Adrenergic regulation of mitogen-activated protein kinase in rat pinealocytes: opposing effects of protein kinase A and protein kinase G

The adrenergic-regulated CRTC1 and CRTC2 phosphorylation and cellular distribution is independent of endogenous SIK1 in the male rat pinealocyte

Salt inducible kinase 1 (SIK1) has been reported to repress cAMP-response element binding protein (CREB)-mediated gene transcription by causing the nuclear export of CREB-regulated transcription coactivators (CRTCs) through phosphorylation. Although the repressor role of SIK1 in suppressing the expression of arylalkylamine N-acetyltransferase, the enzyme that controls the daily rhythm in melatonin production in the rat pineal gland, has been established, whether SIK1 regulates the phosphorylation and localization of CRTC1 and CRTC2 in this tissue remains unclear. The present study found that overexpressing SIK1 in NE-stimulated rat pinealocytes could increase the phosphorylation of CRTC1 and CRTC2, reduced selectively the nuclear level of CRTC2 (but not that of CRTC1), and elevated the cytosolic levels of both CRTC1 and CRTC2. In contrast, transient knockdown of endogenous SIK1 had no effect on the phosphorylation or distribution of CRTC1 and CRTC2 in norepinephrine (NE)-stimulated pinealocytes. Our results also showed that adrenergic blockade during NE stimulation led to a rapid rephosphorylation and decline in the nucleus levels of CRTC1 and CRTC2; however SIK1 knockdown had no effect on this rapid rephosphorylation. Moreover, studies with kinase inhibitors revealed that kinase(s) sensitive to KT5823 appeared to be involved in this rapid rephosphorylation. Together, these results indicate that although overexpressing SIK1 can phosphorylate CRTC1 and CRTC2 in the NE-stimulated pinealocyte, the endogenous SIK1, in spite of its induction by NE, does not appear to be the main regulator of the phosphorylation and intracellular localization of these two coactivators.

Sustained adrenergic stimulation is required for the nuclear retention of TORC1 in male rat pinealocytes

The process involved in relocation of the coactivator, transducer of regulated cAMP-regulated element-binding protein (TORC) to the cytoplasm, unlike its activation, is not well understood. Using cultured pineal cells prepared from male rats, we found that although both α- and β-adrenergic stimulation could cause TORC1 dephosphorylation, only α-adrenergic stimulation was effective in the norepinephrine (NE)-mediated translocation of TORC1 into the nucleus. In contrast, blockade of either the α- or the β-adrenergic receptor after NE stimulation was effective in causing the rephosphorylation and rapid relocation of TORC1 into the cytoplasm. Studies with phosphoprotein phosphatase (PP) inhibitors indicated that although both PP2A and PP2B could dephosphorylate TORC1, only PP2B could cause translocation into the nucleus. However, after NE stimulation, treatment with either PP2A or PP2B inhibitors could cause the rephosphorylation and cytoplasmic relocation of TORC1. These results indicate a requirement of continuous activation of both α- and β-adrenergic receptors as well as PP2A and PP2B activities for the nuclear retention of TORC1 during NE stimulation. Knockdown of salt-inducible kinase 1 (SIK1) had no effect on the phosphorylation or localization of TORC1. Although overexpressing SIK1 could induce TORC1 phosphorylation in the nucleus, it did not reduce TORC1 level in the nucleus, indicating that SIK1-mediated TORC1 phosphorylation may not be sufficient for its relocation into the cytoplasm. Together, these results demonstrate that, in the rat pineal gland, different mechanisms are involved in regulating the nuclear entry and exit of TORC1 and that the SIK1-mediated phosphorylation of TORC1 may not lead to its nuclear exit.

Different signaling mechanisms are involved in the norepinephrine-stimulated TORC1 and TORC2 nuclear translocation in rat pinealocytes

The distribution of transducers of regulated cAMP-response element-binding protein activity (TORC) between the cytoplasm and the nucleus is tightly regulated and represents one of the main mechanisms whereby the cAMP response element activation activities of TORC are controlled. Whereas both cAMP and Ca(2+) pathways can cause translocation of TORC, the relative importance of these two pathways in regulating different TORC within the same cell is unclear. In this study, we determined the mechanism that regulated TORC1 translocation and compared it with that of TORC2 in rat pinealocytes. Stimulation of pinealocytes with norepinephrine (NE), although having no effect on Torc1 transcription, caused rapid dephosphorylation of TORC1. Although NE also caused rapid dephosphorylation of TORC2, pharmacological studies revealed that TORC1 dephosphorylation could be induced by both β-adrenoceptor/cAMP and α-adrenoceptor/intracellular Ca(2+) pathways contrasting with TORC2 dephosphorylation being induced mainly through the β-adrenoceptor/cAMP pathway. PhosTag gel indicated a different pattern of TORC1 desphosphorylation resulting from the selective activation of α- or β-adrenoceptors. Interestingly, only the α-adrenoceptor/intracellular Ca(2+)-mediated dephosphorylation could translocate TORC1 to the nucleus, whereas the β-adrenoceptor/cAMP-mediated dephosphorylation of TORC1 was ineffective. In comparison, translocation of TORC2 was induced predominantly by the β-adrenoceptor/cAMP pathway. Studies with different protein phosphatase (PP) inhibitors indicated that the NE-mediated translocation of TORC1 was blocked by cyclosporine A, a PP2B inhibitor, but that of TORC2 was blocked by okadaic acid, a PP2A inhibitor. Together these results highlight different intracellular signaling pathways that are involved in the NE-stimulated dephosphorylation and translocation of TORC1 and TORC2 in rat pinealocytes.

Adrenergic regulation of the distribution of transducer of regulated cAMP-response element-binding protein (TORC2) in rat pinealocytes

Transducers of regulated cAMP-response element-binding protein (CREB) activity (TORC) are coactivators that can increase CREB transcriptional activity, suggesting that TORC may regulate the transcription of Aanat, a CREB-target gene. In the present study, we focused on the regulation of TORC2 and its role in Aanat transcription in the rat pineal gland. Although there was no endogenous Torc2 mRNA rhythm in the rat pineal gland and treatment of cultured pinealocytes with norepinephrine (NE) had no effect on the mRNA level of Torc2, the phosphorylation state and intracellular distribution of TORC2 protein were regulated by NE. Immunoblot analysis combined with cytosolic/nuclear fractionation or phosphatase treatment showed that TORC2 protein was rapidly dephosphorylated and translocated to the nucleus after NE stimulation in rat pinealocytes. Similar dephosphorylation of TORC2 also occurred nocturnally in the rat pineal gland. The NE-mediated TORC2 dephosphorylation was blocked by cotreatment with propranolol (a β-adrenergic antagonist) but not prazosin (an α(1)-adrenergic antagonist) and mimicked by dibutyryl cAMP, indicating the participation of the β-adrenergic receptor/cAMP pathway. Studies with protein phosphatase inhibitors showed that only okadaic acid and calyculin A were effective in blocking the NE-mediated TORC2 dephosphorylation, suggesting the involvement of protein phosphatase 2A in this dephosphorylation. Moreover, TORC2 overexpression had an enhancing effect on NE-stimulated Aanat transcription. Together, these results indicate that NE stimulation causes nuclear translocation of TORC2 by dephosphorylating the protein through a β-adrenoceptor/cAMP mechanism and that nuclear localization of TORC2 appears to regulate Aanat transcription by NE in the rat pineal gland.

Histone modifications on the adrenergic induction of type II deiodinase in rat pinealocytes

Histone modifications have been shown to play an important role in regulating gene expression. In this study, we investigated the impact of histone modifications on the adrenergic-regulated transcription of type 2 deiodinase (Dio2), a CREB-target gene in the rat pinealocyte. Treatment of pinealocytes with inhibitors of aurora C, a histone kinase, resulted in an inhibitory effect on the adrenergic-stimulated histone H3 Ser10 phosphorylation and Dio2 transcription. Given the established link between histone phosphorylation and acetylation, the role of histone acetylation on the adrenergic-induced Dio2 transcription was investigated. Treatment of pinealocytes with histone deacetylase inhibitors inhibited the adrenergic-induced Dio2 transcription. Chromatin immunoprecipitation with antibodies against acetylated Lys14 of H3 showed an increase in DNA recovery of the promoter region of Dio2 following treatment with trichostatin A. Together, our results indicate that, beside activation of CREB, epigenetic factors such as histone modifications also play an important role in regulating Dio2 transcription.

Salt-inducible kinase 1 in the rat pinealocyte: adrenergic regulation and role in arylalkylamine N-acetyltransferase gene transcription

The recognition of the basic leucine zipper domain in the regulation of transcriptional activity of cAMP response element-binding protein by salt-inducible kinase (SIK) prompted our investigation of the regulatory role of this kinase in the induction of Aa-nat and other cAMP-regulated genes in the rat pineal gland. Here we report Sik1 expression was induced by norepinephrine (NE) in rat pinealocytes primarily through activation of beta-adrenergic receptors, with a minor contribution from activation of alpha-adrenergic receptors. Treatments with dibutyryl cAMP, and to a lesser extent, agents that elevate intracellular Ca(2+) mimicked the effect of NE on Sik1 expression. In parallel to the results of the pineal cell culture studies, a marked nocturnal induction of Sik1 transcription was found in whole-animal studies. Knockdown of Sik1 by short hairpin RNA amplified the NE-stimulated Aa-nat transcription and other adrenergic-regulated genes, including Mapk phosphatase 1, inducible cAMP repressor, and type 2 iodothyronine deiodinase in a time-dependent manner. In contrast, overexpressing Sik1 had an inhibitory effect on the NE induction of Aa-nat and other adrenergic-regulated genes. Together, our results indicate that the adrenergic induction of Sik1 in the rat pineal gland is primarily through the beta-adrenergic receptor --> protein kinase A pathway. SIK1 appears to function as part of an endogenous repressive mechanism that regulates the peak and indirectly the duration of expression of Aa-nat and other cAMP-regulated genes. These findings support a role for SIK1 in framing the temporal expression profile of Aa-nat and other adrenergic-regulated genes in the rat pineal gland.

Protein kinase G is involved in ammonia-induced swelling of astrocytes

Ammonia-induced swelling of astrocytes is a primary cause of brain edema associated with acute hepatic encephalopathy. Previous studies have shown that ammonia transiently increases cGMP in brain in vivo and in cultured astrocytes in vitro. We hypothesized that protein kinase G (PKG), an enzyme activated by cGMP and implicated in regulation of cell shape, size, and/or volume in peripheral and CNS cells, may play a role in the ammonia-induced astrocytic volume increase. Treatment of cultured rat cortical astrocytes with 1 or 5 mM NH4Cl (ammonia) for 24 h increased their cell volume by 50% and 80% above control, respectively, as measured by confocal imaging followed by 3D computational analysis. A cGMP analog, 8-(4-chlorophenylthio)-cGMP, increased the cell volume in control cells and potentiated the increase in 1 mM ammonia-treated cells. A soluble guanylate cyclase inhibitor (1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one) abrogated, and a PKG inhibitor [8-(4-chlorophenylthio)-cGMP-thioate, Rp-isomer] dose-dependently reduced the cell volume-increasing effect of 5 mM ammonia. The results suggest that (i) PKG may play a permissive role in ammonia-induced astrocytic swelling and (ii) elevation of brain cGMP associated with acute exposure to ammonia in vivo may aggravate the ensuing brain edema.

Nocturnal activation of aurora C in rat pineal gland: its role in the norepinephrine-induced phosphorylation of histone H3 and gene expression

We have shown previously that Ser10 phosphorylation of histone H3 occurs in rat pinealocytes after stimulation with norepinephrine (NE) and that histone modifications such as acetylation appear to play an important role in pineal gene transcription. Here we report the nocturnal phosphorylation of a Ser10 histone H3 kinase, Aurora C, in the rat pineal gland. The time profile of this phosphorylation parallels the increase in the level of phospho-Ser10 histone H3. Studies with cultured pinealocytes indicate that Aurora C phosphorylation is induced by NE and this induction can be blocked by cotreatment with propranolol or KT5720, a protein kinase A inhibitor. Moreover, only treatment with dibutyryl cAMP, but not other kinase activators, mimics the effect of NE on Aurora C phosphorylation. These results indicate that Aurora C is phosphorylated primarily by a beta-adrenergic/protein kinase A-mediated mechanism. Treatment with an Aurora C inhibitor reduces the NE-induced histone H3 phosphorylation and suppresses the NE-stimulated induction of arylalkylamine N-acetyltransferase (AA-NAT), the rhythm-controlling enzyme of melatonin synthesis, and melatonin production. The effects of Aurora C inhibitors on adrenergic-induced genes in rat pinealocytes are gene specific: inhibitory for Aa-nat and inducible cAMP repressor but stimulatory for c-fos. Together our results support a role for the NE-stimulated phosphorylation of Aurora C and the subsequent remodeling of chromatin in NE-stimulated Aa-nat transcription. This phenomenon suggests that activation of this mitotic kinase can be induced by extracellular signals to participate in the transcriptional induction of a subset of genes in the rat pineal gland.

Acetylation of histone H3 and adrenergic-regulated gene transcription in rat pinealocytes

In this study we investigated the effect of histone acetylation on the transcription of adrenergic-induced genes in rat pinealocytes. We found that treatment of pinealocytes with trichostatin A (TSA), a histone deacetylase inhibitor, caused hyperacetylation of histone H3 (H3) Lys14 at nanomolar concentrations. Hyperacetylation of H3 was also observed after treatment with scriptaid, a structurally unrelated histone deacetylase inhibitor. The effects of TSA and scriptaid were inhibitory on the adrenergic induction of arylalkylamine-n-acetyltransferase (aa-nat) mRNA, protein, and enzyme activity, and on melatonin production. TSA at higher concentrations also inhibited the adrenergic induction of mapk phosphatase-1 (mkp-1) and inducible cAMP early repressor mRNAs. In contrast, the effect of TSA on the norepinephrine induction of the c-fos mRNA was stimulatory. Moreover, the effect of TSA on adrenergic-induced gene transcription was dependent on the time of its addition; its effect was only observed during the active phase of transcription. Chromatin immunoprecipitation with antibodies against acetylated Lys14 of H3 showed an increase in DNA recovery of the promoter regions of aa-nat, mkp-1, and c-fos after treatment with TSA. Together, our results demonstrate that histone acetylation differentially influences the transcription of adrenergic-induced genes, an enhancing effect for c-fos but inhibitory for aa-nat, mkp-1, and inducible cAMP early repressor. Moreover, both inhibitory and enhancing effects appear to be mediated through specific modification of promoter-bound histones during gene transcription.

The role of repressor proteins in the adrenergic induction of type II iodothyronine deiodinase in rat pinealocytes

In this study, we investigated the transcriptional regulation of the adrenergic induction of type II iodothyronine deiodinase (Dio2) in rat pinealocytes. Treatment of pinealocytes with norepinephrine (NE) caused an increase in the mRNA level of Dio2 that peaked around 2 h and declined over the next 5 h. Both beta- and alpha1-adrenergic receptors contributed to the NE induction of Dio2 expression through a cAMP/protein kinase A mechanism. In pinealocytes that had been stimulated by NE, inhibition of transcription by actinomycin had no discernible effect on Dio2 expression. In contrast, inhibition of protein synthesis by cycloheximide enhanced the NE induction of Dio2 expression, suggesting the involvement of a repressor protein. Transient transfection of pinealocytes with adenovirus expressing small interfering RNA against Fos-related antigen 2 (Fra2) enhanced the NE induction of Dio2 expression, whereas the effect of overexpression of the full-length transcript of Fra2 was inhibitory. Time-course study indicated that preventing the NE induction of Fra2 enhanced the NE induction of Dio2 after 3 h, and the enhancement persisted beyond 6 h after NE stimulation. In comparison, transient transfection of pinealocytes with small interfering RNA against inducible cAMP early repressor (Icer) had no effect on the NE induction of Dio2 expression, whereas overexpression of the full-length transcript of Icer caused a small reduction of the NE-stimulated Dio2 expression. Together, our results support Fra-2 as an important transcriptional repressor that helps shape the time profile of the adrenergic induction of Dio2 expression in the rat pineal gland.

Role of protein turnover in the activation of p38 mitogen-activated protein kinase in rat pinealocytes

Differences in the time profiles of activation between p38MAPK and p42/44MAPK by norepinephrine (NE) in rat pinealocytes suggest involvement of mechanisms other than the phosphorylation cascades in their activation. In the present study we investigated whether protein turnover played a role in regulating p38MAPK activation in the rat pineal gland. NE stimulation caused an increase in MAPK kinase3/6 (MKK 3/6) and p38MAPK phosphorylation that occurred in the absence of changes in the mRNA or protein levels of p38MAPK or MKK3/6. The stimulatory effect of NE on phosphorylated MKK3/6 and p38MAPK, but not phosphorylated p42/44MAPK, was blocked by treatment with actinomycin or cycloheximide, indicating a requirement of transcription and translation in activation of the p38MAPK but not the p42/44MAPK pathway. Moreover, inhibition of proteasomes by clasto-lactacystin beta-lactone or Z-Leu-Leu-Leu-CHO (MG132) selectively increased basal and NE-stimulated phosphorylated MKK3/6 and p38MAPK levels without affecting the mRNA or protein levels of MKK3 or p38MAPK. In contrast, the effect of proteasomal inhibition on NE-stimulated p42/44MAPK phosphorylation was inhibitory. Treatment with MG132 also reduced the decline in the phosphorylated levels of NE-stimulated MKK3/6 and p38MAPK that normally follows beta-adrenergic blockade. Together, our results indicate that p38MAPK but not p42/44MAPK activation in the rat pineal gland is tightly coupled to protein synthesis and degradation. The synthesis of an activator upstream of MKK3/6 is required for the NE-activation of p38MAPK.

Mitogen-activated protein kinase phosphatase-1 (MKP-1): >100-fold nocturnal and norepinephrine-induced changes in the rat pineal gland

The norepinephrine-driven increase in mitogen-activated protein kinase (MAPK) activity is part of the mechanism that regulates arylalkylamine N-acetyltransferase (AA-NAT) activity in the rat pineal gland. We now report a marked nocturnal increase in the expression of a MAPK phosphatase, MAP kinase phosphatase-1 (MKP-1), that was blocked by maintaining animals in constant light or treatment with propranolol. MKP-1 expression was regulated by norepinephrine acting through both alpha- and beta-adrenergic receptors. These results establish a nocturnal increase in pineal MKP-1 expression that is under the control of a photoneural system. Because substrates of MKP-1 can influence AA-NAT activity, our findings suggest the involvement of MKP-1 in the regulation of the nocturnal AA-NAT signal.

Adrenergic regulation and diurnal rhythm of p38 mitogen-activated protein kinase phosphorylation in the rat pineal gland

In this study, we investigated adrenergic and photoneural regulation of p38MAPK phosphorylation in the rat pineal gland. Norepinephrine (NE), the endogenous neurotransmitter, dose-dependently increased the levels of phosphorylated MAPK kinase 3/6 (MKK3/6) and p38MAPK in rat pinealocytes. Time-course studies showed a gradual increase in MKK3/6 and p38MAPK phosphorylation that peaked between 1 and 2 h and persisted for 4 h post NE stimulation. In cells treated with NE for 2 and 4 h, the inclusion of prazosin or propranolol reduced NE-induced MKK3/6 and p38MAPK phosphorylation, indicating involvement of both alpha- and beta-adrenergic receptors for the sustained response. Whereas treatment with dibutyryl cAMP or ionomycin mimicked the NE-induced MKK3/6 and p38MAPK phosphorylation, neither dibutyryl cGMP nor 4beta-phorbol 12-myristate 13-acetate had an effect. The NE-induced increase in MKK3/6 and p38MAPK phosphorylation was blocked by KT5720 (a protein kinase A inhibitor) and KN93 (a Ca(2+)/calmodulin-dependent kinase inhibitor), but not by KT5823 (a protein kinase G inhibitor) or calphostin C (a protein kinase C inhibitor). In animals housed under a lighting regimen with 12 h of light, MKK3/6 and p38MAPK phosphorylation increased in the rat pineal gland at zeitgeber time 18. The nocturnal increase in p38MAPK phosphorylation was blocked by exposing the animal to constant light and reduced by treatment with propranolol, a beta-adrenergic blocker. Together, our results indicate that activation of p38MAPK is under photoneural control in the rat pineal gland and that protein kinase A and intracellular Ca(2+) signaling pathways are involved in NE regulation of p38MAPK.

Inhibition of p38 mitogen-activated protein kinase enhances adrenergic-stimulated arylalkylamine N-acetyltransferase activity in rat pinealocytes

We have previously shown that inhibition of p38(MAPK) increases adrenergic-stimulated p42/44(MAPK) activation in rat pinealocytes. In this study we investigated whether p38(MAPK) played a role in the adrenergic regulation of arylalkylamine-N-acetyltransferase (AA-NAT) induction and melatonin (MT) synthesis. Treatment of pinealocytes with norepinephrine (NE) caused a time-dependent increase in the levels of AA-NAT mRNA, AA-NAT protein, and enzymatic activity as well as MT production. Cotreatment with SB202190, a selective p38(MAPK) inhibitor, although having no effect on AA-NAT activity or protein level 3 h after NE treatment, caused a sustained increase in AA-NAT activity and protein level after 6 h of NE treatment. The increases in NE-stimulated AA-NAT activity and protein level by SB202190 occurred in the absence of an increase in AA-NAT mRNA. Similar results were obtained when AA-NAT was induced by (Bu)(2)cAMP or when SB203580 was used to inhibit p38(MAPK). In comparison, SB202474, the inactive analog, had no effect on NE or (Bu)(2)cAMP-stimulated AA-NAT activity or protein level. SB202190 also increased cumulative NE-stimulated MT production, provided that the medium was supplemented with 5-methoxytryptamine. p38(MAPK) inhibitors had no effect on hydroxyindole-O-methyltransferase activity. These results show that inhibition of p38(MAPK), although having no effect on cAMP-mediated AA-NAT transcription, appears to increase AA-NAT activity either by increasing translation or by reducing degradation of the AA-NAT protein. The lack of effect on NE-stimulated MT accumulation by p38(MAPK) inhibitors in the absence of 5-methoxytryptamine could be secondary to a lack of substrate, or alternatively, hydroxyindole-O-methyltransferase may become limiting.

Regulation of 90-kilodalton ribosomal S6 kinase phosphorylation in the rat pineal gland

In this study we investigated diurnal changes in the activation state of the 90-kDa ribosomal S6 kinase (p90RSK) in the rat pineal gland. In animals housed under a lighting regimen with 12 h of light, we found an increase in phosphorylated p90RSK during the dark phase, and this increase was abolished by treatment with propranolol or continuous exposure to light. To determine the intracellular mechanism involved, rat pinealocytes were treated with norepinephrine. Norepinephrine caused a parallel increase in phosphorylated p42/44 MAPK (p42/44(MAPK)) and p90RSK that was reduced by prazosin or propranolol, indicating involvement of both alpha(1)- and beta-adrenergic receptors. Treatment with dibutyryl cGMP, 4beta-phorbol 12-myristate 13-acetate, or ionomycin mimicked norepinephrine-stimulated p90RSK phosphorylation, whereas dibutyryl cAMP caused a decrease in p90RSK phosphorylation. Inhibition of p42/44(MAPK) activation by UO126 was effective in reducing norepinephrine-stimulated p90RSK phosphorylation. Moreover, UO126 had an inhibitory effect on norepinephrine-stimulated arylalkyl-N-acetyltransferase activity. These results indicate that the adrenergically regulated nocturnal increase in p90RSK phosphorylation is mainly mediated through a cGMP-->p42/44(MAPK)-dependent mechanism.

Tissue-specific transgenic knockdown of Fos-related antigen 2 (Fra-2) expression mediated by dominant negative Fra-2

Fos-related antigen 2 (Fra-2) is a member of the Fos family of immediate-early genes, most of which are rapidly induced by second messengers. All members of this family act by binding to AP-1 sites as heterodimeric complexes with other proteins. However, each appears to have a distinct role. The role and biology of Fra-2 are less well understood than those of its relatives c-Fos, Fra-1, and FosB; moreover, Fra-2 target genes remain largely unknown, as does the basis of its selective effects on transcriptional activity. To pursue these issues, we created a transgenic rat line (NATDNF2) in which a dominant negative fra-2 (DNF2) gene is strongly expressed in the pineal gland; tissue selectivity was achieved by putting the DNF2 gene under the control of the rat arylalkylamine N-acetyltransferase (AANAT) regulatory region, which targets gene expression to a very restricted set of tissues (pineal gland >> retina). Expression of AANAT is normally turned on after the onset of darkness in the rat; as a result, pineal DNF2 expression occurs only at night. This was associated with marked suppression of the nocturnal increase in fra-2 mRNA and protein levels, indicating that DNF2 expression inhibits downstream effects of Fra-2, including the maintenance of high levels of fra-2 gene expression. Analysis of 1,190 genes in the NATDNF2 pineal gland, including the AANAT gene, identified two whose expression is strongly linked to fra-2 expression: the genes encoding type II iodothyronine deiodinase and nectadrin (CD24).