The Association of Polyunsaturated Fatty Acid δ-5-Desaturase Activity with Risk Factors for Type 2 Diabetes Is Dependent on Plasma ApoB-Lipoproteins in Overweight and Obese Adults

Abstract

Background: δ-5 and δ-6 desaturases (D5D and D6D) catalyze the endogenous conversion of n-3 (ω-3) and n-6 (ω-6) polyunsaturated fatty acids (PUFAs). Their activities are negatively and positively associated with type 2 diabetes (T2D), respectively, by unclear mechanisms. Elevated plasma apoB-lipoproteins (measured as plasma apoB), which can be reduced by n-3 PUFA intake, promote T2D risk factors.

Objective: The aim of this study was to test the hypothesis that the association of D5D and D6D activities with T2D risk factors is dependent on plasma apoB.

Methods: This is a pooled analysis of 2 populations recruited for 2 different metabolic studies. It is a post hoc analysis of baseline data of these subjects [n = 98; 60% women (postmenopausal); mean ± SD body mass index (in kg/m2): 32.8 ± 4.7; mean ± SD age: 57.6 ± 6.3 y]. Glucose-induced insulin secretion (GIIS) and insulin sensitivity (IS) were measured using Botnia clamps. Plasma clearance of a high-fat meal (600 kcal/m2, 66% fat) and white adipose tissue (WAT) function (storage of 3H-triolein-labeled substrate) were assessed in a subpopulation (n = 47). Desaturase activities were estimated from plasma phospholipid fatty acids. Associations were examined using Pearson and partial correlations.

Results: While both desaturase activities were positively associated with percentage of eicosapentaenoic acid, only D5D was negatively associated with plasma apoB (r = -0.30, P = 0.003). Association of D5D activity with second-phase GIIS (r = -0.23, P = 0.029), IS (r = 0.33, P = 0.015, in women) and 6-h area-under-the-curve (AUC6h) of plasma chylomicrons (apoB48, r = -0.47, P = 0.020, in women) was independent of age and adiposity, but was eliminated after adjustment for plasma apoB. D6D activity was associated in the opposite direction with GIIS (r = 0.24, P = 0.049), IS (r = -0.36, P = 0.004) and AUC6h chylomicrons (r = 0.52, P = 0.004), independent of plasma apoB. Both desaturases were associated with plasma interleukin-1-receptor antagonist (D5D: r = -0.45, P < 0.001 in women; D6D: r = -0.33, P = 0.007) and WAT function (trend for D5D: r = 0.30, P = 0.05; D6D: r = 0.39, P = 0.027) independent of any adjustment.

Conclusions: Association of D5D activity with IS, lower GIIS, and plasma chylomicron clearance is dependent on plasma apoB in overweight and obese adults.

Figures

FIGURE 1

Percent composition of FAs in fasting plasma phospholipids in overweight and obese postmenopausal women and adult men. Data are means ± SDs. n = 59 women, 39 men unless indicated otherwise. aFor n = 44 women, 28 men; bfor n = 57 women, 27 men; cfor n = 59 women, 36 men; dfor n = 58 women, 39 men; efor n = 42 women, 25 men; ffor n = 32 women, 14 men as the concentration of these FAs were below the detection limit. ALA, α-linolenic acid; ARA, arachidonic acid; DGLA, dihomo-γ-linolenic acid; LA, linoleic acid.

FIGURE 2

Pearson correlation of (A) margaric acid (17:0), (B) total n–6 PUFAs, (C) LA (18:2n–6), (D) DGLA (20:3n–6), (E) estimated D5D activity, and (F) estimated D6D activity measured in fasting plasma phospholipid FAs with plasma apoB, in overweight and obese postmenopausal women and adult men. n = 59 women, 39 men except in (A) where n = 44 women, 28 men and (E) where n = 59 women, 38 men because of missing data, and (F) where n = 42 women, 25 men because of ALA (18:3n–3) concentration being below the detection limit. ALA, α-linolenic acid; D5D, δ-5-desaturase; D6D δ-6-desaturase; DGLA, dihomo-γ-linolenic acid; LA, linoleic acid.

FIGURE 3

Pearson correlation of the estimated D5D activity with (A) IS (GIRclamp), (B) second-phase GIISIVGTT, (C) fasting plasma IL-1Ra, (D) postprandial plasma clearance of chylomicrons (apoB48), (E) postprandial plasma clearance of TGs, and (F) ex vivo WAT function in overweight and obese postmenopausal women and adult men. n = 59 women, 38 men except in (A) where n = 55 women, 37 men, (B) where n = 58 women, 38 men, (D) where n = 24 women, 16 men, (E) where n = 27 women, 18 men and (F) where n = 26 women, 16 men because of missing data. D5D, δ-5-desaturase; GIISIVGTT, glucose-induced insulin secretion measured during intravenous glucose tolerance test; GIRclamp, glucose infusion rate measured during clamp; IL-1Ra, IL-1 receptor antagonist; IS, insulin sensitivity; WAT, white adipose tissue.

FIGURE 4

Pearson correlation of the estimated D6D activity with (A) IS (GIRclamp), (B) second-phase GIISIVGTT, (C) fasting plasma IL-1Ra, (D) postprandial plasma clearance of chylomicrons (apoB48), (E) postprandial plasma clearance of TGs, and (F) ex vivo WAT function in overweight and obese postmenopausal women and adult men. n = 42 women, 25 men except in (A) where n = 40 women, 24 men, (B) where n = 41 women, 25 men, (D) where n = 18 women, 10 men, (E) where n = 21 women, 12 men, and (F) where n = 20 women, 12 men because of missing data. D6D, δ-6-desaturase; GIISIVGTT, glucose-induced insulin secretion measured during intravenous glucose tolerance test; GIRclamp, glucose infusion rate measured during clamp; IL-1Ra, IL-1 receptor antagonist; IS, insulin sensitivity; WAT, white adipose tissue.