obese adults in a 12-week randomized controlled trial

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

YCLNU2717_proof ■ 15 January 2016 ■ 1/8

Clinical Nutrition xxx (2016) 1e8

Contents lists available at ScienceDirect

Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu

Randomized control trials

Q6 Q5

Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/obese adults in a 12-week randomized controlled trial Jennifer E. Lambert a, Jill A. Parnell b, Jasmine M. Tunnicliffe a, Jay Han c, Troy Sturzenegger c, Raylene A. Reimer a, d, * a

Faculty of Kinesiology, University of Calgary, 2500 University Drive NW, Calgary, AB T2N 1N4, Canada Physical Education and Recreation Studies, Mount Royal University, 4825 Mount Royal Gate SW, Calgary, AB T3E 6K6, Canada Food Processing Development Centre, Alberta Agriculture and Rural Development, 6309 e 45 Street, Leduc, AB T9E 7C5, Canada d Department of Biochemistry & Molecular Biology, Cumming School of Medicine, University of Calgary, 3330 Hospital Drive NW, Calgary, AB T2N 4N1, Canada b c

a r t i c l e i n f o

s u m m a r y

Article history: Received 20 April 2015 Accepted 28 December 2015

Background & aims: The purpose of this randomized, double-blind, placebo-controlled study was to assess the effects of yellow pea fiber intake on body composition and metabolic markers in overweight/ obese adults. Methods: Participants (9 M/41 F; age 44 ± 15 y, BMI 32.9 ± 5.9 kg/m2) received isocaloric doses of placebo (PL) or pea fiber (PF; 15 g/d) wafers for 12 weeks. Outcome measures included changes in anthropometrics, body composition (DXA), oral glucose tolerance test (OGTT), food intake (ad libitum lunch buffet), and biochemical indices. Results: The PF group lost 0.87 ± 0.37 kg of body weight, primarily due to body fat (
0.74 ± 0.26 kg), whereas PL subjects gained 0.40 ± 0.39 kg of weight over the 12 weeks (P ¼ 0.022). The PF group consumed 16% less energy at the follow-up lunch buffet (P ¼ 0.026), whereas the PL group did not change. During the OGTT, glucose area under the curve (AUC) was lower in PF subjects at follow-up (P ¼ 0.029); insulin increased in both groups over time (P ¼ 0.008), but more so in the PL group (38% higher AUC vs. 10% higher in the PF group). There were no differences in gut microbiota. Conclusions: In the absence of other lifestyle changes, incorporating 15 g/day yellow pea fiber may yield small but significant metabolic benefits and aid in obesity management. Clinical Trial Registry: ClinicalTrials.gov NCT01719900. © 2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

Keywords: Legume Dried peas Obesity Body weight Functional food

1. Introduction Targeted lifestyle interventions are the safest and most economic approaches for both prevention and treatment of obesity, exhibiting potential for multi-faceted health and metabolic

Q1

Abbreviations: DXA, dual X-ray absorptiometry; tAUC, total area under the curve; OGTT, oral glucose tolerance test; PF, pea fiber; PL, placebo; VAS, visual analog scales. * Corresponding author. University of Calgary, 2500 University Dr. NW, Calgary, AB T2N 1N4, Canada. E-mail addresses: [email protected] (J.E. Lambert), [email protected] (J.A. Parnell), [email protected] (J.M. Tunnicliffe), [email protected] (J. Han), [email protected] (T. Sturzenegger), [email protected] (R.A. Reimer).

benefits. Dietary fiber contributes to weight control and blood glucose management through a variety of mechanisms, including delayed nutrient absorption, increased satiety, and stimulation of gut hormones that regulate food intake [1]. Fiber may also beneficially modulate the intestinal microbiota, the manipulation of which influences whole-body energy metabolism. For example, in a double-blind placebo-controlled trial in obese women, feeding the prebiotic fibers inulin and oligofructose (16 g/d) for 3 months reduced Bacteroides intestinalis and B. vulgatus, which was correlated with glucose homeostasis and a slight reduction in fat mass [2]. As such, there is increasing interest in identifying fibers which can be classified as ‘prebiotics’, meaning they favorably modulate the composition and/or activity of the gut microbiota thereby conferring a health benefit to the host [3].

http://dx.doi.org/10.1016/j.clnu.2015.12.016 0261-5614/© 2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

Please cite this article in press as: Lambert JE, et al., Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/ obese adults in a 12-week randomized controlled trial, Clinical Nutrition (2016), http://dx.doi.org/10.1016/j.clnu.2015.12.016

55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

YCLNU2717_proof ■ 15 January 2016 ■ 2/8

2

J.E. Lambert et al. / Clinical Nutrition xxx (2016) 1e8

Fiber intake in North America falls short of recommendations [4] and consumers are interested in fortified products that could promote weight loss and improve health. Different fiber types (e.g. inulin, beta-glucan) and/or sources (pulses, grains, etc.) have different physical and biological properties [1], which may affect metabolic health in distinct ways. Therefore, there is value in characterizing diverse fiber sources to ensure a broad range of options in the food supply. The hulls of yellow peas are comprised of ~82% fiber making them an excellent fiber source for incorporation into food products [5]. In metabolically unhealthy animals (glucose intolerant, hypercholesterolemic), feeding diets containing whole or fractionated (hulls) peas improved fasting and postprandial glucose and insulin levels and reduced body fat in diet-induced obese rats [6e8]. In metabolically unhealthy humans (overweight hypercholesterolemic), 12 g/d of pea fiber intake for 28 d reduced fasting insulin concentrations and improved postprandial glucose responses after a standardized meal [9], although whether or not the benefits were associated with changes in gut microbiota is not known. Given the importance of determining the effects of specific fiber types on metabolic outcomes in overweight and obese adults, the objective of this randomized, double-blind, placebo-controlled study was to determine the effects of yellow pea fiber on body composition, metabolic markers of obesity, and gut microbiota abundance in overweight/obese adults. 2. Materials and methods 2.1. Subjects Overweight and obese adults (age 18e70 y; BMI 25e38 kg/m2) were recruited from the community in Calgary, Alberta, Canada in 2012 and 2013. Eligible subjects included adults with stable body weight (defined as <3 kg lost or gained within the last 3 months), and exclusion criteria as has been described previously [10]. Eligibility was assessed using a screening questionnaire and phone interview. After screening, participants were randomly assigned using computer generated numbers (and stratified according to age, sex, and BMI) to either placebo (PL) or pea fiber (PF; 15 g/ d yellow pea fiber) for 12 weeks. The randomization sequence was generated by an investigator not involved in recruiting participants and sequences were not revealed to study staff. One research assistant was responsible for product distribution to ensure the correct product was provided to the groups. Study personnel were unaware of treatment allocation prior to the assignment of interventions and participants and research staff were blinded to treatments. As previously described [10], sample size was determined from anticipated changes in body fat based on data from our previous work on prebiotic fiber supplementation in overweight and obese adults [11]. This proposal was approved by the Conjoint Health Research Ethics Board of the University of Calgary, and voluntary, written informed consent was obtained from each participant. 2.2. Study design The detailed design of this double-blind, placebo-controlled parallel group design has been previously described [10] and therefore is described briefly here. Prior to testing days (at baseline and follow-up), subjects completed a 3-day weighed food record (analyzed using FoodWorks software, The Nutrition Company, Long Valley, NJ) and a physical activity record (Godin's Leisure Time Exercise Questionnaire), and collected a stool sample for analysis of gut microbiota. Validated 100 mm visual analog scales (VAS) [12] were completed by participants at home on a weekly basis to

assess subjective ratings of appetite (including hunger, satiety, desire to eat, feeling of fullness) as described previously [10]. On testing days at 0 and 12 wk, anthropometrics (height, weight, BMI, waist circumference) were measured and a fasting blood sample taken to assess inflammatory markers, lipids, and satiety hormones. A standard oral glucose (75 g) tolerance test (OGTT) was performed, and glucose, insulin, and gut hormones assessed at 0, 30, 60, 120, and 180 min. Body composition was measured by whole body dual-energy X-ray absorptiometry (DXA) scan (Hologic QDR 4500, Hologic, Inc., Bedford, MA, USA). An ad libitum lunch buffet [13] consisting of savory (pizza) and sweet (cookies) food options was provided to subjects to objectively assess food intake. Subjects were instructed to eat until “comfortable satisfaction,” and each individual's energy intake was measured. Conversation between subjects was monitored such that topics did not involve food, appetite, or related issues [13]. Subjective ratings of appetite were measured using validated VAS before and after the lunch. 2.3. Dietary intervention The PF group received wafers containing 5 g/serving of yellow pea fiber (‘Best’ Pea Fiber, Best Cooking Pulses Inc., Portage la Prairie, MB, Canada), while the placebo group received an isocaloric dose of control wafers with no pea fiber (204 kcal of total additional energy per day in each group). The wafers were formulated and produced by the Leduc Food Processing Development Centre (Leduc, AB, Canada). The fiber was supplied to the Centre by Best Cooking Pulses Inc. (Portage la Prairie, MB, Canada) and was dry milled from the outer hull of yellow peas, having a 92% total dietary fiber content of which 8% was soluble. Subjects were instructed to consume the wafers approximately 30 min before their 3 largest daily meals. Subjects returned all packages (both empty and unconsumed) to assess compliance. To minimize gastrointestinal discomfort associated with a rapid increase in fiber intake, the dose was increased incrementally during the first 3 weeks of the study (week 1 ¼ 5 g/d; week 2 ¼ 10 g/d; week 3 ¼ 15 g/d). The study was designed to examine the effects of pea fiber supplementation in a real-life setting, independent of any other diet or exercise intervention; therefore, subjects were encouraged to maintain their usual lifestyle and habitual food intake, eat until comfortably full, and not consciously try to gain/lose weight throughout the study. 2.4. Laboratory analyses Blood was centrifuged immediately and plasma and serum aliquots frozen at
80  C for subsequent analyses. Glucose was quantified using a trinder assay (Stan Bio, Boerne, TX, USA). Samples were sent to Calgary Laboratory Services for measurement of plasma HbA1 c, serum lipids (total, LDL- and HDL-cholesterol, and triglyceride), and serum CRP. For gut hormones, blood was drawn into cooled EDTA vacutainers containing inhibitors as previously described [10,11], and centrifuged within 30 min. Milliplex kits (Millipore, Billerica, MA, USA) were used to measure plasma insulin, GIP (total), amylin, ghrelin (active), leptin, and PYY (total). Active GLP-1 was measured with ELISA (Millipore, EGLP-35K). Serum adiponectin, resistin, IFNy, IL12P40, IL-1B, IL-6, IL-8, MCP1, and TNF-a were measured with human cytokine and adipokine Milliplex kits (Millipore). 2.5. Gut microbiota analysis Subjects collected stool specimens as previously described [10]. Samples were kept at
80  C until analysis. Total microbial DNA was extracted from frozen fecal samples using the QIAamp DNA

Please cite this article in press as: Lambert JE, et al., Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/ obese adults in a 12-week randomized controlled trial, Clinical Nutrition (2016), http://dx.doi.org/10.1016/j.clnu.2015.12.016

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

YCLNU2717_proof ■ 15 January 2016 ■ 3/8

J.E. Lambert et al. / Clinical Nutrition xxx (2016) 1e8

Stool Mini Kit (Qiagen, Mississauga, ON) and concentration measured using the Pico-Green DNA Quantification Kit (Invitrogen, Carlsbad, CA). Microbiota were quantified by quantitative PCR (qPCR) and group specific primers using previously described protocols with SYBR green [14,15]. Purified template DNA from reference strains was used to generate standard curves for each primer set using 10-fold serial dilutions of DNA. Melting curve analysis was performed following each assay to confirm the specificity of the PCR products. 2.6. Statistical analysis Data were analyzed with SPSS (v.21; IBM Software, Armonk, NY). Differences between groups at baseline were determined by student's t-test. Intervention data were analyzed by 2-way repeated measures ANOVA to determine the main effects of phase (baseline at week 0 and followup at week 12) and diet group and their interaction. Postprandial OGTT data was analyzed by 3way repeated measures ANOVA to determine main effects of phase (0 and 12 weeks), diet group (PF, PL), and time (0, 30, 60, 120, 180 min) and their interaction. Total area-under-the-curve (tAUC) for glucose, insulin, and gut hormones following the OGTT was analyzed by 2-way repeated measures ANOVA. Data are presented as mean ± SE. 3. Results 3.1. Subjects As described in Fig. 1, 53 adults were recruited and, of these, 47 completed the study. Six individuals dropped out (2 in the PL group

Assessed for eligibility (n= 90)

-

Recruited (n= 53)

Excluded (n= 37) Did not meet including criteria (n= 9) (BMI out of range; pregnancy; medicaƟons) Declined to parƟcipate (n= 28) due to work, travel, or unknown reasons

Placebo (n= 24)

Pea Fiber (n= 29)

Drop out (n= 2)

Drop out (n= 4)

Analyzed (n= 22)

Analyzed (n= 22) - Excluded from analysis (n= 3) due to worsening of preexisƟng disease unrelated to the study or marked change in lifestyle

Fig. 1. Flow diagram of subject recruitment, testing, and analysis.

3

and 4 in the PF group) for reasons including new pregnancy, employment-related, or undisclosed. Three subjects (all in the PF group) were excluded from analysis due to worsening of preexisting disease unrelated to the study or marked change in lifestyle due to family member illness, such that data analysis is presented on 44 subjects (22 in the PL group and 22 in the PF group). 3.2. Body weight and composition Subject characteristics before and after the intervention are presented in Table 1. There were no differences between the PL (n ¼ 22; 4M/18F) and PF (n ¼ 22; 4M/18F) groups at baseline in anthropometric and biochemical characteristics. There was a phase  diet interaction for body weight and body fat mass, such that the PF group lost weight (
0.87 ± 0.37 kg), primarily as fat mass (
0.74 ± 0.26 kg), over the 12 week intervention whereas the PL group gained weight (þ0.4 0 ± 0.39 kg; P ¼ 0.022) and fat mass (þ0.42 ± 0.38 kg; P ¼ 0.014). Correspondingly, BMI was reduced slightly in the PF group but not in the PL group (P ¼ 0.025), but there was no significant change in waist circumference in either group. 3.3. Biochemistries and inflammation HbA1c levels were stable in the PF group but increased by 2.5% in the PL group although this difference was not significant (Table 1) (P ¼ 0.063 for phase  diet group interaction). There was no significant change in plasma lipid levels or most inflammatory markers, except for IL-1B which was numerically but not significantly reduced in the PF group and increased in the PL group (P ¼ 0.083 for phase  diet group interaction), and IL-8, which reduced in both groups (P ¼ 0.01 for main effect of phase). Fasting leptin increased by 25% in the PL but decreased 7% in the PF group (P ¼ 0.011 for phase x diet group interaction). Fasting adiponectin and resistin increased in both groups at follow-up compared to baseline (P < 0.001 for effect of phase for both outcomes). 3.4. Food intake and satiety Self-reported food intake from the 3-day food records showed that both groups reduced energy intake over time (PL group 2048 ± 105 to 1643 ± 117 kcal/d and PF group 1989 ± 120 to 1635 ± 103 kcal/d; P < 0.001 for main effect of phase) but there was no significant effect of diet group on intake of total energy, protein, carbohydrate, fat, fiber (food-derived fiber that was independent of the supplemental fiber provided in pea fiber wafers) or sugars (data not shown; P > 0.05). At the lunch buffet, the PF group consumed 16 ± 6% less energy at follow-up compared to the PL group that consumed ~5% more (Fig. 2). There were no differences in VAS hunger/satiety scores taken either before or after the lunch buffet between groups (data not shown; P > 0.05). Compliance was the same in both groups (88 ± 2% of wafers consumed in the PF group and 85 ± 2% in the PL; P ¼ 0.35), but greater compliance was associated with greater reduction in body weight and body fat (r ¼
0.50, P ¼ 0.020 both) in the PF group but not the PL group (P ¼ 0.77 and P ¼ 0.34, respectively). There was no significant effect of diet group on physical activity; both groups reduced physical activity similarly between baseline and follow-up (data not shown; P < 0.05). 3.5. Oral glucose tolerance test Plasma glucose and hormone (insulin, GIP, GLP-1, ghrelin, amylin, and PYY) responses during the OGTT at baseline and followup are presented over time in Fig. 3 and as tAUC in Fig. 4. The

Please cite this article in press as: Lambert JE, et al., Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/ obese adults in a 12-week randomized controlled trial, Clinical Nutrition (2016), http://dx.doi.org/10.1016/j.clnu.2015.12.016

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

4

J.E. Lambert et al. / Clinical Nutrition xxx (2016) 1e8

Table 1 Anthropometric characteristics and fasting biochemical and inflammatory markers of placebo and pea fiber groups before and after diet intervention. Placebo (n ¼ 22)

Measure

Anthropometrics Weight (kg) BMI (kg/m) Waist circumference (cm) Fat mass (kg) Lean mass (kg) % Body fat Biochemistry HbA1c (%) Total chol (mmol/L) LDL-c (mmol/L) HDL-c (mmol/L) Triglyceride (mmol/L) TC:HDL Adiponectin (ng/mL) Leptin (pg/mL) Resistin (ng/mL) Inflammatory CRP (mg/L) IFNy IL12P40 IL-1B IL-6 IL-8 MCP1 TNF-a *

Pea fiber (n ¼ 22)

RM-ANOVA

Baseline

Follow up

Baseline

Follow up

93.7 ± 4.4 33.3 ± 1.3 109.4 ± 2.9 33.4 ± 2.3 54.2 ± 2.4 36.8 ± 1.7

94.1 ± 4.3 33.4 ± 1.3 108.3 ± 2.4 33.8 ± 2.3 54.5 ± 2.3 37.0 ± 1.7

92.3 ± 4.1 33.1 ± 1.3 109.8 ± 2.9 34.8 ± 2.3 55.2 ± 2.3 37.4 ± 1.6

91.5 ± 4.0 32.8 ± 1.3 107.8 ± 2.8 34.0 ± 2.2 55.2 ± 2.3 37.0 ± 1.6

5.75 5.05 2.96 1.50 1.30 3.54 1.57 4.23 33.1

± ± ± ± ± ± ± ± ±

0.11 0.23 0.19 0.09 0.19 0.29 0.14 (x104) 0.64 (x104) 1.9

4.25 ± 1.02 2.07 ± 0.53 5.0 ± 2.0 0.52 ± 0.12 1.66 ± 0.47 6.85 ± 0.55 548 ± 47 11.9 ± 0.9

5.89 4.93 2.90 1.49 1.20 3.53 1.88 5.29 37.1

± ± ± ± ± ± ± ± ±

0.11 0.25 0.21 0.09 0.14 0.26 0.19 (x104) 0.83 (x104) 1.9

3.08 ± 0.57 2.13 ± 0.53 22.7 ± 8.8 0.80 ± 0.21 1.28 ± 0.29 5.73 ± 0.48 577 ± 47 12.5 ± 0.8

5.68 4.95 2.86 1.41 1.55 3.79 1.71 4.06 31.8

± ± ± ± ± ± ± ± ±

0.11 0.24 0.20 0.09 0.19 0.30 0.20 (x104) 0.61 (x104) 2.1

4.10 ± 0.89 2.25 ± 0.31 21.5 ± 5.5 1.12 ± 0.20 1.27 ± 0.28 6.27 ± 0.60 464 ± 37 13.1 ± 0.8

Q3

5.68 4.95 2.92 1.40 1.48 3.72 2.03 3.78 35.3

± ± ± ± ± ± ± ± ±

0.11 0.25 0.22 0.09 0.14 0.26 0.22 (x104) 0.52 (x104) 1.9

2.80 ± 0.50 1.99 ± 0.43 22.6 ± 6.6 1.01 ± 0.21 0.89 ± 0.20 5.28 ± 0.50 458 ± 41 12.6 ± 0.9

Phase

Diet group

Phase x diet

0.39 0.41 0.083 0.48 0.68 0.52

0.74 0.85 0.99 0.81 0.80 0.89

0.022 0.025 0.64 0.014 0.55 0.12

0.063 0.48 0.97 0.55 0.39 0.69 <0.001 0.13 <0.001

0.36 0.90 0.88 0.45 0.23 0.56 0.59 0.37 0.55

0.063 0.54 0.40 0.96 0.86 0.79 0.96 0.011 0.78

0.99 0.96 0.24 0.13 0.099 0.32 0.084 0.62

0.50 0.90 0.17 0.083 0.79 0.87 0.36 0.22

0.081 0.86 0.12 0.89 0.082 0.010 0.57 0.68

2-way repeated measures ANOVA performed to determine independent effects of diet intervention (phase) and diet group, and interaction; data presented as mean ± SEM.

1400

Energy Intake (kcal)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

YCLNU2717_proof ■ 15 January 2016 ■ 4/8

Phase × Diet P=0.022

1200 1000

PL

800

PF

(Figs. 3E and 4E). Amylin (Fig. 4F) increased in both groups over time (P ¼ 0.003), with a tendency for the increase to be greater in the PL versus PF group (P ¼ 0.062 for phase  diet group interaction for tAUC). Finally, plasma PYY increased in the PL group over time (Fig. 3G) but did not change in the PF group (P ¼ 0.030 for phase  time  diet group interaction).

600 400

3.6. Gut microbiota

200 0 -200 -400

Baseline

FollowUp

*

Change

Fig. 2. Energy intake at the ad libitum lunch buffet. Baseline and follow-up energy intake was analyzed by repeated measures ANOVA. The change in energy intake was calculated as follow-up energy intake minus initial energy intake. * represents a significant difference between PL and PF (P ¼ 0.022). Data presented as mean ± SEM.

interaction between diet and phase influenced plasma glucose levels (Fig. 4A) wherein glucose tAUC was lower in the PF group at follow-up but higher in the PL group (P ¼ 0.029 for phase  diet group interaction). Plasma insulin concentrations were overall higher in both groups at follow-up compared to baseline, reflected in both the timecourse curves (Fig. 3B) (P ¼ 0.008 for main effect of phase) and tAUC (Fig. 4B) (P ¼ 0.009); the magnitude of increase was greater in the PL (~38% higher tAUC at follow-up) than the PF group (~10% higher tAUC at follow-up) but did not reach statistical significance (P > 0.05). Plasma GIP levels were lower in both groups at follow-up, reflected in both the timecourse curves (Fig. 3C) (P ¼ 0.028 for main effect of phase) and tAUC (Fig. 4C) (P ¼ 0.063). The slight increase in GLP-1 (Figs. 3D and 4D) in both groups was not significant (P ¼ 0.072 in 3D). There were no changes in ghrelin

Gut microbial data are presented in Table 2. Clostridium leptum, Clostridium cluster I, and Roseburia spp. increased in both groups between baseline and follow-up (P < 0.05). However, there were no significant effects of pea fiber supplementation on the microbial groups analyzed. 4. Discussion Our findings demonstrate that consumption of a moderate dose of yellow pea fiber (15 g/d) by free-living overweight and obese adults corresponds to small but significant improvements in body fat and glucose tolerance. Although rodent studies have demonstrated changes in gut microbiota, most notably the Clostridiales, with yellow pea fiber [7,8], we were unable to detect any significant changes in microbiota at the dose tested. This suggests that the metabolic benefits of pea fiber may occur independently of microbial changes and instead via reductions in energy intake. Studies on the metabolic effects of pea hull fiber in humans are limited, and the evidence mixed. In the acute setting, adding 10 g pea fiber into a meal did not influence glucose or insulin responses but reduced cholesterol levels in healthy adults [16], while other studies incorporating 10e20 g fiber in the meal have observed reductions in postprandial glucose and/or insulin responses [17e19] as well as in postprandial TG excursions [19]. Incorporating pea fiber-containing products into the diet over the longer term also shows mixed results. For example, in healthy adults the benefit of

Please cite this article in press as: Lambert JE, et al., Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/ obese adults in a 12-week randomized controlled trial, Clinical Nutrition (2016), http://dx.doi.org/10.1016/j.clnu.2015.12.016

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

YCLNU2717_proof ■ 15 January 2016 ■ 5/8

J.E. Lambert et al. / Clinical Nutrition xxx (2016) 1e8

5

Fig. 3. Plasma glucose, insulin and gut hormone concentrations during the oral glucose tolerance test. Panels depict plasma (A) glucose, (B) insulin, (C) GIP, (D) GLP-1, (E) ghrelin, (F) amylin, and (G) PYY concentration time curves. Placebo group baseline, black filled circles; Placebo group follow-up, white filled circles with black outline; Pea Fiber group baseline, gray filled circles; Pea Fiber group follow-up, white filled circles with gray outline. Data presented as mean ± SEM. Time curves were analyzed by 3-way repeated measures ANOVA for the effects of phase (baseline/follow-up), diet group, and time (0 through 180 min). Abbreviations: GIP, gastric inhibitory peptide; GLP-1, glucagon-like peptide 1; PYY, peptide YY.

pea fiber appears to be limited to a reduction in fasting plasma TG [19], which may be expected in normoglycemic and normolipidemic individuals. By contrast, in overweight hypercholesterolemic adults, baked products containing whole or fractionated (hulls only) pea fiber (12e14 g/d) reduced fasting insulin concentrations by 10e14% and improved postprandial glucose responses after a standardized breakfast over 4 weeks [9]. In addition, total (~10%) and LDL-cholesterol (~14%) were also reduced; however, the authors noted that the background diet provided to subjects likely caused this reduction [9]. By comparison, we did not find a significant effect of pea fiber supplementation on plasma lipid levels or inflammatory markers, but the participants were not

hyperlipidemic and therefore this result may be expected. There were indications that glucose tolerance was maintained or improved in the fiber group (e.g. HbA1c, glucose AUC, insulin AUC), while tolerance slightly worsened in the placebo group over time. In contrast to previous reports [9], we found a modest but significant effect of pea fiber on body weight reduction. While the magnitude of weight loss is small, the simple addition of fiber-rich wafers to the free-living diet in subjects (who were instructed not to consciously change other behaviors) appeared to halt the slow but progressive increase in body weight that typically occurs over time. Importantly, body weight reduction occurred largely due to loss of body fat, while lean mass did not change. This is somewhat

Please cite this article in press as: Lambert JE, et al., Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/ obese adults in a 12-week randomized controlled trial, Clinical Nutrition (2016), http://dx.doi.org/10.1016/j.clnu.2015.12.016

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

YCLNU2717_proof ■ 15 January 2016 ■ 6/8

6

J.E. Lambert et al. / Clinical Nutrition xxx (2016) 1e8

Fig. 4. Plasma glucose, insulin and gut hormone concentration total area under the curve (AUC) during the oral glucose tolerance test. Panels depict plasma (A) glucose, (B) insulin, (C) GIP, (D) GLP-1, (E) ghrelin, (F) amylin, and (G) PYY concentration AUC responses. Placebo group baseline, black filled bars; Placebo group follow-up, white filled bars with black outline; Pea Fiber group baseline, gray filled bars; Pea Fiber group follow-up, white filled circles with gray outline. Data presented as mean ± SEM. tAUC data were analyzed by 2way repeated measures ANOVA for the effects of phase (baseline/follow-up) and diet group. Abbreviations: GIP, gastric inhibitory peptide; GLP-1, glucagon-like peptide 1; PYY, peptide YY.

unexpected, but positive given that weight loss typically involves loss of both fat and lean mass; interestingly, previous studies using other fiber types have also observed a relative “sparing” of lean mass in the midst of weight loss in animals [20,21] and humans [11]. The mechanisms for this preservation of lean mass with fiber are not clear, but may include improvements in insulin sensitivity [6] or beneficial effects on gene expression related to fat oxidation that arise directly or via metabolites of microbial fiber fermentation [22]. Previous work has suggested that fiber consumption may reduce food intake and promote satiety. However, in a randomized

crossover study in healthy men, there were no differences in subsequent food intake or appetite score between meals containing pea hull fiber (7 g), pea protein (10 g), pea protein plus fiber, or whole yellow peas [23]. Similarly, providing test meals with 12 g or 22 g of pea fiber did not affect subsequent food intake or appetite in healthy men [24]. Here, we echo these findings, observing no independent effect of chronic pea fiber supplementation on subjective hunger/satiety scores (though pea fiber was not provided in the test meal) or self-reported food intake (as measured by the 3-day food records); however, we did observe a 16% reduction in food intake at the objectively-measured ad

Please cite this article in press as: Lambert JE, et al., Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/ obese adults in a 12-week randomized controlled trial, Clinical Nutrition (2016), http://dx.doi.org/10.1016/j.clnu.2015.12.016

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

YCLNU2717_proof ■ 15 January 2016 ■ 7/8

J.E. Lambert et al. / Clinical Nutrition xxx (2016) 1e8

7

66 67 68 Bacteria Placebo Pea fiber RM-ANOVA 69 Baseline Follow up Baseline Follow up Phase Diet group Phase x diet 70 Total bacteria 7.29 ± 0.03 7.30 ± 0.03 7.29 ± 0.03 7.32 ± 0.02 0.35 0.83 0.58 71 Bacteroidetes/Prevotella spp. 6.72 ± 0.05 6.68 ± 0.07 6.63 ± 0.08 6.70 ± 0.05 0.80 0.73 0.23 72 Bifidobacterium spp. 4.41 ± 0.19 4.20 ± 0.21 4.50 ± 0.19 4.60 ± 0.21 0.60 0.46 0.12 73 Enterobacteriaceae 4.05 ± 0.08 4.14 ± 0.06 4.19 ± 0.11 4.20 ± 0.06 0.41 0.20 0.76 Methanobrevibacter spp. 3.63 ± 0.15 3.59 ± 0.14 3.81 ± 0.17 3.82 ± 0.18 0.75 0.43 0.84 74 Firmicutes 75 Lactobacillus spp. 2.53 ± 0.10 2.29 ± 0.10 2.57 ± 0.13 2.43 ± 0.16 0.15 0.71 0.72 76 Clostridium leptum (C-IV) 6.59 ± 0.08 6.66 ± 0.08 6.69 ± 0.09 6.86 ± 0.05 0.023 0.13 0.29 77 Clostridium coccoides (C-XIVa) 6.71 ± 0.07 6.73 ± 0.04 6.67 ± 0.08 6.81 ± 0.04 0.13 0.84 0.29 Clostridium cluster I 3.84 ± 0.10 4.18 ± 0.06 3.94 ± 0.11 4.29 ± 0.09 <0.001 0.41 0.81 78 Clostridium cluster XI 4.00 ± 0.20 3.97 ± 0.20 4.29 ± 0.20 4.46 ± 0.16 0.58 0.078 0.54 79 Roseburia spp. 5.51 ± 0.15 5.60 ± 0.09 5.26 ± 0.14 5.59 ± 0.10 0.047 0.45 0.26 80 Data presented as mean þ SE. Data represent Log 16S rRNA gene copies/20 ng total genomic DNA, isolated from stool samples. Data analyzed by 2-way RM-ANOVA to 81 determine independent effects of diet intervention (phase) and diet group, and interaction, analyzing for simple effects. 82 83 84 4.1. Limitations libitum test meal in the PF individuals. Lack of agreement between 85 the three different measures of food intake and appetite may 86 A possible limitation of the study was that the subjects were on reflect the inherent limitations of subjective ratings of appetite 87 self-selected diets; however, one of the objectives of the study was and under-reporting known to occur with 3-day food records [25]. 88 to determine the acceptability of pea-fiber containing products in While the ad libitum test meal is designed to be an objective 89 the diets of free-living individuals with ad libitum intake as would measure of food intake, it was nevertheless the only marker of 90 be the case in a real world setting. Further, advances in technology, reduced food intake in the PF participants. Furthermore, both 91 based on global 16S rRNA gene sequencing or microarray, have groups self-reported a reduction of approximately 300 kcal/ 92 allowed recent studies to highlight global differences in taxonomic d based on the 3-day food records, but only the PF group had a 93 groups and thereby provide a more detailed understanding of the small but significant reduction in body weight. This discrepancy 94 deeper compositional changes occurring in response to intervencould potentially be explained in part by greater under-reporting 95 tion or disease state. Nevertheless, the qPCR methods used here are of food intake by the PL group, similar food intake but greater 96 well-established and validated, and this study represents the first energy expenditure by the PF group or compensatory mechanisms 97 to characterize the selected gut microbial response to yellow pea to defend body weight in the face of energy restriction, such as 98 fiber supplementation in humans. energy-sparing in skeletal muscle and thyroid hormone alter99 ations that varied between groups [25,26]. 100 To our knowledge, this is the first study to assess the in vivo 5. Conclusion 101 prebiotic potential of yellow pea fiber in humans. The data sug102 gests that this fiber does not significantly alter the gut microbiota, The present data provides evidence that supplementation of 103 which is in contrast to promising results from animal models and yellow pea fiber into the diet of free-living individuals has modest 104 studies of in vitro fermentation of pulse flour [27]. In particular, but significant benefits on body fat reduction and postprandial 105 yellow peas contain greater galactooligosaccharides (raffinose and glucose tolerance. In this study, pea fiber did not appear to affect 106 stachynose) and total fructans than most other pulses including the selected gut microbiota analyzed, in contrast to effects 107 chickpeas, kidney beans, and lentils [28]. In animal studies, diets observed in animals. Nonetheless, the data suggests that products 108 containing peas or components (fractionated or hulls only) are containing yellow pea fiber are well-received and easy to incor109 associated with greater cecal Bifidobacterium abundance in normal porate into daily food intake, and thus show promise as a tool 110 rats [29], shifts in the cecal microbial profile (particularly the within the context of an overall weight reduction and/or manage111 Firmicutes species), and reduced Bacilli species in hypercholesment program. Further investigations should examine the efficacy 112 terolemic hamsters [7], and greater abundance of ileal Lactobaof pea fiber on body weight, metabolic health, and gut microbial 113 cillus and colonic Bifidobacterium in piglets [30]. Finally, we have composition when dosage is increased and/or when combined with 114 recently shown that in diet-induced obese rats, C. leptum was other nutrients in a focused intervention. 115 reduced with a variety of treatments (yellow pea fiber, flour, or 116 starch), and both pea fiber and flour were associated with reduced Funding sources 117 Firmicutes (as % of total bacteria) [8]. The contrast between these 118 positive animal studies and our observed lack of effect of pea fiber Funding provided by Alberta Innovates Bio Solutions, Alberta 119 in humans could be related to the dose, with the dosage used in Innovates Health Solutions and Alberta Pulse Growers Commission. 120 humans (here 15 g/day) substantially lower than that which JEL is supported by postdoctoral fellowships from the Canadian 121 modulated the gut microbiota in animals (e.g. equivalent of 30 g Institutes of Health Research and the University of Calgary Eyes 122 per 1000 kcal) [7,8] or the specific type/brand of pea fiber. It is also High Postdoctoral program. Q2 123 possible that the well-documented inter-individual differences in 124 microbial profile [31,32] could contribute to variability in the data. 125 The lack of prebiotic effect of pea fiber is in contrast to oligoConflict of interest 126 fructose, a commonly utilized fiber with strong bifidogenic ca127 pacity that appears to more consistently modify the gut Funding was provided in part by Alberta Pulse Growers Com128 microbiota and metabolic markers in overweight humans [2,11]. mission and product provided in-kind by Best Cooking Pulses, Inc. 129 Thus, yellow pea fiber has positive metabolic effects but does not These agencies had no role in study design, data collection, analysis, 130 appear to alter the microbiota at the dose tested. and interpretation, or manuscript preparation. Table 2 Gut microbiota of placebo and pea fiber groups before and after diet intervention.

Please cite this article in press as: Lambert JE, et al., Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/ obese adults in a 12-week randomized controlled trial, Clinical Nutrition (2016), http://dx.doi.org/10.1016/j.clnu.2015.12.016

Q4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

YCLNU2717_proof ■ 15 January 2016 ■ 8/8

8

J.E. Lambert et al. / Clinical Nutrition xxx (2016) 1e8

Statement of authorship JEL analyzed data and wrote the manuscript. JAP participated in research design related to gut microbiota analysis and dietary fiber dosing, and provided manuscript revisions. JH and TS participated in research design related to food product development and packaging, and provided manuscript revisions. JMT conducted the study and provided manuscript revisions. RAR conceived of the study and was involved in research design, data analysis, and writing of the manuscript, and had primary responsibility for final content. All authors read and approved the final manuscript. Acknowledgments The authors would like to thank the individuals who volunteered for this study and the technical assistance of Kristine Lee.

[13]

[14]

[15]

[16]

[17]

[18]

[19]

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.clnu.2015.12.016.

[20]

[21]

References [1] Kristensen M, Jensen MG. Dietary fibres in the regulation of appetite and food intake. Importance of viscosity. Appetite 2011;56(1):65e70. [2] Dewulf EM, Cani PD, Claus SP, Fuentes S, Puylaert PGB, Neyrinck AM, et al. Insight into the prebiotic concept: lessons from an exploratory, double blind intervention study with inulin-type fructans in obese women. Gut 2013;62(8):1112e21. [3] Roberfroid M, Gibson GR, Hoyles L, McCartney AL, Rastall R, Rowland I, et al. Prebiotic effects: metabolic and health benefits. Br J Nutr 2010;104(Suppl): S1e63. [4] Health Canada. Do Canadian adults meet their nutrient requirements through food intake alone?. 2012. Available at: http://www.hc-sc.gc.ca/fn-an/surveill/ nutrition/commun/art-nutr-adult-eng.php#a323. [5] Sosulski FW, Wu KK. High-fiber breads containing field pea hulls, wheat, corn, and wild oat brans. Cereal Chem 1983;65(3):186e91. Available at: http:// www.aaccnet.org/publications/cc/backissues/1988/Documents/CC1988a41. html. [6] Whitlock KA, Kozicky L, Jin A, Yee H, Ha C, Morris J, et al. Assessment of the mechanisms exerting glucose-lowering effects of dried peas in glucoseintolerant rats. Br J Nutr 2012;108(Suppl):S91e102. [7] Marinangeli CPF, Krause D, Harding SV, Rideout TC, Zhu F, Jones PJH. Whole and fractionated yellow pea flours modulate insulin, glucose, oxygen consumption, and the caecal microbiome in Golden Syrian hamsters. Appl Physiol Nutr Metab 2011;36(6):811e20. [8] Eslinger AJ, Eller LK, Reimer RA. Yellow pea fiber improves glycemia and reduces Clostridium leptum in diet-induced obese rats. Nutr Res 2014;34(8): 714e22. [9] Marinangeli CPF, Jones PJH. Whole and fractionated yellow pea flours reduce fasting insulin and insulin resistance in hypercholesterolaemic and overweight human subjects. Br J Nutr 2011;105(1):110e7. [10] Lambert JE, Parnell JA, Han J, Sturzenegger T, Paul HA, Vogel HJ, et al. Evaluation of yellow pea fibre supplementation on weight loss and the gut microbiota: a randomized controlled trial. BMC Gastroenterol 2014;14(1):69. [11] Parnell JA, Reimer RA. Weight loss during oligofructose supplementation is associated with decreased ghrelin and increased peptide YY in overweight and obese adults. Am J Clin Nutr 2009;89(6):1751e9. [12] Stubbs RJ, Hughes DA, Johnstone AM, Rowley E, Reid C, Elia M, et al. The use of visual analogue scales to assess motivation to eat in human subjects: a review

[22]

[23]

[24]

[25] [26]

[27]

[28]

[29]

[30]

[31]

[32]

of their reliability and validity with an evaluation of new hand-held computerized systems for temporal tracking of appetite ratings. Br J Nutr 2000;84(4):405e15. Gregersen NT, Flint A, Bitz C, Blundell JE, Raben A, Astrup A. Reproducibility and power of ad libitum energy intake assessed by repeated single meals. Am J Clin Nutr 2008;87(5):1277e81. Bomhof MR, Saha DC, Reid DT, Paul HA, Reimer RA. Combined effects of oligofructose and Bifidobacterium animalis on gut microbiota and glycemia in obese rats. Obes Silver Spring 2014;22(3):763e71. Parnell JA, Reimer RA. Prebiotic fibres dose-dependently increase satiety hormones and alter Bacteroidetes and Firmicutes in lean and obese JCR: LA-cp rats. Br J Nutr 2012;107(4):601e13. Dubois C, Cara L, Armand M, Borel P, Senft M, Portugal H, et al. Effects of pea and soybean fibre on postprandial lipaemia and lipoproteins in healthy adults. Eur J Clin Nutr 1993;47(7):508e20. Hamberg O, Rumessen JJ, Gudmand-Høyer E. Blood glucose response to pea fiber: comparisons with sugar beet fiber and wheat bran. Am J Clin Nutr 1989;50(2):324e8. Marinangeli CPF, Kassis AN, Jones PJH. Glycemic responses and sensory characteristics of whole yellow pea flour added to novel functional foods. J Food Sci 2009;74(9):S385e9. €m B, Hansen LT, Sørensen A. Pea fiber lowers fasting and postprandial Sandstro blood triglyceride concentrations in humans. J Nutr 1994;124(12):2386e96. Pyra KA, Saha DC, Reimer RA. Prebiotic fiber increases hepatic acetyl CoA carboxylase phosphorylation and suppresses glucose-dependent insulinotropic polypeptide secretion more effectively when used with metformin in obese rats. J Nutr 2012;142(2):213e20. Reimer RA, Maurer AD, Eller LK, Hallam MC, Shaykhutdinov R, Vogel HJ, et al. Satiety hormone and metabolomic response to an intermittent high energy diet differs in rats consuming long-term diets high in protein or prebiotic fiber. J Proteome Res 2012;11(8):4065e74. Islam A, Civitarese AE, Hesslink RL, Gallaher DD. Viscous dietary fiber reduces adiposity and plasma leptin and increases muscle expression of fat oxidation genes in rats. Obes Silver Spring 2012;20(2):349e55. Mollard RC, Luhovyy BL, Smith C, Anderson GH. Acute effects of pea protein and hull fibre alone and combined on blood glucose, appetite, and food intake in healthy young men e a randomized crossover trial. Appl Physiol Nutr Metab 2014;39(12):1360e5. Smith CE, Mollard RC, Luhovyy BL, Anderson GH. The effect of yellow pea protein and fibre on short-term food intake, subjective appetite and glycaemic response in healthy young men. Br J Nutr 2012;108(Suppl):S74e80. Westerterp KR, Goris AH. Validity of the assessment of dietary intake: problems of misreporting. Curr Opin Clin Nutr Metab Care 2002;5(5):489e93. Ebbeling CB, Swain JF, Feldman HA, Wong WW, Hachey DL, Garcia-Lago E, et al. Effects of dietary composition on energy expenditure during weight -loss maintenance. JAMA 2012;307(24):2627e34. Maccaferri S, Klinder A, Cacciatore S, Chitarrari R, Honda H, Luchinat C, et al. In vitro fermentation of potential prebiotic flours from natural sources: impact on the human colonic microbiota and metabolome. Mol Nutr Food Res 2012;56(8):1342e52. Biesiekierski JR, Rosella O, Rose R, Liels K, Barrett JS, Shepherd SJ, et al. Quantification of fructans, galacto-oligosacharides and other short-chain carbohydrates in processed grains and cereals. J Hum Nutr Diet 2011;24(2): 154e76. Da S, Queiroz-Monici K, Costa GE, da Silva N, Reis SMPM, de Oliveira AC. Bifidogenic effect of dietary fiber and resistant starch from leguminous on the intestinal microbiota of rats. Nutrition 2005;21(5):602e8. Chen H, Mao X, He J, Yu B, Huang Z, Yu J, et al. Dietary fibre affects intestinal mucosal barrier function and regulates intestinal bacteria in weaning piglets. Br J Nutr 2013;110(10):1837e48. Eckburg PB, Bik EM, Bernstein CN, Purdom E, Dethlefsen L, Sargent M, et al. Diversity of the human intestinal microbial flora. Science 2005;308(5728): 1635e8. Turnbaugh PJ, Hamady M, Yatsunenko T, Cantarel BL, Duncan A, Ley RE, et al. A core gut microbiome in obese and lean twins. Nature 2009;457(7228): 480e4. Nature Publishing Group.

Please cite this article in press as: Lambert JE, et al., Consuming yellow pea fiber reduces voluntary energy intake and body fat in overweight/ obese adults in a 12-week randomized controlled trial, Clinical Nutrition (2016), http://dx.doi.org/10.1016/j.clnu.2015.12.016

55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108