Abstract
Daily supplementation of blueberries (BBs) reverses age-related deficits in behavior in aged rats. However, it is unknown whether BB is more beneficial to one subset of the population dependent on baseline cognitive performance and inflammatory status. To examine the effect of individual differences on the efficacy of BB, aged rats (17 months old) were assessed for cognition in the radial arm water maze (RAWM) and divided into good, average, and poor performers based on navigation errors. Half of the rats in each cognitive group were then fed a control or a 2% BB diet for 8 weeks before retesting. Serum samples were collected, pre-diet and post-diet, to assess inflammation. Latency in the radial arm water maze was significantly reduced in the BB-fed poor performers (p < .05) and preserved in the BB-fed good performers. The control-fed good performers committed more working and reference memory errors in the post-test than pretest (p < .05), whereas the BB-fed good performers showed no change. An in vitro study using the serum showed that BB supplementation attenuated lipopolysaccharide (LPS)-induced nitrite and tumor necrosis factor-alpha, and cognitive performance was associated with innate anti-inflammatory capability. Therefore, consumption of BB may reverse some age-related deficits in cognition, as well as preserve function among those with intact cognitive ability.
As the number of older adults in the United States continues to increase, there is a strong need to identify modifiable factors that regulate the proper maintenance of brain function and facilitate healthy aging. The brain is highly susceptible to oxidative stress (OS) and inflammation, both of which increase with age; this age-related vulnerability is accompanied by a loss of efficiency in the body’s natural defense mechanisms (1,2). It is this increased susceptibility to the long-term effects of OS and inflammatory insults that is thought to contribute to the decrements in cognitive and/or motor performance seen in aging and neurodegenerative disease (3,–5). Studies in both humans and animals have demonstrated that inflammation and impaired insulin resistance are common features in cardiometabolic and vascular disease, obesity, and age-related cognitive and motor decline (6,–9). Therefore, a key to reducing the incidence of age-related deficits in behavior may be to decrease peripheral OS and inflammation.
Dietary supplementation with individual foods and food constituents has been shown to induce protective effects against some aspects of age-related behavioral and cellular dysfunction (10,11). This protection may be because of non-nutrient phytochemicals contained in these foods, which exhibit antioxidant and anti-inflammatory activities and can reduce the age-related sensitivity to OS or inflammation (12,13). In particular, dark colored fruits and vegetables are rich in polyphenols (12), and evidence has shown that consumption of these foods may be an effective strategy to forestall or even reverse age-related neuronal deficits resulting from neuroinflammation (5,14,15). In rodent studies, supplementation for approximately 2 months with berry fruits, such as blueberry (BB), strawberry, raspberry, and blackberry, significantly attenuated age-related motor and cognitive deficits in senescent rodents (16,–22). Importantly, recent human intervention studies have also shown that consumption of flavonoid and polyphenols is associated with benefits to cognitive function (for reviews see References 23 and 24 and also References 25,–28). Accumulating evidence suggests that the polyphenols found in berry fruit may possess a multiplicity of actions, in addition to their anti-inflammatory and antioxidant activities, such as direct effects on signaling, neurogenesis, and autophagy (reviewed in References 5 and 29).
In humans, the aging process is affected by genetics, nutrition, level of physical activity, and exposure to health risks over the course of life. This leads to individual differences in the rate and the severity of cognitive decline. Even though the benefits of BBs have been well documented in animals (21) and humans (28), it is still unclear whether the brain health benefits of BBs are a global phenomenon or if BBs are more beneficial to a specific subset of the population. If neuroinflammation is an underlying factor in cognitive decline, an aged individual with poor cognitive function might be predicted to have a higher degree of brain inflammation than one with more efficient cognition. A recent study in our laboratory examined whether raspberries differentially improved age-related declines in psychomotor function, depending on baseline motor ability and inflammation status, in aged Fischer 344 rats (19). We found that poor performers showed the greatest improvement in motor function when fed for 8 weeks with 1%–2% raspberry supplemented in the diet, whereas average and good performers were not as likely to benefit from raspberry supplementation. In addition, although pre-diet levels of circulating pro-inflammatory cytokines were not predictive of motor performance in this study, higher levels of interleukin-1β were correlated with reduced motor performance post-diet (19).
The purpose of the present study was to explore the effects of individual differences on the efficacy of dietary BB for enhancing cognition and reducing inflammation. We chose to assess cognition in the radial arm water maze (RAWM) because it allows the ability to assess both reference and working memory errors simultaneously (30,31). Furthermore, the testing protocol for the RAWM was a modification of a multiday form of water maze training that assesses hippocampal-dependent memory and is challenging enough to allow rats to be tested on multiple days (32,33). We hypothesized that BBs would have differential benefits based on cognitive capacity, and that the improvement would be associated with a reduction in the factors that were previously identified as the facilitators of cognitive decline, that is, inflammation.
Methods
Animals
Sixty 17-month-old male Fischer 344 rats, obtained from the NIA colony (Charles River, Frederick, MD), were individually housed in stainless-steel-mesh-suspended cages, provided food and water ad libitum, and maintained on a 12-hour light-dark cycle. Following a 2-week acclimatization period to the facility, which included extensive handling, the rats were pretested in the RAWM (see later. On the basis of their performance on the RAWM (total number of errors), the rats were divided into poor, average, and good performers (average number of errors = 4.24, 3.53, and 2.78, respectively). The rats were then weight-matched and assigned to receive either a control diet or 2% BB diet (described later; n = 10/group). After 8 weeks on their respective diets, the rats repeated RAWM testing.
Body weights were recorded at several time-points throughout the study and a food intake (over a 72-hour period) was assessed during the third week of experimental feeding. There were no differences in average weight (p > .05) or food intake (p > .05) between the groups during the study (Table 1). All rats were observed daily for clinical signs of disease. Four rats died or were killed because of excessive weight loss prior to initiation of the diet. During the course of the study, 10 rats died or were killed because of excessive weight loss: 2 in the 0% poor, 1 in the 0% average, 2 in the 2% poor, 3 in the 2% average, and 2 in the 2% good group. There were no differences in mortality between the groups and these numbers are in line with mortality rates for this strain and age of rats. In addition, data from one weak swimmer (ie, swimming slowly and sinking) in the 2% poor group was not analyzed. Table 1 presents final numbers of rats/group. Animals were used in compliance with all applicable laws and regulations as well as the principles expressed in the National Institutes of Health, USPHS, and the Guide for the Care and Use of Laboratory Animals. This study was approved by the Tufts Medical Center Animal Care and Use Committee.
Average Body Weights and Food Intakes Throughout the Study
| Blueberry | N | Body Weight (g) | Food Intake (g) | ||
|---|---|---|---|---|---|
| Cognitive ability | Poor | 0% | 7 | 431.82 (7.44) | 20.80 (1.03) |
| 2% | 7 | 437.05 (3.73) | 23.52 (0.52) | ||
| Average | 0% | 8 | 433.62 (12.42) | 22.17 (1.02) | |
| 2% | 7 | 429.54 (7.20) | 21.30 (0.91) | ||
| Good | 0% | 9 | 430.50 (4.61) | 22.33 (0.60) | |
| 2% | 7 | 436.93 (8.51) | 23.99 (0.76) | ||
| Blueberry | N | Body Weight (g) | Food Intake (g) | ||
|---|---|---|---|---|---|
| Cognitive ability | Poor | 0% | 7 | 431.82 (7.44) | 20.80 (1.03) |
| 2% | 7 | 437.05 (3.73) | 23.52 (0.52) | ||
| Average | 0% | 8 | 433.62 (12.42) | 22.17 (1.02) | |
| 2% | 7 | 429.54 (7.20) | 21.30 (0.91) | ||
| Good | 0% | 9 | 430.50 (4.61) | 22.33 (0.60) | |
| 2% | 7 | 436.93 (8.51) | 23.99 (0.76) | ||
Note. Data are represented as mean (SEM).
Average Body Weights and Food Intakes Throughout the Study
| Blueberry | N | Body Weight (g) | Food Intake (g) | ||
|---|---|---|---|---|---|
| Cognitive ability | Poor | 0% | 7 | 431.82 (7.44) | 20.80 (1.03) |
| 2% | 7 | 437.05 (3.73) | 23.52 (0.52) | ||
| Average | 0% | 8 | 433.62 (12.42) | 22.17 (1.02) | |
| 2% | 7 | 429.54 (7.20) | 21.30 (0.91) | ||
| Good | 0% | 9 | 430.50 (4.61) | 22.33 (0.60) | |
| 2% | 7 | 436.93 (8.51) | 23.99 (0.76) | ||
| Blueberry | N | Body Weight (g) | Food Intake (g) | ||
|---|---|---|---|---|---|
| Cognitive ability | Poor | 0% | 7 | 431.82 (7.44) | 20.80 (1.03) |
| 2% | 7 | 437.05 (3.73) | 23.52 (0.52) | ||
| Average | 0% | 8 | 433.62 (12.42) | 22.17 (1.02) | |
| 2% | 7 | 429.54 (7.20) | 21.30 (0.91) | ||
| Good | 0% | 9 | 430.50 (4.61) | 22.33 (0.60) | |
| 2% | 7 | 436.93 (8.51) | 23.99 (0.76) | ||
Note. Data are represented as mean (SEM).
Diets
The diets were prepared at Envigo (previously Harlan Teklad, Madison, WI) by adding freeze-dried Tifblue BB powder (T-10711, provided by the U.S. Highbush Blueberry Council, Folsom, CA) to the control diet (20 g/kg diet, 2% w/w). The BB powder, as characterized by Brunswick Labs (Southborough, MA), contained 19.18 mg/g anthocyanins, 36.03 mg/g phenolics, and had an oxygen radical absorbance capacity (ORAC) of 369 μM TE/g. The control diet was a modification of the NIH-31 diet, with the amount of corn adjusted to compensate for the added volume of the BB powder. The control NIH-31 diet was the same as used in earlier studies where other fruits, including BBs, were found to be beneficial in mitigating brain aging (17,21,34,–37).
Cognitive Assessment
The RAWM is an age-sensitive test of spatial learning and memory (30). The maze is composed of a circular pool of water, maintained at 23oC, and fitted with internal walls that form eight channels (arms) radiating from a central zone. The pool, walls, and circular escape platform are all colored black, enabling the platform to be submerged and thus hidden from sight. The maze is located in a room with dimmed lights, surrounded by numerous extramaze cues, which the rat must use to effectively navigate the maze. Once placed in the pool, rats must swim to locate the submerged platform (2 cm below water’s surface), hidden in one of the maze arms, on which to escape. Reference memory errors are committed when rats enter an arm that does not contain the platform and reflect learning the trial-independent procedural aspects of the task (spatial locations), whereas working memory errors are trial-dependent and are committed when rats reenter an arm previously explored during that trial.
The RAWM test was conducted both pre- and post-diet, five trials per day, for 5 consecutive days. The platform was moved to a new arm daily, which forced the rats to unlearn the previous day’s platform location and learn its new location. Rats were tested in quasi-random order, with the restriction that one rat from each group be tested in succession. In a given trial, rats were given 90 seconds to find the submerged platform. If rats failed to locate the platform within 90 seconds, the researcher gently guided them to the platform. Once rats arrived at the platform, they were allowed to remain there for 15 seconds, after which rats were immediately placed in a novel start location (trials 2–4) to again locate the platform. The sequence of start locations changed daily without repetition. After the fourth trial, rats were towel-dried then returned to their home cage for 30 minutes, after which they were run on a fifth trial, to test for retention. Trial 5 is a good assessment of short-term memory and memory consolidation, and data from Trial 5 were used in the statistical analyses. Latency and speed were obtained using image tracking software (HVS Image, Hampton, England), whereas arm entries were recorded and scored manually. For a more detailed description of the maze, see Shukitt-Hale and colleagues (30).
Serum Collection
Tail bleeds were performed prior to the introduction of study diets to obtain serum. During blood collection, the rat was gently wrapped in a towel and its tail was warmed with a heating pad. Then a 21–22 g needle from a winged-infusion set was inserted into a lateral tail vein and approximately 1 mL blood was withdrawn. The blood was then allowed to coagulate for 1 hour at room temperature, and the serum was separated from the clot, centrifuged, and stored at −80°C. After 11 weeks on their respective diets and post-diet cognitive testing, the now 20-month-old animals were killed, and trunk blood was collected in a 50 mL Falcon tube and serum collected as earlier. Brain tissue was also extracted for future research.
Cell Culture
HAPI rat microglial cells (generously provided by Dr. Grace Sun, University of Missouri, Columbia, MO) were maintained in Dulbecco’s modified Eagle’s medium (Invitrogen, Grand Island, NY) supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C in a humidified incubator under 5% CO2. Cells were grown in 100 mm plates and then split into 12-well plates prior to treatment. All treatment groups were assayed in duplicate. For experiments, cells were incubated in serum-free Dulbecco’s modified Eagle’s medium without phenol red. Cells were then pretreated with a concentration of 10% serum from individual animals from each of the groups or with 10% fetal bovine serum for the control groups and allowed to sit overnight (~8 hours). Following pretreatment with the serum, the media were removed and then the cells were stimulated with LPS (Sigma-Aldrich, St. Louis, MO) at 100 ng/mL overnight in Dulbecco’s modified Eagle’s medium without phenol red.
Nitrite Quantification
To assess the production of NO from LPS-treated HAPI cells, extracellular release of nitrite (NO2−) was measured by Griess reagent (Invitrogen) according to manufacturer’s instructions. Absorbance was read at 548 nm and the concentration of nitrite was calculated with the linear equation derived from the standard curve generated by known concentrations of nitrite.
TNF-α ELISA
Quantification of tumor necrosis factor-alpha (TNF-α) in cell-conditioned media was performed with an enzyme-linked immunosorbent assay (ELISA, eBioscience, San Diego, CA) according to manufacturer’s instructions. TNF-α concentration for each sample was calculated from the linear equation derived from the standard curve of known concentrations of the cytokine.
Statistical Analyses
For each measure, the interaction between BB supplementation, baseline cognitive ability (poor, average, or good), and time (pre- vs posttest) was assessed using a three-way analysis of variance to test for statistical significance at a level of p < .05 using Systat (SPSS, Inc., Chicago, IL). Post hoc comparisons were performed using Fisher’s least significant difference test. Correlations between cognitive behavior and cell culture serum results were carried out using Pearson’s r correlation.
Results
Overall, short-term daily consumption of BBs benefited poor cognitive performers and preserved cognitive function in good performers. Cognitive function was assessed by the RAWM prior to the start of diet and after 8 weeks on the diet. As expected, there were differences in latency to find the platform between the poor, average, and good performers, as shown by a significant main effect for cognitive group [F(2,78) = 10.34, p < .001] (Figure 1). Statistical analyses also yielded significant interactions between cognitive group and diet [F(2,78) = 3.31, p < .05], cognitive group and time [F(2,78) = 5.95, p < .01], as well as a triple interaction between cognitive group, diet, and time [F(2,78) = 2.99, p < .05], suggesting that the different cognitive performance groups reacted differently to the diets pre- to post-diet. Specifically, post hoc testing showed that poor learners benefited from BB supplementation as latency was significantly reduced at post-test compared to pretest in the BB-fed poor learners (p < .01), and was less than control-fed poor learners at post-test (p < .05). In addition, good performers fed on control diet had longer latencies post-diet than they did at baseline (p < .05), but BB was able to protect against this increase, as good performers that were fed with BB showed no significant change in performance (p > .05) over the course of the study. Average learners fed the BB diet showed increased latencies compared to control-fed average learners at post-test (p < .05); however, a significant change from baseline performance was not observed among average learners in either diet group. These differences were not due to speed as no group differences in swim speed were observed.
Average Trial 5 latency (mean ± SEM; seconds) to find the hidden platform in the radial arm water maze prior to the start of diet (pretest) and after 8 weeks on the diet (post-test) for poor, average, and good performers in the control (0%) or 2% blueberry-supplemented diet groups. Asterisk (*) indicates a significant difference from pre-diet (p < .05); # indicates a significant difference between the diet groups in the same cognitive performance group (p < .05) as measured by Fisher’s least significant difference post hoc test.
Analysis of errors committed in the maze revealed differences in working memory errors between the poor, average, and good performers, as shown by a significant main effect for cognitive group [F(2,78) = 7.02, p < .01] (Figure 2A). There was also a significant interaction between cognitive group and time [F(2,78) =6.18, p < .01] and a trend for a three-way interaction between cognitive group, diet, and time [F(2,78) = 2.71, p = .07]. Post hoc testing only showed one significant difference in that control-fed good performers committed more working memory errors from pre- to posttest (p < .01), whereas BB-fed good performers did not commit significantly more errors over time (p > .05). Results were similar for commission of reference memory errors in that there was a significant interaction between cognitive group and time [F(2,78) = 4.97, p < .01], due to good control-fed rats making significantly more reference memory errors from pre- to posttest (p < .05), whereas BB-fed rats showed no change in errors pre- to post-diet (p > .05) (Figure 2B).
Average number of working (A) and reference memory (B) errors on Trial 5 (mean ± SEM; seconds) in the radial arm water maze prior to the start of diet (pretest) and after 8 weeks on the diet (post-test) for poor, average, and good performers in the control (0%) or 2% blueberry-supplemented diet groups. Asterisk (*) indicates a significant difference from pre-diet (p < .05) as measured by Fisher’s least significant difference post hoc test.
Overall, results showed that serum from BB-supplemented poor and average cognitive performers significantly attenuated LPS-induced nitric oxide (NO) production (Figure 3A) and TNF-α secretion (Figure 3B) in HAPI microglial cells. Specifically, statistical analyses showed that nitrite concentrations had a significant main effect for diet [F(1,74) = 4.56, p < .05] and time [F(1,74) = 5.76, p < .05], as well as significant interactions between cognitive group and time [F(2,74) = 4.93, p < .05] and diet and time [F(1,74) = 10.76, p < .01] (Figure 3A). Post hoc testing showed that poor learners benefited from BB supplementation as nitrite was significantly reduced at post-test compared to pretest in the BB-fed poor learners (p < .01), and reduced compared to control-fed poor learners at post-test (p < .01). For TNF-α expression, there was a significant main effect for diet [F(1,58) = 4.56, p < .05] and time [F(1,58) = 16.02, p < .001], as well as significant interactions for cognitive group and diet [F(2,58) = 3.41, p < .05], cognitive group and time [F(2,58) = 14.46, p < .001], and diet and time [F(1,58) = 13.04, p < .01] (Figure 3B). In this case, both poor and average learners benefited from BB supplementation as TNF-α concentration was significantly less at post-test compared to pretest in the BB-fed poor (p < .01) and average (p < .01) learners, and reduced at post-test compared to control-fed poor (p < .01) and average (p < .01) learners, respectively, even though TNF-α concentration was also significantly less at post-test compared to pretest in the control-fed poor group (p < .05).
Average (mean ± SEM) lipopolysaccharide-induced nitric oxide (NO) production (A; μM) and tumor necrosis factor-alpha (TNF-α) secretion (B; pg/mg) in HAPI microglial cells treated with serum obtained prior to the start of diet (pretest) and after 11 weeks on the diet (post-test) from poor, average, and good performing rats in the control (0%) or 2% blueberry-supplemented diet groups. Asterisk (*) indicates a significant difference from pre-diet (p < .05); # indicates a significant difference between the diet groups in the same cognitive performance group (p < .05) as measured by Fisher’s least significant difference post hoc test.
Correlations between pre-diet cognitive behavior and LPS-induced inflammatory markers in HAPI microglial cells revealed that the concentrations of nitrite (r = .535, p < .001; Figure 4A) and TNF-α (r = .611, p < .001; Figure 4C) in serum were positively related with latency to locate the hidden platform, with lower inflammation associated with shorter latency and higher inflammation with longer latency to the platform on Trial 5 in the RAWM. However, after 8 weeks on the diet, these inflammatory markers showed a positive relationship with latency only to the platform for animals on the BB diet and nitrite (r = .538, p < .05; Figure 4B), but not TNF-α, r = −.331, p > .05; Figure 4D), and not for animals on the control diet for either inflammatory marker [(nitrite, r = −.410, p > .05; Figure 4B) and (TNF-α, r = −.136, p > .05; Figure 4D)].
Correlations between latency to the platform on Trial 5 in the RAWM (seconds) and concentration of lipopolysaccharide-induced nitrite (μM) and TNF-α (pg/mg) in media produced when HAPI microglia cells were preincubated with serum obtained from rats prior to the start of the diet (nitrite, A; TNF-α, C) and after 11 weeks on either a control or blueberry-supplemented diet (nitrite, B; TNF-α, D).
Discussion
Overall, results from this study showed that consuming BB-supplemented diets led to enhanced spatial working memory performance in animals that were poor performers at baseline and preserved cognition in good performers. These effects were more pronounced in the latency measure, as BB-fed poor learners had reduced latencies to find the platform, and BB-fed good learners did not have increased latencies compared to baseline, in contrast to control-fed rats which took longer to find the platform. Latency was also increased in BB-fed average performers from pre- to posttest. It is not surprising that both raspberry and BB were able to improve psychomotor and cognitive performance, respectively, in poor performers, as there was the largest room for improvement in this group. When examining errors in the maze, BB supplementation was only able to protect again the increase in errors seen in the control-fed good performers. One reason for the lesser effect on errors may be that rats generally do not commit a large number of working or reference memory errors, so there is not much room to reduce these numbers (floor effect). The results of this study are somewhat in contrast to those seen in one of our previous studies, which showed that raspberry supplementation did enhance motor performance in poor performers, but did not preserve motor performance among good performers (19). It could be that the decline in psychomotor ability between 17 and 20 months of age in F344 rats is more difficult to protect against than the decline in cognition during this same period, or it could be that BB better preserves function compared to raspberry.
The improvements observed among BB-fed rats in this study agree with those of a previous study in which BB supplementation led to significant improvements in a different version of the RAWM (36). In that study, the platform was located in the same arm for 3 days, and then moved to a new arm on Days 4 and 5. Rats were given 5 trials/day, and an intertrial interval of 30 minutes. Although the rate of learning was similar on Days 1–3 for the control and BB-fed groups, when the platform was moved on Day 4, BB-fed animals committed fewer reference and working memory errors, suggesting that BB supplementation may have enhanced the rats’ capacity to shift strategies (response inhibition), a skill which is important for solving the maze task in the present study. Response inhibition has been found to be affected by aging in that older adults are more susceptible to interference from irrelevant information because of age-related declines in inhibitory ability (38,39).
Exposing young rats to particles of high energy and charge, a ground-based model for exposure to cosmic rays, disrupts cognitive performance and neuronal function, possibly via production of OS and inflammation, and these changes are similar to those seen in aging (40). However, if the animals are fed diets supplemented with antioxidant-rich berries, specifically BBs or strawberries, prior to irradiation, the deleterious effects of the radiation are blocked, leading to better cognitive outcomes (40,41). Interestingly, we found that the different polyphenols in BBs and strawberries might be acting in different brain regions to produce their beneficial effects (41). The irradiated, BB diet animals tended to have shorter latencies to the platform on Day 4 when the platform was moved to a new position in the Morris water maze, which involves reversal learning dependent on intact striatal function, as well as response inhibition. In contrast, the irradiated, strawberry diet animals showed fewer deficits on the probe trial measures in that they spent more time in the platform quadrant and location and had a greater number of crossings of the platform location on probe trial days compared to radiated controls, suggesting retained place information, which is a hippocampally dependent behavior (41). These results agree with those of the present study in that BB seems to facilitate a type of reversal learning, i.e., ignoring where the platform was on the previous day while learning the new platform location. This type of learning requires the modulation or adaptation of learned associations and is similar to executive function in humans, which is also improved by BB supplementation (28,42). Similar to what we found with aging in this study, another study found that exposure to radiation induced linear dose-dependent increases in the proportions of poor learners and in the severity of their impaired learning, whereas good learners showed little evidence of impaired learning across the same dose range (43). It could be speculated that poor learners may be at higher risk for decline because they are more sensitive to radiation-induced OS and inflammation.
Microglial cells are the resident brain macrophages and they are key mediators of neuroinflammation. Therefore, in this study, we examined the in vitro effects of pretreatment with serum from BB-fed animals on microglia in an attempt to determine if the metabolites present in the circulating blood may be mediating the anti-inflammatory effects of BBs, and if one of the possible mechanisms of action for the BBs’ beneficial effects was reducing inflammation. We found that two measures of inflammation were reduced by BB supplementation, that is, LPS-induced nitrite in poor learners, and TNF-α in both poor and average learners. Pre-diet levels of nitrite and TNF-α both predicted performance in the RAWM at baseline, so this result is not surprising as these two groups had the highest baseline levels of these markers, and thus the largest opportunity for reduction. Two recent studies found similar effects in that pretreating BV-2 microglia with serum from aged rats fed 6% and 9% walnut diets (44) or fed 2% acai diets (45) enhanced protection in the microglia to the inflammatory actions of LPS, determined through the measurement of oxidative- and inflammatory-stress-induced signals, including NO and TNF-α release. Inhibiting microglial activation and production of cytotoxic intermediates are important because activated microglia are involved in a number of neurodegenerative conditions (46,47), and we have shown that eating foods with anti-inflammatory properties, such as BBs and walnuts, can protect against microglial activation by increasing levels of protective metabolites in circulating blood.
The suppression of microglial activation may restore homeostasis in the aged or diseased brain and could therefore be one mechanism by which these foods are able to mitigate cognitive decline. In fact, in the present study, as well as in our previous study (45), we found that nitrite levels, following supplementation, were positively correlated with latency to find the platform in a water maze. Neuroinflammation increases with age, as indicated by increased numbers of activated microglia and upregulation of pro-inflammatory signaling pathways, such as the NF-κB pathway (17). Furthermore, a higher degree of NF-κB activation in the brain was found to be negatively correlated with memory performance, and aged rats maintained on a BB diet had lower NF-κB levels compared to control-fed rats, in some cases similar to the levels of young rats (17). Another recent study showed that BB could reduce microglial activation and increase neuroplasticity in middle-age mice fed a high-fat diet (48). Mice supplemented with a high-fat + 4% BB diet had significantly fewer microglia, less microglial Iba1 staining, and reduced the production of nitric oxide in BV-2 microglial cells treated with their serum, compared to animals fed a high-fat diet without BB (48). The authors postulated that one possible mechanism for the reduction in NO could be that the BB metabolites in the serum were inhibiting the activation by LPS of Toll-like receptor 4. Furthermore, these same high-fat diet fed mice displayed recognition and spatial memory deficits, but supplementation with 4% BB reduced these memory deficits (49). The preceding results suggest that BB has potent anti-inflammatory activity that is effective in combatting age-associated neuroinflammation, leading to improved cognitive function.
One finding from this study was that cognitive performance was associated with innate anti-inflammation capability; this effect was not seen with motor performance in our earlier study (19). It is interesting to speculate that a simple blood test could be used as a screening tool to identify those at risk for cognitive impairment. In addition, as pro-inflammatory measures were correlated with reduced motor (19) and cognitive performance post-diet, this blood test might be used to predict who might benefit most from nutritional interventions.
In conclusion, this study provides additional evidence that consumption of BB may forestall (in animals with intact cognitive ability) or reverse (in animals with impaired cognition) age-related deficits in cognition. In addition, circulating anti-inflammatory factors may predict cognitive function, and daily consumption of BB may have beneficial effects on these factors, depending on the baseline cognition. Therefore, future studies examining the age-related effects of foods or their constituent compounds should give greater consideration to individual differences at baseline to better identify those who will benefit most from dietary intervention.
Funding
This research was supported by USDA Intramural funds. The blueberry powder used in this study was provided at no cost under a Material Transfer Agreement between the USDA and the U.S. Highbush Blueberry Council (USHBC). Dr. Miller received partial support from the National Institute on Aging (T32-AG000029.
Conflict of Interest Statement
None reported.
References
