<p>Asthma is a condition of the airway characterized by 1) a reversible increase in airway resistance; 2) airway hyperresponsiveness; and 3) airway inflammation (Juniper et al., 1981; O'Byrne, 1986; Boushey and Fahy, 1995). Defects of the airway epithelium have been suggested to play a role in the pathogenesis of asthma (Goldie et al., 1986; Knight et al., 1994), and loss of the epithelium is associated with increased reactivity of the underlying smooth muscle (Jongejen et al., 1991; Candenas et al., 1992). It has been proposed that the airway epithelium releases one or more factors which inhibit smooth muscle contraction, termed the epithelial derived inhibitory factor (Tschirhart et al., 1987; Fernandes et al., 1989; Ullman et al., 1991). The inhibitory prostaglandin PGE₂ has been demonstrated to be released from the epithelium (Barrel and Bigby, 1995) and to modulate airway smooth muscle contraction (Braunstein et al., 1988; Abela and Daniel, 1995). In patients with asthma, inhalation of an allergen may result in a biphasic response consisting of an early asthmatic response and a late asthmatic response (Dolovich et al., 1989; Sterk et al., 1993). The late asthmatic response is an indirect measure of allergen-induced airway inflammation (Dolovich et al., 1989; Sterk et al., 1993) and an animal model of the late asthmatic response is produced by inhalation of the allergen Ascaris suum by dogs (Sasaki et al., 1987). in the studies presented in this thesis, the release and actions of PGE₂ from canine airway epithelium, both under unstimulated conditions and following inhalation of the Ascaris suum antigen, were examined. Tracheal and bronchial tissues were excised and studied in the organ bath where contractile responses to agonists and electrical field stimulation, as well as PGE₂ release were measured. Finally, a potential mechanism by which PGE₂ may effect smooth muscle relaxation was examined. In unstimulated animals, PGE₂ was released from tracheal epithelium and inhibited smooth muscle contraction. This release of PGE₂ was dependent upon electrical field stimulation, and was not blocked by the addition of neurotoxins. In the antigen model, tracheal PGE₂ release was increased from animals that inhaled antigen but did not develop late airway hyperresponsiveness comapred to animals that inhaled vehicle or inhaled allergen and did develop a late response. The release of PGE₂ was not dependent on field stimulation. Demonstration of in vitro hyperresponsiveness of the tracheal smooth muscle was dependent on removal of the epithelium. Bronchial smooth muscle from the antigen model did not demonstrate in vitro hyperresponsiveness, even when hyperresponsiveness was observed in vivo. There was an increase in the basal release of PGE₂ from the bronchi of animals that were hyperresponsive in vivo. PGE₂ increases intracellular cAMP concentrations (Madison et al., 1989; Coleman et al., 1994). Our investigations in tracheal smooth muscle demonstrated that cAMP does not lower intracellular Ca²⁺ and cause relaxation of airway smooth muscle by stimulation of the sarcoplasmic reticulum Ca²⁺ pump. The results indicate that PGE₂ was released from airway epithelium and modulated smooth muscle contraction. Alterations in the release of PGE₂ were demonstrated in an animal model of asthma, as PGE₂ played a protective role in the development of airway hyperresponsiveness. Modulation of PGE₂ may be a possible therapy for the treatment of asthma and prevention of the late asthmatic response.</p> / Doctor of Philosophy (PhD)

<p>Human adenoviruses are known to transform rodent cells in culture and these cells are tumorgenic when injected into new born animals. It has been well established that the early region 1 (El) of human adenovirus type 5 is necessary and sufficient for oncogenic transformation. The El region is comprised of two transcription units known as E1A (0 to 4.5% of the genome) and E1B (4.5 to 11.2%), each of which produces multiple species of mRNAs and polypeptides. E1A is also required to activate the transcription of other viral early regions. In the present study anti-peptide sera were used to identify and characterize these viral proteins.</p> <p>Anti-peptide sera specific for the amino- and carboxy-termini of E1A were raised and these two sera precipitated an identical set of four major polypeptides of 52, 50, 48.5, and 45K and two minor species of 37.5 and 35K. Studies using E1A mutant viruses also revealed that 52, 48.5, and 37.5K polypeptides are derived from the 1.1 kb mRNA, and the 50, 45, and 35K species from the 0.9 kb mRNA of E1A. These sera were also used to identify polypeptides that are associated with E1 proteins. A set of five cellular polypeptides consisting of >250K, 105K (doublet), 68K, and 65K species were found to co-precipitate with E1A proteins under various conditions and the nature of this association was investigated using the anti-peptide sera as well as an E1A-specific monoclonal activity.</p> <p>Antisera against synthetic peptides corresponding to the both termini of E1B 58K were also raised and used to identfy 58K from wild-type and mutant-infected cells. It had previously been shown that protein kinase activity was associated with 58K. To ask if protein kinase activity was intrinsic to this viral protein several conventional methods were used to purify 58K and the results suggested that such activity may be intrinsic to this viral protein.</p> <p>The anti-peptide sera were used to purify El proteins. A simple purification procedure using these sera and their corresponding synthetic peptides was developed and highly purified 58K and E1A proteins were obtained. Attempts were made to study protein kinase activity using these purified El proteins, however, the results were inconclusive and it was not possible to unequivocally determine if kinase activity was intrinsic to them.</p> / Doctor of Philosophy (PhD)

<p>These studies were undertaken to characterize cytotoxic cell responses to Pichinde virus (PV; a member of the arenavirus family) in various strains of inbred mice. Emphasis was placed on examining the relationship between cells with natural killer (NK) activity and H-2 restricted, virus-specific cytotoxic T lymphocytes (CTL) that are detected in the spleens of inbred mice after infection with PV.</p> <p>Primary i.v. inoculation of mice with PV resulted in augmented spleen NK activity that peaked at 3- 4 days after infection. This NK response was followed by an H-2 restricted, virus-specific CTL response that peaked 3 days later. Rechallenge of PV-primed mice with homologous virus resulted in a slight but significant increase in spleen NK activity 1 day after reinfection, and this was followed 3 days later by peak CTL activity. Thus, memory cell-mediated immune responses appeared more rapidly after secondary in vivo challenge with PV. Furthermore, the temporal kinetic relationship between virus-induced NK and CTL responses was maintained after both primary and secondary infection with PV, which suggested that virus-induced NK cells may represent pre-CTL.</p> <p>To investigate the relationship between these two cytotoxic cell populations, expression of lineage specific cell-surface antigens on virus-induced NK and CTL effectors was examined. NK cells induced after primary and secondary infection with PV were found to rapidly acquire the pan-T cell marker Thy-1, which was expressed on mature anti-viral CTL. In addition, asialo-GM 1 (a glycolipid which has been considered a marker of NK cells) appears to be expressed on PV-specific CTLp; treatment of PV-primed spleen cells with a polyclonal rabbit antiserum to this marker plus complement prior to secondary in vitro restimulation with PV-infected macrophages prevented the generation of secondary CTL responses to PV. Furthermore, multiple i.v. injections of this antiserum were able to abrogate the in vivo generation of both NK and CTL responses after primary or secondary infection with PV.</p> <p>Secondary NK and CTL responses were generated in mice that had been pretreated with cyclophosphamide (CY), suggesting that memory cell-mediated immune responses can be reactivated in vivo without undergoing cell division. In contrast, treatment with CY before primary infection delayed the appearance of virus-induced NK activity and abrogated the generation of H-2 restricted, virus-specific CTL. Rechallenge of these CY-treated, NK-primed mice resulted in the rapid generation of a secondary NK response that was not followed by either a primary or secondary CTL response. This long-term block in CTL generation was not due to the establishment of a persistent PV infection. Memory CTL generation could be restored by secondarily coinfecting mice with PV and a second arenavirus such as Tacaribe virus (TV) or lymphocytic choriomeningitis virus (LCMV), or by injection of interleukin 2 (IL 2)-containing supernatants after rechallenge with PV. To demonstrate that IL 2 was the responsible lymphokine in these supernatants, highly purified IL 2 was added to in vitro cultures of spleen cells from CY-treated, PV-primed mice. In the presence of PV-infected syngeneic macrophages, addition of purified IL 2 resulted in a dose dependent restoration of H-2 restricted, anti-PV CTL activity. In addition, the CTL precursor frequency of CY-treated, PV-primed mice appeared to be markedly reduced compared to that in normal PV-primed mice. Thus, the long-lasting block in the ability to generate PV-specific memory CTL appears to be due to both a lack of helper T cell activity a significant reduction in the number of CTLp. Furthermore, these results suggest that priming the NK compartment is sufficient to prime for a memory CTL response, provided helper factors such as IL 2 are supplied.</p> / Doctor of Philosophy (PhD)

<p>Two inbred strains of Syrian hamster have been shown to display genetically determined differences in resistance to infections with the arenavirus, Pichinde virus (PV). After intraperitoneal injection, the virus grows to higher titres in the spleens of the susceptible strain, MHA, than in the spleens of the resistant strain, LSH. Preliminary studies examining the basis of susceptibility demonstrated that resistance or susceptibility to the virus did not lie in an inherent difference in target cells to become infected, but suggested that there was a quantitative difference in target cells between the two strains of hamster. The following experiments were conducted in attempts to verify the hypothesis that MHA hamsters are susceptible to infection with Pichinde virus because they possess larger numbers of a splenic lymphocyte that serves as a target cell for virus replication and that also functions as an effector cell of nonspecific cytotoxicity.</p> <p>The spleens and thymi of the high NK strain were found to display greater cellularity than those of the low NK strain. Additionally, thymocytes from MHA hamsters were found to proliferate to a greater extent than those of LSH hamsters in response to ConA-induced conditioned medium or purified interleukin 2 plus mitogen. As well, splenocytes from MHA hamsters showed high levels of lymphokine-activated killer cell (LAK) activity after culture in conditioned medium or interleukin 2. In both the thymus and the spleen, this difference in responsiveness was due to increased numbers of precursor cells responding to lymphokines in MHA organs. When lymphokine production was assessed, it was found that cells from the the high responder, MHA, synthesized less interleukin 2 than cells from LSH hamsters. Interleukin 1 production was equal in the two strains. These results led to the hypothesis that the susceptible hamsters contain immature lymphocytes, possibly because of the reduced interleukin 2 production. Increased numbers of relatively immature cells could account for the increased cellularity of lymphoid organs in these animals, and in the spleen, these cells may be responsible for increased NK activity, increased numbers of LAK precursors, and serve as target cells for PV replication.</p> <p>Splenic cytotoxic cells in the hamster were characterized. Endogenous NK cells, virus-induced NK cells and LAK were all plastic nonadherent cells, as were the precursors for LAK. All populations expressed an antigen homologous to the murine Thy 1.2. Endogenous NK cells and virus-induced NK cells were similar in the expression of an asialo GMl homologue; both were reduced by 50% by treatment with this antiserum plus complement. LAK precursors and LAK effectors were negative for this marker, suggesting that LAK arise from the asialo GM1 negative component of endogenous NK activity. Treatment of hamsters with anti-asialo GM1 serum also reduced splenic NK activity in normal and virus-infected hamsters by 50%.</p> <p>Cells infected with virus were characterized using the same criteria. A preferential infection of nonadherent cells was not evident before day 3 of infection, although MHA spleens already contained twice as much virus as LSH spleens. Both asialo GM1 negative and positive cells were infected. Culture of infected splenocytes in interleukin 2 induced high LAK, but failed to select out a population enriched for infectious centres compared to culture in medium alone, where no cytotoxic activity was evident. However, susceptible MHA hamsters with reduced NK activity after treatment with anti-asialo GM1 serum did display less virus at day 1 after infection with PV, but by day 3, virus loads were at control levels. Treatment with anti-asialo GM1 serum had no effect on the mortality of either strain of hamster. Treatment with purified interleukin 2 slowed mortality of MHA hamsters, although it did not do so by reducing viral replication. In total, t~ese data indicate that cells expressing an antigen detected by anti-asialo GMl serum that mediate some ~K ~ctivity can serve as target cells for PV replication, but these cells alone are not responsible. for the susceptibility of MHA hamsters • • Cytotoxic activity and infected centres could be dissociated, also suggesting that the jnitial hypothesis was incorrect. It could be that some other cell is the relevant target, that MHA hamsters may b~ susceptible because of increased number~ of all splenocyt~s, or that some .factor other than the presence or absence of a target cell accounts for their susceptibility.</p> / Doctor of Philosophy (PhD)

<p>Food restriction has profound effects on various endocrine axes and on amine metabolism. In the present study, the effect of reduced food availability on pineal and serum indole was determined in adult male Wistar rats. Under a lighting regimen of 14 h light and 10 h dark, 3 weeks of 50% food restriction led to a reduction in 24 h mean serum tryptophan and serum serotonin levels but an increase in serum melatonin levels. The duration of the night-time melatonin rise was increased secondary to an earlier rise of both pineal and serum melatonin. Such changes in circulating melatonin may account for the gonadal regression observed in underfed animals. This pineal-gonadal interaction was further investigated after animals were subjected to shortened photoperiod or after pinealectomy. Shortened photoperiod failed to influence either the serum melatonin profile or the undernutrition-related gonadal regression. Pinealectomy, however, was able to reverse though incompletely the gonadal regression in underfed animals. When the pineal responsiveness to beta-adrenergic stimulation was determined in food restricted animals, both the time course and the dose responses were altered. The changes in pineal and serum melatonin post-stimulation, however, were atypical of either a sub- or supersensitive pineal gland.</p> <p>Based on the present study, food availability proves to be another factor that can influence pineal activity. Its effect on the pineal, however, depends on the duration of food restriction and the environmental light/dark cycle.</p> / Doctor of Philosophy (PhD)

<p>Disabled children are increasingly being integrated into the regular school environment. Poor attitudes of able-bodied peers are a major obstacle to the social success of this process. Our knowledge about the determinants of attitudes and methods of improving attitudes has been hampered by the poor quality of available attitude measures. This thesis describes a new measure to overcome this problem.</p> <p>The Chedoke-McMaster Attitudes Toward Children with Handicaps (CATCH) scale is a 36-item self-report measure of children's expressed attitude along three dimension; affective response, behavioral intent, and cognitive understanding. CATCH is intended for children in grades four to eight. Over 800 children have been involved in the testing of the first and then the revised draft of CATCH.</p> <p>Children had no difficulty completing the scale. The items related to their everyday experiences. Descriptive aspects of CATCH were good. Total scores varied from 53 to 143 (possible interval of 0 to 144). The total sample mean was 99.1 with a standard deviation of 16.1. The measure was reliable with a coefficient alpha of .90 for the total score.</p> <p>Factor analysis of CATCH revealed three factors accounting for 83% of the variance. Factor one and factor three were a mixture of affective and behavioral items and accounted for 62% and 6% of the variance, respectively. Factor two contained primarily cognitive items and accounted for 15% of the variance.</p> <p>The construct validity of CATCH was established by testing its ability to discriminate groups based on previously hypothesized differences. Girls had significantly more positive attitudes than boys (104.0 versus 94.5; p=.001). Children who had a friend who was disabled; or who had contact with a disabled child in the last week had significantly better attitudes than children without disabled friends or contact. The mean CATCH score for friend was 106.2 versus 95.6 for no friend. For contact the mean score was 108.1 versus 97.0. Children who volunteered for a buddy program or who had previously taken part in a buddy program with a disabled child had significantly higher CATCH scores.</p> <p>CATCH was used as the primary outcome measure in a randomized controlled trial of a buddy program between able-bodied and disabled children. A criterion score improvement occurred significantly more frequently in buddies than controls (43% versus 18%) p=.05). This intervention also appeared to have a significant effect on parental attitudes with buddy parents having a significant improvement compared to control parents.</p> <p>These results demonstrate that CATCH is both a reliable and valid measure of children's expressed attitudes. While the limitations of this measure are recognized. CATCH will be extremely valuable in the study of attitude determinants and in the evaluation of interventions to improve attitudes.</p> / Doctor of Philosophy (PhD)

<p>Herpes Simplex Virus DNA displays a number of unusual features which have been the subject of intense scrutiny in a number of laboratories. The genome is composed of two segments, each of which is flanked by inverted repeats. These segments invert freely with respect to each other generating equisolar quantities of four different isomers. This phenomenon, called segment inversion, was reputed to be the result of a site specific recombination mechanism operating on the terminal repeat, the 'a' sequence, which is part of the inverted repeats flanking each segment. The 'a' sequence was also implicated as the cleavage/packaging signal utilized by the virus to process viral DNA concatemers. The underlying mechanism of this process was believed to be a double strand break at a specific site between two 'a' sequences. The models of HSV maturation were deficient, however, in explaining several phenomena, namely the tendency of the 'a' sequence to accumulate tandem iterations of itself, the asymmetric distribution of these tandem iterations to one end of the genome, but not to the other, and the ability of defective genomes, which do not have tandemly iterated 'a' sequences, at least initially, to be efficiently packaged. I have shown that the "a" sequence actually contains two signals for cleavage/packaging, not one, that the cleavage occurs it specific distances from these signals, not in specific sequences, and that the cleavage mechanism results in a duplication of the cleavage signal and flanking DNA. Furthermore, I have determined that the 'a' sequence is not a target for site specific recombination, and that there is better evidence to support the idea that segment inversion is accomplished by a number of related, but independent mechanisms, including generalized recombination.</p> / Doctor of Philosophy (PhD)

<p>Cytotoxic T-lymphocytes (CTL) are important in resistance to HIV-1 infection, and modulate disease progression through various effector functions. Conventional CTL recognizing antigenic peptides in the context of major histocompatibility (MHC) class I molecules eliminate virally-infected cells, and have been shown to modulate the course of disease progression. Other functions of CD8⁺ lymphocytes include non-lytic suppresion of viral transcription and replication, and induction of programmed cell death in activated CD4⁺ cells. In the work presented here, the interactions of CD8⁺ lymphocytes with activated CD4⁺ cells were examined. Lysis of activated CD4⁺ cells was unique for CTL derived from HIV-1 infeced persons, and was not restricted by MHC molecules. The interaction was dependent on target cell activation, correlated with cell proliferation, and was independent of viral replication. Furthermore, the lytic process was biochemically consistent with apoptosis, and resulted in nuclear fragmentation in the target cells. In vitro susceptibility to infection by primary HIV-1 isolates was assessed in a cohort of couples consisting of an infected and an uninfected, but exposed, partner. These findings were correlated with the genetic integrity of the HIV-1 co-receptors, and with the presence of CTL specifically directed against the partner's primary isolate. Individuals with partial or complete resistance to infection were identified, as well as some individuals that had CTL recognizing the partner's viral isolate. The latter results suggested that cellular immune responses comprise a major component of resistance to HIV infection, and that they influence resistance even in individuals lacking the main HIV-1 co-receptor. The later studies were extended by examining the in vitro infectivity of CD4⁺ cells from a group of highly exposed, but persistently uninfected, Kenyan sex workers. No barrier to infection of CD4⁺ lymphocytes with primary Kenyan HIV isolates was identified, however, CD8⁺ cells from these subjects were able to efficiently limit viral replication in autologous CD4⁺ cells. In summary, the broadly different CTL effector functions described in these studies illustrate the importance of the cellular immune system in resistance to HIV infection, and in modulating the disease pathogenesis.</p> / Doctor of Philosophy (PhD)

<p>Benzodiazepines (BZ) are a class of drugs that are extensively used for their anxiolytic, anticonvulsant and sedative properties. The therapeutic actions of these drugs may be mediated through the central-type BZ receptors that are linked to the GABA receptor complex (BZ/GABA). When administered in large doses, the pineal hormone, melatonin, can interact with BZ receptors. The pharmacological actions of melatonin, similar to those listed above for BZs, appear to be mediated primarily by BZ/GABA receptors, although other BZ receptors may be involved. Recently, pharmacological doses of melatonin were found to decrease apomorphine-induced rotations in 6-hydroxydopamine lesioned rats. The main objective of this thesis was to determine the mechanism(s) underlying the antidopaminergic effect of melatonin as well as BZs using the 6-hydroxydopamine lesion model of dopaminergic supersensitivity. It was hypothesized that the antidopaminergic action of BZs and melatonin was due to the ability of these drugs to either enhance GABAergic activity (since GABA can suppress striatal dopaminergic activity) or block cyclic AMP (cAMP) production since dopamine enhances cAMP formation in the striatum. The major findings may be summarized as follows: 1) Clonazepam, a central-type BZ agonist, the melatonin blocked apomorphine-induced turning in lesioned animals; 2) intrastriatal injection of the GABA antagonist, bicuculline, caused a significant reduction in the antidopaminergic effect of clonazepam and melatonin; 3) the peripheral-type BZ antagonist, PK 11195, attenuated the antidopaminergic effect of these drugs but with much less potency than bicuculline; 4) the combination of bicuculline and PK 11195, injected directly into the striatum, completely blocked the antidopaminergic action of clonazepam or melatonin; 5) PK 11195 also blocked BZ-induced inhibition of cAMP production which is involved in striatal dopaminergic function. Therefore, in addition to a GAVAergic mechanism, inhibition of a cAMP pathway may be a secondary mechanism in the antidopaminergic action of clonazepam and melatonin. In studying the effects of BZs on the cAMP pathway, a significant increase in the inhibitory effect of diazepam on adenylate cyclase (AC) activity was observed in striatal membranes from lesioned animals. Further studies indicated that this sensitization to the inhibitory effect of diazepam on AC activity may involve upregulation of BZ receptors as well as enhanced functional coupling of these receptors to inhibitory G proteins. Taken together these findings indicate that the antidopaminergic effect of clonazepam and melatonin in lesioned animals involved at least two distinct mechanisms: 1) a predominant GABAergic action and 2) possibly the suppression of cAMP production in the striatum.</p> / Doctor of Philosophy (PhD)

<p>The first objective of this thesis was to investigate the targets of cyclic GMP (cGMP) action in platelets. Proteins that bind cGMP were first detected by photoaffinity labelling with [³²P]cGMP and subsequently identified by molecular, pharmacological and immunological criteria. Since cGMP was already known to exert major effects in platelets through the cGMP-inhibited phosphodiesterase family (PDE3) (Maurice and Haslam, 1990a), an additional objective was to explore the molecular basis of the unique properties of this enzyme by cloning and mutagenesis studies. A photolabelling technique using [³²P]cGMP was modified to permit the rapid detection of cGMP-binding proteins in crude platelet extracts. Five labelled proteins (110, 80, 55, 49 and 38 kDa) were detected in platelet supernatant and four (80, 65, 49 and 38 kDa) in platelet membranes. The sensitivity of photolabelling to PDE3 inhibitors and specific immunoprecipitation established that the 110 kDa photolabelled species was a product of the PDE3 gene family. In addition, the 80 kDa species was identified as cGMP-dependent protein kinase (PKG) by similar methods. Interestingly, cyclic AMP (cAMP) greatly enhanced the labelling of the 80 kDa protein, suggesting the existence of a novel co-operative interaction between cAMP and cGMP. This study also detected a previously unknown 65 kDa cGMP-binding protein in platelet membranes. Since both cAMP and cGMP inhibited labelling of this protein, it may represent a novel target for both cyclic nucleotides in platelets, possibly a subunit of a cyclic nucleotide-gated (CNG) ion channel. The inhibitory effects of variouus compounds on the photolabelling of PDE3 were quantitated by [³²P]cGMP. Thus, concentration-dependent inhibition of photolabelling of PDE3 was observed with trequinsin (IC₅₀ = 13 ± 2 nM), lixazinone (IC₅₀ = 22 ± 4 nM), milrinone (IC₅₀ = 56 ± 12 nM), cilostamide (IC₅₀ = 70 ± 9 nM), siguazodan (IC₅₀ 117 ± 29 nM) and 3-isobutyl 1-methylxanthine (IBMX) (IC₅₀ = 3950 ± 22 nM). The effects of these phosphodiesterase inhibitors on iloproststimulated cAMP accumulation in intact platelets were also investigated. Discrepancies between the abilities of these compounds to inhibit photolabelling of PDE3 and to increase platelet cAMP accumulation are probably related to differences in the rates of entry of the individual inhibitors into the intact platelet. The general applicability of this photolabelling technique and its value in the detection of novel cGMP-binding proteins in crude cell extracts was demonstrated in rat tissues. Distinctive photolabelling patterns were observed in different rat tissues and the 110 kDa protein found in human platelets was replaced by a 115 kDa species in rat platelets. To clarify the molecular mechanisms by which cGMP and inhibitory drugs modulate PDE3 activity, an attempt was made to define the roles of different PDE3 domains in the action of the enzyme. To that end, the C-terminal half of platelet PDE3 was cloned, identified as a product of the PDE3A gene and expressed as an active enzyme in E.coli. Further deletion mutants were generated by removing either an additional 100 amino acids from the N-terminus or the 44 amino acid insert, characteristic of members of the PDE3 family, from the catalytic domain. Site-directed mutagenesis of the 44 amino acid insert was also conducted to explore the function of this region. Kinetic analysis of these mutant enzymes demonstrated that the deletion of N-terminal sequences from PDE3 was accompanied by progressively lower Km values and by an increased Vmax for cGMP relative to that for cAMP. Thus, N-terminal sequences exert modulatory effects on cGMP hydrolysis. Deletion of the 44 amino acid insert abolished enzyme activity, as did site-directed mutagenesis of putative β-turns located at the N- and C-termini of the insert. Mutation of a cluster of negatively charged residues in the insert did not have major effects on the hydrolysis of cAMP or cGMP. The results suggest that this insert is required to preserve an effective catalytic domain structure in PDE3. In conclusion, the major targets of cGMP in platelets were shown to include not only PKG, but also PDE3A and an unidentified membrane protein, possibly a CNG ion channel. This thesis has also identified some of the functionally and structurally important domains within PDE3A.</p> / Doctor of Philosophy (PhD)