Evolution of structure and function in Phenylalanine Hydroxylase. With the regulatory properties in sight
Abstract
In the post-genomic era, an idea of how similar the genomes of different species actually are is on the horizon. Less than 10 years ago, the human genome was estimated to encode 100000 genes. That was an overestimation, as the real number of human genes is 20000-25000. Most genes are expressed as proteins. The 3D structure of a protein is more conserved than its sequence, and therefore the structural context of protein and gene evolution must not be forgotten. By its structure, the protein can propagate its function. In the early 90’s the estimated number of different protein structure classes, so called folds, was predicted to be about 10000. Today there are slightly above 1000 folds and the discovery of new folds has leveled off, despite an increase in the number of protein structures that have been solved over the last few years. Indeed, some folds are used for more than one function, and found in various functional contexts. Then, if the many components are so similar, how is the biological species divergence from same component genomes achieved? One way to study biological diversity is by dividing it into its smaller components, e.g. by studying protein or gene family evolution. Here the evolution and regulation of the aromatic amino acid hydroxylase (AAAHs) have been under examination. This gene family encodes the proteins phenylalanine hydroxylase (PAH), tyrosine hydroxylase (TH), and tryptophan hydroxylase (TPH). These enzymes are highly physiologically important. PAH, expressed in liver, regulates the homeostasis of L-Phe by hydroxylating it into L-Tyr. TH, expressed in the central nervous system, hydroxylates L-Tyr into L-Dopa. L-Dopa is part of two important pathways i) melanogenesis and ii) dopamine production. In humans, dysfunctions in PAH that cause elevated L-Phe concentration can result in phenylketonuria (PKU). Untreated PKU results in neurological damage. TPH produces a precursor of serotonin from LTrp. The end products of these enzymes are neurotransmitters and hormones with increasingly important functions, from e.g. amoeba to nematode to man. As PAH has evolved in mammals its regulation has become increasingly sophisticated, e.g. homotropic positive cooperativity that shifts the conformational equilibrium from dimeric to tetrameric is seen in the mammalian lineage. Nematode PAH is devoid of positive cooperativity, but resembles the tetrameric high-affinity and high-activity mammalian PAH. TH and TPH are always tetrameric and not allosterically regulated. Each AAAH subunit has a regulatory domain, a catalytic domain, and an oligomerization domain. The promotion of positive cooperativity in PAH has been investigated by comparing mammalian PAH to nematode PAH. The low-affinity and low-activity dimer as well as the high-affinity and high-activity tetramer of PAH were modeled. Sequence analysis on a nematode sequence cluster and a mammalian sequence cluster identified sites with high probability of being involved in functional divergence, e.g. change in regulation. Residue specific electrostatic interaction energies were calculated for all ionizible residues in the models. In general, we note important differences in the substrate binding pocket that aids to explain why the active site in nematode PAH is less dynamic than in mammalian PAH. Our results suggest a pathway for the positive cooperativity from one active site to another, involving various predicted hinge regions from human PAH, where we find the nematode PAH more rigid. The regulatory domain in PAH is part of the ACT domain family. The ACT domains are frequently found regulating metabolic enzymes in an allosteric manner. The allosteric effector is often an amino acid that binds to an interface formed by two ACT domains. No contacts are formed between two ACT domains and the stoichiometry of binding is 1:1 for L-Phe in PAH. Therefore the allosteric effect must originate in the active site when the substrate binds. An alternative pathway for aromatic amino acid biosynthesis is present in e.g. plants and bacteria. This pathway has an L-Phe binding ACT domain, which is homologous to the ACT domain in AAAH. The L-Phe binding motif in this domain is also conserved in PAH. A comparative structural analysis of this area shows why L-Phe may not bind in the AAAH regulatory domain and also indicates why it has remained. The ACT domain has an abundant fold, a superfold. A structural approach was used to identify more potential ACT domains to gain further insights to the functional properties that this domain could perform in general, and in PAH in particular. Here we note e.g. two interesting potential domain families that could be homologous to the ACT domain, namely the GlnB-like domains and heavy metal binding domains. The phylogeny of the AAAH family has not been resolved earlier given the lack of a suitable outgroup. As more genome sequences became available, we identified an outgroup candidate and had it experimentally characterized. The phylogeny was resolved, the ancestral function determined, and by comparing the chromosomal gene locations the order of events in AAAH evolution was envisioned.
Has parts
Paper I: Gene 427(1-2), Siltberg-Liberles, J.; Steen, I. H.; Svebak, R. M. & Martinez, A., The phylogeny of the aromatic amino acid hydroxylases revisited by characterizing phenylalanine hydroxylase from Dictyostelium discoideum, pp. 86-92. Accepted version. Copyright 2008 Elsevier B.V. All rights reserved. The published version is available at: http://dx.doi.org/10.1016/j.gene.2008.09.005Paper II: Siltberg-Liberles, J. & Martinez, A., Structural determinants of the regulatory properties in phenylalanine hydroxylase. Draft. Full text not available in BORA.
Paper III: Amino Acids 28(1), Liberles, J. S.; Thorolfsson, M.; & Martinez, A., Allosteric mechanisms in ACT domain containing enzymes involved in amino acid metabolism, pp. 1-12. Accepted version. Copyright 2005 Springer. All rights reserved. The published version is available at: http://dx.doi.org/10.1007/s00726-004-0152-y
Paper IV: Amino Acids 36(2), Siltberg-Liberles, J. & Martinez, A., Searching distant homologs of the regulatory ACT domain in phenylalanine hydroxylase, pp. 235-249. Copyright 2009 Springer. All rights reserved. The published version is available at: http://dx.doi.org/10.1007/s00726-008-0057-2
Publisher
The University of BergenCopyright
Jessica Siltberg-LiberlesAll rights reserved