Cinnamyl alcohol dehydrogenase (CAD) is the generic depiction of a class of enzymes known to catalyze reduction of various phenylpropenyl aldehyde derivatives during the formation of monolignols, the precursors of lignins and lignans. It has often been reported to exist as a multigene family in many plant species although no detailed characterization of the different protein isoforms was ever carried out. In a later and most extraordinary rationale, given its position as the last step on the pathway to the monolignols, CAD was claimed to serve as a key (rate-limiting) step during monolignol biosynthesis (26); however, detailed metabolic flux studies (4), as well as critical analysis of various transgenic (CAD-down-regulated) plant lines (5), have unequivocally demonstrated this not to be the case. Instead, these proteins apparently serve solely to reduce the various aldehydes that are differentially generated during monolignol formation. In this way, while the enzymes apparently display differential substrate versatilities for each of the possible aldehydic precursors, none are strictly substrate specific. Indeed, manifestation of this CAD versatility may have already been demonstrated in vivo during manipulation of monolignol biosynthesis in Arabidopsis, whereby p-coumaryl, coniferyl, and sinapyl alcohol derived lignins were individually obtained via alteration of specific hydroxylation steps (10,27,28). That is, these manipulations had no apparent effect on the ability of the organism to reduce the corresponding aldehydic precursors so formed. These data would suggest, as already noted for previous CAD down-regulation steps, that no single CAD isoform has a strict substrate preference.
To date, there are no reports of the biochemical characterization of any Arabidopsis CAD isoform, even though the Arabidopsis genomic sequence has 17 CAD homologues annotated. All of these, as expected, lack signal peptides and are presumed cytosolic. When compared with each other (relative to the arbitrarily chosen CAD1), these have 60.5 – 55.5% and 51.3 – 44.1% sequence similarities and identities to each other (1). That is, there is a very great variability in this gene family, as noted for most of the other biochemical steps described thus far that have been annotated with putative functions. Furthermore, when compared with a lignin-specific tobacco CAD (29,30), only isoforms CAD4 and CAD5 displayed significant similarity (81.5 and 82.9%) and identity (75.1 and 76.5%) to the tobacco CAD, suggesting that Arabidopsis CAD4 and CAD5 were likely to be lignin-specific. Isoforms CAD6, 7 and 8 also displayed 77.3, 77.6 and 78.0% similarity and 68.5, 71.4 and 72.1% identity, respectively, to the claimed sinapyl aldehyde-specific isoform present in poplar (31). The other eight show no homology to, for instance, the tobacco CAD.
In terms of EST database entries, only five of the CAD homologues (CAD1, 4, 5, 7 and 9) were detected, whereas CAD2, 3, 6 and 8 were not; in the latter cases, this is again perhaps due to low copy numbers, transient expression and/or being inducible only under certain conditions. This notwithstanding, the potential lignin-specific CAD4 and CAD5 isoforms were detected in roots, green siliques and mixed tissues with CAD4 also being present in above-ground organs. Interestingly, CAD7 and CAD9 were also present in above-ground organs, mixed tissues and floral organ tissues, with the latter being detected in developing seeds and green siliques as well. CAD1, by contrast, was only found in developing seeds and green siliques.
As before, the overall strategy being employed is to (i) comprehensively characterize the putative CAD multigene family; (ii) determine individual patterns of putative CAD multigene family using promoter-driven expression analyses; and (iii) analysis of putative CAD knockouts. Progress is shown in Tables 14 – 17.(32).