Biochemistry zyxwvuts 1994,33, 14011-14017 14011 Three-Dimensional Solution Structure of an Immunoglobulin Light Chain-Binding Domain of Protein L. Comparison with the IgG-Binding Domains of Protein zy GT*$ Mats Wikstrom,*s§ Torbjorn Drakenberg,§ Sture ForsCn,§ Ulf Sjobring," and Lars Bjorckl Departments of Physical Chemistry zyxwvutsr 2, Medical Microbiology, and Medical and Physiological Chemistry, Lund University, Lund, Sweden Received July 5, 1994; Revised Manuscript Received September zyxw 16, 1994@ ABSTRACT: Protein L is a multidomain protein expressed at the surface of some strains of the anaerobic bacterial species Peptostreptococcus magnus. It has affinity for immunoglobulin (Ig) through interaction with framework structures in the variable Ig light chain domain. The Ig-binding activity is located to five homologous repeats called B1-B5 in the N-terminal part of the protein. We have determined the three-dimensional solution structure of the 76 amino acid residue long B 1 domain using NMR spectroscopy and distance geometry-restrained simulated annealing. The domain is composed of a 15 amino acid residue long disordered N-terminus followed by a folded portion comprising an a-helix packed against a four-stranded P-sheet. These secondary structural elements are well determined with a backbone atomic root mean square deviation from their mean of 0.54 A. The B domains of protein L show very limited sequence homology to the domains of streptococcal protein G interacting with the heavy chains of IgG. However, despite this fact, and their different binding properties, the fold of the B 1 domain was found to be similar to the fold of the IgG-binding protein G domains [Wikstrom, M., Sjobring, U., Kastem, W., Bjorck, L., Drakenberg, T., & ForsCn, zyxwvutsr S. (1993) Biochemistry 32, 3381-33861. In the present study, the solution structure of the B 1 domain enabled a more detailed comparison which can explain the different Ig-binding specificities of these two bacterial surface proteins. Among the differences observed, the a-helix orientation is the most striking. Thus, in the B1 domain of protein L the helix is almost parallel to the P-sheet, whereas in the protein G domains the helix runs diagonally across the sheet. Over the last decade a number of immunoglobulin- (Ig-)l binding surface proteins have been identified and character- ized in different, mostly Gram-positive, bacterial species [for references see Boyle (1990) and Kehoe (1994)l. With one exception, all of these proteins interact with heavy chains of Ig. The exception, protein L, was identified in some strains of the anaerobic bacterial species Peptostreptococcus magnus (Myhre & Emtell, 1985). The molecule was isolated and found to bind Ig light chains, predominantly K light chains (Bjorck, 1988). Protein L is an elongated protein which, despite its high-affinity interaction with the variable domain of Ig light chains, does not ipterfere with the antigen- binding activity of the antibody (Akerstrom Bjorck, 1989; ' This work was supported by grants from the Swedish Medical Research Council (Projects 7480, 9926, and 10434), the Medical Faculty, Lund University, and the Swedish Research Council for Engineering Sciences (Project 123). The 500-MHz NMR spectrometer was purchased with grants from the Knut and Alice Wallenberg Foundation and the Swedish Council for Planning and Coordination of Research. Atomic coordinates for the protein L B1 domain have been deposited in the Brookhaven Protein Data Bank with accession code 2PTL. * To whom correspondence should be addressed. zyxwvuts 5 Physical Chemistry 2. Medical Microbiology. Medical and Physiological Chemistry. @ Abstract published in Advance ACS Abstracts, November 1, 1994. ' Abbreviations: Ig, immunoglobulin; Fab, antigen binding fragment of Ig; Fc, constant region of the heavy Ig chain; NMR, nuclear magnetic resonance; 2D, two dimensional; NOE, nuclear Overhauser enhance- ment; NOESY, nuclear Overhauser enhancement spectroscopy; PE- COSY, primitive exclusive correlated spectroscopy; SA, simulated annealing; rms, root mean square. Nilson et al., 1992). This is explained by the fact that protein L interacts with the framework region of the variable domain (Nilson et al., 1992). The protein L originally isolated and characterized contains five highly homologous Ig light chain- binding domains (Bl-B5) of 72-76 amino acid residues each (Kastem et al., 1992) and has no albumin-binding activity (de Chateau et al., 1993). However, a recent report describes a protein L variant which apart from four B-type repeats also contains albumin-binding repeats (Murphy et al., 1994) that are closely related t? the albumin-binding domains first identified in protein G (Akerstrom et al., 1987), a surface protein of group C and G streptococci with affinity also for Fc and Fab fragments of IgG (Reis et al., 1984; Bjorck & Kronvall, 1984). In this paper we report the three-dimensional structure of the B 1 domain of protein L determined using NMR methods in combination with distance geometry-restrained simulated annealing. A comparison with the structure of the IgG- binding domains of protein G (Gronenbom et al., 1991; Lian et al., 1992; Achari et al., 1992), reveals similarities and differences with implications for the evolution and biology of Ig-binding bacterial proteins. MATERIALS AND METHODS Protein Purification and Sample Preparation. The ex- pression of the protein L B 1 domain and its purification have been described previously (Wikstrom et al., 1993). The peptide, produced in Escherichia coli, includes the 76 amino acids of the B1 domain (K80-G155), preceded by the two C-terminal residues (E78-N79) from the N-terminal domain of protein L. 0006-2960/94/0433- 14011$04.50/0 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 0 1994 American Chemical Society