Individuals with anorexia nervosa present severe metabolic disturbances as a consequence of abnormal eating behaviors. Alterations in biochemical parameters have been described in AN (cortisol, cholesterol, electrolytes, etc.). However, the metabolic phenotype or fingerprinting of AN has been scarcely studied. Predominantly, plasma and serum samples are analyzed due to the ease of sample acquisition and the information they provide about the metabolic status. Generally, studies are focused on small groups of metabolites such as amino acids, lipids, or carbohydrates. Hence, wide untargeted metabolomics analyses are still lacking in AN. The main metabolomics alterations found in plasma from AN patients are summarized in Figure 1 and detailed below.
Figure 1. Summary of the main metabolomic alterations found in plasma or serum samples from AN patients in the included studies. Altered pathways: (A) glycolysis and gluconeogenesis, (B) methionine and cysteine metabolism, (C) serine and glycine metabolism, (D) lipid metabolism, (E) urea cycle, (F) tricarboxylate cycle, (G) phenylalanine and tyrosine metabolism, (H) glutamate, glutamine, proline and histidine metabolism, (I) branched-chain amino acids metabolism, (J) serotonin pathway, (K) kynurenine pathway, (L) indole pathway, (M) tryptophan metabolism. Metabolites: (1) glucose, (2) pyruvate, (3) alanine, (4) taurine, (5) serine, (6) glycine, (7) methionine, (8) citrate, (9) cis-aconitate, (10) isocitrate, (11) succinate, (12) malate, (13) asparagine, (14) ornithine, (15) arginine, (16) guanidinosuccinate, (17) p-cresyl sulfate, (18) tyrosine, (19) phenylalanine, (20) phenylacetylglutamine, (21) phenylacetate, (22) hippurate, (23) tryptophan, (24) indole-3-acetate, (25) indoxyl sulfate, (26) glutamate, (27) glutamine, (28) histidine, (29) proline, (30) fatty acids, (31) phosphatidylcholines, (32) lysophosphatidylcholines, (33) sphingomyelins, (34) acylcarnitines, (35) oxylipins, (36) leucine, (37) isoleucine.
7. Microbial Metabolites
Recent research has focused the attention on the gut-microbiota-brain axis. The impact of gut microbiota on health and disease has recently been described and appears to be an important biological factor in the development and maintenance of EDs. Gut microbiota is defined as the heterogeneous, unique, and dynamic ecosystem of the intestine that depends on complex interactions between genetic and environmental factors 
. Its role in normal physiology and homeostasis is unquestionable. The microbiota is mainly constituted of bacteria, although there are other organisms such as archaea or protozoa. The composition is highly variable among individuals depending on endogenous and exogenous factors such as sex, age, physical activity, genetic features of the host, and infections, among others. However, it has been demonstrated that the predominant factor determining microbiota composition is the diet 
The numerous implications of gut microbiota on host health and wellness range from nutrient/energy metabolism to brain function and mood regulation pathways 
. Moreover, complex direct and indirect interactions between the microbiota, gut, and brain, have been described constituting the termed “microbiota-gut-brain axis”; and microbiota appears to be involved in the regulation of behaviors and emotions, such as learning, stress, depression, and anxiety, that are common traits in AN 
Once established that the host diet is critical in the gut microbial composition 
and that patients with EDs have altered nutritional patterns, it can be assumed that these patients will present a modified microbiota 
and indeed, this has been described in AN. This dysbiosis results from starvation and malnutrition, but the impact on the onset and progression of the disease needs to be further elucidated 
. Gut microbiota produces a set of bioactive molecules that can induce different responses in the host. Experimental data suggest that an important part of the circulating metabolites in the human body are derived from gut microbiota 
. Some of these metabolites can interact with receptors in enteroendocrine cells (EECs), and some others can enter systemic circulation performing paracrine functions 
. Among those metabolites, short-chain fatty acids (SCFAs), neurotransmitters, and lipopolysaccharides are widely studied due to their autocrine and paracrine effects.
To assess the putative effect of dysbiosis on the physiopathology of AN, a combined analytical strategy that determines the composition of the microbiota and its subproducts should be performed. Metabolomics tools can be applied to analyze the products of bacterial metabolism that develop important functions in the human body. To do so, the patient’s fecal sample constitutes a high-value specimen and should be analyzed. However, alterations in feces have been poorly studied for AN.
SCFAs, such as butyrate, propionate, and acetate, are one of the main products of bacterial metabolism. They come from the fermentation of non-digestible carbohydrates, fiber, and resistant starch. SCFAs can target the ENS stimulating the sympathetic nervous system, which is implicated in energy consumption 
P. Monteleone et al. and A.M. Monteleone et al. performed untargeted
metabolomics of fecal samples by GC-MS. In their comparative analysis between both anorexia types, they found that acetate was decreased in AN-R patients but not in the AN-BP group 
. Moreover, they described increased propionate in AN patients that is restored after treatment, contrary to butyrate that is unchanged in patients and decreases after weight recovery 
Prochazkova et al. performed a multi-omics study with fecal samples from individuals with AN before and after renourishment compared to healthy controls. They determined the composition of gut microbiota and performed targeted metabolomics assays for the analysis of fecal SCFAs and neurotransmitters. Butyrate, acetate, and propionate were analyzed by NMR while the neurotransmitters were determined by MS on selective reaction monitoring (SRM). Butyrate was diminished in the ill patients but showed partial recovery after renourishment therapies, although normal values were not achieved. On the contrary, propionate was significantly decreased in patients after treatment, but there were no significant differences in acute patients compared to controls. Acetate levels were significantly lower in both groups of patients, which implies that renourishment does not restore the normal SCFAs profile.
As suggested previously, the changes in fecal metabolites in patients with AN may result from either their chronic malnutrition and/or changes in their gut microbiota composition 
. Regarding this, butyrate has been related to a reduction of anxiety and depressive-like symptoms and to lower neuroinflammation 
. Thus, decreased butyrate levels might increase susceptibility to depressive-like symptoms. Moreover, the administration of the three SCFAs to mice showed decreased stress-related behaviors 
. Propionate has also been found to exert direct functions in the central nervous system, it can cross the blood-brain barrier acting on different receptors related to the protection of neuroinflammation mainly 
Regarding neurotransmitters, ill patients showed a significant decrease in γ-aminobutyrate (GABA) and dopamine levels. A.M Monteleone et al. also reported significantly decreased GABA levels in both types of AN patients (AN-R and AN-BP) 
. However, serotonin was only significantly lower in renourished patients. Contrary to expected, the comparison between the patients before and after weight restoration did not yield any significant variation. Tyramine, kynurenine, and hydroxytryptophan concentrations did not vary between groups and they did not change during the course of hospitalization. As a result, novel therapeutic approaches are required to be combined with renourishment to improve the metabolic state of patients