Submitted to: Phytochemistry
Publication Type: Peer reviewed journal
Publication Acceptance Date: 6/20/2005
Publication Date: 8/15/2005
Citation: Costa, M.A., Bedgar, D.L., Moinuddin, S.G., Kim, K.W., Cardenas, C.L., Cochrane, F.C., Shockey, J.M., Helms, G.L., Amakura, Y., Takahashi, H., Milhollan, J.K., Davin, L.B., Browse, J., Lewis, N.G. 2005. Characterization in vitro and in vivo of the putative multigene 4-coumarate:coa ligase network in arabidopsis: syringyl lignin and sinapate/sinapyl alcohol derivative formation. Phytochemistry. 66:2072-2091. Interpretive Summary: The thale cress plant, Arabidopsis thaliana, has been used as a model for plant studies for many years. It has proven to be an extremely useful tool for understanding many poorly characterized biochemical processes, in plants and other organisms. The utility of Arabidopsis comes in part from the fact that the entire DNA sequence of all five Arabidopsis chromosomes has recently been determined, and many useful databases that enable fast, efficient gene discovery are available to researchers at the touch of a button. In the current study, this power has been applied to developing a better understanding of the set of enzymes that carry out the first steps in the synthesis of a class of compounds called phenylpropanoids. Phenylpropanoids are used by plants in insect defense, pollinator attraction, protection against damage from ultraviolet light, and for many other purposes. One of the early steps in phenylpropanoid production is the synthesis of activated acid groups, which are any of several different types of acid groups coupled to a molecule called coenzyme A. The enzymes that couple the acid to coenzyme A are called 4-coumaryl-CoA ligases, or 4CLs. Analysis of the Arabidopsis genome sequence revealed a total of 14 genes that resembled 4CLs that had already been cloned from other plants. Three of these had already been characterized in other labs, the other eleven genes were cloned and used to produce the 4CL enzymes in the bacterium E. coli. The 4CL proteins were purified away from all other proteins and then tested for their ability to couple coenzyme A to each of six different acids. Only four of the eleven enzymes were active, suggesting that the other seven enzymes must participate in processes other than this branch of phenylpropanoid metabolism. These findings are significant since they demonstrate that similarity between the sequence of a known gene and unknown genes is not enough to unambiguously determine the true biological function of the unknown gene.
Technical Abstract: A recent in silico analysis revealed that the Arabidopsis genome has 14 genes annotated as putative 4-coumarate:CoA ligase isoforms or homologues. Of these, 11 were selected for detailed functional analysis in vitro, using all known possible phenylpropanoid pathway intermediates (p-coumaric, caffeic, ferulic, 5-hydroxyferulic and sinapic acids), as well as cinnamic acid. Of the 11 recombinant proteins so obtained, four were catalytically active in vitro, with fairly broad substrate specificities, confirming that the 4CL gene family in Arabidopsis has only four members. This finding is in agreement with our previous phylogenetic analyses, and again illustrates the need for comprehensive characterization of all putative 4CLs, rather than piecemeal analysis of selected gene members. All 11 proteins were expressed with a C-terminal His6-tag and functionally characterized, with one, At4CL1, expressed in native form for kinetic property comparisons. Of the 11 putative His6-tagged 4CLs, isoform At4CL1 best utilized p-coumaric, caffeic, ferulic and 5-hydroxyferulic acids as substrates, whereas At4CL2 readily transformed p-coumaric and caffeic acids into the corresponding CoA esters, while ferulic and 5-hydroxyferulic acids were converted quite poorly. At4CL3 also displayed broad substrate specificity efficiently converting p-coumaric, caffeic and ferulic acids into their CoA esters, whereas 5-hydroxyferulic acid was not as effectively utilized. By contrast, while At4CL5 is the only isoform capable of ligating sinapic acid, the two preferred substrates were 5-hydroxyferulic and caffeic acids. Indeed, both At4CL1 and At4CL5 most effectively utilized 5-hydroxyferulic acid with kenz approximately 10-fold higher than that for At4CL2 and At4CL3. The remaining seven 4CL-like homologues had no measurable catalytic activity (at approximately 100 microg protein concentrations), again bringing into sharp focus both the advantages to, and the limitations of, current database annotations, and the need to unambiguously demonstrate true enzyme function. Lastly, although At4CL5 is able to convert both 5-hydroxyferulic and sinapic acids into the corresponding CoA esters, the physiological significance of the latter observation in vitro was in question, i.e. particularly since other 4CL isoforms can effectively convert 5-hydroxyferulic acid into 5-hydroxyferuloyl CoA. Hence, homozygous lines containing T-DNA or enhancer trap inserts (knockouts) for 4cl5 were selected by screening, with Arabidopsis stem sections from each mutant line subjected to detailed analyses for both lignin monomeric compositions and contents, and sinapate/sinapyl alcohol derivative formation, at different stages of growth and development until maturation. The data so obtained revealed that this "knockout" had no significant effect on either lignin content or monomeric composition, or on the accumulation of sinapate/sinapyl alcohol derivatives. The results from the present study indicate that formation of syringyl lignins and sinapate/sinapyl alcohol derivatives result primarily from methylation of 5-hydroxyferuloyl CoA or derivatives thereof rather than sinapic acid ligation. That is, no specific physiological role for At4CL5 in direct sinapic acid CoA ligation could be identified. How the putative overlapping 4CL metabolic networks are in fact organized in planta at various stages of growth and development will be the subject of future inquiry.