The wonders of the modern human brain can be traced to its humble beginnings. Starting with a brain of approximately 470 ml in the Hominini,30 the human brain has grown to about 1350 ml over the past 2 million years.31 The near tripling in the size of the human brain is the result of many factors not the least of which are the external inputs of energy and the molecular building blocks provided by macronutrients (e.g., proteins, carbohydrates and lipids). Although macronutrients are essential for energy and the assembly of neural and non-neural tissues, it is also likely that the micronutrients, biotransformed nutrients, phytochemicals and even anti-nutrients and xenobiotics (i.e., DSMs), triggered the plethora of molecular processes required for the growth, development and differentiation of the modern brain and all its parts. When stimuli from DSMs are integrated neuronally and subjected to the pressures of natural selection, cellular responses can emerge that produce higher-order cognition that is adaptive, sustainable and knowledge generating. This is almost assured considering the extensive feedforward/feedback regulatory controls at play in the brain as discussed later.
Since AHN was first discovered in the mammalian brain by Altman and Das,32 it was often considered a phylogenetic reversion, away from lifelong neurogenesis, in favor of neurological stability within the complexity of the brain.33 We now know that AHN occurs throughout the animal kingdom and while humans have fewer neurogenic zones than fish, within these neurogenic zones, substantial and highly functional neurons can be produced. This suggests that at least for the mammalian dentate gyrus, evolution has moved toward neurogenic plasticity rather than away from it.33 This idea is supported by recent studies using imaging connectomics34 and graph theory showing that normal brain maturation, from infancy to adulthood, involves significant co-evolution and integration of structural (neurons and glial cells) and functional (cognitive processes) networks.35 We propose that this co-evolution of neuronal and synaptic plasticity is supported, if not driven, by the constant interplay between the brain and external stimuli from food. This concept is consistent with the ecological intelligence hypothesis for primate brain evolution.36 According to this hypothesis, “foraging cognition” involving spatial memory, value-based decision making and inhibitory control creates those dynamic feedforward and feedback interactions that are adaptive and lead to higher-quality foods, more productive food sources and larger brains.36,37 We believe the principles of the food–brain axis described below, complement and extend the ecological intelligence hypothesis by connecting the nutritional environment with neurological structures and processes through well-studied signaling pathways and gene regulatory networks.
The food–brain axis
In recent years, it has become apparent that all physiological, metabolic and genetic processes and systems are interconnected and interdependent to some degree. Examples include the inflammation–immunity axis,38 the hypothalamus–pituitary–gonadal axis39 and the gut–brain axis.28,29 In each case, an axis suggests diverse regulatory lineages, orchestrating control and outcomes over other critical processes. This makes all the systems involved highly sensitive to extracellular signals including those from the environment. The food–brain axis represents more than just connectivity and relatedness between what we eat and how our brain grows and functions; it illustrates dynamic interdependencies between food and neurological processes. This semi-quantitative interpretation of “axis” should enable researchers to categorize, quantify and predict neurological changes as a function of food quality and/or quantity.
Here, the food–brain axis is defined as a horizontal line of independent variables thought to be causative (e.g., food) that transects a vertical line of dependent variable thought to be the effects (e.g., neurogenesis, neuroplasticity and neuropathologies). Figure 1 shows three hypothetical examples (a, b and c) depicting the consequences of changing the quantity and/or quality of the food from poor (e.g., -3) to good (e.g., +3) on the X-axis, and its effect on neural growth, differentiation and function on the Y-axis. In addition to being an informative method for displaying the dynamic relationship between diet and brain structure and function, the food–brain axis organizes these interactions into four quadrants (i.e., i–iv). Figure 1b shows the nominal or neuro-typical condition for food–brain interactions whereas Fig. 1a-iv shows a neuro-atypical/challenged condition in which neurological processes are dysfunctional, degenerative and pathological as a result of poor diet. Conversely, Fig. 1c-ii illustrates a neuro-atypical/enhanced condition in which neurological processes and structures are enhanced in response to dietary inputs that are higher in quality and/or quantity.
The neuro-atypical quadrants A-i and C-iii pose interesting questions about food–brain interactions. In the case of quadrant A-i (neuro-atypical/enhanced) one might surmise that brain development and function (as defined by measures of intelligence) may be more dependent on robust genetic factors and age than on the quantity and/or quality of food intake. For example, it is known that the genetic contribution to human intelligence is approximately 80% for adults with additive genetic variance contributed by selective mating based on similar phenotypes.40 Therefore, if prenatal and early postnatal nutrition are adequate (e.g., nursing), the impact of nutritional deficiency later in life may have little or no measurable effect on cognitive performance. For quadrant C-iii, strong genetic determinants like Fragile X syndrome, Huntington’s disease, PKU as well as traumatic brain injury (TBI; Box 1) come into play. These are conditions that make neurological structures and processes refractory to the benefits of abundant, higher-quality dietary inputs. PKU, an autosomal recessive metabolic disorder, causes a toxic accumulation of dietary phenylalanine in the brain. If undiagnosed and left untreated, PKU can cause serious cognitive impairment, behavioral and mental disorders as well as seizures, regardless of food quality and/or quality. However, nutritional intervention with a low phenylalanine diet supplemented with large neutral amino acids, vitamin D and B12 can prevent or mitigate the physical, neurological and developmental problems associated with PKU.41 Therefore, interpreting the impact of food quality and/or quantity (x-axis) in quadrant A-i, age must be taking into account while in quadrant C-iii, early diagnosis of deleterious genetic factors or structural damage to brain tissues are key factors for consideration.
DSMs and the brain
The impact of food on brain development and function has been extensively reviewed, although these reviews are often more descriptive than mechanistic in nature.15,54,55 They emphasize the neuroprotective aspects of nutrition but seldom discuss the mechanistic role of DSMs in AHN and cognition. Some of the first to systematically review research on the mechanistic effects of DSMs on the brain were Gomez–Pinilla55 and Zheng and Berthoud.56 A list of some of the dietary factors discussed in these reviews is shown in Table 1. While not an exhaustive list, Table 1 underscores the multifunctional character of DSMs that range from sources of energy and building blocks for cellular structures to chemical signals that trigger gene regulatory cascades that control transcriptional programs in the brain. One of the best examples of a multifunctional DSM is the long-chain polyunsaturated fatty acid (PUFA), DHA, which can serve as a nutrient, transcription regulator, immuno-modulator and neurotransmitter.
Reduced brain or circulating DHA concentration has been implicated in depression, bipolar disorder and attention deficit (AD) disorder.57,58,59 However, intervention studies with long-chain omega-3 PUFAs have yielded mixed results.5,57,58,59,60 One recent meta-analysis, however, suggested an overall beneficial effect for EPA in major depressive disorder patients, especially at high doses.61 Interestingly, the beneficial effects of EPA were also observed in subjects taking antidepressants. Whether the beneficial effects of high EPA dosage together with antidepressants are additive or synergistic can have significant therapeutic implications and thus requires further study. Most studies designed to assess the benefits of omega-3 PUFAs on children with AD disorder are inconclusive. Another recent study, however, showed significant improvement in working memory for children with attention deficit hyperactive disorder supplemented with EPA62 (see Box 2).
For DSMs to impact various neurological structures and functions in ways that produce neurogenesis, synaptic plasticity and adaptive behaviors, there must be an efficient communication system allowing dietary stimuli to be delivered to the brain from the gut. These connections are provide by the 400–600 million neurons in the human enteric system77 that creates a virtual information highway through which DSMs can communicate critical chemical information from the environment to the brain.78 Alternatively, oxygen and nutrients in peripheral blood can be delivered to the brain via the middle cerebral arteries and their fenestrated capillaries to support hippocampal6 and hypothalamic functions.79 These communication channels permit dietary inputs to be more than just fuel and building blocks for the brain, but also a means for delivering important chemical signals from the extracellular environment to the neuron where they are continually integrated into those signaling pathways and neuronal activity needed for metabolic homeostasis, cognition and overall health.55,56,79 Box 3, Box 4.
Nucleic acid-based regulation of the food–brain axis
Central to our understanding of how the food–brain Axis functions is the premise that homeostasis and the adaptive behaviors associated with cognition are the result of the complex interplay between signal transduction and transcriptional programming in the neuron. Fortunately, evidence is mounting to support the notion that the brain’s cognitive and non-cognitive functions are encoded genomically80,81 and subject to epigenetic1,82,83,84 and epitranscriptomic4 modification. In the case of epigenetics, a recent study demonstrated that transient activation of mature neuronal circuits can up-regulate and down-regulate transcription, particular for early genes like Arc, c-Fos and Jun-b, by dynamic modification of chromatin accessibility (i.e., chromatin condensation or de-condensation).27 The authors of this study concluded that activity-induced reshaping of the transcriptome plays an important role in regulating synaptic plasticity, cognitive function and neurological disorders. Other studies show that activity-induced epitranscriptomics (i.e., post-transcriptional RNA editing and RNA methylation of coding and non-coding RNAs) in the brain can produce experience-dependent plasticity leading to learning and memory.4,26,85
One of the challenges to understanding a nucleic acid-based model for cognition is the temporal scales that can span several orders of magnitude between stimulus sensing and experience-dependent neuronal plasticity. For example, the time to elicit an electrophysiological signal from a neuron is in the range of microseconds, while the initiation of transcription and translation of RNA can span minutes to hours. On the other hand, the time required for learning and durable memory might take days to years.85 This temporal discordance makes the trajectory between stimulus sensing and neuronal plasticity non-linear and difficult to understand.
One possible way to resolve these manifold differences in timing between sensing and response is to view them in the context of molecular processes that affect the rate and magnitude of each step along the trajectory. These molecular processes include epigenetic and epitranscriptomics modifications of chromatin, DNA and RNA (e.g., altered RNA structure, half-life, localization and ligand affinity and, methylation of histones, DNA and RNA85,86,87) and homeostatic scaling.86,88 All of these processes help create the diversity of transcription programs required for proper neuronal function. In the case of homeostatic scaling, both the proteome88 and transcriptome86 are altered to adjust synaptic strength up or down in response to changes in inputs. Deficiencies in homeostatic scaling are associated with neurological disorders such as autism spectrum disorder, epilepsy, Parkinson’s and schizophrenia and underscore the need for tight control over network activity for proper neuronal function.88 Interestingly, many of the transcription factors (e.g., CREB, Elk1, SRF), kinases (e.g., CaMK, CDK5, MAPK) and growth factors (e.g., BDNF) associated with homeostatic scaling are also components of pathways (e.g., ERK) that crosstalk with neurogenesis signaling (Fig. 2, Transactivation of Signaling Pathways, Supplemental Table 1 A comparison of signal transduction pathway relative of signaling for glucose signaling).86 In terms of the potential impact of inputs like DSMs on homeostatic scaling, expression of BDNF, a neurotrophic factor involved in neurogenesis, memory and learning, can be triggered by a variety of polyphenolics compounds found in plant-based foods (Table 2).
Miguel Cruz is a Spanish Director based in Los Angeles with credits in more than 200 TV episodes, including the Top Rated Sitcom "Aída" or the Emmy Winning "Caiga Quien Caiga". His first film "VULNERABLES", is an independent thriller with international distribution, 4th at the Peruvian Box-office and acquired by Sundance Channel. He is currently in development of his second film, a comedy with FOX.
Miguel has also directed dozens of commercials for international brands and combines his extensive professional career with a strong relation with Academia. A Fulbright Scholar, Miguel serves as Vice President of the Fulbright Association in Los Angeles. He is also Associate Chair of the Acting Department of the New York Film Academy, where he shares his experience and knowdledge with aspiring actors and filmmakers.