Cis/Trans Isomerization 

 The distinction between the cis and trans isomers of a molecule originates from a geometry-based classification of molecular structures. When the two substituents X and Y are on the same side of the structural unit [Fig. 1(a)] the isomer is cis. It is designated trans in the opposite arrangement. Ambiguities in the designation of the isomers are avoided by using the more sophisticated E/Z nomenclature, which often has E for the cis form and Z for the trans form. The interconversion between the isomers, the cis/trans isomerization, occurs by rotation about the central linkage [Fig. 1 )a)]. There is a barrier to rotation quantified by the free energy of activation DG^A  that is proportional to a first-order rate constant k_obs, where k_obs = (k_{cis to trans} + k_{trans to cis} for the reversible isomerization. The bond order of the central linkage correlates with the magnitude of the energy barrier DG^A . It ranges from high values of DG ^A > 100 kJ/mol for double bonds, via intermediate range for linkages having a partial double bond character, down to low values in the range of k_BT (where k_B is the Boltzmann constant and T the absolute temperature) for C–C single bonds. Under ambient conditions, a rotational barrier >90 kJ/mol indicates that the individual isomers are not readily interconverted, and they may exhibit quite distinct chemical properties. A corresponding DG^A  value has been used to discriminate semantically between different conformations (barrier to rotation <90 kJ/mole) and configurations of a molecule. 

Figure 1. (a) General representation of cis/trans isomerism (b) Cis/trans isomerism of a prolyl peptide bond.

In a polypeptide chain, both the amide -CONH- and the imide CON < peptide groups are intrinsically competent to undergo cis/trans isomerization by rotation about the torsion angle w of the peptide bond. Among the gene-encoded amino acids, the imide peptide bond is exclusively formed by proline residues. Because the lone electron pair of the nitrogen atom is delocalized over the peptide bond, the C–N linkage has a partial double-bond character. Typically, this destabilizes twisted conformations but stabilizes planar arrangements (trans, w = 180° cis, w = 0°) of the two a-C atoms adjacent to the peptide bond [Fig. 1(b)]. The cis/trans isomerization of the -CONH- moiety is relatively fast (half-time <1 at room temperature) and leads to a very small percentage of cis isomer,  1%, at equilibrium. In contrast, the peptidyl-proline moiety (in the following referred to as a prolyl bond) often has cis/trans isomers in comparable amounts. Steric constraints, which favor a trans arrangement in the case of -CONH-, are similar in both isomers for prolyl bonds, and the cis isomer generally occurs in about 20% of the prolyl peptide bonds. There is a relatively high rotational barrier DG^A  of about 80 kJ/mol for the prolyl bond. The combination of a substantial population of both isomers and their slow interconversion implies that cis prolyl bonds have considerable influence on biochemical reactions of the polypeptide backbone.

In the absence of a folded conformation, polypeptide chains can theoretically form 2ⁿ cis/trans isomers, where n is the number of prolyl bonds in the molecule. In simple cases, as with a four-proline octapeptide derived from the prolactin receptor, all of the possible isomers have been detected and quantified in solution (1). Structural formation in the peptide chain, however, reduces the number of isomers substantially. A particular prolyl bond in native proteins is usually either cis or trans in all the molecules, although cis/trans isomerization in the native structure has been observed for a few proteins by NMR methods in solution. In the latter case, structural alterations propagate through the backbone around the isomeric proline residue, which can be accompanied by distinct biological activities of the isomeric proteins (2, 3). Depending on the polypeptide structure, the half-time for prolyl bond isomerization ranges from seconds to hours.

From stereochemical considerations, the peptidyl transferase center on the ribosome is thought to be constructed for synthesizing all peptide bonds in the trans conformation. However, in native, globular proteins of known three-dimensional structure, about 5 to 6% of prolyl peptide bonds are cis (4, 5) . When proline-containing proteins are unfolded in the presence of high concentrations of denaturants, such as urea or guanidinium chloride (GdmCl), the random coil polypeptides generally equilibrate slowly to a mixture of numerous cis/trans isomers. For the fraction of unfolded molecules that have one or more nonnative isomers of a prolyl bond, subsequent refolding of the protein has to start from different conformational states. If the cis/trans isomerization is slower than refolding, slow kinetic phases of folding may be apparent when the time course of refolding is monitored. For that reason, cis/trans isomerization is the rate-limiting step in folding for some proteins. It is generally most easily detectable when there is a cis prolyl bond in the native state, because then a large fraction of unfolded molecules has the incorrect isomer (6, 7).

Chemical catalysis of prolyl bond isomerization is rare. Most organic solvents and micelles or phospholipid vesicles, cause a moderate decrease in the rotational barrier. The rate constants are independent of the pH value in the physiological range, unless dissociable groups are located adjacent to proline. An increased rotational barrier results from the O-protonation of the peptide bond in acidic solution, characterized by a pKa = –1. In strong acids, N-protonated species become populated (pK_a = –7) that accelerate cis/trans isomerization (8). Many relationships between the peptide structure and prolyl bond isomerization have been elucidated. For example, when measured in oligopeptides, the increased barrier to rotation caused by aromatic amino acids that precede the proline residue is accompanied by an increase in the cis population of up to 40%, whereas small aliphatic side chains in the same position lead to lower DG^A  values in conjunction and lower cis contents of 5 to 10% (9).

 The conformational constraints on the polypeptide backbone induced by the prolyl cis/trans isomerization restrict bimolecular recognition processes. Conformational selectivity was demonstrated by the hydrogen bond directed preferential binding of proline peptides in the cis conformation to b-cyclodextrins (10) and to a synthetic, multidentate terephthaloyl amide (11). For the recognition of opioid peptides of the dermorphin type by m- and ∂-receptors, specificity for the cis conformers was suggested (12). In a biological context, it may be important that endoproteinases, such as chymotrypsin, trypsin, thrombin and clostripain cannot readily cleave a peptide bond adjacent to a cis prolyl moiety, even if the isomeric bond occupies a position remote from the scissile bond (13, 14). Due to this conformational specificity, the rate of the cis to trans isomerization of the prolyl bond limits the rate of proteolysis for good substrates in the presence of high protease

concentrations, and this is a useful assay for isomerization. Even in vivo, the time course of bradykinin (Arg-Pro-Pro-Gly-Phe-Ser-Pro-Phe-Arg) degradation by pulmonary endothelial peptidases is controlled by the conformational specificity of the proteinases (15).

References

1. K. D. Oneal et al. (1996) Biochem. J. 315, 833–844. 

2. A. P. Hinck, E. S. Eberhardt, and J. L. Markley (1993) Biochemistry 32, 11810–11818. 

3. J. Kordel, S. Forsen, T. Drakenberg, and W. J. Chazin (1990) Biochemistry 29, 4400–4409. 

4. M. W. MacArthur and J. M. Thornton (1991) J. Mol. Biol. 218, 397–412. 

5. D. E. Stewart, A. Sarkar, and J. E. Wampler (1990) J. Mol. Biol. 214, 253–260. 

6. J. F. Brandts, H. R. Halvorson, and M. Brennan (1975) Biochemistry 14, 4953–4963. 

7. F. X. Schmid (1986) Methods Enzymol. 131, 70–82. 

8.H. Sigel and B. R. Martin (1982) Chem. Rev. 82, 385–426. 

9.  R. K. Harrison and R. L. Stein (1992) J. Amer. Chem. Soc. 114, 3464–3471. 

10. M. Lin et al. (1995) Anal. Chim. Acta 307, 449–457. 

11.  C. Vicent, S. C. Hirst, F. Garciatellado, and A. D. Hamilton (1991) J. Amer. Chem. Soc. 113, 5466-5467.

12. R. Schmidt et al. (1995) Int. J. Peptide Protein Res. 46, 47–55. 

13. G. Fischer, H. Bang, E. Berger, and A. Schellenberger (1984) Biochim. Biophys. Acta 791, 87–97.

14. S. Meyer, M. Drewello, and G. Fischer (1996) Biol. Chem. 377, 489–495. 

15. M. P. Merker and C. A. Dawson (1995) Biochem. Pharmacol. 50, 2085–2091.