Variable column length method development strategy for amino acid analysis in serum samples of neonates with metabolic disorders

• Variable length method development strategy is presented for the analysis of amino acids.
• Superficially porous particles are used as an alternative to sub-2 μm particles.
• High resolution (>2.5) is achieved for amino acids prominent in metabolic disorders.
• Retention behavior of derivatized amino acids is strongly influenced by pressure.

Accurate quantitative analysis of amino acids forms an important part of targeted metabolic profiling as deviations from normal amino acid profiles can be indicative of several inborn metabolic disorders. A variable-length method development strategy for HPLC analysis on superficially porous particle columns using UV detection is proposed for the separation of 21 primary amino acids. This strategy consists of the evaluation of a large set of gradient and mobile phase compositions on short columns with different stationary phase properties to rapidly determine the best separation conditions leading to the highest number of separated peaks. These conditions are subsequently used to optimize the separation on a longer column length benefitting from its higher separation performance. A minimum critical resolution of 2.0 and a maximum analysis time of 10 min were imposed to allow for accurate, robust and rapid quantitation of all amino acids over a large linear range. For amino acids playing a key role in commonly encountered amino acid and urea cycle disorders, a minimum critical resolution of 2.5 was pursued. Mass spectrometry was used for fast peak tracking during the method development procedure. It was demonstrated that the retention behavior of the studied amino acids is more pressure than temperature dependent. This allowed performing the entire method development strategy at the same maximum pressure on all column lengths, while preserving the retention profile, resulting in a significant time profit for the initial scouting runs. The applicability of the method was demonstrated by analyzing serum samples of several neonates diagnosed with metabolic disorders such as maple syrup urine disorder, phenylketonuria and citrullinemia.