Drug Induced Liver Disease (Hepatotoxicity)-Chapter 7-CYTOCHROME P450 ACTIVATION OF TOXINS AND HEPATOTOXICITY

P450 ENZYMES:-

           The human genome contains 57 P450s, and these are named systematically based on sequence similarity. The sequence of a P450 does not provide direct insight into catalytic function, however. Some assignments are equivocal (e.g., 1B1 and 27A1), but the P450s involved in the metabolism of steroids, eicosanoids, and fat-soluble vitamins are clearly important in development and normal physiology. The P450s in the “xenobiotics” column are not considered critical but these are the enzymes involved in drug metabolism.

           In considering metabolism of all commercial drugs, w75% is attributed to P450s (Fig. 1). Within this group, w95% is catalyzed by five P450s (Fig. 1B).

Humans vary considerably in terms of amounts of each P450 and catalytic activity. Ideally, the same pharmacokinetic pattern would be observed in all individuals, with the increase and decrease in plasma levels (and presumably levels of drug and metabolites in each tissue) being the same. Alternatively, unexpectedly rapid metabolism and low levels of the drug could result in therapeutic effectiveness. Further, if a P450 converts the drug to a toxic metabolite, the patient could be at higher risk.

Figure 7.1. (A) Contribution of individual enzymes in (human) drug metabolism. (B) Contribution of individual P450 enzymes in drug metabolism. Abbreviations: UGT, UDP-glucuronosyl transferase; FMO, flavin-containing monooxygenase; NAT, N-acetyltransferase; MAO, monoamine oxidase 

CONTEXTS OF TOXICITY:-

One way to consider drug toxicity is classification with regard to general mechanism.

Contexts of Drug Toxicity:-

On-target

Hypersensitivity and immune reactions

Off-target

Bioactivation to reactive intermediates

Idiosyncratic

      The first of the five contexts, on-target or mechanism-based toxicity, is relatively

straightforward to deal with. An example is the statins. Most of the toxicities can be explained by the intended action of the parent drug on 3-hydroxymethylglutaryl-CoA reductase, but in a different cell (e.g., muscle). The toxicity can be reduced by lowering the dose or, in this case, by administration of mevalonate, the product of the targeted enzyme.

         The second context is due to immune reactions resulting from modification of a protein by the parent drug. An example is penicillins and related b-lactams. This process is part of a “hapten” hypothesis. To expand this, there is a “danger” hypothesis, basically that the drug–macromolecule complex generates a “danger” signal that ultimately results in antibody/T-cell responses. A further twist on this is the “p-i concept” of “direct pharmacological interaction of drugs with immune receptors”.

         The third case is off-target pharmacology. In this case, the parent drug interacts with additional targets that were not anticipated and yields unexpected pharmacology, which can result in toxicity. A relevant example is terfenadine, an antihistamine that was recalled in 1995 (Fig. 2). In this case, P450 inhibition is also an issue. Normally, terfenadine is rapidly oxidized by P450 3A4 to fexofenadine. Both terfenadine and fexofenadine bind to the H1 receptor and have antihistaminic activity. In most individuals, the oxidation is rapid and no terfenadineis found in plasma. However, individuals who use P450 3A4 inhibitors concurrently (e.g.,erythromycin, ketoconazole) have lowered P450 3A4 activity and terfenadine accumulates. Terfenadine (but not fexofenadine) also binds to the human ether-ago-go related gene (hERG) receptor and inhibits ion channels, resulting in abnormal QT intervals and arrhythmias.

 Terfenadine, the first non-sedating antihistamine, was recalled and replaced with fexofenadine.

Figure 7.2. Terfenadine as an example of off-target toxicity and a role for metabolism. Terfenadine is extensively oxidized in three steps to a carboxylic acid (fexofenadine). Both the terfenadine and the product fexofenadine affect the target, the H1 receptor, and have antihistaminic activity. Terfenadine is normally not found in plasma following oral ingestion because of rapid metabolism. However, inhibitors of P450 3A4 block oxidation and cause terfenadine to accumulate. Terfenadine has off-target pharmacology and interacts with hERG receptors to produce cardiac problems; in some cases, abnormal QT intervals and fatal arrhythmias have been attributed to the increased terfenadine load.M Abbreviation: hERG, human ether-ago-go related gene.

Figure 7.3. Metabolism of acetaminophen and relevance to toxicity. At low doses, the metabolism of acetaminophen is largely through conjugation. Some oxidation does occur but GSH conjugation renders this inactive. At higher doses, the three conjugation pathways are not effective and larger amounts of the protein conjugates are formed and apparently contribute to toxicity.

  

The fourth context of toxicity is bioactivation, which will be discussed later. An example shown is that of acetaminophen (Fig. 3), where some oxidation to a reactive iminoquinone is always detected but can be tolerated, until the dose of the drug exceeds the ability of the conjugation systems to clear the compound. The concept of bioactivation, already presented, is generally associated with the concept of reaction of the products with proteins or other cellular macromolecules to cause damage, except in cases in which the products are themselves dangerous due to altered pharmacology (vide infra) or cause oxidative stress. Thus, this context has some similarity to the second one, except that an immune system component is usually not invoked. Historically, the concept was what can be called the “critical protein hypothesis”, where ablation of the function of an individual protein was responsible for cell toxicity; i.e., that protein acts as a “master switch” for the cell. However, a more accepted current view is that cells have very complex networks and that disruption of those signaling pathways  or any of several of the systems involved in energy production may lead to toxicity, with effects being cell and tissue specific.

The final context in this discussion is idiosyncratic, which is rare (<1/104 patients), a

serious problem, and not understood (by definition). Dihydralazine is usually regarded as an example of an idiosyncratic drug issue, although closer analysis reveals aspects of metabolic activation, covalent binding, and possibly immunological components involved in the response(FIGURE.4)

  

Figure 7.4. Possible mechanisms for dihydralazine-induced hepatitis. Dihydralazine is oxidized by P450 1A2 and the hydroxylamine and/or nitroso product form protein adducts. LKM  antibodies are found in patients. Whether these antibodies are causal in hepatitis is not established. Abbreviation: LKM, liver kidney microsomal

BIOACTIVATION BY P450s AND OTHER ENZYMES:-

Basic Aspects

Bioactivation may seem complex, the fundamentals are quite basic, even if the interpretation may not be straightforward. Seven principles are used to summarize this area:

1. The reactions are described by two basic types of chemistry.

    (a) The first is the reaction of an electrophile (E; formed from the drug) and a nucleophile (Nuc=protein, DNA, or other component in the cell)

  

or

    (b) Free radical propagation, usually resulting from an odd-electron enzymatic process or lipid peroxidation. Radical processes and oxidative stress from reactive oxygen (or nitrogen) species have not been addressed extensively here, but in some cases are probably very important. How much reactive oxygen species contribute to the overall hepatoxicity of drugs is unclear.

2. The first product or the most obvious reactant may not be the one that reacts with the nucleophile. For instance, in some cases, an epoxide may be generated by a P450 but it will undergo a rearrangement to an acyl halide or other product that participates in the reaction.

3. The stability (and also the reactivity) of reactive products may vary considerably. A short-lived product, e.g., aflatoxin B1 (AFB1) exo-8,9-epoxide (t1/2Zone second at neutral pH) can be formed in the endoplasmic reticulum but still migrates to the cell nucleus to react with DNA very efficiently (Fig. 5). Other reactive products may have half-lives on the order of minutes to hours. In contrast to the shorter t1/2 products, these can often migrate from one cell to another and even one tissue to another.

4. In vitro systems are good models for elucidating details but ultimately all covalent binding results must be considered in the context of in vivo situations. In in vitro settings, some factors may be overemphasized or biased.

5. The dose is an issue or, in many cases, the issue, going back 500 years to an axiom of toxicology from Paracelsus.

6. Covalent binding can be an index of toxicity and often correlates. Some pharmaceutical companies are using data from in vitro and in vivo covalent binding studies to at least stratify candidates for development. However, many exceptions are known, i.e., nontoxic compounds show covalent binding and toxic compounds do not bind. Thus, caution is advised in interpretation.

7. Numerous other factors influence the relevance of covalent binding to toxicity. Among these are receptor-medicated events (i.e., with the parent drug itself), the disruption of particular signaling events by the reactive products, the ability of the cells to repair damage, the intervention of immunological processes, and cell proliferation.

 

Figure 7.5. Metabolism of AFB1 and relevance to toxicity. The major human enzymes involved in each reaction are included and rates have been determined. First-order rates are listed in units of reciprocal time. Units of minK1 MK1 indicate either second-order rate constants (for chemical reactions) or catalytic efficiency (kcat/Km) for enzymatic reactions measured using the most relevant purified recombinant human enzymes. The exo-8,9-epoxide plays a central role in this process and appears to be the only product that reacts with DNA. The dialdehyde reacts with proteins (BSA). The dehydrogenase AFAR is now classified as an AKR. Several other oxidation products (AFQ, AFM, endo epoxide) are not shown, which are all less toxic than AFB1. Abbreviations: AFAR, aflatoxin aldehyde reductase; AFB1, aflatoxin B1; AFQ, aflatoxin Q1; AFM, aflatoxin M1; AKR, aldo-keto reductase; BSA, bovine serum albumin.