From Herbal Roots to Synthetic Medicines: A Historical Perspective

Abstract

Alcohol is one of our most ancient intoxicants. A recent anthropological study revealed that a fermented mixed beverage of rice, honey and fruit had existed during the Neolithic era. Written history documented the manufacturing of barley beer and other cereal beverages in Egypt and China as far back as 6000 and 4000 years, respectively (El-Guebaly and el-Guebaly, 1981; McGovern et al., 2004). History also attributed the downfall of empires and dynasties to excessive alcohol use by ruling courts more than 3000 years ago. Grecian scholars 2500 years ago issued warnings about the harm of acute and chronic inebriation arising from drinking too much wine too fast and too often. Plato even provided drinking guidelines: no use under age 18, use in moderation between 18 and 30, and no restriction for those older than 40. Descriptions of modalities for the treatment of acute and chronic inebriation began appearing when alcoholic beverages became more available to the populace. In China, herbal medicines were recommended to combat the progression of what was a common saying: “first a man takes a glass, then the glass takes a glass, finally the glass takes the man” (Sournia, 1990). Many of these herbal medicines contained extracts from the Pueraria plant, commonly known as Kudzu. In Western medicine, disulfiram (Antabuse®), the first medication to treat chronic inebriation, now called alcoholism, alcohol dependence, or alcohol addiction, was approved by the US Food and Drug Administration (FDA) in 1949. Discovered through astute serendipitous observation, Hald and Jacobsen (1948) found that disulfiram (tetraethylthiuram disulphide) used in the rubber industry evoked an aversive reaction in workers who subsequently ingested alcohol. Although the underlying mechanism was unknown at the time, the authors showed that disulfiram plus ingested alcohol produced elevation of blood acetaldehyde concentration and that infusion of acetaldehyde evoked the same alcohol-disulfiram reaction. Subsequent work showed that disulfiram inhibited both cytosolic and mitochondrial aldehyde dehydrogenase (ALDH) in vitro (Kitson, 1977), but it was Deitrich and Erwin (1971) who demonstrated in vivo that disulfiram administration to rats decreased ALDH activity in the liver irreversibly and that recovery was dependent upon new enzyme synthesis. Subsequent to early clinical studies reporting good results, the efficacy of disulfiram in treating all patients diagnosed with chronic relapsing alcoholism was questioned because of its failure to curb craving, its low compliance rate among patients, and its side-effects (e.g., liver toxicity). Disulfiram irreversibly inhibits ALDH2 through reaction with essential sulfhydryl groups of the enzyme; it also inactivates many other enzymes similarly, possibly contributing to undesirable side effects. Nonetheless, as recently reviewed, it may still have a role under supervised administration and in treating subsets of patients who are impulsive or who are older and better motivated (Fuller and Gordis, 2004). ALDH is an attractive medications target; it has been firmly established in East Asian populations, Chinese, Japanese, and Koreans in particular, that functional polymorphisms of alcohol dehydrogenase (ADH) and of mitochondrial aldehyde dehydrogenase (ALDH2) lead to genetically based variation both in alcohol metabolism (pharmacokinetics) and in responses to alcohol (pharmacodynamics) mediated, at least in part, through acetaldehyde. Those individuals who have alleles of ADH that code for a higher activity ADH enzyme (ADH2*2) and for low/absent ALDH2 activity (ALDH 2*2) are protected against heavy drinking and alcoholism because they have the alcohol-flush reaction, a genetic equivalent to the alcohol-disulfiram reaction. Hence alcoholism qualifies as a pharmacogenetic disorder with genetic variation as a major biological reason for individual differences in drinking behavior (Li, 2000). Interestingly, individuals who are heterozygous at the ALDH2 locus (ALDH2*1-ALDH2*2) have reduced risk of developing alcoholism, but some do become alcoholic. A recent study showed that alcoholic and nonalcoholic ALDH2 heterozygotes do not differ in alcohol and acetaldehyde pharmacokinetics, i.e., acetaldehyde is equally elevated in both groups, but the alcoholic subjects exhibit lower alcohol flush reaction responses, i.e., less pharmacodynamic sensitivity to acetaldehyde (Chen et al., 2009). In the last 25 years, investigators began investigating the purported antidipsotropic effects of isolated components of Kudzu. A report appeared in 1989 that the isoflavone fractions from the Pueraria plant when given to mice orally together with ethanol attenuated blood ethanol and acetaldehyde levels and ethanol’s effect on spontaneous motor activity (Niiho et al., 1989). These findings were replicated in rats (Xie et al., 1994). Then several reports appeared that a number of isoflavones isolated from Kudzu root suppressed ethanol intake in Syrian golden hamsters (Keung and Vallee, 1993) and in rats (Lin et al., 1996; Overstreet et al., 1996). The major active compounds that suppressed free-choice drinking were daidzin, daidzein, and puerarin (Lin et al., 1996). In humans, a study of the efficacy of Kudzu extract containing these three major isoflavones has been performed in heavy drinking subjects. The Kudzu extract produced significant reductions in total intake in a drinking session, accompanied by increases in the number of small sips and the time to consume a unit of beverage (Lukas et al., 2005). Although all the animal models studies and the human efficacy trial have shown that the major isoflavone components in Kudzu reduced free-choice alcohol drinking, their mechanism of action was unclear. Initially thought to be mediated by inhibition of ALDH2 and elevation of acetaldehyde as is seen with disulfiram, bioavailability and different effects on ethanol pharmacokinetics when administered by oral or intraperitoneal routes blurred interpretation (Keung et al., 1996; Xie et al., 1994). Since then, a clearer picture has emerged. Importantly, isoflavones are not sulfhydryl-reactive compounds. In vitro studies (Keung and Vallee, 1993) had found daidzin to be a potent and a reversible, competitive inhibitor of ALDH2. Surprisingly, they found that daidzin at the doses effective in suppressing ethanol consumption in the hamsters did not affect overall acetaldehyde metabolism in vivo (Keung et al., 1995). However, ALDH2 also catalyzes the oxidation of other aldehydes. Because ALDH2 together with monoamine oxidase (MAO) in mitochondria are responsible for the oxidative deamination of biogenic amines, Heyman and colleagues (1996) suggested that, rather than ethanol-derived acetaldehyde, it may be an endogenous physiological substrate of ALDH2 that is involved in the regulation of drinking in these experiments. In support of this hypothesis, Keung and colleagues showed that daidzin and its antidipsotropic analogs inhibited the metabolism of serotonin (5HT) and dopamine (DA) to 5-hydroxyindoleacetic acid (5-HIAA) and 3,4-dihydroxyphenylacetic acid (DOPAC) in isolated mitochondria. Concomitantly, there was an accumulation of 5-hydroxyindole-3-acetaldehyde (5HIAL) and 3, 4 dihydroxyphenyl acetaldehyde (DOPAL). Through its inhibitory action on ALDH2, the potential site of action of daidzin is ALDH2 in the MAO-ALDH2 pathway of biogenic amine metabolism. Thus, elevation of biogenic aldehydes such as 5HIAL and DOPAL, might serve as feedback regulators of neuronal 5-HT, and DA synthesis and content (Keung and Vallee, 1998; Rooke et al., 2000). If this were the case, what aspects of ethanol consummatory behavior, thus far measured only with two-bottle free-choice drinking, are being affected by the isoflavonoids in Kudzu, and is there potential for this class of compounds in the development of medications for the treatment of alcoholism? A major step forward in this journey is the article in the November 2009 issue of Alcoholism Clinical and Experimental Research (ACER), the result of collaborative research from multiple laboratories in the United States and Australia, and orchestrated by Ivan Diamond from CV Therapeutics (now Gilead) in Palo Alto, CA (Arolfo et al., 2009). Based on the X-ray crystal structure of daidzin in complex with human ALDH2 (Lowe et al., 2008), more potent and selective ALDH2 inhibitors were synthesized. One of them, CVT-10216, was studied in three different heavy drinking rat strains under several different drinking paradigms: two-bottle free-choice drinking, deprivation-induced drinking, operant response to self-administration, and cue-induced reinstatement. Alcohol gavage (2 g/kg) did increase acetaldehyde concentration in blood 2- to 7-fold indicating that ALDH2 was, indeed, inhibited by CVT-10216 in vivo under conditions of the experiment. However, it should be noted that even the highest drinking rat strains do not usually self-administer this amount of alcohol in a bolus, even in the deprivation-induced paradigm. Although not proven directly, elevation of acetaldehyde is not likely to be the mediating agent in vivo. In support of this conclusion, both operant alcohol-seeking and cue-induced reinstatement of operant alcohol-seeking behaviors were diminished with CVT 10216 administration, even in the absence of alcohol and hence of alcohol-derived acetaldehyde. Is there evidence that neuronal biogenic amine content and response to stimulation are affected? Data are presented showing that CVT 10216 does not alter the basal DA levels in the nucleus accumbens in vivo, but does prevent ethanol-stimulated release, supporting the hypothesis that one mechanism of action of CVT-10216 is, at least in part, its interference with the DA reward pathway. The question is then does the effect of CVT-10216 on biogenic amine metabolism mediated by inhibition of ALDH2 generalize to the self-administration of other drugs such as cocaine? Further, does this class of compounds also affect comorbid entities such as mood and anxiety disorders wherein there is 5HT dysfunction? Tantalizing clues can be culled from reports that: (i) puerarin reduces the anxiety induced by alcohol withdrawal and by the administration of 5 HT agonists (Overstreet et al., 2003); (ii) disulfiram diminishes cocaine-associated subjective effects (“high” and “rush”) in non-treatment-seeking cocaine-dependent volunteers (Baker et al., 2007); and (iii) a beneficial effect of disulfiram in co-occurring cocaine and alcohol dependent subjects has been observed in a number of studies, the latest being the report by Pettinati and colleagues (2008). The neurocircuitry and neurotransmitter/neuromodulator systems that mediate the development of addiction to alcohol and drugs are complex and many. Two outstanding reviews have appeared this year, one emphasizing more the neurocircuitry (Koob and Volkow, 2009) and the other the molecular physiology of alcohol-sensitive sites on receptors and ion channels (Spanagel, 2009). The latter discusses potential targets for medications development and a very comprehensive list is provided. In medications development for alcoholism, behavioral outcomes for a successful candidate should include: attenuation of the positive and negative reinforcement of ethanol self-administration, reduction of the negative affect associated with withdrawal, diminution of craving, and the prevention of relapse. Ideally, this can be accomplished with a single agent or a combination of a minimum number of agents. An ALDH2 inhibitor without the negative side effects of disulfiram attributable to inhibition of other sulfhydryl enzymes might fit this role! Is this a pipedream or attainable reality? Time and more research in animal models as was reported in the article by Arolfo and colleagues (2009) and in human subjects will tell. The assistance of Brenda G. Hewitt in the preparation of this manuscript is gratefully acknowledged.