Transmitter Systems in the Eating Disorders
Timothy D. Brewerton
Investigations into the role of neurotransmitters and other neuromodulators in the eating disorders have been quite productive in the last few years, especially during the last decade. The eating disorders occur in a relatively small cohort of people, primarily young women and girls, across many cultures, and they are known to be associated with significant mortality and morbidity, both medical and psychiatric (Walsh and Devlin, 1998; Becker et al., 1999). Despite popular beliefs, there is no convincing evidence that cultural factors alone cause eating disorders. Recent data clearly identify strong genetic factors in the aetiology of both anorexia nervosa (AN) and bulimia nervosa (BN), which appear to share common genetic vulnerabilities (Lilenfeld et al., 1998; Strober et al., 2000) possibly linked to obsessionality, perfectionism, anxiety and/or behavioural inhibition (Halmi et al., 2000; Kaye et al., 1999a, 1999b). Such information implicates primary neurotransmitter disturbances in both disorders. One powerful piece of evidence to support monoamine involvement in the eating disorders is the observation that antidepressant medications can be beneficial in controlled studies, not only in BN patients, but also in recovered AN patients (Kaye et al., 1998b).
However, it is also clear mat a number of secondary disturbances come about as a result of the profound eating disturbances characteristic of these disorders (Brewerton, 1995), which then in turn only worsen or perpetuate signs and symptoms (Pollice et al., 1997). This perspective, taken together with the disorders' consequences, challenges and costs, compels us towards a better understanding of the biological mechanisms underlying all stages and types of eating disorders. The identification of the psychobiological underpinnings of these conditions may be useful in many ways, including the development of improved medical and psychopharmacologic interventions, improved education and psychotherapy for patients and their families, and improved prevention efforts at a primary level.
It must be emphasized that most measurements of neurotransmitter function provide only a glimpse into the state of the organism at that moment in time. Sorting out what is trait- and what is staterelated has been a challenging focus of neurotransmitter research in the eating disorders.
The classical monoaminergic neurotransmitter systems, including serotonin (5-HT), norepinephrine (NE), and dopamine (DA), have been fairly extensively studied in me eating disorders using available techniques in biological psychiatry. Most of these studies have been conducted during the disease state, which has the disadvantage of being confounded by severe nutritional compromise. Dieting and/or semi-starvation clearly deplete central monoamines and lead to altered neurotransmitter levels and receptor sensitivity in animals and humans (Cowen et al, 1996; Cowen and Smith, 1999; Goodwin et al, 1987a, 1987b, 1990). To avoid this problem a more recent strategy has been to study 'recovered' eating disorder patients, i.e., AN and BN patients who have attained normalization of eating and weight, resumption of menses and/or normalization of gonadal hormone levels, and abatement of typical cognitive features to sub-clinical levels. This strategy attempts to minimize starvationstate-related effects and to reveal potential trait-related disturbances or vulnerabilities. However, the long-term effects of chronic malnutrition and disordered eating behaviours on the brain (similar to substance use disorders) should not be underestimated. Studies of transmitter function in at-risk pre-morbid individuals as well as non-affected identical and fraternal twins, siblings, and other firstdegree relatives of ED patients, could begin to confirm trait-related disturbances.
Neurotransmitter function in patients with EDs have been investigated using a variety of existing techniques and methodologies, each of which has its own advantages and disadvantages. Studies of cerebrospinal fluid (CSF) concentrations of the major metabolites have been a popular strategy and include measures of 5-hydroxyindoleacetic acid (5-HIAA) for serotonin (5-HT), 3-methoxy–4-hydroxyphenylglycol (MHPG) for NE, and homovanillic acid (HVA) for DA. Some studies have also examined actual concentrations of 5-HT and NE, but not DA. Such studies measure transmitter metabolism of the whole brain and spinal cord and lack any anatomical specificity.
Neuroendocrine and other psychobiological response measures have been studied following acute challenges with various agents, including amino acid precursors, e.g., L-tryptophan (L-TRP) and 5-hydroxytryptophan (5-HTP) for 5-HT, pre-synaptic receptor agonists, e.g., dl-fenfluramine (dl-FEN) or d-fenfluramine (d-FEN) for 5-HT, post-synaptic receptor agonists, e.g., mchlorophenylpiperazine (m-CPP) for 5-HT and isoproterenol (ISOP) for NE. Longer term challenges with receptor antagonists, e.g., antipsychotics for DA and 5-HT, and antidepressants, especially the serotonin-specific reuptake inhibitors (SSRIs), also illuminate the role of neurotransmitters in the eating disorders. Acute amino acid precursor depletion, most notably of L-TRP (Delgado et al., 1990; Weltzin et al., 1995; Smith, Fairburn and Cowen, 1999; Kaye et al., 2000), has been another important source of information about the role of central 5-HT function in eating and related disorders.
Platelet (PLT) or leukocyte studies are possibly reflective of central neurotransmitter function but are always at least step removed from the nervous system, e.g., platelet 5-HT reuptake, 3Himipramine binding, 3H-paroxetine binding, platelet MAO, platelet 5-HT content, as well as platelet receptor mediated aggregation (5-HT2 and alpha-adrenergic).
Plasma concentrations of neurotransmitter precursors, e.g., L-TRP, L-tyrosine (L-TYR), and their competing large neutral amino acids (LNAA), neurotransmitters themselves, e.g., NE, DA,