Amberlite XAD-4 is a convenient tool for removing Triton X-100 and Sarkosyl from protein solutions

Detergents are essential for the solubilization, purification and crystallization of membrane proteins. Usually, nonionic detergents are considered to be the most suitable agents for the isolation of membrane proteins under conditions preserving their native properties.

Triton X-100 is a neutral surfactant commonly used for the solubilization of membrane proteins [1,2]. However, the low critical micellar concentration of Triton X-100 (0.22–0.24 mM [3,4]) translates into its difficult removal from protein solutions by dialysis or size exclusion chromatography. Therefore, a number of alternative procedures have been defined to remove this detergent from solubilized proteins. Among these, the efficient adsorption of Triton X-100 by Bio-Beads or Amberlite XAD-2 via on-column or in-batch methods has been reported [5–7].

The overexpression in Escherichia coli of proteins whose natural copy number is low represents an invaluable technology to obtain and purify these proteins. Nevertheless, the overexpression in E. coli of a target protein can lead to its aggregation in vivo in the form of inclusion bodies [8,9], therefore demanding the solubilization of the protein of interest. Quite recently, a mild method to recover overexpressed proteins from inclusion bodies was devised [10]. According to this procedure, the anionic detergent N-lauroylsarcosine (also denoted as Sarkosyl) is used as the solubilizing agent, at concentrations ranging from 1% to 10% (w/v) [10]. Remarkably, it was reported that proteins solubilized with Sarkosyl can maintain their secondary structure [11], with the subsequent removal of Sarkosyl representing the only further step necessary to obtain their refolding. To this aim, Triton X-100 and the zwitterionic detergent CHAPS (Figure 1) are added to the solution containing proteins and Sarkosyl [10]. By this means, the presence of Triton X-100 and CHAPS favors the exchange of Sarkosyl from the solubilized proteins to mixed micelles [10]. It should indeed be noted that both Sarkosyl and CHAPS feature a relatively high critical micelle concentration (CMC) at 25°C (13–14 mM [11] and 6.5 mM [12], respectively).

Nevertheless, removal of the Sarkosyl–Triton–CHAPS (STC) detergent triad from target proteins is a difficult task, the accomplishment of which has not been inspected. Conversely, a simplified procedure for the recovery of aggregated proteins from inclusion bodies has been investigated [13]. In particular, overexpressed subunits of E. coli RNA polymerase were recovered from inclusion bodies using Sarkosyl, which was subsequently removed by dialysis [13]. To devise a simple and inexpensive procedure for the removal of the STC detergent triad or Sarkosyl only from protein solutions, the authors of the present study thought it of interest to investigate the use of Amberlite XAD-4. Indeed, both Amberlite XAD-2 and Amberlite XAD-4 are polystyrene-divinylbenzene copolymers, but the latter features a much higher surface area (about 300 vs ≥750 m²/g, respectively). Accordingly, the authors report here on the competence of Amberlite XAD-4 to adsorb Triton X-100, Sarkosyl and CHAPS, when individually present in solution or as components of a ternary mixture.

To detect the adsorption by Amberlite XAD-4 of Triton X-100, Sarkosyl and CHAPS (Figure 1), the authors selected a spectroscopic analytical method. Accordingly, they first determined the UV absorption spectra of these detergents over the 200–400 nm wavelength interval (Figure 2A). They then tested the adsorption by Amberlite XAD-4 of Triton X-100 at 20°C and observed that the resin did rapidly adsorb the detergent at 0.1% concentration (w/v, 1.6 mM) in 50 mM Tris-HCl, 50 mM NaCl, pH 7.5 (Figure 2B). In addition, a relatively fast and almost complete adsorption of Sarkosyl from a 1 % (w/v, 34.1 mM) solution was also observed under the same conditions (Figure 2C), whereas the performance of Amberlite in removing CHAPS (10 mM starting concentration) was limited to approximately 40% (Figure 2D). Quite a number of years ago, it was shown that Amberlite XAD-2 is very efficient in removing Triton X-100 from aqueous solutions [6]. Therefore, the authors of the present study decided to compare the competence of Amberlite XAD-4 in adsorbing Triton X-100, Sarkosyl and CHAPS with the corresponding performance of Amberlite XAD-2. As expected, the removal of 1.6 mM Triton X-100 by 1.9 g of Amberlite XAD-2 was observed to occur within 30 min after addition of the resin to the sample (Figure 2B). This implies that Amberlite XAD-2 is as efficient as Amberlite XAD-4 in the adsorption of Triton X-100. Surprisingly, the authors detected poor adsorption of Sarkosyl by Amberlite XAD-2, with Amberlite XAD-4 performing much better under the same conditions (Figure 2C). In the case of CHAPS, Amberlite XAD-2 removed only 10% of this detergent (Figure 2D), whereas Amberlite XAD-4 did adsorb ~40% of the zwitterionic surfactant (Figure 2D). According to the observations obtained with Triton X-100, Sarkosyl and CHAPS, Amberlite XAD-4 appears to feature a wider competence in adsorbing detergents when compared with Amberlite XAD-2, the action of which seems quite specific only for Triton X-100.

When the procedure proposed by Park et al. [14] is used to solubilize overexpressed proteins from E. coli inclusion bodies, the target protein/enzyme is finally recovered into a solution containing 0.8% (w/v, 27.3 mM) Sarkosyl, 0.1% (w/v, 1.6 mM) Triton X-100 and 0.06% (w/v, 1 mM) CHAPS (08-01-006 STC triad) [14]. Therefore, the authors of the present study evaluated the effectiveness of Amberlite XAD-4 in adsorbing detergents when 50 ml of the 08-01-006 STC triad in 50 mM Tris-HCl and, 50 mM NaCl (pH 7.5) was incubated with 1.9 g of the resin. Moreover, to determine if proteins were adsorbed by Amberlite XAD-4 when the resin was used to remove the detergents of the STC triad from an aqueous solution, the authors supplemented the sample with bovine serum albumin (BSA) at 1 mg/ml, obtaining the STC–BSA mixture. Under these conditions, and similarly to what was previously observed (Figure 2B), they detected an efficient displacement of Triton X-100 from the protein-containing sample (Figure 3A). Notably, they did not detect a significant adsorption of BSA, the concentration of which in solution appeared to be unaffected by the presence of Amberlite XAD-4, as revealed by SDS-PAGE and absorption spectrum analyses (Figure 3B & C). Remarkably, the spectroscopic analysis did also show that Triton X-100 was completely adsorbed by Amberlite XAD-4: over the 250–350 nm wavelength interval, the absorption spectrum of the sample exposed to the resin for 240 min did indeed reveal the presence of BSA only (Figure 3C). It is also interesting to note that the inability of Amberlite XAD-4 to adsorb BSA is in contrast with observations obtained with Amberlite XAD-8, which was shown to efficiently adsorb BSA from aqueous solutions [15].

To estimate the residual concentration of Sarkosyl in the STC–BSA mixture exposed for 240 min to Amberlite XAD-4, the authors performed multicomponent analyses using previously reported deconvolution procedures (Supplementary Figure 1). Taking into account that Amberlite XAD-4 did not adsorb BSA, the authors first subtracted the absorption spectrum of the protein from that of the STC–BSA mixture (Supplementary Figure 2A). Finally, using the appropriate derivative spectrum of the sample (Supplementary Figure 2B) and the calibration curve for Sarkosyl (Supplementary Figure 1E), they calculated the residual concentration of this detergent to be equal to 5.1 ± 0.4 mM, with this observation indicating that the presence of BSA did not affect the efficiency of Sarkosyl removal by Amberlite XAD-4 (cf.Figure 2C). Notably, when they assayed the removal by Amberlite XAD-4 of the STC triad from a solution also containing recombinant CRM197 (solubilized from E. coli inclusion bodies), they observed a behavior similar to that detected with the STC–BSA mixture (Figure 4). It is also interesting to note that Amberlite XAD-2 performed poorly on the STC–BSA mixture (Supplementary Figure 3 & cf.Supplementary Figure 2A).

According to the observations reported here, Amberlite XAD-4 appears to be an efficient tool to get protein solutions free from Triton X-100. Moreover, it is important to note that when exposing the STC–BSA mixture to Amberlite XAD-4,the concentration of Sarkosyl decreases to a level (about 5 mM) two- to three-times lower than the CMC of this surfactant. Further, it should be considered that the concentration of CHAPS (1 mM) used to solubilize proteins from inclusion bodies according to the procedure proposed by Park et al. [14] is much lower than the CMC of this detergent. Therefore, the authors of the present study suggest the use of Amberlite XAD-4 in protein purification operations requiring the use of detergents to remove Triton X-100 when overexpressed proteins are solubilized from inclusion bodies and then refolded according to the strategy introduced by Tao et al. [10] and to lower the concentration of Sarkosyl below its CMC when this surfactant is used alone or in detergent mixtures to solubilize overexpressed proteins from inclusion bodies.

Finally, it should be noted that Amberlite resins were applied to remove organic [16] and trace metal pollutants [17,18] from aqueous solutions. Moreover, Amberlite XAD-4 has been recognized as a safe material and has been successfully used for the hemoperfusion of plasma affected by acute drug intoxication [19,20]. Accordingly, and considering the observations reported here, the use of Amberlite XAD-4 to remove detergents from proteins of pharmaceutical relevance represents a convenient and inexpensive aid for industrial processes.