Distribution of N-methyld -aspartate (NMDA) receptor mRNAs in the rat suprachiasmatic nucleus

BRAIN RESEARCH ELSEVIER

Brain Research 632 (1993) 329-333

Short Communication

Distribution of N-methyl D-aspartate (NMDA) receptor mRNAs in the rat suprachiasmatic nucleus Jens D. Mikkelsen a,., Philip J. Larsen a, Francis J.P. Ebling b a Institute of Medical Anatomy, University of Copenhagen, Copenhagen, Denmark, b Department of Anatomy, University of Cambridge, Cambridge, UK (Accepted 28 September 1993)

Abstract

Photic entrainment of the circadian oscillator located in the hypothalamic suprachiasmatic nucleus (SCN) is considered to be mediated at least partly by release of glutamate from the retinal presynaptic nerve terminals acting via a NMDA receptor. Several NMDA receptor subtypes have been cloned and expressed in model systems. The NMDA-R1 subtype is essential for the function of the NMDA receptor, and the multiple NMDA-R2(-A, -B, or -C) subunits potentiate and differentiate the function of the NMDA receptor by forming different heteromeric configurations with NMDA-R1. The aim of this study was to use in situ hybridization histochemistry with oligonucleotide sequences (42-48-mer) labeled with 35S to detect whether NMDA receptor mRNA is present in the rat SCN, and if so, to characterize which receptor subtypes occur. In order to identify the precise location of NMDA receptor mRNAs within the SCN, sections were dipped in emulsion and cellular resolution was achieved. The hybridization revealed a high abundance of NMDA-R1 mRNA in the SCN as well as in many other forebrain areas. The NMDA-R1 expressing cells were distributed throughout the SCN. NMDA-R2A and NMDA-R2B mRNAs were found in the hippocampus, but not in the SCN. In contrast, NMDA-R2C mRNA was found in relative high amounts in the rat SCN, but not in other hypothalamic areas. In dipped sections, it was evident that the localization of NMDA-2RC was mostly confined to the dorsomedial part of the SCN. Thus, the rat SCN contains a specific combination of NMDA receptor mRNA subtypes not found in other forebrain structures. These observations are consistent with the hypothesis that glutamate mediates the effect of light on entrainment of the circadian oscillator. Key words: Glutamate receptor; Circadian rhythms; Suprachiasmatic nucleus; In situ hybridisation; Rat

The hypothalamic suprachiasmatic nucleus (SCN) generates circadian rhythms in mammals [8,13]. Entrainment of this oscillator by the environmental light-dark cycle is mediated by a monosynaptic connection from the retina; the retinohypothalamic tract [11]. Several lines of evidence indicate a role for glutamate in transduction of photic information to the SCN mediated by this pathway via a N M D A receptor. First, immunocytochemical studies have indicated that the retinohypothalamic tract contains N-acetylaspartylglutamate- and glutamate-immunoreactivity [3,15]. Both act as agonists for the N M D A receptor subunit NMDA-R1 when expressed in Xenopus oocytes [21].

* Corresponding author. Present address: Institute of Medical Anatomy B, The Panum Institute, Blegdamsvej 3, 2200 Copenhagen N, Denmark. Fax: (45) (3) 536-9612.

Second, in vitro electrophysiological studies in rodents suggest that excitatory amino acids may be involved in the neurotransmission of the R H T [22]. For example in the mouse, optic nerve stimulation of SCN neurons can be blocked by glutamate antagonists [2]. Third, pharmacological manipulations of glutamatergic neurotransmission can disrupt the effects of light on behavioural rhythms [4,24]. Finally, several studies have been carried out using the induction of the immediate early gene c-fos as a cellular marker of photic activation of SCN cells [6,9,20]. The glutamate agonist N M D A can mimic the effects of light in that it induces c-fos expression in the SCN when infused into the cerebral ventricles [6], and several studies indicate that pretreatment of hamsters with a variety of glutamate antagonists will attenuate photic induction of c-fos [1,6]. Although these various lines of evidence support the view that glutamate serves as a neurotransmitter in

0006-8993/93/$06.00 © 1993 Elsevier Science Publishers B.V. All rights reserved SSDI 0006-8993(93)E1318-W

330

JD. Mikkelsen et al. / Bratn Research 632 (1993) 329-333

the retinohypothalamic tract, the alternative possibility that glutamatergic inputs from other sources serve only as a permissive influence for photic activation of the SCN cannot be ruled out. The overall aim of this study was therefore to investigate whether glutamate receptors arc present within the ventrolateral region of the SCN, the region which receives direct retinal and lateral geniculate inputs [11,14]. Such a finding would support the view that' glutamate is directly related to photic input to the SCN. Previous autoradiographic ligand binding studies have only revealed a low density of glutamate receptors in the hypothalamus with poorly defined localization [16]. Recently, cDNAs coding for various NMDA-receptor subtypes have been cloned in rat and mouse [10,12,17,18,25]. The N M D A subtypes of the rat and mouse reflect at least 4 different mRNAs [10,12,17], though alternative splicing may result in an even greater number of translation products [5]. In situ hybridization histochemistry has shown that the mRNA subtypes are differently distributed in the brain [10,17]. NMDA-RI m R N A was found to be widely distributed in the rat brain, NMDA-R2A and NMDA-R2B mRNAs were found mainly in the cortical areas, whereas NMDA-R2C mRNA was found exclusively in the cere-

bellum [17]. These initial studies did not specifically investigate the presence or distribution of NMDA mRNA in the SCN. The receptor channels coded by these genes are heteromeric proteins consisting of the NMDA-R1 product combined with one class of the NMDA-R2 products• These various NMDA receptor subunit complexes are functionally different ion channels when expressed in Xenopus oocytes [10], thus identifying specific mRNAs present in the SCN will provide information about the mechanism of action of glutamate in the SCN. The purpose of the present investigation was to determine the presence and location of N M D A receptor subtypes in the rat SCN by in situ hybridization histochemistry (ISHH). Twenty-four adult male Wistar rats were maintained in a 12L: 12D photoperiod regimen. They were decapitated between 3 and 5 h after light onset and the brains immediately frozen on dry-ice (-80°C). The frozen brains processed for ISHH were cut on a cryostat in 12 /zm thick serial coronal sections through the SCN and hybridized with probes• The sequences of the synthetic DNA oligonucleotide probes (Advanced Biotechnology Center, Charing Cross and Westminster Medical School, London, UK) were similar to those used by Monyer et al. [17]. The probes were labeled with [a-

:-:i ~

:•::,!~:,i

/ .... i l ~ •

Fig. 1. Top: localization of NMDA-R1 mRNA (left) and NMDA-R2C m R N A (right) m representative coronal sections through a rat brain• Bottom: detail of suprachiasmatic nucleus (SCN) region enlarged from complete sections. 12 /zm sections were hybridized with 35S-labelled oligonucleotide probes. Note that NMDA-R1 hybridization signal is present in many areas including the cortex, striatum, SCN and supraoptic nucleus, whereas NMDA-R2C mRNA is restricted to the dorsomedial region of the SCN and the thalamus.

331

J.D. Mikkelsen et al. / Brain Research 632 (1993) 329-333

35S]-thio-ATP ( > 3000 C i / m m o l , N E N ) using terminal deoxynucleotidyl transferase (Boehringer Mannheim, Copenhagen) to a specific activity of 2.0 x 10 ~9 d p m / mol a t 37°C in a buffer of (pH 7.2) containing 50% (v/v). formamide, 4 × SSC (SSC = 0.15 M NaC1, 0.015 M Na-citrate, p H 7.2), 1 x Denhardt's solution, 10% ( w / v ) dextran sulphate, 10 mM dithiotreitol, 0.5 m g / m l salmon sperm D N A and 250 t z g / m l yeast tRNA. After hybridization, slides were washed in 1 × SSC at 55°C for 4 x 15 min and at room temperature for 2 x 30 min. The sections were next dipped into water, blowdried and finally exposed to A m e r s h a m X-ray Hyper-

film for 3 weeks. A serie of slides was dipped in A m e r s h a m LM-1 emulsion and exposed for either 6 weeks ( N M D A - R 1 ) or 9 weeks ( N M D A - R 2 A , -B, and -C) before being developed and counterstained with Cresyl violet. Adjacent sections hybridized with the 4 different oligonucleotide sequences and exposed either on an X-ray film or in fluid emulsion clearly revealed N M D A - R 1 and N M D A - R 2 C m R N A s in the rat SCN (Fig. 1). The N M D A - R 2 B hybridization was in the film measured to be minimal and no hybridization signal with the N M D A - R 2 A probe was detected (not shown).

3-

,". ~ 4 7

© Fig. 2. Microphotographs of coronal sections hybridized and exposed to emulsion fluid film showing the localization of NMDA-RI mRNA (A) and NMDA-R2C (B,C) in the rat SCN. As illustrated in Fig. 2A the NMDA-R1 expressing cells are present throughout the SCN. In the rostral SCN, the NMDA-R2C mRNA expressing cells are accumulated in the dorsal part of the SCN (B). At more caudal levels, the NMDA-R2C transcripts are mostly located in the dorsomedial (dm), but a few labelled cells are found in the ventrolateral (vl) part as well (C). opt, optic chiasm; V, third ventricle. Scale bars: 200/.*m (Fig. 2A); 100/~m (Fig. 2B and C).

332

ZD. Mikkelsen el al. / Brain Research 632 (1993) 329-333

Compared to other regions of the brain in the same coronal section, the SCN contained a moderate to high number of NMDA-R1 transcripts, whereas the SCN stands out as the only region expressing a high level of NMDA-R2C mRNA in the sections, though a low level of hybridization signal was detected in the thalamus (Fig. 1; top right). Compared to the SCN, the abundance of NMDA-RI transcripts was considerably higher in the hippocampus and in the cerebral cortex. Similarly, NMDA-R2A and NMDA-R2B mRNAs were abundant in the hippocampus, but were absent in the SCN as well as in other hypothalamic nuclei. In sections dipped in emulsion, the distribution of the NMDA-R mRNAs could be defined in more detail. The NMDA-R1 rnRNA expressing cells were found in large numbers throughout the SCN. In the rostral SCN, the most densely expressing zone was in the dorsal part. In the middle and caudal SCN the highest expression was observed in the medial SCN, probably reflecting the greater density of cells in this region of the SCN (Fig. 2A). NMDA-R1 mRNA containing ceils were found in the underlying optic chiasm and in the periventricular zone, as well as throughout the forebrain. NMDA-R2A m R N A was not detected in the SCN, whereas a very few NMDA-R2B expressing cells were found. Sections hybridized with an oligonucleotide complimentary to NMDA-R2C mRNA produced a high labeling in the SCN (Fig. 2B, C). In the rostral SCN, the labeled cells occupied the dorsal part of the nucleus (Fig. 2B). More caudally, the transcripts were found in cells in the dorsomedial part of the SCN, and at this level a lower density of cells was found in the ventrolateral portion as well (Fig. 2C). The results presented here show that NMDA-receptor mRNA are present in the rat SCN, and supports the hypothesis that glutamate acts on SCN neurons or astrocytes via a NMDA receptor. The presence of NMDA-R1 m R N A in the SCN is consistent with the functional role of glutamate in photic regulation of SCN activity and metabolism. The protein encoded by the rat NMDA-R1 cDNA possesses four putative transmembrane segments and forms a r e c e p t o r / i o n channel complex in oocytes, that has the electrophysiological and pharmacological properties characteristic of the NMDA receptor. The major effect of these channel is Ca 2+ transport, which has previously been demonstrated to increase intracellularly in response to glutamate in SCN neurons and astrocytes in vitro [23]. However, highly active N M D A receptor channels are formed only when NMDA-R1 and NMDA-R2 subunits are expressed together [10,12]. The predicted diversity of NMDA receptor complexes has functional implications. It appears that the combination of NMDAR 1 / R 2 C subunits, which in the current study predicts to be the major combinations of mRNA in the SCN is more sensitive to glycine than the other combinations

[10]. Pharmacologically the N M D A - R I / R 2 C was more sensitive to 7-chlorokynurenate (7-CK) and less sensitive to MK-801, but most remarkably however, is its resistence to Mg 2+ blockade [10,17]. The functional properties of the N M D A receptor channel are critically determined by the NMDA-R2C subunit, and since the mRNA encoding for this subunit is not widely distributed in the brain, the glutamatergic neurotransmission in the SCN would appear to be specialized. Several groups have noted that various glutamate antagonists including MK-801, D G G and AP5 block photic induction of c-fos in the most ventral region of the SCN in the Syrian hamster, but not in the more dorsal region [1,6,24]. The actions of glutamate antagonists in the ventral SCN correlates well with their ability to block the behavioural effects of light pulses, so the significance of photic induction of c-fos in more doral regions is unclear. The current observations in the rat provide a potential explanation for the differential effect of glutamate antagonists in ventral and dorsal regions, namely that a N M D A - R 1 / 2 C complex which is relatively insensitive to these antagonists [10,17] mediates photic induction of c-los in the more dorsal SCN, whereas in the ventral region either NMDA-R1 alone or NMDA-R1 complexing with an as yet unidentified subunit mediates the effect of glutamate on c-fos induction. In fact, the majority of NMDA-R2C cells appeared to lie in the dorsomedial region of the SCN rather than the retinorecipient ventrolateral zone. This implies that the NMDA-R2C receptor subunit is not involved in photic activation of c-los and entrainment of the circadian oscillator, but is perhaps involved in neurotransmission between the ventrolateral and dorsomedial SCN. Moreover, some light-induced functions of the SCN may be independent of c-los activation. For example, Ohi et al. [19] demonstrated that microinjection of NMDA into the SCN resulted in a very rapid decrease in serotonin N-acetyltransferase activity (NAT) in the pineal gland, an effect presumably mediated via outputs to the parvocellular paraventricular nucleus and the autonomic nervous system. The decline in NAT activity occured within a few minutes of N M D A treatment, before c-los protein would have been synthesized. Thus, the SCN can be considered to be a heterogeneous population of light-responsive cells, one population involved in long-term change (i.e. phase shifts), and others involved in rapid changes in light-induced pineal metabolism. Although the hybridization of NMDA-R1 in the majority of SCN cells and the relatively large size of many hybridized cells provides indirect evidence that the receptor subunit is localized in neurons the presence of NMDA-R1 in glia cannot be discounted. Indeed, glutamate has been shown to increase intracellular Ca 2+ concentrations in SCN astrocytes in vitro [23],

J.D. Mikkelsen et al./ Brain Research 632 (1993) 329-333

and other classes of glutamate receptor mRNAs (GluR, mGluR) have been localized in glia in rat optic nerve [7]. In summary, the current study has identified NMDA-R1 mRNA throughout the rat SCN, consistent with the view that glutamate mediates photic entrainment of this circadian oscillator, and also the communication of information within the nucleus. NMDAR2C mRNA is also present, but is most abundant in the dorsomedial division of the SCN. It is not expressed elsewhere in the hypothalamus, thus this subunit may have a specific function in the circadian timing system. However, it is not clear that NMDA-R2C mRNA has a direct role in transmission of photic information because of a lack of overlap of the cells expressing this subunit and the region of retinal innervation. We wish to acknowledge Tine G6rlen for technical assistance. This study was supported by Landsforeningen til bek~empelse af 0jensygdomme og blindhed, Lzege Eilif Trier-Hansen og hustrus Legat, Eli og Egon Larsens legat, The Danish MRC (12-0236; 12-1642), the UK AFRC (AG8/606), and The European Science Foundation. J.D.M. is recipient of a Hallas-Moller Fellowship financed by the NOVO Foundation, and F.J.P.E. is funded by a Royal Society University Research Fellowship. [1] Abe, H., Rusak, B. and Robertson, H.A., NMDA and nonNMDA receptor antagonists inhibit photic induction of fos protein in the hamster suprachiasmatic nucleus, Brain Res. Bull., 28 (1992) 831-835. [2] Cahill, G.M. and Menaker, M., Effects of excitatory amino acid receptor antagonists and agonists on suprachiasmatic nucleus responses to retinohypothalamic tract volleys, Brain Res., 479 (1989) 76-82. [3] Castel, M., Belenky, M., Cohen, S., Ottersen, O.P. and StormMathisen, J., Glutamate-like immunoreactivity in retinal terminals of the mouse suprachiasmatic nucleus, Eur. J. Neurosci., 5 (1993) 368-381. [4] Colwell, C.S. and Menaker, M., NMDA as well as non-NMDA receptor antagonists can prevent the phase-shifting effects of light on the circadian system of the golden hamster, J. Biol. Rhythms, 7 (1992) 125-136. [5] Durand, G.M., Gregor, P., Zheng, X., Bennett, M.V.L., Uhl, G.R. and Zukin, R.S., Cloning of an apparent splice variant of the rat N-methyl-D-aspartate receptor NMDA-RI with altered sensitivity to polyamines and activators of protein kinase C, Proc. NatL Acad. Sci. USA, 89 (1992) 9359-9363. [6] Ebling, F.J.P., Maywood, E.S., Staley, K., Humby, T., Hancock, D.C., Waters, C.M., Evan, G.I. and Hastings, M.H., The role of N-methyl-D-aspartate-type glutamatergic neurotransmission in the photic induction of immediate-early gene expression in the suprachiasmatic nuclei of the Syrian hamster, J. Neuroendocrinol., 3 (1991) 641-652. [7] Jensen, A.M. and Chiu, S.Y., Expression of glutamate receptor genes in white matter: developing and adult rat optic nerve, J. Neurosci., 13 (1993) 1664-1675. [8] Klein, D.C., Moore, R.Y. and Reppert, S.M. (Eds.), The Suprachiasmatic Nucleus. The Mind's Clock, Oxford University Press, 1991.

333

[9] Kornhauser, J.M., Nelson, D.E., Mayo, K.E. and Takahashi, J.S., Photic and circadian regulation of c-los gene expression in the hamster suprachiasmatic nucleus, Neuron, 5 (1990) 127-134. [10] Kutsuwada, T., Kashiwabuchi, N., Mori, H., Sakimura, K., Kushiya, E., Araki, K., Meguro, H., Masaki, H., Kumanishi, T., Arakawa, M. and Mishina, M., Molecular diversity of the NMDA receptor channel, Nature, 358 (1992) 36-41. [11] Levine, J.D., Weiss, M.L., Rosenwasser, A.M. and Miselis, R.R., Retinohypothalamic tract in the female albino rat: a study using horseradish peroxidase conjugated to cholera toxin, J. Comp. Neurol., 306 (1991) 344-360. [12] Meguro, H., Mori, H., Araki, K., Kushiya, E., Kutsuwada, T., Yamazaki, M., Kumanishi, T., Arakawa, M., Sakimura, K. and Mishina, M. Functional characterization of a heteromeric NMDA receptor channel expressed from cloned cDNAs, Nature, 357 (1992) 70-74. [13] Meijer, J.H. and Rietveld, W.J., Neurophysiology of the suprachiasmatic circadian pacemaker in rodents, Physiol. Rev., 69 (1989) 671-707. [14] Mikkelsen, J.D., Projections from the lateral geniculate nucleus to the hypothalamus of the Mongolian gerbil (Meriones unguiculatus). An anterograde and retrograde tracing study, J. Comp. NeuroL, 299 (1990) 493-508. [15] Moffett, J.R., Williamson, L., Palkovits, M. and Namboodiri, M.A.A., N-Acetylaspartylglutamate: a transmitter candidate for the retinohypothalamic tract, Proc. Natl. Acad. Sci. USA, 87 (1990) 8065-8069. [16] Monaghan, D.T., Cotman, C.W., Distribution of N-methyl-Daspartate sensitive L-3H-glutamate binding sites in rat brain, J. Neurosci., 5 (1985) 2909-2919. [17] Monyer, H., Sprengel, R., Schoepfer, R., Herb, A., Higuchi, M., Lomeli, H., Burnashev, N., Sakmann, B. and Seeburg, P.H., Heteromeric NMDA receptors: molecular and functional distinction of subtypes, Science, 256 (1992) 1217-1221. [18] Moriyoshi, K., Masu, M., Ishii, T., Shigemoto, R., Mizuno, N. and Nakanishi, S., Molecular cloning and characterization of the rat NMDA receptor, Nature, 354 (1991) 31-37. [19] Ohi, K., Takashima, M., Nishikawa, T. and Takahashi, K., N-Methyl-D-aspartate receptor participates in neuronal transmission of photic information through the retinohypothalamic tract, Neuroendocrinology, 53 (1991) 344-348. [20] Rea, M.A., Light increases fos-related protein immunoreactivity in the rat suprachiasmatic nuclei, Brain Res. Bull., 23 (1989) 577-581. [21] Sekiguchi, M., Wada, K. and Wenthold, R.J., N-acetylaspartylglutamate acts as an agonist upon homomeric NMDA receptor (NMDAR1) expressed in Xenopus oocytes, FEBS Lett., 311 (1992) 285-289. [22] Shibata, S., Liou, S.Y. and Ueki, S., Influence of excitatory amino acid receptor antagonsists and of baclofen on synaptic transmission in the optic nerve to the suprachiasmatic nucleus in slices of rat hypothalamus, Neuropharmacology, 25 (1986) 403-409. [23] van den Pol, A.N., Finkbeiner, S.M. and Cornell-Bell, A.H., Calcium excitability and oscillations in suprachiasmatic nucleus neurons and glia in vitro, J. Neurosci., 12 (1992) 2648-2664. [24] Vindlacheruvu, R.R., Ebling, F.J.P., Maywood, E.S. and Hastings, M.H., Blockade of glutamatergic neurotransmission in the suprachiasmatic nucleus prevents cellular and behavioural responses of the circadian system to light, Eur. J. Neurosci., 4 (1992) 673-679. [25] Yamazaki, M., Mori, H., Araki, K., Mori, K.J. and Mishina, M., Cloning, expression an modulation of a mouse NMDA receptor subunit, FEBS Lett., 300 (1992) 39-45.