京都大学 / 新制・課程博士 / 博士(医学) / 甲第23777号 / 医博第4823号 / 新制||医||1057(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 江藤 浩之, 教授 寺田 智祐, 教授 小川 誠司 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

5-Aminolevulinate synthase (ALAS; EC 2.3.1.37) is a pyridoxal 5'-phosphate (PLP)-dependent enzyme that catalyzes the first committed step of heme biosynthesis in animals, the condensation of glycine and succinyl-CoA yielding 5-aminolevuliante (ALA), CoA, and CO2. Murine erythroid-specific ALAS (mALAS2) variants that cause high levels of PPIX accumulation provide a new means of targeted, and potentially enhanced, photosensitization. Transfection of HeLa cells with expression plasmids for mALAS2 variants, specifically for those with mutated mitochondrial presequences and a mutation in the active site loop, caused significant cellular accumulation of PPIX, particularly in the membrane. Light treatment of HeLa cells expressing mALAS2 variants revealed that mALAS2 expression results in an increase in cell death in comparison to aminolevulinic acid (ALA) treatment producing a similar amount of PPIX. Generation of PPIX is a crucial component in the widely used photodynamic therapies (PDT) of cancer and other dysplasias. The delivery of stable and highly active mALAS2 variants has the potential to expand and improve upon current PDT regimes. Mutations in the C-terminus of human ALAS2 (hALAS2) can increase hALAS2 activity and are associated with X-linked erythropoietic protoporphyria (XLEPP), a disease phenotypically characterized by elevated levels or PPIX and zinc protoporphyrin in erythroblasts. This is apparently due to enhanced cellular hALAS2 activity, but the biochemical relationship between these C-terminal mutations and increased hALAS2 activity is not well understood. HALAS2 and three XLEPP variants were studied both in vitro to compare kinetic and structural parameters and ex vivo in HeLa and K562 cells. Two XLEPP variants, delAGTG, and Q548X, exhibited higher catalytic rates and affinity for succinyl-CoA than wild-type hALAS2, had increased transition temperatures, and caused porphyrin accumulation in HeLa and K562 cells. Another XLEPP mutation, delAT, had an increased transition temperature and caused porphyrin accumulation in mammalian cells, but exhibited a reduced catalytic rate at 37[deg]C in comparison to wild-type hALAS2. The XLEPP variants, unlike wild-type hALAS2, were more structurally responsive upon binding of succinyl-CoA, and adopted distinct features in tertiary and PLP cofactor-binding site. These results imply that the C-terminus of hALAS2 is important for regulating its structural integrity, which affects kinetic activity and stability. XLEPP has only recently been identified as a blood disorder, and thus there are no specific treatments. One potential treatment involves the use of the antibiotic isonicotinic acid hydrazide (isoniazid, INH), commonly used to treat tuberculosis. INH can cause sideroblastic anemia as a side-effect and has traditionally been thought to do so by limiting PLP availability to hALAS2 via direct inhibition of pyridoxal kinase, and reacting with pyridoxal to form pyridoxal isonicotinoyl hydrazone. We postulated that in addition to PLP-dependent inhibition of hALAS2, INH directly acts on hALAS2. Using FACS and confocal microscopy, we show here that INH reduces protoporphyrin IX accumulation in HeLa cells expressing either wild-type human hALAS2 or XLEPP variants. In addition, PLP and pyridoxamine 5'-phosphate (PMP) restored cellular hALAS2 activity in the presence of INH. Kinetic analyses with purified hALAS2 demonstrated non-competitive or uncompetitive inhibition with an apparent Ki of 1.5 uM. Circular dichroism studies revealed that INH triggers structural changes in hALAS2 that interfere with the association of hALAS2 with its PLP cofactor. These studies demonstrate that hALAS2 can be directly inhibited by INH, provide insight into the mechanism of inhibition, and support the prospective use of INH in treating patients with XLEPP and potentially other cutaneous porphyrias.

The biosynthesis of tetrapyrolles in eukaryotes and the alpha-subclass of purple photosynthetic bacteria is controlled by the pyridoxal 5?-phosphate (PLP)-dependent enzyme, 5-aminolevulinate synthase (ALAS). Aminolevulinate, the universal building block of these macromolecules, is produced together with Coenzyme A (CoA) and carbon dioxide from the condensation of glycine and succinyl-CoA. The three-dimensional structures of Rhodobacter capsulatus ALAS reveal a conserved active site serine that moves to within hydrogen bonding distance of the phenolic oxygen of the PLP cofactor in the closed, substrate-bound enzyme conformation, and simultaneously to within 3-4 angstroms of the thioester sulfur atom of bound succinyl-CoA. To elucidate the role(s) this residue play(s) in enzyme activity, the equivalent serine in murineerythroid ALAS was mutated to threonine or alanine. The S254A variant was active, but both the KmSCoA and kcat values were increased, by 25- and 2-fold, respectively, suggesting the increase in turnover is independent of succinyl-CoA-binding. In contrast, substitution of S254 with threonine results in a decreased kcat, however the Km for succinyl-CoA is unaltered. Removal of the side chain hydroxyl group in the S254A variant notably changes the spectroscopic properties of the PLP cofactor and the architecture of the PLP-binding site as inferred from circular dichroism spectra. Experiments examining the rates associated with intrinsic protein fluorescence quenching of the variant enzymes in response to ALA binding show that S254 affects product dissociation. Together, the data led us to suggest that succinyl-CoA binding in concert with the hydrogen bonding state of S254 governs enzyme conformational equilibria. As a member of the alpha-oxoamine synthase family, ALAS shares a high degree of structural similarity and reaction chemistry with the other enzymes in the group. Crystallographic studies of the R. capsulatus ALAS structure show that the alkanoate component of succinyl-CoA is bound by a conserved arginine and a threonine. To examine acyl-CoA-binding and substrate discrimination in murine erythroid ALAS, the corresponding residues (R85 and T430) were mutated and a series of CoA substrate analogs were tested. The catalytic efficiency of the R85L variant with octanoyl-CoA was 66-fold higher than that calculated for the wild-type enzyme, suggesting this residue is strategic in substrate binding. Hydrophobic substitutions of the residues that coordinate acyl-CoA-binding produce ligand-induced changes in the CD spectra, indicating that these amino acids affect substrate-mediated changes to the microenvironment of the chromophore. Pre-steady-state kinetic analyses of the R85K variant-catalyzed reaction show that both the rates associated with product-binding and the parameters that define quinonoid intermediate lifetime are dependent on the chemical composition of the acyl-CoA tail. Each of the results in this study emphasizes the importance of the relationship between the bifurcate interaction of the alkanoic acid component of succinyl-CoA and the side chains of R85 and T430. From the X-ray crystal structures of Escherichia coli 8-amino-7-oxonoanoate synthase and R. capsulatus ALAS, it was inferred that a loop covering the active site moved 3-6 Å between the holoenzymic and acyl-CoA-bound conformations. To elucidate the role that the active site lid plays in enzyme function, we shuffled the portion of the murine erythroid ALAS cDNA corresponding to the lid sequence (Y422-R439), and isolated functional variants based on genetic complementation in an ALA-deficient strain. Variants with potentially greater enzymatic activity than the wild-type enzyme were screened for increased porphyrin overproduction. Turnover number and the catalytic efficiency of selected functional variants with both substrates were increased for each of the enzyme variants tested, suggesting that increased activity is linked to alterations of the loop motif. The results of transient kinetics experiments for three isolated variants when compared to wild-type ALAS showed notable differences in the pre-steady-state rates that define the kinetic mechanism, indicating that the rate of ALA release is not rate-limiting for these enzymes. The thermodynamic parameters for a selected variant-catalyzed reaction indicated a reduction in the amount of energy required for catalysis. This finding is consistent with the proposal that, in contrast to the wild-type ALAS reaction, a protein conformational change associated with ALA release no longer limits turnover for this variant enzyme.

ABSTRACT The porphyrias are a diverse group of metabolic disorders arising from diminished activity of enzymes in the heme biosynthetic pathway. They can present with acute neurovisceral symptoms, cutaneous symptoms, or both. The complexity of these disorders is demonstrated by the fact that some acute porphyria patients with the underlying genetic defect(s) are latent and asymptomatic while others present with severe symptoms. This indicates that there is at least one other risk factor required in addition to the genetic defect for symptom manifestation. A systematic review of the heme biosynthetic pathway highlighted the involvement of a number of micronutrient cofactors. An exhaustive review of the medical literature uncovered numerous reports of micronutrient deficiencies in the porphyrias as well as successful case reports of treatments with micronutrients. Many micronutrient deficiencies present with symptoms similar to those in porphyria, in particular vitamin B6. It is hypothesized that a vitamin B6 deficiency and related micronutrient deficiencies may play a major role in the pathogenesis of the acute porphyrias. In order to further investigate the porphyrias, a computational model of the heme biosynthetic pathway was developed based on kinetic parameters derived from a careful analysis of the literature. This model demonstrated aspects of normal heme biosynthesis and illustrated some of the disordered biochemistry of acute intermittent porphyria (AIP). The testing of this model highlighted the modifications necessary to develop a more comprehensive model with the potential to investigated hypotheses of the disordered biochemistry of the porphyrias as well as the discovery of new methods of treatment and symptom control. It is concluded that vitamin B6 deficiency might be the risk factor necessary in conjunction with the genetic defect to trigger porphyria symptoms.