Anatomical biology guides a search for nutrients for the aging brain

Abstract

Considerable evidence have established the importance of specific nutrients that have been found vital for the developing brain. We hypothesize that in a similar manner there should be nutrients vital to the aging brain and that based on its distinct pathophysiology they should be different than those essential to development. Specific brain networks that govern cognition are particularly vulnerable to the aging process, resulting in what is referred to as ‘cognitive aging’. Common late-life disorders, however, such as Alzheimer’s disease also target these same brain networks. Studies have disambiguated cognitive aging from late-life disease by isolating regions and biological pathways within each network differentially linked to one or the other. This anatomical biology anchors a framework to identify nutrients relevant to cognitive aging whose utility is illustrated via a decades-long research program into how dietary flavanols benefit the brain. As we are living longer in cognitively more demanding lives, the framework’s ultimate goal is to generate specific dietary recommendations that will fortify our mind for its golden years.

Introduction

The identification and validation of folate as a nutrient for the nervous system’s development is considered a crowning achievement of nutritional neuroscience. The search for developmentally-sensitive nutrients began after mid-20^th century studies suggested that, in contrast to other developmental malformations, neural tube defects are linked more closely to environmental rather than genetic factors¹. While folate was first implicated by observational studies, its etiological role was ultimately validated by rigorous experimental studies using folate biomarkers and by clarifying its precise mechanisms of action.

Might there exist nutrients similarly important for the aging brain? In considering this question, it is notable that modern nutritional science began at a time when the average life expectancy was approximately 50 years. The early focus on nutrients for nervous system development and not aging is therefore understandable. But as life expectancy dramatically climbed during the 20^th century there emerged a more important barrier in addressing this question: the biological ambiguity between brain dysfunction caused by the aging process versus that caused by diseases of late life, in particular Alzheimer’s disease².

Distinguishing Alzheimer’s disease and other neurodegenerative disorders that are typically genetically based^3,4 from age-related brain dysfunction is critical because the latter is likely linked more closely to environmental rather than genetic factors. Additionally, in contrast to neurodegeneration, age-related brain dysfunction is less pernicious, mediated by dysfunction of neuronal synapses, never resulting in wholesale neuronal death⁵. Synaptic function is in principle more likely to be sensitive to subtle environmental factors such as dietary elements and that unlike neuronal death, synaptic dysfunction can be reversed. These pathophysiological distinctions justify the hypothesis that nutrients are mechanistically more likely to be linked to age-related brain dysfunction, and if identified can help to prevent or even in reverse it.

While age-related brain dysfunction and late-life disorders like Alzheimer’s disease target the same cognitive networks of the brain, as reviewed below, studies have disambiguated one from the other by isolating brain regions and underlying biological pathways differentially linked to each. Accordingly, we propose a framework guided by this anatomical biology to identify and validate nutrients that may be etiologically linked to age-related brain dysfunction. This framework is illustrated by a decades-long research program examining how dietary flavanols benefit the brain, highlighting which future studies are needed to validate nutrients that may be vital for the aging brain.

The anatomical biology of brain aging

Among the many brain networks that govern cognition, two are primarily affected by the deleterious effect of the aging process independent of disease, giving rise to what is referred to as ‘cognitive aging’ ⁶: The prefrontal cortex and its role in executive and related abilities, and the hippocampus, governing the ability to learn and remember new information. The two components of cognitive aging, therefore, are age-related prefrontal dysfunction and age-related hippocampal dysfunction.

The problem is that both brain areas also are affected by neurodegenerative disorders, in particular the most common disorder of late-life, Alzheimer’s disease⁷. The logic of anatomical biology has been used to distinguish cognitive aging from Alzheimer’s disease. Both areas are networks that comprise different but interconnected brain regions. Each houses neuronal populations that are distinct in their molecular and physiological properties. Since aging and disease are assumed to be mechanistically distinct, anatomical biology predicts that aging and disease should differentially target individual regions within each network driven by distinct biological pathways².

Over the course of the recent decades, this assumption has been confirmed, particularly for age-related hippocampal dysfunction and to a lesser degree for age-related prefrontal dysfunction. From a series of observational studies in older adults with and without Alzheimer’s disease, aging non-human primates and rodents, mouse models of Alzheimer’s disease (reviewed in⁸ and more recently⁹), and interventional studies in older adults without disease^10,11, a double-anatomical dissociation has been established within the hippocampal network between aging and Alzheimer’s disease. Across the regions of this network, the dentate gyrus is the hippocampal region identified as differentially vulnerable to aging and a neighboring hippocampal region, the entorhinal cortex, is found differentially resistant to aging (Fig. 1). The reverse pattern is found for Alzheimer’s disease. These anatomical profiles have been used to identify biological pathways whose disruption underlies dentate gyrus dysfunction and drives hippocampal-dependent memory loss in aging^12,13, and pathways that when disrupted underlie entorhinal cortex dysfunction in Alzheimer’s disease^14,15. Confirming the principle of anatomical biology¹⁶, the two pathways are distinct and non-overlapping.

Figure 1. Anatomical biology distinguishes cognitive aging from Alzheimer’s disease.

Both cognitive aging and Alzheimer’s disease target the prefrontal cortex and the hippocampus. As shown in red and biding by the tenets of anatomical biology, compared to disease, cognitive aging differentially targets the inferior frontal gyrus within the prefrontal cortex, and the dentate gyrus within the hippocampus. The entorhinal cortex shown in blue, is the brain’s cortical region most resistant to aging without Alzheimer’s disease, but most vulnerable to Alzheimer’s disease (Adapted from Zeng et al, PLOS One, 2020).

In the case of age-related prefrontal dysfunction, a less extensive series of observational^9,17 and interventional studies¹⁸ in healthy older adults without disease has identified the inferior frontal gyrus as the region that is differentially targeted by age-related prefrontal dysfunction (Fig. 1), although the underlying biological pathway that drives this dysfunction remains unknown.

Collectively, the anatomical biology of cognitive aging establishes anchoring points in constructing a framework for determining whether there are nutrients that are linked to age-related hippocampal dysfunction on the one hand or age-related prefrontal dysfunction on the other.

Identifying candidate nutrients for the aging brain

The framework proposes that any dietary element linked to the dentate gyrus might be a candidate as a nutrient needed to preserve hippocampal function during aging.

On the heels of a cell culture screen of ‘generally recognized as safe’ compounds that suggested that (–)-epicatechin might be linked to dentate gyrus neurons, mouse studies¹⁹ investigated the in vivo effect of (–)-epicatechin, a key component of dietary flavanols commonly found in tea, apples, berries, grapes, cocoa and other fruits and vegetables. These studies showed that (–)-epicatechin improves hippocampal-dependent memory by enhancing angiogenesis and synaptogenesis specifically in the dentate gyrus, and other preliminary studies suggest that it plays a role in improving the biological pathway in the dentate gyrus specifically linked to cognitive aging²⁰.

Inspired by these observations, (–)-epicatechin-enriched dietary flavanols were tested in two randomized-controlled studies to determine whether they target the dentate gyrus to improve hippocampal function in healthy older adults. The first was a relatively small study¹⁰ that was designed to mechanistically confirm observational studies implicating the dentate gyrus as the anatomical hub of age-related hippocampal dysfunction. The second study¹¹ included a greater number of participants, tested different levels of (–)-epicatechin-enriched dietary flavanols, collected more information about baseline diet, and had a washout period that tested cognition after the flavanol interventions were stopped.

Results from these studies support the possibility that (–)-epicatechin may act as a nutrient linked to age-related hippocampal dysfunction. Consistent with the framework’s anatomical biology, the (–)-epicatechin-enriched interventions were found to selectively improve hippocampal-dependent memory. Moreover, as suggested by neuroimaging findings, the interventions’ improvement of hippocampal-dependent memory was mediated primarily via the dentate gyrus acting as the hub of the hippocampal network (Fig. 2A).

Figure 2. Anatomical biology generates mechanistic hypotheses.

A. Left panel: MRI Maps of hippocampal cerebral blood volume (CBV), sensitive to capillary density and/or synaptic function, shows that age-related decline localizes primarily to the dentate gyrus (CBV decreases color-coded in red). Right panel: An (–)-epicatechin-enriched dietary intervention in older people results in the greatest increase in CBV localized mainly to the dentate gyrus (CBV increases color-coded in yellow) (From Brickman et al, Nature Neuroscience, 2014).

B. Two growth factors, BDNF and VEGF, have been linked to (–)-epicatechin and to maintaining synaptic and epithelial health. Across cortical and hippocampal regions BDNF, VEGF, and the VEGF receptor (VEGFR) are localized primarily to the dentate gyrus (from the Human Allen Brain Atlas).

More mechanistic support came from the larger study. By assessing baseline habitual diet using the ‘alternative healthy eating’ index (aHEI), a healthy diet selectively correlated with hippocampal-dependent memory and the flavanol interventions improved cognition most reliably in those participants whose study entry dietary quality was poor. Because the aHEI assesses habitual consumption of foods enriched in (–)-epicatechin, these observations provided the first clue that a relative (–)-epicatechin ‘deficiency’ might drive age-related hippocampal dysfunction in older adults.

The washout period also was informative. During the active phase of the study, the participants were randomly assigned to different levels of flavanol consumption or placebo for a 3-month period, at which point the intervention was stopped and the participants were cognitively evaluated after the 8-week washout After this washout phase, the memory improvements of the participants observed at the end of the intervention returned to their study-entry baseline. This ‘ABA’ experimental design strengthens the mechanistic link between (–)-epicatechin enriched dietary flavanols and hippocampal-dependent memory.

Finally, the second study used a newly validated biomarker that measures the main (–)-epicatechin metabolite produced by the gut microbiome and taken up by the body when consumed in dietary flavanols^21,22. The biomarker documented a ‘dose’ dependent increase of the (–)-epicatechin metabolite across the flavanol intervention conditions, and a reduction to baseline levels after the washout period. More importantly, levels of the biomarker selectively correlated with hippocampal-dependent not prefrontal-dependent cognitive abilities, providing the first evidence linking (–)-epicatechin per se to hippocampal function in people.

Collectively, this series of translational mouse-to-human studies justifies the conclusion that (–)-epicatechin represents a candidate nutrient for the aging brain.

Validating nutrients for the aging brain

The recent development of the (–)-epicatechin biomarker is a major technical milestone for the goal of validating whether dietary flavanols can be considered the first bona fide nutrient for the aging brain. Just as for folate, the goal standard, a reliable measure of any potential nutrient is required to first show that it is correlated with a disorder’s defining readout, as already documented between the (–)-epicatechin biomarker and dentate gyrus-mediated hippocampal function¹¹.

Just as with folate, but more importantly for validation, it is now possible to incorporate this biomarker into flavanol intervention studies. In this way, the causal logic of ‘depletion-repletion’ can be tested by showing that (–)-epicatechin is not simply a nootropic supplement that benefits all takers, but rather, only benefits those older adults with relatively lower baseline levels of (–)-epicatechin consumption. These two interlinked validating assumptions are currently being tested in the context of a third randomized-controlled study (COSMOS-Web, NCT04582617), this time in thousands of older adults assigned either to an (–)-epicatechin-enriched flavanol intervention vs. placebo control, using the (–)-epicatechin biomarker in conjunction with the ‘alternative healthy eating index’.

A parallel but no less important validation step would be mechanistically explaining precisely why (–)-epicatechin is associated with hippocampal function. Here, too, anatomical biology can guide the generation of specific and testable hypotheses. The dentate gyrus is a brain region particularly sensitive to numerous endogenous blood-born factors^23–26 whose levels change during the aging process that, as shown in mice, are associated with age-related hippocampal dysfunction. This differential sensitivity must be explained by cellular and molecular distinctions, which are at least in part explained by the fact that the dentate gyrus is one of the rare brain regions that supports neurogenesis after birth^27,28. This unique developmental feature has been hypothesized to ‘imprint’ upon the cells of the dentate gyrus a distinct molecular expression profile⁸.

(–)-Epicatechin can be considered an exogenous blood-born component of our diets that, as reviewed above, differentially targets the dentate gyrus by inducing regionally-selective angiogenesis and synaptogenesis. Since loss of both capillaries and synapses in the dentate gyrus is precisely what characterizes the cellular pathophysiology of age-related hippocampal dysfunction (Fig. 2A), (–)-epicatechin’s memory-enhancing effect in older adults makes physiological sense. Similar to how an age-related decline in endogenous blood-born factors can specifically target the dentate gyrus, it is now possible to use the (–)-epicatechin biomarker to test whether this exogenous blood-born dietary element declines with age.

What are the molecular mechanisms that link (–)-epicatechin to endothelial and synaptic health in the dentate gyrus? The answer remains unknown and should be considered one cell-type at a time. Since (–)-epicatechin crosses the blood brain barrier, it may directly affect both the endothelium and synapses of the dentate gyrus, perhaps in a synergistic manner.

Starting with endothelial cells, studies have linked (–)-epicatechin to VEGF (Vascular Endothelium Growth Factor) and its secretion^29,30, which in this context can be considered a regulator of endothelial health. In fact, in vivo mouse studies have shown that repleting the age-related decline in one of the endogenous blood-born factors rescues age-related dentate gyrus capillary loss by inducing angiogenesis²⁵. This restorative and normalization benefit is associated with increased VEGF secretion from endothelial cells. Across cortical and hippocampal regions VEGF and its receptor appear to be preferentially expressed in the dentate gyrus (Fig. 2), providing a plausible hypothesis for why (–)-epicatechin is mechanistically linked to maintaining endothelial health in the aging dentate gyrus.

Moving to neurons, here another growth factor, BDNF (Brain-Derived Neurotrophic Factor) a known regulator or synaptic health, almost certainly plays a mechanistic role since mouse studies have linked (–)-epicatechin consumption to increased hippocampal BDNF expression³¹. Even more so than VEGF, BDNF is differentially expressed in the human dentate gyrus (Fig. 2), and so collectively the inescapable hypothesis is that BDNF mediates the link between (–)-epicatechin and synaptic health in the aging dentate gyrus. A second and potentially related hypothesis for how (–)-epicatechin contributes to dentate gyrus synaptic health emerges from the human and mouse studies that have identified a specific histone acetylation pathway that when disrupted is a a causal driver of age-related dentate gyrus dysfunction^12,13.

The hippocampus is highly homologous in mice and humans and dentate gyrus-dependent age-related hippocampal dysfunction is phenocopied in both species. Nevertheless, while more easily tested in mice, it is always better to test mechanistic hypotheses about a human condition in humans. In this regard, genetics can be used to test at least the two hypotheses about the mechanistic links between (–)-epicatechin and synaptic health in humans. BDNF loss-function polymorphisms have been identified that are linked to hippocampal-dependent memory³², and variants in genes encoding key proteins of specific histone acetylation pathway have been identified and are linked to age-related hippocampal dysfunction¹³. Future studies can test whether the cognitive response to (–)-epicatechin interventions will be influenced by those people who carry these genetic polymorphisms and/or variants.

Finally, another mechanistic question worth pursuing is why some people have relatively lower (–)-epicatechin levels. This might simply reflect relatively poorer (–)-epicatechin-enriched diets. Given the fact, however, that even when consumed via a healthy diet, (–)-epicatechin needs to be first metabolized in the gut microbiome before its uptake²¹, lower (–)-epicatechin levels might reflect the established age-related changes to our microbiome³³. If these changes play any role in relative (–)-epicatechin deficiency, even if it somehow interacts with diet quality as would be predicted, future microbiome studies then could isolate the specific gut microbiota that regulate (–)-epicatechin metabolism.

Summary

The framework’s utility for identifying nutrients that are mechanistically linked to cognitive aging was exemplified in the context of dietary flavanols for age-related hippocampal dysfunction. Future studies can rely on this framework to identify other elements in our diets that may fulfill criteria as nutrients for cognitive aging, by targeting either age-related hippocampal or prefrontal dysfunction.

The pathophysiology of cognitive aging supports the hypothesis that, more so than late-life disorders, nutrients not only can drive cognitive aging but that even when initiated nutrients are more likely to ameliorate the process once it has begun. Thus, in contrast to late-life disorders like Alzheimer’s disease, which are more likely to require pharmaceutical interventions, dietary interventions for cognitive aging have the additional bioethical advantage of being more easily accessible to all¹⁶.

Just as there currently exist public health recommendations for dietary interventions to fortify the developing nervous system, the ultimate goal of this research enterprise is to begin assembling the equivalent recommendations to fortify our aging minds so that increases in longer, healthier lives are accompanied by preservation in our cognitive abilities.