Phase 1 Reactions

Phase 1 reactions are catabolic (e.g. oxidation, reduction or hydrolysis), and the products are often more chemically reactive and hence, paradoxically, sometimes more toxic or carcinogenic than the parent drug. Phase 1 reactions often introduce a reactive group, such as hydroxyl, into the molecule, a process known as ‘functionalisation’. This group then serves as the point of attack for the conjugating system to attach a substituent such as glucuronide (Fig. 9.1), explaining why phase 1 reactions so often precede phase 2 reactions (see below). Phase 1 reactions take place mainly in the liver. Many hepatic drug-metabolising enzymes, including CYP enzymes, are embedded in the smooth endoplasmic reticulum. They are often called ‘microsomal’ enzymes because, on homogenisation and differential centrifugation, the endoplasmic reticulum is broken into very small fragments that sediment only after prolonged high-speed centrifugation in the microsomal fraction. To reach these metabolising enzymes in life, a drug must cross the plasma membrane. Polar molecules do this less readily than non-polar molecules except where there are specific transport mechanisms (Ch. 8), so intracellular metabolism is important for lipid-soluble drugs, while polar drugs are at least partly excreted unchanged in the urine.

Fig. 9.1 The two phases of drug metabolism.

The P450 Monooxygenase System

Nature, classification and mechanism of P450 enzymes

Cytochrome P450 enzymes are haem proteins, comprising a large family (’superfamily’) of related but distinct enzymes, each referred to as CYP followed by a defining set of numbers and a letter. These enzymes differ from one another in amino acid sequence, in sensitivity to inhibitors and inducing agents (see below), and in the specificity of the reactions that they catalyse (see Anzenbacher, 2007 for reviews). Different members of the family have distinct, but often overlapping, substrate specificities, and may act on the same substrates but at different rates. Purification of P450 enzymes and complementary DNA cloning form the basis of the current classification, which is based on amino acid sequence similarities. Seventy-four CYP gene families have been described, of which three main ones (CYP1, CYP2 and CYP3) are involved in drug metabolism in human liver. Examples of therapeutic drugs that are substrates for some important P450 isoenzymes are shown in Table 9.1. Drug oxidation by the monooxygenase P450 system requires drug (substrate, ‘DH’), P450 enzyme, molecular oxygen, NADPH and a flavoprotein (NADPH–P450 reductase). The mechanism involves a complex cycle (Fig. 9.2), but the overall net effect of the reaction is quite simple, namely the addition of one atom of oxygen (from molecular oxygen) to the drug to form a hydroxyl group (product, ‘DOH’), the other atom of oxygen being converted to water.

Table 9.1 Examples of drugs that are substrates of P450 isoenzymes

Isoenzyme P450	Drug(s)
CYP1A2	Caffeine, paracetamol (→NAPQI), tacrine, theophylline
CYP2B6	Cyclophosphamide, methadone
CYP2C8	Paclitaxel, repaglinide
CYP2C19	Omeprazole, phenytoin
CYP2C9	Ibuprofen, tolbutamide, warfarin
CYP2D6	Codeine, debrisoquine, S-metoprolol
CYP2E1	Alcohol, paracetamol
CYP3A4, 5, 7	Ciclosporin, nifedipine, indinavir, simvastatin

(Adapted from http://medicine.iupui.edu/flockhart/table.htm.)

Fig. 9.2 The monooxygenase P450 cycle.

Each of the pink or blue rectangles represents one single molecule of cytochrome P450 (P450) undergoing a catalytic cycle. Iron in P450 is in either the ferric (pink rectangles) or ferrous (blue rectangles) state. P450 containing ferric iron (Fe³⁺) combines with a molecule of drug (‘DH’); receives an electron from NADPH–P450 reductase, which reduces the iron to Fe²⁺; combines with molecular oxygen, a proton and a second electron (either from NADPH–P450 reductase or from cytochrome b₅) to form an Fe²⁺OOH–DH complex. This combines with another proton to yield water and a ferric oxene (FeO)³⁺–DH complex. (FeO)³⁺ extracts a hydrogen atom from DH, with the formation of a pair of short-lived free radicals (see text), liberation from the complex of oxidised drug (‘DOH’), and regeneration of P450 enzyme.

page 116

P450 enzymes have unique spectral properties, and the reduced forms combine with carbon monoxide to form a pink compound (hence ‘P’) with absorption peaks near 450 nm (range 447–452 nm). The first clue that there is more than one form of CYP came from the observation that treatment of rats with 3-methylcholanthrene (3-MC), an inducing agent (see below), causes a shift in the absorption maximum from 450 to 448 nm—the 3-MC-induced isoform of the enzyme absorbs light maximally at a slightly shorter wavelength than the un-induced enzyme.

page 117

P450 and biological variation

There are important variations in the expression and regulation of P450 enzymes between species. For instance, the pathways by which certain dietary heterocyclic amines (formed when meat is cooked) generate genotoxic products involves one member of the P450 superfamily (CYP1A2) that is constitutively present in humans and rats (which develop colon tumours after treatment with such amines) but not in cynomolgus monkeys (which do not). Such species differences have crucial implications for the choice of species to be used for toxicity and carcinogenicity testing during the development of new drugs for use in humans.

Within human populations, there are major sources of interindividual variation in P450 enzymes that are of great importance in therapeutics. These include genetic polymorphisms (alternative sequences at a locus within the DNA strand—alleles—that persist in a population through several generations; Ch. 11). Environmental factors (Ch. 56) are also important, since enzyme inhibitors and inducers are present in the diet and environment. For example, a component of grapefruit juice inhibits drug metabolism (leading to potentially disastrous consequences, including cardiac dysrhythmias; Ch. 56), whereas Brussels sprouts and cigarette smoke induce P450 enzymes. Components of St John’s wort (used to treat depression in ‘alternative’ medicine; Ch. 46) induce CYP450 isoenzymes as well as P-glycoprotein (P-gp) (see Ch. 8 and below, and Henderson et al., 2002).

Not all drug oxidation reactions involve the P450 system: some drugs are metabolised in plasma (e.g. hydrolysis of suxamethonium by plasma cholinesterase; Ch. 13), lung (e.g. various prostanoids; Ch. 17) or gut (e.g. tyramine, salbutamol; Chs 14 and 27). Ethanol (Ch. 48) is metabolised by a soluble cytoplasmic enzyme, alcohol dehydrogenase, in addition to CYP2E1. Other P450-independent enzymes involved in drug oxidation include xanthine oxidase, which inactivates 6-mercaptopurine (Ch. 55), and monoamine oxidase, which inactivates many biologically active amines (e.g. noradrenaline [norepinephrine], tyramine, 5-hydroxytryptamine; Chs 14 and 15).

Hydrolytic reactions (e.g. of aspirin; Fig. 9.1) do not involve hepatic microsomal enzymes but occur in plasma and in many tissues. Both ester and (less readily) amide bonds are susceptible to hydrolysis. Reductive reactions are much less common than oxidations, but some are important. For example, warfarin (Ch. 24) is inactivated by conversion of a ketone to a hydroxyl group by CYP2A6.

Phase 2 Reactions

Phase 2 reactions are synthetic (‘anabolic’) and involve conjugation (i.e. attachment of a substituent group), which usually results in inactive products, although there are exceptions (e.g. the active sulfate metabolite of minoxidil, a potassium channel activator used to treat severe hypertension, Ch. 22; morphine-6-glucuronide is an active metabolite of morphine that is being developed as an analgesic agent [Ch. 41]—on acute administration it induces less nausea and vomiting than the parent drug perhaps because, being more polar, it fails to access the vomiting centres). Phase 2 reactions also take place mainly in the liver. If a drug molecule has a suitable ‘handle’ (e.g. a hydroxyl, thiol or amino group), either in the parent molecule or in a product resulting from phase 1 metabolism, it is susceptible to conjugation. The groups most often involved are glucuronyl (Fig. 9.3), sulfate, methyl and acetyl. The tripeptide glutathione can conjugate drugs or their phase 1 metabolites via its sulfhydryl group, as in the detoxification of paracetamol (see Fig. 57.1, p. 701). Glucuronide formation involves the formation of a high-energy phosphate compound, uridine diphosphate glucuronic acid (UDPGA), from which glucuronic acid is transferred to an electron-rich atom (N, O or S) on the substrate, forming an amide, ester or thiol bond. UDP-glucuronyl transferase, which catalyses these reactions, has very broad substrate specificity embracing many drugs and other foreign molecules. Several important endogenous substances, including bilirubin and adrenal corticosteroids, are conjugated by the same system.

Fig. 9.3 The glucuronide conjugation reaction.

A glucuronyl group is transferred from uridine diphosphate glucuronic acid (UDPGA) to a drug molecule.

Acetylation and methylation reactions occur with acetyl-CoA and S-adenosyl methionine, respectively, acting as the donor compounds. Many of these conjugation reactions occur in the liver, but other tissues, such as lung and kidney, are also involved.

Stereoselectivity

Many clinically important drugs, such as sotalol (Ch. 21), warfarin (Ch. 24) and cyclophosphamide (Ch. 55), are mixtures of stereoisomers, the components of which differ not only in their pharmacological effects but also in their metabolism, which may follow completely distinct pathways (see Campo et al., 2009 for a recent review). Several clinically important drug interactions involve stereospecific inhibition of metabolism of one drug by another (Ch. 56). In some cases, drug toxicity is mainly linked to one of the stereoisomers, not necessarily the pharmacologically active one. Where practicable, regulatory authorities urge that new drugs should consist of single isomers to avoid these complications.¹

Inhibition of P450

Inhibitors of P450 differ in their selectivity towards different isoforms of the enzyme, and are classified by their mechanism of action. Some drugs compete for the active site but are not themselves substrates (e.g. quinidine is a potent competitive inhibitor of CYP2D6 but is not a substrate for it). Non-competitive inhibitors include drugs such as ketoconazole, which forms a tight complex with the Fe³⁺ form of the haem iron of CYP3A4, causing reversible non-competitive inhibition. So-called mechanism-based inhibitors require oxidation by a P450 enzyme. Examples include the oral contraceptive gestodene (CYP3A4) and the anthelminthic drug diethylcarbamazine (CYP2E1). An oxidation product (e.g. a postulated epoxide intermediate of gestodene) binds covalently to the enzyme, which then destroys itself (’suicide inhibition’; see Pelkonen et al., 2008 for a fuller review). Many clinically important interactions between drugs are the result of inhibition of P450 enzymes (see Ch. 56).

Induction of Microsomal Enzymes

A number of drugs, such as rifampicin (Ch. 50), ethanol (Ch. 48) and carbamazepine (Ch. 44), increase the activity of microsomal oxidase and conjugating systems when administered repeatedly. Many carcinogenic chemicals (e.g. benzpyrene, 3-MC) also have this effect, which can be substantial; Figure 9.4 shows a nearly 10-fold increase in the rate of benzpyrene metabolism 2 days after a single dose. The effect is referred to as induction, and is the result of increased synthesis and/or reduced breakdown of microsomal enzymes—see Park et al. (1996), Dickins (2004) and Pelkonen et al. (2008) for more detail.

Fig. 9.4 Stimulation of hepatic metabolism of benzpyrene.

Young rats were given benzpyrene (intraperitoneally) in the doses shown, and the benzpyrene-metabolising activity of liver homogenates was measured at times up to 6 days.

(From Conney A H et al. 1957 J Biol Chem 228: 753.)

Enzyme induction can increase drug toxicity and carcinogenicity (Park et al., 2005), because several phase 1 metabolites are toxic or carcinogenic: paracetamol is an important example of a drug with a highly toxic metabolite (see Ch. 57).

The mechanism of induction is incompletely understood but is similar to that involved in the action of steroid and other hormones that bind to nuclear receptors (see Ch. 3). The most thoroughly studied inducing agents are polycyclic aromatic hydrocarbons (e.g. 3-MC). These bind to the ligand-binding domain of a soluble protein, termed the aromatic hydrocarbon (Ah) receptor. This complex is transported to the nucleus by an Ah receptor nuclear translocator and binds Ah receptor response elements in the DNA, thereby promoting transcription of the gene CYP1A1. In addition to enhanced transcription, some inducing agents (e.g. ethanol, which induces CYP2E1 in humans) also stabilise mRNA or P450 protein.

First-Pass (Presystemic) Metabolism

Some drugs are extracted so efficiently by the liver or gut wall that the amount reaching the systemic circulation is considerably less than the amount absorbed. This is known as first-pass or presystemic metabolism and reduces bioavailability (Ch. 8) even when a drug is well absorbed. Presystemic metabolism is important for many therapeutic drugs (Table 9.2 shows some examples), and is a problem because:

Table 9.2 Examples of drugs that undergo substantial first-pass elimination

Pharmacologically Active Drug Metabolites

In some cases (see Table 9.3), a drug becomes pharmacologically active only after it has been metabolised. For example, azathioprine, an immunosuppressant drug (Ch. 26), is metabolised to mercaptopurine; and enalapril, an angiotensin-converting enzyme inhibitor (Ch. 22), is hydrolysed to its active form enalaprilat. Such drugs, in which the parent compound lacks activity of its own, are known as prodrugs. These are sometimes designed deliberately to overcome problems of drug delivery (Ch. 8). Metabolism can alter the pharmacological actions of a drug qualitatively. Aspirin inhibits some platelet functions and has anti-inflammatory activity (Chs 24 and 26). It is hydrolysed to salicylic acid (Fig. 9.1), which has anti-inflammatory but not antiplatelet activity. In other instances, metabolites have pharmacological actions similar to those of the parent compound (e.g. benzodiazepines, many of which form long-lived active metabolites that cause sedation to persist after the parent drug has disappeared; Ch. 43). There are also cases in which metabolites are responsible for toxicity. Hepatotoxicity of paracetamol is one example (see Ch. 57), and bladder toxicity of cyclophosphamide, which is caused by its toxic metabolite acrolein (Ch. 55), is another. Methanol and ethylene glycol both exert their toxic effects via metabolites formed by alcohol dehydrogenase. Poisoning with these agents is treated with ethanol (or with a more potent inhibitor), which competes for the active site of the enzyme. Disulfiram inhibits CYP2E1 and reduces substantially the formation of trifluoroacetic acid during halothane anaesthesia, raising the intriguing possibility that it could prevent halothane hepatitis (see Kharasch, 2008).

Table 9.3 Some drugs that produce active or toxic metabolites

Biliary Excretion and Enterohepatic Circulation

Liver cells transfer various substances, including drugs, from plasma to bile by means of transport systems similar to those of the renal tubule including organic cation transporters (OCTs), organic anion transporters (OATs) and P-glycoproteins (P-gp) (see Ch. 8). Various hydrophilic drug conjugates (particularly glucuronides) are concentrated in bile and delivered to the intestine, where the glucuronide is usually hydrolysed, releasing active drug once more; free drug can then be reabsorbed and the cycle repeated (enterohepatic circulation). The effect of this is to create a ‘reservoir’ of recirculating drug that can amount to about 20% of total drug in the body and prolongs drug action. Examples where this is important include morphine (Ch. 41) and ethinylestradiol (Ch. 34). Several drugs are excreted to an appreciable extent in bile. Vecuronium (a non-depolarising muscle relaxant; Ch. 13) is an example of a drug that is excreted mainly unchanged in bile. Rifampicin (Ch. 50) is absorbed from the gut and slowly deacetylated, retaining its biological activity. Both forms are secreted in the bile, but the deacetylated form is not reabsorbed, so eventually most of the drug leaves the body in this form in the faeces.

Renal Excretion of Drugs and Metabolites

Drugs differ greatly in the rate at which they are excreted by the kidney, ranging from penicillin (Ch. 50), which is cleared from the blood almost completely on a single transit through the kidney, to diazepam (Ch. 43), which is cleared extremely slowly. Most drugs fall between these extremes, and metabolites are nearly always cleared more quickly than the parent drug. Three fundamental processes account for renal drug excretion:

Diffusion Across the Renal Tubule

Water is reabsorbed as fluid traverses the tubule, the volume of urine emerging being only about 1% of that of the glomerular filtrate. Consequently, if the tubule is freely permeable to drug molecules, some 99% of the filtered drug will be reabsorbed passively down the resulting concentration gradient. Lipid-soluble drugs are therefore excreted poorly, whereas polar drugs of low tubular permeability remain in the lumen and become progressively concentrated as water is reabsorbed. Polar drugs handled in this way include digoxin and aminoglycoside antibiotics. These exemplify a relatively small but important group of drugs (Table 9.5) that are not inactivated by metabolism, the rate of renal elimination being the main factor that determines their duration of action. These drugs have to be used with special care in individuals whose renal function may be impaired, including the elderly and patients with renal disease or any severe acute illness (Ch. 56).

Table 9.5 Examples of drugs that are excreted largely unchanged in the urine

Percentage	Drugs excreted
100–75	Furosemide, gentamicin, methotrexate, atenolol, digoxin
75–50	Benzylpenicillin, cimetidine, oxytetracycline, neostigmine
∼50	Propantheline, tubocurarine

The degree of ionization of many drugs—weak acids or weak bases—is pH dependent, and this markedly influences their renal excretion. The ion-trapping effect means that a basic drug is more rapidly excreted in an acid urine which favours the charged form and thus inhibits reabsorption. Conversely, acidic drugs are most rapidly excreted if the urine is alkaline (Fig. 9.5). Urinary alkalinisation is used to accelerate the excretion of salicylate in treating selected cases of aspirin overdose.

Fig. 9.5 The effect of urinary pH on drug excretion.

[A] Phenobarbital clearance in the dog as a function of urine flow. Because phenobarbital is acidic, alkalinising the urine increases clearance about five-fold. [B] Amphetamine excretion in humans. Acidifying the urine increases the rate of renal elimination of amphetamine, reducing its plasma concentration and its effect on the subject’s mental state.

(Data from Gunne & Anggard 1974. In: Torrell T et al. (eds) Pharmacology and pharmacokinetics. Plenum, New York.)