysis of bacterial strains by Raman spectroscopy For each strain to be tested, three to four individual colonies were picked up from the culture plate and spread onto a glass slide. A 400-element grid was superimposed on the selected area of this bacterial smear, and Raman spectroscopy measurements were performed with a HeliumeNeon Laser (632.8 nm; HRP350-ECHeNe Laser; Thorlab, Inc.), an inverted microscope (ECLIPSE Ti, Nikon Corporation), a spectrometer (MS3504i, SOL instruments, Ltd., Minsk, Republic of Belarus, 1200 lines/mm) and a charge coupled device (CCD) camera (Newton DU920-M, Andor Technology plc., U.K.). A 100/NA 1.4 objective lens (Plan Apo VC, Nikon
1007
Corporation, Tokyo, Japan) was used to focus the laser beam on the bacterial smear. For each successive element of the 400-element grid, the backscattered Raman light was routed back through the same objective lens and directed into the cross slit of the spectrometer. Raman signals were successfully collected from approximately 200 of the 400 elements. These signals were averaged by IGOR Pro (WaveMetrics) and used to construct the Raman profiles. 2.4. Immersion challenge with AHPND-causing strains To determine the bacterial density that would produce a final CFU/ml of approximately 104, colonies of bacteria were transferred from the TCBS plates to TSB medium containing 2% NaCl, and incubated for 12e16 h at 30 C. The resulting bacterial cultures were then serially diluted, and for each dilution of the bacterial strain, the optical density (OD) at 600 nm was measured using a spectrophotometer. The CFU/ml of all of these cultures was then determined by serial dilution, plating onto TCBS agar, and subsequently counting the number of colonies on the plate. It was found that for all strains, the desired final value of ~104 CFU/ml was produced by a bacterial culture with an OD600 0.1. The challenge tests were conducted according to a slightly modified version of the bacteria immersion protocols described by Tran et al. [5]. Briefly, after each bacteria strain was cultured overnight in TSB medium containing 2% NaCl at 30 C, aliquots of the cultures were transferred into ~100 ml of fresh TSB medium plus 2% NaCl to a bacterial density of OD600 0.1. Each bacterial suspension was then mixed with 900 ml sea water to give an intermediate bacterial density of approximately 106 CFU/ml. The experimental shrimp (30e40 shrimp per group) were then immersed in this sea water for 15 min. Control groups were immersed in 900 ml sea water with 100 ml TSB medium containing 2% NaCl. The shrimp and 300 ml of the bacterial inoculum were then transferred into another tank containing 30 l of sea water (salinity: 30 ppt) to give a final bacterial density of approximately 104 CFU/ml. At 0, 6, 12, 18, 24, 48, 72 and 96 h post the initial bacteria immersion, stomach, hepatopancreas and hemolymph samples were taken from four of the shrimp in each group. 2.5. Preparation of total DNAs for AHPND diagnostic PCR Total DNA was isolated from shrimp stomach and hepatopancreas by using a DTAB/CTAB DNA extraction kit (GeneReach Biotechnology Corp.). For the bacteria, an individual colony was picked from each culture plate and subjected directly to AHPND diagnostic PCR without DNA extraction. 2.6. AHPND diagnostic PCR Diagnostic PCR for virulent AHPND-causing bacteria was carried out using the primer sets AP1F/AP1R and AP2F/AP2R as described by Lo and her team [11], and the primer set PirF/PirR, which was designed based on the PirBvp gene as it appears in the published genome of the pVA1 plasmid [9]. The primer set Vpara1F/Vpara2R was used to amplify the gyrB gene in V. parahaemolyticus [12]. Sequences of all the specific primer sets used in this study are listed in Table 1. PCR amplification with the bacterial colonies as templates followed a protocol of: denaturing at 94 C for 5 min; 30 cycles of 94 C for 30 s, 60 C for 30 s, 70 C for 1 min; plus a final 10 min extension at 72 C. For the shrimp tissue samples, after quantification of the extracted total DNA, 100 ng of each DNA was used as the PCR template. Except for the primer set PirF/PirR, PCR amplification with the other 3 primer sets was performed as described above. For the primer set PirF/PirR, since 30 cycles failed to produce a signal (data not shown), 40 cycles of amplification were used
1008
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
Table 1 PCR primers used in this study. Gene/sequence
Primer name
Primer sequence (50 e30 )
AP1F AP1R
50 - CCTTGGGTGTGCTTAGAGGATG -30 50 - GCAACCTATCGCGCAGAACACC -30
AP2F AP2R
50 - TCACCCGAATGCTGCTTGTGG -30 50 - CGTCGCTACTGTCTAGCTGAAG -30
PirF PirR
50 - GAGCCAGATATTGAAAACATTTGG -30 50 - CCACGCAGCGAGTTCTGTAATGTA -30
AP1
into each well and incubated for 1 h at room temperature. After developing the enzyme activity signal by using 100 ml/well TMB (3,30 ,5,50 -Tetramethylbenzidine) substrate (Sigma) for 15 min in the dark, the developing was stopped by adding the stop solution (1 N HCl) and the absorbance of each well was detected at a wavelength of 450 nm.
AP2
Pir
gyrB Vpara1F Vpara2R
0
0
5 - CGGCGTGGGTGTTTCGGTAGT -3 50 - TCCGCTTCGCGCTCATCAATA -30
instead. 2.7. Detection of the toxin proteins PirAvp and PirBvp in bacteria using western blotting Individual colonies of each bacterial strain were picked from the TCBS plate, added to 3 ml of TSB medium containing 2% NaCl, and cultured at 30 C overnight. The resulting bacterial broth was adjusted to OD600 1.5, diluted 200 with TSB medium containing 2% NaCl, and incubated again at 30 C for another 12 h. After centrifuging (8000 g) for 5 min at 4 C, the supernatant and bacteria cell pellet were collected. The pellet was dissolved in 200 ml PBS before Western blotting. Samples were mixed with SDS sample buffer, separated on 15% SDS-PAGE and transferred onto a PVDF membrane. Each membrane was then cut into two pieces so that top half contained the heavier bands (including PirBvp) and the bottom half contained the lighter bands (including PirAvp). After blocking with blocking buffer (2% BSA in TBST [50 mM TriseHCl, 500 mM NaCl, 0.2% Tween 20, pH 7.5]) for 1 h at room temperature, the lower and upper halves of the membranes were respectively incubated with primary antibodies against PirAvp and PirBvp [11] in blocking buffer for 1 h at room temperature. After washing twice with TBST, the membranes were incubated with secondary antibody conjugated with horseradish peroxidase (HRP) in TBST for 1 h at room temperature. After the signals were developed using a Western Lightning Plus-ECL reagent kit (Perkin Elmer), immunoblot signals were detected using an ImageQuant LAS 4000 mini (GE Healthcare Life Science). 2.8. Detection of the toxin proteins PirAvp and PirBvp in shrimp tissues and hemolymph using ELISA (enzyme-linked immunoassay) For this assay, the samples of shrimp stomach and hepatopancreatic tissue were homogenized in lysis-PBS buffer (3 diluted PBS) and centrifuged (10,000 g), after which the supernatants (protein lysates) were collected. The hemolymph samples were subjected to centrifugation only (8000 g), and again the supernatants were collected. A Bradford protein assay kit (Bio-Rad) was used to quantify the protein concentrations in the supernatants, and samples (0.75 mg total protein/well) were mixed with 100 ml 0.1 M carbonate/bicarbonate buffer, pH 9.6, and coated onto the microtiter wells of a 96-well ELISA plate for 12 h at 4 C. After removing uncoated proteins by three washes with PBST (0.2% Tween 20 in PBS), blocking buffer (2% BSA in PBST) was added into each well and the mixture was incubated for 1 h at room temperature. The wells were washed again and then incubated with the primary antibodies against PirAvp and PirBvp in blocking buffer for 1 h at room temperature. After washing with PBST, the secondary antibody conjugated with HRP in blocking buffer was then added
2.9. Histological observation of the shrimp hepatopancreas using H&E staining Shrimp used in this experiment were challenged by the same immersion procedure as described above. At 0, 6, 12, 18, 24, 48, 72 and 96 h after the initial bacterial immersion, three shrimp were taken from each group. The cephalothorax of these randomly selected samples were fixed with Davidson's fixative for 48 h and then sectioned and stained with hematoxylin and eosin using methods described by Tran et al. [5]. Microphotographs of the hepatopancreatic structures and AHPND lesions were taken by light microscopy. 2.10. Determination of minimal inhibitory concentration (MIC) In this study, determination of the MIC was performed by the agar plate dilution method modified from Andrews [13]. The following antibiotics were selected for this assay: ampicillin (ABPC), chloramphenicol (CP), kanamycin (KM), streptomycin (SM), tetracycline (TC), nalidixic acid (NA), sulfamethoxazole (SMZ), trimethoprim (TMP), enrofloxacin (ERFX), fosfomycin (FOM), ofloxacin (OFLX) and bicozamycin (BCM). All of the above antibiotics were originally effective either against all bacteria or against Gram-negative bacteria only [14]. However, the inappropriate use of antibiotics can result in the development of resistant bacterial strains, and we note that, among the antibiotics tested here, at least CP, SM, TC, SMZ, TMP, ERFX and OFLX are widely used in shrimp farming [15,16]. After the individual antibiotics were dissolved in their corresponding diluents, they were subjected to 2 serial dilution in ddH2O and added to molten Mueller-Hinton agar with final concentrations that ranged from 0.006 to 12.8 mg/ml. To prepare the bacterial inocula, 3e5 individual colonies of each bacterial strain were taken from TCBS plates and subcultured overnight at 30 C in TSB medium containing 2% NaCl. The resulting bacterial suspension was diluted 500 in TSB and 10 ml of the final bacterial inoculum was placed onto the surface of antibiotic agar plates. The bacterial inoculum was completely absorbed within 10e20 min, and the plates were then cultured overnight at 30 C. The MIC of each bacteria was defined as the concentration of antibiotic which completely inhibited bacterial growth. 3. Results 3.1. Characterization of the bacterial strains used in this study To confirm the presence of the AP1, AP2 and Pir sequences that are associated with the AHPND-causing strains of V. parahaemolyticus [11], PCR was performed to pre-screen the seven bacterial strains used in this study. The three virulent Thailand strains, ThV-1, ThV-16 and 5HP, and one of the Vietnam strains, M1-1, were positive for all three sequences (Fig. 1A). For the two non-AHPND-causing strains from Thailand, ThN-2 and ThN-7, PCR results were negative for AP1, AP2 and Pir. The non-virulent strain M2-36, which was isolated from a pond in Vietnam after an outbreak of AHPND, was positive for AP1 and AP2 but showed a negative result for the Pir amplicon. The Pir amplicon includes the toxin gene pirBvp, and as reported in a previous study [11], the
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
1009
Fig. 1. Characteristics of the AHPND and non-AHPND strains used in this study. (A) PCR detection of the AP1, AP2 and Pir sequences was performed on the indicated V. parahaemolyticus strains. Strains ThN-2 and ThN-7 do not contain the virulence plasmid and were used as a negative controls. Vpara is a primer set that is specific for the V. parahaemolyticus gene gyrB. (B) Western blotting analysis was used to detect PirAvp and PirBvp proteins in the culture medium supernatant (S) and the cell pellet (P) of the indicated strains after overnight culturing at 30 C. The white lines that separate each PVDF membrane into two halves indicate where the membranes were cut prior to incubation with the anti- PirBvp (upper half) and anti- PirAvp (lower half) antibodies.
absence of this gene from M2-36 explains why this strain fails to cause AHPND in shrimp. vp
3.2. Detection of PirAB blotting
toxin in culture medium using western
To investigate whether PirABvp toxin is released into the extracellular environment, the bacteria were incubated overnight at 30 C, and the culture medium and bacteria cell pellets were subjected to Western blotting with antibodies against PirAvp and PirBvp. As shown in Fig. 1B, PirAvp (~12 kDa) and PirBvp (~50 kDa) were detected in both the bacteria cell pellets and the culture medium of strains ThV-1, ThV-16, 5HP and M1-1. Neither PirAvp or PirBvp were detected in the bacteria cell pellets or culture medium of strains ThN-2, ThN-7 and M2-36. All of these results were consistent with the PCR result in Fig. 1A, and confirm that ThV-1, ThV-16, 5HP and M1-1 are virulent strains. In the infection pathway experiments described below, M1-1 and 5HP were used as the AHPND-causing strains.
3.3. Antimicrobial resistance profiles of the seven V. parahaemolyticus strains Evaluation of the antibiotic resistance was performed by detecting the minimal inhibitory concentrations (MICs) of 12 antimicrobial drugs (Table 2). While some antibiotics, such as CP, ERFX and OFLX, still seem to be universally effective, there also seems to be universal resistance to ABPC, SM, SMZ, FOM and BCM. The remaining antibiotics are effective against some strains but not against others.
3.4. The Raman spectral profiles of AHPND-causing strains Raman spectral analysis generates a profile of the organic molecules in a cell, and this method has recently been used to detect and identify bacterial strains via their unique “fingerprints” [17,18]. The Raman spectra of five of the bacterial strains in this study all
Table 2 Minimal inhibitory concentrations of selected antibiotics on V. parahaemolytics. Bacteria
ThV-1 ThV-16 ThN-2 ThN-7 5HP M1-1 M2-36
Antibioticsa (mg/ml) ABPC
CP
KM
SM
TC
NA
12.8 12.8 6.4 12.8 12.8 12.8 12.8
3.2 3.2 0.4 1.6 0.8 1.6 1.6
6.4 3.2 6.4 3.2 12.8 6.4 6.4
12.8 12.8 12.8 12.8 S12.8 12.8 6.4
0.4 3.2 0.4 12.8 0.4 12.8 12.8
12.8 1.6 0.8 1.6 0.8 0.8 0.8
Bacteria
Antibiotics (mg/ml) SMZ
TMP
ERFX
FOM
OFLX
BCM
ThV-1 ThV-16 ThN-2 ThN-7 5HP M1-1 M2-36
S12.8 12.8 6.4 S12.8 S12.8 12.8 12.8
3.2 6.4 1.6 6.4 12.8 3.2 3.2
0.4 0.1 0.1 0.2 0.1 0.1 0.1
S12.8 S12.8 S12.8 S12.8 S12.8 S12.8 S12.8
0.2 0.2 0.2 0.4 0.2 0.4 0.2
S12.8 S12.8 S12.8 S12.8 S12.8 S12.8 S12.8
a Antibiotics: ABPC (Ampicillin), CP (Chloramphenicol), KM (Kanamycin), SM (Streptomycin), TC (Tetracycline), NA (Nalidixic acid), SMZ (Sulfamethoxazole), TMP (Trimethoprim), ERFX (Enrofloxacin), FOM (Fosfomycin), OFLX (Ofloxacin), BCM (Bicozamycin).
1010
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
show a similar pattern (Fig. 2), which indicates a general similarity in the biochemical composition of the tested strains, and suggests that, without more sophisticated mathematical analysis, this method is not suitable for distinguishing between AHPND and nonAHPND causing strains of V. parahaemolyticus. 3.5. Location of AHPND-causing bacteria in shrimp challenged by immersion Shrimps were challenged by immersion in the bacterial strains M1-1 or 5HP or in the TSB control, and for each group, at each time point, the stomach and hepatopancreas were collected. Total DNA was extracted from these tissues, and PCR analysis was performed as described above with the primer sets for AP1, AP2 and Pir in order to detect the presence of AHPND-causing bacteria. In stomach, DNAs extracted from shrimp infected with 5HP gave positive PCR results with the AP1, AP2 and Pir primer sets from at 6e24 hpi through to the death of the challenged shrimp at 24 hpi (Fig. 3A). In general, weaker PCR signals were produced by the M11 strain, and shrimp challenged with this strain survived beyond the end of the experiment (96 hpi). We also note that at 96 hpi, only very weak signals were detected from the AP1, AP2 and Pir sequences. In the hepatopancreas (Fig. 3B), shrimp infected with 5HP gave positive PCR results after 12 hpi, while M1-1 results were almost entirely negative for the entire 96 h (Fig. 3B).
only at 24 hpi in the M1-1 treated group. Thereafter, even when the differences were statically significant, they still represented only small increases over the baseline control values (Fig. 4). In the hepatopancreas of shrimp challenged with 5HP, although the levels of PirAvp started to be elevated only at 18 hpi, PirBvp levels were significantly increased throughout the entire experiment from 6 to 24 hpi (Fig. 5). By contrast, in the M1-1 treated group, the statistically significant increases in the levels of PirAvp and PirBvp were generally smaller and did not appear until later (Fig. 5). Lastly, although significantly elevated levels of PirAvp and PirBvp were observed in the hemolymph at 24 hpi and 48 hpi, respectively (Fig. 6), in general there was little evidence that the toxin or its components were distributed via the hemolymph. 3.7. Histopathology of the hepatopancreas of shrimp challenged with the M1-1 and 5HP strains of V. parahaemolyticus Typical AHPND histopathology was observed in the hepatopancreas of the 5HP-treated group from 6 hpi to 24 hpi (Fig. 7). Almost all of the examined hepatopancreas sections of the 5HPinfected shrimp showed sloughing of the hepatopancreas tubule epithelial cells into the lumen. Hemocyte infiltration was also seen in the areas surrounding the necrotic hepatopancreas tubules. Some bacteria colonization was observed at 24 hpi. By contrast, none of these pathological effects were observed in the M1-1 group (Fig. 7).
3.6. Production and location of the PirABvp toxin
4. Discussion
Based on the absence of large numbers of bacteria in the hepatopancreas even when the characteristic AHPND lesions are clearly seen, it has been proposed [11] that the damage to the hepatopancreatic cells is caused by PirABvp toxin that has been released by AHPND-causing bacteria that are still in the stomach. To further explore this proposed mechanism, we investigated whether PirABvp toxin could be detected in the total protein lysates prepared from stomach, hepatopancreas and hemolymph. In stomach, ELISA assays with antibodies against PirAvp and PirBvp detected the presence of elevated levels of both PirAvp and PirBvp in the 5HP treated group starting at 6 hpi, and continuing through to the death of the shrimp at 24 hpi (Fig. 4). Meanwhile, significant increases in the levels of PirAvp and PirBvp first occurred
V. parahaemolyticus is ubiquitous in the marine environment, and although it is usually only an opportunistic pathogen, the AHPND-causing strains of V. parahaemolyticus are uniquely virulent to shrimp because they contain a large plasmid that encodes the binary insecticidal toxin homologs PirAvp and PirBvp [9,10]. The origin of this plasmid and the toxin genes is still an open question. One intriguing possibility is suggested by the inconsistent results for the TSB group in Fig. 3A. Unexpectedly, in the TSB-treated control group, a number of positive PCR results were found with the AP1 and AP2 primer sets and also with the Pir primer set. The reason for these surprising results is unclear, but it is possible that some unidentified bacteria e probably not V. parahaemolyticus because our initial PCR analysis with the Vpara primer set was negative (data not shown) e may have been present in the stomachs of some of the putatively healthy experimental shrimp. Since we already know that in some bacterial species, during recombination events, entire plasmids or mobile genetic elements within the chromosomal or plasmidic genome significantly contribute to changes in genes, including toxin genes [19,20], it seems likely that a series of recombination events involving plasmids and/or mobile genetic elements may also have created the AHPND virulence plasmid that is now found in V. parahaemolyticus. It also seems possible that these events might have taken place in the stomach of a shrimp. If so, then the positive AP1, AP2 and Pir signals in the stomachs of the TSB-only shrimp might simply be a reflection of these potential events. The transfer of R (resistance) plasmids and mobile genetic elements during recombination events is also used by bacteria to achieve antibiotic resistance [21,22]. Unfortunately, the increased selective pressure caused by the widespread use of antibiotics in shrimp aquaculture further serves to drive and enhance this bacterial antibiotic resistance [23,24]. In this study, the antibiotic resistance of AHPND-causing and non-AHPND causing strains was evaluated for the first time (Table 2). It is alarming that all of the tested bacterial strains were resistant to some of the antibiotics (Table 2). This suggests that increased resistance to antibiotics is
Fig. 2. Raman spectra of single bacterial cells of the indicated strains recorded with 635.8 nm excitation. Each spectrum represents the mean “fingerprint” of approximately 200 bacterial cells.
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
1011
Fig. 3. PCR detection of the indicated AHPND-causing strains in shrimp (A) stomach (Stm) and (B) hepatopancreas (Hep). Shrimp were challenged with the M1-1 and 5HP strains by immersion, and tissues were collected from 4 individual animals at the indicated time points. (Only two animals were collected at 24 hpi in the 5HP group because only two animals were still alive.) After total DNAs were extracted, PCR detection of the AP1, AP2 and Pir sequences was performed. Shrimp treated with TSB medium were used as controls.
already becoming a serious problem. Instead of the unregulated use of antibiotics, we therefore recommend the establishment and maintenance of properly managed biosecure systems as the way forward for an eco-friendly, sustainable shrimp aquaculture industry. The Raman spectroscopy data that was used in this study to generate molecular “whole-organism fingerprints” [25e27] also provided information on the molecular content of AHPND-causing bacteria cells. The Raman spectra in Fig. 2 show bands representing the major cell components, including proteins (1003 cm1 and 1659.9 cm1), nucleic acids (782.9 cm1), lipids (1253.7 cm1) and carbohydrates/lipids (1449.5 cm1) [27,28]. We observed a slight increase in the intensity of the Raman band of the AHPND-causing strain 5HP located between 850 and 950 cm1, although the
significance of this feature, if any, still needs to be clarified. Our main result, however, is negative: although all the AHPND-causing strains contain a large unique plasmid, the presence of this plasmid failed to produce any correspondingly unique Raman spectra features that could be used to distinguish between AHPND-causing strains versus non-AHPND-causing strains (Fig. 2). It will be interesting to investigate whether or not useful differences between the AHPND-causing and non-AHPND-causing strains will be revealed by other techniques with higher sensitivity, such as proteomics and metabolomics. The results obtained in the present study suggest that the pathology of the AHPND-causing strain 5HP is consistent with our previous proposal that the characteristic sloughing of epithelial cells into the hepatopancreatic tubules is due to secreted toxins
1012
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014 6 hpi
***
PirAvp
Relative expression
3.5 3 2.5
***
3 2.5
***
***
3 2.5
**
***
3 2.5
***
***
72 hpi
48 hpi
24 hpi 3.5
3.5
**
*
3 2.5
96 hpi 3.5
3.5 3
3
2.5
2.5
2
2
2
2
2
2
2
1.5
1.5
1.5
1.5
1.5
1.5
1
1
1
1
1
1
1
0.5
0.5
0.5
0.5
0.5
0.5
0.5
TSB
M1-1
***
3.5
PirBvp
***
18 hpi 3.5
1.5
0
Relative expression
12 hpi 3.5
3 2.5
5HP
0
TSB
***
3.5
***
M1-1
3 2.5
5HP
0
M1-1
TSB
**
3.5
***
**
3 2.5
0
5HP
M1-1
TSB
3.5
***
3
**
2.5
5HP
***
0
TSB
M1-1
3.5
*
0
M1-1
TSB
0
3
3
2.5
2.5
2
2
2
2
2
2
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1
1
1
1
1
1
1
0.5
0.5
0.5
0.5
0.5
0.5
0.5
TSB
M1-1
5HP
0
TSB
M1-1
5HP
0
TSB
M1-1
0
5HP
TSB
M1-1
5HP
0
TSB
M1-1
0
TSB
M1-1
3.5
3.5
3 2.5
2
0
*
TSB
M1-1
0
**
TSB
M1-1
Tissue: Stm
Fig. 4. Detection of toxins PirAVP and PirBVP in stomach by ELISA assay. The samples used in this experiment were taken from the same individual shrimps as those used in Fig. 3. Each bar represents the mean ± SD. Asterisks indicate statistically significant differences between the respective groups (Student's t test, *p < 0.05, **p < 0.01, ***p < 0.001).
6 hpi
12 hpi
PirAvp
Relative expression
*
3 2.5
3
3
2.5
2.5
***
***
48 hpi
24 hpi
**
3.5
**
3
3.5
**
2.5
72 hpi
*
3
3
2.5
2.5
96 hpi 3.5
3.5
***
3 2.5
2
2
2
2
2
2
2
1.5
1.5
1.5
1.5
1.5
1.5
1
1
1
1
1
1
1
0.5
0.5 TSB
M1-1
**
7
PirBvp
3.5
1.5
0
Relative expression
18 hpi
3.5
3.5
6
5HP
0
***
7
**
M1-1
5HP
0
7
*
TSB
**
***
0
5HP
M1-1
***
TSB
M1-1
7
*
**
5HP
0
M1-1
0
TSB
M1-1
7
7
***
0.5
0.5 TSB
*
6
0
6
5
5
4
4
4
4
4
3
3
3
3
3
3
3
2
2
2
2
2
2
2
5
4
1
1 0
TSB
M1-1
5HP
0
1 TSB
M1-1
5HP
0
0
5HP
5
1
1 M1-1
TSB
5
TSB
M1-1
5HP
0
M1-1
0
M1-1
6 5
***
4
1 TSB
TSB
7
*
6
6
5
6
0.5
0.5
0.5 TSB
***
1 TSB
M1-1
0
TSB
M1-1
Tissue: Hep
Fig. 5. Detection of toxins PirAVP and PirBVP in hepatopancreas (Hep) by ELISA assay. The samples used in this experiment were taken from the same individual shrimps as those used in Fig. 3. Each bar represents the mean ± SD. Asterisks indicate statistically significant differences between the respective groups (Student's t test, *p < 0.05, **p < 0.01, ***p < 0.001).
6 hpi
*
PirAvp
Relative expression
***
24 hpi
*
*
48 hpi
72 hpi
96 hpi
** 1.75
1.75
1.5
1.5
1.5
1.5
1.5
1.5
1.25
1.25
1.25
1.25
1.25
1.25
1.25
1.75
1.75
1.75
1.75
1
1
1
1
1
1
1
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.25
0.25 0
PirBvp
18 hpi
1.5
1.75
Relative expression
12 hpi
TSB
M1-1
5HP
0
M1-1
5HP
*
0
1.75
1.75
1.5
1.5
1.5
1.25
1.25
1.25
0.25
0.25
0.25 TSB
TSB
1.75
M1-1
*
***
5HP
***
0
TSB
1.75
M1-1
0
0
TSB
M1-1
1.25
0
1.5
1.5
1.25
1.25
1
1
1
1
1
1
1
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0
TSB
M1-1
5HP
0
TSB
M1-1
5HP
0
TSB
M1-1
5HP
0
TSB
M1-1
5HP
0
TSB
M1-1
0
TSB
M1-1
1.75
1.75
*
1.5
1.25
0.25
0.25 M1-1
TSB
1.75
*
*
1.5
5HP
TSB
M1-1
0
TSB
M1-1
Tissue: Hlm
Fig. 6. Detection of toxins PirAVP and PirBVP in hemolymph (Hlm) by ELISA assay. The samples used in this experiment were taken from the same individual shrimps as those used in Fig. 3. Each bar represents the mean ± SD. Asterisks indicate statistically significant differences between the respective groups (Student's t test, *p < 0.05, **p < 0.01, ***p < 0.001).
rather than the presence of the bacteria itself [11]. Thus, our PCR results show that 5HP initially colonizes the stomach at 6 hpi and only later reaches the hepatopancreas (12 hpi; Fig. 3). Meanwhile,
the PirAvp and PirBvp toxins were also detected in the stomach of the 5HP-challenged shrimp at a similarly early time point (6 hpi; Fig. 4). If the PirABvp toxin functions as a pore-forming toxin (as we
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
1013
Fig. 7. H&E stained photomicrographs of hepatopancreas collected from TSB-treated, M1-1 infected, and 5HP-infected groups at 0e96 hpi. Normal hepatopancreas tubules are seen in the TSB-treated and M1-1-infected groups, while typical AHPND lesions with necrotic and sloughed epithelial cells (black arrows) are seen from 6 hpi in the 5HP-infected group. From 6 to 18 hpi in the 5HP group, the small dark dots are hemocytes that have infiltrated between the cells of the hepatopancreas tubules. At 24 hpi, significant bacterial colonization (blue arrows) was also observed in the lumen of the hepatopancreas tubules. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
proposed in a previous publication [11]), it may damage the shrimp stomach epithelium in the same way that the original Pir (Photorhabdus insect related) binary toxin, PirAB, causes swelling and shedding of the epithelial cells in the digestive organs of a range of insects [29e31]. This putative damage to the stomach would then allow both the toxin proteins and, subsequently, the bacteria themselves to spread from the stomach to the hepatopancreas. Sure enough, in the 5HP-challenged shrimp, Fig. 5 shows elevated levels of PirBvp in the hepatopancreas as early as 6 hpi, and at this time there was already sloughing of the hepatopancreatic tubule epithelial cells (Fig. 7). Interestingly, preliminary experiments suggest that infection with 5HP induces significant upregulation of
several immune-related genes (Toll, My88, ALF, PEN2, PEN3 and PEN4) in the hepatopancreas at this time, whereas by contrast, expression of these same genes is suppressed in the stomach and hemocytes (data not shown). The bacteria themselves, on the other hand, were only detected in the hepatopancreas at 12 hpi (Fig. 3). Presumably at this time, the PirABvp toxin is being produced by the 5HP bacteria in both the stomach and the hepatopancreas and released into both of these organs directly. We note, however, that surprisingly, increased levels of PirAvp were not seen in the hepatopancreas until 18 hpi (Fig. 5). The reason for this is unclear, but previous studies on PirB from other (insecticidal) bacteria as well as our own work on PirBvp suggest that under some circumstances
1014
H.-C. Lai et al. / Fish & Shellfish Immunology 47 (2015) 1006e1014
recombinant PirB toxin alone might be sufficient to induce cell damage and mortality in the host [11,32]. It is interesting to note that, although both 5HP and M1-1 were able to secrete the PirAvp and PirBvp toxins in vitro (Fig. 1), neither toxin was detected in the hemolymph except between 24 and 48 hpi (Fig. 6). This suggests that these toxins are not primarily being spread via the shrimp circulatory system, and that they are therefore probably unlikely to cause systemic damage or damage to other remote tissues or organs. In contrast to the 5HP strain, shrimp that were challenged with the M1-1 strain survived through to the end of the experiment (Figs. 3e6) and showed none of the characteristic symptoms of AHPND (Fig. 7). Nevertheless, Fig. 1 suggests that M1-1 was able to produce similar amounts of toxin to 5HP, and we also found that M1-1 caused the hepatopancreas to become pale (which is another clinical sign of AHPND) in postlarvae (data not shown). The reason for M1-1's relatively weak pathogenicity in the 3 g shrimp used here is presently unclear, However, the weaker M1-1 PCR signals from stomach DNA and the almost complete absence of signals from the hepatopancreas (Fig. 3) leads us to tentatively conclude that a relatively slow replication of strain M1-1 in stomach may be the explanation for its weaker pathogenicity. A slower replication rate would also explain why M1-1 generally manages to produce less PirAvp and PirBvp than 5HP in the stomach and hepatopancreas (Figs. 4 and 5). However, we continue to work on this question, and we are presently using a next generation sequencing platform to sequence the genomic DNA and plasmid DNAs of strains M1-1 and 5HP. In the near future, by searching for specific genes/sequences that exist only in M1-1 or 5HP, we hope to better understand on a molecular level the underlying mechanisms that give rise to these differences. Acknowledgments This study was supported financially by the Ministry of Science and Technology (MOST 103-2633-B-006-004 and MOST 104-2622B-006-009 -CC1). One of the Vibrio parahaemolyticus strains, 5HP, was originally provided by Dr. Timothy W. Flegel, Dr. Kallaya Sritunyalucksana, and Dr. Siripong Thitamadee. We would also like to thank Paul Barlow for his helpful criticism of the manuscript. References [1] D.V. Lightner, R.M. Redman, C.R. Pantoja, B.L. Noble, L.H. Tran, Early mortality syndrome affects shrimp in Asia, Glob. Aquac. Advocate 40 (2012). [2] T.W. Flegel, Historic emergence, impact and current status of shrimp pathogens in Asia, J. Invertebr. Pathol. 110 (2012) 166e173. ~ o, C.V. Mohan, Early mortality syndrome threatens Asia's shrimp [3] E.M. Lean farms, Glob. Aquac. Advocate (2012) 38e39. [4] S.A. Soto-Rodriguez, B. Gomez-Gil, R. Lozano-Olvera, M. Betancourt-Lozano, M.S. Morales-Covarrubias, Field and experimental evidence of Vibrio parahaemolyticus as the causative agent of acute hepatopancreatic necrosis disease (AHPND) of cultured shrimp (Litopenaeus vannamei) in northwestern Mexico, Appl. Environ. Microbiol. 81 (2015) 1689e1699. [5] L. Tran, L. Nunan, R.M. Redman, L.L. Mohney, C.R. Pantoja, K. Fitzsimmons, D.V. Lightner, Determination of the infectious nature of the agent of acute hepatopancreatic necrosis syndrome affecting penaeid shrimp, Dis. Aquat. Organ. 105 (2013) 45e55. [6] P. De Schryver, T. Defoirdt, P. Sorgeloos, Early mortality syndrome outbreaks: a microbial management issue in shrimp farming? PLoS Pathog. 10 (2014) e1003919. [7] L. Nunan, D. Lightner, C. Pantoja, S. Gomez-Jimenez, Detection of acute hepatopancreatic necrosis disease (AHPND) in Mexico, Dis. Aquat. Organ 111 (2014) 81e86. [8] H. Kondo, S. Tinwongger, P. Proespraiwong, R. Mavichak, S. Unajak, R. Nozaki, I. Hirono, Draft genome sequences of six strains of Vibrio parahaemolyticus isolated from early mortality syndrome/acute hepatopancreatic necrosis
disease shrimp in Thailand, Genome Announc. 2 (2014) e00221e14. [9] Y.T. Yang, I.T. Chen, C.T. Lee, C.Y. Chen, S.S. Lin, L.I. Hor, T.C. Tseng, Y.T. Huang, K. Sritunyalucksana, S. Thitamadee, H.C. Wang, C.F. Lo, Draft genome sequences of four strains of Vibrio parahaemolyticus, three of which cause early mortality syndrome/acute hepatopancreatic necrosis disease in shrimp in China and Thailand, Genome Announc. 2 (2014) e00816e14. [10] J.E. Han, K.F. Tang, L.H. Tran, D.V. Lightner, Photorhabdus insect-related (Pir) toxin-like genes in a plasmid of Vibrio parahaemolyticus, the causative agent of acute hepatopancreatic necrosis disease (AHPND) of shrimp, Dis. Aquat. Organ. 113 (2015) 33e40. [11] C.T. Lee, I.T. Chen, Y.T. Yang, T.P. Ko, Y.T. Huang, J.Y. Huang, M.F. Huang, S.J. Lin, C.Y. Chen, S.S. Lin, D. Lightner, H.C. Wang, H.J. Wang, H.C. Wang, L.I. Hor, C.F. Lo, Vibrio parahaemolyticus: an opportunistic marine pathogen becomes virulent by acquiring a plasmid that expresses a deadly toxin, Proc. Natl. Acad. Sci. U. S. A. 112 (2015) (2015) 10798e10803. [12] X.Y. Luan, J.X. Chen, X.H. Zhang, J.T. Jia, F.R. Sun, Y. Li, Comparison of different primers for rapid detection of Vibrio parahaemolyticus using the polymerase chain reaction, Lett. Appl. Microbiol. 44 (2007) 242e247. [13] J.M. Andrews, Determination of minimum inhibitory concentrations, J. Antimicrob. Chemother. 48 (2001) 5e16. [14] M.A. Kohanski, D.J. Dwyer, J.J. Collins, How antibiotics kill bacteria: from targets to networks, Nat. Rev. Microbiol. 8 (2010) 423e435. [15] K. Kümmerer, Antibiotics in the aquatic environmentea reviewepart I, Chemosphere 75 (2009) 417e434. € m, S. Gr€ € m, S. Poungshompoo, B.E. Bengtsson, [16] K. Holmstro aslund, A. Wahlstro N. Kautsky, Antibiotic use in shrimp farming and implications for environmental impacts and human health, IJFST 38 (2003) 255e266. [17] J.C. Baritaux, E. Schultz, A.C. Simon, A.G. Bourdat, I. Espagnon, P. Laurent, J.M. Dinten, Fast and robust identification of single bacteria in environmental matrices by Raman spectroscopy, in: SPIE BiOS. International Society for Optics and Photonics, 2015, 93281Ie93281I. [18] X. Wu, Y.W. Huang, B. Park, R.A. Tripp, Y. Zhao, Differentiation and classification of bacteria using vancomycin functionalized silver nanorods array based surface-enhanced Raman spectroscopy and chemometric analysis, Talanta 139 (2015) 96e103. [19] L.S. Frost, R. Leplae, A.O. Summers, A. Toussaint, Mobile genetic elements: the agents of open source evolution, Nat. Rev. Microbiol. 3 (2005) 722e732. [20] K.K. Hill, G. Xie, B.T. Foley, T.J. Smith, A.C. Munk, D. Bruce, L.A. Smith, T.S. Brettin, J.C. Detter, Recombination and insertion events involving the botulinum neurotoxin complex genes in Clostridium botulinum types A, B, E and F and Clostridium butyricum type E strains, BMC Biol. 7 (2009) 66. ~ a, Antibiotic resistance of bacteria from shrimp [21] E.A. Tendencia, L.D. de la Pen ponds, Aquaculture 195 (2001) 193e204. [22] P.M. Bennett, Plasmid encoded antibiotic resistance: acquisition and transfer of antibiotic resistance genes in bacteria, Br. J. Pharmacol. 153 (2008) S347eS357. [23] B.H. Normark, S. Normark, Evolution and spread of antibiotic resistance, J. Intern. Med. 252 (2002) 91e106. n-Aguin ~ aga, P.R. Jensen, Multilocus sequence typing reveals [24] K.C. Freel, N. Milla evidence of homologous recombination linked to antibiotic resistance in the genus Salinispora, Appl. Environ. Microbiol. 79 (2013) 5997e6005. [25] C. Kirschner, K. Maquelin, P. Pina, N.A. Ngo Thi, L.P. Choo-Smith, G.D. Sockalingum, C. Sandt, D. Ami, F. Orsini, S.M. Doglia, P. Allouch, M. Mainfait, G.J. Puppels, D. Naumann, Classification and identification of enterococci: a comparative phenotypic, genotypic, and vibrational spectroscopic study, J. Clin. Microbiol. 39 (2001) 1763e1770. [26] J.F. Almarashi, N. Kapel, T.S. Wilkinson, H.H. Telle, Raman spectroscopy of bacterial species and strains cultivated under reproducible conditions, J. Spectrosc. 27 (2012) 361e365. [27] K. Hamasha, Q.I. Mohaidat, R.A. Putnam, R.C. Woodman, S. Palchaudhuri, S.J. Rehse, Sensitive and specific discrimination of pathogenic and nonpathogenic Escherichia coli using Raman spectroscopy-a comparison of two multivariate analysis techniques, Biomed. Opt. Express 4 (2013) 481e489. [28] E. Staniszewska-Slezak, K. Malek, M. Baranska, Complementary analysis of tissue homogenates composition obtained by Vis and NIR laser excitations and Raman spectroscopy, Spectrochim. Acta A Mol. Biomol. Spectrosc. 147 (2015) 245e256. [29] N. Waterfield, S.G. Kamita, B.D. Hammock, R. Ffrench-Constant, The photorhabdus Pir toxins are similar to a developmentally regulated insect protein but show no juvenile hormone esterase activity, FEMS Microbiol. Lett. 245 (2005) 47e52. [30] M.B. Blackburn, R.R. Farrar, N.G. Novak, S.D. Lawrence, Remarkable susceptibility of the diamondback moth (Plutella xylostella) to ingestion of Pir toxins from Photorhabdus luminescens, Entomol. Exp. Appl. 121 (2006) 31e37. [31] Y. Li, X. Hu, X. Zhang, Z. Liu, X. Ding, L. Xia, S. Hu, Photorhabdus luminescens PirAB-fusion protein exhibits both cytotoxicity and insecticidal activity, FEMS Microbiol. Lett. 356 (2014) 23e31. [32] A. Ahantarig, N. Chantawat, N.R. Waterfield, R. ffrench-Constant, P. Kittayapong, PirAB toxin from Photorhabdus asymbiotica as a larvicide against dengue vectors, Appl. Environ. Microbiol. 75 (2009) 4627e4629.