Albumin domain II mutant with high bilirubin binding affinity has a great potential as serum bilirubin excretion enhancer for hyperbilirubinemia treatment.

BACKGROUND: 4Z,15Z-bilirubin-IXα (BR), an endogenous toxic compound that is sparingly soluble in water, binds human serum albumin (HSA) with high affinity in a flexible manner. Our previous findings suggest that both Lys195 and Lys199 in subdomain IIA are important for the high-affinity binding of BR, and especially Lys199 in stand-alone domain II plays a prominent role in the renal elimination of BR. Our hypothesis is that HSA-domain II with high BR binding would be a useful therapeutic agent to treat hyperbilirubinemia in patients with impaired liver function. METHODS: Unbound BR concentrations were determined using a modified HRP assay. To evaluate the effect of pan3_3-13 domain II mutant in promoting urinary BR excretion, the serum concentration and urinary excretion amount of BR were determined using bile duct ligation mice. RESULTS: After three or six rounds of panning, pan3_3-13 and pan6_4 were found to have a significantly higher affinity for BR than wild-type domain II. Administration of pan3_3-13 significantly reduced serum BR level and increased its urinary excretion in the disease model mice as compared to wild-type domain II treatment. CONCLUSIONS: These results suggest that pan3_3-13 has great potential as a therapeutic agent that promotes urinary BR excretion in hyperbilirubinemia. GENERAL SIGNIFICANCE: This is the first study to be applied to other HSA bound toxic compounds that are responsible for the progression of disease, thereby paving the way for the development of non-invasive and cost effective blood purification treatment methods.

S-Nitrosated human serum albumin dimer is not only a novel anti-tumor drug but also a potentiator for anti-tumor drugs with augmented EPR effects.

Macromolecules have been developed as carriers of low-molecular-weight drugs in drug delivery systems (DDS) to improve their pharmacokinetic profile or to promote their uptake in tumor tissue via enhanced permeability and retention (EPR) effects. In the present study, recombinant human serum albumin dimer (AL-Dimer), which was designed by linking two human serum albumin (HSA) molecules with the amino acid linker (GGGGS)(2), significantly accumulated in tumor tissue even more than HSA Monomer (AL-Monomer) and appearing to have good retention in circulating blood in murine colon 26 (C26) tumor-bearing mice. Moreover, we developed S-nitrosated AL-Dimer (SNO-AL-Dimer) as a novel DDS compound containing AL-Dimer as a carrier, and nitric oxide (NO) as (i) an anticancer therapeutic drug/cell death inducer and (ii) an enhancer of the EPR effect. We observed that SNO-AL-Dimer treatment induced apoptosis of C26 tumor cells in vitro, depending on the concentration of NO. In in vivo experiments, SNO-AL-Dimer was found to specifically deliver large amounts of cytotoxic NO into tumor tissue but not into normal organs in C26 tumor-bearing mice as compared with control (untreated tumor-bearing mice) and SNO-AL-Monomer-treated mice. Intriguingly, S-nitrosation improved the uptake of AL-Dimer in tumor tissue through augmenting the EPR effect. These data suggest that SNO-AL-Dimer behaves not only as an anticancer therapeutic drug, but also as a potentiator of the EPR effect. Therefore, SNO-AL-Dimer would be a very appealing carrier for utilization of the EPR effect in future development of cancer therapeutics.

Biological characteristics of two lysines on human serum albumin in the high-affinity binding of 4Z,15Z-bilirubin-IXα revealed by phage display.

4Z,15Z-bilirubin-IXα (4Z,15Z-BR), an endogenous compound that is sparingly soluble in water, binds human serum albumin (HSA) with high affinity in a flexible manner. A phage library displaying recombinant HSA domain II was constructed, after three rounds of panning against immobilized 4Z,15Z-BR, and eight clones with high affinity for the pigment were found to contain conserved basic residues, such as lysine or arginine, at positions 195 and 199. The wild type and two mutants, K195A and K199A, of whole HSA as well as stand-alone domain II were expressed in Pichia pastoris for ligand-binding studies. The binding of 4Z,15Z-BR to the K195A and K199A mutants was decreased in both whole HSA and the domain II proteins. The P-helicity conformer (P-form) of 4Z,15Z-BR was found to preferentially bind to the wild types and the K195A mutants, whereas the M-form bound to the K199A mutants. Photoconversion experiments showed that the P-form of 4Z,15Z-BR was transformed into highly water-soluble isomers at a much faster rate than the M-form. In addition, the M-form of 4Z,15Z-BR showed higher affinity for domain I than for domain II. The present findings suggest that, whereas both Lys195 and Lys199 in subdomain IIA are important for the high-affinity binding of 4Z,15Z-BR, Lys199 plays a more prominent role in the elimination of 4Z,15Z-BR.

Histidine146 of human serum albumin plays a prominent role at the interface of subdomains IA and IIA in allosteric ligand binding.

Fatty acids are endogenous ligands of human serum albumin (HSA) that induce conformational changes and participate in allosteric ligand binding to HSA. In a previous study, we showed that, when myristate (MYR) is present, the binding of [(14) C]ketoprofen (KP) to subdomain IA of HSA was increased, indicating that, when MYR binds to HSA, a new binding site in formed in that region. Meanwhile, an N-B transition has been reported to increase the binding of ligands at alkaline pH when the status of albumin is the B-conformer. Six histidine single mutants of HSA, H9A, H39A, H67A, H105A, H128A and H146A were produced and photolabeled with [(14) C]KP at pH 6.5, 7.4 and 8.2 and the role of each histidine in causing the N-B transition induced allosteric ligand binding was examined. Cyanogen bromide cleavage of the photolabeled native HSA showed that subdomain IA was the site of the allosteric binding of KP at pH 8.2. From the photolabeling results, H146 was found to play a prominent role whilst H128 played little or no role in the allosteric binding. However, the remaining 4 mutants did not show a clear photolabeling pattern that was similar to either native HSA or H146A and, as a result, no firm conclusions can be made. An additional histidine mutant, H146I, was produced to confirm the results for H146A. A similar experiment using H146I showed that a benzene ring-like structure at position 146 is required for the allosteric ligand binding to occur.