Trypanosoma cruzi: Conformational Preferences of Antigenic Peptides Bearing the Immunodominant Epitope of the B13 Antigen

Experimental Parasitology 93, 38–44 (1999) Article ID expr.1999.4428, available online at http://www.idealibrary.com on

Trypanosoma cruzi: Conformational Preferences of Antigenic Peptides Bearing the Immunodominant Epitope of the B13 Antigen

Ma´rcia A. Duranti,*,§ Lorella Franzoni,† Giorgio Sartor,† Arianna Benedetti,† Le´o K. Iwai,* Arthur Gruber,‡ Bianca Zingales,§ Fanny Guzman,¶ Jorge Kalil,* Alberto Spisni,† and Ede´cio Cunha-Neto*,1 *Laboratory of Transplantation Immunology, Heart Institute, Faculty of Medicine, and ‡Faculty of Veterinary Medicine and Zootechny, and §Institute of Chemistry, University of Sa˜o Paulo, 05403 Sa˜o Paulo, SP, Brazil; †Institute of Biological Chemistry, University of Parma, Parma, Italy; and ¶Institute of Immunology, Hospital San Juan de Dios, National University of Colombia, Bogota´, Colombia

Duranti, M. A., Franzoni, L., Sartor, G., Benedetti, A., Iwai, L. K., Gruber, A., Zingales, B., Guzman, F., Kalil, J., Spisni, A., and CunhaNeto, E. 1999. Trypanosoma cruzi: Conformational preferences of antigenic peptides bearing the immunodominant epitope of the B13 antigen. Experimental Parasitology 93, 38–44. The Trypanosoma cruzi recombinant protein B13 contains tandemly repeated domains and shows high sensitivity in the serological diagnosis of Chagas’ disease. It has been shown that the immunodominant epitope of B13 is contained in the GDKPSLFGQAAAGDKPSLF-NH2 sequence and that the hexapeptide AAAGDK seems to be the “core” of that epitope. Three peptides containing that “core” sequence, one corresponding to the entire repeat motif GDKPSLFGQAAAGDKPSLF-NH2, pB13, and two smaller fragments, FGQAAAGDK-NH2, S4, and QAAAGDKPS-NH2, S5, have been tested in competitive ELISA with recombinant protein B13 in the solid phase against 40 chagasic sera from Brazilian patients. The median percentage inhibition for pB13, S4, and S5 were, respectively, 91, 86, and 68%. The possibility that the distinct antigenic activity of those peptides correlates with the existence of preferential conformational properties has been investigated by CD and NMR spectroscopy. Results indicate their propensity to adopt a helical configuration, centered in the AAAGDK sequence, and whose extent and stability directly correlates with the peptides’ antigenicity. The data are discussed in the light of the existence of conformational preferences involving immunodominant epitopes in tandemly repeated antigens. q 1999 Academic Press Index Descriptors and Abbreviations: Trypanosoma cruzi; Chagas’ disease; B13 antigen; repetitive epitope; antigenicity; immunodominance; peptide conformation; ELISA, enzyme-linked immunosorbent

assay; OD, optical density; TFE, trifluoroethanol; PBS, phosphatebuffered saline solution; 2D 1H NMR, two-dimensional proton nuclear magnetic resonance; CD, circular dichroism.

INTRODUCTION

Protein antigens of several pathogenic protozoa frequently display immunodominant tandemly repeated sequences (Schofield 1991; Frasch et al. 1991). Little is known, yet, about the molecular mechanisms underlying the immunodominance of repeated sequences, except for their multivalency which can make them activate B cells directly and behave as T-independent antigens (Schofield 1991). However, though still a matter of debate, there is much evidence that epitope conformation can be relevant for antigen– antibody binding (Dyson et al. 1985; Dyson and Wright 1995); furthermore, peptide sequences with conformational preferences have been shown to be preferentially recognized by antibodies against native protein epitopes (Craig et al. 1998). The so-called “smoke screen” hypothesis (Kemp et al. 1987) interprets the immunodominance of these tandem repeats as a means developed by the parasites to protect themselves by diverting the immune response away from

1

To whom correspondence should be addressed. Fax: 55-11 282 9350. E-mail: [email protected].

38

0014-4894/99 $30.00 Copyright q 1999 by Academic Press All rights of reproduction in any form reserved.

39

Trypanosoma cruzi: CONFORMATIONAL PREFERENCE OF B13 IMMUNODOMINANT EPITOPE

the antigen functional domains (for Trypanosoma cruzi examples see Cazzulo and Frasch 1992). Chagas’ disease is caused by the protozoan T. cruzi and is endemic in Latin America. The immunodominant recombinant B13 antigen is a tandemly repeated domain of the T. cruzi 140/116 kDa antigen located on the surface of infective trypomastigotes (Gruber and Zingales 1993). Even though the biological properties of the T. cruzi B13 antigen that contains the immunodominant epitope are not known, analogous gene sequences have been isolated from different T. cruzi strains (Iban˜ez et al. 1988; Hoft et al. 1989). The recombinant B13 antigen is recognized by sera from 98% of patients with Chagas’ disease and is formed by 19 tandemly repeated partially degenerate copies of the repeat motif composed of 12 amino acid residues P(L)P(S,A)P(L)FGQAAA(E)G(A)D(G)K, where residues within parentheses can be found replacing the preceding residue in different repeat copies (Gruber and Zingales 1993). Synthetic peptides encompassing the 12 amino acid repeat attain high specificity and sensitivity in serodiagnosis of Chagas’ disease by ELISA (Vergara et al. 1992; Peralta et al. 1994). In previous studies, we tested the antigenicity of a series of B13-derived synthetic peptides on a limited number of sera from chagasic patients by means of competitive ELISA with the B13 recombinant protein immobilized in the solid phase. Results indicated that the peptide FGQAAAGDKNH2 (S4) contained the immunodominant epitope of B13 and that the sequence AAAGDK should be the “core” of that epitope (Cunha-Neto et al. 1995). Furthermore, the three antigenic AAAGDK-containing peptides, pB13, S4, and S5 (Table I), differing for the amino acids flanking the putative “core” epitope, seemed to exhibit a distinct ability to bind to anti-B13 antibodies from patients’ sera (Cunha-Neto et al. 1995). Given the previously demonstrated role of peptide conformation in antigenicity (Craig et al. 1998; Leder et al. 1995), this finding prompted us to verify the existence of a correlation between the conformational properties and the antigenic activity of the three peptides. Here we present data supporting the hypothesis of a direct correlation between the presence of preferential conformational determinants and the immunodominance of the epitope included in the B13 tandemly repeated antigen.

MATERIALS AND METHODS Synthetic peptides. Peptides (Table I) were synthesized by solidphase Merrifield “teabag” technology from t-Boc derivatives, HPLC purified, and checked by mass spectrometry.

TABLE I Amino Acid Sequences of Synthetic Peptides Used in the Assays Name pB13 S4 S5 pNR

Sequence GDKPSLFGQAAAGDKPSLF-NH2 FGQAAAGDK-NH2 QAAAGDKPS-NH2 NKSAKQFSLHIMDSQPDGS

Protein origin B13 B13 B13 TCRVa13

Immunological assays. Competitive inhibition of B13 ELISA was performed essentially as described (Cunha-Neto et al. 1995). Briefly, appropriate dilutions of serum samples (yielding O.D.’s in the 0.3–0.8 range) from Chagas’ disease sero-positive Brazilian patients (20 with Chagas cardiomyopathy and 20 asymptomatic), pre-adsorbed with 8 mg/ml Escherichia coli lysate to reduce background, were preincubated with the synthetic peptides (0.2 M) overnight at 48C. Each serum/ peptide mixture was then incubated 1 h at 378C in duplicate wells of a Corning polypropylene 96-well microtiter plate previously sensitized with 20 ng/well of recombinant B13 protein (Gruber and Zingales 1993). The reaction was developed using an anti-human IgG-peroxidase conjugated with o-phenylenediamine as chromogenic substrate. The ODs were subsequently measured at 490 nm. The antigenic activity for each serum/peptide combination was scored as the percentage inhibition of binding in B13 ELISA test, measured according to the following equation, where the nonrelated peptide sequence NKSAKQFSLHIMDSQPDGS-NH2, derived from the human T-cell receptor V a chain (TCR V a 13), pNR, was used as blank: [OD(serum1pNR)] 2 [OD(serum1test peptide)]/[OD(serum1pNR)] 3 100 The values of the percentage inhibition were then plotted for each peptide/serum combination. Statistical comparison of percentage inhibition values for each peptide was performed with the nonparametric Mann-Whitney’s rank sum test. CD experiments. The spectra for each peptide were recorded on a Jasco J-715 spectropolarimeter in the wavelength range between 188 and 260 nm, using a 1-mm-pathlength cell. Each measure was averaged four times, and an equally signal-averaged solvent baseline was then subtracted. The observed optical activity was converted into the mean residue ellipticity [U] (3 1023 deg cm2 dmol21). NMR experiments. Each peptide was dissolved in 30% PBS (5 mM)/70% TFE-d3, pH 6.0, to a final concentration varying in the range 2–3 mM. All experiments were carried out at 258C, using a specific 5-mm 1H probe-head on a Bruker AMX-400 and/or AMX600 spectrometer, both equipped with an Oxford Instruments superconducting magnet and an Aspect X32 computer. The resonance peaks were referenced to the residual signal of the methylene protons of TFE (3.88 ppm). Sequential resonance assignments were performed using conventional methods (Wutrich 1986), including the acquisition of the two-dimensional (2D) experiments DQF-COSY, TOCSY, NOESY (150 ms mixing time), and ROESY (200 ms spin-locking continuous wave). All 2D experiments were acquired in a phase-sensitive mode using the method of States et al. (1982); 512 t1 experiments with a variable number of transients of 2 K complex points, for each free induction decay, were recorded. In all spectra, the water suppression was achieved by low-power continuous-wave irradiation during the relaxation delay. Data were processed using the Aspect X32 computer and the UXNMR

40 program. To enhance the digital resolution, prior to Fourier transformation, the time-domain data were zero-filled in both dimensions to yield 2 K 3 2 K matrices and apodized by a shifted squared sine bell window function. When required, a fourth-order polynomial baseline correction algorithm was applied after transformation and phasing. The stability and location of helical stretches were assayed by determining the chemical shifts (d) of the a-proton of each amino acid residue, which are strongly dependent on the nature of the polypeptide secondary structure. The d value for the a-proton of an amino acid residue in an a-helical configuration (dahelix) is smaller than the one observed for the same residue within a random coil structure (darc) (Whistart et al. 1992). Thus, for residues within an a-helical stretch, the deviation between the observed a-proton chemical shift (daobs) and the known random coil a-proton chemical shift (darc), defined as Dd 5 daobs 2 darc, yields a negative result. For any peptide, the number of neighboring residues (at least four) displaying a negative deviation, together with the extent of the deviation, correlate with the existence and stability of a helical stretch.

RESULTS AND DISCUSSION

The antigenic activity of pB13, S4, and S5 has been evaluated by measuring, for each peptide, the percentage of inhibition of binding of the anti-B13 antibodies to recombinant protein B13 in the solid phase. The results obtained with 40 sera from Brazilian chagasic patients are reported in Fig. 1. While the values of the median percentage inhibition for the peptides pB13 and S4 cluster at similar but not identical levels, 91 and 86%, respectively (pB13 vs S4, P 5 0.02), the distribution for S5 is quite different, showing a much broader pattern with a median percentage inhibition of 68% (S5 vs pB13 or S4, P , 0.001). No differential behavior between the two Chagas’ disease clinical groups (chronic cardiopathy and asymptomatic) was observed.

FIG. 1. Distribution of the percentage inhibition of B13 ELISA test for peptides pB13, S4, and S5 on sera from 40 Chagas’ disease patients.

DURANTI ET AL.

These data confirm a previous observation, based on a limited number of samples, suggesting that the 9-mer S4 ought to contain the immunodominant epitope of B13 (CunhaNeto et al. 1995). The significant inhibitory activity found in the present work for S5, albeit still lower than the one obtained for S4 and pB13, is in disagreement with previous results (Cunha-Neto et al. 1995). A more detailed analysis, by means of mass spectrometry, automated sequencing, and 2D 1H NMR, of the S5 peptide batch previously used, disclosed that the residue D was missing, the sequence being QAAAG KPS (data not shown). Therefore, the earlier negative results must be attributed to this fact. Since the three peptides pB13, S4, and S5 display different antigenic activities despite containing the “core” AAAGDK sequence, it follows that the sequence may be necessary, but not sufficient, for optimal antibody recognition. Furthermore, it is likely that flanking residues modulate the ability of the peptide to be recognized by anti-B13 antibodies, as the 9-mer peptides S4 and S5 show markedly different antigenic activities. Further in support of that notion, initial results indicate that while the 6-mer AAAGDK is not recognized, the 7-mer peptides QAAAGDK and AAAGDKP, with one single amino acid residue flanking the “core” AAAGDK sequence, are recognized by Chagasic serum anti-B13 antibodies and show median percentage inhibitory values similar to peptide S5 (data not shown). Because such an apparent hierarchy in antigenic activity depending on flanking residues could be associated with a distinct capability of the peptides to assume an ordered secondary structure (Craig et al. 1998) in the AAAGDK region, we investigated this possibility by using CD and NMR spectroscopy. Helices are reported to be among the most frequent conformational determinants recognized by antibodies (Dyson et al. 1985; O’Hern 1991; Dyson and Wright 1995). Thus, we focused our attention on the potential of these peptides to acquire some helical secondary structure: helicity. Knowing that trifluoroethanol (TFE) is a cosolvent able to stabilize helices in polypeptide regions that already have an intrinsic propensity for such a secondary structure (Lehrman et al. 1990; Dyson et al. 1992ab), experiments have been carried out in PBS and in PBS/TFE mixed solvent. In Fig. 2a, the dotted line displays the CD profile for pB13 in 3 mM PBS at pH 6 and 258C, and the solid line presents the spectrum of the same peptide obtained in 30% PBS/70% TFE at 258C. It is known that the presence of a negative band in the region of 220 nm can be diagnostic for the presence of helical stretches, while the presence of a strong negative absorption around 195 nm is associated with random coil organization. Data indicate that in PBS, pB13 peptide exists preferentially in an extended conformation,

Trypanosoma cruzi: CONFORMATIONAL PREFERENCE OF B13 IMMUNODOMINANT EPITOPE

41

FIG. 2. CD spectra of 0.1 mM peptide at 258C in 3 mM PBS, pH 6 (dotted line), and in 30% phosphate buffer/70% TFE, v/v (solid line). (a) pB13; (b) S4; (c) S5.

while in the presence of TFE, pB13 partially folds into a helical structure. For the other two peptides in PBS (Figs. 2b and 2c, dotted lines) the CD profiles again are indicative of an extended conformation. However, in the presence of TFE (Figs. 2b and 2c, solid line) the change in the CD profile suggests a partial structural reorganization associated with a high molecular flexibility. In order to better define the extent and location of those putative structural determinants, a study has been carried out using 2D 1HNMR in 30% PBS/70% TFE. The plots of the deviations of chemical shifts for the a-proton (Dd) of each amino acid residue in the three peptides are reported in Fig. 3. It can be seen that for pB13 and S4, the peptide stretch involved in a helical structure extends over eight

amino acid residues showing negative Dd and encompassing the immunodominant hexapeptide “core”. In the case of S5, instead, we observe a net decrease of the number of sequential residues with negative Dd as well as a decrease of the extent of the Dd values themselves, indicating a reduced capability of the peptide to assume a stable helical configuration. Data suggest that, under appropriate environmental conditions (Dyson et al. 1988), pB13 would be capable of switching from a random to a helical conformation. The fact that peptide helicity is expressed only in the presence of TFE is particularly enlightening if we consider that the dielectric constant of this solvent is similar to the one present inside or at the protein surface (So¨nnischsen et al. 1992; Jasanoff

42

FIG. 3. Differences between the observed and the random coil (Whistart et al. 1992) a-proton chemical shifts (dapeptide - darc) for (a) pB13, (b) S4, and (c) S5 in 30% PBS buffer/70% TFE-d3, (v/v) at 258C, plotted residue by residue against the peptides’ sequence.

DURANTI ET AL.

and Fersht 1994). Thus, one can hypothesize that the conformational changes induced by TFE on B13-derived peptides may also occur at the peptide–anti-B13 antibody interface. It has been suggested that the antigen–antibody interface may force an interconversion of the peptides from an inactive extended configuration to an active structured one (Fieser et al. 1987), in a process dubbed “induced fit by conformational selection” (Leder et al. 1995). Taken together, the data suggest the existence of a direct correlation between the degree of antigenicity of the peptides studied and their propensity to assume helical structures along the hexapeptide “core” region. Results also indicate an effect of the amino acid residues flanking the epitope in modulating the secondary structure of the peptide core that simultaneously affects the antigenicity. Finally, the coincidence between the position of the single epitope of the tandemly repeated domain of B13 and the region that has the ability to assume a helical conformation suggests that this structural feature may be responsible for immunodominance. Taking into account the nature of the competitive ELISA with native B13 in the solid phase, results indicate that peptides possessing both the AAAGDK sequence and the helical propensity in that region (namely, pB13 and S4) can effectively compete for binding antibodies directed against native B13 protein. These observations reinforce the possibility that the native structure of B13 tandemly repeated sequence may also include repetitive helical components, as observed in other antigen systems (Craig et al. 1998). NMR and CD spectroscopy have also been used to probe the conformational ensemble of the tandemly repeated tetrapeptide unit of the circumsporozoite (CS) coat protein of Plasmodium falciparum and the data are consistent with the presence of helical structures in rapid dynamic equilibrium with extended-chain forms (Dyson et al. 1990). These conclusions confirmed previous sequence-based predictions about the secondary structure of CS proteins from different Plasmodium species where a significant bias was observed for b-turns in the immunodominant tandemly repeated regions (Nussenzweig and Nussenzweig 1986). In conclusion, we believe that, at least in the documented cases of T. cruzi B13 antigen and P. falciparum CS protein, the immunodominance of tandemly repeated sequences may be associated with their ability to present multivalent protein epitopes displaying a similar conformation in each repetitive unit, able to be recognized by a single type of antibody. Furthermore, the preferential expansion of B cell clones secreting such complementary antibodies may lead to the accumulation in patients’ sera of antibodies with this peculiar epitope specificity. It can be hypothesized that amino acid sequences with conformational preferences and increased

Trypanosoma cruzi: CONFORMATIONAL PREFERENCE OF B13 IMMUNODOMINANT EPITOPE

43

sequence of proteins. Models for initiation of protein folding. I Myohemerythrin. Journal of Molecular Biology 226, 795–817.

antigenic potential may have been selected for in other immunodominant tandemly repeated antigenic domains of protozoan proteins. Studies are now in progress to better define the molecular aspects of B13 immunodominance. In particular, in the case of B13-derived antigenic peptides, a more detailed characterization of the active secondary structure of the “core” of immunodominant epitope, as well as of the stabilizing role of the flanking amino acids, is under way.

Dyson, H. J., Sayre, J. R., Merutka, G., Shin, H. C., Lerner, R. A., and Wright, P. E. 1992b. Folding of peptide fragments comprising the complete sequence of proteins. Models for initiation of protein folding. II Plastocyanin. Journal of Molecular Biology 226, 819–835.

ACKNOWLEDGMENTS

Fieser, T. M., Tainer, J. A., Geysen, H. M., Houghten, R. A., and Lerner, R. A. 1987. Influence of protein flexibility and peptide conformation on reactivity of monoclonal anti-peptide antibodies with a protein a-helix. Proceedings of the National Academy of Science USA 84, 8568–8572.

The work was partly supported by FAPESP (Brazil), CNPq (Brazil), MURST (Italy), and CNR (Italy). The Interfaculty Centre for Measurements of the University of Parma (Italy) is kindly acknowledged for the use of its facilities.

REFERENCES Cazzulo, J. J., and Frasch A. C. 1992. SAPA/trans-sialidase and cruzipain: Two antigens from Trypanosoma cruzi contain immunodominant but enzymatically inactive domains. The FASEB Journal 6, 3259–3264. Craig, L., Sanschagrin, P. C., Rozek, A., Lackie, S., Kuhn, L. A., and Scott, J. K. 1998. The role of structure in antibody cross-reactivity between peptides and folded proteins. Journal of Molecular Biology 281, 183–201. Cunha-Neto, E., Duranti, M., Gruber, A., Zingales, B., de Messias, I., Stolf, N., Bellotti, G., Pilleggi, F., Patarroyo, M. E., and Kalil, J. 1995. Autoimmunity in Chagas’ disease cardiopathy: Biological relevance of a cardiac myosin-specific epitope crossreactive to an immunodominant Trypanosoma cruzi antigen. Proceedings of the National Academy of Science USA 92, 3541–3545. Dyson, H. J., and Wright, P. E. 1995. Antigenic peptides. The FASEB Journal 9, 37–42. Dyson, H. J., Cross K. J., Houghten, R. A., Wilson, I. A., Wright, P. E., and Lerner, R. A. 1985. The immunodominant site of a synthetic immunogen has a conformational preference in water for a type-II reverse turn. Nature 318, 480–483. Dyson, H. J., Lerner, R. A., and Wright, P. E., 1988. The physical basis for induction of protein-reactive antipeptide antibody. Annual Review Biophysics and Chemistry 17, 305–324. Dyson, H. J., Satterthwait, A. C., Lerner, R. A., and Wright, P. E. 1990. Conformational preferences of synthetic peptides derived from the immunodominant site of the circumsporozoite protein of Plasmodium falciparum by 1H NMR. Biochemistry 29, 7828–7837. Dyson, H. J., Merutka, G., Waltho, J. P., Lerner, R. A., and Wright, P. E. 1992a. Folding of peptide fragments comprising the complete

Frasch, A. C. C., Cazzulo, J. J., Aslund, L., and Petterson, U. 1991. Comparison of genes encoding Trypanosoma cruzi antigens. Parasitology Today 7, 148–151.

Gruber, A., and Zingales, B. 1993. Trypanosoma cruzi: Characterization of two recombinant antigens with potential application in the diagnosis of Chagas’ disease. Experimental Parasitology 76, 1–12. Hoft, D. F., Kim, K. S., Otsu, K., Moser, D. R., Yost, W. J., Blumin, J. H., Donelson, J. E., and Kirchhoff, L. V. 1989. Trypanosoma cruzi expresses diverse repetitive protein antigens. Infection and Immunity 57, 1959–1967. Iban˜ez, C. F., Affranchino, J. L., Medina, R. A., Reyes, M. B., Leguizamon, S., Camargo, M. E., Aslund, L., Pettersson, U., and Frasch, A. C. C. 1988. Multiple Trypanosoma cruzi antigens containing tandemly repeated amino acid sequence motifs. Molecular and Biochemical Parasitology 30, 27–34. Jasanoff, A., and Fersht, A. R. 1994. Quantitative determination of helical propensities from trifluoroethanol titration curves. Biochemistry 33, 2129–2135. Leder, L., Berger, C., Bornhauser, S., Wendt, H., Ackermann, F., Jelesarov, I., and Bosshard, H. R. 1995. Spectroscopic, calorimetric and kinetic demonstration of conformational adaptation in peptideantibody recognition. Biomedical Peptides Proteins Nucleic Acids 1: 109–116. Lehrman, S. R., Tuls, J. L., and Lund, M. 1990. Peptide a-helicity in aqueous trifluoroethanol: Correlations with predicted a-helicity and the secondary structure of the corresponding regions of bovine growth hormone. Biochemistry 29, 5590–5596. Kemp, D. J., Coppel, R. L., and Anders, R. F. 1987. Repetitive proteins and genes of malaria. Annual Review of Microbiology 41, 181–208. Nussenzweig, V., and Nussenzweig, R. S. 1986. Development of a sporozoite malaria vaccine. American Journal of Tropical Medicine and Hygiene 35, 678–688. O’ Hern, P. A. 1991. Immunogenicity of peptides having pre-determined alpha-helical and alpha-alpha fold topologies. Molecular Immunology 28, 1047–1053. Peralta, J. M., Teixeira, M. G. M., Shreffler, W. G., Pereira, J. B., Burns, J. M., Sleath, P. R., and Reed, S. G. 1994. Serodiagnosis of Chagas’ disease by enzyme-linked immunosorbent assay using two synthetic peptides as antigens. Journal of Clinical Microbiology 32, 971–974.

44

DURANTI ET AL.

Schofield, L. 1991. On the function of repetitive domains in protein antigens of Plasmodium and other eukaryotic parasites. Parasitology Today 7, 99–105.

of an enzyme-linked immunosorbent assay for the diagnosis of Chagas’ disease using synthetic peptides. American Journal of Tropical Medicine and Hygiene 46, 39–43.

So¨nnichsen, F. D., Van Eyk, J. E., Hodges, R. S., and Sykes, B. D. 1992. Effect of trifluoroethanol on protein secondary structure: An NMR and CD study using a synthetic actin peptide. Biochemistry 31, 8790–7898.

Whishart, D. S., Sykes, B. D., and Richards, F. M. 1992. The chemical shift index: A fast and simple method for the assignment of protein secondary structure through NMR spectroscopy. Biochemistry 31, 1647–1651.

States, D. J., Hakerborn, R. A., and Ruben, D. J. 1982. A two-dimensional nuclear overhauser experiment with pure absorption phase in four quadrants. Journal of Magnetic Resonance 48, 286–292.

Wutrich, K. 1986. “NMR of Proteins and Nucleic Acids,” Wiley, New York.

Vergara, U., Veloso, C., Gonzalez, A., and Lorca, M. 1992. Evaluation

Received 11 January 1999; accepted with revision 5 May 1999