3-(2,5-Cyclohexadienyl)-L-alanine (1,4-Dihydro-L-phenylalanine)
---
Its Synthesis and Behaviour in the Phenylalanine Ammonia-Lyase Reaction

Introduction
The phenylalanine ammonia-lyase (PAL, EC 4.3.1.5) is one of the best studied plant enzymes. It catalyses a central reaction at the branching point of primary and secondary plant metabolism by converting L-phenylalanine into trans-cinnamic acid, the precursor of phenylpropanoids like flavonoids, coumarins, and lignins [1,2]. These substance classes have functions as dyes, UV-absorbing pigments, or chemical signal-transmitters (scheme 1).
Since the discovery of PAL by KOUKOL and CONN in 1961 [3] this enzyme has met great interest. In 1970 HANSON and HAVIR proposed dehydroalanine to be at the active site of PAL [4], which was supported by chemical and mutation studies. A serine residue has been identified as precursor, that is posttranslationally converted into dehydroalanine [5]. HANSON and HAVIR also made a proposal for the mechanism of the PAL reaction (scheme 2) [4,6,7]. In the key step of the mechanism a Michael-type addition of the amino group of L-phenylalanine to the ß-position of dehydroalanine takes place. This was supposed to enhance the leaving ability of the amino group. A 1,3-hydrogen shift and the abstraction of a benzylic proton by an enzymatic base lead then to trans-cinnamic acid.
This mechanism was accepted as working hypothesis for almost 25 years. But there were some unanswered questions concerning the proposed ionic intermediate. It was not clear how it could be formed or stabilised, because no base is likely to exist in proteins that is strong enough to abstract a non-activated benzylic proton [8].
SCHUSTER and RÉTEY provided evidence against this mechanism by observing the conversion of 4-nitro-L-phenylalanine with inactivated PAL. The inactivation occurred either by reduction of the prosthetic dehydroalanine with sodium borohydride or by site-directed mutagenesis (Ser202 to Ala202) [8]. Their conclusion was that the nitro group in para position can substitute the electrophilic prosthetic group. Consequently SCHUSTER and RÉTEY saw the role of dehydroalanine in the activation of the protons in ß-position of L-phenylalanine to facilitate their abstraction by an enzymatic base and to stabilise the resulting carbanion.
Based on these considerations the following mechanism was proposed: The prosthetic dehydroalanine attacks the ortho position of the aromatic ring of L-phenylalanine in a Friedel-Crafts-type reaction [8,9]. In the resulting carbenium ion the protons in ß-position are acidified and the abstraction by an enzymatic base is facilitated. After removal of the pro-3S-proton elimination of ammonia, rearomatisation of the ring and regeneration of the prosthetic group occur (scheme 3).
The enzymes PAL and HAL (histidine ammonia-lyase, histidase, EC 4.3.1.3) have a high homology (19-29% sequence identity). Hence it can be concluded that the essential part of the active site of both enzymes has the same structure. The X-ray structure of PAL’s “sister enzyme” HAL revealed that the catalytic electrophile was not dehydroalanine but 3,5-dihydro-5-methylidene-4H-imidazol-4-one (MIO), which is autocatalytically formed from the inner tripeptide Ala-Ser-Gly by cyclisation and elimination of two molecules of water [10] (scheme 4). This modification can be interpreted as an enhancement of the electrophilicity of the prosthetic group by preventing the delocalisation of the nitrogen lone pairs into the 2,3-unsaturated carbonyl system.
To provide further support for the proposed mechanism two different non-aromatic L-phenylalanine derivatives were regarded as useful probes: The first is 2,5-dihydro-L-phenylalanine [3-(1,4-cyclohexadienyl)-L-alanine], the other 1,4-dihydro-L-phenylalanine [3-(2,5-cyclohexadienyl)-L-alanine] [11]. Of these the 2,5-dihydro isomer is expected to be a substrate: the electrophilic attack generates a carbenium ion next to the side chain. Thereby the ß-protons are made more acidic and can be abstracted by an enzymatic base (scheme 5).
In the case of the 1,4-dihydro isomer an electrophilic attack is also possible. But the generated positive charge is not located next to the side chain, the ß-protons remain non-acidic and the lyase reaction does not take place (scheme 6). If this prediction turns out to be correct, further evidence for the recently proposed mechanism would be provided.

Title        Abstract       Results and Discussion       Conclusion and Acknowledgements       References