A resistant-starch enriched yogurt: fermentability, sensory characteristics, and a pilot study in children

The rising prevalence of obesity and the vulnerability of the pediatric age group have highlighted the critical need for a careful consideration of effective, safe, remedial and preventive dietary interventions. Amylose starch (RS2) from high-amylose maize (HAM) ferments in the gut and affects body weight. One hundred and ten children, of 7-8 (n=91) or 13-14 (n=19) years of age scored the sensory qualities of a yogurt supplemented with either HAM-RS2 or an amylopectin starch. The amylopectin starch yogurt was preferred to the HAM-RS2-enriched yogurt by 7-8 year old panelists ( P<0.0001). Appearance, taste, and sandiness scores given by 13- to 14-year-old panelists were more favorable for the amylopectin starch yogurt than for HAM-RS2-enriched yogurt ( P<0.05). HAM-RS2 supplementation resulted in acceptable (≥6 on a 1-9 scale) sensory and hedonic ratings of the yogurt in 74% of subjects. Four children consumed a HAM-RS2-enriched yogurt for four weeks to test its fermentability in a clinical trial. Three adolescents, but not the single pre-pubertal child, had reduced stool pH ( P=0.1) and increased stool short-chain fatty acids (SCFAs) ( P<0.05) including increased fecal acetate ( P=0.02), and butyrate ( P=0.089) from resistant starch (RS) fermentation and isobutyrate ( P=0.01) from protein fermentation post-treatment suggesting a favorable change to the gut microbiota. HAM-RS2 was not modified by pasteurization of the yogurt, and may be a palatable way to increase fiber intake and stimulate colonic fermentation in adolescents. Future studies are planned to determine the concentration of HAM-RS2 that offers the optimal safe and effective strategy to prevent excessive fat gain in children.

the sensory qualities of a yogurt supplemented with either HAM-RS2 or an amylopectin starch. The amylopectin starch yogurt was preferred to the HAM-RS2-enriched yogurt by 7-8 year old panelists ( 0.0001). Appearance, P< taste, and sandiness scores given by 13-to 14-year-old panelists were more favorable for the amylopectin starch yogurt than for HAM-RS2-enriched yogurt ( 0.05). HAM-RS2 supplementation resulted in acceptable (≥6 on a 1-9 scale) P< sensory and hedonic ratings of the yogurt in 74% of subjects. Four children consumed a HAM-RS2-enriched yogurt for four weeks to test its fermentability in a clinical trial. Three adolescents, but not the single pre-pubertal child, had reduced stool pH ( =0.1) and increased stool short-chain fatty acids (SCFAs) ( P 0.05) including increased fecal acetate ( =0.02), and butyrate ( =0.089) P< P P from resistant starch (RS) fermentation and isobutyrate ( =0.01) from protein P fermentation post-treatment suggesting a favorable change to the gut microbiota. HAM-RS2 was not modified by pasteurization of the yogurt, and may be a palatable way to increase fiber intake and stimulate colonic fermentation in adolescents. Future studies are planned to determine the concentration of HAM-RS2 that offers the optimal safe and effective strategy to prevent excessive fat gain in children. The rapidly-growing prevalence of obesity in adults and children requires urgent remedial measures to avert individual and societal health care crises 1 . Few adult treatment strategies exist 2 , treating children is more challenging, yet childhood obesity is a growing health concern 3 . Pediatric vulnerability severely limits the use of pharmacological or surgical interventions. Even dietary treatment with energy and nutrient restriction for weight reduction may be detrimental to growth. Dietary resistant starch (RS) supplementation in food may offer a therapeutic opportunity to attenuate excessive fat gain in infants and children by reducing the caloric density while improving dietary quality 1,2 .
RS are dietary carbohydrates that resist cooking processes and enzymatic digestion in the small intestine, are fermented by colonic microbiota and modify the gut flora 4,5 . The amount of RS in the human diet has progressively decreased with modern milling and food preparation methods. RS intake in medieval Europe was 50-100g/day 6 , it is estimated at 30-40g/day 7 in developing countries, and has dropped to 3-8g/day in developed countries 7-9 . It is unlikely that modern human society will return to a diet of coarsely ground grains and legumes high in RS. However, RS is now available as an ingredient that can be incorporated into breads, cereal products and baked goods that are acceptable to the US population.
Microbiota-derived enzymes are needed to digest complex plant polysaccharides 10 and RS-enriched diets increase butyrate-producing Clostridia in rodent feces 11 . A natural, granular, type 2 RS from high-amylose maize (HAM-RS2) decreases plasma cholesterol and triglycerides, increases satiety, increases insulin sensitivity 12-22 , and is anti-adipogenic in adult populations 23-30 . HAM-RS2 fermentation in the colon of rodents produces short chain fatty acids (SCFAs) such as acetate, propionate and butyrate that are absorbed through colonocytes, and change colonic microbiota composition 25,31,32 . Butyrate treatment increases gene expression of peptide tyrosine tyrosine (PYY) and proglucagon in ileal, primary colon and cecal epithelial cells of rats; elevates plasma Glucagon-like peptide-1 and -2 (GLP-1, GLP-2), and raises gene expression and protein production of Glucose transporter 2 (GLUT2) 27,31,33-35 . Clinical studies show that SCFAs increase in response to consumption of HAM-RS2 or RS from potatoes 36-38 , that the microbiota of rodents were modified 39 , and that butyrate was increased in rodents after dietary introduction of human feces 40 . Since yogurt can deliver dietary fibers to treat constipation in children 41 , the aim of the current work is to develop a palatable yogurt delivery vehicle for HAM-RS2 that will withstand pasteurization and demonstrate an increase in fecal pH and SCFAs in children and adolescents.

Methods
The RS yogurt manufacture Yogurt mixes were made by incorporating the starches individually into skim milk. In vitro testing HAM-RS2 30g/237ml yogurt was used for in vitro testing. Six samples were prepared, coded, and tested blindly with half subjected to pasteurization. A modified Englyst method was used to quantify glucose release 4 . Intact granular structure of the starch was evaluated using birefringence light microscopy.

Sensory study
The Institutional Review Board (IRB) granted an exemption #HE13-1 (January 16, 2013) from continued oversight for the sensory study conducted in two groups of children evaluating the two yogurts. Ratings of satisfaction with the appearance, color, aroma, taste, thickness, sandiness, and palatability of each type of yogurt were scored by 110 children without communication. Ninety-one children were 7-8 years old (younger) and 19 were 13-14 years old (elder). The younger children were more willing to volunteer for the sensory study than the elder.
Subjects with no dairy or starch-related allergies were recruited from The Louisiana State University Laboratory School and parental consent to participate was obtained along with the children's assent. Participants were given yogurt samples in 85g cups with a snap-on lid. Cups were coded with a random three-digit number. Disposable plastic spoons and napkins were provided to prevent contamination between samples. Prior to the sensory evaluation, the children were provided with a "warm-up" yogurt sample to avoid the "first sample effect" due to possible previous consumption of other food items, and a cup of drinking water was provided to rinse their palate between samplings. Two evaluation forms were used, one with a face scale for the younger panelists and the other with a preference rating form for the older panelists, and clearly explained to each age group. The younger panelists indicated their yogurt preference by circling "smiling face (  ) as yes", scored as 3, "neutral face (  ) as neither like nor dislike", scored as 2, or "sad face () as no", scored as 1. The elder panelists evaluated the yogurt on a 1-9 scale (1-dislike extremely, 2-dislike very much, 3-dislike moderately, 4-dislike slightly, 5-neither like nor dislike, 6-like slightly, 7-like moderately, 8-like very much, 9-like extremely) for appearance, color, aroma, taste, thickness, and sandiness. The elder panelists evaluated the yogurt thickness by checking 1-too thin, 2-just about right, or 3-too thick; and the sandiness as 1-not grainy, 2-just about right, or 3-too grainy. Elder panelists answered the question "Is this product acceptable?" with 2-"yes" or 1-"no" answer.

Clinical study
The four-week pilot clinical trial was approved by the PBRC Institutional Review Board (IRB28012) and registered (http://clinicaltrials.gov/, NCT01338571) to determine the effects of consuming HAM-RS2-enriched yogurt on fecal pH and fecal SCFAs, pre-and post-consumption, in a healthy child and three healthy adolescents. The subjects (a 6-year-old female, two 10-year-old African-American females, and a 14-year-old Caucasian male) were recruited through the PBRC recruiting department. Parents signed a consent form and subjects signed an assent form. Subjects with gastrointestinal disease, on medications with the potential to alter the intestinal bacterial microbiota such as antibiotics and subjects with allergies to corn were specifically excluded.
Subjects were weighed in the morning on an electronic scale in light street clothing without shoes or outer clothing and with pockets emptied. The electronic scale (Model 450, GSE Inc., Livonia, MI, USA) was calibrated daily using standardized weights and quarterly by an external service. Parents were given stool-collecting kits and instructed to collect a stool specimen from their child for 3 consecutive days at baseline and after 4 weeks of yogurt consumption. They were provided with ice packs, coolers and were instructed to return stool samples in the coolers to the research site on the day they were collected so they could be stored at -70°C until analysis.
Children were given HAM-RS2 10g plus 1g per year of age daily 42 which was 16, 20, or 24g for the four subjects. A fresh supply of yogurt was given to the parents weekly and the daily yogurt was divided into servings at breakfast and dinner.
Measurements of fecal SCFAs and pH were previously published elsewhere 18 . Briefly, the frozen fecal specimens were thawed, homogenized and further diluted to wet sample in distilled water (0.5g/5ml). The pH was measured using a combination electrode. Samples were then acidified with metaphosphoric acid (250g/L, 1ml) containing ethyl-butyric acid (2g/L) as an internal standard. The mixture was vortexed, and centrifuged at 4°C for 10 minutes at 8,000 rpm to remove solids in the homogenized samples and syringefiltered (33mm, Millipore, Billerica, MA). The filtrate was put into a gas chromatograph (GC) auto-sampler vial and capped. SCFAs in the effluent were analyzed using gas-liquid chromatography. The GC conditions (115°C for 0.1 min) were increased to 150°C for 0.1 min in increments of 10°C, then to 170°C for 2 min at increments of 11°C. The injector temperature was 250°C. Helium was the carrier gas with a flow rate of 60 ml/min and splitless injection was 60 ml/min. Single SCFAs were determined by retention time based on standards and the relative concentrations calculated based on the ratio of the peak areas of the sample to the internal standard.

Statistical analysis
The RS content differences of yogurt samples prepared with or without pasteurization were determined using the Student t-test (SAS 9.3, SAS Institute Inc., Cary, NC). The sensory data obtained from children were analyzed with a Randomized Block Design using panelists as blocks (GLM, SAS 9.1), and the paired t-test for HAM-RS2 score minus amylopectin score was performed.
Differences between the types of yogurt were determined by differences of least squares mean ± SEM. The clinical data analysis of feeding yogurt to the four subjects was performed with the Student t-test (weight) and paired t-test (change in weight) (SAS 9.1). Alpha was set at 0.05.

Ethics
All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000.

Resistant starch in yogurt
The glucose release was detected by a modified Englyst method. A light microscope (200x, Leitz Wetzlar, Ortholux II, Ernst Leitz GmbH, Wetzlar, Germany) revealed morphologies of the starch granules in the yogurts as having an equal presence of birefringence indicating an intact granular structure ( Figure 1). The RS content of the six yogurt samples varied minimally (from 45% to 51% on a dry weight basis with values of 51%, 45%, or 48% for the unpasteurized samples, and 45%, 45%, or 46% for the pasteurized samples (P>0.05).  The amylopectin starch yogurt was always judged as acceptable (Table 2) and its acceptability on a 1-2 scale (1-not acceptable or 2acceptable) was significantly higher than for the HAM-RS2-yogurt (P<0.05). The sensory study indicated that children preferred the amylopectin starch yogurt more than the HAM-RS2 added yogurt. Participants were asked to rate amylopectin starch-and HAM-RS2-containing yogurts for preference (7-8 years old; 1-3 scale, 1 being dislike, 3 being like) or appearance, color, aroma, taste, thickness, and sandiness (13-14 year olds; 1-9 scale, 1 being dislike extremely, 9 being like extremely) 45 .

Clinical study
All adolescent participants finished the HAM-RS2-yogurt and returned the empty containers during the weekly clinic visits with no complaints regarding taste or compliance issues related to consumption of the yogurt.  (2,387±645 to 1,889±120μg, P>0.05). The isobutyrate from protein fermentation increased (39%, 285±31 to 471±58μg, P=0.01) (Figure 2). The stool pH of the adolescents was mildly reduced at the end of the fourth week with a trend toward a lower pH (2.8%, from 7.2±0.4 to 7.0±0.35, P=0.1, Figure 3).
The pre-pubertal participant responded to HAM-RS2-enrichedyogurt differently than the three adolescent children with an increase in stool pH (from 6.89 to 7.62). The stool SCFAs were decreased; in ascending order, isobutyrate (35%, from 526 to 186μg), butyrate (39%, from 4,028 to 1,571μg), acetate (52%, from 8,328 to 4,336μg), and propionate (65%, from 2,870 to 1,877μg) over the 4-week study. Four participants (a 6-year-old female, two 10-year-old African-American females, and a 14-year-old Caucasian male) were weighed and their weights recorded. Parents were given stoolcollecting kits and instructed to collect a stool specimen from their child for 3 consecutive days at baseline and after 4 weeks of yogurt consumption. They were provided with ice packs, coolers and were instructed to return stool samples in the coolers to the research site on the day they were collected so they could be stored at -70°C until analysis. Children were given HAM-RS2 10g plus 1g per year of age daily which was 16, 20, or 24g for the four subjects. SCFA content of samples was analysed by GL-chromatography and determined by retention time based on standards and the relative concentrations calculated based on the ratio of the peak areas of the sample to the internal standard 46 .

Discussion
The prevalence of childhood obesity has increased 2-to 3-fold in just the last 25 years globally (see review 2). Childhood obesity is associated with co-morbidities similar to adults including hypertension, dysglycemia, dyslipidemia, inflammation and endothelial dysfunction 2 . Supplementing other treatment approaches with behavioral interventions may increase long term participation and is felt to be more important in the pediatric population than for adults 2 , but meta-analyses show prevention or treatment strategies to be ineffective. Currently, no truly effective pharmacological options are available for weight management, and surgery is restricted to a highly selected subgroup of very obese adolescent individuals. Medication and surgery have safety concerns in growing children and their efficacy is uncertain in the pediatric age group 43 . Novel treatments for childhood obesity offering safety, efficacy and acceptability are urgently needed 44 . Desirable attributes of an intervention for pediatric obesity include a preventive measure that attenuates excess fat accumulation while allowing for normal growth. RS is a natural food ingredient with a low risk profile that attenuates body fat accretion in experimental animal models, and is an excellent candidate to effectively combat childhood obesity. This feasibility study suggests that HAM-RS2-enriched foods likely alter microbiota composition, and this is supported by the increase in fecal SCFA content and lower pH. Yogurt was a generally acceptable vehicle for providing HAM-RS2. The yogurt cultures fermented lactose (milk sugar) and the RS granules in the final yogurt product were not damaged.
Enriching the diet with RS which has been refined out of the US diet will improve dietary quality and may help to ease the severity of pediatric obesity. Although the amylopectin starch yogurt was preferred, our studies confirmed the general acceptability of incorporating HAM-RS2 into yogurt through taste and sensory testing in 91 7-to 8-year-olds and 19 13-to 14-year-old volunteers. The four subjects in our pilot study that consumed the HAM-RS2enriched yogurt twice a day for weeks established the feasibility of feeding the HAM-RS2-enriched yogurt to children. We demonstrated a trend toward a reduction of pH and documented a significant increase in the SCFA content of the stools of the adolescent  children. This agrees with previous studies that have found that adding HAM-RS2 to rodent diets reduced abdominal fat in association with increased fermentation 20,21,37,44 . Supplementing the diet with RS will need to be acceptable and palatable or children are likely to reject it in favor of low-fiber alternatives. Overall, the HAM-RS2-yogurt in the taste testing was acceptable to 74% of the children in the 13-to 14-year-old group, but 24% less acceptable in younger children. The knowledge that RS is healthy may increase the adoption of RS fortified foods, such as yogurt.
People eat for volume and consume fewer calories when food has a lower caloric density 38 . RS and other dietary fibers reduce the caloric density of food 12,13,44 . RS is present in many different sources, which offers the opportunity to choose the RS with the greatest success in reducing or controlling body weight 17 . We have previously shown that the HAM-RS2 supplementation produces a 30% reduction in intestinal fat deposition in wild type C. elegans 28 and in rodents the same also reduced body fat 25,44 . Longer-term controlled studies are needed to determine if the reduced adiposity seen in animal models will occur in human populations. In a human pilot study, a HAM-RS2 (15g/day) supplemented diet enhanced insulin sensitivity by 56.5% in men over two to three months, which suggests that lower amounts of HAM-RS2 may also be efficacious. Beneficial changes in adiposity may occur over longer treatment periods 12,13,21,22 , and lower amounts of RS may further improve palatability -an important factor for long-term consumption.
Children maintain weight loss better than adults 11 . Although it is not clear why the pre-pubertal child in this study did not respond in the same way as the adolescents, it could represent differences in her intestinal microbiota or her pre-treatment diet which was not controlled nor queried. Further research will be necessary to explore the differential role of diet and the intestinal microbiota on the fermentation of RS before puberty. Weight gain in all of the children during the 4-week study may reflect the fact that they were growing.

Conclusion
The current study showed the acceptability and feasibility of using yogurt to deliver RS to adolescents which caused a change in SCFA and probably changed the gut microbiota. These preliminary data suggest the need to evaluate differences that may exist in the microbiota before and after puberty to determine whether the non-response of the pre-pubertal child represented an outlier or a real effect in pre-pubertal children. These preliminary results will need confirmation in a controlled trial so that the effects of growth can be taken into account in evaluating weight changes in longer-term studies using yogurt as a vehicle to deliver the functional food component HAM-RS2 in a range of doses into everyday foods that consumers enjoy. Our data encourage controlled studies in children and adolescents testing insulin sensitivity, effects on body weight, and potential differences between pre-pubertal and adolescent children in their microbiota response to RS. Hopefully, increased consumption of reduced-calorie foods in combination with increased physical activity 12 will reduce weight gain, help to maintain a healthier weight, and lead to future improvements in public health for adults and adolescents alike.

Consent
Informed consent was obtained from all patients being included in the study.

Author contributions
Each individual has contributed to this manuscript as a qualified author and meets ALL of the requirements following the American Medical Association (AMA) manual guidelines. KJ Aryana, C Pelkman, and D Olson contributed to the conception and design of sensory study, acquisition of data, analysis and interpretation of the data, and drafting or editing the manuscript. R Tulley performed stool test and the SCFA analysis. FL Greenway and NV Dhurandhar designed the clinical study, contributed to its conception, acquisition of data, analysis and interpretation of the data, and drafting the manuscript. JW Finley, MJ Keenan, RJ Martin, and J Zheng contributed to designing of the sensory study, data analysis and interpretation, and drafting or editing the manuscript. All authors have approved the final version of the manuscript.   F1000Research they have found some important information about a natural, granular, type 2RS from high-amylose maize (HAM-RS2). Others working in this field can benefit from the fact that they presented their data in what could be characterized as a preliminary study. However, there are two points that bother me about the interpretation of their data. First, they put a very positive spin on the HAM-RS2 intervention, when by almost all accounts this was not a positive outcome. This really should be toned down. Second, the authors make statements about the benefits of fermentability that come across as "facts" when rather this is an open and unresolved issue. For example, when discussing HAM-RS2 they say "decreases plasma cholesterol and triglycerides, increases satiety, increases insulin sensitivity and is anti-adipogenic in adult populations." Although they show several citations that have shown this, they have not reviewed the entire literature, and they make the statement as if it is fact. This needs to be qualified. In a different place they say "People eat for volume and consume fewer calories when food has a lower caloric density." This is Barbara Rolls's hypothesis, but not everyone would agree with this statement. In fact, Rolls herself says it mostly applies to men, not women. So, please go back through your comments that are stated as FACTS and perhaps modify them.

I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. There is much current attention to the role of the gut, especially gut microbiota, in the regulation of body weight. At the same time the US and other countries face escalating rates of childhood and adolescent obesity, suggesting the importance of obesity prevention efforts, of which there have been few shown to have significant impacts. This study examines one innovative potential approach to preventing obesity targeting food components, specifically the incorporation of dietary resistant starch into a common food item, yogurt. The study focused on the acceptability of the sensory characteristics of yogurt enhanced with one such resistant starch, compared to yogurt enhanced with a control amylopectin starch, among pre-adolescent and adolescent subjects. A small pilot 4-week trial also looked at the effects of the resistant starch when consumed by 4 subjects.
The sensory evaluation study was well-designed with tight controls and a sizable N. Assessment of various sensory attributes was rather comprehensive among the adolescent sample, which unfortunately was substantially smaller than the sample of 7-to 8-year olds who could not be expected to make the finer sensory distinctions on which the adolescents were queried. The actual resistant starch content of the various yogurt samples was verified by analyses.
The sensory evaluation study showed that the subjects consistently preferred the control yogurt to that containing the resistant starch. The adolescents found the resistant starch yogurts to be inferior to the control yogurts on appearance, taste, thickness, sandiness and acceptability. About 25% of the adolescent subjects rated the resistant starch as unacceptable. Among the younger children, whose only ratings were a smiling face, neutral face or sad/frowning face, the resistant starch yogurt was more strongly disliked, with nearly two-thirds of subjects awarding it a frowning face. The discussion does not strongly disliked, with nearly two-thirds of subjects awarding it a frowning face. The discussion does not adequately address these results, which show consistently that the sensory characteristics of the resistant starch yogurt preparation are not well received by children and adolescents. The conclusion that "…our studies confirmed the general acceptability of incorporating HAM-RS2 into yogurt…" seems to fly in the face of these results.
The very small pilot clinical trial provided some support for the hypothesis that the resistant starch would produce changes in gut microbiota, findings worth following up with a larger study. It also showed that despite the limited acceptability ratings, the products could be consumed over a 4-week period. However, the weight gain seen among all four subjects is disturbing. Although the discussion dismisses this finding by saying that it may reflect growth, the amounts of weight gain seen over 4 weeks were 0.4, 1.7, 1.9 and 3.2 kg. Certainly the latter three gains, if continued over a year, would be quite excessive. Given that the point of this dietary intervention is to prevent excessive weight gain, these preliminary findings are cause for considerable concern.
The authors have conducted a well-designed study of the feasibility of this innovative dietary intervention which has the purpose of obesity prevention. The findings show that it is possible to get children and adolescents to consume yogurt containing resistant starches, but more development is necessary to produce a food product with adequate acceptability. Aside from that, however, the assumption that this product, if consumed regularly, might avert excess weight gain needs further study, as the limited results here suggest its effect may be just the opposite of that which was expected and desired.

I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. Competing Interests: