Anal Chem.2017 Mar;89(6):3285-3292

Sample Stacking Provides Three Orders of Magnitude Sensitivity Enhancement in SDS Capillary Gel Electrophoresis of Adeno-Associated Virus Capsid Proteins

Chao-Xuan Zhang* and Michael M. Meagher

Department of Therapeutics Production & Quality, St. Jude Children’s Research Hospital, Memphis, Tennessee 38105, United States

PubMed link



Despite recent advancements in analytical techniques, current size-based protein analysis by SDS CGE still suffers from low sensitivity. It requires hundreds of micrograms of proteins loaded in the autosampler. The reported SDS CGE method with sample stacking herein delivers unprecedentedly high sensitivity, requiring only a few nanograms of proteins loaded in the autosampler. The actual amount injected is even less. It is not just a proof of principle study as this method has demonstrated to be robust in solving real-word problems.


Sensitivity is often a challenge in developing analytical strategy for solving real-world problems. In order to improve the sensitivity, a more sensitive and of course more expansive detector such as laser induced fluorescence (LIF) is commonly employed. Online sample pre-concentration techniques in CE provide convenient approaches for sensitivity improvement without modification of the instrument and without additional cost. Although numerous online sample pre-concentration techniques have been reported,[1] only a few of them are effective in real-world applications. Field-amplified samples stacking (FASS) is simple yet effective.[2] Samples prepared in low- conductivity solutions may be injected electrokinetically or hydrodynamically. Multiple terms have been reported for the virtually same technique with minor variations. In practice, stacking techniques employing samples of low conductivity can be classified into head-column FASS (for electrokinetic injection) and in-column FASS (for hydrodynamic injection).[3] In head-column FASS, sample is injected electrokinetically and stacking occurs at the head of the capillary during injection. For in-column FASS, sample is injected hydrodynamically and stacking occurs inside the capillary after injection. Each technique may be realized in a number of different ways, such as employing or not employing power polarity switching, removal or not removal of sample solvent by voltage or pressure, and with or without a pre-injection plug.


Decades of applications have demonstrated that head-column FASS generate much larger sensitivity enhancements than in-column FASS.[4],[5],[6],[7],[8],[9],[10] In-column FASS (also called large volume sample stacking) requires multiple steps of injection, stacking, sample matrix removing and separation with cumbersome programs. Sensitivity enhancement with in-column FASS is limited by the volume of sample that can be injected. The sensitivity enhancement factor of in-column FASS is generally less than two orders of magnitude, even when the whole capillary is filled with sample in order to achieve the largest injection volume. The large volume of sample matrix injected must be removed prior to separation, which is quite a delicate operation. In contrast, there is no limit on sample injection volume in head-column FASS. Only analytes are injected and stacked at the head of capillary with minimum sample solvent introduced.


Although 10s-fold sensitivity enhancements may be easily obtained with head-column FASS, it is not straightforward to achieve 1000s-fold sensitivity enhancements in real-world sample analysis. Introduction of a short pre-injection plug of low conductivity at the head of the capillary prior to the sample injection is critical to achieve high sensitivity in this work. Sample prepared with low conductivity is electrokinetically injected into the capillary filled with high conductivity running buffer zone and low conductivity pre-injection (water, for example) zone. During injection, the water zone will experience a much higher electric field than the running buffer zone. The negatively charged protein-SDS complexes move quickly from the sample vial into the water zone at the head of the capillary and slow down at the boundary to the running buffer. During injection/stacking, the EOF inside the capillary pushes the water zone toward the capillary inlet, avoiding sample matrix entering the capillary.


At the beginning of injection, there are two distinctive zones in the capillary: water zone (length fraction: χ) and SDS gel buffer zone. Based on the equations developed by Chien and Burgi,2 the injection amount of analyte I (Ni) can be expressed by

where Cj1 is the concentration of analyte I in the sample vial, µepi1 is the effective electrophoretic mobility of analyte I in the water zone, χ is the length fraction of water zone, µeo1eo2) is the electroosmotic mobility in the water zone (buffer zone), E0 is the electric field strength when the capillary is filled with homogenous buffer, A is the cross-sectional area of the capillary, and t is the injection time. At the end of injection, the stacked analyte zone width (Ls) is equal to the distance traveled by the analyte front in the buffer zone and given by

where µepi2 is the effective electrophoretic mobility of analyte I in the buffer zone. The sensitivity enhancement factor (SEF), defined as the ratio of analyte concentration in the stacked analyte zone (Ci2) to Ci1, is given by

Equation 3 shows that stacking efficiency is largely determined by the conductivity ratio and water zone length fraction. Note that χ changes with time during injection as the water zone is being pumped out of the capillary by EOF. A computer simulation study shows that the conductivity of water zone (thus, γ) varies with time as well.[11] It is hard to precisely predict the sensitivity enhancement factor based on Equation 3. In reality, a longer water zone will allow more analytes injected before the transient water zone disappears due to diffusion and EOF. However, too large a water zone will decrease the amount of analytes injected due to the reduced electric field strength. Thus, it is important to select an optimum water zone length. In this work, a water zone of around 10 mm generated the highest peaks. In a separate experimental setting, the optimum water zone length is also found to be about 10 mm.11



  1. Breadmore, M. C.; Wuethrich A, T.; Li, F.; Phung, S. C.; Kalsoom, U.; Cabot, J. M.; Tehranirokh, M.; Shallan, A.I.; Abdul Keyon, A. S.; See, H. H.; Dawod, M.; Quirino, J. P. Electrophoresis 2017, 38, 33-59.
  2. Chien R. L.; Burgi, D. S. Anal. Chem. 1992, 64, 489A-496A.
  3. Zhang, C.-X.; Thormann, W. Anal. Chem. 1996, 68, 2523–2532.
  4. Zhang, C.-X.; Aebi, Y.; Thormann, W. Clin. Chem. 1996, 42, 1805–1811.
  5. Song, J. Z.; Chen, J.; Tian, S, J.; Sun, Z. P. J. Pharm. Biomed. Anal. 1999, 21, 569-576.
  6. Chou, Y.-W.; Huang, W.-S.; Chen, C.-C.; Lin, S. J.; Wu, H. L.; Chen, S. H. J. Chromatogr. A 2005, 1087, 189–196.
  7. Yang, Y.; Boysen, R. I.; Hearn, M. T. W. Anal. Chem. 2006, 78, 4752-4758.
  8. Ito, E.; Nakajima, K.; Waki, H. Anal. Chem. 2013, 85, 7859−7865.
  9. Theurillat, R.; Sandbaumhüter, F. A.; Bettschart-Wolfensberger, R.; Thormann, W. Electrophoresis 2015, 37, 1129-1138.
  10. Ciura, K.; Pawelec, A.; Buszewska-Forajta, M.; Markuszewski, M. J.; Nowakowska, J.; Prahl, A.; Wielgomas, B.; Dziomba, S. J Sep Sci. 2017, 40, 1167-1175.
  11. Šesták, J.; Thormann W. J. Chromatogr. A 2017, 1502, 51-61.