Here, we present and evaluate a protocol for making low cost reversed phase nano-flow liquid chromatography columns for peptide characterization using LC-MS/MS proteomic workflows.
The high complexity prevalent in biological samples requires chromatographic separations with high sensitivity and resolution to be effectively analyzed. Here we introduce a robust, reproducible and inexpensive protocol for preparation of a nano-flow reversed phase high performance liquid chromatography (RP-HPLC) columns for on-line separation of analytical peptides before introduction into and detection by a mass-spectrometer in traditional bottom-up proteomics workflows. Depending on the goal of the experiment and the chemical properties of the analytes being separated, optimal column parameters may differ in their internal or outer diameters, length, particle size, pore size, chemistry of stationary phase particles, and the presence or absence of an integrated electrospray emitter at the tip. An in-house column packing system not only enables the rapid fabrication of columns with the desired properties but also dramatically reduces the cost of the process. The optimized protocol for packing a C18 AQ (aqueous) fused silica column discussed here is compatible with a wide range of liquid chromatographic instruments for achieving effective separation of analytes.
HPLC columns have contributed immensely to productivity in the fields of pharmaceutical, medical and environmental research1,2,3,4. Having access to high-quality chromatography columns is a pivotal step in the fractionation of complex analytes. In shotgun proteomics, high analytical sensitivity is routinely accomplished by coupling electrospray ionization (ESI) mass spectrometry (MS) to nanoflow chromatography5,6,7,8. The efficient separation of thousands of peptides is paramount in this application as it allows the mass spectrometer to identify and quantify analytes with high sensitivity and resolution.
The field of column packing for mass-spectrometric applications has witnessed tremendous growth in recent years with advances in the understanding of fundamental column packing principles related to stationary phase morphology, solvent-particle interactions and hardware design, making possible the detailed characterization of a wide range of biomolecules in complex biological settings9,10,11,12,13,14. Efforts highlighting practical considerations in packing analytical columns for LC-MS purposes have paved the way for proteomic laboratories to develop in-house packing systems to meet their specific interests with the promise of maximum performance15,16,17,18.
Nanospray columns with internal diameters in the range of 50-150 μm and tapered ends are well-suited for the purpose of electrospray ionization. In the field of shotgun proteomics, separations are typically carried out using a solvent gradient flowing through a packed non-polar stationary phase, most commonly hydrophobic carbon chain bonded silica (C8-C30) with particle sizes varying between 1.7 to 3.5 μm19,20,21,22. The eluting analytes are emitted through an ESI emitter integrated within the column, which ensures soft ionization of solution phase analytes to gaseous ions. Coupling LC columns with ESI-MS has significantly advanced the application of tandem mass spectrometry to proteomic strategies in biomedical sciences.
LC columns with narrow inner diameters result in narrower chromatographic peaks and higher sensitivity relative to higher bore, microflow columns and hence are particularly advantageous with proteomic workflows. Although commercially available pre-packed LC columns are attractive options due to their convenience and ease-of-use, they can be prohibitively expensive and less flexible than in-house options. The goal of this work is to describe a technically simple and low-cost slurry packing approach to prepare narrow inner diameter reversed phase HPLC columns using fused-silica capillaries and an in-house built pressure bomb system for proteomic applications.
1. Preparation of the capillary tip
2. Polymerization/etching of the tip
3. Preparation of stationary phase
4. Packing the column with stationary phase
5. Finishing the column and making the back-frit
To evaluate the performance of the columns, 750 ng of tryptic peptide digests prepared from whole cell lysates of HEK293 cells were fractionated online using a 25 cm long, 75 µm ID fused-silica capillary packed in-house with bulk ReproSil-Pur 120 C18-AQ particles as described in the protocol. Prior to sample loading, the column was washed using 6 µL of a mixture of acetonitrile, isopropanol and H2O in a ratio of 6:2:2 and pre-equilibrated with buffer A (Buffer A: water with 3% DMSO). The tryptic peptide digest was analyzed using a 70-minute reverse phase gradient. The solvent gradient began with buffer B (Buffer B: acetonitrile with 3% DMSO and 0.1% formic acid) increasing from 0 to 6% over 4 minutes at a flow rate of 400 nL/min. The flow rate was then reduced to 200 nL/min and a linear gradient starting at 6-25% buffer B was applied to the column over the course of next 58 minutes. Buffer B was further increased to 25-32% for a period of 8 minutes, followed by a rapid ramp-up to 85% for washing the column. The gradient composition was dropped to 1% buffer B for the remaining 4 minutes of chromatographic separation.
Peptides were ionized using a distal 2.2 kV spray voltage with an ion transfer capillary temperature of 275 °C and analyzed by tandem mass spectrometry (MS/MS) on an orbitrap mass spectrometer.
Data were acquired by a Data-Dependent Acquisition (DDA) method comprised of a full MS1 scan resolution of 120,000 FWHMat m/z 200 followed by sequential MS2 scans (Resolution = 15,000 FWHM) obtained using higher-energy collisional dissociation (HCD) to induce peptide fragmentation.
In this study, we used the Integrated Proteomics pipeline 2 to generate peptide and protein identifications. MS2 spectra were searched using the ProLuCID algorithm against the EMBL Human reference proteome (UP000005640 9606) followed by filtering by DTASelect using a decoy database-estimated false discovery rate of < 1%.
We evaluated column performance using a series of columns made at different points in time. The extracted chromatograms of 750 ng of HEK293 cell tryptic digest in 70-minute gradient runs are depicted in Figure 8. Retention time alignment, peak width, and peak intensity are reproducible across columns regardless of when the column was prepared suggesting the reproducibility of the protocol. As illustrated in Figure 9, the columns produced using our approach also demonstrate consistent performance in LC-MS/MS runs in terms of the number of peptide and protein identifications.
Figure 1. Representative image of fused-silica capillary portion with removed polyimide coating. Please click here to view a larger version of this figure.
Figure 2. Representative image illustrating loading of the column on the laser puller. Note how the polished portion of the capillary is aligned inside the laser puller. Please click here to view a larger version of this figure.
Figure 3. Comparison of pulled column tips with or without a frit. (A) A microscopic view of laser pulled tip before initiating frit polymerization. (B) A microscopic view of the laser pulled tip after frit polymerization. Please click here to view a larger version of this figure.
Figure 4. HF Etching station. The HF etching station for immersing the emitter after polymerization of the frit. Here, HF neutralizer solution is labeled as "HF EATER" which is used to wash the column tip after HF etching to safeguard against acid contamination. Please click here to view a larger version of this figure.
Figure 5. Schematic of column packing hardware. The hardware system basically consists of pressurized helium gas that is connected through a three-way valve to a high pressure packing chamber housing the vial containing slurry and column with its tip pointing upwards. Please click here to view a larger version of this figure.
Figure 6. Column packing station. (A) An overall view of the column set up on the packing bomb. The red arrows indicate the direction of the flow of helium gas from the helium hose to the 3-way valve and further into the column packing bomb, finally pushing the slurry (upwards) in to the column. (B) A close-up view of the column tip during packing on bomb. Note the solvent droplet formed at the tip of the column. Please click here to view a larger version of this figure.
Figure 7. Visualization of slurry movement through the column while packing on bomb. Note the inset image where movement of the slurry from the bottom to the tip of the capillary is seen due to helium gas flow inside the pressurized chamber. Please click here to view a larger version of this figure.
Figure 8. Evaluation of column performance using a series of columns prepared at different points in time. Comparison of the retention time alignment, peak intensity and width in a series of columns packed at different time points over a period of two years in 70 minute LC-MS/MS analysis of HEK293 cell tryptic digest. Please click here to view a larger version of this figure.
Figure 9. A bar graph representation of number of proteins and peptides identified in a 70 minute LC-MS/MS analysis of HEK293 cell tryptic digest on columns analyzed in Figure 8. Data reflect consistent performance of the columns packed using our approach. Please click here to view a larger version of this figure.
Modern proteomic strategies are reliant upon high quality chromatographic separations to effectively analyze complex biological systems. Hence, high-performing and cost-effective nanoflow LC columns are crucial components of a successful tandem mass-spectrometry regime aimed at characterizing thousands of proteins in a single workflow.
In this study we evaluated the performance and reliability of a range of LC columns for LC-MS/MS made using the protocol described above. The performance of these columns prepared over a period of two years was tested by using them for online fractionation of a HEK293 cell tryptic digest followed by analysis using tandem mass spectrometry. As shown in Figure 8 and Figure 9, a comparison of column parameters such as the alignment of chromatographic elution profiles and numbers of peptide and proteins identified is reproducible displaying less than 10% variability between different columns. These results reflect that columns made in different points in time using the protocol described above exhibit consistent performance and robust run-to-run reproducibility.
Taken together, the protocol presented here produces high-quality columns with low column-to-column variation for proteomic applications. Given the easily available raw materials and low-cost needed to adopt this in-house built column-packing approach, it can be swiftly implemented in many LC-MS laboratories for a wide range of MS-based bioanalytical applications. Further, the protocol provides flexibility for custom optimization such as column length, internal diameter, choice of particles and solvent for column packing, which are often guided by the biological questions being pursued.
The authors have nothing to disclose.
This work was supported by the National Institutes of Health grant GM089778 to J.A.W.
99.99% Formamide acid | Sigma-Aldrich | for making frit | |
alcohol lamp | Any brand | For providing heat | |
Brechbuehler helium pressure cell | BioSurplus | for packing column | |
Ceramic column cutter | Any brand | for cutting silica capillary | |
Dimethyl sulfoxide (DMSO) ≥ 99% | Sigma-Aldrich | Stored in a flammable cabinet | |
Formamide ≥99.5% | Sigma-Aldrich | for making frit | |
Hydrofluoric acid (HF) (50%) | Fisher Scientific | for opening the emitter after polymerization | |
KASIL (Potassium Silicate Solution) | PQ Corporation | for making frit | |
Orbitrap Fusion Lumos | Thermo Fisher Scientific | for MS data acquisition | |
P2000 Laser Puller | Sutter | for pulling capillary | |
PTFE 1/16" Ferrule 0.4 mm ID (long) for Tube Fitting | Chromre | 214104 | For bomb setting |
Reprosil-Pur 120 C18-AQ, 1.9 um, 1g | Dr. Masch GmbH | r119.aq.0001 | Batch 5910 |
Soldering | Any brand | For initiating polimerization | |
Stainless Steel Pipe Fitting, Hex Coupling, 1/4 in. Female NPT | Swagelok | SS-4-HCG | for bomb setting |
TSP075375 fused silica, 75 µm ID x 360 µOD | MOLEX/Polymicro | 1068150019 | For column tubing |
Ultimate 3000 UHPLC | Dionex | HPLC type |
.