www.elsevier.com/locate/ybbrc

Biochemical and Biophysical Research Communications 331 (2005) 1284–1294

BBRC

Partially folded intermediate state of concanavalin A retainsits carbohydrate specificity

Aabgeena Naeem, Afroz Khan, Rizwan Hasan Khan *

Interdisciplinary Biotechnology Unit, Aligarh Muslim University, Aligarh 202002, India

Received 7 April 2005Available online 19 April 2005

Abstract

A systematic investigation of the effect of polyethylene glycol (PEG) 200 and 400 on the solution conformation of concanavalin A(con A) was made using circular dichroism (CD), tryptophan fluorescence, 1-anilino-8-naphthalenesulfonic acid (ANS) binding, andsize-exclusion chromatography. Far-UV CD spectra of con A at 30%(v/v) PEGs show the retention of ordered secondary structureas compared to 70%(v/v) PEGs. Near-UV CD spectra showed the retention of native-like spectral features in the presence of 30%(v/v) PEGs. Intrinsic tryptophan fluorescence studies indicate a change in the environment of tryptophan residues on the addition ofPEG. ANS binding was maximum at 30%(v/v) PEGs suggesting the compact ‘‘molten-globule’’-like state with enhanced exposure ofhydrophobic surface area. Size-exclusion chromatography indicates an intermediate hydrodynamic size at 30%(v/v) PEGs. GdnHCldenaturation of these states was a single-step, two-state transition. To study the possible minimum structural requirement in thespecific binding, the effect of PEGs on the interaction of con A with ligand was investigated by turbidity measurements. The C50

value was less in PEG 400 suggesting the more inhibitory ability of PEG 400. The C50 value of PEGs was highest for dextran fol-lowed by glycogen, ovalbumin, and ovomucoid. From percentage inhibition of con A–ligands at 30%(v/v) PEG, maximum inhibi-tion was in ovalbumin followed by ovomucoid, glycogen, and dextran. To summarize: con A at 30%(v/v) PEGs exists as compactintermediate with molten-globule-like characteristics, viz., enhanced hydrophobic surface area, retention of compact secondary aswell as tertiary structure, and a considerable degree of carbohydrate binding specificity and activity. This result has significant impli-cations on the molten globule state during the folding pathway(s) of proteins in general and quaternary association in the legumelectin in particular, where precise topology is required for their biological activities.� 2005 Elsevier Inc. All rights reserved.

Keywords: Concanavalin A; Circular dichroism; Molten globule; Polyethylene glycol; Specific binding

Protein folding is the process by which amino acid se-quence of a protein determines the three-dimensionalconformation of the functional protein. The elucidationof themolecular mechanism of protein folding from a dis-ordered polypeptide chain to specific native state, that isthe deciphering of second half of the genetic code [1], re-mains one of the major challenges in biochemistry [2].Study of the folding intermediates and denatured statesprovides an insight into understanding how and when

0006-291X/$ - see front matter � 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.bbrc.2005.04.041

* Corresponding author. Fax: +91 571 2721776.E-mail addresses: rizwanhkhan@hotmail.com, rizwanhkhan1@

yahoo.com (R.H. Khan).

various forces come into play in directing protein folding[3–9]. The development of a broad range of techniqueshas led to the identification and characterization of stablefolding intermediates, termed ‘‘molten globules,’’ thathave been shown to be of compact structure with pro-nounced secondary structure but lacking rigid tertiarystructure [10–17]. Lectins are proteins or glycoproteinsof non-immune origin that agglutinate cells and precipi-tate complex carbohydrates or polysaccharides. Theirinteraction with polysaccharides resembles the antibody–antigen and enzyme–substrate reactions. Lectins, whichhave a high degree of carbohydrate specificity, make auseful model for protein carbohydrate interactions

A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294 1285

[18,19]. They have been implicated in cell-to-cell interac-tion, ligand-receptor recognition, bloodgroup typing,and transport of biological macromolecules, and to someextent in the immune recognition process [20,21]. Exten-sive study of sequence homology and 3-D structure ofvarious lectins suggests that they are conserved through-out evolution, and thus may play an important role in thephysiology of plants. Though their exact role in plants isstill unknown, lectins have been implicated in pathogen-esis, cell elongation, defense against fungal attack, andRhizobium legume symbiosis. Unlike animals, plantshave no immune system but lectins are believed to inter-act with invading microorganism thus providing a lineof defense to plants against various bacteria and fungi.The recognition of carbohydrate determinates by lectinsplays an important role inmediation of intercellular bind-ing and elicitation of biosignaling processes [22–24].

Con A is a lectin from jack bean and is a well-studiedprotein because of its molecular structure and manymedical applications. The predominant secondary struc-ture in con A is b pleated sheet structure. More than halfof amino acid residues of the protein exist in b-confor-mation. But still a central question in biological chemis-try is what is the minimal structural requirement of aprotein that would determine specificity and activity.The underlying basis of the entire structural element ofa protein with regard to its activity vis-a-vis the overallintegrity and stability of the protein is lacking. Herenovelty lies in the characterization of an intermediateof con A with the molten-globule-like properties in thepresence of polyethylene glycols that retains its carbohy-drate binding specificity.

Materials and methods

Materials

Con A, ovalbumin, ovomucoid, and glycogen were purchased fromSigma, St. Louis, MO, USA. Dextran was purchased from BDHChemicals, Poole, England. PEG 200 and 400 were purchased fromQualigens fine chemicals. All other chemicals used in this study were ofanalytical grade.

Methods

The purity of commercial preparation was checked by SDS–PAGE. Activity of purified preparation was checked by agglutinationagainst 2% trypsinized rabbit erythrocytes.

CD measurements. CD was measured with a JASCO J-720 spec-tropolarimeter calibrated with ammonium D-10-camphorsulfonate. Acell of pathlengths 0.1 and 1 cm was used for scanning between 250–200 and 300–250 nm, respectively. Protein concentration of the sam-ples was typically 15 and 30 lM in 60 mM phosphate buffer of pH 7.0for the far-UV and near-UV CD studies, respectively. The results wereexpressed as mean residue ellipticity (MRE) in degree cm2/dmol, whichis defined as

MRE ¼ hObs ðmdegÞ=10� n� Cp � l;

where hObs is the observed ellipticity in degrees, Cp is the molar frac-tion, and l is the length of light-path in cm [25].

Fluorescence measurements. Fluorescence spectra were recordedwith a Shimadzu RF 540 spectrofluorometer in a 10 mm pathlengthquartz cell. Samples containing con A with different concentrations ofPEG 200 and PEG 400 were equilibrated at room temperature for30 min before recording for tryptophan fluorescence measurement.The excitation wavelength was 280 nm and emission wavelength wasrecorded from 300 to 400 nm, respectively [26]. The final proteinconcentration was 15 lM. For each sample proper blank was takeninto consideration.

ANS fluorescence measurements. ANS binding was measured byfluorescence emission with excitation at 380 nm and emission was re-corded from 400 to 600 nm, respectively [27]. Typically, ANS con-centration was 100 molar excess of protein concentration and proteinconcentration was in the vicinity of 15 lM in 60 mM phosphate buffer,pH 7.0. It should be noted that PEG itself binds to ANS, therefore toavoid anomaly, proper blank was made for every point.

Size-exclusion chromatography. Size-exclusion chromatography(SEC) experiments were performed on a seralose 6B (74 · 1.1) cmcolumn. The column was pre-equilibrated with 60 mM sodium phos-phate buffer, pH 7.0, and 30% PEG 400 or 200. Two milliliters of 3mg/ml native and protein in 30%(v/v) PEG 400 or 200 was applied tothe column and eluted at 20 ml/h. The eluted fractions were readat 280 nm.

GdnHCl stability studies. To determine the GdnHCl stability of theintermediate state in the presence of PEGs relative to the native,changes in absorbance at 292 nm, fluorescence at 337 nm, and CD at223 nm were measured, as a function of GdnHCl. Protein concentra-tion was 15 lM for CD and fluorescence, respectively, and 30 lM forabsorbance measurements.

Precipitation reaction. The interaction of con A (0.5 mg/ml) withpolysaccharide or glycoprotein (1.5 mg/ml) was studied in 10 mM TM(Tris–HCl metal ion [Ca2+, Mn2+]) buffer (pH 7) by turbidity methodat 350 nm on spectrophotometer model Hitachi U 1500 and at 410 nmShimadzu RF 540 spectrofluorometer [28]. For each sample properblanks of native, polysaccharide or glycoprotein were taken intoaccount.

Inhibition reaction. The inhibition of precipitation was measuredby incubating con A (0.5 mg/ml) with polysaccharide or glycoprotein(1.5 mg/ml) in 10 mM TM buffer, pH 7, in the presence of increasingconcentrations of PEG 200 and PEG 400 separately and subsequentlydetermining the percent decrease in absorbance at 350 and 410 nm[29].

Results

Effect of PEGs on the structural properties of con A as

studied by different optical spectroscopic methods

CD measurements

Fig. 1A shows the far-UV CD spectra in the 250–200 nm range of con A in the presence of different con-centrations (% v/v) of PEG 200 and 400. The absorptionof PEG alone was taken into account and here we reportthe subtracted spectra. As can be seen from the figurethe native con A exhibits a trough (negative peak) at223 nm (curve 4). The position of the trough is a charac-teristic feature of the b-conformation in lectins. At30%(v/v) PEG 200 (curve 5) the position of trough re-mains at 223 nm with a slight increase in negativeMRE while at 30%(v/v) PEG 400 (curve 6), a blue shift

Fig. 1. (A) Far-UV CD spectra of concanavalin A as a function ofPEGs in 60 mM sodium phosphate buffer, pH 7.0. Curve 1 represents6 MGdnHCl-denatured con A, curve 2 represents con A with 70%(v/v)PEG 400, curve 3 represents con A with 70%(v/v) PEG 200, curve 4represents con A at pH 7 alone, curve 5 represents con A with 30%(v/v)PEG 200, and curve 6 represents con A with 30%(v/v) PEG 400. Theprotein concentration was 15 lm and the pathlength was 0.1 cm. (B)Effect of increasing concentrations of %(v/v) polyols, i.e., PEG 400 (�)and PEG 200 (m) on con A at pH 7 as followed by ellipticitymeasurements at 223 nm.

Fig. 2. Near-UV CD spectra of con A (curve 1), at 30%(v/v) PEG 400(curve 2), 30%(v/v) PEG 200 (curve 3), at 70%(v/v) PEG 200 (curve 4),70%(v/v) PEG 400 (curve 5), and 6 M GdnHCl-denatured con A(curve 6) in 60 mM sodium phosphate buffer, pH 7.0. The proteinconcentration was 30 lm and the pathlength was 1 cm.

Table 1Structural propensities of intermediate states of concanavalin A asstudied by far-UV CD

State Conditions MRE223nm

(deg. cm2/dmol)Cm

Native pH 7.0, 25 �C �5100 ndPEG 200 pH 7.0, 25 �C, 30%(v/v) �5610PEG 200 pH 7.0, 25 �C, 70%(v/v) �4602 42%(v/v)PEG 400 pH 7.0, 25 �C, 30%(v/v) �7200PEG 400 pH 7.0, 25 �C, 70%(v/v) �1101 30%(v/v)

1286 A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294

of 4 nm was obtained with further increase in negativeMRE, indicating the existence of a compact state. At70%(v/v) PEG 200 (curve 3) a red shift of 2 nm with de-crease in negative MRE was obtained. PEGs induce anintermediate compact state appearing during the struc-tural transition from the native to an unfolded state.Such types of transition have also been reported earlierin a number of proteins [30,31]. The effect of PEGs, i.e.,200 and 400, on con A was studied by measuring theMRE at 223 nm (fixed wavelength analysis, Fig. 1B).Up to 30%(v/v) PEG there was more induction of sec-ondary structure while at 70%(v/v) there was a signifi-cant loss of secondary structure at this wavelength. An

increase in MRE at 30%(v/v) PEGs can be suggestiveof an increase in secondary structural content, as polyolsinduce a higher secondary structure in native state up toa certain concentration. As the concentration of PEGsincreases, the polarity of the solvent decreases, creatinga hydrophobic environment, which results in disruptionof secondary structure. As PEG is hydrophobic in nat-ure, it may interact favorably with the hydrophobic sidechains exposed upon unfolding. So, most probably thePEGs perturb the structure of the protein surface,partially modifying the layer of water and the microen-vironment of the aromatic residues which is in agree-ment with the observed alterations in CD spectrum(see Table 1).

The CD spectrum in the near-UV region was used toprobe the asymmetry of the proteins aromatic aminoacid environment. Fig. 2 shows the near-UV CD spectrain the 300–250 nm range of native preparation (curve 1),at 30%(v/v) PEG 400 (curve 2), at 30%(v/v) PEG 200(curve 3), at 70%(v/v) PEG 200 (curve 4), at 70%(v/v)

Fig. 3. (A) Tryptophan fluorescence emission spectra of con A in60 mM sodium phosphate buffer, pH 7.0 (curve 1); curve 2 representscon A at 30%(v/v) PEG 200, curve 3 represents con A at 30%(v/v)PEG 400, curve 4 represents con A at 70%(v/v) PEG 200, curve 5represents con A at 70%(v/v) PEG 400, and curve 6 represents 6 MGdnHCl-denatured con A. The protein concentration was 15 lm andthe pathlength was 1 cm. (B) Effect of increasing concentrations ofpolyols %(v/v), i.e., PEG 400 (�) and PEG 200 (m) on con A at pH 7as followed by fluorescence intensity measurements at 337 nm.

A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294 1287

PEG 400 (curve 5), and in the presence of 6 M GdnHCl(curve 6). The two prominent positive peaks at 292 and283 nm characterize the near-UV CD spectrum of nativecon A. As can be seen from figure, at 30%(v/v) PEG 400and 200 there is loss in MRE values with broadenedpeaks at 292 and 283 nm. Near-UV CD spectra showthat the spectra at 30%(v/v) PEGs resemble more thatof native protein as compared to those at 70%(v/v)PEGs where significant structure is lost or is changedto the completely unfolded state in 6 M GdnHCl. At70%(v/v) PEGs loss of tertiary structure was more inPEG 400 than 200 (curves 5 and 4). From our CD datawe conclude that at 30%(v/v) PEGs there was retentionof native secondary as well as partial tertiary structure.These results show that con A at 30%(v/v) PEGs existsas a compact state and is stabilized by inter-chaininteractions.

Intrinsic tryptophan fluorescence

The intrinsic fluorescence maximum (kmax) is anexcellent parameter to monitor the polarity of trypto-phan environment in the protein and is sensitive to theprotein conformation [26]. Contribution of PEG aloneto the emission spectra was taken into account and wereport here the subtracted fluorescence spectra. Fig.3A depicts the tryptophan fluorescence emission spectraof con A in the absence and presence of various concen-trations of PEGs. The emission maximum of native pro-tein was obtained at 337 nm (curve 1) and a red shift of9 nm was observed in the completely unfolded state in6 M Gdn-HC1 as all four tryptophans are exposed tothe solvent (curve 6). At 30%(v/v) PEG 200 (curve 2)and PEG 400 (curve 3) there was 11% and 13% increasein fluorescence intensity with 3 nm red shift relative tonative preparation. On increasing the concentration ofPEG 200 and 400 up to 70%(v/v) (curves 4 and 5), therewas a further increase in fluorescence intensity, suggest-ing exposure of tryptophan residues from non-polar topolar environment. Relative to PEG 200, the increasein fluorescence intensity was more in PEG 400. This ispossibly due to the presence of more hydroxyl groupsin PEG 400 than PEG 200. Fig. 3B depicts the fluores-cence intensity at 337 nm as a function of %(v/v) PEGs.On increasing the concentration of PEGs, there wasenhancement in fluorescence intensity. However, at30%(v/v) PEG, there is only a slight red shift in thewavelength emission maxima (kmax) as compared to asignificantly red-shifted spectrum of the completely un-folded state in 6 M GdnHCl, indicating that the trypto-phan environment resembles the native form of con A[32].

ANS fluorescence

Fig. 4A shows ANS fluorescence emission spectra ofcon A at pH 7 in the presence of PEG 200 and 400. Ascan be seen from the figure, maximum increase in ANS

emission intensity with a blue shift in emission maxi-mum in con A was at 30%(v/v) PEG 400 (curve 5)and 200 (curve 4) followed by 70%(v/v) PEG 200 (curve3) and 400 (curve 2). Native con A in the absence ofPEGs showed negligible ANS binding (curve 1). SinceANS is known to bind to hydrophobic patches on pro-tein, binding of ANS to hydrophobic regions of proteinhas been widely used to study the folding intermediates

Table 2Elution volume of intermediate states of concanavalin A as studied bysize-exclusion chromatography

Properties Conditions Elution volume(Ve) (ml)

Native pH 7.0, 25 �C 145PEG 200 pH 7.0, 25 �C, 30%(v/v) 130PEG 200 pH 7.0, 25 �C, 70%(v/v) 112PEG 400 pH 7.0, 25 �C, 30%(v/v) 118PEG 400 pH 7.0, 25 �C, 70%(v/v) 110GdnHCl pH 7.0, 25 �C, 6 M 109

Fig. 4. (A) ANS fluorescence emission spectra: native con A (curve 1);con A in the presence of 70%(v/v) PEG 400 (curve 2), 70%(v/v) PEG200 (curve 3), 30%(v/v) PEG 200 (curve 4), and 30%(v/v) PEG 400(curve 5) in 60 mM sodium phosphate buffer, pH 7. The proteinconcentration was 15 lm and the pathlength was 1 cm. (B) Effect ofincreasing concentrations of polyols %(v/v), i.e., PEG 400 (�) and PEG200 (m) on con A at pH 7 as followed by ANS fluorescence at 480 nm.

1288 A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294

formed during protein folding. As molecules of ANSbind to hydrophobic surfaces on the protein with greateraffinity, it can be concluded that at 30%(v/v) PEGs acompact intermediate with remarkable increase in thenumber of solvent accessible non-polar clusters of conA exists [33,34]. The emission maximum at 30%(v/v)PEGs blue shifts to 482 nm relative to native (500 nm),further suggesting that ANS binds to protein. Fig. 4Bdepicts the ANS fluorescence intensity at 480 nm as a

function of %(v/v) PEG. The notch in Fig. 4B at30%(v/v) PEGs indicates an intermediate appearing dur-ing the structural transition from native to polyol(PEG)-induced state. There was further decrease inANS fluorescence at 70%(v/v) PEGs because of disrup-tion of hydrophobic patches.

Size-exclusion chromatography

Since con A is a non-covalently associated tetramer atpH 7, one possibility is that the intermediate seen in spec-troscopic studies in the presence of PEGs must be com-pact (resembling native state) as compared to 6 MGdnHCl. To clarify this issue, size-exclusion chromatog-raphy was performed. The gel filtration analysis of nativecon A at pH 7 on seralose 6B column confirms the tetra-meric nature of the protein (the elution volume corre-sponds to a molecular mass of 104 kDa). Gel filtrationanalysis in the presence of 30%(v/v) PEGs showed a de-crease in elution volume corresponding to less compactcon A as compared to native (see Table 2). This indicatesthat the dimensions of the molecule have slightly in-creased in the presence of polyols. The elution volumein the presence of 70%(v/v) approaches 6 M GdnHClwhich further confirms that the state at 70%(v/v) resem-bles the denatured state. At 30%(v/v) PEGs an interme-diate value of elution volume is obtained, suggesting thatthe protein is relatively compact at this PEG concentra-tion and resembles a native state. The state obtained inthe presence of PEG 200 was found to be more compact(less increase in Ve) than that in PEG 400. Thus, the re-sults indicate a slight increase in hydrodynamic dimen-sions of lectin in the presence of PEGs, which could bedue to loosening of tertiary contacts. However, native-like tertiary structure is retained as compared to com-plete loss of structure at 6 M GdnHCl.

GdnHCl stability studies

The fluorescence intensity at 337 nm as a function ofGdnHCl is depicted in Fig. 5A. As can be seen from thefigure, PEG 400 resists the loss of structure in the pres-ence of increasing concentrations of GdnHCl than PEG200 and native. The fraction of protein denatured (fD) ofcon A at pH 7 and in the presence of 30%(v/v) PEGswith increase in GdnHCl denaturation is shown in

Fig. 5. GdnHCl denaturation of con A at pH 7 (j), 30%(v/v) PEG 400 (�), and 30%(v/v) PEG 200 (m): (A) fluorescence intensity at 337 nm versusGdnHCl, (B) data are being plotted between fD versus GdnHCl concentration, fraction of protein denatured by change in fluorescence emissionmaximum as a function of GdnHCl. (C) GdnHCl induced transition as monitored by MRE measurements at 223 nm (D) De versus GdnHCl: changein value of e292 (De) as a function of GdnHCl.

A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294 1289

Fig. 5B. fD of con A at pH 7 and in the presence of 30%PEGs was calculated taking con A at pH 7, in the pres-ence of 30%(v/v) PEGs at 0 M GdnHCl as native point,respectively. At pH 7, no intermediates were present indetectable amount, thus it showed a single-step, two-state cooperative transition. In the presence of PEGstoo, a single-step, two-state weakly cooperative transi-

tion was obtained. The weakly cooperative unfoldingof PEG–con A was indicative of its molten-globule-likenature. There are similar examples in many globularproteins including equine, canine, cytochrome c, andSNase [35–37]. Fraction denaturation was least inPEG 400 followed by PEG 200 and native. This showsthe GdnHCl denaturation was less in the presence of

Fig. 6. (A) Effect of PEG 400 (�) and 200 (m) on con A–dextranquantitative precipitin reaction at 350 nm on spectrophotometer inTM buffer, pH 7. (Inset) Effect of PEG 400 (�) and 200 (m) on con A–dextran quantitative precipitin reaction at 410 nm on spectrofluorom-eter in TM buffer, pH 7. (B) Effect of PEG 400 (�) and 200 (m) on conA–glycogen quantitative precipitin reaction at 350 nm on spectropho-tometer in TM buffer, pH 7. (Inset) Effect of PEG 400 (�) and 200 (j)on con A–glycogen quantitative precipitin 410 nm on spectrofluorom-eter in TM buffer, pH 7.

1290 A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294

PEG 400 than 200. In native con A, a sigmoidal curve isobtained with large variation range in fD while in PEG–con A, the narrow range is obtained. Fig. 5C shows thechange in MRE values at 223 nm. The transition of na-tive was found to be cooperative while of PEG 200 and400 was weakly cooperative. The transitions were con-sistent with the fluorescence results. Fig. 5D shows thechanges in extinction coefficient as a function ofGdnHCl at 292 nm by absorption spectroscopy.30%(v/v) PEG 200 and PEG 400 exhibit a single-step,two-state transition similar to native. From GdnHCldata, we conclude that during unfolding transition ofnative, 30%(v/v) PEG 200 and PEG 400–con A, nodetectable intermediates were obtained. The analysis ofdenaturation profiles of native con A and in the presenceof 30%(v/v) PEGs shows their unfolding was coopera-tive and weakly cooperative, respectively. Further, inthe presence of PEG 400 the loss of structure is less ascompared to PEG 200 as a greater number of hydroxylgroups are present in PEG 400, which resists theGdnHCl denaturation.

Thus, it can be concluded that at 30%(v/v) PEGs, conA exists as molten globule state with the retention of acompact secondary and some tertiary structure, en-hanced hydrophobic surface area, and weakly coopera-tive transition curve of GdnHCl denaturation.

Interaction of con A with polysaccharides (dextran

(a1–6), glycogen (a1–4)), and glycoproteins (ovalbumin

(1-2-linked a-D-mannopyranosyl), ovomucoid (Glc NAcb(1–4) (1–2)-Man a(1–3)))

The precipitin reaction curve of dextran, ovalbumin,ovomucoid, and glycogen was studied by turbiditymethod at 350 and 410 nm using spectrophotometerand spectrofluorometer, respectively. In order to ruleout the possibility of aggregation, proper blanks ofcon A and glycoproteins or polysaccharide were takeninto consideration. The turbidity developed due to inter-action of con A with the above-mentioned ligands wasmaximum in glycogen followed by ovalbumin, ovomu-coid, and dextran (data not shown).

Effect of PEGs on the functional properties of con A

Interaction of con A with polysaccharide (dextran andglycogen) in the presence of PEGs

Effect of PEG 200 and 400 on the precipitin reactionof con A with polysaccharides was investigated by tur-bidity measurements at 350 nm by spectrophotometer(Fig. 6A) and at 410 nm by spectrofluorometer (Fig.6A, inset). The concentration of %(v/v) PEGs corre-sponding to 50% decrease in absorbance at 350 nmyields C50 (per liter). C50 is the concentration of polyolrequired to achieve 50% inhibition of the specific precip-itin reaction. The C50 for PEG 200 (34%) was greater

Fig. 7. (A) Effect of PEG 400 (�) and 200 (m) on con A–ovalbuminquantitative precipitin reaction at 350 nm on spectrophotometer inTM buffer, pH 7. (Inset) Effect of PEG 400 (�) and 200 (m) on con A–ovalbumin quantitative precipitin reaction at 410 nm on spectrofluo-rometer in TM buffer, pH 7. (B) Effect of PEG 400 (�) and 200 (m) oncon A–ovomucoid quantitative precipitin reaction at 350 nm onspectrophotometer in TM buffer, pH 7. (Inset) Effect of PEG 400(�) and 200 (m) on con A–ovomucoid quantitative precipitin reactionat 410 nm on spectrofluorometer in TM buffer, pH 7.

A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294 1291

than that for PEG 400 (24%). Approximately 60% and82% activity was lost in dextran–con A precipitationreaction in the presence of PEG 200 and 400, respec-tively. Fig. 6A (inset) shows the same trend at 410 nmon spectrofluorometer as for con A and dextran in thepresence of PEGs. The activity loss at 30%(v/v) PEG400 and 200 in dextran–con A was 53% and 37%,respectively.

In Fig. 6B, the decrease in turbidity was observed inthe presence of PEG 400 and 200 on con A–glycogen pre-cipitation as monitored by turbidity method at 350 nm.PEG 400 inhibited the precipitation more as comparedto PEG 200. C50 for PEG 400 and 200 glycogen–con Aprecipitation was 18% and 16%(v/v), respectively. Theactivity loss of con A for glycogen was 51% and 60% inthe presence of PEG 400 and 200, respectively. Fig. 6B(inset) shows the inhibition of precipitation of glyco-gen–con A in the presence of PEGs by spectrofluorome-ter at 410 nm. Similar pattern was obtained as seen inFig. 6B. The maximum activity loss in the presence ofPEG 400 and 200 was 71% and 60%, respectively.

Interaction of con A with glycoproteins (ovalbumin and

ovomucoid)

Fig. 7A shows con A–ovalbumin precipitin reactionin the presence of PEG 200 and 400 at 350 nm. Theconcentration of PEG 400 and 200 at which 50% inhi-bition occurred was 14% and 12%(v/v), respectively.The maximum percentage activity loss of con A withovalbumin was 95% in PEG 400 and 91% in PEG200. The activity loss at 30%(v/v) PEG 200 andPEG 400 was 75% and 77%, respectively. Experimentson the inhibition of con A–ovalbumin precipitin reac-tion by PEG 200 and 400 were also performed by tur-bidity methods at 410 nm on spectrofluorometer (seeFig. 7A, inset).

Fig. 7B shows con A–ovomucoid precipitation reac-tion in the presence of PEG 200 and 400. PEG 200 orPEG 400 maximally inhibited the con A–ovalbumin pre-cipitin reaction as compared to the above-mentioned li-gands. 93% decrease in the presence of maximumconcentration of PEG was obtained in the activity ofcon A towards ovomucoid. The loss in the activity ofcon A with ovomucoid was 75% and 77% in the presenceof 30%(v/v) PEG 400 and PEG 200, respectively. C50 va-lue obtained for con A and ovomucoid was 10%. Insetof Fig. 7B shows the same trend at 410 nm by spectroflu-orometer as observed for Fig. 7B.

Discussion

The effect of PEGs on con A at pH 7 was carried outby spectroscopic techniques (fluorescence, ANS binding,and CD), size-exclusion chromatography, and GdnHCl

denaturation as well as carbohydrate binding specificity.PEG 200 and 400, being smaller molecules, penetrate theprotein structure and form contacts with the surface ofprotein. While the larger species of PEGs cannot pene-trate the protein structure, a shell that is impenetrableto co-solvent is formed around the protein molecule.

Table 3Comparison of the structural and functional propensities of interme-diate states of concanavalin A in the presence of PEGs

Properties Native PEG 200 at30%(v/v)

PEG 400 at30%(v/v)

MRE �5100 �5610 �7200ANS fluorescence + ++ +++Activity 1.4a 0.9a 0.12a

1.1b 0.2b 0.18b

0.6c 0.35c 0.25c

1.5d 0.45d 0.3d

% inhibition 0% 51%a 60%a

0% 79%b 80%b

0% 37%c 53%c

0% 75%d 77%d

C50 — 18%a 16%a

— 14%b 12%b

— 34%c 24%c

— 10%d 10%d

a denotes glycogen, b denotes ovalbumin, c denotes dextran, and ddenotes ovomucoid. C50, Concentration of organic solvent required tobring about 50% decrease in absorbance. Activity of native is taken as100%. Activity, the concentration at which maximum turbidity wasobtained with the lectin and ligand. +, least; ++, less; and +++, moreANS binding.

1292 A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294

Higher molecular weight PEGs have larger excludedvolume, leading to a larger protein with preferentialhydration. As a result, due to preferential exclusionhigher PEGs lead to stabilization instead of denatur-ation. The following spectral information was obtained.(1) From far-UV CD spectrum, there was retention ofsecondary structure at 30%(v/v) PEGs, which was loston increasing the concentration further. (2) Fromnear-UV CD band at 292 and 283, assigned to trypto-phan and tyrosine, respectively, some tryptophan andtyrosine side chains were inferred to be constrained byinterhelix interactions. (3) From fluorescence studieswe observed that 30%(v/v) PEGs alter the tryptophanenvironment of native con A. (4) ANS binding was max-imum at 30%(v/v) PEG 400, followed by 30%(v/v) PEG200. (5) From size-exclusion chromatography, an inter-mediate state was obtained which was compact as com-pared to 6 M GdnHCl. (6) The GdnHCl unfolding of30%(v/v) PEG-induced state showed single-step weaklycooperative transition. In all the results obtained, the ef-fect of PEG 400 is more pronounced as compared toPEG 200; this is because PEG 400 has a larger molecularweight and greater number of OH groups. The effect ofpolyols increases with increasing concentration and thenumber of OH groups [38]. Destabilization of humanserum albumin and bovine serum albumin by PEGshas also been reported earlier [39,40]. Thus, intermedi-ates obtained at 30%(v/v) PEGs retain secondary struc-ture with enhanced hydrophobic surface area, asobserved by their strong ability to bind ANS. GdnHCldenaturation studies showed that the transition was asingle-step, two-state process. These species may hencebe described as compact molten-globule-like intermedi-ates, as similar properties have been observed for themolten globule states of a number of proteins. ‘‘Moltenglobules’’ (MGs) are partially structured protein foldingintermediates that adopt a native-like overall backbonetopology in the absence of extensive detectable tertiaryinteractions. It is important to determine the extent ofspecific tertiary structure present in �molten globules�and to understand the role of specific side-chain packingin stabilizing and specifying the MG structure. More-over, the A-state of equine ferricytochrome c has na-tive-like compactness, native-like helix content, nativehelix–helix interactions, and one native heme ligand[41,42]. Tertiary A-state has been reported for a-lactal-bumin [43], RNase H [3], myoglobin [14], and ubiquitin[44]. For many proteins, a native tertiary structure in thepresence of organic solvents has been reported [45].

Yet, unusually, con A intermediate, despite the ex-panded tertiary structure, retains its carbohydrate bind-ing activity to a considerable degree. The precipitinreaction involving con A and specific polysaccharideor glycoprotein resembles an antigen–antibody reaction.Accordingly con A–ligand reaction leading to specificprecipitation reaction can be represented by

con A+L$ complex (Soluble)$ complex (Precipitate)

ð1Þwhere L is a multivalent ligand. It is the secondaryaggregation of the soluble complex, i.e., the secondstep that was monitored by turbidity method in thisstudy. Accordingly, of the two organic solvents usedthe concentration required to achieve maximal sharpinhibition was highest for ovomucoid and minimumfor dextran. This may be due to the slight disruptionof tertiary structure of con A in the presence of highconcentration of PEGs, which leads to a decrease inprecipitation as obtained in native-ligand combina-tions. Since hydroxyl groups are more in PEG 400 thanPEG 200, more inhibition was obtained in all the com-binations of PEG 400 than PEG 200. A comparison ofstructural and functional properties of intermediatestate of con A in the presence of PEGs is summarizedin Table 3. Our studies extend to the characterizationof compact molten globule states of con A in the pres-ence of PEGs, i.e., having high hydrophobic residuesand retention of secondary and tertiary structures.From these studies, it appears that lectin intermediatesin the presence of polyols may have all the necessaryfeatures compatible with carbohydrate binding. Thus,it can be concluded that molten globule is the mini-mum structural requirement of lectins in general, forthe carbohydrate binding specificity. PEG-molten glob-ule complex formation is analogous to the observedinteraction between chaperonins and a molten globuleintermediate [46].

A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294 1293

Acknowledgments

The authors are highly thankful for the facilities ob-tained at AMU Aligarh. Authors are also thankful toDST-FIST for the financial support. A.N. is the recipi-ent of CSIR-SRF. Authors are thankful to CSIR 38/0978/99/EMR-II and UGC F.3-12/2002/SR II for thefinancial grant in the form of project.

References

[1] R. Jacnicke, Protein folding: local structures, domains, subunits,and assemblies, Biochemistry 30 (1991) 3147–3161.

[2] F.U. Hartl, Molecular chaperones in cellular protein folding,Nature 381 (1996) 571–579.

[3] T.M. Raschke, S. Marqusee, The kinetic folding intermediate ofribonuclease H resembles the acid molten globule and partiallyunfolded molecules detected under native conditions, Nat. Struct.Biol. 4 (1997) 298–304.

[4] K. Kuwajima, The molten globule state as a clue for understand-ing the folding and cooperativity of globular-protein structure,Proteins: Struct. Funct. Genet. 6 (1989) 87–103.

[5] D. Barrick, F.M. Hughson, R.L. Baldwin, Molecular mechanismsof acid denaturation. The role of histidine residues in the partialunfolding of apomyoglobin, J. Mol. Biol. 237 (1994) 588–601.

[6] S.K. Haq, S. Rasheedi, R.H. Khan, Characterization of a partiallyfolded intermediate of stem bromelain at low pH, Eur. J.Biochem. 269 (2002) 47–52.

[7] F. Naseem, R.H. Khan, S.K. Haq, A. Naeem, Characterizationof molten globule state of fetuin at low pH, Biochim. Biophys.Acta 1649 (2003) 164–170.

[8] A. Naeem, K.A. Khan, R.H. Khan, Characterization of apartially folded intermediate of papain induced by fluorinatedalcohols at low pH, Arch. Biochem. Biophys. 432 (1) (2004) 79–87.

[9] A. Matousschek, L. Serrano, E.M. Meiering, M. Bycroft, A.R.Ferscht, The folding of an enzyme IV. Structure of an interme-diate in the refolding of barnase analyzed by a protein engineeringprocedure, J. Mol. Biol. 224 (1992) 837–845.

[10] O.B. Ptitsyn, Protein folding: hypotheses and experiments, J.Protein Chem. 6 (1987) 273–293.

[11] Y. Goto, L.J. Calciano, A.L. Fink, Acid-induced folding ofproteins, Proc. Natl. Acad. Sci. USA 87 (1990) 573–577.

[12] K. Wuthrich, NMR, alcohols, protein solvation and proteindenaturation, EXS 71 (1994) 261–268.

[13] M.S. Kay, R.L. Baldwin, Packing interactions in the apomyoglo-bin folding intermediate, Nat. Struct. Biol. 3 (1996) 439–445.

[14] A. Naeem, R.H. Khan, Characterization of molten globule stateof cytochrome c at alkaline, native and acidic pH induced bybutanol and SDS, Int. J. Biochem. Cell Biol. 36 (11) (2004) 2281–2292.

[15] F. Khan, R.H. Khan, S. Muzammil, Alcohol-induced versusanion-induced states of a-chymotrypsinogen A at low pH,Biochim. Biophys. Acta 1481 (2000) 229–236.

[16] P. Gupta, R.H. Khan, M. Saleemuddin, Trifluoroethanol-induced‘‘molten globule’’ state in stem bromelain, Arch. Biochem.Biophys. 413 (2) (2003) 199–206.

[17] A. Naeem, M. Akram, R.H. Khan, Conformational states oftrifluoroacetic acid treated cytochrome c in the presence of saltsand alcohols, J. Protein Chem. 23 (2004) 185–195.

[18] N.M. Young, R.P. Oomen, Analysis of sequence variation amonglectins. A ring of hypervariable residues forms the perimeter of thecarbohydrate-binding site, J. Mol. Biol. 228 (1992) 924–934.

[19] N. Sharon, H. Lis, Lectin-proteins with a sweet tooth: function incell recognition, Essays Biochem. 30 (1995) 59–75.

[20] N. Sharon, Lectin-carbohydrate complexes of plants and animals:an atomic view, Trends Biol. Sci. 18 (1993) 221–226.

[21] S. Ahmad, R.H. Khan, A. Ahmad, Physicochemical character-ization of Cajanus cajan lectin: effect of pH and metal ions onlectin carbohydrate interaction, Biochem. Biophys. Acta 1427(1999) 378–384.

[22] L.A. Lasky, Selectins: interpreters of cell-specific carbohydrateinformation during inflammation, Science 258 (1992) 964–969.

[23] A. Naeem, R.H. Khan, H. Vikram, M. Akif, Purification ofCajanus cajan root lectin and its interaction with Rhizobiallipopolysaccharide, Arch. Biochem. Biophys. 396 (2001) 99–105.

[24] P. Roche, F. Debelle, F. Maillet, P.P. Lerouge, C. Faueher, G.Truechet, J. Denarie, J.C. Prome, Molecular basis of symbiotichost specificity in Rhizobium meliloti: nod H and nod PQ genesencode the sulfation of lipo-oligosaccharide signals, Cell 67 (1991)1131–1143.

[25] Y.H. Chen, J.T. Yang, H.M. Martinez, Determination of thesecondary structure of proteins by circular dichroism and opticalrotatory dispersion, Biochemistry 11 (1972) 4120–4131.

[26] L. Stryer, Fluorescence spectroscopy of proteins, Science 162(1968) 526–540.

[27] D. Matulis, C.G. Baumann, U.A. Bloomfield, R.E. Lovrien, 1-Anilino-8-naphthalene sulfonate as a protein conformationaltightening agents, Biopolymers 49 (6) (1999) 451–458.

[28] A. Waseem, A. Salahuddin, Relevance of hydrophobic interac-tions in specific binding of multivalent ligands to concanavalin A-effect of organic solvents, Indian J. Biochem. Biophys. 31 (1983)253–258.

[29] A. Salahuddin, R. Begum, B.K. Averill, Salt-induced inhibition ofthe precipitin reaction of concanavalin A with polysaccharidesand glycoproteins, Biochem. J. 220 (1984) 639–642.

[30] R. Santucci, F. Polizio, A. Desideri, Formation of a moltenglobule-like state of cytochrome c induced by high concentrationof glycerol, Biochemie 81 (1999) 745–751.

[31] Y. Kumar, S. Tayyab, S. Muzammil, Molten-globule like partiallyfolded states of HSA induced by fluoro and alkyl alcohols at lowpH, Arch. Biochem. Biophys. 426 (2004) 3–10.

[32] D. Dryden, M.P. Weir, Evidence for an acid-induced molten-globule state in interleukin-2; a fluorescence and circular dichro-ism study, Biochim. Biophys. Acta 1078 (1991) 94–100.

[33] S.A. Pawar, V.V. Deshpande, Characterization of acid-inducedunfolding intermediates of glucose/xylose isomerase, Eur. J.Biochem. 267 (2000) 6331–6338.

[34] D. Matulis, R. Lovrien, 1-Anilino-8-naphthalene sulfonateanion–protein binding depends primarily on ion pair formation,Biophys. J. 74 (1998) 422–429.

[35] J.H. Carra, A.A. Elizabeth, P.L. Privalov, Three-state thermody-namic analysis of the denaturation of staphylococcal nucleasemutants, Biochemistry 33 (1994) 10842–10850.

[36] W. Colon, H. Roder, Kinetic intermediates in the formation ofthe cytochrome-c molten globule, Nat. Struct. Biol. 12 (1996)1019–1025.

[37] J.L. Marmorino, M. Lehti, G.J. Pielak, Native tertiary structurein an A-state from resonance Raman spectroscopy, J. Mol. Biol.275 (1998) 379–388.

[38] T. Kamiyama, Y. Sadahide, Y. Nogusa, K. Gekko, Polyol-induced molten globule state of cytochrome c: an evidence forstabilization by hydrophobic interaction, Biochem. Biophys. Acta1434 (1999) 44–57.

[39] B. Farruggia, B. Nerli, H. Di Nuci, R. Rigatusso, G. Pico,Thermal features of the bovine serum albumin unfolding bypolyethylene glycols, Int. J. Biol. Macromol. 26 (1999) 23–33.

[40] B. Farruggia, G. Garcia, C. D�Angelo, G. Pico, Destabilization ofhuman serum albumin by polyethylene glycols studied by

1294 A. Naeem et al. / Biochemical and Biophysical Research Communications 331 (2005) 1284–1294

thermodynamical equilibrium and kinetic approaches, Int. J. Biol.Macromol. 20 (1997) 43–51.

[41] D. Hamada, Y. Kuroda, M. Kataoka, S. Aimoto, T. Yoshimura,Y. Goto, Role of heme axial ligands in the conformationalstability of the native and molten globule states of horsecytochrome c, J. Mol. Biol. 256 (1996) 172–186.

[42] M. Kataoka, Y. Hagihara, K. Mihara, Y. Goto, Molten globuleof cytochrome c studied by small X-ray scattering, J. Mol. Biol.229 (1993) 591–596.

[43] L.C. Wu, P.S. Kim, A specific hydrophobic core in thea-lactalbumin molten globule, J. Mol. Biol. 280 (1998) 175–182.

[44] S. Khorasanizadeh, I.D. Peters, H. Roder, Evidence for a three-state model of protein folding from the kinetic analysis ofubiquitin variants with altered core residues, Nat. Struct. Biol. 3(1996) 193–205.

[45] R.M. Taylor, S.D. Zakharov, J.B. Heymann, M.E. Grivin, W.A.Cramer, Folded state of the integral membrane colicin E1immunity protein in solvents of mixed polarity, Biochemistry 39(2000) 12131–12139.

[46] J.L. Cleland, T.W. Randolph, Mechanism of polyethylene glycolinteraction with the molten globule folding intermediate of bovinecarbonic anhydrase B, J. Biol. Chem. 267 (1992) 3147–3153.