|
Diabetes mellitus is a chronic disease typified by high levels of blood sugars, principally glucose and its metabolites. Glucose in the body is derived from carbohydrates in food and is converted to energy with the aid of insulin. When there is insufficient insulin or it does not function correctly, glucose conversion is compromised, leading to its accumulation in the blood. The excess sugar can induce a series of problems including disease of the retina (retinopathy), kidney disease (nephropathy) and microvascular damage. Another effect is long-term damage to the nerve fibres (neuropathy) which is brought about by the increased conversion of excess glucose to sorbitol. This sugar accumulates in tissues and nerves because it cannot pass through the cell membrane to be exported. Its presence increases osmotic pressure in cells, allowing excess water and sodium ions to enter and cause nerve degeneration. As such, sorbitol is a known biomarker of diabetic neuropathy and there are a number of published methods for measuring its concentration as an aid to diagnosis. However, many of these suffer from limitations, according to a team of scientists from South Korea. For instance, enzymatic methods may also act on other sugars as well as sorbitol to give false data. HPLC and GC/MS procedures generally require derivatisation. So, this group has tried to develop a simpler method for measuring sorbitol, based on HPLC with pulsed amperometric detection. Sun Yeou Kim and Seon-Pyo Hong and colleagues from Kyung Hee University and Chungbuk National University chose this mode of detection because it does not require derivatisation of sorbitol. Instead, it relies on the oxidation of sorbitol in solution at a gold electrode in an electrochemical cell and measures the potential produced. The method was optimised with standard solutions containing seven monosaccharides. As well as sorbitol, they contained other biological sugars that are potential interferents: myoinositol, xylitol, mannitol, glucose, galactose and fructose. HPLC separation was achieved on an anion exchange column. A MetrocepCarb1 column was the best of four tested, producing the greatest resolution and the shortest analysis times. The mobile phase was aqueous 100 mM sodium hydroxide, the concentration being chosen as a compromise between retention time and resolution. At low concentrations, resolution was good but retention times were long, but at too high a concentration the resolution suffered. The solution was degassed with helium before use to remove dissolved carbon dioxide which has a destabilising effect on retention times. The treated mobile phase provided constant retention times and steady baselines. Under these conditions, sorbitol eluted in 5.80 minutes, between xylitol and mannitol. It was clearly separated from glucose which is present in blood at 1000-fold higher concentrations, and no sugars interfered. A second set of potential interferents are amino acids, which normally remain after sample pretreatment but the sorbitol peak stood out clearly in a solution containing 40 free amino acids. The method was tested using the sciatic nerves and salivary glands from the mouse model of diabetes, 50 days after its induction with the toxic glucose analogue alloxan (5-oxobarbituric acid). The tissues were homogenised in 5% trichloroacetic acid containing mannitol as an internal standard and the centrifuged supernatant was analysed. The extraction efficiency was excellent at 98.3 ± 1.8%. Calibration curves constructed by adding known concentrations of sorbitol to the tissues were linear over 0.01-50.0 µg/g. The detection and quantitation limits were 0.03 ng (3 ng/g) and 0.10 ng (10 ng/g), respectively. These values were more sensitive than those of published enzymatic, HPLC or GC/MS methods but were not as low as an APCI LC/MS method. In the analysis of real samples, the sorbitol peaks in the chromatograms were clearly larger for diabetic mice than control mice. The levels in sciatic nerves ranged from 33.6-86.9 µg/g and 6.0-20.9 and µg/g in diabetic and control mice, respectively. The corresponding values in salivary glands were 34.1-60.4 and 3.2-14.4 µg/g. These values agree with published levels of 27 µg/g determined for sorbitol in sciatic nerve determined by an enzymatic method. The high-performance anion exchange method with pulsed amperometric detection is simple to perform and shows high resolution and excellent selectivity for sorbitol in tissues. The research team stated that it will be practically useful in determining sorbitol in the mouse model for the selective screening and monitoring of diabetic neuropathy. Related links:
The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd. |
|
|
|
|
|
