Microbial production of nutrients

Section Introduction:

Shortly after birth or hatching, the gastrointestinal tract of all vertebrates becomes colonized with relatively stable populations of predominantly anaerobic bacteria. Their contributions to the production and conservation of nutrients were reviewed by Stevens and Hume (1998). The number of bacteria in a given gut segment is determined by the pH and retention time of its contents. The low pH of gastric contents and rapid transit of digesta through the midgut or small intestine of most vertebrates inhibit their growth. However, the relatively neutral pH and prolonged digesta retention time of the hindgut or large intestine of terrestrial vertebrates and the foregut of some species result in a larger number of indigenous bacteria. These bacteria are accompanied by colonies of indigenous protozoa in the hindgut or foregut of some species and by indigenous fungi in the forestomach of some herbivorous mammals. The indigenous bacteria can convert carbohydrates into short-chain fatty acids (predominantly acetate, propionate and butyrate), CO2, H2 and CH4, and utilize nitrogenous compounds for the production of ammonia and microbial protein. They can also synthesize the B-complex vitamins required by their host.

The principal substrates for microbial fermentation and nitrogen metabolism in the hindgut of mammals are illustrated in Figure 9.1. Hindgut bacteria ferment dietary carbohydrates that escape digestion in the upper digestive tract and the endogenous carbohydrates in mucous and sloughed epithelial cells for the production of short-chain fatty acids (SCFA). They utilize the nitrogen in dietary compounds, digestive enzymes, and the urea that diffuses into the hindgut for the production of ammonia and microbial protein. Bacteria in the foregut produce the same end products from substrates derived from the diet and the urea that enters the forestomach via the saliva and diffusion. Most of the SCFA and much of the ammonia are absorbed. SCFA provide a major source of the energy required by hindgut or foregut epithelial cells and varying amounts of the maintenance energy of the animal. The ammonia nitrogen is incorporated into microbial protein or absorbed and incorporated into urea or amino acids by the liver. The hindgut bacteria of birds, reptiles, and adult amphibians perform similar functions, except that urea is replaced by uric acid, which arrives in the hindgut via the cloaca. The foregut fermenters and hindgut fermenters that practice coprophagy can also utilize the microbial protein and B-vitamins. The major differences between herbivores and other species are the greater capacity and longer retention time of their gut fermentation chamber.

<img alt="Bacterial fermentation of carbohydrates and metabolism of nitrogen in the hindgut" src="../images/dsv/GITFigures/MicrobialFermentationCarbohydratesCrissey2004.gif">

Figure 9.1. Bacterial fermentation of carbohydrates (left) and metabolism of nitrogen (right) in the hindgut of mammals. Most of the SCFA and ammonia are absorbed from the hindgut, but the microbial protein is lost in the feces of species that do not practice coprophagy. The hindgut bacteria of birds, reptiles and adult amphibians perform similar functions, except that the major waste product of protein metabolism is uric acid, rather than urea, and it enters the hindgut in the urine via the cloaca. (From Wrong & Vince 1984 and Stevens & Hume 1995.)

Characteristics and Distribution of Gut Microbes - Ruminant forestomach:

Much of our understanding of the nutritional contributions of gut microbes comes from the extensive studies of the ruminant forestomach. During the first weeks after birth the ruminant forestomach becomes colonized with E. coli aerogenes and streptococci, which are joined by lactobacilli in the suckling animal (Eadie and Mann 1970). Weaning is followed by development of the extremely complex group of microbiota that are characteristic of adult animals (Wolin 1979; Allison 1984). Culture counts give estimates of 1010 to 1011 predominantly anaerobic bacteria per gram of fluid rumen contents. Microscopic counts, which include organisms that are dead or require specific culture media, give higher numbers. Tables 9.1a and b list the principal bacterial species found in the rumen of sheep and cattle, and their fermentative properties. Collectively, they ferment carbohydrate into short-chain fatty acids, utilize protein and other nitrogenous compounds for production of ammonia, synthesize microbial protein and B-vitamins, hydrolyze lipids, and hydrogenate fatty acids.

Table 9.1a.

<img alt="Fermentation products of rumen bacteria" src="../images/dsv/Tables/MicrobialFermentationProductsRumenA%20T9_01.gif">

Function: C = cellulolytic, X = xylanolytic, D = dextrinolytic, P = pectinolytic, PR = proteolytic, GU = glycerol-utilizing, LU = lactate-utilizing, SS = major soluble sugar fermenting. Products: F = formate, A = acetate, E = ethanol, P = propionate, L = lactate, B = butyrate, S = succinate, H = hydrogen, C = carbon dioxide. (Modified by Allison 1984 from Hespell 1981)

Table 9.1b.

<img alt="Fermentation products of rumen bacteria" src="../images/dsv/Tables/MicrobialFermentationProductsRumenB%20T9_01b.gif">

Function: D = dextrinolytic, P = pectinolytic, PR = proteolytic, L = lipolytic, M = methanogenic, GU = glycerol-utilizing, LU = lactate-utilizing, SS = major soluble sugar fermenting, HU = hydrogen-utilizing. Products: F = formate, A = acetate, E = ethanol, P = propionate, L = lactate, B = butyrate, S = succinate, V = valerate, CP = caproate, H = hydrogen, C = carbon dioxide, M = methane. (Modified by Allison 1984 from Hespell 1981)

Although protozoa are much less numerous (104 to 105/g fluid digesta), they can occupy an almost equal volume of forestomach contents. The predominant species are anaerobic ciliates that belong to the families Isotrichidae and Ophryoscolecidae (Hungate 1966; Ogimoto and Imai 1981). Rumen protozoa ferment carbohydrates, store starch, digest protein, hydrogenate fatty acids, and regulate the numbers of bacteria (Hobson and Wallace 1982; Prins 1991). Although they contribute relatively little to carbohydrate fermentation, they store starch and synthesize protein for their subsequent digestion during passage through the abomasum and midgut.

The uptake of starch and sugars by protozoa has a stabilizing effect in ruminants fed high-grain diets, where rapid bacterial production of SCFA and lactic acid can result in ulceration of the forestomach, systemic acidosis, and dehydration (Dirksen 1970; Ushida et al. 1991). Defaunation of the rumen impaired the absorption of calcium, magnesium, and phosphorus, and changed the peptide patterns of duodenal digesta. Thus, the ciliated protozoa can play an important role in the digestive tract of ruminants. 

Anaerobic fungi in concentrations of 103 to 105 zoospores per g of fluid digesta were reported in animals on high fiber diets (Fonty 1991). They contain relatively high concentrations of protein with an amino acid composition similar to that of alfalfa (Fonty and Joblin 1991) and may be an important source of nutrients for ruminants on low quality diets.

Characteristics and Distribution of Gut Microbes - Gastrointestinal tract of other species:

Foregut: Bacterial counts similar to those of the ruminant forestomach have been reported in the forestomach of other mammalian herbivores, cetaceans, and the crop and distal esophagus of the herbivorous hoatzin (Table 9.2). Those in the forestomach of minke whales included species that could digest the chitinous exoskeleton of krill (Mathieson et al. 1990; Martensson et al. 1994). The bacteria were accompanied by 103 to 106 protozoa/g of digesta in the forestomach of camels, hippos, peccaries, the hoatzin, and some macropod marsupials, but protozoa were absent from the forestomach of sloths and the colobus and langur monkeys. Colonies of fungal sporangia, like those of the rumen, were also found in the forestomach of kangaroos and wallabies (Dellow et al. 1988). Where measured, bacterial counts in the secretory compartment of animals with a forestomach and the simple stomach of other vertebrates tend to be much lower than those of the forestomach. Mean concentrations of 103 to 104/ml of predominantly aerobic bacteria have been reported in the gastric juice of fasting humans (Rambaud 1992).

Table 9.2.

<img alt="Microbial counts in the foregut of herbivorous mammals and birds" src="../images/dsv/Tables/MicrobialCountsForegutHerbMammalsBirds%20%20T9_02.gif">

(From Stevens & Hume 1995)

Midgut: The concentration of bacteria in the midgut are generally much lower than those of the rumen (Table 9.3). Counts of 104 to 108/g of fluid digesta were reported in the human small intestine, and the counts of anaerobic bacteria in grass-fed horses ranged from 106/g of digesta in the duodenum to 108/g in the ileum. The high percentage of proteolytic organisms in the duodenum of horses, suggested that they may digest or compete with endogenous enzymes. Concentrations of 103 to 109/g of digesta were found in the midgut of a variety of teleosts. Many of these bacteria were aerobes and facultative anaerobes, and the number of obligate anaerobes cultured from homogenates of the gastrointestinal tract of the carnivorous rainbow trout was insignificant. However, 107 to 109 anaerobes/g were cultured from the intestine of herbivorous grass carp Table 9.3). Chitinolytic bacteria in counts of 107 to 109/g were reported in the intestine of a number of marine fish, and protozoa in counts of 103 to 106/g of digesta were reported in the midgut of the herbivorous herring cale.

Table 9.3.

<img alt="Microbial counts in the midgut of vertebrates" src="../images/dsv/Tables/MicrobialCountsMidgutVert%20T9_03.gif">

(From Stevens & Hume 1995) 

Hindgut: Bacterial counts of 107 to 1012/g were reported in the hindgut of mammals, birds, reptiles, and leopard frogs (Table 9.4). Hibernation reduced the number and variety of bacteria in the hindgut of leopard frogs, but subcultures of isolates showed that they continued to grow at 4oC (Gossling et al. 1982). Bacterial species in the hindgut of mammals were similar to those found in the rumen (Wolin 1981; Allison 1984). At least 400 species, representing 50 genera of bacteria were isolated from human feces (Topping and Clifton 2001). However, bacterial populations associated with the epithelial surface and lumen contents of the hindgut can differ from one another and from those in the feces. They can also vary between the segments of hindgut. Methanogenic bacteria appear to preferentially colonize the distal colon of humans that are CH4-excretors (Pochart et al. 1993) and the epithelial surfaces of the cecum and proximal colon of the koala were populated with large numbers of bacteria capable of degrading the tannin-protein complexes in their normal diet of Eucalyptus foliage (Osawa et al. 1993). Protozoa were also found in the hindgut of horses, rodents, elephants, green iguanas, and sea bass (Table 9.4).

Table 9.4.

<img alt="Microbial counts in the hindgut of vertebrates" src="../images/dsv/Tables/MicrobialCounts%20HindgutVert%20T9_04.gif">

(From Stevens & Hume 1995)

Fermentation of Carbohydrates - Ruminant forestomach:

The extracellular enzymes of rumen bacteria can ferment starches, cellulose, hemicelluloses and pectins into monosaccharides, which are converted by intracellular enzymes to pyruvate, as in the body cells of vertebrates. However, rather than entering a Krebs cycle for aerobic metabolism to CO2 and H2O, pyruvate is reduced anaerobically to short-chain organic acids, principally acetate, propionate, and butyrate, plus CO2, H2, and CH4 (Fig 9.2). These organic acids were called volatile fatty acids (VFA) in the early literature, due to their ready separation from other components of digesta by steam distillation, but they are now referred to as short-chain fatty acids (SCFA). Although fermentation of starch produces 20% less energy than its conversion to glucose by endogenous enzymes, microbial fermentation of structural carbohydrates is a tremendous advantage to animals on a high-fiber diet.

<img alt="Pathways of carbohydrate metabolism by bacteria in the rumen" src="../images/dsv/GITFigures/MicrobialPathwaysCarbohydrateMetabolism%20F9_02.gif">

Figure 9.2. Pathways of carbohydrate metabolism by bacteria in the ruminant forestomach. (From Van Soest 1994.)

The total concentration of SCFA in the forestomach of sheep and cattle varies between 60 and 120 mmoles/L, depending on the diet and time after feeding (Phillipson 1977). The rate of SCFA production depends on the substrate; soluble carbohydrate (starches and sugars) > pectin > cellulose (Fig. 9.3). In animals fed hay or other roughage, the SCFA consisted of 60 to 70% acetate, 15 to 20% propionate, and 10 to 15% butyrate. The acetate/propionate ratio was reduced by either an increase in the concentration of soluble carbohydrates in the diet or a reduction in rumen pH, and SCFA are replaced by lactic acid at a pH lower than 5.0 (Fig. 9.4).

<img alt="Rate of fermentation of alfalfa in the rumen" src="../images/dsv/Graphs/MicrobialFermentationRateAlfalfa%20F9_03a.gif">

Figure 9.3. Rate of fermentation of alfalfa components in the rumen. (From Baldwin et al. 1977).

<img alt="Rumen pH, and proportions of acetic, propionic, and lactic acid" src="../images/dsv/Graphs/MicrobialpHAceticPropionicLactic%20F9_03b.gif">

Figure 9.4. Relationship between ruminal pH and the proportions of acetic, propionic, and lactic acid produced. (From Kaufmann et al. 1980.)

Rumen gases vary in their rate of production and composition with time after feeding (Fig. 9.5). Carbon dioxide is derived from fermentation of carbohydrate and the neutralization of SCFA with HCO3-. Methane is produced by the reduction of CO2 by formate, succinate, and H2, which accounts for the low concentrations of H2 in the rumen except for the first few days of a fasting period. Methane production is directly proportional to acetate production and inversely proportional to the production of propionate, but it also depends on other factors that affect the growth and replication of methanogenic organisms. Nitrogen and O2 are added from swallowed air, and N2 can diffuse into the rumen from the blood, as well. Rumen microorganisms rapidly reduce O2. Some CO2 is directly absorbed into the blood stream, but much of the CO2 and most of the CH4 produced in the rumen are removed by eructation. Kleiber (1961) found that an adult cow lost 191 liters of CH4/day through eructation and flatulence, which was equivalent to a 10% loss in its daily digestible food energy.

<img alt="Composition of rumen gases in a dairy cow fed hay and grain" src="../images/dsv/Graphs/MicrobialRumenGases%20F9_04.gif">

Figure 9.5. Composition of rumen gases in a dairy cow on a ration of hay and grain (Washburn & Brody 1937.)

Fermentation of Carbohydrates - Forestomach of other species:

SCFA concentrations similar to those in the forestomach of cattle and sheep were reported in the forestomach of other artiodactyls, macropod marsupials, sloths, colobus and langur monkeys, hyraxes, and some rodents, and in the crop and distal esophagus of the hoatzin (Table 9.5). High concentrations were also found in the forestomach of baleen (Herwig et al. 1984, Mathiesen et al. 1990) and toothed (Morii 1972; 1979; Morii and Kanazu 1972) whales. Where measured, the proportions of acetate, propionate and butyrate were generally similar to those in the ruminant forestomach. However, acetate was the predominant SCFA in the hyrax stomach, and this was accompanied by relatively high levels of lactic acid. Savage (1977) found lactobacilli attached to the stratified squamous epithelium lining the forestomach of many rodents, as well.

Table 9.5. Short-chain fatty acids in the foregut of herbivorous birds and mammals.

<img alt="Short chain fatty acids in the foregut of birds and mammals" src="../images/dsv/Tables/MicrobialSCFAForegutBirdsMammals%20T9_05.gif">

Dashes indicate absence of information. Contributions of SCFA to maintenance energy were estimated from the rate of SCFA production by in vitro isotope dilution or measurements of digesta flow. Total maintenance energy was either calculated as twice the BMR or assumed to be equivalent to ad libitum digestible energy intake in captive, nonreproducing, adult animals. (From Stevens & Hume 1995)

Carbon dioxide, H2, and CH4 were found in the forestomachs of macropod marsupials. The forestomach of wallabies contained higher levels of H2 and lower levels of CH4 than that of ruminants, and only negligible amounts of CH4 were present in the forestomach of the eastern gray kangaroo. The principal gasses in the forestomach of langur monkeys were CO2 and CH4, but investigators disagreed over the presence of CH4 in the forestomach of colobus monkeys. The more rapid transit of digesta may inhibit the establishment of slow-growing methanogenic bacteria.

Because SCFA are almost solely the result of anaerobic bacterial fermentation, their presence in the digestive tract is an index of indigenous microbe activity. Therefore, the low levels of SCFA found in the secretory segment of the stomach (abomasum) of ruminants were assumed to be the result of forestomach fermentation. However, low levels of SCFA were also found in the stomachs of dogs, raccoons, bush babies, vervet monkeys, pigs, and ponies (Fig. 9.6), and these were accompanied by substantial concentrations of lactic acid in the stomachs of the dog, pig and pony. Therefore, microbial fermentation occurs to some degree in the stomach of most, if not all vertebrates.

<img alt="Concentrations of short chain fatty acids in the gastrointestinal tract of mammals" src="../images/dsv/Graphs/MicrobialSCFACarnOmniHerb%20F9_05.gif">

Figure 9.6. Concentrations of VFA (SCFA) along the gastrointestinal tracts of mammalian carnivores, omnivores, and herbivores. Animals were fed a at 12 hour intervals. Each value represents the mean (+/- SE) of 12 samples, consisting of three samples collected at two, four, eight, and 12 hours after a meal, from the oral (S1) and aboral (S2) segments of the stomach, three equal-length segments of the small intestine (SI1, SI2, SI3), the cecum (Ce), and two or three equal-length segments of the colon (C1, C2, C3). (Modified from Argenzio et al. 1974b; Clemens et al. 1975a; Clemens & Stevens 1979; Clemens 1980.)

Fermentation of Carbohydrates - Midgut:

SCFA have been found in the midgut or small intestine of all classes of vertebrates (Table 9.6). SCFA concentrations were low in the small intestine of dogs, raccoons, and ponies, but higher concentrations were found in the terminal midgut of the pig, bush baby, vervet monkey (Fig. 9.6, see above). Higher concentrations of SCFA were also reported in the midgut of emus, Florida red-bellied turtles, and a number of fish. The levels in the emu and Florida red-bellied turtle were equivalent to those of the rumen. Concentrations of 7-47 mmoles/L were measured in the midgut of carnivorous rainbow trout, omnivorous common carp, and herbivorous grass carp, and 31 species of tropical marine fish (Clements and Choat 1995). Acetate was the predominant fatty acid in the midgut of emus and fish.

Table 9.6.

<img alt="Short chain fatty acids in the midgut of vertebrates" src="../images/dsv/Tables/MicrobialSCFAMidgutVert%20T9_06.gif">

Dashes indicate absence of information. Contributions of SCFA to maintenance energy were estimated from the rate of SCFA production by in vitro isotope dilution or measurements of digesta flow. Total maintenance energy was either calculated as twice the BMR or assumed to be equivalent to ad libitum digestible energy intake in captive, nonreproducing, adult animals. (From Stevens & Hume 1995)

Fermentation of Carbohydrates - Hindgut:

SCFA levels in the hindgut of the dog, raccoon, bush baby, vervet monkey, pig, and pony (Fig. 9.6, see above) were within the range found in the rumen of cattle and sheep. SCFA concentrations ranging from 65-235 mmoles/L also were found in the hindgut of carnivorous, omnivorous, and herbivorous mammals, geese, ptarmigan, and the green sea turtle (Tables 9.7a and b). Lower concentrations were reported in the hindgut of some marsupials (greater glider and koala), reptiles (caiman, green iguana, and tortoises), and the sea chub. Acetogenic and butyricogenic bacteria were also reported in the hindgut of the leopard frog (Gossling et al. 1982). Acetate/propionate/butyrate ratios were similar to those of the ruminant forestomach, and the addition of fiber to the diet had a similar effect on the acetate/propionate ratio in the pig.

Table 9.7a.

<img alt="Short cahin fatty acids in the hindgut of vertebrates" src="../images/dsv/Tables/MicrobialSCFAHindgutVertA%20T9_07a.gif">

* Absorption from cecum (or ceca) alone.

Dashes indicate absence of information. Contributions of SCFA to maintenance energy were estimated from the rate of SCFA production by in vitro isotope dilution or measurements of digesta flow. Total maintenance energy was either calculated as twice the BMR or assumed to be equivalent to ad libitum digestible energy intake in captive, nonreproducing, and adult animals. (From Stevens & Hume 1995)

Table 9.7b.

<img alt="Short chain fatty acids in the hindgut of vertebrates" src="../images/dsv/Tables/MicrobialSCFAHindgutVertB%20T9_07b.gif">

* Absorption from cecum (or ceca) alone.

Dashes indicate absence of information. Contributions of SCFA to maintenance energy were estimated from the rate of SCFA production by in vitro isotope dilution or measurements of digesta flow. Total maintenance energy was either calculated as twice the BMR or assumed to be equivalent to ad libitum digestible energy intake in captive, nonreproducing, and adult animals. (From Stevens & Hume 1995)

The total concentrations of SCFA in the hindgut of the dog, pig, and pony were relatively unaffected by changes in diet or the time after feeding (Fig 9.6, see above). However, addition of fiber to the diet increased the volume of hindgut contents and, thus, the quantity of SCFA in the hindgut of pigs and ponies. Compartmental analysis of the pony hindgut showed marked, cyclic changes in the net appearance and disappearance of SCFA with time after feeding (Fig. 9.7). Parallel changes in the influx and efflux of water account for the relatively constant levels of SCFA in hindgut contents with time following a meal (Fig. 9.6).

<img alt="Water and volatile fatty acids in the large intestines of ponies after feeding" src="../images/dsv/Graphs/MicrobialVolumeTransmucosalVFA%20F9_06.gif">

Figure 9.7. Volume, net transmucosal flux of water, and net appearance and disappearance of VFA (SCFA) in the large intestine of ponies, with time after feeding. All values, other than volume, are corrected for exchanges between segments that resulted from digesta flow. (Modified from Argenzio et al. 1974 a,b.)

Calloway (1968) reviewed information on the composition of gases in the large intestine of dogs, rats, pigs, cattle, horses and humans. They consisted of the same gasses as those found in the rumen, but with considerable variation among species and with changes in the diet. The human large intestine contained higher percentages of H2 and N2, and lower percentages of CO2 and CH4 than the rumen, and CH4 was absent in about two-thirds of the human population (Levitt and Bond 1970). High levels of H2 and low levels of CH4 were also reported in the hindgut of the green turtle (Bjorndal 1991).

The principal substrates for SCFA production in carnivores and omnivores are dietary starches that escape digestion in the midgut and endogenous carbohydrates (Fig. 9.1). Significant quantities of dietary starch reached the large intestine of humans, rats, mice, hamsters, guinea pigs, rabbits, pigs, cattle, sheep, and ponies (Baker et al. 1950; Karr et al. 1966; Orskov et al. 1971; Hintz et al. 1971; Keys and DeBarthe 1974). Up to 20% of the ingested starch reached the colon of humans. The amount of starch that escaped digestion in the small intestine depended on the dietary source, and was reduced by either boiling or grinding. Endogenous carbohydrates appear to be the principal substrates for SCFA production in the human hindgut (Ehle et al. 1982). Mucus, which is 80% polysaccharide, may be a major substrate (Vercellotti et al. 1978). This may account for the large quantities of mucus that accumulate in the cecum of germ-free rats and guinea pigs (Gordon and Bruckner 1984). Carbohydrates and the carbon skeletons of the amino acids in epithelial cells that are sloughed into the intestine can serve as additional substrates for SCFA production.

<img alt="Bacterial fermentation of carbohydrates and metabolism of nitrogen in the hindgut" src="../images/dsv/GITFigures/MicrobialFermentationCarbohydratesCrissey2004.gif">

Figure 9.1. Bacterial fermentation of carbohydrates (left) and metabolism of nitrogen (right) in the hindgut of mammals. Most of the SCFA and ammonia are absorbed from the hindgut, but the microbial protein is lost in the feces of species that do not practice coprophagy. The hindgut bacteria of birds, reptiles and adult amphibians perform similar functions, except that the major waste product of protein metabolism is uric acid, rather than urea, and it enters the hindgut in the urine via the cloaca. (From Wrong & Vince 1984 and Stevens & Hume 1995.)

Due to a longer digesta retention time, dietary cellulose, hemicellulose and pectin are major substrates in the hindgut of most herbivores. Microbial fermentation accounted for 63-73% of the neutral detergent fiber digested in the hindgut of ponies and 12% of the cellulose digested in the gastrointestinal tract of sheep. Cellulose digestion has been reported in the gut of carp, freshwater catfish, and 16 species of marine detritus feeders. There was no evidence of cellulolytic activity in the intestine of the algae-feeding tilapia, but algae contain little or no cellulose.

Production of Microbial Protein and Recycling of Nitrogen - Ruminant forestomach:

When provided with adequate amounts of dietary nitrogen, the microbial protein synthesized in the forestomach makes ruminants independent of the form in which it is provided. This is because rumen bacteria degrade most forms of dietary nitrogen to peptides, amino acids and ammonia. From 50 to 80% of the nitrogen incorporated into microbial protein comes from ammonia, with most of the remainder coming from peptides. Rumen bacteria have transport systems for the uptake of peptides and ammonia but not for free amino acids.

The amount of microbial protein produced in the rumen is closely related to the rate of SCFA production. About 9-10 g of microbial protein are synthesized per mole of ATP produced by fermentation (Dellow 1982). Only 10% of the substrate energy is released as ATP. Most of the remainder appears as SCFA. The SCFA, which are too highly reduced to be available to the microbes, are absorbed and oxidized in the aerobic environment of the host animal’s tissues. Anaerobiosis sets limits on the amount of amino acids that can be supplied by the rumen microbes to the animal. This is sufficient for maintenance and slow growth. However, for higher levels of production the diet must include protein that escapes degradation in the rumen and is digested in the small intestine. Only 10-20% of the protein of fresh forage escapes ruminal degradation, but this can be increased by treating the forage with heat or with chemicals such as formic acid or formaldehyde.

Much of the nitrogen utilized for microbial protein synthesis is derived from urea. Urea enters the rumen via salivary secretions and diffusion down a steep blood-to-lumen concentration gradient maintained by its conversion to ammonia by ureolytic bacteria attached to the rumen epithelium (Houpt and Houpt 1968; Egan et al. 1986). Therefore, much of the waste from nitrogen metabolism joins the rumen ammonia pool. A portion is incorporated into microbial protein and absorbed as amino acids following its digestion in the small intestine. On low-protein diets, nitrogen recycling can contribute to a large part of the protein flowing out of the ruminant forestomach. The remaining ammonia is absorbed from the forestomach and recycled to the liver for synthesis of amino acids and urea. Numerous studies have demonstrated that urea nitrogen recycling reduces the amount of water required for the renal excretion of urea and, conversely, a restriction of water intake reduces urea excretion and increases the rate of urea-nitrogen recycling through the forestomach (Stevens and Hume 1995).

Production of Microbial Protein and Recycling of Nitrogen - Forestomach of other species:

Camels fed dry, low-protein, desert grass recycled 95% of the urea synthesized by their liver (Mousa et al. 1983). The forestomach of macropod marsupials demonstrates a similar process of urea nitrogen recycling (Dellow and Hume 1982; Dellow et al. 1983). Addition of urea to low-protein diets increased the efficiency of nitrogen utilization by the euro (Brown 1969) and tammar wallaby (Kinnear and Main 1975; Kennedy and Hume 1978).

Production of Microbial Protein and Recycling of Nitrogen - Hindgut:

The major substrates degraded for the production of ammonia and synthesis of microbial protein in the hindgut of mammals are urea, creatinine, digestive enzymes, mucus, sloughed cells, and dietary residues (Fig. 9.1). Most of the amino acids appear to be derived from mucin and amino acids that are least efficiently absorbed from the small intestine. Release of urea into the mammalian hindgut is also due to its passive diffusion down a steep concentration gradient maintained by ureolytic bacteria at the lumen surface. Most of the ammonia is either incorporated into microbial protein or absorbed and recycled to the liver for synthesis of nonessential amino acids or urea. Uric acid, the waste product of protein metabolism in amphibians, reptiles and birds does not defuse across gut epithelium. However, it enters the hindgut via the cloaca and is utilized similarly in the production of ammonia and microbial protein.

<img alt="Bacterial fermentation of carbohydrates and metabolism of nitrogen in the hindgut" src="../images/dsv/GITFigures/MicrobialFermentationCarbohydratesCrissey2004.gif">

Figure 9.1. Bacterial fermentation of carbohydrates (left) and metabolism of nitrogen (right) in the hindgut of mammals. Most of the SCFA and ammonia are absorbed from the hindgut, but the microbial protein is lost in the feces of species that do not practice coprophagy. The hindgut bacteria of birds, reptiles and adult amphibians perform similar functions, except that the major waste product of protein metabolism is uric acid, rather than urea, and it enters the hindgut in the urine via the cloaca. (From Wrong & Vince 1984 and Stevens & Hume 1995.)

Urea nitrogen was extensively recycled through the hindgut of the rabbit, pony, rock hyrax, greater glider, brushtail possum, wombat, and donkey (Stevens and Hume 1995). Nitrogen balance was maintained in rabbits on a low-protein diet by infusion of urea into the cecum (Salse et al. 1977). The recycling of urea nitrogen by the hyrax hindgut increased with either a reduction in dietary protein or restriction of water (Hume et al. 1980), in a manner similar to that seen in the ruminant forestomach.

Conversion of fat jirds, donkeys, and Bedouin goats from the alfalfa or lucerne hay diet to poorly digestible rhodes grass or wheat straw increased the recycling of urea from 53 to 89%, 16 to 75%, and 40 to 69%, respectively, with no loss in body weight (Brosh et al. 1986; Izraely et al. 1989a; Yahav and Choshniac 1989). Black bears recycled 20% of their endogenous urea through their digestive tract during hibernation, when they neither drink, eat, or defecate (Guppy 1986). Recycling of urea was also reported in the gut of sharks (Knight et al. 1988) and the gulf toadfish (Walsh et al. 1990). Recycling of uric acid nitrogen has been demonstrated in chickens (Bell and Bird 1966; Mead and Adams 1975) and ptarmigan (Mortensen and Tindall 1981), and it increased with a reduction in dietary protein in the chickens.

Bacterial degradation of endogenous and microbial protein in the hindgut can also produce amino acids. Although carrier-mediated absorption of amino acids has been demonstrated in the ceca of birds and the colon of chickens on a high-Na diet (Skadhauge 1993), it appears to be absent in the hindgut of other vertebrates. Therefore, most of the microbial protein is voided in the feces and lost by species that do not practice coprophagy. However, feces contain high levels of microbial protein, and much of the nitrogen of cecotrophs is incorporated into essential amino acids (Bjőrnhag 1994).

Synthesis of B-Vitamins - Ruminant forestomach:

Gut microbes can synthesize B-vitamins; a complex of 10 separate water-soluble compounds. Ruminants do not require a dietary source of B-vitamins because they are synthesized by the microbes in their forestomach. However, cobalt is required for the microbial synthesis of vitamin B12, which accounts for the relatively high cobalt requirement of ruminants (Phillipson 1970). The same may be true of other foregut fermenting herbivores. B-complex vitamins are absorbed mainly from the small intestine by Na+-dependent transport mechanisms.

Synthesis of B-Vitamins - Hindgut:

 B-vitamins are also synthesized by hindgut bacteria, but the extent to which they are absorbed from the hindgut is uncertain. There is good evidence that nicotinic acid, riboflavin, pantothenic acid, thiamin, biotin, pyridoxine, folic acid, and vitamin B12 are synthesized by bacteria in the human colon, and all but the first three were absorbed to some degree (Wrong et al. 1981). Pantothenic acid, pyridoxine, and B12 were absorbed equally well following their large intestinal versus oral administration to humans (Sorrell et al. 1971). Thiamin was absorbed from the rat cecum (Kasper 1962), and several studies that showed that rats fed a diet lacking riboflavin, pantothenic acid, biotin, pyridoxine, folic acid, or vitamin B12 showed severe deficiencies unless they were allowed to ingest their feces. The degree to which the mouse, guinea pig, and rabbit required a dietary source of these vitamins was inversely proportional to the degree that they recycle microbially synthesized B-vitamins to their stomach and small intestine by coprophagy (Table 9.8). Rabbits, which are cecotrophic, appeared to be independent of a dietary source of all but three B-vitamins. Guinea pigs, which are coprophagic but not cecotrophic, require more of the vitamins in their diet. Laboratory mice ingest little of their feces and require all of the B-vitamins in their diet. 

Table 9.8.

<img alt="Vitamin requirements for growth of rabbits, guinea pigs, and mice" src="../images/dsv/Tables/MicrobialVitRequRabbitGuineapigMice%20T9_08.gif">

(From NRC 1977, 1978)

Absorption of end products - Short-chain fatty acids:

SCFA are rapidly absorbed from all segments of the gastrointestinal tract. Their absorption provides a major contribution to the nutrition of forestomach and hindgut epithelial cells, and contributes in varying degrees to the total energy requirements of the animal. Acetate and butyrate are metabolized to CO2 and ketone bodies by rumen epithelium (Stevens and Stettler 1966a, b) and the ileal, cecal and colonic mucosa of the rabbit (Roediger 1991). Ketogenesis was low in the ileum and distal colon, but high in the rumen, cecum, and proximal colon. Rumen and colonic epithelial cells also metabolized propionate, but butyrate provides the major source of energy for colonic cells. 

The contributions of SCFA to maintenance energy have been estimated for a number of species. Absorption of SCFA from the forestomach provided a substantial amount of maintenance energy requirements of ruminants and macropod marsupials, and about 10 to 13% of energy required for maintenance of the hyrax (Table 9.5). Absorption from the midgut accounted for most of the daily energy requirement of the Florida red-bellied turtle, but only 11% of the maintenance energy of the emu (Table 9.6). Despite a limited gut capacity, rapid digesta transit time, and low SCFA concentrations, the absorption of SCFA may provide a substantial amount of the energy required by herbivorous fish. Hindgut absorption of SCFA contributed to only a small percentage of the maintenance energy requirement of dogs and humans, but it provided substantial amounts of the maintenance energy required by cecum and colon fermenting mammalian herbivores (Table 9.7a, b).

Table 9.5. Short-chain fatty acids in the foregut of herbivorous birds and mammals.

<img alt="Short chain fatty acids in the foregut of birds and mammals" src="../images/dsv/Tables/MicrobialSCFAForegutBirdsMammals%20T9_05.gif">

Dashes indicate absence of information. Contributions of SCFA to maintenance energy were estimated from the rate of SCFA production by in vitro isotope dilution or measurements of digesta flow. Total maintenance energy was either calculated as twice the BMR or assumed to be equivalent to ad libitum digestible energy intake in captive, nonreproducing, adult animals. (From Stevens & Hume 1995)

Table 9.6.

<img alt="Short chain fatty acids in the midgut of vertebrates" src="../images/dsv/Tables/MicrobialSCFAMidgutVert%20T9_06.gif">

Dashes indicate absence of information. Contributions of SCFA to maintenance energy were estimated from the rate of SCFA production by in vitro isotope dilution or measurements of digesta flow. Total maintenance energy was either calculated as twice the BMR or assumed to be equivalent to ad libitum digestible energy intake in captive, nonreproducing, adult animals. (From Stevens & Hume 1995)

Table 9.7a.

<img alt="Short cahin fatty acids in the hindgut of vertebrates" src="../images/dsv/Tables/MicrobialSCFAHindgutVertA%20T9_07a.gif">

* Absorption from cecum (or ceca) alone.

Dashes indicate absence of information. Contributions of SCFA to maintenance energy were estimated from the rate of SCFA production by in vitro isotope dilution or measurements of digesta flow. Total maintenance energy was either calculated as twice the BMR or assumed to be equivalent to ad libitum digestible energy intake in captive, nonreproducing, and adult animals. (From Stevens & Hume 1995)

Table 9.7b.

<img alt="Short chain fatty acids in the hindgut of vertebrates" src="../images/dsv/Tables/MicrobialSCFAHindgutVertB%20T9_07b.gif">

* Absorption from cecum (or ceca) alone.

Dashes indicate absence of information. Contributions of SCFA to maintenance energy were estimated from the rate of SCFA production by in vitro isotope dilution or measurements of digesta flow. Total maintenance energy was either calculated as twice the BMR or assumed to be equivalent to ad libitum digestible energy intake in captive, nonreproducing, and adult animals. (From Stevens & Hume 1995)

Studies of SCFA transport mechanisms are complicated by their presence as both undissociated acids and anions. Cell membranes are impermeable to the passive diffusion of the water soluble anions. However, they are permeable to passive diffusion of the more lipid-soluble undissociated organic acids down their concentration gradient, and the degree of lipid solubility increases by a factor of approximately 2.8 with each additional CH2 group (acetate < propionate < butyrate). Due to their low pK (4.75 to 4.81), the undissociated forms constitutes only 1 to 6% of the total SCFA in the normal range of digesta pH and less than 1% at the pH of blood. Measurements of concentration gradients of both forms of SCFA across the epithelial cells are further complicated by the lower pH of the intervening epithelial cell contents and the metabolism of SCFA in the course of transport. 

Results from a large number of in vivo and in vitro studies of SCFA transport across rumen, midgut, and hindgut epithelium were reviewed by Engelhardt (1995) and Stevens and Hume (1995). An increase in the rate of SCFA absorption with either a reduction in the pH of lumen contents or increase in the carbon length suggests that they are absorbed as undissociated, lipid-soluble organic acids (Stevens and Stettler 1966a, b). However, their relative rates of absorption were not proportional to the pH gradient between lumen contents and blood, which may be attributed to a micro-layer of fluid at the lumen surface of the epithelial cells that is maintained at a pH different from that of the bulk lumen contents. The increase in their rate of uptake from the gut lumen with an increase in carbon length also could be due to the differences in their rate of intracellular metabolism. Furthermore, studies of SCFA transport across membrane vesicles prepared from epithelial cells of the tilapia midgut (Titus and Ahearn 1991, 1992), rat distal colon (Mascola et al. 1991; Reynolds et al. 1992), and human proximal colon (Harig et al. 1990) suggest that a major fraction of the SCFA may be transported by carrier-mediated exchange with HCO3- across the lumen-facing membrane, and with HCO3-, or Cl- across the basolateral membranes of the epithelial cells. 

Figure 9.8 combines the models that have been proposed for transport of SCFA across rumen and hindgut epithelia. The concentration gradients for passive diffusion of undissociated acids would be increased by the addition of H+ to the lumen by hydration of CO2 in the lumen and the secretion of H+ in exchange for Na+. The interrelationship between SCFA and Na+ absorption could be attributed to intracellular CO2 hydration, resulting in the generation and subsequent secretion of H+ ions in exchange for Na+, and HCO3- in exchange for SCFA- (and Cl-) ions. The contribution of each of these factors could vary among gut segments and species.

<img alt="Mechanisms proposed for the transport of short chain fatty acids across the forestomach and hindgut epithelium" src="../images/dsv/GITFigures/MicrobialMechanismsTransportSCFA%20F9_07.gif">

Figure 9.8. Mechanisms proposed for the transport of SCFA transport across gut epithelium. Hydrogen ions produced by hydration of the CO2 in the lumen or secreted by carrier-mediated Na+/H+ exchange in the lumen-facing membrane may protonate SCFA anions (Ac-) to their undissociated form (HAc), which passively diffuses across these membranes. The H+ and HCO3- produced by carbonic anhydrase-catalyzed intercellular hydration of CO2 produces both H+ for carrier-mediated Na+/H+ exchange and HCO3- for exchanged with SCFA- anions in the lumen. SCFA may be transported across the basolateral membrane by either diffusion of the undissociated form or carrier-mediated exchange of SCFA- anions with blood HCO3-. (Modifications and combinations of models from Stevens et al. 1969; 1986 and Titus & Ahearn 1992.)

SCFA absorption also provides a significant contribution to the absorption of Na+ and water. The mechanisms of Na absorption will be discussed in the next section. However, SCFA and Na+ are the principal anions and cation in the ruminant forestomach and the hindgut of most species. Both are rapidly absorbed and their absorption is partly interdependent. Therefore, absorption of SCFA accounts for the absorption of most of the water that is recovered from these two segments of gut, and a fulminating production SCFA can have adverse effects on both the electrolyte-water and acid base balance of the host animal.

Absorption of end products - Ammonia:

Ammonia is also present in undissociated (NH3) and ionized (NH4+) forms. With a pK of approximately 9, most of the ammonia is present as NH4+ ions at the normal pH of rumen and hindgut contents. As with SCFA, ammonia was assumed to be passively absorbed as the more lipid-soluble NH3. However, the pH and concentration of ammonia at the lumen surface is unknown, and there is evidence for carrier-mediated transport of NH4+ by the rat ileum (Koch and Hall 1992) and the rectal pad of locusts (Phillips et al. 1994).

Absorption of end products - Microbial protein:

The microbial protein produced in the forestomach and in the hindgut of coprophagic species is digested by the gastric and midgut endogenous enzymes and absorbed as amino acids from the midgut. The nutritional significance of microbial protein produced in the foregut was mentioned earlier. The contributions of cecotrophy were reviewed by Hörnicke and Björnhag (1980) and Björnhag (1994). The soft feces of rabbits provided up to 30% of the nitrogen intake, and much of this was microbial protein with a high content of essential amino acids. The nitrogen content of cecotrophs also greatly exceeded that of the diet and normal feces in the Scandinavian lemming, nutria, guinea pig, chinchilla, and Norwegian and kangaroo rats. Cecotrophs provided the ringtail possum with 58% of its daily intake of digestible energy and 81% of its daily intake of nitrogen (Chilcott and Hume 1985).

Digestive Diseases Associated with Diet or Feeding Regimen: 

The normal functions of the indigenous bacteria can be disrupted by a variety of conditions. Bacterial populations can be drastically changed by the prolonged oral administration of antibiotics; especially those that attack a broad spectrum of microorganisms. Any infectious or parasitic condition that leads to maldigestion or malabsorption of carbohydrates in the midgut can result in diarrhea, due to their fermentation into SCFA and lactic acid in the hindgut at rates too rapid for their efficient absorption. This can reduce the digesta pH to levels that destroy the bacteria that are normally present and result in overgrowth of pathogenic species. No effective procedure for rapid repopulation of the gut with its normal compliment of bacteria has been devised.

A number of dysfunctions in microbial fermentation can result from inappropriate diets or feeding practices. The feeding of lactose-containing milk or milk products to lactase-deficient animals, such as the neonate pinnipeds, can produce diarrhea. Rapid fermentation of starch in the forestomach of ruminants fed high levels of grain can initiate a fulminating production of SCFA and depression of digesta pH, which reduces the populations of normal indigenous microorganisms and increases the populations of lactobacilli (Fig. 9.4). The high levels of SCFA and lactic acid can result in hypertonic digesta and systemic dehydration, and their rapid absorption can produce ulceration of the forestomach epithelium and systemic acidosis (Dirksen 1970). Increased production of CO2 and CH4, and their entrapment in the digesta, can also increase intraruminal pressure to levels that produce cardiovascular collapse (Dougherty 1977). High-grain diets can also increase the amounts of SCFA and gas released into the abomasum, which may account for the high incidence of abomasal ulcers in calves and abomasal displacement of cattle on these diets (Svendsen 1969). 

<img alt="Rumen pH, and proportions of acetic, propionic, and lactic acid" src="../images/dsv/Graphs/MicrobialpHAceticPropionicLactic%20F9_03b.gif">

Figure 9.4. Relationship between ruminal pH and the proportions of acetic, propionic, and lactic acid produced. (From Kaufmann et al. 1980.)

Over-production of SCFA, followed by a drop in pH and production of lactic acid, was also observed in the cecum of horses fed intermittent meals of pelleted, high-grain diets for additional energy or convenience in feeding (Clarke et al. 1990b). As noted earlier, ponies fed a pelleted hay-grain diet at 12-hour intervals showed a marked increase in the volume of the cecum and proximal colon the first eight hours after feeding and a return to prefeeding levels during the last four hours prior to the next meal (Fig 9.7). This was accompanied by a 15 % reduction in plasma volume within one hour after feeding, and followed by recovery and a smaller reduction six hours after the meal (Clarke et al. 1990a). The initial reduction in plasma volume was attributed to salivary and pancreatic secretion, and the second reduction was attributed to hindgut secretion. The drop in plasma volume was associated with an increase in the plasma levels of renin and aldosterone (Fig. 9.9). Renin is released from the kidney in response to hypovolemia and generates the release of angiotensin, which reduces urinary excretion and stimulates the more gradual release of aldosterone. Aldosterone stimulates the absorption of Na+ and water by both the kidney and hindgut. Ponies fed the same diet at two hour intervals showed none of these reactions. Therefore, much of the colic, torsion, and volvulus seen in the large intestine of domesticated horses may be due to a diet or feeding regimen for which their digestive system was never designed. Exacerbation of ulcerative colitis inhuman patients was also accompanied by a reduction in pH and butyrate levels, and an increase in lactic acid concentrations

<img alt="Water and volatile fatty acids in the large intestines of ponies after feeding" src="../images/dsv/Graphs/MicrobialVolumeTransmucosalVFA%20F9_06.gif">

Figure 9.7. Volume, net transmucosal flux of water, and net appearance and disappearance of VFA (SCFA) in the large intestine of ponies, with time after feeding. All values, other than volume, are corrected for exchanges between segments that resulted from digesta flow. (Modified from Argenzio et al. 1974 a,b.)

<img alt="Colonic water exchange, plasma renin activity, and aldosterone levels in ponies" src="../images/dsv/Graphs/MicrobialColonicWaterPlasmaReninAldosterone%20F9_08.gif">

Figure 9.9. Relationship between colonic water exchange, plasma rennin activity, and aldosterone levels (+/- SE) in ponies fed a pelleted hay-grain diet at 12-hour intervals. (From Clarke et al. 1990a.) 

Omnivores can also require a minimal amount of dietary fiber for normal function of their large intestine. Burkitt (1971) showed a relationship between low-fiber, high-protein, high-fat diets and the higher incidence of colo-rectal cancer and other diseases of the human large intestine in affluent Western societies. Although the protective effects of cereal fiber against colo-rectal cancer were confirmed by numerous epidemiological studies (Jacobs 1988), the reasons are uncertain. Insoluble fiber (cellulose, lignin, and some hemicelluloses) can reduce the digesta retention time and increase the volume of digesta in the large intestine. Soluble forms of dietary fiber (pectins, gums and other hemicelluloses) can form gels that may sequester potential carcinogens, and all forms but lignin can be fermented to SCFA (Van Soest 1994). Wheat bran and cellulose had a protective effect against colonic cancer in rats, but more rapidly fermentable carbohydrates (corn bran, pectin, carrageenan, agar and metamucil) enhanced the development of chemically-induced tumors (Lupton 1995). A lack of enhancement in germ-free animals suggested that the end products of fermentation were the causative agents. 

Topping and Clifton (2001) pointed out that the diet of African populations that gave rise to the :fiber hypothesis” consume large amounts of starches that are resistant to digestion by endogenous enzymes and subject to microbial fermentation in the large intestine. Although butyrate stimulates the normal growth of colonic cells, it appears to inhibit an excessive rate of growth. Therefore, the resistant starches may inhibit carcinogenesis by providing a slow and steady source of SCFA to the distal colon, which shows the highest incidence of cancer in humans.

The low-fiber diets of affluent societies also tend to be high in protein and fat. Colonic cancer may be due to the toxic effect of high levels of ammonia produced by microbial metabolism of protein (Wrong 1988; Clinton et al. 1988; Hsi-Chiang Lin and Visek 1991a,b), and there is evidence that dietary fat promotes colonic carcinogenesis by influencing microbial metabolism of bile acids and cholesterol (Hill 1983). Therefore, the protective effects of a high-fiber, low-protein, low-fat diet may be due to an increase in the rate of digesta passage, a change in the rate of SCFA or ammonia production, or an increase in the population of microbes available for the destruction of carcinogens.