www.sciencemag.org/cgi/content/full/318/5855/1461/DC1

Supporting Online Material for

Carbon Dioxide Activation at the Ni,Fe-Cluster of Anaerobic Carbon Monoxide Dehydrogenase

Jae-Hun Jeoung and Holger Dobbek*

*To whom correspondence should be addressed. E-mail: holger.dobbek@uni-bayreuth.de

Published 30 November 2007, Science 318, 1461 (2007) DOI: 10.1126/science.1148481

This PDF file includes:

Materials and Methods SOM Text Figs. S1 to S6 Tables S1 to S3 References

1

Supporting Online Material for

Carbon dioxide activation at the Ni,Fe-cluster of a carbon monoxide

dehydrogenase

Jae-Hun Jeoung, Holger Dobbek*

*To whom correspondence should be addressed. Email: holger.dobbek@uni-bayreuth.de

Materials and Methods

Cloning, expression and purification

Genomic DNA was isolated from Carboxydothermus hydrogenoformans Z-2901 (DSM 6008)

and the gene (cooS-II) encoding CODHIICh (S1) was amplified by PCR and ligated into a

linearized pET28a vector (Novagen). The resulting pPKS2 plasmid was transformed into E.

coli Rosseta (DE3) harbouring pRKISC (S2). Cultures were grown in a modified TB medium

with addition of 1.5 % (v/v) glycerol, 0.02 mM NiCl2, 0.1 mM Na2S, 0.1 mM FeSO4, and

2 mM cysteine at 30 °C. Expression of recombinant CODHIICh was induced by addition of

0.2 mM IPTG supplemented with 0.5 mM NiCl2, 1 mM Na2S, 1 mM FeSO4, and 50 mM

KNO3 once the culture reached an OD600 of 0.7 - 0.8. Bacteria were cultivated under N2-

fluxing, harvested anaerobically and cell pellets were frozen in liquid N2. All purification

steps were done in an anoxic glove box (model B, COY Laboratory Products Inc., Michigan,

USA) under an atmosphere of 95 % N2/ 5 % H2 at 17 °C, except that cell breaking by

sonication was carried out under N2 atmosphere on ice. CODHIICh from cell lysates was

solubilized with 0.9 % (v/v) deoxycholic acid for 60 min, centrifuged and purified to

homogeneity by one step affinity chromatography on IMAC medium (Ni Sepharose, GE

Healthcare, Sweden) with standard buffer containing 1 mM Na-dithionite. Imidazole was

2

removed by gel filtration (G-25 material, 10 mm Χ 60 mm). CODHIICh was concentrated

using a centrifugal filter unit (Vivaspin 500, Vivascience AG, Germany) and stored at -80 °C.

Activity Measurement

CO oxidation activity was measured in gas tight cuvettes with CO head space (S3). The

protein batch used for crystallization had a specific activity of 12229 units mg-1 (as isolated

protein) at 70 °C.

Crystallization and manipulation of crystals

Crystallization and soaking experiments were carried out in the anoxic glove box. Crystals of

CODHIICh were grown in 15 - 18 % PEG3350, 200 mM ammonium sulfate, 100 mM Bis-Tris

pH 6.5, and 2 mM DT by hanging drop vapor diffusion. Crystals appeared after a few days.

Reduced methyl viologen (MVred; blue, Eh ≈ –430 mV) or resazurin (colorless, Eh ≈ -110 mV)

in reservoir and/or crystallization drop were used as indicators of reducing conditions during

crystallization.

To obtain different redox states, single crystals of CODHIICh were incubated in buffer A

(25 % PEG3350, and 50 mM Tris-HCl pH 8.0) with different redox potentials adjusted with

either 5 mM Ti(III) citrate solution (S4) to -600 mV vs. SHE (standard hydrogen electrode) or

with 7 mM DTT solution to -320 mV vs. SHE. Redox potentials of the soaking solutions were

measured with an EMC 130 redox electrode containing an integrated Ag/AgCl reference

system (Sensortechnik Meinsberg GmbH, Germany).

The reported states were obtained from crystals grown for 10 days. All crystals were washed

three times with the -600 mV solution and remained in this solution for three hours. After this

procedure one crystal was shock frozen in liquid nitrogen giving the -600 mV state. Another

crystal was soaked for 30 min in the -600 mV solution containing 45 mM NaHCO3 after

which it was shock frozen to give the -600 mV+CO2 state. At a pH of 8.0 approximately 1%

3

of the total carbonate concentration is present as dissolved CO2. The -320 mV state was

prepared by transferring one crystal from the -600 mV solution into a solution containing

7 mM DTT and 0.5 mM MVox as electron acceptor for reoxidation; after no further reduction

of MVox to the blue colored MVred was detectable the crystals were incubated in the -320 mV

solution containing 7 mM DTT for 2 hours. Between all steps crystals were washed

extensively for at least three times with the subsequent soaking solution.

The influence of soaking experiments on CODHIICh activity was monitored by measuring the

specific activities of dissolved crystals. After 40 days of crystallization, all crystals from one

24 well plate were treated in the same way as the crystals for structure determination. Specific

activities of dissolved CODHIICh crystals were: as-isolated state, 10999 units mg-1; -600 mV

state, 11317 units mg-1; -320 mV state, 13427 units mg-1; -600 mV+CO2 state,

13808 units mg-1.

All crystals were shock frozen in soaking solutions containing 15 % (v/v) 2R, 3R-butandiol

(Sigma) and were stored in liquid nitrogen until the diffraction experiments were carried out.

Data collection

For each of the reported states at least two independently prepared crystals were measured

and analyzed. The data statistics reported in Table S1 are given for the crystals with highest

diffraction limit. All crystals diffracted to at least 1.9 Å resolution on an X-ray generator with

rotating Cu-anode. Data were collected at 100 K.

Crystals belonged to space group C2 with one monomer in the asymmetric unit and cell

constants of 112, 74, 70 Å and β=111°.

Structure solution and refinement

Models were built with MAIN 2000 (S5) and COOT (S6). Positional and temperature factor

refinements were carried out using Refmac5 (S7). Restrains for the different states of cluster

4

C were generated by iterative cycles of refinement with weak restrains and generation of new

parameters using Refmac5 and SHELX (S8). Individual occupancies of atoms from cluster C

were adapted to show comparable B factors (Table S2). In the final refinement cycles the B

factors of iron and nickel ions were refined anisotropically, alternative conformations of

several side chains and Fe1 were included and restrains for bond length and angles of cluster

C were practically excluded. The final refinement statistics are shown in Table S1. Simulated

annealing omit maps were used to validate the reported structures.

5

Supporting Text

Geometry of cluster C

In all three states small bond angles (84-101°) are found for the ligands around Fe1 (Table S3).

The change from the distorted T-shaped coordination of Ni in the -320 mV and -600 mV state

to square-planar coordination in the Ni-CO2 state is reflected in an opening of the C526Sγ-Ni-

S3 angle by 12°. The alternative position found for Fe1 is termed Fe1B in Table S2. Fe1B is

very close to the Ni ion (2.3-2.4 Å) and is most likely only occupied in a Ni free form of

cluster C. Cluster bond length and angles are similar in all three states (Fig. S1-S4 and Table

S3). The opening of the C526Sγ-Ni-S3 angle is even more obvious when one compares the

[NiFe4S4OHx] cluster with the [NiFe4S5] cluster (PDB-Id: 1SU8) with four sulfur ligands and

a C526Sγ-Ni-S3 angle of 176 ° (Fig. S5). It is interesting to note that in structures of cluster C

containing a bridging ligand like CO2 or the µ-S ligand in the [NiFe4S5] cluster, less structural

heterogeneity is observed for Fe1 with higher occupancies for Fe1A and lower occupancies

for Fe1B. This may be related to the "healing" effect of CO2 (S9) and the higher specific

activities of dissolved crystals after incubation with CO2 (this work). The beneficial effect of a

bridging ligand for Ni,Fe-CODHase activity may be an indication that the µ-S ligand could

have a physiological function in restoring or preserving the integrity of cluster C.

Water molecules near the OHx ligand of cluster C

Catalytic CO oxidation at cluster C depends on the fast replenishment of the H2O/OH- ligand

at Fe1. A network of water molecules has been found in the direct vicinity of the

water/hydroxo ligand (Fig. S6), which may participate in refilling the empty coordination site

at Fe1 once CO2 has been liberated.

6

Figure S1: Stereo views on the active site cluster. (A) -600 mV state, (B) -600 mV+CO2 state

and (C) -320 mV state. 2Fobs-Fcalc maps contoured at 1σ (blue) and omit Fobs-Fcalc maps

contoured at 5σ for the ligand (OHx in A and C, CO2 in B) at Fe1.

7

Figure S2: Bond length (red) and additional distances (blue) of cluster C in the -600 mV state.

8

Figure S3: Bond length (red) and additional distances (blue) of cluster C in the -600

mV+CO2 state. The lengths of both C-O bonds were restrained throughout the refinement.

9

Figure S4: Bond length (red) and additional distances (blue) of cluster C in the -320 mV state.

10

Figure S5: Superposition of the [NiFe4S4OHx] cluster (-320 mV state) in element colors with

the [NiFe4S5] cluster (PDB-code: 1SU8) (S10) in blue. The positions of the µ-S ligand in the

[NiFe4S5] cluster and the H2O/OH- ligand in the [NiFe4S4OHx] cluster are ~0.6 Å apart in the

superposed structures.

11

Figure S6: Water network close to the H2O/OH- ligand of cluster C. Histidines 93, 96 and 99

have been implicated in turnover associated proton transfer (S11, S12).

12

Table S1. Statistics on diffraction data and structure refinement.

Data set -600 mV state (Ti-Citrate)

-320 mV state (re-oxidized)

-600 mV state + CO2

Total / unique refl. 567,684 / 104,078 315,206 / 81,496 304,202 / 83,938

Rsa (%) 7.0 (19.0) 6.0 (33.0) 8.6 (28.1)

Resolution (Å) 30-1.40 (1.45-1.40) 30-1.48 (1.50-1.48) 30-1.50 (1.60-1.50)

Completeness (%) 97.8 (85.8) 90.4 (85.1) 97.1 (94.4)

(I) / (σI) 15.3 (3.2) 13.5 (5.0) 10.0 (4.4)

Model R / Rfree-factor (%) b 15.8 / 18.3

(32.6 / 34.4)

14.6 / 17.8

(24.5 / 31.3)

15.4 / 18.2

(30.3 / 34.6) Rms deviation from ideal

geometry

Bonds (Å) Angles (°)

0.016 1.91

0.017 1.95

0.014 1.95

ESUc (Å) 0.042 0.053 0.056

For the refinement statistics Friedel mates were merged.

a Rs = Σh Σi | Ii(h) - <I(h)>| / Σh Σi I i(h); where i are the independent observations of reflection h.

b The Rfree factor was calculated from 5% of the data, which were removed at random before the

refinement was carried out.

c Estimated overall coordinate error (ESU) based on maximum likelihood.

Values in parentheses are given for the highest resolution shell.

13

Table S2: Analysis of B factors and occupancies in cluster C. The values given are for the -

600 mV state (top line), the -600 mV+CO2 state (middle line) and the -320 mV state (bottom

line).

Atom Ni Fe1(A) Fe1(B) O1 C O2 S5 S3 S1 Av. B in

Cl. B

B factor

(Å2)

10.8

8.3

13.6

10.7

9.7

12.6

8.2

7.3

12.5

10.2

9.7

10.3

8.4

8.1

7.6

7.4

12.6

8.4

7.6

13.1

8.6

6.5

11.8

4.7

5.1

9.0

Occupancy 0.6

0.6

0.6

0.6

0.7

0.6

0.3

0.1

0.3

0.6

0.8

0.6

0.8

0.8

0.8

0.8

0.8

0.8

0.8

0.8

0.8

0.8

0.8

1.0

1.0

1.0

14

Table S3: Angles between selected atoms of cluster C in the three states.

Angle -600 mV state -600 mV/CO2 state -320 mV state

C526Sγ-Ni-S3

C526Sγ-Ni-S5

C526Sγ-Ni-C

O1-C-O2

Ni-C-O1

Ni-C-O2

H261Nε2-Fe1-O1

H261Nε2 -Fe1-S1

Ni-S5-Fe3

Ni-S3-Fe3

C-O1-Fe1

156 °

94 °

-

-

-

-

88 °

100 °

73 °

76 °

-

168 °

93 °

86 °

133 °

119 °

108 °

84 °

96 °

79 °

83 °

105 °

157 °

94 °

-

-

-

-

87 °

101 °

75 °

75 °

-

15

References

S1. M. Wu et al., PLoS Genet 1, e65 (2005).

S2. M. Nakamura, K. Saeki, Y. Takahashi, J. Biochem. (Tokyo) 126, 10-8 (1999).

S3. V. Svetlitchnyi, C. Peschel, G. Acker, O. Meyer, J. .Bacteriol. 183, 5134-5144 (2001).

S4. L. C. Seefeldt, S. A. Ensign, Anal. Biochem. 221, 379-86 (1994).

S5. D. Turk, TU München (1992).

S6. P. Emsley, K. Cowtan, Acta Crystallogr. D Biol. Crystallogr. 60, 2126-32 (2004).

S7. G. N. Murshudov, A. A. Vagin, E. J. Dodson, Acta Crystallogr. D Biol. Crystallogr.

53, 240-55 (1997).

S8. G. M. Sheldrick, T. R. Schneider, Methods in Enzymology 277, 319-343 (1997).

S9. M. Anderson, E., P. Lindahl, A., Biochemistry 35, 8371-8380 (1996).

S10. H. Dobbek, V. Svetlitchnyi, J. Liss, O. Meyer, J. Am. Chem. Soc. 126, 5382-7 (2004).

S11. C. L. Drennan, J. Heo, M. D. Sintchak, E. Schreiter, P. W. Ludden, Proc. Natl. Acad.

Sci. U S A 98, 11973-11978 (2001).

S12. E. J. Kim, J. Feng, M. R. Bramlett, P. A. Lindahl, Biochemistry 43, 5728-34 (2004).