Extended kalman filter for estimation of parameters in nonlinear state-space models of biochemical networks.

It is system dynamics that determines the function of cells, tissues and organisms. To develop mathematical models and estimate their parameters are an essential issue for studying dynamic behaviors of biological systems which include metabolic networks, genetic regulatory networks and signal transduction pathways, under perturbation of external stimuli. In general, biological dynamic systems are partially observed. Therefore, a natural way to model dynamic biological systems is to employ nonlinear state-space equations. Although statistical methods for parameter estimation of linear models in biological dynamic systems have been developed intensively in the recent years, the estimation of both states and parameters of nonlinear dynamic systems remains a challenging task. In this report, we apply extended Kalman Filter (EKF) to the estimation of both states and parameters of nonlinear state-space models. To evaluate the performance of the EKF for parameter estimation, we apply the EKF to a simulation dataset and two real datasets: JAK-STAT signal transduction pathway and Ras/Raf/MEK/ERK signaling transduction pathways datasets. The preliminary results show that EKF can accurately estimate the parameters and predict states in nonlinear state-space equations for modeling dynamic biochemical networks.

Molecular and biochemical analysis of the {\it Drosophila\/} segmentation gene {\it runt\/}

Genetic evidence has indicated that the segmentation gene runt plays a key role in regulating gene expression of the pair-rule genes hairy, even-skipped, and fushi tarazu. In contrast to other pair-rule genes, sequence data of the runt open reading frame did not reveal homologies to DNA-binding motifs of known transcriptional regulatory proteins. This thesis project examined several properties of the runt gene based on the sequence of the transcription unit, including the subcellular localization of the protein in vivo, its ability to bind DNA, and the functionality of a putative nucleotide binding domain.^ A runt-specific antibody was generated and used to demonstrate that runt is localized in the nucleus. Since the precise overlap of the pair-rule stripes is thought to be critical for the determination of cellular identity along the anterior-posterior axis, phasing of early runt expression in the blastoderm was examined with regard to the segmentation genes hairy, even-skipped, and fushi tarazu. runt was also expressed at later stages of embryogenesis, including expression in neuroblasts, and ganglion mother cells of the developing nervous system. Expression at this stage was required for the subsequent formation of specific neurons and runt was extensively expressed in the central and peripheral nervous systems.^ Several experiments were done to address the biochemical function of the runt protein. A direct interaction of runt with DNA was first examined. Although bacterial expressed runt was found to bind dsDNA-cellulose, subsequent experiments failed to detect sequence-specific interactions with DNA. Inter-species conservation of the putative nucleotide binding domain suggested that this region was functionally important, and runt protein bound a labeled ATP analog with high affinity in vitro. Finally, the effect of substitution of a critical residue of the nucleotide binding domain on runt activity was examined in vivo. Ectopic expression of the mutant protein indicated that this conserved substitution altered, but did not eliminate, runt activity as evaluated by segmentation phenotype and viability. ^

Biochemical characterization of PICK1, a PKC binding protein

Protein kinase C (PKC) is a family of serine-threonine kinases that are activated by a wide variety of hormones, neurotransmitters and growth factors. A single cell type contains multiple isoforms that are translocated to distinct and different subcellular sites upon mitogenic stimulus. Many different cellular responses are attributed to PKC activity though relatively few substrates or binding proteins have been definitively characterized. We used the hinge and catalytic domain of PKC$\alpha$ (PKC7) in a yeast two-hybrid screen to clone proteins that interact with C-kinase (PICKs). One protein which we have termed PICK1 may be involved in PKC$\alpha$-specific function at the level of the nuclear membrane after activation. Binding of PICK1 to PKC$\alpha$ has been shown to be isoform specific as it does not bind to PKC$\beta$II or PKC$\alpha$ in the yeast two-hybrid system. PICK1 mRNA expression level is highest in testis and brain with lower levels of expression in skeletal muscle, heart, kidney, lung and liver. PICK1 protein contains five PKC consensus phosphorylation sites and serves as an in vitro substrate for PKC. The PICK1 protein also contains a P-Loop motif that has been shown to bind ATP or GTP in the Ras family of oncoproteins as well as the G-Protein family. Proteins which bind ATP or GTP using this motif all have some sort of catalytic function although none has been identified for PICK1 as yet. PICK1 contains a DHR/GLGF motif at the N-terminus of the protein. The DHR/GLGF motif is contained in a number of recently described proteins and has been shown to mediate protein-protein interactions at the level of membranes and cytoskeleton. When both PKC$\alpha$ and PICK1 are co-expressed in Cos1 cells the two proteins co-localize to the perinucleus in immunoflouresence studies and co-immunoprecipitate. The binding site for PKC7 has been localized to amino acids 1-358 on PICK1 which contains the DHR/GLGF motif. Binding of PICK1 to PKC$\alpha$ requires the hinge and C-terminal domains of PKC$\alpha$. In vitro, PICK1 binds to PKC$\alpha$ and inhibits its activity as assayed by myelin basic protein phosphorylation. PICK1 also binds to TIS21, a primary response gene that is expressed in response to phorbol ester and growth factor treatment. The Caenorhabditis elegans homologue of PICK1 has been cloned and sequenced revealing a high degree of conservation in the DHR/GLGF motif. A more C-terminal region also shows a high degree of conservation, and the C. elegans PICK1 homologue binds to PKC7 suggesting a conservation of function. Taken together these results suggest that PICK1 may be involved in a PKC$\alpha$-specific function at the level of the nuclear membrane. ^

BIOCHEMICAL PROPERTIES OF GABA (B) RECEPTORS (BACLOFEN, ADENYLATE CYCLASE, CYCLIC-AMP, PHOSPHOLIPASE, PROTEIN KINASE C)

(gamma)-Aminobutyric acid (GABA), a neurotransmitter in the mammalian central nervous system, influences neuronal activity by interacting with at least two pharmacologically and functionally distinct receptors. GABA(,A) receptors are sensitive to blockade by bicuculline, are associated with benzodiazepine and barbiturate binding sites, and mediate chloride flux. The biochemical and pharmacolocal properties of GABA(,B) receptors, which are stereoselectively activated by (beta)-p-chlorophenyl GABA (baclofen), are less well understood. The aim of this study was to define these features of GABA(,B) receptors, with particular emphasis on their possible relationship to the adenylate cyclase system in brain.^ By themselves, GABA agonists have no effect on cAMP accumulation in rat brain slices. However, some GABA agonists markedly enhance the cAMP accumulation that results from exposure to norepinephrine, adenosine, VIP, and cholera toxin. Evidence that this response is mediated by the GABA(,B) system is provided by the finding that it is bicuculline-insensitive, and by the fact that only those agents that interact with GABA(,B) binding sites are active in this regard. GABA(,B) agonists are able to enhance neurotransmitter-stimulated cAMP accumulation in only certain brain regions, and the response is not influenced by phosphodiesterase inhibitors, although is totally dependent on the availability of extracellular calcium. Furthermore, data suggest that inhibition of phospholipase A(,2), a calcium-dependent enzyme, decreases the augmenting response to baclofen, although inhibitors of arachidonic acid metabolism are without effect. These findings indicate that either arachidonic acid or lysophospholipid, products of PLA(,2)-mediated degradation of phospholipids, mediates the augmentation. Moreover, phorbol esters, compounds which directly activate protein kinase C, were also found to enhance neurotransmitter-stimulated cAMP accumulation in rat brain slices. Since this enzyme is known to be stimulated by unsaturated fatty acids such as arachidonate, it is proposed that GABA(,B) agonists enhance cAMP accumulation by fostering the production of arachidonic acid which stimulates protein kinase C, leading to the phosphorylation of some component of the adenylate cyclase system. Thus, GABA, through an interaction with GABA(,B) receptors, modulates neurotransmitter receptor responsiveness in brain. The pharmocological manipulation of this response could lead to the development of therapeutic agents having a more subtle influence than current drugs on central nervous system function. ^

BIOCHEMICAL DETERMINANTS IN THE CYTOTOXIC ACTION OF 9-BETA- D- ARABINOFURANOSYLADENINE

Studies with 9-(beta)-D-arabinfuranosyladenine (ara-A) have illustrated that there are numerous cell functions which are affected by the nucleoside analog and its metabolites. Although the precise mechanism responsible for the cytotoxicity of this drug is not known, it is presently thought tha

CYTOLOGICAL, ULTRASTRUCTURAL, AND BIOCHEMICAL STUDIES OF THE STRUCTURE OF PREMATURELY CONDENSED CHROMOSOMES

The phenomenon of premature chromosome condensation, resulting from fusion between mitotic and interphase cells, includes dissolution of the interphase nuclear framework, thus allowing a direct visualization of interphase chromosomes. Light microscope morphology of prematurely condensed chromosomes (PCC) from synchronized HeLa cells supports the model of an interphase "chromosome condensation cycle". PCC are increasingly attenuated as cells progress through G(,1). A maximum degree of decondensation is observed at active sites of DNA replication during S phase, and a condensed morphology is rapidly resumed following completion of replication of a chromosome segment.^ To permit ultrastructural and biochemical studies of PCC, a procedure was developed to induce premature chromosome condensation at high frequency. This was achieved by polyethylene glycol (PEG)-mediated fusion of a dense monolayer of mitotic and interphase cells induced by centrifugation onto lectin-coated culture dishes. Using this method, PCC induction frequencies of 60-90% are routinely obtained.^ Scanning electron microscope analysis of PCC spreads revealed that the extension of PCC during progression through G(,1) is accompanied by a transition of the basic 30 nm chromatin fiber from tightly packed looping fibers to extended longitudinal fibers. Sites of active DNA replication is S-PCC were indicated to be organized a single longitudinal fibers. Following replication of a chromosome segment, a rapid reorganization from the extended longitudinal fiber to packed looping fibers occurs. The postreplication maturation process appears to include the assembly of a chromosome core consisting of multiple longitudinal fibers.^ The role of histone H1 phosphorylation in PCC formation was investigated by acidurea polyacrylamide gel electrophoresis of total histone extracted from metaphase chromosomes and PCC following high frequency fusion. This investigation failed to demonstrate an extensive phosphorylation of H1 associated with PCC formation. However, significant dephosphorylation of superphosphorylated metaphase chromosome H1 was observed, indicating that interphase H1-phosphatase activity is dominant over metaphase H1 kinase activity. These observations provide evidence against models suggesting a role for H1 superphosphorylation in triggering mitotic condensation of chromosomes. ^

BIOLOGICAL AND BIOCHEMICAL CHARACTERIZATION OF MURINE-TUMOR SPECIFIC TRANSPLANTATION ANTIGENS (ISOTACHOPHORESIS, IMMUNOGENICITY)

Tumor-specific transplantation antigens (TSTA) are individually distinct neoantigens expressed on the cells of chemically-induced neoplasms. TSTA are operationally defined by immunization of syngeneic mice against challenge with viable tumor cells. Immunization with cell surface or extracted TSTA induces specific resistance to transplanted tumor cells. The biological and biochemical nature of TSTA was investigated in the 3-methylcholanthrene-induced fibrosarcomas of female C3H/HeJ mice, MCA-F and MCA-D. Tumor cell suspensions were extracted by treatment with 3M KCl or 2.5% butanol solutions and the TSTA was partially purified by preparative isoelectric focusing. The isoelectric pH of TSTA purified from 3M KCl extracts was 5.8-6.0, and from butanol extracts was 6.4-6.6. Whereas immunization with 10('5) and 10('6) irradiated tumor cells induces complete rejection of tumor cell challenge over a two-fold-log dose range, immunization with ug quantities within a one-fold-log dose range of extracted TSTA induces only partial resistance to tumor challenge. Reduced immunogenicity of extracted TSTA is hypothesized to result from immunization of mice with insufficiently purified TSTA preparations. The hypothesis predicts that immunization with highly purified TSTA, free from interfering substances, induces complete rejection of tumor challenge over a broad dose range. To test the hypothesis preparative isotachophoresis (pITP) was used to purify TSTA from electrofocused TSTA fractions. Significant purification was achieved, as immunization with 15 pg to 1.5 ug (5 logs) of pITP-purified TSTA extracted from the MCA-F, or with 1 pg to 10 ng (4 logs) of TSTA from the MCA-D tumor induced specific resistance to tumor challenge. Despite 50,000 fold purification of TSTA, immunization induced partial, not complete, rejection of transplanted tumor cells. This suggests a clear dissociation of the immunogenicity and purification of extracted TSTA, indicating that the induction of partial immunity to tumor challenge is an intrinsic property of extracted TSTA.^

THE EFFECT OF RELAXIN ON BIOCHEMICAL AND MORPHOLOGICAL PARAMETERS OF RAT MYOMETRIAL CELLS IN CULTURE (CYCLIC-AMP, CELL SHAPE CHANGE)

Relaxin is able to inhibit spontaneous, oxytocin-and prostaglandin-driven uterine contractions. The intracellular mechanism of action of relaxin on uterine relaxation had previously been studied using isometrically suspended uterine strips. Since uterine strips contain stroma as well as myometrium, the changes in biochemical parameters induced by relaxin treatment may not occur in the same cell types responsible for the physical changes. In these studies, cultures of enriched populations of rat myometrial cells were used to investigate the effect of relaxin on biochemical and morphological parameters which are related to relaxation.^ Under optimal culture conditions (initial plating density 1 - 1.5 x 10('6)cells/ml, 3 ml/35 mm dish, 2 days culture), enzymatically isolated rat myometrial cells were able to respond to relaxin with cAMP elevation. Relaxin elevated cAMP levels in the presence but not the absence of 0.1 mM methylisobutylxanthine or 0.4 um forskolin in a time- and concentration-dependent manner. In contrast, isoproterenol was able to elevate cAMP levels in the presence and absence of 0.1 mM methylisobutylxanthine.^ Oxytocin treatment caused a decrease in mean cell length and area of myometrial cells in culture which could be considered analogous to contraction. Under optimal culture conditions, relaxin increased myometrial cell length and area (i.e. analogous to relaxation) of oxytocin-treated cells in a time- and concentration-dependent manner. Other relaxants such as isoproterenol and dibutyryl cAMP also increased cell length and area of oxytocin - treated myometrial cells in culture.^ Under optimal culture conditions, relaxin decreased myosin light chain kinase activity in a time-and concentration-dependent manner by increasing the K(,50) of the enzyme for calmodulin (CaM), i.e. decreasing the affinity of the enzyme for CaM. The decrease in the affinity of myosin light chain kinase for CaM may be due to the phosphorylation of the enzyme by cAMP-dependent protein kinase. Relaxin also decreased the Ca('2+)(.)CaM-independent myosin light chain kinase activity to a lesser extent than that of the Ca('2+)(.)CaM-dependent enzyme activity. This was not attributable to a decrease in the affinity of the enzyme for myosin in myometrial cells in culture, in contrast to the finding of such a change following relaxin treatment of uterine strips. Further studies are required to clarify this point.^ There was a temporal association between the effects of relaxin on elevation of cAMP levels in the presence of 0.4 uM forskolin, increase in cell length and decrease in myosin light chain kinase activity. . . . (Author's abstract exceeds stipulated maximum length. Discontinued here with permission of author.) UMI ^

BIOCHEMICAL CHARACTERIZATION AND GENETIC MAPPING OF WILD TYPE AND MUTANT GLYCERALDEHYDE-3-PHOSPHATE DEHYDROGENASE ALLELES IN CHINESE HAMSTER OVARY CELLS

A UV-induced mutation of the enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPD) was characterized in the CHO clone A24. The asymmetric 4-banded zymogram and an in vitro GAPD activity equal to that of wild type cells were not consistent with models of a mutant heterozygote producing equal amounts of wild type and either catalytically active or inactive mutant subunits that interacted randomly. Cumulative evidence indicated that the site of the mutation was the GAPD structural locus expressed in CHO wild type cells, and that the mutant allele coded for a subunit that differed from the wild type subunit in stability and kinetics. The evidence included the appearance of a fifth band, the putative mutant homotetramer, after addition of the substrate glyceraldehyde-3-phosphate (GAP) to the gel matrix; dilution experiments indicating stability differences between the subunits; experiments with subsaturating levels of GAP indicating differences in affinity for the substrate; GAPD zymograms of A24 x mouse hybrids that were consistent with the presence of two distinct A24 subunits; independent segregation of A24 wild type and mutant electrophoretic bands from the hybrids, which was inconsistent with models of mutation of a locus involved in posttranslational modification; the mapping of both wild type and mutant forms of GAPD to chromosome 8; and the failure to detect any evidence of posttranslational modification (of other A24 isozymes, or through mixing of homogenates of A24 and mouse).^ The extent of skewing of the zymogram toward the wild type band, and the unreduced in vitro activity were inconsistent with models based solely on differences in activity of the two subunits. Comparison of wild type homotetramer bands in wild type cells and A24 suggested the latter had a preponderance of wild type subunits over mutant subunits, and had more GAPD tetramers than did CHO controls.^ Two CHO linkages, GAPD-triose phosphate isomerase, and acid phosphatase 2-adenosine deaminase were reported provisionally, and several others were confirmed. ^

FUNCTIONAL AND BIOCHEMICAL CHARACTERIZATION OF ADHESION DEFECTIVE CHINESE HAMSTER OVARY CELLS

Cell adhesion is an intricate process involving adhesion promoting ligands such as laminin and fibronectin, surface receptors for these ligands and a complex interplay of metabolic and cytoskeletal events (Geiger, BBA 737:305, 1983). Although considerable effort has been directed towards studying adhesion molecules such as fibronectin (Fn), very little is known about the mechanisms regulating the complex process of adhesion.^ I chose to use a CHO adhesion variant clone called AD('v)F11 as a tool to study the various steps which may be involved in adhesion. AD('v)F11 cells unlike wild type (WT), do not adhere to Fn-coated substrata, but will adhere to substrata coated with other extracellular components (Harper and Juliano, J Cell Biol. 91:647, 1981). I have found that although AD('v)F11 cells can bind Fn-coated latex beads to the same extent as WT cells, AD('v)F11 cells also differed from WT cells in that they did not aggregate in the presence of Fn-beads nor internalize Fn-beads. The defect in bead induced cell aggregation and internalization seem to be specific to Fn since lectin coated beads could aggregate AD('v)F11 cells as well as WT cells, and AD('v)F11 cells can also readily internalize lectins. These observations suggest that the defect associated with AD('v)F11 cells is distal to the initial binding to Fn to its cell surface receptor. To further investigate the biochemical defect associated with AD('v)F11 cells, a panel of compounds were examined for their ability to correct the non-adhesive phenotype of AD('v)F11 cells. Among the compounds tested, only those known to increase intracellular cAMP levels were found to be effective in correcting the adhesion defect of F11CA11 cells, a subclone of AD('v)F11 cells.^ Since cAMP effects in eukaryotic cells are mediated through phosphorylation events by the cAMP-dependent protein kinase (cAdPK) system, the phosphorylation pattern and cAdPK system of the F11CA11 cells were analyzed. Comparison between the phosphorylation pattern of intact untreated F11CA11 and WT cells, revealed the presence of a 50 kd phosphoprotein(s) in WT cells but not in F11CA11 cells. Results presented in this dissertation strongly indicate that the adhesion defect in F11CA11 is associated to an altered type I cAdPK that can be corrected by raising intracellular cAMP levels. (Abstract shortened with permission of author.) ^