Drug metabolism by leukocytes and its role in drug-induced lupus and other idiosyncratic drug reactions.

Toxicology destruction of erythrocytes, to aplastic anemia, in which the bone marrow is damaged and there is a deficiency of ail blood cell lines. Isolated deficiencies of other blood cell lines, especially granulocytes, are common. Damage to other organs, especially the liver and kidneys, is also common in idiosyncratic drug reactions. Autoimmune syndromes such as drug-induced lupus can also occur.

Drug Metabolism by Leukocytes and Its Role in Drug-Induced Lupus and Other Idiosyncratic Drug Reactions Jack Uetrecht

1. IDIOSYNCRATIC DRUG REACTIONS Critical Reviews in Toxicology Downloaded from informahealthcare.com by North Carolina State University on 10/01/12 For personal use only.

C. Possible Mechanisms of Idiosyncratic Reactions A. Scope of Idiosyncratic Reactions The term idiosyncratic drug reaction is used here to mean an adverse reaction to a drug which is not expected on the basis of the known pharmacological actions of the drug and which does not occur in most patients who are treated with the drug. A good example of such a reaction is aplastic anemia which occurs with an incidence of about 1 in 30,000 in patients treated with chloramphenicol. Idiosyncratic drug reactions represent a serious medical problem. Although the incidence of an idiosyncratic drug reaction with any specific drug may be relatively low, there are a large number of drugs that are associated with serious, often Jife-threatening reactions. In addition, it is impossible to predict who will have an idiosyncratic drug reaction, and therapy is limited to discontinuation of the drug and supportive therapy. Unfortunately, the reaction often progresses after the drug has been discontinued. With chloramphenicol-induced aplastic anemia, the anemia often begins more than a month after the chloramphenicol has been discontinued, and a long delay between the end of therapy and the onset of anemia represents an unfavorable prognostic sign.* Idiosyncratic drug reactions are also a major problem for the development of new drugs. In general, the ability of a drug to cause a high incidence of idiosyncratic reactions in humans is not detected in animal toxicity tests. The number of patients involved in early clinical tests is also insufficient to detect many serious idiosyncratic reactions. Therefore, the incidence of such reactions is not determined until the drug has been released on the market. Many drugs, such as practolol, benoxaprofen, ticrynafen, zomepirac, and nomifensine, were withdrawn from the market shortly after they were released because of an unacceptable risk of toxicity. Each of these cases represents an enormous cost to the drug company involved as well as to the patients who sustained a serious reaction.

In general, the mechanisms of idiosyncratic drug reactions

are unknown. It is unliiely that many idiosyncratic drug reactions involve direct cytotoxicity of the drug or its metabolites. If such were the case, the reaction would be expected to be more predictable and also to be detected in animal toxicity tests with high doses. Differences in drug metabolism could lead to large interindividual and interspecies differences in toxicity, but this does not appear to be the sole basis for the idiosyncratic character of these drug reactions. Large interindividual and interspecies differences also exist in the pharmacodynamics of a drug, but as stated earlier, most idiosyncratic reactions do not appear to represent an exaggerated response involving the known pharmacological effects of a drug. Some idiosyncratic reactions may be due to the combined effect of a drug and some other factor such as an infection. For example, epidemiological studies suggest that most cases of Reye's syndrome are caused by the combination of aspirin and a viral infection, usually influenza or ~ a r i c e l l a .How ~.~ aspirin and a viral infection interact to cause Reye's syndrome is unknown. Mononucleosis is also associated with a greatly increased incidence of ampicillin rash,5 and AIDS is associated with an increase in the incidence in sulfonamide hypersensitivity reactions . 6 Aspirin hepatotoxicity and other types of idiosyncratic drug reactions appear to be more common in patients with diseases such as The basis for these interactions is unknown. Adverse drug reactions that involve the immune system are referred to as drug hypersensitivity reactions. The pathological processes involved in hypersensitivity reactions were classified by Gel1 and Coombs into four types:" type I, or immediate hypersensitivity reactions which are mediated by IgE; type I1 reactions in which the injury is caused by binding antibodies to specific tissue antigens; type I11 in which injury involves antigen-antibody complexes which activate complement; and type IV,or delayed hypersensitivity, in which injury is mediated by activated cells rather than by antibody. The major cell types involved in type IV reactions are T lymphocytes and macrophages.

B. Manifestations of Idiosyncratic Drug Reactions Different drugs have different patterns of idiosyncratic reactions and the same drug can cause different reactions in different patients; however, certain patterns are common. Probably, the two most common targets of idiosyncratic drug reactions are the skin and blood cells. Involvement of the skin can vary from a mild morbilliform (measles-like)rash to a lifethreatening Stevens-Johnson syndrome which resembles a serious thermal bum. Involvement of the blood cells can vary from a mild hemolytic anemia, in which there is peripheral

J. Uetrecht received his B.S. from the University of Cincinnati, Cincinnati, Ohio; his M.S. and Ph.D. were earned at Cornell University, Ithaca, New York; he also obtained an M.D. degree from The Ohio State University, Columbus. Dr. Uetrecht is an Associate Professor, Faculties of Pharmacy and Medicine, University of Toronto and Sunnybrook Medical Centre, Toronto, Ontario, Canada.

1990

J

2 3

Critical Reviews In

Critical Reviews in Toxicology Downloaded from informahealthcare.com by North Carolina State University on 10/01/12 For personal use only.

Characteristics of many idiosyncratic reactions suggest involvement of the immune system, although in most cases this has not been proven. Such characteristics include

*

A requirement for either prior exposure to the drug, or a delay of more than a week between starting the drug and the development of toxicity. A lack of delay in toxicity on reexposure of a patient to the offending drug. An apparent lack of correlation between dose and the risk of toxicity. This is probably overemphasized, although the range in the toxic dose is greater than with other types of toxicity. An unpredictable nature and lack of an animal model.

There are several excellent reviews of drug hypersensitivity 1.1: but a brief review of the area is presented as a background for understanding how reactive metabolites generated by leukocytes could lead to hypersensitivity reactions. Clnfortunately . with the exception of anaphylactic reactions, not enough is known about the mechanism of many drug hypersensitivity reactions to place them into the Gell and Coombs classification scheme. For a drug to cause a hypersensitivity reaction according to the Gell and Coombs classification, it would probably have to act as a hapten. There are other ways in which a drug could interact with the immune system that would result in injury. and a more general scheme i s presented here: 1,

;!.

3.

A drug or its metabolite could act as a hapten and bind to an endogenous protein. This would alter the protein so that it might be recognized as foreign by the immune system and act as an immunogen to induce the formation of antibodies or a cellular response. If IgE is induced, it would result in a type I reaction. If a tissue-specific antibody is induced, it would result in a type I1 reaction. If antigen-antibody complexes are formed, a type III reaction would be induced. If a cellular response is induced. it would result in a type IV reaction. A drug or its metabolite could directly modify activation of complement or clearance of immune complexes. A drug or its metabolite could alter the function of some cell (i.e.. helper or suppressor T lymphocytes) that is involved in control of the immune system.

1. Toxicity Mediated by Hapten-Induced Immune Reaction Of the above mechanisms, the first has been studied the most extensively. Most adverse drug reactions appear to involve the drug or its metabolite acting as a hapten. In general, for a molecule to be immunogenic. it must have a large molecular weight. usually larger than 1000.7~ii~i2Since the majority of drugs are not this large, they must bind to a large molecule in

214

order to be immunogenic. In general, this binding must be virtually irreversible, usually covalent, for the adduct to be immunogenic. The probable basis for the requirement of a covalent interaction is that for an antigenic response to be induced the immunogen must be processed by a macrophage and presented, along with the class I1 major histocompatibility antigen, to a T lymphocyte.’’ If the interaction between hapten and macromolecule were reversible, the hapten would diffuse away from the macromolecule before the adduct could be processed and presented to a T lymphocyte. Penicillin is a classic example. Penicillin is chemically reactive by virtue of its P-lactam ring, and both it and some of its metabolites can act as haptens and bind to protein.’ In the majority of patients, this leads to the induction of antibodies against penicillin-haptenized protein; however, only a small proportion of patients develop a clinically manifest reaction. Any one of the four types of hypersensitivity reactions can occur, but the most feared is IgE-mediated anaphylaxis (Gell and Coombs type I). It is relatively easy to demonstrate the mechanism of type I reactions because IgE causes characteristic symptoms, and it is easy to demonstrate the presence of specific IgE antibodies. Most drugs are not chemically reactive and therefore cannot act as haptens. However, many drugs are metabolized to chemically reactive metabolites which can act as haptens. Quinidine is associated with thrombocytopenia (depletion of platelets) caused by an antibody that binds to platelets;’.’‘ however, the antibody only binds to platelets if quinidine is present (Gell and Coombs type I1 reaction). It is unknown how this antibody is induced. Since quinidine appears to be part of the structure to which the antibody binds, it seems likely that it acts as a hapten. However, yuinidine does not appear to bind irreversibly to platelets. It may be that the antibody is induced by a metabolite that covalently binds to platelets, and once antibodies have been induced, the antibody recognizes yuinidine loosely bound to platelets. Recent elegant studies have provided compelling evidence that halothane-induced hepatitis is due to an oxidative trifluoroacetyl halide metabolite which acts as a hapten. An antibody is found in the serum of patients that have halothaneinduced hepatitis which binds to trifluoroacetylated hepatic protein.i5.i6This antibody is not found in patients who have been treated with halothane but did not develop toxicity, nor is it found in patients with other types of hepatic disease. The major hepatic proteins that are haptenized have been isolated.” Assuming that this antibody is responsible for the hepatic injury, the mechanism by which it does so is unknown. One factor which makes this antibody-mediated mechanism especially interesting is that many other earlier animal studies suggested that halothane-induced hepatitis is due to a reductive pathway that leads to a reactive metabolite similar to the reductive pathway responsible for the toxicity of carbon tetrachloride. ’* Although this reductive pathway does appear

Volume 20, Issue 4


Toxicology to be responsible for toxicity in the animal model, the animals have to be pretreated with phenobarbital to increase this pathway, and in addition, the rats have to be exposed to halothane under hypoxic conditions to increase the reductive pathway. Even under these artificial conditions, the toxicity in the animal model is not as severe as seen in human halothane hepatitis. It now appears that the animal models of halothane toxicity are completely irrelevant to the mechanism of human halothaneinduced hepatitis. One observation that suggests that clinical halothane hepatitis involves an immunological mechanism is that toxicity is much more common in patients who have had a previous exposure to halothane, especially if it has been recent.’’ This is a common characteristic of an immune response. This example illustrates the danger of trying to reproduce a human idiosyncratic reaction in an animal model by changing some parameter to increase toxicity. Such a procedure may completely change the mechanism of toxicity. 2. Toxicity Mediated by Complement or Alterations

in the Complement System A possible example of the second type of mechanism is the idiosyncratic reaction seen with the radiographic contrast media. Although the syndrome resembles anaphylaxis, it does not appear to be mediated by IgE. The pathogenesis is not completely understood and may be multifactorial, however, the radiographic contrast media predispose to activation of the alternate complement pathway. 20-22 Such activation could explain many of the features associated with the idiosyncratic reactions associated with radiographic contrast media, and this may play an important role in their pathogenesis. A different type of interaction between drugs and the complement system was discovered by Dr. Sim. She found that drugs that are good nucleophiles (e.g., hydrazines) react with C4, and this leads to an increase in the concentration of circulating immune complexes .23,24 She proposes that this may be the mechanism by which some drugs induce 3. Toxicity Involving Changes in Leukocyte Function There are several drugs that alter leukocyte function. The best example is cyclosporin A which depresses helper T lymphocyte function. 26 This pharmacological effect has been used for the suppression of organ rejection in transplant patients, and it even appears to suppress the development of diabetes if used at an early stage of the disease. These are predictable pharmacological effects. Examples of hypersensitivity reactions involving this mechanism are somewhat more difficult to establish. One possible example is the hypersensitivity reactions induced by phenytoin. Phenytoin has been shown to cause a variety of alterations in lymphocyte populations and functions. In an early study, 63 patients on long-term phenytoin therapy were screened It was found for abnormalities in immunological f~nction.~’ that 2 1% had decreased IgA levels, 9% had a failure of antibody

response to Salmonella typhi antigen, 22% had an absence of delayed hypersensitivity to three common skin test antigens, and 27% had a depression of in vitro lymphocyte transformation by phytohemagglutinin. Aarli and Tonder made a similar observation that 25% of patients treated with phenytoin showed a decrease in IgA.28 Another study found a decrease in total lymphocytes and IgA in some patients treated with phenytoin, but they found no significant difference in the parameters of lymphocyte functions that were studied. 29.30 Polymorphonuclear leukocyte (PMN) function was studied by Ricevuti et al., who found impairment of PMN chemotaxis and superoxide generation in phenytoin-treated patients when compared with normal control^.^' Some effects on leukocyte function appear to be idiosyncratic. In one study, a decrease in the number of B lymphocytes and circulating immunoglobulin was detected in a patient after a phenytoin hypersensitivity reaction, and this had not resolved 3 years after the drug had been discontinued. 32 Lymphadenopathy is a relatively common occurrence in patients treated with phenytoin, although the exact incidence is not known.33 In some patients, phenytoin has been associated with a pseudolymphoma syndrome or even malignant lymphoma or multiple myeloma.34.35 Many of the pseudolymphoma patients had other manifestations of a hypersensitivity reaction such as fever, skin rash, arthralgias, and eosinophilia. It is unclear whether these effects are primary or simply manifestations of a Gel1 and Coombs type IV reaction.

II. PROPOSED MECHANISMS OF DRUG HYPERSENSITIVITY REACTIONS THAT INVOLVE LEUKOCYTE-GENERATED METABOLITES Having presented a general scheme of hypersensitivity reactions, an attempt is made to relate types of hypersensitivity reactions to this general scheme and to review data that suggest involvement of chemically reactive metabolites, especially the reactive metabolites generated by leukocytes.

A. Drug-Induced Lupus 1. Background Many drugs have been associated with the induction of a syndrome called lupus. 36 Because it usually requires several months of therapy before lupus is induced, it is unlikely that a drug would cause lupus unless it is used for a chronic illness. Lupus is an autoimmune disease characterized by antibodies that bind to “self”-antigens. In addition to lupus associated with drugs, there is a more common syndrome in which the precipitating cause is unknown and this is referred to as idiopathic lupus or systemic lupus erythematosus (SLE). 37 The idiopathic syndrome is usually more serious than the druginduced variety. In particular, the most serious manifestations involving the central nervous system and kidney are uncommon in drug-induced lupus. However, there is a large degree of 1990

215

____l__l_.___


~

Critical Reviews In

overlap between drug-induced lupus and idiopathic lupus, and it i s impossible to differentiate them in an individual patient solely on the basis of clinical manifestations. For example, renal involvement has been reported in drug-induced The less serious nature of drug-induced lupus may be due, in part, to its usual rapid resolution when the drug is discontinued. In contrast, there is no cure for idiopathic lupus. The mechanism of lupus in unknown, and it is unknown to what degree idiopathic and drug-induced lupus share a common mechanism." It has been suggested that idiopathic lupus is due to exposure to environmental agents such as arylamine~."~ Hydrazine has been associated with the development of lupus in one person,+' and canavanine (an amino acid analog found in legumes such as alfalfa sprouts) appears to cause a lupuslike syndrome in ~nonkeys.'~However. there is no compelling evidence that idiopathic lupus is commonly caused by environmental chemicals. 'The characteristic antibody used in the diagnosis of lupus is the antinuclear antibody (ANA). The specific nuclear antigen recognized by ANAs varies, but the one that appears to correlate best with disease activity is the ANA which recognizes doublestranded or native DNA." The specificity of ANAs is difficult to study because these antibodies are polyclonal, and many that bind to DNA cross-react with other, seemingly disparate, antigens such as cardiolipin. The ANAs in drug-induced lupus are more homogeneous, and most appear to bind to histone protein . 4 7 - 5 1 Several hypotheses have been proposed for the mechanism of idiopathic and drug-induced lupus and are reviewed clsewhere.42One appealing hypothesis for idiopathic lupus is that exposure to some bacterial or viral antigen acts as an tmmunogen and induces an antibody response in which the antibody cross-reacts with DNA. Here it is important to make the distinction between an immunogen and an antigen. An immunogen is the molecule that induces the synthesis of antibodies, while an antigen is any molecule that has a high affinity for or is "recognized" by the antibody. The antibody only recognizes a small part of an antigen (called the epitope), and therefore an immunogen can induce antibodies against many antigens which have very different structures than the immunogen. The finding that ANAs in idiopathic lupus bind strongly to a bacterial surface antigen supports the hypothesis that a bacteria is the source of the immunogen that induces anti-DNA antibodies.52.s' By analogy, drug-induced lupus could result from the drug or its metabolite acting as a hapten with the resulting immunogen being cross-reactive against histone protein. Although drug-induced lupus is associated with autoantibody production, it is unclear what role these antibodies play in the pathology of lupus. At least part of the pathology of idiopathic lupus, especially the kidney diesease, appears to be due to deposition of antigen-antibody complexes and activation of complement"' (Gel1 and Coombs type 111). Although kidney

216

disease and complement depletion is less common in druginduced lupus, as mentioned earlier, they have been reported. The antigen present in the antigen-antibody complexes of idiopathic lupus is usually native or double-stranded DNA. Since antibodies to native DNA are rare in drug-induced lupus, this difference in antigen present in the antigen-antibody complexes may be responsible for some of the differences between the two syndromes. It is likely that multiple types of immune reaction are involved. It is possible that lupus is a clinical syndrome that can involve different mechanisms in different people. Most of the drugs associated with a high incidence of druginduced lupus have an arylamine or hydrazine functional group (Figure l).36,42This structural feature may provide a clue to the mechanism of drug-induced lupus.

NH7

FIGURE 1. Structures of drugs that are arylamines or hydrazines and associated with the induction of a lupus-like syndrome. These are not the only drugs that have been associated with lupus, but this figure contains the structures of several associated with the highest incidence of this syndrome.

2. Procainamide-Induced Lupus Procainamide is a drug used to treat serious cardiac arrhythmias. Its chronic use is limited by a high incidence of drug-induced lupus. Of the drugs associated with lupus, procainamide induces the highest incidence. The incidence of ANA has been estimated at between 50 and 90% in patients on chronic procainamide the rap^;^^.^^ however, most of these patients are asymptomatic, and the incidence of clinically evident lupus is about 290/0.57This is a high incidence for a drug hypersensitivity reaction. As mentioned earlier, several other drugs associated with lupus also contain an arylamine functional group. In procainamide-induced lupus, it is clear that the arylamine functional group is necessary for the induction of lupus. The major metabolic pathway of procainamide is acetylation which blocks the a r y l a n ~ i n eThe . ~ ~ rate at which an individual acetylates procainamide is genetically determined, and patients who are of the rapid acetylator phenotype require, on average, a larger exposure to procainamide before they develop AN As and

Volume 20,Issue 4


Toxicology Furthermore, N-acetylprocainamidecan be used as a drug, and it does not appear to induce lupus, even though some of it is converted back to pr~cainamide.’**’~ In fact, Nacetylprocainamide can usually be used in patients who have had procainamide-induced 1upus.60,61 This is an example of a drug hypersensitivity reaction with a very definite dose response curve that lies within the therapeutic range of the drug. We set out to determine whether procainamide is metabolized to a reactive metabolite which could act as a hapten. We found that hepatic microsomes from both rats and humans oxidized the arylamine of procainamide to a hydroxylamine.62Some of the procainamide became covalently bound to the microsomal The binding was due to a metabolite rather than procainamide because NADPH was necessary, and binding was inhibited by inhibitors of cytochrome P-450. Substitution of the hydroxylamine metabolite for procainamide greatly increased covalent binding, and binding of the hydroxylamine did not require NADPH (in fact, NADPH significantly inhibited covalent binding of the hydroxylamine). However, it does not appear as if the hydroxylamine metabolite is responsible for most of the covalent binding of procainamide because glutathione blocks most of the covalent binding, and the hydroxylamine does not react with glutathione at a significant rate. The hydroxylamine metabolite is further oxidized nonenzymatically to a nitroso metabolite which is more reactive than the hydroxylamine; specifically, it reacts rapidly with glutathione. NADPH decreases the binding of the hydroxylamine because it reduces the nitroso metabolite back to the hydroxylamine. In addition to binding to hepatic microsomes, the nitroso metabolite binds to other proteins such as albumin and histone protein. The majority of the covalent binding to microsomal protein and albumin appears to involve sulfhydryl groups analogous to the reaction to glutathione. In contrast, the binding to histone protein appears to involve some other amino acid, but the identity of this amino acid is unknown at present. We found that the hydroxylamine metabolite (probably because of further oxidation to the nitroso metabolite) was toxic to lymphocyte^.^' In contrast, procainamide and other procainamide metabolites were not toxic. Therefore, it is possible that these reactive metabolites induce lupus by altering lymphocyte function. Although procainamide was metabolized in the liver to a reactive metabolite that covalently bound to protein, when a rat liver was perfused with procainamide, the hydroxylamine and nitroso metabolites could not be detected in the effluent of the liver.62When a liver was perfused with the hydroxylamine metabolite instead of procainamide, most of it was reduced back to procainamide. Thus, it does not appear as if these metabolites would reach the circulation in any significant concentration. Next we set out to determine if procainamide could be metabolized by leukocytes. The first studies did not detect metabolism by human peripheral mononuclear leukocytes (a

mixture of lymphocytes and monocytes). However, we found that procainamide was oxidized by activated human neutrophils to the same hydroxylamine metabolite that was formed in the liver.66The nitro derivative was also detected. and the nitroso metabolite was the presumed intermediate. This metabolism only occurred if the neutrophils were activated so that a respiratory burst occurred. The experiments with mononuclear leukocytes were repeated, but this time the cells were activated with a phorbol ester. Under these conditions, metabolism of procainamide to the hydroxylamine by peripheral mononuclear leukocytes was observed. When neutrophils or monocytes are activated, they undergo what is referred to as a respiratory b u r ~ t .Many ~ ~ -stimuli ~ ~ are capable of activating leukocytes including bacteria, soluble chemotactic peptides (e.g., formyl-L-methionyl-L-leucyl-Lphenylalanine), and miscellaneous agents (e.g., phorbol esters). When the cells are activated, a membrane-bound enzyme, NADPH oxidase, is activated. NADPH oxidase converts molecular oxygen to superoxide radical on the surface of the leukocyte, or on the inside of a phagosome if the cell is activated by the process of phagocytosis. The superoxide is converted to hydrogen peroxide either by the action of superoxide dismutase or spontaneously. In addition, activation leads to the release of the enzyme myeloperoxidase (MPO) from granules. When MPO or some other peroxidase combines with hydrogen peroxide, the peroxidase is converted to a strong oxidant called compound I.’O When chloride is present, hypochlorite, or an oxidant very similar to hypochlorite, is also generated.71This process is a major mechanism for killing bacteria, but it also generates reactive oxygen species which can cause tissue damage and inflammati~n,~~.’~ and it appears to be capable of oxidizing drugs and other x e n o b i o t i c ~ . ~ ~ - ’ ~ When procainamide was incubated with MPO, the hydroxylamine metabolite was again generated.%In addition, in the presence of chloride ion, reactive N-chloroprocainamide was also produced.77 The N-chloroprocainamide could be detected by HPLC but it was too reactive to isolate. Nchloroprocainamide spontaneously rearranges to uchloroprocainamide which can be isolated. It is likely that the reason the N-chloroprocainamide was not observed when procainamide was incubated with activated neutrophils was that it reacted too rapidly with the neutrophils to be detected. The metabolism of procainamide by activated leukocytes or catalyzed by MPO is summarized in Figure 2. If the reactive metabolites of procainamide induce lupus by acting as a hapten to induce autoantibodies, what is the protein that procainamide reacts with to produce the immunogen? Since most of the ANAs induced by procainamide bind to histone protein, histone protein could be the target protein. We have demonstrated that the nitroso metabolite of procainamide does bind to histone protein, but the nitroso metabolite appears to bind to many different proteins,@ so what is unique about histone protein? It may be that the immunogen (the molecule

1990

21 7

--

Critical Reviews In

___


NO

FIGURE 2. 2Metabolisrn of procainamide by activated neutrophils or riiononuclear cells o r catalyzed by myeioperoxidase.

that induced the formation of antibodies) is very different from the antigen (the molecule to which the antibodies bind). This would he analogous to the proposed mechanism for idiopathic lupus described earlier in which a bacterial cell wall protein was the immunogen that induced the formation of antibodies that bound to DNA. It is thought that the reason that the immunogen and antigen can be so different is that, as mentioned earlier, for the immunogen to induce antibody formation it must be taken up by a macrophage, processed, and presented along with the class I1 major histocompatibility antigen (MHC 11) to T lymphocytes. A major part of the processing appears to involve hydrolysis of the immunogen to smaller polypeptides. It is further speculated that, when one of these polypeptides binds to the MHC 11, it is transported to the cell surface for presentation and saved from further hydroly~is.'~.'~ This may explain why the tertiary structure of a protein does not usually appear to have any effect on the specificity of antibodies that a protein induces. The epitope (the actual part of the antigen that is recognized by the antibody) is quite small, and so it is not surprising that two large molecules would share a region that is very similar or even identical. Monocytes are macrophages and therefore are involved in the processing and presentation of immunogens. Since they can also metaboliz,e procainamide to reactive metabolites, they are likely to be involved in the mechanism of procainamideinduced lupus. One possible mechanism by which this could occur is that a reactive metabolite could covalently bind directly to the MHC I1 and bypass the processing step. Others have speculated that the direct binding of a hapten to the MHC 11 would be a strong stimulus to the immune system." Since the reactive metabolite is formed on the surface of the monocyte, it is likely to bind to the MHC 11. This hypothesis still does not explain what is special about histone protein and why most antibodies in procainamideinduced lupus should bind to it. One property of histone protein is that, like DNA. it is found in the nucleus and it is usually 218

hidden from the immune system. It is easier to break immunological tolerance to such sequestered antigens than it is to autoantigens that come into direct contact with the immune s y ~ t e m . 'Another ~ special property of histone protein is that it is a very basic protein with many positve charges. These positve charges neutralize the many negative charges of the phosphate groups on DNA. MPO is also a very basic protein with a PI of about 11" and binds to DNA." Since MPO appears to be responsible for the production of reactive metabolites of procainamide, it should also be exposed to a high concentration of the reactive metabolite and become haptenized to a relatively high degree. If haptenized MPO is the immunogen, it is not surprising that it would induce antibodies that cross-react with histone protein. This hypothesis is presented graphically in Figure 3. In support of this hypothesis, it has been observed that ANAs that bind to neutrophils appear to correlate better with symptoms of lupus than ANAs that bind to all cell nuclei.82 It turns out that the nucleus of neutrophils (and to a lesser extent that of some other leukocytes) contains a small amount of MPO.'' Furthermore, in other studies of anti-MPO antibodies, they "stain" neutrophil nuclei just like other ANAs without staining the cyt~plasm.'~ It is surprising that the cytoplasm is not stained because it contains the bulk of the MPO; however, it may be that antibody cannot penetrate the granules that contain the MPO, or it may be that the cells degranulate during fixation of the cells. There are also other basic proteins in the leukocyte granule^^'**^ that could be haptenized by reactive metabolites and then act as immunogens to induce antihistone ANAs. 3. Other Arylamines Associated with Drug-Induced

Lupus Although drugs with an arylamine functional group are relatively uncommon (probably because of the high incidence of adverse reactions associated with this functional group), there are several other drugs that are arylamines. The most common is the sulfonamide group of antibacterial agents. The sulfonamides are more often associated with generalized hypersensitivity reactions than drug-induced lupus, but they do cause drug-induced lupus. In fact, the first report of druginduced lupus appears to have been due to a sulfonamide.85 We have found that sulfadiazine is metabolized by activated neutrophils to a hydroxylamine.86 Sulfamethoxazole is also metabolized by activated neutrophils, but the metabolism is more complex and it appears as if N-chlorination may be the major pathway.87It is likely that the mechanism of sulfonamideinduced lupus is similar to that of procainamide. Other arylamines associated with drug-induced lupus are nomifensine and aminoglutethimide. Nomifensine is an antidepressant which appeared to be safer than the more common tricyclic antidepressants when taken in an overdose. However, it was withdrawn from the market because of a high incidence of hemolytic anemia.*' It was also reported to cause druginduced lupus.x9 Aminoglutethimide is also associated with a

Volume 20,Issue 4

Toxicology

Activation

+

Osc

N-CH*CHfl(C&Is)


IH

-nPA-MPO

. -

Anti-PA-ProMPO

Monocyte

recognize PA-ProMPO and similar epitopes on other proteins f

FIGURE 3. Proposed mechanism by which induction of an antibody against haptenized MPO could lead to antibodies that cross-react with histone protein because of a similar positively charged epitope (MPO = myeloperoxidase; PA-ProMFQ = procainamide adduct of processed MPO; PAMFQ = pmainamide-MPO adduct; MHC I1 = class I1 major histocompatability antigen).

high incidence of hypersensitivity reactions including druginduced lupus.goPractol01~~-~~ and a c e b ~ t o l o lare ~ * 6-blockers ~~ which have been linked to drug-induced lupus. Although there are case reports linking other @blockers such as propranolol to a lupus-like syndrome,%when h s was carefully investigated, only practolol and acebutolol were associated with a significant increase in the presence of AN AS.^^.^^ These two drugs are not arylamines, but they are amides which are hydrolyzed in vivo to arylamines, and they are the only two P-blockers that are amides of arylamines. In addition, practolol is associated with a serious oculocutaneous syndrome which led to its withdrawal from the market.99-10'Sulfasalazine, which is metabolized to aminosalicylic acid and sulfapyridine, is also associated with drug-induced lupus. 102~103The formation of arylamines from these drugs is shown in Figure 4. There are two drugs that are arylamines which do not appear to be associated with drug-induced lupus. One of these drugs is dapsone. It is oxidized to a hydroxylamine by activated leukocytes104 and causes several different types of

hypersensitivity reactions; yet, with the exception of one case report of discoid Iupus,'05 it does not appear to cause lupus. One difference between the oxidation of dapsone and procainamide is that dapsone does not appear to be N-chlorinated by MPO. lo4 Metoclopramide is an arylamine with a structure similar to that of procainamide. It has not been reported to induce lupus, but it is given in much lower doses than procainamide because of severe CNS side-effects associated with a high dose. If procainamide were given in the same dose as metoclopramide, it is unlikely that it would cause druginduced lupus because N-acetylprocainamide does not induce lupus even though a significant amount is converted to procainamide. Metoclopramide is associated with other types of hypersensitivity reactions that are discussed later. 4. Lupus Caused by Drugs with a Hydrazine

Functional Group The hydrazine functional group is also associated with druginduced lupus. The best example is hydralazine.'" The incidence 1990

219

-_

Critical Reviews In

I


OCHzCHOHCH,NHCH(CH,)z

Q.

CCH,

OCH&HOHCH,NHCH(CH&

i'l

b"02"

T

Q

__t

0=S=0 I

6

(5GGzG-I

hydralazine-induced lupus could also be initiated by a reactive metabolite generated by activated leukocytes such as monocytes. Isoniazid is also a hydrazine derivative and is associated with the induction of lupus. Furthermore, isoniazid is In metabolized by activated leukocytes and MPO/H,O,/CIthis case, the product is isonicotinic acid which may have been produced by a reactive intermediate. However, we have been unable to isolate reactive metabolites of hydrazines as we have done with the arylamines.

5. Lupus Caused by Drugs with a Sulfhydryl or Thiono Sulfur Group Propylthiouracil is a drug used to treat hyperthyroidism. The mechanism appears to involve formation of a reactive metabolite by thyroid peroxidase which inhibits the enzyme and further production of thyroxine. 1 i 2 , 1 i 3 MPO is similar to thyroid peroxidase and propylthiouracil has also been observed to inhibit neutrophil and monocyte chemiluminescence, probably by interaction with MPO.IL4 Therefore, it is likely that propylthiouracil is oxidized to a reactive metabolite by MPOi H,O,/Cl- or activated leukocytes. This is indeed the case. We and others have found that one metabolite is a sulfonic acid.1i5."6 We have found that the sulfonic acid is chemically reactive; specifically, it reacts with glutathione and other sulfhydrylcontaining compounds. Furthermore, it is likely that a sulfenyl chloride is an early intermediate in this reaction, and other intermediates such as thioesters are also reactive. Thus the oxidation of propylthiouracil by MPO appears to produce several reactive intermediates as shown in Figure 5.

FIGURE 4. Umgs that are readily metabolized to arylamines.

of hydralazine-induced lupus is lower than that of procainamide but is still high (about 10%).Like procainamide-induced lupus, the incidence is dose dependent and increases significantly if the daily dose is greater than 200 rng.'" Furthermore, the incidence of hydralazine-induced lupus is more sensitive to acetylator phenotype than is procainamide-induced lupus, and hydralazine-induced lupus is rare in patients of the rapid acetylator phenotype. 3 6 . 1 0 x This is presumably because acetylation makes a larger contribution to the elimination of hydralazine than it does to the elimination of procainamide. Procainamide elimination is dominated by excretion of unchanged drug in the urine. This also suggests that the free hydrazine group is necessary for the induction of lupus by hydralazine. Like arylamines, hydrazines are readily oxidized to reactive metabolites. We have found that hydralazine is metabolized to phthalazinone and phthalazine by activated neutrophils and MPO/H,O,/CI .'"' It has been suggested by others that phthalazinone is the product of a reactive intermediate, and the production of phthalazinone has been linked to the induction of lupus by hydralazine.Il0Thus, although the chemistry is somewhat different, the mechanism of ~

220

r

n

1

FIGURE 5. Metabolism of propylthiouracil by activated leukocytes.

Propylthiouracil has been associated with the induction of lupus. '-1 l 9 Although the incidence appears to be relatively low in humans, propylthiouracil-induced lupus in the cat appears to be the only known animal model of drug-induced lupus and was reported to occur in 50% of treated cats.'20,121 Propylthiouracil is associated with other types of drug hypersensitivity reactions which are discussed later. Other drugs with sulfhydryl or thiono sulfur groups are also associated with drug-induced lupus as well as other types of drug hypersensitivity reactions. Examples include methimazole, ' 19. '22.'23 captopril,124*125 and penicillamine. 126-128

Volume 20, issue 4

Toxicology


6. Lupus Associated with Hydantoins and Related Drugs Phenytoin is an anticonvulsant which is associated with a wide range of hypersensitivity reactions. It appears to modify lymphocyte function as mentioned earlier. It has also been linked with the induction of I u ~ u s . ' ~ ~ - ' ~ ~ There is evidence to suggest that the toxicity of phenytoin, particularly teratogenicity, is related to a reactive arene oxide metabolite. Although this arene oxide has never been isolated, one piece of data to support its existence is the NIH shift observed in the production of a phenol metab01ite.I~~ The presence of large NIH shifts has been taken as evidence of an arene oxide intermediate. However, Ortiz de Montellano has suggested mechanisms of oxidation of an aromatic ring that result in a phenol metabolite and a large NIH shift but do not involve an arene oxide intermediate.'34Other evidence for an arene oxide intermediate consists of covalent binding of phenytoin that appeared to parallel the production of the phenol. I J 5 In addition, metabolite-mediated phenytoin cytotoxicity was potentiated by trichloropropyleneoxide which inhibits epoxide hydrolase.136 A related drug, mephenytoin, is also teratogenic and also It is causes several types of hypersensitivity reactions. also metabolized to a phenol metabolite, and an arene oxide is the presumed inte~mediate.'~~ Unlike phenytoin, mephenytoin is chiral and only the S enantiomer is oxidized to a phenol. If the arene oxide hypothesis is correct, the S enantiomer should be toxic and the R enantiomer should not be toxic. When this was tested in an animal model using fetal toxicity as an endpoint, it was found that the R enantiomer was actually more toxic than the S enantiomer.'40 Because the phenyl ring of the R enantiomer is not oxidized, the half-life of the R enantiomer is longer than that of the S, and more of it is metabolized by N-demethylation. When the N-demethylated metabolite was tested, it was found to be more toxic than either of the two parent enantiomers. This suggests that the nitrogen may be involved in the toxicity. The metabolism and toxicity of the two stereoisomers are contrasted in Figure 6. Another observation which is difficult to explain with the arene oxide hypothesis is that other drugs with a similar heterocyclic ring, such as ethosuximide and trimethadione, do not have an aromatic ring, and yet they have similar toxicities to that seen with In particular, they are phenytoin and mephenytoin. associated with drug-induced lupus.'Z9-'31~'43~1u Although the nitrogen in trimethadione is methylated, it is readily demethylated in vivo and, with chronic therapy, this metabolite is responsible for most of the anticonvulsant activity.145 The structures of these anticonvulsants are shown in Figure 7. We studied the metabolism of phenytoin by activated neutrophils and MPO/H,O,/Cl- to determine if this could have any relationship to its toxicity. Because of the data from the mephenytoin study, we were especially interested in reactive metabolites which involved oxidation of the hydantoin nitrogens.

N-Dernethylation

I

H

more toxic than

FIGURE 6. Comparison of the metabolism and toxicity of the enantiomers of mephenytoin.

I

1373138

1413142

H

[Ethosuximidel y-43

I

CH3

CH3 I

I

CH3

FIGURE 7. Other anticonvulsants with a similar heterocyclic ring and toxicity as phenytoin but without a phenyl ring that could be oxidized to an arene oxide.

We found that phenytoin was chlorinated by MPO/HZ02/CIto N,N'-dichlorophenytoin (Figure 8).146This metabolite was not observed when phenytoin was oxidized by activated neutrophils, but the N,N'-dichlorophenytoin was very reactive. Specifically, synthetic N,N'-dichlorophenytoin reacted rapidly with neutrophils. Although, N,N'-dichlorophenytoin was not observed in incubations of phenytoin with neutrophils, we did observe covalent binding of radiolabeled phenytoin to neutrophils that was increased by activation of the cells. Furthermore, covalent binding of phenytoin to albumin catalyzed by MPO was increased by the presence of chloride ion. This

1990

221

Critical Reviews In

I _

li


)-(

Cl

IN.N‘-Dichloror)henvtoin

1

FIGURE 8. Chlonnation of phenytoin catalyzed by myeloperoxidase

suggest< the ,V-chIonnation of phenytoin was responsible for tt5 covalent binding This provides a possible mechanism for phenytoin toxicity as well a5 for similar drugs, that does not involve an arene oxide metabolite 7. Other Types of Drugs That Cause Lupus

There are many other drugs that have been associated with the induction of lupus. In many cases, the association is based on one or two case reports and it is difficult to know if they really cause lupus. In some cases, the clinical picture is significantly different from that seen with procainamide or hydralazine. In the case of carbamazepine. there is a report that the incidence of ANA is 7 8 8 with chronic therapy;iJ7yet, i i appears as if the incidence of clinically manifest lupus is relatively low. i 4 g . i F ” Carbamazepine is an anticonvulsant that is associated with several different types of hypersensitivity reactions. 1 5 1 - 1 s yWe have found that carbamazepine is oxidized to several metabolites by MPO/H20,/Cl-, 8’ Most of these metabolites are derivatives of iminostilbene, but we have yet to identify many of them. We speculate that even the production of iminostilbene involves a reactive intermediate. The number of metabolites observed when carbamazepine is oxidized by activated neutrophils is less than the number observed with MPO/H,02/CI-. This may be analogous to the metabolism of phenytoin where the reactive metabolite was not observed in neutrophil incubations, presumably because it reacted with the cells.

8. Drug-InducedAgranulocytosis and Aplastic Anemia 1. Background A relatively common type of serious adverse drug reaction involves the bone marrow or mature blood cells.’6o If all of the blood cell lines are depleted (i.e., leukocytes, erythrocytes, and platelets), it is known as aplastic anemia. This condition carries a very high mortality rate. In other cases, there can be selective depletion of only one cell type. If there is peripheral destruction of red cells, it is called hemolytic anemia; if it involves red cell precursors, it is referred to as pure red cell aplasia. Depletion of platelets is known as thrombocytopenia. When granulocytes or polymorphonuclear leukocytes are depleted, it is called agranulocytosis. These cells include neutrophils, eosinophils, and basophils; the most abundant cell

222

being the neutrophils. If only the neutrophils are depleted, it is referred to as neutropenia. These conditions carry a high mortality rate because a patient without neutrophils has a very high risk of infection that cannot be controlled with antibiotics. Pisciotta has classified drug-induced agranulocytosis into three mechanistic categories: antibody-mediated peripheral destruction of neutrophils, agranulocytosis associated with a lupus-like syndrome, and toxic inhibition of bone marrow by phenothiazines. ”’ The category associated with drug-induced lupus is rather artificial because it does not provide much information about the mechanism, and drugs such as procainamide often induce agranulocytosis in the absence of ANA.162-1W it may be that the only association between drugs that cause lupus and agranulocytosis is that drugs in this category are readily oxidized to reactive metabolites by activated leukocytes, and depending on the cell type, monocytes or neutrophils, this could lead to either lupus or agranulocytosis, respectively. Other investigators have described many cases of agranulocytosis as being due to “maturation arrest” based on the appearance of the bone marrow where there is an excess of early forms of cells and a relative lack of mature ~ e 1 l s . l ~ ~ However, the same picture would be expected if the drug damaged neutrophil precursors at a specific stage of development, or if the bone marrow was starting to recover and there was a “wave” of cells going through stages of differentiation together. Other investigators have not found the picture of maturation arrest to be common in drug-induced agranulocytosis. Another reasonable classification would be to differentiate peripheral destruction of mature cells from injury to the bone marrow. Peripheral destruction usually appears to be mediated by antibodies. Bone marrow injury could be subdivided into antibody-mediated damage and “toxic” damage. Although antibody-mediated agranulocytosis appears to be relatively common, it is much more difficult to demonstrate antineutrophil antibodies and prove that they are responsible for agranulocytosis than it is to demonstrate antibodies against red cells and demonstrate that they are responsible for hemolytic Therefore, failure to demonstrate antineutrophil antibodies should not be taken to represent strong evidence against an antibody-mediated mechanism. Conversely, the finding of such antibodies in a case of drug-induced agranulocytosis does not prove that they are responsible for the destruction of neutrophils. The clearest example of a drug that causes antibody-mediated peripheral destruction is aminopyrine. 7,167 The clinical picture is usually dominated by the onset of an infection. The bone marrow can be hyperplastic, reflecting an increase in the production of neutrophils in an attempt to keep up with the peripheral destruction, but in some cases, there is a decrease in the myeloid series. Reexposure to the drug, even at a very low dose. results in an almost immediate onset of fever and

Volume 20,Issue 4

Toxicology


destruction of neutrophils. Antibodies are found in the serum that bind to neutrophils, and acute serum from the patient will destroy neutrophils from normal subjects. This is presumably an example of a Gel1 and Coombs type I1 reaction. The mechanism of most drug-induced agranulocytoses is not as clear.

2. Procainamide-Induced Agranulocytosis There have been many reports of procainamide-induced agranulocytosis.162-161,168-174 The incidence of agranulocytosis appeared to increase when a sustained release form was introduced. One report estimated the incidence of agranulocytosis at 4% with sustained release procainamide, 16* but other reports indicated a much lower incidence of about 0.5%.163 Rocainamide-induced agranulocytosis usually appears to be due to a decreased production of cells in the bone marrow, but myeloid hyperplasia has also been described.'61*'62The decreased production of neutrophils in the bone marrow could be due to direct toxicity of procainamide to bone marrow cells, or it could be due to an antibody which leads to the destruction of granulocyte precursors. The hydroxylamine metabolite of procainamide is toxic to other leukocytes, and so it is likely to be toxic to bone marrow cells.65It is presumed that the hydroxylamine and other reactive metabolites would be produced in the bone marrow if the cells in the bone marrow were activated. Not only could mature neutrophils in the bone marrow generate these metabolites, but other leukocytes, such as monocytes and some immature cells, also contain myeloperoxidase and can generate hydrogen peroxide. Although direct toxicity of the reactive metabolites is a possible explanation, it does not explain why there is a delay between initiation of procainamide therapy and the development of agranulocytosis, or why procainamide does not cause agranulocytosis in animals or in most patients treated with the drug. These characteristics may be due to the lack of cell activation in most patients and animals, and an infection or other inflammatory condition may be a risk factor for procainamide-induced agranulocytosis . The report suggesting that sustained release procainamide was associated with a 4% incidence of agranulocytosis was based on a group of patients who received procainamide for arrhythmias after cardiac surgery. Such major surgery would certainly be expected to lead to a generalized activation of neutrophils. The report estimating the incidence of procainamide-induced agranulocytosis to be only 0.5% did not involve postoperative patients. Another observation more difficult to explain is that in most patients granulocytes are selectively affected, while in others, pancytopenia occurs.175 It would seem likely that if the mechanism involved direct toxicity the reactive metabolites would either be toxic to all of the cell lines or they would not, but it would be the same in all patients. This difference in

selectivity in different patients might be explained by the induction of specific antibodies and that specificity could vary from one patient to another. If procainamide-induced agranulocytosis is mediated by an antibody, it is likely that the antibody recognizes an antigen specific to a neutmphil precursor because destruction of mature neutrophils is not usually observed. Consistent with the antibody hypothesis is a report of two patients with procainamide-induced agranulocytosis who had cytotoxic antibodies against myeloid cells. Reactive metabolites generated by activated leukocytes near the surface of neutrophil precursors may be the stimulus which induces the formation of such antibodies. F'rocainamide has also been reported to cause an antibody-mediated hemolytic anemia, and this was not associated with the presence of ANA.'76 3. Dapsone-Induced Agranulocytosis Agranulocytosis associated with dapsone is usually presented as a case report because of its infrequent use in developed c o ~ n t r i e s . ' ~However, ~ - ' ~ ~ when it was used for malaria prophylaxis in Vietnam, it was responsible for 16 cases of agranulocytosis with 8 deaths, and its routine use was discontinued.IE1Like procainamide-induced agranulocytosis, dapsone-induced agranulocytosis usually appears to be due to decreased production of neutrophils in the bone marrow or maturation arrest rather than to peripheral destruction. The oxidation of dapsone by activated neutrophils and MPO/ H,O,/CI- is very similar to that of procainamide except that, as mentioned earlier, N-chlorination of dapsone was not detected.Io4 Thus, like procainamide, activation of leukocytes in the bone marrow could lead to the metabolism of dapsone to reactive metabolites that could, in turn, directly lead to destruction of neutrophil precursors. The hydroxylamine metabolite of dapsone has been reported to be toxic to bone marrow. Alternatively, the reactive metabolites could induce the production of an antibody that could cause the destruction of neutrophil precursors.

4. Agranulocytosis Associated with Sulfonamides Sulfonamides have long been known to cause agranulocytosis. In many surveys designed to determine the most common causes of agranulocytosis, sulfonamides often top the list. 184,185 However, sulfonamides (usually sulfamethoxazole)are commonly combined with trimethoprim, and in some cases toxicity may be due to trimethopnmlE6or to a synergistic interaction between the sulfonamide and trimethoprim rather than to the sulfonamide. Another complicating factor is that two or more mechanisms may be involved in sulfonamide-induced agranulocytosis and granulocytopenia. A more recent study indicated that 34% of young patients on sulfamethoxazole/trimethoprim therapy develop a significant degree of neutropenia.lE7 They consider the most likely cause for this neutropenia to be the effect of this combination on folate metabolism. In vitro,

1990

223


_ I

Critical Reviews In

sulfarnethoxazoleitrimethoprim is toxic to neutrophils and this toxicity is prevented by adding folinic acid. However, other cases of agranulocytosis are clearly part of a hypersensitivity reaction, and drug-dependent leukocyte agglutinins (antineutrophil antibodies that can cause their agglutination) have been described in agranulocytosis due to sulfapyridine, sulfathiazole, and sulfafurazole. 14-174 It is not clear how often sulfonamides are responsible for hypersensitivity reactions that only involve the bone marrow and do not involve trimethoprim. The portion of sulfonamide-induced agranulocytosis that is part of a hypersensitivity reaction would also fit into the pattern of other arylamines in which the initiating event could be the generation of a reactive metabolite by neutrophils, or other cells in the bone marrow, which contain myeloperoxidase. Even the cases where agranulocytosis appears to be due to trimethoprim rather than sulfamethoxazole could involve a similar mechanism. Trimethopnm is a heterocyclic aromatic amine which may be metabolized to similar reactive metabolites by MPO/H,O,/Cl .

5. Agranulocytosis Caused by Other Arylamines Most other arylamines are associated with a relatively high incidence of agranulocytosis. Aminoglutethimide is associated with a high incidence (about 1 %) of agranulocytosis. 188-1vo Metoclopramide has been reported to cause agranulocytosis even though it is given at a much lower dose than procainamide. 1 9 1 . 1 9 2 4-Aminosalicylic acid193 and sulfasalaZine,194-'96 which is metabolized to sulfapyridine and 5-aminosalicylic acid, are also associated with drug-induced agranulocytosis. Drug-dependent leukocyte agglutinins have been observed in cases of agranulocytosis due to p aminosalicylic acid. l 4 6. Chloramphenicol-InducedAplastic Anemia Chloramphenicol is classically the drug associated with aplastic anemia, and it can also cause agranulocytosis.2 This is one hypersensitivity reaction which is not very sensitive to dose because even chloramphenicol eye drops have been reported to cause aplastic anemia.'91 Chloramphenicol is not an arylamine, but it is an arylnitro compound. Furthermore, the nitro group appears to be involved in the induction of aplastic anemia because its replacement by a methylsulfone leads to a drug (thiamphenicol) which retains most of the antibacterial effects of chloramphenicol, but does not appear to be associated with aplastic anemia. '9R.199The nitro group of chloramphenicol is reduced to an arylamine by anaerobic bacteria in the gut.*"" It has been demonstrated that this arylamine can be oxidized by the liver to hydroxylamine and nitroso which are very toxic to bone marrow cells.202~203 However, it was found that these metabolites were too reactive to get from the liver to the bone marrow.201Gross et al. failed to demonstrate any metabolism of the m i n e metabolite of chloramphenicol by bone marrow cells. 203 Thus, even though the nitroso metabolite was toxic to bone marrow cells, it was 224

doubted that it played any role in the induction of aplastic anemia. However, in the study by Gross et al., the bone marrow cells were not activated, and we have found that MPO/H,O,/ C1- does oxidize the amine metabolite of chloramphenicol. The major metabolite appears to involve N-chlorination. 87 In addition, Yunis et al. have found that another toxic bacterial metabolite of chloramphenicol, which they refer to as dehydrochloramphenicol (the benzylic alcohol is oxidized to a keto group), is more readily reduced than chloramphenicol, and there is circumstantial evidence that it can be reduced to a toxic nitroso metabolite by bone marrow cells.204~20s Thus, the same type of toxic metabolite may be formed in the bone marrow, either from the oxidation of an arylamine metabolite or from the reduction of the nitro group of a different bacterial metabolite of chloramphenicol. These two pathways are shown in Figure 9. Either process could explain the relatively infrequent occurrence of chloramphenicol-induced aplastic anemia because it would require both the appropriate bacteria to form an intermediate metabolite and then, in one case, the activation of neutrophils so that the final toxic metabolite would be formed. This mechanism does not explain the long delay which sometimes occurs between the end of therapy and the development of aplastic anemia - sometimes a delay of 2 months or more. Since precursor cells are involved, some of the delay could result from the maturation of immature cells that have not been effected, but the delay is often longer than can be explained on that basis. This could be an indication that chloramphenicol-induced aplastic anemia is also mediated by an antibody. As before, this antibody may be induced by the reactive metabolites described earlier. Unfortunately, at this point it is unknown how much of either of the intermediate H."CI y

2

0 I

Bone ____)

Marrow

H-C-OH I H-C-CHZOH

H-C-OH I

6Bone

H-CZO I H-C-CHzOH

Marrow

I

H"

-C -CHCI,

d'

H-CZO I

H-C-CHZOH I H"-C-CHCI,

d'

FIGURE 9. Possible mechanisms of forming toxic metabolites of chloramphenicol that could cause aplastic anemia.

Volume 20.Issue 4

Toxicology


bacterial metabolites are formed in patients and that makes it more difficult to guess which is most likely to be responsible for toxicity. 7. Agranulocytosis Caused by Propylthiouracil and Other Sulfur-ContainingDrugs The major serious adverse reaction associated with propylthiouracil therapy is agranulocytosis, and propylthiouracil is one of the most common causes of agranulocytosis.184m In contrast to most of the agranulocytoses associated with arylamines, propylthiouracil-induced agranulocytosis appears to be mediated by an antibody that causes peripheral destruction of mature neutrophils. However, the bone marrow often shows a depletion of neutrophil precursors. The conversion of propylthiouracil to reactive metabolites near the cell membrane of activated neutrophils is a likely initial event in the induction of these antibodies.IL6 Other thiono sulfur- and sulfhydryl-containingdrugs include methimazole, captopril, and penicillamine. These drugs are also associated with a relatively high incidence of agranulocytosis. ’28*21 Metiamide, the first H2-antagonist, had an unacceptable incidence of agranulocytosis that was believed to be due to the thiono ~ u l f u r . ~The ~ ~structure * ~ ‘ ~ was modified to produce cimetidine which is not associated with a significant incidence of agranulocytosis. 1189174,206-210

initiate an immunological reaction whose histological picture could resemble a direct toxic reaction. Agranulocytosis associated with aminopyrine was described earlier. Dipyrone is associated with a high incidence of agranulocytosis similar to that seen with a m i n ~ p y r i n eand ,~~~ even antipyrine has been reported to cause agranulocytosis.226 Although there is a significant difference in the structure of these agents, and metabolism by neutrophils may be somewhat different, it is likely that a reactive metabolite is generated in each case.

10. Agranulocytosis Associated with Carbamazepine and Related Drugs Early clinical studies with carbamazepine suggested that it was associated with a very high incidence of aplastic anemia. Although the incidence is significantly lower that first feared, carbamazepine-induced agranulocytosis is a significant problem,151-153,156,159,227 As mentioned earlier, we have found that carbamazepine is oxidized to several metabolites by activated neutrophils or MPO/H,O,/Cl- , and we speculate that a reactive metabolite is formed. M i a n ~ e r i n ~and ~ * -cloza~~~ pine231-233 have structures similar to that of carbamazepine,and they also appear to be associated with a relatively high incidence of agranulocytosis.

11. Other Drugs Associated with Agranulocytosis Several other drugs have been associated with agranulocytosis, in general with an incidence lower than the Related Drugs ones listed. The hydrazines, hydralazine and isoniazid, have Mild neutropenia is relatively common with phenytoin, but agranulocytosis or aplastic anemia are u n c o m m ~ n In . ~ ~ ~ been ~ ~ ~reported ~ to cause agranulocytosis; however, despite the high incidence of drug-induced lupus associated with these contrast, agranulocytosis is a serious problem with mephenytoin and is the major reason for its lack of common use as an drugs, they are not associated with a high incidence of anticonvulsant. Agranulocytosis also represents a agranulocytosis. Leukocyte agglutinins have been reported in significant toxicity for trimethadione and ethosuximide.221A association with hydralazine-induced agranulocytosis.l4 case of phenytoin-induced pure red cell aplasia has also been reported.222 Since we have demonstrated that phenytoin is C. Generalized Drug Hypersensitivity Reactions oxidized to a reactive metabolite by MPO/H,O,/Cl-, it is likely 1. Sulfonamide Reactions that these related drugs are oxidized in a similar manner.’46 Sulfonamides are a major cause of idiosyncratic drug reactions.z34The exact incidence is unknown. The clinical 9. Agranulocytosis Associated with Phenylbutazone manifestations vary widely from one patient to another. In most patients the skin is involved, usually as a nonspecific rash or and Other Pyrazolone Derivatives urticaria; however, sulfonamides are a common cause of toxic Phenylbutazone is associated with an unacceptable risk of epidermal necrolysis.235 In some patients, the involvement is aplastic anemia and agranulocytosis, especially in the elderly, widespread with fever and various organ involvement such as and it is rarely used t ~ d a y . ~The ~ bone ~ , marrow ~ ~ ~ in , ~these ~ ~ the liver, lungs, and kidneys. cases is usually hypoplastic suggesting a “toxic” reaction. On When the skin involvement is in the form of urticaria, the the other hand, there is also evidence of an immune-mediated mechanism presumably involves an IgE-mediated agranulocytosis associated with phenylb~tazone.~~~ hypersensitivity reaction (typeI). The more generalized reactions Ichihara et al. have demonstrated that phenylbutazone is also have the characteristics common to hypersensitivity oxidized to a hydroperoxide, an alcohol, and it is also chlorinated reactions described in the introduction, but the details of the in the same po~ition.’~ Although these metabolites do not appear mechanisms are unknown. As with procainamide-induced lupus, to have the same intermediate, it appears that a free radical the free arylamine functional group appears to be necessary intermediate is involved. As before, this reactive intermediate and the slow acetylator phenotype is a definite risk factor for could cause direct toxicity to bone marrow cells, or it could 8. Agranulocytosis Associated with Hydantoins and

1373221

1990

225


-

Critical Reviews In

sulfonamide reactions. ’” Analogous to phenytoin hypersensitivity reactions, we have shown that lymphocytes from patients that have had a sulfonamide hypersensitivity reaction are more sensitive to a metabolite of sulfamethoxazole than lymphocytes from normal However, in this case we have synthesized the hydroxylamine, and we use it directly rather than use the parent drug and hepatic microsomes to form the metabolite in situ. As expected, the concentration of hydroxylamine metabolite necessary to cause toxicity is very much lower than the concentration of sulfamethoxazole required when it is used in combination with hepatic microsomes.238.239 Since the reactive metabolite is generated in situ by murine microsomes or added directly to the incubation, the increased sensitivity of the patient’s lymphocytes over control being tested by this system must not involve a difference in the generation of a toxic metabolite. This is such an artificial system that it is also likely that the mechanism of toxicity is significantly different from that involved in the in vivo hypersensitivity reactions. The most likely factor being tested by this system is a detoxication system, such as a glutathione transferase, and lymphocytes from patients with a hereditary deficiency in glutathione synthetase also had an increased sensitivity to sulfonamide metabolite toxicity. 238 These observations support the involvement of a hydroxylamine or similar metabolite in the initiation of these hypersensitivity reactions. Reactive metabolites of sulfonamides are presumably generated in the liver as indicated by the experiments described earlier. Although the liver is sometimes involved in sulfonamide hypersensitivity reactions,240 involvement is usually not limited to the liver. This is in contrast to halothane hypersensitivity reactions that appear to only involve the liver. The probable reason for this difference is that the formation of a reactive metabolite of halothane involves the oxidation of a carbonhydrogen bond. This type of oxidation can probably only occur with the help of cytochrome P-450. Most of the cytochromes P-450 are located in the liver, and the metabolite formed is too reactive to escape the liver; therefore, it is not surprising that toxicity is limited to the liver. However, the formation of a reactive metabolite of procainamide or a sulfonamide involves oxidation of an arylamine. This type of oxidation can also occur via several other types of enzymes such as peroxidases and prostaglandin synthetase. Because the toxicity of procainamide and sulfonamides appears to be mediated by the immune system, it is likely that formation of reactive metabolites by monocytes or similar cells, which are key to the initiation of an immune response, would be more important for the mechanism of hypersensitivity reactions involving these drugs than oxidation by the liver. This could explain the generalized nature of sulfonamide reactions. At the present time, the most common use of the sulfonamides is in combination with trimethoprim. Hypersensitivity reactions associated with this combination are usually attributed to the sulfamethoxazole. However, hypersensitivity reactions to

226

trimethoprim (which is also a heterocyclic aromatic amine) have also been reported.24’Another combination is sulfadoxine, a long-acting sulfonamide, and pyrimethamine, another heterocyclic aromatic amine. This combination is used for malaria prophylaxis in regions where there is a high incidence of chloroquine resistance. This combination is also associated with a high incidence of a hypersensitivity syndrome that is multisystemic and often includes a hypereosinophilic c ~ m p o n e n t . ~ ~This ’ . ’ ~syndrome ~ has a very high mortality rate, and it has greatly tempered the use of this Combination. As mentioned before, sulfasalazine is metabolized to sulfapyridine and 5-aminosalicylic acid. Sulfasalazine is also associated with a relatively high incidence of hypersensitivity reactions. 194.244 Pulmonary fibrosis is a prominent component of many sulfasalazine reaction^.^^^-^^* Analogous to phenytoin, sulfasalazine has also been reported to cause IgA deficiency.249

2. Generalized Hypersensitivity Reactions due to Anticonvulsants Phenytoin is associated with a variety of generalized hypersensitivity reactions,250and the clinical picture is similar to that observed with the sulfonamides. Two major components to phenytoin reactions are liver i n v o l ~ e m e n t ~and ~’-~~~ lymphadenopathy.33.254 Vasculitis and lymphocytic interstitial pneumonia have also been d e s ~ r i b e d . ~As ~ ~in, ’the ~ ~case of sulfonamide reactions, Spielberg et al. have found that a metabolite of phenytoin generated by hepatic microsomes is more toxic to the lymphocytes from patients who have had a hypersensitivity reaction than it is to lymphocytes from normal C O ~ ~ T O In I Saddition, . ~ ’ ~ lymphocytes from “normal” relatives also have a significantly different sensitivity than normal controls. As mentioned before, we have found that phenytoin is chlorinated by MPOIH,O,ICI- to a reactive N , N ’ dichlorophenytoin, and we propose that this metabolite is responsible for the hypersensitivity reactions.‘46 The mechanism could involve its action as a hapten, or this metabolite could have a direct effect on the lymphocyte or macrophage function. The prominent effects of phenytoin on lymphocytes and immune function, as well as the lesser degree of covalent binding of radiolabeled phenytoin to activated neutrophils compared with that observed with the arylamines, make the second hypothesis attractive. Other similar drugs such as mephenytoin and trimethadione have an even higher incidence of hypersensitivity reactions than phenytoin and therefore are seldom used. The most common clinical presentation is a syndrome resembling serum sickness with fever, lymphadenopathy, and arthralgias. This clinical picture suggests a type 111 reaction, but because of their infrequent use, there are few data concerning the mechanism of these reactions. Although the structure of carbamazepine is very different from the other anticonvulsants, it is associated with a similar

Volume 20,Issue 4

1377221


Toxicology spectrum of hypersensitivity reactions.’54~’s5~’58 Shear and Spielberg have also found that the same in v i m assay used for other anticonvulsants and the sulfonamides also predicts hypersensitivity reactions to carbamazepine. 157 In fact, there is some degree of “cross-reactivity” between the anticonvulsants; that is, if a patient has a hypersensitivity reaction to phenytoin and is positive to phenytoin in the in vitro assay, he is more likely to be positive in the in virro assay to carbamazepine and have a hypersensitivity reaction to carbamazepine. It is unknown what factor is responsible, but it appears to be inherited and to involve detoxication pathways. Although the structure of carbamazepine appears very different from phenytoin and the other hydantoins, it is a substituted urea with a nitrogen which is relatively easy to oxidize, and we have found that carbamazepine is oxidized to iminostilbene and other metabolites by activated neutrophils or MPO/H,O,/Cl- and this could generate reactive metabolite^.^^ 3. Hypersensitivity Reactions Associated with Dapsone and Other Arylamines Dapsone is associated with a syndrome resembling mononucleosis, and it is referred to as the “dapsone syndrome” . 2 5 8 - 2 6 0 Its major manifestations are fever, lymphadenopathy, and circulating atypical lymphocytes. At least superificially, this appears to be just a minor variation on the theme of hypersensitivity reactions that have already been described. Dapsone is also associated with a high incidence of hemolytic anemia, but this is due to the “oxidative stress” caused by the redox cycling of the hydroxylamine and nitroso metabolites, and it does not represent a hypersensitivity reaction. 261 In contrast to the hemolytic anemia due to dapsone, nomifensine is associated with high incidence of an antibodymediated hemolytic anemia.88 It is likely that this toxicity is initiated by a reactive metabolite of nomifensine acting as a hapten. Aminoglutethimide is associated with a serum sickness-like hypersensitivity reaction. I9O Even though procainamide is associated with a high incidence of drug-induced lupus and agranulocytosis, surprisingly, it rarely causes a generalized hypersensitivity reaction. However, there are case reports of an antibody-mediated hemolytic anemia and granulomatous hepatitis associated with procainamide .176.262.263

4. Hypersensitiwify Reactions Associafed with Thiono Sulfur- and Sulfhydryl-ContainingDrugs Propylthiouracil and methimazole are associated with hypersensitivity reactions. These reactions most often affect the skin, but vasculitis,264-266 p ~ l y a r t h r i t i s , ’ ~and ~ , ’“allergic” ~~ hepatitis have also been reported. 267 Other sylfhydryl-containing drugs, such as penicillamineiZ8 and captopril,268 are also associated with generalized hypersensitivity reactions. One interesting reaction is myasthenia

gravis which is associated with the use of penicillamine.i2*In myasthenia gravis, an antibody is induced that binds t(i thc acetylcholine receptor on motor neurons and leads to paralysis. This same reaction has been reported with trimethadione.2h” This is an example of a type I1 hypersensitivity reaction. lr is interesting to speculate how these drugs induce antibodies with this specificity.

D. Toxicity involving the Thyroid As mentioned earlier. the mechanism by which propylthiouracil and methoimazole exert their therapeutic effect of inhibiting thyroxine synthesis appears to involve the formation of a reactive metabolite by thyroid peroxidase which inhibits the enzyme.”* This type of inhibition is sometimes referred to as suicide inhibition. Thyroid peroxidase is similar to myeloperoxidase, and it is likely that many of the drugs that cause hypersensitivity reactions and are metabolized bj myeloperoxidase are also metabolized by thyroid peroxidase. Arylamines are classically associated with the inhibition of’ thyroid function.z70The sulfonamides cause hypothyroidism in ratsz7’ and have been reported to decrease thyroxine an3 triiodothyronine in humans, but this does not appear to hi. significant at clinical doses in human~.’~’Aminoglutethimide can lead to hypothyroidism at therapeutic doses.2 7 3 m Dapsonc has not been reported to cause hypothyroidism in man. but i t does cause thyroid cancer in rats.*” Another interesting aspect of this problem is that autoimmune thyroiditis has been observed in patients who were recovering from severe hypersensitivity reactions. 27h27i In our study. this occurred in young patients who would otherwise be at very low risk of developing this condition. Antimicrosomal antibtdies were observed, and although this is common in autoimniune thyroiditis, microsomes in the thyroid are chiefly thyroid peroxidase and would be the expected target of an antibody that was induced by a thyroid peroxidase-generated reactivc metabolite acting as a hapten.

111. ANTIINFLAMMATORY EFFECTS MEDIATED BY LEUKOCYTE-GENERATED METABOLITES In addition to the drug hypersensitivity reactions and hypothyroidism that may be mediated by peroxidase-generated reactive metabolites, a similar mechanism may be involved in the antiinflammatory effects associated with several of these drugs. Dapsone is the drug of choice for the treatment of several inflammatory diseases, especially those in which the inflammation is mediated by n e ~ t r o p h i l s27‘, . ~ The ~ ~ best example is dermatitis herpetiformis.280This is a disease often associated with gluten enteropathy, and it involves an infiltration of neutrophils into the skin. This leads to small vesicles and severe itching. Treatment with dapsone leads to a prompt and com-

1990

227


Critical Reviews In plete amelioration of the condition. However, discontinuation of the dapsone usually leads to a recurrence of the condition. Dapsone is effective in a wide variety of other bullous skin diseases and several types of vasulitis. Even rheumatoid arthritis responds to dapsone, although other drugs are somewhat more effective, and dapsone is associated with a high incidence of side effects.28'.282Sulfapyridine is effective in the same conditions, and it is often used when dapsone is not tolerated. The therapeutic action of dapsone in dermatitis herpetiformis has been studied. It had little effect on most aspects of neutrophil action, and the major activity appeared to be inhibition of MP0.283*284 Presumably, the activation of neutrophils with generation of hydrogen peroxide and release of MPO is responsible for much of the inflammation which occurs in conditions such as dermatitis herpetiformis. Since we have found that dapsone is metabolized by MPO, it seems logical that the mechanism of MPO inhibition by dapsone involves either competitive inhibition of MPO or suicide inhibition of MPO by a reactive metabolite of dapsone. Io4 Another example of a possible connection between oxidation of a drug by MPO and that drug's therapeutic effect is the use of sulfasalazine and 5-aminosalicylic in the treatment of ulcerative colitis and other inflammatory bowel diseases.285There is evidence that sulfasalazine and 5-aminosalicylic acid inhibit formation of neutrophil-generated activated oxygen,286-288 and furthermore, that neutrophils are involved in the pathology of inflammatory bowel disease.289Sulfasalazine also has a therapeutic effect in rheumatoid arthritis.290.291 Phenylbutazone and tenoxicam, which are used as nonsteroidal antiinflammatory drugs, are metabolized by neutrophils, and it has been suggested that this is involved in their action as antiinflammatory agents.74*75 Other antiinflammatory drugs also scavenge reactive species such a hypochlorite that are produced by neutrophils, and the drugs are presumably oxidized in the process.292*293 Thus, a portion of the antiinflammatory effect of several drugs may be due to their interaction with MPO or the reactive species produced by activated neutrophils.

IV. SUMMARY This review presents a unifying hypothesis that provides a connection between several types of hypersensitivity reactions associated with several types of drugs and explains some of the therapeutic effects (antiinflammatory activity and antithyroid effects) of these same drugs. This hypothesis centers on the oxidation of these drugs to chemically reactive metabolites by peroxidases. The drugs of interest have functional groups that are easily oxidized. The major peroxidase involved in this hypothesis is MPO because of its critical location in leukocytes which play a key role in the function of the immune system. However, thyroid peroxidase can probably also oxidize many of the same drugs to reactive metabolites, and this may be

228

responsible for the thyroid autoimmunity observed in connection with some hypersensitivity reactions. Peroxidases have also been described in the skin2" and in platelets,295and their presence may be responsible for the high incidence of skin reactions in the hypersensitivity response and the occurrence of immune-mediated thrombocytopenia, respectively. Involvement of other peroxidases, such as prostaglandin peroxidase, may also be important for antiinflammatory effects of drugs. In addition, leukocytes contain prostaglandin synthetase, and the activation of leukocytes leads to the release of arachidonic acid and the production of prostaglandins. This process may also lead to the metabolism of drugs to reactive metabolites. In studies of the metabolism of procainamide and dapsone, aspirin and indomethacin did not inhibit the formation of the hydroxylamine by neutrophils and mononuclear leukocytes. This is evidence against the involvement of prostaglandin synthetase in these oxidation; however, preliminary studies with other drugs suggest that prostaglandin synthetase may contribute to the metabolism of some drugs by leukocytes. Furthermore, the metabolism of phenylbutazone, phenytoin, and tenoxicam, as well as our preliminary work with other drugs such as carbamazepine, suggests that the range of drugs that are metabolized to reactive metabolites by peroxidases may be broader than initially suspected. There are several other drugs that do not fit into the functional group classes covered in this review but have similar properties. A good example is alphamethyldopa, which is associated with drug-induced lupus, immune-mediated hemolytic anemia, and other hypersensitivity reactions. Such drugs may also be metabolized to reactive metabolites by peroxidases. Another aspect of the hypothesis is that an infection, or other inflammatory condition, may be an important risk factor for a hypersensitivity reaction because such a stimulus leads to activation of leukocytes which can lead to formation of reactive metabolites from certain drugs. This may explain why some types of hypersensitivityreactions are more common in patients with certain types of infection or inflammatory disease such as SLE. Other possible risk factors, such as binding of haptenized protein to class I1 MHC mtigens, which will be different for each patient, are much more difficult to predict. It should also be noted that this hypothesis is limited to the initial steps of hypersensitivity reactions, and even if correct, does not explain how antibodies lead to the observed pathology. In addition, it does not explain why reactive metabolites from different drugs lead to a different spectrum of hypersensitivity reactions. Procainamide and dapsone are similar arylamines, and yet, procainamide causes a high incidence of drug-induced lupus and agranulocytosis, but it seldom causes a generalized hypersensitivity reaction. In contrast, dapsone causes agranulocytosis and hypersensitivity reactions, but seldom, if ever, causes drug-induced lupus. In general, the data presented in this review were selected to emphasize common patterns, but it should be understood that such patterns belie the complexity of drug hypersensitivity reactions.

Volume 20,Issue 4

Toxicology ACKNOWLEDGMENT


This work was supported by grants from the Medical Research Council of Canada (MA-9336 and MA-10036) and a grant from the Sunnybrook Trust for Medical Research.

REFERENCES 1. Wallerstein, R. O., Condit, P. K., Kasper, C. K., Brown, J. W., and Morrison, F. R., Statewide study of chloramphenicol therapy and fatal aplastic anemia, JAMA 208, 2045, 1969. 2. Polak, B. C. P., Wesseling, H., Schut, D., Herxheimer, A., and Meyler, L., Blood dyscrasias attributed to chloramphenicol: a review of 576 published and unpublished cases, Acta Med. Scand., 192,409, 1972. 3. Pinsky, P. F., Hurwitz, E. S., Schonberger, L. B., and Gunn, W. J., Reye's syndrome and aspirin: evidence for a dose-response effect, JAMA, 260, 657, 1988. 4. Waldman, R. J., Hall, W. N., McGee, H.,and Van Amburg, G., Aspirin as a risk factor in Reye's syndrome, JAMA, 247, 3089, 1982. 5. Pullen, H., Wright, N., and Murdoch, J., Hypersensitivityreactions to antibacterial drugs in infectious mononucleosis, Lancet, 2, 1176, 1967. 6. Fischl, M. A., Dickinson, G. M., and La Voie, L., Safety and efficacy of sulfamethoxazole and trimethoprim chemoprophylaxis for Pneumocystis carinii pneumonia in AIDS, JAMA, 259, 1185, 1988. 7. Parker, C. W., Allergic reactions in man, Pharmacol. Rev., 34, 85, 1982. 8. Seaman, W. E., Ishak, K. G., and Plotz, P. H.,Aspirin-induced hepatotoxicity in patients with systemic lupus erythematosus, Ann. Intern. Med.. 80, 1, 1974. 9. Seaman, W. E. and Plotz, P. H., Effect of aspirin on liver tests in patients with RA or SLE and in normal volunteers, Arthritis Rheum., 19, 155, 1976. 10. Coombs, R. R. A. and Gell, P. G . H., The classification of allergic reactions underlying disease, in Clinical Aspects oflmmunology, Gell, P. G . H. and Cwmbs, R. R. A., Eds., Blackwell Scientific F'ublications, Oxford, 1963, 317. 11. Park, B. K., Coleman, J. W., and Kitteringham, N. R., Drug disposition and drug hypersensitivity,Biochem. Pharmacol., 36, 581, 1987. 12. Pohl, L. R., Satoh, H.,Christ, D. D., and Kenna, J. G., The immunologicand metabolic basis of drug hypersensitivities,Annu. Rev. P h a m c o l . , 28, 367, 1988. 13. Unanue, E. R. and Allen, P. M., The hasis for the immunoregulatory role of macrophages and other accessorycells, Science, 236,551,1987. 14. Petz, L. D. and Fudenberg, H. H., Immunologic mechanisms in drug-induced cytopenias, Prog. Hemarol., 9, 185, 1975. 15. Vergani, D., Mieli-Vergani, G., Alberti, A., Neuberger, J., Eddleston, A., Davis, M., and Williams, R., Antibodies to the surface of halothane-altered rabbit hepatocytes in patients with severe halothane-associated hepatitis, N . Engl. J. Med., 303, 66, 1980. 16. Kenna, J. G., Neuberger, J., and Williams, R., Evidence for the expression in human liver of halothane-inducedneoantigens recognized by antibodies in sera from patients with halothane hepatitis, Hepatology, 8, 1635, 1988. 17. Kenna, J. G., Satoh, H., Christ, D. D., and Pohl, L. R., Metabolic basis for a drug hypersensitivity: antibodies in sera from patients with halothane hepatitis recognize liver neoantigens that contain the trifluo-

1990

-

roacetyl group derived from halothane,J . Pharmacol. Exp. Ther.,245, 1103, 1988. 18. McLain, G. E., S i p , I. G., and Brown, B. R., An animal model of halothane hepatotoxicity, Anesthesiology, 51, 321, 1979. 19. Moult, P. J. A. and Sherlock, S., Halothane related hepatitis. A clinical study of 26 cases, Q. J . Med., 44,99, 1975. 20. Arroyave, C. M., Bhat, K. N., and Crown, R., Activation of the alternative pathway of the complement system by radiographiccontrast media, J . Immunol., 117, 1866, 1976. 21. Kolb, W. P., Lang, J. H.,and Lasser, E. C., Nonimmunologic complement activation in normal human serum induced by radiographic contrast media, J. Immunol., 121, 1232, 1978. 22. von Zabern, I., Pnyklenk, H., Vogt, W., and Sachsenheimer, W., The effect of radiographic contrast media on complement components C3 and C4: generation of C3b-like C3 and C4b-like C4, Int. J. Immunopharmac., 5, 503, 1983. 23. Sim, E. and Law, S. A., Hydralazine hinds covalently to complement component C4: different reactivity of C4A and C4B gene products, FEBS. Lett., 184, 323, 1985. 24. Sim, E., Stanley, L., Gill, E. W., and Jones, A., Metabolites of procainamide and practolol inhibit complement components C3 and C4, Biochem. J., 251, 323, 1988. 25. Sim, E., Gill, E. W., and Sim, R. B., Drugs that induce systemic lupus erythematosus inhibit complement component C4, Lancet, 2, 422, 1984. 26. American Medical Association, Drug Evaluations, W. B. Saunders, Philadelphia, 1986, 1154. 27. Sorrell, T. C., Forbes, I. J., Burness, F. R., and Rischbieth, R. H. C., Depression of immunological function in patients treated with phenytoin sodium (sodium diphenylhydantoin), Lancet, 2, 1233, 1971. 28. Aarli, J. A. and Tonder, O., Effect of antiepileptic drugs on serum and salivary IgA, Scand. J . Immunol., 4, 391, 1975. 29. Seager, J., Jamison, D. L., Wilson, J., Hayward, A. R., and Soothill, J. F., IgA deficiency, epilepsy and phenytoin treatment, Lancet, 2, 632, 1975. 30. Gilhus, N. E., Matre, R., and Aarli, J. A., Lymphocyte subpopulations and lymphocyte function in phenytoin-treatedpatients with epilepsy, Int. J . Imrnunophannac., 4, 43, 1982. 31. Ricevuti, G., Marcoli, M., Gatti, G., Mazzone, A., hechini, S., and Frigo, G. M., Assessment of polymorphonuclearleukocyte functions in adult epileptic patients undergoing long-term phenytoin treatment, Hum. Toxicol., 5, 237, 1986. 32. Guerra, I. C., Fawcett, W. A., Redmon, A. H., Lawrence, E. C., Rosenblatt, H. M., and Shearer, W. T., Permanent intrinsic B cell immunodeficiency caused by phenytoin hypersensitivity, J . Allergy Clin. Immunol., 77, 603, 1986. 33. Abratt, A. P., Sealy, R., Uys, C. J., and Lawson, R., Lymphadenopathy associated with diphenylhydantoin therapy, Clin. Oncol., 8, 351, 1982. 34. Rosenthal, C. J., Noguera, C. A., Coppola, A., and Kapetner, S. N., Pseudolymphoma with mycosis fungoides manifestations, hyperrespnsiveness to diphenylhydantoin, and lymphocyte disregulation, Cancer, 49, 2305, 1982. 35. Aymard, J. P., M e r l i n , P., Witz, F., Colomb, J. N., Faure, G., Guerci, O., and Herbeuval, R., Multiple myeloma after phenytoin therapy, Scand. J. Haematol., 26, 330, 1981. 36. Uetrecht, J. P. and Woosley, R. L., Acetylator phenotype and lupus erythematosus, Clin. Phartnacokin., 6, 118, 1981. 37. Wallace, D. J. and Duboii, E. L., Dubois' Lupus Erythematosus. Lea & Febiger, Philadelphia, 1987. 38. Levo, Y., Pick, A. I., Avidor, I., and Ben-Bassat, M.,Clinicopathological study of a patient with procainamide-induced systemic lupus erythematosus, Ann. Rheum. Dis.,35, 181, 1976.

229


_

_

_

_

_

_

_

I

~

Critical Reviews In

39. Whittle, T. S. and Ainsworth, S. K., Procainamide-induced systemic lupus erythematosus, Arch. Puthol. Lab. Med., 100, 469, 1976. 4FJ. k h , P., Colon, S., Labeeuw, M., Bernheim, J., and Blanc-Brunat, N., Nephrotic syndrome in procainamide induced lupus, Clin. Nephrol., 11, 218, 1979. 41. Foucar, E., Erickson, D. G., and Tung, K. S. K., Glomerulonephntis in procainamide induced lupus erythematosus: report of a case and review of the literature. J . Clin. Lnb. Immunol., 2, 79, 1979. 42. Uetrecht, J. P., Mechanism of drug-induced lupus, Chem. Res. Toxt d . , 1 , 133, 1988. 43. Reidenberg, M. M., Aromatic amines and the pathogenesis of lupus erythematosus, Am. J. Med.. 75. 1037. 1983. 44. Reidenberg, M. M., Durant, P. J., Harris, R. A., De Boccardo, G . , Lahita, R.. and Stenzel, K. H., Lupus erythematosus-like disease due to hydrazine. Am. J . Med.. 75. 365, 1983.

45, Malinow, M. R., Bardana, E. J., Pirofsky, B., Craig, S., and McLaughlin, P., Systemic lupus erythematosus-like syndrome in monkeys fed alfalfa sprouts: role of a nonprotein amino acid, Science, 216, 415. 19x2. 46, Isenberg, D. A., Dudeney, C., Williams, W., Todd-Pokropek, A., and Stollar, B. D., Disease activity in systemic lupus erythematosus related to a range of antibodies binding DNA and synthetic polynucleotides. Ann. Rheum. D i s . , 47. 717, 1988. 47. Fritzler, M. J. and Tan, E. M., Antibodies to histones in druginduced and idiopathic lupus erythematosus. J . Clin. Invest., 62, 560, 1978.

48. Portanova, J. P., Rubin, R. L., Joslin, F. G., AgneUo, V. D., and Tan, E. M., Reactivity of anti-histone antibodies induced by procainamide and hydralazine, Clin. Immunol. Immunopathol., 25, 67, 1982. 49. Gohill, J. and Fritzler, M. J., Antibodies in procainamide-induced and systemic lupus erythematosus bind the C-terminus of histone 1 (HI), Mol. Immunol., 24, 275, 1987. SO. Portanova, J. P., Arndt, R. E., Tan, E. M., and Kotzin, B. L., Anti-histone antibodies in idiopathic and drug-induced lupus recognize distinct intrahistone regions, J. Immunol., 138. 446, 1987. 51. Totoritis, M. C., Tan, E. M., McNally, E. M., and Rubin, R. L.,

Association of antibody to histone complex H2A-HZB with symptomatic procainamide-induced lupus, N . Engl. J . Med., 318. 1431, 1988. 57. Schwartz, R. S. and Stollar, B. D., Origins of anti-DNA autoantibodies, J . Clin. Invest., 75. 321, 1985. 53. Harkiss, G . D., Hendrie, F. M., and Thompson, D., Anti-Klebsiella K30 phospholipid antibodies in systemic lupus erythematosus: antigen cross-reactions and idiotypic sharing with antibodies to DNA and Kleb.siellu K30 plysaccharide, J. Clin. Immunol., 8, 319, 1988. 54. Pincus, T., Schur, P. H., Rose, J. A., Decker, J. L., and Talal, N., Measurenieni of semm DNA-binding activity in systemic lupus erythematosus. A’. Engl. J. Med.. 281, 701, 1969. 55. Blomgren, S. E., Condemi, J. J., Bignall, M. C., and Vaughan, J. H., Antinuclear antibody induced by procainamide: a prospective study. N . Engl. J , Med., 281, 64, 1969. 56. Wmley. R. L., Drayer, D. E., Reidenberg, M. M., Nies, A. S., Cam, K., and Oates, J. A.. Effect of acetylator phenotype on the rate at which procainamide induces antinuclear antibodies and the lupus syndrome, N . EngI. J . Med.. 298, 1157, 1978. 57. Henningsen, N. C., Cederberg, A., Hanson, A., and Johansson, B. W., Effects of long-term treatment with procaine amide, Acta Med. Scund., I98. 47.5. 1975. .%. Lahita, R., Kluger, J., Drayer, D. E., Koftler, D., and Reidenberg, M. M., Antibodies to nuclear antigens in patients treated with procainamide or acetylprocainamide. N . Engl. J . Med., 301, 1382, 1979. 50. Roden, D. M., Reele, S. B., Higgins, S. B., Willdnson, G. R., Smith, R. F., Oates, J. A., and Woosley, R. L., Antiarrhythmic efficacy, phannacokinetics and safety of N-acetylprocainarnide in hu-

230

man subjects: comparison with procainamide, Am. J . Cardiol., 46, 463, 1980. 60. Stec, G. P., Lertora, J. J. L., Atkinson, A. J., Nevin, M. J., Kushner, W., Jones, C., Schmid, F. R., and Askenazi, J., Remission of procainamide-induced lupus erythematosus with N-acetylprocainamide therapy, Ann. Intern. Med., 90, 799, 1979. 61. Kluger, J., Drayer, D. E., Reidenberg, M. M., and Lahita, R., Acetylprocainamide therapy in patients with previous procainamideinduced lupus syndrome, Ann. Intern. Med., 95, 18, 1981. 62. Uetrecht, J. P., Sweetman, B. J., Woosley, R. L., and Oates, J. A., Metabolism of procainamide to a hydroxylamine by rat and human hepatic microsomes, Drug Metab. Dispos., 12, 77, 1984. 63. Freeman, R. W., Uetrecht, J. P., Woosley, R. L., Oates, J. A., and Harbison, R. D., Covalent binding of procainamide in vitro to hepatic protein in mice, Drug Metab. Dispos., 9, 188, 1981. 64. Uetrecht, J. P., Reactivity and possible significance of hydroxylamine and nitroso metabolites of procainamide, J . Pharmacol. Exp. Ther., 232, 420, 1985. 65. Rubin, R. L., Uetrecht, J. P., and Jones, J. E., Cytotoxicity of oxidative metabolites of procainamide, J . Pharmacol. Exp. Ther., 242, 833, 1987. 66. Uetrecht, J., Zahid, N., and Rubin, R.,Metabolism of procainamide to a hydroxylamine by human neutrophils and mononuclear leukocytes, Chem. Res. Toxicol., 1, 74, 1988. 67. Klebanoff, S. J. and Clark, R. A., The Neutrophil: Function and Clinical Disorders, North-Holland, New York, 1978. 68. Sberra, A. J. and Strauss, R. R., The Respiratory Burst and Its Physiological Significance, Plenum Press, New York, 1988. 69. van Furth, R., Mononuclear Phagocytes: Functional Aspects, Martinus Nijhoff, The Hague, 1980. 70. Harrison, J., Araiso, T.,Palcic, M., and Dunford, H., Compound Iof myeloperoxidase, Biochem. Biophys. Res. Commun., 94,34,1980. 71. Winterborn, C., Comparative reactives of various biological compounds with myeloperoxidase-hydrogenperoxide-chloride, and similarity of the oxidant to hypochlorite, Biochim. Biophys. Acta, 840, 2 0 4 , 1985. 72. Hammerschmidt, D. E., Activation of the complement system and of granulocytes in lung injury: the adult respiratory distress syndrome, in Advances in Inflammation Research, Weissmann, G . , Ed., Raven Press, New York, 1983, 147. 73. Johnson, R. J., Cower, W. G., Chi, E. Y., Adler, S., and Klebanoff, S., New mechanism for glomerular injury: myeloperoxidasehydrogen peroxide-halide system, J . Clin. Invest., 79, 1379, 1987. 74. Ichihara, S., Tomisawa, H., Fukazawa, H., and Tateishi, M., involvement of leukocyte peroxidases in the metabolism of tenoxicam, Biochem. Pharmacol., 34, 1337, 1985. 75. Ichihara, S., Tomisawa, H., Fukazawa, H., Tateishi, M., Joly, R., and Heintz, R., Involvement of leukocytes in the oxygenation and chlorination reaction of phenylbutazone, Biochem. Pharmacol., 35, 3935, 1986. 76. Tsuruta, Y., Subrahmanyam, V. V., Marhsall, W., and O’Brien, P. J., Peroxidase-mediated irreversible binding of arylamine carcinogens to DNA in intact polymorphonuclear leukocytes activated by a tumor promoter, Chem.-Biol. Interact., 53, 25, 1985. 77. Uetrecht, J. and Zahid, N., Procainamide (PA) is N-chlorinated by myeloperoxidase (MFQ) - implications for toxicity, Pharmacologist, 30, A98, 1988. 78. Werdelin, O., Mouritsen, S., Peterson, B. L., Sette, A., and Buus, S., Facts on the fragmentation of antigens in presenting cells, on the association of antigen fragments with MHC molecules in cell-free systems, and speculation on the cell biology of antigen processing, Immunol. Rev., 106, 181, 1988. 79. Francoeur, A. and Heitzmann, J. G., Autoantibodies: terms and

Volume 20,Issue 4


Toxicology concepts, Clin. Immunol. Immunopathol., 47, 245, 1988. 80. Andersen, M. R., Atkin, C. L., and Eyre, H. J., Intact form of myeloperoxidase from normal human neutrophils, Arch. Biochem. Biophys.. 214, 273, 1982. 81. Murao, S., Stevens, F. J., Ito, A., and Huberman, E., Myeloperoxidase: a myeloid cell nuclear antigen with DNA-binding properties, Proc. Natl. Acad. Sci. U.S.A., 85, 1232, 1988. 82. Gorsulowsky, D. C., Bank, P. W., Goldberg, A. D., Lee, T. G., Heinzerling, R. H., and Burnham, T. K., Antinuclear antibodies as indicators for the procainamide-induced systemic lupus erythematosuslike syndrome and its clinical presentations, J . Am. Acad. Dermatol., 12, 245, 1985. 83. Fa&, R. J. and Jennette, J. C., Anti-neutrophil cytoplasmic autoantibodies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and crescentic glomerulonephritis, N. Engl. J. Med., 318, 1651, 1988. 84. Weller, P. F., Ackerman, S. J., and Smith, J. A., Eosinophil granule cationic proteins: major basic protein is distinct from the smaller subunit of eosinophil peroxidase, J. Leukocyte Biol.. 43, 1, 1988. 85. Hoffman, B. J., Sensitivity to sulfadiazine resembling acute lupus erythematosus, Arch. Dermatol. Syph., 51, 190, 1945. 86. Uetrecht, J. P., Shear, N., and Biggar, W., Dapsone is metabolised by human neutrophils to a hydroxylamine, Pharmacologist, 28, 239, 1986. 87. Uetrecht, J., unpublished observation. 88. Salama, A. and Mueller-Edrhardt, C., The role of metabolite-specific antibodies in nomifensine-dependent immune hemolytic anemia, N. Engl. J. Med., 313, 469, 1985. 89. Garcia-Morteo, 0. and Maldonado-Cocco, J. A., Lupus-like syndrome during treatment with nomifensine, Arthritis Rheum., 26, 936, 1983. 90.McCraken, M., Benson, E. A., and Hickling, P., Systemic lupus erythematosus induced by aminoglutethimide, Br. Med. J . , 281, 1254, 1980. 91. Raffery, E. B. and Denman, A. M., Systemic lupus erythematosus syndrome induced by practolol, Br. Med. J . , 2, 452, 1973. 92. Stewart, M. W. and Clarke, S. W., Practolol-induced SLE-like syndrome, Proc. R . Soc. Med., 69, 61, 1976. 93. Milner, G. R., Holt, P. J. L., Bottomley, J., and Maciver, J. E., Practolol therapy associated with a systemic lupus erythematosus-like syndrome and an inhibitor to factor XIII, J. Clin. Pathol., 30, 770, 1977. 94. Cody, R. J., Calabrese, L. H., Clough, J. D., Tarazi, R. C., and Bravo, E. L., Development of antinuclear antibodies during acebutolol therapy, Clin.Pharmacol. Ther., 25, 800, 1979. 95. Booth, R. J., Bullock, J. Y., and Wilson, J. D., Antinuclear antibodies in patients on acebutolol, Br. J. Clin. Pharmacol., 9, 515, 1980. 96. Harrison, T., Sisca, T. S., and Woad, W. H., Propranolol-induced lupus syndrome?, Postgrad. Med., 59, 241, 1976. 97. Wilson, J. D., Bullock, J. Y., Sutherland, D. C., Main, C., and O’Brien, K. P., Antinuclear antibodies in patients receiving nonpractolol beta-blockers, Br. Med. J., 1, 14, 1978. 98. Booth, R. J., Wilson, J. D., and Bullock, J. Y., P-Adrenergicreceptor blockers and antinuclear antibodies in hypertension, Clin. Pharmacol. Ther., 31, 555, 1982. 99. Sutherland, D. C. and Wilson, J. D., Practolol and the safety of other beta blockers, N.Z. Med. J., 84, 282, 1976. 100. Amos, H. E., Lake, B. G., and Artis, J., Possible role of antibody specific for practolol metabolite in the pathogenesis of oculomucocutaneous syndrome, Br. Med. J., I , 402, 1978. 101. Amos, H. E., Immunological aspects of practolol toxicity, Int. J. Immunopharmac., 1, 9, 1979. 102. Griffiths, I. D. and Kane, S. P., Sulphasalazine-induced lupus syndrome in ulcerative colitis, Br. Med. J., 2, 1188, 1977.

1990

103. Vanheule, B. A. and Carswell, F., Sulphasalazine-induced systemic lupus erythematosus in a child, Eur. J. Pediatr., 140, 66, 1983. 104. Uetrecht, J., Zahid, N., Shear, N. H., and Biggar, W. D., Metaholism of dapsone to a hydroxylamine by human neutrophils and mononuclear cells, J. Pharmacol. Exp. Ther., 245, 274, 1988. 105. Vandersteen, P. R. and Jordan, R. E., Dermatitis herpetiformis with discoid lupus erythematosus: Occurrence of sulfone-induced discoid lupus erythematosus, Arch. Dermatol., 110, 95, 1974. 106. Perry, H. M., Late toxicity to hydralazine resembling systemic lupus erythematosus or rheumatoid arthritis, Am. J. Med., 54, 58, 1973. 107. Cameron, H. A. and Ramsay, L. E., The lupus syndrome induced by hydralazine: a common complication with low dose therapy, Br. Med. J . , 289, 410, 1984. 108. Harland, S. J., Facchini, V., and Timbrell, J. A., Hydrallazineinduced lupus erythematosus-like syndrome in a patient of the rapid acetylator phenotype, Br. Med. J., 281, 273, 1980. 109. Hofstra, A. and Uetrecht, J., Oxidation of hydralazine by monocytes as a mechanism of hydralazine-induced lupus, Pharmacologist, 30, A99, 1988. 1 10. Timbrell, J. A., Facchini, V., Harland, S. J., and Mansilla-Tinoco, R., Hydralazine-induced lupus: is there a toxic pathway?, Eur. J . Clin. Pharmacol., 27, 555, 1984. 111. Li, A. and Uetrecht, J., Metabolism of isoniazid by activated neutrophils, Pharmacologist, 30, A99, 1988. 112. Engler, H., Taurog, A., and Nakashima, T.,Mechanism of inactivation of thyroid peroxidase by thioureylene drugs, Biochem. Pharmacol., 31, 3801, 1982. 113. Engler, H.,Taurog, A., Luthy, C., and Dorris, M., Reversible and irreversible inhibition of thyroid peroxidase-catalyzed iodination by thioureylene drugs, Endocrinology, 112, 86, 1983. 114. Imamura, M., Aoki, N., Saito, T., Ohno, Y., Maruyama, Y., Yamaguchi, J., and Yamamoto, T., Inhibitory effects of antithyroid drugs on oxygen radical formation in human neutrophils, Acra Endocrinol., 112, 210, 1986. 115. Lee, E., Miki, Y., Hosokawa, M., Sayo, H., and Kariya, K., Oxidative metabolism of propylthiouracil by peroxidases from rat bone marrow, Xenobiotica, 18, 1135, 1988. 116. Waldhauser, L. and Uetrecht, J., Propylthiouracil (PTU) is metabolized by activated neutrophils - implications for agranulocytosis, FASEBJ., 2, A1134, 1988. 117. Amrhein, J. A., Kenny, F. M., and Ross, D., Granulocytopenia, lupus-like syndrome, and other complications of propylthiouracil therapy, J . Pediatr., 76, 54, 1970. 118. Berkman, E. M., Orlin, J. B., and WoIfsdorf, J., An anti-neutrophil antibody associated with a propylthiouracil-induced lupus-like syndrome, Transfusion, 23, 135, 1983. 119. Takuwa, N., Kojima, I., and Ogata, E., Lupus-like syndrome - a rare complication in thionamide treatment for Graves’ disease, Endocrinol. Jpn., 28, 663, 1981. 120. Aucoin, D. P., Peterson, M. E., Hurvitz, A. I., Drayer, D. E., Lahita, R. G., Quimby, F. W., and Reidenberg, M. M.,Propylthiouracil-induced immune-mediated disease in the cat, J. Pharmacol. Exp. Ther., 234, 13, 1985. 121. Aucoin, D. P., Rubin, R. L., Peterson, M. E., Reidenberg, M. M., Drayer, D. E., Hurvitz, A. I., and Lahita, R. G., Dosedependent induction of anti-native DNA antibodies in cats by propylthiouracil, Arthritis Rheum., 31, 688, 1988. 122. Farbman, K., Wheeler, M. F., and Glick, S. M., Arthritis induced by antithyroid medication, N.Y. State J. Med., 69, 826, 1969. 123. Kurtz, M. D., Migratory polyarthritis Occumng with methimazole therapy, N.Y. Stare J. Med., 68, 2810, 1968. 124. Reidenberg, M. M., Case, D. B., Drayer, D. E., Reis, S., and Lorenzo, B., Development of antinuclear antibody in patients treated with high doses of captopril, Arthritis Rheum., 27, 579, 1984. 125. Patri, P., Nigro, A., and Rebora, A., Lupus erythematosus-like

231


-.

.

Critical Reviews In

eruption from captopril, Acta Dermatol. Venereol., 65, 447, 1985. 126. Harkcom, T. M., Conn, D. L., and HoUey, K. E., D-Penicillamine and lupus erythematosus-likesyndrome, Ann. Intern. Med.. 89, 1012, 1978. 127. Chalmers, A., Thompson, D., Stein, H. E., Reid, G., and Patterson, A. C., Systemic lupus erythematosus during penicillaminetherapy for rheumatoid arthritis, Ann. Intern. Med., 97, 659, 1982. 128. Netter, P., Bannwarth, B., Faure, G., Trechot, P., and Royer, R. J., Adverse effects of D-penicihnine. A cooperative study by the French regional drug surveillance centers., J . Rheumatol., 15, 1730, 1988. 129. Benton, J. W., Tynes, B., Register, H. B., Alford, C., and Holley, H. L., Systemic lupus erythematosus occurring during anticonvulsive drug therapy, JAMA, 180. 115, 1962. 130. Beernink, D. H. and Miller, J. J., Anticonvulsant-induced antinuclear antibodies and lupus-like disease in children, J . Pediatr.. 82, 113, 1973. 131. Singsen, B. H., Fishman, L., and Hanson, V., Antinuclear antibodies and lupus-like syndromes in children receiving anticonvulsants, Pediatrics, 57, 529, 1976. 132. Gleichmann, H., Systemic lupus erythematosus triggered by diphenylhydantoin, Arthrifis Rheum., 25, 1387, 1982. 133. Claesen, M., Moustafa, M. A. A., Adline, J., Vandervorst, D., and Poupaert. J. H., Evidence for an arene oxide-NIH shift pathway in the metabolic conversion of phenytoin to 5-(4-hydroxyphenyl)-5phenylhydantoin in the rat and man, Drug Metab. Dispos., 10, 667, 1982. 134. Ortiz de MonteUano, P. R., Oxygen activation and transfer, in Cytochrome P 4 5 0 : Structure, Mechanism. and Biochemistry, Ortiz de Montellano, P. R., Ed., Plenum Press, New York, 1986, 217. 135. Pantarotto, C., Arboix, M.,Sezzano, P., and Abbruzzi, R., Studies on 5,5-diphenylhydantoin irreversible binding to rat liver microsomal proteins, Biochem. Pharmacol., 31. 1501,1982. 136. Spielberg, S. P., Gordon, G. B., Blake, D. A., Mellits, E. D., and Bross, D. S., Anticonvulsant toxicity in vitro: possible role of arene oxides, 1. Pharmacol. Exp. Ther., 217, 386, 1981. 137. Kupferberg, H. J., Other hydantoins: mephenytoin and ethotoin, in Antiepileptic Drugs, Woodbury, D. M., Penry, J. K., and Pippenger, C. E., Eds.. Raven Press. New York, 1982, 283. 138. Muller-Kuppers, M., Zur frage der fruchtschadigung in der schwangerschaft durch einnahme von antiepileptica, Acta Paedopsychiatry. 30, 401, 1963. 139. Kupfer, A,, Roberts, R. K., Schenker, S., and Branch, R. A., Sterioselectivemetabolism of mephenytoin in man, J. Pharmacol. Exp. Ther., 218, 193, 1981. 140. W e b , P. G., Kupfer, A., Lawson, J. A., and Harbison, R. D., Relation of in vivo drug metabolism to stereoselective hydantoin toxicity in mouse: evaluation of mephenytoin and its metabolite, nirvanol. J . Pharmacol. Exp. Ther.. 221, 228, 1982. 141. Feldman, G. L., Weaver, D. D., and Lovrien, E. W., The fetal trimethadione syndrome: report of an additional family and further delineation of this syndrome, Am. J . Dis. ChiId.. 131, 1389, 1977. 142. Dreifuss, F. E., Ethosuximide, in Antiepileptic Drugs, Woodbury, D. M., Penry, J. K.. and Pippenger, C. E., Eds., Raven Press, New York, 1982, 647. 143. Dabbous, I. A. and Idriss, H. M., Occurrence of systemic lupus erythematosus in association with ethosuximide therapy, 1. Pediatr., 76. 617, 1970. 1 4 4 Teoh, P. C. and Chan, H. L., Lupus-sclerodermasyndrome induced by ethosuximide, Arch. Dis. Child., 50, 658, 1975. 145. Rall, T. W. and Schieifer, L. S., Drugs effective in the therapy of epilepsies. in Goodman and Gilman’s the Pharmacological Basis of Thcrapeurics, Gilman, A. G . , Goodman, L. S., Rall, T. W.,and Murad, F.. Eds., Macmillan, New York. 1985. 446.

232

146. Uetrecht, J. and Zahid, N., N-Chlorination of phenytoin by myeloperoxidase to a reactive metabolite, Chem. Res. Toxicol., 1, 148, 1988. 147. Alarcon-Segovia, D., Fishbein, E., Reyes, P. A., Dies, H., and Shwadsky, S., Antinuclear antibodies in patients on anticonvulsant therapy, Clin. Exp. Immunol., 12, 39, 1972. 148. Simpson, J. R., “Collagen disease” due to carbamazepine(Tegretol), Br. Med. J., 2, 1434, 1966. 149. Bateman, D. E., Carbamazepine induced systemic lupus erythematosus: case report, Br. Med. J . , 291, 632, 1985. 150. Alballa, S., Fritzler, M., and Davis, P., A case of drug induced lupus due to carbamazepine, J . Rheumatol., 14, 599, 1987. 151. Livingston, S., Villamater, C., Sakata, Y., and Pauli, L. L., Use of carbamazepine in epilepsy, JAMA, 200, 204, 1967. 152. Killian, J. M., Tegretol in trigeminal neuralgia with special reference to hemopoietic side effects, Headache, 9, 58, 1969. 153. Livingston, S., Pauli, L. L., and Berman, W., Carbamazepine (Tegretol) in epilepsy, Dis. Nerv. Sysf., 35, 103, 1974. 154. Rosenow, E. C., Drug-induced bronchopulmonary pleural disease, J. Allergy Clin. Immunol.,80, 780, 1987. 155. Severson, G. S., Harrington, D. S., Burnett, D. A., and Linder, J., Dematopathic lymphadenopathy associated with carbamazepine: a case mimicking a lymphoid malignancy, Am. J . Med., 83, 597, 1987. 156. Luchins, D.J., Fatal agranulocytosisin a chronic schizophrenicpatient treated with carbamazepine, Am. J . Psychiatry, 141, 687, 1984. 157. Shear, N. H. and Spielberg, S. P., Anticonvulsant hypersensitivity syndrome, J. Clin. Invesf., 82, 1826, 1988. 158. Lewis, I. J. and Rosenbloom, L., Glandular fever-like syndrome, pulmonary eosinophilia and asthma associated with carbamazepine, Postgrad. Med. J., 58, 100, 1982. 159. Hart, R. G. and Easton, J. D., Carbamazepine and hematological monitoring, Ann. Neurol., 11, 309, 1982. 160. Horler, A. R., Blood disorders, in Textbook of Adverse Drug Reactions, Davies, D . M., Ed., Oxford University Press, Oxford, 1985, 578. 161. Pisciotta, V., Drug-induced agranulocytosis, Drugs, 15, 132, 1978. 162. Ellrodt, A. G., Murata, G. H., Reidinger, M. S., Stewart, M. E., Mochizuki, C., and Gray, R., Severe neutropenia associated with sustained-release procainamide, Ann. Intern. Med., 100. 197, 1984. 163. Meyers, D. G., Gonzalez, E. R., Peters, L. L., Davis, R. B., Feagler, J. R., Egan, J. D., and Nair, C. K., Severe neutropenia associated with procainamide: comparison of sustained and conventional preparations, Am. Heart J., 109, 1393, 1985. 164. Thompson, J. F., Robinson, C. A., and Segal, J. L., Procainamide agranulocytosis: a case report and review of the literature, Curr. Ther. Res., 44, 872, 1988. 165. Vincent, P. C., Drug-induced aplastic anaemia and agranulocytosis: incidence and mechanisms, Drugs, 31, 52, 1986. 166. Heit, W., Heimpel, H., Fisher, A., and Frickhofen, N., Druginduced agranulocytosis: evidence for commitment of bone marrow haematopoiesis, Scand. J. Haematol., 35, 459, 1985. 167. Pisciotta, A. V., Immune and toxic mechanisms in drug-induced agranulocytosis, Semin. Hematol., 10, 279, 1973. 168. Azocar, J., Anti-myeloid-cell antibody in procainamide-induced agranulocytosis, Lancet, 1, 1069, 1984. 169. Berger, B. and Hauser, D., Agranulocytosis due to new sustainedrelease procainamide, Am. Heart J., 105, 1035, 1983. 170. Bluming, A. Z., Plotkin, D., Rosen, P., and Thiessen, R., Severe transient pancytopenia associated with procainamide ingestion, JAMA, 236, 2520, 1976. 17I . Netson, J. C., Lutton, J. D., and Fass, A. E., Procainamide-induced agranulocytosis with reversible myeloid sensitivity, Am. J. Hematol., 17, 427, 1984. 172. Rothman, I. K. and Amorosi, E. L., Procainamide-inducedagran-

Volume 20, Issue 4


Toxicology ulocytosis and thrombocytopenia,Arch. Intern. Med., 139,246,1979. 173. Wang, R. I. H. and Schuller, G., Agranulocytosis following procainamide administration, Am. Heart J . , 78, 282, 1969. 174. Weitzman, S. A. and Stossel, T. P., Drug-induced immunological neutropenia, Lancet, 1, 1068, 1978. 175. Shields, A. F. and Berenson, J. A., Procainamide-associatedpancytopenia, Am. J. Hemutol., 27, 299, 1988. 176. Kleinman, S., Nelson, R., Smith, L., and Goldfiger, D., Positive direct antiglobulin tests and immune hemolytic anemia in patients receiving procainamide, N . Engl. J . Med., 311, 809, 1984. 177. Firkin, F. C. and Mariani, A. F., Agranulocytosis due to dapsone, Med. J. Aust., 2, 247, 1977. 178. Levine, P. H. and Weintraub, L. R., Pseudoleukemia during recovery from dapsone-induced agranulocytosis, Ann. Intern. Med., 68, 1060, 1968. 179. Wilson, J. R. and Harris, J. W., Hematologic side-effects of dapsone, Ohio State Med. J . , 73, 557, 1977. 180. Potter, M. N., Yates, P., Slade, R., and Kennedy, C. T. C., Agranulocytosis caused by dapsone therapy for granuloma annulare, J. Am. Acad. Dermatol., 20, 87, 1989. 181. Ognibene, A. J., Agranulocytosis due to dapsone, Ann. Intern. Med., 72,521, 1970. 182. Weetman, R. M., Boxer, L. A., Brown, M. P., Mantich, N. M., and Baehner, R. L., In vitro inhibition of granulopoiesis by 4-amino4’-hydroxylaminodiphenylsulfone, Br. J . Haematol., 45, 361, 1980. 183. Rinkoff, S. S. and Spring, M., Toxic depression of the myeloid elements following therapy with the sulfonamides: report of 8 cases, Ann. Intern. Med., 15, 89, 1941. 184. Arneborn, P. and Palmblad, J., Drug-induced neutropenia in the Stockholm region 1973-75, Acra Med. Scand., 204, 283, 1978. 185. Arneborn, P. and Palmblad, J., Drug-induced neutropenias in the Stockholm region 1976-1977, Acra Med. Scand., 206, 241, 1979. 186. Sheehan, J., Trimethoprim-associated marrow toxicity, Lancet, 2, 692, 1981. 187. Asmar, B. I., Maqbool, S., and Dajani, A. S., Hematologic abnormalities after oral trimethoprim-sulfamethoxazoletherapy in children, Am. J . Dis. Child.,135, 1100, 1981. 188. Lawrence, B., Santen, R. J., Lipton, A., Harvey, H. A., Hamilton, R., and Mercurio, T., Pancytopenia induced by aminoglutethimide in the treatment of breast cancer, Cancer Treat. Rep., 62, 1581,1978. 189. Gez, E. and Sulkes, A., Aminoglutethimide-induced leucopenia: a case report and review of the literature, Oncology, 41, 399, 1984. 190. Lonning, P. E. and Kvinnsland, S., Mechanisms of action of aminoglutethimide as endocrine therapy of breast cancer, Drugs, 35,685, 1988. 191. Harvey, R. L. and Luzar, M. J., Metoclopramide-inducedagranulocytosis, Ann. Intern. Med., 108,214, 1988. 192. Manoharan, A., Metoclopramide-induced agranulocytosis, Med. J . Aust., 149,508, 1988. 193. Rab, S. M. and A i m , M. N., Severe agranulocytosis during paraaminosalicylic acid therapy, Br. J . Dis. Chest, 64, 164, 1970. 194.Das, K. M., Eastwood, M. A., McManus, J. P. A., and Sircus, W., Adverse reactions during salicylazosdfapyridhe therapy and the relation with drug metabolism and acetylator phenotype, N . Engl. J . Med., 289, 491, 1973. 195.Jacobson, I. M., Kelsey, P. B., Blyden, G. T., Deinkjhn, Z. N., and Isselbacher, K. J., Sulfasalazine-induced agranulocytosis, Am. J . Gasrroenterol., 80, 118, 1985. 196. Wheelan, K. R., Cooper, B., and Stone, M. J., Multiple hematologic abnormalities associated with sulfasalazine, Ann. Intern. Med., 97, 726, 1982. 197. Abrams, S. M.,Degnan, T.J., and Vinceguerra, V., Marrow aplasia following topical application of chloramphenicol eye ointment, Arch. Intern. Med., 140,576, 1980.

1990

198. Keiser, G., Cooperative study of patients treated with thiamphenicol. comparative study of patients treated with chloramphenicoland thiamphenicol, Postgrad. Med. J . , 50, 143, 1974. 199. Manyan, D. R., Arimura, G. K., and Yunis, A. A., Comparative metabolic effects of chloramphenicol analogues,Mol. Pharmacol., 11, 520, 1975. 200. Scheline, R. R., Metabolism of foreign compounds by gastrointestinal microorganisms, Pharmucol. Rev., 25,451, 1973. 201. Ascherl, M., Eyer, P., and Kampffmeyer, H.,Formation and disposition of nitrosochloramphenicolin rat liver, Biochem. P h a m c o l . , 34, 3755, 1985. 202. Yunis, A. A., Miller, A. M., Salem, Z., Corbett, M. D., and Arimura, G. K., Nitroso-chloramphenicol:possible mediator in chlorampbenicol-inducedaplastic anemia, J. Lab. Clin. Med., 96,36,1980. 203. Gross, B. J., Branchtlower, R. V., Burke, T. R., Lees, D. E., and Pow, L. R., Bone marrow toxicity in vitro of chloramphenicol and its metabolites, Toxicol. Appl. Pharmacol., 64, 557, 1982. 204. Jimenez, J. J., Arimura, G. K., Abou-Khalil, W. H.,Isildar, M., and Yunis, A. A., Chloramphenicol-induced bone marrow injury: possible role of bacterial metabolites of chloramphenicol, Blood, 70, 1180, 1987. 205. Isildar, M., Abou-Khalil, W. H.,Jimenez, J. J., Abou-Khaliil, A., and Yunis, A. A., Aerobic nitroreduction of dehydrochloramphenicol by bone marrow, Toxicol. Appl. Pharmacol., 94, 305, 1988. 206. Bilezikian, S. B., Laleli, Y., Tsan, M., Hodkinson, B. A., Ice, S., and McIntyre, P. A., Immunological reactions involving leukocytes. 111. Agranulocytosis induced by antithyroid drugs, Johns Hopkins Med. J . , 138, 124, 1976. 207. Guffy, M. M., Goeken, N. E., and Burns, C. P., Granulocytotoxic antibodies in a patient with propylthiouracil-induced agranulocytosis, Arch. Intern. Med., 144, 1687, 1984. 208. Wall, J. R., Fang, S. L., Kuroki, T.,Ingbar, S. H., and Braverman, L. E., In vitro immunoreactivity to propylthiouracil, methimazole, and carbimazole in patients with Graves’ disease: a possible cause of antithyroid drug-induced agranulocytosis,J. Clin.Endocrinol. Metab., 58, 868, 1984. 209. Fibbe, W. E.,Claas, F. H. J., Van der Star-Dijrstra, W., sehaafsma, M. R., Meyboom, R. H. B., and Falkenburg, J. H. F., Agranulocytosis induced by propyltbiouracil: evidence of a drug dependent antibody reacting with granulocytes, monocytes and haematopoietic progenitor cells, Br. J . H a e m t o l . , 64, 363, 1986. 210. Toth, E. L., Mant, M. J., Shamim, S., and Ginsberg, J., Propylthiouracil-induced agranulocytosis: an unusual presentation and a possible mechanism, Am. J . Med., 85, 725, 1988. 211. Moreb, J., Shemesh, O., Shilo, S., Manor, C., and Hershko, S., Transient methimazole-inducedbone marrow aplasia: in vitro evidence for a humoral mechanism of bone marrow suppression, Acfa Haematol., 69, 127, 1983. 212. Amann, F. W., Buhler, F. R., Conen, D., Brunner, F., Ritz, R., and Speck, B., Captopril-associated agranulocytosis, Lancet, 1, 150, 1980. 213. van Brummelen, P.,Willemze, R., Tan, W. D., and Thompson, J., Captopril-associated agranulocytosis, Lancet. 1, 150, 1980. 214. Forslund, T., Borgmastars, H., and Fybrquist, F., Captopril-associated leukopenia confirmed by rechallenge in patient with renal failure, Lancet, 1, 166, 1981. 215. Hammond, W. P., Miller, J. E., Starkebaum, G., Zweerink, H. J., Rosenthal, A. S., and Dale, D. C., Suppression of in vitro granulocytopoiesis by captopril and penicillamine, Exp. Hematol., 16, 674, 1988. 216. Cuthbert, M. F., Adverse reactions to non-steroidal antirheumatic drugs, Curr. Med. Res. Opinion, 2, 600,1974. 217. Feldman, E. J. and Isenberg, J. I., Effects of metiamide on gastric acid hypersecretion, steatorrhea and bone-marrow function in a patient

233


Critical Reviews In with systemic mastocytosis, X . Enpl. J . Mad., 295. 1178, 1976. 218. Fleischer, D. and Samloff, I. M., Cimetidine therapy in a patient with metiamide-induced agranulocytosis. N . Engl. J . Med.. 296, 342, 1977. 2 19. Pisciotta, A. V., Phenytoin: hematological toxicity, in Antiepileptic Drugs, Woodbury. D. M . . Ed.. Raven Press. New York, 1982. 257. 220. Taetle, R., Lane, T. A., and Mendelsohn, J., Drug-induced agranulocytosis: in virro evidence for immune suppression of granulopoiesis and a cross-reacting lymphocyte antibody, Blood. 54, 501. 1979. 221. Ahbott, J. A. and Schwab, R. S., The serious side effects of the newer antiepileptic drugs: their control and prevention, N. Engl. J . Med., 242. 943, 1950. 222. Dessypris. E. N., Redline, S., Harris, J. W., and Krantz, S. B., Diphenylhydantoin-inducedpure red cell aplasia, Blood. 65, 789, 1985. 223. Inman, W. H. W., Study of fatal bone marrow depression with special reference to phenylbutazone and oxyphenbutazone, Br. Med. J . , I , 1500, 1977. 224. International Agranulocytosis and Aplastic Anemia Study, Risks of apranulrxytosis and aplastic anemia: a first report with special reference to analgesics, JAMA, 256, 1749, 1986. 225 Weissmann, G . and Xefteris, E. D., Phenylbutazone leukopenia, Arch Intern. Med.. 103. 957, 1959. 222h. Kadar. D. and Kalow, W., Acute and latent leukopenic reaction to antipyrine. Clin. Phurmcicol. Ther., 28. 820, 1980. 227. Servant, D., Danel, T., and Goudemand, M., Lithium therapy in carbamazepine-induced agranulocytosis, Am. J . Psychiatry, 145. 381, 19x8

22X. Curson, D. A, and Hale, A. S., Mianserin and agranulocytosis, Br. Med. J . , 1, 378, 1979. 229. McHarg, A. M. and McHarg, J. F., Leukopenia in association with mianserin, Br Med. J . , I . 623. 1979. 230. Page, C. E., Mianserin-induced agranulocytosis. Br. Med. J . , 284, 1912, 1982. 23 I . Senn, H. J., Jungi, W.F., Kunz, H., and Poldinger, W., Clozapine and agranulocytosis, Lancer. 1. 547, 1977. 2 3 2 . Idanpaan-Heikkila, J., Alhava, E., Olkinuora, M., and Palva, I. P., Agrdnulrxytosis during treatment with clozapine, Eur. J . Clin. Phurmacol.. 11, 193. 1977. 233. Mindham, R. H. S., The major tranquilizers. in Side Effects ofDrugs Annual 2, Dukes. M. N . G . , Ed.. Excerpta Medica. Amsterdam, 1978, 45 234. Bigby, M., Jick, S., Jick, H., and Arndt, K., Drug-induced cutaneous reactions: a report from the Boston Collaborative Drug Surveillance Progam on 15.438 consecutive inpatients, 1975-1982, JAMA, 256, 3358. 1986. 235. Lyell, A., A review of toxic epidermal necrolysis in Britain. Br. J . Dermatol.. 79. 662, 1967. 2.36. Shear, N. H.. Spielberg. S. P., Grant, D. M., Tang, B. K., and Kalow, W., Differences in metabolism of sulfonamides predisposing to idiosyncratic toxicity. Ann. Intern. Med.. 105, 179, 1986. 2.17 Rieder, iM. J., Uetrecht, J., Shear, N. H., Cannon, M., Miller, M.,and Spielberg, S. P., Diagnosis of sulfonamide hypersensitivity reactions bv in-vitro “rechallenge” with hydroxylamine metabolites. Ann. Intern. Mrd., 110. 286. 1989. 238. Shear, N. H. and Spielherg, S. P., In vitro evaluation of a toxic metabolite of sulfadiazine, Can. J . Phyuiol. Pharmacol., 63, 1370, 1985. 239. Rieder, M. J.. Uetrecht, J., Shear, N. H.,and Spielberg, S. P., Synthesis and in vitro toxicity of hydroxylamine metabolites of sulfonamides. J . Phurmacol. Exp. Ther.. 244, 724, 1988. 240. Dujovne, C. A., Chan, C. H., and Zimmerman, H. J., Sulfonamide hepatic injury: review of the literature and report of a case due to sulfamethoxazole, N . Engl. J . Med., 277, 785, 1967. 241. Harle, D. G., Baldo, B. A., Smal, M. A., and van Nunen, S. A.,

234

An immunoassay for the detection of IgE antibodies to trimethoprim in the sera of allergic patients, Clin. Allergy, 17, 209, 1987. 242. Svanbom, M., Rombo, L., and Gustafsson, L., Unusual pulmonary reaction during short term prophylaxis with pyrimethamine-sulfadoxine (Fansidar), Br. Med. J . , 288, 1876, 1984. 243. Selby, C. D., Ladusans, E. J., and Smith, P. G., Fatal multisystemic toxicity associated with prophylaxis with pyrimethamine and sulfadoxine (Fansidar), Br. Med. J . , 290, 113, 1985. 244. Pearl, R. K., Nelson, R. L., Prasad, M. L., Orsay, C. P., and Abcarian, H., Serious complications of sulfasalazine, Dis. Colon Recrum, 29, 201, 1986. 245. Jones, G. R. and Malone, D. N. S., Sulphasalazine induced lung disease, Thorax, 27, 713, 1972. 246. Davies, D. and MacFarlane, A., Fibrosing alveolitis and treatment with sulfasalazine, Gut, 15, 185, 1974. 247. Thomas, P., Seaton, A., and Edwards, J., Respiratory disease due to sulfasalazine, Clin. Allergy, 4, 41, 1974. 248. Tydd, T. F. and Dyer, N. H., Sulphasalazine lung, Med. J . Ausr., 1. 570, 1976. 249. Leickly, F. E. and Buckley, R. H.,Development of IgA and IgG2 subclass deficiency after sulfasalazine therapy, Clin. Lab. Obser., 108, 481, 1986. 250. Powers, N. G. and Carson, S. H., Idiosyncratic reactions to phenytoin, Clin. Pediarr., 26, 120, 1987. 251. Kahn, H. D., Faguet, G. B., Agee, J. F., and Middleton, H. M., Drug-induced liver injury: in virro demonstration of hypersensitivity to both phenytoin and phenobarbital, Arch. Intern. Med., 144, 1677, 1984. 252. Taylor, J. W., Stein, M. N., Murphy, M. J., and Mitros, F. A., Cholestatic liver dysfunction after long-term phenytoin therapy, Arch. Neurology, 41, 500, 1984. 253. Aaron, J. S., Bank, S., and Ackert, G., Diphenylhydantoin-induced hepatotoxicity, Am. J . Gastroenferol, 80, 200, 1985. 254. Brown, M. and Schubert, T., Phenytoin hypersensitivity hepatitis and mononucleosis syndrome, J . Clin. Gastroenrerol., 8, 469, 1986. 255. Gaffey, C. M.,Chun, B., Harvey, J. C., and Manz, H. J., Phenytoin-induced systemic granulomatous vasculitis, Arch. Pathol. Lab. Med., 110, 131, 1986. 256. Chamberlain, D. W., Hyland, R. H., and Ross, D. J., Diphenylhydantoin-induced lymphocytic interstitial pneumonia, Chest, 90, 458, 1986. 257. Spielberg, S. P., Gordon, G. B., Blake, D. A., Goldstein, D. A., and Herlong, H. F., Predisposition to phenytoin hepatotoxicity assessed in virro, N . Engl. J . Med.. 305, 722, 1981. 258. Frey, H. M., Gershon, A. A., Bordowsky, W., and Bullock, W. E., Fatal reaction to dapsone during treatment of leprosy, Ann. Intern. Med., 94, 717, 1981. 259. Kromann, N. P., Vilhelmsen, R., and Stahl, D., The dapsone syndrome, Arch. Dermatol, 118, 531, 1982. 260. Wille, R. C. and Morrow, J. D., Case report: Dapsone hypersensitivity syndrome associated with treatment of the bite of a brown recluse spider, Am. J . Med. Sci., 296, 270, 1988. 261. Grossman, S. J. and Jollow, D. J., Role of dapsone hydroxylamine in dapsone-induced hemolytic anemia, J . Pharmacol. Exp. Ther., 244, 118, 1988. 262. Jones, G. W., George, T. L., and Bradley, R. D., Procainamideinduced hemolytic anemia, Transfusion, 18, 224, 1978. 263. Rotmensch, H. H., Yust, I., Siegman-Igra, Y., Liron, M., Ilie, B., and Vardmon, N., Granulomatous hepatitis: a hypersensitivity response to procainamide, Ann. Intern. Med., 89, 646, 1978. 264. Houston, B. D., Crouch, M. E., Brick, J. E., and DiBartolomeo, A. G., Apparent vasculitis associated with propylthiouracil use, Arrhriris Rheum., 22, 925, 1979. 265. Vasily, D. B. and Tyler, W. B., Propylthiouracil-inducedcutaneous

Volume 20, Issue 4


Toxicology vasculitis: case presentation and review of the literature, JAMA, 243, 458, 1980. 266. Carrasco, M. D., Riera, C., Clotet, B., Grifol, M., and Foz, M., Cutaneous vasculitis associated with propylthiouracil therapy, Arch. Intern. Med., 147, 1677, 1987. 267. Mihas, A. A., Holley, P., Koff, R. S., and Hirschowitz, B. I., Fulminant hepatitis and lymphocyte sensitization due to propylthiouracil, Gastroenterology, 70, 770, 1976. 268. Coleman, J. W., Yeung, J. H. K., Roberts, D. H., Breckenridge, A. M., and Park, B. K., Drug-specific antibodies in patients receiving captopril, Br. J. Clin. Pharm., 22, 161, 1986. 269. Peterson, H., Association of trimethadione therapy and myastenia gravis, N. Engl. J. Med., 274, 506, 1966. 270. Haynes, R. C. and Murad, F., Thyroid and antithyroid drugs, in Goodman and Gilman’s the Pharmacological Basis of Therapeutics, Gilman, A. G.,Goodman, L. S., Rall, T. W., and Murad, F., Eds., Macmillan, New York, 1985, 1389. 271. Takayama, S., Aihara, K., Onodera, T., and Akimoto, T., Antithyroid effects of propylthiouracil and sulfamonomethoxine in rats and monkeys, Toxicol. Appl. Pharmacol., 82, 191, 1986. 272. Smellie, J. M., Bantock, H. M., and Thompson, B. D., Co-trimoxazole and the thyroid, Lancet, 2, 96, 1982. 273. Hughes, S. W. M. and Burley, D. M., Aminoglutethimide: a ‘sideeffect’ turned to therapeutic advantage, Postgrad. Med. J., 46, 409, 1970. 274. Brown, C. G., Fowler, K. L., Nicholls, P. J., and Atterwill, C., Assessment of thyrotoxicity using in vitro cell culture systems, Food Chem. Toxicol., 24, 557, 1986. 275. Griciute, L. and Tomatis, L., Carcinogenicity of dapsone in mice and rats, Int. J. Cancer, 25, 123, 1980. 276. Nishiyama, S., Matsukura, M., Fujimoto, S., and Matsuda, I., Reports of two cases of autoimmune thyroiditis while receiving anticonvulsant therapy, Eur. J. Pediatr., 140, 116, 1983. 277. Gupta, A., Waldhauser, L., Reider, M., Harper, P., Daneman, D., Eggo, M., Shear, N., Uetrecht, J., and Spielberg, S., Druginduced hypothyroidism: the thyroid as a target organ in hypersensitivity reactions, SOC. Pediatr. Res., 23, 277, 1988. 278. Lang, P.G., Sulfones and sulfonamides in dermatology today, J. Am. Acad. Dermatol., 1, 479, 1979. 279. Bernstein, J. E. and Lorincz, A. L., Sulfonamides and sulfones in dennatologic therapy, Int. J. Dermatol., 20, 81, 1981. 280. Katz, S. I., Commentary: sulfoxone (diasone) in the treatment of dermatitis herpetifomis, Arch. Dermatol., 118, 809, 1982. 281. Swinson, D. R., Zlosnick, J., and Jackson, L., Double-blind trial of dapsone against placebo in the treatment of rheumatoid arthritis, Ann. Rheum. Dis., 40,235, 1981. 282. Grindulis, K. A. and McConkey, B. Rheumatoid arthritis: the effects of treatment with dapsone on hemoglobin, J. Rheumatol., 11, 776, 1984. 283. Stendahl, O., Molin, L., and Dahlgren, C., The inhibition of polymorphonuclear leukocyte cytotoxicity by dapsone: a possible mechanism in the treatment of dermatitis herpetifomis, J. Clin. Invest., 62, 214, 1978. 284. Wozel, G. and Barth, J., Current aspects of modes of action of dapsone, Int. J. Dermatol., 27, 547, 1988. 285. Klotz, U., Mauer, K., Fischer, C., and Heinkel, K., Therapeutic efficacy of sulfasalazine and its metabolites in patients with ulcerative colitis and Crohn’s disease, N. Engl. J. Med., 303, 1499, 1980. 286. Dull, B. J., Salata, K., Van Langenhove, A., and Goldman, P., 5-Aminosalicylate: oxidation by activated leukocytes and protection of cultured cells from oxidative damage, Biochem. Pharmacol., 36,2467, 1987. 287. Craven, P. A., Pfanstiel, J., Saito, R., and DeRubertis, F. R., Actions of sulfasalazine and 5-aminosalicylic acid as reactive oxygen

scavengers in the suppression of bile acid-induced increases in colonic epithelial cell loss and proliferative activity, Gastroenrerology, 92, 1998, 1987. 288. Miyachi, Y., Yoshioka, A., Imamura, S., and Niwa, Y., Effect of sulfasalazine and its metabolites on the generation of reactive oxygen species, Gut, 28, 190, 1987. 289. Hallgren, R., Colombel, J. F., Dahl, R., Fredens, K., Kruse, A., Jacobsen, N. O., Venge, P., and Rambaud, J. C., Neutrophil and eosinophil involvement of the small bowel in patients with celiac disease and Crohn’s disease: studies on the secretion rate and immunohistochemical localization of granulocyte granule constituents, Am. J. Med., 86, 56, 1989. i l l , 290. Neumann, V. C., Taggart, A. J., Le Gallez, P., Astbury, C., H J., and Bud, H. A., A study to determine the active moiety of sulphasalazine in rheumatoid arthritis, J. Rheumatol., 13, 285, 1986. 291. Pinals, R. S., Kaplan, S. B., Lawson, J. G., and Hepburn, B., Sulfasalazine in rheumatoid arthritis: a double-blind, placebo-controlled trial, Arthritis Rheum., 29, 1427, 1986. 292. Cuperus, R. A., Muijsers, A. O., and Wever, R., Antiarthritic drugs containing thiol groups scavenge hypochlorite and inhibit its formation by myeloperoxidase from human leukocytes: a therapeutic mechanism of these drugs in rheumatoid arthritis, Arthritis Rheum., 28, 1228, 1985. 293. Wasil, M., Halliwell, B., Morehouse, C. P., Hutchison, D. C. S., and Baum, H., Biologically-significant scavenging of the myeloperoxidase-derived oxidant hypochlorous acid by some anti-inflammatory drugs, Biochern. Pharmacol., 36, 3847, 1987. 294. Walzer, C. and Frenk, E., Cytochemical expression of epidermal peroxidase and cytochrome oxidase activities in pathological skin conditions of man, Arch. Dermatol. Res., 278, 460, 1986. 295. Khalaf, M. R. and Hayhoe, F. G. J., Cytochemical demonstration of ‘platelet’ peroxidase at the light microscope level, Eur. J. Haematol., 39, 128, 1987.

1990

235

Idiosyncratic drug reactions: a mechanistic evaluation of risk factors.

The role of leukocyte-generated reactive metabolites in the pathogenesis of idiosyncratic drug reactions.

In vitro testing for diagnosis of idiosyncratic adverse drug reactions: Implications for pathophysiology.

Idiosyncratic drug hepatotoxicity: a 2008 update.

Autoantibody specificity in drug-induced lupus and neutrophil-mediated metabolism of lupus-inducing drugs.

Colourants and drug reactions.

Nuclear Receptors in Drug Metabolism, Drug Response and Drug Interactions.

Pathogenesis of idiosyncratic drug-induced liver injury and clinical perspectives.

Drug-drug interactions with oral anti-HCV agents and idiosyncratic hepatotoxicity in the liver transplant setting.

Epidemiology and risk factors for idiosyncratic drug-induced liver injury.

Clinical perspectives in drug safety and adverse drug reactions.

Role of genetics and drug metabolism in human cancer risk.

Role of cytochromes P450 in drug metabolism and hepatotoxicity.

Adverse drug reactions and secrecy.

Drug reactions and the lung.

Drug monitoring and adverse reactions.

Fetal drug metabolism and its possible clinical implications.

Influence of xenobiotics on drug metabolism and its sensitive detection.

Molecular docking to identify associations between drugs and class I human leukocyte antigens for predicting idiosyncratic drug reactions.

Drug metabolism and hepatotoxicity.

The idiosyncratic drug-induced gene expression changes in HepG2 cells.

Chrysin and its emerging role in cancer drug resistance.

Adverse reactions caused by drug metabolites.

[Drug reactions caused by chemotherapy agents].