Food restriction reduces neurogenesis in the avian hippocampal formation

Abstract

The mammalian hippocampus is particularly vulnerable to chronic stress. Adult neurogenesis in the dentate gyrus is suppressed by chronic stress and by administration of glucocorticoid hormones. Post-natal and adult neurogenesis are present in the avian hippocampal formation as well, but much less is known about its sensitivity to chronic stressors. In this study, we investigate this question in a commercial bird model: the broiler breeder chicken. Commercial broiler breeders are food restricted during development to manipulate their growth curve and to avoid negative health outcomes, including obesity and poor reproductive performance. Beyond knowing that these chickens are healthier than fully-fed birds and that they have a high motivation to eat, little is known about how food restriction impacts the animals' physiology. Chickens were kept on a commercial food-restricted diet during the first 12 weeks of life, or released from this restriction by feeding them ad libitum from weeks 7-12 of life. To test the hypothesis that chronic food restriction decreases the production of new neurons (neurogenesis) in the hippocampal formation, the cell proliferation marker bromodeoxyuridine was injected one week prior to tissue collection. Corticosterone levels in blood plasma were elevated during food restriction, even though molecular markers of hypothalamic-pituitary-adrenal axis activation did not differ between the treatments. The density of new hippocampal neurons was significantly reduced in the food-restricted condition, as compared to chickens fed ad libitum, similar to findings in rats at a similar developmental stage. Food restriction did not affect hippocampal volume or the total number of neurons. These findings indicate that in birds, like in mammals, reduction in hippocampal neurogenesis is associated with chronically elevated corticosterone levels, and therefore potentially with chronic stress in general. This finding is consistent with the hypothesis that the response to stressors in the avian hippocampal formation is homologous to that of the mammalian hippocampus.

Fig 1. Photomicrographs of the HF at different levels.

A-C Photomicrographs of a rostral (A), a medium (but still part of the rostral pole; B), and a caudal (C) coronal section through the hippocampal formation (HF). D. BrdU⁺/Hu⁺ (white arrow), BrdU⁺/Hu^- (black arrow), and BrdU^-/Hu⁺ (open arrows) cells from the rostral section of the HF. E. BrdU⁺ cell in the ventricular wall of the caudal section of the HF. The scale bar in C applies to A and B as well. The scale bar in D also applies to E.

Fig 2. Body size.

FR birds had lower body mass (A; 1467±81g) and shorter tarsometatarsal length (B; 75.4±0.8mm) than AL birds (3458±81g and 85.3±0.8mm resp.).*** p < 0.001.

Fig 3. Endocrinological and molecular stress markers.

FR birds had higher baseline CORT levels in plasma than AL birds (A). There were no differences in the expression of pro-opiomelanocortin (POMC) in the pituitary gland (B), nor of steroidogenic acute regulatory protein (STAR) (C) and ADP ribosylation factor like GTPase 10 (ARL10) (D) in the adrenal glands. However, Interleukin-6 (IL6) expression in the spleen was significantly higher in FR than in AL birds (E). *** p<0.001; * p<0.05.

Fig 4. Brain size measures.

There were no differences between FR and AL birds in telencephalon volume (A), hippocampal volume (B), total number of hippocampal neurons (C) or density of hippocampal neurons (D). In panel B and C, values from birds for which the caudal-most section of the HF was present are dark, and those for which that section was missing are plotted in a lighter shade.

Fig 5. Neurogenesis.

(A) There were significantly more BrdU⁺/Hu⁺ new neurons in the rostral HF of AL birds than FR birds. This trend was not significant in the caudal pole of the HF. (B) There were significantly more BrdU⁺/Hu^- cells in the AL than in the FR birds, in both the rostral and caudal poles of the HF. (C) There is no effect of treatment on the number of BrdU⁺ cells in the ventricular zone, but there are more ventricular BrdU⁺ cells in the rostral than in the caudal HF, and this effect is the same for both treatments. Different symbols and shades of grey represent brains that were stained in different staining batches. Staining batch was used as a random factor in the Linear Mixed Model. * p<0.05.