Abstract
In the present work, pI-separation of proteins by pH-gradient ion-exchange chromatography (IEC) on packed capillary columns is demonstrated. An important technical challenge for adapting this technique to capillary dimensions, i.e. the development of a miniaturized flow-through pH probe for reliable pH monitoring of the column effluent, was solved by designing a low ÝL/min flow rate housing to a commercially available micro pH probe. Highly linear outlet pH gradients within the pH range 8.5 to 4.0 were obtained when applying simple buffering species consisting solely of piperazine, N-methylpiperazine and imidazole on 0.32 mm ID x 10 cm fused silica capillaries packed with anion-exchange polystyrene divinylbenzene (PS-DVB)-based macroporous materials, i.e. 10 Ým Mono P and 10 Ým PL-SAX. When using a pH gradient from pH 6.8 to 4.0, both columns were able to baseline separate the A and B genetic variants of £]-lactoglobulin, which differ with two amino acid residues only. However, the PL-SAX column provided almost a two-fold decrease in peak width compared to the Mono P column.
The influence of varying the buffer concentration, injection volume and column temperature on peak widths and resolution of £]-lactoglobulins was investigated, e.g., a 100 ÝL sample of dilute £]-lactoglobulins was injected directly on the column with practically no increase in peak width as compared to what obtained with conventional injection volumes. In addition, slightly better resolution was obtained with 10 mM buffers than with the other concentrations, and the narrowest peaks and best resolution were obtained between 30 and 50 oC.
Furthermore, a pH gradient, which was set to decrease from the pH of the start buffer (i.e. 10 mM piperazine and N-methylpiperazine, pH = 6.8) to the pH of the eluting buffer (i.e. 10 mM piperazine and N-methylpiperazine, pH = 4.3) in 25 min, was used to fractionate proteins in skimmed bovine milk on the PL-SAX column. The milk was simply diluted 1:10 (v/v) with water and filtrated before injection (5 ÝL), and the within-day repeatability (n = 6) of the retention time, peak width and peak area ranged from 0.3-3.0%, 3.1-6.9%, and 4.0-11.5%, respectively.
Finally, heart-cut 2D separation was performed, where bovine milk pI-fractions from the pH-gradient IEC column (first dimension) were injected on-line to a C8 reversed phase column (second dimension) where further separation according to hydrophobicity was achieved.