Nutrition bars that claim to have high amounts of protein but very low levels of net or available carbohydrate, (the carbohydrate left over after dietary fiber and sugar alcohols are subtracted out) have become wildly popular in certain segments of the fitness community. A key ingredient in some of these bars is the carbohydrate isomalto-oligosaccharide (IMO), which is promoted as a prebiotic dietary fiber with a light sweetness profile that has functional properties (moisture retention, low viscosity) well-suited for nutrition bars. Bars made with IMO can include over 15g of this ingredient, leading to claims of high dietary fiber. Consumers see this as a win. They have a nutrition bar that has a low level of net carbohydrate and relatively low calories with the added benefit of a prebiotic dietary fiber to go along with the desired protein component. Plus, IMO’s do occur naturally in some foods such as soy sauce, miso, and honey (Goffin et al., 2011), making it appealing to consumers as a “natural” ingredient. So far, what’s not to like about IMO? A deeper dive into the scientific research shows that IMO, while still having potential health benefits, might not live up to its sterling reputation as a calorie-free, guilt-free dietary fiber.
First of all, what is IMO? To understand what IMO is, it helps to first understand the structure of typical food starches and how they are digested in the body. Starches from corn and wheat, for example, consist of long chains of glucose molecules (polymers) linked together with α-1,4 chemical bonds. In the initial phase of starch digestion in the small intestine, the enzyme α-amylase (salivary and pancreatic) breaks these glucose polymers into much shorter chains, called oligosaccharides, that can have anywhere from 2 to maybe 10 glucose units. Then, the combined action of α-amylase and disaccharidase enzymes present in the brush border of the small intestine (e.g., sucrase-isomaltase and maltase-glucoamylase) reduces these oligosaccharides down to individual glucose units that are subsequently absorbed. One common disaccharide that results from the initial phase of starch digestion is maltose -- two glucose units linked together by an α-1,4 chemical bond, which is later hydrolyzed by the maltase-glucoamylase enzyme. Related to maltose is isomaltose, which is found in honey and, like maltose, in beer. Isomaltose, like maltose, is made up of two glucose units, but the two are linked via an α-1,6 versus α-1,4 bond. This shifts the primary digestion of isomaltose to the sucrase-isomaltase enzyme. This is an important consideration because the digestion of α-1,6 bonds in the small intestine is considerably slower than for α-1,4 bonds and is sometimes incomplete. The isomaltose backbone forms the foundation for several different IMOs (see table).
This brings us to IMO syrups that are used in food products like nutrition bars. An IMO syrup will contain a blend of di- and oligosaccharides that usually have one or more α-1,6 chemical bonds between monosaccharide units. These syrups can also contain small amounts of glucose and maltose (Kohmoto, et al., 1992). Thus, it is a misconception that a commercial IMO syrup contains just a fixed single molecule of one particular isomalto-oligosaccharide of a given chain length and bond structure.
Sugars commonly found in IMO syrups
|Sugar ||Chemical structure |
|Isomaltose ||α-D-glucose-(1→6)-α-D-glucose |
|Isomaltulose ||α-D-glucose-(1→6)-α-D-fructose |
|Panose ||α-D-glucose-(1→6)-α-D-glucose-(1→4)-D-glucose |
|Isomaltotriose ||α-D-glucose-(1→6)-α-D-glucose-(1→6)-D-glucose |
|Kojibiose (trehalose) ||α-D-glucose-(1→2)-D-glucose |
|Nigerose ||α-D-glucose-(1→3)-D-glucose |
|Glucose-maltotriose ||α-D-glucose-(1→6)-(α→D-glucose-(1→4))2-D-glucose |
|Glucose-panose ||α-D-glucose-(1→6)-α-D-glucose-(1→6)-α-D-glucose-(1→4)-D-glucose |
|Isomaltotetraose ||α-D-glucose-(1→6)-(α-D-glucose-(1→6))2-D-glucose |
Source: Goffin et al. (2011)
So, is IMO syrup natural? Is it a completely non-digestible dietary fiber source? Is IMO calorie-free, as consumers might assume when food manufacturers claim it as a dietary fiber? Is IMO syrup a prebiotic fiber, as is claimed?
Question 1: Is IMO syrup really “natural”? Although IMO’s do occur naturally in some foods, it is not economically feasible to extract IMO from foods on a large scale. As such, commercially-available IMO syrups are enzymatically synthesized from starch (Goffin et al., 2011). This process converts starch into smaller chains of glucose units (typically 2-4 glucose units) that are linked together mainly via alpha-1,6 and other chemical bonds. Thus, it is not correct to say that the IMO syrup added to foods is natural.
Question 2: Is IMO syrup truly a completely nondigestible dietary fiber, as claimed? Essentially, no. Early studies using an in vitro digestive modeling system and human salivary or hog pancreatic alpha-amylase showed a lack of hydrolysis of IMO, suggesting that it may be a dietary fiber that would largely resist hydrolysis in the small intestine and reach the colon intact (Goffin et al., 2011). However, the small intestinal brush border contains the sucrase-isomaltase enzyme that is capable of breaking down alpha-1,6-glucosidic bonds. Repeated studies in the in vitro system with rat intestinal mucosa or isolated studies with an in situ rat jejunal loop method found that IMO was digested to some degree (Kaneko et al., 1992; Kaneko et al., 1995). Kohmoto et al. (1992) conducted human studies on 6 fasted subjects who ingested 25 g IMO syrup and 50 mg 13C-IMO (an isotopically-labeled form of IMO used to determine its metabolic fate). Under resting conditions, mean serum glucose levels increased from 109 mg/dL pre-ingestion to a peak of 136 mg/dL at 30 min postingestion. Serum insulin also rose in parallel with glucose, reaching a peak of 32 μU/mL at 30 min postingestion, up from 4.8 μU/mL pre-ingestion. Breath 13CO2 recovery (an index of the metabolism of the labeled IMO to carbon dioxide) for 8 h after IMO ingestion was 83% that of an equivalent dose of maltose (a fully digestible carbohydrate). Subjects also performed similar testing after a brief period of exercise and the breath 13CO2 recovery was 69% maltose. These data indicate that IMO was about 83% as digestible as maltose under resting conditions and about 69% as digestible after the exercise period. Thus, the majority of the carbohydrate in the IMO syrup was, in fact, digested, absorbed, and metabolized.
The authors of this paper also tested breath hydrogen excretion in these subjects after IMO ingestion. When the bacteria in the large intestine ferment carbohydrates, hydrogen gas is produced, absorbed into the blood, and ultimately excreted via the lungs. Therefore, a rise in breath hydrogen indicates some degree of carbohydrate malabsorption. There was a small increase in breath hydrogen in the first 2 h postingestion, suggesting that a minor portion of the IMO did reach the colon and was fermented. No IMO was detected in feces from the subjects, so the fermentation of any malabsorbed IMO was essentially complete.
In a similar human study of 8-13 subjects, 20 g doses of fructo-oligosaccharide (FOS) and galactosyl-sucrose (GS) (each nondigestible oligosaccharides) were compared with 20 g IMO with regard to their ability to increase breath hydrogen excretion (Oku and Nakamura, 2003). The breath hydrogen areas under the curve for GS and FOS were 8.5 and 14.2 times larger than that of the IMO, respectively. Thus, while the breath hydrogen response to IMO was quantifiable, it was nowhere near as large as the breath hydrogen response to GS and FOS. In addition, neither a 10- nor 20-g dose of IMO caused appreciable gastrointestinal intolerance symptoms. This again indicates that the vast majority of the carbohydrate in IMO is, in fact, digestible.
Question 3: Is IMO calorie-free? No, it is not calorie-free. Kohmoto et al. (1992) used their breath 13CO2 recovery data showing that IMO is 69-83% as digestible as maltose to estimate IMO’s caloric value. Maltose has an energy density of 3.947 kcal/g, so 69 to 83% of that number results in estimated energy density of IMO of 2.7 to 3.3 kcal/g. Thus, it would be wrong to assume that the carbohydrates from IMO don’t count as available carbohydrates to the body simply because IMO is often referred to as a dietary fiber. It is also misleading for manufacturers to list the entire carbohydrate content of IMO as dietary fiber, since these studies clearly show that this is not the case. And, finally, consumers should not assume that all dietary fibers are calorie-free since they are not digested in the small intestine. The reason for this is that some dietary fibers are fermented by colonic bacteria, producing short-chain fatty acids that can be absorbed into the blood from the colon and used for energy by other tissues. As a general rule of thumb, highly fermentable dietary fibers provide about 2 kcal/g due to the energy utilization of the end products of their colonic fermentation.
Question 4: Is IMO a prebiotic? Possibly. Despite the fact that most IMO is digested before it reaches the colon, a portion of IMO carbohydrate is fermented by colonic bacteria. The fermentation of some types of carbohydrate selectively increases the populations of healthy bacteria, such as bifidobacteria. This type of selective favoring of the growth of healthy bacteria by a particular dietary fiber is known as a prebiotic effect. A couple of studies have shown that feeding doses of IMO ranging from 5-20 g/d for 12-14 d significantly increased fecal bifidobacteria concentrations (Kaneko et al., 1994; Kohmoto et al., 1988). However, another study (Bouhnik et al., 2004) reported no bifidogenic effects of the feeding of 10 g IMO per day for 7 days. More studies are needed to determine the potential prebiotic effects of IMO.
IMO syrups used in food products consist of a variety of isomalto-oligosaccharides and other sugars. Most of these carbohydrates are fully digested, while others are not. As such, it is estimated that the caloric density of IMO syrup is 2.7-3.3 kcal/g, about 25% less than for maltose on average. As such, it is an error to list the entire contents of IMO as dietary fiber on the labels of food products. A small fraction of the carbohydrates in IMO syrups does escape digestion and undergoes colonic bacterial fermentation. Those IMO carbohydrates escaping digestion may result in potential prebiotic effects, but more studies are needed to confirm this.
Bouhnik Y, Raskine L, Simoneau G, Vicaut E, Neut C, Flourie B, Brouns F, Bornet F. The capacity of nondigestible carbohydrates to stimulate fecal bifidobacteria in healthy humans: a double-blind, randomized, placebo-controlled, parallel-group, dose-response relation study. Am J Clin Nutr 2004;80:1658-1664.
Goffin D, Delzenne N, Blecker C, Hanon E, Deroanne C, Paquot M. Will isomalto-oligosaccharides, a well-established functional food in Asia, break through the European and American market? The status of knowledge on these prebiotics. Crit Rev Food Sci Nutr 2011;51:394-409.
Kaneko T, Kohmoto T, Kikuchi H, Fukui F, Shiota M, Yatake T, Takaku H, Iino H. Digestibility of isomaltooligosaccharides by rats and effects on serum lipids. Nippon Nogeikagaku Kaishi 1992;66:1211-1220.
Kaneko T, Kohmoto T, Kikuchi H, Shiota M, Iino H, Mitsuoka T. Effects of isomaltooligosaccharides with different degrees of polymerization on human fecal bifidobacteria. Biosci Biotech Biochem 1994;58:2288-2290.
Kaneko T, Yokoyama A, Suzuki M. Digestibility characteristics of isomaltooligosaccharides in comparison with several saccharides using the rat jejunum loop method. Biosci Biotech Biochem 1995;59:1190-1194.
Kohmoto T, Fukui F, Takaku H, Machida Y, Arai M, Mitsuoka T. Effect of isomalto-oligosaccharides on human fecal flora. Bifidobacteria Microflora 1988;7:61-69.
Kohmoto T, Tsuji K, Kaneko T, Shiota M, Fukui F, Takaku H, Nakagawa Y, Ichikawa T, Kobayashi S. Metabolism of 13C-isomaltooligosaccharides in healthy men. Biosci Biotech Biochem 1992;56:937-940.
Oku T, Nakamura S. Comparison of digestibility and breath hydrogen gas excretion of fructo-oligosaccharide, galactosyl-sucrose, and isomalto-oligosaccharide in healthy human subjects. Eur J Clin Nutr 2003;57:1150-1156.