Haloperidol (HP) has been used for the amelioration of some of the symptoms of schizophrenia since its introduction into clinical practice in the 1950s. Unfortunately, in up to 70% of patients, long-term treatment with HP leads to the development of movement disorders, such as parkinsonism and tardive dyskinesia. In some instances these side effects do not dissipate upon drug withdrawal, suggesting the induction of permanent damage.
It has long been known that HP undergoes carbonyl reduction to reduced HP (RHP) and N-dealkylation to the products, FBPA and CPHP. More recently it was shown that HP is converted to two pyridinium metabolites, HPP+ and RHPP+, which are structural analogues of the well-characterised parkinsonian agent, MPP+. This led to the hypothesis that HP metabolites may contribute to the development of side effects. The back oxidation of RHP to HP, apparently by P450 2D6, has been reported in vivo and in vitro. Previous in vitro studies have also implicated cytochrome P450 enzymes in the biotransformation of HP to HPP+; however, conflicting evidence has been reported for the existence of an intermediate, HPTP, in this pathway.
The aims of the work conducted in this thesis were to: 1) examine whether HPP+ and RHPP+ are present in the brain of HP-treated patients and HPTP-treated baboons; 2) investigate whether HPTP is an intermediate in the biotransformation of HP to HPP+, in vivo and in vitro; 3) investigate the metabolism of HP and RHP using human liver microsomal preparations and specifically, to identify the cytochrome P450 enzymes responsible for the metabolism of HP and RHP; 4) characterise the kinetics of these metabolic pathways using human liver microsomes and recombinant P450s; and 5) investigate the catalytic ability of mutant P450 enzymes to metabolise HP and RHP.
Although HP is converted to HPP+ in brain mitochondrial preparations, and HPP+accumulated in the brains of rats following repeated administration of HP, the presence of HPP+ and RHPP+ in the brains of HP-treated patients has not been reported. The current investigation reports the presence of HPP+ and RHPP+ in the brains of HPTP-treated baboons and HP-treated humans in vivo.
The biotransformation of HP and RHP to pyridinium metabolites was proposed to occur via the tetrahydropyridine intermediates, HPTP or RHPTP. However, the results of a study reported here found no evidence of those compounds in the urine or plasma of any patient treated with high dose intravenous HP, suggesting that HPTP and RHPTP are not intermediates in pyridinium formation.
In vitro studies were performed using human liver microsomes and recombinant P450s (rP450s 1A1, 1A2, 1B1, 2A6, 2B6, 2C9, 2C19, 2D6, 2E1, 3A4, 3A5, 3A7) to identify the enzymes involved in the conversion of HP to HPP+, RHP to RHPP+, RHP to HP, and the N-dealkylation of both HP and RHP. P450 3A4 was the most active enzyme for all pathways, although P450s 3A5 and 3A7 also demonstrated some catalytic activity. P450 2D6 was also capable of catalysing the oxidation of RHP to HP at a low level. In support of the in vivo findings, no in vitro evidence was obtained to support the hypothesis that HPTP and RHPTP are intermediates in the biotransformation of HP.
The kinetics of HP and RHP metabolism by human liver microsomes and rP450 3A4 were analysed by non-linear regression using the Michaelis-Menten (MM) equation. The average Km value determined for the three human liver microsomal preparatioons corresponded well with the estimated Km for rP450 3A4. Using rP450 3A4, all metabolic pathways displayed typical MM kinetics, however, deviations consistent with substrate inhibition were apparent for some metabolic pathways with some human liver microsomes.
While P450s 3A4 and 3A5 share greater than 80% sequence homology, the catalytic activity of P450 3A4 towards HP and RHP was considerably higher than that of P450 3A5. Sequence alignment and homology modelling of P450s has led to the identification of six putative substrate recognition sites (SRSs 1-6) that may be important in substrate binding; P450s 3A4 and 3A5 differ by only 17 amino acids in these 6 SRSs. The effect of interchanging these SRS regions on catalytic activity was investigated in preliminary studies using chimeric enzymes. An important role for SRS1 in conferring P450 3A4-like activity was demonstrated. Unexpectedly, replacing SRS 6 of P450 3A4 with SRS 6 of P450 3A5 resulted in catalytic activity equivalent to or greater than that of P450 3A4 for all metabolic pathways.
In summary these studies indicate that P450 3A4 is the primary P450 involved in HP metabolism and suggest that HP and RHP may be effective probe substrates for investigating the structure-function relationships in P450 3A enzymes.