Division of Comparative Physiology and Biochemistry, Society for Integrative and Comparative Biology Digestive Organ Sizes and Enzyme Activities of Refueling Western Sandpipers ( Calidris mauri ): Contrasting Effects of Season and Age

We examined seasonal and age-related variation in digestive organ sizes and enzyme activities in female western sandpipers (Calidris mauri) refueling at a coastal stopover site in southern British Columbia. Adult sandpipers exhibited seasonal variation in pancreatic and intestinal enzyme activities but not in digestive system or organ sizes. Spring migrants had 22% higher total and 67% higher standardized pancreatic lipase activities but 37% lower total pancreatic amylase activity than fall migrants, which suggests that the spring diet was enriched with lipids but low in glycogen. Spring migrants also had 47% higher total intestinal maltase activity as well as 56% higher standardized maltase and 13% higher standardized aminopeptidase-N activities. Spring migrants had higher total enzymic capacity than fall migrants, due primarily to higher total lipase and maltase activities. During fall migration, the juvenile’s digestive system was 10% larger than the adult’s, and it was composed differently: juveniles had a 16% larger small intestine but a 27% smaller proventriculus. The juvenile’s larger digestive system was associated with lower total enzymic capacity than the adult’s due to 20% lower total chitinase and 23% lower total lipase activities. These results suggest that juvenile western sandpipers may process food differently from adults and/or have a lower-quality diet. * Corresponding author; e-mail: [email protected]. Physiological and Biochemical Zoology 78(3):434–446. 2005. 2005 by The University of Chicago. All rights reserved. 1522-2152/2005/7803-5051$15.00 Introduction For birds, long-distance migration often is characterized by alternating periods of endurance flight and residence at stopover sites, where refueling occurs. Although fat is the primary fuel for these extended flights, some lean mass is also catabolized (Jenni and Jenni-Eirmann 1998), particularly from the digestive system (Karasov and Pinshow 1998; Battley et al. 2000). The digestive system serves two potentially conflicting functions during migration: it provides the means of extracting energy and nutrients from the diet while refueling at stopover sites, and it also serves as a source of lean mass that is catabolized in flight. After arrival at a stopover site, the digestive system may need to be reconstituted before maximal refueling rates can be attained (Hume and Biebach 1996; Karasov and Pinshow 2000). In association with these alternating demands, the digestive organs of long-distance migrants exhibit pronounced phenotypic flexibility, the rapid, reversible, and repeatable modulation of organ size or function (Piersma and Lindström 1997), during migration (Piersma 1998; McWilliams and Karasov 2001). The digestive system is a good model for evaluating refueling capacity because it is responsive to changes in the quantity, composition, and quality of the diet (Karasov 1996; McWilliams and Karasov 2001). The digestive system is also an energetically expensive organ system to maintain (Starck 1996, 1999), and it is expected to exhibit economical design (Diamond and Hammond 1992). Maximal refueling rates are achieved primarily by hyperphagia (Karasov 1990); however, diet composition and quality are also important (Bairlein 1998). The primary response of the digestive system to periods of hyperphagia is an increase in size, which results in an increase in volumetric digestive capacity (Karasov 1996). When diet composition changes or diet quality increases, the rate of nutrient assimilation can be increased by modulating the activities of inducible digestive enzymes (Karasov 1996; McWilliams and Karasov 2001). Dietary modulation studies in birds provide general support for the hypothesis that the major pancreatic and intestinal digestive enzymes are modulated in relation to their dietary substrate levels (reviewed in Karasov 1996; Sabat et al. 1998; Levey et al. 1999; Caviedes-Vidal et al. 2000). Inducible digestive enzymes may, therefore, provide information on diet composition or quality (Karasov and Hume 1997). Digestive capacity is an important determinant of refueling rate, and two important indices of digestive capacity are digestive system size and total hydrolytic capacity. We define the This content downloaded from 142.58.26.133 on Wed, 13 May 2015 17:39:11 PM All use subject to JSTOR Terms and Conditions Digestive Organs and Enzymes of Migrating Western Sandpipers 435 digestive system as consisting of the proventriculus, gizzard, pancreas, and small intestine because the primary function of these organs is associated with the breakdown and assimilation of dietary nutrients. Although the esophagus, ceca, and large intestine contribute to the overall size of the alimentary canal, they are not included in our definition of the digestive system. The esophagus may provide limited space as a storage organ for ingested food; however, it has limited secretory activity (Duke 1986). The primary roles of the ceca and large intestine are related to microbial digestion and water absorption, respectively (Duke 1986). We use the size of the digestive system (excluding the pancreas) as an index of volumetric capacity. We assess enzymic capacity by measuring the activities of a series of enzymes that are involved in the breakdown of the diet’s structural (e.g., chitin) and nutritional (fat, carbohydrate, and protein) components. Proventricular chitinase (EC 3.2.1.14; Bairoch 1993) is an important component of the digestive machinery of birds that consume arthropods because their exoskeletons are composed largely of chitin (Place 1996a). The primary benefit of chitinolysis appears to be an increase in the accessibility of the nutrients contained in or concealed by the exoskeleton (Jackson et al. 1992). The pancreatic lipases (pancreatic lipase, EC 3.1.1.3, and nonspecific carboxyl ester lipase, EC 3.1.1.1) hydrolyze triglycerides into fatty acids and glycerol in the lumen of the small intestine, but full lipolytic capacity is bile dependent in birds (Place 1992). Bile salts emulsify dietary triglycerides, allowing the pancreatic lipases to efficiently hydrolyze fatty acids at the lipid-water interface (Place 1996b). Pancreatic amylase (EC 3.2.1.1) hydrolyzes a-1-4-glucosidic bonds of complex soluble carbohydrates, and one of the primary products is maltose (Alpers 1987), which is hydrolyzed to glucose by maltase (EC 3.2.1.20), an intestinal disaccharidase. Aminopeptidase-N (EC 3.4.11.2; also known as leucine-aminopeptidase and aminooligopeptidase [Vonk and Western 1984]), an intestinal dipeptidase, has broad specificity in hydrolyzing oligopeptides into amino acids and accounts for almost all of the peptidase activity of the brush border membrane (Maroux et al. 1973). We examined seasonal and age-related variation in digestive organ sizes and enzyme activities in refueling western sandpipers (Calidris mauri Cabanis), a small-bodied (22–35 g) longdistance migrant that breeds mainly in subarctic Alaska and overwinters primarily on the Pacific coast, between California and Peru (Wilson 1994). Western sandpipers use a short-hop migration strategy to move between coastal stopover sites en route to and from the breeding grounds (Wilson 1994; Iverson et al. 1996), and the spring migration is somewhat more protracted than fall migration. In fall, the juveniles initiate their first southward migration after completing growth; this delays the onset of migration by approximately 1 mo relative to the adults (Wilson 1994). Western sandpipers are considered to be invertebrate generalist and are known to consume a variety of benthic invertebrates, primarily arthropod crustaceans, polycheate annelids, and bivalve molluscs, while refueling at coastal stopover sites (Wilson 1994). The digestive system of adult western sandpipers attains its maximal size in refueling migrants; however, the digestive system of the later-migrating juveniles is substantially larger than the adults’ (Guglielmo and Williams 2003). Age-dependent differences in digestive physiology may influence refueling rate and, thereby, the tempo or duration of migration; however, the functional significance of this age-dependent difference in digestive system size is unclear. For example, it is not known whether the juvenile’s larger digestive system is simply an enlarged version of the adult’s or if it is composed differently. Similarly, it is not known whether the enzymic capacity of the juvenile’s larger digestive system is higher, lower, or equal to the adult’s. We evaluate these alternatives through age-related comparisons of (1) the size of the digestive system and its component organs; (2) the standardized and total activities of five digestive enzymes; and (3) an index of the residual dietary energy contained in feces. To place these age-related comparisons in the context of migration during the annual cycle, we also examine seasonal variation in the size of the digestive system and the activities of digestive enzymes in adults. To enhance our interpretation of the seasonal comparisons, we predict the relationships between three pairs of inducible digestive enzymes including gallbladder bile and then examine the validity of these predictions for refueling adult western sandpipers. First, we predict a positive correlation between pancreatic lipase and gallbladder bile because bile salts are essential to the efficient utilization of dietary fats (Place 1992, 1996b). Second, we predict a positive correlation between pancreatic amylase and intestinal maltase because the substrate of maltase, maltose, is one of the primary products of the hydrolysis of complex soluble carbohydrates by amylase (Alpers 1987). Third, we predict a positive correlation between two intestinal enzymes, maltase and aminopeptidase-N. Although maltase and aminopeptidase-N utilize rather different substrates, Sabat et al. (1998) reported a positive correlation between these enzymes in two species of passerine birds. Material and Methods Fieldwork and Sample Collection Sandpipers were captured with mist nets (1 1/4 inch mesh, Avinet, Dryden, NY) and collected in accordance with permits from Environment Canada. Animal handling protocols were approved by the Simon Fraser University Animal Care Committee (B529) and conformed to the Canadian Committee for Animal Care Guidelines. Refueling sandpipers were captured at Boundary Bay, British Columbia, Canada (49 10 N, 123 05 W) during the fall migration of 1999 and the spring and fall migrations of 2000. Immediately after capture each bird was weighed (capture mass; 0.001 g) and culmen length was measured to assign sex: males ≤ 24.2 mm ! unknown (Page and Fearis 1971). During sex ! 24.8 mm ≤ females This content downloaded from 142.58.26.133 on Wed, 13 May 2015 17:39:11 PM All use subject to JSTOR Terms and Conditions 436 R. W. Stein, A. R. Place, T. Lacourse, C. G. Guglielmo, and T. D. Williams the fall migrations of 1999 and 2000, fecal samples were collected from refueling sandpipers to determine the amount of residual dietary organic matter, which would have contained any residual dietary energy that was excreted. In 1999, fecal samples were collected from 19 sandpipers: 4 males (1 adult and 3 juveniles), 7 females (5 adults and 2 juveniles), and 8 unknown-sex birds (1 adult and 7 juveniles). In 2000, fecal samples were collected from 26 sandpipers: 14 males (8 adults and 6 juveniles), 9 females (2 adults and 7 juveniles), and 3 unknown-sex birds (2 adults and 1 juvenile). Fecal samples were collected by placing individual sandpipers in ventilated plastic containers fitted with raised hardware cloth bottoms for 30–40 min. Fecal samples were rinsed from the containers with distilled water and frozen for subsequent analysis. During the spring and fall migrations of 2000, 53 refueling female sandpipers were collected to obtain organ samples for enzyme assays: 13 spring adults (May 1–9), 22 fall adults (July 4–28), and 18 fall juveniles (August 9–27). Collected individuals were transported to Simon Fraser University for processing within 2 h of capture. Immediately before dissection, culmen and tarsus were measured with digital calipers ( 0.01 mm), and each bird was reweighed (dissection mass; 0.001 g). Birds were killed by an overdose (4 mL/25 g) of a 1 : 1 mixture of ketamine hydrochloride (100 mg/mL) and xylazine (20 mg/mL) administered via intramuscular injection into the breast muscle. Birds were dissected immediately after death to collect gallbladder bile and the digestive organs. Bile was extracted from the gallbladder with a sterile 1-mL tuberculin syringe. The proventriculus was separated from the esophagus distal to the bifurcation of the trachea and at its connection with the gizzard. The small intestine was separated from the gizzard at the pylorus and from the large intestine immediately proximal to the ceca. The pancreas was delicately separated from the duodenal loop of the small intestine. Gastroliths and grit were removed from the gizzard by cutting it open and rinsing it with ice-cold physiological saline (350 mOsm/kg H2O). Digesta was purged from the lumen of the small intestine by suturing a gavauge needle to its proximal end and gently flushing the contents with icecold physiological saline. The evacuated small intestine was cut in half at Meckel’s diverticulum, and each resulting section was cut in half again; this resulted in four intestinal sections: the duodenum and the proximal, mid, and distal ileum. After excision, the organs were rinsed in ice-cold physiological saline, adherent fat and mesenteries were removed, and the organs were blotted dry, weighed ( 0.001 g), and flash frozen (liquid N2, 196 C) in labeled cryovials. At the end of each dissection the sex of the bird was verified, keel length was measured with digital calipers ( 0.01 mm), and the carcass was stored at 20 C. For proventricular chitinase, sample sizes were reduced by 2 for spring adults and by 1 for fall adults due to lost tissue. Total gallbladder bile was not reported because bile was lost from several individuals during collection. For intestinal maltase and aminopeptidase-N, sample size was reduced by 4 for the fall juveniles because the intestines from these birds were used for histological studies by Stein and Williams (2003). Measurement of Proventricular Chitinase Protein Extraction and Assay. Individual proventriculi were homogenized for 30 s using an OMNI 5000 homogenizer, setting 6, in 10 mL of a 1.0 M acetate buffer, pH 4.5. Homogenates were centrifuged at 10,000 g for 10 min, and the supernatant was stored at 80 C until it was assayed. Proventricular protein, used to standardize chitinase activity, was determined using a Pierce BCA (bicinchoninic acid) kit, adapted for use in microtiter plates. Absorbance was recorded at 590 nm on a microplate reader (Vmax; Molecular Devices, Menlo Park, CA) using bovine serum albumin as a standard. Assays were performed in triplicate, with a mean coefficient of variation of 3.5%. Chitinase Activity Assay. Chitinase activity was measured using the tritiated chitin method of Molano et al. (1977) with modifications by Cabib and Sburlati (1988). Proventricular extracts were thawed on ice, and 100 mL was diluted 1 : 10 with the 1.0 M acetate buffer, pH 4.5. The reaction mixture consisted of 70 mL distilled water, 15 mL of suspended acetyl-[H]-chitin (0.5 , ), 5.0 mL mCi/mg 0.23 mCi/mM N-acetyl-d-glucosamine [NAG] of the 1.0 M acetate buffer, and 10 mL of diluted proventricular extract. The reaction was initiated by the addition of the proventricular extract, incubated at 41 C on a shaker for 60 min, and stopped by the addition of 300 mL of 10% (w/v) trichloroacetic acid. The suspension was centrifuged for 5 min at 500 g, and 200 mL of the supernatant was transferred to a scintillation vial. Radioactivity (dpm) was determined with a Beckman LS6500 autoanalyzer using the window preset for tritium. Activity was calculated from dpm, which was due to the liberation of tritiated NAG. Assays were performed in triplicate, with a mean coefficient of variation of 6.8%. Chitinase activity is reported as total activity (mg NAG/min) and as specific activity (mg NAG/min/mg protein). Analysis of Gallbladder Bile Bile samples were thawed on ice and centrifuged at 20,000 g at 4 C for 10 min. Biliary protein was precipitated by combining 10 mL of bile with 10 mL of methanol, and the resulting methanolic bile was diluted 1 : 50 with distilled H2O. Bile salt concentrations were assayed using 3 a-hydroxysteroid dehydrogenase (E.C. 1.1.150; Coleman et al. 1979). The reaction mixture contained 170 mL of a hydrazine buffer (33 mM sodium pyrophosphate and 0.33 M hydrazine sulfate), pH 9.5, 20 mL of 0.33 mM NAD , and 5 mL of diluted methanolic bile. The reaction was started by the addition of 5 mL (0.2 Units) This content downloaded from 142.58.26.133 on Wed, 13 May 2015 17:39:11 PM All use subject to JSTOR Terms and Conditions Digestive Organs and Enzymes of Migrating Western Sandpipers 437 of 3 a-hydroxysteroid dehydrogenase in a potassium phosphate buffer, pH 7.5. After a 1-h incubation at 25 C, absorbance was recorded at 340 nm on a microplate reader at 25 C with taurocholate as the standard. Concentration was calculated from the increase in absorbance at 340 nm, which was due to the production of NADH. Bile salt assays were performed in duplicate with a mean coefficient of variation of 3.4%. Measurement of Pancreatic Hydrolytic Enzymes Protein Extraction and Assay. Each pancreas was homogenized for subsampling by powdering it in a stainless steel mortar and pestle that was partially submerged in liquid N2 ( 196 C). The resulting powder was placed in a labeled cryovial and frozen at 80 C until extraction. For amylase and lipase activity measurements, subsamples of the powdered pancreas, 10–0 mg, were extracted in 1–3 mL of a 10 mM sodium acetate buffer, pH 4.8 (buffer volume depending on tissue mass), for 15 min at 4 C. The buffer contained 0.9% (w/v) NaCl, 0.2% (w/v) Triton X-100, 3 mM sodium taurocholate, 2 mM hydrocinnamic acid, and 1 mM benzamidine. The extract was sonicated (Branson model 460) on ice for 1 min at 50% power and 50% duty cycle with a microtip and then centrifuged at 20,000 g for 20 min at 4 C. The supernatant was removed and frozen at 80 C until it was assayed. Total pancreatic protein in each extract, used to standardize amylase and lipase activity, was determined using a Pierce BCA kit, adapted for use in microtiter plates. Absorbance was recorded at 590 nm on a microplate reader using bovine serum albumin as a standard. Protein assays were performed in triplicate, with a mean coefficient of variation of 6.4%. Lipase Activity Assay. Lipase activity was measured at 41 C and pH 8.0 using a pH-stat (TTT80 Radiometer) with olive oil, a long-chain triglyceride, as the substrate and 0.02 N NaOH as the neutralizing base. Long-chain fatty acids ( ) are inpk ∼ 6.8 completely ionized below pH 9.0; consequently, lipolytic activity is underestimated at pH 8.0 when a long-chain triglyceride is the substrate. Although pH 8.0 is more realistic physiologically than pH 9.0, the reported activities are not absolute. Titrametric assay of bile-dependent lipolytic activity was performed using a 0.5% (w/v) gum arabic–stabilized emulsion of olive oil, which was sonicated for 3 min at a power setting of 30 on a Branson sonicator immediately before use. The 10-mL assay volume contained 0.5 mL of substrate emulsion in a 2 mM tris-maleate buffer (pH 8.0), 150 mM NaCl, 1 mM CaCl2, and 0.02% mM sodium azide. Nonenzymatic base uptake was measured by incubating the reaction mixture for 2 min without enzyme. The reaction was initiated by the addition of 50 mL of 200 mM sodium taurocholate, a common avian bile salt, followed by 100 mL of proventricular extract. The presence of bile salts fully activates bile-dependent carboxyl-ester lipase and the colipase-pancreatic lipase. Preliminary studies demonstrated that maximal activation by taurocholate occurred at concentrations above 1.0 mM. Activity was calculated from the amount of base required to neutralize liberated fatty acids and maintain pH 8.0. Assays were performed in duplicate with a mean coefficient of variation of 1.6%. Lipase activity is reported as total activity (mM/min) and as specific activity (mM/min/mg protein). Amylase Activity Assay. Amylase activity was determined with an amylase assay kit (Sigma Diagnostics). The reaction was conducted at pH 7.0, using 4,6-ethylidine (G7)-p-nitrophenyl (G1)-a, d-maltoheptaside as a substrate, which results in aglucosidic release of p-nitrophenol. The reaction was initiated by adding 5 mL of pancreatic extract, diluted 1 : 10, to 200 mL of reagent. After 2 min, the rate of increase in absorbance at 405 nm is directly proportional to amylase activity. After allowing the reaction to equilibrate for 2 min, absorbance was recorded continuously at 405 nm for 5 min at 41 C on a microplate reader. Activity was calculated from the mean rate of increase in absorbance at 405 nm, which was due to the production of p-nitrophenol. Amylase activity is reported as total activity (mM/min) and as specific activity (mM/min/mg protein). Measurement of Intestinal Hydrolytic Enzymes Tissue Homogenates and Standardization. The entire small intestine was homogenized for 30 s using an OMNI 5000 homogenizer, setting 6, in 6–14 mL of 350-mM mannitol in 1 mM Hepes/KOH, pH 6.5 (buffer volume depending on tissue mass). The final volume of the homogenate was calculated as the sum of intestine mass (g) and buffer volume (mL), assuming a density of 1.0 (g/mL) for the small intestine. For three fall adults and three fall juveniles, the four intestinal sections were homogenized separately and were used to examine variation in enzyme activity along the length of the small intestine. Intestinal enzyme assays were conducted using tissue homogenates rather than isolated vesicles because yields and activities of membrane preparations are often low and variable (Martı́nez del Rio et al. 1995 and references therein). Intestinal enzyme activities were calculated on the basis of absorbance standards constructed for glucose and p-nitroanilide. Martı́nez del Rio et al. (1995) provide justification for our choice of standardization. Maltase Activity Assay. Maltase activity was measured according to the methodology of Dahlqvist (1964), with modifications by Martı́nez del Rio et al. (1995). Assay mixtures, 56 mM, were prepared by adding 0.202 g maltose to 10 mL of maleate/NaOH, pH 6.5. The stop-development reagent was a mixture of 250 mL of 0.5 M monobasic/dibasic phosphate buffer (pH 6.5), 250 mL 1 M Tris/HCl (pH 6.5), and one 500-mL bottle of glucose (Trinder) reagent powder (Sigma Diagnostics). Tissue This content downloaded from 142.58.26.133 on Wed, 13 May 2015 17:39:11 PM All use subject to JSTOR Terms and Conditions 438 R. W. Stein, A. R. Place, T. Lacourse, C. G. Guglielmo, and T. D. Williams Figure 1. From the western sandpiper, pHoptima of proventricular chitinase (squares), intestinal maltase (triangles), and intestinal aminopeptidase-N (circles). homogenates were thawed and then diluted 1 : 50 in 350 mM mannitol in 1 mM Hepes/KOH, pH 6.5. To start the reaction, 100 mL of diluted homogenate was combined with 100 mL of 56 mM maltose, and then the tube was vortexed and incubated at 40 C. After 15 min, 3.0 mL of the stop-development reagent was added, and the reaction mixture was allowed to develop for 20 min at room temperature. Absorbance was measured at 505 nm in a Beckman DU-64 spectrophotometer. Activity was calculated from the absorbance at 505 nm, which was due to the production of red chinonimin dye from the Trinder reaction. Assays were performed in duplicate, with a mean coefficient of variation of 2.6%. Maltase activity is reported as total activity (mM/min) and as tissue-specific activity (mM/min/ g tissue). Aminopeptidase-N Activity Assay. Aminopeptidase-N activity was determined using the methodology of Roncari and Zuber (1969) with l-alanine-p-nitroanilide as the substrate. A 2.04 mM l-alanine-p-nitroanilide assay mixture was prepared by dissolving 125 mg l-alanine-p-nitroanilide in double distilled water, 250 mL final volume. The dissolved substrate was combined with 250 mL of 0.2 M phosphate buffer, pH 6.5. Tissue homogenates were thawed on ice, and 10 mL of undiluted homogenate was combined with 1.0 mL of the l-alanine-pnitroanilide assay mixture at 40 C. The tube was vortexed and then incubated at 40 C for 15 min. The reaction was stopped by the addition of 3.0 mL of ice-cold 2N acetic acid and vortexing. Absorbance was measured at 384 nm in a Beckman DU-64 spectrophotometer. Activity was calculated from the increase in absorbance at 384 nm, which was due to the production of p-nitroanilide. Assays were performed in triplicate, with a mean coefficient of variation of 2.5%. AminopeptidaseN activity is reported as total activity (mM/min) and as tissuespecific activity (mM/min/g tissue). Determination of pHoptima The optimal reaction pH for proventricular chitinase and those for intestinal maltase and aminopeptidase-N were determined by running the assays as described above, while varying the pH. The resulting activities were standardized to the enzyme’s maximal activity to facilitate presentation (Fig. 1). The pHoptima for each enzyme was determined by fitting a quadratic to the data, taking the derivative of the fitted equation, and solving for the positive pH value where the slope was equal to 0. The chitinase , the maltase , and the pH p 3.1 pH p 5.8 optima optima aminopeptidase-N . The activities reported for pH p 6.9 optima these three enzymes are adjusted to their pHoptima and, therefore, represent maximal values.