Speculation of hemoglobin A3 peak position in clinical cation exchange high performance liquid chromatogram of the diabetic blood sample with microarray isoelectric focusing

Hemoglobin A1c (HbA1c) is a major component of glycated hemoglobin in human red blood cells. It has been proven to be a significant biomarker for the diagnosis of diabetes; its content in fresh red cells in diabetes blood reflects the average level of blood glucose over the previous three months. Thus, HbA1c level has been used for the assessment of long-term glycemic control in diabetes; the level of 6.5% HbA1c has been certified as a critical cut-off for the diabetes diagnosis. The current commonly used method for HbA1c quantification is based on cation-exchange high performance liquid chromatography (CX-HPLC). The method has advantages such as high stability, rapidity, and automation, but there are still some unidentified peaks of Hb species in CX-HPLC (VARIANT Ⅱ system); in particular, the presence of HbA3 (a glutathiolated Hb) affects the accurate determination of HbA1c. HbA3 is usually present in healthy adult blood samples at 2%-4%, but the concentration of HbA3 increases due to the protection of erythrocytes from oxidation, resulting in decreased HbA1c. However, the relative location of the HbA3 peak in the CX-HPLC clinical chromatogram has not been established. To address this issue, we extracted Hb species from fresh blood samples obtained from a hospital in an anaerobic environment to avoid possible redox reactions of Hb and glutathione. After the extraction, the Hb samples were analyzed using two methods: a low-resolution CX-HPLC (5/50 mm column) currently used for diabetes diagnosis and a high-resolution cationic exchange HPLC (Mono-S 5/50 mm column), to identify the peak corresponding to HbA3. The CX-HPLC analysis of fresh blood samples indicated that the unknown peak P3 located between HbA1c and HbA0 peaks corresponded to the HbA3 peak between HbA1c and HbA0 in the Mono-S-HPLC. Microarray isoelectric focusing (IEF) was used for the micro-preparation of HbA3, HbA1c, and HbA0 in healthy blood samples; then, the micro-prepared species of HbA3, HbA1c, and HbA0 were individually identified via Mono-S-HPLC. The results of the CX-HPLC, Mono-S-HPLC, and microarray IEF experiments indicated that the P3 peak might correspond to HbA3. To confirm this, glutathiolated Hb samples were synthesized via acetylphenylhydrazine and analyzed using both the Mono-S- and CX-HPLC systems. The results showed that the content of both glutaminated hemoglobin of HbA3 in Mono-S-HPLC and P3 in CX-HPLC increased, implying the peak of P3 with the retention time of 1.50 min in CX-HPLC was the peak corresponding to HbA3 in Mono-S-HPLC and microarray IEF. Based on the above experiments and our previous results, the influence of HbA3 on both the analysis of HbA1c in blood samples and the diabetes diagnosis needs to be considered and discussed. The study results are significant for the tentative assignment of peak P3 and for offering more information on diabetes diagnosis using CX-HPLC in the clinical setting.