Rhythmic Diurnal Synthesis and Signaling of Retinoic Acid in the Rat Pineal Gland and Its Action to Rapidly Downregulate ERK Phosphorylation

Abstract

Vitamin A is important for the circadian timing system; deficiency disrupts daily rhythms in activity and clock gene expression, and reduces the nocturnal peak in melatonin in the pineal gland. However, it is currently unknown how these effects are mediated. Vitamin A primarily acts via the active metabolite, retinoic acid (RA), a transcriptional regulator with emerging non-genomic activities. We investigated whether RA is subject to diurnal variation in synthesis and signaling in the rat pineal gland. Its involvement in two key molecular rhythms in this gland was also examined: kinase activation and induction of Aanat, which encodes the rhythm-generating melatonin synthetic enzyme. We found diurnal changes in expression of several genes required for RA signaling, including a RA receptor and synthetic enzymes. The RA-responsive gene Cyp26a1 was found to change between day and night, suggesting diurnal changes in RA activity. This corresponded to changes in RA synthesis, suggesting rhythmic production of RA. Long-term RA treatment in vitro upregulated Aanat transcription, while short-term treatment had no effect. RA was also found to rapidly downregulate extracellular signal-regulated kinase (ERK) 1/2 phosphorylation, suggesting a rapid non-genomic action which may be involved in driving the molecular rhythm in ERK1/2 activation in this gland. These results demonstrate that there are diurnal changes in RA synthesis and activity in the rat pineal gland which are partially under circadian control. These may be key to the effects of vitamin A on circadian rhythms, therefore providing insight into the molecular link between this nutrient and the circadian system.

Keywords: Circadian; ERK; Pineal gland; RARgamma; Retinoic acid; Retinol; rdh10; rdh12.

Fig. 1

The components necessary for retinoic acid signaling are present in the rat pineal gland. mRNA expression of the genes encoding the synthetic enzymes retinol dehydrogenase, Rdh10, and the three retinaldehyde dehydrogenases, Raldh1, Raldh2, and Raldh3; the retinal reductase enzyme Rdh12; the retinol transporter Stra6; the retinoic acid receptors, Rara, Rarb and Rarg; and the catabolic enzymes cytochrome P450 family 26 (Cyp26) a1 and Cyp26b1; determined by PCR and gel electrophoresis (a). Protein for RALDH1, RALDH2, RARα and RARβ is present, determined by western blotting (b)

Fig. 2

Immunohistochemistry for detection of RARα in the rat pineal gland. Representative images of double-labeling of paraffin sections of the rat pineal gland with antibodies against RARα and S-antigen (SAG; a, b) or GFAP (c), at × 20 (a) or × 40 (b, c) magnification. RARα-immunoreactivity was detected in the cytoplasm of cells throughout the pineal parenchyma, indicated by arrows. No co-localization was detected of RARα with SAG, whereas a small subset of RARα-positive cells co-expressed GFAP, indicated by arrowhead

Fig. 3

Immunohistochemistry for detection of RALDH1 in the rat pineal gland. Representative images of double-labeling of paraffin sections of the rat pineal gland with antibodies against RALDH1 and S-antigen (SAG; a, b) or GFAP (c), at × 20 (a) or × 40 (b, c) magnification. RALDH1-immunoreactivity was detected in the cytoplasm of cells throughout the pineal parenchyma, indicated by arrows. No co-localization was detected of RALDH1 with SAG or GFAP

Fig. 4

Retinoic acid signaling genes exhibit diurnal changes in expression in the rat pineal gland. qPCR analysis of expression of Aanat, Bmal1 and retinoic acid signaling genes in Sprague Dawley rat pineal glands collected at different zeitgeber times (ZT) throughout a 24 h cycle; the horizontal bar represents the light/dark cycle, white bar = light phase, black bar = dark phase; grey shading indicates the dark period. Values represent mean mRNA expression relative to Gapdh, ± SEM; with the exception of Raldh2 which represent median mRNA expression relative to Gapdh, with interquartile range, as these data are not normally distributed. N ≥ 6 glands per time-point. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 5

Production of retinoic acid in the rat pineal gland exhibits a diurnal change. Retinoic acid concentration [RA] measured in spent media from pineal glands collected at zeitgeber time (ZT) 0 and 12 and cultured for 2 h, measured by a retinoic acid reporter cell line. The horizontal bar represents the light/dark cycle, white bar = light phase, black bar = dark phase; grey shading indicates the dark period. Values represent median with interquartile range. N = 4 glands per time-point. *P < 0.05

Fig. 6

Diurnal changes in Cyp26a1 expression suggest there are diurnal changes in retinoic acid activity in the rat pineal gland. a Retinoic acid (RA) rapidly induces upregulation of RA-responsive genes Cyp26a1 and Cyp26b1 in the rat pineal gland. qPCR analysis of cultured rat pineal glands following 4-h treatment with vehicle control or RA. Values represent the fold change in mean mRNA expression compared to control, ± SEM. N = 3 glands per treatment. *P < 0.05; **P < 0.01. b Cyp26a1 exhibits diurnal changes in expression in the rat pineal gland. qPCR analysis of Sprague Dawley rat pineal glands collected at different zeitgeber times (ZT) during the light/dark cycle; white bar = light phase, black bar = dark phase; grey shading indicates the dark period. Values represent median mRNA expression relative to Gapdh, with interquartile range. N ≥ 6 glands per time-point. *P < 0.05 (Kruskal-Wallis test)

Fig. 7

Diurnal changes in retinoic acid signaling genes, Rarg and Rdh12, persist in constant darkness in the rat pineal gland. qPCR analysis of expression of retinoic acid signaling genes in pineal glands of Sprague Dawley rats maintained in constant darkness for 2 days. Pineal glands were collected at different circadian times (CT) throughout a 24 h cycle; CT0 and CT12 correspond to subjective light on and light off, respectively; horizontal black bar and grey shading indicate the dark period. Values represent mean mRNA expression relative to Gapdh, ± SEM. N = 5 glands per time-point. **P < 0.01; ***P < 0.001

Fig. 8

Norepinephrine represses Rarg and Rdh12 gene expression. qPCR analysis of cultured rat pineal glands following 4-h treatment with vehicle control or norepinephrine (NE). Values represent the fold change in mean mRNA expression compared to control, ± SEM. N ≥ 3 glands per treatment. *P < 0.05; **P < 0.01

Fig. 9

Retinoic acid does not rapidly influence Aanat transcription. qPCR analysis of cultured rat pineal glands following 4-h treatment with vehicle control, norepinephrine (NE), NE + retinoic acid (RA), or RA. Values represent the fold change in mean mRNA expression compared to control, ± SEM. N = 3–4 glands per treatment. *P < 0.05; ***P < 0.001, compared to control treatment

Fig. 10

Long-term retinoic acid treatment induces upregulation of Aanat transcription. qPCR analysis of cultured P10–12 rat pineal glands following 4-h (a) or 48-h treatment (b) with vehicle control or retinoic acid (RA). Values represent the fold change in mean mRNA expression compared to control, ± SEM. N = 3 (a) or 6 (b) glands per treatment. **P < 0.01

Fig. 11

Retinoic acid rapidly downregulates ERK phosphorylation. Activation of ERK1/2 (a) and Akt (b) was determined by western blotting following 10 min treatment of cultured rat pineal glands with vehicle control or retinoic acid (RA). EGF was tested as a positive control (n = 1). Phosphorylated levels were normalized to total ERK or Akt; values represent the fold change in mean compared to control, ± SEM. N ≥ 3 glands per treatment. ***P < 0.001, compared to control

Fig. 12

Summary schematic of the proposed retinoic acid rhythm in the rat pineal gland. At the start of the light period, zeitgeber time (ZT) 0, measured retinoic acid (RA) concentration is at its highest. At this time, the expression of the rate-limiting enzyme for the conversion of retinol to retinoic acid, Rdh10, is lowest. While there is a peak in the expression of Rdh12, which encodes an enzyme that converts retinaldehyde to retinol, therefore reducing the amount of retinaldehyde available for conversion to RA. These changes in expression of the RA synthetic enzymes are likely to lead to the sevenfold reduction in RA synthesis during the day, resulting in the low RA concentration measured at ZT12. At this time, Rdh10 expression rises to its peak, which is expected to increase RA synthesis again during the night, returning RA concentration to the high levels detected at the end of night, at ZT0. This may be accompanied by an increase in RA signaling during the night, with a peak in expression of Rarg at ZT18, which encodes one of the RA receptors, and increase in expression of the RA-responsive gene, Cyp26a1