As the older adult population expands, the economic burden of care and treatment of age-related health disorders also rises. Between 2015 and 2060, the United States will experience significant growth of its older population, with the size of the population aged 65 and over more than doubling from an estimated 46 million in 2015 to 98 million in 2060 (Mather et al., 2015). Therefore, successful strategies to promote healthy brain aging are of significant interest to public health initiatives in the United States.

Nutrition is a promising target for intervention efforts to support healthy brain aging (Zamroziewicz and Barbey, 2016). Accumulating evidence indicates that particular nutrients may slow or prevent aspects of age-related cognitive decline by targeting specific features of brain aging. Studies that couple neuroimaging techniques with neuropsychological testing provide insight into mechanisms of action through which particular nutrients might influence specific aspects of age-related cognitive decline (Bowman et al., 2012; Zamroziewicz et al., 2015; Boespflug et al., 2016; Gu et al., 2016). While some nutrients may be effective at preventing late-life changes in the brain, other nutritional factors may accumulate across the lifespan and therefore confer neuroprotection in the aging brain (Söderberg et al., 1990; Coyle and Puttfarcken, 1993). Identifying the mechanisms through which nutrients provide neuroprotective effects will help guide the development of successful lifelong dietary strategies for healthy brain aging.

Carotenoids are naturally occurring pigments made by plants, and can only be acquired through the diet (Erdman et al., 2015). Xanthophylls are a subclass of carotenoids, which have a polar molecular structure and therefore possess unique membrane-spanning properties (Erdman et al., 2015; Widomska et al., 2016). As compared to the other dietary carotenoids, xanthophylls preferentially accumulate in neural tissue, with lutein accounting for the majority of carotenoid accumulation in the brain (Craft et al., 2004; Johnson, 2012; Johnson et al., 2013; Li et al., 2014; Vishwanathan et al., 2014b; Widomska et al., 2016). As the most prevalent carotenoid in the brain, lutein is thought provide a variety of neuroprotective benefits. Candidate mechanisms of action are primarily based on the unique membrane spanning properties of lutein and include influencing membrane properties, such as fluidity, interneuronal communication via gap junctions, ion exchange, oxygen diffusion, membrane stability, and preventing oxidation and inflammation (Stahl and Sies, 1996, 2001; Paiva and Russell, 1999; Krinsky, 2002; Izumi-Nagai et al., 2007; Johnson, 2014; Widomska and Subczynski, 2014; Erdman et al., 2015). Lutein accretes in the brain across the entire lifespan, and may therefore contribute to brain health in a lifelong manner (Renzi et al., 2014; Bovier and Hammond, 2015; Lieblein-Boff et al., 2015). Notably, lutein is selectively distributed in gray matter, and has been detected in the prefrontal cortex, the temporal cortex, and the hippocampus (Craft et al., 2004; Vishwanathan et al., 2014b). Blood levels of lutein correlate with brain concentrations of lutein in older adults, suggesting that blood concentrations can serve a measure of lutein status in the brain (Johnson et al., 2013).

Lutein status has been linked to cognitive performance across the lifespan (Feeney et al., 2013; Johnson et al., 2013; Renzi et al., 2014; Vishwanathan et al., 2014a; Bovier and Hammond, 2015). Of particular interest, lutein levels have been linked to memory and general intelligence (Feeney et al., 2013; Johnson et al., 2013; Vishwanathan et al., 2014a), which are cognitive constructs closely related to crystallized intelligence. Crystallized intelligence refers to the ability to retrieve and use information that has been acquired throughout life (Horn and Cattell, 1966). Although most aspects of cognitive function undergo age-related decline, certain aspects of cognition—like crystallized intelligence—are spared and even show improvement with age (Craik and Bialystok, 2006; Park and Reuter-Lorenz, 2009). Measuring crystallized intelligence may therefore be a way to assess the lifelong impact of nutritional factors, rather than the immediate effects of nutrients on preventing age-related decline (Craik and Bialystok, 2006).

Crystallized intelligence is dependent upon the temporal cortex, and particular regions of the temporal cortex may play key roles in implementing this cognitive function (Colom et al., 2009; Barbey et al., 2012). In general, structural integrity of the temporal cortex is associated with performance on tasks of crystallized intelligence, but integrity of specific regions within the temporal cortex may underlie the preservation of this cognitive function in aging (Choi et al., 2008). For example, the parahippocampal cortex plays a role in mediating the storage of knowledge about objects (Ricci et al., 1999; Aminoff et al., 2013). This region of the temporal cortex shows resistance to age-related structural decline in the absence of neurodegenerative disease, unlike other regions within the temporal cortex that show significant cortical thinning even in healthy aging (Salat et al., 2004; Jiang et al., 2014). Although the sparing of particular regions within the temporal cortex may be linked to the preservation of crystallized intelligence in healthy aging, the question remains: do particular neuroprotective nutrients like lutein underlie this sparing of cognition and brain health?

In summary, increasing evidence suggests that lutein may be a reliable nutrient biomarker for healthy brain aging: (i) among the carotenoids, lutein accounts for the majority of accumulation in the brain and provides unique neuroprotective benefits, (ii) lutein status has been linked to cognitive performance across the lifespan, and (iii) lutein is selectively distributed in gray matter of brain regions known to underlie the preservation of cognitive function in healthy brain aging. However, the core brain structures upon which lutein may act to preserve cognition have not been investigated. Prior research suggests that lutein plays a neuroprotective role across the lifespan, crystallized intelligence is resistant to age-related decline, and specific regions of the temporal cortex may underlie the preservation of crystallized intelligence. This study aims to explore the role of structures within the temporal cortex in mediating the relationship between serum lutein levels and crystallized intelligence in healthy older adults.

This study revealed that gray matter volume of the right parahippocampal cortex mediates the relationship between serum lutein and crystallized intelligence. This report provides a novel link between lutein, gray matter volume of a specific cortical region, and crystallized intelligence. The individual relationships reported within the mediation, including those between serum lutein levels and parahippocampal cortex (Figure 2, path a), between serum lutein levels and crystallized intelligence (Figure 2, path c), and between parahippocampal cortex and crystallized intelligence (Figure 2, path b), are each supported by prior findings reviewed in turn below.

Prior findings support each relationship reported within the mediation model. The first relationship demonstrated a positive association between serum lutein and gray matter volume of the parahippocampal cortex in the right hemisphere (Figure 2, path a). Four lines of evidence support this finding: (i) serum lutein correlates with brain concentrations of lutein in older adults, (ii) lutein preferentially accumulates in brain tissue, (iii) xanthophylls are selectively distributed in gray matter, as opposed to white matter, and (iv) lutein is found in the temporal cortex (Craft et al., 2004; Johnson, 2012; Johnson et al., 2013; Vishwanathan et al., 2014b). Second, higher serum levels of lutein were associated with higher scores of crystallized intelligence (Figure 2, path c). Prior work demonstrates that intake of green leafy and cruciferous vegetables—both major dietary sources of lutein—is positively associated with cognitive function (Kang et al., 2005; Morris et al., 2006). Indeed, macular pigment optical density (MPOD), a biomarker of lutein concentrations in the brain, has been reported to be significantly related to cognitive function in three independent research groups (Feeney et al., 2013; Vishwanathan et al., 2013, 2014a, 2016; Renzi et al., 2014). Among the carotenoids, lutein most consistently associates with measures of intelligence and memory (Johnson et al., 2008, 2013). Third, the indirect pathway of mediation indicated a mediatory effect of right parahippocampal gray matter volume on the relationship between lutein and crystallized intelligence (Figure 2, path a to b). Previous studies indicate that greater temporal cortex gray matter volume correlates with general intelligence, and the parahippocampal cortex in particular shows activity during tasks that require auditory-verbal recall (Grasby et al., 1993; Colom et al., 2009). Furthermore, healthy aging is characterized by the preservation of crystallized intelligence as well as structural integrity of the parahippocampal cortex (Horn and Cattell, 1966; Salat et al., 2004; Jiang et al., 2014). The unilateral nature of this mediation is supported by work suggesting that the right hemisphere is more resistant to age-related and disease-related neurodegeneration than the left hemisphere (Chételat et al., 2005; Querbes et al., 2009; Risacher et al., 2010; Mosconi et al., 2014).

The underlying physiological mechanisms of the relationship between serum lutein levels, crystallized intelligence, and cortical integrity of the parahippocampal cortex are 3-fold. First, lutein is differentially localized to membrane domains high in polyunsaturated fatty acids, and is therefore well positioned to prevent oxidation of vulnerable lipids (Widomska and Subczynski, 2014). Second, by preventing lipid oxidation, lutein may help preserve long-chain polyunsaturated fats, such as docosahexaenoic acid (DHA), for cleavage and conversion into anti-inflammatory compounds (Miller et al., 2014). Third, lutein is a polar and soluble molecule, and can therefore span the membrane and influence membrane properties, such as fluidity, ion exchange, oxygen diffusion, membrane stability, and interneuronal communication via gap junctions (Stahl and Sies, 2001). Importantly, these neuroprotective effects are relevant across the lifespan, and may therefore be particularly important for the preservation of cognitive functions that are enhanced with increasing age, such as crystallized intelligence (Horn and Cattell, 1966; Bovier and Hammond, 2015; Lieblein-Boff et al., 2015).

The partial nature of this mediation is supported by the specificity of this analysis, which focused on one aspect of cognition and its underlying cortical structures. Given that lutein is thought to influence a wide range of biological processes in the brain, it is unlikely that the role of lutein in the aging brain is limited to crystallized intelligence and the parahippocampal cortex (Erdman et al., 2015). Rather than claiming the parahippocampal cortex as the sole mediator of the relationship between lutein and crystallized intelligence, this study suggests that preserving structural integrity of the parahippocampal cortex is one mechanism through which lutein contributes to the preservation of cognitive function. These findings add to a growing line of evidence which suggests that particular nutrients may slow or prevent aspects of age-related cognitive decline by targeting specific features of brain aging (Bowman et al., 2012; Zamroziewicz et al., 2015; Gu et al., 2016). In the case of lutein, future studies are needed to assess whether lutein is uniquely protective of crystallized intelligence and temporal cortex structure, or whether other carotenoids contribute neuroprotective effects as well. Another promising direction for future work is to examine the interactive effects among nutrients through the use of nutrient biomarker pattern analysis—a technique that enables an investigation of the beneficial effects of broader nutrient profiles on healthy brain aging. Ultimately, this line of research can inform a comprehensive and personalized approach to nutritional intervention that takes into account dietary patterns and individual variability in nutritional status and brain health.

The strengths of the present study include: (i) the use of blood biomarkers to measure physiological status of lutein, (ii) the use of structural magnetic resonance imaging to measure regional cortical integrity with high spatial resolution, and (iii) the assessment of a particular component of cognitive function known to be highly variable across individuals, rather than a global cognitive function measure with little variability within a sample. The limitations of this study include: (i) relatively small sample size, (ii) cross-sectional design, (iii) lack of MPOD measurement to better reflect lutein concentrations in neural tissue (Vishwanathan et al., 2013, 2016), and (iv) lack of assessment of other carotenoids. Thus, directions for future research include: (i) larger samples to confirm these results, (ii) longitudinal studies to investigate the relationship between serum lutein and crystallized intelligence in different stages of life, (iii) MPOD measures of lutein in neural tissue, and (iv) assessment of other carotenoids to determine potential neuroprotective benefits.

MZ is the primary author on this manuscript. AB is the primary investigator and contributed to study design and manuscript drafting. EP, CZ, EJ, MK, and NC contributed to study design and manuscript drafting.