Probing Peptidylprolyl Bond cis/trans Status Using Distal 19F NMR Reporters

Abstract A method for measuring peptidylprolyl bond cis‐trans conformational status in peptide models is described, using 4‐fluorophenylalanine (4FPhe) as a distal reporter for 19F NMR. The %cis‐Pro population was measured for peptides of the general structure Ac‐X‐Pro‐Z‐Ala‐Ala‐4FPhe (X and Z are proteinogenic amino acids) at pH 7.4, and provided conformational populations consistent with literature values obtained by more complex methods. This approach was applied to probe the prolyl bond status in pentapeptide models of the intrinsically disordered C‐terminal region of α‐synuclein, which mirrored the preferences in the Ac‐X‐Pro‐Z‐Ala‐4FPhe models. Advantageously, the 19F reporter group does not need to be adjacent to or attached to proline to provide quantifiable signals and distal 4‐fluorophenylalanines can be placed so as not to influence prolyl bond conformation. Finally, we demonstrated that the prolyl bond status is not significantly affected by pH when there are ionisable amino acid residues at the carboxyl side of proline, which makes 19F NMR an invaluable tool with which to study proline isomerism at a range of pHs and in different solvents and buffers.


Introduction
The peptidylprolyl bond is a key facilitator of conformational change and peptidylprolyl bond cis-trans isomerism is often the slow-phase or rate-limiting step in protein folding, with a ~80 kJ/mol energy barrier for interconversion. [1]Standard backbone amide bonds, comprised of non-cyclic canonical amino acids have a very high thermodynamic preference for a trans conformation. [2]However, with its sidechain forming part of the backbone, proline contains a tertiary amide in peptides and proteins, which makes the difference in energy between the cis and trans forms lower than for standard amino acids, affording a higher population of the cis conformation (Figure 1A) at proline compared with non-prolines.These features result in unique properties that are fundamental to the three-dimen-sional structure of peptides and proteins.Indeed, the prevalence of proline residues in intrinsically disordered protein motifs (approx.double the average [3] ) has prompted further investigation into the role of prolines in adopting partially folded states and promoting disorder or compaction. [4]Indeed, the peptidylprolyl cis-trans isomerases CypD and FK506 binding proteins (FKBPs) have been shown to interact directly with the acidic proline-rich C-terminal region of α-synuclein (αSyn), which forms a constituent part of the protein aggregates in Lewy bodies in a range of synucleinopathies, for example, Parkinsons Disease. [5,6]FKBPs were shown to accelerate the rate of aggregation, and aggregation was abolished by removal of the proline residues from the C-terminus.However, CypD binding to soluble αSyn prevented its aggregation, whilst the addition of CypD to preformed αSyn fibrils led to fibril disassembly.In both cases, this behaviour was linked to an increased rate of isomerase-catalysed peptidylprolyl isomerisation.Moreover, Pro-to-Ala mutations at C-terminal region are reported to accelerate the rate of uncatalysed aggregation, [7] highlighting a key role of proline conformation in Parkinson's Disease pathology.
11][12][13] Most of the studies described have used NMR-based approaches to detect and/or quantify prolyl conformers.NMR techniques provide structural information at the molecular level and can be useful in understanding the finer details of protein and peptide folding at specific sites, particularly with respect to proline isomerism. [14,15]][12]16] However, this can create complexity in the analysis and in many cases, 2D NMR approaches are required to resolve specific resonances.Compared with the use of 1 H NMR for studies of peptide conformation, 19 F NMR exhibits very little-to-no spectral crowding and will detect separate signals for the different forms that are in chemical exchange.The fluorine-19 isotope ( 19 F) exhibits higher local environmental sensitivity compared with the 1 H nucleus, providing defined and distinct chemical shifts for individual fluorinated molecules, isomers or conformers. [17]In addition, it is only slightly less sensitive compared with the 1 H nucleus. Therefore, 19 F NMR analysis of fluorinated peptides enables far simpler identification and quantification of conformer populations and over a wider range of timescales as compared with traditional methods.20][21] Prolyl bond conformational status in proteins has been measured previously using a combination of fluorinated prolines and 19 F NMR.In particular, a number of groups have used fluorinated amino acids derivatives to study cis-trans isomerism by 19 F NMR. Thomas and colleagues studied 3S and 3Rfluoroproline measured the kinetic and thermodynamic parameters of proline cis-trans isomerisation, [22] whilst Renner performed a similar analysis of 4S, 4R and 4,4-difluoroprolines in Ac-(F)Pro-OMe using 19 F NMR. [23] It was clear from this work (and others) that the 4R-fluoroproline has a greater preference for the trans-Pro conformation, whilst the 4S-diastereomer has a relatively increased cis-Pro population compared to the parent proline.These studies highlighted the connection between the fluorine configuration and cis:trans conformer populations, albeit these observations were only in only in simple model systems.Kawahara was able to incorporate 4R-fluoroproline into collagen polypeptides and study triple helix assembly and the relationship to proline conformation using 19 F NMR. [24] Torbeev demonstrated that 4,4-dIfluoroproline, which has nearly the same cis-trans preference as proline but an increased rate of isomerisation, caused pronounced destabilization of macroglobulin; indicating that the cis-trans equilibrium and isomerisation dynamics of a single proline affect the rate of protein folding. [25]Increasing the number of equivalent fluorine atoms enhances the signal-to-noise ratio in 19 F NMR analysis.Trifluoromethylproline was incorporated into a simple Ac-TfmPro-OMe model and used with 19 F NMR to provide an accurate reporter of cis:trans and to determine the K trans/cis . [26]hilst Tressler and co-workers developed perfluoro-tert-butyl 4hydroxyproline (9 fluorine atoms) as a highly-sensitive 19 F NMR reporter for proline isomerism, permitting quantification at concentrations as low as 200 nM. [27]However, these studies also highlighted that fluorine atoms exert an influence on the conformation of prolines, and fluoroproline, is therefore, a biased proline conformational reporter. [8]4-Fluorophenylalanine has been used to study the formation of oligomers in the misfolding of prion proteins using 19 F NMR. [28] Multiple 4fluorophenylalanine residues have also been incorporated into a single protein, rat intestinal fatty acid binding protein (IFABP), to monitor the folding process and providing information on the conformational status at various locations simultaneously. [29]-Fluorophenylalanine was also used as a reporter of the enzyme catalysed cis-to-trans isomerisation rate of [4-fluoro-Phe]bradykinin measured by magnetisation-transfer 19 F NMR. [30] However, most of the reported examples of using 4-fluorophenylalanine as a 19 F NMR reporter in peptide and protein systems have focused on its applications in probing global conformational effects and little attention has been paid to its use for probing prolyl bond status.
Here we examine the use of 4-fluorophenylalanine-as a reporter to measure the relative cis:trans prolyl bond conformer ratio in X-Pro-Z systems contained in pentapeptide models.Moreover, we explore how the proximity of this reporter group and the nature of proximal amino acids impacts upon peptidylprolyl cis:trans populations (Figure 1B).

Results and Discussion
Employing 4-fluorophenylalanine as a reporter group, the initial aim was to understand how the proximity of the 4-fluorophenylalanine residue to the proline, affected the relative cis:trans ratio and how far away this could be moved within a peptide sequence and still provide distinguishable 19 F NMR resonances for each conformer.The Tyr-Pro motif exhibits a relatively high cis-Pro population of around 20-39 % in, for example, Ac-Ala-Tyr-Pro-Ala-Lys-NH 2 , [9] Ac-Gly-Tyr-Pro-Gly [12] and Ac-Thr-Tyr-Pro-Asn, [13] depending on other proximal residues. [9,12,13]Therefore, this was used as a starting point for model design and optimisation, and a series of peptides containing Ac-Tyr-Pro-(Ala) n -4FPhe (where n = 0-7) were synthesised using standard microwave-assisted solid phase peptide synthesis, sequentially moving the 4-fluorophenylalanine reporter one residue at a time away from proline.Peptides were prepared as C-terminal amides and were N-terminally capped as an acetamide to avoid ionisable termini influencing the cis:trans through electrostatic interactions with one another or ionisable sidechains, as noted in other reports. [31,32]he first point to note was the relative simplicity of the 19 F NMR spectrum compared with the 1 H NMR spectrum for the same model peptide (Figure 2), which significantly deconvolutes the conformational analysis.An advantage of the 19 F NMR method is that it functions readily at physiological pH (7.4), whereas 1 H NMR approaches often quantify cis-Pro in the amide region, which disappears due to exchange above ~pH 7 or in D 2 O (absent in Figure 2). 19F NMR can also report cis-Pro populations in any non-fluorinated solvent or buffer without need for water-suppression techniques.Peptide prolyl bond cis: trans was analysed by 19 F NMR (relaxation delay time (D1) set at 10 s; with 1 H-decoupling) in (1 mg/mL in aqueous 10 mM phosphate buffered saline (10 % D 2 O), pH 7.4) at 23 °C and afforded two 19 F resonances with the major being assigned as the thermodynamically favoured trans and minor as cis conformer (e. g., Figure 3A).Pleasingly, the control peptide containing no proline, only afforded a single resonance.The signal integrations were consistent with the reported range of cis-Pro populations in related systems as mentioned earlier and were largely similar for each probe (Figure 3B), confirming that 4fluorophenylalanine can report on cis-trans conformer populations over a range of amino acids (up to > 6 residues away).Models containing six and seven alanine spacer residues exhibited distinct signals but were poorly resolved and not quantifiable.Slightly lower cis-Pro populations were noted for cases when the 4-fluorophenylalanine was two (27 %) or three (32 %) residues away from proline, whilst the model containing 4FPhe adjacent to Pro exhibited a slightly greater cis-Pro population (37 %).In some cases, aromatic (Aro) amino acids in an Aro-Pro-Aro sequence can increase the propensity for a cis prolyl bond conformation but this effect may be less significant when the aromatic residue is electron poor, for example, 4fluorophenylalanine. [13] Despite the Ala-Ala spacer model appearing to slightly under-report the %cis conformer (27 %), we later found that this provided a good match with the literature conformer populations (discussed subsequently).
The signal resolution between cis-Pro and trans-Pro resonances is an important criterion for modelling proline isomerism by NMR.By 19 F NMR analysis the chemical shift difference for conformers varied with relative proximity of the reporter to the proline (Figure 3C; also see Supporting Information, Figure S1).Although it was possible to quantify cis-trans prolyl  populations up to at least 6 amino acids away from proline, the optimal signal resolution (0.07 ppm) was obtained in the model containing two alanine residues between the 4-fluorophenylalanine and proline (AcYPAA-4FPhe, Figure 3A, C).Therefore, this was used as the basis for our the model system to probe the effects of amino acids X and Z in X-Pro-Z systems.
To demonstrate that this model system could be used to study the influence of amino acids proximal to proline on cis: trans ratio, 39 peptides of the form Ac-X-Pro-Ala-Ala-4FPhe and Ac-Tyr-Pro-Z-Ala-4FPhe were synthesised by systematically replacing the amino acid X or Z immediately adjacent to proline.As before, peptidylprolyl cis:trans conformer ratios were determined by integration of characteristic resonances obtained from 19 F NMR (386 MHz, 23 °C, 1 mg/mL aqueous 10 mM phosphate buffered saline (10 % D 2 O), pH 7.4) analysis (Figure S2).As expected, the relative ratio of signal integration varied according to the amino acid at the proline nitrogen (Figure 4, Table S1) -indicating that this was reporting the cis: trans ratio.Some pairs of signals were poorly resolved despite the model optimisation and required line-fitting analysis using the Mnova line-shape tool.Chemical shifts also were observed to change slightly, and indeed there was a weak correlation for higher cis-Pro preferences to also exhibit greater signal resolution in 19 F NMR spectra, perhaps due to slower rates of isomerism.
Comparison of the cis-Pro populations measured by 19 F NMR with those obtained by 1 H NMR in related models published by Reimier et al. [9] and Dyson et al., [10] respectively, afforded good agreement (Figure 4A, B and Table S2).The 19 F NMR measurements are also consistent with the earlier reports of local conformational effects and the role of neighbouring amino acids upon proline cis-trans conformational preferences, albeit in slightly different models. [9,10]In general, the X-Pro motif exhibits the greatest influence over the conformation of proline, particularly when X is an aromatic amino acid: Phe (22.5 %), Tyr (26.5 %) and Trp (34.5 %).This was further enhanced in Tyr-Pro-Aro motifs (Aro = aromatic amino acids: Phe (38 %), Tyr (38.5 %) and Trp (35.5 %)) exhibiting the highest propensity for a cis prolyl bond.[35] These observations are attributed to favourable CH-π interactions between relatively acidic proline H α or H δ atoms and electronrich aromatic rings; and aromatic rings interacting with both faces of proline, respectively. [9,12,13,34] Hwever, in both types of model His was found to provide a lower cis-Pro population than other aromatic amino acids (His-Pro-Ala (14 %) and Tyr-Pro-His (31 %), behaving more like a basic sidechain.Indeed, there was a clear trend for electron-rich aromatics to promote higher cis-Pro in the order: Trp-Pro > Tyr-Pro > Phe-Pro > His-Pro. [9,13,36]It was also clear that polar side chains, for example, Ser (12 %), Thr (13 %) Asp (13.5 %) promote a slightly higher cis-Pro population in X-Pro systems compared with Ala-Pro (10.5 %), while Arg (8 %) or Lys (9 %) in this position leads to a subtle decrease in the cis-Pro population.These results corroborate the observations from Sebak and co-workers in their statistical analysis of cis-Pro populations in intrinsically disordered proteins. [15]Another interesting observation was the relatively high %cis for the Tyr-Pro-Gly (31 %) and Tyr-Pro-Pro (39 %) peptides, which are approximately equivalent to the Aro-Pro-Aro cis-populations.
A small number of discrepancies were noted between the reported literature cis-Pro populations and those measured using 19 F NMR. Notably, in our hands when basic amino acids were proximal to proline in Tyr-Pro-Z models we observed significantly higher cis-Pro populations (Z = Arg 29 %, Lys 29.5 %, His 31 %) compared with Tyr-Pro-Ala (26.5 %) (Figure 4B). [37]Our systems appeared to exhibit higher %cis than the corresponding models by Dyson et al., [10] wherein the basic amino acids exhibited rather low cis-Pro (vs.16, 17 and 13 %, respectively,) compared with Tyr-Pro-Ala.However, our data are mostly well aligned with Reimer et al. for Ac-X-Pro models (X = basic amino acid), who reported cis-Pro populations of X = Arg (7.2 %), Lys (6.8 %) and His (16.5 %) at pH 6 but noted that the histidine peptide Ac-Ala-His-Pro-Ala-Lys-NH 2 exhibited a lower cis-Pro population (9.5 %) at pH 8.
Given that the literature experiments were conducted at different pHs -often to observe non-proline amide NHs for 1 H NMR analysis, we were concerned that basic-Pro or Pro-basic motifs may be sensitive to ionisation status or intramolecular electrostatic interactions.To try to understand this better, we resynthesized exemplar peptides related to those from Dyson et al. but containing 4-fluorophenylalanine (AcYPZD(4FPhe) where Z = ionisable residues D, H, K or R) and measured the % cis-Pro at two different pHs (4.1 and 7.4) using 19 F NMR.
In our analyses, the cis-Pro populations were observed to align closely with our earlier models, dominated by the Tyr-Pro motif (Figure 5), and importantly, the prolyl conformer popula- ) [9] at pH 6 where 'X' is at the prolyl amine.B) Comparison with model H 2 N-Tyr-Pro-Z-Asp-Val-COOH (Dyson et al.) [10] at pH 4.1 where 'Z' is at the proline carboxamide.Note: blank columns indicate no possible comparison within the literature.
tions were found to be largely insensitive to modest pH changes.This underlines that the use of 19 F NMR at pH 7.4 accurately reflects what others have seen at acidic pH.This is a key advantage over 1 H NMR methods that are difficult to evaluate at physiological pH.Therefore, the differences between this work and the Dyson work may be related to previous models containing a second adjacent ionisable group (i.e., Asp), which may provide a repulsive interaction.We also demonstrated that the %cis was unaffected by changes in salt (NaCl) concentration up to 1.2 M in the buffer solution (see Supporting Information, Figures S9-12).
We briefly investigated whether the effect of temperature on cis-Pro population could be studied using 19 F NMR, by heating NMR samples of peptide models Ac-LPAA(4FPhe)-CONH 2 and Ac-YPGA(4FPhe)-CONH 2 in pH 7.4 buffer between 10 °C-60 °C in 10 °C steps.In simple model prolines, this is shown to lead to an increase in the % cis-Pro, [23] whereas in small peptides, this was suggested to be a largely insignificant effect. [38]In our hands, resonances relating to cis and trans-Pro AcLPAA(4FPhe) was found to appear to coalesce towards a single signal with increasing temperature and correspondingly, the % cis-Pro after line shaping analysis appeared to slightly increase above 30 °C (7 % at 10 °C to 12 % at 60 °C).Conversely, the peptide AcYPGA(4FPhe) seemed to trend towards marginally lower % cis-Pro (26 % at 10 °C to 22 % at 60 °C) (see Supporting Information, Figure S7, Table S3).Albeit, in both cases, this was a minor change, this further emphasises that 19 F NMR measurements can perhaps provide additional insight into subtle effects that are challenging to observe by 1 H NMR.
α-Synuclein is a generally disordered protein, but can assume stable misfolded protein oligomers and aggregates, which is a hallmark of Parkinson's disease pathophysiology.To demonstrate that the observed prolyl bond populations are relatively conserved in different peptides and are governed mainly by local sequences, we prepared a series of pentapeptide models of the proline-rich (5 proline residues in full length protein) intrinsically disordered C-terminal region of α-synuclein (Figure 6, Table 1).These models contained disorder-promoting amino acids for example Q, E, D, N rather than alanines between the proline and 19 F reporter.It was clear from these experiments, that the %cis-Pro in each model was similar to that observed in our earlier models and the observations of Sebak et al. in intrinsically-disordered proteins.This indicates that prolyl bond conformation is controlled predominantly by the proline N-acyl substituent in these models and that %cis-Pro may be generally predictable within an unstructured polypeptide.Importantly, this also highlights that the 19 F NMR approaches are broadly useful and sensitive to small changes in conformer populations.

Conclusion
In summary, the prolyl bond conformational trends reported herein using 19 F NMR are broadly similar to the previous observations using different techniques, and importantly, align with the general trends for prolyl bond conformation in folded protein structures -indicating that prolyl bond conformation is predominantly governed by local effects.Furthermore, very similar prolyl bond conformational populations were observed to occur in short models of the intrinsically disordered Cterminal region of α-synuclein, indicating that these prolyl bond populations are conserved in flexible peptides and proteins.
This work has demonstrated that incorporating fluorinated reporter groups via 4-fluorophenylalanine is a reliable way to quantify prolyl bond status in peptide models.If positioned carefully, it does not influence the prolyl bond status and advantageously, does not need to be adjacent or attached to the proline to provide quantifiable and distinguishable signals, providing flexibility in model design.Furthermore, Fmoc-L-4fluorophenylalanine is commercially available at low cost, which makes this approach useful for a wide range of researchers.The simplicity of the 19 F NMR read-out compared to the correspond-   ing 1 H NMR spectrum and the ability to quantify cis-Pro populations at a range of temperatures, physiological pH and in a variety of solvents and buffers without solvent-suppression and in complex mixtures such as cell lysate, will be of broad and important use in biological chemistry.Specifically, this will be utilised in our future studies as a straightforward tool to understand the consequences of point mutations or the role of specific residues upon cis-trans isomerism at specific proline residues in larger, more complex and intrinsically disordered and proline-rich systems.

Synthesis of model peptides
Fmoc solid-phase peptide synthesis was performed using automated Fmoc-SPPS on a Liberty Blue microwave-assisted peptide synthesiser (CEM).SPPS was conducted using Rink amide ProTide resin (0.56 mmol/g, 0.1 mmol, 179 mg).Fmoc protected amino acids were used in 5 equivalents (0.2 M in DMF) with DIC (1 M stock solution in DMF; 10 equiv.) and Oxyma Pure (1 M stock solution, 5 equiv) as activators.Standard coupling procedures employed single coupling of each amino acid (2.5 min, 90 °C).Amino acids bearing thermally-sensitive protecting groups, for example, Fmoc-L-His(Trt)-OH and Fmoc-L-Cys(Trt)-OH, were coupled under milder conditions (50 °C for 10 min).Fmoc deprotection was performed by treatment of the resin with piperidine (20 % v/v in DMF, 4 mL).Following the removal of the final Fmoc-protecting group, peptides were acetylated (where relevant) using acetic anhydride (20 % in DMF) at 37 °C for 15 min and repeated once, followed by washing the resin with DMF three times and drying under vacuum.The resin was washed several times with Et 2 O (5 mL) and cleaved from the resin using a cocktail of TFA/TIPS/H 2 O (95/2.5/2.5) for 2 h at 37 °C or 4 h at RT in a disposable fritted syringe.Afterwards, the suspension was filtered and precipitated in ice-cold Et 2 O.The resulting suspension was centrifuged (5 min at 5000 rpm) and the pellet was resuspended with Et 2 O (10 mL).This was centrifuged again (5 min 5000 rpm) and the Et 2 O was discarded.Crude peptide mixtures were purified using RP-HPLC (Agilent 1260), equipped with a Waters Xbridge peptide BEH C18 Prep column (19 × 100 mm, 5 μm, 130 Å) using a linear gradient from 5-95 % MeOH in water containing 0.1 % TFA over 20 minutes at a flow rate of 8 mL/min.

Characterisation of model peptides
High-resolution mass spectrometry method A: performed using an Agilent 6530 accurate mass QToF system.Separation was per-formed on an Agilent ZORBAX Eclipse Plus C18 Rapid Resolution HD analytical C18 column (1.8 μm particle size, 2.1 × 50 mm) at 25 °C.The mobile phase of the system consisted of water (with 0.1 % TFA, eluent A) and Methanol (with 0.1 % TFA, eluent B). the 12.5 min gradient was programmed as follows: 0-1 min 95 % A, 1.0-7.0min to 0 % A, 7.0-9.0min hold 0 %, 9.0-9.5 min to 95 % A, 9.5-12.5 min hold 95 % A. The injection volume was set to 2 μL and the flow rate to 0.5 mL/min, respectively.Monitoring was performed at λ = 215 nm.Mass spectral analysis of peptides was performed in the positive ionisation (ESI +) mode in full scan mode (m/z 100-3200).Operating pressures were in the range of 2000-3000 PSI.Electrospray ionisation mass spectrometry was conducted in positive ion mode (m/z range: 50-3200) using a fragmentor voltage of 150 V, gas temperature of 325 °C (flow 10 L/min) and sheath gas temperature of 400 °C (flow 11 L/min).Exact mass measurements of the products were based on the protonated ions [M + H] + and were detected as sodiated adducts [M + Na] + .
Analytical HPLC method B (retention times b ): Separation was performed on a Waters XBridge C18 analytical column (3.5 μm particle size, 4.6 × 50 mm) at 40 °C.The mobile phase of the system consisted of water (with 0.1 % TFA, eluent A) and Methanol (with 0.1 % TFA, eluent B).The 15 min gradient was programmed as follows: 0-10 min 95 % A to 20 % A, 10-11 min to 5 % A, 11-12 min hold 5 % A, 12-13 min to 95 % A, 13-15 min hold 95 % A. The injection volume was set to 2 μL and the flow rate to 1 mL/min, respectively.Monitoring was performed at λ = 215 nm.Variable Temperature Experiment Procedure: Purified peptides (2 mg/ mL) were dissolved in 10 % D 2 O/H 2 O.The NMR acquisition parameters were the same as the general procedure other than being conducted over a range of temperatures from 10 °C to 60 °C over 10 °C steps.

Figure 1 .
Figure 1.Peptidylprolyl cis-trans isomerism and the design of model systems containing fluorine reporter groups for 19 F NMR to measure prolyl bond status.

Figure 2 .
Figure 2. Comparaison of an example 1 H NMR (top) and 19 F NMR (bottom) spectrum of peptide Ac-Tyr-Pro-Ala-Ala-Ala-4FPhe.Experiments were acquired at 2 mg/mL in D 2 O, room temperature.

Figure 3 .
Figure 3. A) An example 19 F NMR spectrum of peptide Ac-Tyr-Pro-Ala-Ala-(4FPhe) (2 alanine spacer) exhibiting prolyl bond cis-trans conformers.Probing the effect of proximity of the 19 F NMR reporter (4FPhe = 4fluorophenylalanine) on B) measured cis-Pro populations in model peptides using 19 F NMR, and C) chemical shift difference between resonances.* Poorly resolved: not accurately quantifiable but still distinguishable 19 F NMR spectrum.

Figure 4 .
Figure 4. Comparison of 19 F NMR derived cis-Pro conformational preference (dark blue) with earlier published data (light blue) obtained by 1 H NMR analysis.A) Comparison with model Ac-Ala-X-Pro-Ala-Lys-NH 2 (Reimer et al.)[9] at pH 6 where 'X' is at the prolyl amine.B) Comparison with model H 2 N-Tyr-Pro-Z-Asp-Val-COOH (Dyson et al.)[10] at pH 4.1 where 'Z' is at the proline carboxamide.Note: blank columns indicate no possible comparison within the literature.

Figure 6 .
Figure 6.The intrinsically disordered C-terminal region of α-synuclein (PDB: 1XQ8), which consists of amino acids 96 to 140 (proline residues in red).Tetramer sequences, indicated with bold lines, informed the design of the 5 pentapeptide models used in this study.