Apparent cooperativity in the folding of multidomain proteins depends on the relative rates of folding of the constituent domains -- Batey and Clarke 103 (48): 18113 Data Supplement - HTML Page - index.htslp -- Proceedings of the National Academy of Sciences

Batey et al. 10.1073/pnas.0604580103.

Supporting Information

Files in this Data Supplement:

Supporting Text
Supporting Figure 7
Supporting Figure 8
Supporting Figure 9
Supporting Figure 10

Supporting Figure 7

Fig. 7. Equilibrium denaturation curves of R1617 under conditions where cooperativity is lost. The unfolding of R1617 wild-type (black), the unfolding of R1617 at 45°C (red), the unfolding of the F117L mutant of R1617 (blue) and the unfolding of R1617 in 0.5 M Na₂SO₄ (purple), (solid circles, fluorescence; open circles, CD). Loss of cooperativity results in a decrease in the equilibrium m-value and, in the cases shown here, non-coincidence in fluorescence and CD unfolding data. Data taken from ref. 1

1. Batey S, Randles LG, Steward A, Clarke J (2005) J Mol Biol 349:1045-1059.

Supporting Figure 8

Fig. 8. Kinetic properties of R1516. (a) Plot of the natural logarithm of the observed rate constants from single jump experiments of R1516. The only observable unfolding phase and the major refolding phase are shown in black, the refolding phase with 30% of the amplitude is shown in blue and the isomerisation-limited folding phase shown in red (fluorescence, filled circles; CD, open circles). (b) Time course of the fluorescence amplitudes of the observed unfolding rate constants from interrupted refolding experiments. For discussion, see Supporting text. (c) The urea dependence of the unfolding of the folding intermediate. The rate constants observable by single jump are shown as open circles. The two observable unfolding rates after the protein was allowed to refold for 60 ms are shown as filled circles.

Supporting Figure 9

Fig. 9. Modeled data showing the effect of cooperativity on equilibrium m values for a hypothetical two-domain protein. Each domain has been assigned an m value of 1.85 kcal mol^-1 M^-1. Black, red and blue; the two domains unfold entirely independently, and an intermediate with one domain denatured (D) and one domain folded (N) is formed: . Black: the [urea]_50% values are very different (2 and 4 M), 2 separate transitions are seen, each has an m-value of 1.85 kcal mol^-1 M^-1. Blue: the [urea]_50% values are different but close (3.25 and 4 M). A single transition is observed with a lower apparent m-value than that of the constituent domains alone (1.48 kcal mol^-1 M^-1). Red: the 2 domains have identical [urea]_50% values (4 M). There is a single transition with an apparent m-value of 1.85 kcal mol^-1 M^-1. Green: the 2 domains unfold as a single unit in an all-or-none fashion and no intermediate accumulates: . There is a single transition with an m-value which is double that of the constituent domains (3.7 kcal mol^-1 M^-1).

Supporting Figure 10

Fig. 10. Alignment of R15, R16, R17. The residues in bold are the 106 residue spectrin repeats as defined by Pascual et al. (1). The N- and C-terminal extensions as used in this study are shown in plain text. The R1516 and R1617 constructs also contain the N- and C-terminal extensions.

1. Pascual J, Pfuhl M, Walther D, Saraste M, Nilges M (1997) J Mol Biol 273:740-751.

Supporting Text

Kinetic studies of R1516. Assigning the kinetic phases.

The folding and unfolding of R1516 was followed by fluorescence and CD spectroscopy. Fluorescence and CD stopped-flow experiments gave the same rate constants. The refolding data of R1516 fit to three exponential phases (Fig. 8a). The faster of the three phases had an amplitude of ~30% of the total, the middle phase had an amplitude of ~60%, and the slower phase ~10%, when followed by fluorescence. When followed by the change in CD, the two fastest rate constants have between 40% and 50% of the amplitude each and the slowest rate has less than 10%. The unfolding data of R1516 fitted to a single exponential equation at all concentrations of urea.

Double-jump interrupted refolding experiments (D® N® D) (1, 2) were used to investigate the three refolding phases. Unfolded protein (in 8 M urea) was refolded in 4 M urea for a variety of delay times (50 ms to 100 s) before unfolding in 6 M urea, this final unfolding reaction was followed and the amplitudes plotted (Fig. 8b). The unfolding data were globally fit and at all delay times the data were well described by a double exponential process with rate constants of 17 s^-1 and 0.9 s^-1. In single jump unfolding at 6 M only one unfolding phase is observed, with a rate constant of 1 s^-1, therefore the phase with the unfolding rate of 17 s^-1 must be due to the unfolding of a species that we cannot observe in single jump experiments. The rate of appearance of each of the unfolding phases (shown by a plot of delay time vs. amplitude) shows the rate at which the species which is unfolding has been formed (at 4 M urea). The species which unfolds in the double-jump experiments with a rate constant of 0.9 s^-1, appears in two phases. The first stage has an apparent rate constant of 1 s^-1 and the second an apparent rate constant of 6 ´ 10^-3 s^-1. Therefore, this species folds in two stages. At 4 M urea in single jump experiments the observed folding rates are 0.9 s^-1 and 7 ´ 10^-3 s^-1. Thus the two slower refolding rate constants are the formation of the same species.

The species which is only observed in double jump experiments and unfolds at 17 s^-1 has a maximum amplitude at the shortest delay time (~40 ms) and disappears as the delay times increase (Fig. 8b). This species cannot be seen to unfold in single jump experiments. This is characteristic of the presence of a refolding intermediate. To investigate this refolding intermediate further the urea dependence of its unfolding was studied. Unfolded protein (in 8 M urea) was refolded at 4 M urea for 60 ms (where there is a maximum concentration of the intermediate) before unfolding in a range of urea concentrations. The logarithm of this unfolding rate constant of the intermediate has linear dependence on urea concentration as shown in Fig. 8c.

Assignment of the R1516 folding pathway
The minor, slow folding rate is due to proline isomerization.
A combination of single and double-jump kinetic experiments reveal three refolding rate constants for R1516. R1516 has one proline in the R16 domain. This proline has been shown to result in a low amplitude isomerization-limited folding phase in both R16 (3) and R1617 (4). Interrupted unfolding experiments (data not shown) show that slowest of the three rate constants in R1516 can also be assigned to isomerization-limited folding. This phase is now excluded from further discussion. The fast folding rate due to the folding of R15
. R15 has only one tryptophan whereas R16 has two. As the change in fluorescence is due to the change in environment of aromatic residues (mainly tryptophan) the folding of R16 would be expected to have twice the amplitude of R15 when followed by fluorescence. The fast folding rate has 30% of the total amplitude and the middle rate 60%. This indicates that the fast rate is the folding of R15 and the middle rate the folding of R16. When following the folding by the change in secondary structure (CD) the fast and middle rate have approximately the same change in amplitude as would be expected for two domains of the same size with the same structure. To confirm this assignment we have made mutations in R1516. Proteins with a mutation in the R15 domain of R1516 lead to a change in the fast folding phases and proteins with mutation in the R16 domain affect only the slower phases. Assignments of the unfolding rates.
Mutant studies were used to confirm the identity of the unfolding phases. Mutations in R15 of R1516 affected only the fast unfolding rate specific to double jump experiments and mutation in the R16 domain of R1516 affect only the slow single jump phase. Therefore, the phase observable in single jump experiments is the unfolding of R16 and the faster double jump only phase is the unfolding of R15. It is important to note that in the first stage of the double jump experiments only R15 has had time to fold, therefore in the unfolding we are observing R15 unfolding in the presence of unfolded R16.

We cannot observe the unfolding of R15 in the presence of folded R16. This must mean that folded R16 slows the folding of R15 still more - the unfolding must be slower than the unfolding of the R16 domain. Now, once R16 has unfolded, this extra stabilization is lost and R15 unfolds at a fast rate.

1. Kiefhaber T (1995) Proc Natl Acad Sci USA 92:9029-9033.

2. Mayr LM, Odefey C, Schutkowski M, Schmid FX (1996) Biochemistry 35:5550-5561.

3. Scott KA, Batey S, Hooton KA, Clarke J (2004) J Mol Biol 344:195-205.

4. Batey S, Scott KA, Clarke J (2006) Biophys J 90:2120-2130.