Affinity-tagged green fluorescent protein (GFP) extraction from a clarified E. coli cell lysate using a two-phase aqueous micellar system

Green fluorescent protein (GFP) has been proposed as an ideal choice for a protein-based biological indicator for use in the validation of decontamination or disinfection treatments. In this article, we present a potentially scalable and cost-effective way to purify recombinant GFP, produced by fermentation in Escherichia coli, by affinity-enhanced extraction in a two-phase aqueous micellar system. Affinity-enhanced partitioning, which improves the specificity and yield of the target protein by specific bioaffinity interactions, has been demonstrated. A novel affinity tag, family 9 carbohydrate-binding module (CBM9) is fused to GFP, and the resulting fusion protein is affinity-extracted in a decyl beta-D-glucopyranoside (C10G1) two-phase aqueous micellar system. In this system, C10G1 acts as phase forming and as affinity surfactant. We will further demonstrate the implementation of this concept to attain partial recovery of affinity-tagged GFP from a clarified E. coli cell lysate, including the simultaneous removal of other contaminating proteins. The cell lysate was partitioned at three levels of dilution (5x, 10x, and 40x). Irrespective of the dilution level, CBM9-GFP was found to partition preferentially to the micelle-rich phase, with the same partition coefficient value as that found in the absence of the cell lysate. The host cell proteins from the cell lysate were found to partition preferentially to the micelle-poor phase, where they experience less excluded-volume interactions. The demonstration of proof-of-principle of the direct affinity-enhanced extraction of CBM9-GFP from the cell lysate represents an important first step towards developing a cost-effective separation method for GFP, and more generally, for other proteins of interest.     

Glucose-6-phosphate dehydrogenase partitioning in two-phase aqueous mixed (nonionic/cationic) micellar systems

The enzyme glucose-6-phosphate dehydrogenase (G6PD) plays an important role in maintaining the level of NADPH and in producing pentose phosphates for nucleotide biosynthesis. It is also of great value as an analytical reagent, being used in various quantitative assays. In searching for new strategies to purify this enzyme, the partitioning of G6PD in two-phase aqueous mixed (nonionic/cationic) micellar systems was investigated both experimentally and theoretically. Our results indicate that the use of a two-phase aqueous mixed micellar system composed of the nonionic surfactant C(10)E(4) (n-decyl tetra(ethylene oxide)) and the cationic surfactant C(n)TAB (alkyltrimethylammonium bromide, n = 8, 10, or 12) can improve significantly the partitioning behavior of G6PD relative to that obtained in the two-phase aqueous C(10)E(4) micellar system. This improvement can be attributed to electrostatic attractions between the positively charged mixed (nonionic/cationic) micelles and the net negatively charged enzyme G6PD, resulting in the preferential partitioning of G6PD to the top, mixed micelle-rich phase of the two-phase aqueous mixed micellar systems. The effect of varying the cationic surfactant tail length (n = 8, 10, and 12) on the denaturation and partitioning behavior of G6PD in the C(10)E(4) /C(n)TAB/buffer system was investigated. It was found that C(8)TAB is the least denaturing to G6PD, followed by C(10)TAB and C(12)TAB. However, the C(10)E(4)/C(12)TAB/buffer system generated stronger electrostatic attractions with the net negatively charged enzyme G6PD than the C(10)E(4)/C(10)TAB/buffer and the C(10)E(4)/C(8)TAB/buffer systems, when using the same amount of cationic surfactant. Overall, the two-phase aqueous mixed (C(10)E(4)/C(10)TAB) micellar system yielded the highest G6PD partition coefficient of 7.7, with a G6PD yield in the top phase of 71%, providing the optimal balance between the denaturing effect and the electrostatic attractions for the three cationic surfactants examined. A recently developed theoretical framework to predict protein partition coefficients in two-phase aqueous mixed (nonionic/ionic) micellar systems was implemented, and the theoretically predicted G6PD partition coefficients were found to be in reasonable quantitative agreement with the experimentally measured ones.     

Separating lysozyme from bacteriophage P22 in two-phase aqueous micellar systems

This communication demonstrates that two-phase aqueous mixed (nonionic/ionic) micellar systems have the potential for improving the separation of proteins from viruses. Specifically, two separation experiments were performed to show that the addition of the anionic surfactant sodium dodecyl sulfate (SDS) to the two-phase aqueous nonionic n-decyl tetra(ethylene oxide) (C(10)E(4)) micellar system increases the yield of a model net positively charged protein, lysozyme, in the micelle-rich phase from 75 to 95%, while still maintaining approximately the same yield of a model net negatively charged virus, bacteriophage P22, in the micelle-poor phase (97% vs. 98%).     

Understanding viral partitioning in two-phase aqueous nonionic micellar systems: 1. Role of attractive interactions between viruses and micelles

The partitioning behavior of viruses in the two-phase aqueous nonionic n-decyl tetra(ethylene oxide) (C10E4) micellar system cannot be fully explained by considering solely the repulsive, steric, excluded-volume interactions that operate between the viruses and the nonionic C10E4 micelles. Specifically, an excluded-volume theory developed recently by our group is not able to quantitatively predict the observed viral partition coefficients, even though this theory is capable of providing reasonable quantitative predictions of protein partition coefficients. To shed light on the discrepancy between the theoretically predicted and the experimentally measured viral partition coefficients, a central assumption underlying the excluded-volume theory that the viruses and the C10E4 micelles interact solely through repulsive, excluded-volume interactions was challenged in this study. In particular, utilizing bacteriophage P22 as a model virus, a competitive inhibition test and a partitioning study of the capsids of bacteriophage P22 were conducted. Based on the results of these two experimental studies, it was concluded that any attractive interactions between the tailspikes of bacteriophage P22 and the C10E4 micelles are negligible. Another experimental study was carried out wherein the partition coefficients of the model viruses, bacteriophages P22 and T4, were measured at various temperatures, and compared with those previously obtained for bacteriophage phiX174. This comparison also indicated that possible attractive, electromagnetic-induced interactions between the bacteriophage particles and the C10E4 micelles cannot be invoked to rationalize the observed discrepancy between the theoretically predicted and the experimentally measured viral partition coefficients.     

Isolation of plasmid DNA from cell lysates by aqueous two-phase systems

This work presents a study of the partitioning of a plasmid vector containing the cystic fibrosis gene in polyethylene glycol (PEG)/salt (K2HPO4) aqueous two-phase systems (ATPS). The plasmid was extracted from neutralized alkaline lysates using PEG with molecular weights varying from 200 to 8000. The effects of the lysate mass loaded to the ATPS (20, 40, and 60% w/w) and of the plasmid concentration in the lysate were evaluated. The performance of the process was determined by qualitative and quantitative assays, carefully established to overcome the strong interference of impurities (protein, genomic DNA, RNA), salt, and PEG. Plasmid DNA partitioned to the top phase when PEG molecular weight was lower than 400. The bottom phase was preferred when higher PEG molecular weights were used. Aqueous two-phase systems with PEG 300, 600, and 1000 were chosen for further studies on the basis of plasmid and RNA agarose gel analysis and protein quantitation. The recovery yields were found to be proportional to the plasmid concentration in the lysate. The best yields (>67%) were obtained with PEG 1000. These systems (with 40 and 60% w/w of lysate load) were able to separate the plasmid from proteins and genomic DNA, but copartitioning of RNA with the plasmid was observed. Aqueous two-phase systems with PEG 300 concentrated both plasmid and proteins in the top phase. The best system for plasmid purification used PEG 600 with a 40% (w/w) lysate load. In this system, RNA was found mostly in the interphase, proteins were not detected in the plasmid bottom phase and genomic DNA was reduced 7.5-fold.     

Characteristics of protein partitioning in an aqueous micellar-gel system

Partitioning of proteins has been studied experimentally in a system combining a gel-bead phase and a nonionic micellar phase. The micellar phase consists of cylindrically shaped micelles, which are completely excluded from the gel-bead phase. Partitioning of single-component protein solutions (myoglobin, ovalbumin, and BSA) is determined by excluded-volume interactions in the micellar phase, and as a result the proteins prefer the gel-bead phase to the micellar phase. The protein concentration inside the gel beads increases with an increase in volume fraction of the micelles and increases with an increase in the size of the proteins. The protein partition coefficients obtained for a binary mixture of myoglobin and bovine serum albumin (BSA) show the same protein concentration dependence as the single-component protein partition coefficients.     

Phase separation rates of aqueous two-phase systems: correlation with system properties

The kinetics of phase separation in aqueous two-phase systems have been investigated as a function of the physical properties of the system. Two distinct situations for the settling velocities were found, one in which the light, organic-rich (PEG) phase is continuous and the other in which the heavier, salt-rich (phosphate) phase is continuous. The settling rate of a particular system is a crucial parameter for equipment design, and it was studied as a function of measured viscosity and density of each of the phases as well as the interfacial tension between the phases. Interfacial tension increases with increasing tie line length. A correlation that describes the rate of phase separation was investigated. This correlation, which is a function of the system parameters mentioned above, described the behavior of the system successfully. Different values of the parameters in the correlation were fitted for bottom-phase-continuous and top-phase-continuous systems. These parameters showed that density and viscosity play a role in the rate of separation in both top continuous- and bottom continuous-phase regions but are more dominant in the continuous top-phase region. The composition of the two-phase system was characterized by the tie line length. The rate of separation increased with increasing tie line length in both cases but at a faster rate when the bottom (less viscous) phase was the continuous phase. These results show that working in a continuous bottom-phase region is advantageous to ensure fast separation.     

Concentration of mammalian genomic DNA using two‐phase aqueous micellar systems

The concentration of biomarkers, such as DNA, prior to a subsequent detection step may facilitate the early detection of cancer, which could significantly increase chances for survival. In this study, the partitioning behavior of mammalian genomic DNA fragments in a two‐phase aqueous micellar system was investigated using both experiment and theory. The micellar system was generated using the nonionic surfactant Triton X‐114 and phosphate‐buffered saline (PBS). Partition coefficients were measured under a variety of conditions and compared with our theoretical predictions. With this comparison, we demonstrated that the partitioning behavior of DNA fragments in this system is primarily driven by repulsive, steric, excluded‐volume interactions that operate between the micelles and the DNA fragments, but is limited by the entrainment of micelle‐poor, DNA‐rich domains in the macroscopic micelle‐rich phase. Furthermore, the volume ratio, that is, the volume of the top, micelle‐poor phase divided by that of the bottom, micelle‐rich phase, was manipulated to concentrate DNA fragments in the top phase. Specifically, by decreasing the volume ratio from 1 to 1/10, we demonstrated proof‐of‐principle that the concentration of DNA fragments in the top phase could be increased two‐ to nine‐fold in a predictive manner. Biotechnol. Bioeng. 2009;102: 1613–1623. © 2008 Wiley Periodicals, Inc.     

Parameters influencing protein extraction for whole broths in detergent based aqueous two-phase systems

The parameters important for an optimisation of cloud point extraction in technical scale were investigated using a genetically engineered fusion protein derived from endoglucanase I expressed in Trichoderma reesei and the nonionic polyoxyethylene Agrimul NRE 1205. The key parameters are temperature, detergent concentration, and additional salts. These parameters are interdependent, thus there is an optimum in the partition coefficient with respect to detergent concentration and a maximum for the partition coefficient and the yield with respect to temperature. These results were confirmed for the detergent C12E5 to demonstrate that these optima are due to the nature of polyoxyethylenes. Cloud point extraction was found to be only slightly affected by pH. In the case studied extraction of whole broth is favourable for a high yield and partition coefficient, since fusion protein adhering to the cells can be solubilized. However some loss of detergent which remains in the fungal biomass was observed. This revised version was published online in July 2006 with corrections to the Cover Date.     

Extraction of peptide tagged cutinase in detergent-based aqueous two-phase systems

Detergent-based aqueous two-phase systems have the advantage to require only one auxiliary chemical to induce phase separation above the cloud point. In a systematic study the efficiency of tryptophan-rich peptide tags was investigated to enhance the partitioning of an enzyme to the detergent-rich phase using cutinase as an example. Up to 90% enzyme activity could be extracted in a single step from whole broth of recombinant Saccharomyces cerevisiae expressing cutinase variants carrying a (WP)₄ tag. In contrast, the extraction yield of wild type cutinase was 2–3% only. The detergent concentration and the temperature are the main parameters to optimize the extraction yield. Considering availability, extraction yields, and price the detergent Agrimul NRE 1205 served best for enzyme recovery.     

Intensification of mass transfer in aqueous two-phase systems

A novel technique which intensifies conventional aqueous two-phase extraction by conversion of dispersed phase into colloidal gas aphrons (CGAs) has been developed for extraction of an enzyme. In the present work, amyloglucosidase (1,4-alpha-D-glucan glucohydrolase) was extracted using a polyethylene glycol-sodium sulfate-water system. The lighter phase, i.e., polyethylene glycol (PEG) rich phase, was converted into CGAs which were then dispersed into a salt rich phase. The effect of type of surfactant and its concentration, dispersed phase velocity, phase composition, and type of sparger on the dispersed phase mass transfer coefficient was investigated. The results suggests 9-16 times higher values of mass transfer coefficient compared to spray column. The multiorifice sparger at concentrations of 0.33 g/L of cetyl trimethyl ammonium chloride yielded best results. (c) 1993 John Wiley & Sons, Inc.     

Extraction of β-galactosidase fused protein a in aqueous two-phase systems

A method for the purification of proteins hybridized with β-galactosidase and produced in Escherichia coli is suggested. The method is based on the dominating properties of the β-galactosidase part of the molecule that are utilized for extraction in a poly(ethylene glycol) 4000/potassium phosphate aqueous two-phase system. The purification of the hybrid protein Staphylococcal protein A-Escherichia coli-β-galactosidase (SpA-βgal) produced in Escherichia coli is described. The partitioning of the cell debris and SpA-βgal depended on the distance to the critical point, i.e., the length of the tie line. A poly(ethylene glycol) top phase and an interface free from cell debris were obtained for a composition close to the binodial with a relatively short tie line. At this composition no Spa-βgal partitioned to the interface. When the length of the tie line was increased, more of the SpA-βgal was caught by the interface. The partitioning of SpA-βgal to the top phase was also affected by the salts present during the extraction. The utilization of SpA-βgal for affinity extraction has been investigated. Experiments with SpA-βgal and fluorescence-labeled human IgG(hIgG-F) in a poly(ethylene glycol) 4000/potassium phosphate aqueous two-phase system showed that the complex SpA-βgal-hlgG-F was partitioned to the interface, probably as a precipitate.     

Continuous cell partitioning using an aqueous two-phase flow system in microfluidic devices

We present a novel microfluidic system in which an aqueous two-phase laminar flow is stably formed, and the continuous partitioning of relatively large cells can be performed, eliminating the influence of gravity. In this study, plant cell aggregates whose diameters were 37-96 microm were used as model particles. We first performed cell partitioning using a simple straight microchannel having two inlets and two outlets and examined the effects of the flow rate and the phase width on partitioning efficiency. Second, by using a microchannel with a pinched segment, the partitioning efficiency was successfully improved. This microscale aqueous two-phase flow system can further be incorporated into micro total analysis systems (microTAS) or lab-on-a-chip technology, owing to its simplicity, applicability, and biocompatibility.     

On the use of mild hydrophobic interaction chromatography for "method scouting" protein purification strategies in aqueous two-phase systems: a study using model proteins

The behavior of a series of pure proteins partitioned in aqueous two-phase systems is compared with their behavior during mild hydrophobic interaction chromatography (HIC). A simple theoretical rationale for this comparison is presented based upon solvophobic theory. Similarities were found in the behavior of the model proteins in the two forms of partition chromatography. This indicates that HIC may be employed as a rapid instrumental technique for the broad characterization of protein behavior, which may be of benefit in the development of liquid-liquid partitioning strategies. However, it has proved difficult to completely account for this behavior on the basis of the known physical and structural properties of the proteins used. The variety in the detailed partitioning behavior of this small sample of protein types suggests that partition in aqueous two-phase systems is uniquely sensitive to subtle differences in surface properties of complex macromolecules. (c) 1994 John Wiley & Sons, Inc.     

Purification of plasmid DNA with polymer-salt aqueous two-phase system: optimization using response surface methodology

An experimental design was used to optimize plasmid purification from an alkaline lysate of Escherichia coli cells using PEG-sodium citrate aqueous two-phase systems (ATPS), and to evaluate the influence of pH, PEG molecular weight, tie line length, phase volume ratio, and lysate load. To build the mathematical model and minimize the number of experiments for the design parameters, response surface methodology (RMS) with an orthogonal rotatable central composite design was defined based on the conditions found for the highest purification by preliminary tests. The adequacy of the calculated models for the plasmid recovery and remaining RNA were confirmed by means of variance analysis and additional experiments. Analysis of contours of constant response as a function of pH, PEG molecular weight, tie line length, and cell lysate load for three different phase volume ratios revealed different effects of these five factors on the studied parameters. Plasmid recovery of 99% was predicted for a system with PEG 400, pH 6.9, tie line length of 38.7%, phase volume ratio of 1.5, and lysate load of 10% (v/v). Under these conditions the predicted RNA removal was 68%.     

Purification of recombinant cutinase by extraction in an aqueous two-phase system facilitated by a fatty acid substrate

Purification of recombinant wild-type cutinase from the culture supernatant of Saccharomyces cerevisiae by extraction in aqueous two-phase system was investigated. The partition of the enzyme in a polyethylene glycol (PEG)-potassium phosphate system to the top phase was increased with lower molecular weight PEG. Enzyme partition in a 20% PEG/15% phosphate two-phase system was studied in the presence of detergents, fatty acids, and alcohols, respectively. Addition of 0.5% (w/w) butyrate increased the partition coefficient from 17 to 135 and the purification factor from 10 to 23. The effect of butyrate was also confirmed by using the countercurrent mode of extraction. Recovery of cutinase from the top phase was achieved by a secondary extraction into a new salt phase at a lower pH or a lower temperature. A specific interaction of butyrate to the active site of the enzyme was demonstrated by fluorescence spectroscopy. Size exclusion chromatography showed the cutinase-butyrate complex to be over two times the size of the free enzyme.     

Ionic liquids for aqueous two-phase extraction and stabilization of enzymes

The ionic liquid (IL) Ammoeng110 contains cations with oligoethyleneglycol units and was found to be highly effective for the formation of aqueous two-phase systems (ATPS) that can be used for the biocompatible purification of active enzymes. Above critical concentrations of the IL and an inorganic salt in aqueous solution, phase separation takes place resulting in the formation of an IL-enriched upper and a salt-enriched lower phase. For the optimization of the composition of IL-based ATPS with regard to the extraction of catalytically active enzymes, the Box-Wilson method of experimental design was successfully applied; IL-based ATPS proved to be suitable for the purification and stabilization of two different alcohol dehydrogenases (from Lactobacillus brevis and a thermophilic bacterium). Both enzymes were enriched in the IL-containing upper phase resulting in an increase of specific activity by a factor of 2 and 4 respectively. Furthermore, the presence of IL within the system provided the opportunity to combine the extraction process with the performance of enzyme-catalyzed reactions. The IL was found to exhibit a stability improving effect on both enzymes and a solubility enhancing effect on hydrophobic substrates. Thus the conversion and volumetric productivity of ADH catalyzed reduction of acetophenone could be increased significantly.     

Effect of glycosylation on the partition behavior of a human antibody in aqueous two‐phase systems

Human proteins are expressed in some hosts wrongly glycosylated or nonglycosylated. Although it is accepted that glycosylation contributes to the stability of the protein in solution, the effect of glycosylation on the stability of human antibodies is not fully understood. In this work, we present solubility studies of two human antibodies that have the same primary structure but different glycosylation pattern. The studies were done by monitoring the partitioning behavior of both proteins in a series of aqueous two‐phase systems at and away the isoelectric point of the proteins and at different temperatures. Our studies show that in the absence of direct electrostatic forces, the partitioning behavior of the antibodies depends on the presence or absence of the polysaccharide chains. Overall, the nonglycosylated protein is less soluble than the glycosylated one. The potential of aqueous two‐phase systems for the separation of the glycosylated and nonglycosylated proteins was also explored. A simple series of extractions seems to be enough to separate the glycosylated variety from the nonglycosylated one at high purity but low yields. © 2013 American Institute of Chemical Engineers Biotechnol. Prog., 29:943–950, 2013     

Correlation for the partition behavior of proteins in aqueous two-phase systems: effect of surface hydrophobicity and charge

Correlations to describe the effect of surface hydrophobicity and charge of proteins with their partition coefficient in aqueous two-phase systems were investigated. Polyethylene glycol (PEG) 4000/phosphate, sulfate, citrate, and dextran systems in the presence of low (0.6% w/w) and high (8.8% w/w) levels of NaCl were selected for a systematic study of 12 proteins. The surface hydrophobicity of the proteins was measured by ammonium sulfate precipitation as the inverse of their solubility. The hydrophobicity values measured correlated well with the partition coefficients, K, obtained in the PEG/salt systems at high concentration of NaCl (r = 0.92-0.93). In PEG/citrate systems the partition coefficient correlated well with protein hydrophobicity at low and high concentrations of NaCl (r = 0.81 and 0.93, respectively). The PEG/citrate system also had a higher hydrophobic resolution than other systems to exploit differences in the protein's hydrophobicity. The surface charge and charge density of the proteins was determined over a range of pH (3-9) by electrophoretic titration curves; PEG/salt systems did not discriminate well between proteins of different charge or charge density. In the absence of NaCl, K decreased slightly with increased positive charge. At high NaCl concentration, K increased as a function of positive charge. This suggested that the PEG-rich top phase became more negative as the concentration of NaCl in the systems increased and, therefore, attracted the positively charged proteins. The effect of charge was more important in PEG/dextran systems at low concentrations of NaCl. In the PEG/dextran systems at lower concentration of NaCl, molecular weight appeared to be the prime determinant of partition, whereas no clear effect of molecular weight could be found in PEG/salt systems.