On-line buffer removal and fraction selection in gradient capillary high-performance liquid chromatography prior to electrospray mass spectrometry of peptides and proteins

Instrumental requirements for nanoscale liquid chromatography

Nanoscale liquid chromatography (nano-LC), with packed columns of typically 75 μm i.d. × 15 cm length, packed with C18, 5 μm of stationary phase, and optimal flow rates of 180 nL/min, can be considered as a miniaturized version of conventional HPLC. Using the down-scaling factor, which corresponds to the ratio of the column diameter in square, (d(conv)/d(micro))(2), excellent agreement between the theoretically calculated values and the values obtained using the down-scaling factor (∼3800) has been observed. This factor was applied to all system components, including flow rate, injection and detection volumes, and connecting capillaries. Down-scaling of a conventional HPLC system to a nano-LC system is easy to realize in practice and involves using a microflow processor for nanoflow delivery (50-500 nL/min), a longitudinal nanoflow cell (≤3 nL), a microinjection valve (≤ 20 nL), low-dispersion connecting tubing, and zero dead volume connections. Excellent retention time reproducibility was measured with RSD values of ±0.1% for isocratic and ±0.2% for gradient elution. Plates counts of more than 100 000/m indicate the excellent performance of the entire nano-LC system. With minimal detectable amounts of proteins in the low femtomole and subfemtomole ranges (e.g., 520 amol for bovine serum albumin), high mass sensitivity was found, making nano-LC attractive for the microcharacterization of valuable and/or minute proteinaceous samples. Coupling nano-LC with concomitant mass spectrometry using nanoscale ion spray or electrospray interfaces looks very promising and is obviously the next step for future work.

Capillary electrophoresis–mass spectrometry as a characterization tool for therapeutic proteins

With the increasing use of capillary electrophoresis (CE) in the biotechnology industry, there is a demand for analytical tools and methodology that can be used to characterize CE profiles. This article describes the implementation and optimization of a robust online CE-mass spectrometry (CE-MS) system used for the characterization of several CE assays developed at Genentech Inc. These assays include CE as a complement to reverse-phase peptide mapping for the identification of small peptides eluting in the void volume, profiling N-linked glycopeptide heterogeneity, and determining O-linked site occupancy. In addition, CE-MS was used to confirm major 8-aminopyrene-1,3,6-trisulfonate (APTS)-labeled glycans released from recombinant antibodies that are routinely profiled by CE-laser-induced fluorescence (CE-LIF). For each study, CE-MS was able to successfully identify components seen in UV or LIF electropherograms, thereby expanding the capability of CE and CE-MS for profiling biomolecules.

The use of a porous graphitic carbon column for desalting hydrophilic peptides prior to matrix-assisted laser desorption/ionization time-of-flight mass spectrometry

Reversed-phase high-performance liquid chromatography is the standard method for separating peptides generated from proteolytic digests of proteins. Typically, the small hydrophilic peptides from a proteolytic digest are recovered in the flowthrough fraction along with nonvolatile buffers and salts. Unfortunately, the presence of these salts can interfere with subsequent mass spectrometric analysis or Edman sequencing. To overcome this limitation, and thus enable these small peptides to be identified and characterized, we have investigated a porous graphitic carbon (PGC) column for desalting the peptides found in the unretained fraction. Using a volatile mobile phase combined with a PGC column (Hypercarb), we demonstrate that small hydrophilic peptides at the picomole level can be desalted and characterized by matrix-assisted laser desorption and ionization-time-of-flight-mass spectrometry. Even after desalting, we show that the choice of matrix still plays a significant role in distinguishing the small peptides from the matrix background. The utility of this approach is demonstrated with the flowthrough fraction of an endoproteinase Lys-C digest of a recombinant immunoglobulin. In addition, we demonstrate that a PGC column offers an alternative approach for the separation of hydrophilic, phosphorylated peptides from their unphosphorylated counterparts.

Use of nonvolatile buffers in liquid chromatography/mass spectrometry: advantages of capillary-scale particle beam interfacing

The use of nonvolatile buffers like phosphate and citrate is usually incompatible with mass spectrometric detection or may heavily interfere with ion generation. In the case of conventional HPLC flow rates, the continuous buffer deposition produces fouling of the mass spectrometer ion source and may worsen the performance of any LC/MS interface as well. Our research group demonstrated that reducing the mobile phase flow rate in a particle beam interface improves the nebulization and enhances the overall performance. We have also assumed that such a modification may better handle potentially instrument harmful solvents or nonvolatile HPLC buffers. Since the presence of salts does not interfere with the electron ionization process, the particle beam interface can be considered, in principle, particularly suitable for those chromatographic separations that benefit from nonvolatile buffers. The microscale flow rate interface, developed in our laboratory, generates an aerosol from 1 microL/min of mobile phase flow rate. Under these conditions, the absolute amount of a nonvolatile buffer actually introduced into the system is approximately 1/1000 of that carried by a conventional HPLC column, slowing its deposition. This assumption was verified with a real-world application requiring the use of a phosphate buffer at a maximum concentration of 10 mM. It involved the determination of dexamethasone, an anti-inflammatory drug, in human blood after its administration to a patient. Several crucial mass spectral and chromatographic parameters were monitored during a 35-day period of intense use. Neither appreciable signal modification nor evident buffer deposition was observed during the test period, with only a normal and limited run-by-run and day-by-day variation of the instrument response.