J Comp Physiol B (1992) 162:552-560

Journal of

Comparative

Systemic, Biochemical, and Environ-

P h y s i o l o g y B "~"~' Physiology 9 Springer-Verlag 1992

Digestive tract morphology and digestion in the wombats (Marsupialia: Vombatidae) P.S. Barboza 1 and I.D. Hume z Department of Biochemistry, Microbiology and Nutrition, University of New England, Armidale, NSW 2351, Australia Accepted February 20, 1992

Summary. Wombats consume grasses and sedges which are often highly fibrous. The morphology of the digestive tract and the sequence of digestion were studied in two species of wombats from contrasting habitats: Vombatus ursinus from mesic habitats and Lasiorhinus latifrons from xeric regions. Studies were performed on wild wombats consuming their natural winter diets, and on captive wombats fed a high-fibre pelleted straw diet. Vombatus had a shorter digestive tract (9.2 vs 12.5 times body length) of greater capacity (wet contents 17.9 vs 13.7% body weight) than Lasiorhinus. The most capacious region of the digestive tract was the proximal colon (62-79 % of contents). The proportional length and surface area of the proximal colon were greater in Vombatus, but those of the distal colon were greater in Lasiorhinus. These digestive morphologies may reflect adaptations for greater capacity and longer retention of digesta in Vombatus, but greater absorption and lower faecal water loss in Lasiorhinus. Apparent digestion along the digestive tract was estimated by reference to lignin. The proximal colon was the principal site of fibre and dry matter digestion, whereas nitrogen was mainly digested in the small intestine. Depot fats in captive wombats were highly unsaturated and reflected those in the diet. Therefore, lipids, proteins and soluble carbohydrates in the plant cell contents were digested and absorbed in the stomach and small intestine. Conversely, dietary fibre was probably retained and digested by microbial fermentation along the proximal colon. Key words: Herbivore - Hindgut - Digestion - Marsupial - Wombat Abbreviations: ADF acid detergent fibre; DM dry matter; NDF

neutral detergent fibre; SD standard deviation Present addresses: 1 Department of Zoological Research, National

Zoological Park, Smithsonian Institution, Washington, DC 20008-2598, USA, and 2 School of Biological Sciences A08, University of Sydney, NSW 2006, Australia Correspondence to: P.S. Barboza

Introduction The complex interactions between the structures and functions of the vertebrate digestive tract were recently discussed by Stevens (1988). Herbivorous mammals exhibit a variety of digestive adaptations to utilize bulky fibrous diets, partly because the capacity to retain and digest herbage is constrained by body size and metabolic requirements (Demment and Van Soest 1985). Herbivorous marsupials include two groups of large grazers: the kangaroos (family Macropodidae) and the wombats (family Vombatidae). Although the digestive physiology of the macropod marsupials has been investigated in several studies recently reviewed by Hume (1989), the digestive adaptations of the wombats have received little attention. Unlike the macropods, which feature a well-developed forestomach (Hume 1989), the vombatid stomach is small and simple, whereas the colon is well developed (Mackenzie 1918; Wells 1973). The wombats are represented by two species from contrasting habitats; Vombatus ursinus (the common wombat) from mesic regions and Lasiorhinus latifrons (the southern hairy-nosed wombat) from xeric habitats. Vombatus inhabits forests and heathlands from the coast to sub-alpine regions of south-eastern Australia, whereas Lasiorhinus inhabits the plains of mainland southern Australia. Adult body masses range from 22 to 50 kg in Vombatus, and from 18 to 35 kg in Lasiorhinus. Wombats excavate complex burrows from which they emerge in the dark or half-light to forage in feeding territories adjacent to the burrow (Lasiorhinus) or further away in a forest clearing or pasture (Vombatus) (Mcllroy 1976; Wells 1978). Perennial grasses are the principal component of the diets of both species. This paper describes experiments comparing digestive tract morphology and digestion in Vombatus and Lasiorhinus. The two species were studied under captive conditions on a high-fibre pelleted diet, and in the wild on their natural winter diets. The morphology and capacity of the digestive tract were described, the function of tract segments was assessed from the composition of digesta,

P.S. Barboza and I.D. Hume: Digestion in wombats

and the apparent digestion of DM, fibre and N was estimated by reference to lignin. Digestion and absorption of dietary fats was assessed by comparing the fatty acid composition of the diet with that of fat stores in captive wombats.

Materials and methods Captive study. A captive population of wombats was maintained at the University of New England for 3 years prior to this study. Vombatus were transferred from other captive populations near Sydney, New South Wales. Lasiorhinus were taken from the wild near Blanchetown, South Australia. All animals were fed a maintenance diet consisting of mixed hay and a pelleted mixture of milled oat (Arena spp.) straw (40.0% air-dry basis), chopped lucerne (Mediea9o sativa) hay (13.9%), corn (Zea mays) (13.9%), oats (15.0%) and a mineral/vitamin supplement (Janos Chemicals, Forbes, NSW) (1.1%) provided ad libitum. Four Vombatus and four Lasiorhinus were moved from their enclosures to metabolism cages (0.62 x 1.28 • 1.00 m high, with woven mesh floors 30 x 30 • 7 mm thick) and allowed to adjust to experimental conditions over 4 weeks. Ambient temperature was maintained at 22 + 2 ~ with the daily average relative humidity at 47 a: 14%. Natural incident lighting was provided during the day and dim blue fluorescent lighting during the night (16:00-06:00 hours). Fresh feed and water were available ad libitum and provided daily at 10:00 hours and 17:00 hours. The pelleted experimental diet was 79% barley (Hordeum vulgate) straw, 15.9% corn, 4.0% casein (hydrochloric) and 1.1% mineral/vitamin supplement on an air-dry basis. This diet contained 89.9% D M which was 89.5% organic matter, 65.3% N D F , 55.9% ADF, 6.7% lignin and 1.3% N. Feed intakes and body masses of wombats were monitored for consistency over 7 days before beginning a collection period of 5 days. Wombats were weighed at the beginning and end of the collection period, during which all feed offered and refused, and faeces voided were weighed, subsampled (20%), and stored at - 20 ~ Subsamples from each animal were subsequently combined and thoroughly mixed together for analysis. The D M content of feed offered was determined daily in a sample of the morning feeding. At the end of the collection period, the wombats were euthanased 1 h after the morning feeding, one on each of 8 successive days in alternate order of species. Wombats were first restrained manually and anaesthetized with 4% halothane (Fluothane, ACI) in O2 at 5 1 - m i n - 1 through a face mask. The anaesthetized animal was then killed humanely by an overdose of sodium pentobarbitone injected into the heart. Body length was measured from nose to tail on the ventral surface. The digestive tract was removed and divided into stomach, small intestine, proximal colon (regions 1M), and distal colon (Fig. 1). The stomach was ligated at the pyloric valve and the small intestine at the ileo-caecal junction. The proximal colon was divided into four sections of approximately equal length between the ileocaecal junction and the point where faecal pellets were distinguishable; this was taken to be the junction between proximal and distal colons. Proximal colon 3 included the mid-colonic sacculations (Fig. 1). The contents of the digestive tract segments were removed and stored immediately at - 20 ~ for analysis. The capacity of each segment was estimated as the difference between the intact and empty mass of the segment. Digesta adhering to the tissue were removed by rinsing the segment in water. Excess water was removed by dabbing the tissue with paper towels before weighing and measuring the length of each segment. The area of a 15-cm section of each intestinal segment was determined by cutting the gut section along its length and laying it flat on a clear plastic board. The outline of the tissue was drawn on the board, which was then placed on a 1 cm x 1 cm grid, and the

553 area determined as the number of square centimetres bounded by the outline. The total area of the segment was calculated from the area per unit length of section and total length of the segment. Since the stomach was not tubular, its area was measured directly by tracing the whole segment without opening, and doubling the bound area to sum the dorsal and ventral surfaces. The fatty acid composition of the maintenance mixture and the experimental diet were compared with that of the depot fats of four Vombatus and two Lasiorhinus. Depot fats were dissected from the lumbar region and stored at - 2 0 ~ for analysis.

Field study. Wild wombats were collected from pastoral properties in typical habitats within their distributions during the winter of 1986. Five Lasiorhinus were collected near Blanchetown, South Australia, during mid-August. Similarly, four Vombatus were collected on the edges of Riamukka State Forest in the northern tablelands of New South Wales during July. These wombats were killed whilst foraging at night by a single shot to the head from a 0.222 calibre rifle. Carcasses were rapidly conveyed to a field laboratory for dissection and sampling of digesta as described for the captive animals. Digesta samples were kept on ice for less than 8 h until they were stored at - 2 0 ~ Although segment lengths and capacities were measured, mucosal surface areas were not determined in these animals. Chemical analyses. Feed, faeces and digesta were dried for analysis at 50 ~ and ground to pass through a l-ram screen (Retsch Crossbeater Mill). D M content was determined by drying to constant weight at 80 ~ in a convection oven. Ash content was determined by combustion in a muffle furnace at 500 ~ for 3 h. Dietary fibre fractions were estimated by the methods of Goering and Van Soest (1970). The N D F fraction was extracted with the addition of sodium sulphite (Van Soest et al. 1991). Estimates of N D F and A D F were performed on separate samples, but acid lignin was determined on the A D F residue. Total N content, which ineludes N in both protein and non-protein compounds, was determined by the Kjeldahl method using a Tecator (Model 2030) autodistillation apparatus. The recovery of N by this technique was estimated routinely (>98%) and used as a correction factor. The fatty acid composition of feeds and depot fats was determined by the method of Lepage and Roy (1986). Calculations of digestibility. Digestibilities of dietary components from the collection period on captive wombats were calculated from their apparent excretion in the faeces. Percentage digestibilities with reference to lignin were calculated by the method of DeUow and Hume (1982) as follows: Cd/LdJ 9 100, where C, = component concentration in digesta or faeces; L s = lignin concentration in digesta or faeces; Cd = dietary concentration of the component; and L d = dietary concentration of lignin. The percentage digestible component remaining at any segment in the digestive tract was calculated as:

1 Ds.g

100,

where D , ~ = maximum digestibility in the tract, and Ds~g= digestibility from mouth to segment inclusive. The dietary concentrations for wild wombats were estimated from stomach or small intestinal contents. Small intestinal contents were used for calculations in two wild Vombatus where stomach digesta were insufficient for complete chemical analysis. Since small intestinal contents of wild Lasiorhinus were also insufficient for complete analysis, only digestibilities along the hindgut of wild animals were calculated relative to the stomach or small intestine.

554

Statistical analysis. Measures of digestive tract length and capacity were compared by analysis of variance with species and population (wild or captive) as factors between groups (Winer 1971). Digestibility data from wild and captive animals were not compared statistically, because of differencesin the diet and the calculation of relative digestibility. However, the species were compared by analysis of variance within each population group. The means for mucosal surface areas between sites were compared by t-test (Winer 1971; Zar 1974). All percentages were arcsine transformed before statistical analysis (Zar 1974). Values given are means:t: SD.

P.S. Barboza and I.D. Hume: Digestion in wombats Vombatus

Oesophagus ;tomach

3mall ntestine PC

Results PC:

lecum

PC,

C3

Digestive morphology and capacity The stomach of both Vombatus and Lasiorhinus featured a thick, rugose mucosa with a cardio-gastric gland (Hingson and Milton 1968; Langer 1988). The small intestine of both species lacked large folds or haustra. Conversely, the proximal colon was extensively haustrated, with several permanent sacculations near the ileo-caecal junction (proximal colon 1) and two large sacculations or "fixed pouches" (Mackenzie 1918) in proximal colon 3 where there is mesenteric attachment with the duodenum. The sacculations and haustra of the proximal colon were less pronounced in Lasiorhinus than in Vombatus (Fig. 1). The caecum was small and variable in both species. Vombatus had a larger digestive tract capacity than Lasiorhinus as a proportion of body mass (Table 1). Conversely, Lasiorhinus had a longer digestive tract, with a ratio of tract to body length 30% greater than Vornbatus (Table 1). The long, narrow tract of Lasiorhinus and the short, wide tract of Vombatus had similar total mucosal surface areas among captive wombats (Vombatus." 4789+741 cm2; Lasiorhinus: 4790:t: 564 cm2). The colon was the longest region of the tract in both species. The proportional length of the proximal colon was greater in Vombatus, but the distal colon was proportionately longer in Lasiorhinus (Table 1). These differences in proportional tract length were reflected in the distributions of mucosal surface area (Fig. 2). Although surface areas of the stomach and small intestine were similar between the species, Vornbatus had a greater area of proximal colon, whereas Lasiorhinus had a greater surface area of distal colon. The apparent mucosal surface area will underestimate the total surface area of the gut segment if microscopic structures such as villi or microvilli are present. Consequently, the total area of the small intestine may be 50-70 times greater than the apparent area of mucosa (Diamond 1987). In contrast, the hindgut has few microscopic infoldings (Stevens 1988), and therefore the apparent mucosal surface of this region probably closely reflected its total surface area. The proximal colon was the most capacious region of the digestive tract in both species, greater in Vombatus than in Lasiorhinus (Table 1). Wild wombats had proportionately heavier small intestinal contents but lighter proximal colon contents than captive wombats. In captive Vombatus and Lasiorhinus stomach contents were relatively dry, whereas the small intestine had

Di C, a Lasiorhinus

lOcm

Oesophagus ch

!he Caed F

PC

b Fig. la, b. The digestivetracts of wombats. Lines indicate points of ligation dividing the tract into stomach, small intestine, proximal colon (PC) regions 14 and distal colon. * Mid-colonic sacculations the wettest contents of any tract segment. All four regions of the proximal colon were similar in D M content, with the highest D M levels in the distal colon (Fig. 3). Although wild wombats had a similar pattern of D M content to that of the captive animals, digesta in the stomach and distal colon were wetter than those of cap-

P.S. Barboza and I.D. Hume: Digestion in wombats

555

Table 1. Capacity (wet contents) and length (empty) of the digestive tract of captive and wild Vombatus (V) and Lasiorhinus (L) Measure

Species

Captive

Wild

Body mass (kg)

V L V L V L V L

27.8 24.6 90 88 17.2 14.5 8.7 12.2

(1.9) (2.8) (1) (5) (2.9) (0.7) (0.5) (1.1)

39.5 21.9 100 85 18.5 12.8 9.6 12.8

(5.8) (4.0) (5) (6) (4.1) (2.5) (1.t) (1.3)

V L V L V L V L

1.6 1.1 40.7 37.1 43.8 24.8 13.8 36.9

(0.2) (0.1) (3.7) (3.3) (2.5) (1.1) (2.2) (3.2)

1.6 1.2 35.6 39.7 42.9 24.4 19.9 34.7

(0.4) (0.2) (3.3) (3.5) (2.7) (6.0) (3.2) (4.4)

S**

V L V L V L V L

6.5 5.2 4.6 5.7 79.3 73.6 9.7 15.5

(5.3) (5.2) (2.6) (4.9) (6.3) (6.9) (2.0) (3.8)

4A 8.9 9.3 15.2 72.0 62.1 14.3 13.8

(2.6) (5.1) (2.7) (4.3) (6.8) (2.6) (4.1) (5.2)

ns

Body length (cm) Tract contents (% body mass) Tract/body length ratio Distribution of length (%) Stomach Small intestine Proximal colon Distal colon Distribution of contents (%) Stomach Small intestine Proximal colon Distal colon

Statisticsa S***, P* I** S**, I* S** S***

I* S*** S***, I*

P** S*, P** ns

" Effects of species (S), population (wild or captive; P) and species by population interaction (I). Significanceof the F statistic: *** P < 0.001, ** P 0.05). Proximal colon 1 was the principal site of D M digestion in Lasiorhinus (39%). In Vombatus, the inflection in the sequence o f digestion in the small intestine precluded estimation of relative digestion in the small intestine and proximal colon 1. The remaining 15-19% of D M was digested in the distal regions of the proximal colon and the distal colon in both species (Fig. 4). Digestion in the stomach accounted for 22-23% of digestible N D F in both species. Lasiorhinus digested 37 %

556

P.S. Barboza and I.D. Hume: Digestion in wombats

70

30 27

S0 24 5O

21 v

40

18

s -~ 3O

Y_

~5 20

9

10

0 8t

81

PC

DO

0

St

Digestive tract segment

Digestion in wild wombats Vombatus was much larger than Lasiorhinus in the wild: 39.5 vs 21.3 kg ( P < 0.01). The composition o f the digesta from wild wombats was more variable than that o f captive animals (Table 2) but exhibited the same relative

PC1

PC2

PC3

PC4

DC

Digestive tract segment

Fig. 2. Distribution of the apparent mucosal surface area of the digestive tract in captive Vombatus (stippled bars) and Lasiorhinus (solid bars) (n = 4). Percentage contribution of the stomach (St), small intestine (S/), proximal colon (PC), and distal colon (DC). Areas significantly different across segments (t-test, P 0.05) at all segments; b wild wombats: n = 5 for Lasiorhinus and n = 4 for Vombatus, with n = 3 for stomach digesta from Vombatus. ** Species significantly different for proximal colon 4 (P