LOCAL ORGANIZATION AT SERINE-PROLINE AND PHOSPHOSERINE-PROLINE SEQUENCES: STABILIZATION OF THE cis-AMIDE CONFORMATION VIA C–H/O INTERACTIONS
Date
2024-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Serine-Proline sequences are abundant in the human proteome, particularly in
intrinsically disordered proteins. Ser-Pro motifs are also highly overrepresented as
phosphorylation sites. This trend indicates that Ser-Pro sequences play a significant
role in the stabilization and modulation of secondary structures. To comprehensively
examine transient and native folded structures at Ser-Pro and pSer-Pro sequences, a
series of SP and pSP dipeptides were synthesized with the proline derivative 2S,4S hydroxyproline-(4-iodophenyl)-methyl ester (hyp(4-I-Ph)-OMe) and subsequently
analyzed via NMR and X-ray crystallography. Experimental structural methods were
performed in conjunction with DFT calculations and bioinformatics analysis of the
PDB to generalize the observed stabilizing interactions and structural preferences.
hyp(4-I-Ph)-OMe was synthesized via C-terminal methylation of Boc-2S,4R hydroxyproline-OH, Mitsunobu reaction at the position, and subsequent acidic
deprotection to generate the amine. The amine was coupled to the C-terminus of
Boc-serine-OH to make the Boc N-capped dipeptide (Boc-Ser-hyp(4-I-Ph)-OMe).
Further N-terminal deprotection and acetylation produced the acetyl N-capped
dipeptide (Ac-Ser-hyp(4-I-Ph)-OMe). The X-ray crystal structure of Boc-Ser-hyp(4-
I-Ph)-OMe contained two molecules in the unit cell, one with cis-Pro and one with
trans-Pro. The molecule with cis-Pro adopted a type VIa2 -turn (BcisD), and the
Pro cis-amide conformation was stabilized via a C–H/O interaction between Pro C–
H and Ser O. The molecule with trans-Pro adopted the PPII conformation in both
residues. Two conformations were also observed in the X-ray crystal structure of Ac Ser-Pro(4-I-Ph)-OMe: one with Ser in the conformation and one with Ser in the
PPII conformation, while both had trans-Pro in the conformation. Stabilization of
the Pro cis-amide via a C–H/O interaction was consistent with both larger downfield
1H- chemical shifts of cis-Pro H from trans-Pro H (cis–trans) and lower Ktrans/cis
values compared to Ala-Pro containing controls. Boc/Ac-Ser-hyp(4-I-Ph)-OMe were
also chemoselectively phosphorylated. pH-dependent NMR revealed a dependence
of the downfield chemical shift cis-Pro H on phosphate ionization state, consistent
with stabilization of pSer-cis-Pro via a C–H/O interaction. The cis–trans of Pro H
was larger in Ac-pSer-hyp(4-I-Ph)-OMe than Ac-Ser-hyp(4-I-Ph)-OMe in water.
Bioinformatics analysis of the PDB and DFT geometry optimization of model
peptides were used to more thoroughly examine structures at both Ser-cis-Pro and
Ser-trans-Pro. Comparison to Ala-Pro sequences offered insights into structure
stabilization mediated by local interactions with the Ser side chain. Analysis of
structures containing Ser-cis-Pro revealed that these sequences were almost all type
VI -turns, with 45% of structures containing C–H/O interactions. C–H/O
interactions were also found to be exclusive to the Pro cis-amide conformation.
Alternatively, Seri-trans-Proi+1 sequences favor C=Oi/H–Ni+3 or C=Oi/H–Ni+4 main
chain hydrogen-bonded secondary structures, found in 53% of all Ser-trans-Pro
sequences. These sequences primarily adopt type I -turns or act as N-terminal caps
of 310- and -helices. Both structures are stabilized by intramolecular solvation of
amide hydrogens by Ser O. When in a type I -turn or acting as a 310-helix Ncap,
Ser O is capable of hydrogen bonding with the i+2 amide N–H in the 1 trans
rotamer, leaving no solvent-exposed N-terminal amide hydrogens. When acting as an
-helix Ncap, Ser O is capable of hydrogen bonding with the i+3 amide N–H in the
1 g rotamer, leaving only one solvent-exposed N-terminal amide hydrogen. C–
H/O interactions are also capable of stabilizing non-Pro cis-amide bonds