Accurate prediction of X-ray pulse properties from a free-electron laser using machine learning.

Free-electron lasers providing ultra-short high-brightness pulses of X-ray radiation have great potential for a wide impact on science, and are a critical element for unravelling the structural dynamics of matter. To fully harness this potential, we must accurately know the X-ray properties: intensity, spectrum and temporal profile. Owing to the inherent fluctuations in free-electron lasers, this mandates a full characterization of the properties for each and every pulse. While diagnostics of these properties exist, they are often invasive and many cannot operate at a high-repetition rate. Here, we present a technique for circumventing this limitation. Employing a machine learning strategy, we can accurately predict X-ray properties for every shot using only parameters that are easily recorded at high-repetition rate, by training a model on a small set of fully diagnosed pulses. This opens the door to fully realizing the promise of next-generation high-repetition rate X-ray lasers.

Sound and speech detection and classification in a Health Smart Home.

Improvements in medicine increase life expectancy in the world and create a new bottleneck at the entrance of specialized and equipped institutions. To allow elderly people to stay at home, researchers work on ways to monitor them in their own environment, with non-invasive sensors. To meet this goal, smart homes, equipped with lots of sensors, deliver information on the activities of the person and can help detect distress situations. In this paper, we present a global speech and sound recognition system that can be set-up in a flat. We placed eight microphones in the Health Smart Home of Grenoble (a real living flat of 47m(2)) and we automatically analyze and sort out the different sounds recorded in the flat and the speech uttered (to detect normal or distress french sentences). We introduce the methods for the sound and speech recognition, the post-processing of the data and finally the experimental results obtained in real conditions in the flat.

Platelet-derived growth factor stimulates phospholipase C-gamma 1, extracellular signal-regulated kinase, and arachidonic acid release in rat myometrial cells: contribution to cyclic 3',5'-adenosine monophosphate production and effect on cell proliferation.

In the present study, we examined downstream signaling events that followed exposure of cultured rat myometrial cells to platelet-derived growth factor (PDGF) and their effect on cell proliferation. PDGF-BB induced tyrosine phosphorylation of PDGF-beta receptors and increased inositol trisphosphate production via the tyrosine phosphorylation of phospholipase (PL)C-gamma 1. PDGF-BB also increased cAMP synthesis. This increase was potentiated by forskolin and reduced by indomethacin, a cyclooxygenase inhibitor, reflecting a Gs protein-mediated process via prostaglandin biosynthesis. The prostaglandin produced by PDGF was characterized as prostacyclin (PGI(2)). PDGF-BB increased arachidonic acid (AA) release, which, similarly to cAMP accumulation, was abolished in the presence of AACOCF3, a cytosolic PLA(2) inhibitor, and in the absence of Ca(2+). U-73122, a potent inhibitor of PLC activity, blocked both the production of inositol phosphates and the AA release triggered by PDGF-BB. Extracellular signal-regulated kinases (ERKs) 1 and 2 are expressed in myometrial cells, and PDGF-BB selectively activated ERK2. PD98059, an inhibitor of the ERK-activating kinase, blocked PDGF-BB-mediated ERK2 activation, AA release, and cAMP production. The results demonstrate that PDGF-BB stimulated cAMP formation through both PLC activation and ERK-dependent AA release and PGI(2) biosynthesis. PDGF-BB also increased cell proliferation and [(3)H]thymidine incorporation. This was abolished by PD98059, demonstrating that the ERK cascade is required for the mitogenic effect of PDGF-BB. Forskolin, which potentiated the cAMP response to PDGF-BB, attenuated both DNA synthesis and ERK activation triggered by PDGF-BB, suggesting the presence of a negative feedback regulation.

Direct transactivation of c-Ha-Ras gene by p53: evidence for its involvement in p53 transactivation activity and p53-mediated apoptosis.

p53 protein is a sequence-specific transcriptional activator which induces the expression of a number of cellular genes involved in different metabolic pathways. We report that the computer-selected sequence in human and mouse C-Ha-Ras gene confers to a reporter gene the ability to be directly transactivated by wild-type p53 either overexpressed or activated in response to a cellular stress. By analysing human transformed cell lines, we showed, at both mRNA and protein level, that the endogenous c-Ha-Ras gene expression is positively regulated by wt p53 protein. The stimulation of c-Ha-Ras gene expression in Saos-2Ts cells by a temperature shift down to the permissive temperature for the p53-wt conformation is associated with a significant increase in the activated form of p21c-Ha-Ras protein. Furthermore, in human transformed cell lines, the transient expression of a dominant interfering mutant of c-Ha-Ras greatly reduced the ability of p53 to induce apoptosis and inhibited the p53-dependent transactivation. This is due, at least in part, to a decrease in the protein (but not mRNA) level of the transiently expressed p53, indicating that inactivation of p21c-Ha-Ras signalling pathways led to a specific degradation of p53 protein. We therefore suggest that, by inducing c-Ha-Ras, p53 activates a positive feedback loop that counteracts the negative feedback loop mediated by Mdm2.

Tissue and cell-specific expression of the p53-target genes: bax, fas, mdm2 and waf1/p21, before and following ionising irradiation in mice.

The mechanisms by which the p53 tumour suppressor protein would, in vivo, co-ordinate the adaptive response to genotoxic stress is poorly understood. p53 has been shown to transactivate several genes that could be involved in two main cellular responses, growth arrest and apoptosis. To get further insight into the tissue-specific regulation of p53 transcriptional activity, we performed an extensive study looking at the expression of four well characterized p53-responsive genes, before and after gamma-irradiation in p53 wild-type (p53+/+) and p53-deficient (p53-/-) mice. The waf1, bax, fas and mdm2 genes were chosen for their different potential roles in the cellular response to stress. Our data demonstrate the strict p53-dependence of mRNA up-regulation for bax, fas and mdm2 in irradiated tissues and confirm such findings for waf1. They further highlight complex levels of regulatory mechanisms that could lead, in vivo, to selective transcriptional activation of genes by p53. In addition, our results provide arguments for the involvement of p53 in the basal mRNA expression of the four genes in some organs. Finally, in situ expression of Bax and p21Waf-1 protein suggests, at least in lymphoid organs, a direct correlation between selective p53-target gene expression and a particular response of a cell to ionising radiation.

Translational diffusion of globular proteins in the cytoplasm of cultured muscle cells.

Modulated fringe pattern photobleaching (MFPP) was used to measure the translational diffusion of microinjected fluorescein isothiocyanate (FITC)-labeled proteins of different sizes in the cytoplasm of cultured muscle cells. This technique, which is an extension of the classical fluorescence recovery after photobleaching (FRAP) technique, allows the measurement of the translational diffusion of macromolecules over several microns. Proteins used had molecular masses between 21 and 540 kDa. The results clearly indicated that the diffusivity of the various proteins is a decreasing function of their hydrodynamic radius. This decrease is more rapid with globular proteins than with FITC-labeled dextrans (, Biophys. J. 70:2327-2332), most likely because, unlike globular proteins, dextrans are randomly coiled macromolecules with a flexible structure. These data do not exclude the possibility of a rapid diffusion over a short distance, unobservable with our experimental set-up, which would take place within the first milliseconds after bleaching and would correspond to the diffusion in restricted domains followed by impeded diffusion provoked by the network of microtubules, microfilaments, and intermediate filaments. Thus our results may complement rather than contradict those of Verkman and collaborators (, J. Cell Biol. 138:1-12). The biological consequence of the size-dependent restriction of the mobility of proteins in the cell cytoplasm is that the formation of intracellular complexes with other proteins considerably reduces their mobility.

Interactions between beta-enolase and creatine kinase in the cytosol of skeletal muscle cells.

We studied interactions in vivo between the cytosolic muscle isoform of creatine kinase (M-CK) and the muscle isoform of 2-phospho-D-glycerate hydrolyase (beta-enolase) in muscle sarcoplasm by incubating glycerol-skinned fibres with FITC-labelled beta-enolase in the presence or absence of free CK. A small amount of bound beta-enolase was observed in the presence of large concentrations of CK. The mobility of enolase was measured in cultured satellite cells by modulated-fringe-pattern photobleaching. FITC-labelled beta-enolase was totally mobile in both the presence and the absence of CK but its diffusion coefficient was slightly lower in the presence of CK. This suggests a weak interaction in vivo between enolase and CK.

A tyrosine kinase signaling pathway, regulated by calcium entry and dissociated from tyrosine phosphorylation of phospholipase Cgamma-1, is involved in inositol phosphate production by activated G protein-coupled receptors in myometrium.

Our experiments were conducted to evaluate, in rat myometrium, the potential contribution of a protein tyrosine kinase (PTK) pathway in the hydrolysis of phosphatidylinositol-4,5-bisphosphate mediated by bombesin, endothelin-1 (ET-1), and carbachol. The production of inositol phosphates (InsP) by agonists and AlF4- was partly inhibited (35-40%) by genistein and tyrphostins, two PTK inhibitors. Genistein attenuated uterine contractions elicited by the stimulation of muscarinic and bombesin receptors, whereas pervanadate, a protein tyrosine phosphatase inhibitor, potentiated receptor-mediated contraction. Tyrosine-phosphorylated proteins were detected in detergent extracts from agonist- and pervanadate-stimulated myometrium. The amount of InsP produced in response to pervanadate was related to the tyrosine phosphorylation status of phospholipase C-gamma1. In contrast, with ET-1 and bombesin, phosphorylated phospholipase C-gamma1 made a minor contribution. Additional findings were rather consistent with a role for Ca2+. In fura-2-loaded cells, genistein partly decreased both the transient and sustained intracellular Ca2+ concentration phases induced by bombesin. The removal of extracellular Ca2+ or the addition of nifedipine inhibited (35%) InsP production due to bombesin and ET-1. The inhibitory effects of genistein and tyrphostins were abolished in Ca2+-depleted medium, were not additive with that of nifedipine, and (as for nifedipine) were counteracted by the Ca2+ channel agonist Bay K 8644. The data are consistent with a PTK-mediated process in the activation of the voltage-gated Ca2+ influx that is involved in the production of InsP by stimulated G protein-coupled receptors.

Presence of enolase in the M-band of skeletal muscle and possible indirect interaction with the cytosolic muscle isoform of creatine kinase.

Glycerol-skinned skeletal muscle fibres retain the defined sarcomeric structure of the myofibrils. We show here that a small fraction of two enzymes important for energy metabolism, the cytosolic muscle isoform of creatine kinase (EC 2.7.3.2), MM-creatine kinase (MM-CK), and enolase (EC 4.2.1.11), remains bound to skinned fibres. CK is slowly exchangeable, whereas enolase is firmly bound. Two-dimensional gel electrophoresis followed by Western blot analyses demonstrates that both alpha (ubiquitous) and beta (muscle-specific) subunits of enolase are present in these preparations. Enolase and CK were co-localized at the M-band of the sarcomeres, as observed by indirect immunofluorescence and confocal microscopy. Cross-linking experiments were performed on skinned fibres with three bifunctional succinimidyl esters of different lengths and yielded a protein complex of 150 kDa that reacted with antibodies directed against either M-CK or beta-enolase. The cross-linking efficiency was greatest for the longest reagent and zero for the shortest one. The length of the cross-linker giving a covalent complex between the two enzymes does not support the notion of a direct interaction between M-CK and enolase. This is the first demonstration of the presence of an enzyme of energy metabolism other than CK at the M-band of myofibres.

Mobility of creatine phosphokinase and beta-enolase in cultured muscle cells.

The diffusion of beta-enolase and creatine phosphokinase in muscle cells has been studied by modulated fringe pattern photobleaching. Beta-enolase is mobile in the sarcoplasm. At 20 degrees C, the diffusion coefficient is 13.5 +/- 2.5 microm2 s(-1) in the cytosol and 56 microm2 s(-1) in aqueous media. As in the case of dextrans of the same hydrodynamic radius, its mobility is hindered by both the crowding of the fluid phase of the cytoplasm and the screening effect due to myofilaments. A fraction of creatine phosphokinase is mobile in the sarcoplasm. Its diffusion coefficient in the cytosol, 4.5 +/- 1 microm2 s(-1), is lower than that of the dextran of equivalent size. The other fraction (20 to 50%) is very slightly mobile, with an apparent diffusion coefficient varying from 0.0035 to 0.043 microm2 s(-1). This low mobility might be attributed to exchange between free and bound creatine phosphokinase. The bound fraction of the endogenous enzyme was localized by immunocytofluorescence on the cultured muscle cells. Our results favor a localization of bound cytosolic creatine phosphokinase on the M-line and a diffuse distribution in all myotubes.

Effect of constant and modulated electrical charges applied to the culture material on PGI2 and TXA2 secretion by endothelial cells.

Electrical stimulation of an endothelial cell culture applied through the culture support biomaterial induced strong secretion of prostaglandin I2 (PGI2) and, to a lesser extent, thromboxane A2 (TXA2), which varied with time, the sign of the stationary electrical charge and the amplitude and frequency of a superimposed sine wave. In stationary conditions, the response is more rapid at positive electrical charges (+ some tens of microC cm-2) than at negative ones (-some tens of microC cm-2). In sine wave conditions, the ratio [PGI2]/[TXA2] is strongly increased and is maximum for an amplitude of 50 mV and a frequency of 1 Hz. These results suggest that the extracellular matrix is not only involved in cellular anchoring but also participates actively in secretion of thrombomodulatory substances, an effect apparently related to the piezoelectric properties of its components.

Limited proteolysis of myelin basic protein in a system mimetic of the myelin interlamellar aqueous space.

We have investigated the early steps of myelin basic protein (MBP) degradation in a membrane mimetic system (reverse micelles), resembling the interlamellar aqueous spaces where the protein is located in the myelin sheath. MBP, unfolded in buffer, refolds on incorporation into the micelles, resulting in reduced accessibility to three proteolytic enzymes, trypsin, cathepsin D, and Staphylococcus aureus V8 protease, in comparison with aqueous solution. Eleven cleavage sites seen in buffer are removed from proteolytic attack in micellar solution. These sites delineate a protected protein domain displaying a potential beta-sheet structure capable of interacting with the myelin membrane. An additional site not seen in buffer is attacked in the micelles. Experiments with a structure inducer, 15% 1-propanol in buffer, reveal that the refolding pattern of MBP in reverse micelles is specific to the membrane biomimetic system and is not produced by organic solvent per se. Micellar digestions of MBP generate long peptides, two of which, isolated after tryptic digestion, have been found to be immunodominant in multiple sclerosis patients. The findings suggest the structure induced in MBP by the micelles resembles that leading to production of the self-peptides recognized by T cells during proteolytic breakdown of MBP in autoimmune demyelinating diseases.

Antigen-antibody binding in reverse micelles: interaction of monoclonal antibodies with a myelin basic protein peptide.

Reverse micelles can be used to mimic biological processes occurring at interfaces. To investigate antigen-antibody binding in a membrane-like environment, we first obtained Fab fragments from monoclonal antibodies against bovine myelin basic protein (MBP), an encephalitogenic protein. The binding of the fragments to a dansylated synthetic human MBP peptide gly(119)-gly(131), presenting sequence homologies with a viral protein, was measured in buffer and for the first time in reverse micelles of sodium bis(2-ethylhexyl) sulfosuccinate, in isooctane. Analysis of the fluorescence polarisation titration curves discloses that the Fab fragments in reverse micelles have retained the high affinity for the peptide found in buffer, and similar to that for intact MBP.

Myelin proteins in reverse micelles: tight lipid association required for insertion of the Folch-Pi proteolipid into a membrane-mimetic system.

The solubility and reactivity of the Folch-Pi proteolipid from bovine CNS have been studied in reverse micelles of sodium bis(2-ethylhexyl)sulfosuccinate, isooctane, and water. Such a membrane-mimetic system resembles the aqueous spaces of the native myelin sheath in terms of its physicochemical properties. Although the proteolipid is completely insoluble in water, it can be inserted into the water-containing micellar system. In contrast, the lipid-depleted protein failed to be incorporated into these organized assemblies. The lipid requirements for insertion of the proteolipid were studied, therefore, after delipidation by several precipitations with isooctane, a nondenaturing solvent. Novel extraction procedures and quantitative analyses by HPLC of the protein-bound lipids revealed the persistence of a lipid-protein complex (6 +/- 1 mol of lipid/mol of protein) displaying optimal micellar solubilization. Competition experiments carried out with brain lipids provide evidence for a preference of the myelin protein for sulfatide, phosphatidylinositol, and phosphatidylserine, in that order. The resulting proteolipid, although differing in relative composition, showed good solubility in the membrane-mimetic system. In contrast, reconstitution experiments carried out with the lipid-depleted protein resulted in weak lipid binding and poor micellar incorporation. These results suggest that the tightly bound acidic lipids may stabilize a protein conformation required for insertion into the micellar system.

Proteins in membrane mimetic systems. Insertion of myelin basic protein into microemulsion droplets.

The insertion of myelin basic protein into microemulsion droplets of sodium bis (2-ethylhexyl) sulfosuccinate (AOT) has been studied by quasi-elastic light scattering. Measurements were made at both low and high molar ratios of water to surfactant, as a function of protein occupancy. The hydrodynamic radii of filled and empty droplets were experimentally evaluated. These were compared to values calculated using a water shell model of protein encapsulation, and excellent agreement was obtained. At low molar ratio of water to surfactant (w0 = 5.6), the hydrodynamic radius of filled droplets is significantly larger than the radius of empty ones. Under these conditions, about three empty (water-filled) droplets are required to build up a droplet of sufficient size to accommodate a single protein molecule. At maximum solubilization, which occurs at w0 = 5.6, a small fraction of droplets are found containing protein aggregates. In contrast, results at high values of w0 (22.4) reveal radii for empty and occupied droplets of comparable dimension, and the absence of aggregates. The results are discussed in terms of the model and the mechanism of interaction of this protein with the aqueous interfaces provided by these membrane-mimetic systems.

Membrane proteins in reverse micelles: myelin basic protein in a membrane-mimetic environment.

The solubility, reactivity, and conformational dynamics of myelin basic protein (MBP) from bovine brain were studied in reverse micelles of sodium bis(2-ethylhexyl) sulfosuccinate (AOT)-isooctane and water. Such a membrane-mimetic system resembles the aqueous spaces of native myelin sheath in terms of physicochemical properties as reflected in the high affinity of MBP for interfacial bound water. This is marked by the unusual profile of the solubility curve of the protein in reverse micelles, which shows optimal solubility at a much lower molar ratio of water to surfactant ([ H2O]/[AOT] = w0) than that reported for other water-soluble proteins. The role of counterions and/or charged polar head groups in the solubilization process is revealed by comparison of the solubility of MBP in nonionic surfactant micellar solutions. Whereas MBP is unfolded in aqueous solutions, insertion into reverse micelles generates a more folded structure, characterized by the presence of 20% alpha-helix. This conformation is unaffected by variations in the water content of the system (in the 2.0-22.4 w0 range). The reactivity of epsilon-amino groups of lysine residues with aqueous solutions of o-phthalaldehyde demonstrates that segments of the peptide chain are accessible to water. Similar results were obtained with the sequence involved in heme binding. In contrast, the sole tryptophan residue, Trp-117, is shielded from the aqueous solvent, as indicated by lack of reaction with N-bromosuccinimide. The invariance of the wavelength maximum emission in the fluorescence spectra as a function of w0 is consistent with this result.(ABSTRACT TRUNCATED AT 250 WORDS)

Solubilization and insertion into reverse micelles of the major myelin transmembrane proteolipid.

The Folch-Pi proteolipid has been isolated from bovine white matter and characterized with respect to phospholipid and glycolipid composition. The protein-lipid complex has been solubilized in aqueous reverse micelles of di(2-ethylhexyl) sodium sulfosuccinate and isooctane. Solubilization of this otherwise water-insoluble proteolipid requires small amounts of water, the percent of solubility being maximum for a low molar ratio of water to surfactant (Wo = 5.6). Unlike hydrophilic proteins, the extent of incorporation into the micellar system is negligible at 50 mM surfactant and reaches 90Vo only at 300 mM. However, the conformation of the proteolipid in reverse micelles as studied by fluorescence emission spectroscopy and circular dichroism was not affected by variations of the surfactant concentration. These results are consistent with the peculiar properties of the aqueous environment of the proteolipid within the reverse micelles and may reflect the membrane-like character of these bio-assemblies.

A heme binding site on myelin basic protein: characterization, location, and significance.

Myelin basic protein (MBP), an extrinsic membrane protein from the myelin sheath, binds dicyanohemin. The binding generates absorption bands in the Soret region and quenches the fluorescence emitted by the sole tryptophan residue. The absorption titration curves in the Soret demonstrate that the binding is stoichiometric, one heme per protein, with a large value of the extinction coefficient (8 X 10(4) M-1 cm-1 at 420 nm). Fluorescence quenching titration curves indicate an identical stoichiometry and a low quenching efficiency of 20%. From the heme titration curve the association constant between dicyanohemin and MBP is estimated to be greater than or equal to 10 nM-1 in 50 mM 4-morpholinepropanesulfonic acid buffer, pH 7.0, at 20 degrees C. Digestion of MBP by Staphylococcus aureus V8 protease yields a peptide (38-118) whose heme binding properties are identical to those of MBP. In contrast, peptides obtained by digestion of MBP with cathepsin D do not exhibit any specific binding of dicyanohemin. The cleavage of the Phe-Phe (42-43) bond appears to be critical in this respect. A comparison of the sequence immediately preceding, including these residues with a probable heme binding site of a mitochondrial cytochrome b, reveals a high degree of homology. The possible significance of heme binding is discussed.

The thiol groups of the Folch-Pi protein from bovine white matter. Exposure, reactivity and significance.

The number and the reactivity of accessible thiol groups of the Folch-Pi apoprotein and proteolipid (50% of myelin proteins) were studied, by using a specific thiol-disulphide interchange reaction, in connection with the known solubility of this protein in organic and aqueous solvents. The high reactivity of 2,2'-dipyridyl disulphide towards thiol groups leads to the titration of 4.8 mol of SH groups/mol of protein (Mr 30000) in alkaline and acidic chloroform/methanol (2:1, v/v). Unlike previous findings, this value was consistently found from batch to batch and remained stable with time. In the proteolipid 1 mol of SH groups/mol was not accessible as compared with the apoprotein. In aqueous solvents, a similar number of 4.4 mol of SH groups/mol was also found. For the first time, kinetic studies carried out in chloroform/methanol discriminated between two classes of thiol groups. The reaction of 2 mol of SH groups/mol was characterized by apparent second-order rate constants whose values were 5-10-fold higher than those of the other class. Kinetic studies and cyanylation experiments in aqueous solvents also indicated the high reactivity of these thiol groups with Ellman's reagent. Together with kinetic results, studies on the stoichiometry of the interchange reaction of equimolar solutions of protein and disulphide indicate that these highly reactive thiol groups are near to each other in the amino acid sequence. The location of the thiol groups at the boundary between hydrophilic and hydrophobic domains of the Folch-Pi protein is suggested in connection with their possible structural and biological significance.

The molecular size and shape of the Folch-Pi apoprotein in aqueous and organic solvents.

In 1% acetic acid, sedimentation velocity measurements and equilibrium ultracentrifuge experiments demonstrate that the Folch-Pi apoprotein is not monodisperse. The weight-average molecular weight calculated from ultracentrifuge experiments and combining sedimentation coefficient and viscosity measurements, ranged from 64000 to 80000. The intrinsic viscosity value suggests an asymetric shape for the apoprotein if a low value of hydration is considered. In dioxan/1% acetic acid (2:3, v/v) a smaller sedimentation coefficient was found, the intrinsic viscosity value remaining identical to that in 4% acetic acid. In pure 2-chloroethanol, light-scattering experiments led to a molecular weight of 165000 indicating that even in this solvent the protein is not monomeric. Intrinsic viscosity and light scattering measurements on the one hand, primary sequence on the other hand (six proline residues per monomer of Mr 23500) suggest that the molecule in 2-chloroethanol may consist of rod-like segments with flexible junctions.