Kuniaki Sano

Modulation of AP-1 activity by nitric oxide (NO) in vitro: NO-mediated modulation of AP-1

To understand the role of nitric oxide (NO) in controlling the specific DNA‐binding activities of transcriptional factors, we investigated the in vitro effect of the NO‐donor sodium nitroprusside (SNP) on the AP‐1 activity of cultured mouse cerebellar granule cells. A gel‐mobility assay showed that SNP inhibited AP‐1 activity in the presence, but not the absence of dithiothreitol (DTT). This DTT‐dependent inhibition of AP‐1 activity by SNP corresponded with the activation of the chemical reactivity of SNP with DTT, which can be monitored by the production of nitrite (NO− 2). In contrast, diamide, a typical sulfhydryl oxidizing agent, inhibited AP‐1 activity in the absence of DTT and its inhibitory effect was reversed competitively by DTT. Studies using structurally or functionally related analogues of SNP demonstrated that S‐nitrosylation of the AP‐1 moiety mediated by some NO‐carriers but not by free NO, which can be produced by the chemical reaction of SNP with DTT, was responsible for the inhibition of AP‐1 activity, suggesting NO‐mediated regulation of the AP‐1 transcriptional factor.

Stimulation of Cultured Cerebellar Granule Cells via Glutamate Receptors Induces TRE‐ and CRE‐Binding Activities Mediated by Common DNA‐Binding Complexes

Abstract: By use of nuclear mini‐extracts prepared from cultured cerebellar granule cells in a gel‐mobility assay, exogenous N‐methyl‐D‐aspartate (NMDA) or kainate was shown to increase both 12‐O‐tetradecanoylphorbol 13‐acetate‐responsive element (TRE)‐ and cyclic AMP‐responsive element (CRE)‐binding activity. These increases were specifically prevented by the NMDA receptor antagonist D,L‐2‐amino‐5‐phosphonovalerate and the non‐NMDA receptor antagonist 6‐cyano‐7‐nitroquinoxaline‐2,3‐dione, respectively. The increase of TRE‐binding activity was dependent on de novo protein synthesis, and its inductions by both NMDA and kainate required extracellular Ca2+. TRE‐binding activity was competitively inhibited by the CRE, and vice versa, showing higher DNA‐binding affinity to the CRE than to the TRE. A proteolytic clipping bandshift assay demonstrated that the increase in CRE‐binding activity could be mediated by the TRE‐binding activity. Thus, the TRE‐binding activity cross‐binding to the CRE could be activated by NMDA or kainate stimulation. The involvement of c‐Fos or Fos‐related proteins in the TRE‐ and CRE‐binding complexes was shown by a supershift gel‐mobility assay using anti‐c‐Fos antiserum.

Immunohistochemical analyses of DNA topoisomerase II isoforms in developing rat cerebellum.

In mammalian cells, two isoforms of DNA topoisomerase II (topo IIα and topo IIβ) have been identified. Topo IIα is essential in mitotic cells, whereas the function of topo IIβ remains unclear. In the present study, we investigated the developmental control of topo II isoforms in two different neuronal lineages, cerebellar Purkinje cells and granule cells, by immunohistochemical analysis with isoform‐specific monoclonal antibodies. As expected, proliferating cells in the neuroepithelium and in the external germinal layer (EGL) were topo IIα immunopositive. The migrating as well as differentiating Purkinje cells and granule cells showed an enhanced topo IIβ immunoreactivity. The postmitotic granule cells in the postnatal EGL showed an abrupt transition of expressed topo II isoforms from IIα to IIβ. The transition was clearly coincident with the completion of final cell division and the initiation of terminal differentiation because no increase of the topo IIβ immunoreactivity was observed in the spreading EGL cells that are still in the cell division cycle. The topo IIβ signal was detected in both nucleoplasm and nucleolus of differentiating cells. However, the nucleoplasmic signal decreased significantly as the cells reached terminal differentiation. The residual topo IIβ in nucleoli was shown to occupy an unique location with respect to other nucleolar proteins, nucleolin and DNA topoisomerase I. Our findings indicate that both Purkinje cells and granule cells express the topo II isoforms in a similar timing during the cerebellar development and also suggest that topo IIβ localized in nucleoplasm is the functional entity involved in neuronal differentiation. J. Comp. Neurol. 431:228–239, 2001.

Topoisomerase IIβ Activates a Subset of Neuronal Genes that Are Repressed in AT-Rich Genomic Environment

DNA topoisomerase II (topo II) catalyzes a strand passage reaction in that one duplex is passed through a transient brake or gate in another. Completion of late stages of neuronal development depends on the presence of active β isoform (topo IIβ). The enzyme appears to aid the transcriptional induction of a limited number of genes essential for neuronal maturation. However, this selectivity and underlying molecular mechanism remains unknown. Here we show a strong correlation between the genomic location of topo IIβ action sites and the genes it regulates. These genes, termed group A1, are functionally biased towards membrane proteins with ion channel, transporter, or receptor activities. Significant proportions of them encode long transcripts and are juxtaposed to a long AT-rich intergenic region (termed LAIR). We mapped genomic sites directly targeted by topo IIβ using a functional immunoprecipitation strategy. These sites can be classified into two distinct classes with discrete local GC contents. One of the classes, termed c2, appears to involve a strand passage event between distant segments of genomic DNA. The c2 sites are concentrated both in A1 gene boundaries and the adjacent LAIR, suggesting a direct link between the action sites and the transcriptional activation. A higher-order chromatin structure associated with AT richness and gene poorness is likely to serve as a silencer of gene expression, which is abrogated by topo IIβ releasing nearby genes from repression. Positioning of these genes and their control machinery may have developed recently in vertebrate evolution to support higher functions of central nervous system.

Nuclear protein LEDGF/p75 recognizes supercoiled DNA by a novel DNA-binding domain

Lens epithelium-derived growth factor (LEDGF) or p75 is a co-activator of general transcription and also involved in insertion of human immunodeficiency virus type I (HIV-1) cDNA into host cell genome, which occurs preferentially to active transcription units. These phenomena may share an underlying molecular mechanism in common. We report here that LEDGF/p75 binds negatively supercoiled DNA selectively over unconstrained DNA. We identified a novel DNA-binding domain in the protein and termed it ‘supercoiled DNA-recognition domain’ (SRD). Recombinant protein fragments containing SRD showed a preferential binding to supercoiled DNA in vitro. SRD harbors a characteristic cluster of lysine and glutamic/aspartic acid residues. A polypeptide mimicking the cluster (K9E9K9) also showed this specificity, suggesting that the cluster is an essential element for the supercoil recognition. eGFP-tagged LEDGF/p75 expressed in the nucleus distributed partially in transcriptionally active regions that were identified by immunostaining of methylated histone H3 (H3K4me3) or incorporation of Br-UTP. This pattern of localization was observed with SRD alone but abolished if the protein lacked SRD. Thus, these results imply that LEDGF/p75 guides its binding partners, including HIV-1 integrase, to the active transcription site through recognition of negative supercoils generated around it.

The SUMO pathway is required for selective degradation of DNA topoisomerase IIβ induced by a catalytic inhibitor ICRF-1931

DNA topoisomerase I and II have been shown to be modified with a ubiquitin‐like protein SUMO in response to their specific inhibitors called ‘poisons’. These drugs also damage DNA by stabilizing the enzyme–DNA cleavable complex and induce a degradation of the enzymes through the 26S proteasome system. A plausible link between sumoylation and degradation has not yet been elucidated. We demonstrate here that topoisomerase IIβ, but not its isoform IIα, is selectively degraded through proteasome by exposure to the catalytic inhibitor ICRF‐193 which does not damage DNA. The β isoform immunoprecipitated from ICRF‐treated cells was modified by multiple modifiers, SUMO‐2/3, SUMO‐1, and polyubiquitin. When the SUMO conjugating enzyme Ubc9 was conditionally knocked out, the ICRF‐induced degradation of topoisomerase IIβ did not occur, suggesting that the SUMO modification pathway is essential for the degradation.

cDNA cloning and expression of mutant catalase from the hypocatalasemic mouse: comparison with the acatalasemic mutant.

Mutant catalase cDNAs from the hypocatalasemic and acatalasemic mice were cloned and expressed in bacteria. A novel missense mutation, Asp (AAT) to Ser (AGT), was identified at amino acid position 439 of the hypocatalasemic catalase. Analysis of recombinant catalase mutants revealed that the mutation is responsible for the reduced activity of hypocatalasemic catalase and the unstable tetrameric structure of acatalasemic catalase was also suggested.

Involvement of oxidative stress in hydroquinone-induced cytotoxicity in catalase-deficient Escherichia coli mutants

Hydroquinone is a benzene-derived metabolite. To clarify whether the reactive oxygen species (ROS) are involved in hydroquinone-induced cytotoxicity, we constructed transformants of Escherichia coli (E. coli) strains that express mammalian catalase gene derived from catalase mutant mice (Csb, Csc) and the wild-type (Csa) using a catalase-deficient E. coli UM255 as a recipient. Specific catalase activities of these tester strains were in order of Csa > Csc > Csb > UM255, and their susceptibility to hydrogen peroxide (H2O2) showed UM255 > Csb > Csc > Csa. We found that hydroquinone exposure reduced the survival of catalase-deficient E. coli mutants in a dose-dependent manner significantly, especially in the strains with lower catalase activities. Hydroquinone toxicity was also confirmed using zone of inhibition test, in which UM255 was the most susceptible, showing the largest zone of growth inhibition, followed by Csb, Csc and Csa. Furthermore, we found that hydroquinone-induced cell damage was inhibited by the pretreatment of catalase, ascorbic acid, dimethyl sulfoxide (DMSO), and ethylenediaminetetraacetic acid (EDTA), and augmented by superoxide dismutase (both CuZnSOD and MnSOD). The present results suggest that H2O2 is probably involved in hydroquinone-induced cytotoxicity in catalase-deficient E. coli mutants and catalase plays an important role in protection of the cells against hydroquinone toxicity.

Nuclear dynamics of topoisomerase IIβ reflects its catalytic activity that is regulated by binding of RNA to the C-terminal domain

DNA topoisomerase II (topo II) changes DNA topology by cleavage/re-ligation cycle(s) and thus contributes to various nuclear DNA transactions. It is largely unknown how the enzyme is controlled in a nuclear context. Several studies have suggested that its C-terminal domain (CTD), which is dispensable for basal relaxation activity, has some regulatory influence. In this work, we examined the impact of nuclear localization on regulation of activity in nuclei. Specifically, human cells were transfected with wild-type and mutant topo IIβ tagged with EGFP. Activity attenuation experiments and nuclear localization data reveal that the endogenous activity of topo IIβ is correlated with its subnuclear distribution. The enzyme shuttles between an active form in the nucleoplasm and a quiescent form in the nucleolus in a dynamic equilibrium. Mechanistically, the process involves a tethering event with RNA. Isolated RNA inhibits the catalytic activity of topo IIβ in vitro through the interaction with a specific 50-residue region of the CTD (termed the CRD). Taken together, these results suggest that both the subnuclear distribution and activity regulation of topo IIβ are mediated by the interplay between cellular RNA and the CRD.

Novel splice variants of amphiphysin I are expressed in retina.

Using RT‐PCR‐based cDNA cloning, we identified novel splice variants of amphiphysin I, termed amph Ir, that are expressed specifically in retina. In comparison with the prototype amphiphysin I, amph Ir contained two novel insertions (inserts A and B) and one deletion. Insert A is only 9 bp in length but appears to be a determinant for the retina‐specific expression. In contrast, insert B is a large domain of 1740 bp and two shorter transcripts with 3′‐truncated insert B were also expressed. All the insert sequences were present as unidentified exons in the amphiphysin I gene on human chromosome 7. Western blot analysis of various rat tissues with anti‐insert B antibody confirmed the presence and tissue specificity of the variant proteins.