|Home | About | Journals | Submit | Contact Us | Français|
Conceived and designed the experiments: CN PJ IS JMT KS. Performed the experiments: CN PJ SM. Analyzed the data: CN PJ IS SM JMT KS. Contributed reagents/materials/analysis tools: SM JT. Wrote the paper: CN PJ IS JMT KS.
Aeromonas hydrophila is the most common Aeromonas species causing infections in human and other animals such as amphibians, reptiles, fish and crustaceans. Pathogenesis of Aeromonas species have been reported to be associated with virulence factors such as lipopolysaccharides (LPS), bacterial toxins, bacterial secretion systems, flagella, and other surface molecules. Several mutant strains of A. hydrophila AH-3 were initially used to study their virulence in two animal species, Pacifastacus leniusculus (crayfish) and Tenebrio molitor larvae (mealworm). The AH-3 strains used in this study have mutations in genes involving the synthesis of flagella, LPS structures, secretion systems, and some other factors, which have been reported to be involved in A. hydrophila pathogenicity. Our study shows that the LPS (O-antigen and external core) is the most determinant A. hydrophila AH-3 virulence factor in both animals. Furthermore, we studied the immune responses of these hosts to infection of virulent or non-virulent strains of A. hydrophila AH-3. The AH-3 wild type (WT) containing the complete LPS core is highly virulent and this bacterium strongly stimulated the prophenoloxidase activating system resulting in melanization in both crayfish and mealworm. In contrast, the ΔwaaE mutant which has LPS without O-antigen and external core was non-virulent and lost ability to stimulate this system and melanization in these two animals. The high phenoloxidase activity found in WT infected crayfish appears to result from a low expression of pacifastin, a prophenoloxidase activating enzyme inhibitor, and this gene expression was not changed in the ΔwaaE mutant infected animal and consequently phenoloxidase activity was not altered as compared to non-infected animals. Therefore we show that the virulence factors of A. hydrophila are the same regardless whether an insect or a crustacean is infected and the O-antigen and external core is essential for activation of the proPO system and as virulence factors for this bacterium.
Aeromonas hydrophila is a Gram-negative bacterium living in aquatic environments. It can be found in freshwater, seawater, and also chlorinated-drinking water. This bacterium has been considered as a food-borne pathogen since it is found in many food products, for example sea food, shrimp cocktail, ground meat and raw vegetables –. A. hydrophila is the most common species of Aeromonas that causes infections in human and other animals such as amphibian reptile, fish and crayfish –. Infection of this bacterium is also a major problem in carp aquaculture in India . Furthermore, A. hydrophila was isolated from several rainbow trout farms and found resistant to antibiotics used in aquaculture in Australia , and recently it was also isolated from freshwater crayfish (Pacifastacus leniusculus) and was found to be highly virulent to this animal .
The pathological conditions found in fish infected with A. hydrophila are usually hemorrhagic septicemias (reddish eyes, skin, gills, and fins) and tail and fin rot . Catfish infected with this bacterium exhibited hemorrhagic fins and had larger spleen, kidney and liver . Crayfish infected with A. hydrophila showed necrotic injury in gill, heart and hepatopancreas . The pathogenesis of Aeromonas species have been reported to be associated with virulence factors such as lipopolysaccharides (LPS), bacterial toxins, bacterial secretory system, flagella and capsules. These factors are believed to be important in both resistance of bacteria to host immune responses and bacterial virulence , , –. Some other factors, such as siderophores (high-affinity iron chelating molecules) and porins (pore-forming proteins), are also reported to be involved in bacterial growth and A. hydrophila pathogenesis –.
LPS have been widely studied and reported to contribute an important role in resistance of Aeromonas spp. to the host immune system as well as in inducing harmful effects to the host –. The O-antigen, LPS core (external and internal), and lipid A, are assembled to form a complete LPS structure and are considered to play a role in bacterial pathogenesis , . Several genes of A. hydrophila involved in LPS biogenesis have been studied. For example, the waaL gene encodes a ligase protein required for ligation of the O-antigen to the LPS core, wzz gene is responsible for the length of the O-antigen, waaE gene plays a role in LPS core synthesis, and msbB gene is involved in lipid A biosynthesis. These genes are believed to be associated with virulence of A. hydrophila –. The chemical structure of the most relevant LPS structure from A. hydrophila AH-3 strain and mutants is shown in Figure 1.
Previously, crayfish has been used as an experimental model to study the host immune response after A. hydrophila challenge and melanization, e.g. phenoloxidase activity was found to be important for crayfish immune defense against A. hydrophila infection , . The details of prophenoloxidase (proPO) activating cascade have been studied in Manduca sexta and Tenebrio molitor, and the induction of melanization in Tenebrio larvae can be easily observed after microbial infection , . In this study we used A. hydrophila AH-3 wild type strain and several mutants on crayfish and Tenebrio experimental models to study the A. hydrophila most determinant virulence factor of in these animal models. Furthermore, we also studied the immune responses of the host to infection with A. hydrophila AH-3 virulent or non-virulent strains in both animal models.
To investigate how surface molecules on bacteria or secretion systems influence pathogenicity, the virulence of A hydrophila AH-3 wild type and several mutant strains was studied in two animal species, P. leniusculus (Table 1) and T. molitor (Table 2). These bacterial strains have been mutated in genes involved in the synthesis of cell surface structures and molecules, or secretion systems. No major differences in growth curves could be observed among the mutants when they grow in LB or in minimal Davies medium with glucose as a single carbon and energy source. In this study, A. hydrophila B1 was used as a positive control, since this strain was previously reported as a highly virulent bacterium to freshwater crayfish . Since the bacterial species used in this study had mutations in genes encoding some component of the cell walls, this in some cases had effects on the character of the bacterial mutants, which made it difficult to adjust the bacterial dose for injection to be exactly the same for each mutant species. As a consequence, there are some variations in injected dose or CFU between different bacterial mutant strains as shown in Table 1 and Table 2. The virulence of all bacterial strains can be determined from the injected CFU and the time of death after bacteria injection. Similar results for the different strains were obtained in crayfish and mealworm. A. hydrophila B1 used as a positive control was the most virulent strain causing death within 4 h and 21 h in crayfish and mealworm, respectively. The A. hydrophila AH-3 WT was less virulent than A. hydrophila B1, and this wild-type strain caused death within 25 h in both crayfish and mealworm. A. hydrophila AH-3 which lacks a Type VI secretion system (T6SS) and the AH-3 strain which lacks lateral flagella  showed similar virulence as the A. hydrophila AH-3 WT. In contrast, the mutant strains AH-3 Δuge and Δwzz  seemed to be more virulent than the WT in crayfish since all challenged animals died especially if a lower CFU of these two mutants was used for challenge. However, infection with AH-3 Δuge or Δwzz mutants took approximately 60 h or 45 h to cause crayfish or mealworm death, respectively. This is about two-fold longer time when compared to the WT. Therefore, A. hydrophila AH-3 which lacks a Type III secretion system (T3SS) , the strain which lacks polar flagellum  and all other mutant strains, excluding the ΔwaaE (O:34 antigen and external core negative strain) , were all virulent but to a lesser extent than the A. hydrophila AH-3 WT. The ΔwaaE was not virulent at all and could not kill any animals within the experimental period in this study. Moreover, it is important to notice that AH-3 ΔwaaL  mutant showed a decrease in virulence but not to the same extent as that of the AH-3 ΔwaaE mutant. Therefore, the virulent A. hydrophila AH-3 WT and the non-virulent ΔwaaE were selected for further detailed studies. The AH-3 ΔwaaE mutant showed a generation time in rich or minimal medium never superior to 20% of the corresponding wild type strain AH-3.
To investigate whether the toxins or extracellular products produced from the highly virulent A. hydrophila WT and the non-virulent ΔwaaE play a role in pathogenicity the bacterial extracellular products were prepared from these two strains and were used for cytotoxicity test using the crayfish hematopoietic cells (Figure 2). Extracellular products from both bacterial strains exhibited cytotoxic effects to the cells. Hematopoietic cells were completely lysed within 30 and 60 min after incubation with the extracellular products from the A. hydrophila WT and the ΔwaaE, respectively.
The results in Figure 3 clearly show that mealworm could not eliminate the virulent AH-3 WT from their hemolymph and the bacteria grew very fast at 12 h after injection, from 0.4–0.6×104 CFU/animal at the time of injection to an average of 2×107 CFU/ml hemolymph (Figure 3a). In contrast, the worms completely cleared the non-virulent ΔwaaE mutant within 12 h after injection, and no bacterial colonies were observed after 12 h injection of 1.8–2.9×104 CFU/animal (Figure 3b).
The virulent and non-virulent A. hydrophila strains were tested to see whether they have any effect on the expression of antimicrobial peptide genes in crayfish. The expression levels of crayfish antimicrobial peptide genes were determined 6 h after injection of A. hydrophila WT and the ΔwaaE mutant. The results in Figure 4 show that there was no change in expression level of Plcrustin1 in hepatopancreas or hemocytes after injection with WT, but the expression of this gene was increased in both tissues of crayfish injected with the ΔwaaE mutant strain. Plcrustin2 expression was not changed after injection with WT or the ΔwaaE mutant. In the case of Plcrustin3, expression of this gene in hepatopancreas was decreased after injection with WT, but the expression was not changed after injection with ΔwaaE mutant. The expression level of LGBP did not seem to be changed after either injection of WT or ΔwaaE mutant. The ALF gene expression was obviously decreased at least in hemocytes following AH-3 WT injection, but not changed after ΔwaaE mutant injection while the expression of astacidin2 was not changed after injection with either AH-3 WT or ΔwaaE mutant.
The results presented in Figure 5 clearly show that melanin was formed in mealworms injected with the virulent AH-3 WT. On the other hand, melanin formation could not be observed in mealworms injected with control buffer or the non-virulent ΔwaaE mutant strain.
Activation of the proPO system in crayfish after injection with the AH-3 WT or the ΔwaaE mutant was studied and the results are shown in Figure 6. Activity of PO in a hemocyte lysate prepared from crayfish injected with AH-3 WT was significantly increased compared to the buffer injected control group (P<0.05). However, activity of PO was increased to a low extent in crayfish injected with the ΔwaaE mutant. The proPO and pacifastin, a proPO activating enzyme inhibitor , transcripts were examined (Figure 7). The proPO and pacifastin light chain (proteinase subunit) transcripts were decreased in WT infected crayfish but there were no obvious changes in the ΔwaaE mutant infected animals. However, both AH-3 WT and ΔwaaE mutant showed a similar effect on the expression of the pacifastin heavy chain (transferrin subunit)  when compared to the CFS control animals.
Aeromonas hydrophila is a pathogenic bacterium for both terrestrial and aquatic animals. Several mutant strains of A. hydrophila AH-3 were used to study their virulence in two different animal species, Pacifastacus leniusculus (crayfish) and Tenebrio molitor larvae (mealworm). The A. hydrophila AH-3 strains used in this study have mutations in genes involved in the synthesis of flagella, LPS structures, secretion systems, and some other factors, which have been reported to be involved in A. hydrophila pathogenicity , –.
Polar flagellum and lateral flagella are factors suspected to be involved in Aeromonas spp. virulence. Many studies show that mutation in genes involved in polar flagellum and lateral flagella syntheses lead to reduction in bacterial motility and adhesion of Aeromonas spp. to Hep-2 cell , , . There is at least 50% of Aeromonas spp. which have two flagella systems. These flagella systems have been mentioned to be associated with bacterial pathogenesis by increasing host cell adhesion ability and swarming motility of A. hydrophila –. Moreover, some other structures of A. hydrophila such as siderophores, porins, and capsules are also believed to be associated with bacterial pathogenesis. The genes encoding siderophores, high-affinity iron chelating molecules, are present in the A. hydrophila ATCC 7966 genome. These molecules have also been reported to affect Aeromonas spp. pathogenesis and growth , , . A pore-forming protein, porin, on the outer membrane of A. hydrophila has also been shown to correlate with bacterial resistance to the host immune system. Lack of this protein in O-antigen depleted A. hydrophila strains (serum sensitive) increases bacterial serum resistant ability and this results in induction of bacterial survival in human serum . Although the flagella, siderophores, porin, and bacterial capsules have been previously reported to have correlations with A. hydrophila pathogenesis, the mutation of these factors in this study did not show any major effects on the virulence of A. hydrophila AH-3 in P. leniusculus or in T. molitor larvae.
This study also used A. hydrophila AH-3 strains with mutations in several LPS genes and found that the bacterium completely lost its virulence when the waaE gene was mutated. The ΔwaaE mutant did not cause death in P. leniusculus at all and only two T. molitor larva died after injection with this mutant strain. The death of T. molitor larva, however, could be an effect of injection since some animals also died after being injected with buffer control. The results are correlated with a study performed in mice which showed the LD50 for intraperitoneal injection of AH-3ΔwaaL mutant was approximately 20-fold higher than that of the AH-3 wild type, while the AH-3ΔwaaE mutant was completely avirulent (data not shown). The results, therefore, indicate that the LPS structure is important for A. hydrophila AH-3 virulence especially in the invertebrate models used in this study, and the most important part of this structure is the O:34 antigen LPS and the external core (Figure 1 shows the relevant LPS chemical structures).
Some secretion systems are involved in the release of bacterial toxins and effector proteins, some of which have been reported to play a role in bacterial virulence , –, , . One of the secretion systems which have been widely studied is a type III secretion system (T3SS). This secretion system is present in both clinical and environmental strains of A. hydrophila . Mutation of some genes involved in the function of T3SS reduced the virulence of A. hydrophila SSU and A. veronii in mice and that of A. hydrophila AH-1 in blue gourami fishes , , . Another secretion system which has been recently found in A. hydrophila is a type VI secretion system (T6SS) , . The study done by Suarez et al.  showed that mutation of two genes, vasH and vasK, present in the T6SS gene cluster of A. hydrophila SSU resulted in reduction of virulence of this bacterium both in vitro and in a mouse model. However, our study shows that T3SS and T6SS are probably not the determinant virulence factors of A. hydrophila AH-3 in the experimental models used in this study because mutation of these two secretion systems did not alter the virulence of A. hydrophila AH-3 infection in P. leniusculus or in T. molitor larvae as the AH-3 ΔwaaE mutant did.
Several toxins like aerolysins, phospholipases, β-hemolysin, HlyA-like hemolysin have been reported to be important for virulence of A. hydrophila. Phospholipase and hemolysin genes are present in A. hydrophila and their products were reported to have the toxic effects to rainbow trout, crayfish and mice , , . The A. hydrophila AH-3 used in this study has at least the hemolysin genes. β-hemolysin and HlyA-like hemolysin genes present in both the WT and the ΔwaaE mutant. The hemolytic, cytotoxic, and caseinolytic extracellular activities showed similar values for the AH-3 ΔwaaE mutant and the wild type strain (data not shown). Although the WT is virulent whereas the ΔwaaE mutant is non-virulent in in vivo systems, the extracellular products obtained from these two strains showed cytotoxic effects to crayfish hematopoietic cells even though the extracellular products obtained from the waaE negative strain showed a slightly lower effect. This indicates that the toxins produced by A. hydrophila do not seem to play a major role in A. hydrophila virulence in vivo in our models. Considering the expression of other virulence factors which might be affected by waaE mutation, the AH-3 ΔwaaE was able to express either polar or lateral flagellum like the AH-3 WT (data not shown). The WaaL LPS mutation in A. hydrophila AH-3 downregulated the expression of T3SS . A similar effect was observed for the AH-3ΔwaaE (data not shown). This indicated that LPS in some cases has a role in regulation of T3SS system . However, T3SS mutation alone has no drastic effects on virulence of A. hydrophila AH-3 in our study.
The study on bacterial clearance in T. molitor larvae showed different capacity of the worms in handling infection with virulent or non-virulent bacterial strains. The WT continuously grew inside the worms with a statistically significant increase of CFU at 12 h after injection (P<0.05) and finally caused death of the worms whereas the non-virulent ΔwaaE mutant was completely eliminated by T. molitor larvae within 12 h. The difference between these two bacterial strains is mainly in their LPS structure. The WT has complete LPS, but the ΔwaaE mutant which is non-pathogenic lacks the O-antigen and the external core (Figure 1). This indicates that both components of the LPS molecule are important for A. hydrophila to escape from the host defense system. This assumption is supported by some previous studies which showed that serum-resistant strains of A. hydrophila with complete LPS had high percent survival in non-immune human serum and low binding capacity to complement components , .
In addition to having different immune-resistant capacity, WT and ΔwaaE mutant induced different host immune responses. Expression of antimicrobial peptide (AMP) transcripts was investigated in crayfish after injection with WT and ΔwaaE mutant, and only minor changes in AMP expression was detected. Expression of Plcrustin3 and ALF (anti-lipopolysacharide factor) decreased after WT infection but no change in other AMP expressions could be observed. Upregulation of Plcrustin1 was observed after ΔwaaE mutant infection. Expression of this gene have been reported to be upregulated after infection with either pathogenic or non-pathogenic bacteria  but an interesting observation in this study is that the upregulation of this gene was observed only following an infection with the non-pathogenic ΔwaaE mutant.
The AH-3 ΔwaaE had an ability to induce PO activity in crayfish at a low level but could not activate melanization in T. molitor larvae or it might activate melanization to a very low extent which could not be observed. In contrast, the AH-3 WT significantly induced PO activity (P<0.05) and subsequent melanization in vivo. The AH-3 WT has a complete LPS with O-antigen to protect itself from host defense system , ,  whereas ΔwaaE mutant lacks its defense structure (LPS devoid of O-antigen and external core) and it was cleared very fast from T. molitor larvae. Since there were small changes in AMP gene expression but strong activation of the proPO activating system, PO may play a major role as defense mechanism of the host to fight the invading pathogenic A. hydrophila AH-3. This hypothesis could be supported by the study in 2007 which showed that PO was important for P. leniusculus to defend against A. hydrophila B1 infection .
In addition to induction of PO activity, the expression level of mRNA of the zymogen proPO and pacifastin was investigated. The expression of proPO was slightly down-regulated in WT infected crayfish, although these animals had a high PO activity, and as expected there was no obvious change in expression of this gene or the PO activity in ΔwaaE mutant infected crayfish. This higher activity of PO seems to be a result of that the pacifastin transcripts were decreased in the WT infected animals. Pacifastin is an efficient inhibitor of ppA (prophenoloxidase-activating enzyme) and this inhibitor prevents activation of the proPO system. Then if there is no pacifastin, activation of proPO can occur. In a study by Liu et al., , knockdown of pacifastin resulted in high PO activity and low bacterial number in crayfish hemolymph at 3 h after bacterial infection. Low expression of pacifastin in WT infected crayfish and high induction of PO activity in this study correlates well with this previous study. The reason for the slight down-regulation of proPO might occur because the proPO system can be activated continuously in the absence of pacifastin and the production of toxic products by PO may have reached levels in which those products are toxic to the animals and as a consequence the proPO transcripts were down-regulated.
Pacifastin of crayfish consist of two subunits containing a proteinase inhibitor light chain and a transferrin heavy chain. These two subunits are encoded by different mRNA transcripts , and as described above the expression level of the light chain transcript was decreased in WT infected crayfish but not in ΔwaaE mutant infected crayfish. In contrast, the heavy chain subunit was down regulated in crayfish infected with either WT or ΔwaaE mutant. The crayfish pacifastin heavy chain consists of three transferrin lobes of which two can bind iron . Iron is essential for bacterial metabolism and growth and A. hydrophila has iron chelating molecules called siderophores for iron acquisition from host iron-binding molecules such as transferrin or lactoferrin . A recent study in fish showed a reduction in serum iron and saturated transferrin following a bacterial infection . Therefore, the down regulation of transferrin subunits of pacifastin might be explained by the low iron level caused by WT or ΔwaaE mutant AH-3 infection.
In conclusion, mutation in several known virulent factors in an A. hydrophila strain in this study provides us a clear picture that O-antigen and external core of the LPS molecule are the A. hydrophila AH-3 most important factor for virulence in our animal models. A. hydrophila AH-3 WT, which contains a complete LPS core, strongly stimulated the proPO-activating system and melanization in P. leniusculus and T. molitor larvae, respectively. LPS, therefore, triggers host immune responses and protects bacteria against host defense process which then results in continuous growth of bacteria in the host circulating system and finally leads to death of the host.
Pacifastacus leniusculus (crayfish) was maintained in aquarium at 10°C with aerated running fresh water. Intermolt crayfish with weights ranging from 30–40 g were selected for the experiments. Tenebrio molitor larvae (mealworm) were maintained at room temperature (20–22°C) in beakers placed on the laboratory bench and they were fed with wheat bran. Mealworms with sizes from 2.5–3.5 cm were used for the experiments.
An inner DNA fragment of vasH, msbB, CirA, and uge was independently obtained from AH-3 chromosomal DNA using appropriate primers and subcloned in the pir replication dependent plasmid pSF100  through initial cloning in pGEM-T plasmid. These plasmid constructs (pSF-VasH, pSF-MsbB, pSF-CirA, pSF-Uge, respectively) were used to obtain vasH, msbB, cirA, uge deficient mutants from A. hydrophila strain AH-3 by a single recombination event leading to the generation of two incomplete copies of wild type genes in the chromosome of these mutants, as previously described . Plasmids pSF-VasH, pSF-MsbB, pSF-CirA, and pSF-Uge, were independently isolated, transformed into E. coli SM10 (λpir) , and transferred by conjugation from E. coli SM10 to the AH-3 rifampin-resistant (Rifr) mutant as previously described . Kmr Rifr transconjugants arising from the different conjugations using pSF-VasH, pSF-MsbB, pSF-CirA, and pSF-Uge, should contain the mobilized plasmid integrated onto the chromosome by homologous recombination between the wild type gene screened and the plasmid, leading to two incomplete copies of the wild type gene studied (defined insertion mutant). Chromosomal DNA from 10 transconjugants obtained were independently analyzed by Southern blot hybridization with appropriate vasH, msbB, CirA, and uge DNA probes to obtain the following defined insertion mutants: AH3ΔvasH (T6SS negative), AH3ΔmsbB (lipid alteration in LPS), AH3:: CirA (siderophore negative), and AH3::uge (lacking capsule), as previously described .
This study was performed in both crayfish and mealworm. All mutant and wild type strains of A. hydrophila AH-3 were grown at 28°C in tryptic soy broth (Fluka, 20 µg/ml of kanamycin was required for growing all mutant strains except for the T3SS negative one) until they reached a bacteria suspension with an optical density of 0.5–0.6 at 600 nm. The bacterial pellets were then collected and washed 3 times with 0.85% NaCl, and resuspended and diluted in buffer CFS (0.2 M NaCl, 5.4 mM KCl, 10 mM CaCl2.2H2O, 2.6 mM MgCl2.6H2O, 2 mM NaHCO3, pH 6.8) or in insect Ringer solution (128 mM NaCl, 18 mM CaCl2, 1.3 mM KCl, 2.3 mM NaHCO3, pH 7) for injection in crayfish or mealworm, respectively. Bacterial cell concentration for injection was determined and adjusted by performing viable plate counts and is reported as mean of colony-forming unit (CFU) ± SE.
The bacterial suspension (100 µl) was injected into the base of a crayfish walking leg, and 10 µl of the bacterial suspension or insect Ringer solution was injected using Ultra-Fine needle, 31G×8 mm, (BD Micro-Fine) into the abdominal part of Tenebrio larvae at the position between the third and the second segments. After injection, animals were normally maintained and fed, and were observed for 7 days. The time that animals died after the injection was recorded and reported as mean ± SE. The mortality of the animal is given as number of animals dead/total number of animals tested (No. dead/No. tested).
Two bacterial strains, A. hydrophila AH-3 WT (WT) and A. hydrophila AH-3 ΔwaaE mutant (ΔwaaE mutant) were chosen for further study because the WT is highly virulent whereas the ΔwaaE mutant completely lost its virulence. Extracellular products of the bacteria were prepared fresh as described by Jiravanichpaisal et al. . Briefly, bacteria, WT and ΔwaaE mutant, were cultured overnight in 10 ml TSB, and then the culture medium was centrifuged at 8000 g for 10 min at 4°C. The supernatant was collected and sterile filtered through 0.22-µm-pore membranes (Millipore). Then, 10 µl of sterile extracellular products of bacteria or 10 µl of TSB was incubated at room temperature (20–22°C) with freshly isolated crayfish hematopoietic (hpt) cells, which had been prepared as previously described by Söderhäll et al. . Briefly, the hpt was dissected out from crayfish and digested with 0.1% of collagenase type I and type IV at room temperature for 45 min, to get single cells. The cells were then washed 3 times with CPBS (10 mM Na2HPO4, 10 mM KH2PO4, 0.15 M NaCl, 10 µM CaCl2, 10 µM MnCl2, 2.7 µM KCl, pH 6.8). The cells were finally resuspended in modified L-15 culture medium  and seeded into 96-well plate with the amount of 3×104 cells/well. During the experiments, the morphology of hpt cells was observed every 15 min for 2 h. This experiment was repeated 3 times.
This experiment was performed in mealworm. A. hydrophila, WT (0.4–0.6×104 CFU/animal) or ΔwaaE mutant (1.8–2.9×104 CFU/animal) were injected into mealworm (n=3–7) as described above. At 1 h and 12 h after injection, the mealworms were bled, and 10 µl of hemolymph was collected from each individual worm to perform viable plate counts to determine CFU of bacteria in each mealworm. This experiment was repeated 2 times.
Crayfish (n=3 for each experimental group) was injected with bacteria WT (3.3×106 CFU/animal) or with ΔwaaE mutant (10.2×106 CFU/animal), and 6 h after injection hepatopancreas and hemolymph were collected and were kept separately. Total RNA was isolated from hemocytes and hepatopancreas using Trizol reagent (Gibco BRL) following the manufacturer's instruction. Each RNA sample was then treated with 2 units of RNase-free DNase I (Ambion) at 37°C for 30 min. Then, RNA was purified again using phenol-chloroform which was followed by ethanol precipitation.
First strand cDNA synthesis was performed using Oligo (dT)20 primer and all other reagents were from ThermoScript-PCR kit (Invitrogen) according to manufacturer's instruction. One microgram of RNA obtained from hepatopancreas or 100 ng obtained from hemocytes were used as starting material. cDNAs were then subjected to PCR using the primers shown in Table 3. The 40S ribosomal protein gene was used as an internal loading control for RT-PCR analysis. PCR was performed using the following condition: 95°C for 2 min, 22–38 cycles (see figure 4 and and77 legends for cycles of each gene) of 95°C for 20 s, 58°C for 20 s, 72°C for 30 s, and followed by a cycle of 72°C final extension for 3 min. The PCR products were then subjected to 1.5% agarose gel electrophoresis, followed by ethidium bromide staining and visualized under ultraviolet light.
To study the effect of a bacterial infection on melanization, the experiments were performed in mealworm because melanin formation could be seen easily in this animal. A. hydrophila WT (1×104 CFU/animal) or ΔwaaE mutant (1.7×104 CFU/animal) were injected into mealworms (n=10 each), and melanin formation was observed for 7 days or until the animals died.
A study of the effect on the prophenoloxidase (proPO) activating system after bacterial challenge was performed in crayfish because it was more convenient to get enough hemolymph from crayfish for the study when compared to that from mealworm. Activation of the proPO system in crayfish after bacterial challenge was done by preparing hemocyte lysate (HLS) and then determined LPS-activated PO activity in the HLS samples. A. hydrophila WT (5.8×106 CFU/animal) or ΔwaaE mutant (10.1×106 CFU/animal) or CFS was injected into 5 crayfishes each. Six hours after injection, crayfishes were bled and 10 drops of hemolymph from each individual were collected and pooled together in one experimental group and this sample was then centrifuged at 800×g for 20 min at 4°C and the resulting hemocyte pellet was homogenized and then centrifuged again at 16000×g for 20 min at 4°C. The obtained supernatant was HLS and was used for PO activity assay. The assay was performed by incubating 50 µl of HLS, 50 µl of 1 mg/ml LPS (E. coli 005:B5; Sigma), and 50 µl of 3 mg/ml L-3,4-dihydroxyphenylalanine (L-DOPA) (Sigma) at room temperature (20–22°C) for 10 min. As a control reaction, sterile distilled water was used instead of LPS. PO activity was determined by measuring the absorbance of dopachrome at 490 nm. The protein content in HLS was determined and PO activity was reported as ΔA490/mg protein/10 min. This experiment was repeated 2 times with a total amount of 10 crayfishes for each experimental group.
All experiments were repeated 2–3 times with 3–10 animals (for each experimental group) used as indicated in each experiment. All animals used in this study were independent from each other, and each animal was used for the measurement or for our study only one time. The T-test was performed when two experimental groups were compared, and One-way ANOVA followed by Turkey's multiple comparison was used for multiple comparisons. The P-value <0.05 was considered as a statistically significant difference.
Competing Interests: The authors have declared that no competing interests exist.
Funding: This work has been supported by the Swedish Science Research Council (K.S.), The Swedish Research Council Formas (K.S.) and The Swedish Foundation for International Cooperation in Research and Higher Education (K.S. and P.J.). Part of this work was supported by Plan Nacional de I+D and FIS grants (Ministerio de Educación, Ciencia y Deporte and Ministerio de Sanidad, Spain) and from Generalitat de Catalunya (Centre de Referència en Biotecnologia) to J.M.T.. The funders had no role in study design, data collection and analysis, decision of publish, or preparation of the manuscript.