Alligators and Crocodiles Have High Paracellular Absorption of Nutrients, But Differ in Digestive Morphology and Physiology.

Integrative and Comparative Biology Advance Access published June 9, 2015

Integrative and Comparative Biology Integrative and Comparative Biology, pp. 1–19 doi:10.1093/icb/icv060

Society for Integrative and Comparative Biology

SYMPOSIUM

Alligators and Crocodiles Have High Paracellular Absorption of Nutrients, But Differ in Digestive Morphology and Physiology Christopher R. Tracy,1,*,†,‡ Todd J. McWhorter,§ C. M. Gienger,*,ô J. Matthias Starck,k Peter Medley,# S. Charlie Manolis,** Grahame J. W. Webb*,** and Keith A. Christian*

From the symposium ‘‘Integrated Biology of the Crocodilia’’ presented at the annual meeting of the Society for Integrative and Comparative Biology, January 3–7, 2015 at West Palm Beach, Florida. 1

E-mail: [email protected]

Synopsis Much of what is known about crocodilian nutrition and growth has come from animals propagated in captivity, but captive animals from the families Crocodilidae and Alligatoridae respond differently to similar diets. Since there are few comparative studies of crocodilian digestive physiology to help explain these differences, we investigated young Alligator mississippiensis and Crocodylus porosus in terms of (1) gross and microscopic morphology of the intestine, (2) activity of the membrane-bound digestive enzymes aminopeptidase-N, maltase, and sucrase, and (3) nutrient absorption by carrier-mediated and paracellular pathways. We also measured gut morphology of animals over a larger range of body sizes. The two species showed different allometry of length and mass of the gut, with A. mississippiensis having a steeper increase in intestinal mass with body size, and C. porosus having a steeper increase in intestinal length with body size. Both species showed similar patterns of magnification of the intestinal surface area, with decreasing magnification from the proximal to distal ends of the intestine. Although A. mississippiensis had significantly greater surface-area magnification overall, a compensating significant difference in gut length between species meant that total surface area of the intestine was not significantly different from that of C. porosus. The species differed in enzyme activities, with A. mississippiensis having significantly greater ability to digest carbohydrates relative to protein than did C. porosus. These differences in enzyme activity may help explain the differences in performance between the crocodilian families when on artificial diets. Both A. mississippiensis and C. porosus showed high absorption of 3-O methyl D-glucose (absorbed via both carrier-mediated and paracellular transport), as expected. Both species also showed surprisingly high levels of L-glucose-uptake (absorbed paracellularly), with fractional absorptions as high as those previously seen only in small birds and bats. Analyses of absorption rates suggested a relatively high proportional contribution of paracellular (i.e., non-mediated) uptake to total uptake of nutrients in both species. Because we measured juveniles, and most paracellular studies to date have been on adults, it is unclear whether high paracellular absorption is generally high within crocodilians or whether these high values are specific to juveniles.

Introduction Much of our knowledge regarding crocodilian nutrition, growth, and biology has come from propagation and farming of captive individuals for use in the

leather industry (Coulson and Hernandez 1983; Garnett and Murray 1986; Webb et al. 1987; Garnett 1988). Attempts to optimize growth rates of individuals, and therefore production, by altering

ß The Author 2015. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved. For permissions please email: [email protected].

Downloaded from http://icb.oxfordjournals.org/ at Mount Allison University on June 10, 2015

*Research Institute for the Environment and Livelihoods, Charles Darwin University, Darwin, NT 0909, Australia; † Department of Zoology, University of Melbourne, Parkville, VIC 3010, Australia; ‡Department of Biological Science, California State University Fullerton, Fullerton, CA 92831, USA; §School of Animal and Veterinary Sciences, University of Adelaide, Roseworthy Campus, Adelaide, SA 5371, Australia; ôDepartment of Biology and Center of Excellence for Field Biology, Austin Peay State University, Clarksville, TN 37044, USA; jjDepartment of Biology, University of Munich (LMU), Munich, Germany; #Department of the Environment, Environmental Research Institute of the Supervising Scientist, GPO Box 461, Darwin, NT 0801, Australia; **Wildlife Management International Pty. Limited, Berrimah, NT 0828, Australia

2

Materials and methods Animal husbandry In February 2007, 7 American alligators (A. mississippiensis) and 10 saltwater crocodiles (C. porosus) (all juveniles approximately 1-year old 2 months and with no more than 2 months variation in age among individuals) were selected from long-term captive populations at Crocodylus Park, Berrimah, Northern Territory, Australia. Subsets of these were used for each experiment, as described below. After recording body mass (A. mississippiensis: 817.3 236.0 g; C. porosus: 695.5 109.4 g) and total length (A. mississippiensis: 681.6 73.7 mm; C. porosus: 648.1 25.3 mm), individuals were placed in 1000 L fiberglass tanks in conspecific groups of four to five per tank. Animals were maintained outdoors under shade at ambient temperatures; average daily minimum temperature was 24.68C and average daily maximum temperature was 31.58C (Australian Bureau of Meteorology). Animals were fed ad libitum three times per week on a diet of diced red meat (beef or horse). Because the intestinal epithelium of crocodylians undergoes significant morphometric and histological changes in response to feeding (e.g., Starck et al. 2007), we fed all animals at the same time so they were all in the same condition for comparisons of epithelium morphometry. Measurements of nutrient-absorption We conducted in vivo measurements of nutrient uptake in animals that were fed ad libitum. We used 3H-labeled L-glucose as a probe to measure paracellular (i.e., non-mediated) transport of nutrients. L-Glucose does not interact with the SGLT protein, a transmembrane glucose transporter, so the paracellular pathway is the only way for this probe to appear in the bloodstream (Wright et al. 1980; Ikeda et al. 1989; Fine et al. 1993; Uhing and Kimura 1995; Lane et al. 1999). L-Glucose is also not metabolized once absorbed, so acts as a metabolically inert probe (Karasov and Cork 1994). We used 14 C-labeled 3-O methyl D-glucose as a probe for transport both by the paracellular pathway and by carrier-mediated transport simultaneously; this probe is also not metabolized once absorbed. The combination of these two probes allowed us to calculate the relative importance of paracellular transport (see below). Each animal (n ¼ 6 per species) received two rounds of treatment, one an oral dose and the other an intraperitoneal injection. The sequence of treatments was randomized for each individual. For


diet have revealed differences in the utilization of food among species of crocodilians (Garnett 1986, 1988; Staton 1988; Webb et al. 1991, 2013). It has been suggested that these differences in food utilization could fall along taxonomic lines between the Alligatoridae (alligators and caimans) and Crocodylidae (true crocodiles). However, patterns in digestive performance, including differences in the efficiency of assimilating energy, protein and dry matter, also seem closely aligned with physiological (e.g., passage-time of digesta) or ecological (e.g., consumption of gastroliths) differences among species (Davenport et al. 1990, 1992). Farmed alligators seem to be more efficient at converting food to body mass than farmed crocodiles (Staton 1988; Webb et al. 1991). Alligators also grow well on a variety of different diets, including feeds containing inexpensive carbohydrates and plant protein (Kercheval and Little 1990; Staton et al. 1990a, 1990b). Crocodiles, however, seem to require a narrower range of foods, and only thrive when fed a diet of lean animal protein (Garnett and Murray 1986; Read 2000; Peucker and Jack 2006; Webb et al. 2013). Comparative studies of crocodilians’ digestive morphology and physiology are reasonably scarce. As a consequence, it is unclear whether species from the Crocodylidae and Alligatoridae have differences in structure and/or function of the gut that correlate with their relative abilities to cope with diets containing different substrates, particularly vegetable protein and carbohydrates. In the present study we explored the mechanisms and anatomy that underlie digestive performance in American alligators (Alligator mississippiensis) and saltwater crocodiles (Crocodylus porosus). We predicted that differences in morphology of the digestive tract and/or in the physiology of digestive processes could explain the differences in performance on various diets exhibited by these two crocodilian species. To determine whether alligators and crocodiles have fundamental differences in their digestive systems that might affect their digestive function, we examined a series of indicators of digestive morphology and function, including the general morphology of the intestine, surface-area magnification and architecture of intestinal tissue, activity of membrane-bound proteases and disaccharidases, and mechanisms of absorbing nutrients. We measured the relative contributions of carrier-mediated and non-mediated (i.e., paracellular) uptake of nutrients in order to determine whether differences in the relative importance of these mechanisms could partially account for observed differences in digestive performance between these species.

C. R. Tracy et al.

3

Digestive physiology of crocodilians

1986). Fractional absorption (f) of the probes was calculated as: f ¼

AUCoral : AUCinjection

ð1Þ

Following standard pharmacokinetic procedures (Welling 1986), the area from t ¼ 0 to t ¼ x min (when the final blood sample was taken) was calculated using the trapezoidal rule. The area from t ¼ x to t ¼ 1 was calculated as: AUCx ! 1 ¼ ðC at t ¼ xÞ=Kel ;

ð2Þ

where Kel (hr1) is the elimination-rate constant for removal of the probe from plasma, estimated as the slope of the linear regression of the final four to six log-transformed plasma concentrations against time. The total AUC0!1 was obtained by summing the two areas. The time-course of carbohydrate-probe absorption was calculated for individual animals following Loo and Riegelman (1968), using kinetic constants derived by fitting probe-elimination data to biexponential models (Fig. 7, Table 1) and the plasma concentrations following oral administration of each compound (Fig. 7). Apparent probe-absorption rate was calculated at each time-point as the percentage of total uptake absorbed per hour. In order to estimate the proportion of 3-O methyl D-glucose that was passively absorbed, we assumed that the rate of absorption of L-glucose can serve as a proxy for the non-mediated absorption of 3-O methyl D-glucose, once adjusted for the small difference in molecular mass (MM). Because diffusion in water declines with MM1/2 (Smulders and Wright 1971), each value of absorption of L-glucose was decreased by 3.6% [100 (1941/2–1801/2)/1941/2]. Assuming that the absorption of 3-O methyl D-glucose represents the sum of passive and carrier-mediated absorption, the ratio of the amounts absorbed (L-glucose/3-O methyl D-glucose) indicates the proportion of absorption of 3-O methyl D-glucose that occurs via nonmediated pathways. Chang and Karasov (2004) showed that this ratio was not significantly affected when measurements were conducted under conditions that are non-saturating versus relatively saturating for nutrient transporters, so it provides a robust estimate of the proportional contribution of passive to total nutrient uptake. Dissection for studying gross morphology of the intestine Just before a regular feeding (i.e., 3.5 days after their last meal), four A. mississippiensis and five C. porosus


both treatments, the dose was approximately 5.55 105 Bq kg1 for each probe. All animals were force-fed a sausage of finely minced beef at the time of treatment so that they all had an equivalent meal that was equal to 2% of body mass and was similar in size to meals these animals would voluntarily eat. The oral treatment was given by injecting the probe solution into the sausage before force-feeding. After the day of treatment, animals were fed ad libitum on the normal schedule of three times per week. Blood samples (300–400 mL) were drawn from the spinal venous sinus, just posterior to the head (Elsey et al. 2008), for a background measurement before the administration of the probes, and subsequently at 4 h after the treatment, then at 12 h intervals until 136 h, 24 h intervals until 232 h, and finally at 48 h intervals until 328 h after treatment (total of 19 samples). The second treatment was administered 42 days after the first (28 days after the last blood sample from the first treatment). Blood plasma was separated from cells by centrifugation and 100 mL sub-samples were transferred to 7-mL glass scintillation vials. A 6 mL aliquot of Ultima Gold LLT scintillation cocktail (Perkin Elmer) was added to each vial and mixed thoroughly with the plasma. Scintillation vials were counted using a lowbackground liquid scintillation counter (Tri-Carb model 3100TR, PerkinElmer) with a count time of 10 min per sample. Efficiency and quench correction standards were prepared in the same manner as samples; water spiked with the 3H-labeled L-glucose or 14 C-labeled 3-O methyl D-glucose probes (1130 Bq) was used instead of plasma. Nitromethane (from 0 to 100 mL) was used as a quenching agent in the quench-correction standards. Efficiency and quench-correction standards were measured under the same conditions as samples and counted for 30 min. Beta activity was determined in the channel region 0–12 keV for 3H, and in the channel region 12–156 keV for 14C. The internal software on the liquid scintillation counter was used to determine the gross count rates of samples and backgrounds in disintegrations per minute (dpm). The activity of each probe in plasma, C (dpm probe mg1 plasma dpm1 dose), was normalized to dose and plotted as a function of time since dosing, t (h). The absorbed amounts of the probes were calculated from the areas under the curves for the injection and oral-dose treatments (AUC ¼ area under the curve of plasma activity versus time). This simple method does not require assumptions about the number or size of the pools, or about the kinetics of the process (e.g., first or second order) (Welling

4

C. R. Tracy et al.

Table 1 Summary of bi-exponential fits of plasma-probe concentration versus time from injection/elimination experiments shown in Fig. 7 C. porosus Parameter A (dpm plasma g1 dosed dpm1) 1

L-glucose

0.00311 0.00013

A. mississippiensis

3-O methyl D-glucose 0.00314 0.00026

L-glucose

3-O methyl D-glucose

0.00232 0.00012

0.00313 0.00005

a (min )

0.03028 0.00174

0.03112 0.00302

0.04387 0.00293

0.04449 0.0020

B (dpm plasma g1 dosed dpm1)

0.00119 0.00014

0.00156 0.00028

0.00118 0.00013

0.00067 0.00010

b (min1)

0.00565 0.00064

0.00699 0.00099

0.00975 0.00080

0.00846 0.00108

Note: The model was Ct ¼ Aeat þ Bebt.

Assays of digestive enzymes Preparation of tissues and assays of enzymes were conducted using procedures as outlined by Iglesias et al. (2009). Subsamples of intestinal sections were

thawed to near room temperature and mixed with an ice-cold buffer of 300 mmol L1 mannitol in 1 mmol L1 Hepes/KOH, pH 7.5. Tissues were ground for 30 s at 22,000 rpm using a T25 Ultra Turrax disperser (Ika Werke GmbH and Co. KG, Staufen, Germany), and the tissue homogenate was then frozen at 808C for later analysis. Aminopeptidase-N (APN) activity was assayed to give an index of the digestion of protein. APN activity was determined using the general procedures of Roncari and Zuber (1969), using L-alanine-p-nitroanilide as a substrate for the reaction. Homogenates of intestinal tissue were thawed to near room temperature and were diluted using a 1:2 ratio with 300 mmol L1 mannitol in 1 mmol L1 Hepes/ KOH, pH 7.5. A 10 mL sample of the diluted homogenate was then added to 200 mL of assay solution (2.0 mmol L1 L-alanine-p-nitroanilide in 0.2 mol L1 NaH2PO4/Na2HPO4 buffer, pH 7). The reaction was incubated at 328C for 20 min and then arrested by adding 600 mL of chilled 2 N acetic acid. Absorbance of the reaction products was measured using Beckman DU-64 spectrophotometer set to 384 nm. Under these experimental conditions, the hydrolysis of L-alanine-p-nitroanilide proceeds linearly with time (Martıńez del Rio et al. 1995). Maltase and sucrase activity were assayed to give an index of disaccharide digestion by these membranebound enzymes. Disaccharidase activity was determined via the methods of Dahlqvist (1984), as modified by Martıńez del Rio et al. (1995). Homogenates of sections of crocodiles’ intestines were assayed for maltase activity without dilution, but the homogenates of sections from alligators’ intestines were diluted to a 1:2 ratio with a buffer of 300 mmol L1 mannitol in 1 mmol L1 Hepes/KOH, pH 7.5. Sucrase assays were performed on undiluted homogenates of intestinal sections. A 30 -mL sample of the homogenate (or diluted homogenate) was then added to 30 mL of 56 mmol L1 maltose or sucrose substrate solution in a 0.1 mol L1 maleate/ hydroxide buffer, pH 6.5. After incubation at 328C for 20 min, 400 mL of a glucose-assay reagent was added. The glucose-assay reagent (GAGO20,


were euthanized with an injection of Euthasol (sodium pentobarbital) into the heart, and were immediately dissected. The age of the animals at euthanasia was 16 2 months. The intestinal tract from the stomach to the cloaca was excised from the body cavity, cut longitudinally, and rinsed with a Ringers solution containing 0.1 mol L1 NaCl, 1.8 mmol L1 KCl, 2.0 mmol L1 CaCl2, 1.0 mmol L1 MgCl2, 5.0 mmol L1 Hepes–NaOH, pH 7.6. Total mass and length of the intestine were recorded and the intestine was divided into four sections corresponding to general morphology. The small intestine was divided roughly into thirds (proximal, middle, and distal sections), and the large intestine comprised the final, and much shorter, colonic section. Each section was blotted dry and weighed (1 mg). Nominal surface area of each section was calculated from measurements (0.01 mm) of length and width, after the section was cut longitudinally and opened flat. For each of the sections, two subsamples of intestine 5–10 mm in length were dissected out, weighed (1 mg), measured for length and width (0.01 mm), flash frozen in liquid nitrogen, and stored at 808C for later use in determining enzymatic activity. An additional two subsamples were taken at this time for examination of micro-structure of the gut (see below). To allow a comparison of the scaling of intestinal morphology over a wider range of body sizes than just the juveniles used in the experiments on enzyme activity and paracellular uptake, larger animals (A. mississippiensis: n ¼ 8; C. porosus n ¼ 7) were intermittently obtained from production activities at Crocodylus Park. These were larger individuals culled from the captive population used for skin production. Totalbody length, and length and mass of the intestine were measured as above (see Statistical Analysis below for details on conversion of total body-length to snoutvent length [SVL] for these individuals).

5


Micro-structure of the gut Tissue samples were preserved for histology in 5% paraformaldehyde in 0.1 mol L1 phosphate buffer at pH 7.4 and 48C for at least 48 h. They were then washed in buffer, dehydrated through a graded series of ethanol to 96% ethanol and embedded in hydroxyethyl methacrylate (Historesin, Fa. Leica, Germany). Embedded material was sectioned to a thickness of 2 mm, mounted on slides, and stained with Methylene Blue–Thionin (¼Ruedeberg stain). Histological sections were studied using an Olympus BX51TF microscope (Olympus, Hamburg, Germany) equipped with a CT5 digital camera (Olympus). We used the software analySIS (Olympus) to capture images, which were then imported into SigmaScan 5.0 (Systat Software, San Jose´, USA) for morphometric analysis. Histological morphometry For the proximal, middle, and distal intestine, we measured the epithelial-surface magnification as the epithelial surface over a baseline defined by the inner circular muscle layer. Measurements were made by tracing the epithelial surface with a cursor on a digitizing tablet and calculating its total length divided by the length of the baseline, expressed as a dimensionless ratio. For each animal, five sections were measured for the surface-magnification. An average value from those five measurements per individual was calculated to avoid pseudoreplication of data. Although fixation of tissue in isotonic and buffered paraformaldehyde, dehydration to 96% ethanol and embedding in methacrylate minimizes artefacts

arising from embedding, the procedure may result in about 10% shrinkage of tissue from its original size (Bo¨ck 1989). However, all samples of tissue were treated identically and thus can be compared directly. Statistical analysis Results are presented as mean standard deviation, unless otherwise noted. Comparisons of length and mass of the intestine between species were conducted using analysis of covariance (ANCOVA) using SVL as a covariate. Length was used as a covariate rather than mass because of the logistic difficulties in obtaining accurate measurements of mass from the largest animals (42 m). For some individuals (n ¼ 4 A. mississippiensis, n ¼ 9 C. porosus), SVL data were not available, so total body length (TL) was converted to SVL using allometric relationships from data collected on harvested animals from the farmed population (Webb et al. 2013). For C. porosus: SVL ¼ 0.57554 þ 0.49258 * TL; n ¼ 630, P ¼ 0.0001, R2 ¼ 0.997; size range 390–1950 mm SVL. For A. mississippiensis: SVL ¼ 0.821 þ 0.477 * TL; n ¼ 331, P ¼ 0.0001, R2 ¼ 0.988, size range 130– 590 mm SVL. When data came from repeated measures of individuals, they were analyzed using linear mixed-effect models (JMP 8.0, SAS Institute, Cary, North Carolina; SPSS 14.0, IBM Corporation, Somers, New York). Individuals were considered a random effect and all treatment main effects (e.g., species, intestinal section) were considered fixed. Post-hoc comparisons of levels of treatment were conducted using Tukey’s honest significant difference (HSD) with Bonferroni correction for multiple comparisons and a ¼ 0.05 for all tests. We used one-factor ANCOVA with intestinal segment (proximal, middle, distal) as factor and body length as covariate for statistical analysis of histological morphometry (SPSS 22.0, IBM Corporation). When body length was not significant as a covariate, we removed it from the model and re-ran the test as an ANOVA followed by a Tukey’s HSD post-hoc test for comparison of multiple means. We used two-way ANOVA to compare surface magnification factor between species (Graph Pad Prism 6, Graph Pad Software, La Jolla, California, USA). Fractional absorption (f) values for paracellular probes were arcsine-square root transformed prior to statistical comparisons (Sokal and Rohlf 1995). Fits of the monoexponential and biexponential elimination-model to semi-log plots of paracellular probe-concentration versus time after injection were


Sigma–Aldrich, St Louis, MO, USA) was made up with equal parts of 1 mol L1 Tris/HCl, pH 7, and 0.5 mol L1 phosphate buffer (NaH2PO4/Na2HPO4, pH 7). After a further 30 min of incubation at 328C, the reaction was arrested by adding 400 mL of 12 N H2SO4 and the absorbance of the resulting solution was measured at 540 nm. Intestinal enzymatic activity was calculated on the basis of absorbance and expressed as mmol min1 mg1 or mmol min1 cm2 for each intestinal section. Karasov (1988) and Martıńez del Rio (1990) provided justification for using nominal area and wet-weight of tissue to standardize intestinal hydrolysis and rates of transport. Total enzymatic activity was calculated by summing the totals for each section and expressing them as mmol min1. We assumed that any decline in enzymatic activity due to the post-feeding interval before sampling was similar between the species.

6

C. R. Tracy et al. Table 2 Gut morphology and the activity of membrane-bound enzymes for Alligator mississippiensis (n ¼ 4) and Crocodylus porosus (n ¼ 5)

Intestinal mass (g)

Intestinal length (cm)

Nominal surface area (cm2)

Villous magnification factor

17.8 1.7

30.2 9.2

13.4 1.8

Maltase activity (nmol min1)

Aminopeptidase activity (nmol min1)

5.8 1.3

39.6 12.7

Alligator mississippiensisa Proximal

8.0 2.3

Middle

6.0 2.4

21.8 5.9

26.1 10.0

9.9 1.3

3.5 1.8

29.6 5.9

Distal

3.4 1.7

19.5 7.7

21.4 14.2

4.5 1.5

0.4 0.3

7.4 4.4

Colon

4.5 1.3

6.2 0.8

16.8 4.1

Total

21.9 7.4

65.2 14.1

97.6 29.6

20.6 2.6

25.9 3.1

—

0.4 0.1

5.5 1.1

10.1 2.1

81.9 4.1

1.2 0.5

38.6 9.2

Crocodylus porosusb Proximal

5.2 0.7

9.3 2.0

2.8 0.2

23.3 5.3

21.0 3.1

6.4 1.2

0.5 0.2

24.8 7.1

2.1 0.2

19.9 4.5

21.9 7.2

4.5 1.0

0.2 0.1

10.7 10.1

Colon

2.6 0.8

4.7 0.7

6.2 1.8

Total

12.8 0.7

68.5 9.8

72.4 4.1

—

0.1 0.06

6.7 1.9

2.0 0.3

82.9 10.2

Notes: Values are mean standard deviation. a Body mass ¼ 1312.5 280.0 g; total body length ¼ 784.8 52.7 mm. b Body mass ¼ 812.0 107.6 g; total body length ¼ 683.4 33 mm.

compared with F-tests (Motulsky and Ransnas 1987). Apparent rates of paracellular probe-absorption were compared within species individually using twofactor repeated measures ANOVA (probe and time as factors) with Bonferroni correction on post-hoc tests (Graph Pad Prism 5, Graph Pad Software). Permits and disposition of samples All research was conducted under approval from the Charles Darwin University Animal Ethics Committee (permit No. A06017). Histological slides are deposited at the Department of Biology, University of Munich.

Results The alligators and crocodiles that were used in the studies of paracellular uptake and enzymatic activity were of similar size, and did not differ in intestinal length (F1,5 ¼ 2.36, P ¼ 0.18) or intestinal mass (F1,5 ¼ 0.16, P ¼ 0.71) (Table 2). However, the allometry of the length of the digestive tract did differ between alligators and crocodiles when we looked at the broader range of body sizes (Fig. 1). There was a significant interaction term with body size (SVL) in the ANCOVA for intestinal length (F1,20 ¼ 36.02, P50.0001) with crocodile intestinal length increasing more quickly with body size, and a significant main effect of species (F1,20 ¼ 79.14, P50.0001), with crocodiles generally having longer intestines at a particular body size. However, there

was no difference in the allometry of intestinal mass (F1,14 ¼ 3.46, P ¼ 0.08), nor was there any difference in overall intestinal mass between the species (F1,14 ¼ 4.20, P ¼ 0.06). Intestinal morphometry Surface-magnification of the intestine was highest in the proximal section, where the villi increased the absorptive surface by a factor of 13.4 1.8 in alligators and 9.3 2.0 in crocodiles. The surface-magnification was less in the middle and distal sections (Figs. 2–4, Table 2). Within both species, intestinal section had a significant effect on surface-magnification, but body length was only significant as a covariate in crocodiles (alligators: intestinal section F2,9 ¼ 34.16, P50.001; body length F1,8 ¼ 1.78, P ¼ 0.219; crocodiles: intestinal section F2,9 ¼ 18.3, P ¼ 0.001; body length F1,9 ¼ 5.82, P ¼ 0.039). In both crocodiles and alligators, epithelial surface-magnification was significantly larger at the anterior end of the small intestine (proximal section) than at the posterior end (distal section). The middle section was significantly different from both the proximal and distal sections in alligators but was only significantly different from the proximal section in crocodiles (Fig. 2). Surface magnification was significantly higher in alligators than in crocodiles (F1,19 ¼ 18.71, P ¼ 0.0004) and there was a significant interaction between species and intestinal section (F2,19 ¼ 4.99, P ¼ 0.0182). Across intestinal sections, surface magnification was


Middle Distal


7

Fig. 1 Allometry of intestinal length (top) and mass (bottom) with SVL for American alligators (Alligator mississippiensis) and saltwater crocodiles (Crocodylus porosus). The allometric relationship between body size and intestinal length was different between species, with crocodiles having both a steeper slope and greater length over the range in body size measured (C. porosus: y ¼ 0.377x 58.218, R2 ¼ 0.99, P50.0001; A. mississippiensis: y ¼ 0.212 x 17.012, R2 ¼ 0.85, P50.0001), but the allometry of intestinal mass was not significantly different between species (y ¼ 0.304x 96.078, R2 ¼ 0.95, P50.0001).

significantly higher in alligators than in crocodiles in the proximal and middle sections (P50.001) but not in the distal section (P40.05; Fig. 2). We multiplied the surface-magnification of each intestinal segment of each individual by the nominal surface area of that segment and summed these to estimate total surface area for each animal. We did not measure surface-magnification for the large intestine, so assumed a magnification factor of 1 (i.e., no magnification) for this segment. The total areas were compared between species with ANCOVA, using total body length as the covariate. There were no significant differences between species (F1,8 ¼ 0.0008, P ¼ 0.978). Intestinal histology The epithelium of the proximal segment of the small intestines showed the typical features of an epithelium

found in actively digesting animals (Figs. 3–5). In both species, the enterocytes of the mucosa’s epithelium were arranged as a single-layered epithelium of high columnar cells with a prominent brush border at the apical end. The enterocytes were loaded with lipid droplets, but the degree of loading differed between cells, with some enterocytes appearing overloaded with lipid droplets while others contained only few. Numerous lymphocytes and mast cells populated the interstitial space of the epithelium. Large paracellular spaces could be recognized between the enterocytes of the villi (Fig. 5). Especially toward the tip of the villi, the paracellular spaces were a conspicuous feature of the epithelium, even under light microscopy. The lamina propria mucosae contained large capillaries and swollen lymphatic vessels (lacteals). A characteristic feature of the apical region of the villi was frequent lesions that disrupted epithelial integrity. The degree of epithelial disintegration was not quantified, but qualitative inspection of slides from all individuals indicated quite clearly that the disintegration was strongest at the tip of the villi, while, the bases of the villi usually were in good condition. The observed epithelial disintegration was much more expressed in crocodiles than in alligators. Within species, the height of the villi (Figs. 3 and 4) and the surface-area magnification both declined from anterior to posterior (Fig. 2). We also


Fig. 2 Intestinal surface-magnification factors in the proximal, middle, and distal sections of the small intestine in Alligator mississippiensis and Crocodylus porosus. Both species showed a significant reduction in surface-magnification from the proximal to the distal end of the intestine, but alligators overall had significantly greater surface-magnification than did crocodiles. Bars are means SD. Different letters denote significant differences (P50.05) among sections within a species (capital letters for comparisons within A. mississippiensis; lower case letters for comparisons within C. porosus), and asterisks indicate significant differences between species within an intestinal section (**P50.001).

8

C. R. Tracy et al.

observed that the loading of enterocytes with lipid droplets was more intense in the anterior segments than in the middle and posterior segments. The enterocytes of the posterior segment contained no obvious lipid droplets. While this observation was based on inspection of sections of all individuals, the lipid-loading was not quantified. An anterior to posterior gradient was also observed for epithelial disintegration. While the anterior segments showed a high degree of disintegration, the middle and posterior segments showed little or none. Because all tissue samples were treated identically we can safely assume that the observed epithelial disruptions were condition-dependent and not methods-dependent. In the latter case, we would have seen the same damage

to tissue in all sections of all individuals. Another (unquantified) anterior–posterior gradient was seen in the distribution of the lymphatic tissue. In the more anterior segments the lymphocytes were distributed equally throughout the tissue, with occasional aggregations between the intestinal crypts. The lamina propria mucosae of the posterior segments contain more distinct lymphatic follicles (nodule lymphatici aggregati). The number of goblet cells increased in the more posterior position of the sections. Activity of digestive enzymes Activity of aminopeptidase was higher overall for crocodiles than for alligators, regardless of whether


Fig. 3 Light-microscopic micrographs of histological sections through different segments of the small intestine of Alligator mississippiensis. (A) Anterior region of the small intestine under low-power magnification (LP), (B) anterior region of the small intestine under highpower magnification (HP), (C) middle region of the small intestine under LP, (D) middle region of the small intestine under HP magnification (E), posterior region of the small intestine under LP, (F) posterior region of the small intestine under HP. The staining in all slides is Ruedeberg-stain as described in Materials and methods section. Abbreviations: c, capillary; lpm, lamina propria mucosae; lv, lymphatic vessel; me, lamina epithelialis mucosae; de, disintegrating epithelium; tm, tunica muscularis. Scale bars: A, C, E ¼ 40 mm; B, D, F ¼ 10 mm.


9

activity was normalized by mass of the intestinal segment (per mg of intestine: F1,7 ¼ 6.29, P ¼ 0.04) or by nominal surface area (per cm2 of intestine: F1,7 ¼ 6.70, P ¼ 0.04), although post-hoc tests indicated that they were not statistically different between species for any individual section of the intestine (Fig. 6, Table 2). Activity of aminopeptidase also was significantly different between intestinal sections (per mg intestine: F3,21 ¼ 12.11, P50.0001; per cm2 intestine: F3,21 ¼ 7.07, P ¼ 0.002; Fig. 6).

However, when the mass-specific activity for a segment was multiplied by mass of that segment, and then summed over the whole intestine, total activity of aminopeptidase over the whole intestine was not statistically different between species (F1,7 ¼ 0.031, P ¼ 0.865; Table 2). Activity of maltase differed between species (per mg intestine: F1,7 ¼ 62.17, P50.0001; per cm2 intestine: F1,7 ¼ 16.80, P ¼ 0.0003) and also by section of the intestine (per mg intestine: F3,21 ¼ 27.22,


Fig. 4 Light-microscopic micrographs of histological sections through different segments of the small intestine of Crocodylus porosus. (A) Anterior region of the small intestine under low-power magnification (LP), (B) anterior region of the small intestine under highpower magnification (HP), (C) middle region of the small intestine under LP, (D) middle region of the small intestine, under HP, (E) posterior region of the small intestine under LP, (F) posterior region of the small intestine under HP. The staining in all slides is Ruedeberg-stain as described in Materials and methods section. Abbreviations: lpm, lamina propria mucosae; lf, lymphatic follicle; me, lamina epithelialis mucosae; tm, tunica muscularis; ts, tunica serosa. Scale bars: A, C, E ¼ 40 mm; B, D, F ¼ 10 mm. (This figure is available in black and white in print and in color at Integrative and Comparative Biology online.)

10

C. R. Tracy et al.

P50.0001; per cm2 intestine: F3,21 ¼ 13.14, P50.0001; Fig. 6, Table 2); it was significantly higher for alligators than for crocodiles in the proximal sections of the intestine (regardless of normalization), and in the middle sections of the intestine (normalized to mass), but was not statistically different between species for distal, large intestinal sections (Fig. 6). Additionally, total activity summed over the whole intestine was significantly higher in alligators than in crocodiles (F1,7 ¼ 74.5, P50.0001; Table 2). Initial results showed that activity of sucrase in all intestinal sections of both species was at (alligators) or below (crocodiles) the lower limits of the detection-levels of the assay, even when homogenates of tissue were assayed undiluted, so we did not conduct further assays for sucrase. As an index of the ability to digest carbohydrate relative to protein, we calculated the ratio of maltase to aminopeptidase. For crocodiles, this was 0.025 0.007 and for alligators it was 0.124 0.032 (F1,7 ¼ 41.3, P ¼ 0.0004).

Absorption of nutrients After the oral force-fed treatment, the mean activity of in plasma peaked at 28 h both for alligators and for crocodiles (Fig. 7). The peak activity for 3-O methyl D-glucose in plasma peaked 28 h after the treatment for crocodiles, and at 16 h after treatment for alligators (Fig. 7). After injection, activities in plasma peaked at 4 h for both probes in both species, and declined exponentially thereafter (Fig. 7). Fractional absorption (f) of both probes was high in both species. For 3-O methyl D-glucose, fractional absorption was 0.72 0.20 for alligators and 0.67 0.14 for crocodiles; fractional absorption of Lglucose was 0.64 0.17 for alligators and 0.84 0.21 for crocodiles. There were no significant differences between species for fractional absorption of either probe (F1,9 ¼ 1.70, P ¼ 0.23), nor were there significant differences in fractional absorption between probes (F1,9 ¼ 0.18, P ¼ 0.68). Plots of post-injection data for alligators were significantly a better fit by a model of bi-exponential L-glucose


Fig. 5 (A) Alligator mississippiensis, small intestine, proximal region; high-power magnification showing paracellular spaces (*), intracellular droplets of lipid (arrow heads), and intraepithelial lymphocytes. (B) Crocodylus porosus, small intestine, proximal region; high-power magnification showing paracellular spaces (*), intracellular droplets of lipid (arrow heads), and intraepithelial lymphocytes. (C) Crocodylus porosus, high-power magnification of the mucosa epithelium of the proximal part of the small intestine showing paracellular spaces. Abbreviations: bb, brush-border of enterocytes; c, capillary; arrow heads, intracellular droplets of lipid; lc, intraepithelial lymphocytes; lv, lymphatic vessels; *, paracellular spaces. Scale bar ¼ 10 mm. (This figure is available in black and white in print and in color at Integrative and Comparative Biology online.)

11


decline than by one of mono-exponential decline for both probes (3-O methyl D-glucose: F2,14 ¼ 158.3, P50.001; L-glucose: F2,14 ¼ 176.1, P50.001). Biexponential models were also a better fit for postinjection data for crocodiles for both probes (3-O methyl D-glucose: F2,14 ¼ 84.9, P50.001; L-glucose: F2,14 ¼ 211.6, P50.001). The parameters from the bi-exponential fits of post-injection data (Table 1) and the mean concentrations of plasma following oral administration of each probe were thus used to calculate the time-course for absorption, following Loo and Riegelman (1968). Apparent probe absorption rate (% absorbed per hour) followed a similar pattern for both probes in both species, declining rapidly during the initial absorptive phase (Fig. 8A, C). For both alligators and crocodiles, there was no significant effect of probe on apparent absorption rate (F1,170 ¼ 0.86, P ¼ 0.38 and F1,136 ¼ 3.56, P ¼ 0.096, respectively), but there was a highly significant effect of time (F17,170 ¼ 32.51, P50.0001 and F17,136 ¼ 47.94, P50.0001, respectively).

The ratio of absorption of L-glucose to that of 3-O methyl D-glucose provides an estimate of the proportion of total uptake of glucose that occurs passively (Fig. 8B, D). Averaged over the initial absorptive phase (0–50 h), when the bulk of uptake is occurring, this ratio was 0.72 in alligators and 0.85 in crocodiles (i.e., 72% and 85% of the total uptake of 3-O methyl D-glucose was occurring via the passive pathway). The ratios exceeded 0.57 in alligators and 0.74 in crocodiles at all sampling time-points, indicating significant non-mediated uptake of glucose in both species. At later times of sampling in crocodiles this ratio exceeded 1; we believe that this is simply a mathematical artifact of the modeling process, caused by the slightly (but not significantly) higher f value for L-glucose relative to 3-O methyl D-glucose in this species.

Discussion Our results support the hypothesis that alligators and crocodiles differ in structural and functional aspects


Fig. 6 Activities of the membrane-bound enzymes aminopeptidase (upper panels) and maltase (lower panels), normalized to intestinal mass (left panels) and to nominal intestinal surface area (right panels). Aminopeptidase activity was higher for crocodiles than for alligators, regardless of normalization; however, maltase showed the opposite pattern, with alligators having significantly higher activity. For both enzymes, there were significant differences among segments along the intestine. Asterisks indicate segments where there were significant differences (P50.05) between species by Tukey’s HSD post-hoc test.

12

C. R. Tracy et al.

related to digestive function. Alligators and crocodiles both have similarly high rates of paracellular (non-mediated) absorption of nutrients, which is surprising, given their relatively sedentary lifestyle. Crocodile guts have a steeper allometry with body size (i.e., longer guts at body size greater than hatchlings), but a smaller absorptive surface-magnification and lower enzymatic function for the digestion of carbohydrates (i.e., lower ratio of maltase to aminopeptidase, an index of ability to digest carbohydrates relative to protein). In contrast, alligators have more robust guts and a shallower allometric slope (i.e., shorter guts at most sizes), but their absorptive-surface magnification is greater and they have higher

enzymatic function for the digestion of carbohydrates (i.e., higher ratio of maltase to aminopeptidase). These differences may explain why alligators and crocodiles can exploit different diets, and why they differ in digestive performance in captivity. Intestinal morphometrics and histology Although the juvenile animals that we used for physiological comparisons between species had similar gross morphology (length, mass) of the intestine, the allometry of the digestive tract over a larger range of body sizes was significantly different between the two species (Fig. 1), with alligators having shorter, but more massive, intestines. These allometries predict


Fig. 7 Mean concentrations (decays per minute, dpm) of 3-O methyl D-glucose (A, C) and L-glucose (B, D) in the plasma, normalized to the dose given, as a function of the time since each probe was consumed (oral) or injected. Insets display the same results on a semilog plot. The line in the insets through the injection data is the bi-exponential fit of the model Ct ¼ Aeat þ Bebt (see Table 1 for specific values of parameters).


13

larger intestinal surface area for adult crocodiles than for adult alligators; however, we found that total intestinal surface areas (based on multiplying nominal surface area by the villus-magnification factor) were not significantly different in juvenile animals. Crocodiles seem to have an overall digestive efficiency similar to that of alligators, with overall energy digestibility (the energy content of the ingested food minus that of the excreta as a percentage of the energy content of ingested food) reported to be 86.4% for juvenile crocodiles (Garnett 1988) and 84.3% for juvenile alligators (Staton et al. 1990a), an observation which is consistent with the similarity in total intestinal surface area. Very few data are available for comparison of intestinal morphometry of crocodilians. In an earlier study of Caiman latirostris (Alligatoridae), Starck et al. (2007) used the same methods for measuring absorptive-surface magnification and found a surface-

magnification of 15 in fasting caimans and an average of 25 in those that were digesting. This is higher than our results for A. mississippiensis, and even more extreme when compared with C. porosus. However, it is unclear whether this difference was because our animals were young juveniles; Starck et al. (2007) measured larger, older animals. Surface-area magnification may change with ontogeny, as seen in some birds (Starck 1996; Mitjans et al. 1997). The lower surfacearea magnification may suggest a selective pressure for increased paracellular absorption; with lower surface area and a relatively more permeable epithelium (see below), there could be lesser need for energetically expensive carrier-mediated uptake of nutrients. Starck et al. (2007) observed epithelial disintegration similar to that described here in the mucosal epithelium of C. latirostris, but the degree of disintegration was not as strong as we found in A.


Fig. 8 Analysis of carbohydrate-probe uptake (A, C) and proportional contribution of non-mediated absorption to total uptake (B, D). Apparent rates of absorption of L-glucose and 3-O methyl D-glucose were statistically indistinguishable from each other in both species. Because 3-O methyl D-glucose is absorbed both by carrier-mediated and non-mediated pathways, the statistical similarity suggests extensive non-mediated absorption. Insets in panels A and C show the time-course of absorption. The ratio of absorption of L-glucose to that of 3-O methyl D-glucose provides an indication of the proportional contribution of non-mediated absorption to total uptake, which exceeded 74% at all time-points in crocodiles and 57% at all time-points in alligators.

14

Activitiy of digestive enzymes Our two species differed significantly in their ability to digest carbohydrates relative to protein, as indicated by different ratios of maltase activity to aminopeptidase activity. This ratio was on average

five-fold higher in alligators relative to crocodiles. This suggests that there might be dietary differences between the species that would provide a selective pressure for different digestive capacities. Activities of specific enzymes often show a correlation with the presence or absence of their substrates in the diet (Martıńez del Rio and Stevens 1988, 1989; Martıńez del Rio et al. 1988; Martıńez del Rio 1990). Activities of disaccharidases, for example, appear to be broadly related to diet (Vonk and Western 1985). However, crocodilians generally have similar diets that include a wide taxonomic range of prey (Webb et al. 1982; Magnusson et al. 1987). In general terms, small crocodilians eat primarily insects and crustaceans, crocodilians of intermediate size add fish to their diets, and large crocodilians also include birds and mammals, and this holds both for A. missippienesis (Delany and Abercrombie 1986; Delany 1990; Rice et al. 2007), and C. porosus (Taylor 1979; Webb et al. 1982, 1991; Sah and Stuebing 1996). Although these diets are primarily carnivorous, they are not necessarily carbohydrate free; some molluscs consumed by alligators can be up to 50% glycogen by dry mass (Goddard and Martin 1966), and some crustaceans use trehalose as a storage sugar in their hemolymph (Chang and O’Connor 1983). Indeed, many crocodilians, including both species studied here, are known to consume fruits and other plant matter (Platt et al. 2013), albeit at a relatively small percentage of total intake of food. Thus, there is no obvious evolutionary explanation for their different physiological capacities, so the reasons behind the differences remain unanswered. Regardless of the explanation for differences in enzymatic activity among the species, the differences themselves may explain why alligators can effectively digest the complex, commercial diets used in the farming industry that include both animal and plant protein and carbohydrates. Alligator-farming has successfully used diets of dried pellets containing up to one-third carbohydrate material, including corn and rice (Staton et al. 1990b). However, the significantly lower activities of disaccharidases that we measured in crocodiles, relative to alligators, suggest that crocodiles have low capacity to digest carbohydrates, which could explain why farmed crocodiles seem to require a more strictly carnivorous diet. Martinez del Rio and Stevens (1988) also found modest disaccharidase activities in American alligators. Thus, enzymatic activities seemingly reflect the animals’ different abilities to utilize different diets in captivity. We assumed that any changes in activity of digestive enzymes due to time since last


mississippiensis or C. porosus. In particular, their study showed that lesions occurred at the tips of the villi in caimans that were digesting and were associated with swollen cell-bodies, lysis of the cytoplasm, structural dissociation of the cells, and breakdown of cellular membranes. Because methods of tissue-sampling, histology, and morphometry were exactly the same as used here, we are confident that the differences between the species are real and not based on different methods. However, the lower surface-area magnification we observed in this study compared with Starck et al. (2007) is consistent with the greater level of epithelial disintegration we observed. We note that the observed leaks in epithelial contingency do not relate to any of the known pathologies (e.g., hemorrhagic diarrhea). Rather, the intestinal epithelium appears to be healthy and fully functional, and animals were feeding well, were in good condition, and grew in mass and length throughout the study. We suggest that the epithelial disintegration as observed in crocodilians during digestion is a physiological processes associated with digestion and regeneration of the mucosal epithelium, and is consistent with high levels of paracellular uptake of nutrients. Similar observations of epithelial disintegration have been made in birds (Karasov et al. 2004), but these have never been studied in detail. Starck et al. (2007) measured the effects of fasting on mucosal thickness in caimans’ intestines. They found that mucosal thickness in the distal small intestine of fasted animals was not significantly lower than that of animals in the fed state until 6 days post-feeding, and in the duodenum (where most absorption of nutrients occurs) not significantly lower until 9 days post-feeding. Because animals in the present study were fed 3.5 days before euthanasia and because the passage-time of digesta is approximately 24–36 h (C. Gienger, unpublished data), we considered them to be in the fed state. Intestinal histology is consistent with the fed state. However, we acknowledge that post-feeding changes in gastrointestinal morphology or function may become important before they become statistically significant. We assumed that because regimes of feeding and experimental procedures were identical for both species any changes in morphology or in enzymatic activity would be similar across species.

C. R. Tracy et al.


feeding were similar between the species because they were fed ad libitum on the same diet and schedule. Absorption of nutrients

Pappenheimer (1993) suggested that high paracellular (i.e., non-mediated) absorption may be selectively advantageous because it provides a pathway for nutrient absorption that is energetically inexpensive. Thus, animals with a reduced absorptive capacity or high relative energetic demand would benefit from having high paracellular absorption. This hypothesis is supported by the tendency for small, flying vertebrates (birds, bats) that tend to have shorter guts and rapid passage of digesta to also have relatively high paracellular absorption (Caviedes-Vidal et al. 2007; Price et al. 2015). The extent of paracellular absorption observed here for alligators and crocodiles is similar to that of passerine birds, hummingbirds, and bats (Fig. 9) in which high paracellular absorption is assumed to be an adaptation for flight (Caviedes-Vidal et al. 2007, 2008; Karasov et al. 2012; Price et al. 2015), and was substantially different from the only other reptile measured to date, the Egyptian spiny-tailed lizard (McWhorter et al. 2013). High paracellular absorption may allow small, volant species to have shorter and lighter digestive tracts and thereby reduce wing-loading during flight (Lavin and Karasov 2008). Clearly, adaptations for flight are not relevant to extant crocodilians, and there seems to be no obvious teleological explanation for this phenomenon. However, crocodilians and birds do share a common ancestor (Benton and Clark 1988), so a shared evolutionary history may partially explain similar mechanisms of nutrient absorption. From that point of view the high paracellular uptake of birds might appear rather as an exaptation (sensu Gould and Vrba 1982) of an ancestral archosaurian trait than as an adaptation to flight. Larger birds (approaching 1 kg in body mass) have relatively lower paracellular nutrient uptake that is comparable to non-flying mammals (Fig. 9). However, greater paracellular absorption may also result in increased absorption and exposure to water-soluble toxins (Diamond 1991), providing a potential selective pressure for increased reliance on specific transport-proteins. That selective pressure to reduce exposure to toxins could have resulted in a secondary loss of an ancestral state of high paracellular absorption. Thus, the question of whether high paracellular absorption is an ancestral condition that has been lost secondarily in larger birds, or has arisen independently in multiple vertebrate lineages, remains to be answered. Another related question that remains largely unanswered is whether herbivorous sauropsids rely to any significant extent on paracellular uptake of nutrients. It is likely that as a group they would be exposed to higher levels of water-soluble toxins in their diets than are the


Alligators and crocodiles in this study exhibited high fractional absorption (f) values for carbohydrate probes of both carrier-mediated (3-O methyl D-glucose) and non-mediated (L-glucose) uptake. While these results imply significant reliance on paracellular uptake, data on the extent of absorption (f) alone do not provide the most robust evidence (Schwartz et al. 1995). If absorption of L-glucose occurred much more slowly relative to 3-O methyl D-glucose, but over the entire length of the intestine, then f values could be similar. However, the analyses of absorption rate that we conducted in the present study confirmed that these probes were absorbed at similar rates, and that paracellular uptake can account for a significant proportion of the total uptake of carbohydrate (Fig. 8). Indeed, our data on absorption rates for the two probes in both species is consistent with theoretical predictions of relative apparent absorption rates when passive absorption is dominant (cf. Fig. 8A and C with Fig. 1 in Chang and Karasov 2004). That is, the curves of apparent absorption rate for probes of carrier-mediated and non-mediated uptake should overlap, with absorption rate initially high and declining rapidly. We did not find any significant differences in the values of fractional absorption either between probes within a species or for the same probe between species, and the estimated proportional contributions of paracellular uptake to that of total nutrients during the initial absorptive phase was high and similar between species (72% in alligators and 85% in crocodiles). Thus, it appears that this apparent similarity in the mechanisms of absorbing nutrients is consistent with the similarly high digestive efficiency in juvenile animals of these species (Garnett 1986, 1988; Staton 1988; Webb et al. 1991). We hasten to point out that uptake of L-glucose as measured in the present study does not represent uptake of carbohydrate specifically, but rather uptake of small, hydrosoluble, nonelectrolyte compounds. Other hydrosoluble compounds, such as amino acids, would also be absorbed by this pathway (Pappenheimer et al. 1994), although there is some effect of charge (if present) on uptake (Chediack et al. 2006). Taken together, data in the present study suggest extensive reliance on non-mediated uptake of nutrients by juvenile crocodilians of both species.

15

16

C. R. Tracy et al.

carnivorous crocodilians measured in this study, and so we would predict relatively less reliance on paracellular uptake. Indeed, McWhorter et al. (2013) found that reliance on paracellular uptake was low (accounting for approximately 24% of total glucose uptake) in the herbivorous Egyptian spiny-tailed lizard, Uromastyx aegyptia. To our knowledge, this is the only herbivorous lizard in which such measurements have been done to date. The individual crocodilians used in this portion of the present study were juveniles and growth in crocodilians is fast during their first 2–3 years (Webb et al. 1991; Richardson et al. 2002). Growing juvenile vertebrates would also have relatively high energetic

demands (compared with adults), and may also have reduced digestive capacity, if the functioning of the digestive system is not fully developed at hatching or birth. The altricial nestlings of house sparrows (Passer domesticus) show an increase in overall digestive function from 3 to 12 days post-hatching (Brze k et al. 2010). Fractional absorption of L-glucose (a paracellular probe) increased from 0.7 to 1.0 over that time, when the nestlings were growing rapidly (Brze k et al. 2010). In contrast, fractional absorption of L-glucose in adult house sparrows is approximately 0.6 (Chang and Karasov 2004), suggesting that there may be increased paracellular absorption while energetic demands for growth are highest.


Fig. 9 Fractional absorption of paracellular probes as a function of body mass for birds, fruit bats, non-flying mammals, and crocodilians. Each point represents the mean value for a species. Because diffusion in water declines with MM1/2 (Smulders and Wright 1971), ƒ values of the probes were adjusted for probe-mass, relative to the MM of L-glucose, using a multiplier calculated from the formula: multiplier ¼ 1((Lglu0.5 Prb0.5)/Lglu0.5), where Lglu is the MM of L-glucose and Prb is the MM of the paracellular probe used. Lines connect the points within a group to the median values (for both axes) for that group (e.g., birds, non-flying mammals). The two species of crocodilians measured in this study have paracellular fractional absorptions much higher than would be expected, based on their body size and their having terrestrial habits (non-flying). Plotted points came from the review by McWhorter et al. (2013).


Therefore, it is not clear whether high paracellular absorption is a general characteristic of crocodilians, or whether it is only present in small, rapidly growing individuals absorbing nutrients at a high rate. Hence, in vivo studies of nutrient uptake in crocodilians at different developmental stages and body sizes are needed. Conclusions

Acknowledgments Ian Hunt, Jacob Bar-Lev, Dave Ottway, Rick Perry, Akira Matsuda, and the rest of the staff at Crocodylus Park provided help with the wrangling and care of crocodiles. The Australian Reptile Park provided alligators to Crocodylus Park. Andreas Bollhoefer provided support for scintillation counting at the Environmental Research Institute of the Supervising Scientist (NT). Bill Karasov helped with the initial concepts and experimental setup. The comments of Hal Heatwole and two anonymous re-viewers assisted in improving the manuscript.

Funding This work was supported by the Charles Darwin University and grants from the Australian Government Rural Industries Research and Development Corporation (PRJ-000660) and the Australian Research Council (LP0882478).

References Benton MJ, Clark JM. 1988. Archosaur phylogeny and the relationships of the Crocodylia. In: Benton MJ, editor. The phylogeny and classification of the tetrapods. Vol 1.

Amphibians, reptiles and birds. Oxford: Clarendon Press. p. 295–338. Bo¨ck P. 1989. Romeis’ mikroskopische technik. Mu¨nchen: Urban and Schwarzenberg. Brze k P, Caviedes-Vidal E, Hoefer K, Karasov WH. 2010. Effect of age and diet on total and paracellular glucose absorption in nestling house sparrows. Physiol Biochem Zool 83:501–11. Caviedes-Vidal E, Karasov WH, Chediack JG, Fasulo V, CruzNeto AP, Otani L. 2008. Paracellular absorption: A bat breaks the mammal paradigm. PLoS One 3:e1425. Caviedes-Vidal E, McWhorter TJ, Lavin SR, Chediack JG, Tracy CR, Karasov WH. 2007. The digestive adaptation of flying vertebrates: High intestinal paracellular absorption compensates for smaller guts. Proc Natl Acad Sci USA 104:19132–7. Chang ES, O’Connor JD. 1983. Metabolism and transport of carbohydrates and lipids. In: Bliss DE, editor. The biology of Crustacea. Vol. 5. London: Academic Press. p. 263–87. Chang MH, Karasov WH. 2004. How the house sparrow Passer domesticus absorbs glucose. J Exp Biol 207:3109–21. Chediack JG, Caviedes-Vidal E, Karasov WH. 2006. Electroaffinity in paracellular absorption of hydrophilic D-dipeptides by sparrow intestine. J Comp Physiol B 176:303–9. Coulson RA, Hernandez T. 1983. Alligator metabolism: Studies on chemical reactions in vivo. Oxford: Pergamon Press. Dahlqvist A. 1984. Assay of intestinal disaccharidases. Scand J Clin Lab Invest 44:69–72. Davenport J, Andrews TJ, Hudson G. 1992. Assimilation of energy, protein and fatty acids by the spectacled caiman Caiman crocodilus crocodilus L. Herpetol J 2:72–6. Davenport J, Grove DJ, Cannon J, Ellis TR, Stables R. 1990. Food capture, appetite, digestion rate and efficiency in hatchling and juvenile Crocodylus porosus. J Zool 220:569–92. Delany MF. 1990. Late summer diet of juvenile American alligators. J Herpetol 24:418–21. Delany MF, Abercrombie CL. 1986. American alligator food habits in northcentral Florida. J Wildl Manag 50:348–53. Diamond J. 1991. Evolutionary design of intestinal nutrient absorption: Enough but not too much. News Physiol Sci 6:92–96. Elsey R, Tracy CR, Lance V, Manolis C. 2008. Collecting blood from crocodilians. Crocodile Specialist Group Newsletter 27:21–2. Fine KD, Santa Ana CA, Porter JL, Fordtran JS. 1993. Effect of D-glucose on intestinal permeability and its passive absorption in human small intestine in vivo. Gastroenterology 105:1117–25. Garnett ST. 1986. Metabolism and survival of fasting estuarine crocodiles. J Zool 208:493–502. Garnett ST. 1988. Digestion, assimilation and metabolism of captive estuarine crocodiles, Crocodylus porosus. Comp Biochem Physiol A Physiol 90:23–9. Garnett ST, Murray RM. 1986. Parameters affecting the growth of the estuarine crocodile, Crocodylus porosus, in captivity. Aust J Zool 34:211–23. Goddard CK, Martin AW. 1966. Carbohydrate metabolism. In: Wilbur KM, Yonge CM, editors. Physiology of Mollusca. Vol. II. New York: Academic Press. p. 275–308.


American alligators and saltwater crocodiles exhibited distinct differences in intestinal morphometrics over a large range of body sizes, with the alligators having shorter, more massive guts and crocodiles having longer, less massive guts. In juvenile animals similar in body size, significant differences in surface-area magnification between the species compensated for different dimensions of the gut, so total surface area was not significantly different. Crocodiles had relatively lower ability to digest carbohydrates relative to protein when compared with alligators. These patterns in enzymatic activity, while apparently not correlated with diet in the wild, may explain differences in performance while on various diets in captivity. No clear differences in mechanisms of uptake of nutrients are apparent in juvenile animals similar in body size; however, given diverging allometries of gut size this may translate into important differences in function in adult animals.

17

18

McWhorter TJ, Pinshow B, Karasov WH, Tracy CR. 2013. Paracellular absorption is relatively low in the herbivorous Egyptian spiny-tailed lizard, Uromasyx aegyptia. PLoS One 8:e61869. Mitjans M, Barniol G, Farerrer R. 1997. Mucosal surface area in chicken small intestine during development. Cell Tissue Res 190:71–8. Motulsky HJ, Ransnas LA. 1987. Fitting curves to data using nonlinear regression: A practical and nonmathematical review. FASEB J 1:365–74. Pappenheimer JR. 1993. On the coupling of membrane digestion with intestinal absorption of sugars and amino acids. Am J Physiol 265(Part 1):G409–17. Pappenheimer JR, Dahl CE, Karnovsky ML, Maggio JE. 1994. Intestinal absorption and excretion of octapeptides composed of D amino acids. Proc Natl Acad Sci USA 91:1942–5. Peucker SKJ, Jack RH. 2006. Crocodile farmng research: Onfarm research of pelleted feed for crocodiles. Australian Government Rurual Industries Research and Development Corporation. RIRDC Publication No. 06/016. RIRDC, Canberra. Platt SG, Elsey RM, Liu H, Rainwater TR, Nifong JC, Rosenblatt AE, Heithaus MR, Mazzotti FJ. 2013. Frugivory and seed dispersal by crocodilians: An overlooked form of saurochory? J Zool 291:87–99. Price ER, Brun A, Caviedes-Vidal E, Karasov WH. 2015. Digestive adaptations of aerial lifestyles. Physiology 30:69–78. Read M. 2000. Aspects of protein utilisation and metabolism by post-hatchling estuarine crocodiles (Crocodylus porosus Schneider) [PhD thesis]. [Brisbane, Qld, Australia]: University of Queensland. Rice AN, Ross JP, Woodward AR, Carbonneau DA, Percival HF. 2007. Alligator diet in relation to alligator mortality on Lake Griffin, FL. Southeast Nat 6:97–110. Richardson KC, Webb GJW, Manolis SC. 2002. Crocodiles: Inside out. A guide to the crocodilians and their functional morphology. Chipping Norton, NSW, Australia: Surrey Beatty & Sons. p. 172. Roncari G, Zuber H. 1969. Thermophilic aminopeptidases from Bacillus stearothermophilus. I. Isolation, specificity, and general properties of the thermostable aminopeptidase I. Int J Protein Res 1:45. Sah SAM, Stuebing RB. 1996. Diet, growth and movements of juvenile crocodiles Crocodylus porosus Schneider in the Klias River, Sabah, Malaysia. J Trop Ecol 12:651–62. Schwartz RM, Furne JF, Levitt MD. 1995. Paracellular intestinal transport of 6-carbon sugars is negligible in the rat. Gastroenterology 109:1206–13. Smulders AP, Wright EM. 1971. The magnitude of nonelectrolyte selectivity in the gallbladder epithelium. J Membr Biol 5:297–318. Sokal RR, Rohlf FJ. 1995. Biometry. New York: W. H. Freeman. Starck JM. 1996. Intestinal growth in altricial European starling (Sturnus vulgaris) and precocial Japanese quail (Coturnix coturnix japonica). Cells Tissues Organs 156:289–306. Starck JM, Cruz-Neto AP, Abe AS. 2007. Physiological and morphological responses to feeding in broad-nosed caiman (Caiman latirostris). J Exp Biol 210:2033–45.


Gould SJ, Vrba ES. 1982. Exaptation—A missing term in the science of form. Paleobiology 8:4–15. Iglesias S, Tracy CR, Bedford GS, McWhorter TJ, Christian KA. 2009. Seasonal effects on intestinal enzyme activity in the Australian agamid lizard, Lophognathus temporalis. Comp Biochem Physiol B Biochem Mol Biol 153:89–94. Ikeda TS, Hwang E-S, Coady MJ, Hirayama BA, Hediger MA, Wright EM. 1989. Characterization of a Naþ/glucose cotransporter cloned from rabbit small intestine. J Membr Biol 110:97–5. Karasov WH. 1988. Nutrient transport across vertebrate intestine. In: Gilles RE, editor. Advances in comparative and environmental physiology. Vol. 2. Berlin: Springer. p. 131–73. Karasov WH, Caviedes-Vidal E, Hartman Bakken B, Izhaki I, Samuni-Blank M, Arad Z. 2012. Capacity for absorption of water-soluble secondary metaboites greater in birds than rodents. PLoS One 7:e32417. Karasov WH, Cork SJ. 1994. Glucose absorption by a nectarivorous bird: The passive pathway is paramount. Am J Physiol 267:G18–G26. Karasov WH, Pinshow B, Starck JM, Afik D. 2004. Anatomical and histological changes in the alimentary tract of migrating blackcaps (Sylvia atricapilla): A comparison among fed, fasted, food-restricted and refed birds. Physiol Biochem Zool 77:149–60. Kercheval DR, Little PL. 1990. Comparative growth rates of young Alligators utilizing rations of plant and/or animal origin. Crocodiles. Proceeding of the 10th Working Meeting of the Crocodile Specialist Group of the Species Survival Commission of the IUCN—The World Conservation Union, 23–27 April 1990. Gland: IUCN. p. 286–312. Lane JS, Whang EE, Rigberg DA, Hines OJ, Kwan D, Zinner MJ, McFadden DW, Diamond J, Ashley SW. 1999. Paracellular glucose transport plays a minor role in the unanesthetized dog. Am J Physiol 276:G789–94. Lavin SR, Karasov WH. 2008. Allometry of paracellular absorption in birds. Physiol Biochem Zool 81:551–60. Loo JCK, Riegelman S. 1968. A new method for calculating intrinsic absorption rates of drugs. J Pharm Sci 57:918–28. Magnusson WE, da Silva EV, Lima AP. 1987. Diets of Amazonian crocodilians. J Herpetol 21:85–95. Martıńez del Rio C. 1990. Dietary, phylogenetic, and ecological correlates of intestinal sucrase and maltase activity in birds. Physiol Zool 63:987–1011. Martıńez del Rio C, Brugger KE, Rios JL, Vergara ME, Witmer MC. 1995. An experimental and comparative study of dietary modulation of intestinal enzymes in European starlings (Sturnus vulgaris). Physiol Zool 68:490–511. Martıńez del Rio C, Stevens BR. 1988. Intestinal brush border membrane-bound disaccharidases of the American alligator, Alligator mississippiensis. Comp Biochem Physiol B 91:751–4. Martıńez del Rio C, Stevens BR. 1989. Physiological constraint on feeding behavior: Intestinal membrane disaccharidases of the starling. Science 243:794–6. Martıńez del Rio C, Stevens BR, Daneke DE, Andreadis PT. 1988. Physiological correlates of preference and aversion for sugars in three species of birds. Physiol Zool 61:222–9.

C. R. Tracy et al.


Webb GJW, Hollis GJ, Manolis SC. 1991. Feeding, growth, and food conversion rates of wild juvenile saltwater crocodiles (Crocodylus porosus). J Herpetol 25:462–73. Webb GJW, Manolis SC, Buckworth R. 1982. Crocodylus johnstoni in the McKinlay River area, N.T. I. Variation in the diet, and a new method of assessing the relative importance of prey. Aust J Zool 30:877–799. Webb GJW, Manolis SC, Whitehead PJ. 1987. Wildlife management: Crocodiles and alligators. Chipping Norton, NSW: Surrey Beatty & Sons PTY Limited. Webb GJW, Reynolds S, Brien ML, Manolis SC, Brien JJ, Christian KA. 2013. Improving Australia’s crocodile industry productivity. Nutritional requirements, feed ingredients and feeding systems for farmed crocodile production. Canberra, NSW, Australia: RIRDC. Welling PG. 1986. Pharmacokinetics: Processes and mathematics. Washington, DC: American Chemical Society. Wright EM, van Os CH, Mircheff AK. 1980. Sugar uptake by intestinal basolateral membrane vesicles. Biochim Biophys Acta 597:112–24.


Staton MA. 1988. Studies on the use of fats and carbohydrates in the diet of American alligators (Alligator mississippiensis) [PhD thesis]. [Athens, Georgia]: University of Georgia. p. 160. Staton MA, Edwards HM Jr, Brisbin IL Jr, Joanen T, McNease L. 1990a. Protein and energy relationships in the diet of the American alligator (Alligator mississippiensis). J Nutr 120:775–85. Staton MA, Edwards HM Jr, Brisbin IL Jr, McNease L, Joanen T. 1990b. Dietary energy sources for the American alligator, Alligator mississippiensis (Daudin). Aquaculture 89:245–61. Taylor JA. 1979. The foods and feeding habits of subadult Crocodylus porosus Schneider in northern Australia. Aust Wildl Res 6:347–59. Uhing MR, Kimura RE. 1995. Active transport of 3-Omethyl-glucose by the small intestine in chronically catheterized rats. J Clin Invest 95:2799–805. Vonk HJ, Western JRH. 1985. Comparative biochemistry of enzymatic digestion. London: Academic Press. p. 255–95.

19

New dimensions in digestive physiology.

Hydrodynamics, Fungal Physiology, and Morphology.

Hydrogen Sulfide in Physiology and Diseases of the Digestive Tract.

Physiology of Intestinal Absorption and Secretion.

Physiology and pathophysiology of intestinal absorption.

Significant water absorption goes paracellular in kidney proximal tubules.

Multidisciplinary team meeting in digestive oncology: when opinions differ.

Digestive tract morphology and digestion in the wombats (Marsupialia: Vombatidae).

Chickens from lines artificially selected for juvenile low and high body weight differ in glucose homeostasis and pancreas physiology.

Role of intestinal basolateral membrane in absorption of nutrients.

Correlations between abnormal hippocampal morphology and prefrontal physiology in schizophrenia.

Sensory Neurons that Detect Stretch and Nutrients in the Digestive System.

Emulsion-based colloidal nanosystems for oral delivery of doxorubicin: improved intestinal paracellular absorption and alleviated cardiotoxicity.

What roles have anti-interferon antibodies in physiology and pathology?

Morphology, physiology, and molecular biology of renin secretion.

Poxvirus infection in two crocodiles.

Digestive Physiology of Octopus maya and O. mimus: Temporality of Digestion and Assimilation Processes.

Faecal particle size: digestive physiology meets herbivore diversity.

Changes in intestinal absorption of nutrients and brush border glycoproteins after total parenteral nutrition in rats.

Effects of benzoic acid (VevoVitall®) on the performance and jejunal digestive physiology in young pigs.

Duodenal absorption and tissue utilization of dietary heme and nonheme iron differ in rats.

Growth morphology of Streptomyces akiyoshiensis in submerged culture: influence of pH, inoculum, and nutrients.

Roles of the calcium sensing receptor in digestive physiology and pathophysiology (review).

Digestive absorption of silicon, supplemented as orthosilicic acid-vanillin complex.