Antonio Brun

Digestive Adaptations of Aerial Lifestyles

Flying vertebrates (birds and bats) are under selective pressure to reduce the size of the gut and the mass of the digesta it carries. Compared with similar-sized nonflying mammals, birds and bats have smaller intestines and shorter retention times. We review evidence that birds and bats have lower spare digestive capacity and partially compensate for smaller intestines with increased paracellular nutrient absorption.

Gut microbial ecology of lizards: insights into diversity in the wild, effects of captivity, variation across gut regions and transmission

Animals maintain complex associations with a diverse microbiota living in their guts. Our understanding of the ecology of these associations is extremely limited in reptiles. Here, we report an in‐depth study into the microbial ecology of gut communities in three syntopic and viviparous lizard species (two omnivores: Liolaemus parvus and Liolaemus ruibali and an herbivore: Phymaturus williamsi). Using 16S rRNA gene sequencing to inventory various bacterial communities, we elucidate four major findings: (i) closely related lizard species harbour distinct gut bacterial microbiota that remain distinguishable in captivity; a considerable portion of gut bacterial diversity (39.1%) in nature overlap with that found on plant material, (ii) captivity changes bacterial community composition, although host‐specific communities are retained, (iii) faecal samples are largely representative of the hindgut bacterial community and thus represent acceptable sources for nondestructive sampling, and (iv) lizards born in captivity and separated from their mothers within 24 h shared 34.3% of their gut bacterial diversity with their mothers, suggestive of maternal or environmental transmission. Each of these findings represents the first time such a topic has been investigated in lizard hosts. Taken together, our findings provide a foundation for comparative analyses of the faecal and gastrointestinal microbiota of reptile hosts.

Physiological and microbial adjustments to diet quality permit facultative herbivory in an omnivorous lizard.

ABSTRACT While herbivory is a common feeding strategy in a number of vertebrate classes, less than 4% of squamate reptiles feed primarily on plant material. It has been hypothesized that physiological or microbial limitations may constrain the evolution of herbivory in lizards. Herbivorous lizards exhibit adaptations in digestive morphology and function that allow them to better assimilate plant material. However, it is unknown whether these traits are fixed or perhaps phenotypically flexible as a result of diet. Here, we maintained a naturally omnivorous lizard, Liolaemus ruibali, on a mixed diet of 50% insects and 50% plant material, or a plant-rich diet of 90% plant material. We compared parameters of digestive performance, gut morphology and function, and gut microbial community structure between the two groups. We found that lizards fed the plant-rich diet maintained nitrogen balance and exhibited low minimum nitrogen requirements. Additionally, lizards fed the plant-rich diet exhibited significantly longer small intestines and larger hindguts, demonstrating that gut morphology is phenotypically flexible. Lizards fed the plant-rich diet harbored small intestinal communities that were more diverse and enriched in Melainabacteria and Oscillospira compared with mixed diet-fed lizards. Additionally, the relative abundance of sulfate-reducing bacteria in the small intestine significantly correlated with whole-animal fiber digestibility. Thus, we suggest that physiological and microbial limitations do not sensu stricto constrain the evolution of herbivory in lizards. Rather, ecological context and fitness consequences may be more important in driving the evolution of this feeding strategy. Summary:The digestive system and gut microbiota of lizards are highly responsive to diet, and thus intrinsic physiological limitations may not limit the evolution of herbivory in lizards.

Intestinal perfusion indicates high reliance on paracellular nutrient absorption in an insectivorous bat Tadarida brasiliensis

Flying vertebrates have been hypothesized to have a high capacity for paracellular absorption of nutrients. This could be due to high permeability of the intestines to nutrient-sized molecules (i.e., in the size range of amino acids and glucose, MW 75-180 Da). We performed intestinal luminal perfusions of an insectivorous bat, Tadarida brasiliensis. Using radio-labeled molecules, we measured the uptake of two nutrients absorbed by paracellular and transporter-mediated mechanisms (L-proline, MW 115 Da, and D-glucose, MW 180 Da) and two carbohydrates that have no mediated transport (L-arabinose, MW 150 Da, and lactulose, MW 342 Da). Absorption of lactulose (0.61±0.06 nmol min(-1) cm(-1)) was significantly lower than that of the smaller arabinose (1.09±0.04 nmol min(-1) cm(-1)). Glucose absorption was significantly lower than that of proline at both nutrient concentrations (10mM and 75 mM). Using the absorption of arabinose to estimate the portion of proline absorption that is paracellular, we calculated that 25.1±3.0% to 66.2±7.8% of proline absorption is not transporter-mediated (varying proline from 1 mM to 75 mM). These results confirm our predictions that 1) paracellular absorption is molecule size selective, 2) absorption of proline would be greater than glucose absorption in an insectivore, and 3) paracellular absorption represents a large fraction of total nutrient absorption in bats.

High paracellular nutrient absorption in intact bats is associated with high paracellular permeability in perfused intestinal segments

Water-soluble nutrients are absorbed by the small intestine via transcellular and paracellular mechanisms. Based on a few previous studies, the capacity for paracellular nutrient absorption seems greater in flying mammals than in nonflying mammals, but there has been little investigation of the mechanisms driving this difference. Therefore, we studied three species each of bats (Artibeus lituratus, Sturnira lilium and Carollia perspicillata) and nonflying mammals (Akodon montensis, Mus musculus and Rattus norvegicus). Using standard pharmacokinetic techniques in intact animals, we confirmed the greater paracellular nutrient absorption in the fliers, comparing one species in each group. Then we conducted in situ intestinal perfusions on individuals of all species. In both approaches, we measured the absorption of 3OMD-glucose, a nonmetabolizable glucose analog absorbed both paracellularly and transcellularly, as well as l-arabinose, which has no mediated transport. Fractional absorption of l-arabinose was three times higher in the bat (S. lilium: 1.2±0.24) than in the rodent (A. montensis: 0.35±0.04), whereas fractional absorption of 3OMD-glucose was complete in both species (1.46±0.4 and 0.97±0.12, respectively). In agreement, bats exhibited two to 12 times higher l-arabinose clearance per square centimeter nominal surface area than rodents in intestinal perfusions. Using l-arabinose, we estimated that the contribution of the paracellular pathway to total glucose absorption was higher in all three bats (109–137%) than in the rodents (13–39%). These findings contribute to an emerging picture that reliance on the paracellular pathway for nutrient absorption is much greater in bats relative to nonflying mammals and that this difference is driven by differences in intestinal permeability to nutrient-sized molecules.

A Comparison of mucosal surface area and villous histology in small intestines of the Brazilian free-tailed bat (Tadarida brasiliensis) and the mouse (Mus musculus)

Studies on birds have led to the hypothesis that increased intestinal absorption between enterocytes (paracellular) evolved as a compensation for smaller intestinal size in fliers, which was perhaps selected to minimize the mass of digesta carried. This hypothesis predicts that bats will also exhibit relatively reduced intestinal size and high paracellular absorption, compared with nonflying mammals. Published studies on three bat species indicate relatively high paracellular absorption. One mechanism for increasing paracellular absorption per cm2 small intestine (SI) is increased number of tight junctions (TJs) across which paracellular absorption occurs. To our knowledge, we provide the first comparative analysis of enterocyte size and number in flying and nonflying mammals. Intestines of insectivorous bats Tadarida brasiliensis were compared with Mus musculus using hematoxylin and eosin staining method. Bats had shorter and narrower SIs than mice, and after correction for body size difference by normalizing to mass3/4, the bats had 40% less nominal surface area than the mouse, as predicted. Villous enhancement of surface area was 90% greater in the bat than in the mouse, mainly because of longer villi and a greater density of villi in bat intestines. Bat and mouse were similar in enterocyte diameter. Bats exceeded mice by 54.4% in villous area per cm length SI and by 95% in number of enterocytes per cm2 of the nominal surface area of the SI. Therefore, an increased density of TJs per cm2 SI may be a mechanistic explanation that helps to understand the high paracellular absorption observed in bats compared to nonflying mammals. J. Morphol. 276:102–108, 2015.

Intestinal Water Absorption Varies with Expected Dietary Water Load among Bats but Does Not Drive Paracellular Nutrient Absorption

Rapid absorption and elimination of dietary water should be particularly important to flying species and were predicted to vary with the water content of the natural diet. Additionally, high water absorption capacity was predicted to be associated with high paracellular nutrient absorption due to solvent drag. We compared the water absorption rates of sanguivorous, nectarivorous, frugivorous, and insectivorous bats in intestinal luminal perfusions. High water absorption rates were associated with high expected dietary water load but were not highly correlated with previously measured rates of (paracellular) arabinose clearance. In conjunction with these tests, we measured water absorption and the paracellular absorption of nutrients in the intestine and stomach of vampire bats using luminal perfusions to test the hypothesis that the unique elongated vampire stomach is a critical site of water absorption. Vampire bats’ gastric water absorption was high compared to mice but not compared to their intestines. We therefore conclude that (1) dietary water content has influenced the evolution of intestinal water absorption capacity in bats, (2) solvent drag is not the only driver of paracellular nutrient absorption, and (3) the vampire stomach is a capable but not critical location for water absorption.

Small intestinal epithelial permeability to water-soluble nutrients higher in passerine birds than in rodents

In the small intestine transcellular and paracellular pathways are implicated in water-soluble nutrient absorption. In small birds the paracellular pathway is quantitatively important while transcellular pathway is much more important in terrestrial mammals. However, there is not a clear understanding of the mechanistic underpinnings of the differences among taxa. This study was aimed to test the hypothesis that paracellular permeability in perfused intestinal segments is higher in passerine birds than rodents. We performed in situ intestinal perfusions on individuals of three species of passerine birds (Passer domesticus, Taeniopygia guttata and Furnarius rufus) and two species of rodents (Mus musculus and Meriones ungiculatus). Using radio-labelled molecules, we measured the uptake of two nutrients absorbed by paracellular and transcellular pathways (L-proline and 3-O-methyl-D-glucose) and one carbohydrate that has no mediated transport (L-arabinose). Birds exhibited ~2 to ~3 times higher L-arabinose clearance per cm2 epithelium than rodents. Moreover, paracellular absorption accounted for proportionally more of 3-O-methyl-D-glucose and L-proline absorption in birds than in rodents. These differences could be explained by differences in intestinal permeability and not by other factors such as increased retention time or higher intestinal nominal surface area. Furthermore, analysis of our results and all other existing data on birds, bats and rodents shows that insectivorous species (one bird, two bats and a rodent) had only 30% of the clearance of L-arabinose of non-insectivorous species. This result may be explained by weaker natural selection for high paracellular permeability in animal- than in plant-consumers. Animal-consumers absorb less sugar and more amino acids, whose smaller molecular size allow them to traverse the paracellular pathway more extensively and faster than glucose.

Gut microbes limit growth in house sparrow nestlings (Passer domesticus) but not through limitations in digestive capacity

Abstract Recent research often lauds the services and beneficial effects of host‐associated microbes on animals. However, hosting these microbes may come at a cost. For example, germ‐free and antibiotic‐treated birds generally grow faster than their conventional counterparts. In the wild, juvenile body size is correlated with survival, so hosting a microbiota may incur a fitness cost. Avian altricial nestlings represent an interesting study system in which to investigate these interactions, given that they exhibit the fastest growth rates among vertebrates, and growth is limited by their digestive capacity. We investigated whether reduction and restructuring of the microbiota by antibiotic treatment would: (i) increase growth and food conversion efficiency in nestling house sparrows (Passer domesticus); (ii) alter aspects of gut anatomy or function (particularly activities of digestive carbohydrases and their regulation in response to dietary change); and (iii) whether there were correlations between relative abundances of microbial taxa, digestive function and nestling growth. Antibiotic treatment significantly increased growth and food conversion efficiency in nestlings. Antibiotics did not alter aspects of gut anatomy that we considered but depressed intestinal maltase activity. There were no significant correlations between abundances of microbial taxa and aspects of host physiology. Overall, we conclude that microbial‐induced growth limitation in developing birds is not driven by interactions with digestive capacity. Rather, decreased energetic and material costs of immune function or beneficial effects from microbes enriched under antibiotic treatment may underlie these effects. Understanding the costs and tradeoffs of hosting gut microbial communities represents an avenue of future research.

Aminopeptidase activity is related to the amino acids composition of the food in passerine birds

31 Background. Passerine birds exploit different kinds of feeding habits and they have to 32 face seasonal changes in food availability. Therefore, the composition of the principal nutrient 33 in their food differs from the usual. In consequence the digestive function – enzyme 34 hydrolysis and absorption – have to adapt to these nutrients. These changes in digestive 35 physiology could respond to the adaptive modulation hypothesis which postulated that the 36 activities of digestive enzymes should match the levels of their substrates in their diet so 37 energy is not wasted on enzymes that are no need. Thus, we decide to measure intestinal 38 enzymes activities of two species of passerine birds that differ in natural diet. Overall we 39 hypothesized that species with different feeding habits present enzyme activity according to 40 the mainly component of the diet (e.g., carbohydrates, proteins). Our prediction is that the 41 individuals will present enzyme activity proportionally to the primary components of the 42 diets. 43 Methods. We select for study: red ovenbirds (Furnarius rufus), which are strict 44 insectivores and zebra finches (Taeniopygia guttata), which are specialist granivores. We 45 complete the analysis with publish data for house sparrows (Passser domesticus) feed on high 46 starch from the literature. To examine intestinal enzyme activities, we measured the activity 47 of two disaccharidases (sucrase-isomaltase and maltase-glucoamilase) and one dipeptidase 48 (aminopeptidase-N). 49 Results. The average intestinal activity of sucrase shows that the omnivorous P. 50 domesticus presents almost 4 times more activity than the granivorous T. guttata and more 51 than 11 times than the insectivorous F. rufus. This difference is also reflected in the total 52 sucrase hydrolytic capacity where P. domesticus has roughly 10 times more than the other two 53 birds. Surprisingly in F. rufus we found maltase and aminopeptidase activity while sucrase 54 activity was close to zero. In the case of the average activity of maltase for the omnivorous P. 55 domesticus is approximately 40 % more than the granivorous T. guttata and more than 5 times 56 than the insectivorous F. rufus. Although the total maltase hydrolytic capacity of P. 57 domesticus is 5 times more than T. guttata and F. rufus. The average of aminopeptidase-N 58 activity for F. rufus and T. guttata almost doubled the P. domesticus ones. Also F. rufus 59 roughly doubles the other two birds in total aminopeptidase hydrolytic capacity. 60 Discussion. This study has shown that exist a relationship between the levels of amino 61 acids in the diet and the total aminopeptidase capacity, but in the case of carbohydrates this 62 relationship is not evident. 63 64 INTRODUCTION 65 Birds possess the capacity to exploit a broad diversity of resources and they have to 66 face seasonal changes in food availability. The switch to available food modifies the 67 predominant nutrients in their food intake. Consequently the consumption of food with 68 PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3443v2 | CC BY 4.0 Open Access | rec: 5 Dec 2017, publ: 5 Dec 2017 different nutrients requires that the optimization of the digestive function – enzyme hydrolysis 69 and absorption – is adapted to these nutrients (Karasov & Martínez del Río 2007). In several 70 vertebrates was reported a modulation of disaccharidases activity correlated with substrates in 71 their diets (Biviano et al. 1993; Harpaz & Uni 1999; Hernández & Martínez del Río 1992; 72 Sabat et al. 1995). These changes in digestive physiology could respond to the adaptive 73 modulation hypothesis which postulated that activities of digestive enzymes should match the 74 levels of their substrates in their diet so energy is not wasted on enzymes that are no need 75 (Karasov 1992; Karasov & Diamond 1988). In birds, at the interspecific level it has been 76 observed that the hydrolytic capacity of the individuals is related with the level of the 77 substrate in the feeding habits (Kohl et al. 2011; Ramírez-Otarola & Sabat 2011). In 78 consequence, we decide to measure intestinal enzymes activities of two species of passerine 79 birds that differ in natural diet. 80 To examine intestinal enzyme activities, we measured the activity of two 81 disaccharidases and one dipeptidase. Digestion of carbohydrates implicates glucosidase 82 enzymes located on the brush border of the small intestine; between them we can find 83 sucrase-isomaltase (EC 3.2.1.10) and maltase-glucoamilase (EC 3.2.1.3) (Hunziker et al. 84 1986; Palmer 1971). Maltase-glucoamilase hydrolyzes maltose in two molecules of glucose, 85 while sucrase-isomaltase hydrolyzes sucrose in one molecule of fructose and another of 86 glucose. In reference of protein digestion we have chosen to measure aminopeptidase-N (E.C. 87 3.4.11.2) that account for almost all peptidase activity in the brush border membrane (Maroux 88 et al. 1973). This enzyme cleaves oligopeptides to produce dipeptides and amino acids 89 (Sjostrom et al. 1978). 90 We select for study: red ovenbirds (REDO; Furnarius rufus), which are strict 91 insectivores (Fraga 1980b) and zebra finches (ZEBF; Taeniopygia guttata), which are 92 specialist granivores (Zann 1996a). We complete the analysis with publish data for the house 93 PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3443v2 | CC BY 4.0 Open Access | rec: 5 Dec 2017, publ: 5 Dec 2017 sparrows (HOSP; Passser domesticus) feed on high starch from the literature (Caviedes-Vidal 94 et al. 2000). We choose the high starch because represent most likely the natural diet 95 (Anderson 2006). 96 Overall we hypothesized that species with different feeding habits present enzyme 97 activity according to the mainly component of the diet (e.g., carbohydrates, proteins). Thus 98 we predicted that the individuals will present enzyme activity proportionally to the primary 99 components of the diets. 100 MATERIALS AND METHODS 101 Animals 102 Zebra finches (ZEBF) were purchased in San Luis, Argentina and house sparrows 103 (HOSP) and red ovenbirds (REDO) were captured on the campus of Universidad Nacional de 104 San Luis (UNSL), San Luis. HOSP and ZEBF were housed in cages indoors under constant 105 environmental conditions (25 ± 1°C, relative humidity of 50 ± 10%) on a photoperiod of 106 14:10 (L:D) with water and food ad libitum (alpist, millet, vitamins and minerals). REDO 107 were used on the same day of capture in order not to alter their eating habits. All animal 108 procedures adhered to institutional animal use regulations and approved animal use protocols 109 by the Animal Care and Use Committee of the UNSL, protocol number B212/15. Captured 110 animals were approved by the Environmental Office of the state of San Luis, resolution 111 number 75-PBD-2015. 112 Intestinal Enzyme Assays 113 Disaccharidases activities, sucrase-isomaltase and maltase-glucoamilase were assayed 114 using the colorimetric method developed by Dahlqvist (Dahlqvist 1984) and modified by 115 Martínez del Río (Martínez del Río 1990). Briefly, tissues were thawed at 4 °C and 116 homogenized for 30 s using a manual homogenizer (Fisher ScientificTM Laboratory 117 PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.3443v2 | CC BY 4.0 Open Access | rec: 5 Dec 2017, publ: 5 Dec 2017 Homogenizer, Model 125) in mannitol buffer (350 mM for birds) in 1 mM Hepes-KOH, pH 118 7.0. Aliquots of 40 L of diluted intestinal homogenates were incubated with 40 L of 56 119 mM sucrose or 56 mM maltose in 0.1 M maleate/NaOH buffer, pH 6.5, at 40 °C for 20 min. 120 After 20 min of incubation the reaction was stopped by adding 1 mL of enzymatic glucose 121 assay (Glucosa Liquid plus reagent-GT Laboratorios S.R.L.). Sample solutions were allowed 122 to stand for 5 min at room temperature and the absorbance was measured at 505 nm and 123 activity was determined using a glucose standard curve. 124 Aminopeptidase-N activity was assayed using L-alanine-p-nitroanilide as a substrate 125 (Maroux et al. 1973). Aliquots of 10 μL of the intestinal homogenate were added to 1 mL 126 assay solution (2.0 mM L-alanine-p-nitroanilide in 0.2 M phosphate buffer 127 (NaH2PO4/Na2HPO4, pH 7). The reaction was incubated for 20 min at 40 °C and then stopped 128 with 3 mL of chilled 2 M acetic acid. Absorbance was measured at 384 nm, and activity was 129 determined using a p-nitroanilide standard curve. 130 On the basis of absorbance measurements and glucose and p-nitroanilide standard 131 curve we calculated activities of each intestinal section normalized to the wet mass of the 132 section. Activities of intestinal enzymes were expressed in micromoles per minute per gram 133 of wet tissue. 134 We calculated the summed hydrolysis activity of the entire small intestine, an index of 135 the total hydrolysis capacity, by multiplying activity per gram of wet tissue in each region by 136 its respective mass, and summed over the three regions. 137