Keywords: T. dioica lectin, ribosome-inactivating proteins, denaturation, thermostability
 |  |
| ||||
Copyright © International Union of Crystallography 2006 Crystallization and preliminary characterization of a highly thermostable lectin from Trichosanthes dioica and comparison with other Trichosanthes lectins Correspondence e-mail: cg.suresh/at/ncl.res.in Received December 16, 2005; Accepted January 23, 2006. | |||||||||||
Abstract A lectin from Trichosanthes dioica seeds has been purified and crystallized using 25%(w/v) PEG 2K MME, 0.2 M ammonium acetate, 0.1 M Tris–HCl pH 8.5 and 50 µl 0.5%(w/v) n-octyl β-d-glucopyranoside as thick needles belonging to hexagonal space group P64. Unit-cell parameters were a = b = 167.54, c = 77.42 Å. The crystals diffracted to a Bragg spacing of 2.8 Å. Both the structures of abrin-a and T. kirilowii lectin could be used as a model in structure determination using the molecular-replacement method; however, T. kirilowii lectin coordinates gave better values of reliability and correlation parameters. The thermal, chemical and pH stability of this lectin have also been studied. When heated, its haemagglutination activity remained unaffected up to 363 K. Other stability studies show that 4 M guanidinium hydrochloride (Gdn–HCl) initiates unfolding and that the protein is completely unfolded at 6 M Gdn–HCl. Treatment with urea resulted in a total loss of activity at higher concentrations of denaturant with no major structural changes. The protein remained stable over a wide pH range, from pH 6 to pH 12, except for partial unfolding at extremely alkaline pH. The role of disulfide bonds in the protein stability was found to be insignificant. Rayleigh light-scattering studies showed no molecular aggregation in any of the extreme treated conditions. The unusual stability of this lectin resembles that of type II ribosome-inactivating proteins (type II RIPs), which is also supported by structure determination. The structural features observed in a preliminary electron-density map were compared with the other two available Trichosanthes lectin structures. Keywords: T. dioica lectin, ribosome-inactivating proteins, denaturation, thermostability | |||||||||||
Lectins are carbohydrate-binding proteins that occur ubiquitously in nature and exhibit important biological properties such as blood-group specificity, preferential agglutination of tumour cells and mitogenicity (Lis & Sharon, 1991 From our studies, we have found that the two Trichosanthes lectins, one from T. dioica and the other from T. anguina, belonging to the Cucurbitaceae family show unusual stability towards denaturation conditions. T. anguina lectin has been independently studied in detail (Komath et al., 1998 In this paper, we describe the crystallization and preliminary characterization of T. dioica lectin, the evidence of its highly thermostable nature and also the effect of chemical denaturants and pH on its stability. The structure determination using molecular replacement and the gross structural features in comparison with the known structures of T. anguina and T. kirilowii lectins are highlighted. | |||||||||||
2.1. Materials Guanidinium hydrochloride (Gdn–HCl) was obtained from Sigma (USA). The lectin from T. dioica was purified following a procedure described by Anuradha & Bhide (1999 ![[triangle]](corehtml/pmc/pmcents/rtrif.gif) ) for T. anguina lectin. Buffers used were glycine–HCl for the pH range 2–3, acetate for pH 4, citrate–phosphate for pH 5, phosphate for pH 7, Tris–HCl for pH 8–9 and glycine–NaOH for pH 10–12 (all at 100 mM concentration). The stock solutions of 8 M Gdn–HCl and 10 M urea were prepared in phosphate buffer pH 7 and filtered through a 0.45 µm filter. Crystal Screen solutions were obtained from Hampton Research (USA). The Tris, PEG 2K MME, n-octyl β-d-glucopyranoside and glycerol used for crystallization experiments were obtained from Sigma (USA). Other chemicals used were obtained locally and were of analytical reagent grade.2.2. Fluorescence studies The effect of temperature on stability was studied by incubating the protein samples (1.8 µM) at temperatures in the range 303–368 K for 15 min each. Haemagglutination activity was checked simultaneously by removing aliquots of the samples. The protein samples (1.8 µM) were equilibrated for 4 h at the required denaturant (Gdn–HCl or urea) concentration at 303 K. The intrinsic tryptophan fluorescence emission of the protein was monitored in 1 cm quartz cell in the 300–400 nm range, when excited at 280 nm, in a Perkin–Elmer LS 50B spectrofluorimeter with attached circulating water bath. Excitation and emission band passes of 5 nm each were used. The activity of the sample was measured at the same time. These studies were also carried out with the protein samples in the presence of the reducing agents DTT and β-mercaptoethanol (5 mM).2.3. Determination of the lectin activity Twofold serial dilution of the lectin in saline medium was carried out in a microtitre plate. A 3%(v/v) erythrocyte suspension was prepared by repeated washing of red blood cells (RBCs) with saline until all proteins were removed and diluting the pellet with saline. 100 µl of this suspension was added to the serially diluted lectin in saline and incubated for 30–60 min at room temperature. Haemagglutination was recorded after 1 h.2.4. Light-scattering studies Rayleigh light-scattering experiments were carried out using the spectrofluorimeter to monitor protein aggregation during Gdn–HCl-induced and thermal denaturation. Both excitation and emission wavelengths were set at 400 nm and the time-dependent change in scattering intensity was monitored.2.5. Renaturation studies 200 µl aliquots were drawn from the samples treated with different concentrations of Gdn–HCl as described above and diluted ten times with 100 mM buffer pH 7.0. The fluorescence spectra of the diluted samples were recorded after 24 h and the activity was assayed. A protein sample in the absence of Gdn–HCl, but treated under identical conditions, was used as a control. The protein samples treated at high temperatures were cooled back to 303 K and scans were repeated.2.6. Crystallization Initial crystallization trials were carried out using Crystal Screens (Hampton Research, USA). This resulted in forked crystals which grew in condition No. 22 of Crystal Screen I, the composition of which was 0.2 M sodium acetate trihydrate, 0.1 M Tris–HCl pH 8.5, 30%(w/v) PEG 4000. The crystal quality was improved by varying the choice and concentrations of salt and PEG. Good crystals were obtained by employing the hanging-drop vapour-diffusion method, in which 1 µl protein solution mixed with 1 µl well solution comprising of 0.2 M ammonium acetate, 0.1 M Tris–HCl pH 8.5, 25%(w/v) PEG 2K MME was equilibrated against 1 ml well solution. The well solution also contained 50 µl 0.5%(w/v) n-octyl β-d-glucopyranoside as an additive. All crystallization experiments were carried out at 295 K.2.7. Diffraction data collection, processing and structure determination X-ray diffraction data were collected on an R-AXIS IV++ image plate using Cu Kα radiation generated by a Rigaku rotating-anode X-ray generator (RUH-3R) operated at 50 kV and 100 mA. X-rays were focused using a confocal mirror system (Osmic, USA). The crystal was kept frozen at 113 K in a liquid-nitrogen cryostream produced by an X-Stream system (Rigaku–MSI, USA) during data collection, using 30%(v/v) glycerol in the crystallization solution as cryoprotectant. The crystal-to-detector distance was kept at 200 mm and an oscillation of 0.5° per frame was used. The programs DENZO and SCALEPACK (Otwinowski & Minor, 1997 ) were used for processing and scaling the data. The structure was determined using the molecular-replacement method implemented in the AMoRe (Navaza, 1994 ) program from the CCP4 suite (Collaborative Computational Project, Number 4, 1994 ). Coordinates of related type II RIPs were used for modelling. | |||||||||||
3.1. Effect of temperature The fluorescence emission spectra of the native lectin showed an emission maximum at 336 nm characteristic of Trp residues in a nonpolar environment. The lectin is active up to 363 K and the fluorescence intensity gradually decreases with increase in temperature. At 363 K the intensity is almost 25% of the original value. On cooling to 303 K the original fluorescence intensity is restored. Although Trp residues are buried, as indicated by the λmax at 336 nm, the microenvironments of these residues seem to be polar. No light scattering was observed under the above conditions, indicating that no aggregation of the protein molecules took place as a consequence of heating. The hydrophobic interactions in the interior of the protein molecules seem to be sufficiently strong to stabilize their respective native structures.3.2. Effect of Gdn–HCl On treatment with increasing concentrations of Gdn–HCl, the lectin slowly started losing activity at a 3 M concentration of the denaturant. There was no shift in the wavelength corresponding to the emission maximum. At a 4 M concentration of Gdn–HCl λmax shifted to 340 nm, indicating a slight increase in the polarity of the Trp environment. A red shift in the λmax to 355 nm observed at 6 M must be a consequence of unfolding. This is accompanied by a major activity loss. At concentrations above 3 M Gdn–HCl, even the intensity of fluorescence has reduced substantially, indicating a change in the microenvironment of the Trp residues. Renaturation or refolding of the protein was measured by the extent of reappearance of the original spectrum and estimation of the degree of recovery of sugar-binding activity. On diluting the lectin treated with 4 and 5 M Gdn–HCl, the activity was partially regained (with a 10–15% increase in the activity) and correspondingly a blue shift of λmax to 340–342 nm was observed after 24 h. This kind of partial renaturation and reactivation was observed also in the case of Artocarpus hirsuta lectin when 3–5 M Gdn–HCl-treated protein was diluted (Gaikwad et al., 2002 ).3.3. Effect of urea Just as in the case of Gdn–HCl-mediated denaturation, the lectin initially remained stable up to a 3 M concentration of urea and then slowly started losing activity when the urea concentration was increased to 4 M. A major loss of activity occurred at a urea concentration of 7 M, but this was not accompanied by any major structural changes in the protein, as no significant shift in λmax could be detected in the fluorescence. Although no major structural changes in lectin treated with urea could be observed, a comparable loss of activity as in the case of Gdn–HCl-treated lectin was observed. The reason could be that the high concentration of urea is either preventing the access of the lectin-binding site to the RBC surface or causing changes in the geometry of the binding site of the lectin itself.3.4. Effect of pH
T. dioica lectin remains stable in the pH range 6–12 and is only partially stable at pH 4, while 60% activity is lost at pH 2. The fluorescence intensity is much lower at highly acidic or highly alkaline pHs compared with the neutral pH range. Thus, except for some minor changes in the microenvironment of the Trp residue, no structural transition occurrs over a wide pH range. The decrease in fluorescence intensity may arise from protonated and deprotonated forms of the amino acids on the surface that affect the emission of the Trp residues.To assess the role of disulfide linkages in the unusual stability of the lectin structure, thermal and Gdn–HCl denaturation studies were carried out in the presence of 5 mM DTT as well as 5 mM β-mercaptoethanol. In both cases, proteins were found to be stable up to 358 K, ruling out involvement of disulfide linkages in stabilizing the structures of Trichosanthes lectins. Previous reports and our own structural investigations indicate that both the lectins contain two non-identical subunits linked by a disulfide bond (Anuradha & Bhide, 1999 3.5. Crystallization and crystal characterization Needle-shaped crystals of T. dioica lectin (Fig. 1 ) were grown against 1 ml well solution consisting of 0.2 M ammonium acetate, 0.1 M Tris–HCl pH 8.5, 25%(w/v) PEG 2K MME along with 50 µl 0.5%(w/v) n-octyl β-d-glucopyranoside as an additive. 30%(v/v) glycerol introduced into the crystallization solution acted as cryoprotectant. Crystals diffracted to 2.8 Å Bragg spacing. Data-collection parameters and intensity statistics are listed in Table 1 .
3.6. Structure determination and comparison Initially, we could not accurately determine the space group of T. dioica crystals because of difficulty in fixing the screw symmetry from systematic absences of reflections. However, on running AMoRe for the translation function separately for all possible sixfold rotation axes, inputting the correct rotation-function solutions, the exact screw translation could be identified through higher correlation and lower R factors of the solution (Table 2 ).
On carrying out the molecular-replacement procedure for structure determination, inputting the coordinates of various available type II RIPs, the coordinates of T. kirilowii lectin (PDB code 1ggp) gave the best results (Table 3
In the absence of sequence information, no extensive refinement could be attempted on the initial model. However, an estimated R factor of 0.38 and R
free of 0.46 were obtained on applying one cycle of refinement (subsequent to rigid-body refinement) prior to electron-density calculation using REFMAC (Murshudov et al., 1997 One major difference between the present crystal structure and those of the T. anguina and T. kirilowii lectins is the association of the two heterodimers in the asymmetric unit and the crystal packing. The association of the two molecules of the T. dioica lectin in the asymmetric unit takes place exclusively through interactions involving the A chains (Fig. 2
Unfortunately, in the absence of confirmed amino-acid sequence for any of the three Trichosanthes lectins, any detailed attempts to correlate stability with structural features at this time may be premature and inconclusive. However, the study reported is highly suggestive of a close relationship between the structural and biochemical characteristics and the type II RIP features common to the three Trichosanthes lectins of T. dioica, T. anguina and T. kirilowii. | |||||||||||
Type II ribosome-inactivating proteins (type II RIPs) are toxic N-glycosidases that depurinate the universally conserved sarcin loop of larger rRNAs. RIPs are widely distributed among different plant genera and are present in a variety of different tissues. Some of them are highly thermostable (Lam & Ng, 2001 | |||||||||||
Acknowledgments PDD is Senior Research Fellow of the Council of Scientific and Industrial Research, New Delhi. PA acknowledges the Department of Biotechnology, New Delhi, India for a Post-Doctoral Fellowship. | |||||||||||
References
| |||||||||||
