Artocarpus hirsuta lectin Differential modes of chemical and thermal denaturation Sushama M. Gaikwad, Madhura M. Gurjar* and M. Islam Khan Division of Biochemical Sciences, National Chemical Laboratory, Pune, India Unfolding, inactivation and dissociation o f the lectin from Artocarpus hirsuta seeds were studied by chemical (guanidine hydrochloride, GdnHCl) and thermal denaturation. Con- formational transitions were monitored by intrinsic fluor- escence and circular dichroism. The gradual red shift in t he emission maxima of the native p rotein from 33 5 to 356 nm, change in the e llipticity at 218 nm and simultaneous decrease in the sugar binding activity were observed with increasing concentration o f G dnHCl in the pH range between 4.0 and 9.0. The unfolding and inactivation by GdnHCl were par- tially reversible. Gel filtration of the lectin in presence o f 1–6 M GdnHCl showed that the protein dissociates rever- sibly into partially unfolded dimer and then irreversibly into unfolded inactive monomer. Thermal denaturation was irreversible. The l ectin loses activity rapidly above 45 °C. The exposure of h ydrophobic patches, distorted secondary structure and formation o f insoluble aggregates of the thermally inactivated protein probably leads to the irre- versible d enaturation. Keywords: Artocarpus lectin; denaturation; intrinsic fluores- cence; unfolding; aggregation. Proteins that bind carbohydrates specifically and reversibly are termed as lectins. They occur ubiquitously in nature and have diverse role in plants, animals and microbes. T he recognition of c arbohydrate m oieties b y l ectins h as import- ant applications in a number of biological processes such a s cell–cell i nteractions, signal transduction, and cell growth and differentiation [1]. a-Galac toside specific lectin present in the s eeds of Artocarpus hirsuta [2–4], is a homotetrameric protein with molecular mass of 60 000 Da and high specificity f or methyl a- D -galactopyranoside (Me a-gal). The folding pathways of oligomeric proteins involve both intramolecular and intermolecular interactions. The dena- turation of pea and peanut lectins, both oligomeric proteins, has b een studied in great detail [ 5–7]. In t his paper we show the progressive unfolding and inactivation of the lectin in presence of GdnHCl and heat, and refolding and reactiva- tion under renaturing conditions. MATERIALS AND METHODS Materials GdnHCl was a product o f Sigma Chemical Co. A ll other chemicals were of t he highest purity available. The l ectin from A. hirsuta was purified as described previously [2]. Stocks of 7 M GdnHCl were freshly prepared in appropriate buffers and filtered through 0.45-lm filter. Buffers used were acetate f or pH 4.0, citrate/phosphate for pH 5.0 and 6.0, phosphate for p H 7.0 and Tris/HCl for pH 8.0 and 9.0 (all 100 m M ). Fluorescence studies Protein samples (1.5 l M ) were equilibrated for 4 h at the desired denaturant concentration at 30 °C in the pH range of 4.0–9.0. Unfolding as a function of GdnHCl concentr a- tion was monitored by intrinsic tryptophan fluorescence emission in a 1-cm quartz cell in the 300–400 nm region, when excited at 280 nm, in a Perkin-Elmer LS 50B spectrofluorimeter with attached circulating water bath. Excitation and emission band passes of 5 n m were u sed. Activity of the sample was measured at the same time. Unfolding as a function of temperature w as carried out by incubating the protein samples (1.5 l M ) in duplicates at the temperature from 30–70 °C for 15 min. One of the duplicates was u sed to r ecord the spectra and activity a t the respective temperature. The other sample was brought to 35 °C, centrifuged to remove any particulate matter, spectra were recorded and activity was estimated. Circular dichroism studies Far UV CD (210–250 nm) spectra of the protein samples (15 l M ) t reated with different concentrations GdnHCl in respective buffers of pH 4.0–9.0 and incubated for 14 h were recorded in a 1-mm path length cell, on a Jasco J715 spectropolarimeter connected to a circulating water bath. For thermal denaturation studies, the protein sample was incubated at various temperatures for 10 min and the spectra were recorded. The spectra were collected with response time of 4 s, and scan speed of 100 n mÆs )1 .Each data point was an average of three accumulations. Correspondence to S. M. Gaikwad, Division of Biochemical Sciences, National Chemical Laboratory, Pune, 411008, India. Fax: + 9 1 20 5884032, Tel.: + 91 20 5893034, E-mail: gaikwad@ems.ncl.res.in Abbreviations: GdnHCl, guanidine hydrochloride; Me a-gal, methyl a- D -galactopyranoside; ANS, 1-anilino-8-naphthalene sulfonate. *Present address: University of Medic ine and D e ntistry New J ersey, Robert Wood J ohn son Medical School, D ivision of Biochemistry (Research t ower), Piscataway, Ne w Jersey 0 8854, USA. (Received 21 December 2 001, acc epted 14 Ja nuary 2002) Eur. J. Biochem. 269, 1413–1417 (2002) Ó FEBS 2002 Hydrophobic dye binding studies 8-Anilino-1-naphthalene sulfonate (ANS) emission spectra were recorded in the range of 400–500 nm with excitation at 375 n m using slit widths of 5 nm. The changes in the ANS fluorescence induced by the binding to the lectin were followed by r ecording the spectra at constant concentration of protein (1 l M )andANS(50l M ), in different concentra- tions of GdnHCl (1–5 M ). Determination of the lectin activity Sugar binding activity of the samples was measured by the enhancement in the intrinsic flu orescence of the prote in at 335 n m after addition of the nonfluorescent ligand Me a-gal (4 m M ) at saturating c oncentration. Increase in the fl uores- cence of the native lectin after adding Me a-gal was taken as 100% activity [3]. Light scattering studies Rayleigh light scattering experiments w ere carried out with the s pectrofluorimeter t o f ollow p rotein aggregation during GdnHCl and thermal denaturation. Both excitation and emission wavelengths were set at 400 nm and the time dependent change in scattering intensity was followed. Renaturation studies Two-hundred microliters aliquot was removed from the samples treated with different c oncentrations of GdnHCl (3.0–5.0 M )for4hat30°C a t p H 7 .0 and d iluted 1 0 t imes with 100 m M buffer of pH 7.0. After 30 min, the fluores- cence spectra and activity of the original (treated w ith GdnHCl) a s well as diluted samples were recorded. Protein sample without GdnHCl tre ated under identical conditions was taken as control. The renaturation of thermally denatured protein was followed b y c ooling t he heate d samples to 3 5 °C, re moving any p articulate matter by centrifuging, and t hen recording the fluorescence spectra and the activity. Gel filtration studies Lectin samples (10 l M in 100 lL) were incubated for 14 h with GdnHCl (1.0–6.0 M ) a nd injected onto Protein Pak 300SW HPLC column (7.8 · 300 mm) connected to a Waters HPLC syste m preequilibrated a nd eluted with different concentrations of GdnHCl (1.0–6.0 M )in 100 m M buffer of pH 7 .0 at a flow rate of 0.5 mLÆmin )1 . Elution was monitored by absorbance at 280 nm. The standard molecular mass markers run in the presence of buffer were, BSA (66 kDa), o valbumin (45 kDa), carbonic anhydrase (29 kDa) and cytochrome c ( 14.5 k Da). RESULTS AND DISCUSSION Unfolding studies The fluorescence emission spectrum of the native lectin showed a maximum a t 335 nm which characterizes no n- polar environment o f t he tryptoph an residues. On dena- turation of the lectin with increasing concentration of GdnHC l (1–5 M ) a lthough fluorescence intensity at 335 nm does not change much, the fluorescence at 356 n m increases significantly thus changing the emission maxi- mum from 335 to 356 nm (Fig. 1A) indicating that due to unfolding, most of the tryptophans in the protein are getting exposed to th e solvent. The ratio of fluorescence intensities at 3 35 and 356 F(335/356) (Fig. 1B) decreases from 1.35 to 0.82. Similar trend of denaturation with GdnHCl is observed in the pH range of 5.0–8.0, while it is more drastic at pH 4.0 and 9.0. The concentration of GdnHCl required for 50% unfolding of the protein between the pH range 6.0–8.0 is higher (3.2 M )thanat pH 4.0 and pH 9.0 (2.2 M ). The l ectin showed a typical far UV CD spectrum observed for proteins with high proportion of b sheet content, with minimum at 218 nm [2]. The relative percentage of structural elements calculated using CDPro software package for analyzing protein CD spectra was a helix 2%, b sheet 44%, turns 2 3% and random coil 30% for native protein. The CD spectra of the G dnHCl treated protein when analysed using the above programs did not show any significant change in the different structural elements compared to the native protein, while there was visible difference in the respective CD spectra. This was probably due to the incompatibility o f the data with the programmes used. Because GdnHCl was interfering the CD s pectra below 2 10 nm, d ata in the ran ge of 210–250 nm could be collected. The negative ellipticity of the protein at 218 nm increases in 1–2 M GdnHCl and then decreases at h igher c oncentration ( Fig. 1C). The change in the structure at 1–2 M GdnHCl is concomitant with loss of activity and therefore cannot be a stable Fig. 1. GdnHCl-induced unfolding of A. hirsuta lectin at 30 °C. Protein (1.5 l M ) at the required GdnHCl concentration was incubated for 4 h and the fluorescence emission spectra were re co rded between 300 and 400 nm with the excitation wavelength of 280 nm (A), shift in the emission max (B), ratio F(335/356) (C), mean residue ellipticity at 218 nm in far UV region and (D) activity of the GdnHCl treated protein. The s ym bols used for all the figure s a re p H 4.0 (.), pH 5.0 (d), pH 6.0 ( m), pH 7.0 (,), pH 8.0 (s)andpH9.0(n). 1414 S. M. Gaikwad et al. (Eur. J. Biochem. 269) Ó FEBS 2002 conformation. When Rayleigh light scattering studies of the s amples were carried out, the sample incubated at 1 M GdnHCl at 30 °C showed lower light scattering intensity than the native p rotein. At 2.0 M GdnHCl, there was lower light scattering than that with 1.0 M and at s till higher concentrations of GdnHCl, there was n o light scattering at all (data not shown). Thus, the increase in the negative ellipticity at 218 n m at low concentrations of GdnHCl could be due to the solubilization of the aggregates in the protein. There is substantial loss in the secondary structure of t he lectin as indicated b y the decrease in the n egative ellipticity at 218 nm with increasing concentration of GdnHCl. Similar trend was observed for denaturation between p H 5.0–8.0, while the rate of unfolding was faster at pH 4 .0 and 9.0. The inactivation of the lectin was proportional to the concentration of GdnHCl (Fig. 1D). The maximum enhancement in the intrinsic fluorescence of the lectin due to the binding of sugar, Me a-gal, taken as measure of 100% activity of the lectin [3] determined at different pH was diffe rent. T he percentage decrease in the enhancement, i.e. activity w ith increasing concentration o f GdnHCl, however, was equivalent in the pH range of 4.0–9.0 a nd the loss in the activity of the lectin is concomitant with the unfolding of the protein. A t 3 M GdnHCl, more than 5 0% activity was lost with 60% decrease in the ratio (F335/356) and 25–35% shift in the emission maximum. Refolding of the protein Renaturation or re folding of t he protein w as measured as the extent of reappearance of the original spectra (F 335/ 356) and recovery of the sugar binding activity. After dilution of the r eaction mixture containing lectin and GdnHCl (10 t imes), partial reactivation o f the lectin was observed. The lectin treated with 3 , 4, and 5 M GdnHCl had 45, 13 and 7% activity, which increased to 75, 37, and 23%, respectively ( Table 1 ) on renaturation. Re fold- ing of the protein w as indicated by s ubstantial increase i n the F(335/356) ratio. GdnHCl probably unfolds the protein in such a way that substantial interactions are reformed after removal of the denaturant, leading to the significant reformation of the structure and regaining of activity. Gel filtration studies The native protein gets dissociated first into dimer (M r 30 000) and then into monomer (M r 14 000) with increasing concentration of G dnHCl (Fig. 2). At 3–4 M GdnHC l, a single peak at 10.4 min appears that seems arise f rom the totally denatured monomer. Complete dissociation of the tetramer does not take place even at 6 M GdnHCl. The protein components corresponding to the peaks 1, 2, 3 and 0 were analysed separately for sugar binding activity. Peak 1 was found to be the f olded and active fraction of the total population o f the lectin molecules t reated with GdnHCl. Peak 2 is p artially unfolded form of t he lectin with traces of activity. Peak 3 is unfolded, inactive monomer. Peak 0 is the totally denatu red m onomer similar t o t hat observed in c ase of peanut lectin [6]. The dissociation of the native protein in presence of GdnHCl into dimer is reversible, that into monomer is irreversible as observed by rechromatography of the individual peaks on gel filtration column under renaturing conditions (data not shown). Based on the dissociation pattern, the following scheme can be written: T () D () M () M* where T is tetramer, D is dimer, M is monomer and M* is totally denatured monomer. The monomer seems to be unstable and the conforma- tional stability of the oligomer seems to be contributed wholly by the q uaternary i nteractions. I n case of p eanut lectin, folded m onomer is obtained after dissociation of the protein [7] and the molten globule-like state of the monomer was d etected during its unfolding [6], both of which retain the sugar binding activity. Thermal denaturation The A. hirsuta lectin loses sugar binding activity and starts precipitating above 45 °C. The fluorescence emission spec- trum broadens, but the emission maxima does not shift from 335 to 356 even at 70 °C where almost total inactivation of the lectin takes place. T he decrease in the ratio F(335/356) observed for thermally denatured protein, from 1.36 (native) to 1.03 (70 °C, 15 min) (Table 1) was less than that observed with GdnHCl denaturation (at pH 7.0), 1.35 (native) to 0.82 (5 M GdnHCl) ( Fig. 1B). Table 1. Effect of treatment GdnHCl and thermal denaturation and renaturation on A. hirsuta lectin. The samples treated w ith GdnHCl were diluted 10 times with 100 m M phosphate buffer, pH 7.0, the spectra were recorded and activity was estimated as described in Materials and methods. The lectin samples i ncubated at respective temperatures were cooled to 35 °C, spectra were recorded and activity was estimated. Treatment Activity (%) F 335/356 On denaturation On Renaturation On denaturation On Renaturation Lectin + GdnHCl (0 M ) 100 100 1.35 1.35 Lectin + GdnHCl (3 M ) 45 75 1.15 1.26 Lectin + GdnHCl (4 M ) 13 37 0.86 1.2 Lectin + GdnHCl (5 M ) 7 23 0.82 1.16 Lectin Þ 35 °C,15 min 100 100 1.36 1.36 Lectin Þ 45 °C,15 min 75 70 1.26 1.18 Lectin Þ 50 °C,15 min 53 35 1.23 1.1 Lectin Þ 60 °C,15 min 13 5 1.09 0.98 Lectin Þ 70 °C,15 min 7 0 1.03 0.95 Ó FEBS 2002 Denaturation studies of Artocarpus hirsuta lectin (Eur. J. Biochem. 269) 1415 On thermal denaturation, the p rotein forms insoluble aggregates before total unfolding and loses its sugar binding activity. When the temperature of the samples incubatedfrom45to70°C was brought down slowly to 35 °C, the activity was not restored and no refolding was observed a s there is decrease in the F(335/356) ratio (Table 1) indicating that the thermal denaturation is irreversible. Because the protein s tarts a ggregating o n thermal denaturation, Rayleigh light scattering studies were carried out. The protein shows higher light scattering intensity a t 45 °C (Fig. 3A) which goes on increasing with further i ncrease i n the temperature. ANS binding studies Binding of ANS to the proteins occurs upon the exposure of hydrophobic clusters during the unfolding process. ANS does not bind to the native or t he denatured states of t he A. hirsuta lectin but binds at the intermediate stage (at 2 M GdnHCl), showing increase in t he fluorescence intensity, indicating temporary exposure of the hydrophobic patches of the protein during unfolding (Fig. 3B). The p ossibility of the occurrence of the molten globule during unfolding of A. hirsuta lectin as observed in t he peanut lectin [6], was ruled out because a significant amount of th e tertiary and secondary structure was intact. T he ANS binding to the protein samples exposed to 50, 60, and 70 °C was more than those incubated a t 30 a nd 40 °C (Fig. 3C) i ndicating the exposure of hydrophobic patches are due to therm al denaturation. The tendency of the protein to aggregate increases a s the hydrophobic patches get exposed due to thermal denaturation. The CD s pectra of the protein exposed at 45–70 °C for 10 min s hows progressive loss in the secondary structure (Fig. 3D). There s eem to be two different modes o f d enaturation o f the A. hirsuta lectin with GdnHCl and heat. The former unfolds and inactivates the protein, allowing it to fold back and reactivate to certain extent after removal of the r eagent. Thermal denaturation l eads to unfolding and simultaneous formation of insoluble aggregates and is therefore irrevers- ible. Different modes of folding and unfolding observed under different conditions could be due to the unusual Fig. 2. Gel filtration of A. hirsuta lectin in presence of GdnHCl in 100 m M potassium phosphate buffer (pH 7.0). Molarity of GdnHCl is indicated o n t he figure. M r values of the standards used were as f o l- lows, 1, BSA 6 6 kDa, 2 , ov albumin, 45 kDa, 3, carbonic anhydrase, 29 kDa and 4, cytochrome c, 14. 5 k Da. Fig. 3. Rayleigh light scattering (A) and ANS fluorescence (B,C) studies of A. hirsuta lectin. (A) The l ectin (1.5 l M ) was incubated at d iff erent temperatures for 10 min an d the light scattering was m onitored by setting kex ¼ kem ¼ 400nm1,50m M bufferofpH7.0,2,30°C, 3, 35 °C, 4, 40 °C and 5, 45 °C. (B) C hange in A NS fluorescence in the presence of A. hirsuta lectin an d GdnHCl. T he fluorescence emission spectra of the lectin (2.0 l M ) in the presence of ANS (50 l M ). (kex, 375 nm). Numbers on the curves indicate the molarity of GdnHCl. (C) Change in ANS fl uorescence in the p resence of t he A. hirsuta lectin a t various temperatures. T he spectra were t aken as described in (B) prote in sam ples treated a t, 1, 30 °C, 2, 40 °C, 3, 50 °C, 4, 60 °C, and 5, 70 °C. (D) N ear UV CD s pectra of A. hirsuta lectin (15 l M ), lectin exposed at 1, 35 °C , 2, 45 °C, 3, 50 °C, 4, 55 °C, 5, 60 °C, 6, 65 °C and 7, 70 °C f or 15 min. 1416 S. M. Gaikwad et al. (Eur. J. Biochem. 269) Ó FEBS 2002 folding and association of subunits of the lectin as compared to other plant lectins [5,6]. REFERENCES 1. Lis, H. & Sharon, N. (1991) lectin–carbohydrate interactions. Curr. Opin. Struct. Biol. 1, 741–749. 2. Gurjar, M.M., Khan, M.I. & Gaikwad, S.M. (1998) a-Galactoside binding lectin from Artocarpus hirsuta: c haracterization of the sugar specificity and binding site. Biochim. Biophys. Acta 1381, 256–264. 3. Gaikwad, S.M., Gurjar, M.M. & Khan, M.I. (1998) Fluorimetric studies on saccharide binding to the b asic lectin fr om Artocarpus hirsuta Biochem. Mol . Biol. I nt. 46,1–9. 4. Rao, K.N., Gurjar, M.M., Gaikwad, S.M., Khan, M.I. & Suresh, C .G. (1999) Crystallization and preliminary X-ray studies of the basic lectin from the seeds of Artocarpus hirsuta. Acta Crystallo. D 55, 120 4–1205. 5. Ahmed, N ., Srinivas, V .R., Reddy, G.B. & Surolia, A. (1998) Thermodynamic characterization of the confor mational stability of th e homodimeric protein, Pea lectin. Bi oc hemi stry 37, 16765– 16772. 6. Reddy, G .B., Srinivas, V.R., Ahmed, N . & Surolia, A. (1999) Molten globule like state of peanut l ectin monomer retains its carbohydrate specificity. J. Biol. Chem. 274, 4500–4503. 7. Reddy, G.B., Bharadwaj, S. & Surolia, A. (1999) Thermal stability and m ode o f oligomerization of the tetrameric Peanut agglutinin: a d ifferent scanning calorimetric stud y. Bioc hemistry 38 , 4 464–4470. Ó FEBS 2002 Denaturation studies of Artocarpus hirsuta lectin (Eur. J. Biochem. 269) 1417 . Artocarpus hirsuta lectin Differential modes of chemical and thermal denaturation Sushama M. Gaikwad, Madhura M. Gurjar* and M. Islam Khan Division of. bind carbohydrates specifically and reversibly are termed as lectins. They occur ubiquitously in nature and have diverse role in plants, animals and microbes.