Satiety Hormone LEAP2 After Low-Calorie Diet With/Without Endobarrier Insertion in Obesity and Type 2 Diabetes Mellitus

Abstract

Context: The liver/foregut satiety hormone liver-expressed antimicrobial peptide 2 (LEAP2) is an inverse agonist at the acyl ghrelin receptor (GHSR), increasing after food intake and decreasing after bariatric surgery and short-term nonsurgical weight loss, but effects of long-term dietary weight loss are unknown.

Objective: The objective of this study was to examine and compare the effects of these interventions on fasting and postprandial plasma LEAP2 and investigate potential metabolic mediators of changes in plasma LEAP2.

Methods: Plasma LEAP2 was measured in a previously published 2-year trial comparing standard medical management (SMM) (including 600-kcal/day deficit) with duodenal-jejunal bypass liner (DJBL, Endobarrier) insertion (explanted after 1 year) in adults with obesity and inadequately controlled type 2 diabetes mellitus.

Results: In the SMM group (n = 25-37), weight decreased by 4.3%, 8.1%, 7.8%, and 6.4% at 2, 26, 50, and 104 weeks and fasting plasma LEAP2 decreased from baseline mean ± SD 15.3 ± 0.9 ng/mL by 1.7, 3.8, 2.1, and 2.0 ng/mL, respectively. Absolute/decreases in fasting plasma LEAP2 positively correlated with absolute/decreases in body mass index, glycated hemoglobin A_1c, fasting plasma glucose, serum insulin, homeostatic model assessment for insulin resistance, and serum triglycerides. Despite greater weight loss in the DJBL group (n = 23-30) at 26 to 50 weeks (10.4%-11.4%), the decrease in fasting plasma LEAP2 was delayed and attenuated (vs SMM), which may contribute to greater weight loss by attenuating GHSR signaling. Plasma LEAP2 did not increase with weight regain from 50 to 104 weeks after DJBL explant, suggesting a new set point with weight loss maintenance. Increases in plasma LEAP2 after a 600-kcal meal (10.8%-16.1% at 1-2 hours) were unaffected by weight loss, improved glucose metabolism, or DJBL insertion (n = 9-25), suggesting liver rather than duodenum/jejunum may be the primary source of postprandial LEAP2 secretion.

Conclusion: These findings add to our understanding of the regulation and potential physiological role of plasma LEAP2.

Keywords: LEAP2; appetite; duodenal-jejunal bypass liner; glucose; insulin; obesity; weight loss.

Figure 1.

Longitudinal analysis of body mass index (BMI) and change in percentage weight, glycated hemoglobin A_1c (HbA_1c), fasting glucose, HOMA-IR, fasting triglycerides, fasting plasma liver/foregut satiety hormone liver-expressed antimicrobial peptide 2 (LEAP2), and appetite ratings across visits. A, BMI, change in B, percentage weight loss; C, HbA_1c; D, fasting glucose; E, HOMA-IR; F, fasting triglycerides; G, fasting plasma LEAP2; and H, fasting appetite VAS ratings vs −2 weeks, across time for participants undergoing standard medical management alone (SMM, dashed line) and duodenal-jejunal bypass liner (DJBL, solid line) groups. Black bar indicates duration of DJBL being in situ (0-52 weeks). N indicates number of participants at each time point. Gray bar indicates duration of liquid diet (−1 to 2 weeks). Statistical results below graph indicates linear mixed model analysis including group (SMM, DJBL), sweet taste detection task (done in the taste subgroup at weeks −2, 2, and 26 only, for fasting glucose, HOMA-IR, and fasting LEAP2 only) as between-participant, and weeks (−2, 2, 26, 50, and 104) as within-participant fixed factors, to examine effects of interventions on dependent variables over time. DJBL vs SMM: *P less than .05, **P less than .01, ***P less than .005, ****P less than .001; DJBL vs baseline week −2: # P less than .05, ## P less than .01, ### P less than .005, #### P less than .001; SMM vs baseline week −2: $ P less than .05, $$ P less than .01, $$$ P less than .005, $\$\$\$\$$  P less than .001. All data given as mean ± SEM. Abbreviations: HOMA-IR, homeostatic model assessment for insulin resistance; kcal, kilocalorie; VAS, visual analogue scale.

Figure 1.

Longitudinal analysis of body mass index (BMI) and change in percentage weight, glycated hemoglobin A_1c (HbA_1c), fasting glucose, HOMA-IR, fasting triglycerides, fasting plasma liver/foregut satiety hormone liver-expressed antimicrobial peptide 2 (LEAP2), and appetite ratings across visits. A, BMI, change in B, percentage weight loss; C, HbA_1c; D, fasting glucose; E, HOMA-IR; F, fasting triglycerides; G, fasting plasma LEAP2; and H, fasting appetite VAS ratings vs −2 weeks, across time for participants undergoing standard medical management alone (SMM, dashed line) and duodenal-jejunal bypass liner (DJBL, solid line) groups. Black bar indicates duration of DJBL being in situ (0-52 weeks). N indicates number of participants at each time point. Gray bar indicates duration of liquid diet (−1 to 2 weeks). Statistical results below graph indicates linear mixed model analysis including group (SMM, DJBL), sweet taste detection task (done in the taste subgroup at weeks −2, 2, and 26 only, for fasting glucose, HOMA-IR, and fasting LEAP2 only) as between-participant, and weeks (−2, 2, 26, 50, and 104) as within-participant fixed factors, to examine effects of interventions on dependent variables over time. DJBL vs SMM: *P less than .05, **P less than .01, ***P less than .005, ****P less than .001; DJBL vs baseline week −2: # P less than .05, ## P less than .01, ### P less than .005, #### P less than .001; SMM vs baseline week −2: $ P less than .05, $$ P less than .01, $$$ P less than .005, $\$\$\$\$$  P less than .001. All data given as mean ± SEM. Abbreviations: HOMA-IR, homeostatic model assessment for insulin resistance; kcal, kilocalorie; VAS, visual analogue scale.

Figure 2.

Correlations of absolute and change in fasting plasma liver/foregut satiety hormone liver-expressed antimicrobial peptide 2 (LEAP2) with absolute and change in weight, fasting glucose, HOMA-IR, and appetite ratings across visits in the SMM group. Correlations of absolute fasting plasma LEAP2 with A, body mass index (BMI); C, absolute fasting plasma glucose; E, HOMA-IR; G, absolute fasting appetite VAS ratings across visits −2, 2, 26, 50, and 104 weeks; and correlations of change in fasting plasma LEAP2 with B, percentage weight change; change in D, fasting plasma glucose; F, HOMA-IR; and H, fasting appetite VAS ratings between weeks 2, 26, 50, and 104 vs −2 weeks using linear mixed model analysis including sweet taste detection task (done in the taste subgroup at weeks −2, 2, and 26 only, for fasting glucose, HOMA-IR and LEAP2 only) as between-participant, and weeks (−2, 2, 26, 50, 104) as within-participant fixed factors, to examine effects of interventions on absolute and change in fasting plasma LEAP2 over time with pseudo r² and β parameters. *P less than .05, **P less than .01, ***P less than .005, ****P less than .001. n = 22-30. Abbreviations: HOMA-IR, homeostatic model assessment for insulin resistance; SMM, standard medical management; VAS, visual analogue scale.

Figure 3.

Correlations of absolute and change in fasting plasma liver/foregut satiety hormone liver-expressed antimicrobial peptide 2 (LEAP2) with absolute and change in weight, fasting glucose, HOMA-IR, and appetite ratings across visits in the DJBL group. Correlations of absolute fasting plasma LEAP2 with A, body mass index (BMI); C, absolute fasting plasma glucose; E, HOMA-IR; and G, absolute fasting appetite VAS ratings across visits −2, 2, 26, and 50 weeks; and correlations of change in fasting plasma LEAP2 with B, percentage weight change; change in D, fasting plasma glucose; F, HOMA-IR; and H, fasting appetite VAS ratings between weeks 2, 26, and 50 vs −2 weeks using linear mixed model analysis including sweet taste detection task (done in the taste subgroup at weeks −2, 2, and 26 only, for fasting glucose, HOMA-IR and LEAP2 only) as between-participant, and weeks (−2, 2, 26, 50) as within-participant fixed factors, to examine effects of interventions on absolute and change in fasting plasma LEAP2 over time with pseudo r² and β parameters. *P less than .05, **P less than .01, ***P less than .005, ****P less than .001. n = 22-30. Abbreviations: DJBL, duodenal-jejunal bypass liner; HOMA-IR, homeostatic model assessment for insulin resistance; VAS, visual analogue scale.

Figure 4.

Changes in postprandial plasma liver/foregut satiety hormone liver-expressed antimicrobial peptide 2 (LEAP2). A and E, Longitudinal analysis of A, percentage weight loss and E, postprandial plasma LEAP2 iAUC from t = 0 minutes to t = 120 minutes after consumption of a fixed 600-kcal test meal (49% carbohydrate, 35% fat, 16% protein) across visits for participants undergoing standard medical management alone (SMM, dashed line) and duodenal-jejunal bypass liner (DJBL, solid line) groups. Black bar indicates duration of DJBL being in situ (0-52 weeks). Gray bar indicates duration of liquid diet (−1 to 2 weeks). N indicates number of participants at each time point. Statistical results below graph indicates linear mixed model analysis including group (SMM, DJBL), sweet taste detection task (done at weeks −2, 2, and 26 only), as between-participant, and weeks (−2, 2, 26, 50) as within-participant fixed factors, to examine effects of interventions on dependent variables over time: DJBL vs SMM *P less than .05, **P less than .01, ***P less than .005, ****P less than .001; DJBL vs baseline week 0: # P less than .05, ## P less than .01, ### P less than .005, #### P less than .001; SMM vs baseline week 0: $ P less than .05, $$ P less than .01, $$$ P less than .005, $\$\$\$\$$  P less than .001. B, Postprandial plasma LEAP2 after 600-kcal fixed test meal (49% carbohydrate, 35% fat, 16% protein) at baseline −2 week visit. Comparisons made using one-way repeated-measures analysis of variance including time (t = 0, 60, 120 minutes) as within-participant factors with post hoc Fisher least significant difference: *P less than .05, **P less than .01, ***P less than .005, ****P less than .001, t = 60 or 120 minutes vs 0 minutes, n = 40. C and D, Change in absolute postprandial plasma LEAP2 response at time points t = 60 and 120 minutes vs t = 0 minutes in C, DJBL group and D, SMM group, after 600-kcal fixed test meal at t = 0 minutes across visits. Statistical results given in Supplementary Tables S26-27 [63] made using linear mixed model analysis including time (t = 0, 60, 120 minutes) and weeks (−2, 2, 26, 50) as within-participant fixed factors, to examine effects of interventions on postprandial plasma LEAP2 over time. n = 9-25. F, Correlation of change in absolute plasma LEAP2 (ng/mL) concentrations between fed (+60 to +150 minutes) and fasted (0 minutes) time points (Δ fed-fasted) and preload meal size (kcal) consumed across different studies using Spearman (r_S) correlation coefficient. Black dots indicate participants with obesity (BMI ≥30 kg/m²), black triangle with arrow indicates obesity with type 2 diabetes mellitus in the present study, open dots indicate participants without obesity (BMI < 30 kg/m²). Meals are nutritional supplements with similar macronutrient composition: carbohydrate: 49% to 55%, fat 30% to 35%, protein 14% to 18% energy. List of individual studies given in Supplementary Table S25 [63]. All data given as mean ± SEM. Abbreviations: BMI, body mass index; iAUC, incremental area under the curve; kcal, kilocalorie; OGTT, oral glucose tolerance test.