Structural and physical properties of type wild type and various selected mutants of the vnd/NK-2 homeodomain, the protein product of the homeobox, and the implication in genetic disease are reviewed. The structure, dynamics and thermodynamics have been Investigated by NMR and by calorimetry. The interactions responsible for the nucleotide sequence-specific binding of the homeodomain to its consensus DNA binding site have been identified. There is a strong correlation between significant structural alterations within the homeodomain or its DNA complex and the appearance of genetic disease. Mutations in positions known to be important in genetic disease have been examined carefully For example, mutation of position 52 of vnd/NK-2 results in a significant structural modification and mutation of position 54 alters the DNA binding specificity and amity The $^{15}N$ relaxation behavior and heteronuclear Overhauser effect data was used to characterize and describe the protein backbone dynamics. These studies were carried out on the wild type and the double mutant proteins both in the free and in the DNA bound states. Finally, the thermodynamic properties associated with DNA binding are described for the vnd/NK-2 homeodomain. These thermodynamic measurements reinforce the hypothesis that water structure around a protein and around DNA significantly contribute to the protein-DNA binding behavior. The results, taken together, demonstrate that structure and dynamic studies of proteins combined with thermodynamic measurements provide a significantly more complete picture of the solution behavior than the individual studies.

One of the innate immune reactions in invertebrates is the pro-phenoloxidase (pro-PO) activation system that is involved in the generation of superoxide, melanin synthesis, and the subsequent sequestration of foreign matter entering the hemocoel of the invertebrates. However, the molecular mechanism of this biological reaction is still obscure. To expand our understanding of the biological roles of the pro-PO activation system in invertebrates, we performed a yeast two-hybrid screening by using three regions of pro-PO as bait and a yeast two-hybrid cDNA library from Tenebrio molitor larvae as prey We isolated a novel partial cDNA clone that encodes a glycine-rich protein that interacted with the active phenoloxidase (termed phenoloxidase interacting protein, POIP). POIP consists of two domains: One is an N-terminal unique domain and the other is a C-terminal glycine-rich domain. The C-terminal glycine-rich domain showed sequential homology with those of insect antifungal proteins. Also, the yeast two-hybrid screen in a reverse orientation (using POIP as bait) yielded PO, suggesting that the PO-POIP interaction is specific. By using a 315 bP PCR fragment of the N-terminal unique region of POIP, we cloned the full-length cDNA of POIP from the Tenebruo cDNA library constructed by using E. coli injected larvae. The interaction analysis between PO, and a truncated fragment lacking the N-terminal unique region of POIP, indicated that the N-terminal unique region is necessary for interaction between PO and POIP. The expression level of the POIP mRNA is increased by bacterial injection into T. molitor larvae. This suggests that POIP might be engaged in the humoral defense reaction.

The structural stability of brain pyrydoxine-5'-phosphate (PNP) oxidase and the catalytic properties of the monomeric species were investigated. The unfolding of brain pyridoxine-5'-phosphate (PNP) oxidase by guanidine hydrochloride (GuHCl) was monitored by means of fluorescence and circular dichroism spectroscopy Reversible dissociation of the dimeric enzyme into subunits was attained by the addition of 2 M GuHCl. The perturbation of the secondary structure under the denaturation condition resulted in the release of the cofactor FMN. Separation of the processes of refolding and reassociation of the monomeric species was achieved by the immobilization method. Dimeric PNP oxidase was immobilized by the covalent attachment to Affi-gel 15 without any significant lass of its catalytic activity. Matrix-bound monomeric species were obtained from the reversible refolding processes. The matrix bound-monomer was found to be catalytically active, possessing only a slightly decreased specific activity when compared to the refolded dimeric enzyme. In addition, limited chymotrypsin digestion of the oxidase yields two fragments of 12 and 161 kDa with a concomitant increase of catalytic activity The catalytically active fragment was isolated by ion exchange chromatography and analyzed for association of two subunits using the FPLC gel filtration analysis. The retention time indicated that the catalytic fragment of 16 kDa behaves as a compact monomer. Taken together, these results are consistent with the hypothesis that the native quaternary structure of PNP oxidase is not a prerequisite for catalytic function, but it could play a role in the regulation.

The Fos-Jun heterodimers are part of the regulatory network of gene expression and nuclear proteins encoded by proto-oncogenes. The activation of Fos-Jun is important in the transmission of the tumor-promoting signal from the extracellular environment to the nuclear transcription mechanism. To search for the inhibitors of the Fos-Jun DNA complex formation, several natural products were screened and water-soluble paeoniflorin reduced the binding activity of the Fos-Jun heterodimer. This active compound was purified by silica gel column chromatography and HPLC. The electrophoresis mobility shift assay and reverse-phase HPLC test showed that paeoniflorin reduced the AP-l function. The cytotoxic effect of paeoniflorin was observed in HL-60. These results indicate that paeoniflorin blocks the Fos-Jun heterodimer-binding site of the AP-l DNA and it also has cytotoxic effects on human leukemia cell lines.

The human PTK6 (also known as Brk) polypeptide, which is deduced from its full-length cDNA, represents a non-receptor protein tyrosine kinase (PTK). It contains SH3, SH2, and tyrosine kinase catalytic domains that are closely related to Src family members. We generated an antihuman PTK6 antibody by immunizing rabbits with a PTK6-specific oligopeptide conjugated to BSA, which corresponds to 11 amino acid residues near the C-terminus. An immunoblot analysis with the antibody detected an expected 52-kDa band in various mammalian transformed cell lines. Immunoprecipitation and immunoblot analyses demonstrated that PTK6 is phosphorylated on the tyrosine residues) and interacts with approximately a 23-kDa tyrosine-phosphorylated polypeptide (most likely a substrate of PTK6) in breast carcinoma T-47D cells. An immunofluorescence analysis demonstrated that PTK6 is localized throughout the cytoplasm of T-47D cells. These results support a possible role for PTK6 in the intracellular signal transduction through tyrosine phosphorylation.

Previous studies indicate that B. subtilis protoporphyrinogen oxidase is poorly inhibited by diphenyl ether herbicides. To better understand the basis of this insensitivity, the enzyme was overexpressed as a soluble protein in E. coli, purified and characterized. The mechanism of oxidation of B. subtilis protoporphyrinogen IX was studied and the enzyme kinetic parameters were determined for protoporpyrinogen IX; $K_m$, and $k_{cat}$ were $6.3\;{\mu}M$ and $0.028\;h-^1$, respectively. The enzyme required flavin adenine dinucleotide as a cofactor and its activity was enhanced by 1 mM n-octylglucopyranoside. The nonenzymatic oxidation rate was dependent on the concentration of protoporphyrinogen IX, suggesting that the reaction involves a pre-equilibrium step followed by a rate-limiting step.

$\alpha$-Phenyl-N-t-butylnitrone (PBN) is one of the most widely used spin-trapping compounds for investigating the existence of free radicals in biological systems. Recently, there has been considerable interest in the antioxidant nature of PBN on degenerative diseases, presumably related to oxidative stress. In the present study, the protective effect of PBN on the HepG2 cell line under oxidative stress was investigated. When the HepG2 cells were exposed to oxidant, such as hydrogen peroxide, menadione, or ethanol, the protective role of PBN was manifested as a reduction in trypan blue uptake and a decrease in the endogenous production of oxidants, as measured by the oxidation of 2',7'-dichlorodihydrofluorescin. The modulation of activity of major antioxidant enzymes, such as superoxide dismutase and catalase, was not significantly different either in the presence or in the absence of PBN. This indicates that PBN acts as a direct scavenger of reactive oxygen species.

A one-step, two-site enzyme immunoassay was developed for measuring human alpha-fetoprotein (AFP) in serum and amniotic fluid using monoclonal antibodies (McAb) by eliminating the high-dose hook effect. Three McAbs that recognize different epitopes were selected among 16 different clones on the basis of epitope mapping, two for immobilization and one for horseradish peroxidase conjugation. This one-step immunoassay system is more convenient and rapid compared to a conventional two-step sandwich immunoassay system. It did not exhibit the hook effect to around 2.7 mg/ml of AFP, which is probably one of the highest concentrations of AFP in the serum. The dose-response curve of the system was linear to 500 mg/ml of AFP and the system could differentiate as low as 1 mg/ml of AFP The intra- and inter-assay variations were in an acceptable range; 95~104% and 97~105% respectively Its correlation with other commercial systems was around 95%.

Attempts using in vitro and in vivo selection procedures have been made to search for hammerhead ribozymes that have higher activities than the wild-type ribozyme and also to determine whether other sequences might be possible in the catalytic core of the hammerhead ribozyme. Active sequences selected in the past conformed broadly to the consensus core sequence except at A9, and no sequences were associated with higher activity than that of the hammerhead with the consensus core, an indication that the consensus sequence derived from viruses and virusoids is probably the optimal sequence [Vaish et al. (1997) Biochemistry 36, 6495-6501]. Recently, during construction of ribozyme expression vectors, we isolated a mutant hammerhead ribozyme, with an insertion of G between A9 and G10.1, that appeared to show significant activity [Kawasaki et al. (1996) Nucleic Acids Res. 24, 3010-3016; Kawasaki et al. (1998) Nature 393, 284-289]. We, therefore, characterized kinetic properties of the G-inserted mutant ribozymes in terms of the NUX rule. We demonstrate that the NUX rule is basically applicable to the G-inserted ribozymes and, more importantly, one type of G-inserted ribozyme was very active with $k_{cat}$, value of $6.4\;min^{-1}$ in 50 mM Tris-HCl (pH 8.0) and 10 mM $MgCl_2$ at $37^{\circ}C$.

In subtractive hybridization, target sequences in the tester are enriched by hybridizing with an excess amount of driver, followed by removing the tester hybridized with the driver. All of existing subtractive cloning methods are designed to remove the tester/driver hybrid. The removal of hybrid, however, is often unsatisfactory For various reasons. In this study we developed a subtractive enrichment protocol in which the tester/driver can be completely removed by selecting only the tester/tester after hybridization. In this protocol both the tester and driver DNAs are ligated with same linker DNAs and amplified by polymerase chain reaction (PCR). The tester DNA is then digested with two different enzymes and used in subsequent hybridization with an excess driver. After hybridization, the DNA is ligated with the adaptor that is only compatible with the tester/tester. Since only the tester/tester can have the new adaptor, no tester/driver can be amplified by PCR in this protocol. Unlike other methods, a 100% subtraction efficiency can be achieved even though the enzymatic treatments used in the enrichment procedure are incomplete. Furthermore, only the hybridized tester DNA can have the new adaptor and be amplified by PCR, resulting in 100% denaturation in effect. The efficacy of this novel method was verified with the model system in which a known amount of the target sequence is included.

To investigate the mechanism of adriamycin (ADM) resistance in the ADM resistant subline PC-14/ADM, we examined the expressions of p-glycoprotein (P-gp), topoisomerase I (Topo I) and II (Topo II), glutathione-S-transferases (GSTs), tissue transglutaminase (t-TG), epidermal growth factor receptor (EGFR), and E-cadherin and the activity of superoxide dismutase (SOD) in PC-14 and PC-14/ADM cells. There was no change in the cellular levels of P-gp, Topo I, Topo II, and the two isoforms of GSTs. However, SOD activity in PC-14/ADM cells was 2.38 fold higher than that in PC-14 cells. A marked induction of the t-TG expression was also observed in PC-14/ADM cells. In addition to those changes, expressions of EGFR and E-cadherin were down regulated in PC-14/ADM cells. Therefore, molecular modifications such as an increase in SOD activity, induction of the t-TG expression, and down regulation of EGFR and E-cadherin expressions may play important roles in PC-14/ADM cells during the development of ADM resistance.

Transcriptional activation domains are known to be inherently "unstructured" with no tertiary structure. A recent NMR study, however, has shown that the transactivation domain in human p53 is populated with an amphipathic helix and two nascent turns. This suggests that the presence of such local secondary structures within the overall "unstructured" structural framework is a general feature of acidic transactivation domains. These pre-existing local structures in p53, formed selectively by positional conserved hydrophobic residues that are known to be critical for transcriptional activity, thus appear to constitute the specific structural motifs that regulate recognition of the p53 transactivation domain by target proteins. Here, we report the results of a NMR structural comparison between the native human p53 transactivation domain and an inactive mutant (22L,23W$\rightarrow$22R,23S). Results show that the mutant has an identical overall structural topology as the native protein, to the extent that the amphipathic helix formed by the residues 18T 26L within the native p53 transactivating domain is preserved in the double mutant. Therefore, the lack of transcriptional activity in the double mutant should be ascribed to the disruption of the essential hydrophobic contacts between the p53 transactivation domain and target proteins due to the (22L,23W$\rightarrow$22R,23S) mutation.

The human papillomavirus E7 protein can form a specific complex with a retinoblastoma tumor suppressor gene product (p105-Rb) that results in the release of the E2F transcription factor, which is critical for the growth-deregulation and transforming properties of the viral E7 oncoprotein. In an attempt to apply interaction between the E7 oncoprotein and a target cellular protein Rb for an in vitro screening system for drugs against human papillomavirus infection, we primarily investigated the E7Rb binding through a pull down assay and enzyme-linked immunosorbent assay. The pull down assay showed that both glutathione S-transferase-tagged E7 and His-tagged E7 immobilized on resins specifically produced complexes with bacterially expressed Rb in a dose-dependent manner, as determined by immunoblot analyses. This result coincided with that of an enzyme-linked immunosorbent assay, which is a useful system for the mass screening of potential drugs. Taken together, this screening system (based on the interaction between E7 and Rb) can be a promising system in the development of drugs against cervical cancers caused by human papillomavirus infection.

Both topoisomerase $II{\alpha}$ and $II{\beta}$ east as phosphoproteins in the cells. Recently it was reported that DNA topoisomerase $II{\alpha}$ associates with and is phosphorylated by the extracellular signal-regulated kinase 2 (ERK2). Also, ERK2 stimulates the activity of topoisomerase II by a phosphorylation-independent manner [Shapiro et al., (1999) Mol. Cell. Biol. 19, 3551-3560]. In this study, a yeast two-hybrid system was used to investigate the binding site between topoisomerase $II{\alpha}$ or $II{\beta}$ and ERK2. The two-hybrid test clearly showed that topoisomerase $II{\beta}$ residues 1099-1263, and topoisomerase $II{\alpha}$ residues 1078-1182, mediate the interaction with ERK2, and that the leucine zipper motifs of topoisomerase $II{\alpha}$ and $II{\beta}$ are not required for its physical binding to ERK2. Our results suggest that topoisomerase $II{\beta}$ residues 1099-1263, and topoisomerase $II{\alpha}$ residues 1078-1182, may be common binding sites for activator proteins.

The mutagenic activity of chitosan oligosaccharide and its antimutagenic effect against 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) were investigated using the Salmonella/Ames test. No mutagenic activity was found in the Salmonella typhimurium strains TA 98 and TA 100, either with or without S9 activation. In contrast, chitosan oligosaccharide showed an inhibitory effect on the mutagenic activity of the cooked food mutagen, MeIQx, in the presence of S9. The influence of chitosan oligosaccharide on the genotoxicity of MeIQx was examined using a host-mediated assay in mice. The oligosaccharide was administered for 14 consecutive days (intragastric application at doses of 0.1 or 0.5 g/kg body wt) to mice. S. typhimurium TA 98 was given intravenously before an oral dose of MeIQx (4.5 mg/kg body wt.). The number of $his^+$ revertants were determined from the Ever of mice. The intragastric application of oligosaccharide led to a 47% reduction in the number of mutants induced by MeIQx (p<0.05). These results suggested that chitosan oligosaccharide had antimutagenic properties against MeIQx in vitro and in vivo.