Method-Unifying View of Loop-Formation Kinetics in Peptide and Protein Folding

Abstract

Protein folding can be described as a probabilistic succession of events in which the peptide chain forms loops closed by specific amino acid residue contacts, herein referred to as loop nodes. To measure loop rates, several photophysical methods have been introduced where a pair of optically active probes is incorporated at selected chain positions and the excited probe undergoes contact quenching (CQ) upon collision with the second probe. The quenching mechanisms involved triplet-triplet energy transfer, photoinduced electron transfer, and collision-induced fluorescence quenching, where the fluorescence of Dbo, an asparagine residue conjugated to 2,3-diazabicyclo[2.2.2]octane, is quenched by tryptophan. The discrepancy between the loop rates afforded from these three CQ techniques has, however, remained unresolved. In analyzing this discrepancy, we now report two short-distance FRET methods where Dbo acts as an energy acceptor in combination with tryptophan and naphtylalanine, two donors with largely different fluorescence lifetimes of 1.3 and 33 ns, respectively. Despite the different quenching mechanisms, the rates from FRET and CQ methods were, surprisingly, of comparable magnitude. This combination of FRET and CQ data led to a unifying physical model and to the conclusion that the rate of loop formation in folding reactions varies not only with the kind and number of residues that constitute the chain but also in particular with the size and properties of the residues that constitute the loop node.

Two Orders of Magnitude Variation of Diffusion-Enhanced Förster Resonance Energy Transfer in Polypeptide Chains

“…The analysis yielded a diffusion coefficient of 55.4 Å 2 /ns in the absence of ethylene glycol, and a skewed Gaussian distance distribution, p ( r ) = cr 2 ∙exp(− a ( r − b ) 2 , with a = 1.27 × 10 −3 Å −2 and b = −25.3 Å (Figure 7b) at all solution conditions. Because the optimization could not directly provide confidence intervalls of the parameters of interest and because we have repeatedly expressed our skepticism, whenever very sharp values of diffusion coefficients were reported in the literature [13,36], we tested the robustness of the analysis and the results by also analyzing random subsets, limited to 15–35 points, of the entire set of 48 data points. This procedure and the analysis of 200 optimizations yielded D = 53.4 ± 6.0 Å, a = 1.28 ± 0.06 × 10 −3 Å −2 , and b = −25.7 ± 2.3 Å, which is very close to the results obtained from all 48 points.…”

PET and FRET utility of an amino acid pair: tryptophan and 4-cyanotryptophan

“…First, we employed it to probe the loop formation rate of a short peptide. Recently, Jacob et al 19 applied multiple techniques (e.g., FRET and PET) to assess the loop formation rate (k L ) of a series of flexible and unstructured L 1 -(GS) n -L 2 peptides, where L 1 and L 2 represent two different probes and n varies from 0 to 10. Their findings indicate that k L depends not only on n but also on the identity of L 1 and L 2 , highlighting the importance of using non-interacting or weakly interacting probes in this type of experiments.…”

“…In addition, their study provides valuable experimental data that can be used as a reference to validate the applicability of new PET and FRET pairs. Therefore, to make a direct comparison, we studied the 4CN-Trp-(GS) 4 -Trp peptide, which is an analogue of the Trp-(GS) 4 -MR121 peptide of Jacob et al 19 As shown (Fig. 7), compared to the fluorescence decay kinetics of a reference peptide that lacks the Trp quencher Assuming that this change in the fluorescence lifetime of 4CN-Trp is only due to the addition of the PET quenching channel, the effective quenching rate constant (k Q ) is calculated to be 9.9 × 10 7 s −1 .…”

“…7), compared to the fluorescence decay kinetics of a reference peptide that lacks the Trp quencher Assuming that this change in the fluorescence lifetime of 4CN-Trp is only due to the addition of the PET quenching channel, the effective quenching rate constant (k Q ) is calculated to be 9.9 × 10 7 s −1 . As described by Jacob et al, 19 k Q is proportional to k L although the exact proportionality factor depends on the model used. Interestingly, the k Q value determined by Jacob et al 19 for Trp-(GS) 4 -MR121 is faster, ca.…”

“…As described by Jacob et al, 19 k Q is proportional to k L although the exact proportionality factor depends on the model used. Interestingly, the k Q value determined by Jacob et al 19 for Trp-(GS) 4 -MR121 is faster, ca. 1.7 × 10 8 s −1 .…”