The causes of depression

The causes of major depressive disorder (MDD) are still uncertain and not fully understood; however, numerous studies point to several underlying and contributing factors, such as genetic^{(Reference Lebowitz and Ahn24)}, biological^{(Reference Lynch, Gunning and Liston25)}, psychological^{(Reference Yang, Zhao, Wang, Liu, Zhang and Li26)}, social^{(Reference Keles, McCrae and Grealish27)}, lifestyle^{(Reference Raboch, Ptacek, Vnukova and Tkacova28)}, environmental^{(Reference Lopresti, Hood and Drummond29)}, and socio-economic factors^{(Reference Freeman, Tyrovolas, Koyanagi, Chatterji, Leonardi and Ayuso-Mateos30)}. Many of these factors can be evaluated via biomarkers, which are medical signs within the body that provide an objective measure of a patient’s medical state^{(Reference Strawbridge, Young and Cleare31)}. Biomarkers influence or predict the occurrence or outcome of disease and include any substance, structure, process or their products that are measurable in the body^{(Reference Gururajan, Clarke, Dinan and Cryan32)}. Potential biomarkers implicated in the development and progression of depression and depressive symptoms include nutritional status, genetics and gene expression, hormone levels, and markers of inflammation and oxidative stress^{(Reference Strawbridge, Young and Cleare31)}. Dysregulation of any of these biomarkers may influence depression mental health outcomes and warrant investigation^{(Reference Strawbridge, Young and Cleare31)}.

The role of social nutritional factors in depression

When exploring which nutritional factors may be influencing depressive symptoms, it is crucial that a holistic and person-centred approach is used. Therefore, assessing and acknowledging individual, social, and psychological factors which may influence food intake and mood, must also be carefully considered^{(Reference Lee, Angus, Walsh and Sargeant33)}. For example, gathering information on appetite, traditional or cultural food consumption, food security, intuitive eating practises and their relationship with food is crucial to deepen our understanding of the food–mood relationship for each individual person. Previous studies have highlighted the significant way that these social factors can influence food and mood^{(Reference Lee, Angus, Walsh and Sargeant33,Reference Lee, Bradbury, Yoxall and Sargeant34)} . These include: improving mood by removing food restriction^{(Reference Lee, Bradbury, Yoxall and Sargeant34)}, the effect of familial and cultural influences of food and mood^{(Reference Lee, Angus, Walsh and Sargeant33)}, and how food nostalgia can impact mood^{(Reference Lee, Angus, Walsh and Sargeant33)}. Therefore, it is important that these factors are evaluated thoroughly, alongside the functional tests presented in this review. We will now outline several nutritional factors, their role in the aetiology of depression, and the biomarkers which can be tested (see Figs. 1 and 2).

Fig. 1. Nutritional factors which impact depression.

The role of genetic factors in depression

There is a complex interaction between genetic and non-genetic factors regarding depression^{(Reference López-León, Janssens, Gonzalez-Zuloeta Ladd, Del-Favero, Claes and Oostra35)}. Research shows that susceptibility genes potentially interact with each other and with the environment, which is often referred to as a gene–environment interaction^{(Reference Lopizzo, Bocchio Chiavetto, Cattane, Plazzotta, Tarazi and Pariante36)}. Mounting evidence suggests a relationship between genetic variants (polymorphisms) and depression^{(Reference López-León, Janssens, Gonzalez-Zuloeta Ladd, Del-Favero, Claes and Oostra35)}, indicating that a predisposition towards the development of depression can be inherited^{(Reference Harold, Rice, Hay, Boivin, Van Den Bree and Thapar37)}. A meta-analysis on all MDD case–control gene association studies found 393 polymorphisms in 102 genes associated with depression^{(Reference López-León, Janssens, Gonzalez-Zuloeta Ladd, Del-Favero, Claes and Oostra35)}. The review found six susceptibility genes which may be implicated in MDD. These genes include: apolipoprotein E (APOE), guanine nucleotide-binding protein (GNB3) 825T, dopamine transporter (SLC6A3) 40 bp VNTR, dopamine receptor D4 (DRD4), methylenetetrahydrofolate reductase (MTHFR) 677T, and serotonin transporter (SLC6A4) 44 bp ins/del ^{(Reference López-León, Janssens, Gonzalez-Zuloeta Ladd, Del-Favero, Claes and Oostra35)}.

Specific genes have been researched in regard to their exposure to environmental factors and how these factors influence subsequent outcomes^{(Reference Halldorsdottir and Binder38)}. For example, the serotonin transporter (5-HTT) gene, and its two promotor variants have been studied in regard to its interaction with stressful life events and predisposing potential towards depression^{(Reference Caspi, Sugden, Moffitt, Taylor, Craig and Harrington39)}. It is speculated that the interaction of only a few, rather than many genes and their variants, such as the 5-HTT gene, are conditionally affected impacted by an individual’s exposure to environmental risks, and may be an underlying cause of depression and other multifactorial disorders^{(Reference Caspi, Sugden, Moffitt, Taylor, Craig and Harrington39)}.

The role of metabolic factors in depression

Disturbances within metabolic pathways, including the dysregulation of glucose, insulin, leptin, tryptophan and serotonin, are commonly observed in patients with depression^{(Reference Sanchez-Villegas and Martínez-González50–Reference Farooqui, Farooqui, Panza and Frisardi52)}. Furthermore, studies suggest metabolic disturbances along these pathways may provide a link between depression and metabolic syndrome, obesity and CVD^{(Reference Sanchez-Villegas and Martínez-González50)} due to the sharing of these common metabolic pathways^{(Reference Sanchez-Villegas and Martínez-González50,Reference Watson, Nasca, Aasly, McEwen and Rasgon53)} .

It has been proposed that hyperglycaemia may modify hypothalamic–pituitary–adrenal (HPA) axis function, which leads to an increased risk of depressive symptoms^{(Reference Akbaraly, Kumari, Head, Ritchie, Ancelin and Tabák54)}. Additionally, hyperglycaemia and insulin resistance may lead to depression due to increased oxidative stress^{(Reference Essmat, Soliman, Mahmoud and Mahmoud55)}. Hyperglycaemia can lead to a reduction of antioxidant enzymes in the brain, resulting in an accumulation of reactive oxygen species (ROS)^{(Reference Essmat, Soliman, Mahmoud and Mahmoud55)}. ROS cause cell damage and cellular death, resulting in cerebral injury and inhibition of neurogenesis. Furthermore, ROS activate NF-κB, thereby up-regulating the expression of pro-inflammatory cytokines such as TNF-α, interferon gamma, cyclooxygenase-2 and IL-1β^{(Reference Shih, Wang and Yang56)}. This increase in neuroinflammation is thought to be a key contributor in the pathogenesis of depression due to the resulting activation of the HPA axis, the reduction of brain-derived neurotrophic factor (BDNF) and the induction of indoleamine 2, 3-dioxygenase (IDO)^{(Reference Essmat, Soliman, Mahmoud and Mahmoud55)}. IDO plays a crucial role in the tryptophan–kynurenine pathway and prolonged activation is thought to contribute to reduced serotonin production^{(Reference Dogrul57)}. Therefore, testing patients fasting blood glucose and HbA1c should be considered in patients with depression where hyperglycaemia is suspected to be a contributing factor. Assessing dietary intake of refined, added and excess sugar is also warranted.

Leptin is an adipokine that exerts neuroendocrine regulatory functions on energy balance and glucose metabolism^{(Reference Cernea, Both, Huţanu, Şular and Roiban58)}. Diet-induced obesity can lead to leptin resistance, and further exacerbate overeating. Leptin also influences several psychological functions including cognition, motivation, and memory, in addition to its effects on neuronal structure, survival and plasticity^{(Reference Stieg, Sievers, Farr, Stalla and Mantzoros59)}. Leptin receptors are extensively dispersed throughout the brain, impacting several neurotransmitter systems. Leptin has both metabolic and neurological properties involved in neuroprotection, cognition and mood, and therefore has been associated with depression^{(Reference Lang and Borgwardt5)}. It modulates the mesolimbic dopamine pathway, HPA axis function, hippocampal synaptic plasticity and serotonin synthesis^{(Reference Cernea, Both, Huţanu, Şular and Roiban58)}. Leptin influences serotonin via inhibition of nitric oxide synthase, which facilitates the normal action of serotonin on its receptors and its reuptake by serotonin transporters^{(Reference Lu60)}. Consequently, leptin resistance and the associated decreased nitric oxide synthase leads to the inhibition and transformation of serotonin into inactive dimers^{(Reference Lu60)}. Therefore, underlying leptin resistance may play a role in depressive symptoms^{(Reference Milaneschi, Simonsick, Vogelzangs, Strotmeyer, Yaffe and Harris61)}. While serum blood tests can ascertain leptin levels, healthcare providers do not routinely test leptin levels and the test is not available at all laboratories. As dietary factors, including increased fat and sucrose intake and low protein consumption, are drivers of leptin resistance, a thorough dietary analysis should be undertaken.

As mentioned previously, alterations to the tryptophan–kynurenine (KYN) pathway^{(Reference Kopschina Feltes, Doorduin, Klein, Juárez-Orozco, Dierckx and Moriguchi-Jeckel62)} can result in depleted levels of serotonin^{(Reference Kopschina Feltes, Doorduin, Klein, Juárez-Orozco, Dierckx and Moriguchi-Jeckel62)}. Under normal circumstances, less than 3% of tryptophan metabolism is directed down the serotonin branch, while approximately 95% is directed down the kynurenine pathway^{(Reference Qin, Wang, Zhang, Han, Zhai and Lu63)}. The activation of the kynurenine pathway and the consequent generation of neurologically active metabolites are implicated in several psychiatric disorders including depression, schizophrenia, and bipolar disorder^{(Reference Qin, Wang, Zhang, Han, Zhai and Lu63)}. The enzymes indoleamine 2,3-dioxygenase (IDO) and tryptophan 2,3-dioxygenase (TDO) are the first and rate-limited step of catalysing the conversion of l-tryptophan into n-formylkynurenine^{(Reference Qin, Wang, Zhang, Han, Zhai and Lu63)}. n-Formylkynurenine is then further degraded to either picolinic acid (neuroprotective) or quinolinic acid (neurotoxic)^{(Reference Hestad, Alexander, Rootwelt and Aaseth64)}. Quinolinic acid is considered neurotoxic due to its activation of n-methyl-d-aspartate receptors and free radical production^{(Reference Hestad, Alexander, Rootwelt and Aaseth64)}. It may also promote interferon gamma-induced inflammation and has been strongly associated with depression and neurodegenerative diseases^{(Reference Hestad, Alexander, Rootwelt and Aaseth64)}.

Neuroinflammation can lead to excessive production of quinolinic acid and further alterations to the picolinic acid/quinolinic acid ratio^{(Reference Hestad, Alexander, Rootwelt and Aaseth64)}. Specific nutritional factors have been shown to influence quinolinic acid levels and reduce quinolinic induced damage. These include probiotics^{(Reference Fukngoen, Sivamaruthi, Sirilun, Lalitsuradej, Khongtan and Peerajan65)}, polyphenols^{(Reference dos Santos, Visentin, Scariot, Echeverrigaray, Salvador and Branco66)} and essential fatty acids^{(Reference Morales-Martínez, Sánchez-Mendoza, Martínez-Lazcano, Pineda-Farías, Montes and El-Hafidi67)}. Assessing intake of these dietary components may therefore be valuable. Currently, there are no widely available tests for IDO and TDO. Some functional pathology labs assess tryptophan and associated metabolites via urine organic acid tests; however, many of these tests have not been adequately assessed for validity and reliability, so caution should be used when interpreting results. Further, urine levels of quinolinic acid are not a reflection of its levels in the brain^{(Reference Lugo-Huitrón, Ugalde Muñiz, Pineda, Pedraza-Chaverrí, Ríos and Pérez-de la Cruz68)}. The same applies for organic acid test of 5-hydroxyindoleacetic acid (5-HIAA), a metabolite of serotonin. Levels measured in urine are often used as a marker to determine the levels of serotonin in the body. However, urinary levels of 5-HIAA, do not reflect levels found in the brain or cerebrospinal fluid^{(Reference Aizenstein and Scavone69)}.

The role of inflammation and Immunological factors in depression

Interactions between the central nervous system and immune system have been studied to assess immune activation in depression. It has been suggested that, from an evolutionary perspective, the inflammation response and symptoms of depression formed an integrated adaptive response to pathogens that enabled wound healing and allowed individuals to avoid subsequent pathogen exposure^{(Reference Miller and Raison70)}. Crosstalk between the various inflammatory pathways and neurocircuits within the brain can result in altered behaviour, such as avoidance and anxiety^{(Reference Miller and Raison70)}. This may have provided an evolutionary advantage to early humans in regard to their interactions with predators and pathogens. However, in today’s modern world, sustained high levels of inflammatory cytokines may be driving increased levels of depression^{(Reference Miller and Raison70)}.

Cytokines, such as IL-1β, IL-4, IL-6, IL-10, TNF-α and C-reactive protein (CRP)^{(Reference Lang and Borgwardt5,Reference Sánchez-Villegas, Ruíz-Canela, de la Fuente-Arrillaga, Gea, Shivappa and Hébert71)} , are signalling molecules involved in immune regulation, specifically regulating a host’s response to immune processes, inflammation, infection, and trauma. Cytokines can have either pro- or anti-inflammatory actions within the body. An increase in the levels of inflammatory cytokines and the induction of their signalling pathways, in addition to the activation of various immune cell subsets have been observed in the peripheral blood and brain of depression patients^{(Reference Miller and Raison70)}. CRP, TNF-α, IL-1β and IL-6 are the most consistently raised inflammatory markers in patients with depression^{(Reference Miller and Raison70)}. Assessing CRP may therefore be a useful tool in determining the extent that inflammation may be contributing to patient symptoms and may assist with monitoring the effectiveness of treatment strategies^{(Reference Pariante72)}. Extensive research demonstrates that diet can affect circulating markers of inflammation. Findings from a recent meta-analysis indicated that CRP concentration is positively associated with sugar intake, and negatively associated with the consumption of vitamins, minerals and polyunsaturated fatty-acids, suggesting inflammation is likely influenced by dietary intake^{(Reference Mazidi, Kengne, Mikhailidis, Cicero and Banach73)}.

Accumulating evidence has also demonstrated that the Kelch-like ECH-associated protein 1 nuclear factor (erythroid 2-derived)-like 2 (Keap1-Nrf2) system may be involved in depression pathogenesis, via its key role in regulating inflammation and oxidative stress^{(Reference Hashimoto74)}. The Keap1-Nrf2 system can influence the body’s response to oxidative stress by regulating the activity of Nrf2, a transcription factor that promotes the expression of antioxidant genes and other protective mechanisms^{(Reference Hashimoto74)}. When this system is disrupted or impaired, it can lead to increased oxidative stress and inflammation, which may contribute to the neurobiological changes associated with depression^{(Reference Hashimoto74)}. Some studies have shown that dysfunction in the Keap1-Nrf2 pathway may be associated with depressive symptoms and that dietary interventions, such as sulforaphane, aimed at enhancing Nrf2 activity could potentially have antidepressant effects^{(Reference Yao, Zhang, Ishima, Dong, Yang and Ren75,Reference Zheng, Li, Zhang and Wang76)} . However, there are currently no widely used clinical tests specifically designed to measure the activity of the Keap1-Nrf2 system in a routine medical setting.

The role of hormones in depression

Impairments in hormones and endocrine function may play an important role in the underlying pathophysiology of depression^{(Reference Dwyer, Aftab, Radhakrishnan, Widge, Rodriguez and Carpenter77)}. Abnormalities in adrenal (HPA axis) and thyroid (HPT axis) function are often associated with altered mood, and medications which target endocrine function, often lead to mood-related and cognitive effects^{(Reference Dwyer, Aftab, Radhakrishnan, Widge, Rodriguez and Carpenter77)}.

Research has demonstrated that in patients with depression, glucocorticoid receptor (GR) signalling is abnormal and associated with chronic hypersecretion from the corticotrophin releasing hormone neurons of the HPA axis^{(Reference Nandam, Brazel, Zhou and Jhaveri78)}. It is thought that this chronic hypersecretion may shift HPA activity toward increasingly higher set points, which may result in persistently elevated HPA activity often observed in patients with depression^{(Reference Nandam, Brazel, Zhou and Jhaveri78)}. Unhealthy dietary habits, and the consumption of unhealthy foods, for example, foods high in carbohydrates or fat, have also been connected to HPA axis hyperactivity, and represent a significant contribution to stress accumulation^{(Reference Pearlmutter, DeRose, Samson, Linehan, Cen and Begdache79)}. Hypercortisolaemia can cause excitotoxicity leading to loss and atrophy of dendrites, as well as inhibition of neurogenesis in the dentate gyrus of the hippocampus^{(Reference Yang, Zhao, Wang, Liu, Zhang and Li26)}. Numerous studies conducted in patients with depression have shown abnormalities in the suppression of cortisol after pharmacological and psychological challenges^{(Reference Yang, Zhao, Wang, Liu, Zhang and Li26)}. However, the relationship between depression and cortisol is highly complex, and appears dependent on a range of different factors^{(Reference Nandam, Brazel, Zhou and Jhaveri78)}. These factors include the stage and severity of illness, and the type of challenge faced^{(Reference Nandam, Brazel, Zhou and Jhaveri78)}. For example, research demonstrates that depressive symptom severity is proportionate to cortisol levels. However, the HPA response appears to be unaffected in chronic depressive states, such as in those experiencing symptoms for longer than two years, therefore measuring cortisol levels will not provide prognostic information in chronic major depressive disorder^{(Reference Nandam, Brazel, Zhou and Jhaveri78)}.

Thyroid hormones are responsible for a number of important functions within the body. Their primary function is the regulation of cell differentiation, metabolism and nervous system development^{(Reference Baksi and Pradhan80)}. Altered levels of thyroid hormones can lead to nervous system-related problems linked to cognition, vision, motor skills, language and memory^{(Reference Baksi and Pradhan80)}. Thyroid hormones have also been associated with neuropsychiatric disorders including bipolar disorder, schizophrenia, depression and anxiety^{(Reference Baksi and Pradhan80)}. Clinical hypothyroidism is commonly associated with depressive symptoms, and subclinical hypothyroidism is commonly reported in treatment-resistant depression^{(Reference Dwyer, Aftab, Radhakrishnan, Widge, Rodriguez and Carpenter77)}. However, most patients experiencing depression do not have clear biochemical evidence of thyroid disease^{(Reference Hage and Azar81)}. Although, research does show that patients with depression often display low T3 levels, raised T4 levels, raised reverse T3 levels, a blunted thyroid-stimulating hormone response to thyrotropin-releasing hormone and positive antithyroid antibodies^{(Reference Hage and Azar81)}. Further research exploring the various mechanisms involved in the interaction between thyroid function and depression needs to be conducted before firm conclusions can be drawn^{(Reference Hage and Azar81)}. However, routine screening of patients presenting with depression for potential thyroid dysfunction should still be considered.

Several key nutrients are involved in healthy thyroid function. For example, thyroid peroxidase is a haem-dependent enzyme required for thyroid hormone synthesis, which only becomes active at the apical surface of thyrocytes after it binds a prosthetic haem group, therefore, adequate iron is essential for the synthesis of thyroid hormones^{(Reference Rayman82)}. Additionally, selenoproteins are required for healthy thyroid function. In the thyroid, several ROS, such as hydrogen peroxide (H₂O₂) are generated during iodine organification, which is the incorporation of iodine into thyroglobulin for the production of thyroid hormones^{(Reference Gheorghiu and Badiu83)}. The thyroid is protected from the harmful effects of ROS primarily by the antioxidant enzymatic system, comprising the antioxidant elements selenium, copper, manganese and zinc^{(Reference Gheorghiu and Badiu83)}. Assessing these trace elements via a food diary and blood pathology may therefore also be of value.

The role of neurological factors in depression

BDNF is a peptide involved in the maintenance and survival of neurons, synaptic integrity and synaptic plasticity^{(Reference Rana, Behl, Sehgal, Srivastava and Bungau84)}. The production and secretion of BDNF is compromised by inflammatory processes and endothelial dysfunction, resulting in low BDNF levels^{(Reference Patterson85)}. Research demonstrates that low levels of BDNF have been reported in patients with depression⁷⁹ and those with suicidal ideation^{(Reference Khan, Wu, Reus, Hough, Lindqvist and Westrin86)}. Emerging evidence also suggests that the therapeutic action of many anti-depressive medications is due to their ability to reverse this decrease^{(Reference Rana, Behl, Sehgal, Srivastava and Bungau84)}. Numerous factors have been associated with low BDNF levels. For example, stress has been shown to modulate BDNF expression^{(Reference Notaras and van den Buuse87)}, and low BDNF levels have also been linked to food cravings, overeating and weight gain^{(Reference Lang and Borgwardt5,Reference Bumb, Bach, Grosshans, Wagner, Koopmann and Vollstädt-Klein88)} . Several nutritional factors have been explored for their potential to increase BDNF levels^{(Reference Gravesteijn, Mensink and Plat89)}. These include polyphenols, omega-3 fatty acids, zinc and probiotics^{(Reference Gravesteijn, Mensink and Plat89)}. Assessing dietary intake of these nutrients may therefore be of value.

Depression has been associated with neurotransmitter imbalances, specifically serotonin, dopamine, noradrenaline and glutamate^{(Reference Maletic, Robinson, Oakes, Iyengar, Ball and Russell90)}. Deficiencies in serotonin availability, serotonin receptor abnormalities^{(Reference Benjamin and Klein91)} and an increase in monoamine oxidase which metabolises serotonin in the brain^{(Reference Meyer, Ginovart, Boovariwala, Sagrati, Hussey and Garcia92)}, all support the neurotransmitter imbalance hypothesis for MDD^{(Reference Lopresti, Hood and Drummond29)}. The theory that depression is caused by low serotonin was first suggested in the 1960s and has been heavily promoted since the discovery of selective serotonin reuptake inhibitor antidepressants^{(Reference Moncrieff, Cooper, Stockmann, Amendola, Hengartner and Horowitz93)}. However, this theory of depression has received growing scrutiny in recent years. For example, this hypothesis does not explain why drugs such as tianeptine, which reduce rather than increase serotonin availability at the synaptic cleft, are effective antidepressants^{(Reference Hindmarch94)}. Further, a large meta-analysis of monoamine depletion studies concluded that monoamine depletion appears to only decrease mood in subjects with a family history of depression and in drug-free patients whose depression is in remission^{(Reference Ruhé, Mason and Schene95)}. However, monoamine depletion does not appear to decrease mood in healthy individuals and fails to demonstrate a causal relationship^{(Reference Ruhé, Mason and Schene95)}. Additionally, a large umbrella review published in 2022 found that research investigating the role of serotonin in depression concluded that there is no convincing evidence to support that depression is associated with, or caused by, reduced serotonin levels or activity^{(Reference Moncrieff, Cooper, Stockmann, Amendola, Hengartner and Horowitz93)}.

Although several antidepressants alter monoamine levels within a few hours, changes in mood are typically not observed for approximately 3–4 weeks^{(Reference Hill, Sahay and Hen96)}. This discrepancy, coupled with the fact that antidepressants appear to increase the quantity of adult-born neurons, which take approximately 4 weeks to form synaptic connections, has created the hypothesis that antidepressants may affect mood via increasing the levels of adult hippocampal neurogenesis^{(Reference Hill, Sahay and Hen96)}. This plasticity hypothesis links depression with reduced hippocampal neurogenesis and neurotrophin levels^{(Reference Penn and Tracy97)}. The delayed therapeutic onset of antidepressant medications may therefore be explained by this hypothesis, in which the time lag is attributed to antidepressants changing various intracellular enzymes, for example, cyclic adenosine monophosphate, protein kinase A and adenylyl cyclase^{(Reference Penn and Tracy97)}. These enzymes activate the expression of the neuroprotective BDNF^{(Reference Penn and Tracy97)}.

The role of the gut–brain axis in depression

An emerging field of study is demonstrating a strong link between the gut microbiota and the brain. Referred to as the gut–brain axis, an understanding of the role of the gut microbiota and its influence on alterations of the immune, neural and endocrine systems continues to develop^{(Reference Arneth98)}. However, influences are not unidirectional. The central nervous system through the activation of the HPA axis, autonomic nervous system and neuroendocrine systems, alters the composition, motility and secretion, and therefore gut microbiota equilibrium, thus directly impacting the gut microbiota^{(Reference Carabotti, Scirocco, Maselli and Severi99)}.

The health of the gut and its microbiota have been linked with numerous neuropsychiatric disorders^{(Reference Appleton100)}. Gut bacteria are involved in the production of active metabolites via enzymatic reactions which result in products having therapeutic activity within the body^{(Reference Paul, Barnes, Demark-Wahnefried, Morrow, Salvador and Skibola101)}. For example, the monoamines serotonin, dopamine and norepinephrine can be derived from certain microbes. Escherichia coli, Candida and Streptococcus produce serotonin, Bacillus and Serratia produce dopamine, and E. coli, Bacillus and Saccharomyces produce norepinephrine^{(Reference Evrensel and Ceylan102)}. Additionally, gamma-aminobutyric acid is synthesised by Lactobacillus and Bifidobacterium ^{(Reference Evrensel and Ceylan102)}.

The body’s immune system is supported by the role of the gut microbiota and its diversity^{(Reference Evrensel and Ceylan102)}. Regulation of pro-inflammatory cytokines and chemokines, such as the interleukins IL-1, IL-8, IL-10 and transforming growth factor B occurs due to the interaction of gut bacteria with the gut mucosa. The gut mucosa covers the largest surface of epithelium in the body, and along with tight junctions in the gut epithelium, they form a physical barrier against bacteria, harmful substances produced by bacteria and antigens^{(Reference Evrensel and Ceylan102)}.

Alterations to the composition of resident gut microbiota, or dysbiosis, leave the gut epithelium wall susceptible to micro-damage and increase its permeability, thereby allowing harmful substances to enter systemic circulation^{(Reference Evrensel and Ceylan102)} and instigating an immune response. As the brain and stomach are directly linked via the vagus nerve, bacterial, hormonal and neuronal changes in the gut are transmitted directly to the brain. The vagus nerve, also referred to as cranial nerve number ten, is a principal component of the parasympathetic nervous system^{(Reference Tan, Yan, Ma, Fang and Yang103)}. It is comprised of roughly 80% afferent and 20% efferent fibers. It plays a crucial role in interoceptive awareness, whereby it senses microbial metabolites through its afferents, and transfers this information to the central nervous system^{(Reference Tan, Yan, Ma, Fang and Yang103)}. This information is then integrated into the central autonomic network, where a response is then generated.^{(Reference Tan, Yan, Ma, Fang and Yang103)}. The gut microbiota and its metabolites can therefore affect the brain’s development and plasticity by secreting neurotrophins and proteins, such as post-synaptic density (PSD)-95, BDNF and synaptophysin^{(Reference Evrensel and Ceylan102)}.

Additionally, stress is known to inhibit the vagus nerve and has harmful effects on the gastrointestinal tract and the microbiota^{(Reference Foster, Rinaman and Cryan104)}. Brief exposure to stress can have a notable impact on the composition of the microbiota community, causing shifts in the proportions of major microbiota phyla^{(Reference Foster, Rinaman and Cryan104)}. Moreover, when the gut microbiota is deliberately manipulated in experiments, it can affect an individual’s responsiveness to stress and the threshold at which the neuroendocrine HPA axis becomes activated^{(Reference Luna and Foster105)}. These alterations are associated with changes in microbiota-related metabolites and immune signalling pathways, suggesting that these systems could play a significant role in stress-related conditions, including depression^{(Reference Luna and Foster105)}.

Microbiome mapping has gained increasing popularity in recent years, with several companies offering a range of tests aimed at assessing the quality and diversity of the gastrointestinal tract microbiome. Some biotechnology companies market personalised microbiome testing directly to the public^{(Reference Staley, Kaiser and Khoruts106)}. These tests range greatly in complexity and costs and aim to provide personalised diagnosis and therapies. Many companies provide detailed reports on the gut microbiota diversity, beneficial and pathogenic microorganisms which effect health and disease, and personalised dietary, supplemental and lifestyle advice^{(Reference Mills, Lane, Smith, Grimaldi, Ross and Stanton107)}. However, current research evidence is not adequate to allow for meaningful diagnoses to be concluded, due to the high degree of inter-individual variability in the microbiome and the substantial limitations of the analytic methods most commonly used^{(Reference Staley, Kaiser and Khoruts106)}. Over-extrapolation of results by the service provider and over-interpretation of results by patients poses a significant risk which may cause unnecessary stress and anxiety, dietary changes and supplement use, which may ultimately do more harm than good^{(Reference Mills, Lane, Smith, Grimaldi, Ross and Stanton107)}.

Although microbiome data are not diagnostic at present, data from commercial tests do provide patients with a snapshot of their microbiome in comparison to other individuals from different environments and backgrounds^{(Reference Staley, Kaiser and Khoruts106)}. For example, abundances of particular genera which may contain true pathogens, may be informative regarding risk for specific illnesses^{(Reference Staley, Kaiser and Khoruts106)}. More research is needed to explore changes to the microbiome in major depressive disorder to draw any firm conclusion about a patient’s risk or diagnosis based on their microbiome profile.

The role of minerals in depression

Magnesium is an essential mineral required for the appropriate activity of many biochemical and physiological processes, including DNA replication, transcription and translation^{(Reference Botturi, Ciappolino, Delvecchio, Boscutti, Viscardi and Brambilla108)}. Previous research has demonstrated that magnesium is involved in various brain regions in the limbic system, consequently implicating magnesium in the aetiology and progression of depression^{(Reference Botturi, Ciappolino, Delvecchio, Boscutti, Viscardi and Brambilla108)}. However, current research is conflicting, with some studies demonstrating an increased risk of depression in those with low magnesium intake^{(Reference Tarleton and Littenberg109,Reference Stanisławska, Szkup-Jabłońska, Jurczak, Wieder-Huszla, Samochowiec and Jasiewicz110)} , and others finding no association^{(Reference Derom, Martínez-González, Sayón-Orea, Bes-Rastrollo, Beunza and Sánchez-Villegas111,Reference Martínez-González and Sánchez-Villegas112)} . The biological mechanisms involved between depression and low serum magnesium levels is currently uncertain; however, it is thought to include the central nervous system, HPA axis and oxidative stress pathways^{(Reference Wang, Um, Dickerman and Liu113)}. Research has shown that magnesium deficiency leads to changes in glutamatergic transmission in the limbic system and cerebral cortex. Additionally, magnesium’s role as an antagonist of the n-methyl-d-aspartate glutamate receptor, a significant component involved in synaptic potentiation, learning and memory, is well established^{(Reference Wang, Um, Dickerman and Liu113)}. Research indicates that magnesium intake from the diet is insufficient in approximately 60% of adults, and that subclinical magnesium deficiency is widely prevalent in western populations^{(Reference Fiorentini, Cappadone, Farruggia and Prata114)}. Unfortunately, the evaluation of magnesium status is complicated due to a number of factors, with serum magnesium concentrations demonstrating no reliable correlation with total body magnesium levels or concentrations in specific tissues^{(Reference Fiorentini, Cappadone, Farruggia and Prata114)}.

Iron is an important mineral which acts as a co-factor in the production of several neurotransmitters, including dopamine and serotonin. There has been a growing interest in the role of iron in neurodevelopment and its implication in psychiatric and neurological conditions^{(Reference Berthou, Iliou and Barba115)}. Several studies have highlighted that iron deficiency is associated with an increased depression risk^{(Reference Hidese, Saito, Asano and Kunugi19,Reference Li, Wang, Xin, Song and Zhang116,Reference Lee, Chao, Huang, Chen and Yang117)} , and recent randomised clinical trials have demonstrated that supplementing with iron leads to reduced depressive symptoms^{(Reference Berthou, Iliou and Barba115)}. It is therefore recommended that patients iron levels, particularly ferritin, are assessed to determine if iron deficiency could be a causative or contributing factor in depressive symptomology^{(Reference Berthou, Iliou and Barba115)}.

Another essential trace element required for numerous vital biochemical and physiological processes needed for brain growth and function, is zinc^{(Reference Szewczyk, Kubera and Nowak118)}. The ratio of intracellular and extracellular zinc levels is critical for zinc homeostasis in the brain, especially in regions linked with the pathophysiology of depression, for example the amygdala, cerebral cortex and the hippocampus^{(Reference Wang, Um, Dickerman and Liu113)}. A meta-analysis concluded that blood zinc concentrations were roughly 0·12 µg/ml lower in depressed patients compared to healthy controls^{(Reference Swardfager, Herrmann, Mazereeuw, Goldberger, Harimoto and Lanctôt119)}. Additionally, randomised controlled trials conducted in individuals with depression have observed improvements in symptoms of depression when participants combined antidepressant drug therapy with zinc supplements, compared with antidepressants alone^{(Reference Lai, Moxey, Nowak, Vashum, Bailey and McEvoy120,Reference Siwek, Dudek, Paul, Sowa-Kućma, Zięba and Popik121)} . It has been suggested that low zinc is associated with depression due to decreased neurogenesis and neural plasticity, increased cortisol levels, and a disturbance in glutamate homeostasis. Zinc’s antidepressant action is thought to be due to increased expression of BDNF^{(Reference Wang, Um, Dickerman and Liu113)}. Zinc status is often measured via a blood test, urine test or hair analysis, which have all demonstrated to be reliable biomarkers of zinc status^{(Reference Lowe, Fekete and Decsi122)}.

The role of polyunsaturated fatty acids in depression

Polyunsaturated fatty acids (PUFA), particularly omega 3 and omega 6, are found in high concentrations in the brain^{(Reference Liao, Xie, Zhang, He, Guo and Subramanieapillai123)}. Clinical studies have demonstrated that patients diagnosed with depression or anxiety display significantly lower levels of PUFA in the blood and in the brain^{(Reference Parletta, Zarnowiecki, Cho, Wilson, Procter and Gordon124,Reference McNamara, Jandacek, Tso, Dwivedi, Ren and Pandey125)} . Furthermore, the cytochrome P soluble epoxide hydrolase pathway has been suggested to play a key role in depression^{(Reference Borsini126)}. Soluble epoxide hydrolase are enzymes which metabolise cytochrome P derived epoxy fatty acids to their corresponding diols. Evidence suggests that soluble epoxide hydrolase plays a key role in the anti-inflammatory effects involved in the metabolism of omega-3 PUFA^{(Reference Borsini126)}. Evidence also demonstrates that in patients with major depressive disorder, the protein expression of soluble epoxide hydrolase in the brain is higher than in healthy controls^{(Reference Borsini126)}.

PUFA intake, particularly the omega-3 eicosapentaenoic acid (EPA) has been linked with depression via a number of different mechanisms. For example, increased intake leads to decreased production of proinflammatory cytokines, such as TNF-α, IL-1β, IL-2 and IL-6^{(Reference Liao, Xie, Zhang, He, Guo and Subramanieapillai123)}, and has been linked with increases in N-acetyl-aspartate in the brain, a key marker for neuronal homeostasis, which suggests a role as a neuroprotective agent^{(Reference Frangou, Lewis, Wollard and Simmons127)}. EPA supplements have also been shown to increase the ratio of cerebral phosphomonoesters to phosphodiesters, a crucial marker of phospholipid turnover, in addition to reversing brain atrophy in patients with MDD^{(Reference Puri, Counsell, Hamilton, Richardson and Horrobin128)}. EPA supplementation has also demonstrated an ability to increase BDNF levels after traumatic brain injury^{(Reference Liao, Xie, Zhang, He, Guo and Subramanieapillai123)}. Lastly, EPA can increase dopaminergic and serotonergic neurotransmission^{(Reference McNamara, Able, Liu, Jandacek, Rider and Tso129)}. PUFA can be measured via an Omega-3 Index test, which measures erythrocyte EPA and DHA content. This test is offered by a number of functional medicine pathology labs and has shown to be a reliable and valid measure of PUFA status^{(Reference Harris and Polreis130,Reference Gurzell, Wiesinger, Morkam, Hemmrich, Harris and Fenton131)} .

Recommendations and key implementation strategies

Given the significant role that nutritional factors play in the pathophysiology of major depressive disorder, it is important that these factors are thoroughly and comprehensively investigated. For example, identifying if an undiagnosed iron deficiency or thyroid condition is a contributing factor to depressive symptomology may prevent unnecessary antidepressant medication prescriptions. Therefore, based on the latest evidence discussed in this review article, we recommend the following nutrients and associated biochemistry be assessed in patients displaying symptoms of depression (see Fig. 2).

Genetic factors: testing for homocysteine, folate and vitamin B12 can be evaluated via blood pathology and are routine tests available with a general practitioner referral. If patients present with elevated homocysteine and abnormal folate results, then testing MTHFR polymorphisms may also be warranted.

Metabolic factors: testing patients’ fasting blood glucose and HbA1c levels should be considered in patients with depression where hyperglycaemia is suspected to be a contributing factor. Assessing the diet for refined, added and excess sugar intake is also warranted. Testing urine levels of quinolinic acid and 5-hydroxyindoleacetic acid (5-HIAA) is not recommended at this time.

Inflammation: much evidence shows that CRP, TNF-α, IL-1β and IL-6 are the most consistently raised inflammatory markers in patients with depression. Therefore, testing these markers may be a useful tool in determining the extent that inflammation may be contributing to patient symptoms and may assist with monitoring the effectiveness of treatment strategies.

Hormones: the association between depression and cortisol is highly complex and appears dependent on a range of different factors. HPA response does not appear to be affected in chronic depressive states, such as in those experiencing symptoms for longer than two years, therefore cortisol levels will not provide accurate prognostic data in chronic MDD. In patients presenting with signs and symptoms of thyroid dysfunction, a full thyroid panel including TSH, T3, T4, rT3 and thyroid antibodies should be conducted.

Gut–brain axis: at this stage, microbiome testing is not advanced enough to accurately determine if the microbiome composition indicates MDD. Testing may be warranted to rule out gastrointestinal pathogens, but should not be relied upon to direct treatment strategies for depression at this time.

Nutritional medicine and diet: in the first instance, we recommend that a thorough dietary analysis should be conducted from a food diary, 24-h food recall or food frequency questionnaire to identify potential nutrient deficiencies (e.g. iron, magnesium, zinc, vitamin B12, folate, PUFAs) or inadequate intakes. Nutrition analysis software programmes such as FoodZone⁽¹³²⁾ or FoodWorks⁽¹³³⁾ may assist in accurately capturing this information. If additional clinical signs and symptoms of deficiency are also present, then further testing may also be necessary. Blood tests to assess iron, zinc, vitamin B12, folate and omega 3 are recommended; however, blood tests for magnesium are not recommended at this time.

Fig. 2. Functional testing for depression.