Chitin is one of the main complex polysaccharides in naturewhere it is found in the cuticle of arthropods and in fungal cellwalls. In the crustacean cuticle, this polymer of β 1, 4 N-acetyl-glucosamine accounts for over 50% of the organic content(Muzzarelli, 1977). Chitin has been found to be either a growth-enhancing or a growth-suppressive dietary component in crus-taceans and fish (e.g. Shiau and Yu, 1998). Observations fromthe wild have revealed that some crustaceans eat their shedcuticle following ecdysis suggesting a beneficial dietary effect.The addition of 5% chitin to a basal diet in the shrimp, Penaeusmonodon, enhanced the feed efficiency, protein efficiency ratioand growth during an eight week experiment (Shiau and Yu,1998). Dietary supplementationwith chitin and chitosan (formedby acetylation of chitin under alkaline conditions) has, however,been found to significantly depress the growth of tilapia (Shiau
and Yu, 1999) suggesting that the effects of these two compo-nents may differ dramatically from species to species.
As well as the effect of chitin in nutrition, both chitin andchitosan have been found to have immune stimulatory activityin fish and shellfish (reviewed by Sakai, 1999). For instance,white shrimp, Litopenaeus vannamei, injected with chitin orchitosan showed a short-term enhancement in survival follow-ing challenge with Vibrio alginolyticus and higher blood cell(hemocyte) counts, respiratory burst and phagocytic activity(Wang and Chen, 2005). Several reports have shown immunestimulatory activity of chitin and/or chitosan in a range of fishincluding rainbow trout, Oncorhynchus mykiss (Sakai et al.,1992) and gilthead seabream, Sparus aurata (Esteban et al.,2001).
This current study examines the potential effects of dietaryincorporation of chitin on the survival and immune potential ofthe European green shore crab, Carcinus maenas, as a modelspecies. Although this species is currently not subject to largescale commercial culture, recent developments have sought toascertain the feasibility of such culture for a range of decapodcrustaceans (Wilkins and Lee, 2002) and therefore there is a
123A. Powell, A.F. Rowley / Comparative Biochemistry and Physiology, Part A 147 (2007) 122–128
requirement for the development of cost effective diets for suchanimals.
2. Materials and methods
2.1. Animals
Shore crabs, C. maenas, were caught in baited traps fromQueen′s Dock, Swansea, UK. Only adult males (carapace width60±10 mm) that had both chelae present were used. Prior totheir use, all crabs were held for ca. 2 weeks and were fed adlibitum on edible mussels, Mytilus edulis. Three different dietgroups were created, each consisting of 20 randomly selectedindividuals. The groups were housed separately in identicalaquaria. Seawater was supplied using a recirculating systemunder constant aeration. Individuals were marked with a uniquepattern, colour coded with respect to diet group. Aquaria werechecked and cleaned daily for uneaten feed and dead animals.During the trials, the seawater temperature in the aquariumvaried from 15 to 17 °C.
2.2. Diets
Sufficient diet was prepared to last the duration of the diettrial. Any skin or bone was removed from defrosted haddock(Melanogrammus aeglefinus) fillets and the flesh briefly disso-ciated using a food processor. Excess water was removed byblotting. Chitin powder from crab shells (Sigma-Aldrich, Poole,Dorset, UK) was added in powdered form to give 5 or 10% finalconcentrations. Control feed (i.e. 0% chitin supplement) con-tained haddock flesh only. The solids were then placed in plasticfood containers to a depth of ca. 4 cm, mixed with a solution of5% dissolved 300 bloom porcine gelatin (Sigma) at 45 °C, thatwas left to solidify at 4 °C and stored at −15 °C. Feed wasdefrosted for approximately 3 h, cut into 4×1×1 cm cubes.Each crab was fed ca. 3 g on every alternate day throughout thetrial.
2.3. Bleeding regime
Hemolymph was removed from the unsclerotised membranebetween the dorsal carapace and the base of the fourth walkinglimb. This reduced the risk of autotomy of the chelae, and henceany reduction in feeding efficiency. The membrane was swab-bed with 70% ethanol before bleeding, and crabs were bled inrotation according to number rather than feed group, so thattime of bleeding did not bias the results. Crabs were bled at astandard time once a week for the duration of the trial.
2.4. Total (THC) and differential (DHC) hemocyte counts
Hemolymph (200 μL) was obtained via a 21-gauge needleand syringe containing an equal volume of ice-cold sterilemarine anticoagulant (MAC; 0.45 M NaCl, 0.1 M glucose,30 mM sodium citrate, 26 mM citric acid, 10 mM EDTAsodium salt, pH 4.6) containing 8% formalin (final concentra-tion). The syringe was emptied and the contents mixed carefully
into sterile Eppendorf tubes. Aliquots were then counted on aNeubauer hemocytometer. DHC were performed using slidesprepared by placing 100 μL (ca. 3×106 cells/slide) of fixedhemolymph into a Shandon cytocentrifuge and centrifuged at170 g for 5 min to sediment the cells. These hemocyte prepa-rations were then stained with Wright′s stain (1:1 dilution ofstain with Wright′s buffer). The number and percentages ofthree main crab hemocyte types for each individual were calcu-lated using a minimum number of 200 cells/slide.
2.5. Hemolymph bacterial load
At the end of the trial (week 11), 100 μL of hemolymph weretaken aseptically from 10 individuals from each group, anddiluted ten-fold in sterile 3% NaCl solution. Aliquots (100 μL)were spread plated in triplicate onto tryptic soy agar (Difco,Becton Dickinson) supplemented with 2% NaCl. Colonies werecounted after 7-d incubation at 20 °C.
2.6. Serum protein and glucose assays
All assays were performed using prepared serum from thesame bleeding session. Hemolymph (400 μL) was taken and leftfor 2 h at 4 °C to clot, centrifuged (5000 g; 5 min), and thesupernatant (serum) was split into 100 μL fractions and stored at−80 °C. All assays were performed on every individual crab intriplicate in 96 well flat bottom plates.
Serum protein concentration was determined using a bicin-chromic acid assay kit (Pierce and Warriner, Chester, UK) inaccordance with the manufacturer's instructions for use in amicrotitre plate assay. Defrosted serumwas diluted one hundred-fold in 3% sodium chloride solution. Samples were calibratedagainst a bovine serum albumin standard curve (100–1000 μgmL−1) run on the same plate. Glucose concentrations in serumwere determined using the glucose oxidasemethod adapted fromBergmeyer (1984) and Reed et al. (2003).
2.7. Hepatopancreatic glycogen assay
Hepatopancreatic glycogen was assayed using the anthronemethod of Van Handel (1965). Five crabs per diet group weresacrificed by placing at −15 °C for ca. 40 min, and the crabs werethen opened at the pleural line on the posterior abdomen. Intriplicate, homogenised hepatopancreas (0.2 g wet mass) wasmixed with 4 mL distilled water and 1 mL 30% potassiumhydroxide solution. After boiling in a water bath for 20 min, thereagents were cooled on ice and homogenised for 30 s. Analyticalgrade ethanol (6 mL) and 100 μL of saturated sodium sulphatesolution were added and boiled for a further 1 min. Aftercentrifugation (2000g; 20min) the supernatant was discarded andpellet dried overnight. After re-suspension in 5 mLwater, 500 μLwas mixed in triplicate with 2 mL anthrone reagent (anthrone75mg, concentrated sulphuric acid 38mL, distilled water 15mL)and incubated at 90 °C for 12 min. The optical densities weredetermined at 640 nmon amicroplate reader. Glycogen content ofthe serum samples were calibrated against a standard curve ofType II oyster glycogen (Sigma) (0–500 μg mL−1).
Fig. 1. Cumulative mortality of shore crabs (C. maenas) in the three dietarygroups on 0, 5 and 10% chitin-supplemented diets. Each group consisted of 20adult male crabs at the beginning of the trial. ⁎Pb0.05 compared with control.
124 A. Powell, A.F. Rowley / Comparative Biochemistry and Physiology, Part A 147 (2007) 122–128
2.8. Bacterial load of hepatopancreas
Five crabs per diet group were sacrificed at the end of week11 of the trial and opened at the pleural line of the posteriorabdomen under aseptic conditions. Portions of hepatopancreas(ca. 0.2 g) were removed under aseptic conditions and liquefiedby repeated mixing in a 1-mL sterile plastic syringe. Liquefiedtissue (100 μL) was diluted ten and one hundred-fold in sterile3% NaCl solution. Aliquots (100 μL) of each concentrationwere spread-plated in duplicate onto tryptic soy agar supple-mented with 2% NaCl solution. Bacterial colonies were countedafter 7-d incubation at 20 °C.
2.9. In vitro phagocytosis assay
The assay was performed with fluorescein isothiocyanate(FITC)-labelled Bacillus cereus, using the method of Rohloffet al. (1994) adapted by Wootton et al. (2003). Hemolymph(400 μL) was removed using a 21-gauge needle into an equalvolume of ice-cold sterile MAC (without formalin). The syringewas emptied and the contents weremixed carefully with a further400 μL ice-cold MAC. This was washed twice by centrifugation(150 g; 5 min; 4 °C) and the hemocytes were carefully re-suspended in calcium-containing marine saline (0.58 M NaCl,12 mM KCl, 20 mM Ca Cl2, 0.05 M Tris, 0.56 mM disodiumphosphate, pH 7.4) so that 5×105 cells were deposited intriplicate onto 5 mm diameter circular coverslips contained in 96well flat bottomed plates, and incubated at 15 °C. After 20 min,the saline was removed, and 5×106 FITC-labelled bacteriasuspended in calcium containing marine saline were added, andincubated for 1 h at 15 °C. Excess liquid was removed and thefluorescence of extracellular bacteria was quenched with 150 μLTrypan blue (2 mg mL−1 in sterile marine saline) for 5 min.Wells underwent a standardised rinse with calcium containingmarine saline and were fixed for 5 min in this saline containing8% formalin. The coverslips were then removed from the wells,inverted and placed onmicroscope slides withKaiser's glycerinejelly. Slides were observed using a Zeiss Photomicroscope IIequipped with a UV light source and a FITC filter set. Thenumbers and percentages of cells containing fluorescing bacteria(i.e. the intracellular forms) were calculated using a minimumnumber of 200 cells per slide.
2.10. Bacterial challenge
V. alginolyticus (NCIMB 1339) was grown overnight (18 h;25 °C) in tryptic soy broth supplemented with 2% sodiumchloride, and then washed twice with sterile 3% sodium chlo-ride by centrifugation (2800 g; 10 min; 4 °C). Serial dilutionswere made on ice using sterile 3% sodium chloride solution andbacterial cells were counted using a bacterial counting chamber.Appropriate dilutions were made to give a bacterial suspensionof 1×108 bacteria/mL to be used in the challenge. All injectionsinto crabs were made through the surface sterilized soft mem-brane where the fifth pereiopod joins the body. Total hemo-lymph volume of individual C. maenas were estimated usingthe amaranth dye dilution technique (Smith and Ratcliffe, 1980)
and the volume of bacterial suspension (1×108 bacteria/mL) tobe injected was calculated to give a circulating blood dose of2×106 bacteria/mL hemolymph. Mortality of injected crabswas determined for up to 26 h post-challenge.
2.11. Statistical analyses
All data are shown±1 SEM. Total and differential hemocytecounts were analysed using the Bonferroni multiple comparisontest. Bacterial loading, tissue chemistry and phagocytosis datawere analysed using ANOVA followed by Tukey's post-test.The cumulative mortality and bacterial challenge data wereanalysed using contingency χ2.
3. Results
3.1. Cumulative mortality
Crabs fed a 10% chitin-supplemented diet had lower mortal-ity throughout the trial than those in the control (no chitin)group (Fig. 1). At the end of the trial, the control diet group hada mean mortality rate of 30%. This was significantly higherwhen compared to the 10% chitin diet group, which had only5% mortality (Pb0.05). Crabs in the 5% chitin diet groupexperienced 10% mortality, which was not significantly lowerthan that of the control group at all time points.
3.2. Bacterial load of the hemolymph and hepatopancreas
The mean hepatopancreatic bacterial numbers in the controlcrabs were 20,405 CFU/g tissue which was significantlyhigher (ANOVA, Pb0.05) than the mean hepatopancreaticbacterial loads of the 5 and 10% chitin-fed groups (8258 and6838 CFU/g hepatopancreas respectively) (Fig. 2A). Themean hemolymph bacterial load of crabs on control dietdetermined in week 11 of the trial was 46 CFU mL−1 hemo-lymph (Fig. 2B). This was not significantly different to thebacterial numbers in the blood of crabs fed on diets containingeither 5 or 10% chitin.
Fig. 2. Numbers of cultivatable bacteria (CFU/g or CFU/mL) in the hepato-pancreas (A) and hemolymph (B) of crabs fed a diet supplemented with 0, 5 or10% chitin. Mean values+SEM, n=5–10. ⁎Pb0.05 compared with control.
Fig. 3. (A). Changes in the total hemocyte counts (THC) of crabs-fed dietscontaining 0, 5 or 10% chitin. Mean values±SEM, n=5. (B). Changes in thetotal number of hyaline cells in the hemolymph of crabs-fed diets supplementedwith 0, 5 or 10% chitin. ⁎⁎Pb0.01 compared with control group. Bars sharingthe same letters are also significantly different compared with the equivalent dietat time 0 (Pb0.05 or Pb0.01 for a and Pb0.001 for b; Bonferroni post-test).Mean values±SEM, n=5.
125A. Powell, A.F. Rowley / Comparative Biochemistry and Physiology, Part A 147 (2007) 122–128
3.3. Total (THC) and differential (DHC) hemocyte counts
The THC was not altered significantly between diet groupsthroughout the duration of the trial (Fig. 3A). Between weeks 0and 8, all groups had a THC of 2–4×107 cells/mL hemolymph.In the case of the granular cells, the cell counts did not changeduring the trial, with this type of blood cell remaining at ca.15% of the THC (data not shown). Similarly, the semigranularcells at ca. 80% of the THC also showed no significantchanges in numbers with time. In the case of the hyaline cells,however, significant reductions were found in their number inanimals on the control diet at weeks 2, 6 and 8 compared withtime 0 (Pb0.05; ANOVA and Bonferroni post-test). A similarreduction in the number of hyaline cells was also observed atweeks 2, 6 and 8 in animals fed the 5% chitin-containing diet(Pb0.001; ANOVA and Bonferroni post-test). A comparisonof the number of hyaline cells between the three dietary groupsat week 6 showed a significant difference between crabs in the0 and 10% group (Pb0.01; Bonferroni multiple comparisontest; Fig. 3B). No differences were observed in the numbers ofsemigranular and granular hemocytes taken from animals onthe three different dietary regimes at all times tested (data notshown).
3.4. Serum chemistry
Crabs fed either control diet or a diet containing chitinshowed similar feeding behaviour throughout the trial. Todetermine if chitin had any effect of food uptake by thecrabs, the levels of total protein and glucose in the hemo-lymph, and glycogen in the hepatopancreas were determined.None of these assays showed any significant differencesbetween diet groups. The serum protein concentration wasgenerally stable throughout the experiment remaining at 25–50 mg mL−1 throughout (Fig. 4A). Serum glucose concen-trations fluctuated during the first few weeks of the trial butfrom week 3, it stabilised until week 8 (Fig. 4B). The meanhepatopancreatic glycogen concentration in the final week ofthe trial was 4716 μg mL−1 wet mass for the control group,and 5699 and 5548 μg mL−1 for 5 and 10% chitin-fedgroups respectively (Fig. 4C). These values were not statis-tically different.
126 A. Powell, A.F. Rowley / Comparative Biochemistry and Physiology, Part A 147 (2007) 122–128
3.5. Phagocytosis
The in vitro phagocytic activity of the hemocytes from crabson the three diets was assessed. There was no statisticallysignificant difference in the phagocytic activity of these cellsfrom crabs on the three dietary regimes. For example, thepercentage phagocytic activity of hemocytes from crabs in thecontrol group was 23.7±1.1% compared to 26.1±1.1% and26.2±1.5% in the hemocytes from crabs on the 5 and 10%chitin diets respectively (mean values±SEM, n=5, PN0.05).
Fig. 4. Effect of chitin supplementation with 0, 5 or 10% chitin on the serumprotein (A), glucose (B) and hepatopancreatic glycogen (C) levels.Mean values±SEM, n=5–8.
Fig. 5. Effect of challenge of crabs with Vibrio alginolyticus. Values showcumulative mortality over a 26-h period post-challenge, n=9–14.
3.6. Bacterial challenge
Crabs exposed by injection to a standardised dose of V.alginolyticus started to die ca. 10 h after challenge (Fig. 5).Mortality continued at a similar rate until 20 h, although by theend of the experiment the 5% chitin-fed group had the lowestmortality of 61.5%. In the control group themortalitywas 77.8%,whilst in the 10% chitin-supplemented group the mortality was85.7%. None of these values were statistically different.
4. Discussion
The main finding of note from this current study was that thedietary addition of chitin dramatically reduced the natural mor-tality of crabs (C. maenas) during the experiment. Furthermore,chitin also caused a significant reduction in the number ofcultivatable bacteria in the hepatopancreas of C. maenas. Thereare several explanations for these observations. For example,chitin and chitosan are thought to have some antibacterialactivity (e.g. Tsai and Hwang, 2004). Secondly, chitin couldalso purge bacteria from the gut as a result of the attachment ofthese micro-organisms to chitin via chitin-binding proteins(Signoretto et al., 2005; Vaaje-Kolstad et al., 2005). This obser-vation could be of interest in the development of probiotictreatment regimes for fish and shellfish in that introduction ofchitin to the diet could act as a selective agent to removepotentially harmful bacteria such as vibrios from the digestivetract prior to feeding probiotic bacteria to recolonise this area. Itis perhaps significant that Tsai and Hwang (2004) found in ahamster model that chitin was more active against pathogenicbacteria such as Clostridium perfringens than a range of pro-biotic forms including Lactobacillus and Bifidobacterium sug-gesting that a combination of chitin and probiotics could bedelivered together in the feed without adversely affecting theviability of these ‘good’ bacteria. It should be remembered,however, that chitin could purge potentially useful bacteria fromthe alimentary canal if such forms also express chitin-bindingproteins.
The addition of chitin to a fish-based diet as used in thecurrent study had no obvious nutritional effect as evidenced by
127A. Powell, A.F. Rowley / Comparative Biochemistry and Physiology, Part A 147 (2007) 122–128
either serum protein, glucose or hepatopancreatic glycogenlevels. Indeed, the faecal material produced by crabs on thechitin-containing feeds had a distinctive whitish/grey granularappearance like that of chitin suggesting that only a limitedamount, if any, of this substance was digested. Other studies onchitin digestibility in crustaceans have shown varying results.For instance, Clark et al. (1999) reported on an increase in N-acetyl glucosamine in faecal matter from various species ofpenaeid shrimp fed chitin-containing diets. They concluded thatchitin was digested in all three species of Penaeus studied but tovarying degrees. Fox (1999) however, found no compellingevidence for any significant digestion of dietary chitin in ju-venile P. monodon. Chitinase [EC 3.2.1.14] is one of the mainenzymes involved in the breakdown of chitin to dimers andtrimers of N-acetyl glucosamine. This enzyme has been foundin the hepatopancreas of several species of crustaceans includ-ing C. maenas (Johnston and Freeman, 2005), P. monodon (Tanet al., 2000; Lehnert and Johnson, 2002), P. japonicus (Kogaet al., 1990; Watanabe et al., 1996) and the Red King crab,Paralithodes camtschaticus (Novikov and Mukhin, 2003). Inthe case of P. monodon, chitinase 1 is expressed in the F-cellswithin the hepatopancreatic tubules (Lehnert and Johnson,2002). These latter studies appear to show that crustaceans canprobably utilise chitin in diets and Tan et al. (2000) found thatchitinase expression peaked just before ecdysis, suggesting thatit is involved in the degradation and recycling of endogenouschitin prior to moulting. Similarly, chitinase is present in speciesthat eat other crustaceans, conspecifics or chitin-containing feed(Johnston and Yellowlees, 1998; Johnston, 2003; Wilde et al.,2004). The large microbial population of the hepatopancreasfound in the current study probably includes chitinolyticbacteria that are commonly found associated with crustaceans(e.g. Vogan et al., 2002) and therefore at least some of thechitinase found in this organ may originate from these bacteria.Although not considered in the current study, it would be ofinterest to ascertain if any change in the population of chiti-nolytic bacteria in the hepatopancreas of crabs exists followingdietary supplementation with chitin.
A wide range of glucans, polysaccharides and microbialproducts has been claimed to act as immunostimulants incrustaceans although the efficacy of these has been ques-tioned (Smith et al., 2003). While it has been shown thatchitin and chitosan have often potent immunostimulatoryproperties in various species of fish (e.g. Esteban et al., 2001)only a limited number of studies have sought to determine ifthese substances have such activity in invertebrates. Wangand Chen (2005) in their study with L. vannamei found thatintrahemocoelic injection of chitin and chitosan resulted inan increase in the THC and hemocyte phagocytic activity.Similarly, Furukawa et al. (1999) reported that injection ofchitin oligomers (dimers–hexamers) triggered gene expres-sion for antibacterial peptides in an insect, Bombyx mori. Ourcurrent studies with C. maenas found no increase in eitherthe THC or the phagocytic activity of hemocytes followingdietary supplementation with chitin. Furthermore, challengeof crabs with a pathogenic bacterium (V. alginolyticus) showedno increase in resistance to such infection in the chitin-fed
animals that would have been expected if the immunedefences were stimulated by the presence of this substance.Hence our results do not concur with those of Wang andChen (2005) although the route of administration of chitindiffered between the two studies. The current study exposedthe crabs to chitin in the diet while the studies of Wang andChen (2005) and Furukawa et al. (1999) were by intrahe-mocoelic injection. As it is unlikely that chitin is transportedacross the alimentary canal into the hemolymph withoutsome digestion, this could be an explanation of the differentresults. One of the few reports of dietary chitin stimulationof the immune system comes from studies with giltheadseabream (S. aurata) where significant increases in a range ofnon-specific humoral and cellular immune parameters werestimulated (Esteban et al., 2001). The mechanism of how adietary factor of limited digestibility could cause such sti-mulation was not considered by these authors and clearlywould benefit from further study.
Overall, it is suggested that the observed increased survivalof crabs fed a chitin-based diet is linked to selective removal ofpotentially pathogenic bacteria from the alimentary canal andnot stimulation of the immune system.
Acknowledgements
A.P. was supported by a University of Wales Swansea stu-dentship bursary. We thank J.W. Aquaculture (Research) Ltd.for advice and support and Ian Tew for the help with thephagocytosis assay.
Esteban, M.A., Cuesta, A., Ortuño, J., Meseguer, J., 2001. Immunomodulatoryeffects of dietary intake of chitin on gilthead seabream (Sparus aurata L.)innate immune system. Fish Shellfish Immunol. 11, 303–315.
Fox, C.J., 1999. The effect of dietary chitin on the growth, survival and chitinaselevels in the digestive gland of juvenile Penaeus monodon (Fab.). Aqua-culture 109, 39–49.
Furukawa, S., Taniai, K., Yang, J., Shono, T., Yamakawa, M., 1999. Induction ofgene expression of antibacterial proteins by chitin oligomers in the silk-worm, Bombyx mori. Insect Mol. Biol. 8, 145–148.
Johnston, D.J., 2003. Ontogenetic changes in digestive enzyme activity of thespiny lobster, Jasus edwardsii. Mar. Biol. 143, 1071–1082.
Johnston, D.J., Freeman, J., 2005. Dietary preference and digestive enzymeactivities as indicators of trophic resource utilization by six species of crab.Biol. Bull. Woods Hole 208, 36–46.
Johnston, D.J., Yellowlees, D., 1998. Relationship between dietary preferencesand digestive enzyme complement of the slipper lobster Thenus orientalis(Decapoda: Scyllaridae). J. Crustac. Biol. 18, 126–135.
Koga, D., Mizuki, K., Ide, A., Kono, M., Matsui, T., Shimizu, C., 1990. Kineticsof chitinase from a prawn, Penaeus japonicus. Agric. Biol. Chem. 54,2505–2512.
Lehnert, S.A., Johnson, S.E., 2002. Expression of hemocyanin and digestiveenzyme messenger RNAs in the hepatopancreas of the black tiger shrimpPenaeus monodon. Comp. Biochem. Physiol. 133B, 163–171.
Muzzarelli, R.A.A., 1977. Enzymatic synthesis of chitin and chitosan. Occur-rence of chitin. In: Muzzarelli, R.A.A. (Ed.), Chitin. Pergamon Press, NewYork, pp. 5–44.
128 A. Powell, A.F. Rowley / Comparative Biochemistry and Physiology, Part A 147 (2007) 122–128
Novikov, V.Y., Mukhin, V.A., 2003. Chitosan depolymerization by enzymesfrom the hepatopancreas of the crab, Paralithodes camtschaticus. Appl.Biochem. Microbiol. 39, 464–468.
Reed, R., Holmes, D., Weyers, J., Jones, A., 2003. Practical Skills in Biomo-lecular Sciences. Pearson Prentice–Hall, Harlow, p. 538.
Rohloff, L.-H., Weisner, A., Goetz, P., 1994. A fluorescence assay demonstrat-ing stimulation of phagocytosis by hemolymph molecules of Galleriamellonella. J. Insect Physiol. 40, 1045–1049.
Sakai, M., 1999. Current research status of fish immunostimulants. Aquaculture172, 63–92.
Sakai, M., Kamiya, H., Ishii, S., Atsuta, S., Kobayashi, M., 1992. The immu-nostimulating effects of chitin in rainbow trout, Oncorhynchus mykiss. Dis.Asian Aquac. 1, 413–417.
Shiau, S.Y., Yu, Y.-P., 1998. Chitin but not chitosan supplementation enhancesgrowth of grass shrimp, Penaeus monodon. J. Nutr. 128, 908–912.
Shiau, S.Y., Yu, Y.-P., 1999. Dietary supplementation of chitin and chitosandepresses growth in tilapia,Oreochromis niloticusXO. aureus. Aquaculture179, 439–446.
Smith, V.J., Ratcliffe, N.A., 1980. Host defence reactions in the shore crab,Carcinus maenas (L.): clearance and distribution of test particles. J. Mar.Biol. Assoc. UK 60, 89–102.
Smith, V.J., Brown, J.H., Hauton, C., 2003. Immunostimulation in crustaceans:does it really protect against infection? Fish Shellfish Immunol. 15, 71–90.
Tan, S.H., Degnan, B.M., Lehnert, S.A., 2000. The Penaeus monodon chitinase1 gene is differentially expressed in the hepatopancreas during the moltcycle. Mar. Biotechnol. 2, 126–135.
Tsai, G.-J., Hwang, S.-P., 2004. In vitro and in vivo antibacterial activity ofshrimp chitosan against some intestinal bacteria. Fish. Sci. 70, 675–681.
Vaaje-Kolstad, G., Horn, S.J., van Aalten, D.M.F., Synstad, B., Eijsink,V.G.H., 2005. The non-catalytic chitin-binding protein CBP21 fromSerratia marcescens is essential for chitin degradation. J. Biol. Chem.280, 28492–28497.
Van Handel, E., 1965. Microseparation of glycogen, sugars and lipids. Anal.Biochem. 11, 266–271.
Vogan, C.L., Costa-Ramos, C., Rowley, A.F., 2002. Shell disease syndrome inthe edible crab, Cancer pagurus — isolation, characterization and patho-genicity of chitinolytic bacteria. Microbiology 743–754.
Wang, S.-H., Chen, J.-C., 2005. The protective effect of chitin and chitosanagainst Vibrio alginolyticus in white shrimp, Litopenaeus vannamei. FishShellfish. Immunol. 19, 191–204.
Watanabe, T., Kono, M., Aida, K., Nagasawa, H., 1996. Isolation of a cDNAencoding a putative chitinase precursor in the kurama prawn Penaeusjaponicus. Mol. Mar. Biol. Biotechnol. 5, 299–303.
Wilde, J.E., Linton, S.M., Greenaway, P., 2004. Dietary assimilation and thedigestive strategy of the omnivorous anomuran land crab Birgus latro(Coenobitidae). J. Comp. Physiol. B 174, 299–308.
Wootton, E., Dyrynda, E.A., Ratcliffe, N.A., 2003. Bivalve immunity: com-parisons between the marine mussel (Mytilus edulis), the edible cockle(Cerastoderma edule) and the razor shell (Ensis siliqua). Fish ShellfishImmunol. 15, 195–210.
