Supporting Information

Files in this Data Supplement:

Supporting Figure 6
Supporting Figure 7
Supporting Text
Supporting Table 1

Fig. 6. Kinetic scheme and concentration profile that apply to the target analysis. (Upper) Kinetic scheme used to describe the spectral evolution of Slr1694 by means of a target analysis. Here k_i is a rate constant, F_i denotes the fractional contribution of the FAD*_i state. F is the yield by which Q₁ interconverts to Q₂ and is set at 50%. (Right Lower) Concentration profiles for FAD*_1–4 (blue), Q₁ (black), Q₂ (magenta), and Slr_RED (orange) as follow from the target analysis of time-resolved data on Slr1694 in H₂O (solid lines) and D₂O (dotted lines).

Fig. 7. Singular value decomposition (SVD) of the matrix of residuals using a model with only one intermediate (Q₁) preceding the product state Slr_RED (top) and with the inclusion of a second intermediate Q₂ (bottom). (a and d) First left singular vector, u_res,1, H₂O in black and D₂O in red. (b and e) First right singular vector, w_res,1. (c and f) First 10 singular values, s_res,1 on a logarithmic scale.

Table 1. Rate constants and fractional contributions resulting from target analysis
of the ultrafast time-resolved data on the Slr1694 BLUF domain in terms of the
kinetic scheme shown in Fig. 6 Upper

The unprimed and primed rate constants refer to Slr1694 in H₂O and D₂O, respectively.

Supporting Text

Detailed Description of the Target Analysis. The global and target analysis techniques used in this work have been extensively reviewed in ref. 1. The kinetic model used in this analysis to disentangle the contribution of the FAD excited state, the intermediate states, and the signaling state consists of seven compartments: four FAD*, intermediate Q₁, intermediate Q₂, and the long-lived signaling state Slr_RED, as indicated in Fig. 6. The decay of FAD* is highly heterogeneous in Slr1694 (note that the stimulated emission from FAD* is gradually decreasing in the first four EADS in the sequential model; Fig. 2A) as it is in the related BLUF protein, AppA, where four FAD* lifetimes were observed with a synchroscan streak camera system (2). This result is very likely related to the presence of conformational substates in the dark, as revealed by NMR spectroscopy (3). The heterogeneity in the singlet excited-state decay of FAD is accounted for by assuming four compartments with different decay times but identical spectra, denoted as FAD^*_1–4. Relaxation from FAD₁^*, FAD₂^*, and FAD₃^* occurs through Q₁, whereas FAD₄^* decays directly to the ground state. From intermediate Q₁, intermediate Q₂ is generated, which in turn forms the signaling state Slr_RED.

A kinetic model with only one intermediate (Q₁) preceding the product-state Slr_RED was first tested. However, the target analysis showed a significant correlated structure in the matrix of residuals, in Slr1694 in H₂O as well as in D₂O. This finding is illustrated by the results from the singular value decomposition (SVD) of the matrix of residuals (1) in Fig. 7. The clear structure in the first left singular vector in Fig. 7a is no longer present in Fig. 7d, justifying the inclusion of a second intermediate Q₂. The rms error of the fit decreased from 0.116 to 0.100 mOD.

The following assumptions in the kinetic scheme have been made: (i) the lifetimes of FAD* in Slr1694 in H₂O and D₂O are identical, as discussed in the main text; (ii) the species-associated difference spectrum (SADS) of the signaling state Slr_RED of Slr1694 in H₂O and D₂O are identical; (iii) the SADS of the intermediate species Q₁ in Slr1694 are identical above 535 nm in H₂O and D₂O and are allowed to vary slightly in the region 420–535 nm; and (iv) same as (iii) for the SADS of Q₂.

The SADS of FAD*, Q₁, and Q₂ look very similar in H₂O and D₂O (compare the solid and dotted lines in Fig. 4 Upper), which demonstrates that the spectral evolution in H₂O and D₂O can be described by the same molecular states, solidifying the applied kinetic model. The rate constants and FAD decay fractions estimated from the target analysis on Slr1694 in H₂O and D₂O are summarized in Table 1. Decay of FAD^*₁, FAD^*₂, FAD^*₃, and FAD^*₄ takes place with time constants of 7, 40, 180, and 210 ps, respectively. The biggest fractional contribution comes from FAD^*₁ (47%), whereas FAD^*₂, FAD^*₃, and FAD^*₄ show 28%, 17%, and 8% contributions, respectively. After the decay of excited flavin, the intermediate Q₁ is formed, which evolves in Q₂ in 6 ps in H₂O and 18 ps in D₂O. The intermediate Q₂ lives for 65 ps in H₂O and 156 ps in D₂O. The kinetic isotope effect for the transition from Q₁ to Q₂ amounts to » 3, whereas for the transition Q₂ to Slr_RED it amounts to » 2.4, demonstrating that these reactive events involve proton transfer(s).

1. van Stokkum, I. H. M., Larsen, D. S. & van Grondelle, R. (2004) Biochim. Biophys. Acta 1657, 82–104.

2. Gauden, M., Yeremenko, S., Laan, W., van Stokkum, I. H. M., Ihalainen, J. A., van Grondelle, R., Hellingwerf, K. J. & Kennis, J. T. M. (2005) Biochemistry 44, 3653–3662.

3. Grinstead, J. S., Hsu, S. T. D., Laan, W., Bonvin, A. M. J. J., Hellingwerf, K. J., Boelens, R. & Kaptein, R. (2006) ChemBioChem 7, 187–193.