Use of Photoaffinity Labeling and Site-directed Mutagenesis for Identification of the Key Residue Responsible for Extraordinarily High Affinity Binding of UCN-01 in Human α1-Acid Glycoprotein

[Elucidation of Drug Transport Mechanism by Serum Protein and Development for Pancreatic Cancer Treatment]

Human serum albumin (HSA) and α1-acid glycoprotein (AGP) are the major drug-binding proteins in the blood and regulate the tissue transfer of bound drugs. We succeeded in clarifying the three-dimensional structure of AGP for the first time in the world from X-ray crystal structure analysis. Using a site-directed mutagenesis method by constructing yeast expression systems as well as the three-dimensional structure, we elucidated the properties of drug binding sites of AGP. We also found that structural change due to the interaction between AGP and cell membranes causes the release of bound drugs and reported an "AGP-mediated drug transport process." Pancreatic cancer has an extremely low response rate to anticancer drugs compared to other cancers and is resistant to starvation of nutrients including fatty acids. We clarified that glutamine metabolism is involved in this tolerance. Furthermore, aiming at efficient drug delivery and effective treatment for pancreatic cancer, we focused on nitric oxide (NO) which increases pancreatic blood flow and has a cell-killing effect on tumors and surrounding stromal tissues. We successfully synthesized nitrated phenylbutyrate (NPB), which binds to HSA and has an antitumor effect in vitro and vivo. The binding of NPB to HSA is considered to be useful for delivery to tumors through the enhanced permeability and retention (EPR) effect and HSA receptors.

A method for overcoming plasma protein inhibition of tyrosine kinase inhibitors

Plasma protein binding reduces potency of staurosporine-derived tyrosine kinase inhibitors against Flt3-mutant AML. “Decoy” drugs interfering with the binding, including mifepristone, can be harnessed to restore the antileukemia activity.

FMS-like tyrosine kinase 3 (FLT3) is the most frequently mutated gene in acute myeloid leukemia and a target for tyrosine kinase inhibitors (TKI). FLT3 TKIs have yielded limited improvements to clinical outcomes. One reason for this is TKI inhibition by endogenous factors. We characterized plasma protein binding of FLT3 TKI, specifically staurosporine derivatives (STS-TKI) by alpha-1-acid glycoprotein (AGP), simulating its effects upon drug efficacy. Human AGP inhibits the antiproliferative activity of STS-TKI in FLT3/ITD-dependent cells, with IC₅₀ shifts higher than clinically achievable. This is not seen with nonhuman plasma. Mifepristone cotreatment, with its higher AGP affinity, improves TKI activity despite AGP, yielding IC₅₀s predicted to be clinically effective. In a mouse model of AGP drug inhibition, mifepristone restores midostaurin activity. This suggests combinatorial methods for overcoming plasma protein inhibition of existing TKIs for leukemia as well as providing a platform for investigating the drug–protein interaction space for developing more potent small-molecule agents.

Revisiting Chiral Recognition Mechanism on Chicken Alpha 1-Acid Glycoprotein: Location of Chiral Binding Sites and Insight into Chiral Binding Mechanism

Chiral stationary phases based on chicken alpha 1-acid glycoprotein (cAGP) have been used for enantioseparations of various compounds. However, the chiral binding sites and mechanism have not been clarified yet. Based on chromatographic properties of native and W26-modified cAGP columns and docking simulations of studied compounds into the generated model structure of cAGP, the chiral binding sites were located on cAGP and the chiral binding mechanism was discussed. On cAGP, there existed a binding cavity lined with H25, W26, Y47, R128, T129, D161 and E168, which contribute electrostatic or hydrogen bonding interactions. Benzoin and chlorpheniramine enantiomers interacted with cAGP at almost the same sites a little away from W26, while propranolol enantiomers docked, slightly shifting toward H25 and W26. Furthermore, in addition to hydrophobic interactions, ionic interactions between amino groups of chlorpheniramine enantiomers and a carboxy group of D161 or E168 played an important role in the chiral recognition, while hydrophobic interactions and hydrogen bonding interactions worked for the chiral recognition of benzoin and propranolol enantiomers.

Photoaffinity labeling of plasma proteins

Photoaffinity labeling is a powerful technique for identifying a target protein. A high degree of labeling specificity can be achieved with this method in comparison to chemical labeling. Human serum albumin (HSA) and α₁-acid glycoprotein (AGP) are two plasma proteins that bind a variety of endogenous and exogenous substances. The ligand binding mechanism of these two proteins is complex. Fatty acids, which are known to be transported in plasma by HSA, cause conformational changes and participate in allosteric ligand binding to HSA. HSA undergoes an N-B transition, a conformational change at alkaline pH, that has been reported to result in increased ligand binding. Attempts have been made to investigate the impact of fatty acids and the N-B transition on ligand binding in HSA using ketoprofen and flunitrazepam as photolabeling agents. Meanwhile, plasma AGP is a mixture of genetic variants of the protein. The photolabeling of AGP with flunitrazepam has been utilized to shed light on the topology of the protein ligand binding site. Furthermore, a review of photoaffinity labeling performed on other major plasma proteins will also be discussed. Using a photoreactive natural ligand as a photolabeling agent to identify target protein in the plasma would reduce non-specific labeling.

Brazilian propolis-derived components inhibit TNF-α-mediated downregulation of adiponectin expression via different mechanisms in 3T3-L1 adipocytes

BACKGROUND

Previous reports suggest that Brazilian propolis has multiple biological functions and may help to restore adiponectin expression and insulin sensitivity. However, little is known about the molecular mechanisms by which these compounds inhibit the downregulation of adiponectin.

METHODS

The effect of various Brazilian propolis-derived components on inhibition of tumor necrosis factor-α (TNF-α)-mediated downregulation of adiponectin expression in 3T3-L1 adipocytes and molecular mechanism was investigated.

RESULTS AND CONCLUSIONS

Pretreatment with either artepillin C (C3) or its derivative (C4) significantly inhibited TNF-α-mediated downregulation of adiponectin expression in 3T3-L1 adipocytes. Interestingly, C3 strongly activated peroxisome proliferator-activated receptor γ (PPARγ) transcriptional activity. Treatment of adipocytes with C3 resulted in the upregulation of adiponectin and fatty acid-binding protein 4 expression, but C4 did not significantly induce PPARγ transactivation. C4 did, however, inhibit the TNF-α-induced c-Jun-NH(2)-terminal kinase (JNK) signaling that is involved in adiponectin expression. Molecular docking studies based on hPPARγ with C3 and JNK1 with C4 clearly supported our experimental results. These data demonstrate that 1) both C3 and C4 significantly inhibit the TNF-α-mediated downregulation of adiponectin in adipocytes, 2) C3 functions as a PPARγ agonist, and its inhibition of the effect of TNF-α is due to this PPARγ transactivation, and 3) C4 is an effective inhibitor of JNK activation, thus inhibiting the TNF-α-mediated downregulation of adiponectin.

GENERAL SIGNIFICANCE

Brazilian propolis-derived components (C3 and C4) can significantly inhibit TNF-α-mediated downregulation of adiponectin in adipocytes, although they do so via different mechanisms.

Structural insights into differences in drug-binding selectivity between two forms of human alpha1-acid glycoprotein genetic variants, the A and F1*S forms

Human α(1)-acid glycoprotein (hAGP) in serum functions as a carrier of basic drugs. In most individuals, hAGP exists as a mixture of two genetic variants, the F1*S and A variants, which bind drugs with different selectivities. We prepared a mutant of the A variant, C149R, and showed that its drug-binding properties were indistinguishable from those of the wild type. In this study, we determined the crystal structures of this mutant hAGP alone and complexed with disopyramide (DSP), amitriptyline (AMT), and the nonspecific drug chlorpromazine (CPZ). The crystal structures revealed that the drug-binding pocket on the A variant is located within an eight-stranded β-barrel, similar to that found in the F1*S variant and other lipocalin family proteins. However, the binding region of the A variant is narrower than that of the F1*S variant. In the crystal structures of complexes with DSP and AMT, the two aromatic rings of each drug interact with Phe-49 and Phe-112 at the bottom of the binding pocket. Although the structure of CPZ is similar to those of DSP and AMT, its fused aromatic ring system, which is extended in length by the addition of a chlorine atom, appears to dictate an alternative mode of binding, which explains its nonselective binding to the F1*S and A variant hAGPs. Modeling experiments based on the co-crystal structures suggest that, in complexes of DSP, AMT, or CPZ with the F1*S variant, Phe-114 sterically hinders interactions with DSP and AMT, but not CPZ.

A site-directed mutagenesis study of drug-binding selectivity in genetic variants of human alpha(1)-acid glycoprotein

Human alpha(1)-acid glycoprotein (AGP), a major carrier of many basic drugs in circulation, consists of at least two genetic variants, namely A and F1*S variant. Interestingly, the variants of AGP have different drug-binding properties. The purpose of this study was to identify the amino acid residues that are responsible for the selectivity of drug binding to genetic variants of AGP using site-directed mutagenesis. First, we screened amino acid residues in the region proximal to position 100 that are involved in binding of warfarin and dipyridamole, which are F1*S-specific ligands, and of propafenone, which is an A-specific ligand, using ultrafiltration. In the F1*S variant, His97, His100, and Trp122 were involved in either warfarin- or dipyridamole-binding, while Glu92, His100, and Trp122 participated in the binding of propafenone in the A variant. Exchange of the residue at position 92 between AGP variants reversed the relative strength of propafenone binding to the two variants, but had a markedly different effect on binding of warfarin and dipyridamole. These findings indicate that the amino acid residue at position 92 plays a significant role in drug-binding selectivity in AGP variants, especially for drugs that preferentially bind to the A variant.

[Study on binding of drug to serum protein]

After being distributed in the circulating blood, drugs bind to serum proteins varying degrees. In general, such binding is reversible, and a dynamic equilibrium exists between the bound and unbound molecular species. It is believed that unless there is a specific transport system (e.g. receptor-mediated endocytosis, protein-mediated transport), only unbound drugs are able to penetrate through biomembranes, are distributed to tissues, and undergo metabolism and glomerular filtration. It is also believed that only unbound molecules present in target tissues can exert their pharmacological effects, and that the concentration of unbound molecules in tissues is in proportion to the drug serum concentration. Therefore, drug-serum protein binding is critically involved in the manifestation of the pharmacological effects of a drug as well as its pharmacokinetics. Among serum proteins, human serum albumin (HSA) and alpha(1)-acid glycoprotein (AGP) play important roles in protein binding for many drugs, which is of key importance to drug distribution in the body. In addition, they are widely used in clinical settings as blood preparations and drug delivery system carriers. It is thus of great importance from the viewpoint of pharmaceutical science to clarify the structure, function, and pharmaceutical properties of HSA and AGP. Accordingly, since starting my laboratory, the focus of my research has involved molecular pharmaceutical studies on the interactions of drugs and HSA and AGP for the purpose of applying these findings to clinical fields, such as drug treatment, diagnosis and drug discovery. In this review, the molecular properties of HSA and AGP will be briefly outlined. The static and dynamic topology of drug binding sites on these proteins, investigated by various spectroscopic techniques, X-ray crystallography, quantitative structure-activity relationships, molecular modeling, photo affinity labeling, site-directed mutagenesis etc., changes in the serum protein binding of drugs in pathological conditions, such as liver and kidney failure and various inflammation diseases and factors contributing to the changes will then be summarized. Finally, cases in which protein binding displacement can be applied to medical fields will also be introduced.

The drug binding site of human α1-acid glycoprotein: Insight from induced circular dichroism and electronic absorption spectra

Human alpha(1)-acid glycoprotein (AGP) is an important drug binding plasma protein which affects pharmacokinetical properties of various therapeutic agents. For the first time, interpretation of the induced circular dichroism (ICD) spectra of drug-AGP complexes is presented yielding valuable information on the protein binding environment. ICD spectra were obtained by novel ligands of which AGP induced optical activity have never been reported (primaquine, mefloquine, propranolol, terazosin, carbamazepine, rhodamine B) and by re-investigation of ICD spectra of protein-bound drugs published earlier (chlorpromazine, dipyridamole, prazosin). Spectroscopic features of the ICD and absorption bands of drugs combined with native AGP indicated chiral non-degenerate exciton coupling between the guest chromophore and the indole ring of an adjacent tryptophan (Trp) residue. Results of additional CD experiments performed by using recombinant AGP mutants showed no changes in the ligand binding ability of W122A in sharp contrast with the W25A which was unable to induce extrinsic CD signal with either ligand. Thus, these findings unequivocally prove that, likely via pi-pi stacking mechanism, Trp25 is essentially involved in the AGP binding of drugs studied here as well as of related compounds. Survey of the AGP binding data published in the literature support this conclusion. Our results provide a fast and efficient spectroscopic tool to determine the inclusion of ligand molecules into the beta-barrel cavity of AGP where the conserved Trp25 is located and might be useful in ligand-binding studies of other lipocalin proteins.

Selective binding of imatinib to the genetic variants of human α1-acid glycoprotein

Imatinib is a selective tyrosine kinase inhibitor, successfully used for the treatment of chronic myelogenous leukaemia. Its strong plasma protein binding referred to alpha1-acid glycoprotein (AGP) component was found to inhibit the pharmacological activity. AGP shows genetic polymorphism and the two main genetic variants have different drug binding properties. The binding characteristics of imatinib to AGP genetic variants and the possibility of its binding interactions were investigated by various methods. The results proved that binding of imatinib to the two main genetic variants is very different, the high affinity binding belongs dominantly to the F1-S variant. This interaction is accompanied with specific spectral changes (induced circular dichroism, UV change, intrinsic fluorescence quenching), suggesting that the bound ligand has chiral conformation that would largely overlap with other ligands inside the protein cavity. Binding parameters of Ka=1.7(+/-0.2)x10(6)M(-1) and n=0.94 could be determined for the binding on the F1-S variant at 37 degrees . Imatinib binding on the A variant is weaker and less specific. The binding affinity of imatinib to human serum albumin (nKa approximately 3 x 10(4)M(-1)) is low. Pharmacologically relevant binding interactions with other drugs can be expected on the F1-S variant of AGP.

Involvement of disulfide bonds and histidine 172 in a unique beta-sheet to alpha-helix transition of alpha 1-acid glycoprotein at the biomembrane interface

Human alpha(1)-acid glycoprotein (AGP), which is comprised of 183 amino acid residues and 5 carbohydrate chains, is a major plasma protein that binds to basic and neutral drugs as well as to steroid hormones. It has a beta-sheet-rich structure in aqueous solution. Our previous findings suggest that AGP forms an alpha-helix structure through an interaction with biomembranes. We report herein on a study of the mechanism of alpha-helix formation in AGP using various modified AGPs. The disulfide reduced AGP (R-AGP) was extensively unfolded, whereas asialylated AGP (A-AGP) maintained the native structure. Intriguingly, reduced and asialylated AGP (RA-AGP) increased the alpha-helix content as observed in the presence of biomembrane models, and showed a significant decrease in ligand binding capacity. This suggests that AGP has an innate tendency to form an alpha-helix structure, and disulfide bonds are a key factor in the conformational transition between the beta-sheet and alpha-helix structures. However, RA-AGP with all histidine residues chemically modified (HRA-AGP) was found to lose the intrinsic ability to form an alpha-helix structure. Furthermore, disulfide reduction of the H172A mutant expressed in Pichia pastoris also caused a similar loss of folding ability. The present results indicate that disulfide bonds and the C-terminal region, including H172 of AGP, play important roles in alpha-helix formation in the interaction of the protein with biomembranes.

Multiple ligand-binding properties of the lipocalin member chicken α1-acid glycoprotein studied by circular dichroism and electronic absorption spectroscopy: The essential role of the conserved tryptophan residue

Multiple ligand-binding properties of the 30-kDa chicken alpha(1)-acid glycoprotein (cAGP), a member of the lipocalin protein family, were investigated for the first time by using circular dichroism (CD) and UV/Vis absorption spectroscopy methods. By measuring induced CD (ICD) spectra, high-affinity binding (K(a) approximately 10(5)-10(6) M(-1)) of several drugs, dyes and natural compounds to cAGP was demonstrated including antimalarial agents (quinacrine, primaquine), phenotiazines (chlorpromazine, methylene blue), propranolol, non-steroidal antiinflammatory drugs (ketoprofen, diclofenac), tamoxifen, diazepam, tacrine, dicoumarol, cationic dyes (auramine O, thioflavine T, ethidium bromide), benzo[a]pyrene, L-thyroxine, bile pigments (bilirubin, biliverdin), alkaloids (piperine, aristolochic acid), saturated and unsaturated fatty acids. Analysis of the extrinsic CD spectra with the study of the covalently modified protein and CD displacement experiments revealed that a single Trp26 residue of cAGP conserved in the whole lipocalin family is part of the binding site, and it is essentially involved in the ligand-binding process via pi-pi stacking interaction resulting in the appearance of strong induced CD bands due to the non-degenerate intermolecular exciton coupling between the pi-pi* transitions of the stacked indole ring-ligand chromophore. The finding that cAGP is able to accommodate a broad spectrum of ligands belonging to different chemical classes suggests that its core beta-barrel cavity is unusually wide containing overlapping sub-sites. Significance of these new data in understanding of the ligand-binding properties of other lipocalins, especially that of human AGP, and potential practical applications are briefly discussed. Overall, cAGP serves as a simple, ultimate model to extend our knowledge on ligand-binding properties of lipocalins and to study the role of tryptophan residues in molecular recognition processes.

Antitumor activity of ZSTK474, a new phosphatidylinositol 3-kinase inhibitor

BACKGROUND

We previously synthesized a novel s-triazine derivative, ZSTK474 [2-(2-difluoromethylbenzimidazol-1-yl)-4,6-dimorpholino-1,3,5-triazine], that strongly inhibited the growth of tumor cells. We identified its molecular target, investigated its effects on cellular signaling pathways, and examined its antitumor efficacy and toxicity in vivo.

METHODS

We used COMPARE analysis of chemosensitivity measurements from 39 human cancer cell lines and identified phosphatidylinositol 3-kinase (PI3K) as a molecular target for ZSTK474. PI3K was immunoprecipitated from A549 cell lysates, and its activity was measured by assessing the incorporation of 32P into phosphatidylinositol. We used the crystal structure of the PI3K-LY294002 complex to model the binding of ZSTK474 to PI3K (where LY294002 is a known PI3K inhibitor). PI3K downstream activity was analyzed by immunoblotting. Antitumor activity of ZSTK474 was examined against A549, PC-3, and WiDr xenografts in nude mice. Phosphorylation of Akt, a serine/threonine protein kinase and a major signaling component downstream of PI3K, was assessed in vivo by immunohistochemistry.

RESULTS

PI3K was identified as a molecular target for ZSTK474 by COMPARE analysis. We confirmed that ZSTK474 directly inhibited PI3K activity more efficiently than the PI3K inhibitor LY294002. At concentrations of 1 microM, ZSTK474 and LY2194002 reduced PI3K activity to 4.7% (95% confidence interval [CI] = 3.2% to 6.1%) and 44.6% (95% CI = 38.9% to 50.3%), respectively, of the untreated control level. Molecular modeling of the PI3K-ZSTK474 complex indicated that ZSTK474 could bind to the ATP-binding pocket of PI3K. ZSTK474 inhibited phosphorylation of signaling components downstream from PI3K, such as Akt and glycogen synthase kinase 3beta, and mediated a decrease in cyclin D1 levels. ZSTK474 administered orally to mice had strong antitumor activity against human cancer xenografts without toxic effects in critical organs. Akt phosphorylation was reduced in xenograft tumors after oral administration of ZSTK474.

CONCLUSION

ZSTK474 is a new PI3K inhibitor with strong antitumor activity against human cancer xenografts without toxic effects in critical organs. ZSTK474 merits further investigation as an anticancer drug.

A molecular functional study on the interactions of drugs with plasma proteins

The binding of drugs to plasma proteins, such as albumin and alpha1-acid glycoprotein (AGP) is a major determinant in the disposition of drugs. A topology analysis of drug binding sites on HSA and AGP was determined using various methods, including spectroscopy, QSAR, photoaffinity labeling and site directed mutagenesis. Recombinant albumin was found to be useful for rapidly identifying drug binding sites. The binding sites on AGP are not completely separated but are partially overlapped, and Trp, Tyr, Lys and His residues in the drug binding pockets play important roles in this process. Drug displacement is somewhat complex, due to the involvement of multiple effects. The reduced binding in uremic patients may be explained by a mechanism that involves a combination of direct displacement by free fatty acids as well as cascade effects of free fatty acids and unbound uremic toxins for significant inhibition in serum binding. Albumin-containing dialysate is useful for the extracorporeal removal of endogenous toxins and in the treatment of drug overdoses. Oxidized albumin is a useful biomarker for the quantitative and qualitative evaluation of oxidative stress. Interestingly, AGP undergoes a structural transition to a unique structure that differs from the native and denatured states, when it interacts with membranes.

Comparison of Consensus Scoring Strategies for Evaluating Computational Models of Protein−Ligand Complexes

Here, the comparisons of performance of nine consensus scoring strategies, in which multiple scoring functions were used simultaneously to evaluate candidate structures for a protein-ligand complex, in combination with nine scoring functions (FlexX score, GOLD score, PMF score, DOCK score, ChemScore, DrugScore, PLP, ScreenScore, and X-Score), were carried out. The systematic naming of consensus scoring strategies was also proposed. Our results demonstrate that choosing the most appropriate type of consensus score is essential for model selection in computational docking; although the vote-by-number strategy was an effective selection method, the number-by-number and rank-by-number strategies were more appropriate when computational tractability was taken into account. By incorporating these consensus scores into the FlexX program, reasonable complex models can be obtained more efficiently than those selected by independent FlexX scores. These strategies might also improve the scoring of other docking programs, and more-effective structure-based drug design should result from these improvements.

Characterization of benzodiazepine binding site on human α1-acid glycoprotein using flunitrazepam as a photolabeling agent

The binding of flunitrazepam (FNZP) by human alpha1-acid glycoprotein (hAGP) and the relationships between the extent of drug binding and desialylation and the genetic variants of hAGP were examined. The photolabeling specificity of [3H]FNZP was confirmed by findings in which other hAGP-binding ligands inhibited the formation of covalent bonds between [3H]FNZP and hAGP. The photolabeling of asialo-hAGP suggested that sialic acid does not involve in the binding of [3H]FNZP. No difference in the labeling could be found between the F1*S variants and A variant. Similarly, FNZP did not show a difference in binding affinity to the two genetic variants of hAGP. Sequence analysis of the photolabeled peptide indicated a sequence corresponding to Tyr91-Arg105 of hAGP.

The relationship between the pharmacology of antiepileptic drugs and human gene variation: an overview

Individual differences in clinical responsiveness to antiepileptic drugs are due to a complex interaction between environmental factors and genetic variation. Considerable interest has arisen in exploiting advances in molecular genetics to improve drug therapy for epilepsy and many other diseases; however, practical application of pharmacogenetics has been difficult to realize. Attempts to define gene variants that are associated with therapeutic (or adverse) effects of antiepileptic drugs rely currently on the prior identification of candidate genes and the subsequent evaluation of the distribution of allelic variants between individuals who have a "good" versus a "poor" clinical response. Many factors can adversely affect interpretation of such data, and careful consideration must be given to the design of genetic association studies involving candidate genes. Candidate genes may be identified in a number of ways; however, for studies of drugs, application of knowledge derived from basic pharmacology can suggest focused and testable hypotheses that are based on the fundamental principles of drug action. Thus, studies of genetic variation as they relate to proteins involved in antiepileptic drug kinetics and dynamics will identify key polymorphisms in endogenous molecules that determine degrees of drug efficacy and toxicity. Delineation of these effects in the coming years will promote enhanced success in the treatment of epilepsy.