Introduction
Although a great deal of research was stimulated throughout the world by the Burkitt and Trowell's hypothesis (84), it is still early to assign clear health claims to dietary fibre. This difficulty is due in great part to the fact that dietary fibre includes many complex substances, each having unique chemical structure and physical properties. Moreover, dietary fibre is often intimately associated in the plant cell structure with other organic compounds, such as vitamins, phyto-oestrogens, flavonoids, etc., displaying their own biological activity. Nevertheless, numerous prospective and well-designed experimental studies have highlighted several physiological and metabolic effects of dietary fibre which may be important for human health.
Digestive fate of dietary fibre
It is now well-established that dietary fibre reaches the large intestine and is fermented by the colonic microflora with the production of short chain fatty acids (SCFA), hydrogen, carbon dioxide and biomass. This fermentative process dominates human large bowel function and provides a means whereby energy is obtained from carbohydrates not digested in the small bowel, through absorption of SCFA.
Fermentation of fibre in the colon
Polysaccharides cannot penetrate in the bacterial cells. They are first hydrolysed in monosaccharides, by membranous or extra-cellular enzymes secreted by bacteria. Metabolism of these monomeric sugars continue in the bacterial cells using the Embden-Meyerhoff pathway which leads to pyruvate. Pyruvate does not appear in the large bowel because it is immediately converted in end-products. These are SCFA, mainly acetate, propionate and butyrate, and gases: carbon dioxide (CO2), hydrogen (H2), and methane (CH4).
Colonic fermentation is an efficient digestive process since starch is almost totally degraded, as well as lactose, alcohol-sugars and fructans if the intake of these sugars is not too high. More than half of the usually consumed fibres are degraded in the large intestine, the rest being excreted in the stool (see Table 9). A number of factors are likely to affect the utilization of fermentable carbohydrates in the colon. Among these is solubility. The more soluble substrates, being more accessible to hydrolytic enzymes, are likely to be degraded more rapidly. Nevertheless, some soluble fibres such as alginates or carragheenans are poorly fermented. Other factors involving digestive motility and individual differences in microflora could also modulate fermentation. Furthermore, certain metabolic pathways can be modified by the repeated occurrence of some sugars (lactose, lactulose, fructans) in the colon. The mechanisms and the physiological consequences of this adaptation are not completely identified.
TABLE 9 Colonic fermentability of dietary fibres in humans
|
Dietary fibre |
Fermentability (%) |
|
Cellulose |
20-80 |
|
Hemicelluloses |
60-90 |
|
Pectins |
100 |
|
Guar gum |
100 |
|
Ispaghula |
55 |
|
Wheat bran |
50 |
|
Resistant starch |
100 |
|
Inulin, oligosaccharides |
100 (if they are not in excess) |
Absorption and metabolism of end-products
Reducing the rate of digestion of carbohydrate spreads the absorption of carbohydrate along a longer portion of the small intestine (159,160), and tends to increase the amount of carbohydrate which escapes digestion in the small intestine (161). For example, the amount of carbohydrate from lentils entering the colon is 2.5 times as great as carbohydrate from bread. Increasing the delivery of starch to the colon has many implications which include those on the health of the colon itself and on systemic metabolism. It is believed that starch entering the colon is completely and rapidly fermented, mostly in the cecum (162). The fermentation of starch produces relatively more butyrate than the fermentation of dietary fibre (162), and resistant starch produces somewhat different fermentation products than readily digested starch (163).
A part of the products of fermentation are utilized by bacteria yielding energy and carbon necessary for synthesis and growth of the flora. Another part is eliminated in the stool or rectal gases, but the major part is absorbed by the colonic mucosa. Absorption of SCFA is rapid and leads to accumulation of bicarbonates and increase of pH in the lumen. Butyrate is considered to be the primary nutrient for the epithelial cells lining the colon (164), and SCFA stimulate proliferation of colonic epithelial cells and growth of the colon in general (165). Butyrate is the preferred substrate of colonocytes. SCFA which are not metabolized in the mucosa are oxidized in the liver, a part of acetate being also metabolised in the peripheric tissues.
Only a fraction of gases produced during fermentation is available for absorption. Hydrogen and methane are excreted in the breath gases. A large part of gases are consumed in the colonic lumen by 'bacteria. Acetogenic bacteria produce acetate from CO2 and H2. Methanogenic bacteria produce CH4 by reduction of CO2 with H2. Finally, sulfate reducing bacteria utilise H2 to reduce sulfates and produce sulfites or hydrogen sulfide. Unused gases are excreted through the anus.
Effects of dietary fibre on gut microflora
The composition of microflora appears to be influenced to some degree by diet, age and geographic considerations, but these factors are not thought to be particularly significant, at least as far as the commonly studied bacterial groups are concerned. Recent studies have shown, however, that the ingestion of certain oligosaccharides, such as fructo-oligosaccharides, could modify bacterial composition of the dominant flora by increasing bifidobacteria. Some studies suggest that these bifidobacteria, which are saccharolytic bacteria naturally occurring in the normal colonic flora, might be beneficial to host health. At the present time, however, this has not been conclusively established.
Ingestion of fructo-oligosaccharides have increased faecal counts of endogenous bifidobacteria by a factor of 10, without changing the total anaerobes concentration (85). The similarity of effects of chemically different substrates is likely due to the capacity of bifidobacteria to hydrolyze all these substrates and to metabolize the produced monomeric sugars (glucose, galactose, fructose). The exact mechanisms whereby only some substrates could stimulate preferentially the growth of bifidobactria are not known. A recent in vitro study suggested that the polymerization degree could be more determinant than the chemical nature of oligosaccharides (86). The metabolic consequences of the changes in faecal flora composition are unknown. Ingestion of oligosaccharides had no effect on stool weight and pH.
Effects of dietary fibre on gut function
In the gastrointestinal tract, some fibres form a matrix with fibrous characteristics. That is, some fibres, because of their ability to swell within the aqueous medium, can trap water and nutrients, especially water-soluble ones such as sugars. The physical characteristics of the gastric and small intestinal contents are altered by fibre sources. The bulk or amount of material in the gastrointestinal tract is greater because fibre is not digestible and hence remains during the transit of digesta through the small intestine. The volume increase is due to the water-holding capacity of certain fibres. The viscosity of the intestinal contents increases due to the presence of fibre sources containing viscous polysaccharides.
The changes in the physical characteristics of the intestinal contents may influence gastric emptying, dilute enzymes and absorbable compounds in the gut, prevent starch from hydrolyzing, and slow the diffusion or mobility of enzymes, substrates and nutrients to the absorptive surface. These effects result in the slower appearance of nutrients such as glucose and some lipid molecules in the plasma following a meal.
The effects of purified dietary fibres on bowel function may or may not be similar to those of intact fibres in whole foods. This is presumably due, at least in part, to interactions between fibre and starch, and the presence of fibre associated substances such as phytate and lectins which are present in the whole food. This makes it very difficult to make valid generalizations about the physiologic effects of fibre based simply on fibre analysis. For example, when considering the effect of fibre on postprandial blood glucose responses, purified viscous fibres have been found to produce a significant reduction in glycemic response in 33 of 50 studies (66%) reviewed in 1992, compared to only 3 of 14 (21%) studies with insoluble fibre (166). The effects of purified fibres appear to be directly related to their viscosity (167,168). This would suggest that the blood glucose responses of foods should be more closely related to their soluble than insoluble fibre content, however the opposite is the case. For 52 foods, the food glycemic index (as the indicator of rise in blood glucose) was weakly related to the amount of total fibre per 50g carbohydrate, and insoluble fibre explained a larger proportion of the variance in glycemic index, 17%, than soluble fibre, 9% (169).
Effects on carbohydrate digestion and absorption
Gastric emptying
Dietary fibres may affect gastric emptying in several ways (87). First, they may slow gastric filling, due to their bulking and energetic dilution capacity, which might in turn slow gastric emptying. Secondly, when certain soluble fibres are mixed in liquid meals or in liquid/solid meals, they delay emptying of gastric liquids by increasing viscosity of gastric contents. Such an increase in the viscosity of chyma could also slow the gastric emptying of solid components of the meal. On this issue, results are very controversial. Moreover, by acting as an emulsifier, viscous fibre can stabilize the gastric chyma and prevent separation of the solid from the liquid phase, impairing selective retention of the largest particles, and thereby increasing their rate of passage into the small intestine. Besides the effects of soluble fibres, insoluble fractions may also alter gastric emptying by mechanisms depending on their water retention capacity or size of particles.
Enzyme-substrate interaction
Available evidence suggests that fibre has little, if any, direct acute effect on the secretory function of the exocrine pancreas suggesting that the primary effect of fibre on carbohydrate digestion is exerted in the intestinal lumen. In the lumen, enzymes and substrates may be diluted with the addition of non-digestible material. Evidence from in vitro studies and from duodenal aspirates suggest that most of the tested fibres can alter the activity of pancreatic amylase (88). The inhibitory effects of fibre on pancreatic enzyme activities have been attributed to various factors including pH changes, ion-exchange properties, enzyme inhibitors and adsorption. Rather than a chemical enzyme-fibre interaction, the presence of fibre, through its particulate or viscous nature, probably impedes enzyme-substrate interaction.
The presence of fibre in a form that restricts starch gelatinization or access of the hydrolytic enzymes to starch can slow the rate of digestion of the starch. For instance, the slow rate of digestion of legumes may be related to the entrapment of starch in fibrous thick-walled cells, which prevents its complete swelling during cooking. In addition, resistance of starch to pancreatic hydrolysis may result from the presence of intact cell walls, which survive processing and cooking and insulate starch in such a manner that portions of it cannot be digested or absorbed.
Small intestinal motility
There is evidence that viscous fibres can influence accessibility of available carbohydrates to the mucosal surface and slow their absorption. One of the major mechanisms of this action is related to the effects of dietary fibre on small intestinal motility (89). Small intestinal contractions create turbulences and convective currents which cause fluid circulation and mixing of luminal contents. These movements allow glucose to be brought from the centre of the lumen close to the epithelium. When it reaches proximity to the epithelium, glucose must then diffuse across the unstirred water layer (UWL). This layer is created by a gradient of progressively poorer stirring as the mucosa is approached and forms an aqueous diffusion barrier separating mixed bulk luminal contents from the brush border. Thickness of the UWL depends on small intestinal contractions and is inversely related to the magnitude of the stir rate. When there is no contraction, fluid moves through the small intestine with laminar flow comparable to that occurring in a pipe. In this flow, there is no movement in the radial direction (from the centre of the lumen toward the epithelium), and consequently the stirring is very poor and the UWL very thick. On the contrary, normal motility generates both longitudinal and radial convection currents, hence creating turbulences and stirring of luminal fluid. Beside the effects of mixing contractions on glucose movement, small intestinal motility may alter absorption by influencing transit rate which determines area and time of contact between glucose and the epithelium.
Dietary fibres which alter small intestinal motility could thus influence glucose absorption by this mechanism. Viscous fibres, such as guar gum, stimulate motility but decrease transit rate, because they resist propulsive contractions. However, though guar gum slows transit it does not affect the distribution of glucose in the human upper small intestine. It is thus unlikely that guar gum delays glucose absorption by reducing contact area. As they resist propulsion, viscous fibres should similarly resist mixing contractions, hence inhibiting the effects of motility on fluid stirring. This is probably the mechanism by which they increase thickness of the UWL, and diminish passage of glucose across the epithelium.
Effects of dietary fibre on large bowel function
The major effects of dietary fibre occur in the colon. Here each type of dietary fibre interacts with the microflora, and the colonic mucosa and muscle to produce several possible effects. The actions of an individual fibre source depends to a large extent on its fermentability. The range of fermentability of different fibre is great and difficult to predict. Dietary fibre, however, can be roughly divided into those which are rapidly fermented, such as oligosaccharides, those which are more slowly fermented, such as gums, and those which are hardly fermented at all, such as wheat bran. The least fermentable fibres are the most likely to increase stool output. Dietary fibre which is highly fermentable is unlikely to have much effect on stool output but will affect bacterial fermentation products in the proximal colon and hence colonic and systemic physiology. Fibres which are slowly fermented may have a major influence in the distal colon even if they do not increase stool output significantly. Furthermore, the effect of each type of fibre is determined by dose.
Stool output
The dietary fibres which have the greatest effects on stool output are in general the least fermentable (90) These fibres probably act by virtue of their water holding capacity (WHC). The relationship between WHC and stool output is not simple. Dietary fibres with high WHC are those which are the most fermentable and are lost before they reach the rectum. There are exceptions such as ispaghula which has high WHC but resists fermentation. Moreover, one of the most reliable stool bulkers is wheat bran which has a WHC that is as low as the rest of faecal contents on a normal low fibre diet. It appears that the most important factor for a large effect on stool output is simply for the fibre to appear in stool. The effect is then dependent on the amount of fibre present as well as its residual WHC. The contribution of bacterial cells to faecal mass should not be forgotten, as the water content of bacteria is high. The effects of fibre are not restricted to increasing output. Dietary fibre has also a role in changing the consistency of the stool by increasing the water content and the plasticity, and increasing stool frequency.
Colonic motility and transit time
Certain fibres are known to have a laxative effect, in that their presence in the colon affects its motility and modifies colonic transit time. Two major mechanisms to explain this effect depend on the physicochemical properties and fermentative fate of fibre (91). These mechanisms refer to stimulation by the bulking effect of fibre as well as changes in the contractile activity and secretion of the colon (see Figure 6).
Increasing the volume of colonic contents distends the colon wall and stimulates propulsion of digesta through the activation of intramuscular mechanoreceptors. Dietary fibre can increase the faecal bulk by several mechanisms. First, the volume occupied by undegraded fibres adds to the volume of the rest of contents. This explains why the least fermentable fibres, such as wheat and corn bran, ispaghula or some algal polysaccharides, are particularly efficient laxatives. Also, these residues can trap water within their matrix, thus leading to a greater bulk. A third possible mechanism to increase intraluminal volume and stretch colonic muscle is the production of gases occurring during the fermentation of fibre.
Figure 6 Mechanisms of action of dietary fibre on colonic transit time
Source: Adapted from Salvador, V., Cherbut, C. (91)
In addition to their bulking effects, dietary fibre can reduce transit time by modulating contractile activity and water movements in the colon. Here again, they can act in several ways. First, the edges of solid particles can stimulate mechanoreceptors located in the submucosa and by that, modify the contractile pattern of the colon in favour of a greater propulsion of digesta, as has been shown with plastic particles. Fibre could also release compounds trapped in the small intestine (such as biliary salts or fatty acids) into the colon during fermentation. Such compounds have been shown to stimulate secretion and rectosigmoid motility.
Finally, a large part of fibre is fermented by microflora yielding several metabolites which can themselves influence colonic motility (92). For instance, SCFA stimulate contractions in the terminal ileum of humans and may also affect colonic motility as has been demonstrated with rats.
It has recently been appreciated that dietary starch bulks the stool (143,144,145), presumably because undigested starch provides energy for colonic bacterial growth. Thus, some of the faecal bulking effect of dietary fibre, at least in intact foods, could be due to the associated increase in starch delivery to the colon.
Vitamin and mineral absorption
Purified dietary fibres may reduce acutely the absorption of some vitamins and minerals by binding or entrapping them in the small intestinal lumen, however, there is little evidence that population groups consuming nutritionally adequate diets rich in high fibre foods, such as vegetarians, have any problems with vitamin or mineral deficiencies (170,171). Recent studies with calcium suggest that purified fibres reduce calcium availability in the small intestine, but that at least some of the calcium carried into the colon, bound to or entrapped by fibre, is released when the fibre is fermented (172) with the short chain fatty acid products of digestion facilitating calcium absorption from the distal colon and rectum (173).
