HDAC4 represses p21(WAF1/Cip1) expression in human cancer cells through a Sp1-dependent, p53-independent mechanism.

Cancer cells have complex, unique characteristics that distinguish them from normal cells, such as increased growth rates and evasion of anti-proliferative signals. Global inhibition of class I and II histone deacetylases (HDACs) stops cancer cell proliferation in vitro and has proven effective against cancer in clinical trials, at least in part, through transcriptional reactivation of the p21(WAF1/Cip1)gene. The HDACs that regulate p21(WAF1/Cip1) are not fully identified. Using small interfering RNAs, we found that HDAC4 participates in the repression of p21(WAF1/Cip1) through Sp1/Sp3-, but not p53-binding sites. HDAC4 interacts with Sp1, binds and reduces histone H3 acetylation at the Sp1/Sp3 binding site-rich p21(WAF1/Cip1) proximal promoter, suggesting a key role for Sp1 in HDAC4-mediated repression of p21(WAF1/Cip1). Induction of p21(WAF1/Cip1) mediated by silencing of HDAC4 arrested cancer cell growth in vitro and inhibited tumor growth in an in vivo human glioblastoma model. Thus, HDAC4 could be a useful target for new anti-cancer therapies based on selective inhibition of specific HDACs.

CREB-1 and AP-1 transcription factors JunD and Fra-2 regulate bone sialoprotein gene expression in human breast cancer cells.

Bone sialoprotein (BSP) expression is detected in a variety of human osteotropic cancers. High expression of BSP in breast and prostate primary carcinomas is associated with progression and bone metastases development. In this study, we examined the transcriptional regulation of BSP gene expression in MDA-MB-231 and MCF-7 human breast cancer cells compared with Saos-2 human osteoblast-like cells. BSP human promoter deletion analyses delineated a -56/-84 region, which comprises a cAMP response element (CRE) that was sufficient for maximal promoter activity in breast cancer cell lines. We found that the basic fibroblast growth factor response element (FRE) also located in the proximal promoter was a crucial regulator of human BSP promoter activity in Saos-2 but not in breast cancer cells. Promoter activity experiments in combination with DNA mobility shift assays demonstrated that BSP promoter activity is under the control of the CRE element, through CREB-1, JunD and Fra-2 binding, in MDA-MB-231, MCF-7 and in Saos-2 cells. Forskolin, a protein kinase A pathway activator, failed to enhance BSP transcriptional activity suggesting that CRE site behaves as a constitutive rather than an inducible element in these cell lines. Over-expression of JunD and Fra-2 increased BSP promoter activity and upregulated endogenous BSP protein expression in MCF-7 and Saos-2 cells while siRNA-mediated inhibition of both factors expression significantly reduced BSP protein level in MDA-MB-231. Collectively, these data provide with new transcriptional mechanisms, implicating CREB and AP-1 factors, that control BSP gene expression in breast cancer cells.

Ultrahigh resolution drug design I: details of interactions in human aldose reductase-inhibitor complex at 0.66 A.

The first subatomic resolution structure of a 36 kDa protein [aldose reductase (AR)] is presented. AR was cocrystallized at pH 5.0 with its cofactor NADP+ and inhibitor IDD 594, a therapeutic candidate for the treatment of diabetic complications. X-ray diffraction data were collected up to 0.62 A resolution and treated up to 0.66 A resolution. Anisotropic refinement followed by a blocked matrix inversion produced low standard deviations (<0.005 A). The model was very well ordered overall (CA atoms' mean B factor is 5.5 A2). The model and the electron-density maps revealed fine features, such as H-atoms, bond densities, and significant deviations from standard stereochemistry. Other features, such as networks of hydrogen bonds (H bonds), a large number of multiple conformations, and solvent structure were also better defined. Most of the atoms in the active site region were extremely well ordered (mean B approximately 3 A2), leading to the identification of the protonation states of the residues involved in catalysis. The electrostatic interactions of the inhibitor's charged carboxylate head with the catalytic residues and the charged coenzyme NADP+ explained the inhibitor's noncompetitive character. Furthermore, a short contact involving the IDD 594 bromine atom explained the selectivity profile of the inhibitor, important feature to avoid toxic effects. The presented structure and the details revealed are instrumental for better understanding of the inhibition mechanism of AR by IDD 594, and hence, for the rational drug design of future inhibitors. This work demonstrates the capabilities of subatomic resolution experiments and stimulates further developments of methods allowing the use of the full potential of these experiments.

Crystallization of the 43 kDa ATPase domain of Thermus thermophilus gyrase B in complex with novobiocin.

The 43 kDa ATPase domain of Thermus thermophilus gyrase B was overproduced in Escherichia coli and a three-step purification protocol yielded large quantities of highly purified enzyme which remained stable for weeks. Crystals of the 43 kDa domain in complex with novobiocin, one of the most potent inhibitors of bacterial topoisomerases, were obtained. Crystals obtained in the presence of PEG 8000 do not diffract, but a different crystal form was obtained using sodium formate as a precipitating agent. The plate-shaped crystals, which were less than 10 microm in thickness, could be cryocooled directly from the mother liquor and a full diffraction data set was collected to 2.3 A allowing the determination of the first structure of a gyrase B 43K domain in complex with a coumarin.

Hyperthermophilic topoisomerase I from Thermotoga maritima. A very efficient enzyme that functions independently of zinc binding.

Topoisomerases, by controlling DNA supercoiling state, are key enzymes for adaptation to high temperatures in thermophilic organisms. We focus here on the topoisomerase I from the hyperthermophilic bacterium Thermotoga maritima (optimal growth temperature, 80 degrees C). To determine the properties of the enzyme compared with those of its mesophilic homologs, we overexpressed T. maritima topoisomerase I in Escherichia coli and purified it to near homogeneity. We show that T. maritima topoisomerase I exhibits a very high DNA relaxing activity. Mapping of the cleavage sites on a variety of single-stranded oligonucleotides indicates a strong preference for a cytosine at position -4 of the cleavage, a property shared by E. coli topoisomerase I and archaeal reverse gyrases. As expected, the mutation of the putative active site Tyr 288 to Phe led to a totally inactive protein. To investigate the role of the unique zinc motif (Cys-X-Cys-X(16)-Cys-X-Cys) present in T. maritima topoisomerase I, experiments have been performed with the protein mutated on the tetracysteine motif. Strikingly, the results show that zinc binding is not required for DNA relaxation activity, contrary to the E. coli enzyme. Furthermore, neither thermostability nor cleavage specificity is altered in this mutant. This finding opens the question of the role of the zinc-binding motif in T. maritima topoisomerase I and suggests that this hyperthermophilic topoisomerase possesses a different mechanism from its mesophilic homolog.

Characterization of crystal content by ESI-MS and MALDI-MS.

A general approach based on mass spectrometry is described for the rapid identification of the content of macromolecular crystals. The experimental procedure was established using lysozyme crystals and then successfully applied to various systems containing specifically bound molecules not easily detectable by other classical techniques. This procedure can be carried out on crystals containing macromolecules of a different nature, such as proteins, nucleic acids and small organic molecules and their non-covalent complexes, grown under various crystallization conditions including PEGs and salts. It can be applied very early on in the crystallization process - as soon as the crystals can be handled. It allows the biologist to control precisely the sequence integrity and homogeneity of the crystallized proteins (in particular at the C-terminus) as well as to verify whether the protein has crystallized with all its expected partners or ligands (nucleic acid molecules, cofactor or small organic molecules).

The structure of human aldose reductase bound to the inhibitor IDD384.

The crystallographic structure of the complex between human aldose reductase (AR2) and one of its inhibitors, IDD384, has been solved at 1.7 A resolution from crystals obtained at pH 5.0. This structure shows that the binding of the inhibitor's hydrophilic head to the catalytic residues Tyr48 and His110 differs from that found previously with porcine AR2. The difference is attributed to a change in the protonation state of the inhibitor (pK(a) = 4.52) when soaked with crystals of human (at pH 5.0) or pig lens AR2 (at pH 6.2). This work demonstrates how strongly the detailed binding of the inhibitor's polar head depends on its protonation state.

Production of crystals of human aldose reductase with very high resolution diffraction.

As the action of human aldose reductase (hAR) is thought to be linked to the pathogenesis of diabetic complications, much effort has been directed towards the analysis of the catalytic mechanism and the development of specific inhibitors. Here, the crystallization of recombinant hAR with its cofactor NADP+ at 277 K in the presence of the precipitating agent PEG 6000 is reported. The crystals diffract to high resolution (1.1 A) and belong to the P21 space group with unit-cell parameters a = 49.97, b = 67.14, c = 48. 02 A, beta = 92.2 degrees with one molecule per asymmetric unit. Seleno-substituted hAR crystals were also produced and diffract to 1. 7 A on a conventional X-ray source.

The HIR protein family: isolation and characterization of a complete murine cDNA.

A full-length cDNA has been isolated for the murine homolog of the human HIRA protein, a member of the HIR family of nuclear proteins that is encoded from the chromosome 22 region critical for the DiGeorge syndrome. This family also contains Hir1p and Hir2p, two proteins identified as regulators of histone gene transcription in yeast. The murine and human amino acid sequences are 95.3% identical, with a striking 99.2% identity in the N-terminal WD repeat domain that is characteristic of the family. The two cDNAs are highly conserved within the coding regions, but also in the entire 5' untranslated region and in a strikingly long stretch of nucleotides in the 3' untranslated region.

Structural Organization of the WD repeat protein-encoding gene HIRA in the DiGeorge syndrome critical region of human chromosome 22.

The human gene HIRA lies within the smallest critical region for the DiGeorge syndrome (DGS), a haploinsufficiency developmental disorder associated with instertitial deletions in most patients in a juxtacentromeric region of chromosome 22. The HIRA protein sequence can be aligned over its entire length with Hir1 and Hir2, two yeast proteins with a regulatory function in chromatin assembly. The HIRA transcription unit was found to spread over approximately 100 kb of the DGS critical region. The human transcript is encoded from 25 exons between 59 and 861 bp in size. Domains of highest conservation with Hir1 and Hir2 are encoded from exons 1-11 and 13-25, respectively. The amino- and carboxy-terminal regions of homology are separated from each other by a domain unique to HIRA that is encoded from a single exon. Seven WD repeats are conserved between yeast and man in the amino-terminal region of the HIR proteins. Individual repeats were found to be encoded from one, two, or three exons of the HIRA gene. End sequences have been obtained for all 24 introns, opening the way to PCR amplification of the entire coding sequence starting from genomic DNA. Point mutations can also be sought in 16 of the 24 introns that are readily PCR-amplifiable.

A human homolog of the S. cerevisiae HIR1 and HIR2 transcriptional repressors cloned from the DiGeorge syndrome critical region.

The DiGeorge syndrome (DGS) is a developmental disorder affecting derivatives of the third and fourth pharyngeal pouches. DGS patients present an interstitial deletion in one of their two chromosomes 22. Cosmid DAC30 was mapped to the DGS smallest critical region. Iterative cDNA library screening initiated with a DAC30 gene fragment candidate yielded a cDNA contig whose assembled nucleotide sequence is consistent with the widely transcribed, 4.2-4.4 kb long, messengers detected by northern analysis. The deduced protein sequence, 1017 amino acids in length, entirely encompasses the 766 amino acids previously designated as TUPLE1. The completed protein has been renamed HIRA because it contains various features matching those found in HIR1 and HIR2, two repressors of histone gene transcription characterized in the yeast Saccharomyces cerevisiae. Strikingly alike in their N-terminal third, HIRA and HIR1 contain seven copies of the WD repeat, a motif implicated in protein-protein interactions, suggesting that they might define a new subfamily of functionally homologous proteins. The remainder of the human polypeptide highly resembles a corresponding fragment in HIR2. We propose that HIRA, alone, could have a part in mechanisms of transcriptional regulation similar to that played by HIR1 and HIR2 together. The presence of a single copy of the HIRA gene in DGS patients possibly accounts for some of the abnormalities associated with this syndrome.

Evolution of the Glx-tRNA synthetase family: the glutaminyl enzyme as a case of horizontal gene transfer.

An important step ensuring the fidelity in protein biosynthesis is the aminoacylation of tRNAs by aminoacyl-tRNA synthetases. The accuracy of this process rests on a family of 20 enzymes, one for each amino acid. One exception is the formation of Gln-tRNA(Gln) that can be accomplished by two different pathways: aminoacylation of tRNA(Gln) with Gln by glutaminyl-tRNA synthetase (GlnRS; EC 6.1.1.18) or transamidation of Glu from Glu-tRNA(Gln) mischarged by glutamyl-tRNA synthetase (GluRS; EC 6.1.1.17). The latter pathway is widespread among bacteria and organelles that, accordingly, lack GlnRS. However, some bacterial species, such as Escherichia coli, do possess a GlnRS activity, which is responsible for Gln-tRNA(Gln) formation. In the cytoplasm of eukaryotic cells, both GluRS and GlnRS activities can be detected. To gain more insight into the evolutionary relationship between GluRS and GlnRS enzyme species, we have now isolated and characterized a human cDNA encoding GlnRS. The deduced amino acid sequence shows a strong similarity with other known GlnRSs and with eukaryotic GluRSs. A molecular phylogenetic analysis was conducted on the 14 GlxRS (GluRS or GlnRS) sequences available to date. Our data suggest that bacterial GlnRS has a eukaryotic origin and was acquired by a mechanism of horizontal gene transfer.

The E subunit of vacuolar H(+)-ATPase localizes close to the centromere on human chromosome 22.

As part of a general effort to identify new genes mapping to disease-associated regions of human chromosome 22, we have isolated heterogeneous nuclear RNA from somatic cell hybrids selected for their chromosome 22 content. Inter-Alu PCR amplification yielded a series of human DNA fragments which all detected evolutionarily-conserved sequences. The centromere-most gene fragment candidate, XEN61, was shown to lie centromeric to the chromosome 22 breakpoint in the X/22-33-11TG somatic cell hybrid. This region, which is still devoid of characterized genes, overlaps with the critical region for the cat eye syndrome (CES), a developmental disorder associated with chromosomal duplication within 22pter-q11.2. Gene dosage analysis performed on DNA from six CES patients consistently revealed the presence of four copies of XEN61. A fetal brain cDNA clone, 61EW, was identified with XEN61 and entirely sequenced. The deduced protein is the E subunit of vacuolar H(+)-ATPase. This 31 KDa component of a proton pump is essential in eukaryotic cells as it both controls acidification of the vacuolar system and provides it with its main protonmotive force. RT-PCR experiments using oligonucleotides designed from the 61EW cDNA sequence indicated that the corresponding messenger is widely transcribed.

Involvement of 22q13.3 in chromosomal-anomalies.

In a series of neoplasms involvement of chromosome 22, mainly concerning loci within bands 22q11-q12 has been reported. Yet, little is known about chromosomal anomalies in 22q13. As loss of heterozygosity in two neurofibromatosis type 2 patients was described in a 22q13.3 locus and deletions in the 22q13.3 chromosomal region were noted in a set of 7 patients, we decided to apply several newly isolated cosmids from 22q13 to analyse additional cases with chromosome 22 anomalies. In addition, the study was aided by centromeric probes and chromosome 22 painting. Fluorescent in situ hybridization with new cosmids mapping to 22q13.1 and 22q13.3 did not indicate deletions or rearrangements in one neurofibromatosis type 2 case [r(22)], a bisatellited chromosome 22 and in a translocation case [t(Y;22)].

Isolation of cosmids and fetal brain cDNAs from the proximal long arm of human chromosome 22.

The proximal portion of human chromosome 22q appears to carry genes implicated in the pathogenesis of various developmental disorders, including the cat eye syndrome (CES) and the DiGeorge syndrome (DGS). A cosmid library was prepared from a radiation hybrid selected for its content in chromosome 22 fragments. A large fraction of cosmids containing human DNA were found to derive from the juxtacentromeric region of chromosome 22, as shown by fluorescence in situ hybridization (FISH) performed using individual cosmids or cosmid pools as probes. Finer mapping was obtained for individual cosmids by hybridization to a somatic cell hybrid mapping panel which splits the long arm of the chromosome into 14 bins numbered 1 to 14 from the centromere to the telomere. Of the 10 cosmids mapped, eight belonged to group 1, the other two to group 14, in agreement with FISH data. Rare endonuclease sites and fragments conserved between species were searched in single cosmids, resulting in the selection of seven cosmid fragments which were used to screen a human fetal brain cDNA library. Three cDNAs were identified, encoded from two chromosome 22 genes which appeared to be novel, as determined from partial end sequence and comparison with the database entries. Fine localization of the 30.9 cDNA indicated that the corresponding gene was located in a segment of proximal 22q overlapping with the critical DGS region.

Localization of 15 cosmids on human chromosome 22 by fluorescence in situ hybridization.

Fluorescence in situ hybridization has been used in a cytogenetic analysis to map 15 cosmids on human chromosome 22. Thirteen cosmids were localized on the long arm of chromosome 22 (22q) while two other probes displayed a hybridization signal on 22p and the short arm of the acrocentric chromosomes of groups D and G. The regional assignment of these new chromosome markers will improve the mapping of chromosome 22; they can be used to detect numerical and structural aberrations of this chromosome involved in numerous pathologies.