Photoinducible and rhythmic ICER–CREM immunoreactivity in the rat suprachiasmatic nucleus

Neuroscience Letters 385 (2005) 87–91

Photoinducible and rhythmic ICER–CREM immunoreactivity in the rat suprachiasmatic nucleus William J. Schwartz a,∗ , Neil Aronin b , Paolo Sassone-Corsi c b

a Department of Neurology, University of Massachusetts Medical School, Worcester, MA, USA Departments of Medicine and Cell Biology, University of Massachusetts Medical School, Worcester, MA, USA c IGBMC, Centre National de la Recherche Scientifique, Illkirch-Strasbourg, France

Received 9 March 2005; received in revised form 26 April 2005; accepted 7 May 2005

Abstract Several genes expressed in the suprachiasmatic nucleus (SCN) are induced by light and are candidate links in the photic entrainment pathway of the SCN’s circadian clock. Since the cAMP response element binding protein (CREB) and CRE-mediated gene transcription in the SCN appears to be crucial for light-induced phase shifts of circadian rhythmicity, we analyzed the immunohistochemical expression of proteins encoded by the cAMP response element modulator (CREM) gene, including a repressor isoform (inducible cAMP early repressor [ICER]). ICER–CREM immunoreactivity was detected in cells of the ventrolateral subdivision of the rat SCN after light administration during the subjective night in constant darkness; but only late after light onset (at 240 min), following earlier successive peaks of phosphorylated CREB protein (by 5 min), c-fos mRNA (by 40 min), per1 mRNA (by 55 min), and c-Fos protein (by 60 min). In constant darkness, there was a modest but significant endogenous rhythm of ICER–CREM immunoreactivity, with a two-fold difference between high levels at circadian time (CT) 10 and low levels at CT 22. Our data raise the possibility that ICER–CREM might be involved in downregulating the SCN expression of immediate-early and “clock” genes after their induction by phase-shifting light pulses. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Circadian; CRE; CREB; c-fos; per

The circadian pacemaker in the suprachiasmatic nucleus (SCN) of the hypothalamus can function as a clock because its endogenous period is adjusted to the external 24 h light–dark (LD) cycle, primarily by light-induced phase shifts that reset the pacemaker’s oscillation (for review, see [13]). The responsible input pathway is believed to include a specialized group of SCN-projecting retinal ganglion cells that synaptically release glutamate onto SCN neurons. How the resulting membrane depolarization and Ca2+ influx lead to a shift in the clock remains uncertain, but much attention has been focused on the expression of photoinducible genes in SCN cells, sp., immediate early genes, like c-fos, and oscillating “clock” genes, like per1. A universal mediator of cAMP- and Ca2+ -dependent gene expression is the cAMP response element binding protein ∗

Corresponding author. Tel.: +1 508 856 4147; fax: +1 508 856 6778. E-mail address: [email protected] (W.J. Schwartz).

0304-3940/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2005.05.018

(CREB), which binds to a cis-regulatory element (the cAMP response element [CRE]) on the promoters of target genes; CREB acts as a transcriptional activator upon its phosphorylation by a variety of kinases at a common site (Ser-133) (for reviews, see [4,11]). The promoters of the c-fos and per1 genes both contain functional CREs that bind CREB from SCN tissue extracts [9,29]. SCN CREB is phosphorylated within minutes of photic or glutamatergic stimulation during the subjective night (but not during the subjective day) in vivo or in vitro [7,9,12,30]. Nighttime light pulses lead to CRE-mediated gene transcription in the SCN of a CREß-galactosidase reporter mouse strain [19], and CRE activation appears to be necessary for light- and glutamate-induced phase advances of the locomotor rhythm in mice in vivo and the SCN firing rate rhythm in rat slices in vitro, respectively [27]. Recently, Gau et al. [8] have shown a phase-dependent phosphorylation of SCN CREB at an additional amino acid site (Ser-142) by light in vivo and glutamate in vitro.

88

W.J. Schwartz et al. / Neuroscience Letters 385 (2005) 87–91

Another set of CRE-binding proteins is encoded by the cAMP response element modulator (CREM) gene (for reviews, see [4,21]). Cell-specific alternative splicing generates isoforms that act as transcriptional activators (CREM␶) or repressors (CREM␣, -␤, -␥) upon phosphorylation, as well as an additional repressor isoform that can be induced via an intronic promoter within the CREM gene (inducible cAMP early repressor [ICER]) [17]. Of note, ICER–CREM mRNA levels are increased in the rat SCN after a nighttime light pulse [24], while they appear constitutively low over the 12 h:12 h LD cycle [23] (although an mRNA rhythm has been reported in the hamster SCN, with relatively high levels during the light phase of the LD cycle [14,15]). An ICER-like immunoreactivity has also been detected in rat and mouse SCN [10]. In contrast to CREB, a role for ICER–CREM in the SCN remains undefined. Since ICER lacks an amino terminal phosphorylation domain, its repressor activity is believed to be determined by the amount of induced protein [17]. To analyze photoinducible and spontaneous expression of ICER–CREM proteins within the SCN, we have used an antiserum directed against a bacterially expressed CREM␶ protein (recognizing all CREM isoforms and ICER) [6] to perform immunohistochemistry on tissue sections of rat SCN. Here we report our observations on the spatial and temporal pattern of the immunoreactive protein and its relationship to phosphorylated CREB (pCREB) and c-Fos immunoreactivities as well as to c-fos and per1 mRNA levels (by in situ hybridization). Adult male Sprague–Dawley rats (Harlan-Sprague–Dawley, Indianapolis, IN) were housed in clear polycarbonate cages contained within light-proof environmental compartments. Light within each compartment was provided by 15 W cool-white fluorescent tubes automatically controlled by a 24-h timer; intensity varied within the cages but was on the order of 300–400 lx at the mid-cage level. A single 15 W safe light with a dark red (Kodak series 2) filter was used to allow for routine care and anesthetic injection; rats were exposed to approximately 30 lx maximally but usually <1 lx. Food and water were freely available and replenished once every week at irregular hours. Lighting cycles within different compartments could be set so that every animal (irrespective of circadian phase or lighting condition) could be killed within a 2- to 3-h interval of real time, allowing us to treat brains at different time points together in the same immunohistochemical (n = 4–6 rats per time point) or in situ (n = 3 rats per time point) run. For immunohistochemistry, rats were deeply anesthetized with pentobarbital (50 mg, i.p.) and perfused rapidly (@ 60 ml/min by peristaltic pump) through the ascending aorta with 300 ml of freshly prepared cold 4% paraformaldehyde in 0.15 M phosphate-buffered saline (PBS). Brains were removed and immersed in fixative for 2–4 h at 4 ◦ C, 60-␮m thick coronal sections were cut on a vibratome, and individual sections were incubated with either rabbit anti-CREM␶ (1:5000), anti-pCREB123–136 (1:1000; graciously supplied by Dr. David Ginty, Johns Hopkins University, Baltimore, MD),

or anti-c-Fos3–17 (1:2000; SC52; Santa Cruz Biotech., Inc.) antisera in 0.1% bovine serum albumin and 0.2% Triton X100 in PBS overnight at 4 ◦ C. Sections were processed for immunohistochemistry using the avidin–biotin method (Vector Laboratories, Burlingame, CA, USA) with diaminobenzidine as the chromogen, as previously described [22]. On three to four sections/animal through the middle of the SCN, labeled cell nuclei in the unilateral ventrolateral subdivision (irrespective of the intensity of staining) were counted by one of us (N.A.) without knowledge of lighting conditions or time of day. For in situ hybridization, linearized recombinant plasmids were used as templates for the generation of antisense cRNA probes (rat c-fos from a 2.3 kb cDNA insert in pSP65 and rat per1 from a 981-bp cDNA insert in pGEM® -T Easy Vector) and transcribed in the presence of [35 S] UTP with T7 RNA polymerase using the MaxiScript in vitro transcription kit (Ambion, Austin, TX, USA). Rats were decapitated (guillotine), brains were rapidly removed and frozen in 2methylbutane cooled to −30 ◦ C with dry ice, 15-␮m-thick coronal sections through the SCN were cut on a cryostat and mounted onto slides coated with Vectabond (Vector Laboratories), and slides were processed and exposed to Kodak Biomax MR film (Eastman Kodak Company, Rochester, NY, USA), as previously described [5,22]. Optical density (OD) of the autoradiographic hybridization signal was measured using a Zeiss (Kontron) Image Processing System. The average OD for each rat SCN was derived from at least three sections through the nucleus, each expressed as a relative OD (ratio OD SCN/OD surrounding hypothalamus). Rats were entrained to a 12 h:12 h LD cycle for 2 weeks, then left in darkness for one circadian cycle after the last lights-off, and exposed to light during the second half of the subjective night (lights-on at circadian time (CT) 19, or 31 h after the last lights-off). Groups of rats were killed before the light pulse and 5, 30, 60, 120, or 240 min (for immunohistochemistry) or 10, 25, 40, 55, 90, or 120 min (for in situ hybridization) after lights-on. We observed CREM immunoreactivity in the ventrolateral SCN; but only late after light onset, after the earlier successive peaks of pCREB (at 5 min) and c-Fos (at 60 min) had subsided (Fig. 1). Immunoreactive cell counts are plotted in Fig. 2, along with the hybridization signals for c-fos (appearing at 10 min and peaking by 40 min) and per1 (appearing at 25 min and peaking by 55 min) mRNAs. For CREM, the mean values for the six time points were significantly different (p < 0.001, one-way ANOVA on ranks), and pairwise comparisons using Dunn’s method showed that CREM levels at 240 min were significantly higher than the levels at 0, 5, and 30 min (p < 0.05). A comparably delayed and quantitatively similar rise in immunoreactive CREM at 240 min was also observed when we performed this experiment during the first half of the subjective night (lights-on at CT 14; data not shown). In an additional set of animals, we showed that 60 min of light (from CT 19 to 20) was as effective as 240 min (from CT 19 to 23) in elevating late CREM expression (when cells

W.J. Schwartz et al. / Neuroscience Letters 385 (2005) 87–91

89

Fig. 1. Photic stimulation of immunoreactive pCREB, c-Fos, and CREM expression in the ventrolateral SCN. Representative coronal brain sections of the unilateral SCN from rats killed in darkness and 5, 60, and 240 min after light onset at CT 19 and processed for immunohistochemistry. Immunoreactive pCREB appears early (5 min), followed by c-Fos (60 min), and later by CREM (240 min) (arrows). Scale bar represents 0.1 mm.

were counted at CT 23: 53 ± 10 after 60 min of light, 61 ± 4 after 240 min of light, 10 ± 3 in the absence of light; mean number of labeled cells/section in the unilateral ventrolateral SCN ± S.E.M., n = 4 rats per group). To test for an endogenous (circadian) rhythm of CREM immunoreactivity in the SCN, we studied a final set of rats that was first entrained to the 12 h: 12 h LD cycle, then left in darkness after the last lights-off, and sacrificed in darkness at 4 h intervals spanning the circadian cycle (beginning at CT 2 [14 h after the last lights-off], and at CT 6, 10, 14, 18, and 22). Immunoreactive cells were observed throughout the SCN, with only a modest rhythm (∼2-fold peak-to-trough amplitude; Fig. 3). However, the mean values for the six time points

Fig. 2. Time course of photo-inducible gene expression in the ventrolateral SCN. Graphed is (A) expression of immunoreactive pCREB, c-Fos, and CREM as number of labeled cells/section in the ventrolateral subdivision of the unilateral SCN and (B) expression of c-fos and per1 mRNA levels as SCN autoradiographic relative optical density. Each point represents mean ± S.E.M. from four to six animals (for immunohistochemistry) or three animals (for in situ hybridization).

Fig. 3. Endogenous circadian rhythm of immunoreactive CREM expression in the ventrolateral SCN. Graphed is expression of immunoreactive CREM as number of labeled cells/section in the ventrolateral subdivision of the unilateral SCN. Each point represents mean ± S.E.M. from four animals.

90

W.J. Schwartz et al. / Neuroscience Letters 385 (2005) 87–91

were signficantly different (p < 0.02, one-way ANOVA), and pairwise comparisons using Bonferroni t-statistics showed that the CT 10 value was significantly higher than the value at CT 22 (p < 0.05). Our results show that immunoreactive CREM in the SCN is induced as part of a transcriptional program [3] activated by a nighttime light pulse. We surmise that this immunoreactivity represents photoinducible ICER, since it is the only cyclic-AMP-inducible isoform of the CREM family. Our data complement the previous demonstration by Stehle et al. [24] of a light-induced upregulation of ICER–CREM mRNA by in situ hybridization of rat SCN sections. They reported increased mRNA levels 60 min after light presented during the late night and 180 min after light presented during the early night. In the rat, photic stimuli during the late night appear to induce higher counts of c-Fos immunoreactive cells than photic stimuli during the early night [1,20,26,28]. Of note, our data do not distinguish whether ICER-activated cells are a subset of, or different from, the cells that express pCREB and c-Fos, nor whether individual ICER cells are both photoresponsive and endogenously rhythmic. Multiple stimuli lead to CRE-mediated gene transcription in a variety of brain regions; like phase-resetting light pulses in the SCN, they are known to be associated with the early phosphorylation of CREB and late induction of ICER (for review, see [16]). How CREB and ICER might work to determine the initiation and termination of gene expression has been studied extensively in the pineal gland, where they play an important role in regulating rhythmic melatonin synthesis (for review, see [25]). In the SCN, it is conceivable that the decay of light-induced, CRE-dependent mRNAs is not only due to autorepression and mRNA instability but also to transcriptional inhibition by ICER. While at least c-fos mRNA levels appear to fall well before immunoreactive ICER levels peak, there are other photoinducible mRNAs (e.g., fra-2 [22], per2 [31], and egr-3 [18]) that begin to decline only after a delay of at least 2 h after light onset, at a time that ICER levels are beginning to rise. Whether or not ICER actually contributes to the downregulation of these and other genes is unknown, but it is important to note that its high level a few hours after a light pulse does not appear to diminish the SCN’s photo-responsiveness to a second light pulse at this time [2].

Acknowledgements We thank Robin Peters and Jennifer Meyer for their expert technical assistance, and Drs. Tom Curran, Hitoshi Okamura, and Shigenobu Shibata for their previous gifts of recombinant plasmids. This publication reports research supported by the National Institute of Neurological Diseases and Stroke (NINDS R01 NS46605 to W.J.S.), and its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NINDS.

References [1] C. Beaul´e, S. Amir, Photic entrainment and induction of immediateearly genes within the rat circadian system, Brain Res. 821 (1999) 95–100. [2] J.D. Best, E.S. Maywood, K.L. Smith, M.H. Hastings, Rapid resetting of the mammalian circadian clock, J. Neurosci. 19 (1999) 828–835. [3] C. Crosio, N. Cermakian, C.D. Allis, P. Sassone-Corsi, Light induces chromatin modification in cells of the mammalian circadian clock, Nat. Neurosci. 3 (2000) 1241–1247. [4] D. De Cesare, G.M. Fimia, P. Sassone-Corsi, Signaling routes to CREM and CREB: plasticity in transcriptional activation, Trends Biochem. Sci. 24 (1999) 281–285. [5] H.O. de la Iglesia, J. Meyer, W.J. Schwartz, Using Per gene expression to search for photoperiodic oscillators in the hamster suprachiasmatic nucleus, Brain Res. Mol. Brain Res. 127 (2004) 121– 127. [6] V. Delmas, F. van der Hoorn, B. Mellstr¨om, B. Jegou, P. SassoneCorsi, Induction of CREM activator proteins in spermatids: downstream targets and implications for haploid germ cell differentiation, Mol. Endocrinol. 7 (1993) 1502–1513. [7] J.M. Ding, L.E. Fairman, W.J. Hurst, L.R. Kuriashkina, M.U. Gillette, Resetting the biological clock: mediation of nocturnal CREB phosphorylation via light, glutamate, and nitric oxide, J. Neurosci. 17 (1997) 667–675. [8] D. Gau, T. Lemberger, C. von Gall, O. Kretz, N. Le Minh, P. Gass, W. Schmid, U. Schibler, H.W. Korf, G. Sch¨utz, Phosphorylation of CREB Ser142 regulates light-induced phase shifts of the circadian clock, Neuron 34 (2002) 245–253. [9] D.D. Ginty, J.M. Kornhauser, M.A. Thompson, H. Bading, K.E. Mayo, J.S. Takahashi, M.E. Greenberg, Regulation of CREB phosphorylation in the suprachiasmatic nucleus by light and a circadian clock, Science 260 (1993) 238–241. [10] C.A. Kell, F. Dehghani, H. Wicht, C.A. Molina, H.-W. Korf, J.H. Stehle, Distribution of transcription factor inducible cyclicAMP early repressor (ICER) in rodent brain and pituitary, J. Comp. Neurol. 478 (2004) 379–394. [11] B.E. Lonze, D.D. Ginty, Function and regulation of CREB family transcription factors in the nervous system, Neuron 35 (2002) 605–623. [12] S. McNulty, I.L. Schurov, P.J. Sloper, M.H. Hastings, Stimuli which entrain the circadian clock of the neonatal Syrian hamster in vivo regulate the phosphorylation of the transcription factor CREB in the suprachiasmatic nucleus in vitro, Eur. J. Neurosci. 10 (1998) 1063–1072. [13] J.H. Meijer, W.J. Schwartz, In search of the pathways for lightinduced pacemaker resetting in the suprachiasmatic nucleus, J. Biol. Rhythms 18 (2003) 235–249. [14] S. Messager, D.G. Hazlerigg, J.G. Mercer, P.J. Morgan, Photoperiod differentially regulates the expression of Per1 and ICER in the pars tuberalis and the suprachiasmatic nucleus of the Siberian hamster, Eur. J. Neurosci. 12 (2000) 2865–2870. [15] S. Messager, A.W. Ross, P. Barrett, P.J. Morgan, Decoding photoperiodic time through Per1 and ICER gene amplitude, Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 9938–9943. [16] B. Mioduszewska, J. Jaworski, L. Kaczmarek, Inducible cAMP early repressor (ICER) in the nervous system—a transcriptional regulator of neuronal plasticity and programmed cell death, J. Neurochem. 87 (2003) 1313–1320. [17] C.A. Molina, N.S. Foulkes, E. Lalli, P. Sassone-Corsi, Inducibility and negative autoregulation of CREM: an alternative promoter directs the expression of ICER, an early response repressor, Cell 75 (1993) 1–20. [18] M.E. Morris, N. Viswanathan, S. Kuhlman, F.C. Davis, C.J. Weitz, A screen for genes induced in the suprachiasmatic nucleus by light, Science 279 (1998) 1544–1547.

W.J. Schwartz et al. / Neuroscience Letters 385 (2005) 87–91 [19] K. Obrietan, S. Impey, D. Smith, J. Athos, D.R. Storm, Circadian regulation of cAMP response element-mediated gene expression in the suprachiasmatic nuclei, J. Biol. Chem. 274 (1999) 17748–17756. [20] H.J. Romijn, A.A. Sluiter, C.W. Pool, J. Wortel, R.M. Buijs, Differences in co-localization between Fos and PHI, GRP, VIP and VP in neurons of the rat suprachiasmatic nucleus after a light stimulus during the phase delay versus the phase advance period of the night, J. Comp. Neurol. 372 (1996) 1–8. [21] P. Sassone-Corsi, Coupling gene expression to cAMP signalling: role of CREB and CREM, Intl. J. Biochem. Cell Biol. 30 (1998) 27–38. [22] W.J. Schwartz, A. Carpino Jr., H.O. de la Iglesia, R. Baler, D.C. Klein, Y. Nakabeppu, N. Aronin, Differential regulation of fos family genes in the ventrolateral and dorsomedial subdivisions of the rat suprachiasmatic nucleus, Neuroscience 98 (2000) 535–547. [23] J.H. Stehle, N.S. Foulkes, P. P´evet, P. Sassone-Corsi, Developmental maturation of pineal gland function: synchronized CREM inducibility and adrenergic stimulation, Mol. Endocrinol. 9 (1995) 706– 716. [24] J.H. Stehle, M. Pfeffer, R. K¨uhn, H-W. Korf, Light-induced expression of transcription factor ICER (inducible cAMP early repressor) in rat suprachiasmatic nucleus is phase-restricted, Neurosci. Lett. 217 (1996) 169–172. [25] J.H. Stehle, C. von Gall, H-W. Korf, Analysis of cell signalling in the rodent pineal gland deciphers regulators of dynamic transcription in neural/endocrine cells, Eur. J. Neurosci. 14 (2001) 1–9.

91

[26] E.L. Sutin, T.S. Kilduff, Circadian and light-induced expression of immediate early gene mRNAs in the rat suprachiasmatic nucleus., Brain Res. Mol. Brain Res. 15 (1992) 281–290. [27] S.A. Tischkau, J.W. Mitchell, S.-H. Tyan, G.F. Buchanan, M.U. Gillette, Ca2+ /cAMP response element-binding protein (CREB)dependent activation of Per1 is required for light-induced signaling in the suprachiasmatic nucleus circadian clock, J. Biol. Chem. 278 (2003) 718–723. [28] Z. Tr´avn´ıckov´a, A. Sumov´a, R. Peters, W.J. Schwartz, H. Illnerov´a, Photoperiod-dependent correlation between light-induced SCN c-fos expression and resetting of circadian phase, Am. J. Physiol. 271 (1996) R825–R831. [29] Z. Tr´avn´ıckov´a-Bendova, N. Cermakian, S.M. Reppert, P. SassoneCorsi, Bimodal regulation of mPeriod promoters by CREBdependent signaling and CLOCK/BMAL1 activity, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 7728–7733. [30] C. von Gall, G.E. Duffield, M.H. Hastings, M.D.A. Kopp, F. Dehghani, H.-W. Korf, J.H. Stehle, CREB in the mouse SCN: a molecular interface coding the phase-adjusting stimuli light, glutamate, PACAP, and melatonin for clockwork access, J. Neurosci. 18 (1998) 10389–10397. [31] L. Yan, S. Takekida, Y. Shigeyoshi, H. Okamura, Per1 and Per2 gene expression in the rat suprachiasmatic nucleus: circadian profile and the compartment-specific response to light, Neuroscience 94 (1999) 141–150.