LDL particles exhibit heterogeneity in density, size, chemical composition, and charge (1). Lipoperoxidation, oxidation, and glycosylation increase the net negative charge and may enhance LDL atherogenicity with important metabolic consequences. A relevant role of more electronegative LDL in atherogenesis is supported by the observation that it is elevated in subjects at high risk, such as familial hypercholesterolemic and type 1 diabetic patients (2).

We reported the precise measurement of the electrophoretic mobility of LDL as an indicator of modification by capillary electrophoresis and the UV absorption at 234 nm that results from the formation of conjugated dienes in constituent polyenoic fatty acids in 14 type 1 diabetic patients (7 normoalbuminuric and 7 microalbuminuric patients) and in 6 nondiabetic subjects. In type 1 diabetic patients with normoalbuminuria (six men and one woman; mean age 38 ± 12 years) the mean duration of diabetes was 25 ± 7 years, and they were in stable glycemic control (HbA1c = 7.1 ± 0.6%). The seven diabetic patients with microalbuminuria (six men and one woman; mean age 52 ± 9 years, P < 0.01 vs. normoalbuminuric patients) had a mean duration of diabetes of 22 ± 14 years and a mean HbA1c value of 8.8 ± 1% (P < 0.01 vs. normoalbuminuric patients). Diabetic patients had significantly higher BMI (25 ± 2 kg/m2) (P < 0.01 for normoalbuminuric subjects, 25 ± 3 kg/m2; P < 0.05 for microalbuminuric vs. control group, 21 ± 2 kg/m2) and fasting glucose levels (215 ± 83 mg/dl) (P < 0.01 for normoalbuminuric subjects, 197 ± 91 mg/dl; P < 0.01 for microalbuminuric vs. control group, 99 ± 18 mg/dl) than control subjects. There was no difference in triglycerides, total cholesterol, LDL cholesterol, and HDL cholesterol levels between diabetic subjects and the control group.

LDL was isolated by preparative sequential ultracentrifugation at the density of 1.063 g/ml. Dialyses, capillary electrophoresis (CE), and the electrophoretic mobility (μ) of LDL were performed as described by Stock and Miller (3). Migration of LDL particles was monitored at 200 and 234 nm. The amount of conjugated dienes is obtained from the percentage of the height of LDL peak at 234 nm related to the height of LDL peak at 200 nm. Student’s t test and Pearson’s correlation were used to assess statistical significance.

The electrophoretic mobility (mean ± SD) for the diabetic LDL was −1.249 ± 0.065 · 10−4 · cm2 · vol−1 · s−1, while that for the control LDL was −1.032 ± 0.121 (P = 0.0001). The diabetic group, subdivided into normoalbuminuric and microalbuminuric subjects, presented an electrophoretic mobility mean of −1.234 ± 0.068 and −1.263 ± 0.064 · 10−4 · cm2 · vol−1 · s−1, respectively. When each group was compared with the control, the differences were always statistically significant in both cases (P = 0.0032 for normoalbuminuric patients vs. control subjects; P = 0.0001 for microalbuminuric patients vs. control subjects). Diabetic subjects have LDL with significantly higher migration rates, which were independent from microalbuminuria.

In LDL obtained from the diabetic patients the content of diene conjugates was not statistically different from the control group (6.22 ± 1.199% for diabetic subjects vs. 5.509 ± 0.219% for control subjects).

The difference between diabetic and control subjects was still not statistically significant when the content of diene conjugates in normoalbuminuric (6.235 ± 1.544%) and microalbuminuric (6.214 ± 0.854%) subjects was individually compared with that of the control group. In the diabetic group, the electrophoretic mobility was not significantly correlated with HbA1c, duration of diabetes, the subjects’ age, or fasting glucose levels.

The finding of electronegative LDL in type 1 diabetic subjects could be related to the increase of the so-called LDL(−), which is also detectable in normal subjects, although in small amounts (4). Capillary electrophoresis cannot separate the fraction LDL(−) from the bulk of plasma LDL. It gives an estimate of the algebraic sum of the electronegative charges distributed on the surface of LDL particles.

Nonenzymatic glycosylation should, surprisingly, be excluded as a cause of higher LDL electronegativity. In this regard, we found no significant correlation between electrophoretic mobility and HbA1c and the fasting plasma glucose levels in the diabetic group. Furthermore, neither the duration of diabetes nor subject age had effects on LDL mobility. Thus, the increased negative charge could be related to compositional abnormalities or other modifications not evaluated in this report, such as an enrichment in sialic acid. Desialylated LDL is more resistant to copper oxidation than native LDL (5).

In conclusion, the finding of more electronegative LDL in diabetic subjects could be an additional risk factor for atherosclerosis in diabetes. Investigations are under way to assess if electrophoretic mobility of LDL in type 1 diabetes can be decreased by further lowering HbA1c levels.

La Belle M, Blanche PJ, Krauss M: Charge properties of low density lipoprotein subclasses.
J Lipid Res
Sánchez-Quesada JL, Pérez A, Caixàs A, Ordóñez-Llanos J, Carreras G, Payés A: Electronegative low density lipoprotein subform is increased in patients with short-duration IDDM and is closely related to glycaemic control.
Stock J, Miller NE: Capillary electrophoresis to monitor the oxidative modification of LDL.
J Lipid Res
Demuth K, Myara I, Chappey B, Vedie B. Pech-Amsellem MA, Haberland ME, Moatti N: A cytotoxic electronegative LDL subfraction is present in human plasma.
Arterioscl Thromb Vasc Biol
Myara I, Haberland ME, Demuth K, Chappey B, Moatti N: Susceptibility to copper oxidation of neuraminidase-treated LDL.
Clin Chem Acta