Journal of Chromatographic Science 2015;53:1078– 1083 doi:10.1093/chromsci/bmu169 Advance Access publication December 11, 2014
Article
Recovery of Proteins Affected by Mobile Phase Trifluoroacetic Acid Concentration in Reversed-Phase Chromatography Bala´zs Boba´ly, Vivien Mikola, Eniko´´ Sipko´, Zolta´n Ma´rta and Jeno´´ Fekete* Department of Inorganic and Analytical Chemistry, Budapest University of Technology and Economics, Szt. Gelle´rt te´r 4, Budapest 1111, Hungary *Author to whom correspondence should be addressed. Email: [email protected] Received 5 February 2014; revised 26 August 2014
It was found that recoveries of proteins depend on trifluoroacetic acid concentration in the mobile phase and showed maximum in the range of 0.01 –0.1 v/v%. Transferrin and lysozyme were used to evaluate the recoveries of proteins from dedicated reversed-phase columns. Different types of reversed-phase columns were evaluated, such as core shell type materials (Aeris Widepore with C4, C8 and C18 modification) as well as fully porous hybrid particles (Waters BEH, modified with C4 and C18 alkyl chains). Recoveries ranged between 60.7– 95.2% for transferrin and 72.1 –99.8% for lysozyme. Based on the data presented, at least two different adsorption effects, the wellknown hydrophobic and silanophilic/polar interaction might influence the recovery. In addition to this, conformational effects due to ion pairing with the acidic mobile phase additive might change them.
Introduction Reversed-phase chromatography is a widely used separation technique for the characterization of proteins in the pharmaceutical field (1, 2), for isolation tasks (3, 4), in proteomics (5 –7) and related research areas (8 – 12). In the separation of biological macromolecules, the stationary phase, the mobile phase composition, the experimental conditions and even the conformation and microheterogeneity of the solute play primary roles. Fekete et al. (2) and Nugent et al. (13) published expansive data on the key factors of reversed-phase protein separation. Modern instrumentation and column technology allows very efficient separation (14 – 17), whereas different column chemistries can be considered during the method development. Today’s chromatographic modeling software help the analyst to tune the selectivity through finding the appropriate experimental conditions (18). In the separation of proteins and peptides, a prerequisite of a successful analysis is to recover the injected solute from the stationary phase. Several studies reported moderate or weak recoveries of proteins from the bonded phases (4, 13, 19 – 25). This can be attributed to irreversible adsorption, or very slow desorption of the macromolecules from the surface of the particles. Irreversible adsorption can be related to (i) strong ion-exchange and/or H-bonding interactions between the unreacted surface silanols and the protein, even if ion-pairing agents, like trifluoroacetic acid (TFA), are used in the mobile phase. (ii) Pronounced hydrophobic interaction between highly hydrophobic proteins and the alkyl chains. Nugent et al. (13), Dillon et al. (22) and Fekete et al. (23, 24) investigated the effect of mobile phase temperature on the recovery of monoclonal
antibodies and other proteins. The recovery of the antibodies and their heavy chain fragments showed a significant increase at elevated temperatures. Hamada et al. (25) mentioned weak recoveries of hydrophobic proteins from long chain alkyl modified stationary phases. Eschelbach and Jorgenson (19) found that application of ultrahigh pressure can enhance the recovery of model proteins. Reh et al. (21) published a comparative study between the recoveries of different proteins on various reversedphase columns. Absolute mass recoveries ranged between 27 and 100%, depending on the stationary phase and the protein. Serious carry over effects also have been reported, using conventional mobile phase compositions (0.1% TFA in the eluents). Acidic ion-pairing agents and their concentration in the mobile phase play a double role in the separation. They prevent strong ion-exchange interactions by suppressing silanol ionization and charge the basic side chains of the protein, which form ion pairs with acid anion (26). The hydrophobic ion pair is supposed to be retained on the bonded phase (27). Several studies described the influence of ion-pairing additive and its concentration on peak shape and retention profile of proteins and peptides (19, 20, 25, 28 – 32). These studies reflected that ion-pairing additives and their concentration in the mobile phase generally determine the protein retention (the interactions between the macromolecules and the stationary phase.) To the best of our knowledge and in spite of the wide use of TFA, the influence of additive concentration on protein recovery has not been studied in a wide concentration range yet. The aim of this work was to evaluate the recovery of proteins from recently introduced reversed phases using TFA as an ion-pairing additive in a wide concentration range. Various columns packed with core shell (Phenomenex Aeris Widepore C4, XB-C8 and XB-C18), and hybrid (Waters BEH C4 and C18) particles have been used for the recovery study. Lysozyme and transferrin were used as model proteins.
Experimental Instrumentation and reagents Chromatographic experiments were performed on a Waters Acquity UPLC system (Waters, Milford, MA, USA). The instrument was equipped with a binary solvent manager; the maximum delivery flow rate was 2 mL/min. The autosampler was equipped with an 5-mL loop, operating in a full loop injection mode. The Waters PDA detector operated with an 500-nL flow cell. PDA spectra were recorded in the 210- to 300-nm range and at a
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
20-Hz sampling rate. Chromatograms were evaluated at 280 nm. Data acquisition, data handling and instrument control were performed by Empower Pro (Waters) software. Data were processed by MS Excel. The columns investigated in this study were Acquity UPLC BEH300 C18 (2.1 mm 100 mm, 1.7 mm) and C4 (2.1 mm 150 mm, 1.7 mm) obtained from Waters, and Aeris WIDEPORE XB-C18, XB-C8 and C4 (2.1 mm 150 mm, 3.6 mm) obtained from Phenomenex (Torrance, CA, USA). Water was obtained from a Milli-Q Purification System (Bedford, MA, USA). Acetonitrile of HPLC gradient grade and TFA were purchased from Sigma-Fluka (Budapest, Hungary). Protein standards such as transferrin (human, 77 kDa) and lysozyme (hen egg, 14.3 kDa) were obtained from Sigma-Fluka. Proteins were dissolved in water at 20 pmol/mL (1.54 mg/mL of transferrin and 0.286 mg/mL of lysozyme). To study the recovery of small injected amounts, 20 pmol/mL of transferrin solution was diluted 3, 10, 50 and 100 times by water. The concentrations of the resulting solutions were 513, 154, 77, 30.8 and 15.4 ng/mL. Protein solutions were kept at 48C between and under the chromatographic runs.
eluent composition. Each runs were repeated two times, then the average of the peak areas was used for calculation. Peak areas were integrated and compared with peak areas injected without the column (the column was replaced by a zero dead volume connector unit). Absolute mass recoveries of the injected proteins were calculated from the peak areas using the following form:
Methods Eluent A was water, and eluent B was acetonitrile. TFA was added to both eluents in the range of 0.005 –0.3 v/v%. Gradient elution was applied from 3% B to 55% B with a gradient slope of 5% B/min. The flow rate was 0.3 mL/min in each cases. Column temperature was set to 508C. Chromatographic runs were recorded from low TFA to high TFA concentrations. When changing the eluents, columns were flushed with 10 column volumes of the next eluent. To avoid carry over effects, a blank was run before injecting the proteins onto the columns. At low TFA concentration levels (below 0.1– 0.05 v/v%), the first blank after the protein samples showed 1 – 2% carry over, whereas the second blank showed only 0.1 – 0.3% in each cases. This was below the relative deviation of the method [see equation (2), below 0.5%]. Carry over effect in 55% B final acetonitrile concentration was very similar to carry overs observed with 100% B as final
Results
R¼
Areaon column 100 ½%: Areawithout column
ð1Þ
When studying the recovery of small injected amounts, a similar calibration method has been applied as described by Eschelbach and Jorgenson (19). Calibration has been performed using the chromatographic column, then the column was replaced by a zero dead volume connector unit. Using the connector unit, the samples were introduced directly into the detector cell. Recovery has been calculated using equation (1). When injecting small protein amounts, shortened gradients have been run from 20% B to 55% B with 5% B/min gradient steepness.
Recovery as the function of TFA concentration When no TFA was added to the mobile phase, proteins have not been eluted from the stationary phase, even at relatively high (80% B) organic content. This suggested irreversible adsorption of the macromolecules to the stationary phase. Results of duplicate runs have been averaged, and the calculated recovery plotted against the TFA concentration in the mobile phase (Figure 1). The relative deviation of the recoveries calculated by equation (2) was ,0.5% in all cases. RD ¼
jRi Rj 100 ½%; R
ð2Þ
where Ri is the particular injection and R is the average of the repeated injections.
Figure 1. Recovery of the proteins from different columns as the function of TFA concentration in the mobile phase.
Recovery of Proteins by TFA Concentration 1079
Maximal curve recoveries have been observed. When TFA concentration was set to 0.005 v/v%, partial recoveries of the model proteins have been observed. A further increase in the additive concentration to 0.01 v/v% led to better recoveries in all cases. The maximum absolute recovery of the proteins varied between 0.01 and 0.1 v/v% TFA in almost all cases (except transferrin on the BEH C4 column). Above 0.1 v/v% TFA, recoveries showed a slight decrease. Maximum recoveries varied with the columns and were protein-dependent. Average maximum recoveries were 90.4% in the case of transferrin, and 98.8% for lysozyme. As shown in Figure 1, columns showed similar behavior in the terms of TFA-dependent recovery, except the AWP C18 column. In the case of the AWP C18 column, the recoveries of both the lysozyme and transferrin were somewhat lower in the 0.05–0.3 v/v% TFA concentration range. The trend of recovery functions is presumably influenced by the nature of interaction between the stationary phase and the protein chain. However, the explanation and evaluation of interaction mechanisms for proteins in reversed-phase conditions is not straightforward. As reported by Fekete et al. (16), it is a mixture of hydrophobic, H-bonding and ion-exchange mechanisms. The effect of these interactions is also influenced by (i) accessibility of the pores (considering molecular weight and conformation), (ii) by the hydrophobic side chains, charged residues and the pI of the protein itself, and (iii) by the quality of the surface modification (surface coverage, endcapping and residual silanol activity). It has also been reported that proteins of similar molecular weight and pI behaved differently, when comparing their retention on AWP and BEH columns. In the present case, any significant difference has been observed in the recovery properties as the function of the modifying alkyl chain length. To study the recovery of small injected amounts, 0.077, 0.154, 0.77 and 2.57 mg of transferrin were injected onto the Aeris WP C18 column, then recoveries were calculated as described in Methods. Three different TFA concentrations have been investigated, namely the lowest (0.005 v/v%), the highest (0.3 v/v%) and a middle concentration (0.05 v/v%), where acceptable recoveries have been observed. Results showed when injecting small protein amounts (below 1 mg) recovery decreases. Above 1 mg of recoveries remained more or less constant and were comparable to the results obtained with different TFA concentrations and 7.7 mg injected amount. The lowest recovery was 47.7%, when 7.7 ng of transferrin was injected using 0.005 v/v% TFA in the mobile phase. Highest recovery has been observed with 0.05 v/v% TFA and 2.57 mg injected amount. Results are shown in Figure 2.
Peak shape and retentivity as the functions of TFA concentration Peak shape is a good indicator for monitoring strong interactions between the stationary phase and the solute (33). To evaluate possible interaction mechanisms influencing recovery, peak widths at half height of the peaks were plotted against TFA concentration in the mobile phase. Figure 3 shows a dramatic increase in peak widths if TFA concentration was smaller than 0.05 v/v%. This suggests the presence of strong interactions, possibly ion-exchange and/or H-bonding between the stationary phase and the solute. Transferrin showed increased peak widths at low TFA concentration, compared with lysozyme. Above 0.05 1080 Boba´ly et al.
Figure 2. Recovery of transferrin from the Aeris WP C18 column, when injecting small protein amounts.
v/v% TFA concentration peak width remained more or less constant and peaks showed an acceptable shape (Figure 4). Retentivity of the proteins increased with the TFA concentration in the well-known logarithmic type (Figure 5). The retentivity (measured by the apparent retention factor) of the BEH-C18 stationary phases was significantly greater than the BEH-C 4. This is reasonably driven by the more expressed hydrophobic retention mechanism, observed on the BEH columns, where the hydrophobicity increases with the alkyl chain length (16). The Aeris columns showed similar retention properties, compared with the BEH-C18. However, the retention mechanisms characterized on the Aeris columns possesses rather silanophilic nature. In the case of AWP columns, the modifying alkyl chain length did not show any significant contribution to retentivity. This might due to the mixed effect of different surface coverage and silanol activity of the Aeris bonded phases (16).
Discussion Maximum curve recoveries have been observed with both of the proteins as the functions of TFA concentration in the mobile phase. Nugent et al. reported similar observations in the case of phosphoric acid as a mobile phase additive, and higher recoveries if lower (i.e. 0.05 – 0.1%) TFA concentration is used. According to our knowledge, maximum curve recoveries for TFA have not been reported yet (13). As it was shown in this study, maximum curve behavior can be spread to TFA and probably to other usual acidic ion-pairing additives, if they were investigated in a wider concentration range. This may be due to multiple effects influencing the adsorption –desorption kinetics between the stationary phase and the macromolecules. At low TFA concentrations (below ca. 0.05 v/v%), strong interactions, like ion-exchange and H-bonding properties of the stationary phases, might decrease recovery. Below the pI of the proteins [ pIlysozyme ¼ 11 and pItransferrin ¼ 5.3 –6.1, depending on iron saturation (34)], basic amino acid residues (lysine, arginine, histidine and also the N-terminus) will become protonated, which results in carrying positive net charge and having relatively
Figure 3. Peak widths at half height of the protein peaks as the function of TFA concentration in the mobile phase.
Figure 4. UV chromatograms of lysozyme and transferrin at 280 nm on the AWP C18 column using different TFA concentrations in the mobile phase. Darker chromatograms represent lower TFA concentration.
Figure 5. Apparent retention factors of the proteins on different columns as the function of TFA concentration in the mobile phase.
extended conformation (35). As reported by Fekete et al. (16), acidic activity of the stationary phases was significantly greater at pH 2.7 than in the case of 0.1% TFA ( pH 1.9) in the mobile phase. Between 0.005 and 0.05 v/v% TFA, pH of the mobile phase
ranged between 3.2 and 2.2, measured in eluent A. In this pH range, the acidic activity of the silanol groups supposed to result in relatively strong interaction between the proteins and the surface of the particles, which decreased recovery. Decreasing peak Recovery of Proteins by TFA Concentration 1081
widths as the function of increasing TFA concentration are in line with this assumption, as shown in Figure 3. Geng and Regnier published a solvent displacement theory, in which it was assumed, that three-dimensional structure of a protein is a major factor affecting protein retention (36). At higher TFA concentrations (above 0.05 – 0.1 v/v%), proteins undergo a conformational change. Trifluoroacetate anions bind ion pairs with the charged basic residues, repulsive electronic forces will be masked and the overall protein becomes more hydrophobic. This ion pair has been described as a molten globule type structure (35). As assumed in the solvent displacement theory, increasing TFA concentration might result in more pronounced ion-pairing. The more TFA anions bind ion pairs with the protein ( probably depending also on the accessibility of the charged residues, which may change with the conformation), the more hydrophobic the TFA –protein ion pair might be. This may result in pronounced hydrophobic interaction with the bonded phase, decreasing recovery. This is in line with increasing retentivity as the function of TFA concentration in the mobile phase (Figure 5). As it is discussed in the section Recovery as the function of TFA concentration, the comparison of recovery trends in the case of different columns and proteins is not evident. Nevertheless, refining the increments of TFA concentrations might slightly move the maxima of the curves. From the information published earlier (16), the followings can be concluded. The BEH columns showed mainly hydrophobic retention properties and low retentivity when separating proteins, whereas the Aeris columns provided more expressed silanophilic (H-bonding and ion-exchange) activity, and were much retentive at 0.1 v/v% TFA. The higher silanophilic activity and relatively low surface coverage of the AWP C18, compared with AWP C4, might led to decreased recovery above 0.05 –0.1 v/v% TFA concentration in our study. Note that the columns were overloaded when studying the effect of mobile phase TFA concentration on recovery (16). About 100 pmol injected protein amount equals to 7.7 mg of transferrin and 1.4 mg of lysozyme. In some cases of lower injected protein amounts—probably due to the binding activity of the phases— lower recoveries have been observed. This effect was very similar at different TFA concentrations. Above 1 mg the active sites might overload; therefore, the recovery remains practically independent from the injected amount. The incomplete recovery is presumably due to slow desorption kinetics of the macromolecules, and therefore it is supposed to be reversible. If the sorption of the molecules on the active sites would be irreversible, the recovery should increase by repeated injections. This was not observed in this study. Relatively low carry over suggested that desorption occurs during the chromatographic run, even if desorbing species cannot be recognized as chromatographic peaks. Conclusion In this study, it was shown that TFA concentration in the mobile phase influences protein recovery from reversed phases, presumably due to conformational effects related to ion pairing. The results have been validated on different columns and model proteins. The contact area of the stationary phase is thought to play an important role also. The evaluation of the stationary phases in the terms of relative and absolute recovery is not straightforward. Their effect might depend on the surface area, the 1082 Boba´ly et al.
carbon load and type of the modification, the quantity of the remaining surface silanols and their activity. Therefore, the influencing properties of stationary phases on recovery should be investigated further. The recoveries were different for the two proteins. In general, transferrin has been less recovered than lysozyme. This might depend on the size and also on the pI of the protein. In this study, two different effects are reported influencing protein recovery from the stationary phase. At low TFA concentrations (below 0.05 v/v%), strong ion-exchange and H-bonding effects due to remaining silanols are assumed to decrease recovery. At higher TFA concentrations (above 0.1 v/v%), enhanced hydrophobic interaction between the stationary phase and the protein – TFA ion-pair is supposed to decrease recovery. In this study, we reported partial recoveries of the proteins, even if the columns have been overloaded. In the case of small injected protein amounts (below 1 mg), relative recoveries might be lower. The results showed significant differences between the stationary phases and TFA concentrations belonging to maximum recovery. Recovery might also depend on the protein and the injected amount. This study reflected that among other chromatographic parameters, such as temperature (22 – 24), pressure (19) and modifying alkyl chain length (25), mobile phase additive concentration also influences recovery when working with proteins in reversed-phase chromatography. This is supposed to be driven by different types of interaction mechanisms and probably also by conformational effects. When developing new reversed-phase chromatographic methods for proteins also, the recovery of macromolecules and its optimization through chromatographic conditions is recommended to be investigated.
Acknowledgments The authors thank Dr Tivadar Farkas (Phenomenex, Inc.) for providing the Aeris WP columns, and Dr Attila Gali (Waters Hungary Ltd) for providing the Waters BEH columns.
Funding This work was supported by Chemical Works of Gedeon Richter Plc. under Gedeon Richter Talent Foundation.
References 1. Staub, A., Guillarme, D., Schappler, J., Veuthey, J.-L., Rudaz, S.; Intact protein analysis in the biopharmaceutical field; Journal of Pharmaceutical and Biomedical Analysis, (2011); 55(4): 810–822. 2. Fekete, S., Veuthey, J.-L., Guillarme, D.; New trends in reversed-phase liquid chromatographic separations of therapeutic peptides and proteins: theory and applications; Journal of Pharmaceutical and Biomedical Analysis, (2012); 69: 9–27. 3. Jin, W.-H., Dai, J., Li, S.-J., Xia, Q.-C., Zou, H.-F., Zeng, R.; Human plasma proteome analysis by multidimensional chromatography prefractionation and linear ion trap mass spectrometry identification; Journal of Proteome Research, (2005); 4(2): 613– 619. 4. Zolotarjova, N., Mrozinski, P., Chen, H., Martosella, J.; Combination of affinity depletion of abundant proteins and reversed-phase fractionation in proteomic analysis of human plasma/serum; Journal of Chromatography A, (2008); 1189(1– 2): 332– 338. 5. Can˜as, B., Pin˜eiro, C., Calvo, E., Lo´pez-Ferrer, D., Gallardo, J.M.; Trends in sample preparation for classical and second generation proteomics; Journal of Chromatography A, (2007); 1153(1–2): 235–258.
6. Di Palma, S., Hennrich, M.L., Heck, A.J.R., Mohammed, S.; Recent advances in peptide separation by multidimensional liquid chromatography for proteome analysis; Journal of Proteomics, (2012); 75(13): 3791– 3813. 7. Bodzon-Kulakowska, A., Bierczynska-Krzysik, A., Dylag, T., Drabik, A., Suder, P., Noga, M., et al.; Methods for samples preparation in proteomic research; Journal of Chromatography B, (2007); 849(1 – 2): 1– 31. 8. Wuhrer, M., Deelder, A.M., Hokke, C.H.; Protein glycosylation analysis by liquid chromatography – mass spectrometry; Journal of Chromatography B, (2005); 825(2): 124– 133. 9. Everley, R.A., Croley, T.R.; Ultra-performance liquid chromatography/ mass spectrometry of intact proteins; Journal of Chromatography A, (2008); 1192(2): 239–247. 10. Jungbauer, A.; Chromatographic media for bioseparation; Journal of Chromatography A, (2005); 1065(1): 3– 12. 11. Urban, J., Jandera, P., Kucˇerova´, Z., van Straten, M.A., Claessens, H.A.; A study of the effects of column porosity on gradient separations of proteins; Journal of Chromatography A, (2007); 1167(1): 63 –75. 12. Ozohanics, O., Turia´k, L., Puerta, A., Ve´key, K., Drahos, L.; Highperformance liquid chromatography coupled to mass spectrometry methodology for analyzing site-specific N-glycosylation patterns; Journal of Chromatography A, (2012); 1259: 200–212. 13. Nugent, K.D., Burton, W.G., Slattery, T.K., Johnson, B.F., Snyder, L.R.; Separation of proteins by reversed-phase high-performance liquid chromatography: II. Optimizing sample pretreatment and mobile phase conditions; Journal of Chromatography A, (1988); 443: 381– 397. 14. Fekete, S., Kohler, I., Rudaz, S., Guillarme, D.; Importance of instrumentation for fast liquid chromatography in pharmaceutical analysis; Journal of Pharmaceutical and Biomedical Analysis, (2014); 87: 105–119. 15. Fekete, S., Dong, M.W., Zhang, T., Guillarme, D.; High resolution reversed phase analysis of recombinant monoclonal antibodies by ultrahigh pressure liquid chromatography column coupling; Journal of Pharmaceutical and Biomedical Analysis, (2013); 83: 273–278. 16. Fekete, S., Berky, R., Fekete, J., Veuthey, J.-L., Guillarme, D.; Evaluation of recent very efficient wide-pore stationary phases for the reversedphase separation of proteins; Journal of Chromatography A, (2012); 1252: 90 –103. 17. Fekete, S., Berky, R., Fekete, J., Veuthey, J.-L., Guillarme, D.; Evaluation of a new wide pore core– shell material (AerisTM WIDEPORE) and comparison with other existing stationary phases for the analysis of intact proteins; Journal of Chromatography A, (2012); 1236: 177–188. 18. Fekete, S., Rudaz, S., Fekete, J., Guillarme, D.; Analysis of recombinant monoclonal antibodies by RPLC: toward a generic method development approach; Journal of Pharmaceutical and Biomedical Analysis, (2012); 70: 158– 168. 19. Eschelbach, J.W., Jorgenson, J.W.; Improved protein recovery in reversed-phase liquid chromatography by the use of ultrahigh pressures; Analytical Chemistry, (2006); 78(5): 1697– 1706. 20. Garcı´ a, M.C., Hogenboom, A.C., Zappey, H., Irth, H.; Effect of the mobile phase composition on the separation and detection of intact proteins by reversed-phase liquid chromatography – electrospray mass spectrometry; Journal of Chromatography A, (2002); 957(2): 187–199. 21. Reh, E., Hahn, B., Lamotte, S.; Evaluation of stationary phases for 2-dimensional HPLC of proteins: Part 1. Validation of commercial RP-columns; Journal of Chromatography B, (2006); 844(2): 204–212. 22. Dillon, T.M., Bondarenko, P.V., Rehder, D.S., Pipes, G.D., Kleemann, G.R., Ricci, M.S.; Optimization of a reversed-phase high-performance
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
liquid chromatography/mass spectrometry method for characterizing recombinant antibody heterogeneity and stability; Journal of Chromatography A, (2006); 1120(1– 2): 112–120. Fekete, S., Rudaz, S., Veuthey, J.-L., Guillarme, D.; Impact of mobile phase temperature on recovery and stability of monoclonal antibodies using recent reversed-phase stationary phases; Journal of Separation Science, (2012); 35(22): 3113– 3123. Fekete, S., Veuthey, J.-L., Eeltink, S., Guillarme, D.; Comparative study of recent wide-pore materials of different stationary phase morphology, applied for the reversed-phase analysis of recombinant monoclonal antibodies; Analytical and Bioanalytical Chemistry, (2013); 405(10): 3137–3151. Hamada, T., Tanaka, H., Izumine, H., Ohira, M.; Evaluation of an embedded polar C4 phase for hydrophobic protein analysis by reversedphase liquid chromatography; Journal of Chromatography A, (2004); 1043(1): 27– 32. Corradini, D., Kalghatgi, K., Horva´th, C.; Effect of mobile phase additives on peptide retention in reversed-phase chromatography with pellicular and totally porous sorbents; Journal of Chromatography A, (1996); 728(1–2): 225– 233. Horvath, C., Melander, W., Molnar, I., Molnar, P.; Enhancement of retention by ion-pair formation in liquid chromatography with nonpolar stationary phases; Analytical Chemistry, (1977); 49(14): 2295–2305. Shibue, M., Mant, C.T., Hodges, R.S.; The perchlorate anion is more effective than the trifluoroacetate anion as an ion-pairing reagent for reversed-phase chromatography of peptides; Journal of Chromatography A, (2005); 1080(1): 49–57. Shibue, M., Mant, C.T., Hodges, R.S.; Effect of anionic ion-pairing reagent concentration (1– 60 mM) on reversed-phase liquid chromatography elution behaviour of peptides; Journal of Chromatography A, (2005); 1080(1): 58– 67. Shibue, M., Mant, C.T., Hodges, R.S.; Effect of anionic ion-pairing reagent hydrophobicity on selectivity of peptide separations by reversed-phase liquid chromatography; Journal of Chromatography A, (2005); 1080(1): 68–75. Chen, Y., Mehok, A.R., Mant, C.T., Hodges, R.S.; Optimum concentration of trifluoroacetic acid for reversed-phase liquid chromatography of peptides revisited; Journal of Chromatography A, (2004); 1043(1): 9– 18. Guo, D., Mant, C.T., Hodges, R.S.; Effects of ion-pairing reagents on the prediction of peptide retention in reversed-phase high-resolution liquid chromatography; Journal of Chromatography A, (1987); 386: 205– 222. Grand-Guillaume Perrenoud, A., Boccard, J., Veuthey, J.-L., Guillarme, D.; Analysis of basic compounds by supercritical fluid chromatography: attempts to improve peak shape and maintain mass spectrometry compatibility; Journal of Chromatography A, (2012); 1262: 205– 213. Legendre, J.M., Moineau, M.P., Menez, J.F., Turzo, A.; Fast analysis of transferrin-iron complexes using isoelectric focusing and in situ radioactive densitometry; Analyse rapide des complexes transferrinefer par isoelectrofocalisation et densitometrie radioactive sur le gel, (1985); 33(7): 741–745. Boba´ly, B., To´th, E., Drahos, L., Zsila, F., Visy, J., Fekete, J., et al.; Influence of acid-induced conformational variability on protein separation in reversed phase high performance liquid chromatography; Journal of Chromatography A, (2014); 1325: 155– 162. Geng, X., Regnier, F.E.; Retention model for proteins in reversedphase liquid chromatography; Journal of Chromatography A, (1984); 296: 15 –30.
Recovery of Proteins by TFA Concentration 1083
Article
Recovery of Proteins Affected by Mobile Phase Trifluoroacetic Acid Concentration in Reversed-Phase Chromatography Bala´zs Boba´ly, Vivien Mikola, Eniko´´ Sipko´, Zolta´n Ma´rta and Jeno´´ Fekete* Department of Inorganic and Analytical Chemistry, Budapest University of Technology and Economics, Szt. Gelle´rt te´r 4, Budapest 1111, Hungary *Author to whom correspondence should be addressed. Email: [email protected] Received 5 February 2014; revised 26 August 2014
It was found that recoveries of proteins depend on trifluoroacetic acid concentration in the mobile phase and showed maximum in the range of 0.01 –0.1 v/v%. Transferrin and lysozyme were used to evaluate the recoveries of proteins from dedicated reversed-phase columns. Different types of reversed-phase columns were evaluated, such as core shell type materials (Aeris Widepore with C4, C8 and C18 modification) as well as fully porous hybrid particles (Waters BEH, modified with C4 and C18 alkyl chains). Recoveries ranged between 60.7– 95.2% for transferrin and 72.1 –99.8% for lysozyme. Based on the data presented, at least two different adsorption effects, the wellknown hydrophobic and silanophilic/polar interaction might influence the recovery. In addition to this, conformational effects due to ion pairing with the acidic mobile phase additive might change them.
Introduction Reversed-phase chromatography is a widely used separation technique for the characterization of proteins in the pharmaceutical field (1, 2), for isolation tasks (3, 4), in proteomics (5 –7) and related research areas (8 – 12). In the separation of biological macromolecules, the stationary phase, the mobile phase composition, the experimental conditions and even the conformation and microheterogeneity of the solute play primary roles. Fekete et al. (2) and Nugent et al. (13) published expansive data on the key factors of reversed-phase protein separation. Modern instrumentation and column technology allows very efficient separation (14 – 17), whereas different column chemistries can be considered during the method development. Today’s chromatographic modeling software help the analyst to tune the selectivity through finding the appropriate experimental conditions (18). In the separation of proteins and peptides, a prerequisite of a successful analysis is to recover the injected solute from the stationary phase. Several studies reported moderate or weak recoveries of proteins from the bonded phases (4, 13, 19 – 25). This can be attributed to irreversible adsorption, or very slow desorption of the macromolecules from the surface of the particles. Irreversible adsorption can be related to (i) strong ion-exchange and/or H-bonding interactions between the unreacted surface silanols and the protein, even if ion-pairing agents, like trifluoroacetic acid (TFA), are used in the mobile phase. (ii) Pronounced hydrophobic interaction between highly hydrophobic proteins and the alkyl chains. Nugent et al. (13), Dillon et al. (22) and Fekete et al. (23, 24) investigated the effect of mobile phase temperature on the recovery of monoclonal
antibodies and other proteins. The recovery of the antibodies and their heavy chain fragments showed a significant increase at elevated temperatures. Hamada et al. (25) mentioned weak recoveries of hydrophobic proteins from long chain alkyl modified stationary phases. Eschelbach and Jorgenson (19) found that application of ultrahigh pressure can enhance the recovery of model proteins. Reh et al. (21) published a comparative study between the recoveries of different proteins on various reversedphase columns. Absolute mass recoveries ranged between 27 and 100%, depending on the stationary phase and the protein. Serious carry over effects also have been reported, using conventional mobile phase compositions (0.1% TFA in the eluents). Acidic ion-pairing agents and their concentration in the mobile phase play a double role in the separation. They prevent strong ion-exchange interactions by suppressing silanol ionization and charge the basic side chains of the protein, which form ion pairs with acid anion (26). The hydrophobic ion pair is supposed to be retained on the bonded phase (27). Several studies described the influence of ion-pairing additive and its concentration on peak shape and retention profile of proteins and peptides (19, 20, 25, 28 – 32). These studies reflected that ion-pairing additives and their concentration in the mobile phase generally determine the protein retention (the interactions between the macromolecules and the stationary phase.) To the best of our knowledge and in spite of the wide use of TFA, the influence of additive concentration on protein recovery has not been studied in a wide concentration range yet. The aim of this work was to evaluate the recovery of proteins from recently introduced reversed phases using TFA as an ion-pairing additive in a wide concentration range. Various columns packed with core shell (Phenomenex Aeris Widepore C4, XB-C8 and XB-C18), and hybrid (Waters BEH C4 and C18) particles have been used for the recovery study. Lysozyme and transferrin were used as model proteins.
Experimental Instrumentation and reagents Chromatographic experiments were performed on a Waters Acquity UPLC system (Waters, Milford, MA, USA). The instrument was equipped with a binary solvent manager; the maximum delivery flow rate was 2 mL/min. The autosampler was equipped with an 5-mL loop, operating in a full loop injection mode. The Waters PDA detector operated with an 500-nL flow cell. PDA spectra were recorded in the 210- to 300-nm range and at a
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
20-Hz sampling rate. Chromatograms were evaluated at 280 nm. Data acquisition, data handling and instrument control were performed by Empower Pro (Waters) software. Data were processed by MS Excel. The columns investigated in this study were Acquity UPLC BEH300 C18 (2.1 mm 100 mm, 1.7 mm) and C4 (2.1 mm 150 mm, 1.7 mm) obtained from Waters, and Aeris WIDEPORE XB-C18, XB-C8 and C4 (2.1 mm 150 mm, 3.6 mm) obtained from Phenomenex (Torrance, CA, USA). Water was obtained from a Milli-Q Purification System (Bedford, MA, USA). Acetonitrile of HPLC gradient grade and TFA were purchased from Sigma-Fluka (Budapest, Hungary). Protein standards such as transferrin (human, 77 kDa) and lysozyme (hen egg, 14.3 kDa) were obtained from Sigma-Fluka. Proteins were dissolved in water at 20 pmol/mL (1.54 mg/mL of transferrin and 0.286 mg/mL of lysozyme). To study the recovery of small injected amounts, 20 pmol/mL of transferrin solution was diluted 3, 10, 50 and 100 times by water. The concentrations of the resulting solutions were 513, 154, 77, 30.8 and 15.4 ng/mL. Protein solutions were kept at 48C between and under the chromatographic runs.
eluent composition. Each runs were repeated two times, then the average of the peak areas was used for calculation. Peak areas were integrated and compared with peak areas injected without the column (the column was replaced by a zero dead volume connector unit). Absolute mass recoveries of the injected proteins were calculated from the peak areas using the following form:
Methods Eluent A was water, and eluent B was acetonitrile. TFA was added to both eluents in the range of 0.005 –0.3 v/v%. Gradient elution was applied from 3% B to 55% B with a gradient slope of 5% B/min. The flow rate was 0.3 mL/min in each cases. Column temperature was set to 508C. Chromatographic runs were recorded from low TFA to high TFA concentrations. When changing the eluents, columns were flushed with 10 column volumes of the next eluent. To avoid carry over effects, a blank was run before injecting the proteins onto the columns. At low TFA concentration levels (below 0.1– 0.05 v/v%), the first blank after the protein samples showed 1 – 2% carry over, whereas the second blank showed only 0.1 – 0.3% in each cases. This was below the relative deviation of the method [see equation (2), below 0.5%]. Carry over effect in 55% B final acetonitrile concentration was very similar to carry overs observed with 100% B as final
Results
R¼
Areaon column 100 ½%: Areawithout column
ð1Þ
When studying the recovery of small injected amounts, a similar calibration method has been applied as described by Eschelbach and Jorgenson (19). Calibration has been performed using the chromatographic column, then the column was replaced by a zero dead volume connector unit. Using the connector unit, the samples were introduced directly into the detector cell. Recovery has been calculated using equation (1). When injecting small protein amounts, shortened gradients have been run from 20% B to 55% B with 5% B/min gradient steepness.
Recovery as the function of TFA concentration When no TFA was added to the mobile phase, proteins have not been eluted from the stationary phase, even at relatively high (80% B) organic content. This suggested irreversible adsorption of the macromolecules to the stationary phase. Results of duplicate runs have been averaged, and the calculated recovery plotted against the TFA concentration in the mobile phase (Figure 1). The relative deviation of the recoveries calculated by equation (2) was ,0.5% in all cases. RD ¼
jRi Rj 100 ½%; R
ð2Þ
where Ri is the particular injection and R is the average of the repeated injections.
Figure 1. Recovery of the proteins from different columns as the function of TFA concentration in the mobile phase.
Recovery of Proteins by TFA Concentration 1079
Maximal curve recoveries have been observed. When TFA concentration was set to 0.005 v/v%, partial recoveries of the model proteins have been observed. A further increase in the additive concentration to 0.01 v/v% led to better recoveries in all cases. The maximum absolute recovery of the proteins varied between 0.01 and 0.1 v/v% TFA in almost all cases (except transferrin on the BEH C4 column). Above 0.1 v/v% TFA, recoveries showed a slight decrease. Maximum recoveries varied with the columns and were protein-dependent. Average maximum recoveries were 90.4% in the case of transferrin, and 98.8% for lysozyme. As shown in Figure 1, columns showed similar behavior in the terms of TFA-dependent recovery, except the AWP C18 column. In the case of the AWP C18 column, the recoveries of both the lysozyme and transferrin were somewhat lower in the 0.05–0.3 v/v% TFA concentration range. The trend of recovery functions is presumably influenced by the nature of interaction between the stationary phase and the protein chain. However, the explanation and evaluation of interaction mechanisms for proteins in reversed-phase conditions is not straightforward. As reported by Fekete et al. (16), it is a mixture of hydrophobic, H-bonding and ion-exchange mechanisms. The effect of these interactions is also influenced by (i) accessibility of the pores (considering molecular weight and conformation), (ii) by the hydrophobic side chains, charged residues and the pI of the protein itself, and (iii) by the quality of the surface modification (surface coverage, endcapping and residual silanol activity). It has also been reported that proteins of similar molecular weight and pI behaved differently, when comparing their retention on AWP and BEH columns. In the present case, any significant difference has been observed in the recovery properties as the function of the modifying alkyl chain length. To study the recovery of small injected amounts, 0.077, 0.154, 0.77 and 2.57 mg of transferrin were injected onto the Aeris WP C18 column, then recoveries were calculated as described in Methods. Three different TFA concentrations have been investigated, namely the lowest (0.005 v/v%), the highest (0.3 v/v%) and a middle concentration (0.05 v/v%), where acceptable recoveries have been observed. Results showed when injecting small protein amounts (below 1 mg) recovery decreases. Above 1 mg of recoveries remained more or less constant and were comparable to the results obtained with different TFA concentrations and 7.7 mg injected amount. The lowest recovery was 47.7%, when 7.7 ng of transferrin was injected using 0.005 v/v% TFA in the mobile phase. Highest recovery has been observed with 0.05 v/v% TFA and 2.57 mg injected amount. Results are shown in Figure 2.
Peak shape and retentivity as the functions of TFA concentration Peak shape is a good indicator for monitoring strong interactions between the stationary phase and the solute (33). To evaluate possible interaction mechanisms influencing recovery, peak widths at half height of the peaks were plotted against TFA concentration in the mobile phase. Figure 3 shows a dramatic increase in peak widths if TFA concentration was smaller than 0.05 v/v%. This suggests the presence of strong interactions, possibly ion-exchange and/or H-bonding between the stationary phase and the solute. Transferrin showed increased peak widths at low TFA concentration, compared with lysozyme. Above 0.05 1080 Boba´ly et al.
Figure 2. Recovery of transferrin from the Aeris WP C18 column, when injecting small protein amounts.
v/v% TFA concentration peak width remained more or less constant and peaks showed an acceptable shape (Figure 4). Retentivity of the proteins increased with the TFA concentration in the well-known logarithmic type (Figure 5). The retentivity (measured by the apparent retention factor) of the BEH-C18 stationary phases was significantly greater than the BEH-C 4. This is reasonably driven by the more expressed hydrophobic retention mechanism, observed on the BEH columns, where the hydrophobicity increases with the alkyl chain length (16). The Aeris columns showed similar retention properties, compared with the BEH-C18. However, the retention mechanisms characterized on the Aeris columns possesses rather silanophilic nature. In the case of AWP columns, the modifying alkyl chain length did not show any significant contribution to retentivity. This might due to the mixed effect of different surface coverage and silanol activity of the Aeris bonded phases (16).
Discussion Maximum curve recoveries have been observed with both of the proteins as the functions of TFA concentration in the mobile phase. Nugent et al. reported similar observations in the case of phosphoric acid as a mobile phase additive, and higher recoveries if lower (i.e. 0.05 – 0.1%) TFA concentration is used. According to our knowledge, maximum curve recoveries for TFA have not been reported yet (13). As it was shown in this study, maximum curve behavior can be spread to TFA and probably to other usual acidic ion-pairing additives, if they were investigated in a wider concentration range. This may be due to multiple effects influencing the adsorption –desorption kinetics between the stationary phase and the macromolecules. At low TFA concentrations (below ca. 0.05 v/v%), strong interactions, like ion-exchange and H-bonding properties of the stationary phases, might decrease recovery. Below the pI of the proteins [ pIlysozyme ¼ 11 and pItransferrin ¼ 5.3 –6.1, depending on iron saturation (34)], basic amino acid residues (lysine, arginine, histidine and also the N-terminus) will become protonated, which results in carrying positive net charge and having relatively
Figure 3. Peak widths at half height of the protein peaks as the function of TFA concentration in the mobile phase.
Figure 4. UV chromatograms of lysozyme and transferrin at 280 nm on the AWP C18 column using different TFA concentrations in the mobile phase. Darker chromatograms represent lower TFA concentration.
Figure 5. Apparent retention factors of the proteins on different columns as the function of TFA concentration in the mobile phase.
extended conformation (35). As reported by Fekete et al. (16), acidic activity of the stationary phases was significantly greater at pH 2.7 than in the case of 0.1% TFA ( pH 1.9) in the mobile phase. Between 0.005 and 0.05 v/v% TFA, pH of the mobile phase
ranged between 3.2 and 2.2, measured in eluent A. In this pH range, the acidic activity of the silanol groups supposed to result in relatively strong interaction between the proteins and the surface of the particles, which decreased recovery. Decreasing peak Recovery of Proteins by TFA Concentration 1081
widths as the function of increasing TFA concentration are in line with this assumption, as shown in Figure 3. Geng and Regnier published a solvent displacement theory, in which it was assumed, that three-dimensional structure of a protein is a major factor affecting protein retention (36). At higher TFA concentrations (above 0.05 – 0.1 v/v%), proteins undergo a conformational change. Trifluoroacetate anions bind ion pairs with the charged basic residues, repulsive electronic forces will be masked and the overall protein becomes more hydrophobic. This ion pair has been described as a molten globule type structure (35). As assumed in the solvent displacement theory, increasing TFA concentration might result in more pronounced ion-pairing. The more TFA anions bind ion pairs with the protein ( probably depending also on the accessibility of the charged residues, which may change with the conformation), the more hydrophobic the TFA –protein ion pair might be. This may result in pronounced hydrophobic interaction with the bonded phase, decreasing recovery. This is in line with increasing retentivity as the function of TFA concentration in the mobile phase (Figure 5). As it is discussed in the section Recovery as the function of TFA concentration, the comparison of recovery trends in the case of different columns and proteins is not evident. Nevertheless, refining the increments of TFA concentrations might slightly move the maxima of the curves. From the information published earlier (16), the followings can be concluded. The BEH columns showed mainly hydrophobic retention properties and low retentivity when separating proteins, whereas the Aeris columns provided more expressed silanophilic (H-bonding and ion-exchange) activity, and were much retentive at 0.1 v/v% TFA. The higher silanophilic activity and relatively low surface coverage of the AWP C18, compared with AWP C4, might led to decreased recovery above 0.05 –0.1 v/v% TFA concentration in our study. Note that the columns were overloaded when studying the effect of mobile phase TFA concentration on recovery (16). About 100 pmol injected protein amount equals to 7.7 mg of transferrin and 1.4 mg of lysozyme. In some cases of lower injected protein amounts—probably due to the binding activity of the phases— lower recoveries have been observed. This effect was very similar at different TFA concentrations. Above 1 mg the active sites might overload; therefore, the recovery remains practically independent from the injected amount. The incomplete recovery is presumably due to slow desorption kinetics of the macromolecules, and therefore it is supposed to be reversible. If the sorption of the molecules on the active sites would be irreversible, the recovery should increase by repeated injections. This was not observed in this study. Relatively low carry over suggested that desorption occurs during the chromatographic run, even if desorbing species cannot be recognized as chromatographic peaks. Conclusion In this study, it was shown that TFA concentration in the mobile phase influences protein recovery from reversed phases, presumably due to conformational effects related to ion pairing. The results have been validated on different columns and model proteins. The contact area of the stationary phase is thought to play an important role also. The evaluation of the stationary phases in the terms of relative and absolute recovery is not straightforward. Their effect might depend on the surface area, the 1082 Boba´ly et al.
carbon load and type of the modification, the quantity of the remaining surface silanols and their activity. Therefore, the influencing properties of stationary phases on recovery should be investigated further. The recoveries were different for the two proteins. In general, transferrin has been less recovered than lysozyme. This might depend on the size and also on the pI of the protein. In this study, two different effects are reported influencing protein recovery from the stationary phase. At low TFA concentrations (below 0.05 v/v%), strong ion-exchange and H-bonding effects due to remaining silanols are assumed to decrease recovery. At higher TFA concentrations (above 0.1 v/v%), enhanced hydrophobic interaction between the stationary phase and the protein – TFA ion-pair is supposed to decrease recovery. In this study, we reported partial recoveries of the proteins, even if the columns have been overloaded. In the case of small injected protein amounts (below 1 mg), relative recoveries might be lower. The results showed significant differences between the stationary phases and TFA concentrations belonging to maximum recovery. Recovery might also depend on the protein and the injected amount. This study reflected that among other chromatographic parameters, such as temperature (22 – 24), pressure (19) and modifying alkyl chain length (25), mobile phase additive concentration also influences recovery when working with proteins in reversed-phase chromatography. This is supposed to be driven by different types of interaction mechanisms and probably also by conformational effects. When developing new reversed-phase chromatographic methods for proteins also, the recovery of macromolecules and its optimization through chromatographic conditions is recommended to be investigated.
Acknowledgments The authors thank Dr Tivadar Farkas (Phenomenex, Inc.) for providing the Aeris WP columns, and Dr Attila Gali (Waters Hungary Ltd) for providing the Waters BEH columns.
Funding This work was supported by Chemical Works of Gedeon Richter Plc. under Gedeon Richter Talent Foundation.
References 1. Staub, A., Guillarme, D., Schappler, J., Veuthey, J.-L., Rudaz, S.; Intact protein analysis in the biopharmaceutical field; Journal of Pharmaceutical and Biomedical Analysis, (2011); 55(4): 810–822. 2. Fekete, S., Veuthey, J.-L., Guillarme, D.; New trends in reversed-phase liquid chromatographic separations of therapeutic peptides and proteins: theory and applications; Journal of Pharmaceutical and Biomedical Analysis, (2012); 69: 9–27. 3. Jin, W.-H., Dai, J., Li, S.-J., Xia, Q.-C., Zou, H.-F., Zeng, R.; Human plasma proteome analysis by multidimensional chromatography prefractionation and linear ion trap mass spectrometry identification; Journal of Proteome Research, (2005); 4(2): 613– 619. 4. Zolotarjova, N., Mrozinski, P., Chen, H., Martosella, J.; Combination of affinity depletion of abundant proteins and reversed-phase fractionation in proteomic analysis of human plasma/serum; Journal of Chromatography A, (2008); 1189(1– 2): 332– 338. 5. Can˜as, B., Pin˜eiro, C., Calvo, E., Lo´pez-Ferrer, D., Gallardo, J.M.; Trends in sample preparation for classical and second generation proteomics; Journal of Chromatography A, (2007); 1153(1–2): 235–258.
6. Di Palma, S., Hennrich, M.L., Heck, A.J.R., Mohammed, S.; Recent advances in peptide separation by multidimensional liquid chromatography for proteome analysis; Journal of Proteomics, (2012); 75(13): 3791– 3813. 7. Bodzon-Kulakowska, A., Bierczynska-Krzysik, A., Dylag, T., Drabik, A., Suder, P., Noga, M., et al.; Methods for samples preparation in proteomic research; Journal of Chromatography B, (2007); 849(1 – 2): 1– 31. 8. Wuhrer, M., Deelder, A.M., Hokke, C.H.; Protein glycosylation analysis by liquid chromatography – mass spectrometry; Journal of Chromatography B, (2005); 825(2): 124– 133. 9. Everley, R.A., Croley, T.R.; Ultra-performance liquid chromatography/ mass spectrometry of intact proteins; Journal of Chromatography A, (2008); 1192(2): 239–247. 10. Jungbauer, A.; Chromatographic media for bioseparation; Journal of Chromatography A, (2005); 1065(1): 3– 12. 11. Urban, J., Jandera, P., Kucˇerova´, Z., van Straten, M.A., Claessens, H.A.; A study of the effects of column porosity on gradient separations of proteins; Journal of Chromatography A, (2007); 1167(1): 63 –75. 12. Ozohanics, O., Turia´k, L., Puerta, A., Ve´key, K., Drahos, L.; Highperformance liquid chromatography coupled to mass spectrometry methodology for analyzing site-specific N-glycosylation patterns; Journal of Chromatography A, (2012); 1259: 200–212. 13. Nugent, K.D., Burton, W.G., Slattery, T.K., Johnson, B.F., Snyder, L.R.; Separation of proteins by reversed-phase high-performance liquid chromatography: II. Optimizing sample pretreatment and mobile phase conditions; Journal of Chromatography A, (1988); 443: 381– 397. 14. Fekete, S., Kohler, I., Rudaz, S., Guillarme, D.; Importance of instrumentation for fast liquid chromatography in pharmaceutical analysis; Journal of Pharmaceutical and Biomedical Analysis, (2014); 87: 105–119. 15. Fekete, S., Dong, M.W., Zhang, T., Guillarme, D.; High resolution reversed phase analysis of recombinant monoclonal antibodies by ultrahigh pressure liquid chromatography column coupling; Journal of Pharmaceutical and Biomedical Analysis, (2013); 83: 273–278. 16. Fekete, S., Berky, R., Fekete, J., Veuthey, J.-L., Guillarme, D.; Evaluation of recent very efficient wide-pore stationary phases for the reversedphase separation of proteins; Journal of Chromatography A, (2012); 1252: 90 –103. 17. Fekete, S., Berky, R., Fekete, J., Veuthey, J.-L., Guillarme, D.; Evaluation of a new wide pore core– shell material (AerisTM WIDEPORE) and comparison with other existing stationary phases for the analysis of intact proteins; Journal of Chromatography A, (2012); 1236: 177–188. 18. Fekete, S., Rudaz, S., Fekete, J., Guillarme, D.; Analysis of recombinant monoclonal antibodies by RPLC: toward a generic method development approach; Journal of Pharmaceutical and Biomedical Analysis, (2012); 70: 158– 168. 19. Eschelbach, J.W., Jorgenson, J.W.; Improved protein recovery in reversed-phase liquid chromatography by the use of ultrahigh pressures; Analytical Chemistry, (2006); 78(5): 1697– 1706. 20. Garcı´ a, M.C., Hogenboom, A.C., Zappey, H., Irth, H.; Effect of the mobile phase composition on the separation and detection of intact proteins by reversed-phase liquid chromatography – electrospray mass spectrometry; Journal of Chromatography A, (2002); 957(2): 187–199. 21. Reh, E., Hahn, B., Lamotte, S.; Evaluation of stationary phases for 2-dimensional HPLC of proteins: Part 1. Validation of commercial RP-columns; Journal of Chromatography B, (2006); 844(2): 204–212. 22. Dillon, T.M., Bondarenko, P.V., Rehder, D.S., Pipes, G.D., Kleemann, G.R., Ricci, M.S.; Optimization of a reversed-phase high-performance
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
liquid chromatography/mass spectrometry method for characterizing recombinant antibody heterogeneity and stability; Journal of Chromatography A, (2006); 1120(1– 2): 112–120. Fekete, S., Rudaz, S., Veuthey, J.-L., Guillarme, D.; Impact of mobile phase temperature on recovery and stability of monoclonal antibodies using recent reversed-phase stationary phases; Journal of Separation Science, (2012); 35(22): 3113– 3123. Fekete, S., Veuthey, J.-L., Eeltink, S., Guillarme, D.; Comparative study of recent wide-pore materials of different stationary phase morphology, applied for the reversed-phase analysis of recombinant monoclonal antibodies; Analytical and Bioanalytical Chemistry, (2013); 405(10): 3137–3151. Hamada, T., Tanaka, H., Izumine, H., Ohira, M.; Evaluation of an embedded polar C4 phase for hydrophobic protein analysis by reversedphase liquid chromatography; Journal of Chromatography A, (2004); 1043(1): 27– 32. Corradini, D., Kalghatgi, K., Horva´th, C.; Effect of mobile phase additives on peptide retention in reversed-phase chromatography with pellicular and totally porous sorbents; Journal of Chromatography A, (1996); 728(1–2): 225– 233. Horvath, C., Melander, W., Molnar, I., Molnar, P.; Enhancement of retention by ion-pair formation in liquid chromatography with nonpolar stationary phases; Analytical Chemistry, (1977); 49(14): 2295–2305. Shibue, M., Mant, C.T., Hodges, R.S.; The perchlorate anion is more effective than the trifluoroacetate anion as an ion-pairing reagent for reversed-phase chromatography of peptides; Journal of Chromatography A, (2005); 1080(1): 49–57. Shibue, M., Mant, C.T., Hodges, R.S.; Effect of anionic ion-pairing reagent concentration (1– 60 mM) on reversed-phase liquid chromatography elution behaviour of peptides; Journal of Chromatography A, (2005); 1080(1): 58– 67. Shibue, M., Mant, C.T., Hodges, R.S.; Effect of anionic ion-pairing reagent hydrophobicity on selectivity of peptide separations by reversed-phase liquid chromatography; Journal of Chromatography A, (2005); 1080(1): 68–75. Chen, Y., Mehok, A.R., Mant, C.T., Hodges, R.S.; Optimum concentration of trifluoroacetic acid for reversed-phase liquid chromatography of peptides revisited; Journal of Chromatography A, (2004); 1043(1): 9– 18. Guo, D., Mant, C.T., Hodges, R.S.; Effects of ion-pairing reagents on the prediction of peptide retention in reversed-phase high-resolution liquid chromatography; Journal of Chromatography A, (1987); 386: 205– 222. Grand-Guillaume Perrenoud, A., Boccard, J., Veuthey, J.-L., Guillarme, D.; Analysis of basic compounds by supercritical fluid chromatography: attempts to improve peak shape and maintain mass spectrometry compatibility; Journal of Chromatography A, (2012); 1262: 205– 213. Legendre, J.M., Moineau, M.P., Menez, J.F., Turzo, A.; Fast analysis of transferrin-iron complexes using isoelectric focusing and in situ radioactive densitometry; Analyse rapide des complexes transferrinefer par isoelectrofocalisation et densitometrie radioactive sur le gel, (1985); 33(7): 741–745. Boba´ly, B., To´th, E., Drahos, L., Zsila, F., Visy, J., Fekete, J., et al.; Influence of acid-induced conformational variability on protein separation in reversed phase high performance liquid chromatography; Journal of Chromatography A, (2014); 1325: 155– 162. Geng, X., Regnier, F.E.; Retention model for proteins in reversedphase liquid chromatography; Journal of Chromatography A, (1984); 296: 15 –30.
Recovery of Proteins by TFA Concentration 1083
