Affinity Precipitation and Macroaffinity Ligand FacilitatedThree-Phase Partitioning for Refolding and SimultaneousPurification of Urea-Denatured Pectinase

Aparna Sharma, Ipsita Roy, and Munishwar Nath Gupta*Chemistry Department, Indian Institute of Technology, Delhi, Hauz Khas, New Delhi 110016, India

Protein refolding is an integral step in the recovery of protein activity from inclusionbodies. It is shown that affinity precipitation and macroaffinity ligand facilitated three-phase partitioning (MLFTPP) led to refolding of urea-denatured pectinase present ina commercial preparation, with simultaneous purification. Affinity precipitation con-sists of precipitation of the desired enzyme by complexing it with a suitable stimulus-sensitive macroaffinity ligand. This ligand in this case was alginate/esterified alginate.The complex of the polymer-pectinase could be precipitated by adding calcium ions.In MLFTPP (carried out by adding tertiary butanol and ammonium sulfate to theaqueous solution of crude enzyme and the polymer), the polymer or its complex withthe enzyme form an interfacial precipitate between tert-butyl alcohol phase andaqueous phase. It is believed that in both processes, while molecular recognition ofalginate/esterified alginate to pectinase facilitates their selective binding to the enzyme,the correct refolding is facilitated by preventing molecular aggregation of unfoldedenzyme molecules. Three-phase partitioning with esterified alginate as the macroaf-finity ligand gave 100% recovery with 4-fold purification. Affinity precipitation with1% alginate gave 52% yield with 18-fold purification. On the other hand, use of 0.5%esterified alginate gave only 7-fold purification but with 75% recovery of activity.

Introduction

Recovery of active protein from overexpressed proteinsin Escherichia coli requires denaturation of inclusionbodies (generally carried out by 8 M urea or 6 Mguanidinium hydrochloride), followed by a refolding step(1). Attempts continue to be made for developing anefficient refolding protocol (2, 3). In this work, it is shownthat two processes, affinity precipitation and macroaf-finity ligand facilitated three-phase partitioning (MLFT-PP) showed considerable promise in refolding of pectinasedenatured by 8 M urea. The pectinase preparation usedwas a commercial one, and refolding was accompaniedby significant purification of active pectinase.

The advantage of developing a strategy for simulta-neous purification and renaturation is relevant becauseinclusion bodies generally consist of various contami-nants including nonproduct polypeptides, nucleic acids,and cell envelope components. Thus, unfolding/refoldinggenerally has to be followed by purification (4). Anotherviewpoint is that the presence of impurities in solubilizedinclusion bodies may interfere with the refolding processand prior purification may be required (5).

Purification of pectinase by affinity precipitation hasbeen reported by elution of active enzyme from theprecipitated complex of alginate-pectinase (6). MLFTPPof pectinase has also been reported and consists of mixingalginate, ammonium sulfate, and tert-butyl alcohol withthe crude preparation of the enzyme (7). The enzymeappears in the interfacial precipitate (between loweraqueous and upper tert-butyl alcohol rich layers) as an

affinity complex with alginate. The active enzyme couldbe stripped off the precipitated alginate-pectinase com-plex as in affinity precipitation using salt solution.

Materials and MethodsMaterials. Polygalacturonic acid (from oranges) was

purchased from Sigma Chemical Co., St. Louis, MO,USA. Pectinex Ultra SP-L (a highly purified preparationof pectolytic enzymes from a selected strain of Aspergillusniger) was from Novo Nordisk, Denmark and was ob-tained from Arun and Co., Mumbai, India. Urea (molec-ular biology grade) was obtained from Sisco ResearchLaboratories, Mumbai, India. Alginate (sold as ProtanalLF 10/60) and esterified alginate (sold as Protanal esterSDH) were from Pronova Biopolymer, Norway and wereobtained as gifts from Prof. Bo Mattiasson, Lund, Swe-den. All other chemicals used were of analytical grade.

Methods. Estimation of Pectinase Activity andAmount of Protein. Pectinase activity was estimatedusing polygalacturonic acid as the substrate (8). One unitof pectinase activity is described as the amount of enzymerequired to produce 1 µmol of reducing sugar (measuredas galacturonic acid) per minute under assay conditions.Protein was estimated by the dye binding method, usingbovine serum albumin as the standard protein (9).

Denaturation of Pectinase with 8 M Urea. Pecti-nase (16 µL containing 111 U) was incubated with 16 mLof 0.05 M Tris acetate buffer, pH 9.2 containing 10 Murea (final concentration 8 M) and 100 mM dithiothreitol(DTT), the final volume made up to 20 mL with 0.05 MTris acetate buffer, pH 9.2 and incubated at 25 °C for 4h. This denatured enzyme was subjected to MLFTPP andaffinity precipitation to check for refolding of the enzyme,after decreasing the pH to 3.8.

* To whom correspondence should be addressed. Fax: 91-11-2658 1073. Email: mn_gupta@hotmail.com.

1255Biotechnol. Prog. 2004, 20, 1255−1258

10.1021/bp0342295 CCC: $27.50 © 2004 American Chemical Society and American Institute of Chemical EngineersPublished on Web 01/17/2004

Macroaffinity Ligand Facilitated Three-PhasePartitioning. The denatured enzyme (1.6 mL) wasadded to 0.4 mL of alginate (stock solution of 5%, pHadjusted to 3.8), followed by addition of 30% (w/v)ammonium sulfate. tert-Butyl alcohol (4 mL) was addedto this, and the three phases formed at 37 °C wereseparated after 1 h, as described in ref 7. The boundpectinase was recovered from alginate, as described inref 6. Refolding of denatured pectinase was also at-tempted by incubating the enzyme (1.6 mL) with 0.4 mLof esterified alginate (stock solution of 2.5%, pH adjustedto 3.8), followed by addition of 20% (w/v) ammoniumsulfate. Volumetric equivalent of tert-butyl alcohol wasadded, and three-phase partitioning was carried out at37 °C, as described earlier. The enzyme was recoveredas described in ref 6.

Affinity Precipitation of Denatured Enzyme. Thedenatured enzyme (1.6 mL) was incubated with 0.4 mLof alginate (stock solution of 5%, pH adjusted to 3.8) and0.4 mL of esterified alginate (stock solution of 2.5%, pHadjusted to 3.8). The precipitation of the polymer/polymer-enzyme complex and recovery of the enzyme activity fromthe precipitate was carried out as described in ref 6.Refolding of the denatured enzyme was also attemptedat various other concentrations of the polymers.

A control was run where the urea-denatured enzyme(pH adjusted to 3.8) was incubated at 37 °C for 1 h andthen assayed for enzyme activity. The enzyme activityinitially taken has been considered as 100%.

ResultsTable 1 shows the results obtained with urea-dena-

tured pectinase by employing MLFTPP. Initially, theconditions reported earlier for MLFTPP (7) were usedand 1% alginate was employed. In that case, MLFTPPresulted in renaturation as well as purification of pecti-nase with 85% activity recovery and 2-fold purification.At this point, it was thought prudent to use a modifiedalginate that was more hydrophobic than alginate. Therationale was that stimuli-sensitive polymers such aspoly(N-isopropylacrylamide) are believed to work bybinding to hydrophobic patches on the unfolded proteins,facilitating prevention of aggregation and promotingcorrect folding (10). Even poly(ethylene glycol) and cy-clodextrin are reported to function via similar mechanism(5). Thus, hydrophobic interaction between modifiedalginate and pectinase was expected to be an additionalfavorable interaction for refolding of the enzyme. Hencea partially esterified alginate was used. The concentra-tion of the esterified alginate during MLFTPP was chosenas 0.5% since an earlier result had shown that theesterified alginate gave maximum precipitation duringthree-phase partitioning at this concentration (11). In thiscase (Table 1), 100% recovery of biological activity with4-fold purification was obtained.

Thus, although refolding could be carried out success-fully, fold purification was quite low. Hence it was

decided to try the alternative strategy of affinity precipi-tation. In this case, the separation of alginate-enzymecomplex is carried out by addition of calcium to formprecipitates of calcium alginate and the calcium salt ofthe alginate-enzyme complex (6). Table 2 show that thepercent recovery of the enzyme activity (and fold purifi-cation) was critically dependent upon the concentrationof the polymer during the refolding. It is useful to takeinto account the fact that a control (from which alginatehas been completely omitted) gave 32% recovery ofactivity. As this control had no precipitation step (sincealginate was absent), there was no purification involved.Thus, it is important to note that the presence of 0.3%alginate gave less recovery of activity than even control(Table 2). As this was known to be sufficient for affinityprecipitation of all active pectinase molecules (6), theconcentration of alginate required for optimum foldingis obviously greater than that required for just affinityprecipitation of active pectinase molecules. A comparisonof results obtained with 0.6% and 1% alginate is usefulto gain some insight into the process (Table 2). Theenzyme recovery was not grossly different, but the foldpurification was. The latter originated in substantialreduction in protein recovery, i.e., greater selectivity ofthe separation component of the process. Affinity pre-cipitation with urea-denatured pectinase with 1% algi-nate and 0.5% esterified alginate resulted in 56% yield(with 20-fold purification) and 74% yield (with 9-foldpurification), respectively (Table 2).

Figure 1 (panel A) shows the SDS-PAGE of crudepectinase, renatured and purified pectinase with MLFT-PP with 1% alginate, the same with 0.5% esterified algin-ate, renatured and purified pectinase with affinity precip-itation with 1% alginate, and the same with 0.5% ester-ified alginate. The data are in agreement with the datain the purification tables (Tables 1 and 2). The affinityprecipitation preparation, which gave greater fold purifica-tion, also showed greater purity on SDS-PAGE. Lessthan 100% recovery of the activity implies that aggre-grates and/or denatured proteins were also present in thepurified preparation. Neither SDS-PAGE (Figure 1,panel A) nor PAGE (Figure 1, panel B) shows thesemolecules.

DiscussionIn MLFTPP, the alginate-pectinase complex floats to

the interface. Thus, the first step in the most likelysequence of events seems to be binding of alginate tothose pectinase molecules that have folded into correctlyfolded intermediates. The exact mechanism of molecularrecognition of alginate by pectinase is not clear. It ispossible that these folding intermediates also possessstructural elements required for somewhat selectivebinding of the enzyme to alginate. Alginate at this stagealso suppresses aggregation and facilitates correct foldingof the enzyme molecule into native structure. This is also

Table 1. Recovery of Denatured Pectinase Activity,Using Macroaffinity Ligand Facilitated Three-PhasePartitioning

activity(U)

protein(µg)

spec act.(U mg-1)

yield(%)

foldpurif

starting samplebefore denaturation

111 115 965 100 1

recovered frominterfacial precipitate

using 1% alginate 95 44 2159 86 2using 0.5% esterified

alginate112 26 4307 100 4

Table 2. Recovery of Denatured Pectinase Activity,Using Affinity Precipitation with Alginate (A) andEsterified Alginate (B)

activity(U)

protein(µg)

spec act.(U mg-1)

yield(%)

foldpurif

starting beforedenaturation

111 115 965 100 1

recovered A B A B A B A B A B0.15%a 16 20 8.42 8 1904 714 14 18 2 0.70.3%a 16 27 6.0 25 2667 1080 14 24 3 10.6%a 47 90 6.0 11 7833 8182 42 81 8 81%a 62 82 3.2 9 19375 9111 56 74 20 9a Concentration of polymer.

1256 Biotechnol. Prog., 2004, Vol. 20, No. 4

likely to shift the equilibrium in favor of correctly foldedintermediates. Such a sequence of events is very similarto the well-accepted mechanism of folding catalysts-chaperones or other polymeric additives such as poly-(ethylene glycol) (12) or poly(N-isopropylacrylamide) (10).Almost simultaneously, partitioning of alginate moleculesto interfacial phase starts. This includes alginate mol-ecules bound to pectinase.

In affinity precipitation also the sequence of eventsmay have been similar to what has been discussed abovein the context of MLFTPP. The precipitation in this casewas triggered by addition of calcium. So, the presence ofenough alginate led to a majority of the enzyme moleculescomplexing with the polymer. Before precipitation, therefolding was complete. The selectivity is likely to haveoperated at both binding as well as elution stages as inall affinity based processes (13). One important issue hereis the role played by the affinity of alginate towardpectinase in the folding process. In many other cases suchas polyeythlene glycol (12) or poly(N-isopropylacrylamide)(10), the role of additives has been quite nonspecific. Alikely scenario seems to be that as specific bindingbetween the matrix and the enzyme generally has ahigher binding constant, the available alginate was firstused up in forming a complex with pectinase molecules.Alginate may have facilitated folding of other proteinmolecules as well. This is, in fact, more likely with theless hydrophilic form, i.e., esterified alginate. With es-terified alginate (Table 2), there was not much differencein the selectivity of the polymer at 0.6% and 1% concen-trations. In both the cases, similar levels of proteinconcentration were recovered. The higher concentrationfacilitated greater recovery, presumably by facilitatingfolding in a more effective way. The esterified alginatewas a better additive for folding (as reflected in therecovery of activities) but a poor choice for effective

purification. This is in line with the hydrophobic struc-tural motifs leading to nonspecific binding in severalseparation processes (13).

The outcome of these two somewhat similar strategiesis that we have two options. MLFTPP gives higheractivity recovery (100%) with lower fold purification (4-fold). On the other hand, affinity precipitation giveshigher fold purification (18-fold or 7-fold) but reducedactivity yields of 52% and 75% with alginate and esteri-fied alginate as affinity ligands, respectively. It may alsobe added that (i) enzymes such as pectinases, which aremostly used in food processing industries, are not re-quired in highly pure form (14), so renatured and purifiedpreparations obtained through both strategies reachacceptable level of purity, and (ii) unlike affinity precipi-tation, MLFTPP allows one to work directly with un-clarified crude extracts (15). If the starting broth/homogenate is not clear, it may be the preferred option.

Many additives that suppress aggregration and facili-tate correct refolding have been described in the litera-ture (3, 5). The range includes low molecular weightsubstances, e.g., arginine hydrochloride, cyclodextrins,and polymers such as poly(ethylene glycol) and poly(N-isopropylacrylamide). In some cases, such as argininehydrochloride, it is not clear how aggregation is pre-vented (5). Alginate, which has been successfully shownin this work to facilitate correct refolding is, as far as weknow, the first anionic polymer that has been found tobe a useful additive in this regard. Finally, the strategiesthat combine renaturation and purification may help indesigning more efficient downstream processing forproteins obtained as overexpressed proteins in Escheri-chia coli (4). The correct refolding of solubilized inclusionbodies and denatured proteins still involves a trial anderror approach (3, 5). Many protocols have been describedin the literature (1-3, 5). Some approaches are alreadyavailable for simultaneous purification and refolding.Apart from two-phase-based strategies, immobilizedmetal ion affinity chromatography has also been em-ployed (12, 16, 17). In most of the cases (where simulta-neous purification and refolding has been attempted),recovery of biological activity is usually very poor (16,17). For example, only 10% recovery of enzyme activityhas been reported for recombinant oxoglutarate/malatecarrier (17). A large number of successful affinity pre-cipitation studies for protein/enzyme purification havebeen described in the literature (6, 18). Thus, simplesmart macroaffinity ligands are available for a largenumber of proteins/enzymes such as xylanase (15), amy-lases (19), phospholipase D (20), and immunoglobulins(21). Hence the simple strategy outlined here may be yetanother approach for obtaining active and correctly foldedproteins/enzymes from their denatured states/inclusionbodies.

AcknowledgmentThe partial financial support provided by Department

of Science and Technology, Council for Scientific andIndustrial Research (Extramural Division and Technol-ogy Mission on Oilseeds, Pulses and Maize), all Govern-ment of India organizations is gratefully acknowledged.

References and Notes(1) Jaenicke, R.; Rudolph, R. Folding proteins. In Protein

Structure. A Practical Approach; Creighton, T. E., Ed.; IRLPress: Oxford, 1989; pp 191-223.

(2) Betts, S.; Speed, M.; King, J. Inhibition of aggregation sidereactions during in vitro protein refolding. In Methods inEnzymology; Wetzel, R., Ed.; Academic Press: New York,1999; Vol. 309, pp 333-344.

Figure 1. (A) SDS-PAGE of renatured pectinase. Electro-phoresis was carried out according to ref 22 using an 8% cross-linked polyacrylamide gel. Lane 1: crude extract of PectinexUltra SP-L (20 µg protein). Lane M: molecular weight markerproteins. Lane 2: renatured and purified pectinase with MLFT-PP with 1% alginate. Lane 3: renatured and purified pectinasewith MLFTPP with 0.5% esterified alginate. Lane 4: renaturedand purified pectinase with affinity precipitation with 1%alginate. Lane 5: renatured and purified pectinase with affinityprecipitation with 0.5% esterified alginate. The gel was stainedwith Coomassie Blue for 45 min and then destained. (B) PAGEof renatured pectinase. Electrophoresis was carried out undernondenaturing and nonreducing conditions, in a similar manneras described in the legend above. Lane C: crude extract ofPectinex Ultra SP-L (20 µg protein). Lane 1: renatured andpurified pectinase with MLFTPP with 1% alginate. Lane 2:renatured and purified pectinase with MLFTPP with 0.5%esterified alginate. Lane 3: renatured and purified pectinasewith affinity precipitation with 1% alginate. Lane 4: renaturedand purified pectinase with affinity precipitation with 0.5%esterified alginate. The gel was stained with Coomassie Bluefor 45 min and then destained.

Biotechnol. Prog., 2004, Vol. 20, No. 4 1257

(3) Middelberg, A. P. J Preparative protein refolding. TrendsBiotechnol. 2002, 20, 437-443.

(4) Middelberg, A. P.; O’Neill, B. K. Harvesting recombinantproteins from inclusion bodies. In Bioseparation and Biopro-cessing; Subramanian, G., Ed.; Wiley-VCH: Weinheim, 1998;Vol. 2, pp 81-106.

(5) Tsumoto, K.; Ejima, D.; Kumagai, I.; Arakawa, T. Practicalconsiderations in refolding proteins from inclusion bodies.Protein Expression Purif. 2003, 28, 1-8.

(6) Gupta, M. N.; Guoqiang, D.; Mattiasson, B. Purification ofendopolygalacturonase by affinity precipitation using algi-nate. Biotechnol. Appl. Biochem. 1993, 18, 321-324.

(7) Sharma, A.; Mondal, K.; Gupta, M. N. Separation of enzymesby sequential macroaffinity ligand-facilitated three-phasepartitioning. J. Chromatogr. A 2003, 995, 127-134.

(8) Bailey, M. J.; Poutanen, K. Production of xylanolytic en-zymes by strains of Aspergillus. Appl. Microbiol. Biotechnol.1989, 30, 5-10.

(9) Bradford, M. M. A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal. Biochem. 1976, 72,248-254.

(10) Chen, Y.-J.; Huang, L-.W.; Chiu, H.-C.; Lin, S.-C. Tem-perature-responsive polymer-assisted protein refolding. En-zyme Microb. Technol. 2003, 32, 120-130.

(11) Sharma, A.; Gupta, M. N. Three phase partitioning ofcarbohydrate polymers - Separation and purification ofalginates. Carbohydr. Polym. 2002, 48, 391-395.

(12) Kuboi, R.; Morita, S.; Ota, H.; Umakoshi, H. Proteinrefolding using stimuli-responsive polymer-modified aqueoustwo-phase systems. J. Chromatogr. B 2000, 743, 215-223.

(13) Roy, I.; Gupta, M. N. Selectivity in affinity chromatogra-phy. In Isolation and Purification of Proteins; Mattiasson, B.,Kaul-Hatti, R., Eds.; Marcel Dekker Inc.: New York, 2003;pp 57-94.

(14) Grassin, C. and Fauquembergue, P. Fruit juices. In In-dustrial Enzymology, Godfrey, T., West, S., Eds.; MacmillanPress Ltd.: London, 1998; pp 233-236.

(15) Sharma, A.; Gupta, M. N. Macroaffinity ligand facilitatedthree phase partitioning (MLFTPP) for purification of xyla-nase. Biotechnol. Bioeng. 2002, 80, 228-232.

(16) Glynou, K.; Ioannou, P. C.; Christopoulos, T. K. One-steppurification and refolding of recombinant photoprotein ae-quorin by immobilized metal-ion affinity chromatography.Protein Expression Purif. 2003, 27, 384-390.

(17) Smith, V. R.; Walker, J. E. Purification and folding ofrecombinant bovine oxoglutarate/malate carrier by immobi-lized metal-ion affinity chromatography. Protein ExpressionPurif. 2003, 29, 209-216.

(18) Roy, I.; Gupta, M. N. Macroaffinity ligands in biosepara-tion. In Methods for Affinity-Based Separation of Proteins/Enzymes; Gupta, M. N., Ed.; Birkhauser Verlag: Basel, 2002;pp 130-147.

(19) Sharma, A.; Sharma, S.; Gupta, M. N. Purification of wheatgerm amylase by precipitation. Protein Expression Purif.2000, 18, 111-114.

(20) Sharma, S.; Sharma, A.; Gupta, M. N. One step purificationof peanut phospholipase D by precipitation with alginate.Bioseparation 2000, 9, 93-98.

(21) Taipa, M. A.; Kaul, R.; Mattiasson, B.; Cabral, J. M.Preliminary studies on the purification of a monoclonalantibody by affinity precipitation with Eudragit S-100. J. Mol.Recognit. 1998, 11, 240-242.

(22) Hames, B. D. An introduction to polyacrylamide gel elec-trophoresis. In Gel Electrophoresis of Proteins: A PracticalApproach; Hames, B. D., Rickwood, D., Eds.; IRL Press:Oxford, 1986; pp 1-86.

Accepted for publication December 11, 2003.

BP0342295

1258 Biotechnol. Prog., 2004, Vol. 20, No. 4