Association Between Circulating Gremlin 2 and β-Cell Function Among Participants With Prediabetes and Type 2 Diabetes

Abstract

Aim: Circulating Gremlin 2 (Grem2) has recently been linked to human obesity, but its role in type 2 diabetes (T2D) remains unclear. This study aims to explore the association of circulating Grem2 with β-cell function.

Methods: A post hoc analysis was conducted using data from three clinical trials, in which all participants underwent the oral glucose tolerance test (OGTT). Circulating Grem2 levels were measured at 0, 1, and 2 h during the OGTT. In Trial 1, Grem2 levels were compared between participants with T2D (n = 59) and without T2D (n = 119). We further examined changes in Grem2 levels in response to oral antidiabetic drugs in participants with T2D in Trial 2 (n = 67) and calorie restriction in participants with prediabetes in Trial 3 (n = 231). The relationship between Grem2 levels and β-cell function was analyzed across all trials.

Results: Fasting and 1-h Grem2 levels were lower in participants with T2D compared with those without T2D (728 ± 25 vs. 649 ± 31 pg/mL, p = 0.020; 631 ± 26 vs. 537 ± 31 pg/mL, p = 0.007). Fasting Grem2 levels were restored after antidiabetic treatment (550 ± 12 vs. 575 ± 12 pg/mL, p = 0.019), and 1-h Grem2 levels increased following calorie restriction (1118 ± 89 vs. 1144 ± 90 vs. 1253 ± 89 pg/mL, p for trend = 0.002). The 1-h Grem2 levels were positively associated with β-cell function assessed by the oral disposition index and HOMA-β.

Conclusion: Reduced circulating Grem2 levels are associated with impaired β-cell function in T2D, and could be restored through antidiabetic interventions.

Trial registration: ClinicalTrials.gov: NCT01959984, NCT01758471, NCT03856762.

Keywords: Gremlin 2; calorie restriction; oral antidiabetic drug; prediabetes; type 2 diabetes; β‐cell function.

FIGURE 1

Flowchart of the study population in Trial 1 (A), Trial 2 (B), and Trial 3 (C).

FIGURE 2

Grem2 levels decline in participants with type 2 diabetes. (A) Trial 1 comprising participants (n = 178) without type 2 diabetes (n = 119) and with type 2 diabetes (n = 59) who underwent a 3‐h, 75‐g oral glucose tolerance test. (B–D) Circulating Grem2 levels at 0 (B), 1 (C), and 2 h (D) in participants without and with type 2 diabetes. Data are expressed as estimated marginal mean ± SEM. Statistical significance was determined by multivariable linear regression models, adjusted for age, sex, BMI, lipid‐lowering drug use, and antihypertensive drug use. (E) Results of mediation analysis, demonstrating that oral disposition index mediated 43% of the effects of 1‐h Grem2 on type 2 diabetes. The β‐coefficients and p values are provided, along with the percentage of causal mediation effect. DI, disposition index; T2D, type 2 diabetes.

FIGURE 3

Grem2 levels are restored by antidiabetic treatment in participants with treatment‐naïve type 2 diabetes. (A) Trial 2 comprising 67 participants with treatment‐naïve type 2 diabetes who underwent a 3‐month treatment with the oral antidiabetic drug. (B) Elevation of fasting Grem2 levels after antidiabetic treatment. Data are expressed as estimated marginal mean ± SEM. Statistical significance was determined by linear mixed‐effects model, adjusted for age, sex, drug types, changes of BMI, lipid‐lowering drug use, and antihypertensive drug use, and baseline values.

FIGURE 4

Grem2 levels increase after calorie restriction in participants with prediabetes. (A) Trial 3 comprising 231 participants with prediabetes who underwent a 6‐month 25% calorie restriction. (B–D) Circulating Grem2 levels at 0 (B), 1 (C), and 2 h (D) after calorie restriction. Data are expressed as estimated marginal mean ± SEM. Statistical significance was determined by linear mixed‐effects models, adjusted for age, sex, diet regimes, changes of BMI, lipid‐lowering drug use, antihypertensive drug use, and baseline values. (E) Results of mediation analysis, demonstrating that 1‐h Grem2 mediated 24% of the effects of caloric restriction on oral disposition index. The β‐coefficients and p values are provided, along with the percentage of causal mediation effect. CR, calorie restriction; DI, disposition index; T2D, type 2 diabetes.