Crystal structure of rat heme oxygenase-1 in complex with biliverdin-iron chelate. Conformational change of the distal helix during the heme cleavage reaction

The crystal structure of rat heme oxygenase-1 in complex with biliverdin-iron chelate (biliverdin(Fe)-HO-1), the immediate precursor of the final product, biliverdin, has been determined at a 2.4-A resolution. The electron density in the heme pocket clearly showed that the tetrapyrrole ring of heme is cleaved at the alpha-meso edge. Like the heme bound to HO-1, biliverdin-iron chelate is located between the distal and proximal helices, but its accommodation state seems to be less stable in light of the disordering of the solvent-exposed propionate and vinyl groups. The middle of the distal helix is shifted away from the center of the active site in biliverdin(Fe)-HO-1, increasing the size of the heme pocket. The hydrogen-bonding interaction between Glu-29 and Gln-38, considered to restrain the orientation of the proximal helix in the heme-HO-1 complex, was lost in biliverdin(Fe)-HO-1, leading to relaxation of the helix. Biliverdin has a distorted helical conformation; the lactam oxygen atom of its pyrrole ring-A interacted with Asp-140 through a hydrogen-bonding solvent network. Because of the absence of a distal water ligand, the iron atom is five-coordinated with His-25 and four pyrrole nitrogen atoms. The coordination geometry deviates considerably from a square pyramid, suggesting that the iron may be readily dissociated. We speculate that the opened conformation of the heme pocket facilitates sequential product release, first iron then biliverdin, and that because of biliverdin's increased flexibility, iron release triggers its slow dissociation.     

Two serine residues distinctly regulate the rescue function of Humanin,an inhibiting factor of Alzheimer's disease-related neurotoxicity: functional potentiation by isomerization and dimerization

The 24-residue peptide Humanin (HN), containing two Ser residues at positions 7 and 14, protects neuronal cells from insults of various Alzheimer's disease (AD) genes and A beta. It was not known why the rescue function of (S14G)HN is more potent than HN by two to three orders of magnitude. Investigating the possibility that the post-translational modification of Ser14 might play a role, we found that HN with D-Ser at position 14 exerts neuroprotection more potently than HN by two to three orders of magnitude, whereas D-Ser7 substitution does not affect the rescue function of HN. On the other hand, S7A substitution nullified the HN function. Multiple series of experiments indicated that Ser7 is necessary for self-dimerization of HN, which is essential for neuroprotection by this factor. These findings indicate that the rescue function of HN is quantitatively modulated by d-isomerization of Ser14 and Ser7-relevant dimerization, allowing for the construction of a very potent HN derivative that was fully neuroprotective at 10 pM against 25 microM A beta1-43. This study provides important clues to the understanding of the neuroprotective mechanism of HN, as well as to the development of novel AD therapeutics.     

ik3-2, a relative to ik3-1/cables, is associated with cdk3, cdk5, and c-abl

A cDNA coding for ik3-2 (designated as ik3-2 from an interactor-2 with cdk3) was cloned by cross-hybridization with ik3-1 and RT-PCR. Analysis of amino acid sequence indicated that ik3-2 has the C-terminal cyclin-box-like region highly homologous to that of ik3-1 (identity in amino acids: 78%). On the other hand, the remainder of ik3-2 gene is not so similar to that of ik3-1. There are several regions other than the C-terminal cyclin-box-like region that are conserved between ik3-1 and ik3-2. In vivo binding assay indicated that like ik3-1, ik3-2 binds to cdk3, cdk5, and c-abl, although ik3-2 binds to cdk3 weakly as compared with ik3-1. The C-terminal cyclin-box-like region of ik3-2 (123 amino acids) is able to be associated with cdk5. Accordingly, ik3-2 is very similar to ik3-1 concerning its molecular interaction with other molecules, suggesting that ik3-2 function in the same biological field as ik3-1. Northern blot analysis indicated that ik3-2 is expressed ubiquitously all over tissues.     

Polymorphisms of estrogen receptor alpha gene in endometrial cancer

It is hypothesized that polymorphisms of estrogen receptor-alpha (ERalpha) gene are involved in endometrial cancer. To test this hypothesis, the genotype distributions of six different loci (codon 10 T-->C, codon 87 G-->C, codon 243 C-->T, codon 325 C-->G, codon 594 G-->A, and intron 1 C-->G) of the ERalpha gene were investigated and their association with endometrial cancer was determined. The DNA from 113 cases of human endometrial cancer was analyzed by sequence-specific polymerase chain reaction. The relative risk of variant genotype was calculated by comparison with 200 healthy controls. The frequency of variant genotype on codon 10 was significantly lower in endometrial cancer patients as compared to controls. Nine of 113 endometrial cancer patients (8.0%) showed genotype 10C/C compared to 27 of 200 healthy controls (13.5%). The relative risk of genotype 10C/C was calculated as 0.44, compared to wild-type. Forty-five of 113 endometrial cancer patients (39.8%) showed genotype T/C on codon 10 compared to 111 of 200 healthy controls (55.5%). The relative risk of genotype 10T/C was calculated as 0.67, compared to wild-type. The polymorphism on codon 87 was not detected both in endometrial cancer patients and in healthy control. Other loci, intron 1, and codons 243, 325, and 594, did not show a correlation with endometrial cancer. The frequency of alleles on codon 10 was also significantly lower in endometrial cancer patients as compared to controls. Sixty-three of 226 alleles (27.9%) of endometrial cancer patients showed allele C compared to 165 of 400 (41.2%) of healthy controls. The relative risk of allele 10C was calculated as 0.67, compared to wild-type. Other loci, intron 1, and codons 243, 325, and 594, did not show a difference between cancer patients and controls. All genotype and allelic distributions were in accordance with the Hardy-Weinberg equilibrium. The present study demonstrates for the first time a protective effect of 10C allele against endometrial cancer. Thus, inherited alterations in ERalpha may be associated with changes in estrogen metabolism and thereby may possibly explain inter-individual differences in disease incidences of endometrial cancer.     

Effects of parenteral lipid infusion on DNA damage in very low birth weight infants

Very low birth weight (VLBW) infants are known to have poorly developed antioxidant system and may be at increased risk for radical damage. Previous studies have reported higher levels of lipid peroxide products in lipid emulsion used for parenteral nutrition. To examine the direct effects of parenteral lipid infusion on DNA damage in VLBW infants, we measured urinary 8-hydroxydeoxyguanosine (8-OHdG) levels in VLBW infants before, during, and after the parenteral lipid infusion. In both the lipid-infused and lipid-free groups, urinary 8-OHdG excretion levels at 14 days old were significantly (p < 0.01) lower than those at 2 and 7 days old. However, there were no significant differences in urinary 8-OHdG excretion levels between the lipid-infused and lipid-free groups at 2, 7, and 14 days old. Our results suggest that parenteral lipid infusion does not cause oxidative DNA damage, but irrespective of the infusion DNA damage during the first week of life is enhanced when compared with 14 days after birth in VLBW infants.