The existence of D-amino acids in mammals has been increasingly reported, and clarified that D and L enantiomers have different biological functions, distributions and metabolic pathways. Some D-amino acids were also reported to have a correlation with diseases, and the screening of biomarkers based on quantitative enantioselective metabolomics is highly expected. However, the amounts of most D-amino acids, especially the metabolic-related chiral amino acids, are usually at trace levels in biological samples. Therefore, highly selective and sensitive analytical methods are essential for their accurate determination. In this focusing review, multi-dimensional high-performance liquid chromatographic (HPLC) systems for the precise determination of metabolic-related chiral amino acids are introduced with their application to biological samples. For the highly-sensitive determination, target amino acids were precolumn-derivatized with 4-fluoro-7-nitro-2,1,3-benzoxadiazole (NBD-F). The multi-dimensional HPLC systems were composed of a reversed-phase column in the 1^st dimension and an enantioselective column in the 2^nd dimension. By using these systems, various proteinogenic amino acids were clearly observed in the tissues and physiological fluids of rodents. The existence of metabolic-related chiral amino acids in various biological samples were also elucidated.

Glucuronyl- and sulpho-conjugates of intestinal putrefactive compounds (IPC) are considered kidney disease-associated toxic agents. Recently, several simultaneous methods of detection by liquid chromatography tandem mass spectrometry (LCMS/MS) of the conjugated forms of glucuronyl- and sulpho-conjugates in blood serum were reported. To analyze these conjugated forms in urine, we modified and validated the LC-MS/MS detection methods. The range of the calibration curves were determined using phenyl glucuronide (PhG; 1.56-25.0 μM), indoxyl glucuronide (IndG; 1.25-20.0 μM), p-cresyl glucuronide (PCG; 0.625-10.0 μM), phenyl sulfate (PhS; 3.13-50.0 μM), indoxyl sulfate (IndS; 7.81-250 μM) and p-cresyl sulfate (PCS; 1.56-25.0 μM). Urine was diluted 2-4 fold to determine PhG, IndG and PCG concentrations, and 4-16 fold to determine PhS, IndS and PCS concentrations. We validated the newly-modified detection methods by a recovery test, and intra-day and inter-day repeatability. We analyzed six clinical urine samples, PCS and IndS, major toxic agents, which were mostly detected. Creatinine-corrected values of PCS and IndS were high in six conjugates. In summary, in the present work, the concentrations of six urinary conjugated forms of IPC were successfully measured by LC-MS/MS.

A method for the simultaneous determination of 13 slimming drugs in dietary supplements by liquid chromatography/time-of-flight mass spectrometry was developed and applied to real-time surveys of imported health supplements. The target compounds were mutually well separated on a reversed-phase LC Shodex ODP2-HP 2D column using a combination of step-wise and linear gradient elution of a mixed mobile phase solution of ammonium formate buffer, methanol, and formic acid. The detection limits were in the range 1-50 ng/mL, and the lower limits of quantification were 2-100 ng/mL. The average recoveries of the 13 types of pharmaceutical ingredients in the diet supplements were 93-110%, and the RSD was within 7.1%. Applying the developed method to the survey of imported dietary supplements revealed that 32 out of the 39 Chinese products tested contained pharmaceutical ingredients such as sibutramine. On the other hand, out of 30 US products, synephrine was detected in 26 specimens, and yohimbine was detected in 8 specimens. These results proved that a number of products were contaminated by drugs, which was considered to be intentional. The developed method is applicable for the survey of imported health supplements suspected to be contaminated with pharmaceutical ingredients.

For the highly chemoselective separations of lactate (LA) and 3-hydroxybutyrate (3HB) enantiomers in mammalian urine, an online three-dimensional HPLC system (3D-HPLC) composed of reversed-phase (KSAARP, 1.0 mm i.d. x 250 mm), mixed-mode (KSAAMX, 1.5 mm i.d. x 250 mm) and enantioselective (Chiralpak AD-H, 2.0 mm i.d. x 250 mm) columns has been developed. Following the pre-column derivatization with 4-nitro-7-piperazino-2,1,3-benzoxadiazole (NBD-PZ), NBD-LA and 3HB were separated in the first and second dimensions as their scalmic D plus L mixtures and the enantiomer separations were performed in the third dimension with high separation factors (α = 1.70 for LA; α = 1.43 for 3HB). The present system was applied to the urine of mice and humans. For the mice urine, the D and L-forms of NBD-LA and 3HB were well separated without visible interfering compounds, and trace levels of D-LA (less than 5% of total LA) and non-negligible amounts of L-3HB (34.4% of the total 3HB) were clearly observed. Trace levels of the LA and 3HB enantiomers were detected for the human urine samples, although the existence of a few interfering peaks were observed.

Aqueous formic acid (FA) was determined by purge-and-trap extraction with a miniaturized extraction capillary using ion chromatography (IC). The extraction capillary was prepared by packing particulate activated carbon in a stainless-steel capillary (0.8 mm i.d., 1.6 mm o.d.). FA was purged from the sample solution and extracted on the extraction medium in its molecular form. Following extraction, the extraction capillary was connected directly to a six-port valve of the IC. The extracted analyte was eluted with sodium hydroxide solution as the formate ion and introduced to a conventional IC. The extraction capillary can be reused after conditioned by passing hydrochloric acid solution. The limit of detection of FA in a standard aqueous sample is 0.5 mg/L. The method was successfully applied to determine FA in fruit juices without significant interference.