Bora phosphorylation substitutes in trans for T-loop phosphorylation in Aurora A to promote mitotic entry.

Polo-like kinase 1 (Plk1) is instrumental for mitotic entry and progression. Plk1 is activated by phosphorylation on a conserved residue Thr210 in its activation segment by the Aurora A kinase (AURKA), a reaction that critically requires the co-factor Bora phosphorylated by a CyclinA/B-Cdk1 kinase. Here we show that phospho-Bora is a direct activator of AURKA kinase activity. We localize the key determinants of phospho-Bora function to a 100 amino acid region encompassing two short Tpx2-like motifs and a phosphoSerine-Proline motif at Serine 112, through which Bora binds AURKA. The latter substitutes in trans for the Thr288 phospho-regulatory site of AURKA, which is essential for an active conformation of the kinase domain. We demonstrate the importance of these determinants for Bora function in mitotic entry both in Xenopus egg extracts and in human cells. Our findings unveil the activation mechanism of AURKA that is critical for mitotic entry.

Trapping by amylose of the aliphatic chain grafted onto chlorogenic acid: importance of the graft position.

5-Caffeoylquinic acid (chlorogenic acid), is classified in acid-phenols family and as polyphenolic compounds it possesses antioxidant activity. The oxydative modification of chlorogenic acid in foods may lead to alteration of their qualities; to counteract these degradation effects, molecular encapsulation was used to protect chlorogenic acid. Amylose can interact strongly with a number of small molecules, including lipids. In order to enable chlorogenic acid complexation by amylose, a C16 aliphatic chain was previously grafted onto the cycle of quinic acid. This work showed that for the two lipophilic derivatives of chlorogenic acid: hexadecyl chlorogenate obtained by alkylation and 3-O-palmitoyl chlorogenic acid obtained by acylation; only the 3-O-palmitoyl chlorogenic acid complexed amylose. The chlorogenic acid derivatives were studied by X-ray diffraction, differential scanning calorimetry and NMR to elucidate the interaction. By comparing the results with previous work on the complexation of amylose by 4-O-palmitoyl chlorogenic acid, the importance of the aliphatic chain position on the cycle of the quinic acid is clearly highlighted. A study in molecular modeling helped to understand the difference in behavior relative to amylose of these three derivatives of chlorogenic acid.

Coupling lipophilization and amylose complexation to encapsulate chlorogenic acid.

Chlorogenic acid (5-caffeoylquinic acid) is a hydrophilic phenolic compound with antioxidant properties. Because of its high polarity, these properties may be altered when formulated in oil-based food. There is therefore an interest in trying to protect the natural antioxidant by molecular encapsulation. Amylose, the linear fraction of starch with essentially α(1-4) linkages, is well known for its ability to form semi-crystalline complexes with a variety of small ligands. Monoacyl lipids, as well as smaller ligands such as alcohols or flavor compounds, are able to induce the formation of left-handed amylose single helices. In contrast, chlorogenic acid is a bulky molecule whose topology requires the amylose helix to be distorted, which could prevent amylose complexation. An innovative strategy has been developed to overcome this problem by grafting an aliphatic chain onto chlorogenic acid then trapping this chain in the helical cavity. The lipophilization reaction was used to obtain a palmitoyl chlorogenic acid derivative and the amylose-palmitoyl chlorogenic acid assemblies were studied by X-ray diffraction, differential scanning calorimetry and NMR to elucidate the interaction. The results showed that such interactions between amylose and palmitoyl chlorogenic acid are effective.

Lipase-catalyzed synthesis of two new antioxidants: 4-O- and 3-O-palmitoyl chlorogenic acids.

Chlorogenic acid (5-caffeoyl quinic acid (CQA)) extracted from Hydrangea macrophylla (44%, w/w) with 98% purity, was acylated with palmitic acid by Novozym 435 to yield mono-acylated CQA. Acylation of CQA was achieved in 2-methyl-2-butanol at 60°C, and yielded two mono-acylated products: a major product acylated at the C-4 of the quinic moiety (4-O-palmitoyl chlorogenic acid) and a minor product acylated at the C-3 (3-O-palmitoyl chlorogenic acid). The bioconversions obtained in 7 days ranged from 14 to 60% and were influenced by the molar ratio of palmitic acid/CQA, which ranged from 10 to 80. The regioselectivity (4-O-palmitoyl/3-O-palmitoyl ratio) of the reaction was also affected by the molar ratio, and ranged from 90 to 70%. The scavenging activities against 1,1-diphenyl-2-picryl-hydrazyl radicals demonstrated that these palmitoyl CQA derivatives are associated with antioxidant activity (70% vs CQA).

RSK2 is a kinetochore-associated protein that participates in the spindle assembly checkpoint.

The spindle assembly checkpoint (SAC) prevents anaphase onset until all the chromosomes have successfully attached to the spindle microtubules. The MAP kinase (MAPK) is an important player in this pathway, however its exact role is not fully understood. One major target of MAPK is the p90 ribosomal protein S6 kinase (RSKs) family. In this study, we analyse whether Rsk2 could participate in the activation of the SAC. Our data indicate that this protein is localized at the kinetochores under checkpoint conditions. Moreover, it is essential for the SAC activity in Xenopus egg extracts as its depletion prevents metaphase arrest as well as the kinetochore localization of the other SAC components. We also show that this kinase might also participate in the maintenance of the SAC in mammalian cells as Rsk2 knockdown in these cells prevents the kinetochore localization of Mad1, Mad2 and CENP-E under checkpoint conditions.

Chfr interacts and colocalizes with TCTP to the mitotic spindle.

Chfr is a checkpoint protein that plays an important function in cell cycle progression and tumor suppression, although its exact role and regulation are unclear. Previous studies have utilized overexpression of Chfr to determine the signaling pathway of this protein in vivo. In this study, we demonstrate, by using three different antibodies against Chfr, that the endogenous and highly overexpressed ectopic Chfr protein is localized and regulated differently in cells. Endogenous and lowly expressed ectopic Chfr are cytoplasmic and localize to the spindle during mitosis. Higher expression of ectopic Chfr correlates with a shift in the localization of this protein to the nucleus/PML bodies, and with a block of cell proliferation. In addition, endogenous and lowly expressed ectopic Chfr is stable throughout the cell cycle, whereas when highly expressed, ectopic Chfr is actively degraded during S-G2/M phases in an autoubiquitination and proteasome-dependent manner. A two-hybrid screen identified TCTP as a possible Chfr-interacting partner. Biochemical analysis with the endogenous proteins confirmed this interaction and identified beta-tubulin as an additional partner for Chfr, supporting the mitotic spindle localization of Chfr. The Chfr-TCTP interaction was stable throughout the cell cycle, but it could be diminished by the complete depolymerization of the microtubules, providing a possible mechanism where Chfr could be the sensor that detects microtubule disruption and then activates the prophase checkpoint.

Immobilisation of oligo-peptidic probes for microarray implementation: characterisation by FTIR, atomic force microscopy and 2D fluorescence.

Proteomic microarrays show a wide range of applications for the investigation of DNA-protein, enzyme-substrate as well as protein-protein interactions. Among many challenges to build a viable "protein microarray", the surface chemistry that will allow to immobilised various proteins to retain their biological activity is of paramount importance. Here we report a chemical functionalisation method allowing immobilisation of oligo-peptides onto silica surface (porous silica, glass, thermal silicon dioxide). Substrates were first derivatised with a monofunctional silane allowing the elaboration of dense and uniform monolayers in highly reproducible way. Prior to the oligo-peptides grafting, this organic layer was functionalised with an amino-polyethyleneglycol. The coupling step of oligo-peptides onto functionalised supports is achieved through activation of the C-terminal function of the oligo-peptides. Chemical surface modifications were followed by FTIR spectroscopy, AFM measurements and fluorescence scanning microscopy. A systematic study of the oligo-peptide grafting conditions (time, concentration, solvent) was carried out to optimise this step. The oligo-peptides grafting strategy implemented in this work ensure a covalent and oriented grafting of the oligo-peptides. This orientation is ensured through the use of fully protected peptide except the terminal primary amine. The immobilized peptides will be then deprotected before biological recognition. This strategy is crucial to retain the biological activity of thousands of oligo-probes assessed on a microarray.

Cyclin B/cdc2 induces c-Mos stability by direct phosphorylation in Xenopus oocytes.

The c-Mos proto-oncogene product plays an essential role during meiotic divisions in vertebrate eggs. In Xenopus, it is required for progression of oocyte maturation and meiotic arrest of unfertilized eggs. Its degradation after fertilization is essential to early embryogenesis. In this study we investigated the mechanisms involved in c-Mos degradation. We present in vivo evidence for ubiquitin-dependent degradation of c-Mos in activated eggs. We found that c-Mos degradation is not directly dependent on the anaphase-promoting factor activator Fizzy/cdc20 but requires cyclin degradation. We demonstrate that cyclin B/cdc2 controls in vivo c-Mos phosphorylation and stabilization. Moreover, we show that cyclin B/cdc2 is capable of directly phosphorylating c-Mos in vitro, inducing a similar mobility shift to the one observed in vivo. Tryptic phosphopeptide analysis revealed a practically identical in vivo and in vitro phosphopeptide map and allowed identification of serine-3 as the largely preferential phosphorylation site as previously described (Freeman et al., 1992). Altogether, these results demonstrate that, in vivo, stability of c-Mos is directly regulated by cyclin B/cdc2 kinase activity.

Mps1 is a kinetochore-associated kinase essential for the vertebrate mitotic checkpoint.

The mitotic checkpoint acts to inhibit entry into anaphase until all chromosomes have successfully attached to spindle microtubules. Unattached kinetochores are believed to release an activated form of Mad2 that inhibits APC/C-dependent ubiquitination and subsequent proteolysis of components needed for anaphase onset. Using Xenopus egg extracts, a vertebrate homolog of yeast Mps1p is shown here to be a kinetochore-associated kinase, whose activity is necessary to establish and maintain the checkpoint. Since high levels of Mad2 overcome checkpoint loss in Mps1-depleted extracts, Mps1 acts upstream of Mad2-mediated inhibition of APC/C. Mps1 is essential for the checkpoint because it is required for recruitment and retention of active CENP-E at kinetochores, which in turn is necessary for kinetochore association of Mad1 and Mad2.

Part of Xenopus translin is localized in the centrosomes during mitosis.

During oogenesis, maternal mRNAs are synthesised and stored in a translationally dormant form due to the presence of regulatory elements at the 3' untranslated regions (3'UTR). In Xenopus oocytes, several studies have described the presence of RNA-binding proteins capable to repress maternal-mRNA translation. The testis-brain RNA-binding protein (TB-RBP/Translin) is a single-stranded DNA- and RNA-binding protein which can bind the 3' UTR regions (Y and H elements) of stored mRNAs and can suppress in vitro translation of the mRNAs that contain these sequences. Here we report the cloning of the Xenopus homologue of the TB-RBP/Translin protein (X-translin) as well as its expression, its localisation, and its biochemical association with the protein named Translin associated factor X (Trax) in Xenopus oocytes. The fact that this protein is highly present in the cytoplasm from stage VI oocytes until 48 h embryos and that it has been described as capable to inhibit paternal mRNA translation, indicates that it could play an important role in maternal mRNA translation control during Xenopus oogenesis and embryogenesis. Moreover, we investigated X-translin localisation during cell cycle in XTC cells. In interphase, although a weak and diffuse nuclear staining was observed, X-translin was mostly present in the cytoplasm where it exhibited a prominent granular staining. Interestingly, part of X-translin underwent a remarkable redistribution throughout mitosis and associated with centrosomes, which may suggest a new unknown role for this protein in cell cycle.

Regulation of Xenopus p21-activated kinase (X-PAK2) by Cdc42 and maturation-promoting factor controls Xenopus oocyte maturation.

Signal transduction cascades involved in regulation of the cell cycle machinery are poorly understood. In the Xenopus oocyte model, meiotic maturation is triggered by MPF, a complex of p34(cdc2)-cyclin B, which is activated in response to a progesterone signal by largely unknown mechanisms. We have previously shown that the p21-activated kinase (PAK) family negatively regulates the MPF amplification loop. In this study, we identify the endogenous PAK2 as a key enzyme in this regulation and describe the pathways by which PAK2 is regulated. We show that the small GTPase Cdc42 is required for maintenance of active endogenous X-PAK2 in resting stage VI oocytes, whereas Rac1 is not involved in this regulation. During the process of maturation, X-PAK2 phosphorylation results in its inactivation and allows maturation to proceed to completion. Activation of mitogen-activated protein kinase and cyclin B-p34(cdc2) is coincident with X-PAK2 inactivation, and purified active MPF inhibits X-PAK2, demonstrating the existence of a new positive feedback loop. Our results confirm and extend the importance of p21-activated kinases in the control of the G(2)/M transition. We hypothesize that the X-PAK2/Cdc42 pathway could link p34(cdc2) activity to the major cytoskeleton rearrangements leading to spindle migration and anchorage to the animal pole cortex.

Control of G2/M transition in Xenopus by a member of the p21-activated kinase (PAK) family: a link between protein kinase A and PAK signaling pathways?

X-PAKs are involved in negative control of the process of oocyte maturation in Xenopus (). In the present study, we define more precisely the events targetted by the kinase in the inhibition of the G2/M transition. We show that microinjection of recombinant X-PAK1-Cter active kinase into progesterone-treated oocytes prevents c-Mos accumulation and activation of both MAPK and maturation-promoting factor (MPF). In conditions permissive for MAPK activation, MPF activation still fails. We demonstrate that a constitutive truncated version of X-PAK1 (X-PAK1-Cter) does not prevent the association of cyclin B with p34(cdc2) but rather prevents the activation of the inactive complexes present in the oocyte. Proteins participating in the MPF amplification loop, including the Cdc25-activating Polo-like kinase are all blocked. Indeed, using active MPF, the amplification loop is not turned on in the presence of X-PAK1. Our results indicate that X-PAK and protein kinase A targets in the control of oocyte maturation are similar and furthermore that this negative regulation is not restricted to meiosis, because we demonstrate that G2/M progression is also prevented in Xenopus cycling extracts in the presence of active X-PAK1.

Fizzy is required for activation of the APC/cyclosome in Xenopus egg extracts.

The Xenopus homologue of Drosophila Fizzy and budding yeast CDC20 has been characterized. The encoded protein (X-FZY) is a component of a high molecular weight complex distinct from the APC/cyclosome. Antibodies directed against FZY were produced and shown to prevent calmodulin-dependent protein kinase II (CaMKII) from inducing the metaphase to anaphase transition of spindles assembled in vitro in Xenopus egg extracts, and this was associated with suppression of the degradation of mitotic cyclins. The same antibodies suppressed M phase-promoting factor (MPF)-dependent activation of the APC/cyclosome in interphase egg extracts, although they did not appear to alter the pattern or extent of MPF-dependent phosphorylation of APC/cyclosome subunits. As these phosphorylations are thought to be essential for APC/cyclosome activation in eggs and early embryos, we conclude that at least two events are required for MPF to activate the APC/cyclosome, allowing both chromatid segregation and full degradation of mitotic cyclins. The first one, which does not require FZY function, is the phosphorylation of APC/cyclosome subunits. The second one, that requires FZY function (even in the absence of MAD2 protein and when the spindle assembly checkpoint is not activated) is not yet understood at its molecular level.

A member of the Ste20/PAK family of protein kinases is involved in both arrest of Xenopus oocytes at G2/prophase of the first meiotic cell cycle and in prevention of apoptosis.

We have identified new members (X-PAKs) of the Ste20/PAK family of protein kinases in Xenopus, and investigated their role in the process that maintains oocytes arrested in the cell cycle. Microinjection of a catalytically inactive mutant of X-PAK1 with a K/R substitution in the ATP binding site, also deleted of its Nter-half that contains the conserved domains responsible for binding of both Cdc42/Rac GTPases and SH3-containing proteins, greatly facilitates oocyte release from G2/prophase arrest by progesterone and insulin. Addition of the same X-PAK1 mutant to cell cycle extracts from unfertilized eggs induced apoptosis, as shown by activation of caspases and cytological changes in in vitro-assembled nuclei. This was suppressed by adding Bcl-2 or the DEVD peptide inhibitor of caspases, and rescued by competing the dominant-negative mutant with its constitutively active X-PAK1 counterpart. Such results indicate that X-PAK1 (or another member of the Xenopus Ste20/PAK family of protein kinases) is involved in arrest of oocytes at G2/prophase and prevention of apoptosis; thus death by apoptosis and release of healthy oocytes from cell cycle arrest may be linked. That cell cycle arrest protects oocytes from apoptosis is consistent with the finding that extracts from metaphase II-arrested oocytes are less sensitive to apoptotic signals than those from activated eggs.

DNA polymorphisms in linkage disequilibrium at the 3' end of the human APO AII gene: relationships with lipids, apolipoproteins and coronary heart disease.

Two investigations were undertaken to analyze the 3' region of the apolipoprotein AII (Apo AII) gene in patients with myocardial infarction (MI) and controls. Previous studies have suggested that a MspI polymorphism in this gene may be associated with hypertriglyceridaemia, high levels of HDL cholesterol and Apo AII. To verify this hypothesis, the distribution of MspI genotypes and their possible associations with several plasma lipid variables were studied in 882 subjects (411 cases with MI and 471 controls) from the ECTIM study. There were no differences in genotype and allele frequencies between cases and controls, and no differences in lipid variable levels in controls carrying the less frequent MspI allele vs other controls. Using single-strand conformation polymorphism (SSCP) analysis, we detected a new polymorphism which caused by a C-to-T transition located in the third intron near the splice junction site (acceptor). This polymorphism modifies a Bst N1 restriction site. The ECTIM population was screened for this new marker, and no significant associations with MI and plasma lipid levels were found. Our results suggest that these two variants located in the coding region of the Apo AII gene are unlikely to contribute significantly to the level of plasma lipid variables and the risk of coronary heart disease (CHD) in the European population.

Structural analysis of the minisatellite present at the 3' end of the human apolipoprotein B gene: new definition of the alleles and evolutionary implications.

The internal structure of different alleles of the minisatellite present at the 3' end of the apolipoprotein B (ApoB) gene has been analysed by different approaches including sequencing. The repeat unit arrangements of the minisatellite on 570 chromosomes belonging to European and African populations were thus determined. It was possible to group the alleles using this structural criterion much more clearly than by the number of repeat units which can in some cases be misleading in case-control genetic epidemiological studies using such DNA sequences as markers. We were thus able to define five types (a to e) of alleles and their subtypes and to recognize clearly those which are, respectively, specific of the African and Caucasian populations. A phylogeny of the different alleles found in all human populations could also be deduced by this approach. The different putative mutational events leading from one type, or subtype, to the other were simply determined as point mutations, expansion/contraction and conversion events. Sequencing of one chimpanzee's allele suggested that the ApoB minisatellite was present before divergence between great apes and humans. It was determined also that a particular ApoB gene haplotype was in linkage disequilibrium with the minisatellite (a) type of alleles. This and the observation that the potential scaffold attachment regions (SAR) and topoisomerase II binding sites present in this minisatellite have a different distribution between the Caucasian and the African specific alleles suggest that the minisatellite could be involved in the epidemiology of coronary diseases.

Polymorphism at VNTR locus 3 to the apolipoprotein B gene in a Tunisian population: difference from other ethnic groups.

The Hypervariable region (HVR) detected at the 3' end of the apolipoprotein B (Apo B) locus has been the subject of numerous studies. As for many VNTR (variable number of tandem repeat), this locus is highly polymorphic and until now about 20 alleles have been described. The genotype distribution in all populations follows the Hardy-Weinberg predictions. A bimodal pattern of allele frequency distribution is apparent in all Caucasoid populations. We have analyzed the frequencies of different alleles in a Tunisian population (123 individuals) by the polymerase chain reaction technique and compared our results to those obtained in several ethnic groups. It appears that the distributions of the allele frequencies are very different: for Caucasoid populations, there are two peaks of frequencies for alleles with 36 and 48 repeats, but alleles of intermediate lengths are more frequent. Hixson et al. [(1993) Hum Genet 91:475-479] have shown a similar difference between black and white American populations. We found the same results in a black African group. Some of the repeat units of this HVR contain a Ssp I restriction site and digestion of the PCR products by this enzyme gives different patterns on gradient acrylamide gel [Desmarais et al., 1993, Nucleic Acids Res 21:2179-2184.] The DNA of African individuals (42) has been analyzed to discover the origin of this new allele. Preliminary results indicate that these particular alleles probably arose by introgression from the African population into the Tunisian one.

Variant mapping of the Apo(B) AT rich minisatellite. Dependence on nucleotide sequence of the copy number variations. Instability of the non-canonical alleles.

Because of its variations in length, the AT rich Hyper-Variable Region (HVR) of the 3' end of the Apolipoprotein B gene is used as a polymorphic maker in genetic studies. It contains a SspI site in its repeated motif and we used this feature to precisely analyse the internal structure of the different alleles found at this locus in a Caucasian population. We performed total digestion on 194 alleles as well as Minisatellite Variant Repeat mapping (MVR mapping: partial digestion) on 54. The results show that the level of length variability (in copy number) of the 5' end of this locus is at least two times higher than that of the 3' end. This could be correlated with the difference in nucleotide sequence between the two parts of the HVR and suggests the dependence on the primary structure of the mechanism that produces length variability. A molecular model is proposed to explain this result. Moreover, the sharp analysis of the minisatellite structure by the distribution of SspI sites reveals differences between long and short alleles, indicating that in most cases, no recombination occurs between alleles of different sizes. Finally the rare alleles exhibit a non-canonical structure. These important points could explain the bimodal distribution of the frequencies of the alleles in the population.