J. Mol. Biol. (1991) 217, 337-352
Structure of Rubredoxtn from Desulfovibrio vulgaris at l-5 A Resolution Elinor T. Adman?, Larry C. Sieker and Lyle H. Jensen Department
of Biological Structure, SM-20 and Department of Biochemistry University of Washington, Seattle, WA 98195, U.S.A. (Received
25 May
1990; accepted 6 August
1990)
The X-ray model of rubredoxin from Desulfovibrio vulgaris has been refined against 15 A X-ray diffraction data collected on a diffractometer. The final model comprises 395 nonhydrogen protein atoms, and 180 solvent 0 atoms. The final R-value for the model with calculated H atom positions included as fixed contributions is 9098 over all reflections greater than 20, from infinity to 1.5 A. The error in co-ordinates is estimated to be 908 A. The solvent model was twice redetermined during the later stages of refinement and was instrumental in its success. One sequence error has been detected and corrected (Thr21 + Asp). The iron-sulfur site bond angles are distorted from true tetrahedral symmetry, as found in other rubredoxin structures. A significant deviation from tetrahedral angles is seen at C” atoms 9, 10, 42 and 43, interior angles of the loops binding the iron atom. The planes of two aromatic groups, Tyr4 and Trp37, are nearly parallel to, and lie under, an extended system of atoms that includes the peptide bonds preceding the first cysteine residue of each cysteine loop as well as the cysteine side-chain, the iron, and the cysteine side-chain of the opposite loop, forming a previously unrecognized extended system that may function in electron transfer.
reported (Adman et al., 1977). Additional data have been collected to 1.5 A, and the model refined, reaching an overall agreement factor, R = @098, for data between infinity and 1.5 A. Details of the refinement and the resulting model are discussed in this paper.
1. Introduction Rubredoxins are low molecular weight non-heme iron proteins that are believed to function in electron transfer. They have been isolated from several sulfate-reducing bacteria (Desulfovibrio) and some aerobic bacteria. The X-ray structures of four of these have been reported; Clostridium pasteurianum (RdCpS, 54 amino acid residues, 1.2 A resolution (1 A = 01 nm); Watenpaugh et al., 1979) Desulfovibrio vulgaris (RdDv, 52 amino acid residues, 2.0 A; Adman et al., 1977), Desulfovibrio gigas (RdDg, 52 amino acid residues, 1.4 A; Frey et al., 1987) and Desulfovibrio sulfuricans (RdDd, 45 amino acid residues, 1.5 A; Sieker et al., 1986; Stenkamp et al., 1990). Comparisons of these structures have shown only small structural differences, except for the last, in which an entire loop is missing. Comparisons of these structures at high resolution reveals small, but significant, deviations from expected bond angles. The solution and preliminary refinement of the structure of RdDv at 2.0 A resolution has been
2. Experimental Procedures (a) Diffractometer data
RdDv crystallizes in space group P2,, with cell dimensions a = 19993 A, b = 41.505 A, c = 24404 A, fl= 107.6” (Adman et al., 1977). A crystal of dimensions 92 mm x 92 mm x 92 mm was mounted in a thin-walled capillary with crystallization buffer on either side, and data collected as described (Adman et al., 1977). The X-ray source was a sealed Cu target tube operating at 40 kV, 30 mA. The Picker FACS-1 diffractometer was set to operate in the w/20 mode at S”/min in 20 and backgrounds were measured for 10 s at both limits of the scan range. Data were collected in 2 shells. 2.1 to 1.5 A, followed by infinity to 2.0 A. Crystal decay was followed by monitoring 8 reflections at intervals of 200 reflections, and 3 others at intervals of 2000 reflections. The maximum decay correction was 1.2. Coincidence and absorption corrections were made (Sletten et al., 1969; North et al., 1968), the latter by measuring the 0 10 0 reflection at x = 90” as a function of 4; the maximum absorption was 1.12. The scale factor to bring the measurements to an absolute scale was 1.19, and the overall B estimated to be
t Author to whom all correspondence should be sent. $ Abbreviations used: Rd, rubredoxin; Cp, Clostridium pasteurianum; Dv, Desuljovibrio vulgaris; Dg, Desulfovibrio gigas; Dd, Desulfovibrio sulfwicans; r.m.s.. root-mean-square. 0022-2836/91/020337-16$03.09/O
337
0 1991 Academic Press Limited
E. T. Adman
338
et al.
Table 1 Data cdection statistics Diffractometert DUti#l
npred
n,,,,
590 300 250 260 1.75 1.62 1.55 150
200 667 578 1324 1336 1039 689 539
199 667 578 1324 1336 1039 689 96
Area detector n > 2a
l$
Area detector q
(Flu)
seas
?l > 20
39 39 37 35 29 23 18 16
166 611 490 836 17
166 603 486 822 17
99 99 64 48 45
196 661 573 1301 1276 979 627 86
%lc..
175 624 522 945 187
n>2a 172 617 515 927 177
Structure of Rubredoxtn from Desulfovibrio vulgaris at l-5 A Resolution Elinor T. Adman?, Larry C. Sieker and Lyle H. Jensen Department
of Biological Structure, SM-20 and Department of Biochemistry University of Washington, Seattle, WA 98195, U.S.A. (Received
25 May
1990; accepted 6 August
1990)
The X-ray model of rubredoxin from Desulfovibrio vulgaris has been refined against 15 A X-ray diffraction data collected on a diffractometer. The final model comprises 395 nonhydrogen protein atoms, and 180 solvent 0 atoms. The final R-value for the model with calculated H atom positions included as fixed contributions is 9098 over all reflections greater than 20, from infinity to 1.5 A. The error in co-ordinates is estimated to be 908 A. The solvent model was twice redetermined during the later stages of refinement and was instrumental in its success. One sequence error has been detected and corrected (Thr21 + Asp). The iron-sulfur site bond angles are distorted from true tetrahedral symmetry, as found in other rubredoxin structures. A significant deviation from tetrahedral angles is seen at C” atoms 9, 10, 42 and 43, interior angles of the loops binding the iron atom. The planes of two aromatic groups, Tyr4 and Trp37, are nearly parallel to, and lie under, an extended system of atoms that includes the peptide bonds preceding the first cysteine residue of each cysteine loop as well as the cysteine side-chain, the iron, and the cysteine side-chain of the opposite loop, forming a previously unrecognized extended system that may function in electron transfer.
reported (Adman et al., 1977). Additional data have been collected to 1.5 A, and the model refined, reaching an overall agreement factor, R = @098, for data between infinity and 1.5 A. Details of the refinement and the resulting model are discussed in this paper.
1. Introduction Rubredoxins are low molecular weight non-heme iron proteins that are believed to function in electron transfer. They have been isolated from several sulfate-reducing bacteria (Desulfovibrio) and some aerobic bacteria. The X-ray structures of four of these have been reported; Clostridium pasteurianum (RdCpS, 54 amino acid residues, 1.2 A resolution (1 A = 01 nm); Watenpaugh et al., 1979) Desulfovibrio vulgaris (RdDv, 52 amino acid residues, 2.0 A; Adman et al., 1977), Desulfovibrio gigas (RdDg, 52 amino acid residues, 1.4 A; Frey et al., 1987) and Desulfovibrio sulfuricans (RdDd, 45 amino acid residues, 1.5 A; Sieker et al., 1986; Stenkamp et al., 1990). Comparisons of these structures have shown only small structural differences, except for the last, in which an entire loop is missing. Comparisons of these structures at high resolution reveals small, but significant, deviations from expected bond angles. The solution and preliminary refinement of the structure of RdDv at 2.0 A resolution has been
2. Experimental Procedures (a) Diffractometer data
RdDv crystallizes in space group P2,, with cell dimensions a = 19993 A, b = 41.505 A, c = 24404 A, fl= 107.6” (Adman et al., 1977). A crystal of dimensions 92 mm x 92 mm x 92 mm was mounted in a thin-walled capillary with crystallization buffer on either side, and data collected as described (Adman et al., 1977). The X-ray source was a sealed Cu target tube operating at 40 kV, 30 mA. The Picker FACS-1 diffractometer was set to operate in the w/20 mode at S”/min in 20 and backgrounds were measured for 10 s at both limits of the scan range. Data were collected in 2 shells. 2.1 to 1.5 A, followed by infinity to 2.0 A. Crystal decay was followed by monitoring 8 reflections at intervals of 200 reflections, and 3 others at intervals of 2000 reflections. The maximum decay correction was 1.2. Coincidence and absorption corrections were made (Sletten et al., 1969; North et al., 1968), the latter by measuring the 0 10 0 reflection at x = 90” as a function of 4; the maximum absorption was 1.12. The scale factor to bring the measurements to an absolute scale was 1.19, and the overall B estimated to be
t Author to whom all correspondence should be sent. $ Abbreviations used: Rd, rubredoxin; Cp, Clostridium pasteurianum; Dv, Desuljovibrio vulgaris; Dg, Desulfovibrio gigas; Dd, Desulfovibrio sulfwicans; r.m.s.. root-mean-square. 0022-2836/91/020337-16$03.09/O
337
0 1991 Academic Press Limited
E. T. Adman
338
et al.
Table 1 Data cdection statistics Diffractometert DUti#l
npred
n,,,,
590 300 250 260 1.75 1.62 1.55 150
200 667 578 1324 1336 1039 689 539
199 667 578 1324 1336 1039 689 96
Area detector n > 2a
l$
Area detector q
(Flu)
seas
?l > 20
39 39 37 35 29 23 18 16
166 611 490 836 17
166 603 486 822 17
99 99 64 48 45
196 661 573 1301 1276 979 627 86
%lc..
175 624 522 945 187
n>2a 172 617 515 927 177
