Evolutionary dynamic analyses on monocot flavonoid 3′-hydroxylase gene family reveal evidence of plant-environment interaction

Background Flavonoid 3′-hydroxlase (F3’H) is an important enzyme in determining the B-ring hydroxylation pattern of flavonoids. In monocots, previous studies indicated the presence of two groups of F3’Hs with different enzyme activities. One F3’H in rice was found to display novel chrysoeriol-specific 5′-hydroxylase activity. However, the evolutionary history of monocot F3’Hs and the molecular basis for the observed catalytic difference remained elusive. Results We performed genome-wide survey of 12 common monocot plants, and identified a total of 44 putative F3’H genes. The results showed that F3’H gene family had underwent volatile lineage-specific gene duplication and gene loss events in monocots. The expansion of F3’H gene family was mainly attributed to dispersed gene duplication. Phylogenetic analyses showed that monocot F3’Hs have evolved into two independent lineages (Class I and Class II) after gene duplication in the common ancestor of monocot plants. Evolutionary dynamics analyses had detected positive natural selection in Class II F3’Hs, acting on 7 specific amino acid sites. Protein modelling showed these selected sites were mainly located in the catalytic cavity of F3’H. Sequence alignment revealed that Class I and Class II F3’Hs displayed amino acid substitutions at two critical sites previously found to be responsible for F3’H and flavonoid 3′5’-hydroxylase (F3’5’H) activities. In addition, transcriptional divergence was also observed for Class I and Class II F3’Hs in four monocot species. Conclusions We concluded that monocot F3’Hs have evolved into two independent lineages (Mono_F3’H Class I and Class II), after gene duplication during the common ancestor of monocot plants. The functional divergence of monocot F3’H Class II has been affected by positive natural selection, which acted on specific amino acid sites only. Critical amino acid sites have been identified to have high possibility to affect the substrate specificity of Class II F3’Hs. Our study provided an evolutionary and protein structural explanation to the previously observed chrysoeriol-specific 5′-hydroxylation activity for CYP75B4 in rice, which may also be true for other Class II F3’Hs in monocots. Our study presented clear evidence of plant-environmental interaction at the gene evolutionary level, and would guide future functional characterization of F3’Hs in cereal plants