The Origin and Evolution of Baeyer-Villiger Monooxygenases (BVMOs): An Ancestral Family of Flavin Monooxygenases

Abstract

The Baeyer-Villiger Monooxygenases (BVMOs) are enzymes belonging to the "Class B" of flavin monooxygenases and are capable of performing exquisite selective oxidations. These enzymes have been studied from a biotechnological perspective, but their physiological substrates and functional roles are widely unknown. Here, we investigated the origin, taxonomic distribution and evolutionary history of the BVMO genes. By using in silico approaches, 98 BVMO encoding genes were detected in the three domains of life: Archaea, Bacteria and Eukarya. We found evidence for the presence of these genes in Metazoa (Hydra vulgaris, Oikopleura dioica and Adineta vaga) and Haptophyta (Emiliania huxleyi) for the first time. Furthermore, a search for other "Class B" monooxygenases (flavoprotein monooxygenases--FMOs--and N-hydroxylating monooxygenases--NMOs) was conducted. These sequences were also found in the three domains of life. Phylogenetic analyses of all "Class B" monooxygenases revealed that NMOs and BVMOs are monophyletic, whereas FMOs form a paraphyletic group. Based on these results, we propose that BVMO genes were already present in the last universal common ancestor (LUCA) and their current taxonomic distribution is the result of differential duplication and loss of paralogous genes.

Fig 1. Taxonomic distribution of BVMO genes.

A. Distribution of BVMO encoding genes in genomes from Bacteria and Archaea domains. B. Distribution of BVMO encoding genes in genomes from Eukarya domain. Green triangles show clades containing BVMOs; grey triangles show clades where BVMOs were not detected. Numbers in brackets show the number of different species where BVMOs were found. The tree topology is based on the tree of life web project (Maddison DR and Schulz KS (eds.) 2007. Available at: http://tolweb.org (Accesed 17 November 2014)).

Fig 2. Phylogenetic analysis of BVMOs protein-family by Maximum Likelihood (ML) method.

The tree was constructed by using the ML method, employing the alignment constructed with MAFFT 7 on-line tool and the best model parameters calculated with ProTest 3.4. Evolutionary analyses were conducted in PhyML 3.0 on-line server. Bootstrap values (> 45) are shown next to the branches. Colored triangles show the source of BVMO genes as follows: Fungi (black), Bacteria (orange), Green plants (dark green), Rhodophyta (light green), Metazoa (red), Haptophyta (purple) and Archaea (blue). Empty circles indicate BVMO encoding sequences where strong evidences of contamination artifacts were found: Pantholops hodgsonii (red) and Hordeum vulgare (dark green). Sequences marked with asterisks have been previously characterized.

Fig 3. Rooted phylogenetic tree of flavin monooxygenases “Class B” by Maximum Likelihood (ML) method.

The tree was constructed by using the ML method, employing the alignment constructed with MAFFT 7 on-line tool and the best model parameters calculated with ProTest 3.4. Evolutionary analyses were conducted in PhyML 3.0 on-line server. Bootstrap values (> 45) are shown next to the branches. Colored branches display: BVMOs (yellow), NMOs (blue) and FMOs (black). As outgroup, hydroxylases belonging to “Class A” flavin monooxygenases were employed (orange).

Fig 4. Scheme of the proposed model for the evolutionary history of “Class B” flavin monooxygenases.

Hypothetical ancestral flavin monooxygenases genes were present in a pre-LUCA time. These genes gave origin to the monophyletic group of BVMOs and to a group of FMOs/NMOs-like genes, which later originated the monophyletic group of NMOs and a paraphyletic group of modern FMOs. We propose that BVMOs, FMOs and NMOs were already present in LUCA. After crossing the Darwinian threshold different paralogous of flavin monooxygenase genes were inherited by the three life domains: Bacteria, Archaea and Eukarya.