Proc. Nati. Acad. Sci. USA Vol. 90, pp. 750-754, January 1993 Biochemistry Crystal structure of STEVEN L. EDWARDS*t, REETTA RAAG*, HIROYUKI WARHSHII, MICHAEL H. GOLDt, AND THOMAS L. POULOS*§¶ *Center for Advanced Research in Biotechnology, 9600 Gudelsky Drive, Rockville, MD 20850; *Department of Chemical and Biological Sciences, Oregon Graduate Institute of Science and Technology, Beaverton, OR 97006-1999; and §Departments of Molecular Biology & Biochemistry and Physiology & Biophysics, University of California, Irvine, CA 92717 Communicated by I. C. Gunsalus, September 18, 1992

ABSTRACT The crystal structure of In this scheme, R is and P is . LiP (LIP) from the basidiomycete chrysosporium compound I (LiP-I) carries both oxidizing equivalents of has been determined to 2.6 A resolution by using multiple H202, one as an oxyferryl (Fe4+-O) center and one as a isomorphous replacement methods and simulated analig porphyrin ir cation (P ), whereas LiP compound II refinement. Of the 343 residues, residues 3-335 have been (LiP-I) carries only one oxidizing equivalent. The substrate accounted for in the electron density map, induding four R is oxidized by compound I to an aryl cation radical with bonds. The overall three-dimensional structure is very subsequent nonenzymatic reactions yielding the final prod- similar to the only other peroxidase in this group for which a ucts (2, 3, 7-9). P. chrysosporium also is able to degrade a high-resolution crystal structure is available, cytochrome c variety of aromatic pollutants (10-12). Overall, the LiP peroxidase, despite the fact that the sequence identity is only reaction cycle is very similar to that of horseradish peroxi- -20%, LiP has four disulfde bonds, while cytochrome c dase (13). peroxidase has none, and LIP is larger (343 vs. 294 residues). Despite the similarities and differences between LiP and The basic helical fold and connectivity defined by 11 helical other , important questions regarding the mech- segments with the sandwiched between the distal and anism of LiP and its exact role in lignin degradation remain proximal helices found in is main- unanswered. For example, the unique capacity of LIP to tained in LIP. Both have a histidine as a proximal oxidize nonphenolic compounds with high potentials heme ligand, which is hydrogen bonded to a buried aspartic (2, 3, 7-9), the basis for the low pH optimum near pH 3.0 is (14-17), and the sensitivity to excess H202 (18, 19) are not side chain. The distal or peroxide binding pocket also understood in detail. Moreover, the mechanism of degrada- similar, including the distal arginine and histidine. The most tion by LiP of polymeric lignin (2, 3, 20) either directly or striking difference is that, whereas cytochrome c peroxidase possibly indirectly via a mediator also is not clearly under- has tryptophans contacting the distal and proximal heme stood. To elucidate the detailed molecular structure and surfaces, LIP has phenylalanines. This in part explains why, in mechanism of LiP, we have determined the crystal structure the reaction with peroxides, cytochrome c peroxidase forms an of this peroxidase. II -centered free radical, whereas LIP forms a por- phyrin ir cation radical. METHODS AND MATERIALS Crystalization. One of the major isozymes of LIP (LiP-2) Lignin is the most abundant aromatic polymer on earth (1). from P. chrysosporium strain OGC101 was purified accord- It comprises 15-30% of woody plant cell walls, forming a ing to earlier procedures (14, 19). Crystals were grown by matrix surrounding the cellulose. This matrix significantly microseeding in hanging drops using polyethylene glycol as retards the microbial depolymerization of cellulose and, the precipitant at pH 4.5 (E. L. Winborne, S.L.E., M.H.G., hence, lignin plays a key role in the earth's carbon cycle'(2, and T.L.P., unpublished data). The crystals belong to 3). White rot basidiomycete fungi are the only known orga- space group P21 with a = 44.7 A, b = 77.5 A, c = 100.0 A, and nisms that are capable ofdegrading lignin to CO2 and H20 (2, (3 = 1010 with two LiP molecules per asymmetric unit. 3). The best-studied white rot fungus, Phanerochaete chry- Intensity data were collected with a Siemens area detector sosporium, secretes two extracellular heme peroxidases, and Rigaku rotating anode x-ray source. lignin peroxidase (LiP) and , which are Structure Determination. The structure was solved by its LIP conventional multiple isomorphous replacement (MIR) pro- major components of lignin degradative system (2-5). cedures, solvent leveling, and noncrystallographic averag- is a glycoprotein with a molecular mass of =41 kDa, contains ing. A total of five heavy atom derivatives using three 1 mol of protoporphyrin IX per mol of , and different reagents was used. A single crystal was used for exists as a series of isozymes of pI 3.2-4.0 (2, 3, 5, 6). LiP each data set and all were >90%o complete to 3.0 A (Table 1). catalyzes the H202-dependent oxidation ofa variety oflignin The initial MIR electron density at 3.0 A was not of very good model compounds in the following multistep reaction se- quality. We subsequently found that this was due to incom- quence: plete identification of heavy atom sites. Nevertheless, non- crystallographic averaging (21) using a suite of programs LiP(Fe3+)P + H202 -- LiP-I(Fe4+-O)F + H20. Abbreviations: LiP, lignin peroxidase; CCP, cytochrome c peroxi- LiP-I(Fe4+-O)P + R -- LiP-II(Fe4+-O)P + RI dase; MIR, multiple isomorphous replacement; F. and Fc, observed and calculated structure factors, respectively. tPresent address: National Institutes of Health, Building 6, Room LiP-II(Fe4+-O)P + R + 2H+ -- LIP(Fe3+)P + R7 + H20 114, Bethesda, MD 20892. ITo whom reprint requests should be addressed at the University of California. The publication costs of this article were defrayed in part by page charge IThe atomic coordinates and structure factors have been deposited payment. This article must therefore be hereby marked "advertisement" in the Data Bank, Chemistry Department, Brookhaven in accordance with 18 U.S.C. §1734 solely to indicate this fact. National Laboratory, Upton, NY 11973 (reference 1LGA). 750 Downloaded by guest on October 1, 2021 Biochemistry: Edwards et al. Proc. Natl. Acad. Sci. USA 90 (1993) 751 Table 1. Summary of structure determination I/al at Maximum maximum Total Unique % Rmerge Data set resolution resolution observed data complete intensity Native 2.03 2.2 123,543 38,748 98 0.074 Uranyl 1 3.0 3.0 24,081 8,538 98 0.14 Uranyl 2 2.5 2.6 26,440 18,252 80 0.08 Platinum 1 2.2 1.9 79,357 26,829 90 0.09 Platinum 2 2.7 2.8 51,322 26,215 82 0.07 Mercury 3.0 3.0 52,782 13,619 99 0.12

written by J. K. Mohan Rao (Frederick Cancer Research asymmetric unit was extensively rebuilt and the noncrystal- Institute) did improve the map. An estimate of how the two lographic rotation matrix was used to generate molecule 2. molecules are related in the asymmetric unit was made from The model was subjected to 10 more rounds of restrained the relationship of heavy atom sites. least-squares refinement, giving an R factor of 0.40. The averaged MIR map allowed for a preliminary chain At this stage simulated annealing refinement using X-PLOR tracing by using the published sequence (22). It became (29) was used. One round of simulated annealing using the evident during this stage that the overall fold of LiP closely slow-cooling protocol recommended by Brunger (30) lowered follows that of cytochrome c peroxidase (CCP), the only R from 0.42 to 0.27. The procedure was repeated with a other heme peroxidase for which a high-resolution structure starting temperature of4000 K instead of3000 K, followed by is available (23) even though the sequence homology is <20o refinement ofan overall isotropic temperature factor. X-PLOR (24). Nevertheless, once it became clear that the helices in was most useful in fitting molecule 2 in the asymmetric unit both structures were similarly arranged, model building and and in obtaining a better orientation matrix that relates refinement proceeded more smoothly. The model was sub- molecule 2 to molecule 1. As a result, model building became jected to several rounds of restrained least-squares refine- less time consuming since only one molecule need be fit to the ment using the fast Fourier transform version of PROLSQ (25, electron density map, with the second generated by the 26). Sim weighting (27) was used to combine the probability orientation matrix, and X-PLOR was used to "force" a better distributions ofthe MIR phases and calculated phases (28). A fit for molecule 2. The overall major advantage in using new map using figure-of-merit weighted Fos and combined X-PLOR was in the reduction of time spent fitting to the centroid phases then was used for the next round of model electron density map. building. In the early stages ofphase combination, only those The final R factor is 0.25 for 19,356 reflections between 8 reflections for which there were MIR phases were combined. and 2.6 A and .-2a. Individual temperature factors have not This procedure was continued for four rounds of model been refined nor has solvent been included in the refinement. building, refinement, and phase combination. Nevertheless, The "annealed" model exhibited excellent geometry with refinement did not converge and several regions remained deviations ofbond distances of 0.027 A and the resulting 2FO difficult to interpret. In an attempt to improve the initial MIR - Fc map was ofexcellent quality. The quality ofthe electron map, derivative-parent difference Fouriers were reexamined density of the molecule 1 heme is shown in Fig. 1. by using calculated phases based on the model, which included -80%o of the expected protein atoms. As antici- RESULTS AND DISCUSSION pated, these maps confirmed the location ofheavy atom sites Molecular models of CCP and LiP shown in Figs. 2 and 3 but also indicated two more sites in the uranyl 1 and three in illustrate the striking similarities between the two peroxi- the mercury derivatives. The occupancy of the new uranyl dases despite the fact that the sequence identity is <20%o (24). site was -0.28 after refinement to be compared with the two Most striking is the similar arrangement of 11 helical seg- main sites, which exhibited occupancies of0.53 and 0.56. The ments and a limited amount of, structure in the proximal new mercury sites also yielded refined occupancies about (lower) domain. The extra residues in LiP not present in CCP half that of the two main mercury sites. MIR refinement was extend from an extra C-terminal segment that traverses over repeated with the PHASES package of programs** with these the surface ofthe enzyme with no regular extended elements additional sites leading to a higher figure of merit at 2.6 A resolution (Table 2). This MIR map was considerably better Table 2. Heavy atom refinement and phasing than the original map. The newer MIR map was subjected to Phase Phase power solvent leveling with a 30% solvent cutoff resulting in an No. of No. of power at maximum average change in acentric phase of230. The leveled map was Derivative sites reflections overall resolution next averaged about the noncrystallographic twofold axis. RClijS For refinement of the noncrystallographic rotation/ Uranyl 1 8 8,331 0.59 1.78 1.45 at 2.6 A translation matrix, a 20-A box ofdensity centered on the iron Uranyl 2 6 13,739 0.55 1.83 1.50 at 2.7 A atoms was used since this region represented the most highly Platinum 1 2 15,502 0.64 1.46 1.21 at 2.7 A ordered and clearest parts of the MIR map. The correlation Platinum 2 2 12,900 0.68 1.44 1.34 at 2.9 A coefficient after refinement was 0.56. Mercury 8 10,118 0.64 1.49 1.11 at 3.oA The resulting orientation matrix then was used to average Derivatives were prepared by soaking crystals in U03(NO3)2, volumes of densities corresponding to both LiP molecules. Pt(NH3)(NO2)CI2, or Hg(CH3CO2)2. Two data sets were obtained for The model was rebuilt using the new MIR, leveled/averaged, the uranyl and platinum soaks and were treated as separate deriva- and combined phased maps. Despite the low solvent cutoff tives. Reflections were accepted for MIR phasing only ifat least four due to the relatively low solvent content of the crystals, derivatives contributed and had amplitudes at least 6 times above background level (i.e., .6of) to a maximum resolution of2.6 A for all solvent leveling did improve the map and allowed for suc- derivatives. The final figure of merit for 10,287 phased reflections to cessful interpretation of difficult regions. Molecule 1 in the 2.6 A was 0.68. RcuuiS, R factor for centric reflections, XIIFpHobsI - Fpobs + FHcalcII/XjjFpHobs - Fpobsl. Phase power, mean heavy **Furey, W. & Swaminathan, S., 14th American Crystallography atom structure factor amplitude divided by mean residual lack of Association Meeting, 1990, New Orleans, abstr. PA33. closure error, I(FH)/I(FpHobs - FpHcalc). Downloaded by guest on October 1, 2021 752 Biochemistry: Edwards et al. Proc. Natl. Acad Sci. USA 90 (1993)

FIG. 2. LiP and CCP molecules with helices colored red. Disul- fide bridges in LiP are indicated and are paired as follows: 3:15, 14:285, 34:120, and 249:317. MIR map of secondary structure. The last 6 residues have not been included in the current LiP model. LiP differs from CCP in having 4 disulfide bridges (Fig. 2). In both peroxidases, the heme is sandwiched between helices B (distal) and F (proximal; Fig. 3). In both , one heme edge is situated at the bottom of a crevice formed by the surfaces of both domains. Nevertheless, in LiP this crevice is smaller, owing to side-chain interactions, whereas the same pocket in CCP is more open. This structural difference was anticipated from the work ofDePillis et al. (31, 32), who found that the heme edge is available for modifica- tion with aryl hydrazines in CCP (31) but not in LiP (32). The LiP structure now shows that the meso position of the heme facing the open end of the crevice, which is attacked by hydrazines in other peroxidases, is not readily accessible in LiP, whereas in CCP this region remains open. This also raises an interesting question concerning how substrates such as veratryl alcohol interact with LiP. This small opening connecting the surface of the enzyme to the distal pocket is the best candidate. 2Fo-Fc omit map The distal pocket is very similar in both proteins. Each contains a distal histidine and arginine, which have been postulated to operate in the formation of CCP compound I. F0 - F, difference maps clearly indicate the presence of water molecules. One of these waters sits directly over the heme iron atom (Fig. 1C). The average distance of this water from the iron atom in the two LiP molecules in the asymmetric unit is 2.5 A compared to 2.4 A in CCP. This distance is consistent with a variety of spectroscopic studies, which indicate that, like CCP, LiP is predominantly high spin and pentacoordinate (33-35). The proximal pocket also is very similar in that in both LiP and CCP the proximal histidine ligand is hydrogen bonded to a buried aspartic acid side chain. We have postulated elsewhere that this hydrogen bond imparts a greater anionic character to the histidine ligand than in other heme proteins, such as the globins, and therefore provides additional stability to the ferryl (Fe4+) iron in compound I (36, 37). Overall, these similarities support the view that the detailed mechanism of compound I formation proposed for CCP (36, 37) operates in LiP as well. The most striking difference in the active site was antici- pated from sequence comparisons-that is, the substitution of tryptophan residues in both the proximal and distal pock- ets of CCP with phenylalanine in LiP (Fig. 4). This probably FIG. 1. Fit of molecule 1 heme to electron density before (A) and explains in part why LiP compound I has a porphyrin ircation after (B) refinement. Both the MIR and 2F. - Fc maps were radical and in CCP compound I the radical is centered on an contoured at la, to 2.6 A. Heme was excluded from the phase amino acid side chain. Moreover, CCP is relatively rich in calculation in generating the 2F. - Fc omit map. (C) Heme is viewed edge on with the 2F. - Fc map (dashed lines) contoured at la, and amino that can be oxidized (7 tryptophans, 14 tyrosines, the F. - Fc map (solid lines) contoured at 3cr. The large lobe of 5 methionines, 1 cysteine), whereas LiP has no tyrosines, 3 positive difference density clearly indicates the presence of water tryptophans, 8 methionines, and no free cysteines. Only directly over the iron atom. The F. - F_ map was generated with the Met-172 actually contacts the heme in LiP, while both Trp-51 heme included in the structure factor calculation. and -191 contact the heme in CCP. Downloaded by guest on October 1, 2021 Biochemistry: Edwards et al. Proc. Natl. Acad. Sci. USA 90 (1993) 753

LIP LIP

FIG. 3. Stereo Ca backbone models of LiP (Upper) and CCP (Lower). Helices are labeled A-J according to the labeling scheme used earlier for CCP (22). Note that one ofthe helices is labeled b' to be consistent with the CCP no- menclature. Because helix G is somewhat obscured in this view, this helix is not labeled. This view is looking down into the open channel, which connects the sur- face ofthe enzymes with the heme CCP CCP distal pocket. LiP also exhibits some unusual properties not shared by within the heme pocket, although, as with CCP, the proximal CCP, including the types ofsubstrates oxidized by compound histidine ligand hydrogen bonds with a buried aspartic acid I. LiP can oxidize nonphenolic aromatics such as veratryl side chain (Fig. 4). The distal helix (helix B) does contain a alcohol (3,4-dimethoxybenzyl alcohol) and, although it is not highly conserved aspartic acid (Asp-48; refs. 22 and 24), known whether LiP directly attacks lignin, LiP very likely is which immediately follows the distal His-47, yet the Asp-48 directly involved in lignin degradation (2, 3, 20). In sharp side chain points away from the active site and does not contrast, CCP is not very effective at oxidizing small aro- interact with the distal His-47. However, LiP does have an matic molecules and appears to be the only peroxidase in this unusual carboxylate-carboxylate hydrogen bond not present group to be specifically designed to interact with another in CCP, which might lead to some subtle conformational macromolecule, cytochrome c. LiP also exhibits a very low and/or stability changes that could indirectly affect electron pH maximum (near pH 3), which controls the reduction but transfer from or binding to small substrates as the pH is not the formation of compound 1 (14, 17, 38). Such a low pK raised. One heme propionate is hydrogen bonded to a surface for activity would suggest a side chain aspartic acid residue (Asp-183; Fig. 4), whereas in CCP the somehow participating in the reaction (17, 39). While we analogous propionate is hydrogen bonded to Asn-184. We cannot rule out this possibility, there are no carboxylates also note that aspartic acid at this position is unique to the LiP

His52 His47 \ ; nrg48X 184 ~~ / ~~~Asnli84 Arg43

FIG. 4. Comparison ofLiP and Asp 183 CCP active sites. Note that where LiP has phenylalanine residues contacting the heme and proximal histidine ligand, CCP has tryp- tophan. Note, too, that where His176 CCP has Asn-184 hydrogen bond- Phe 193 ing with a heme propionate, Asp- 183 serves this role in LiP. This may explain in part the low pH optimum of LiP since disruption ofthe aspartic acid-propionate hy- CCP drogen bond would be expected to LIP destabilize the heme pocket. Downloaded by guest on October 1, 2021 754 Biochemistry: Edwards et al. Proc. Natl. Acad. Sci. USA 90 (1993) family of peroxidases (24). Disruption of this hydrogen bond 12. Valli, K., Wariishi, H. & Gold, M. H. (1992) J. Bacteriol. 174, at elevated pH could adversely affect activity by causing 2131-2137. 13. Renganathan, V. & Gold, M. H. (1986) Biochemistry 25, 1626- structural perturbations. Such perturbations, however, could 1631. not be so large as to alter compound I formation since the rate 14. Marquez, L., Wariishi, H., Dunford, H. B. & Gold, M. H. of compound I formation is, as with CCP, insensitive to pH (1988) J. Biol. Chem. 263, 10549-10552. from pH 3-8 (14, 38, 40). Alternatively, the pH dependence 15. Tien, M., Kirk, T. K., Bull, C. & Fee, J. A. (1986) J. Biol. could have as much to do with the control of the LiP Chem. 261, 1687-1693. compound I redox potential. In both CCP (40) and horserad- 16. Renganathan, V., Miki, K. & Gold, M. H. (1987) Biochemistry 26, 5127-5132. ish peroxidase (41), the redox potential of compound I 17. Wariishi, H., Huang, J., Dunford, H. B. & Gold, M. H. (1991) increases with decreasing pH making compound I a better J. Biol. Chem. 266, 20694-20699. oxidizing agent at low pH. If the same pattern holds for LiP 18. Valli, K., Wariishi, H. & Gold, M. H. (1990) Biochemistry 29, compound I, then the low pH optimum for LiP may simply 8535-8539. reflect the higher redox potential needed to oxidize nonphe- 19. Wariishi, H. & Gold, M. H. (1990) J. Biol. Chem. 265, 2070- nolic substrates. 2077. 20. Hammel, K. E. & Moen, M. A. (1991) Enzyme Microb. Tech- As this manuscript was being reviewed, the refinement of nol. 13, 15-18. LiP continued and the current R factor is 0.15 at 2 A. While 21. Bricogne, G. (1976) Acta Crystallogr. Sect. A 32, 832-845. many more details are now visible, the primary conclusions 22. Ritch, T. G., Jr., Nipper, V. J., Akileswaran, L., Smith, A., based on the 2.6-A structure described in this paper are Pribnow, D. & Gold, M. H. (1991) Gene 107, 119-126. consistent with the higher-resolution refined structure. A 23. Finzel, B. C., Poulos, T. L. & Kraut, J. (1984) J. Biol. Chem. detailed description of the higher-resolution model will ap- 259, 13027-13036. 24. Henrissat, B., Saloheimo, M., Lavaitte, S. & Knowles, pear elsewhere (42). J. K. C. (1990) Proteins 8, 251-257. 25. Hendrickson, W. A. & Konnert, J. H. (1980) in Computing in We are indebted to Dr. J. K. Mohann Rao (Frederick Cancer Crystallography, eds. Diamond, R., Ramaseshan, S. & Ven- Research Institute) for providing us with the suite of programs for katesan, K. (Indian Inst. of Sci., Bangalore), pp. 13.01-13.23. noncrystallographic averaging and to Michael Tung for adapting the 26. Finzel, B. C. (1987) J. Appl. Crystallogr. 20, 53-55. software to be compatible with the PHASES package of crystallo- 27. Sim, G. A. (1960) Acta Crystallogr. 5, 535-542. graphic programs and local workstations. This work was supported 28. Hendrickson, W. A. & Lattman, E. E. (1970)Acta Crystallogr. in part by National Science Foundation Grants DMB 9104960 Sect. B 26, 136-143. (T.L.P.) and DMB 890438 (M.H.G.) and by Grant DE-FG-08-87-ER 29. Brunger, A. T., Krukowski, A. & Erickson, J. W. (1989) Acta 13715 (M.H.G.) from the Division of Energy Biosciences, U.S. Crystallogr. Sect. A 46, 585-593. Department of Energy. 30. Brunger, A. T. & Krukowski, A. (1990) Acta Crystallogr. Sect. A 46, 585-593. 1. K. V. & C. H. (1971) : Occurrence, 31. DePillis, G. D., Sishta, B. P., Mauk, A. G. & Ortiz de Mon- Sarkanen, Ludwig, tellano, P. R. (1991) J. Biol. Chem. 266, 19334-19341. Formation, Structure, and Reactions (Wiley-Interscience, 32. DePillis, G. D., Wariishi, H., Gold, M. H. & Ortiz de Montel- New York). lano, P. R. (1990) Arch. Biochem. Biophys. 280, 217-223. 2. Gold, M. H., Wariishi, H. & Valli, K. (1989) in Biocatalysis 33. Andersson, L. A., Renganathan, V., Chiu, A. A., Loehr, in Agricultural Biotechnology, ACS Symposium Series, eds. T. M. & Gold, M. H. (1985) J. Biol. Chem. 260, 6080-6087. Whitaker, J. R. & Sonnet, P. (Am. Chem. Soc., Washington), 34. Kuila, D., Tien, M., Fee, J. A. & Ondrias, M. R. (1985) Vol. 389, pp. 127-140. Biochemistry 24, 3394-3397. 3. Kirk, T. K. & Farrell, R. L. (1987) Annu. Rev. Microbiol. 41, 35. Andersson, L. A., Renganathan, V., Loehr, T. M. & Gold, 465-505. M. H. (1987) Biochemistry 26, 2258-2263. 4. Buswell, J. A. & Odier, E. (1987) CRC Crit. Rev. Biotechnol. 36. Poulos, T. L. & Finzel, B. C. (1984) in Peptide and Protein 6, 1-60. Review (Dekker, New York), Vol. 4, pp. 115-177. 5. Renganathan, V., Miki, K. & Gold, M. H. (1985) Arch. 37. Poulos, T. L. (1987) Adv. Inorgan. Biochem. 7, 1-36. Biochem. Biophys. 241, 304-314. 38. Andrawis, A., Johnson, K. A. & Tien, M. (1988) J. Biol. Chem. 6. Leisola, M. S. A., Kozulic, B., Meussdoerfer, F. & Fiechter, 263, 1195-1198. A. (1987) J. Biol. Chem. 262, 419-424. 39. Cai, D. & Tien, M. (1991) J. Biol. Chem. 266, 14464-14469. 7. Higuchi, T. (1990) Wood Sci. Technol. 24, 23-63. 40. Purcell, W. L. & Erman, J. E. (1976) J. Am. Chem. Soc. 98, 8. Shoemaker, H. E. (1990) Recl. Trav. Chim. Pays-Bas 109, 7033-7037. 255-272. 41. Hayashi, Y. & Yamazaki, I. (1979) J. Biol. Chem. 254, 9101- 9. Miki, K., Kondo, R., Renganathan, V., Mayfield, M. & Gold, 9106. M. H. (1988) Biochemistry 27, 4787-4794. 42. Bernstein, F. C., Koetzle, T. F., Williams, G. J. B., Meyer, 10. Hammel, K. E. (1989) Enzyme Microb. Technol. 11, 776-777. E. F., Jr., Brice, M. D., Rogers, J. R., Kennard, O., Shima- 11. Valli, K. & Gold, M. H. (1991) J. Bacteriol. 173, 345-352. nouchi, T. & Tasumi, M. (1977) J. Mol. Biol. 112, 535-542. Downloaded by guest on October 1, 2021