High Performance Liquid Chromatography (HPLC) is a vital tool in all biological sciences. In metabolomics, HPLC has evolved to become, in conjunction with mass spectrometry (MS), the central technique. Metabolomics has already contributed to a greater understanding of diabetes, insulin resistance, and cardiovascular disease, and has lead to the discovery of biomarkers for early detection of cancer. However, further advances in this field are heavily dependent on advances in analytical instrumentation that will improve identification of large numbers of metabolites. HPLC retention is a chemically-specific metric that is very orthogonal to MS data. Unfortunately, in metabolomics it has largely gone unused because the mechanism of LC retention is complex and more importantly, gradient elution retention times are very dependent on many differences between HPLC instruments. Preliminary work shows that use of LC retention data in combination with MS exact mass dramatically improves analyte identification probability; using a low-end LC-single quadrupole MS, unambiguous identifications rose from 88 (of 7,307 unknowns) to 3,232-bettering the identification power of LC-FTICR MS on the same set of unknowns. In this proposal, a novel approach to a robust HPLC retention database is outlined along with a novel retention projection system. This system uses the retention times of a set of 30 judiciously chosen standard solutes spiked into the sample to back-calculate what the effective gradient and flow rate profiles must have been to give those retention times. Those profiles can then be used to calculate the retention of other compounds with unprecedented accuracy. They account for differences between HPLC instruments, gradients, flow rates, and column dimensions. Additionally, we will extend the system to account for retention drift with column aging. This easy-to-use retention projection system is accurate to 0.23% of the gradient time (e.g. 2.8 seconds in a 20 min gradient). There are three specific aims: 1) Select an optimal set of standard stationary phases and calibration solutes and determine the ultimate accuracy of the HPLC retention projection system, 2) Develop standard protocol for building and using the retention database and determine the reproducibility of the projection system among multiple labs/operators, and 3) Generate a 1,000 metabolite retention database and determine its value for metabolite identification by LC-MS. Due to its high accuracy, wide applicability, and ease of use, the HPLC retention projection system will considerably accelerate research in metabolomics and related fields that rely on the identification of unknowns by LC-MS. Such fields are on the brink of extraordinary discoveries in the causes, diagnosis, prevention, and cure of human diseases. This system will offer far more identification power out of even low-end LC-MS instruments, significantly broadening the base of labs capable of performing these cutting-edge experiments.
The work describes a methodology which will drastically improve the ability of existing LC-MS instrumentation, a vital tool in clinical analysis and biological research, to identify large numbers of small molecules in blood, urine, tissue, and other sources of human origin without any additional cost. It will not only improve high-end instrumentation, but will also allow the large number of researchers with only basic LC-MS instruments to perform cutting-edge analyses. This will broaden the base of laboratories capable of performing research in the growing field of metabolomics that promises extraordinary discoveries in the causes, diagnosis, prevention, and cure of human diseases such as cancer, diabetes, and cardiovascular disease.
Showing the most recent 10 out of 12 publications
