Nano-liquid chromatography-mass spectrometry and recent applications in omics investigations

Abstract

Liquid chromatography coupled to mass spectrometry (LC-MS) is one of the most powerful tools in identifying and quantitating molecular species. Decreasing column diameter from the millimeter to micrometer scale is now a well-developed method which allows for sample limited analysis. Specific fabrication of capillary columns is required for proper implementation and optimization when working in the nanoflow regime. Coupling the capillary column to the mass spectrometer for electrospray ionization (ESI) requires reduction of the subsequent emitter tip. Reduction of column diameter to capillary scale can produce improved chromatographic efficiency and the reduction of emitter tip size increased sensitivity of the electrospray process. This improved sensitivity and ionization efficiency is valuable in analysis of precious biological samples where analytes vary in size, ion affinity, and concentration. In this review we will discuss common approaches and challenges in implementing nLC-MS methods and how the advantages can be leveraged to investigate a wide range of biomolecules.

Figure 1.

Van Deemter curves generated for hexylbenzene on particles that are 1.7 μm (red diamonds), 3.5 μm (green squares), and 5.0 μm (blue triangles) in diameter. The column dimensions were 2.1 × 50 mm operated at 25°C with mobile phase 7:3 acetonitirile: water. Reprinted with permission from: U. D. N. Jeffrey R. Mazzeo, Marianna Kele, and Robert S. Plumb, Analytical Chemistry, 2005, 77, 460 A-467 A.

Figure 2.

Mechanism of electrospray ionization operating in positive ion mode. The applied voltage results in the formation of a Taylor cone followed by coulombic explosion resulting in the desolvation and ionization. Reprinted with permission from: L. Konermann, E. Ahadi, A. D. Rodriguez and S. Vahidi, Analytical Chemistry, 2013, 85, 2-9.

Figure 3.

Schematic illustrating common nano-injection methods (A) four-port injector method. (B) Gas pressure driven injection. (C) Split flow injection method. (D) trap column injection method.

Figure 4.

Groove injection method using a 12-port system in the load (left) and inject (right) position. Reprinted with permission from : V. Berry and K. Lawson, Journal of High Resolution Chromatography, 1988, 11, 121-123.

Figure 5.

Examples of different shapes of needles obtained by pulling a 170/100 μm (OD/ID) fused silica capillary. The needles were generated using (a) low plasma intensity and high pulling speed; (b) low plasma intensity and medium pulling speed; (c) medium plasma intensity and high pulling speed; (d) medium plasma intensity and medium pulling speed; (e) high plasma intensity and medium pulling speed; (f) high plasma intensity and high pulling speed. Reprinted with permission from: P. Ek and J. Roeraade, Analytical Chemistry, 2011, 83, 7771-7777.

Figure 6.

Chromatograms of endocannabinoids and N-acylethanolamine MS/MS fragmentations overlaid on each other from a representative standard mix. Reprinted with permission from: V. Kantae, S. Ogino, M. Noga, A. C. Harms, R. M. van Dongen, G. L. J. Onderwater, A. M. J. M. van den Maagdenberg, G. M. Terwindt, M. van der Stelt, M. D. Ferrari and T. Hankemeier, Journal of Lipid Research, 2017, 58, 615-624.

Figure 7.

MetPA analysis pathways. Node size and color indicate the degree of importance. Large red nodes are pathways with the highest level of change in diabetes. Orange, yellow, and white nodes represent moderate, slight, and zero importance, respectively. Reprinted with permission from: L. A. Filla, W. Yuan, E. L. Feldman, S. Li and J. L. Edwards, Journal of Proteome Research, 2014, 13, 6121-6134.

Figure 8.

(A) The sensitivity and the linear dynamic range of the nLC-MS method were benchmarked against a corresponding narrow-bore HPLC method in both positive- and negative-ionization mode for different lipid standards that were supplemented to a yeast lipid extract. The dynamic ranges of the respective classes were extrapolated using following standards: LCB_a from LCB 17:1;2 LCB_b from LCB 17:0;2, LCBP_a from LCBP 17:0;2; LCBP _b from LCBP 17:1;2; Cer and CerP from 18:1;2/12:0;0; LPS, LPG and LPA from their 17:1 species, LPC from LPC 13:0; PA, PC, PE, PG, PI and PS from their 17:0/14:1 species; CL from CL 15:0(3)/16:1; DAG from DAG 17:0/17:0 d5; TAG from TAG 17:0–17:1–17:0 d5. (B) and (C) Show the MS response (peak area) of the standards GlcCer 18:1;2/12:0;0 and PE 17:0/14:1 in dependency of the injection amount. Reprinted with permission from N. Danne-Rasche, C. Coman and R. Ahrends, Analytical Chemistry, 2018, 90, 8093-8101.

Figure 9.

(A) TIC of proteins extracted from swine heart tissue separated by C8@BTSEY column. (B) Top-down mass spectra and deconvoluted mass spectra of selected proteins with different masses. Italic “p” in red refers to phosphorylation. In panel g, the two peaks in the deconvoluted mass spectrum of αactin are two different isoforms, α-cardiac actin (41 813.82 Da) and α-skeletal actin (41 845.77 Da). Reprinted with permission from: Y. Liang, Y. Jin, Z. Wu, T. Tucholski, K. A. Brown, L. Zhang, Y. Zhang and Y. Ge, Analytical Chemistry, 2019, 91, 1743-1747.

Figure 10.

Schematic for two-column labeling setup. Valve 1 is housed on the autosampler, Valve 2 is housed in the LC column oven, and Valve 3 is an external valve housed on the MS stage. Valve 2 is a 10-port valve but is depicted with 6-ports for clarity. White indicates a valve that is switching. Labeling is performed in the following way: (A) A 5 μL sample loop on Valve 2 is filled with microdialysate (MD), and upon switching, the sample is directed onto the precolumn (PC) by the loading pump (LP). (B) A series of five autosampler (AS) injections delivers reagents from vials (In sequential order: light labeling reagent, formic acid, 100 pM peptide aqueous standards, heavy labeling reagent, and formic acid) onto the PC, which react with peptides that have adsorbed onto the PC. (C) Valve 3 switches, directing flow from the analytical pump (AP) through the PC and eluting peptides off the PC and onto the analytical column-mass spectrometer (LC-MS) by gradient elution. Reprinted with permission from R. E. Wilson, A. Jaquins-Gerstl and S. G. Weber, Analytical Chemistry, 2018, 90, 4561-4568.

Figure 11.

Oligosaccharide distribution in porcine milk. Results are expressed as the average of the relative abundance (%) of OS found in porcine milk from seven sows throughout the lactation period. Neutral, neutral-fucosylated, and acidic OS distributions exhibited changes by day (P < 0.05) throughout lactation. a,b Indicate statistical differences among time of milk collection during the lactation period, where means that do not share a common superscript letter differ (P < 0.05, HSD-Tukey Test). Reprinted with permission from A. T. Mudd, J. Salcedo, L. S. Alexander, S. K. Johnson, C. M. Getty, M. Chichlowski, B. M. Berg, D. Barile and R. N. Dilger, Frontiers in Nutrition, 2016, 3.