University of Southern Denmark

Acute effects of delayed-release hydrolyzed pine nut oil on glucose tolerance, incretins,ghrelin and appetite in healthy humans

Sørensen, Karina V.; Korfitzen, Svend S.; Kaspersen, Mads H.; Ulven, Elisabeth Rexen;Ekberg, Jeppe H.; Bauer-Brandl, Annette; Ulven, Trond; Højlund, Kurt

Published in:Clinical Nutrition

DOI:10.1016/j.clnu.2020.09.043

Publication date:2021

Document version:Accepted manuscript

Document license:CC BY-NC-ND

Citation for pulished version (APA):Sørensen, K. V., Korfitzen, S. S., Kaspersen, M. H., Ulven, E. R., Ekberg, J. H., Bauer-Brandl, A., Ulven, T., &Højlund, K. (2021). Acute effects of delayed-release hydrolyzed pine nut oil on glucose tolerance, incretins,ghrelin and appetite in healthy humans. Clinical Nutrition, 40(4), 2169-2179.https://doi.org/10.1016/j.clnu.2020.09.043

Go to publication entry in University of Southern Denmark's Research Portal

Terms of useThis work is brought to you by the University of Southern Denmark.Unless otherwise specified it has been shared according to the terms for self-archiving.If no other license is stated, these terms apply:

• You may download this work for personal use only. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying this open access versionIf you believe that this document breaches copyright please contact us providing details and we will investigate your claim.Please direct all enquiries to puresupport@bib.sdu.dk

Download date: 14. Mar. 2022

Journal Pre-proof

Acute effects of delayed-release hydrolyzed pine nut oil on glucose tolerance,incretins, ghrelin and appetite in healthy humans

Karina V. Sørensen, Svend S. Korfitzen, Mads H. Kaspersen, Elisabeth Rexen Ulven,Jeppe H. Ekberg, Annette Bauer-Brandl, Trond Ulven, Kurt Højlund

PII: S0261-5614(20)30515-X

DOI: https://doi.org/10.1016/j.clnu.2020.09.043

Reference: YCLNU 4484

To appear in: Clinical Nutrition

Received Date: 29 November 2019

Revised Date: 22 August 2020

Accepted Date: 27 September 2020

Please cite this article as: Sørensen KV, Korfitzen SS, Kaspersen MH, Ulven ER, Ekberg JH, Bauer-Brandl A, Ulven T, Højlund K, Acute effects of delayed-release hydrolyzed pine nut oil on glucosetolerance, incretins, ghrelin and appetite in healthy humans, Clinical Nutrition, https://doi.org/10.1016/j.clnu.2020.09.043.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the additionof a cover page and metadata, and formatting for readability, but it is not yet the definitive version ofrecord. This version will undergo additional copyediting, typesetting and review before it is publishedin its final form, but we are providing this version to give early visibility of the article. Please note that,during the production process, errors may be discovered which could affect the content, and all legaldisclaimers that apply to the journal pertain.

© 2020 Published by Elsevier Ltd.

1

Acute effects of delayed-release hydrolyzed pine nut oil on glucose tolerance, incretins, ghrelin and 1

appetite in healthy humans 2

Karina V. Sørensena,b, Svend S. Korfitzenb, Mads H. Kaspersenc, Elisabeth Rexen Ulvend, Jeppe H. Ekberge, 3

Annette Bauer-Brandlc, Trond Ulvenc,d, Kurt Højlunda,b 4

a Steno Diabetes Center Odense, Odense University Hospital, Denmark 5

b Department of Clinical Research, University of Southern Denmark, Odense, Denmark 6

c Department of Physics, Chemistry and Pharmacy, University of Southern Denmark, Odense, Denmark 7

d Department of Drug Design and Pharmacology, University of Copenhagen, Copenhagen, Denmark 8

e Section for Metabolic Receptology, Novo Nordisk Foundation Center for Basic Metabolic Research, University 9

of Copenhagen, Copenhagen, Denmark 10

Corresponding Author 11

Professor Kurt Højlund, e-mail: kurt.hoejlund@rsyd.dk 12

Kløvervænget 10, 5. floor, 5000 Odense, Denmark 13

Journ

al Pre-

proof

2

ABSTRACT 14

Bacground and aim Pinolenic acid, a major component (~20%) of pine nut oil, is a dual agonist of the free fatty 15

acid receptors, FFA1 and FFA4, which may regulate release of incretins and ghrelin from the gut. Here, we 16

investigated the acute effects of hydrolyzed pine nut oil (PNO-FFA), delivered to the small intestine by delayed-17

release capsules, on glucose tolerance, insulin, incretin and ghrelin secretion, and appetite. 18

Methods In two cross-over studies, we evaluated 3 g unhydrolyzed pine nut oil (PNO-TG) or 3 g PNO-FFA 19

versus no oil in eight healthy, non-obese subjects (study 1), and 3 g PNO-FFA or 6 g PNO-FFA versus no oil in 20

ten healthy, overweight/obese subjects (study 2) in both studies given in delayed-release capsules 30 min prior to 21

a 4-h-oral glucose tolerance test (OGTT). Outcomes were circulating levels of glucose, insulin, GLP-1, GIP, 22

ghrelin, appetite and gastrointestinal tolerability during OGTT. 23

Results Both 3 g PNO-FFA in study 1 and 6 g PNO-FFA in study 2 markedly increased GLP-1 levels (p<0.001) 24

and attenuated ghrelin levels (p<0.001) during the last two hours of the OGTT compared with no oil. In study 2, 25

these effects of PNO-FFA were accompanied by an increased satiety and fullness (p<0.03), and decreased 26

prospective food consumption (p<0.05). PNO-FFA caused only small reductions in glucose and insulin levels 27

during the first two hours of the OGTT. 28

Conclusions Our results provide evidence that PNO-FFA delivered to the small intestine by delayed-release 29

capsules may reduce appetite by augmenting GLP-1 release and attenuating ghrelin secretion in the late 30

postprandial state. 31

Keywords Free fatty acid receptors, pine nut oil, pinolenic acid, incretins, appetite, oral glucose tolerance test 32

Clinical Trial registry numbers: NCT03062592 and NCT03305367 33

Journ

al Pre-

proof

3

ABBRIVIATIONS 34

PNO-FFA Hydrolyzed pine nut oil 35

PNO-TG Unhydrolyzed pine nut oil 36

FFA1 Free fatty acid receptor 1 37

FFA4 Free fatty acid receptor 4 38

GLP-1 Glucagon-like Peptide-1 39

GIP Gastric inhibitory polypeptide 40

FFA Free fatty acid 41

RYGB Roux-En-Y Gastric Bypass 42

OGTT Oral glucose tolerance test 43

AUC Area under the curve 44

VAS Visual Analogue Scale 45

Journ

al Pre-

proof

4

INTRODUCTION 46

Type 2 diabetes and its serious complications remain a major health care problem (1, 2). Therefore, the 47

development of preventive and treatment strategies for type 2 diabetes is of paramount importance. Lifestyle 48

modifications focusing on physical activity and a healthy diet are recommended as first line therapies from the 49

time of diagnosis (3-5). However, long term compliance to dietary lifestyle interventions is known to be poor. 50

Moreover, there is an ongoing debate as to what constitutes a healthy diet for prevention and treatment of type 2 51

diabetes that is related to the fact that the metabolic effects of specific food items remain elusive. One of these is 52

the anti-diabetic potential of dietary fatty acids through activation of free fatty acid receptors. 53

The free fatty acid receptor 1 (FFA1) and the free fatty acid receptor 4 (FFA4), formerly known as 54

GPR40 and GPR120, respectively, have attracted considerable attention as potential targets for antidiabetic drug 55

development due to their modulatory effects on glucose metabolism and low-grade inflammation (6-10). FFA1 56

is activated by medium to long-chain fatty acids in both human and rodent cell lines and the receptor is highly 57

expressed in human and rodent pancreatic tissue, especially pancreatic beta-cells (11-14). FFA1 is also found in 58

murine enteroendocrine cells of the small intestine (14). Furthermore, activation of FFA1 by specific free fatty 59

acids (FFA) in MIN6 cells enhances glucose-stimulated insulin secretion (GSIS) (13). Insulin secretion is also 60

stimulated indirectly through the release of the incretin hormones glucagon-like-peptide-1 (GLP-1) and gastric 61

inhibitory polypeptide (GIP) from murine enteroendocrine cells (14-16). An increase in GLP-1 is known to also 62

reduce appetite and body weight (17, 18). FFA4 is also activated by medium to long-chain fatty acids and has 63

been reported to be highly expressed in enteroendocrine cells, regulating GIP and GLP-1 secretion (19, 20), 64

although its ability to promote GLP-1 release has been questioned (21, 22). Furthermore, there is evidence that 65

activation of FFA4 on gastric cells inhibits the secretion of the appetite stimulating orexigenic hormone ghrelin 66

(23-25), which could contribute to an appetite reducing effect of FFA4 agonists. Finally, FFA4 has been linked 67

to anti-inflammatory and insulin sensitizing effects in mice through its presence in adipose tissue (26-28). 68

The activity of a broad panel of dietary fatty acids was recently screened on human FFA1 and FFA4 in 69

vitro and demonstrated that the polyunsaturated fatty acid pinolenic acid is a dual FFA1/FFA4 agonist with a 70

Journ

al Pre-

proof

5

combined high potency and high efficacy on both FFA1 and FFA4 (29). This was supported by follow-up 71

experiments in mice showing that in particular purified pinolenic acid but also Siberian pine nut oil (containing 72

~20% pinolenic acid) reduced blood glucose levels during an oral glucose tolerance test (OGTT) in mice when 73

compared to maize oil (29). The more pronounced effect of purified pinolenic acid suggests that hydrolysis of 74

pine nut oil into FFAs might enhance the effect. To investigate the long term effect of hydrolyzed pine nut oil, a 75

longitudinal study of diet-induced obese mice evaluated hydrolyzed Siberian pine nut oil versus a water control 76

and found significantly lower fasting insulin and OGTT glucose levels after 21 days of supplementation 77

(Wargent, E. et al., personal communication, unpublished data in preparation). 78

As described, enteroendocrine fatty acid activation of FFA1 and FFA4 are able to regulate GLP-1 and 79

ghrelin secretion, and one important function mediated by these hormones is appetite regulation (30-33). In 80

accordance, it was shown that 3 g pine nut oil given as FFA but not as triglycerides (TG) was able to increase 81

GLP-1 and reduce feelings of prospective food consumption compared to olive oil, however no differences in 82

ghrelin was observed (34). Moreover, another study showed reduced food intake after pine nut oil given as FFA 83

but not as TG (35). Consistent with a study by Verhoef et al. reporting no effect of unhydrolyzed pine nut oil on 84

either satiety or energy intake, this provide further evidence that hydrolysis of pine nut oil may be a key element 85

(36). 86

Another possible important factor, that may enhance the effect of pine nut oil FFA mediated incretin 87

secretion through FFA1 and FFA4, is the delivery of FFA distal to the normal absorption site in the upper 88

gastrointestinal tract. This gut section is bypassed after Roux-en-Y gastric bypass (RYGB), resulting in increased 89

loads of nutrients, including FFAs, to the distal intestine, which strongly amplifies GLP-1 secretion (37). It is 90

possible that activation of FFA1 and FFA4 on the enteroendocrine L-cells contributes to this amplification, due 91

to their increased abundance in this gut section (38). In support of this, purified lauric acid, ingested as enteric-92

coated pellets, designed to release its content throughout the ileum and colon was demonstrated to reduce 93

glucose levels and enhance GLP-1 secretion compared to placebo (39). 94

Journ

al Pre-

proof

6

In the present study, we hypothesized that hydrolyzed Siberian pine nut oil (containing ~20% pinolenic 95

acid) delivered to the small intestine by delayed-release capsules, mimicking RYGB, would increase incretin 96

release and insulin secretion by activation of FFA1 and FFA4 in enteroendocrine cells and FFA1 in pancreatic 97

beta-cells resulting in improved glucose tolerance and together with an FFA4-mediated inhibition of ghrelin 98

release reduce appetite. The main objectives were to investigate the effect of hydrolyzed Siberian pine nut oil 99

(PNO-FFA) versus unhydrolyzed Siberian pine nut oil (PNO-TG) and the dose-response relationship of PNO-100

FFA compared to no oil intake on glucose tolerance, insulin, incretin and ghrelin secretion, gastrointestinal 101

tolerability and appetite in healthy humans. 102

Journ

al Pre-

proof

7

MATERIALS AND METHODS 103

The studies were conducted at Odense University Hospital, Department of Endocrinology, Denmark, and were 104

approved by The Regional Committees on Health Research Ethics for Southern Denmark (s-20150060). They 105

were conducted in accordance with the principles of the Helsinki Declaration and written informed consent was 106

obtained from all participants prior to enrollment. 107

Inclusion and exclusion criteria 108

In study 1, the objective was to compare the effect of PNO-FFA versus PNO-TG. Inclusion was based on the 109

following criteria: age between 20-50 years, body mass index (BMI) of 18.5-30 kg/m2, normal glucose tolerance 110

(two hours OGTT plasma glucose < 7.8 mmol/L), normal blood pressure (< 140/90 mmHg) and normal 111

screening blood samples (kidney and liver function, lipids and hematology within Danish reference ranges) and 112

written informed consent. Exclusion criteria were: any chronic disease including past gastrointestinal diseases or 113

gastrointestinal surgery, first degree relatives with diabetes, food allergies of relevance, need for prescriptive 114

medicine, smoking, body weight changes (> 3 kg within three month prior), intake of dietary supplements (<1 115

month prior) or any type of restrictive diet for example calorie restriction, vegan diet etc. In study 2, the 116

objective was to study the dose-response relationship of PNO-FFA compared with no oil. Inclusion and 117

exclusion criteria were similar to study 1, except for age between 40-70 years and a BMI of 27.5-40 kg/m2. 118

Study 2 was performed after study 1, and the change in criteria for age and BMI was chosen to rule out that lack 119

of an effect of pine nut oil on plasma glucose in study 1 was caused by the inclusion of young and healthy 120

individuals. Eight out of nine included subjects completed the protocol in study 1, and 10 out of 11 in study 2. In 121

both studies, one subject dropped out due to personal time constraints. Baseline characteristics for participants in 122

both studies are shown in Table 1. 123

Experimental procedure 124

In addition to the screening OGTT (no oil intake), two OGTTs were conducted in combination with either 3 g 125

PNO-TG or 3 g PNO-FFA in study 1, and either 3 g PNO-FFA or 6 g of PNO-FFA in study 2, in both studies 126

Journ

al Pre-

proof

8

using a randomized cross-over design. Supplementary Fig. 1 show a design chart of the two studies. 127

Randomization was computerized using www.randomizer.org. Prior to all experimental days, each subject was 128

instructed to consume a similar evening meal of their own preference (no calorie restriction). They showed up 129

after an overnight fast of at least ten hours; a small amount of water was allowed. Moreover, alcohol 130

consumption and physical exercise was not allowed 48 hours prior to the OGTTs. Blood was drawn using a 131

peripheral venous catheter at time points -30, -15 and 0 min prior to consumption of a 75 g glucose solution (250 132

mL), and hereafter every 30 min until 240 min of testing. The second and third OGTT were similar to the 133

screening OGTT with the exception of oil intake 30 minutes prior to the test. The oils were administered in 134

delayed-release capsules consumed with water (320 mL). Washout between tests were at least one and maximum 135

four weeks. Moreover, all subjects were instructed to maintain their habitual life style throughout the study 136

period. We measured body weight and body composition at every visit using a Tanita Body Composition 137

Analyzer (model TBF-300GS). Study 1 was conducted from February to April 2016 and study 2 from January to 138

August 2017. 139

Visual analog scale (VAS) questionnaires 140

During each OGTT a set of standard VAS questionnaires were handed out to the participants to assess 141

gastrointestinal tolerability and appetite (40, 41). In these, the degree of symptoms were indicated on a 100 mm 142

horizontal line, best feeling indicated as 0 mm and worst feeling as 100 mm. VAS on gastrointestinal symptoms 143

(flatulence, nausea/vomiting, bloating, diarrhea, constipation, and abdominal pain) were provided before (-15 144

min) and after (240 min) the OGTT. Moreover, two additional retrospective questionnaires were completed at 8 145

PM and at 8 AM after the OGTT to detect any longer-term effects. Appetite symptoms (satiety, hunger, fullness, 146

prospective food consumption and overall well-being) were assessed at minutes: -15, 15, 30, 45, 60 and every 147

half hour throughout the OGTT. 148

Fasting and OGTT derived indices of insulin sensitivity and beta-cell function 149

Journ

al Pre-

proof

9

Homeostatic model assessment of insulin resistance (HOMA-IR) was calculated as: fasting glucose mmol/L x 150

fasting insulin mU/L/22.5 (42). Indices of insulin sensitivity were: BIGTTSi, based on plasma glucose and serum 151

insulin measured at 0, 30 and 120 min, MATSUDA, based on plasma glucose and serum insulin measured at 0, 152

30, 60, 90 and 120 min, and OGIS from 0-180 min, based on plasma glucose measured at 0, 120 and 180 min 153

and serum insulin measured at 0 and 120 min (43-45). Indices of beta-cell function were: BIGTTAIR, based on 154

plasma glucose and serum insulin measured at 0, 30 and 120 min, CIR30min, based on plasma glucose and serum 155

insulin at 30 min and IGI, based on plasma glucose and serum insulin at 0 and 30 min (43, 46, 47). Finally, 156

disposition indexes, that is, beta-cell function adjusted for insulin sensitivity were: BIGTTSi x BIGTTAIR, CIR x 157

MATSUDA and IGI x MATSUDA. 158

Biochemical analyses 159

Blood glucose was measured using the ABL800 FLEX blood gas analyzer. Insulin and C-peptide were analyzed 160

using an electrochemiluminescence immunoassay (ECLIA) on Cobas e 411. The insulin intra-assay CV% was 161

1.9-2.0 and the inter-assay CV% was 2.5-2.6. For C-peptide the intra-assay CV% was 1.3-4.6 and the inter-assay 162

CV% was 1.8-5.0. Moreover, we determined GLP-1 levels using Total GLP-1 (ver. 2) Assay Kit (Meso Scale 163

Discovery). GIP and ghrelin were measured using the corresponding Human GIP (Total) or Human Ghrelin 164

(Total) ELISA Kit (Millipore). The GIP kit had an intra-assay CV% of 3.0-8.8 and an inter-assay CV% of 1.8-165

6.1 and the ghrelin kit had an intra-assay CV% of 0.90-1.91 and an inter-assay CV% of 5.18-7.81. 166

Investigational product and blinding 167

Hydrolysis of Siberian pine nut oil (The Siberian Pines Company) was conducted in the following manner: a 168

Blue Cap flask containing PNO-TG (0.16 mol, 141 g was charged with aqueous NaOH (Panreac Applichem, 169

pharmagrade 2M, 1 mmol, 480 mL.) and a magnetic stir bar. The biphasic solution was stirred at 300 rpm 170

(stirring rate is crucial to avoid emulsion) for 24 hours after which a white solid had formed. The aqueous phase 171

was decanted off and the remaining white solid was suspended in aqueous citric acid (Sigma-Aldrich, Ph. Eur. 172

grade, 2 M, 0.95 mmol, 480 mL). The suspension was stirred at 400 rpm (stirring rate is crucial to avoid 173

Journ

al Pre-

proof

10

emulsion) for 24 hours resulting in a biphasic solution. The golden oil was siphoned off and carefully washed 174

with H2O (40 mL). The biphasic solution was stirred slowly until complete separation of the two phases. The 175

golden oil was siphoned off resulting in the pure PNO-FFA. PNO-FFA was subjected to H1 NMR analysis to 176

ensure complete hydrolysis. The major fatty acid types in the Siberian pine nut oil is pinolenic acid (20.9 %), 177

linoleic acid (47.7 %) and oleic acid (21.8 %) (29). 178

To mimic the effects of RYGB, the pine nut oil (PNO-FFA and PNO-TG) was hand-filled in 179

semitransparent, acid resistant capsule shells made of hydroxypropyl methylcellulose with delayed release 180

properties (DRcaps™ from Capsugel®, size 00 batch 5332311). Doses of 3 g and 6 g amounted to five and ten 181

capsules containing a total of ~114 and 228 kJ of fat, respectively. The encapsulated oil was stored at 5 °C and 182

for a maximum of 3 months until use. Subjects in study 1 were blinded towards oil type, however differences in 183

number of capsules did not allow for blinding of subjects in study 2. DRcaps™ capsule shells were originally 184

designed for dry powder filling. To test the release profile of oils, we did a simple in vitro evaluation of capsules 185

containing mannitol (dry matter control), PNO-TG or PNO-FFA, supplemented with the dyes rhodamine B or 186

sulphorhodamine B (in the case of PNO-FFA), in triplicates. The capsules were placed in individual vials with 187

magnetic stirring containing 10 mL fasted state simulated gastric fluid without pepsin (biorelevant.com) (pH = 188

1.5, 37 °C, 400 rpm) and the release was followed by UV-measurements (550 nm). We observed a similar, yet 189

slower release profile for mannitol powder as previously described (i.e. in vivo disintegration time approx. 50 190

min, and content release completed after another 20 min)(48). For the oils, we observed a slow release over four 191

hours, where approximately half of the oil was released at 90 minutes (Supplementary Fig. 2). 192

Outcomes 193

Outcomes for both studies were treatment differences in 4-h total area under the curve (AUCtotal) of glucose, 194

insulin, C-peptide, GLP-1, GIP and ghrelin, as well as the total AUC for the first two hours (AUC0-120) and the 195

second two hours (AUC120-240). Additional outcomes were treatment differences in OGTT based indices of 196

Journ

al Pre-

proof

11

insulin sensitivity and beta-cell function and VAS scores of gastrointestinal symptoms and appetite. All AUCs 197

were calculated using the trapezoidal rule. 198

Statistics 199

All statistics were conducted using Stata/IC 14.2 package. Statistical significance was set to a p-value of < 0.05. 200

Mixed model linear regression analysis was chosen for hypothesis testing of treatment differences between oil 201

types (study 1) or oil doses (study 2). The explanatory and fixed effect variable was oil type or dose and the 202

random variable was subject ID. Non-normal data were transformed using the natural logarithm. Model 203

validation was conducted by visual assessment of q-q plots of the obtained residuals and plots of residuals versus 204

estimated values. Blood biomarkers below detection limit were imputed to the detection limit of the assay. In 205

study 2, one subject was omitted from all analyses after 3 g PNO-FFA treatment due to suspected non-206

compliance. In case of missing appetite responses (missing completely at random) these remained missing 207

during analysis and results were reported as model estimations. Treatment differences in VAS scores of 208

gastrointestinal discomfort were analyzed by the use of non-parametric Wilcoxon signed rank test adjusted for 209

multiple comparison by the Bonferroni-Holm method. 210

Journ

al Pre-

proof

12

RESULTS 211

Glucose, insulin and C-peptide 212

In study 1, neither glucose AUCtotal nor insulin AUCtotal differed significantly between treatments, whereas C-213

peptide AUCtotal tended to be increased in response to PNO-TG versus no oil (p=0.075) (Fig. 1 a-f). However, 214

glucose AUC0-120 tended to be lower in response to PNO-FFA compared to both PNO-TG (p=0.064) and no oil 215

(p=0.072), whereas no treatment differences in glucose AUC120-240 were observed (Fig. 1 a-b). Insulin AUC0-120 216

and C-peptide AUC0-120 were reduced (~10-16%) in response to PNO-FFA versus both PNO-TG and no oil (all 217

p<0.05), whereas insulin AUC120-240 and C-peptide AUC120-240 were increased (~22-35%) after intake PNO-FFA 218

compared to no oil (all p<0.05) with only a tendency to increased C-peptide AUC120-240 after PNO-TG versus no 219

oil (p=0.059) (Fig. 1 c-f). 220

In study 2, there was no significant treatment differences in glucose AUCtotal, whereas insulin AUCtotal 221

decreased (~13%; p=0.03) and C-peptide AUCtotal tended to decrease (p=0.059) after intake of 6 g PNO-FFA 222

versus no oil (Fig. 2 a-f). Glucose AUC0-120 was reduced (~9-10%) in response to 6 g PNO-FFA compared to 223

both 3 g PNO-FFA and no oil (all p<0.05), whereas glucose AUC120-240 was increased (~13%) after 6 g PNO-224

FFA versus no oil (p=0.002) and tended to be increased in response to 3 g PNO-FFA versus no oil (p=0.057). 225

Consistent with study 1, insulin AUC0-120 and C-peptide AUC0-120 were reduced (~17-22%) in response to 6 g 226

PNO-FFA compared to no oil (all p<0.05) (Fig. 2 c-f). 227

GLP-1, GIP and ghrelin 228

In study 1, GLP-1 AUCtotal was increased (~63%) in response to PNO-FFA compared to no oil (p=0.003) and 229

tended to be increased after PNO-FFA versus PNO-TG (p=0.09) (Fig. 3 a-b). These effects of PNO-FFA were 230

mainly explained by increased GLP-1 AUC120-240 (~86-156%) compared with both PNO-TG (p=0.004) and no 231

oil (p<0.001), whereas only a tendency of increased GLP-1 AUC0-120 after PNO-FFA versus no oil (p=0.079) 232

was observed. Moreover, GLP-1 AUC120-240 tended to be increased after PNO-TG versus no oil (p=0.055) (Fig. 3 233

a-b). While GIP AUCtotal was unaltered by treatment, GIP AUC0-120 decreased (~20%; p=0.021) whereas GIP 234

Journ

al Pre-

proof

13

AUC120-240 increased (~43%; p=0.011) after PNO-FFA versus no oil (Fig. 3 c-d). The orexigenic hormone 235

ghrelin showed reduced AUCtotal (~13-14%; all p<0.05) and AUC120-240 (~15-25%, all p<0.01) in response to 236

both PNO-FFA and PNO-TG compared to no oil (Fig. 3 e-f). 237

In study 2, GLP-1 AUCtotal was increased (~26%) in response to 6 g PNO-FFA compared to no oil 238

(p=0.024). Consistent with study 1, these effects of 6 g PNO-FFA were mainly explained by increased GLP-1 239

AUC120-240 (~35-52%) compared with both 3 g PNO-FFA (p=0.002) and no oil (p<0.001), whereas no treatment 240

differences were observed in the first two hours (Fig. 4 a-b). GIP AUC responses showed no differences between 241

treatments (Fig. 4 c-d). As observed in study 1, ghrelin AUCtotal were reduced (~30%) after intake of 6 g PNO-242

FFA compared to no oil (p=0.02), and ghrelin AUC120-240 decreased (~30-36%) in response to 3 g PNO-FFA 243

(p=0.013) and 6 g PNO-FFA (p<0.001) compared to no oil (Fig. 4 e-f). 244

OGTT indices 245

In study 1, insulin sensitivity estimated as BIGTTSi was slightly higher after PNO-FFA than after PNO-TG 246

(p=0.015), but there was no effect of either PNO-FFA or PNO-TG on BIGTTSi compared with no oil. Otherwise, 247

we observed no differences in insulin sensitivity measured as MATSUDA or OGIS, β-cell function measured as 248

BIGTTAIR, CIR30min or IGI, or the three estimated disposition indices between treatments. 249

In study 2, we observed no effects of 6 g PNO-FFA and 3 g PNO-FFA on any of these markers of 250

insulin sensitivity, insulin secretion and disposition indices compared with no oil. Data are presented in 251

supplementary Table 1. 252

Appetite 253

In study 1, satiety AUCtotal and AUC120-240 were increased (~17-24%) after PNO-TG versus no oil (all p<0.05), 254

whereas hunger AUCtotal and AUC120-240 were reduced (~10-11%) in response to PNO-TG versus no oil (all p 255

<0.05). Moreover, hunger AUC120-240 was also decreased (~9%) after PNO-TG versus PNO-FFA (p=0.007). 256

Lastly, fullness AUC120-240 was increased (~22%) after PNO-TG versus no oil (p=0.02), and correspondingly 257

Journ

al Pre-

proof

14

prospective food consumption AUC120-240 was reduced (~8%) after PNO-TG versus no oil (p=0.001). All other 258

symptoms did not differ between treatments (Supplementary Table 2). 259

In study 2, satiety AUCtotal and AUC120-240 were increased (~17-40%) after both 3 g PNO-FFA and 6 g 260

PNO-FFA compared with no oil (all p<0.03). Also, satiety AUC120-240 was increased (~14%) after 6 g PNO-FFA 261

versus 3 g PNO-FFA (p=0.046). Fullness AUCtotal, AUC-15-120 and AUC120-240 were increased (~22-51%) after 262

both 3g PNO-FFA and 6 g PNO-FFA versus no oil (all p<0.01). In agreement, the AUCtotal for prospective food 263

intake was reduced (~7-10%) after both 3 g PNO-FFA and 6 g PNO-FFA versus no oil (all p<0.05). After 3 g 264

PNO-FFA prospective food intake AUC-15-120 was reduced versus no oil (~11%; p=0.014), whereas prospective 265

food intake AUC120-240 was reduced after 6 g PNO-FFA versus no oil (~3%; p=0.03). Finally, overall well-being 266

AUCtotal and AUC-15-120 were increased after no oil compared to 6 g PNO-FFA (p=0.01) (Supplementary Table 267

2). 268

Gastrointestinal tolerability 269

In study 1, there were no differences in changes of the degree of gastrointestinal symptoms during the OGTT nor 270

at 8PM or at 8AM between any of the treatments (Supplementary Table 3). 271

In study 2, the degree of abdominal pain was increased after 6 g PNO-FFA versus 3 g PNO-FFA 272

(p=0.048) and no oil (p=0.02), and also after 3 g PNO-FFA versus no oil (p=0.048). Otherwise, no differences in 273

changes of the degree of gastrointestinal symptoms during the OGTT, at 8PM or at 8AM were seen between the 274

treatments (Supplementary Table 3). 275

Journ

al Pre-

proof

15

DISCUSSION 276

The major aim of the present study was to explore the effect of PNO-FFA containing ~20% of the dual 277

FFA1/FFA4 agonist pinolenic acid on metabolic outcomes relevant in obesity and type 2 diabetes, including 278

glucose tolerance, insulin response, incretin and ghrelin secretion, and appetite. We found increased GLP-1 279

secretion from 120-240 min of the OGTT after both 3 g PNO-FFA (study 1) and 6 g PNO-FFA (study 2), which 280

was accompanied by decreased circulating levels of the orexigenic hormone ghrelin versus no oil. Interestingly, 281

in study 2, these changes were accompanied by subjective appetite reducing effects. Gastrointestinal tolerability 282

was high, with only a minor increase in abdominal pain after both 3 g and 6 g PNO-FFA in study 2. 283

Unexpectedly, insulin levels were slightly decreased from 0-120 min of the OGTT after both 3 g PNO-FFA 284

(study 1) and 6 g PNO-FFA (study 2). After 6 g PNO-FFA this was accompanied by reduced glucose levels. 285

Overall, our results suggest that delayed-release PNO-FFA has an appetite reducing effect, which may be 286

mediated by increased GLP-1 secretion and reduced ghrelin levels in glucose-tolerant individuals. 287

Based on the study by Christiansen et al, demonstrating a high agonistic potency and superior efficacy of 288

pinolenic acid on FFA1 and FFA4 in vitro, as well as improved glucose tolerance in mice after intake of pine nut 289

oil given as TG and, in particular, as purified pinolenic acid [27], we hypothesized that hydrolyzed pine nut oil 290

delivered to the small intestine by delayed-release capsules would increase insulin secretion by activation of 291

FFA1/FFA4 either on enteroendocrine cells leading to increased incretin secretion or on pancreatic beta-cells 292

stimulating insulin release directly. Only a limited number of previous intervention studies have investigated the 293

effect of pine nut oil on glucose, insulin, appetite regulating hormones and subjective appetite (VAS) (34-36). 294

None of these studies used delayed-release formulation. However, in a recent study, it was reported that purified 295

lauric acid (2.35 g) administered twice in the form of enteric coated pellets for delivery to ileum and colon, in 296

combination with a mixed meal breakfast and lunch, was able to reduce postprandial breakfast and lunch AUC 297

of glucose and increase postprandial lunch AUC of GLP-1 in type 2 diabetic subjects (n=8) (39). These findings 298

partly correspond to our results, showing a pronounced increase in GLP-1 levels from 120-240 min during the 299

OGTT in both studies for the 3 g PNO-FFA (study 1) and 6 g PNO-FFA (study 2) treatment versus no oil. 300

Journ

al Pre-

proof

16

Despite the delayed increase in GLP-1 AUC from 120-240 min for the 3 g PNO-FFA in study 1 and the 6 g 301

PNO-FFA in study 2, we observed only a small increase in insulin AUC from 120-240 min of the OGTT after 3 302

g PNO-FFA in study 1. The lack of an incretin effect, evaluated as increased insulin secretion, could be 303

explained by the fact that the insulinotropic effect of GLP-1 is mediated by glucose levels above normoglycemia 304

(49). In our studies, glucose concentrations generally returned to fasting levels after 120 min of the OGTTs, after 305

which the augmented GLP-1 secretion in response to PNO-FFA was observed. It could be speculated, that in 306

individuals with pre-diabetes or type 2 diabetes, the insulinotropic effect of GLP-1 would have been more 307

pronounced, as they exhibit higher and longer lasting postprandial glucose levels compared to healthy, glucose 308

tolerant individuals. However, this was not the case in the study of lauric acid in patients with type 2 diabetes 309

(39), in which no effect on insulin secretion was observed despite reduced glucose levels and increased GLP-1 310

secretion (39). Though not suggested by Ma et al, an explanation of the reduction in glucose by lauric acid might 311

be mediated by an acute improvement of insulin sensitivity. However, in our studies, we did not observe any 312

consistent effects of PNO-FFA or PNO-TG on fasting or OGTT derived indices of insulin sensitivity or insulin 313

secretion compared with no oil. Instead, Ma et al argued that the glucose lowering effects could be mediated by 314

GLP-1 through activation of GLP-receptors in the hepatic portal vein shown to stimulate glucose uptake without 315

any alteration in insulin levels (39). The lack of changes in indices of beta-cell function in response to pine nut 316

oil was observed in the absence of an early increase in incretins (0-120 min). This argues against a direct FFA1-317

mediated effect of pinolenic acid on beta-cell function in humans, at least in the design and doses used in the 318

present study. 319

Interestingly, in study 1, we observed only a tendency of PNO-TG to increase GLP-1 from 120-240 min 320

of the OGTT versus no oil, whereas, the same 3 g dose of PNO-FFA caused a significant and more pronounced 321

increase in GLP-1 AUC versus both PNO-TG and no oil. These results are similar to findings reported by 322

Pasmann et al (34), where 3g of pine nut oil FFA in combination with a standard breakfast increased GLP-1 323

secretion at 60 min and non-significantly from 120-240 min compared to olive oil in overweight women (n=18), 324

while no difference in effect was observed for the same dose of pine nut oil triglycerides (34). These results 325

Journ

al Pre-

proof

17

provide evidence that pine nut oil given in the form of FFAs may be more powerful to stimulate GLP-1 release 326

than genuine pine nut oil (triglycerides). 327

Ghrelin is a known orexigenic hormone (50), and therefore a reduction in the postprandial release may 328

be beneficial in overweight or type 2 diabetic subjects, including a reduction in appetite and energy intake, 329

which may induce negative energy balance and over time body weight loss. Ghrelin secretion is inhibited by 330

activation of FFA4, but also other unknown mechanisms may contribute as ghrelin inhibition was reported to be 331

preserved in FFA-4 knockout models (23). Pinolenic acid may be superior in causing a FFA4-mediated 332

inhibition of ghrelin release due to its high potency on the receptor compared to other fatty acids (29). In both 333

studies, we demonstrated a decreased ghrelin AUC from 120-240 min in response to all pine nut oil interventions 334

versus no oil. These results suggest that pine nut oil given either as triglycerides or FFAs during an OGTT are 335

capable of reducing ghrelin levels more than glucose alone. As we did not include a control oil, we cannot 336

conclude that the observed ghrelin suppression is superior to other types of oil. It has been shown that ghrelin 337

inhibition is energy dependent (51) and therefore, we cannot exclude the possibility that the observed reduction 338

in ghrelin levels in response to pine nut oil could be explained by the extra energy (8-15%) consumed. However, 339

it has recently been suggested that mono and polyunsaturated fatty acids, which is the main type of fatty acids 340

contained in pine nut oil (~90%) may be superior in suppressing ghrelin secretion compared to saturated fatty 341

acids (52). Thus, it is possible that pine nut oil may be a better inhibitor of ghrelin secretion compared to other 342

oils depending on the amount of unsaturated fatty acids. The ghrelin response to pine nut oil was also evaluated 343

in the study by Pasmann et al (34). Here, none of the tested pine nut oils reduced ghrelin levels more than the 344

placebo oil (olive). Olive oil contains ~80% unsaturated fatty acids (53) and may therefore be equally effective 345

in inhibiting ghrelin release compared to pine nut oil or pinolenic acid. However, in our studies, ghrelin was 346

suppressed until 120 min and did not reach fasting levels within the 240 min of testing, while ghrelin was only 347

suppressed until 90 min and returned to fasting levels by 180 min in the study by Pasmann et al (34). Thus, 348

ghrelin was suppressed for a longer duration of time in our studies, which may be explained by enhanced FFA4 349

activation likely obtained by the delayed release formulation. Finally, in study 2, we detected no differences 350

Journ

al Pre-

proof

18

between the 3 g and 6 g PNO-FFA treatments in ghrelin suppression. This suggests that 3g PNO-FFA was 351

sufficient to cause this suppression compared to glucose alone. 352

Consistent with the study by Pasmann et al, (34), we observed an increase in satiety and fullness after 3 353

g and 6 g PNO-FFA compared to no oil in study 2. For satiety, this increase was observed for the total period of 354

testing and from 120-240 min corresponding to the observed postprandial increase in GLP-1 and decrease in 355

ghrelin secretion after intake of 6 g PNO-FFA. However, in study 2 fullness showed a significant increase in 356

both periods (-15-120 min and 120-240 min) indicating that the intake of capsules versus no oil rather than 357

changes in circulating levels of GLP-1 and ghrelin contributes to fullness also in the early phase of an OGTT. 358

Intriguingly, only 3g PNO-TG significantly increased satiety and fullness and reduced hunger in the direction of 359

appetite reduction in study 1. However, in line with our results in study 2, Hughes et al demonstrated that 2 g 360

pine nut oil FFA, but not TG, given 30 min prior to an ad libitum buffet lunch induced a significant 9% 361

reduction in food intake and tended to reduce energy intake (35). Lending further support to a superior effect of 362

pine nut oil FFAs on appetite reduction, a study by Verhoef et al. reported no effect of 3 g or 6 g pine nut oil TG 363

on subsequent ad libitum energy intake compared with 6 g milk fat (36). When considering the studies 364

investigating effects of PNO-FFA or PNO-TG on appetite as a whole, including our results, there are a number 365

of important differences between the designs including timing of oil intake, formulation, sample size and 366

metabolic state of participants. The latter is also a factor when comparing outcomes from our studies, as we 367

included older subjects with a higher BMI in study 2 compared to study 1. These differences might explain the 368

inconsistencies we observed in subjective appetite. Nonetheless, there is suggestive evidence that PNO-FFA may 369

reduce appetite, even when given in small doses in more physiological situations than during an OGTT as 370

reflected in studies using mixed meals and buffet style energy intake assessments (34, 35). However, more 371

robust, long-term studies are needed, to fully establish the potential appetite lowering effect of PNO-FFA, and to 372

what extent this effect might translate into reduced energy intake and weight loss over time. 373

Importantly, we detected no major negative effects of pine nut oil on gastrointestinal tolerability during 374

the OGTT, at 8 PM or 8 AM after. However, in study 2, we did observe a small increase in abdominal pain 375

Journ

al Pre-

proof

19

during the OGTT after 6 and 3 g PNO-FFA versus no oil, and also between doses. The mean treatment 376

difference in changes of abdominal pain was 11.8 mm between 6 g PNO-FFA and no oil, which we considered 377

small, but still of potential relevance. However, the treatment difference of 0.7 mm between 3 g PNO-FFA and 378

no oil is in our opinion not clinically relevant. In the study by Ma et al, investigating lauric acid, they did not 379

evaluate gastrointestinal symptoms; however, they reported that no adverse events occurred during the study 380

(39). As previously discussed their oil dose was 2.35 g, which almost corresponds to the 3 g PNO-FFA doses in 381

our studies. Thus, it is likely that smaller doses do not induce relevant gastrointestinal tolerability issues, but we 382

cannot exclude the possibility that giving the larger 6 g dose may pass a threshold, which may induce a few 383

adverse gastrointestinal symptoms. On the other hand, in study 2 subjects were unblinded, which may have 384

contributed to subjects being more susceptible to record a worsening in symptoms on the 6 g dose treatment. 385

Our studies have some limitations including the relatively small sample sizes, which may have 386

compromised the power of the studies to detect relevant differences. At the same time, the possibilities of type 1 387

errors were present, due to the potential influence of outlying results. Therefore, the studies should be seen as 388

explorative and hypothesis generating. While our studies demonstrated relevant effects of PNO-FFA on GLP-1, 389

ghrelin and appetite measures, there were some inconsistencies with respect to 3 g PNO-FFA in the two studies. 390

Thus, in study 2, no effect of 3 g PNO-FFA on GLP-1 (all time periods) or ghrelin (0-240 min) was observed, 391

whereas in study 1, intake of 3 g PNO-FFA showed no appetite reducing effects. Moreover, the changes in the 392

circulating levels of GLP-1 and ghreline did not match the self-reported appetite scores in several of the tested 393

conditions. Whether these differences could be explained by inclusion of older individuals with a higher BMI 394

and fat mass in study 2 as well as the large variation in the VAS-scores needs to be addressed in future studies. 395

Another potential limitation is that we did not measure the active forms of GLP-1, GIP or ghrelin. However, at 396

least for GLP-1 it has been argued that in praxis measurement of total GLP-1 is also of biological relevance (37). 397

We did not measure in vivo gastric emptying or release time of oil content from the capsules, which may be 398

important explanatory factors. Nonetheless, we believe that the manner in which ghrelin suppression differed 399

from other studies seemed plausible to depend on delayed and slow release of oil content. Lastly, we did not 400

Journ

al Pre-

proof

20

measure energy intake, thus whether the observed changes in subjective appetite would lead to a reduction in 401

energy intake is unknown. 402

In conclusion, PNO-FFA delivered to the small intestine by delayed-release capsules augmented GLP-1 403

secretion 120-240 min after glucose intake, which was accompanied by a concomitant attenuation of ghrelin 404

levels as compared with intake of glucose alone. These changes in circulating levels of GLP-1 and ghrelin could 405

explain some of the observed reductions in subjective appetite, supporting an appetite reducing effect of PNO-406

FFA. Finally, the oil interventions were well tolerated, with only the 6 g PNO-FFA dose causing minor, but 407

potentially clinically relevant changes in self-reported abdominal pain. Importantly, the suggested beneficial 408

effects of delayed-release PNO-FFA observed in our studies need to be evaluated and confirmed in long term 409

studies, preferably including patients with type 2 diabetes, in whom potential beneficial effects beyond weight 410

loss may include improved glucose tolerance, insulin sensitivity and reduced low-grade inflammation. 411

Journ

al Pre-

proof

21

FUNDING SOURCES 412

This work was supported by the Danish Research Council for Strategic Research and Innovation Fund Denmark 413

[grant number 11-116196 and 0603-00452B]. In addition, the study was supported by grants from Odense 414

University Hospital, The Region of Southern Denmark, the University of Southern Denmark and Ingemann O. 415

Bucks Fund. 416

Journ

al Pre-

proof

22

CONFLICT OF INTEREST 417

The authors declare that they have no conflict of interest. 418

Journ

al Pre-

proof

23

REFERENCES 419

1. Zheng Y, Ley SH, Hu FB. Global aetiology and epidemiology of type 2 diabetes mellitus and its 420

complications. Nature reviews Endocrinology. 2018;14(2):88-98. 421

2. Sortso C, Green A, Jensen PB, Emneus M. Societal costs of diabetes mellitus in Denmark. Diabetic 422

medicine : a journal of the British Diabetic Association. 2016;33(7):877-85. 423

3. Davies MJ, D'Alessio DA, Fradkin J, Kernan WN, Mathieu C, Mingrone G, et al. Management of 424

hyperglycaemia in type 2 diabetes, 2018. A consensus report by the American Diabetes Association (ADA) and 425

the European Association for the Study of Diabetes (EASD). Diabetologia. 2018;61(12):2461-98. 426

4. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al. Management of 427

hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the 428

American Diabetes Association and the European Association for the Study of Diabetes. Diabetes care. 429

2015;38(1):140-9. 430

5. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al. Management of 431

hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes 432

Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes care. 433

2012;35(6):1364-79. 434

6. Hara T, Ichimura A, Hirasawa A. Therapeutic role and ligands of medium- to long-chain Fatty Acid 435

receptors. Front Endocrinol (Lausanne). 2014;5:83. 436

7. Ichimura A, Hasegawa S, Kasubuchi M, Kimura I. Free fatty acid receptors as therapeutic targets for the 437

treatment of diabetes. Frontiers in pharmacology. 2014;5:236. 438

8. Watterson KR, Hudson BD, Ulven T, Milligan G. Treatment of Type 2 Diabetes by Free Fatty Acid 439

Receptor Agonists. Frontiers in Endocrinology. 2014;5. 440

Journ

al Pre-

proof

24

9. Li Z, Qiu Q, Geng X, Yang J, Huang W, Qian H. Free fatty acid receptor agonists for the treatment of type 441

2 diabetes: drugs in preclinical to phase II clinical development. Expert opinion on investigational drugs. 442

2016;25(8):871-90. 443

10. Riddy DM, Delerive P, Summers RJ, Sexton PM, Langmead CJ. G Protein-Coupled Receptors Targeting 444

Insulin Resistance, Obesity, and Type 2 Diabetes Mellitus. Pharmacological reviews. 2018;70(1):39-67. 445

11. Briscoe CP, Tadayyon M, Andrews JL, Benson WG, Chambers JK, Eilert MM, et al. The orphan G protein-446

coupled receptor GPR40 is activated by medium and long chain fatty acids. The Journal of biological chemistry. 447

2003;278(13):11303-11. 448

12. Ichimura A, Hirasawa A, Hara T, Tsujimoto G. Free fatty acid receptors act as nutrient sensors to 449

regulate energy homeostasis. Prostaglandins Other Lipid Mediat. 2009;89(3-4):82-8. 450

13. Itoh Y, Kawamata Y, Harada M, Kobayashi M, Fujii R, Fukusumi S, et al. Free fatty acids regulate insulin 451

secretion from pancreatic beta cells through GPR40. Nature. 2003;422(6928):173-6. 452

14. Edfalk S, Steneberg P, Edlund H. Gpr40 is expressed in enteroendocrine cells and mediates free fatty 453

acid stimulation of incretin secretion. Diabetes. 2008;57(9):2280-7. 454

15. Hara T, Kimura I, Inoue D, Ichimura A, Hirasawa A. Free Fatty Acid Receptors and Their Role in 455

Regulation of Energy Metabolism. In: Nilius B, Amara SG, Lill R, Offermanns S, Gudermann T, Petersen OH, et 456

al., editors. Reviews of Physiology, Biochemistry and Pharmacology. 164. Cham: Springer International 457

Publishing; 2013. p. 77-116. 458

16. Reimann F, Gribble FM. Mechanisms underlying glucose-dependent insulinotropic polypeptide and 459

glucagon-like peptide-1 secretion. Journal of diabetes investigation. 2016;7 Suppl 1:13-9. 460

17. Andersen A, Lund A, Knop FK, Vilsboll T. Glucagon-like peptide 1 in health and disease. Nature reviews 461

Endocrinology. 2018;14(7):390-403. 462

Journ

al Pre-

proof

25

18. Nauck MA, Meier JJ. Incretin hormones: Their role in health and disease. Diabetes, obesity & 463

metabolism. 2018;20 Suppl 1:5-21. 464

19. Hirasawa A, Tsumaya K, Awaji T, Katsuma S, Adachi T, Yamada M, et al. Free fatty acids regulate gut 465

incretin glucagon-like peptide-1 secretion through GPR120. Nat Med. 2005;11(1):90-4. 466

20. Iwasaki K, Harada N, Sasaki K, Yamane S, Iida K, Suzuki K, et al. Free fatty acid receptor GPR120 is highly 467

expressed in enteroendocrine K cells of the upper small intestine and has a critical role in GIP secretion after fat 468

ingestion. Endocrinology. 2015;156(3):837-46. 469

21. Paulsen SJ, Larsen LK, Hansen G, Chelur S, Larsen PJ, Vrang N. Expression of the Fatty Acid Receptor 470

GPR120 in the Gut of Diet-Induced-Obese Rats and Its Role in GLP-1 Secretion. PLoS ONE. 2014;9(2):e88227. 471

22. Ekberg JH, Hauge M, Kristensen LV, Madsen AN, Engelstoft MS, Husted AS, et al. GPR119, a Major 472

Enteroendocrine Sensor of Dietary Triglyceride Metabolites Coacting in Synergy With FFA1 (GPR40). 473

Endocrinology. 2016;157(12):4561-9. 474

23. Engelstoft MS, Park WM, Sakata I, Kristensen LV, Husted AS, Osborne-Lawrence S, et al. Seven 475

transmembrane G protein-coupled receptor repertoire of gastric ghrelin cells. Molecular metabolism. 476

2013;2(4):376-92. 477

24. Lu X, Zhao X, Feng J, Liou AP, Anthony S, Pechhold S, et al. Postprandial inhibition of gastric ghrelin 478

secretion by long-chain fatty acid through GPR120 in isolated gastric ghrelin cells and mice. American journal of 479

physiology Gastrointestinal and liver physiology. 2012;303(3):G367-76. 480

25. Gong Z, Yoshimura M, Aizawa S, Kurotani R, Zigman JM, Sakai T, et al. G protein-coupled receptor 120 481

signaling regulates ghrelin secretion in vivo and in vitro. American journal of physiology Endocrinology and 482

metabolism. 2014;306(1):E28-35. 483

26. Oh DY, Talukdar S, Bae EJ, Imamura T, Morinaga H, Fan W, et al. GPR120 is an omega-3 fatty acid 484

receptor mediating potent anti-inflammatory and insulin-sensitizing effects. Cell. 2010;142(5):687-98. 485

Journ

al Pre-

proof

26

27. Oh DY, Walenta E, Akiyama TE, Lagakos WS, Lackey D, Pessentheiner AR, et al. A Gpr120-selective 486

agonist improves insulin resistance and chronic inflammation in obese mice. Nat Med. 2014;20(8):942-7. 487

28. Azevedo CM, Watterson KR, Wargent ET, Hansen SV, Hudson BD, Kepczynska MA, et al. Non-Acidic 488

Free Fatty Acid Receptor 4 Agonists with Antidiabetic Activity. Journal of medicinal chemistry. 489

2016;59(19):8868-78. 490

29. Christiansen E, Watterson KR, Stocker CJ, Sokol E, Jenkins L, Simon K, et al. Activity of dietary fatty acids 491

on FFA1 and FFA4 and characterisation of pinolenic acid as a dual FFA1/FFA4 agonist with potential effect 492

against metabolic diseases. The British journal of nutrition. 2015;113(11):1677-88. 493

30. Flint A, Raben A, Astrup A, Holst JJ. Glucagon-like peptide 1 promotes satiety and suppresses energy 494

intake in humans. The Journal of clinical investigation. 1998;101(3):515-20. 495

31. Gutzwiller JP, Drewe J, Goke B, Schmidt H, Rohrer B, Lareida J, et al. Glucagon-like peptide-1 promotes 496

satiety and reduces food intake in patients with diabetes mellitus type 2. The American journal of physiology. 497

1999;276(5 Pt 2):R1541-4. 498

32. Verdich C, Flint A, Gutzwiller JP, Naslund E, Beglinger C, Hellstrom PM, et al. A meta-analysis of the 499

effect of glucagon-like peptide-1 (7-36) amide on ad libitum energy intake in humans. The Journal of clinical 500

endocrinology and metabolism. 2001;86(9):4382-9. 501

33. Mihalache L, Gherasim A, Nita O, Ungureanu MC, Padureanu SS, Gavril RS, et al. Effects of ghrelin in 502

energy balance and body weight homeostasis. Hormones (Athens, Greece). 2016;15(2):186-96. 503

34. Pasman WJ, Heimerikx J, Rubingh CM, van den Berg R, O'Shea M, Gambelli L, et al. The effect of Korean 504

pine nut oil on in vitro CCK release, on appetite sensations and on gut hormones in post-menopausal 505

overweight women. Lipids in health and disease. 2008;7:10. 506

Journ

al Pre-

proof

27

35. Hughes GM, Boyland EJ, Williams NJ, Mennen L, Scott C, Kirkham TC, et al. The effect of Korean pine 507

nut oil (PinnoThin) on food intake, feeding behaviour and appetite: a double-blind placebo-controlled trial. 508

Lipids in health and disease. 2008;7:6. 509

36. Verhoef SP, Westerterp KR. No effects of Korean pine nut triacylglycerol on satiety and energy intake. 510

Nutrition & metabolism. 2011;8(1):79. 511

37. Holst JJ, Madsbad S. Mechanisms of surgical control of type 2 diabetes: GLP-1 is key factor. Surg Obes 512

Relat Dis. 2016;12(6):1236-42. 513

38. Gribble FM, Diakogiannaki E, Reimann F. Gut Hormone Regulation and Secretion via FFA1 and FFA4. 514

Handbook of experimental pharmacology. 2017;236:181-203. 515

39. Ma J, Checklin HL, Wishart JM, Stevens JE, Jones KL, Horowitz M, et al. A randomised trial of enteric-516

coated nutrient pellets to stimulate gastrointestinal peptide release and lower glycaemia in type 2 diabetes. 517

Diabetologia. 2013;56(6):1236-42. 518

40. Flint A, Raben A, Blundell JE, Astrup A. Reproducibility, power and validity of visual analogue scales in 519

assessment of appetite sensations in single test meal studies. International journal of obesity and related 520

metabolic disorders : journal of the International Association for the Study of Obesity. 2000;24(1):38-48. 521

41. Kristensen M, Juul SR, Sorensen KV, Lorenzen JK, Astrup A. Supplementation with dairy calcium and/or 522

flaxseed fibers in conjunction with orlistat augments fecal fat excretion without altering ratings of 523

gastrointestinal comfort. Nutrition & metabolism. 2017;14:13. 524

42. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model 525

assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in 526

man. Diabetologia. 1985;28(7):412-9. 527

Journ

al Pre-

proof

28

43. Hansen T, Drivsholm T, Urhammer SA, Palacios RT, Volund A, Borch-Johnsen K, et al. The BIGTT test: a 528

novel test for simultaneous measurement of pancreatic beta-cell function, insulin sensitivity, and glucose 529

tolerance. Diabetes care. 2007;30(2):257-62. 530

44. Mari A, Pacini G, Murphy E, Ludvik B, Nolan JJ. A model-based method for assessing insulin sensitivity 531

from the oral glucose tolerance test. Diabetes care. 2001;24(3):539-48. 532

45. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: 533

comparison with the euglycemic insulin clamp. Diabetes care. 1999;22(9):1462-70. 534

46. Sluiter WJ, Erkelens DW, Reitsma WD, Doorenbos H. Glucose tolerance and insulin release, a 535

mathematical approach I. Assay of the beta-cell response after oral glucose loading. Diabetes. 1976;25(4):241-536

4. 537

47. Seltzer HS, Allen EW, Herron AL, Jr., Brennan MT. Insulin secretion in response to glycemic stimulus: 538

relation of delayed initial release to carbohydrate intolerance in mild diabetes mellitus. The Journal of clinical 539

investigation. 1967;46(3):323-35. 540

48. ProductDescription_DRcaps™. A matter of timing. Ensure optimal delivery for acid-sensitive products. 541

2016 [Available from: https://s3.amazonaws.com/cpsl-web/kc/library/c1a-32029_DRCaps-A4_FIN.PDF. 542

49. Holst JJ. The physiology of glucagon-like peptide 1. Physiological reviews. 2007;87(4):1409-39. 543

50. Wren AM, Bloom SR. Gut hormones and appetite control. Gastroenterology. 2007;132(6):2116-30. 544

51. Callahan HS, Cummings DE, Pepe MS, Breen PA, Matthys CC, Weigle DS. Postprandial suppression of 545

plasma ghrelin level is proportional to ingested caloric load but does not predict intermeal interval in humans. 546

The Journal of clinical endocrinology and metabolism. 2004;89(3):1319-24. 547

52. Stevenson JL, Clevenger HC, Cooper JA. Hunger and satiety responses to high-fat meals of varying fatty 548

acid composition in women with obesity. Obesity (Silver Spring, Md). 2015;23(10):1980-6. 549

Journ

al Pre-

proof

29

53. Quintero-Florez A, Sinausia Nieva L, Sanchez-Ortiz A, Beltran G, Perona JS. The Fatty Acid Composition 550

of Virgin Olive Oil from Different Cultivars Is Determinant for Foam Cell Formation by Macrophages. Journal of 551

agricultural and food chemistry. 2015;63(30):6731-8. 552

Journ

al Pre-

proof

30

Table 1 Fasting baseline characteristics of subjects who completed study 1 and 2 553

Study 1

(n=8)

Study 2

(n=10)

Age (years) 32 ± 9 53 ± 8

Sex, m/f (%) 50 / 50 40 / 60

Height (cm) 175 ± 11 170 ± 10

Weight (kg) 77 ± 18 95 ± 13

BMI (kg/m2) 25.0 ± 3.0 31.3 ± 2.9

Fat mass (kg) 18.0 ± 6.7 34.9 ± 8.3

Fat % 23.3 ± 5.4 38.0 ± 7.8

Fat free mass (kg) 58.8 ± 13.8 57.0 ± 9.7

Systolic blood pressure (mmHg) 128 ± 20 130 ± 11

Diastolic blood pressure (mmHg) 73 ± 8 83 ± 9

Plasma glucose (mmol/L) 4.8 ± 0.6 5.3 ± 0.4

Creatinine (umol/L) 81 ± 8 79 ± 10

Triglycerides (mmol/L) 0.9 ± 0.3 1.3 ± 0.8

HDL-cholesterol (mmol/L) 1.4 ± 0.2 1.4 ± 0.6

LDL-cholesterol (mmol/L) 2.5 ± 0.4 3.4 ± 0.6

Total cholesterol (mmol/L) 4.4 ± 0.5 5.2 ± 0.7

HOMA-IR 1.2 ± 0.9 2.2 ± 0.8

Values are mean ± SD 554

Journ

al Pre-

proof

31

FIGURE LEGENDS 555

Fig. 1. Glucose (a-b), insulin (c-d) and C-peptide (e-f) levels during a 4h OGTT and corresponding AUCs 556

calculated from 0-120, 120-240 and 0-240 min in response to no oil (glucose alone) (black bars and black 557

circles), 3 g pine nut oil (PNO-TG, striped bars and grey triangles) or 3 g pine nut oil free fatty acids (PNO-558

FFA, dotted bars and white squares) in study 1. Values are mean ± SEM. *p<0.05, **p<0.01, (n = 8). 559

560

Fig. 2. Glucose (a-b), insulin (c-d) and C-peptide (e-f) levels during a 4h OGTT and corresponding AUCs 561

calculated from 0-120, 120-240 and 0-240 min in response to no oil (glucose alone) (black bars and black 562

circles), 3 g pine nut oil free fatty acids (PNO-FFA, dotted bars and white squares) or 6 g PNO FFA (checked 563

bars and grey diamonds) in study 2. Values are mean ± SEM. *p<0.05, **p<0.01, (n=10 for no oil and 6 g PNO-564

FFA treatments, and n=9 for 3g PNO-FFA, due to exclusion based on suspected non-compliance). 565

566

Fig. 3. GLP-1 (a-b), GIP (c-d) and ghrelin (e-f) levels during a 4h OGTT and corresponding AUCs calculated 567

from 0-120, 120-240 and 0-240 min in response to no oil (glucose alone) (black bars and black circles), 3 g pine 568

nut oil (PNO-TG, striped bars and grey triangles) or 3 g pine nut oil free fatty acids (PNO-FFA, dotted bars and 569

white squares) in study 1. Values are mean ± SEM. *p<0.05, **p<0.01, (n = 8). 570

571

Fig. 4 GLP-1 (a-b), GIP (c-d) and ghrelin (e-f) levels during a 4h OGTT and corresponding AUCs calculated 572

from 0-120, 120-240 and 0-240 min in response to no oil (glucose alone) (black bars and black circles), 3 g pine 573

nut oil free fatty acids (PNO-FFA, dotted bars and white squares) or 6 g PNO-FFA (checked bars and grey 574

diamonds) in study 2.Values are mean ± SEM. *p<0.05, **p<0.01, (n=10 for no oil and 6 g PNO-FFA 575

treatments, and n=9 for 3g PNO-FFA, due to exclusion based on suspected non-compliance). 576

Journ

al Pre-

proof

0

200

400

600

800

1000

1200

1400

0-120 120-240 0-240

Glu

co

se

AU

C (

mm

ol/L

*min

)

Time (min)

A

3

4

5

6

7

8

0 30 60 90 120 150 180 210 240

Glu

co

se

(m

mo

l/L

)

Time (min)

B

0

100

200

300

400

500

0 30 60 90 120 150 180 210 240

Insu

lin (

pm

ol/L

)

Time (min)

D

0

10000

20000

30000

40000

50000

60000

0-120 120-240 0-240

Insu

lin A

UC

(p

mo

l/L

*min

)

Time (min)

*

*

*

C

0

500

1000

1500

2000

2500

0 30 60 90 120 150 180 210 240

C-p

ep

tid

e (

pm

ol/L

)

Time (min)

No oil PNO-TG PNO-FFA

F

0

100000

200000

300000

400000

500000

0-120 120-240 0-240

C-p

ep

tid

e (

pm

ol/L

*min

)

Time (min)

No oil PNO-TG PNO-FFA

*

*

**

E

Journ

al Pre-

proof

Journ

al Pre-

proof

0

200

400

600

800

1000

1200

1400

1600

0-120 120-240 0-240

Glu

co

se

AU

C (

mm

ol/L

*min

)

Time (min)

*

**

*

A

3

4

5

6

7

8

9

0 30 60 90 120 150 180 210 240

Glu

co

se

(m

mo

l/L

)

Time (min)

B

0

10000

20000

30000

40000

50000

60000

70000

80000

0-120 120-240 0-240

Insu

lin A

UC

(p

mo

l/L

*min

)

Time (min)

*

*

C

0

100

200

300

400

500

600

700

0 30 60 90 120 150 180 210 240

Insu

lin (

pm

ol/L

)

Time (min)

D

0

100000

200000

300000

400000

500000

600000

700000

0-120 120-240 0-240

C-p

ep

tid

e (

pm

ol/L

*min

)

Time (min)

No oil 3g PNO-FFA 6g PNO-FFA

**

E

0

1000

2000

3000

4000

5000

0 30 60 90 120 150 180 210 240

C-p

ep

tid

e (

pm

ol/L

)

Time (min)

No oil 3g PNO-FFA 6g PNO-FFA

F

Journ

al Pre-

proof

0

1000

2000

3000

4000

5000

0-120 120-240 0-240

GL

P-1

AU

C (

pg

/mL

*min

)

Time (min)

**

**

** A

0

10

20

30

40

50

0 60 120 180 240

GL

P-1

AU

C (

pg

/mL

)

Time (min)

B

0

50

100

150

200

250

300

350

0 60 120 180 240

GIP

AU

C (

pg

/mL

)

Time(min)

D

0

10000

20000

30000

40000

50000

60000

0-120 120-240 0-240

GIP

AU

C (

pg

/mL

*min

)

Time (min)

* *

C

0

200

400

600

800

1000

1200

0 60 120 180 240

Gh

relin

(p

g/m

L)

Time (min)

No oil PNO-TG PNO-FFA

F

0

30000

60000

90000

120000

150000

180000

0-120 120-240 0-240

Gh

relin

AU

C (

pg

/mL

)

Time (min)

No oil PNO-TG PNO-FFA

* *

**

**

E

Journ

al Pre-

proof

0

1000

2000

3000

4000

5000

6000

7000

0-120 120-240 0-240

GL

P-1

AU

C (

pg

/mL

*min

)

Time (min)

**

*

**

* * A

0

5

10

15

20

25

30

35

40

45

0 60 120 180 240

GL

P-1

(p

g/m

L)

Time (min)

B

0

10000

20000

30000

40000

50000

60000

70000

0-120 120-240 0-240

GIP

AU

C (

pg

/mL

*min

)

Minutes

C

0

100

200

300

400

500

0 60 120 180 240

GIP

(p

g/m

L)

Time (min)

D

0

200

400

600

800

1000

1200

1400

0 60 120 180 240

Gh

relin

(p

g/m

L)

Time (min)

No oil 3g PNO-FFA 6g PNO-FFA

F

0

50000

100000

150000

200000

250000

0-120 120-240 0-240

Gh

relin

AU

C (

pg

/mL

*min

)

Time (min)

No oil 3g PNO-FFA 6g PNO-FFA

E

*

**

*

Journ

al Pre-

proof