sign in sign up
<<
Home , Chitin, Chitinase

The Auk 109(4):758-770, 1992

CHITIN DIGESTION AND ASSIMILATION BY SEABIRDS

SUE JACKSON?4 ALLEN R. PLACE?AND LINDSAYJ. SEIDERER3's •PercyFitzPatrick Institute of AfricanOrnithology, University of CapeTown, PrivateBag, Rondebosch 7700, South Africa; 2Universityof Maryland,Center of MarineBiotechnology, 600 East Lombard St., Baltimore,Maryland 21202,USA; and 3ZoologyDepartment, University of CapeTown, Private Bag, Rondebosch 7700, South Africa

ABSTRACT.--ASa structural componentof crustaceanexoskeletons, is the most im- portantcarbohydrate in the dietsof many marine carnivores.To investigatethe physiological and biochemicaladaptations that may enable seabirdsto break down this "prey defense," we estimatedchitin digestibilitiesfor Sooty Albatrosses(Phoebetria fusca), White-chinned Petrels (Procellariaaequinoctialis), Rockhopper Penguins (Eudypteschrysocome), Gentoo Pen- guins (Pygoscelispapua), King Penguins(Aptenodytes patagonicus) and Leach'sStorm-Petrels (Oceanodromaleucorhoa) fed Antarctickrill (Euphausiasuperba). These species retain a substantial proportion (46.5 + $D of 13.1%,39.1 + 4.9%, 52.8 + 37.6%,45.3 + 5.6%, 84.8 + 11.7%and 35 + 12.2%, respectively) of ingested chitin. We also obtained preliminary estimatesof chitinolytic activity in the gastricmucosae of the abovesix speciesby incubating extractsof tissuesamples with a chitin substrateand measuring the production of the end product of chitin hydrolysis,N-acetyl-D- (NAG). Chitinolytic activity (up to 5,000•tg NAG h • g • expressedper gram tissue)was measuredfrom proventriculartissue and within the activity range (1,350 to 61,650•tg NAG h • g-l) reported for eight other avian species.In order to assessthe energeticand nutritional benefitsof chitinolytic activity in seabirds,we studied gastrointestinalabsorption of the end productsof chitinolysisin Leach'sStorm- Petrels.The overall absorptionefficiency of NAG and its deacetylatedprecursor, glucosamine (Gin), in this specieswas 44.0 + 3.0% and 11.0 + 1.9%,respectively. These absorptioneffi- ciencieswere significantlyless than for ,which was absorbedwith an efficiencyof 90.6 + 2.5%. No absorptionof NAG and Gin occurred in the proventriculus. Overall, we showed that seabirdshave a capacityto assimilatea considerableportion of the and present as chitin in the exoskeletonof their prey, but we have not demonstrated that assimilationactually occurs. The potentialcosts and benefitsof chitin hydrolysis,as well as the absorption of the breakdown products, need to be assessed.Received 24 May 1991, accepted18 December1991.

THE MUCOPOLYSACCHARIDEpolymer chitin .Although chitin digestion and as- ([ 1/•4-2]-acetamido-2-deoxy-/•-D-) in similation in marine are well studied (e.g. crustaceanexoskeletons represents a substantial Fange et al. 1976, Lindsay et al. 1984, Lindsay sourceof potential energy and carbohydratefor and Gooday1985), little attentionhas been paid marine predators (Anderson et al. 1978, Reh- to utilization of dietary chitin by seabirdswhose bein et al. 1986). Estimates of biomass, for daily mass-specificenergy demandsfar exceed example,range from 80 to 500 million metric those of marine fish. tons (Sahrage and Steinberg 1975), the equiv- Seabirds such as those that feed on crusta- alent of approximately!.6 to 10 million tons of ceans in the Southern Ocean would benefit chitin, and 0.24 to 3 million tons of other car- greatly from an ability to penetratethe "struc- bohydrates(Clarke 1980). Hence, chitin is the tural defenses"of their prey, and to utilize the mostimportant carbohydratein the dietsof ma- energy residing in chitinous .By rine predatorsof krill, and probably of other studying specieswith a range of natural diets (Table 1), we testedthe predictionthat seabirds known to habitually eat crustaceansalso secrete 4 Departmentof Physiology,UCLA Schoolof Med- chitinolytic ,which enhancetheir abil- icine, Center for the Health Sciences, 10833 Le Conte ity to break down and assimilate this carbo- Ave., Los Angeles, California 90024, USA. hydrate.Although a studyof microbialand ver- 5 Marine EcologicalSurveys Ltd., P.O. Box 6, Fav- tebrate chitin degradation by crabeater seals ersham,Kent, ME13 AAA, United Kingdom. (Lobodoncarcinophagus) and Ad•lie Penguins

758 October1992] ChitinDigestion by Seabirds 759

T^I•LE1. Predominantprey types of seabirdsstu•tied (less-frequently-eaten prey in parentheses).

Species Predominantprey Reference Leach's Storm-Petrel (Oceanodroma Fish (crustacea) Linton (1978) leucorhoa) White-chinned Petrel (Procellariaae- Fish (, crustacea) Jackson(1988), A. Berruti (unpubl. quinoctialis) data) SootyAlbatross (Phoebetria fusca) Squid (fish) Cooper and Klages(unpubl. data) RockhopperPenguin (Eudypteschry- Crustacea(fish) Brown and Klages(1987) SO½ODle) Gentoo Penguin (Pygoscelispapua) Fish (crustacea) Adams and Wilson (1987) King Penguin (Aptenodytespatagoni- Fish (squid) Adamsand Klages(1987), Croxall and Prince (1980)

(Pygoscelisadeliae) is currentlyunder way (Stal- radation route, involving the breakdown of chi- ey 1986), we are not aware of published data tin into glucosamineby specific deacetylases other than our own (for a preliminary report, (Reyeset al. 1986),has not been shown to occur see Place 1990, Place and Jackson 1991) on the in vertebrates.There are four recognizedroles levels of chitinolytic activity in seabirdguts. for chitinasesin nature, namely morphogenic, Most forms of chitin are insoluble in many nutritional, digestive and defensive (Gooday solvents, highly hydrophobic, and relatively 1990). In this , we use a variety of tech- inert to (Pangburn et al. 1984, niques to assesswhich of these roles Gooday 1990). With only one exception,the plays in seabirddigestion. •-chitin of , chitin is always found We report the resultsof chitinolytic assayson crosslinked or associated with other structural stomachcontents and tissuesamples from five components. In insectsand other invertebrates, speciesof seabirdsoccurring in the Southern chitin is associatedwith specific Ocean,and one speciesthat breedson the North through covalent and noncovalentbonding, Atlantic coast. We examine the nutritional im- producing an ordered structure(Blackwell and portanceof chitin using two approaches,the Weih 1984).Chitin often exhibitsvarying de- first of which is a balancestudy estimatingap- grees of mineralization, in particular, calcifi- parent digestibilitiesof chitin in three penguin cation or sclerotizationinvolving interactions speciesand three procellariiform speciesfound with phenolic and lipid molecules(Peter et al. to have gastric chitinase activity. Second, we 1986,Poulicek et al. 1986).In nature, varying demonstratethe capacityof a procellariiform to degreesof deacetylationof the result absorb the hydrolytic products of chitin, in a structuralcontinuum between chitin (fully N-acetyl-D-glucosamineand glucosamine,and acetylated) and (fully deacetylated; comparethis capacityto that for absorptionof Aruchami et al. 1986, Datema et al. 1977, Davis glucose,a structurally similar and Bartnicki-Garcia 1984, Gowri et al. 1986). that is probablythe secondmost important car- Once consideredthe action of a nonspecific bohydrate in seabirddiets after chitin itself. In enzymefor the catabolismof chitin,chitinolysis addition, we investigated some in vitro condi- is now known to involve a complexgroup of tions (i.e. pH and Ca concentrations)that may enzymeswith discretephylogenies and subtle regulate chitinolytic activity and comparedthe differencesin specificity,role, and biochemical effects with in vivo conditions found in seabird properties.Chitin degradationrequires the ac- stomachs. tion of at least two enzymes, chitinase (E.C. 3.2.1.14; poly-•-l,4-[2-acetamido-2-deoxy]-D- METHODS glucoside glycanohydrolyase)and chitobiase (E.C. 3.2.1.30; •-N-acetyl-D-glucosaminidase; We purchasedthe following chemicalsfrom Sigma Chemical(St. Louis,Missouri): chitin; chitosan;D-(+ )- Jeuniaux1961). Chitinase hydrolyzes chitin to glucosamine hydrochloride; N-acetyl-D-glucos- yield trimers and dimers (chitotriose and chi- ;N,N'-diacetylchitobiose; D-glucose; polyeth- tobiose)of N-acetyl-D-glucosamine(NAG), and ylene glycol 4000;Serratia marcescens chitinase; Helix chitobiasehydrolyzes these trimers and dimers pomatiafl-glucuronidase; and almond fl-glucosidase. to the monomerof NAG. An alternativedeg- Aceticanhydride was purchasedfrom Aldrich Chem- 760 JACKSON,PLACE, AND SEIDERER [Auk,VoL 109 ical (Milwaukee, Wisconsin). All other chemicals were weight) olive oil. Nightly mealsof 5 to 8 cc (approx- reagent grade unlessspecified otherwise. All solvents imately 88 to 141 kJ/meal) were warmed to 37øCand were either pesticideor FIPLC grade.Whatman 3-MM delivered to the stomachvia the esophaguswith a chromatographicfilter paper wasobtained from VWR disposable5-ml syringe attachedto a 10-cmlength of Scientific(Philadelphia, Pennsylvania). polyethylene tubing. All chicks took the feeding Radiolabelsand fiuors.--We used acetic anhydride without regurgitation, and their daily wing-chord [•FI] (50 mCi/mmol), D-['4C(U)] glucose(4.28 mCi/ growth indicated normal development.After inges- retool), acetyl-N-[glucosamine-l,6-3H(N)]-D-glucos- tion, each chick was placed on a polyethylene mesh amine (44.2 mCi/mmol), D-[1,6-3H(N)] glucosamine (6-ram mesh size) platform suspendedin a 9-L Baine hydrochloride(24 mCi / retool), and [1,2-•4C]polyeth- Marie polyethylenecontainer (Cole-Palmer) to collect ylene glycol 4000 (PEG) (0.77 mCi/g) from NEN Re- excreta. Containers were kept in the dark and main- search Products (Boston, Massachusetts). All of these tained at 14 _+ 3øC to simulate the nest environment were found to have radiopuritiesgreater than 98%by as much as possible.Meals and daily collectedexcreta thin-layer chromatography(TLC). were treated as above. Fluors were Pico-Safe (Packard, Downers, Illinois) Digestibilities estimated in this manner are here- and BiosafeII (ResearchProducts International, Mount after referred to as "apparent" digestibilitiesbecause Prospect,Illinois). Sampleswere counted on a Pack- such balance studies do not take into account reten- ard Tri-Carb 1600A liquid scintillation analyzer. A tion of unassimilatedchitin within the gut (seeDis- correction ("quench") curve was derived to deter- cussion).The chitin content of food and excreta from mine counting efficiency for different extractsand all feeding trials was estimated by equating chitin tissuetypes. Counting efficiencyfor •4Cin the samples with fiber, and using the method for crude fiber de- varied from 78 to 82%, and that for 3FI from 38 to 46%. termination in animal feeds describedby FIorwitz et Counting times (3-5 rain) were chosento ensure at al. (1975).The acid-detergent-fiber(ADF) method has least 95% counting accuracy.The coefficientof vari- been shown to recover chitin completely, at least in ation for replicate samples averaged 3.8 _+ 1.3% tri- shellfish meals (Stelmock et al. 1985, Watkins et al. tium and 2.1 _+ 0.9% for carbon-14. All radioactivities 1982). We validated our estimatesby assessingthe are expressedin •tCi (1.00 •tCi = 37.0 kilobequerals). chitin content of the crude "fiber" using two hydro- Apparentchitin digestibilities.--Chitin digestibilities lysis routes:enzymatic hydrolysiswith Serratiamar- were determined for: adult and fledgling Sooty Al- cescenschitinase and either almond/•-glucosidaseor batrosses;fledgling White-chinned Petrels;and adult snail gut/•-glucuronidase(Cabib and Sburlati 1988); GentooPenguins and King Penguinsat Marion Island or acid hydrolysis and reacetylation(Lindsay et al. (46ø52'S, 37ø51'E);adult Rockhopper Penguins at 1984). The N-acetylglucosamine thus formed was Gough Island (39ø21'S,09ø53'W); and Leach's Storm- quantified colorimetrically according to the proce- Petrel chicks at Little Duck Island (44ø10'N, 68ø15'W). dure of Reissigand Leloir as modifiedby Boiler and Procellariiform chicks were used because:they are Mauch (1988). easierto capture than are the adult of many species; Acid hydrolysisresulted in significantdestruction they must eat what their parent presents to them as of glucosaminewith resulting low levels of NAG be- food; the energeticand nutritional demandsfor chick ing detected.We recoveredonly 24.2 _+5.0% (n = 11) growth are greater than for adult maintenance; and of the predictedquantity of amino sugarsfrom pu- we wished to lessenthe impactof our studyon breed- rified chitin after acid hydrolysis for 12 h at 105øC ing adults as far as possible.Removal of a chick from (Lindsayet al. 1984).Enzymatic hydrolysis (Cabib and the nest for two to four days does not result in de- Sburlati 1988) was far more accurate in our hands, sertion in White-chinned Petrels and Sooty Alba- yielding nearly 82.2 _+22.1% (n = 6) of the predicted trosses.For the studieson Marion and Gough Island, NAG from reacetylatedchitin after 4.5 h of treatment. five adult birds of each speciesand age class were We estimate that 85% of the weight of the fiber was fastedfor 48 h before receivingan homogenizedAnt- NAG releasedby enzymatichydrolysis. The ashcon- arctic krill (Euphausiasuperba, hereafter referred to as tent of the acid extractedfiber averaged0.14 _+0.3%. krill) meal with a wet mass of 6 to 8% of the birds' Apparentdigestibilities of chitin for individual birds body mass.The birds were then fasted for an addi- were calculatedas percentagesusing the formula: tional 48 h, during which time all excreta were col- 100 (C,, - Cou,)/C,, (1) lected separatelyfor each bird. Samplesof food (n = 6) and excretawere dried at 45øCfor five days, and where C,, and C(•ut,respectively, are the total dry mass- then weighed and ground finely in an electriccoffee es (g) of chitin ingested in the food and chitin ex- grinder. creted. For studies on Little Duck Island, 12 Leach's Storm- In vivo chitinolysisin Leach'sStorm-Petrel chicks.- Petrel chicksinitially 30 days old were hand-reared We used an isotope-ratiotechnique to measurein vivo for 25 days on homogenized krill. A 20% (weight/ chitinolysisof tritiated chitin in sevenLeach's Storm- volume) homogenateof freeze-dried krill was made Petrel chicks between 45 and 60 days old. Tritiated with water and supplemented with 35% (weight/ chitin wasprepared by reacetylationof chitosanwith October1992] ChitinDigestion bySeabirds 761

aceticanhydride, asdescribed by Molano et al. (1977). [•H]-NAG or [•H]-D-glucosaminewere measuredin The birds were fed a homogenizedmeal (4.6 + 0.9 g, the excreta of five birds fed a meal of avian Ringer n = 3; or 9.6 + 0.6 g, n = 4) containing tritiated chitin solutioncarrying the labeledglucose and one of the (0.15 •tCi/ml, 0.5 •tCi/mg, 0.23 •tCi/•tmoleNAG), plus two labeled . The concentration of [•4C]-PEG(0.22 •tCi/ml) in a carrier suspensionof 50 eachsugar was 25 mM. Assimilationefficiencies of mg/ml unlabeled reacetylatedchitin and t0 mg/ml the labeled substanceswere expressedas percentages unlabeled PEG in avian Ringer solution. Both PEG of the total label ingested. (polyethyleneglycol 4000) and the productsof chi- In all of the above experiments, the meals were tinolysisare water soluble,but unhydrolyzedchitin administeredper os (no chicksregurgitated after feed- is not. We validatedthis by centrifugationof the meal ing), and the fecescollected as describedabove. Pro- suspensionfor 5 min at 200 x g; more than 98% of ventricular samplesremoved by intubation at t- and the tritiated chitin and less than 1% of the •4C-labeled 2-h postingestion.Accumulated excreta in eachcon- PEG sedimentedout of the suspension.With the as- tainer were removed at 2-, 12- and 24-h intervals, sumption that PEG is nonabsorbableby the proven- homogenized,centrifuged, and analyzedas described triculus,we thus quantifiedin vivohydrolysis of the above. tritiated chitin by measuring the ratio of dissolved Chitinolyticactivity.--Tissue samples were taken from tritium-labeled chitin oligomers to [•C]-PEG in cen- the stomach(proventriculus) linings of adult Rock- trifuged proventricular liquids, because98% of the hopper,Gentoo and King penguins(in all casesn = undigested chitin in the proventricular liquid is not 4) 12 h after a krill meal, using Olympus FB-t5K E in solution and is removed by centrifugation.More- biopsyforceps passed through an OlympusGIF type over, use of the PEG marker circumventsbias arising P fiber-opticgastroscope with an Olympus CLE 4U! from dilution of the original meal by gastricor in- 4E cold light source.This method did not necessitate testinal secretions. killing the study animals (Jacksonand Cooper 1988). After feeding, feces were collected as described The sampleswere rinsed a minimum of five times above. Proventricular samples (ca. 0.2 ml) were ob- with 67 mM phosphatebuffer (pH 7.0), and immersed tained by intubation at 30-, 60- and t20-min post- in a solution of 20 •tl of PMSF (phenylmethanesul- ingestion,centrifuged, and an aliquot removed for phonylfluoride,Merck, RSA) in t0 ml of phosphate scintillation counting. Accumulated excreta (2-, 12- buffer. The sampleswere frozen and storedat -20øC and 24-h postingestion) in each container were ex- for one to two months before analysis.Two samples tractedwith 50 cc of distilled water and homogenized were taken from every individual and samplespooled with a Polytron homogenizer to uniform composi- for each species. tion. Samples were then removed for scintillation Whole sections of stomach and intestinal wall were counting. taken from one Sooty Albatross,two White-chinned Absorptionof productsof chitinolysisin Leach'sStorm- Petrel, and six Leach's Storm-Petrel fledglings that Petrelchicks.--We estimated glucose,N-acetyl-D-glu- had been killed immediatelyprior to dissectionby cosamine(NAG) and D-glucosamine(Gln) absorption intravenousinjection of "Euthanaze," a stable solu- efficienciesusing both an isotope-ratiotechnique, and tion containing 200 mg sodium pentobarbitoneper by measuringtotal isotoperecovery. The first method ml (Centaur Laboratories,Johannesburg, South Af- involved feeding five birds t0 •tCi of one of the tri- rica).These samples were frozen separatelyand stored tiated monosaccharidesof interest (NAG, D-glucos- at -20øC or -70øC, each in t0 ml of a solution with amine or D-glucose)simultaneously with 5 •tCi of the sameproportions of phosphatebuffer and PMSF labeled PEG, in a carrier meal of either avian Ringer as that used above. solution, or homogenizedwhole krill. Both types of Both squid and krill synthesize chitin (Jeuniaux carrier meal also contained50 mg of unlabeled PEG, 1963,Spindler and Buchholz 1988,Grisley and Boyle and 0.2 mmol of unlabeled monosaccharide (NAG, 1990).To investigatevariation in gastricchitinolytic Gln or D-glucose), a quantity equivalent to the total activity resultingfrom the presenceof this activity amount of chitin containedin a krill meal of 8.6 g, in the prey itself, the proventriculuscontents of cap- assumingcomplete hydrolysis of chitin. tive Sooty Albatrosseswere sampled through the Gastrointestinal absorption of [3H]-N-acetyl glu- esophagusby intubation, 12 h after a krill or a squid cosamineor [3H]-glucosaminewas calculated from the meal (in each case, four individuals were sampled). ratio of [•4C]to [•H] in the daily fecal collection using It was necessaryto introduce 30 ml of distilled water the formula: into each bird's stomach to facilitate withdrawal of the sample.We added 20 •tl of PMSF (a serine protease Percent absorbed = (t - R,n/Rout)x tOO, (2) inhibitor) per 10 ml of stomachcontents to each of where R•nequals the ratio •4C/3H in the meal, and the samples,which were storedas describedabove. Ro•tequals the ratio •C/•H in fecal collection. Total In vitro chitinolysis.--The tissue samples were recoveryof •C and •H in accumulatedfeces also were thawed, crushed by hand over ice in 4 ml of phos- measured. phate buffer using glasstissue grinders, centrifuged Secondly,total recoveryof •C-labeled glucoseand at 10,000x g for 10 min, and the supernatantdecanted 762 JACKSON,PLACE, AND $EIDERER [Auk, Vol. 109 for assays.Samples of stomachcontents were thawed •g NAG = 7.4 _+0.63/rag (crude extract) and centrifuged in the same manner, but were not + 0.336 (R2 = 0.98, n = 10), homogenized. All assayswere performed in dupli- and to time, cate. Two methods were used: the end-product col- orimetric measurement (Jeuniaux 1966); and the tri- •g NAG = 0.009 -+ 0.0004/min + 0.083 tiated-chitin method (Molano et al. 1977). The final (R 2 = 0.995, n = 14) -activity values estimatedusing both methods for at least 180 min. The production of the soluble were expressedas •g NA h 'mg ' protein. tritium label did not result from deacetylationof [3H] For the colorimetric measurement, chitin substrate chitin, becauseonly 0.5 _+0.1% (n = 12) of the tritium was prepared using the method of Reichenbachand was ethyl-acetate extractable(Araki and Ito 1988). Dworkin (1981). Sampleswere incubatedon a shaker Protein content of the tissuesamples was determined at room temperature, with 1.5 ml of enzyme solution usingthe method of Bradford(1976) when we assayed and 600 •1 of chitin slurry (10 g/100 ml of phosphate using the [SH] chitin method. buffer). We added 100 •1 of toluene every 24 h to pH and calciumdependence of chitinolyticactivity.- inhibit bacterial activity. At 0, 4, 24, 48 and 72 or 96 The mucosallinings from samplesof proventricular h after the start of incubation, 250-•1 aliquots of the tissue from two White-chinned Petrels and six Leach's solution were removed and assayedfor the end prod- Storm-Petrels were stripped away from the outer uct N-acetylglucosamine(NAG). The sampleswere muscleby scrapingwith a glassmicroscope slide. The read on a BeckmanDU-40 spectrophotometerat wave- lining was weighed, minced, and homogenized with length 585 nm, and absorbancewas converted to chi- an Ultra-Turrax T25 homogenizerin a volume of 2% tinolytic activity (•g NAG) using the equation: aceticacid adjustedto double the massof tissue.The Y = 9.43X ø87, (3) extract was centrifuged at 20,000 x g for 60 rain and the supernatant collected and stored at -70øC. The where Y is •g NAG and X is absorbanceat 585 nm. extractsprepared with 2% acetic acid showed higher The equation was calculated from a dilution series specificactivities than did thoseextracted with phos- with standardsolutions of NAG (Sigma) in phosphate phate buffer, because200- to 300~fold less protein is buffer (N = 16, r2 = 0.99). This assay, without the obtained with the former technique. The acid extracts addition of added chitobiase, measures the extent to lose activity with storage,even when kept at -70øC. which chitin is completely hydrolyzed to its mono- Lysozymeactivity was measuredusing a 0.25 mg/ml roetic constituent. The chitinolytic activities mea- suspensionof freeze-driedMicrococcus luteus as a sub- sured in proventricular tissue and luminal contents strate.Upon addition of the mucosalextract, clearing are based on the production of NAG without added of the suspensionreflects lysing of the Micrococcus N-acetylhexosamidaseto the assay mixture. Protein walls, which can be quantified by recording the content of the tissue extractswas determined using change in absorbanceof the suspensionat 540 nm the method of Lowry et al. (1951) to permit the stan- with a BeckmanDU8 spectrophotometerat 37øC(Dob- dardized expressionof NAG concentrationper mg of son et al. 1984). A unit of activity is defined as a 1% protein analyzed. change in absorbanceper minute. For studiesof ac- For the tritiated-chitin method, substrate was pre- tivity as a function of pH at constantionic strength pared by reacetylationof chitosanwith aceticanhy- of 0.133, a seriesof buffers were made by combining dride, as describedby Molano et al. (1977). Samples five parts of the appropriate buffer (HCI!NaCI, pH were incubated at 41øC on a shaker, in 100 •1 of a 1-2.5, Acetate, pH 3-4.5, Phosphate,pH 6-7.5, Miller mixture containing0.05 M buffer (sodiumphosphate, and Golder 1950) at ionic strength of 0.1, with one pH 6.3; sodium acetate,pH 4.1; or KC1/HC1, pH 1.0 part of 0.03 M NaCI. or 2.0; depending on the desired pH); 3.6 mg/ml of acetyl[3H] chitin (0.5 •Ci/mg, 0.23 •Ci/•mole NAG); and tissue extract. After incubation, 0.3 ml of 10% (w! v) trichloroaceticacid was added, and the suspensions RESULTS centrifugedfor 5 rain at 500 x g. We removed 200 •1 of the supernatant with a micropipette, and trans- Apparent chitin digestibilities.--Thekrill used ferred this aliquot to a scintillation vial for radioac- in our diet contained 2.9 + SD of 0.4% chitin tivity determination. This assayprocedure measures (dry mass)based on the fiber content, and the the production of all chitin oligomers with chain wet-mass-to-dry-massratio for the krill was 4.25 lengths less than 7 and, thus, does not require the _+0.04. The calciumcontent (percentdry mass) production of NAG for activity determi- was 1.8 + 0.2% and the energy content was nations. The activities of both endochitinase (which cleaveschitin within the polysaccharidechain) and 17.87 _+1.91 (k] g-• dry mass).These figures are exochitinase(which cleavesthe dimer chitobiosefrom consistentwith other analysesof krill compo- the ends of the chitin polymer) are thus determined. sition (Clarke 1980, Fricke et al. 1984). All six The release of soluble chitin oligomers was linear seabirdspecies exhibited appreciable,although both with respectto enzyme addition, highly variable, apparent chitin digestibilities, October1992] ChitinDigestion bySeabirds 763 with values for King Penguins and Sooty Al- batrosschicks higher than those for other spe- King Penguins cies (Fig. 1). Gentoo penguins

In vivo chitinolysisin Leach's Storm-Petrel RockhopperPenguins chicks.--In the two sets of experiments where Leach'sStorm-Petrel chicks' chicks were fed different-sized meals of radio- labeled chitin, average rates of proventricular White-chinnedPetrelchicks ' chitinolysisdetermined by in situsolubilization SootyAlbatross chicks' of tritiated chitin in the first hour postingestion SootyAlbatross adults were 28.5 + 11.2 (n = 3) and 220 + 87.7 (n = 0 20 40 60 80 100 4)/•mole NAG equivalents h -• ml -• of proven- tricular liquid for the two meal sizes(4.6 and Assimilation Efficiency 9.6 ml containing 50 mg/ml chitin), respective- Fig. 1. Mean apparent digestibilities(with SD in- ly. If we assumethe averagemolecular weight dicatedby whiskers;in all casesn = 5, exceptfor King of a purified chitin chain is 1,000,000,then ap- Penguins where n = 6) of chitin in King Penguins, proximately4,500 molesof NAG make up each Gentoo Penguins, Rockhopper Penguins, Leach's Storm-Petrels,White-chinned Petrels,and Sooty Al- chain. This translatesto a chitinolytic rate (ex- batrosses fed meals of . pressedas a percent of ingested chitin) of ap- proximately 13%/h and 98%/h for the two meal sizes,respectively. Very little NAG (< 10%)was of 3H in NAG to •4C in PEG in the meal (2.55 produced;the major product being a polymer _+ 0.05) is statisticallyindistinguishable from six or more units in length as assessedby TLC that in the proventriculusat 1-h (2.60 _+0.05) of the radioactiveproducts (Molano et al. 1979). and 2-h (2.57 _+0.08) postingestion.Similar re- These chitinolytic activities are not diet-de- suits were obtained for glucosamineand glu- rived, becausepurified chitin was usedas food. cose. In vivo chitindigestibility in Leach's Storm-Petrel In contrast,absorption of all three sugarswas chicks.--The aqueous marker (PEG) emptied proven to occur acrossthe entire gut. When rapidly from the proventriculus (Place et al. estimated by the marker ratio method, 46 _+ 1989), with recoveryof 38 _+13.5% of the orig- 8.1% (n = 3) of ingested NAG fed in avian Ring- inal dosein excreta2 h after feeding. High pro- er solution is absorbed, whereas the corre- portions of the original doseswere recovered spondingestimates using the total radio-isotope 24 h after feeding in excreta(88.4 _+3.0% and recovery method is 39.4 _+ 6.9%. The aqueous 76.2 + 5.4% for the 4.6 g and 9.6 g meals, re- marker(PEG) is rapidly evacuatedfrom the gas- spectively). Tritiated chitin was retained in the fro-intestinal tract, evidenced by the recovery proventriculus,with only I% recovered in ex- of 46.3 _+ 8.1% of this marker in the excreta 2 creta after 2 h. After 24 h, 42.6 _+ 5.7% and 49.4 h after ingestion.After 24 h, 78.1 _+4.7% of the + 16.3% of the ingested tritiated chitin was re- ingesteddose of this marker was excreted. covered in excreta, for the two meal sizes, re- When fed as a mixture with equal molar con- spectively.The recoveredtritiated chitin in ex- centrationsof [•4C]-D-glucose,a similar absorp- creta was sufficiently nondegraded to be tion efficiency was recorded for [3H]-NAG. completely precipitated by 10% trichloroacetic D-glucoseis absorbedat 90.6 _+2.5% of the in- acid. Apparent digestibilities of radio-labeled gested dose, and NAG at 44.3 _+ 3.0% of the chitin estimated from these excretion rates (50- ingesteddose. Similarly, when [•*C]-D-glucose 58%) are similar to those obtained using the and [3H]-D-glucosamineare fed asan equal mo- crude-fibermeasurement with krill meals (Fig. lar solution, absorptionefficiencies of 89.7 _+ 1). The reacetylatedchitin is a better substrate 1.2% and 12.5 _+ 11.0% were observed. for chitinolysis than is the native-protein-as- The absorptionefficiency for NAG increased sociatedchitin in the krill meals,accounting for to 59.9 + 17.9% when this substance was fed the slightly higher apparent digestibilities with a meal of homogenizedkrill, whereasthe yielded by the former technique. correspondingvalue for D-glucosaminewas Absorptionof D-glucoseand products of chitino- unchanged at 11 _+ 1.9%. The [•4C]-PEGrecov- lysisin Leach's Storm-Petrels.--Theproventricu- eries after 48 h were 116 _+ 20.6% and 94.5 + lus of this speciesexhibits no capacityto absorb 21.5%,respectively, when fed with krill meals. N-acetyl-D-glucosamine (for all results given In vitro chitinolyticactivity.--Four of the five below, n = 5 unless otherwise stated). The ratio Southern Ocean speciesand the single North- 764 JACr,SON, PLACE, ANY $EIDERER [Auk, Vol. 109

3000' Sooty Albatross Stomach contents (krlll) Stomach contents (squid) Whlte-chinnedPetrel 2OOO Leach'sStorm-Petrel KingPenguin GentooPenguin RockhopperPenguin . 4 8 12 Chitinolytic Activity (pg NAG h-•mg-• protein) 0 Fig. 2. Chitinolytic activities(•g NAG h -• rag • I :• 3 4 5 6 7 protein; ! SD indicated)in the gastricmucosa and pH stomach contents of Sooty Albatrosses,and in the gastricmucosae of White-chinnedPetrels, Leach's Fig. 3. pH dependenceof chitinolytic activity (•g Storm-Petrels,King Penguins,Gentoo Penguins, and NAG equivalentsh-' rag-• protein; SD indicated)ex- RockhopperPenguins. In all casesn = 4, exceptLeach's tracted from the proventricular mucosa of White- Storm-Petrels (n = 6) and White-chinned Petrels (n chinned Petrels. Activity measuredat indicated pH's = 2). at constant ionic strength using acetyl [•I-I] chitin (Molano et al. 1977). Specific activities represented here (and in Fig. 4) are three orders of magnitude ern Hemisphere speciesshowed similar levels greater than those measured with the colorimetric of chitinolyticactivity in the gastricmucosa (Fig. method (Fig. 2), becausethe tritium assay is more 2, overall mean of 0.77 + 0.38/•g NGA h • mg -• sensitivefor two reasons:it detectsall chitin oligo- protein), despitewide differencesin their nat- mers with chain lengths lessthan 7, not just toOhO- ural diets (Table 1). The chitinolytic activity in mers; and the acetic acid extract assayedhas 200- to 300-fold lessprotein (see Resultssection for greater the stomach contents of krill- and squid-fed detail). SootyAlbatrosses was significantlyhigher than were mucosalactivity levels, and was highly variable among the krill-fed seabirds. 0.011 + 0.004 units/mg protein for the phos- Recall that the colorimetric assay measures phate-buffered extracts. the production of NAG monomers,and does not detect the higher NAG resulting DISCUSSION from enzymatichydrolysis unless chitobiase is added. The values presented in Figure 2 un- Exogenousversus endogenous chitinolytic activity derestimate total chitinolytic activity; when in seabirds.--Whenexamining the digestive ma- similar mucosal extracts from Leach's Storm- chinery responsiblefor assimilationof a partic- Petrels were assayedusing acetyl [3H] chitin, ular dietary component, it is important not to the chitinolytic activity increasednearly four- underestimate the possiblecontribution of en- fold (3.8 + 1.6 /•g NAG equivalents h • mg • zymespresent in the food to an item's eventual protein). digestion(Vonk and Western 1984).We believe pH andcalcium dependence of chitinolyticactiv- that the exogenouschitinolytic activity present ity.--Chitinolysis is fully functionaleven with in the food items contributes substantially to the pH at lessthan 2 (Fig. 3). There appearsto complete chitin digestion, especially in the hy- be a bimodal pH dependency,although in gen- drolysis of chitin oligosaccaharidesto product eral chitinolytic activity declinesas neutrality free NAG. Both and chitobiases have is approached.The chitinolytic activity is en- been describedin krill (Spindler and Buchholz hancednearly threefold when calciumis added 1988)and cephalopods(Grisley and Boyle 1990, to the reaction mixture (Fig. 4). Jeuniaux 1963). The high levels of chitinolytic Boththe acidextracts and the phosphate-buff- activity (as evidenced by production of NAG) ered extracts from White-chinned Petrels and we found in stomachcontents (of Sooty Alba- Leach'sStorm-Petrels exhibited lysozymeactiv- trossesfed krill or squid; Fig. 2) as opposedto ity. The specificactivity varied from 0.322 + the mucosaltissue of this and all other species 0.007 units/mg protein for the acid extractsto studied,suggest dietary origin in thesesamples. October1992] ChitinDigestion bySeabirds 765

However, we also have demonstratedendoge- nouschitinolytic activity in the proventricular tissue of Leach's Storm-Petrels and White- chinned Petrels (see below). 12OO Endogenous chitinase synthesis by verte- brateswas first reportedby Jeuniaux(1961), and has since been documented in marine fish 8OO (Okutani 1966, Rehbein et al. 1986, Seiderer et al. 1987), as well as terrestrial mammals and 4OO birds (Cornelius et al. 1975, 1976, Jeuniaux and Cornelius 1978). The synthesisof chitinolytic enzymes (chitinase and chitobiase)is thought o to be a primitive, retained characteristicrather o 5 lO 15 20 30 than one that has evolved secondarily,and is prevalentamong invertebrates (Jeuniaux 1971). • CalciumMolarity (mM) Chitinase production is thought not to be in- Fig. 4. Calcium dependenceof chitinolytic activ- ducible in vertebratesthat habitually eat food ity (•g NAG equivalentsh • mg • protein; SD indi- devoid of chitin, and synthesisof this enzyme cated)extracted from proventricularmucosa of White- persistsin animals that normally feed on chi- chinned Petrels.Activity measuredat pH 2.0 using tinousprey, oncechitin hasbeen excluded from [•H] chitin. their diets (Jeuniaux 1971). The mucosalchitinolytic activitiespresented the conversion factor is approximately 2.5. here (seealso Place 1990, Place and Jackson1991) Hence, the chitinaseactivity of White-chinned are the first evidence of substantialcarbohy- Petrelsis in the rangeof 2,000to 5,000#g NAG drase activity in any seabird. Kerry (1969) re- h • g-• expressedper gram tissue.Bird gastric ported very low levels of disaccharidaseactivity chitinaseshave an optimum pH of 4.7 to 5.4, in five Southern Ocean seabirds:Rockhopper retaining considerableactivity at lower pH val- Penguins, Gentoo Penguins, King Penguins, ues,but showing a sharp decreasein activity at Brown Skuas(Stercorarius skua) and Kelp Gulls pHs above 6. Seabird stomachpH is between (Larus dominicanus). 0.50 and 2.75 (van Dobben 1952, Place et al. Jeuniaux (1961, 1963) and Jeuniaux and Cor- 1986),below the optimum levels for the chitin- nelius (1978) reported chitinase activities in olytic enzymesdescribed by Jeuniauxand Cor- eight bird speciesranging from 1,350 to 61,650 nelius (1978). However, the pH optimum of the #g NAG liberated h -• g-• fresh massof tissue. endogenouschitinolytic activity that we report The highestvalues that they reportedwere for in White-chinned Petrels and Leach's Storm- the EuropeanStarling (Sturnusvulgaris), and the Petrelsis well within the pH range for seabird lowest were for the chicken (Gallusgallus). The stomachs. main sourceof the avian chitinasewas the pro- Roleof chitinolyticbacteria in seabirdguts.--Chi- ventricularmucosa, with levelsdecreasing pro- tinolytic bacteriahave been implicated in the gressivelyfrom the and intestinal lu- digestionof chitin by fish (Okutani 1966,Good- men contents to the intestinal mucosa (Jeuniaux rich and Morita 1977),although severalstudies 1963). These chitinase activities measuredby usingantibiotics have shown that fish gut flora Jeuniaux and his colleaguesuse added chito- may be insignificant in chitin digestion (Dan- biase in the assayto releaseNAG for colori- ulat 1986). Soucek and Mushin (1970) isolated metric determination. Our colorimetric assays gram-negative from the guts of eight of chitinolytic activity reported in Figure 2 do speciesof penguin and two speciesof skua.Al- not incorporate added chitobiase.To compare though Escherichiacoli predominated, 7 of 95 their activitieswith values reported by us for Ad•lie Penguins and 1 of 10 RockhopperPen- seabirds,we must comparethe seabirdactivities guins sampled had Enterobacterin their intes- measuredwith acetyl [3H] chitin as substrate. tines. Enterobacteris known to produce large In addition, since our values are reported on a amountsof chitinase (Monreal and Reese1969). mg-protein basis,a conversionto grams of tis- In our study, tissuebiopsies yielded positive sue mass must be made. For the White-chinned chitinaseassay results for Rockhopperand Gen- Petrels activities reported in Figures 3 and 4, too penguins, whereas stomachcontents sam- 766 JACKSON,PLACE, AND $EIDERER [Auk, Vol. 109 pled within minutes of the biopsiesand from exhibit prolonged retention of neutral lipids the same individual birds yielded negative re- when comparedwith adult conspecificsfed the sults.This suggeststhat the chitinasewas se- same diet (Jackson and Place 1990). Retention cretedby the birds' gastricmucosae, rather than of exoskeletalmaterial for longer than the du- producedby bacteriain the stomachlumen. In ration of the fecal-collectionperiod might have addition, we have used tritiated chitin to dem- resulted in underestimation of chitin content onstrate chitinolytic activity in extractsfrom of the fecesand, consequently,overestimation the mucosa of Leach's Storm-Petrels and White- of digestibilities.However, apparentchitin di- chinned Petrels.Finally, all tissuesamples were gestibilitiesas high as 90%have been recorded thoroughly rinsed before freezing, and assays for chickens(Hirano et al. 1984),and equally were carried out in the presence of toluene, high values for these two seabird speciesare which inhibits bacterial activity. Overall, we possible. attribute a substantialcomponent of initial chi- The seabirdsthat we studied are able to hy- tin breakdown (i.e. cleavageof chain into small- drolyze a substantialproportion and to retain er fragments)to the endogenousgastric activ- nearly one-half of the chitin they ingest.Chitin ities possessedby the seabirds. itself may not be a significantsource of energy Chitinolyticactivity in relationto naturaldiet.-- to seabirds, because crustaceans such as krill Chitinase activity has been demonstratedin contain a small proportion of chitin (between seven speciesof raptors (Leprince et al. 1979) 2.1 and 2.9%,dry mass;Clarke 1980,this study). and in insectivorous birds, but is absent in two Even assumingthat chitin has the sameenergy graminivores,the domesticpigeon (Columbaliv- value (17.9 kJg-• dry mass)as whole krill (Kara- ia) and the African Grey Parrot (Psittacuseritha- sov 1990), an overestimate, seabirds able to di- cus; Jeuniaux 1961, 1963, Jeuniaux and Corne- gest40 to 90%of ingestedchitin and fully as- lius 1978).Of the five speciesof SouthernOcean similate all digestiveproducts would derive a seabirdsthat we sampled,all but the King Pen- maximum of between 1.2 and 2.7% of their total guin showedchitinolytic activity in the gastric energy gain from the chitin fraction of each mucosa.Among seabirds,secretion of this en- meal. Our data on absorptionof NAG and glu- zyme appearswidespread irrespectiveof the cosaminesuggest that intestinal absorptionof proportion of crustaceansin the natural diet, the productsof chitinolysisis far from complete but further studyof speciesthat do not eat crus- in at least one planktivorousseabird species, taceans would be instructive. further reducingthe likelihood that chitin is of Why digestchitin?--There are few published significantenergetic value to seabirds.Chitin- data on the digestibilityof chitin in birds. Jap- olysis probably increasesthe rapidity of me- anese Nightingales (i.e. Red-billed Leiothrix, chanical breakdown of the exoskel- Leiothrixlutea) digest 56.8% of the chitin in meal- eton, thereby facilitating digestionof soft prey worm Tenebriomolitar larvae, and 15-day-old tissues in shorter time than otherwise would be chickensdigest 23.5 to 31.7%of purified possible.Also, cleavageof the chitin polymer chitin added to their normal diets (Jeuniaux and in the exoskeletonwould allow easieraccess by Cornelius 1978). Digestibilitiesas high as 92% proteolyticenzymes to associatedproteins in havebeen reportedfor adult and juvenile chick- the . Increased gastric evacuation rates ens fed chitin and chitosandietary supplements may be the majorbenefit of chitinasesecretion (Hirano et al. 1984).In our study,most Southern accruedby seabirds,given the energeticadvan- Ocean speciesexhibit digestibilities that agree tages of rapid passageof digesta in all avian closelywith the values for Leach'sStorm-Pet- species(Sibly 1981). rels that were estimatedusing purified chitin Chitin and its breakdownproducts also may and a double-isotopetechnique to circumvent influencethe gut flora of seabirds.As a dietary biases arising from retention of . supplement,chitin improvesthe utilization ef- However, digestibility estimatesfor King Pen- ficiencyof whey by chickens(Spreen et al. 1984), guinsand SootyAlbatross chicks are higher than apparentlyby enhancingintestinal growth of thosefor all seabirds,and may be artifactsre- -producingBifidiobacteria spp. sulting from the extendedgastric retention of Despitethese benefits of chitin digestion,ab- undigestedkrill exoskeletonwhich has been sorptionof the productsof chitinolysismay car- observedin thesespecies (Jackson and Cooper ry a cost;juvenile chickensfed chitin and chi- 1988,Jackson 1992). Sooty Albatross chicks also tosan supplements gained mass at a rate October1992] ChitinDigestion bySeabirds 767 equivalent respectively to only 12 and 77% of mean relative absorption for NAG as a per- the rate shown by birds in a control group fed centageof glucoseabsorption was 34%, and for an otherwise identical diet free of these com- glucosamine (Gin) was 22%. Similar findings pounds,despite apparent chitin digestibilities were obtained when rats were fed molar solu- as high as 92% (Hirano et al. 1984). Similarly, tions of NAG and Gin (Carts et al. 1966). More the growth rates of rainbow trout (Salmogaird- recently, Karasov(1989) documented low NAG neri) smeltsfed diets containing 4, 10 and 25% uptake rates in mice (Mus musculus).Is this in- chitin were significantly lower over a 12-week efficiencyadaptive for a speciesingesting chi- period than the growth rates of fish fed a diet tin? Further comparativestudies using equica- containing25% ; however, high chitinase loric diets differing in chitin content are needed and chitobiaseactivities were found in the guts to assessthe influence of the end products of of thesefish (Lindsayet al. 1984).These reduced chitinolysis on the growth of juvenile birds. growth rates may have resulted from nutrient Do birdsdigest chitin moreefficiently than do or energy deprivation arising from "dilution" mammals?--Amongmammals, gastric chitinases of the digestiblecomponents in the diet by chi- are found in omnivorousand insectivorousspe- tin, but the high levels of chitinolytic enzymes cies belonging to the orders Insectivora, Chi- in trout guts and high apparent chitin digest- roptera, Carnivora, Rodentia and primates, and ibilities in chickens suggestthat the energy in pancreatic chitinaseshave been found in the chitin is available for assimilationin both spe- pig (Susdomesticus; Jeuniaux and Cornelius1978). cies.Alternatively, absorption of the end prod- There is a close relationship between chitinase uctsof chitin digestion may have an inhibitory secretion and the normal diet of the species effect on growth. In mammals, amino considered,and the ability to secret chitinase like NAG and Gin are respired much less rap- is not inducible by manipulation of the diet idly than is D-glucose (Kohn et al. 1962). Con- (Jeuniaux and Cornelius 1978). Chitinase activ- sideration of chitobiaseactivities in avian guts ity in the gastric mucosais generally lower in may be illuminating here; recall that this en- mammals than in birds (Jeuniaux 1963, Jeu- zyme accomplishesthe final breakdown of NAG niaux and Cornelius 1978, LePrince et al. 1979). dimers and trimers into absorbable monomers. Few data are available on chitin digestibilities Chitobiasehas a predominantly luminal dis- among mammals: values for mice are 19.4 + tribution in the and intestinesof five 27.6% (Jeuniaux and Cornelius 1978), and for bird speciesstudied by Jeuniaux (1963), with musk shrews (Suncus murinus) and pygmy activity levels ranging from 0 to 200 •g NAG hedgehogtenrecs (Echinops telfairi), respective- h -• g-• tissue. Jeuniaux (1963) considered the ly, are 1.8% and 19.8% (Allen 1989), substan- highest values that he reported (1,160 •g NAG tially lower than both the overall mean value h -• g-• tissue in the cecal lumen of the chicken) of 56.8 + 23.2 that we report for seabirds,and to be of bacterialorigin. On the basisof the low values among terrestrial birds (Jeuniaux and chitobiaseconcentrations in all speciesthat they Cornelius 1978). studied, Jeuniaux and Cornelius (1978) con- Furtherconsiderations and caveats.--We clearly cluded that the extent of metabolic utilization have documented that seabirds have the capac- of the products of chitin hydrolysis by verte- ity to digesta considerableportion of the chitin brates is uncertain. Their conclusionis highly in the exoskeletons of their prey. Our study significant in view of possiblegrowth-related yields no evidencethat seabirdsactually assim- costsof chitin end-product absorption. ilate the carbon and nitrogen thus released;we Furthermore,our resultssuggest low absorp- merely demonstratethat a significantportion tion efficienciesof the productsof chitinolysis, of dietary chitin is retained within the gut, and D-glucosamine in particular, reinforcing the that much of the biochemicalmachinery exists conclusion that the major role of chitinase in for assimilation of the breakdown products of seabirds is its acceleration of exoskeleton break- this polymer by seabirds.It is possiblethat a down rather than the freeing of assimilablenu- significant portion of NAG is utilized by gut trients in the chitin polymer. Low absorption microflora for cell-wall synthesis and/or as a efficiencies are consistent with earlier measure- metabolic fuel source (e.g. Carts et al. 1966). ments on the rate of intestinal absorption of Until a speciesis fed uniformly labeledcarbon- carbohydratesfrom isolatedintestinal segments 14 chitin, and respired carbon-14 is recovered of the rat (Rattusnorvegicus; Crane 1968). The in addition to the label depositedin tissue,there 768 JACKSON,PLACE, AND SEIDERER [Auk, Vol. 109 is no evidence for chitin assimilation by an an- Antarctic krill (Euphausiasuperba Dana) asa source imal. of chitin and chitosan.Pages 5-10 in Proceedings Further researchinto the chitinolytic activity of the First International Conference on Chitin/ extracted from the proventriculus of Leach's Chitosan (R. A. A. Muzzarelli and E. R. Pariser, Storm-Petrels and White-chinned Petrels also Eds.). MassachusettsInstitute of Technology, Cambridge,Massachusetts. is warranted. Some portion of this chitinolytic ARAKI, Y., AND E. Ito. 1988. Chitin deacetylase. activity may be attributable to a . Re- Methods Enzymol. 161:510-514. cruitmentof lysozymeas a bacteriolyticenzyme ARUCHAMI, M., G. SUNDARA-RAJULU,AND N. GOWRI. in the stomachs of both ruminants and colobine 1986. Distribution of deacetylasein monkeys is well documented (Stewart et al. species.Pages 263-265 in Chitin in nature and 1987), as is the chitinolytic activity of somely- technology (R. A. A. Muzzarelli, C. Jeuniauxand sozymes(Joll&s and Joll•s 1984). We propose G. W. Gooday,Eds.). Plenum Press,New York. that seabirdsmay have recruited lysozymeas a BLACKWELL,J., AND M. A. WEIH. 1984. The structure gastricdigestive enzyme in addition to its nor- of chitin-protein complexes.Pages 257-272 in Chitin, chitosan and related enzymes (J.P. Zi- mal bacteriolytic function. Current efforts to kakis, Ed.). Academic Press, New York. purify and characterizethis activity will pro- BOLLER,T., AND F. MAUCH. 1988. Colorimetric assay vide a test for this hypothesis. for chitinase.Methods Enzymol. 161:430-435. BRADFORD,g.g. 1976. A rapid and sensitive meth- ACKNOWLEDGMENTS od for the quantitation of microgram quantities of protein utilizing the principle of protein-dye We aregrateful to FrankT. Robbfor his enthusiastic building. Anal. Blochem. 72:248-254. advice,and to John Cooper,W. Roy Siegfried,Doug BROWN, C. R., AND N. T. KLAGES. 1987. Seasonal and J. Levey, and an anonymousreviewer for their com- annual variation in diets of Macaroni (Eudyptes ments on the manuscript. We thank E. R. Bateman chrysolophuschrysolophus) and Southern Rockhop- and M. Kautskyfor assistancein the laboratory,and per (Eudypteschrysocome chrysocome) penguins at V. Siegeland F. Buchholtzof the Institut f•ir Seeils- Marion Island. J. Zool. (Lond.) 212:7-28. cherei, Hamburg-Altona,for providing frozen - CABIB,E., AND g. SBURLATI.1988. Enzymatic deter- arctickrill. S. Jacksonwas a recipientof a grant from mination of chitin. Methods Enzymol. 161:457- the Frank M. ChapmanMemorial Fund of the Amer- 459. ican Museum of Natural History, and of a postgrad- CAPES,J. C., M. R. SHEILAR, AND R. H. BRADFORD. uate bursaryfrom the SouthAfrican Council for Sci- 1966. Hexosaminemetabolism. I. The absorption entific and Industrial Research. A. R. Place was a and ,in vivo, of orally administered recipientof a NationalScience Foundation grant (USA; D-glucosamineand acetyl-D-glucosaminein the DCB87-09340).The South African Departments of rat. Biochim. Biophys.Acta 127:194-204. Transportand EnvironmentAffairs provided logistic CLARKE,g. 1980. The biochemical compositionof supportat Marion and GoughIslands, where all re- krill, Euphausiasuperba Dana, from SouthGeorgia. searchis doneunder the auspicesof the SouthAfrican J. Exp. Mar. Biol. Ecol. 43:221-236. Scientific Committee for Antarctic Research. We are CORNELIUS,C., G. DANDRIFOSSE,AND C. JEUNIAUX. gratefulto the Administratorand IslandCouncil of 1975. Biosynthesisof chitinasesby mammalsof Tristan da Cunha for their approval of researchat the order Carnivora. Biochem. Syst. Ecol. 3:121- GoughIsland. This is contributionnumber 129 from 122. the Center of Marine Biotechnologyat the University CORNELIUS,C., G. DANDRIFOSSE,AND C. JEUNIAUX. of Maryland. 1976. Chitinolytic enzymesof the gastricmucosa of Perodicticuspotto (primate prosimian): Purifi- LITERATURE CITED cation and enzyme specificity.Int. J. Blochem. 7:445-448. ADAMS,N.J., ANDN. T. KLAGES.1987. Seasonalvari- CR•qE, R. K. 1968. Absorptionof sugars.Page 1343 ation in the diet of the King Penguin (Aptenodytes in Handbook of physiology, section6: Alimen- patagonicus)at sub-Antarctic Marion Island.J. Zool. tary canal,vol. 3. Intestinalabsorption (C. F. Code (Lond.) 212:303-324. and W. Heidel, Eds.).American Physiological So- ADAMS,N.J., AND M.-P. WILSON. 1987. Foraging ciety, Washington,D.C. parametersof Gentoo PenguinsPygoscelis papua CROXALL,J. P.,ANDP. A. PRINCE.1980. Food,feeding at Marion Island. Polar Biol. 7:51-56. ecologyand ecologicalsegregation of seabirdsat ALLEN,M.E. 1989. Nutritional aspectsof insectiv- South Georgia. Biol. J. Linn. Soc. 14:103-131. ory.Ph.D. thesis, Michigan State University, East DANULAT,E. 1986. Role of bacteria with regard to Lansing. chitin degradationin the digestivetract of the ANDERSON,C. G., N. DEPABLO,AND C. R. ROMO. 1978. cod Gadus morhua. Mar. Biol. 90:335-343. October1992] ChitinDigestion bySeabirds 769

DATEMA,R., M. VAN r•EN ENDE,AND J. G. H. WESSELS. JEUNIAUX,C. 1961. Chitinase: An addition to the list 1977. The hyphal wall of Mucor mucedo2. Hex~ of hydrolasesin the digestivetract of vertebrates. osamine-containingpolymers. Eur. J. Biochem. Nature (Lond.) 192:135. 80:621-626. JEUNIAUX,C. 1963. Chitin et chitinolyse, un chapitre DAVIS, B., AND $. BARTNICKI-GARCIA. 1984. The co- de la biologique mol'culaire. Masson, Paris. ordination of chitosanand chitin synthesisin JEUNIAUX,C. 1966. Chitinases. Methods Enzymol. Mucor rouxii. J. Gen. Microbiol. 130:2095-2102. 8:644-650. DOBSON,D. E., E. M. PRAGER,AND A. C. WILSON. 1984. JEUNIAUX,C. 1971. On somebiochemical aspects of Stomachlysozymes of ruminants.I. Distribution regressive evolution in animals. Pages 304-313 and catalyticproperties. J. Biol. Chem. 259:11607- in Biochemicalevolution and the origin of life 11616. (E. Schoffeniels,Ed.). North-Holland Publishing FANGE,R., G. LUNDBLAD,AND J. LIND. 1976. Lyso- Co., Amsterdam. zyme and chitinasein and lymphomyeloid JEUNIAUX,C., AND C. CORNELIUS. 1978. Distribution tissues of marine fish. Mar. Biol. (Berl.) 36:277- and activity of chitinolytic enzymes in the di- 282. gestive tract of birds and mammals. Pages 542- FRICKE, H., G. GERKEN, W. SCHREIBER, AND J. 549 in Proceedingsof the First International Con- OENLENSCHLAGER.1984. Lipid, sterol, and fatty ference on Chitin/Chitosan (R. A. A. Muzzarelli acid compositionof Antarctic krill (Euphausiasu- and E. R. Pariser, Eds.). Massachusetts Institute perbaDana). Lipids 19:821-827. of Technology,Cambridge, Massachusetts. Goor•^¾,G. W. 1990. The ecologyof chitin degra- JoLL•s,P., ^ND J. JOLL•S.1984. What's new in ly- dation.Pages 387-430 in Advancesin microbial sozymeresearch? Mol. Cell. Biochem.63:165-189. ecology,vol. 11 (K. C. Marshall, Ed.). Plenum KAP,ASOV, W. H. 1989. Can terrestrial vertebrates Press, New York. digest and absorbchitin? Am. Zool. 29:24A. GOODRICH, T. D., AND R. Y. MORITA. 1977. Bacterial KARASOV,W. H. 1990. Digestion in birds: Chemical chitinase in the stomachs of marine from and physiologicaldeterminants and ecological Yaquina Bay, Oregon, USA. Mar. Biol. (Berl.) 41: implications.Stud. Avian Biol. 13:391-415. 355-360. GOWRi, N., g. ARUCHAMI, AND G. $UNDARA-RAJULU. KERRY,K. R. 1969. Intestinal disaccharidaseactivity in a monotremeand eight speciesof marsupials 1986. Infrared and X-ray diffractiondata on chi- (with an added note on the of tins of variable structure. Carbohydr. Res. 165: 105-110. five speciesof seabirds). Comp. Biochem.Phys- iol. 29:1015-1022. GRISLEY,M. 5., ^ND P. R. BOYLE. 1990. Chitinase, a KOHN, P., R. J. WINSLER,AND R. C. HOFFMANN. 1962. new enzyme in octopussaliva. Comp. Biochem. Metabolism of D-glucosamineand N-acetyl-D- Physiol. B Comp. Biochem. 95B:311-316. HIRANO, S., H. 5ENDA,Y. YAMAMOTO,AND A. WATANA- glucosaminein the intact rat. J. Biol. Chem. 237: 304-308. BE. 1984. Several novel attemptsfor the use of the potential functions of chitin and chitosan. LEPRINCE,iv., G. DANDRIFOSSE,AND E. 5CHOFFENIELS. Pages 77-95 in Chitin, chitosan and related en- 1979. The digestive enzymes and acidity of the zymes (J.P. Zikakis, Ed.). Academic Press, Or- pellets regurgitatedby raptors.Biochem. Syst. Ecol. 7:223-227. lando. HORWITZ, W., A. 5ENZEL,H. REYNOLDS,AND D. L. PARK LINDSAY,G. J. H., AND G. W. GOODAY. 1985. Chitin- (Eds.). 1975. Pages 135-137 in Official methods olyticenzymes and the bacterial microflora in the of analysisof the Associationof Analytical Chem- digestive tract of cod, Gadusmorhua. J. Fish. Biol. ists.Association of Official Analytical Chemists, 26:255-265. Washington, D.C. LINDSAY,G. J. H., M. J. WALTON, J. W. ADRON, T. C. JACKSON,5. 1988. Diets of the White-chinned Petrel FLETCHER,C. Y. CHO, AND C. B. COwEY. 1984. and Sooty Shearwater in the southern Benguela The growthof rainbowtrout (Salmogairdneri) giv- region, South Africa. Condor 90:20-28. en diets containingchitin, and its relationshipto JACKSON,5. 1992. Do seabird gut sizes and mean chitinolytic enzymes and chitin digestibility. retention times reflectadaptation to diet and for- Aquaculture 37:315-334. aging method? Physiol. Zool. 65:674-697. LINTON,A. 1978. The food and feeding habits of JACKSON,5., ANDJ. COOPER.1988. Use of fibre-optic Leach'sStorm-Petrel Oceanodromaleucorhoa (Viel- endoscopesin studiesof gastricdigestion in car- lot) at Island, Nova Scotia and Middle Lawn nivorousvertebrates. Comp. Biochem.Physiol. A Island, Newfoundland. M.S. thesis, Dalhousie Comp. Physiol. 91A:303-308. Univ., Halifax. JACKSON,5., AND A. R. PLACE. 1990. Gastrointestinal LOWRY,O. H., N.J. ROSEBROUGH,A. L. FARR,AND R. transit and lipid assimilationefficiencies in three N. RANDALL. 1951. Protein measurements with speciesof sub-Antarcticseabird. J. Exp. Zool. 255: the folin phenol reagents.J. Biol. Chem. 193:265- 141-154. 275. 770 JACKSON,PLACE, AND SEIDERER [Auk, Vol. 109

MILLER,G. L., AND R. H. GOLDER. 1950. Buffers of otes, vol. 1 (M.P. Starr, H. Stolp, H. G. Truper, pH 2 to 12 for usein electrophoresis.Arch. Bio- A. Balows, and H. C. Schlegel, Eds.). Springer- chem. 29:420-423. Verlag, Berlin. MOLANO,J., A. DURAN,AND E. CABIB. 1977. A rapid REYES,F., M. J. MARTINEZ,J. CALATAYUD,AND R. LAHOZ. and sensitiveassay for chitinaseusing tritiated 1986. Degradationof chitin from the chitin. Anal. Biochem. 83:648-656. Aspergillusnidans during autolysis.Pages MOLANO, J., I. POLACHtECK,A. DURAN, AND E. CABIB. 99-102 in Chitin in nature and technology (R. A. 1979. An endochitinasefrom germ. J. Biol. A. Muzzarelli, C. Jeuniaux,and G. W. Gooday, Chem. 254:4901-4907. Eds.). Plenum Press, New York. MONREAL,J., AND E. T. REESE. 1969. The chitinase of SAHtP,XGE, L. J., AND R. STEINBERG.1975. Dem krill Serratiamarcescens. Can. J. Microbiol. 15:689-696. auf der spur. Umschau75:627-631. OKUTANI,K. 1966. Studiesof chitinolytic systemsin SEIDERER,L. J., C. L. DAVIS, F. T. ROBB,AND R. C. the digestive tracts of Lateolabraxjaponicus. Bull. NEWELL.1987. Digestiveenzymes of the ancho- Misaki Mar. Biol. Inst., Kyoto Univ. 10:1-47. vy Engrauliscapensis in relationto diet. Mar. Ecol. PANGBURN,S. H., P. V. TRESCONY,AND J. HELLER. 1984. Prog. Set. 35:15-23. Partially deacetylated chitin: Its use in self-reg- SIBLY,R.M. 1981. Strategiesof digestionand defe- ulateddrug delivery systems.Pages 3-19 in Chi- cation. Pages 109-139 in Physiologicalecology: tin, chitosanand related enzymes(J.P. Zikakis, An evolutionaryapproach to resourceuse (C. R. Ed.). Academic Press, Orlando. Townsend and P. Calow, Eds.). Sinauer Associ- PETER,M. G., G. KEGEL, AND R. KELLER. 1986. Struc- ates, Sunderland, Massachusetts. tural studies on scelrotized cuticle. Pages SOUCEK,Z., AND R. MUSHIN. 1970. Gastrointestinal 21-28 in Chitin in nature and technology (R. A. bacteria of certain Antarctic birds and mammals. A. Muzzarelli, C. Jeuniaux and G. W. Cooday, App1. Microbiol. 20:561-566. Eds.). Plenum Press, New York. SPINDLER,K.-D., AND F. BUCHHOLZ. 1988. Partial PLACE,A.R. 1990. Chitin digestionin nestlingLeach's characterization of chitin degrading enzymes Storm-Petrels, Oceanodromaleucorhoa. Mount Des- from two euphausiids, Euphausiasuperba and ert Island Biol. Bull. 29:139-142. Meganyctiphaesnorvegica. Polar Biol. 9:115-122. PLACE,A. R., ANDS. JACKSON.1991. Chitin digestion SPREEN,K. A., J. A. Z•KAKIS,AND P. R. AUSTIN. 1984. and assimilationby seabirds.Pages 1378-1383 in The effect of chitinous materials on the intestinal Acta XX CongressusInternationalis Ornitholo- microfloraand the utilization of whey in mono- gici. Christchurch,New Zealand,1990. New Zea- gastricanimals. Pages 57-75 in Chitin, chitosan land Ornithol. Congr. Trust Board, Wellington. and relatedenzymes (J. A. Zikakis,Ed.). Academ- PLACE,g. g., N. STOYAN, AND g. E. RICKLEFS. 1986. ic Press, Orlando. The physiologicalbasis of stomachoil formation STALEY,J. T. 1986. Microbial and vertebrate chitin in Leach's Storm-Petrel. Mount Desert Island Biol. degradation.Antarct. J. U.S. 21:5. Lab. Bull. 26:73-76. STELMOCK,g. L., F. M. HUSBY, AND g. B. BRUNDAGE. PLACE, A. R., N. STOYAN, R. E. RICKLEFS,AND R. G. 1985. Application of Van Soestacid detergent BUTLER.1989. Physiologicalbasis of stomachoil fiber method for analysis of shellfish chitin. J. formation in Leach's Storm-Petrel (Oceanodroma Dairy Sci. 68:1502-1506. leucorhoa).Auk 106:687-699. STEWART,C. B., J. W. SCHILLING, AND A. C. WILSON. POULICEK,M., M. F. VOSS-FOUCART,AND J. JEUNIAUX. 1987. Adaptive evolution in the stomachlyso- 1986. Chitinoproteiccomplexes and mineraliza- zymesof foregutfermenters. Nature (Lond.)330: tion in mollusc skeletal structures.Pages 7-12 in 401-404. Chitin in nature and technology (R. A. A. Muz- VAN DOBBEN,W.H. 1952. The food of the Cormorant zarelli, C. Jeuniaux,and G. W. Gooday,Eds.). Ple- in the Netherlands. Ardea 40:1-63. num Press, New York. VONK, H. J., AND J. R. H. WESTERN. 1984. Compar- REHBEIN,H., E. DANULAT, AND M. LEINEMAN. 1986. ative biochemistryand physiologyof enzymatic Activities of chitinaseand proteaseand concen- digestion.Academic Press, London. tration of fluoride in the digestivetract of fishes WATKINS, B. E., J. ADAIR, AND J. E. OLDFIELD. 1982. feeding on krill (Euphausiasuperba Dana). Comp. Evaluation of shrimp and king processing Blochem.Physiol. A Comp. Physiol.85A:545-551. by productsas feed supplementsfor mink. J. REICHENBACH,H., AND M. DWORKIN. 1981. The order Anim. Sci. 55:578-584. Myctobacterales.Pages 360-376 in The prokary-