isolation and characterization of a chitinase from the nine-banded ...

ISOLATION AND CHARACTERIZATION OF A CHITINASE FROM THE NINE-BANDED ARMADILLO, DASYPUS NOVEMCINCTUS STEPHANIE A. SMITH, LYNN W. ROBBINS, AND JOHN G. STEIERT

Southwest Missouri State University, Department of Biology, 901 S. National, Springfield, MO 65804

The insectivorous diet of the nine-banded armadillo, Dasypus novemcinctus, is abundant in the polysaccharide chitin. To detennine if armadillos produce a chitinase that aids in digestion of chitinous exoskeletons, armadillos were collected, and samples of their gastric and pancreatic tissues were removed. Tissues were lyophilized and stored at -20°C until resuspended in appropriate buffers for studies of pH and temperature optima. Extract from resuspended tissues was assayed for chitinase activity using a colorimetric procedure that detected N-acetyl-D-glucosamine (NAG), the monomeric unit in chitin. Chitinase activity was present only in gastric tissues. Optimal pH for chitinase activity from gastric tissues of armadillo was 5.0, and optimal temperature was 50-60°C. This chitinase was purified five-fold. Key words: Dasypus novemcinctus, nine-banded armadillo, chitin, chitinase, insectivory, myrmecophagy Chitin is the second most abundant biopolymer in nature, ranking only behind cellulose. Chitin is found in the integument of arthropods, namely crustaceans and insects, molluscs, algae. protozoa, and hyphal cell walls of fungi (Muzzarelli, 1977). N-acetylD-glucosamine (NAG) is the monomeric unit within chitin. The NAG polymerizes via 13-1,4 linkages to form the dimer chitobiose, and the same linkage is assumed by chitobiose when it polymerizes into chitin. Chitin is broken down by two naturally occurring enzymes---chitinase and chitobiase. These are "true" chitinases, not general lysozymes with chitinolytic activity (Jeuniaux and Cornelius, 1978). Chitobiose is the main product of hydrolysis by chitinase, and chitobiase hydrolyzes chitobiose and residual chitotriose into NAG monomers (Jeuniaux, 1966). Although bacteria are the most convenient models for chitinase studies, the discovery of vertebrate chitinases by leuniaux (1961) pioneered a new field of studies of chitinase, and the range of vertebrate classes possessing the enzyme is broad. Trout, Journal of Mammalogy, 79(2):486-491, 1998

lizards, frogs, pigeons, bats, pigs, and mice are among some of the chitinase-producing vertebrates identified by leuniaux (1961), and the primate Perodicticus pollo also has been found to produce chitinase (Jeuniaux and Cornelius, 1978). Noticeable trends in the production of chitinase by vertebrates include localization of chitinases in gastric and pancreatic tissues of members of the higher taxa, and the lack of chitobiase in these taxa. All vertebrates identified as having chitinases are insectivorous or omnivorous. Chitinase production is not induced by diet. as was shown in experiments where animals fed a diet devoid of chitin continued to produce chitinase (Jeuniaux, 1963). Since 1972, the nine-banded armadillo, Dasypus novemcinctus, has extended its range from the south and southeastern United States into the southern half of Missouri and the southeastern one-third of Kansas (Taulman and Robbins, 1996). Abundance of ants and termites in most habitats makes them the primary food of many insectivorous mammals-an eating pattern referred to as myrmecophagy. Grubs and other lar486

May 1998

SMITH ET AL.-ARMADILLO CHLTINASE

val forms tend to have a higher content of fat and accessible protein than mature insects and are presumably the armadillo's preferred food. Availability and abundance, however, are the primary detenninants of diet composition of armadillos and other insectivorous mammals (Redford and Dorea, 1983). The insectivorous diet of the armadillo in the southernmost portions of its range has been described (Layne and Glover, 1985; McDonough, 1992; Talmage and Buchanan, 1954). The armadillo's diet in Missouri is myrmecophagous, containing ::529.3% ants during summer (Lippert, 1994). In this paper, we present evidence of endogenous production of chitjnase that may facilitate efficient digestion of the annadillo's predominantly arthropod diet. MATERIALS AND METHODS

Tissue collection and storage.---Gastric and pancreatic tissues were removed from freshly killed armadillos. Tissues were rinsed with 0.8% saline. placed in a ziploc bag. and stored on ice for transport to the laboratory. After thorough washing with saline, tissues were homogenized in 0.1 M phosphate buffer (pH 7.0) using a tissue homogenizer (Lourdes Instrument Corp., Lourdes, IL). The homogenate was lyophilized and stored in sealed containers at - 20 C. Samples were labeled according to the armadillo from which they were taken and for the tissue type (gastric or pancreatic). Crude enzyme extract was prepared by resuspending lyophilized tissues in 20 ml of cold 0.02 M citrate-phosphate buffer (pH 5.0) containing sodium azide. The resuspended tissue was centrifuged at 20.000g at 3°C for 30 min, and the supernatant was used immediately for chitinase assays or frozen at - 20°C for purification. Little loss of activity was observed by storage of supernatants at -20°C. Chitin medium.-Colloidal chitin was prepared from crude chitin according to the method of Hunter-Cevera et al. (1986). Chitin medium was prepared by suspending ca. IS g (wet weight) of colloidal chitin in SOO ml of distilled water. In a separate flask, minimal-salts medium was prepared by combining 0.7 g K2 HP0 4 , 0.3 g KH 2P04 , O.S g MgS04 X SH 20, 0.01 g FeS04 X 7H2 0, 0.001 g ZnS0 4 , 0.001 g MnCI 2 • and D

487

IS.0 g agar and adjusting the volume to 500 ml with distilled water. The pH of both flasks was adjusted to 6.8-7.0. After autoclaving for 20 min at 121°C, flasks were combined, swirled to mix, and used to prepare agar plates (Hunter-Cevera et al., 1986). Chitin was the only source of carbon and nitrogen in this medium. Isolation of chitinolytic bacteria.-Samples from stomach, small intestine, colon. and feces of freshly killed armadillos were inoculated onto two plates of chitin media using sterile swabs. Half of the plates were incubated aerobically at 30°C, and half were incubated anaerobically in BBL Gas-Pak bags (Beckton-Dickinson. Cockeysville. MD) at 30°C. Isolates were maintained on chitin medium and characterized physiologically for taxonomic classification (Smith, 1996). NAG assay for chitinase activity.-Reissig's colorimetric procedure for detennination of Nacetylamino sugars was used to estimate chitinase activity (Reissig et al., 1955) with modified reagent compositions. Reagents were 5% potassium tetraborate and p-dimethylaminobenzaldehyde (DMAB) reagent that contained 1.25 ml of 12 N HCI, 48.75 ml of glacial acetic acid, and 1.6 g of DMAB. The DMAB reagent was freshly prepared for each assay. A standard curve was generated using known concentrations of N-acetyl-D-glucosamine. Protein concentrations were detennined using the Bio-Rad Protein Assay Kit (Richland. CA) with a standard of bovine serum albumin. Specific chitinase activity is defined as I-lg NAG released h- l mg protein-I. Estimation of pH and temperature optima.About 0.05 g of colloidal chitin was suspended in 1 ml of the appropriate buffer. A 0.1 mI sample of crude enzyme extract from lyophilized tissues was added, and the final volume was adjusted to 1.5 ml with the appropriate buffer: 0.02 M HCVKCl (pH 2.0), 0.02 M citrate-phosphate (pH 3.0). 0.02 M citrate-phosphate (pH 4.0), 0.02 M citrate-phosphate (pH 5.0), 0.02 M citrate-phosphate (pH 6.0). or 0.02 M phosphate (pH 7.0). Tubes were incubated at 55°C for 12 h. Controls in which buffer replaced extract were incubated simultaneously. For determination of temperature optima, extracts were added to 0.05 g of colloidal chitin suspended in 1 ml of 0.02 M citrate-phosphate buffer (pH 5.0). Final volume was adjusted to 1.5 ml using the same buffer, and tubes were incubated at 5°C intervals from 30°C to 70°C.

488

Vol. 79, No.2

JOURNAL OF MAMMALOGY

TABLE

I.-Specific activity of armadillo chitinase at varying temperatures. Temperature ("C)

Sample no.

22

30

35

40

45

50

55

12G 12G 12G 8G

26.7

54,7 46,1

36.5

44.S 58.3

60.7

54.3

67,8 89.1 59.3 3.9

85.7 80.8 63.9

35.1

4.2

60

65

70 86.1

85,0 5.3

76.5 3.0

• Not tested in this trial.

Samples from one armadillo were incubated for 3 h, and samples from another armadillo were incubated for 20.25 h. Chitin was removed by centrifugation. and samples were assayed for NAG. Controls contained buffer in place of extract. Partial purification of chitinase.-About 0.7 g (dry weight) of lyophilized armadillo tissue was resuspended in 25 mt of 0.02 M citratephosphate buffer (pH 5.0), A 2.5-ml sample was removed and stored on ice for NAG and protein assays. One gram of colloidal chitin (wet weight) was added to the remaining 22.5 ml of crude extract and stirred on ice for 1 h to facilitate adsorption of the enzyme onto substrate. Enzyme bound to the chitin was collected by centrifugation (17.OCX>g, 20 min. 3°C) and the supernatant (supernatant "A") was stored at - 20°C. The chitin was suspended in 15 ml of 0.02 M TRIS-HCl buffer (pH 9.0) to elute enzymes. The eluate was kept on ice and gently agitated manually about every 5 min for 30 min. Chitin was removed by centrifugation and discarded. A 2.5-ml sample of supernatant was removed and stored on ice for NAG and protein assays. The remaining supernatant was saturated with 70% ammonium sulfate and stirred for 1 h on ice. The precipitate was collected by centrifugation, and the pellet was resuspended in 12.5 ml of 0.02 M citrate-phosphate buffer (pH 5.0). A 2.5-ml sample was removed for assays and stored on ice. and the remaining 10 ml was equilibrated to 10% ammonium sulfate saturation. stirred on ice for 1 h. and centrifuged. The pellet was resuspended in 2.5 ml of 0.02 M citrate-phosphate buffer (pH 5.0). All samples. except the crude extract. were passed through a PD-IO column (Pharmacia, Uppsala, Sweden) for desalting or equilibration with the citrate-phosphate buffer. Equilibration of columns and elution of samples were done

with 0.02 M citrate-phosphate buffer (pH 5.0). All samples were assayed for protein. Chitinase activity was detennined for each sample by suspending 0.05 g of colloidal chitin in 1.0 ml of 0.02 M citrate-phosphate buffer (pH 5.0) and 0.5 ml of enzyme samples from purification steps or 0.1 ml of crude extract. Final volume was adjusted to 1.5 ml with the buffer. Controls were ~oiled for 5 min. and all samples were incubated for 13.5 or 14 h at 55°C. RESULTS

Optimal pH and temperature.-Chitinase activity was apparent in gastric tissue samples. but no activity was detected in pancreatic tissues. The possibility that detected activity was microbial was ruled out by adding sodium azide to resuspended tissues to control microbial growth and streaking samples onto colloidal chitin agar. Lack of microbial growth on this medium indicated that there were no chitinolytic microorganisms in tissue samples. At each temperature and pH tested, three separate samples and a control, in which buffer replaced extract, were incubated. Resulting absorbances from colorimetric NAG assays were averaged, and those averages were used to determine NAG production at each temperature or pH using standard curves. Protein concentrations were determined from the original extract used for NAG assays. Results of temperature assays on gastric tissues indicate that the optimal temperature for armadillo chitinase lied between 50-60°C (Table I). The optimal pH of 5.0 was clearly indicated by these experiments (Fig. 1).

Partial purification of armadillo chiti-

SMITH ET AL.-ARMADILLO CHITINASE

May 1998

pH optimum for gastric samples

--- ---t--- - - - - - - ,

pH

FIG. I.-The relationship between pH and chitinase-specific activity of the gastric tissue of the armadillo.

nase.-Chitinase was purified 4.6-fold. with a yield of 3.1 % relative to chitinase activity in crude samples (Table 2). Controls contained extract but were boiled to eliminate chitinase activity. Supernatant collected after the adsorption step presumably contained chitobiase and was added to samples and controls after they had been incubated. That was done to ensure that lack of chitobiase in extracts would not limit release of NAG, which would prevent detection of chitinase activity (Cornelius et al., 1976). The adsorption step was where ca. 95% of chitinase activity in the original sample was lost (Table 2). DISCUSSION

The armadillo produces a gastric chitinase that has optimal activity at a pH of 5.0 and a temperature between 50-60°e. While optimal pH was consistent with those previously reported for chitinases. optimal temperature was high, although a temperature optimum of 55-60°C has been reported

TABLE

489

for chitinase of the fish Lateolabrax japonicus (Okutani, 1978). Optimal activity of chitinase at a pH of 5.0, with low activity at a pH of 3.0 and a high temperature optimum, lead to speculation of how active armadillo chitinase is in vivo. The pH in stomachs of dogs, ponies, and pigs can range from below 3.0, 2 h after feeding, to >5.0 within 8 h after feeding (Stevens and Hume, 1995). The armadillo stomach, however, is aglandular and consists of stratified squamous epithelium cranially and pyloric mucosa in the caudal region. It lacks gastric tissue responsible for secretion of pepsinogen and HCl (Stevens and Hume. 1995). The pH therefore may not be as greatly reduced after feeding, allowing for chitinase activity. Most mammals maintain a body temperature at ca. 37°C. McNab (1980) showed that body temperature of the armadillo remained at ca. 34.5°C at ambient temperatures from 5 to 30°C. In a study of armadillos in the field. over a wider range of ambient temperatures (-14-30°C), Schell (1994) found that body temperature was maintained at ca. 35°e. Activity of the armadillo's chitinase was shown to be low :::::::35°C (Table 1). Chitinase is therefore released into an environment with temperatures that are relatively low compared with conditions for its optimal activity in vitro. An exochitinase cleaves dimers or monomers from the end of the chitin polymer. Detection of NAG using the colorimetric assay described would not be possible unless the armadillo's chitinase is an exochitinase. chitobiase is present, or both. Assay incubations with partially purified samples

2.-Purification of armadillo chitinase.

Volume (ml)

Protein (mg)

Total activity (units)

Specific activity (U/mg)

Yield

Step Crude extract Adsorption 70% ppt.a 10% ppt. •

22.5 12.5 10.0 3.5

33.50 1.60 0.77 0.22

1,283 64 44 40

38.3 39.9 57.3 181.5

100.0 5.0

• ppt.

=

ammonium Rulfate precipitation.

(%)

3.4

3.1

Purification fold 1.0 1.0 1.5 4.7

490

Vol. 79, No.2

JOURNAL OF MAMMALOGY

showed increased NAG production when supernatant A was added, which is indicative of chitobiase activity (chitinase activity in supernatant A was corrected for in controls). It was not detennined if the cleavage mechanism of the armadillo's chitinase is exochitinolytic or endochitinolytic, in which the chitin polymer would be cleaved randomly. Chitinase could not be sufficiently purified by our methods to allow separation of activities of chitinase and chitobiase. A feature preventing basic kinetic studies of chitinase is that its activity plateaus as a function of time, which results from inhibition of the enzyme by NAG (Smucker, 1986). The non-linearity of enzymatic chitin hydrolysis also may be due to the structural heterogeneity of chitin (Tilburg and Thomas, 1994). An assay for which chitobiase activity is not required, such as use of tritiated chitin as a substrate ( Molano et al., 1977), shorter incubation times, and greater purification of the armadillo's chitinase would meet more closely requirements for accurate kinetic studies. Several lysozymes have been shown to have chitinolytic activity (Rupley and Gates. 1967). Enzymatic extracts from the primate P. potto had chitinolytic and lysozymic activities (Cornelius et aI., 1976). Contribution of the lysozyme to chitinolysis could not be separated from chitinase-specific activity in crude extracts from P. potto. After adsorption of extracts from P. potto onto colloidal chitin, however, lysozyme activity was not detectable, but chitinase activity remained. It is possible that lysozyme is contributing to chitinolytic activity found in crude extracts from the armadillo's gastric tissues. However, increase in chitinasespecific activity after the adsorption step would not be expected if the majority of NAG being released was due lysozyme activity, because lysozyme does not adsorb to chitin (Cornelius et aI., 1976). Presence of chitinase provides no evidence of the value of this enzyme to the armadillo's metabolic needs. Low absorp-

tion efficiencies of NAG and NAG oligomers in tissues of several seabirds suggests that chitinase may function to increase rate of breakdown of prey exoskeletons, increase digesta transit, or facilitate access to proteins complexed with chitin rather than serving a direct nutritive function (Jackson et aI., 1992). In the same seabirds, it was shown that a substantial component of chitin degradation in vivo was due to endogenous chitinase activity. The remaining chitinolytic activity was attributed to bacteria in the gut or enzymes present in the food consumed by these birds. Seabirds are consumers of krill and cephalopods, and chitinases produced by these prey items could contribute to breakdown of their own exoskeleton in digestive tracts of seabirds studied (Jackson et al., 1992). In contrast, the intestine of the shark Scylliorhinus canicula reportedly absorbs NAG more efficiently than glucose (Alliot, 1967). Studies of NAG absorption in the armadillo would provide information about the role of chitinase. Degradation of chitin by microorganisms seems possible in the gut of the armadillo, and these microorganisms possibly could use NAG for cell wall synthesis or metabolize it further by fermentation (Capps et aI., 1966). We have obtained one microbial isolate from the armadillo's small intestine that is a candidate for a symbiotic chitinolytic function. Besides being chitinolytic and facultatively anaerobic, it is a member of the Enterobacteriaceae, which commonly reside in digestive tracts of vertebrates (Smith, 1996). Additional work needs to be done before the role of this isolate in the armadillo can be understood clearly, and the possibility that it is simply a transient soil microorganism that has been ingested by the armadillo cannot be ruled out. ACKNOWLEDGMENTS

We thank the Department of Biology at Southwest Missouri State University for financial support of this research and the Missouri Department of Conservation personnel at the

May 1998

SMITH ET AL-ARMADILLO CHITINASE

Drury-Mincy Wildlife Reserve near Branson, Missouri, for allowing collection of armadillos. Special thanks to P. Steiert for editing comments. LITERATURE CITED ALUOT. E. 1967. Absorption intestinale de I'N-acetylglucosamine chez la petite roussette: Scylliorhinus canicula. Comptes Rendus des Seances de la Societe de Biologie et de ses FiliaIes, 161:2544-2546. CAPPS. J. C, M. R. SHETLAR, AND R. H. BRADFORD. 1966. Hexosamine metabolism, I. The absorption and metabolism, in vivo, of orally administered Dglucosamine and acetyl-D-glucosamine in the rat. Biochimica et Biophysica Acta, 127:194-204. CORNELIUS, C, G. DANDRIFOSSE, AND C JEUNIAUX. 1976. Chitinolytic enzymes of the gastric mucosa of Perodicticus potto (Primate Prosimian): purification and enzyme specificity. International lournal ofBiochemistry, 7:445-448. HUNTER-CEVERA, J. C, M. E. FONDA, AND A. BELT. 1986. Isolation of cultures. pp. 14, in Manual of industrial microbiology and biotechnology (A. L. Demain and N. A. Solomon, eds.). American Society for Microbiology, Washington, D.C., 466 pp. JACKSON, S., A. R. PLACE, AND L. J. SEIDERER. 1992. Chitin digestion and assimilation by seabirds. The Auk, 109:758-770. JEUNIAUX, C 1961. Chitinase: an addition to the list of hydrolases iv the digestive tract of vertebrates. Nature, 192:135-136. - - - . 1963. Chitin et Chitinolyse, un Chapitre de la Biologie Moleculaire. Masson, Paris, 181 pp. - - - . 1966. Chitinases. Pp. 644-650, in Methods in enzymology. Volume 8 (E. E Neufield and V. Ginsburg, eds.). Academic Press, New York, 759 pp. IEUNIAUX, C, AND C CORNELIUS. 1978. Distribution and activity of chitinolytic enzymes in the digestive tract of birds and mammals. pp. 542-549, in Proceedings of the first international conference on chitinlchitosan (R. A. A. Muzzarelli and E. R. Pariser, eds.). Massachusetts Institute of Technology, Cambridge, 652 pp. LAYNE, J. N., AND D. GLOVER. 1985. Activity patterns of the common long-nosed armadillo Dasypus nov· emcinctus in south Central Florida. Pp. 407-417, in The evolution and ecology of armadillos, sloths, and vennilingua" (G. G. Montgomery, ed.). Smithsonian Institution Press, Washington, D.C, 451 pp. LIPPERT, K. J. 1994. Food habits, distribution and impact of the nine-banded armadillo in Missouri. M.S. thesis, Southwest Missouri State University, Springfield, 88 pp. McDONOUGH, C. M. 1992. The behavior and ecology of nine-banded armadillos (Dasypus novemcinctus) in South Texas. Ph.D. dissertation, University of California, Davis, 101 pp. McNAB, B. K. 1980. Energetics and the limits to tem-

491

perate distribution in armadillos. Journal of Mammalogy, 61:606-627. MOLANO, J., I. POLACHECK, AND E. CABIB. 1977. A rapid and sensitive assay for chitinase using tritiated chitin. Analytical Biochemistry, 83:648-656. MUZZARELLl, R. A. A. 1977. Chitin. Pergamon Press, Inc., Elmsford, New York, 309 pp. OKUTANI, K. 1978. Chitin digestion in the digestive tract of fish. Pp. 554-56, in Proceedings of the first international conference on chitinlchitosan (R. A. A. Muzzarelli and E. R. Pariser, cds.). Massachusetts Institute of Technology, Cambridge, 652 pp. REDFORD, K. H., AND 1. G. DOREA. 1983. The nutritional value of invertebrates with emphasis on ants and tennites as food for mammals. Journal of Zoology (London), 203:385-395. REISSIG, 1. L., J. L. STROMINGER, AND L. F. LELOIR. 1955. A modified colorimetric method for the estimation of N-acetylamino sugars. Journal of Biological Chemistry, 217:959-966. RUPLEY, J. A., AND V. GATES. 1967. Studies on the enzymic activity of lysozyme-H. The hydrolysis and transfer reactions of N-acetylglucosamine oligosaccharides. Proceedings of the National Academy of Sciences, 57:496-510. SCHELL, P. T. 1994. Home range, activity period, burrow use, and body temperatures of the nine-banded armadillo (Dasypus novemcinctus) on the northern edge of its range. M.S. thesis, Southwest Missouri State University, Springfield, 35 pp. SMITH, S. A. 1996. Isolation and characterization of chitinase from Dasypus novemcinctus, the ninebanded armadillo. M.S. thesis, Southwest Missouri State University, Springfield, 74 pp. SMUCKER, R. A. 1986. Interpretive model for enzymatic chitin hydrolysis. Pp. 254-258, in Chitin in nature and technology (R. A. A. Muzzarelli, C. Jeuniaux, and G. W. Gooday, eds.). Plenum Press, New York, 583 pp. STEVENS, C. E., AND I. D. HUME. 1995. Comparative physiology of the vertebrate digestive system. Cam· bridge University Press, Cambridge, United Kingdom, 400 pp. TALMAGE, R. V, AND G. D. BUCHANAN. 1954. The armadillo (Dasypus novemcinctus). A review of its natural history, ecology, anatomy and reproductive physiology. Rice Institute Pamphlet, 41:1-135. TAULMAN, J. E, AND L. W. ROBBINS. 1996. Recent range expansion and distributional limits of the ninebanded armadillo (Dasypus novemcinctus) in the United States. lournal of Biogeography, 23:635648. TILBURG, A. U. B., AND M. D. THOMAS. 1994. Extracellular (3-N-acetylhexosaminidases of Gliocladium virens RVI4. Enzyme and Microbial Technology, 16:849-854. Submitted 2 February 1997. Accepted 5 September 1997. Associate Editor was Janet K. Braun.