The hippocampus is receiving increasing attention for its potential role in energy regulation [¹]. The hippocampus is part of a neural circuit involved with reward and energy regulation [²] and is sensitive to satiety signals associated with learning and memory [³]. Recent findings indicate that palatable high-fat diets promote excessive food intake and weight gain and interfere with hippocampal functioning. This is supported by epidemiological data linking diets high in saturated fat with weight gain and memory deficits [⁴–⁶]. Furthermore, rats and humans with diabetes mellitus show age-related performance impairments on memory tasks [⁷]. More recent studies demonstrate that high-fat diet-induced maternal obesity impairs offspring hippocampal BDNF production [⁸], alters fetal hippocampal development [⁹], and reduces hippocampal neurogenesis during the early life of their offspring [¹⁰]. Moreover, adult male rats fed with a high-fat diet show impaired hippocampal neurogenesis [¹¹]. Neurogenesis induced by ciliary neurotrophic factor (CNTF) or brain-derived neurotrophic factor (BDNF) in feeding centers of the murine hypothalamus is associated with weight loss in obese rodents [¹², ¹³]. Because both hippocampal neurogenesis and proxy metabolic parameters are connected with stress and mood disorders [¹⁴–¹⁶], the aforementioned hypothalamic data raise important questions regarding the relationship between hippocampal neurogenesis and the regulation of peripheral metabolic parameters. We present a pilot study of the relationship between hippocampal neurogenesis and metabolic parameters in adult nonhuman primates.

The age range for all subjects was 9 to 12 years old with VFD-reared subjects (mean age 11 yrs, SD = 1.41) and LFD-reared subjects (mean age 11 yrs SD = 1.41) showing no difference (df = 6, P = 1.0). The weight range was 5.5 to 14 kg with VFD-reared (mean weight 11.5 kg, SD = 2.79) and LFD-reared subjects (mean weight 9.37 kg, SD = 2.90) showing no difference (df = 6; P = 0.33). Table 1 shows means and standard deviations of individual parameters for all subjects, and Figure 1 displays representative images of DCX, BCL-2, and Ki67 positive cells.

There was a significantly inverse correlation of body weight with the number of DCX-stained neurons with tertiary dendrites (r = −0.76, P = 0.03, n = 8) and BCL2-ratings of density (r = −0.73, P = 0.041, n = 8) (Figure 2). Using a general linear model, it was demonstrated that the relationship between DCX with tertiary dendrites and weight remained all but significant [F(1 : 5) = 6.34; P = 0.053] for both LFD and VFD reared animals when adjusted for age [F(1 : 5) = 1.18; P = 0.32]. Furthermore, BCL-2 expression correlated with body weight [F(1 : 5) = 21.05; P = 0.006] for both LFD and VFD reared animals, when adjusted for age [F(1 : 5) = 5.46; P = 0.067]. These findings indicate that neuroplastic effects in the dentate gyrus were dependent upon a relationship to body mass, which was independent of the effects of age. There were also significant inverse correlations between the crown-rump length and the number of DCX-stained neurons with tertiary dendrites (r = −0.98; P = 0.02, n = 4) and BCL2-ratings of density (r = −0.99; P = 0.006, n = 4) (Table 2). In addition, the fasting serum glucose levels were positively correlated with the DCX-stained neurons with tertiary dendrites (r = 0.97; P = 0.021, n = 4) and the BCL2-ratings of density (r = 0.99; P = 0.01, n = 4) (Table 2). The same was true for the blood lipid ratio, but not individual lipid parameters, which also positively correlated with DCX (r = 0.96; P = 0.042, n = 4), but only at trend levels for BCL-2 (Table 2). There was no significant correlation between the cell proliferation marker, Ki67, and metabolic parameters.

To our knowledge, this is the first report to show that markers of dentate gyrus neuroplasticity correlated with metabolic parameters. However, precursor proliferation, as indicated by Ki67, did not correlate with the metabolic parameters. The results suggest that neurogenesis, but not precursor proliferation, is associated with metabolic processes. Indicating that the findings of the current study were not an artifact of a possible relationship between body weight and central parameters, our previous monkey MRI study revealed no relationship between grey matter volume and body weight [²¹].

Studies in rodents demonstrate that high-fat diet causing weight gain impairs hippocampal neurogenesis through increased lipid peroxidation associated with decreased BDNF in the hippocampus [²², ²³] and/or through glucocorticoid-mediated effects on immature or mature new neurons [²⁴, ²⁵]. In mice, genetic models deficient of BDNF signaling exhibit hyperphagia and obesity [²⁶] in addition to suppressing hippopcampal neurogenesis [²⁷]. In humans, obese children and adolescents exhibit reduced serum BDNF levels relative to their normal weight counterparts [²⁸, ²⁹]. Both central and peripheral administration of BDNF decrease food intake, increase energy expenditure, and ameliorate hyperglycemia in diabetic mice by a central nervous-system-mediated mechanism [²⁵, ²⁸, ³⁰]. The antidepressant effects of BDNF infusion are closely linked to the induction of hippocampal neurogenesis [²²]. Indeed, BDNF plays an important role in regulation of neurogenesis, angiogenesis, and synaptogenesis in the hippocampus [²², ³¹]. Therefore, BDNF may be one of the common pathways mediating both neurogenesis in the hippocampus and systemic metabolic process, such as insulin resistance. Together these preclinical and clinical findings imply an inverse correlation between hippocampal neurogenesis and body weight.

Our data also showed that the hippocampal neurogenesis correlated positively with blood glucose level and lipid ratio, indicating that increased neurogenesis is associated with elevated metabolic activity [³¹]. Under physiological conditions, adult neurogenesis occurs within an angiogenic niche—a microenviroment, providing metabolic support for neurogenesis—within the hippocampus. The microenvironment is modulated by the circulating system that supplies mesenchymal stem cells and nutrients (e.g., glucose) [³¹]. Specific mechanisms linking hippocampal neurogenesis interacting with the hypothalamus and regulating appetite and intake behavior remain unclear.

One potential candidate for the current link between hippocampal neurogenesis and metabolic markers is leptin. Leptin is a hormone secreted by adipose tissue and has effects at hypothalamic receptors that control food intake. Deficits in leptin or leptin receptors result in obesity, indicating the importance of the hormone in body mass homeostasis [³²]. Leptin levels positively correlate with fasting insulin, fasting glucose, and triglycerides in nondiabetic men [³³]. Studies show leptin increases adult hippocampal neurogenesis mainly by increasing cell proliferation and not by promoting cell differentiation and survival and reduces body weight (inversely correlated with cell differentiation) of animals [³⁴]. By contrast, our study demonstrated that weight was inversely correlated with differentiation, as reflected by the generation of new neurons and survival, indicated by BCL-2 expression, but there was no correlation between weight and cell proliferation. The discrepancy between our findings and others may be due to multiple factors, such as differences in species, studies, and interventions.

To our knowledge, this is the first study to show correlation between neurogenesis and metabolic parameters in nonhuman primates. There are several limitations of this study, including the small sample size and the inability to reveal a causal link between central and metabolic process. Further studies should investigate the relationship between neuroplasticity and peripheral metabolic processes.

The authors wish to thank Ms. Laura Anderson, Ms. Katharine Celentano, and Dr. Andrew Dwork.