The design of fault tolerant software for loosely coupled distributed systems

Requirements for systems to continue to operate satisfactorily in the presence of faults has led to the development of techniques for the construction of fault tolerant software. This thesis addresses the problem of error detection and recovery in distributed systems which consist of a set of communicating sequential processes. A method is presented for the `a priori' design of conversations for this class of distributed system. Petri nets are used to represent the state and to solve state reachability problems for concurrent systems. The dynamic behaviour of the system can be characterised by a state-change table derived from the state reachability tree. Systematic conversation generation is possible by defining a closed boundary on any branch of the state-change table. By relating the state-change table to process attributes it ensures all necessary processes are included in the conversation. The method also ensures properly nested conversations. An implementation of the conversation scheme using the concurrent language occam is proposed. The structure of the conversation is defined using the special features of occam. The proposed implementation gives a structure which is independent of the application and is independent of the number of processes involved. Finally, the integrity of inter-process communications is investigated. The basic communication primitives used in message passing systems are seen to have deficiencies when applied to systems with safety implications. Using a Petri net model a boundary for a time-out mechanism is proposed which will increase the integrity of a system which involves inter-process communications.

Mechanism of attenuation of skeletal muscle atrophy by zinc-α2-glycoprotein

The mechanism by which the adipokine zinc-a2-glycoprotein (ZAG) increases the mass of gastrocnemius, but not soleus muscle of diabetic mice, has been evaluated both in vivo and in vitro. There was an increased phosphorylation of both double-stranded RNA-dependent protein kinase and its substrate, eukaryotic initiation factor-2a, which was attenuated by about two-thirds in gastrocnemius but not soleus muscle of ob/ob mice treated with ZAG (50 µg, iv daily) for 5 d. ZAG also reduced the expression of the phospho forms of p38MAPK and phospholipase A2, as well as expression of the ubiquitin ligases (E3) muscle atrophy F-box/atrogin-1 and muscle RING finger protein, and the increased activity of both caspase-3 and casapse-8 to values found in nonobese controls. ZAG also increased the levels of phospho serine-threonine kinase and mammalian target of rapamycin in gastrocnemius muscle and reduced the phosphorylation of insulin receptor substrate-1 (Ser307) associated with insulin resistance. Similar changes were seen with ZAG when murine myotubes were incubated with high glucose concentrations (10 and 25 mm), showing that the effect of ZAG was direct. ZAG produced an increase in cAMP in murine myotubes, and the effects of ZAG on protein synthesis and degradation in vitro could be replicated by dibutyryl cAMP. ZAG increased cAMP levels of gastrocnemius but not soleus muscle. These results suggest that protein accretion in skeletal muscle in response to ZAG may be due to changes in intracellular cAMP and also that ZAG may have a therapeutic application in the treatment of muscle wasting conditions.

Mechanism of attenuation of muscle protein degradation induced by tumor necrosis factor-alpha and angiotensin II by beta-hydroxy-beta-methylbutyrate

Both tumor necrosis factor-alpha (TNF-alpha)/interferon-gamma (IFN-gamma) and angiotensin II (ANG II) induced an increase in total protein degradation in murine myotubes, which was completely attenuated by treatment with beta-hydroxy-beta-methylbutyrate (HMB; 50 microM). There was an increase in formation of reactive oxygen species (ROS) within 30 min, as well as an increase in the activity of both caspase-3 and -8, and both effects were attenuated by HMB. Moreover, inhibitors of caspase-3 and -8 completely attenuated both ROS formation and total protein degradation induced by TNF-alpha/IFN-gamma and ANG II. There was an increased autophosphorylation of double-stranded RNA-dependent protein kinase (PKR), which was attenuated by the specific caspase-3 and -8 inhibitors. Neither ROS formation or protein degradation occurred in myotubes expressing a catalytically inactive PKR variant, PKRDelta6, in response to TNF-alpha/IFN-gamma, compared with myotubes expressing wild-type PKR, although there was still activation of caspase-3 and -8. HMB also attenuated activation of PKR, suggesting that it was important in protein degradation. Formation of ROS was attenuated by rotenone, an inhibitor of the mitochondrial electron transport chain, nitro-l-arginine methyl ester, an inhibitor of nitric oxide synthase, and SB 203580, a specific inhibitor of p38 mitogen-activated protein kinase (p38 MAPK), which also attenuated total protein degradation. Activation of p38 MAPK by PKR provides the link to ROS formation. These results suggest that TNF-alpha/IFN-gamma and ANG II induce muscle protein degradation by a common signaling pathway, which is attenuated by HMB, and that this involves the initial activation of caspase-3 and -8, followed by autophosphorylation and activation of PKR, which then leads to increased ROS formation via activation of p38 MAPK. Increased ROS formation is known to induce protein degradation through the ubiquitin-proteasome pathway.

Mechanism of the attenuation of proteolysis-inducing factor stimulated protein degradation in muscle by β-hydroxy-β-methylbutyrate

The leucine metabolite β-hydroxy-β-methylbutyrate (HMB) prevents muscle protein degradation in cancer-induced weight loss through attenuation of the ubiquitin-proteasome proteolytic pathway. To investigate the mechanism of this effect, the action of HMB on protein breakdown and intracellular signaling leading to increased proteasome expression by the tumor factor proteolysis-inducing factor (PIF) has been studied in vitro using murine myotubes as a surrogate model of skeletal muscle. A comparison has been made of the effects of HMB and those of eicosapentaenoic acid (EPA), a known inhibitor of PIF signaling. At a concentration of 50 μmol/L, EPA and HMB completely attenuated PIF-induced protein degradation and induction of the ubiquitin-proteasome proteolytic pathway, as determined by the "chymotrypsin-like" enzyme activity, as well as protein expression of 20S proteasome α- and β-subunits and subunit p42 of the 19S regulator. The primary event in PIF-induced protein degradation is thought to be release of arachidonic acid from membrane phospholipids, and this process was attenuated by EPA, but not HMB, suggesting that HMB might act at another step in the PIF signaling pathway. EPA and HMB at a concentration of 50 μmol/L attenuated PIF-induced activation of protein kinase C and the subsequent degradation of inhibitor κBα and nuclear accumulation of nuclear factor κB. EPA and HMB also attenuated phosphorylation of p42/44 mitogen-activated protein kinase by PIF, thought to be important in PIF-induced proteasome expression. These results suggest that HMB attenuates PIF-induced activation and increased gene expression of the ubiquitin-proteasome proteolytic pathway, reducing protein degradation.

Development of an in-vitro model system to investigate the mechanism of muscle protein catabolism induced by proteolysis-inducing factor

The mechanism of muscle protein catabolism induced by proteolysis-inducing factor, produced by cachexia-inducing murine and human tumours has been studied in vitro using C2C12 myoblasts and myotubes. In both myoblasts and myotubes protein degradation was enhanced by proteolysis-inducing factor after 24 h incubation. In myoblasts this followed a bell-shaped dose-response curve with maximal effects at a proteolysis-inducing factor concentration between 2 and 4 nM, while in myotubes increased protein degradation was seen at all concentrations of proteolysis-inducing factor up to 10 nM, again with a maximum of 4 nM proteolysis-inducing factor. Protein degradation induced by proteolysis-inducing factor was completely attenuated in the presence of cycloheximide (1 μM), suggesting a requirement for new protein synthesis. In both myoblasts and myotubes protein degradation was accompanied by an increased expression of the α-type subunits of the 20S proteasome as well as functional activity of the proteasome, as determined by the 'chymotrypsin-like' enzyme activity. There was also an increased expression of the 19S regulatory complex as well as the ubiquitin-conjugating enzyme (E214k), and in myotubes a decrease in myosin expression was seen with increasing concentrations of proteolysis-inducing factor. These results show that proteolysis-inducing factor co-ordinately upregulates both ubiquitin conjugation and proteasome activity in both myoblasts and myotubes and may play an important role in the muscle wasting seen in cancer cachexia. © 2002 Cancer Research UK.

Mechanism of attenuation of skeletal muscle protein catabolism in cancer cachexia by eicosapentaenoic acid

Cancer cachexia is characterized by selective depletion of skeletal muscle protein reserves. Soleus muscles from mice bearing a cachexia-inducing tumor (MAC16) showed an increased protein degradation in vitro, as measured by tyrosine release, when compared with muscles from nontumor-bearing animals. After incubation under conditions that modify different proteolytic systems, lysosomal, calcium-dependent, and ATP-dependent proteolysis were found to contribute to the elevated protein catabolism. Treatment of mice bearing the MAC16 tumor with the polyunsaturated fatty acid, eicosapentaenoic acid (EPA), attenuated loss of body weight and significantly suppressed protein catabolism in soleus muscles through an inhibition of an ATP-dependent proteolytic pathway. The ATP-ubiquitin-dependent proteolytic pathway is considered to play a major role in muscle catabolism in cachexia, and functional proteasome activity, as determined by “chymotrypsin-like” enzyme activity, was significantly elevated in gastrocnemius muscle of mice bearing the MAC16 tumor as weight loss progressed. When animals bearing the MAC16 tumor were treated with EPA, functional proteasome activity was completely suppressed, together with attenuation of the expression of 20S proteasome a-subunits and the p42 regulator, whereas there was no effect on the expression of the ubiquitin-conjugating enzyme (E214k). These results suggest that EPA induces an attenuation of the up-regulation of proteasome expression in cachectic mice, and this was correlated with an increase in myosin expression, confirming retention of contractile proteins. EPA also inhibited growth of the MAC16 tumor in a dose-dependent manner, and this correlated with suppression of the expression of the 20S proteasome a-subunits in tumor cells, suggesting that this may be the mechanism of tumor growth inhibition. Thus EPA antagonizes loss of skeletal muscle proteins in cancer cachexia by down-regulation of proteasome expression, and this may also be the mechanism for inhibition of tumor growth.

A structural method for the design of fault tolerant distributed control systems

Distributed digital control systems provide alternatives to conventional, centralised digital control systems. Typically, a modern distributed control system will comprise a multi-processor or network of processors, a communications network, an associated set of sensors and actuators, and the systems and applications software. This thesis addresses the problem of how to design robust decentralised control systems, such as those used to control event-driven, real-time processes in time-critical environments. Emphasis is placed on studying the dynamical behaviour of a system and identifying ways of partitioning the system so that it may be controlled in a distributed manner. A structural partitioning technique is adopted which makes use of natural physical sub-processes in the system, which are then mapped into the software processes to control the system. However, communications are required between the processes because of the disjoint nature of the distributed (i.e. partitioned) state of the physical system. The structural partitioning technique, and recent developments in the theory of potential controllability and observability of a system, are the basis for the design of controllers. In particular, the method is used to derive a decentralised estimate of the state vector for a continuous-time system. The work is also extended to derive a distributed estimate for a discrete-time system. Emphasis is also given to the role of communications in the distributed control of processes and to the partitioning technique necessary to design distributed and decentralised systems with resilient structures. A method is presented for the systematic identification of necessary communications for distributed control. It is also shwon that the structural partitions can be used directly in the design of software fault tolerant concurrent controllers. In particular, the structural partition can be used to identify the boundary of the conversation which can be used to protect a specific part of the system. In addition, for certain classes of system, the partitions can be used to identify processes which may be dynamically reconfigured in the event of a fault. These methods should be of use in the design of robust distributed systems.

Modulation of tissue transglutaminase in tubular epithelial cells alters extracellular matrix levels: a potential mechanism of tissue scarring

The up-regulation and trafficking of tissue transglutaminase (TG2) by tubular epithelial cells (TEC) has been implicated in the development of kidney scarring. TG2 catalyses the crosslinking of proteins via the formation of highly stable e(?-glutamyl) lysine bonds. We have proposed that TG2 may contribute to kidney scarring by accelerating extracellular matrix (ECM) deposition and by stabilising the ECM against proteolytic decay. To investigate this, we have studied ECM metabolism in Opossum kidney (OK) TEC induced to over-express TG2 by stable transfection and in tubular cells isolated from TG2 knockout mice. Increasing the expression of TG2 led to increased extracellular TG2 activity (p < 0.05), elevated e(?-glutamyl) lysine crosslinking in the ECM and higher levels of ECM collagen per cell by 3H-proline labelling. Immunofluorescence demonstrated that this was attributable to increased collagen III and IV levels. Higher TG2 levels were associated with an accelerated collagen deposition rate and a reduced ECM breakdown by matrix metalloproteinases (MMPs). In contrast, a lack of TG2 was associated with reduced e(?-glutamyl) lysine crosslinking in the ECM, causing reduced ECM collagen levels and lower ECM per cell. We report that TG2 contributes to ECM accumulation primarily by accelerating collagen deposition, but also by altering the susceptibility of the tubular ECM to decay. These findings support a role for TG2 in the expansion of the ECM associated with kidney scarring.

Mechanism of action of the tetraflex accommodative intraocular lens

PURPOSE:To investigate the mechanism of action of the Tetraflex (Lenstec Kellen KH-3500) accommodative intraocular lens (IOL). METHODS:Thirteen eyes of eight patients implanted with the Tetraflex accommodating IOL for at least 2 years underwent assessment of their objective amplitude-of-accommodation by autorefraction, anterior chamber depth and pupil size with optical coherence tomography, and IOL flexure with aberrometry, each viewing a target at 0.0 to 4.00 diopters of accommodative demand. RESULTS:Pupil size decreased by 0.62+/-0.41 mm on increasing accommodative demand, but the Tetraflex IOL was relatively fixed in position within the eye. The ocular aberrations of the eye changed with increased accommodative demand, but not in a consistent manner among individuals. Those aberrations that appeared to be most affected were defocus, vertical primary and secondary astigmatism, vertical coma, horizontal and vertical primary and secondary trefoil, and spherical aberration. CONCLUSIONS:Some of the reported near vision benefits of the Tetraflex accommodating IOL appear to be due to changes in the optical aberrations because of the flexure of the IOL on accommodative effort rather than forward movement within the capsular bag.

Mechanism of action of a tumour derived lipid mobilising factor

Cancer cachexia comprises unintentional and debilitating weight loss associated with certain tumour types. Fat loss in cachexia is mediated by a 43kDa Lipid Mobilising Factor (LMF) sharing homology with endogenous Zinc-α2-Glycoprotein (ZAG). LMF and ZAG induced significant lipolysis in isolated epidydimal adipose tissue. This is attenuated by co-incubation with 10μM of antagonist SR59230A and partially attenuated by 25μM PD098059 (indicating β3-AR and MAPK involvement respectively). LMF/ZAG induced in vitro lipid depletion in differentiated 3T3-L1 adipocytes that seen to comprise a significant increase in lipolysis (p<0.01), with only a modest decrease in lipid synthesis (p=0.09). ZAG significantly increased in vitro protein synthesis (p<0.01) in C2C12 myotubes (without an effect on protein degradation). This increase was activated at transcription and attenuated by co-incubation with 10μM SR59230A. Proteolytic digestion of ZAG and LMF followed by sephadex G50 chromatography yielded active fragments of 6-15kDa, indication the entire molecule was not required for bioactivity. Cachexigenic MAC16 cells demonstrated significant in vitro ZAG expression over non-cachexigenic MAC13 cells (p<0.001). WAT and BAT excised from MAC16 mice of varying weight loss demonstrated increased ZAG expression compared to controls. Dosing of NMRI mice with s/c ZAG failed to reproduce this up-regulation, thus another cachectic factor is responsible. 0.58nM LMF conferred significant protection against hydrogen peroxide, paraquat and bleomycin-induced oxidative stress in the non-cachexigenic MAC13 cell line. This protection was attenuated by 10μM SR59230A indicating a β3-AR mediated effect. In addition, 0.58nM LMF significantly up regulated UCP2 expression (p<0.001), (a mitochondrial protein implicated in the detoxification of ROS) implying this to be the mechanism by which survival was achieved. In vitro, LMF caused significant up-regulation of UCP1 in BAT and UCP2 and 3 in C2C12 myotubes. This increase in uncoupling protein expression further potentiates the negative energy balance and wasting observed in cachexia.

Studies on the mechanism of action on antitumour imidazotetrazinones

The imidazotetrazinones are clinically active antitumour agents, temozolomide currently proving successful in the treatment of melanomas and gliomas. The exact nature of the biological processes underlying response are as yet unclear.This thesis attempts to identify the cellular targets important to the cytotoxicity of imidazotetrazinones, to elucidate the pathways by which this damage leads to cell death, and to identify mechanisms by which tumour cells may circumvent this action. The levels of the DNA repair enzymes O6-alkylguanine-DNA-alkyltransferase (O6-AGAT) and 3-methyladenine-DNA-glycosylase (3MAG) have been examined in a range of murine and human cell lines with differential sensitivity to temozolomide. All the cell lines were proficient in 3MAG despite there being 40-fold difference in sensitivity to temozolomide. This suggests that while 3-methyladenine is a major product of temozolomide alkylation of DNA it is unlikely to be a cytotoxic lesion. In contrast, there was a 20-fold variation in O6-AGAT levels and the concentration of this repair enzyme correlated with variations in cytotoxicity. Furthermore, depletion of this enzyme in a resistant, O6-AGAT proficient cell line (Raji), by pre-treatment with the free base O6-methylguanine resulted in 54% sensitisation to the effects of temozolomide. These observations have been extended to 3 glioma cell lines; results that support the view that the cytotoxicity of temozolomide is related to alkylation at the O6-position of guanine and that resistance to this drug is determined by efficient repair of this lesion. It is clear, however, the other factors may influence tumour response since temozolomide showed little differential activity towards 3 established solid murine tumours in vivo, despite different tumour O6-AGAT levels. Unlike mitozolomide, temozolomide is incapable of cross-linking DNA and a mechanism by which O6-methylguanine may exert lethality is unclear. The cytotoxicity of the methyl group may be due to its disruption of DNA-protein interactions, or alternatively cell death may not be a direct result of the alkyl group itself, but manifested by DNA single-strand breaks. Enhanced alkaline elution rates were found for the DNA of Raji cells treated with temozolomide following alkyltransferase depletion, suggesting a relationship between O6-methylguanine and the induction single-strand breaks. Such breaks can activate poly(ADP-ribose) synthetase (ADPRT) an enzyme capable of rapid and lethal depletion of cellular NAD levels. However, at concentrations of temozolomlde relevant in vivo little change in adenine nucleotides was detected in cell lines, although this enzyme would appear important in modulating DNA repair since inhibition of ADPRT potentiated temozolomide cytotoxicity in Raji cells but not O6-AGAT deficient GM892A cells. Cell lines have been reported that are O6-AGAT deficient yet resistant to methylating agents. Thus, resistance to temozolomide may arise not only by removal of the methyl group from the O6-position of guanine, but also from another mechanism involving caffeine-sensitive post-replication repair or mismatch repair activity. A modification of the standard Maxam Gilbert sequencing technique was used to determine the sequence specificity of guanine-N7 alkylation. Temozolomide preferentially alkylated runs of guanines with the intensity of reaction increasing with the number of adjacent guanines in the DNA sequence. Comparable results were obtained with a polymerase-stop assay, although neither technique elucidates the sequence specificity of O6-guanine alkylation. The importance of such specificity to cytotoxicity is uncertain, although guanine-rich sequences are common to the promoter regions of oncogenes. Expression of a plasmid reporter gene under the control of the Ha-ras proto~oncogene promoter was inhibited by alkylation with temozolomide when transfected into cancer cell lines, However, this inhibition did not appear to be related to O6~guanine alkylation and therefore would seem unimportant to the chemotherapeutic activity of temozolomide.

Mechanism of the hydropyrolysis of Manvers coal

Manvers coal has been pyrolysed to 500ºC in a stirred autoclave under various pressures of nitrogen (pyrolysis) and hydrogen (hydropyrolysis). All products were investigated. Pyrolysis of coals involves the transfer of hydrogen atoms from one part of their structure to another. In the above experiments there was no way of labelling the hydrogen or of distinguishing between hydrogen which was initially part of the coal and hydrogen originating in the external atmosphere. Consequently, Manvers coal has been pyrolysed in an atmosphere of deuterium in order to obtain greater insight into the mechanism of hydropyrolysis. In particular it was hoped to distinguish between direct hydrogenation (deuteration!) of the coal and the products of pyrolysis and the 'shuttling' of hydrogen atoms between different parts of the pyrolysing coal. The addition to the coal of 5% (wt.% of coal) of either tetralin or pyrite was also studied. A variety of techniques were used to analyse the products of pyrolysis: gas chromatography - mass spectrometry and high performance liquid chromatography for tars; thermal conductivity gas chromatography and high resolution mass spectrometry for gases; methanol densities, microporosities and diffuse reflectance infra red spectroscopy for the cokes (chars); refractive index to determine deuterium in the liquor. An attempt has been made to apply basic thermodynamics to reactions which are likely to occur in the hydropyrolysis of coals. Diffusion and effusion rates for hydrogen and tar molecules have also been estimated.