A proposed mechanism for p-aminoclonidine allergenicity based on its relative oxidative lability.

p-Aminoclonidine (apraclonidine) is a selective alpha 2 adrenergic agonist used to reduce intraocular pressure in the treatment of glaucoma. Use of apraclonidine is frequently associated with severe local allergic effects which warrant discontinuation of the drug in affected patients. We have assessed the oxidative lability of apraclonidine relative to a panel of adrenergic agonists and/or known allergens; amodiaquine, epinephrine, clonidine, and brimonidine. These compounds were compared by their electrochemical potentials as well as their oxidative lability in the presence of several oxidative enzyme systems (i.e., horseradish peroxidase, lactoperoxidase, myeloperoxidase, and diamine oxidase). The half-lives for enzymatic oxidation of these compounds were found to parallel the electrochemical oxidation potentials in the order: amodiaquine approximately epinephrine < apraclonidine << clonidine approximately brimonidine. The production of a reactive electrophilic intermediate of apraclonidine was demonstrated through the formation of two glutathione apraclonidine adducts from the horseradish peroxidase/H2O2-mediated oxidation of apraclonidine in the presence of glutathione. A mechanism for apraclonidine allergenicity in vivo is proposed wherein apraclonidine is bioactivated through oxidation to the bis-iminoquinone followed by protein conjugation to form an apraclonidine-protein hapten that elicits the immune response.

Mechanisms of adrenergic agonist induced allergy bioactivation and antigen formation.

Reduction of elevated intraocular pressure with alpha 2 agonists has proved to be an exciting new therapeutic approach for the treatment of glaucoma. We have studied the chemical reactivities of several alpha 2 agonists and known allergens to elucidate the origin of the observed ocular allergic response to the alpha 2 agonist apraclonidine. The oxidation potentials of clonidine, apraclonidine, brimonidine, and two known allergens, amodiaquine, and epinephrine, were measured vs. a standard calomel electrode.. Agents that were oxidatively labile were treated with both chemical and enzymatic oxidants. Clonidine and brimonidine proved to be oxidatively stable in sharp contrast to apraclonidine which had an oxidation potential similar to that observed with epinephrine and amodiaquine, two known allergy-inducing agents. In addition, two glutathione-apraclonidine conjugates formed by the in-situ reaction of glutathione with an enzymatically oxidized apraclonidine intermediate were isolated and their structures determined using spectroscopic methods. The structures were shown to be analogous to those obtained with amodiaquine and epinephrine. Apraclonidine, like amodiaquine and epinephrine, possesses a hydroquinone-like subunit and can be readily oxidized and conjugated with thiols modeling well known hapten-forming reactions. Brimonidine, like clonidine, lacks the hydroquinone subunit and does not undergo the thiol conjugation reactions.

Synthesis and evaluation of 2-(arylamino)imidazoles as alpha 2-adrenergic agonists.

A series of 2-(arylamino)imidazoles was synthesized and evaluated for activity at alpha 1- and alpha 2-adrenoceptors. This class of agents has been shown to have potent and selective agonist activity at the alpha 2-adrenoceptors. The most potent member of this class, 2-[(5-methyl-1,4-benzodioxan-6yl)amino]imidazole, proved efficacious for the reduction of intraocular pressure upon topical administration and for the reduction of blood pressure upon intravenous administration. During the course of our studies, we developed a new reagent that allowed rapid assembly of the target compounds. This reagent, N-(2,2-diethoxyethyl)carbodiimide, was convenient to prepare and was stable under low-temperature storage conditions.

Characterization of brimonidine metabolism with rat, rabbit, dog, monkey and human liver fractions and rabbit liver aldehyde oxidase.

1. In vitro metabolism of 14C-brimonidine by the rat, rabbit, dog, monkey and human liver fractions was studied to assess any species differences. In vitro metabolism with rabbit liver aldehyde oxidase and human liver slices, and in vivo metabolism in rats were also investigated. The hepatic and urinary metabolites were characterized by liquid chromatography and mass spectrometry. 2. Up to seven, six, 11 and 14 metabolites were detected in rat liver S9 fraction, human liver S9 fraction, human liver slices and rat urine respectively. Rabbit liver aldehyde oxidase catalysed the metabolism of brimonidine to 2-oxobrimonidine and 3-oxobrimonidine, and further oxidation to the 2,3-dioxobrimonidine. Menadione inhibited the liver aldehyde oxidase-mediated oxidation. 3. Hepatic oxidation of brimonidine to 2-oxobrimonidine, 3-oxobrimonidine and 2,3-dioxobrimonidine was a major pathway in all the species studied, except the dog whose prominent metabolites were 4',5'-dehydrobrimonidine and 5-bromo-6-guanidinoquinoxaline. 4. These results indicate extensive hepatic metabolism of brimonidine and provide evidence for aldehyde oxidase involvement in brimonidine metabolism. The species differences in hepatic brimonidine metabolism are likely related to the low activity of dog liver aldehyde oxidase. The principal metabolic pathways of brimonidine are alpha(N)-oxidation to the 2,3-dioxobrimonidine, and oxidative cleavage of the imidazoline ring to 5-bromo-6-guanidinoquinoxaline.

Synthesis and evaluation of 2-[(5-methylbenz-1-ox-4-azin-6-yl)imino]imidazoline, a potent, peripherally acting alpha 2 adrenoceptor agonist.

We have synthesized 2-[(5-methylbenz-1-ox-4-azin-6-yl)imidazoline, 3, a potent, peripherally acting alpha 2 adrenoceptor agonist. The agent is conveniently prepared in five steps from 2-amino-m-cresol. The agent has demonstrated good selectivity for alpha 2 adrenoceptors in binding and functional studies. When applied topically to eyes, the agent is efficacious for the reduction of intraocular pressure. The agent does not penetrate the blood-brain barrier and, as a consequence, does not lower blood pressure or induce sedation when administered topically or intravenously. We have determined the pKa and log P in water versus both octanol and dodecane of 3 and a set of related agents. The best physical parameter to explain its lack of central nervous system penetration appears to be log P measured in octanol versus water.

Demonstration of a pronounced effect of noncovalent binding selectivity on the (+)-CC-1065 DNA alkylation and identification of the pharmacophore of the alkylation subunit.

Studies on the structural origin of the DNA alkylation selectivity of the antitumor antibiotic (+)-CC-1065 are detailed. The sites of alkylation of double-stranded DNA were examined for simple derivatives of 7-methyl-1,2,8,8a-tetrahydrocycloprop[1,2-c]pyrrolo[3,2-e]indol- 4(5H)-one (CPI), (+)-CC-1065, and agents incorporating the parent 1,2,7,7a-tetrahydrocycloprop[1,2-c]indol-4-one (CI) left-hand subunit. The CI subunit of the agents is a much more reactive alkylating agent than the natural CPI alkylation subunit of CC-1065. Consequently, simple derivatives of CI were found to alkylate double-stranded DNA under milder conditions than were simple derivatives of CPI, and the marked similarities in the CI and CPI DNA alkylation profiles illustrate that CI represents the minimum pharmacophore of CPI. Comparisons of the DNA alkylation profiles of (+)-N-butyloxycarbonyl-CPI, (+)-N-acetyl-CPI, and (+)-CC-1065 revealed distinctions in the CPI and (+)-CC-1065 sites of alkylation, whereas the incorporation of the reactive CI electrophile into an analog of CC-1065 (CI-CDPI2) (CDPI, N3-carbamoyl-1,2-dihydro-3H-pyrrolo[3,2-e]indole-7-carboxylic acid) provided an agent that possesses the characteristic CC-1065 DNA alkylation profile (site selectivity and relative site intensity). These observations suggest that the noncovalent binding selectivity of the agents may restrict the number of available DNA alkylation sites and play a productive role in controlling the sequence-selective alkylation by effectively delivering the electrophile to A + T-rich minor groove regions of DNA possessing accessible adenine N-3 alkylation sites. In turn, the noncovalent binding selectivity may be derived from preferential binding within the narrower, sterically more accessible A + T-rich minor groove of double-stranded DNA.

An alternative and convenient strategy for generation of substantial quantities of singly 5'-32P-end-labeled double-stranded DNA for binding studies: development of a protocol for examination of functional features of (+)-CC-1065 and the duocarmycins that contribute to their sequence-selective DNA alkylation properties.

Development of an alternative strategy for securing substantial quantities of singly 5'-32P-end-labeled double-stranded DNA suitable for binding studies is described based on M13 cloning techniques and offers advantages of production of replenishable quantities of singly 5'-32P-end-labeled double-stranded DNA of homogenous length without need for DNA isolation (restriction fragment), dephosphorylation, and lengthy preparative gel electrophoresis procedures. The 32P label is introduced onto the free 5'-hydroxyl group of a chemically synthesized universal primer [5'-32P-d(GTAAAACGACGGCCAGT)-3'] which is used to initiate DNA synthesis on M13-derived single-stranded DNA templates. Following DNA synthesis, a restriction enzyme cleavage reaction produces a uniform length duplex suitable for agent binding studies. The strategy further permits the use of the Sanger dideoxynucleotide sequencing technique for direct and unambiguous identification of cleavage sites introduced by an agent on the end-labeled DNA. The use of the procedure in the examination of the DNA alkylation properties of (+)-CC-1065 (1) and a series of synthetic analogs is reviewed. From these studies a refined definition of the alkylation selectivity of (+)-CC-1065 is detailed. Employing agents possessing the parent 1,2,7,7a-tetrahydrocycloprop[1,2-c]indol-4-one (CI) alkylation subunit constituting the minimum pharmacophore of the CC-1065 alkylation subunit (CPI), comparative DNA alkylation studies illustrate that the activated cyclopropane is not obligatory for observation of the CI/CPI characteristic alkylation, highlight the relative nonselectivity of the alkylation event in the absence of noncovalent binding selectivity, illustrate a prominent role for agent binding selectivity for agents that possess such capabilities, and demonstrate that a sequence dependent autocatalytic phosphate activation of the alkylation event may not be uniquely responsible for the nonselective or selective alkylations. The ease with which the procedure may be extended to the rapid and convenient examination of additional agents is illustrated with the demonstration of the strikingly similar DNA alkylation properties of the duocarmycins (3-8) and (+)-CC-1065 (1) which suggest that the agents may be acting by a common mechanism.