Stress is a term that is widely used in modern society. Stress activates two main physiological pathways: the activation of the autonomic nervous system and hypothalamic-pituitary-adrenal (HPA) axis, which results in an increase in plasma glucocorticoids levels. HPA axis controls reaction to stress and regulates many body processes, including the immune system, digestion, energy storage and expenditure, sexuality, and mood and emotions. Activation of the HPA axis is considered as the key way to response to stresses to maintain the steady states in the organism. There is a large body of evidence to support that activation of the HPA axis is regulated by signaling through molecules like prostaglandins (PGs), especially PGE2, in the brain. PGs are produced by metabolism of arachidonic acid by cyclooxygenase (COX), which exists in two distinct isoforms, COX-1 and COX-2. COX-1 has the characteristics of a housekeeping gene, which is constitutively expressed in most cells and tissues, whereas COX-2 is a rate-limiting enzyme in PGs synthesis, which is generally expressed at low basal levels and rapidly induced in response to various stress stimuli. Furthermore, our laboratory has previously shown that glucocorticoids maintain gonadotropin secretion through the suppression of COX-2 under various acute stress conditions, suggesting that COX-2 plays a role as a common mediator of stresses in the brain. However, the mechanisms by which different acute stresses converge to regulate COX-2 and PGs synthesis and the relationship between COX-2 and PGs synthesis and activation of the HPA axis under various acute stress conditions are not fully understood. This dissertation was conducted to investigate the relationship and mechanism between hypothalamic COX-2-related signaling and activation of the HPA axis under three different acute stress conditions, namely infectious, hypoglycemic and restraint stresses.

In Chapter 1, I evaluated the involvement of hypothalamic COX-2-related signaling in activation of the HPA axis. Three different acute stresses, i.e. infectious (lipopolysaccharide, LPS), hypoglycemic (2-Deoxy-D-glucose, 2DG) and restraint stresses, were applied to male intact and adrenalectomized (ADX) rats. First, I used an unselective COX inhibitor (indomethacin) or a selective COX-2 inhibitor (NS-398) to elucidate the roles of COX-2-related signaling pathway in activation of the HPA axis under three different acute stress conditions. Subsequently, the gene expression of corticotrophin-releasing hormone (CRH), COX-2, microsomal prostaglandin E synthase-1 (mPGES-1), and interleukin-1β (IL-1β) in the hypothalamus were examined by real-time RT-PCR using intact and ADX rats. The results showed that both indomethacin and NS-398 significantly attenuated the increase of serum corticosterone levels after LPS and restraint stresses, but not after 2DG injection. All the three acute stresses induced significant increases in serum corticosterone levels in intact rats at the both time points examined (30 and 120 min), while COX-2 and mPGES-1 mRNA levels in the hypothalamus were not increased at 30 min in intact rats. CRH mRNA levels were not changed throughout the experimental period under all the three acute stress conditions. COX-2 and mPGES-1 mRNA levels were significantly increased 120 min after LPS injection, but were not up-regulated after 2DG and restraint stresses in intact rats. In ADX rats, only LPS and restraint stresses increased CRH mRNA levels at 120 min, which was completely restored by corticosterone treatment. COX-2 and mPGES-1 mRNA expression was significantly increased 120 min after 2DG and restraint stresses, which was blocked by treatment with corticosterone. In contrast, following LPS injection, COX-2 and mPGES-1 mRNA levels in ADX rats were not different from those in intact rats at both 30 and 120 min. IL-1β mRNA levels in the hypothalamus in intact rats were increased at 120 min only by LPS injection, though those in ADX rats were increased by all the three acute stress stimuli. These results suggest that the relationship between COX-2-related signaling and activation of the HPA axis is stress-specific, and that COX-2-related signaling preferably mediates infectious and restraint stresses. Furthermore, the expression of COX-2 and mPGES-1 mRNA under the infectious stress condition were not negatively regulated by endogenous glucocorticoids, likely due to an increase in IL-1β levels.

The results in Chapter 1 suggest that activation of the HPA axis is dependent on COX-2-related signaling under infectious and restraint stress conditions, but less dependent on it under hypoglycemic stress condition. In Chapter 2, I evaluated the mechanisms underlying a stress-specific relevance between COX-2-related signaling and activation of the HPA axis. First, I outlined the stress-induced pattern of neuronal activation via the expression of the immediate early gene product c-Fos by immunostaining in the brain under infectious, hypoglycemic and restraint stress conditions. Subsequently, I used a selective COX-2 inhibitor (NS-398) to elucidate the roles of COX-2-related signaling pathway in the neuronal activation. The results showed that all the three acute stresses increased the number of c-Fos-immunoreactive (c-Fos-IR) cells in several brain regions, including the paraventricular hypothalamic nucleus (PVN), supraoptic nucleus (SON), arcuate nucleus, central amygdaloid nucleus, medial amygdaloid nucleus, hippocampus and piriform cortex. Quantitative analysis showed that the number of c-Fos-IR cells in the PVN and SON at 120 min were higher than those of other 2 time points (0 and 30 min) under all the three acute stress conditions. Moreover, LPS and restraint stresses increased the number of c-Fos-IR cells in the ventromedial hypothalamic nucleus (VMH), while 2DG injection increased it in the lateral hypothalamic area (LHA), supporting the notion that the VMH and LHA play a role as satiety and feeding centers, respectively. In the PVN, NS-398 did not decrease the number of c-Fos-IR cells at any time points after three different acute stresses. Furthermore, in the SON, the number of c-Fos-IR cells in NS-398-treated rats was significantly higher than those in vehicle-injected rats at 30 min after LPS and restraint stresses, whereas there was no such difference at any time points after 2DG stress. These results indicate that, among the brain regions responding to stresses, the PVN and SON are strongly activated by stresses. In addition, it is suggested that, while CRH neurons in the PVN are involved in activation of the HPA axis, oxytocin neurons, which are known to be inhibitory to the HPA axis, in the SON as well as PVN are suppressed through COX-2-related signaling and involved in the activation of the HPA axis under infectious and restraint stress conditions. This stress-specific difference in the activation of the brain nuclei following treatment with COX-2 inhibitor may at least partially account for a stress-specific relevance between COX-2-related signaling and activation of the HPA axis.

In conclusion, one of the major findings of this thesis is that activation of the HPA axis is at least partially dependent on COX-2-related signaling under infectious and restraint stress conditions, while less dependent on it under hypoglycemic stress condition. The mechanisms underlying this stress-specific relevance between COX-2-related signaling and activation of the HPA axis may involve stress-specific activation of CRH neurons in the PVN and oxytocin neurons in the PVN and SON. The second major finding of this thesis is that endogenous glucocorticoids negatively regulate COX-2-related signaling under hypoglycemic and restraint stress conditions, but not under infectious stress condition, probably because infectious stress stimulation is mediated by cytokines including IL-1β in the hypothalamus. Thus, the present study revealed that, although COX-2-related signaling is potentially up-regulated by all the three acute stresses used, the role and regulatory mechanism of it varies among stresses, which may be important for animals to effectively respond to each stress condition.

哺乳類では一般にストレスに暴露された場合、視床下部-下垂体-副腎(HPA)軸が活性化され、グルココルチコイド分泌が亢進して体内の恒常性の維持、ひいては生命の維持に必須の役割を果たしている。申請者の研究室では、様々なストレス刺激は脳内でシクロオキシゲナーゼ-2(COX-2)の発現を誘導し、COX-2により合成されるプロスタグランジン(PGs)が各種ストレスの共通の仲介物質となっていることを明らかにした。さらに、ストレス刺激により大量に分泌されたグルココルチコイドは、COX-2の発現を抑制することによりストレス反応を制御していることが示された。しかし、各種のストレス条件下におけるCOX-2関連シグナルの制御機構や、そのHPA軸活性化に対する関与については十分に理解されていない。本論文では、感染ストレスモデルとしてリポ多糖(LPS)投与群、低血糖ストレスモデルとして2-デオキシ-グルコース(2-DG)投与群、および拘束ストレスモデル群の3群を設け、これらの急性ストレスに対する反応における視床下部COX-2関連シグナルの役割とその作用機序について解明することを目的とした。

第1章ではHPA軸活性化におけるCOX-2関連シグナルの役割を明らかにすることを目的に、ストレス負荷後0分、30分、120分の血中コルチコステロン濃度をラジオイムノアッセイで測定するとともに、視床下部のCOX-2、コルチコトロピン放出ホルモン(CRH)、膜結合型 PGE2 合成酵素(mPGES-1)、インターロイキン-1β(IL-1β)のmRNA発現量をリアルタイムPCRで測定した。まず、非選択的 COX 阻害剤インドメタシンあるいはCOX-2選択的阻害剤NS398の血中コルチコステロン濃度に対する影響を調べたところ、LPSと拘束ストレスではこれらの阻害剤によりコルチコステロン濃度が有意に低下したが、2-DGでは阻害剤の効果が認められなかった。視床下部CRHのmRNA発現量はLPSと拘束ストレスでは有意な上昇が認められたが、コルチコステロンの投与によりこの上昇が抑制された。COX-2とその下流因子であるmPGES-1のmRNAの動態はほぼ同様で、インタクト群ではLPSにおいてのみ上昇が見られ、その発現は副腎摘出(ADX)やコルチコステロン投与によっても変化しなかったが、2-DGと拘束ストレスではADX群でのみ軽度な上昇が見られた。視床下部のIL-1β mRNA発現量は、インタクト群ではLPSによってのみ有意な上昇が見られ、またストレスの種類に関わらずADX群ではその発現量が有意に高くなっていた。これらの結果より、LPSと拘束ストレスではHPA軸の活性化がCOX-2関連シグナルに依存しているが、2-DGでは他のストレスモデルよりもCOX-2関連シグナルに対する依存度が低いことが示唆された。さらに、拘束ストレスと2-DGでは内因性のグルココルチコイドがCOX-2関連シグナルの活性化を抑制しているがLPSでは抑制しないことが示され、その理由としてLPSではIL-1βを介してCOX-2関連シグナルの活性化が起こるためであることが考えられた。

第2章では3種のストレス条件下での脳内の神経活動のパターンにCOX-2関連シグナルがどのように関与しているかを解明することを目的に、c-Fosの発現を免疫染色により調べた。その結果、全てのストレス条件下で特に視床下部の室傍核(PVN)と視索上核(SON)においてc-Fos陽性細胞数が有意に増加した。COX-2選択的阻害剤NS398は全てのストレス条件下でPVNのc-Fos陽性細胞数に影響を与えなかった。一方、SONにおいてはLPSと拘束ストレスではCOX-2阻害剤はc-Fos陽性細胞数を増加させるが、2-DGは影響を与えないことが示された。これらの結果より、LPSと拘束ストレスではCOX-2関連シグナルはHPA軸に対して抑制的な作用をもつPVNとSONのオキシトシンニューロンの興奮性を抑制することにより、HPA軸の活性化に関与していることが考えられた。LPSと拘束ストレスにおいては第1章の結果からCOX-2阻害剤によりPVNに存在するCRHニューロンの興奮性は低下することが期待されるが、PVNのc-Fos陽性細胞数に変化は認められなかったことから、これらのストレス条件下ではCOX-2阻害剤はPVN のCRHニューロンの興奮性を低下させるがオキシトシンニューロンの興奮性を上昇させるため、c-Fos陽性細胞の全数に変化を生じないことが考えられた。一方、2-DGではCOX2阻害剤によってもPVNとSONのc-Fos陽性細胞の数に変化が認められなかったことから、他のストレスモデルよりも両神経核のニューロンの興奮性はCOX-2関連シグナルに対する依存度が低いということが示唆された。

以上のように、本研究において用いた全てのストレスにより潜在的にはCOX-2関連シグナルの活性化が生じるが、その制御機構やHPA軸の活性化における役割はストレスの種類により異なることが明らかとなった。これらの知見は各種のストレスに対する生体の対応メカニズムの解明に大きく貢献するとともに、ストレスに関連する病態の発生機構の解明や治療法の開発にも資するもので、学術的、応用的意義は少なくない。よって、審査委員一同は本論文が博士(獣医学)の学位論文として価値あるものと認めた。