Desenvolvimento e caracterização de sistemas líquido-cristalinos para aplicação tópica de metotrexato: estudos de liberação, retenção e permeação in vitro

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Análise de estabilidade de taludes de aterro de resíduos urbanos utilizando parâmetros geotécnicos de propostas bibliográficas e correlações com Nspt

Pós-graduação em Geociências e Meio Ambiente - IGCE

Farmacocinética pré-clínica e avaliação toxicológica do novo composto α2-adrenérgico 3-(2-cloro-6-fluorobenzil)-imidazolidina-2,4-diona - PT-31 GIRSUPAN

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Calculation of parameter ranges for robust gain tuning of power system controllers

This work proposes a computational tool to assist power system engineers in the field tuning of power system stabilizers (PSSs) and Automatic Voltage Regulators (AVRs). The outcome of this tool is a range of gain values for theses controllers within which there is a theoretical guarantee of stability for the closed-loop system. This range is given as a set of limit values for the static gains of the controllers of interest, in such a way that the engineer responsible for the field tuning of PSSs and/or AVRs can be confident with respect to system stability when adjusting the corresponding static gains within this range. This feature of the proposed tool is highly desirable from a practical viewpoint, since the PSS and AVR commissioning stage always involve some readjustment of the controller gains to account for the differences between the nominal model and the actual behavior of the system. By capturing these differences as uncertainties in the model, this computational tool is able to guarantee stability for the whole uncertain model using an approach based on linear matrix inequalities. It is also important to remark that the tool proposed in this paper can also be applied to other types of parameters of either PSSs or Power Oscillation Dampers, as well as other types of controllers (such as speed governors, for example). To show its effectiveness, applications of the proposed tool to two benchmarks for small signal stability studies are presented at the end of this paper.

[Lys40(Ahx-DTPA-111In)NH2]exendin-4, a very promising ligand for glucagon-like peptide-1 (GLP-1) receptor targeting

High levels of glucagon-like peptide-1 (GLP-1) receptor expression in human insulinomas and gastrinomas provide an attractive target for imaging, therapy, and intraoperative tumor localization, using receptor-avid radioligands. The goal of this study was to establish a tumor model for GLP-1 receptor targeting and to use a newly designed exendin-4-DTPA (DTPA is diethylenetriaminepentaacetic acid) conjugate for GLP-1 receptor targeting. METHODS: Exendin-4 was modified C-terminally with Lys(40)-NH(2), whereby the lysine side chain was conjugated with Ahx-DTPA (Ahx is aminohexanoic acid). The GLP-1 receptor affinity (50% inhibitory concentration [IC(50)] value) of [Lys(40)(Ahx-DTPA)NH(2)]exendin-4 as well as the GLP-1 receptor density in tumors and different organs of Rip1Tag2 mice were determined. Rip1Tag2 mice are transgenic mice that develop insulinomas in a well-defined multistage tumorigenesis pathway. This animal model was used for biodistribution studies, pinhole SPECT/MRI, and SPECT/CT. Peptide stability, internalization, and efflux studies were performed in cultured beta-tumor cells established from tumors of Rip1Tag2 mice. RESULTS: The GLP-1 receptor affinity of [Lys(40)(Ahx-DTPA)NH(2)]exendin-4 was found to be 2.1 +/- 1.1 nmol/L (mean +/- SEM). Because the GLP-1 receptor density in tumors of Rip1Tag2 mice was very high, a remarkably high tumor uptake of 287 +/- 62 %IA/g (% injected activity per gram tissue) was found 4 h after injection. This resulted in excellent tumor visualization by pinhole SPECT/MRI and SPECT/CT. In accordance with in vitro data, [Lys(40)(Ahx-DTPA-(111)In)NH(2)]exendin-4 uptake in Rip1Tag2 mice was also found in nonneoplastic tissues such as pancreas and lung. However, lung and pancreas uptake was distinctly lower compared with that of tumors, resulting in a tumor-to-pancreas ratio of 13.6 and in a tumor-to-lung ratio of 4.4 at 4 h after injection. Furthermore, in vitro studies in cultured beta-tumor cells demonstrated a specific internalization of [Lys(40)(Ahx-DTPA-(111)In)NH(2)]exendin-4, whereas peptide stability studies indicated a high metabolic stability of the radiopeptide in beta-tumor cells and human blood serum. CONCLUSION: The high density of GLP-1 receptors in insulinomas as well as the high specific uptake of [Lys(40)(Ahx-DTPA-(111)In)NH(2)]exendin-4 in the tumor of Rip1Tag2 mice indicate that targeting of GLP-1 receptors in insulinomas may become a useful imaging method to localize insulinomas in patients, either preoperatively or intraoperatively. In addition, Rip1Tag2 transgenic mice represent a suitable animal tumor model for GLP-1 receptor targeting.

Regulation of mRNA splicing in Moloney murine sarcoma virus mutants

Cells infected with the conditionally defective MuSVts110 mutant of Moloney murine sarcoma virus are transformed at 33$\sp\circ$C but appear morphologically normal at 39$\sp\circ$C. The molecular basis for this phenotype is as follows: MuSVts110 contains a 1487 nucleotide central deletion that has truncated the 3$\sp\prime$ end to the gag gene and the 5$\sp\prime$ end of the mos gene. The resulting gag-mos junction is out-of-frame and the v-mos protein is not expressed. At 33$\sp\circ$C or lower, a splicing event is activated such that a 431 base intron is removed to realign the gag and mos gene in-frame, allowing the expression of a transforming protein P85$\sp{gag-mos}$. Temperature-dependent splicing appeared to be an intrinsic property of MuSVts110 transcripts and not a general feature of pre-mRNA splicing in 6m2 cells since splicing activity of a heterologous transcript in the same cells did not vary with temperature. The possibility that the splice event was not temperature-sensitive, but that the accumulation of spliced transcript at the lower growth temperatures was due to its selective thermolability was ruled out as stability studies revealed that the relative turnover rates of the unspliced and spliced MuSVts110 transcripts were not affected by temperature.^ The consensus sequences containing the splice sites activated in the MuSVts110 mutant (5$\sp\prime$ gag and 3$\sp\prime$ mos) are present, but not utilized, in wild-type MuSV-124. To test the hypothesis that it was the reduction of the 1919 base intervening sequence in MuSV-124 to 431 bases in MuSVts110 which activated splicing, the identical 1487 base deletion was introduced into cloned wild-type MuSV-124 DNA to create the MuSVts110 equivalent, ts32.^ To examine conditions permissive for splicing, we assayed splice site activation in a series of MuSV-124 "intron-modification" mutants. Data suggest that splicing in wild-type MuSV-124 may be blocked due to the lack of a proximal branchpoint sequence, but can be activated by those intron mutations which reposition a branch site closer to the 3$\sp\prime$ splice site. (Abstract shortened with permission of author.) ^

Caracterización multivariada de los temporales para su aplicación en el dimensionamiento del manto principal

En esta tesis se presenta una metodología para la caracterización del oleaje, dentro del marco de las nuevas Recomendaciones para Obras Marítimas (ROM 0.0.-00 y ROM 1.0-09), por ser una de las principales acciones que afectan a la estabilidad de las estructuras marítimas. Debido al carácter aleatorio intrínsecamente multivariado de la acción considerada, las tormentas, su caracterización paramétrica se realiza en términos de funciones cópula uniparamétricas. Las variables consideradas son altura de ola significante del pico de la tormenta, el periodo medio asociado y la magnitud, o número de olas, de todo el ciclo de solicitación. Para establecer un patrón teórico de evolución de la tormenta que permita extrapolar las muestras fuera de la región con datos se analizan los modelos teóricos existentes, comprobándose que no reproducen adecuadamente las tormentas constituidas por estados de mar con un peso importante de oleaje swell. Para evitar esta limitación se proponen cuatro modelos teóricos de evolución de tormentas con distintas formas geométricas. El análisis de los modelos existentes y los propuestos pone de relieve que el Modelo Magnitud Equivalente de Tormenta (EMS= Equivalent Magnitude Storm) con la forma triangular es el que mejor adapta las tormentas constituidas por estados de mar típicos del viento. Para tormentas con un mayor grado de desarrollo, el modelo teórico de tormenta EMS con la forma trapezoidal es el adecuado. De las aproximaciones propuestas para establecer el periodo medio de los sucesivos estados de mar del ciclo de solicitación. la propuesta por Martín Soldevilla et al., (2009) es la más versátil y , en general , mejor reproduce la evolución de todo tipo de tormentas. La caracterización de las tormentas se complementa con la altura de ola máxima. Debido a la mayor disponibilidad y longitud temporal de los datos sintéticos frente a las registros, la práctica totalidad de los análisis de extremos se realizan con tormentas sintéticas en las que la distribución de olas individuales es desconocida. Para evitar esta limitación se utilizan modelos teóricos de distribución de olas acordes a las características de cada uno de los estados de mar que conforman la tormenta sintética. Para establecer dichas características se utiliza la curtosis y en función de su valor la altura de ola máxima se determina asumiendo una determinada distribución de olas. Para estados de mar lineales la distribución de olas individuales de Rayleigh es la considerada. Para condiciones no lineales de gran ancho de banda el modelo de distribución de olas propuesto por Dawson, (2004) es el utilizado y si es de banda estrecha las predicciones de (Boccotti, (1989), Boccotti et al., (2013)) se compara con las resultantes del modelo de Dawson. La caracterización de la evolución de las tormentas en términos multivariados es aplicada al estudio de la progresión del daño del manto principal de diques en talud, y al rebase de las olas. Ambos aspectos cubren el segundo objetivo de la tesis en el que se propone una nueva formulación para el dimensionamiento de mantos constituidos por bloques cúbicos de hormigón. Para el desarrollo de esta nueva formulación se han utilizado los resultados recogidos en los estudios de estabilidad del manto principal de diques talud realizados en modelo físico a escala reducida en el Centro de Estudios de Puertos y Costas (CEDEX) desde la década de los 80 empleando, en su mayoría, bloques paralelepípedos cúbicos de hormigón. Por este motivo y porque los últimos diques construidos en la costa Española utilizan este tipo de pieza, es por lo que la formulación planteada se centra en este tipo de pieza. Después de un primer análisis de las fórmulas de cálculo y de evolución existentes, se llega a la conclusión de que es necesario realizar un esfuerzo de investigación en este campo, así como ensayos en laboratorio y recogida de datos in-situ con base a desarrollar fórmulas de evolución de daño para mantos constituidos por piezas diferentes a la escollera, que tenga en cuenta las principales variables que condiciona su estabilidad. En esta parte de la tesis se propone un método de análisis de evolución de daño, que incluye el criterio de inicio de avería, adecuada para diques en talud constituidos por bloque cúbicos de hormigón y que considera la incidencia oblicua, el daño acumulado y el rebase. This thesis proposes a methodology to estimate sea waves, one of the main actions affecting the maritime structures stability, complying with (ROM 0.0.-00 & ROM 1.0-09.Due to the multivariate behavior of sea storms, the characterization of the structures of sea storms is done using copula function. The analyzed variables are the significant height wave, mean period and magnitude or number of waves during the storm history. The storm evolution in terms of the significant height wave and the mean period is also studied in other to analyze the progressive failure modes. The existing models of evolution are studied, verifying that these approximations do not adjust accurately for developed waves. To overcome this disadvantage, four evolution models are proposed, with some geometrical shapes associated to fit any development degree. The proposed Equivalent Magnitude Storm model, EMS, generally obtains the best results for any kind of storm (predominant sea, swell or both). The triangle is recommended for typical sea storms whereas the trapezoid shape is much more appropriate for more developed storm conditions.The Martín Soldevilla et al., (2009) approach to estimate the mean period is better than others approaches used.The storm characterization is completed with the maximum wave height of the whole storm history. Due to synthetic historical waves databases are more accessible and longer than recorded database, the extreme analyses are done with synthetic data. For this reason the individual waves’ distribution is not known. For that limitation to be avoided, and depending on the characteristics of every sea states, one theoretical model of waves is choose and used. The kurtosis parameter is used to distinguish between linear and nonlinear sea states. The Rayleigh model is used for the linear sea states. For the nonlinear sea states, Dawson, (2004) approach is used for non-narrow bandwidth storms, comparing the results with the Boccotti, (1989), Boccotti et al., (2013) approach, with is used for narrow bandwidth storms. The multivariate and storm evolution characterization is used to analyze of stone armour damage progression and wave overtopping discharge. Both aspects are included in the second part of the thesis, with a new formula is proposed to design cubes armour layer. The results the stability studies of armour layer, done in the Centre for Harbours and Coastal Studies (CEDEX) laboratory are used for defining a new stability formula. For this reason and because the last biggest breakwater built in Spain using the cube, the damage progression is analyze for this kind of concrete block. Before to analyze the existing formulae, it is concluded that it is necessary more investigation, more tests in laboratory and data gathering in situ to define damage evolution formulae to armour of other kind of pieces and that takes to account the principal variables. This thesis proposed a method to calculate the damage progression including oblique waves, accumulated damage, and overtopping effect. The method also takes account the beginning of the movement of the blocks.

Stabilization of tetanus and diphtheria toxoids against moisture-induced aggregation.

The progress toward single-dose vaccines has been limited by the poor solid-state stability of vaccine antigens within controlled-release polymers, such as poly(lactide-co-glycolide). For example, herein we report that lyophilized tetanus toxoid aggregates during incubation at 37 degrees C and elevated humidity--i.e., conditions relevant to its release from such systems. The mechanism and extent of this aggregation are dependent on the moisture level in the solid protein, with maximum aggregation observed at intermediate moisture contents. The main aggregation pathway is consistent with formaldehyde-mediated cross-linking, where reactive electrophiles created and stored in the vaccine upon formalinization (exposure to formaldehyde during vaccine preparation) react with nucleophiles of a second vaccine molecule to form intermolecular cross-links. This process is inhibited by the following: (i) succinylating the vaccine to block reactive amino groups; (ii) treating the vaccine with sodium cyanoborohydride, which presumably reduces Schiff bases and some other electrophiles created upon formalinization; and (iii) addition of low-molecular-weight excipients, particularly sorbitol. The moisture-induced aggregation of another formalinized vaccine, diphtheria toxoid, is also retarded by succinylation, suggesting the generality of this mechanism for formalinized vaccines. Hence, mechanistic stability studies of the type described herein may be important for the development of effective single-dose vaccines.

Modeling of Distributed Generation under Next Generation Interconnection Requirements

Thesis (Master's)--University of Washington, 2016-06

Solid dispersions: formulation, characterisation, permeability and genomic evaluation

Poor water solubility is characterised by low dissolution rate and consequently reduced bioavailability. Formulation of solid dispersion of the drug has attracted considerable interest as a means of improving dissolution process of a range of poorly water soluble drugs. This current study investigates the formulation of solid dispersion for a range of poorly water soluble drugs with varying physicochemical properties including paracetamol, sulphamethoxazole, phenacetin, indomethacin, chloramphenicol, phenylbutazone and succinylsulphathiazole. Solid dispersions were prepared using various drugs to polymer ratios. PEG 8000 was selected as a carrier in the solid dispersions. The study revealed that inclusion of drug within the polymeric matrix, ratio of drug to polymer and physicochemical properties of the drug molecules enhance the dissolution rate. Characterisations of the solid dispersions were performed using DSC, FTIR and SEM. These studies revealed that all seven drugs were present in the amorphous form within the solid dispersions and there was a lack of interaction between the PEG 8000 and drug. Stability studies for solid dispersions showed that all seven drugs studied were unstable at accelerated conditions (40°C±2°C/75%RH±5%RH) whereas, they were found to be stable for 12 months at room conditions. Permeability of indomethacin, phenacetin, phenylbutazone and paracetamol were higher for solid dispersions as compared to drug alone across Caco-2 cell monolayers. From the cell uptake studies it was shown that PEG 8000 enhanced rhodamine123 uptake which suggested that PEG 8000 may increase the permeability of these drugs in solid dispersions. Gene expression profiles analyzing the expression changes in the ABC and solute carrier transporter during permeability studies.ABCA10, ABCB4, ABCC12, SLC12A6, MCT13, SLC22A12 and SLC6A6 gene expression were increased by indomethacin alone whereas solid dispersion of indomethacin resulted in a slight increase in expression. ABCC12 and SAMC gene expression was increased in case of paracetamol alone but slightly increased when exposed to solid dispersion of paracetamol.

Evaluation of DNA enzymes targeted against the RNA component of human telomerase

Glioblastoma Multiforme (GBM) is a highly malignant form of brain cancer for which there is currently no effective cure. Consequently, developing new therapies and elucidating effective targets is crucial for this fatal disease. In recent years, DNA enzymes, deoxyribonucleic acid molecules with enzymatic activity, have emerged. In the same manner as ribozymes, DNA enzymes are able to effect cleavage of RNA in a sequence-specific manner, and operate with catalytic efficiency. In this study, two DNA enzymes were designed to target the template region of human telomerase RNA (hTR), utilising the 10-23 and 8-17 catalytic motifs elucidated by Santoro and Joyce (1997). Telomerase is an RNA-dependent DNA polymerase, which stabilises telomere lengths by adding hexameric repeats (TTAGGG in humans) to chromosome termini, thus preventing the telomere shortening that usually occurs during mitotic cell division. Telomerase activity, whilst absent in normal somatic tissues, is present in almost 90% of all tumours. Thus, there is speculation that telomerase may be the much sought universal target for therapeutic intervention in cancer. In vitro cleavage assays showed both DNA enzymes to be catalytically competent. Unmodified phosphodiester (PO) backbone DNA enzymes were rapidly degraded in the presence of serum, with a half-life of 10 minutes. The common approach of introducing phosphorothioate (PS) linkages was used in an effort to overcome this instability. As a result of concurrent activity and stability studies on the DNA enzymes with various numbers of PS linkages, the DNA enzymes with a PO core and PS arms were chosen for use in further cell work. The cleavage activity of both was shown to be specific and affected by temperature, pH, MgCI2 concentration and enzyme concentration. Both DNA enzyme motifs reduced telomerase activity in cell lysates, as assessed by the telomerase repeat amplification protocol (TRAP) with an IC50 of 100nM. DNA enzymes being polyanionic molecules do not readily cross biological barriers. Cellular association of naked DNA enzyme was inefficient at less than 2%. Cellular delivery of the DNA enzymes was effectively improved using commercial cationic lipid formulations. However, the lipid-mediated delivery of DNA enzymes to U87-MG cells over a 4-hour period did not significantly inhibit cell proliferation compared to controls. This is possibly due to an expected lag period between the inhibition of telomere maintenance and cell death. Therefore, biodegradable polymer microspheres were investigated as a potential delivery option for prolonged and sustained delivery. In vitro release profiles showed that after an initial burst, sustained release of DNA enzymes was observed over 35 days. Finally, the efficacy and specificity of the DNA enzymes were demonstrated in a luciferase based reporter assay. Specific inhibition of luciferase expression was displayed at 10nM. Thus DNA enzymes have potential against endogenous cellular targets.

High-performance liquid chromatography and pharmacokinetics of some antibiotics

The principles of High Performance Liquid Chromatography (HPLC) and pharmacokinetics were applied to the use of several clinically-important drugs at the East Birmingham Hospital. Amongst these was gentamicin, which was investigated over a two-year period by a multi-disciplinary team. It was found that there was considerable intra- and inter-patient variation that had not previously been reported and the causes and consequences of such variation were considered. A detailed evaluation of available pharmacokinetic techniques was undertaken and 1- and 2-compartment models were optimised with regard to sampling procedures, analytical error and model-error. The implications for control of therapy are discussed and an improved sampling regime is proposed for routine usage. Similar techniques were applied to trimethoprim, assayed by HPLC, in patients with normal renal function and investigations were also commenced into the penetration of drug into peritoneal dialysate. Novel assay techniques were also developed for a range of drugs including 4-aminopyridine, chloramphenicol, metronidazole and a series of penicillins and cephalosporins. Stability studies on cysteamine, reaction-rate studies on creatinine-picrate and structure-activity relationships in HPLC of aminopyridines are also reported.

Novel biological roles for pyrimidines

The development of classical and lipophilic inhibitors of dihydrofolate reductase (DHFR) as antitumour agents is reviewed and the advantages and problems associated with each class are discussed. The antitumour activity, pharmacokinetics and metabolism of m-azido-pyrimethamine (MZP), a novel lipophilic inhibitor, are considered and compared with metoprine, the prototype lipophilic antifolate. Evidence for a folate-independent target for lipophilic DHFR inhibitors is presented. Synthetic studies centred on three principal objectives. Firstly a series of structural analogues of MZP were prepared encompassing alkoxy, chloro and alkylamino substituents and evaluated, as the ethanesulphonate salts, for activity against mammalian DHFR. Inhibitory constant (KI) determinations were conducted by a Zone B analysis, the corresponding 4'-azido isomer of MZP proving more potent than the parent compound. Secondly, to facilitate metabolism and stability studies on MZP, a range of possible reference compounds were synthesised and characterised. Finally, a series of diaminopyrimidine derivatives were synthesised embracing structural features incompatible with DHFR inhibitory activity, in order that such compounds may serve as biochemical probes for the unidentified folate-independent target for lipophilic diaminopyrimidines discussed previously. Inactivity against DHFR was achieved via introduction of an ionic or basic group into a normally hydrophobic region of the molecule and compounds were screened against a mammalian DHFR and thymidylate synthase to confirm the abolition of activity. Several derivatives surprisingly proved potent inhibitors of DHFR exhibiting KI values comparable to that of methotrexate. Analogues were screened for antitumour activity in vitro and in vivo against murine leukaemia cell lines in order to identify potential lead compounds. Several derivatives virtually inactive against DHFR exhibited a disparate cytotoxicity and further biochemical studies are warranted. The nobreak hitherto unreported debenzylation of 2,4-diamino-5-(N-alkyl-benzylaminophenyl) pyrimidines was discovered during the course of the synthetic studies, treatment of these compounds with nitrous acid affording the corresponding benzotriazoles.

Physicochemical characteristics of chlorofluorohydrocarbon based inhalation aerosols

The aim of this work was to gain a better understanding of the physiochemical factors which affect the formulation of suspension inhalation aerosols. This has been attempted by applying the principles of colloid science to aerosol formulation. Both a drug system and a model colloid system have been used. The adsorption of six nonionic and cationic surfactants onto Spherisorb has been investigated. The results were analysed by calculating the area occupied by one adsorbed molecule at the surface and by comparing these values for each surfactant. The amount of each surfactant adsorbed was correlated with the number of sites on that surfactant molecule which could interact with the surface. The stability of suspensions, produced by both the model colloid Spherisorb, and by the drug isoprenaline sulphate, after adsorption of the surfactants, has been assessed by measuring settling times and rising times. The most stable suspensions were found to be those which had the greatest amounts of long chain fatty acid surfactant adsorbed on their surface. A comparison was made between the effective stabilising properties of Span 85 and oleic acid on various drug suspensions. It was found that Span 85 gave the most stable suspensions. Inhalation aerosol suspensions of isoprenaline sulphate were manufactured using the same surfactants used in the adsorption and suspension stability studies and were analysed by measuring the particle size distributions of the suspension and the emitted doses. The results were found to correlate with the adsorption and suspension stability studies and it was concluded that a deflocculated suspension was preferable to a flocculated suspension in inhalation aerosols provided that the drug density was less than the propellant density. The application of this work to preformulation studies was also discussed.

Formulation of immunomodulatory agents

Reversed-phase high-performance liquid chromatography procedures were developed for the analysis of pyrimidine-based drugs bropirimine and its derivatives (2-N-acetyl- and 2-N-propanoyl-) and for pyrimethamine and its 2/4- substituted derivatives (2, N-propanoyl and 2,4-N, N-dipropanoyl-) and its 6- substituted (methyl-, ethyl-, propyl- and isopropyl- carboxylates) analogues. Stability studies indicated that these derivatives were not sufficiently labile to act as potential prodrugs. Solubility-pH profiles were constructed from which the dissociation constants were calculated. The physicochemical properties of these compounds were studied and attempts were made to increase the poor aqueous solubility of bropirimine (35μg/mL) by prodrug synthesis, solvate formation (acetic acid, N, N-dimethylformamide and N-methylformamide) and the use of co-solvents and additives. The first two methods proved to be fruitless whereas the latter method resulted in an intravenous formulation incorporating 32mg/mL of bropirimine. An in-vitro method for the detection of precipitation was developed and the results suggested that by using low injection rates (< 0.24mL/min) and high mobile phase flow rates (> 500mL/hr) precipitation could be minimised. Differential scanning calorimetry showed that bropirimine debrominates in the presence of a number of additives commonly used in formulation work but the temperature at which this occurred were usually > 200oC. In-vitro work gave encouraging results for the possibility of rectal delivery of bropirimine but in-vivo work on rabbits showed considerable variations in the resulting plasma levels and pharmacokinetic parameters.