Khat (Catha edulis) and Obesity: A Scoping Review of Animal and Human Studies

Obesity has become a real concern for individuals and communities, with consequences ranging from complications in the field of aesthetics to increased risk of major health diseases that include cardiovascular disease, diabetes mellitus and cancer [1,2,3]. The alarming global statistics on the incidence of obesity has led to a number of national and international initiatives and organisational activities aimed at minimising the risk of obesity [4,5,6,7].

Adhering to healthy food habits and lifestyles is the key to weight management, combined with the available pharmacotherapy [8,9]. Nonetheless, the withdrawal of 2 centrally acting drugs - rimonabant in 2009 and sibutramine in 2010 due to serious adverse effects [10,11] - has left physicians with only a few choices of suitable anti-obesity drugs. Therefore, there is a need for the development of new drugs with alternative mechanisms or targets. We should probably direct our attention toward studies on medicinal plants, which may direct us at promising findings of a novel drug with new or unknown molecular targets [12]. One should also consider the rise in the prevalence of obesity in low- and middle-income countries where resources are limited [13] and where the belief in complementary medicine is widespread [14].

Our review aims (1) to assess the nature of the research activity on khat and obesity and (2) to summarise and disseminate the research findings and identify research gaps in the available existing literature. This review assesses the role of dopamine and serotonin as a main focus of possible anti-obesity effects through which khat acts. Khat (Catha edulis) studies in literature, either in animals or humans, are collected, summarised and appraised to address the current knowledge of obesity-related physical and hormonal effects that khat could change.

The nature of this review is to establish the state of knowledge on khat regarding obesity, and therefore we used the approach of scoping the literature [15]. We considered quantitative studies. A systematic search for khat literature to meet the objectives of our review was performed in September 2016 using 4 electronic databases (PubMed, CINHAL, Cochrane Library and Web of Science) from the time of their inception. Keywords denoting khat (Catha, mirra, qat, khat and kath) were used as search terms from a relevant current systematic review [16]. In addition, khat constituents (cathinone, cathine, norephedrine and pseudoephedrine) were taken into consideration for pooling the relevant literature. The abstracts from the gathered studies were screened for the following words: body weight OR body mass index OR fat mass OR cholesterol OR triglyceride OR food intake OR appetite OR insulin OR leptin OR cortisol. We included quantitative (observational, experimental and clinical trials) studies irrespective of whether they were animal or human studies (fig. 1). Studies that were duplicates or those that were neither in English nor in Arabic or those that had no data or those that were related to toxicity, cancer, infections, pregnancy, children, or neonates and letters to the editor and reviews were excluded.

Dopamine and serotonin neurotransmitters play an important role in regulating the food intake and energy in the body. A state of dopamine dysregulation had been observed in obese rats [17,18], where diet-induced obesity causes dopamine activity to be increased in the hypothalamus due to the overexpression of tyrosine hydroxylase (TH; enzyme catalysing the rate limiting step) and dopamine transporters. An inverse relation in the striatum exists, whereby dopamine activity is attenuated due to the downregulation of these genes in addition to the low dopamine receptors (D1 and D2) of obese rats [19]. This results in an excessive food intake, which is the result of motivational stimuli. On the other hand, serotonin release in the brain stem and hypothalamus showed an inverse relationship to food intake, where serotonin depletion in the TH knockout animals is associated with high food intake, and a high serotonin level is associated with anorexia [20].

The complex neuroendocrine mechanism related to obesity also highlights that dopamine and serotonin pathways are targets for hormone and peptide regulation. Briefly, dopamine activation by an agonist to D1 or D2 enhances insulin sensitivity [21]. Moreover, insulin is also known to regulate reward pathways to food by increasing the dopamine transporter in the striatum [22], while inhibition of serotonin reuptake is associated with insulin resistance [23]. Leptin is another important hormone that regulates food intake and energy balance. Oury and Karsenty [24] described a number of studies that link the action of leptin mediated by the inhibition of serotonin synthesis and release in the brain stem and, as a consequence, alters the serotonin level in the hypothalamus. Similar to the function of insulin, leptin also suppresses the dopaminergic system in the reward pathway [25,26]. An opposite effect of the orexigenic hormone ghrelin seems to motivate food intake by increasing basic dopamine release in nucleus accumbens (NAc) [27]. Ghrelin was found to decrease serotonin release in the hippocampus area [28].

In hypothalamus, serotonin had been found to mediate appetite suppression by increasing 2 anorexigenic peptides, pro-opiomelanocortin [29] and cocaine and amphetamine-related transcript (CART) [30]. Central injection of CART showed reduced locomotor sensitivity to cocaine through the inhibition of the upregulation of the dopamine receptor [31]. Nonetheless, orexigenic peptides or agouti-related peptide injection to the brain reduces TH in dopaminergic neurons [32], while peptide YY (PYY) reduces the dopamine release in the hypothalamus [33].

A number of drugs have been investigated for treating obesity, either for the short term or the long term. These drugs act by modulating the release and uptake of monoamines mainly in the brain. The most popular drug for treating obesity during the 1940s to 1960s was amphetamine [34]. Amphetamine had been known to increase monoamine release and inhibit the reuptake of monoamines. Withdrawal of the amphetamine for this indication was due to its multiple side effects, dependence, and sudden death [34]. Following α-methyl-β-phenylethylamine (amphetamine), the development of many centrally acting drugs ensued [35]. For the sake of more clarity, a few centrally acting drugs are reviewed to highlight the common effects of these drugs on monoamines, stating the most potential adverse effects of each one to emphasise the diverse and unshared effects.

Phentermine is an amphetamine derivative that acts by the inhibition of monoamine oxidase A, that is, an enzyme that metabolises monoamines [36], thereby increasing serotonin release in the hypothalamus [37] and increasing dopamine levels in the NAc [38]. Fluoxetine (a selective serotonin reuptake inhibitor), which also has anti-obesity effects [39], has been shown to antagonise the phentermine effect [37]. Phentermine was approved as an anti-obesity agent by the FDA in 1959 for short-term management [40] and was approved again when combined with topiramate in 2012 [40]. A recent study for long-term use has shown that phentermine is not associated with addiction and drug-craving behaviour [41]. Phentermine is not associated with serious risk, although high blood pressure due to vasoconstriction and vasculopathies has been reported [42].

Fenfluramine (dexfenfluramine and levofenfluramine) is a monoamine-releasing agent. Fenfluramine releases serotonin in the striatum [43], hippocampus, and hypothalamus [44]. Dopamine release associated with fenfluramine was secondary to serotonin release [44]. In a meta-analysis of fenfluramine and dexfenfluramine, results showed an effective reduction of body weight compared to the placebo, but their efficacy lasted only for 3 months. Therefore, its long-term use has not been justified [45]. The major drawback of fenfluramine and its derivatives was the way I which they blocked potassium voltage-gated channels [46] because of which users had to experience pulmonary hypertension and cardiac valvular diseases [47].

Sibutramine is a reuptake inhibitor of serotonin and noradrenaline and, to a lesser extent, of dopamine [35]. Sibutramine gained popularity because of its effect as an appetite suppressant, resulting in the reduction of food intake and increased thermogenesis (energy expenditure) in animals [48]. Clinical studies revealed that sibutramine had better efficacy for weight reduction than the peripherally acting orlistat [49]. The maximum effect occurred in the first 4 months, but following that, weight gain was shown to occur, which was more than that observed as a result of orlistat [50]. In 2010, sibutramine was withdrawn from the market due to high incidence of cardiovascular risks associated with its use [51].

Lorcaserin is a drug that was introduced in 2012 and it was considered to be an agonist at serotonin receptors with high selectivity to the 5-HT2C receptor [52] and with minimal activity on other serotonin receptors. Lorcaserin has minimal action on dopamine and noradrenaline transporters [52,53]. The BLOOM-DM study (behavioural modification and lorcaserin for obesity and overweight management in diabetes mellitus), which was a one-year clinical trial study, reported that lorcaserin was an effective drug for obesity and is associated with better glycaemic control in diabetic patients [54]. The main adverse effect of lorcaserin as reported by the 3 clinical trials was headache, while the risk of valvulopathy was comparable to that of the placebo [55].

To summarise the above paragraphs, centrally acting anti-obesity drugs (e.g., sibutramine) seem to be more effective than peripherally acting anti-obesity drugs (e.g., orlistat). The common problem is the serious side effects they produced. The distinct action of each drug was associated with peculiar adverse effects, which may reflect the diversity of the monoamine target interactions in regulating food intake and energy expenditure. Therefore, in the light of above-mentioned literature, khat will be discussed in the following sections. It shares a very similar mechanism with the mentioned drugs with some differences. The khat action and effects on body will be discussed in depth.

Khat (Catha edulis), is a plant that has deep roots in the cultural life of East African and Southwestern Arabian populations, where traditionally most people have the habit of chewing the tender fresh leaves in the cheek for hours. In this way, it permits the mucosal absorption of 2 of the active alkaloids of khat, namely, cathinone (in fresh) and cathine (non-fresh), both of which are considered to be amphetamine-like substances. Figure 2 shows the similarity of the chemical structure of cathinone, amphetamine, and other anti-obesity drugs. Khat chewers usually favour its use in a group session at specially designed rooms in houses or public gathering places [56,57,58]. Traditionally, there are strong beliefs among khat chewers that khat has health benefits for treatment of diabetes mellitus and that it has anti-obesity effects due to the suppression of appetite. Nevertheless, there is agreement on its central effect [59], as recently, khat alkaloids have gained more attention as synthetic ‘bath salts' with stimulant effects. It has been referred to as a new drug with potential for abuse [60,61].

The phytochemical constituents of khat are alkaloids, flavonoids, tannins and sterols. These constituents vary in concentration based on the time and ground of harvesting [62]. The leaf is still under exploration; main alkaloids are cathinone, cathine and norephedrine [63]. All alkaloids showed 2-compartment model pharmacokinetics, probably due to oral and intestinal absorption. Initial absorption occurs after 12-15 min and reaches a peak plasma level within 2 h. Cathine has a longer elimination half-life (4 h) compared to that of cathinone (1.5 h) [63]. Norephedrine shows slower absorption (2.8 h). Cathinone metabolism forms R-S-norephedrine; therefore, it is difficult to determine its elimination duration [63]. A review on khat pharmacology and toxicology had reported the cardiovascular (high blood pressure, tachycardia), gastrointestinal (anorexia, constipation) and CNS (insomnia, dependence) effects [64].

The findings of the search on khat and obesity did not identify any systematic review assessing the role of khat on obesity. The available literature in the following sections explains the monoaminic action of khat or its alkaloids, followed by a scoping review of studies that addressed physical, biochemical (table 1), or hormonal (table 2) changes associated with khat, cathinone, or cathine, which are related to obesity as described in the method and research strategy section above.

Kalix [59] demonstrated that cathinone, an active compound of khat leaves, has the ability to trigger the release of labelled dopamine in a rabbit's caudate nucleus (in vitro) and that dopamine release was blocked by cocaine (known to inhibit monoamine release). Later, Kalix [65] studied the attenuated dopamine release in the NAc in the presence of uptake inhibitor drugs and explained that an intracellular penetration of cathinone is necessary to produce exocytosis of dopamine from the dopaminergic neuron. In vivo studies had indicated a reduction in the amount of 3,4-dihydroxyphenylacetic acid, a major cytoplasmic metabolite of dopamine, indicating a blockade of dopamine uptake back to neurons in the dopaminergic areas of the brain [66]. Pehek [67] replicated these results and concluded that different cathinone concentrations produced different effects on dopamine release and uptake, which reflect the heterogeneity of the striatum of the rat response to cathinone. Upon chronic administration of cathinone, dopamine depletion was observed in NAc, and this was accompanied by a reduced number of dopamine transporters [68,69].

Khat also had quite similar effects on serotonin. In vitro studies, for instance, have shown that cathinone enhances the release of serotonin from the striatum [70,71] and can bind to serotonin receptors [72]. A study by Fleckenstein et al. [73] demonstrated serotonin uptake inhibition by cathinone following single or multiple dosing. These differential effects that can be produced by psychostimulants on dopamine and serotonin from in vivo and in vitro findings are interesting [73] due to the following 2 reasons: the inconsistency between in vivo and in vitro findings and the differences in short- and long-term effects of psychostimulants on different regions of the brain (NAc vs. hypothalamus) [74]. Some studies have shown that cathinone has no effect on brain serotonin [68], while other recent studies have suggested that there is a slight effect [75,76].

Previously, we addressed the effects of khat on monoamines and intentionally overlooked the behavioural and locomotor effects and comparisons to other psychostimulants, which is out of the scope of this review. However, all related studies provided strong evidence that khat and its active alkaloid have a plausible direct and indirect influence on monoamine release and uptake, a mechanism that has been known to affect food intake. The cellular short- and long-term effects stated previously necessitate further studies at the genetic and molecular levels.

The effect of khat on the physical and biochemical parameters of obesity is summarised in table 1. Here, we critically review the findings from both animal and human studies.

Body Weight

It was clear from the available studies that khat reduces body weight in the rat animal model. The reduction of body weight was sustained in studies that used high doses or long durations. Small doses produced a short-term reduction of weight followed by the tolerance of the effect. Both cathinone and cathine (similar to many psychoactive drugs) demonstrated an acute tolerance mechanism in the previous research [77,78]. The effects of khat extracts on the body have been reported within 14 days of its use [79], while different doses of cathine had shown no effect for a period of 11 days [80]. This suggests that cathine is less active compared to the S-cathinone isoform [81].

In healthy humans, a cross-sectional study in Ethiopia reported that khat chewers were associated with less mean body weight compared to non-chewers [82]. However, khat chewers with type-2 diabetic mellitus showed higher mean body weight compared to non-chewer diabetic patients [83]. Data from both cross-sectional studies lacked previous nutritional history as well as family history of obesity. It is difficult to conclude any effect of khat on body weight in humans. However, the aforementioned studies postulate that the effect of khat extract - cathinone - or cathine could produce a change in body weight if given at high doses for a sufficient period to observe the effect.

Food Intake

Although khat use and its effect on appetite has been reviewed recently [75], it is worth re-emphasising the dosage and duration of khat. Acute studies (hours) showed consistent findings concerning the suppressant effect of cathinone and cathine (table 1). The effect of sub-chronic studies demonstrated a reduction in food intake of animals that lapsed over the first 2 weeks with low doses [79,84,] but it could be sustained with high- and slow-release doses [84].

Body Mass Index and Fat Mass Index

The body mass index (BMI) measurement was included in human studies. Most studies were cross-sectional studies that compared fresh regular khat leaf chewers for one year or more to non-chewers or ex-chewers. There was only a cohort study [85] that considered a measurement of BMI both as a baseline and after one year of khat chewing. The latter study showed an initial reduction of BMI among khat chewers compared to non-khat chewers at the baseline (6%) but increased significantly to 10.7% after a year. Another study [86] reported a gender difference where females showed higher BMI in khat chewers compared to non-chewers. All these studies did not use a rigorous standard to ensure that the nutritional status of khat chewers and non-chewers were equal or assess the presence of any family history of obesity. Nonetheless, the accumulative evidence suggests an association between khat chewing and low BMI. We suggest that any study attempting to consider BMI must consider relevant confounding factors that include previous dietary history, family history and socioeconomic status, since these factors would better reflect the comparison of BMI between khat chewers and non-chewers. The fat mass index was stated only in one study that considered the measurement of fat mass [82] and reported that khat chewing was associated with low fat mass. It is interesting to mention here that most of these studies found that khat chewers smoke more during khat sessions. Therefore, we can conclude that the effect of khat in these studies may be confounded by the concurrent smoking effect.

Cholesterol and Triglycerides

There was an agreement in most animal studies about the effect of khat extract in reducing cholesterol and triglycerides. Mahmood and Lindequist [79] showed a non-significant increase in plasma cholesterol of rats with genetically inherited metabolic syndrome, probably due to a significant increase in plasma HDL during khat administration. In the later study, the khat was used for 14 days and observation followed for another 14 days after cessation of khat intake. The triglyceride count was raised after cessation of khat, but it was not significant. These studies used different khat extraction techniques, and the plant was harvested from different regions in Yemen; this could have changed the constituent concentrations of cathinone and cathine in each extract. A difference in the part of the khat extract was a phenomenon that had been observed by Mahmood and Lindequist [79]. The finding could also reflect gender differences in responses between female and male rats. The previous sustained effect of khat when it is given as a high dosage was obvious from the listed studies. A study on humans reported that in one measurement, which was taken 4 h after khat administration, it was found that there was no effect of fresh khat on the plasma cholesterol and triglycerides of male khat chewers [87]. This makes us suspicious about this results that was manifested after this short time, but it could be inferred that chronic khat use may cause the reduction of both plasma cholesterol and triglycerides.

There are many hormones associated with obesity. Unfortunately, few studies have been conducted on the effect of khat on hormones. Hormonal changes associated with khat are summarised in table 2.

Leptin and Insulin

Leptin is considered an important discovery (1994) that has changed our understanding of obesity. Leptin is a cytokine-like hormone, secreted from white adipose tissue to inhibit anabolism and enhance catabolism, primarily by its action on specific leptin receptors (LEPR-R) [88,89]. There are only 2 studies that link khat effects to leptin. In animals [79], khat extract was shown to reduce circulating plasma leptin. On the other hand, the human study [87] demonstrated an increased plasma leptin in khat chewers compared to non-chewers, which was marked with high doses. Both studies tried to explain the modulating gut hormone and gastric-emptying effect on the changing leptin secretion. However, this explanation is inconsistent with the findings of a study that reported no effect of fresh khat on ghrelin and PYY levels [90]. Murray et al. [90] recruited 6 habitual khat chewers and khat administration was done in a biweekly manner.

This finding highlighted a number of questions regarding the complex regulation and effect of leptin at different doses of khat as well as species variation. The molecular effect of khat or its active alkaloids - cathinone, cathine, and norpseudoephedrine - on adipocyte or the stomach when secreting leptin has not yet been studied. Leptin's anorexic effect is suggested to be mediated by increased dopamine concentration in NAc to decrease reward stimuli as well as the negative effect on brainstem serotonin release as stated earlier. The peripheral effect of khat, cathinone, cathine, or norpseudoephedrine cannot be ruled out.

Our review of insulin level and khat use found a study that measured plasma insulin [79]. This study reported a decrease of the insulin plasma level associated with a decrease in the plasma glucose in rats. The effect was more pronounced with the dark khat than light khat extracts. Probably khat increased motor activity in the animals [91,] making muscles to adsorb glucose in the absence of insulin. An in depth review regarding khat and glycaemic control has been published recently [62].

Cortisol and Sex Hormones

Cortisol is an important hormone secreted by the adrenal gland and it plays an important role in energy expenditure and digestion [92.] It is regulated by the corticotrophin releasing factor (CRF; hypothalamic regulation) and adrenocorticotropic hormone (ACTH; pituitary regulation) [92]. Some amount of ACTH is produced from the melanocyte stimulating hormone [92]. It is known that the hypothalamic-pituitary-adrenal axis is modulated by the sex hormone [93]. Serotonin levels in the brain were associated with the modulation of sex hormones and cortisol [94].

Cortisol and khat have shown conflicting results. In monkeys, cortisol is reduced in male baboons fed with khat extract [95], and cathinone administration is associated with time- and dose-dependent reduction of cortisol in vervet monkeys [96]. In the former study, baboons also showed an increased testosterone level, which suggests the inhibition of cortisol release [94]. In contrast, rabbits showed a dose-dependent increase in cortisol with khat extract administration, simultaneously with the reduction of luteinising hormone (LH) and testosterone [97]. In rats, the findings were similar to rabbits. At a high dose of khat extract without cathinone, cortisol was high with low testosterone [98]. Fourie et al. [99] explained a variation of the cortisol level in both sexes of yellow and olive baboons in relation to gender and the environment. In rodents, cortisol tends to be higher in females due to glucocorticoid stimulation by ACTH [93]. The observed increase of cortisol in male rabbits [97] and male rats [98] was probably caused by CRF stimulation as a response to decreased testosterone and LH under the influence of khat-induced high serotonin levels in the brain.

In humans, Al'Absi et al. [100] described dysphorphic mood changes in khat chewers and a blunted diurnal pattern of cortisol. The interesting finding was in the latter study, although it was not significant. However, the elevation of cortisol was higher among females than males. A similar trend of cortisol variation in association with stress, but not with khat, has been described by [101]. Lovallo et al. [102] also found cortisol variation due to genetic polymorphisms.

All the aforementioned studies are inconclusive regarding the effect of khat or cathinone on cortisol and sex hormone levels; therefore, predicting a relationship would be a complex task due to the inability of differentiating between cortisol changes caused by khat action or stress induced by its use [100]. It would be recommended to evaluate the khat effect on sex hormones and cortisol separately, such as in cell line studies. In addition, more precise research about the action of khat extract or its alkaloids on cortisol regulation should be conducted. Any study should consider the confounding factors of stress and cortisol.

Prolactin

Prolactin is another important hormone that plays a role in energy expenditure and lipid metabolism in adipose tissue [103]. Prolactin is mainly regulated by the level of dopamine in the hypothalamus; thus, drugs that reduce dopamine produce hyperprolactinemia [104]. All available studies of khat were in agreement with the described mechanism of khat on dopamine. Additionally, it was observed that different doses of khat extract or cathinone were associated with a reduction in the prolactin level (table 2).

The reviewed studies provide relatively acceptable evidence that different khat extracts or cathinone produced changes in weight, fat mass, appetite, lipid biochemistry and hormones. These changes were more pronounced at high doses and long durations of intervention. The most suggested mechanism of these changes is the central action that produces changes in the physiology of dopamine and serotonin. Nonetheless, there are a number of variations in the study design, species, doses and durations of intervention that would make it difficult to arrive at a final conclusion about khat's role in reducing obesity. We suggest further studies could be carried out in the future to overcome these limitations.

The authors would like to thank the University of Malaya, Malaysia for supporting this study by grant No. RG539-13HTM.

The authors declare no conflicts of interest.