- Email: [email protected]
Activation of the nitric oxide response in gilthead seabream after experimental infection with Photobacterium damselae subsp. piscicida
Fish & Shellfish Immunology 16 (2004) 581e588 www.elsevier.com/locate/fsi
Activation of the nitric oxide response in gilthead seabream after experimental infection with Photobacterium damselae subsp. piscicida F. Acostaa,), C.M. Ruiz de Galarretab, A.E. Ellisc, R. Dı´ azb, V. Go´meza, D. Padillaa, F. Reala a
Department of Animal Pathology, University of Las Palmas de Gran Canaria, Veterinary Faculty, Arucas, Las Palmas, Spain Department of Biochemistry, Molecular Biology and Physiology, University of Las Palmas de Gran Canaria, School of Medicine, Las Palmas, Spain c Marine Laboratory, Victoria Road, Aberdeen AB11 9DB, Scotland, UK
b
Received 2 June 2002; revised 18 August 2003; accepted 15 September 2003
Abstract Inoculation of small gilthead seabream (Sparus aurata) (30e75 g body weight) with a sublethal dose of different Photobacterium damselae subsp. piscicida (Pdp) strains (DI-21 and 94/99) induced an increase in serum concentrations of stable nitric oxide (NO) metabolites lasting from 6 h to six days post-infection, with a peak at 24 h. In contrast, no such response was detected in larger fish (150e600 g). Since the virulence of Pdp correlates with the presence of a polysaccharide capsular layer which can be induced by growing the bacteria in medium supplemented with 1% glucose (C+ forms), the effect of the presence of an enhanced capsular layer on the NO response in small fish was also evaluated. Although, all bacteria induced a similar rapid (6 h) and sustained (up to six days) NO response, serum concentrations of nitrites and citrulline were significantly increased in fish infected with the Pdp strains grown in glucose-supplemented medium. When the NO response of fish infected with the C+ form of Pdp was blocked by prior injection of the inhibitor L-NAME, the LD50 was reduced by over 10-fold and the mean time to death was also markedly reduced. Considering that (i) pasteurellosis only affects gilthead seabream with body weights below 100 g; (ii) capsulated Pdp are more resistant to the bactericidal action of NO and peroxynitrites than non-capsulated strains; and (iii) blocking the NO response of the fish results in greater susceptibility to Pdp, it seems reasonable to propose that the sustained NO response reported in this study represents a relevant protective mechanism of juvenile gilthead seabream against pasteurellosis. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Nitric oxide; Citrulline; Capsule; Photobacterium damselae subsp. piscicida; Gilthead seabream; LD50
) Corresponding author. University of Las Palmas de Gran Canaria, Veterinary Faculty, Trasmontan˜a s/n, 35416, Arucas, Spain. Tel.: +34-928451093; fax: +34-928451142. E-mail address: [email protected] (F. Acosta). 1050-4648/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.fsi.2003.09.010
582
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
1. Introduction It is now well established that the diffusible and highly reactive gaseous moiety nitric oxide (NO) is produced in living organisms by different and widely expressed NO-synthases (NOS) which catalyse the conversion of arginine to equimolar concentrations of citrulline and free NO [1]. Two NOS isoforms predominantly expressed in neurones (nNOS) and endothelial cells (eNOS) are constitutively expressed enzymes which generate transient and limited amounts of NO required to maintain the vascular tone [2], retrograde synaptic signalling in neurons [3e5] and tissue homeostasis [6]. In contrast, a third inducible enzyme (iNOS) is regulated at the transcriptional and transductional level in macrophages and a variety of non-immunocompetent cells and once expressed generates large and sustained amounts of NO involved in tissue remodelling, inflammation and host-defence [7,8]. The bactericidal action of macrophage-derived NO has been related to its capacity to directly react and inactivate important haem-containing enzymes such as cytochrome oxidases [9] and cytochrome P450 [10]. It can also combine with superoxide anions (O) 2 ) generated during the respiratory burst of phagocytes to produce the more toxic peroxynitrite (ONOO) radical, which attacks ironesulphur clusters of such important enzymes for bacterial survival as hydrogenases [11], aconitase [12] and ribonucleotide reductase [13]. It is becoming evident that fish macrophages respond to pathogens with a similar bactericidal armamentarium i.e. generation of reactive oxygen species (ROS), and reactive nitrogen species (RNS) as their mammalian counterparts [14e17]. A physiological role for iNOS as a component of the non-specific immune response in fish is further sustained by evidence obtained in cell-free experiments which demonstrated the bactericidal action of NO and other RNS on fish pathogens such as Renibacterium salmoninarum [18] or Aeromonas salmonicida [19]. Likewise using a macrophage-free assay [20], we recently reported the bactericidal action of NO and the more toxic peroxynitrite on different strains of Photobacterium damselae subsp. piscicida, the causative agent of pasteurellosis, a serious disease which affects a variety of commercial marine fish species [21]. With respect to pasteurellosis in gilthead seabream, it is apparent that only juvenile fish, less than 100 g in body weight succumb to infection, while larger fish are resistant [21]. The virulence of Pdp correlates with the expression of a polysaccharide capsular layer [22] which can be induced by culturing the pathogen in glucose-supplemented medium [23]. Furthermore, the bactericidal effect of NO and peroxynitrites was more apparent on strains cultured in normal medium (C forms) than in the same strains grown in medium supplemented with glucose to enhance capsule formation (C+ forms) [20]. Although these in vitro data suggest a role for RNS in the host response against Pdp, the mechanism(s) whereby fish macrophages protect against the infective action of the pathogen remains controversial [24], and so far the involvement of NO during the development of pasteurellosis has not been explored. The aim of the present study, therefore, was to determine whether activation of the L-arginineeNO pathway participates in the non-specific host-defence during experimentally induced pasteurellosis in gilthead seabream (Sparus aurata) of different sizes infected with different strains of Pdp cultured with (C+ forms) or without (C forms) glucose supplementation. Furthermore, the effect of blocking the NO response in fish by injection of the Nitric Oxide Synthase inhibitor, L-NAME, on the LD50 of the C+ form was investigated.
2. Materials and methods 2.1. Chemicals 2-Phenoxyethanol, Nu-nitro-L-arginine methyl ester (L-NAME), L-lactic dehydrogenase (LDH), Griess reagent, urease, diacetylmonoxime and thiosemicarbazide were obtained from Sigma Chemical Co
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
583
(St Louis, MO). Nitrate reductase, flavine adenine dinucleotide and other reagents used in this study were purchased from Calbiochem (San Diego, CA).
2.2. Bacteria The strains of P. damselae subsp. piscicida used in this study were 94/99 from our own collection and DI21 [25] generously provided by Dr. Alicia E. Toranzo (Department of Microbiology and Parasitology, University of Santiago de Compostela, Spain). Both Pdp strains were grown and maintained in Brain Heart Infusion Broth (BHIB) (Panreac SA, Barcelona, Spain) supplemented either with 1% NaCl alone (for culture of normal capsulated bacteria, C forms) or in combination with 1% glucose (enhanced capsulated bacteria, C+ forms) as recently described [20].
2.3. Experimental infection of fish Gilthead seabream (S. aurata) of different sizes were purchased from a local fish farm (GRAMACAN, S.A., Arguineguin, Spain) and transported to the research facilities located at the Instituto Canario de Ciencias Marinas, in continuously oxygenated 500 l tanks containing a low dose of the sedative 2-phenoxyethanol (50 mg P1) to prevent stress. Upon arrival, equal numbers of fish were placed in 1000 l tanks according to body weight (group A: 30e75 g; group B: 150e300 g; group C: 400e600 g) and maintained under standard oxygenation conditions at 22 (C with a commercial gilthead seabream pelleted diet (ProAqua, Palencia, Spain) provided once a day. Before starting the experiments, fish were acclimated for a minimum of 10 days to confirm the absence of any disease and thereafter placed (24e48 h) in oxygenated 500 l tanks according to size (groups A, B and C) and experimental group (30 fish/tank). Fish were anaesthetised (2-phenoxyethanol 300 ml l1) and injected intraperitoneally (i.p.) with either 100 ml phosphate buffered saline (PBS) (controls) or an equivalent volume of PBS containing 3!103 colony forming units (c.f.u.) of the different Pdp strains used. In selected experiments, fish were injected with the non-specific NOS inhibitor L-NAME (3 mM) (100 ml/fish) and infected with the different Pdp strains 1 h later. A group injected with L-NAME but not infected served as control. After the time-periods indicated (6 h and up to 20 days) the fish were anaesthetised and blood (0.5 ml) was collected from the caudal vein using sterile insulin syringes. Samples were maintained at 4 (C for 5 h, centrifuged at 5000!g, 15 min at 4 (C and the sera were stored at 20 (C for later analysis.
2.4. Serum nitrite/nitrate levels Total NO metabolites (nitrite + nitrate) in serum were determined by using the nitrate reductase method essentially as described previously [26]. Briefly, serum samples (100 ml, diluted 1:20 in PBS) and serial dilutions of sodium nitrate (1e100 mM) used as standard were added to 96-well culture plates (Bio-Rad, Hercules, USA), and thereafter incubated (37 (C for 1 h) with 3 ml of a freshly prepared reaction mixture containing nitrate reductase (1.5 mg ml1), NADPH (3.3 mg ml1) and flavine adenine dinucleotide (0.13 mg ml1). After reduction of nitrates, the excess of NADPH, which interferes with the nitrite determination, was eliminated by incubating (37 (C for 1 h) with a solution (11.5 ml) containing 25 mg ml1 LDH and sodium pyruvate (100 mM). Incubations were terminated by adding 100 ml of Griess reagent (1% sulphanilamide/0.1% naphthylethylene diamine dihydrochloride/2.5% H3PO4) and total NO metabolites were determined at 550 nm using a microplate reader (EL-312, Biotek Instruments, Winooski, VT).
584
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
2.5. Serum L-citrulline determination The citrulline by-product of NO biosynthesis was determined by a standard colorimetric assay [27]. Briefly, 400 ml fish serum or serial dilutions of L-citrulline (2.5e100 mM) were incubated with urease (45 UI ml1) for 30 min at 37 (C, and thereafter deproteinized by mixing with an equivalent volume of ice-cold 10% trichloroacetic acid. After centrifugation (10,000!g, 15 min at 4 (C) the clear supernatants (400 ml) were mixed with 3 ml of a chromogenic solution obtained by mixing one volume of 0.5% diacetylmonoxime/0.01% thiosemicarbazide and two volumes of acid-ferric solution (0.025% of FeCl3 in a solution containing 25% sulphuric acid and 20% phosphoric acid). The samples were boiled (96 (C for 5 min), allowed to reach room temperature and then 200 ml aliquots were pipetted into 96-well plates and the absorbance of standards and unknowns was determined at 570 nm using an EL-312 microplate reader (Biotek Instruments, Winooski, VT). 2.6. Effect of L-NAME on LD50 of Pdp Groups of small-sized fish (35e70 g; eight individuals/group) were i.p. injected with 100 ml PBS or 100 ml 3 mM L-NAME and 1 h later serial 10-fold dilutions of C+ Pdp 94/99 were injected i.p. (107e101 cfu/fish). Control groups were injected with either PBS or L-NAME and left unchallenged. Groups were maintained in separate tanks and mortalities were monitored for 20 days. Kidney from dead fish was sampled by plating onto Brain Heart Infusion agar + 1% NaCl to confirm that mortalities were due to Pdp. LD50 values were calculated by the method of Reed and Muench [29]. 2.7. Statistical analysis Results (meanGSD) are pooled data from three fish per sample time, four replicate readings per sample. Data analysis was performed applying a unifactorial ANOVA test and p ! 0:05 were considered significant and p ! 0:01 were considered very significant.
3. Results 3.1. Effect of infection with C form Pdp strains 94/99 and DI-21 on NOS activation in gilthead seabream (S. aurata) specimens of different sizes Twenty-four hours after infection with two strains of Pdp (94/99 and DI-21) grown in normal medium, there was a significant and equimolar increase in total serum concentrations of nitrites and L-citrulline (to 16 mM, from less than 1 mM in control fish) in the small fish specimens (group A, 35e70 g). In contrast, no changes in serum concentrations of NOS metabolites were observed in the medium-sized (group B, 150e300 g) and large-sized (group C, 400e600 g) fish (results not shown). 3.2. Time-dependent activation of the arginineeNO pathway in small-sized specimens of gilthead seabream (S. aurata) infected with C or C+ forms of Pdp strain 94/99 Small-sized fish (30e75 g) were used to determine a possible relationship between Pdp expressing different levels of capsule and NOS activation. To evaluate this, the 94/99 strain was grown under normal conditions (C form) or in the presence of 1% glucose to obtain enhanced capsulated Pdp (C+ form) and injected into the fish. As compared with PBS-injected controls, serum concentrations of nitrites and citrulline increased in a time-dependent and equimolar manner in the infected fish. The two by-products of
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
585
the NOS-catalysed reaction increased in the serum of gilthead seabream as early as 6 h post-inoculation with the C form (3:0G0:2 mM for nitrites and 2:9G0:4 mM for citrulline vs 0:5G0:13 mM and 0:45G0:1 mM, respectively, in control fish), reached maximum levels after 24 h and remained higher than in the PBStreated controls even after three days (p ! 0:01) and six days (p ! 0:05) (Fig. 1). Fish infected with the C+ bacteria had significantly greater serum concentrations of nitrites and citrulline by 6 h and 12 h (p ! 0:05 at both time-points) as compared with fish inoculated with the C form. This difference was even more apparent by 24 h (p ! 0:01). Serum concentrations of NOS metabolites remained higher in fish treated with the C+ form even at 48 h (p ! 0:01) when the serum concentrations of nitrites and citrulline started to decline. A significant difference between fish infected with C+ and C forms of Pdp was still apparent (p ! 0:05) at three and six days after inoculation, with no further differences between experimental groups observed at longer time-periods (up to 20 days) (results not shown). Regardless of the form of Pdp used, pretreatment (1 h) of fish with the non-specific NOS inhibitor L-NAME completely abolished the rise in serum nitrites and citrulline concentrations which were similar to those of PBS-treated fish at all time-periods studied (Fig. 1). The control group injected with L-NAME but not infected showed similar basal levels of NOS metabolites as the PBS-injected control fish (data not shown).
Fig. 1. Small-sized seabream (35e70 g) were injected with PBS (B) or an equivalent volume (100 ml) of PBS containing 3!103 cfu of either C form (P) or C+ form (,) Pdp 94/99 strain. Similar groups of fish of the same weight were pretreated with L-NAME (3 mM) 1 h before PBS treatment or infection with C or C+ pathogens (inset window in the upper and lower panel). Results (meanGSEM of three fish/sample point) represent serum concentrations of NOS metabolites (mM) determined at the time-periods indicated for total nitrites (upper panel) and L-citrulline (lower panel). Standard errors were less than the size of the data point symbols drawn and statistical differences are represented by *p ! 0:05 and **p ! 0:01.
586
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
During the course of these experiments, five fish died in the PBS-pretreated group and 12 fish died in the group following infection with the C+ form of Pdp. No mortalities occurred in any of the other groups during the course of the experiment. L-NAME-pretreated
3.3. Effect of L-NAME on LD50 of C+ Pdp 94/99 The LD50 of the C+ form of Pdp in fish pre-injected with PBS was 2!104:5 and the mean time to death was 8.5 days with a range of 3e14 days. The LD50 in fish pre-injected with L-NAME was 2!103 and the mean time to death was 4.5 days with a range of 12 h to nine days. Pdp was isolated from the kidney of all dead fish in pure culture.
4. Discussion While there is abundant information concerning pathogen-induced activation of the arginineeNO pathway in fish species such as the channel catfish [14] or rainbow trout [15e17], this is the first report demonstrating a similar response in the salt-water fish gilthead seabream (S. aurata) infected with sublethal concentrations of two different strains of Pdp. The response, measured as a significant increase in serum concentrations of stable metabolites of NO (NO 2 þ NO3 ) and equimolar concentrations of the inactive by-product citrulline, was induced only in small-sized seabream (30e70 g), 24 h following infection with Pdp. Fish larger than 150 g did not show any response. It is of interest that gilthead seabream are only susceptible to pasteurellosis when less than 100 g in body weight [21]. Larger seabream are resistant and the present data indicate that the clearance of sublethal doses of this pathogen in large fish may not require induction of the NO response. The presence of a polysaccharide capsule in Pdp has been associated with antiphagocytic properties and exacerbated pathogenic capacity [28]. Expression of the capsule can be enhanced by growing the bacterium in glucose-supplemented medium [23] and such forms (C+ forms) are less sensitive to the bactericidal actions of NO and RNS than the bacteria grown in normal medium [20]. It was therefore of interest to evaluate a possible relationship between the level of expression of the capsule and the NO response. For this purpose, the bacteria (strain 94/99) were grown in normal medium (C form) or medium supplemented with 1% glucose to induce enhanced synthesis of the polysaccharide capsule (C+ form) [23] and injected into small-sized fish (35e70 g). Both forms of the bacterium induced a rapid (6 h) and sustained (up to six days) increase in NO metabolites, and this effect was completely abolished in fish pretreated with the NOS inhibitor L-NAME. Moreover, serum levels of nitrite and citrulline were significantly higher in small gilthead seabream inoculated with the C+ bacteria than in those inoculated with the C form. It is to be expected that the C+ Pdp would be more resistant to being cleared by phagocytes than the C forms because of the antiphagocytic properties of the capsule [28]. If the C+ forms persisted longer after injection, they may produce a stronger stimulus for the NO response. A similar enhanced NO response has been reported in rainbow trout inoculated with a virulent strain of R. salmoninarum compared with avirulent strains [17]. An important virulence factor of R. salmoninarum (the causative agent of Bacterial Kidney Disease, BKD) is a surface protein (p57 protein) which also confers autoaggregating properties on the bacterium. Non-autoaggregating forms are avirulent [17]. Following injection of an avirulent strain into rainbow trout, serum nitrite levels increased from day 2 to day 12. However, following injection of the virulent strain, serum nitrite levels did not increase until day 8 and then continued to rise until day 21 when most of the fish died from BKD [17]. Thus, it appears that the NO response to R. salmoninarum subsides after the avirulent strains are cleared but continues until death when fish are given a lethal dose of the virulent strain.
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
587
In the present study, the dose of Pdp used to infect the fish was intended to be sublethal but during the course of the experiment to monitor the production of NOS metabolites over time in the serum of smallsized fish, five fish died in the group pretreated with PBS and 12 fish died in the group pretreated with L-NAME when they were infected with the C+ form of Pdp. No mortalities occurred in the uninfected control groups injected with PBS or L-NAME or in the groups infected with the C form of Pdp. This result suggested that blocking the NO response of the fish with L-NAME resulted in a greater susceptibility of the fish. In a further experiment to test this, the LD50 of the C+ Pdp 94/99 strain was determined in fish pretreated with either PBS or L-NAME. The LD50 in the latter group was over 10-fold lower and the mean time to death was reduced from 8.5 days in the PBS group to 4.5 days in the L-NAME-treated fish. This strongly implies that the NO response is involved in the innate defences of the small-sized seabream against pasteurellosis. Furthermore, the apparently greater virulence of the C+ form of Pdp in small-sized seabream found in this study is consistent with the previous work where we found that the C+ form is more resistant to RNS-mediated killing in a cell-free assay [20]. In conclusion, the present study provides the first evidence for the involvement of NO during the development of pasteurellosis in juvenile seabream of 30e70 g body weight but not in larger fish. Since Pdp is susceptible to being killed by RNS [20] and causes an acute disease only in gilthead seabream with body weights below 100 g [21], and the LD50 of Pdp is reduced in fish by blocking the NO response, it seems reasonable to propose that the rapid and sustained NO response reported in this study represents a relevant protective mechanism of the juvenile form of gilthead seabream against pasteurellosis.
Acknowledgements This work was supported by grants from the CICYT PM98/0233 (CMRG) and Consejerı´ a de Educacio´n del Gobierno de Canarias PI 2002/163 (FRV).
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
[13] [14]
Lancaster JR. A tutorial on the diffusibility and reactivity of free nitric oxide. Nitric Oxide 1997;1:18e30. Titheradge MA. Nitric oxide in septic shock. Biochim Biophys Acta 1999;5:437e55. Nathan C, Xie QW. Regulation of biosynthesis of nitric oxide. J Biol Chem 1994;269:13725e8. Mayer B, Hemmens B. Biosynthesis and action of nitric oxide in mammalian cells. Trends Biochem Sci 1997;22:477e81. Prast H, Philippu A. Nitric oxide as modulator of neuronal function. Prog Neurobiol 2001;64(1):51e68. Wiklund NP, Iversen HH, Leone AM, Cellek S, Brundin L, Gustafsson LE, et al. Visualization of nitric oxide formation in cell cultures and living tissue. Acta Physiol Scand 1999;167(2):161e6. Xie QW, Nathan C. The high-output nitric oxide pathway: role and regulation. J Leukoc Biol 1994;56:576e82. Schmidt HHHW, Walter U. NO at work. Cell 1994;78:919e25. Stuehr DJ, Nathan CF. Nitric oxide: a macrophage product responsible for cytostasis and respiratory inhibition in tumor target cells. J Exp Med 1989;169:1543e55. Wink DA, Osawa Y, Darbyshire JF, Jones CR, Eshenaur SC, Nims RW. Inhibition of cytochromes-P450 by nitric oxide and a nitric oxide releasing agent. Arch Biochem Biophys 1993;300:115e23. Hyman MR, Arp DJ. Kinetic analysis of the interaction of nitric oxide with the membrane-associated, nickel and irone sulfur-containing hydrogenase from Azotobacter vine-landii. Biochim Biophys Acta 1991;1076:165e72. Drapier JC, Hibbs JB. Differentiation of murine macrophages to express nonspecific cytotoxicity for tumor cells results in L-arginine dependent inhibition of mitochondrial ironesulfur enzymes in the macrophage effector cells. J Immunol 1988;140:2829e38. Lepoivre M, Fieschi F, Coves J, Thelander L, Fontecave M. Inactivation of ribonucleotide reductase by nitric oxide. Biochem Biophys Res Commun 1991;179:442e8. Schoor WP, Plumb JA. Indication of nitric oxide synthase in channel catfish Ictalurus punctatus by Edwarsiella ictaluri. Dis Aquat Organ 1994;19:153e5.
588
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
[15] Wang R, Neumann NF, Shen Q, Belosevic M. Establishment and characterisation of a macrophage cell line from goldfish. Fish Shellfish Immunol 1995;5:329e46. [16] Laing KJ, Hardie LJ, Aartsen W, Grabowski PS, Secombes CJ. Expression of an inducible nitric oxide synthase gene in rainbow trout Oncorhynchus mykiss. Dev Comp Immunol 1999;12:60e74. [17] Campos-Pe´rez JJ, Ward MP, Grabowski PS, Ellis AE, Secombes CJ. The gills are an important site of iNOS expression in rainbow trout Oncorhynchus mykiss after challenge with the Gram-positive pathogen Renibacterium salmoninarum. Immunology 2000;99:153e61. [18] Campos-Pe´rez JJ, Ellis AE, Secombes CJ. Toxicity of nitric oxide and peroxynitrite to bacterial pathogens of fish. Dis Aquat Organ 2000;43:109e15. [19] Barnes AC, Horne MT, Ellis AE. Effect of iron on expression of superoxide dismutase by Aeromonas salmonicida and associated resistance to superoxide anion. FEMS Microbiol Lett 1996;142:19e26. [20] Acosta F, Real F, Ruiz de Galarreta CM, Dı´ az R, Padilla D, Ellis AE. Toxicity of nitric oxide and peroxynitrite to Photobacterium damselae subsp. piscicida. Fish Shellfish Immunol 2003;3:241e3. [21] Daly JG. Other bacterial pathogens. In: Woo PTK, Bruno DW, editors. Fish disease and disorders. New York: CAB International; 1999. p. 577e98. [22] Magarin˜os B, Bonet R, Romalde JL, Martı´ nez MJ, Congregado F, Toranzo AE. Influence of the capsular layer on the virulence of Pasteurella piscicida for fish. Microb Pathog 1996;21:289e97. [23] Bonet R, Magarin˜os B, Romalde JL, Simon-Pujol MD, Toranzo AE, Congregado F. Capsular polysaccharide expressed by Pasteurella piscicida grown in vitro. FEMS Microbiol Lett 1994;124:285e9. [24] Barnes AC, Balebona MC, Horne MT, Ellis AE. Superoxide dismutase and catalase in Photobacterium damselae subsp. piscicida and their roles in resistance to reactive oxygen species. Microbiology 1999;145(2):483e94. [25] Toranzo AE, Barreiro S, Casal JF, Figueras A, Magarin˜os B, Barja J. Pasteurellosis in cultured gilthead seabream (Sparus aurata): first report in Spain. Aquaculture 1991;99:1e15. [26] Schmidt HHHW, Seifert R, Bohme EJN. Formation and release of nitric-oxide from human-neutrophils and HL-60 cells induced by a chemotactic peptide, platelet activating factor and leukotriene-B4. FEBS Lett 1989;244:357. [27] Boyde TR, Rahmatullah M. Optimization of conditions for the colorimetric determination of citrulline, using diacetyl monoxime. Anal Biochem 1980;15:424e31. [28] Arijo S, Borrego JJ, Zorilla I, Balebona MC, Morin˜igo MA. Role of the capsule of Photobacterium damselae subsp. piscicida in protection against phagocytosis and killing by gilt-head seabream Sparus aurata macrophages. Fish Shellfish Immunol 1998;8:63e72. [29] Reed LJ, Muench H. A simple method of estimating fifty percent endpoints. Am J Hyg 1938;27:493e7.
Activation of the nitric oxide response in gilthead seabream after experimental infection with Photobacterium damselae subsp. piscicida F. Acostaa,), C.M. Ruiz de Galarretab, A.E. Ellisc, R. Dı´ azb, V. Go´meza, D. Padillaa, F. Reala a
Department of Animal Pathology, University of Las Palmas de Gran Canaria, Veterinary Faculty, Arucas, Las Palmas, Spain Department of Biochemistry, Molecular Biology and Physiology, University of Las Palmas de Gran Canaria, School of Medicine, Las Palmas, Spain c Marine Laboratory, Victoria Road, Aberdeen AB11 9DB, Scotland, UK
b
Received 2 June 2002; revised 18 August 2003; accepted 15 September 2003
Abstract Inoculation of small gilthead seabream (Sparus aurata) (30e75 g body weight) with a sublethal dose of different Photobacterium damselae subsp. piscicida (Pdp) strains (DI-21 and 94/99) induced an increase in serum concentrations of stable nitric oxide (NO) metabolites lasting from 6 h to six days post-infection, with a peak at 24 h. In contrast, no such response was detected in larger fish (150e600 g). Since the virulence of Pdp correlates with the presence of a polysaccharide capsular layer which can be induced by growing the bacteria in medium supplemented with 1% glucose (C+ forms), the effect of the presence of an enhanced capsular layer on the NO response in small fish was also evaluated. Although, all bacteria induced a similar rapid (6 h) and sustained (up to six days) NO response, serum concentrations of nitrites and citrulline were significantly increased in fish infected with the Pdp strains grown in glucose-supplemented medium. When the NO response of fish infected with the C+ form of Pdp was blocked by prior injection of the inhibitor L-NAME, the LD50 was reduced by over 10-fold and the mean time to death was also markedly reduced. Considering that (i) pasteurellosis only affects gilthead seabream with body weights below 100 g; (ii) capsulated Pdp are more resistant to the bactericidal action of NO and peroxynitrites than non-capsulated strains; and (iii) blocking the NO response of the fish results in greater susceptibility to Pdp, it seems reasonable to propose that the sustained NO response reported in this study represents a relevant protective mechanism of juvenile gilthead seabream against pasteurellosis. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Nitric oxide; Citrulline; Capsule; Photobacterium damselae subsp. piscicida; Gilthead seabream; LD50
) Corresponding author. University of Las Palmas de Gran Canaria, Veterinary Faculty, Trasmontan˜a s/n, 35416, Arucas, Spain. Tel.: +34-928451093; fax: +34-928451142. E-mail address: [email protected] (F. Acosta). 1050-4648/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.fsi.2003.09.010
582
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
1. Introduction It is now well established that the diffusible and highly reactive gaseous moiety nitric oxide (NO) is produced in living organisms by different and widely expressed NO-synthases (NOS) which catalyse the conversion of arginine to equimolar concentrations of citrulline and free NO [1]. Two NOS isoforms predominantly expressed in neurones (nNOS) and endothelial cells (eNOS) are constitutively expressed enzymes which generate transient and limited amounts of NO required to maintain the vascular tone [2], retrograde synaptic signalling in neurons [3e5] and tissue homeostasis [6]. In contrast, a third inducible enzyme (iNOS) is regulated at the transcriptional and transductional level in macrophages and a variety of non-immunocompetent cells and once expressed generates large and sustained amounts of NO involved in tissue remodelling, inflammation and host-defence [7,8]. The bactericidal action of macrophage-derived NO has been related to its capacity to directly react and inactivate important haem-containing enzymes such as cytochrome oxidases [9] and cytochrome P450 [10]. It can also combine with superoxide anions (O) 2 ) generated during the respiratory burst of phagocytes to produce the more toxic peroxynitrite (ONOO) radical, which attacks ironesulphur clusters of such important enzymes for bacterial survival as hydrogenases [11], aconitase [12] and ribonucleotide reductase [13]. It is becoming evident that fish macrophages respond to pathogens with a similar bactericidal armamentarium i.e. generation of reactive oxygen species (ROS), and reactive nitrogen species (RNS) as their mammalian counterparts [14e17]. A physiological role for iNOS as a component of the non-specific immune response in fish is further sustained by evidence obtained in cell-free experiments which demonstrated the bactericidal action of NO and other RNS on fish pathogens such as Renibacterium salmoninarum [18] or Aeromonas salmonicida [19]. Likewise using a macrophage-free assay [20], we recently reported the bactericidal action of NO and the more toxic peroxynitrite on different strains of Photobacterium damselae subsp. piscicida, the causative agent of pasteurellosis, a serious disease which affects a variety of commercial marine fish species [21]. With respect to pasteurellosis in gilthead seabream, it is apparent that only juvenile fish, less than 100 g in body weight succumb to infection, while larger fish are resistant [21]. The virulence of Pdp correlates with the expression of a polysaccharide capsular layer [22] which can be induced by culturing the pathogen in glucose-supplemented medium [23]. Furthermore, the bactericidal effect of NO and peroxynitrites was more apparent on strains cultured in normal medium (C forms) than in the same strains grown in medium supplemented with glucose to enhance capsule formation (C+ forms) [20]. Although these in vitro data suggest a role for RNS in the host response against Pdp, the mechanism(s) whereby fish macrophages protect against the infective action of the pathogen remains controversial [24], and so far the involvement of NO during the development of pasteurellosis has not been explored. The aim of the present study, therefore, was to determine whether activation of the L-arginineeNO pathway participates in the non-specific host-defence during experimentally induced pasteurellosis in gilthead seabream (Sparus aurata) of different sizes infected with different strains of Pdp cultured with (C+ forms) or without (C forms) glucose supplementation. Furthermore, the effect of blocking the NO response in fish by injection of the Nitric Oxide Synthase inhibitor, L-NAME, on the LD50 of the C+ form was investigated.
2. Materials and methods 2.1. Chemicals 2-Phenoxyethanol, Nu-nitro-L-arginine methyl ester (L-NAME), L-lactic dehydrogenase (LDH), Griess reagent, urease, diacetylmonoxime and thiosemicarbazide were obtained from Sigma Chemical Co
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
583
(St Louis, MO). Nitrate reductase, flavine adenine dinucleotide and other reagents used in this study were purchased from Calbiochem (San Diego, CA).
2.2. Bacteria The strains of P. damselae subsp. piscicida used in this study were 94/99 from our own collection and DI21 [25] generously provided by Dr. Alicia E. Toranzo (Department of Microbiology and Parasitology, University of Santiago de Compostela, Spain). Both Pdp strains were grown and maintained in Brain Heart Infusion Broth (BHIB) (Panreac SA, Barcelona, Spain) supplemented either with 1% NaCl alone (for culture of normal capsulated bacteria, C forms) or in combination with 1% glucose (enhanced capsulated bacteria, C+ forms) as recently described [20].
2.3. Experimental infection of fish Gilthead seabream (S. aurata) of different sizes were purchased from a local fish farm (GRAMACAN, S.A., Arguineguin, Spain) and transported to the research facilities located at the Instituto Canario de Ciencias Marinas, in continuously oxygenated 500 l tanks containing a low dose of the sedative 2-phenoxyethanol (50 mg P1) to prevent stress. Upon arrival, equal numbers of fish were placed in 1000 l tanks according to body weight (group A: 30e75 g; group B: 150e300 g; group C: 400e600 g) and maintained under standard oxygenation conditions at 22 (C with a commercial gilthead seabream pelleted diet (ProAqua, Palencia, Spain) provided once a day. Before starting the experiments, fish were acclimated for a minimum of 10 days to confirm the absence of any disease and thereafter placed (24e48 h) in oxygenated 500 l tanks according to size (groups A, B and C) and experimental group (30 fish/tank). Fish were anaesthetised (2-phenoxyethanol 300 ml l1) and injected intraperitoneally (i.p.) with either 100 ml phosphate buffered saline (PBS) (controls) or an equivalent volume of PBS containing 3!103 colony forming units (c.f.u.) of the different Pdp strains used. In selected experiments, fish were injected with the non-specific NOS inhibitor L-NAME (3 mM) (100 ml/fish) and infected with the different Pdp strains 1 h later. A group injected with L-NAME but not infected served as control. After the time-periods indicated (6 h and up to 20 days) the fish were anaesthetised and blood (0.5 ml) was collected from the caudal vein using sterile insulin syringes. Samples were maintained at 4 (C for 5 h, centrifuged at 5000!g, 15 min at 4 (C and the sera were stored at 20 (C for later analysis.
2.4. Serum nitrite/nitrate levels Total NO metabolites (nitrite + nitrate) in serum were determined by using the nitrate reductase method essentially as described previously [26]. Briefly, serum samples (100 ml, diluted 1:20 in PBS) and serial dilutions of sodium nitrate (1e100 mM) used as standard were added to 96-well culture plates (Bio-Rad, Hercules, USA), and thereafter incubated (37 (C for 1 h) with 3 ml of a freshly prepared reaction mixture containing nitrate reductase (1.5 mg ml1), NADPH (3.3 mg ml1) and flavine adenine dinucleotide (0.13 mg ml1). After reduction of nitrates, the excess of NADPH, which interferes with the nitrite determination, was eliminated by incubating (37 (C for 1 h) with a solution (11.5 ml) containing 25 mg ml1 LDH and sodium pyruvate (100 mM). Incubations were terminated by adding 100 ml of Griess reagent (1% sulphanilamide/0.1% naphthylethylene diamine dihydrochloride/2.5% H3PO4) and total NO metabolites were determined at 550 nm using a microplate reader (EL-312, Biotek Instruments, Winooski, VT).
584
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
2.5. Serum L-citrulline determination The citrulline by-product of NO biosynthesis was determined by a standard colorimetric assay [27]. Briefly, 400 ml fish serum or serial dilutions of L-citrulline (2.5e100 mM) were incubated with urease (45 UI ml1) for 30 min at 37 (C, and thereafter deproteinized by mixing with an equivalent volume of ice-cold 10% trichloroacetic acid. After centrifugation (10,000!g, 15 min at 4 (C) the clear supernatants (400 ml) were mixed with 3 ml of a chromogenic solution obtained by mixing one volume of 0.5% diacetylmonoxime/0.01% thiosemicarbazide and two volumes of acid-ferric solution (0.025% of FeCl3 in a solution containing 25% sulphuric acid and 20% phosphoric acid). The samples were boiled (96 (C for 5 min), allowed to reach room temperature and then 200 ml aliquots were pipetted into 96-well plates and the absorbance of standards and unknowns was determined at 570 nm using an EL-312 microplate reader (Biotek Instruments, Winooski, VT). 2.6. Effect of L-NAME on LD50 of Pdp Groups of small-sized fish (35e70 g; eight individuals/group) were i.p. injected with 100 ml PBS or 100 ml 3 mM L-NAME and 1 h later serial 10-fold dilutions of C+ Pdp 94/99 were injected i.p. (107e101 cfu/fish). Control groups were injected with either PBS or L-NAME and left unchallenged. Groups were maintained in separate tanks and mortalities were monitored for 20 days. Kidney from dead fish was sampled by plating onto Brain Heart Infusion agar + 1% NaCl to confirm that mortalities were due to Pdp. LD50 values were calculated by the method of Reed and Muench [29]. 2.7. Statistical analysis Results (meanGSD) are pooled data from three fish per sample time, four replicate readings per sample. Data analysis was performed applying a unifactorial ANOVA test and p ! 0:05 were considered significant and p ! 0:01 were considered very significant.
3. Results 3.1. Effect of infection with C form Pdp strains 94/99 and DI-21 on NOS activation in gilthead seabream (S. aurata) specimens of different sizes Twenty-four hours after infection with two strains of Pdp (94/99 and DI-21) grown in normal medium, there was a significant and equimolar increase in total serum concentrations of nitrites and L-citrulline (to 16 mM, from less than 1 mM in control fish) in the small fish specimens (group A, 35e70 g). In contrast, no changes in serum concentrations of NOS metabolites were observed in the medium-sized (group B, 150e300 g) and large-sized (group C, 400e600 g) fish (results not shown). 3.2. Time-dependent activation of the arginineeNO pathway in small-sized specimens of gilthead seabream (S. aurata) infected with C or C+ forms of Pdp strain 94/99 Small-sized fish (30e75 g) were used to determine a possible relationship between Pdp expressing different levels of capsule and NOS activation. To evaluate this, the 94/99 strain was grown under normal conditions (C form) or in the presence of 1% glucose to obtain enhanced capsulated Pdp (C+ form) and injected into the fish. As compared with PBS-injected controls, serum concentrations of nitrites and citrulline increased in a time-dependent and equimolar manner in the infected fish. The two by-products of
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
585
the NOS-catalysed reaction increased in the serum of gilthead seabream as early as 6 h post-inoculation with the C form (3:0G0:2 mM for nitrites and 2:9G0:4 mM for citrulline vs 0:5G0:13 mM and 0:45G0:1 mM, respectively, in control fish), reached maximum levels after 24 h and remained higher than in the PBStreated controls even after three days (p ! 0:01) and six days (p ! 0:05) (Fig. 1). Fish infected with the C+ bacteria had significantly greater serum concentrations of nitrites and citrulline by 6 h and 12 h (p ! 0:05 at both time-points) as compared with fish inoculated with the C form. This difference was even more apparent by 24 h (p ! 0:01). Serum concentrations of NOS metabolites remained higher in fish treated with the C+ form even at 48 h (p ! 0:01) when the serum concentrations of nitrites and citrulline started to decline. A significant difference between fish infected with C+ and C forms of Pdp was still apparent (p ! 0:05) at three and six days after inoculation, with no further differences between experimental groups observed at longer time-periods (up to 20 days) (results not shown). Regardless of the form of Pdp used, pretreatment (1 h) of fish with the non-specific NOS inhibitor L-NAME completely abolished the rise in serum nitrites and citrulline concentrations which were similar to those of PBS-treated fish at all time-periods studied (Fig. 1). The control group injected with L-NAME but not infected showed similar basal levels of NOS metabolites as the PBS-injected control fish (data not shown).
Fig. 1. Small-sized seabream (35e70 g) were injected with PBS (B) or an equivalent volume (100 ml) of PBS containing 3!103 cfu of either C form (P) or C+ form (,) Pdp 94/99 strain. Similar groups of fish of the same weight were pretreated with L-NAME (3 mM) 1 h before PBS treatment or infection with C or C+ pathogens (inset window in the upper and lower panel). Results (meanGSEM of three fish/sample point) represent serum concentrations of NOS metabolites (mM) determined at the time-periods indicated for total nitrites (upper panel) and L-citrulline (lower panel). Standard errors were less than the size of the data point symbols drawn and statistical differences are represented by *p ! 0:05 and **p ! 0:01.
586
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
During the course of these experiments, five fish died in the PBS-pretreated group and 12 fish died in the group following infection with the C+ form of Pdp. No mortalities occurred in any of the other groups during the course of the experiment. L-NAME-pretreated
3.3. Effect of L-NAME on LD50 of C+ Pdp 94/99 The LD50 of the C+ form of Pdp in fish pre-injected with PBS was 2!104:5 and the mean time to death was 8.5 days with a range of 3e14 days. The LD50 in fish pre-injected with L-NAME was 2!103 and the mean time to death was 4.5 days with a range of 12 h to nine days. Pdp was isolated from the kidney of all dead fish in pure culture.
4. Discussion While there is abundant information concerning pathogen-induced activation of the arginineeNO pathway in fish species such as the channel catfish [14] or rainbow trout [15e17], this is the first report demonstrating a similar response in the salt-water fish gilthead seabream (S. aurata) infected with sublethal concentrations of two different strains of Pdp. The response, measured as a significant increase in serum concentrations of stable metabolites of NO (NO 2 þ NO3 ) and equimolar concentrations of the inactive by-product citrulline, was induced only in small-sized seabream (30e70 g), 24 h following infection with Pdp. Fish larger than 150 g did not show any response. It is of interest that gilthead seabream are only susceptible to pasteurellosis when less than 100 g in body weight [21]. Larger seabream are resistant and the present data indicate that the clearance of sublethal doses of this pathogen in large fish may not require induction of the NO response. The presence of a polysaccharide capsule in Pdp has been associated with antiphagocytic properties and exacerbated pathogenic capacity [28]. Expression of the capsule can be enhanced by growing the bacterium in glucose-supplemented medium [23] and such forms (C+ forms) are less sensitive to the bactericidal actions of NO and RNS than the bacteria grown in normal medium [20]. It was therefore of interest to evaluate a possible relationship between the level of expression of the capsule and the NO response. For this purpose, the bacteria (strain 94/99) were grown in normal medium (C form) or medium supplemented with 1% glucose to induce enhanced synthesis of the polysaccharide capsule (C+ form) [23] and injected into small-sized fish (35e70 g). Both forms of the bacterium induced a rapid (6 h) and sustained (up to six days) increase in NO metabolites, and this effect was completely abolished in fish pretreated with the NOS inhibitor L-NAME. Moreover, serum levels of nitrite and citrulline were significantly higher in small gilthead seabream inoculated with the C+ bacteria than in those inoculated with the C form. It is to be expected that the C+ Pdp would be more resistant to being cleared by phagocytes than the C forms because of the antiphagocytic properties of the capsule [28]. If the C+ forms persisted longer after injection, they may produce a stronger stimulus for the NO response. A similar enhanced NO response has been reported in rainbow trout inoculated with a virulent strain of R. salmoninarum compared with avirulent strains [17]. An important virulence factor of R. salmoninarum (the causative agent of Bacterial Kidney Disease, BKD) is a surface protein (p57 protein) which also confers autoaggregating properties on the bacterium. Non-autoaggregating forms are avirulent [17]. Following injection of an avirulent strain into rainbow trout, serum nitrite levels increased from day 2 to day 12. However, following injection of the virulent strain, serum nitrite levels did not increase until day 8 and then continued to rise until day 21 when most of the fish died from BKD [17]. Thus, it appears that the NO response to R. salmoninarum subsides after the avirulent strains are cleared but continues until death when fish are given a lethal dose of the virulent strain.
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
587
In the present study, the dose of Pdp used to infect the fish was intended to be sublethal but during the course of the experiment to monitor the production of NOS metabolites over time in the serum of smallsized fish, five fish died in the group pretreated with PBS and 12 fish died in the group pretreated with L-NAME when they were infected with the C+ form of Pdp. No mortalities occurred in the uninfected control groups injected with PBS or L-NAME or in the groups infected with the C form of Pdp. This result suggested that blocking the NO response of the fish with L-NAME resulted in a greater susceptibility of the fish. In a further experiment to test this, the LD50 of the C+ Pdp 94/99 strain was determined in fish pretreated with either PBS or L-NAME. The LD50 in the latter group was over 10-fold lower and the mean time to death was reduced from 8.5 days in the PBS group to 4.5 days in the L-NAME-treated fish. This strongly implies that the NO response is involved in the innate defences of the small-sized seabream against pasteurellosis. Furthermore, the apparently greater virulence of the C+ form of Pdp in small-sized seabream found in this study is consistent with the previous work where we found that the C+ form is more resistant to RNS-mediated killing in a cell-free assay [20]. In conclusion, the present study provides the first evidence for the involvement of NO during the development of pasteurellosis in juvenile seabream of 30e70 g body weight but not in larger fish. Since Pdp is susceptible to being killed by RNS [20] and causes an acute disease only in gilthead seabream with body weights below 100 g [21], and the LD50 of Pdp is reduced in fish by blocking the NO response, it seems reasonable to propose that the rapid and sustained NO response reported in this study represents a relevant protective mechanism of the juvenile form of gilthead seabream against pasteurellosis.
Acknowledgements This work was supported by grants from the CICYT PM98/0233 (CMRG) and Consejerı´ a de Educacio´n del Gobierno de Canarias PI 2002/163 (FRV).
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
[13] [14]
Lancaster JR. A tutorial on the diffusibility and reactivity of free nitric oxide. Nitric Oxide 1997;1:18e30. Titheradge MA. Nitric oxide in septic shock. Biochim Biophys Acta 1999;5:437e55. Nathan C, Xie QW. Regulation of biosynthesis of nitric oxide. J Biol Chem 1994;269:13725e8. Mayer B, Hemmens B. Biosynthesis and action of nitric oxide in mammalian cells. Trends Biochem Sci 1997;22:477e81. Prast H, Philippu A. Nitric oxide as modulator of neuronal function. Prog Neurobiol 2001;64(1):51e68. Wiklund NP, Iversen HH, Leone AM, Cellek S, Brundin L, Gustafsson LE, et al. Visualization of nitric oxide formation in cell cultures and living tissue. Acta Physiol Scand 1999;167(2):161e6. Xie QW, Nathan C. The high-output nitric oxide pathway: role and regulation. J Leukoc Biol 1994;56:576e82. Schmidt HHHW, Walter U. NO at work. Cell 1994;78:919e25. Stuehr DJ, Nathan CF. Nitric oxide: a macrophage product responsible for cytostasis and respiratory inhibition in tumor target cells. J Exp Med 1989;169:1543e55. Wink DA, Osawa Y, Darbyshire JF, Jones CR, Eshenaur SC, Nims RW. Inhibition of cytochromes-P450 by nitric oxide and a nitric oxide releasing agent. Arch Biochem Biophys 1993;300:115e23. Hyman MR, Arp DJ. Kinetic analysis of the interaction of nitric oxide with the membrane-associated, nickel and irone sulfur-containing hydrogenase from Azotobacter vine-landii. Biochim Biophys Acta 1991;1076:165e72. Drapier JC, Hibbs JB. Differentiation of murine macrophages to express nonspecific cytotoxicity for tumor cells results in L-arginine dependent inhibition of mitochondrial ironesulfur enzymes in the macrophage effector cells. J Immunol 1988;140:2829e38. Lepoivre M, Fieschi F, Coves J, Thelander L, Fontecave M. Inactivation of ribonucleotide reductase by nitric oxide. Biochem Biophys Res Commun 1991;179:442e8. Schoor WP, Plumb JA. Indication of nitric oxide synthase in channel catfish Ictalurus punctatus by Edwarsiella ictaluri. Dis Aquat Organ 1994;19:153e5.
588
F. Acosta et al. / Fish & Shellfish Immunology 16 (2004) 581e588
[15] Wang R, Neumann NF, Shen Q, Belosevic M. Establishment and characterisation of a macrophage cell line from goldfish. Fish Shellfish Immunol 1995;5:329e46. [16] Laing KJ, Hardie LJ, Aartsen W, Grabowski PS, Secombes CJ. Expression of an inducible nitric oxide synthase gene in rainbow trout Oncorhynchus mykiss. Dev Comp Immunol 1999;12:60e74. [17] Campos-Pe´rez JJ, Ward MP, Grabowski PS, Ellis AE, Secombes CJ. The gills are an important site of iNOS expression in rainbow trout Oncorhynchus mykiss after challenge with the Gram-positive pathogen Renibacterium salmoninarum. Immunology 2000;99:153e61. [18] Campos-Pe´rez JJ, Ellis AE, Secombes CJ. Toxicity of nitric oxide and peroxynitrite to bacterial pathogens of fish. Dis Aquat Organ 2000;43:109e15. [19] Barnes AC, Horne MT, Ellis AE. Effect of iron on expression of superoxide dismutase by Aeromonas salmonicida and associated resistance to superoxide anion. FEMS Microbiol Lett 1996;142:19e26. [20] Acosta F, Real F, Ruiz de Galarreta CM, Dı´ az R, Padilla D, Ellis AE. Toxicity of nitric oxide and peroxynitrite to Photobacterium damselae subsp. piscicida. Fish Shellfish Immunol 2003;3:241e3. [21] Daly JG. Other bacterial pathogens. In: Woo PTK, Bruno DW, editors. Fish disease and disorders. New York: CAB International; 1999. p. 577e98. [22] Magarin˜os B, Bonet R, Romalde JL, Martı´ nez MJ, Congregado F, Toranzo AE. Influence of the capsular layer on the virulence of Pasteurella piscicida for fish. Microb Pathog 1996;21:289e97. [23] Bonet R, Magarin˜os B, Romalde JL, Simon-Pujol MD, Toranzo AE, Congregado F. Capsular polysaccharide expressed by Pasteurella piscicida grown in vitro. FEMS Microbiol Lett 1994;124:285e9. [24] Barnes AC, Balebona MC, Horne MT, Ellis AE. Superoxide dismutase and catalase in Photobacterium damselae subsp. piscicida and their roles in resistance to reactive oxygen species. Microbiology 1999;145(2):483e94. [25] Toranzo AE, Barreiro S, Casal JF, Figueras A, Magarin˜os B, Barja J. Pasteurellosis in cultured gilthead seabream (Sparus aurata): first report in Spain. Aquaculture 1991;99:1e15. [26] Schmidt HHHW, Seifert R, Bohme EJN. Formation and release of nitric-oxide from human-neutrophils and HL-60 cells induced by a chemotactic peptide, platelet activating factor and leukotriene-B4. FEBS Lett 1989;244:357. [27] Boyde TR, Rahmatullah M. Optimization of conditions for the colorimetric determination of citrulline, using diacetyl monoxime. Anal Biochem 1980;15:424e31. [28] Arijo S, Borrego JJ, Zorilla I, Balebona MC, Morin˜igo MA. Role of the capsule of Photobacterium damselae subsp. piscicida in protection against phagocytosis and killing by gilt-head seabream Sparus aurata macrophages. Fish Shellfish Immunol 1998;8:63e72. [29] Reed LJ, Muench H. A simple method of estimating fifty percent endpoints. Am J Hyg 1938;27:493e7.