Aspirin, one of non-steroidal anti-inflammatory drugs (NSAIDs), has been widely used for its analgesic, anti-inflammatory and anti-thrombotic effects. However, the serious gastrointestinal damage remains a significant problem. The inhibition of cyclooxygenase (COX) enzymes and subsequent reduction of gastric mucosal protective prostaglandin (PG) were first proposed to be the reason of aspirin-induced gastropathy. Recently, there is increasing evidence that COX-independent effects, such as direct irritation, influence of inflammatory mediators and the involvement of microcirculation, might also be important for aspirin-induced gastropathy.

Numerous strategies have been undertaken to decrease the incidence of gastropahty, including co-administration of protective substances such as sucralfate, proton pump inhibitors or prostaglandin analogues. However, the protection effects were not totally satisfied. Except co-administration, investigators also tried to develop safer NSAIDs, for instance, selective COX-2 inhibitors, however some of this type of drugs were withdrawn from the market because of the adverse effect on cardiovascular system. Among these various strategies, the combination of NSAIDs with small protective molecules such as nitric oxide, hydrogen sulfide, and phosphatidylcholine, which help to maintain mucosal integrity, has been highly emphasized recently.

Trehalose, a natural disaccharide, is widely used as a food ingredient and in the cosmetics industry. Recent researches discovered several novel anti-inflammatory and cell-protective functions of trehalose. Trehalose could protect the protein integrity, reduce oxidative damage to cells, inhibit inflammatory cytokine production, and suppress apoptosis. Therefore, trehalose is expected to be a potential powerful therapeutic material for various diseases. Our research group had previously undergone several clinical trials of trehalose on adhesion prevention after abdominal surgery, inhibition of xenosis during dental treatment, and suppressing inflammation and vasospasm induced in an experimental subarachnoid hemorrhage model by its protective effects of cell membranes. Our group further hypothesized that lyophiliaztion of aspirin with trehalose, by which the specific interaction mediated by hydrogen bonding changes, may decrease the severity of aspirin-induced gastropathy.

The aim of this study was to determine whether lyophilization of aspirin with trehalose would decrease the incidence of aspirin-induced gastropthy. In this present study, the author investigated the effect of lyophilized aspirin/trehalose compared with other NSAIDs compounds on gastric mucosa in vitro and in vivo.

In the first chapter, the effects of NSAIDs on gastric cells were evaluated in vitro. A human gastric cell line (AGS) cultured in normal F-12 medium was exposed to aspirin, lyophilized aspirin, mixture of (mix) aspirin/trehalose, and lyophilized aspirin/trehalose at the aspirin concentration of 1-10 mM for 24 hr. After exposure, cytotoxicity was measured by MTT assay, and apoptosis-inducing potency was evaluated by DNA fragmentation analysis and TUNEL assay.

In the MTT assay, lyophilized aspirin/trehalose induced less cell death compared to other three test compounds at the same aspirin concentration. DNA fragmentation in AGS cells treated with aspirin, lyophilized aspirin or mix aspirin/trehalose was shown a ladder pattern on agarose gel at 10 mM aspirin concentration. On the contrary, there was no fragmented DNA pattern shown in any concentrations of lyophilized aspirin/trehalose treatment. Quantification of apoptosis by TUNEL assay also revealed that lyophilized aspirin/trehalose caused less apoptosis than other three test compounds at the same concentration. These results indicated that lyophilized aspirin/trehalose was much less cytotoxic and low apoptosis-inductive in vitro compared to aspirin, lyophilized aspirin and mix aspirin/trehalose, suggesting that the cytoprotective effect did not appear either by simple lyophilization of aspirin nor by mixture of trehalose with aspirin.

In the second chapter, the effects of aspirin, mix aspirin/trehalose, and lyophilized aspirin/trehalose on gastric mucosa were evaluated in a rodent acute gastric ulceration model. Acute gastric ulceration was induced by a single oral administration of 200mg/kg aspirin, 960mg/kg mix aspirin/trehalose (200 mg/kg aspirin), 960mg/kg lyophilized aspirin/trehalose (200mg/kg aspirin), or vehicle (0.5% carboxymethylcellulose aqueous solution) in rats. Rats (n=6-8) were euthanized 5 hr after drug administration, and the stomach was exicised and grossly evaluated for the ulcerative lesions. The tissues were prepared for histology for gastric injuries and immunohistochemistry. In situ TUNEL assay and cleaved caspase-3 immunohistochemistry were performed to evaluate the apoptosis-inducing potency of each treatment. The acute ulceration was also induced in other groups of rats (n=6-8) by the same regimens, and the stomach was excised for the mRNA expression of inflammatory mediators by real-time polymerase chain reaction (PCR) and detection of mucosal PGE2 concentration by EIA kit.

In this rodent acute ulceration model, lyophilized aspirin/trehalose was proved to induce less extent of gastropathy than aspirin and mix aspirin/trehalose macroscopically and microscopically. The results of in situ TUNEL assay and cleaved caspase-3 immunohistochemistry revealed that lyophilized aspirin/trehalose was less potent to induce apoptosis despite profound inhibition of mucosal prostaglandin synthesis. The inhibition of PGE2 synthesis elicited by the lyophilized aspirin/trehalose was comparable with that of aspirin, which indicated the lyophilization procedure did not interfere the ability to inhibit COX activity. Moreover, lack of ulceration with presence of profound suppression of gastric PG synthesis suggested that lyophilized aspirin/trehlaose exerted protective effects that counteracted the potential damaging effects of COX inhibition. However, there were no significant differences between each group in the relative mRNA expression of inflammatory mediators, which might be due to the sampling time of gastric mucosa.

In the third chapter, the effects of aspirin and lyophilized aspirin/trehalose on gastric mucosa were evaluated in a canine ulceration model. Thirteen beagle dogs were assigned into aspirin (n=5), lyophilized aspirin/trehalose (n=5), and control groups (n=3). The dogs received oral administration of 25 mg/kg aspirin, lyophilized aspirin/trehalose (25 mg/kg aspirin), or vehicle every 12 hr for 28 consecutive days. The NSAID-induced gastropathy was evaluated on gastroscopy and scored using the modified Lanza scale 7 days before drug administration and 5, 14, 28 days after initiation of drug administration. Another 6 dogs were used to measure the plasma aspirin concentration by high-performance liquid chromatography (HPLC) at 1, 2, 4, 6, 8, 12,16, 24 hr after aspirin or lyophilized aspirin/trehalose administration.

The gastric lesion scores of the aspirin group rapidly increased after administration on day 5 and maintained high on day 14, while that of the lyophilized aspirin/trehalose group was significantly lower. The scores of the aspirin group slightly decreased from day 14 to day 28, and this decrease may be due to the adaptation effect of long-term aspirin administration. Although there was no significant difference in scores between the two groups on day 28, the aspirin group attained a higher lesion score than the lyophilized aspirin/trehalose group. In addition, the plasma aspirin level increased rapidly after administration and peaked 2 hr after administration in both groups. The mean peak plasma aspirin concentration and the area under curve showed no significant difference between both groups. These results indicated that lyophilized aspirin/trehalose induced less gastric ulceration than aspirin alone while maintaining therapeutic concentrations of plasma aspirin -.

In conclusion, lyophilized aspirin/trehalose could reduce gastric mucosal injuries mainly through the inhibition of apoptosis while maintaining the anti-inflammatory effect. These effects were proved both in vitro and in vivo, and also appear in a canine model. These results indicated that lyophilized aspirin/trehalose might be a solution to decrease aspirin-induced gastropathy, and encouraged the clinical application, both for medical and veterinary medicine.

アスピリンは代表的な非ステロイド系抗炎症薬で、鎮痛薬として広く用いられてきたが、消化器に対する重篤な副作用が現在でも問題となっている。胃粘膜を保護するプロスタグランジン(PG)の産生が、シクロオキシゲナーゼ(COX)抑制を通じて減少することが、消化器障害の原因だと考えられてきたが、近年、直接刺激や炎症メディエイターの影響、微小循環の関連などの別経路も原因と考えられるようになった。

消化器障害を減少させるために、粘膜保護剤の併用やCOX-2選択的阻害NSAIDsの開発などが行われてきたが、いまだ十分ではない。最近では、一酸化窒素や硫化水素などの低分子物質をNSAIDsに結合させて、粘膜保護効果を発揮させる方法が注目されている。

トレハロースは、天然の二糖類で食品添加物などに使用されているが、近年、抗炎症効果や細胞保護効果を持つことが見いだされた。すなわち細胞に対する酸化障害を防ぎ、炎症性サイトカインの産生を防止し、アポトーシスを抑制することが示され、さまざまな疾患の治療薬として期待されている。

今回の実験では、トレハロースとアスピリンの組み合わせにより、アスピリン誘発性胃潰瘍を減少できるかどうかを検討した。なおトレハロースとアスピリンは、単なる混合物と凍結乾燥品の二つを比較した。

第1章ではトレハロースとアスピリンが胃粘膜細胞に与える影響をin vitroで検討した。ヒト胃粘膜細胞株(AGS)にアスピリン(A)、凍結乾燥アスピリン(lyoA)、アスピリンとトレハロースの混合物(MixA/T)、アスピリンとトレハロースの凍結乾燥品(LyoA/T)の4種類の物質をアスピリン相当量1~10mMの濃度にして加えた。24時間の培養後に、MTT法で評価したところ、LyoA/Tは他の3物質に比べ、低い細胞毒性を示した。またDNA断片化アッセイではA、LyoA、MixA/Tが10mM濃度でDNA断片化を示したのに対し、LyoA/Tでは断片化は現れなかった。またTUNEL法でもLyoA/Tでは他の3物質に比べ、アポトーシスが明らかに少なかった。以上の結果から、LyoA/Tはin vitroではA、LyoA、MixA/Tにくらべ、細胞毒性が少なく、アポトーシス誘導能も低いことが示された。

第2章ではアスピリンとトレハロースの効果を、ラットを用いた急性胃潰瘍モデルを用いて検証した。ラットを4群にわけ、対照群(カルボキシメチルセルロース投与)、A、MixA/T、LyoA/T(いずれもアスピリン相当量200mg/kg)を設定した。各薬剤の投与5時間後に安楽殺し、胃を採取した。

肉眼的及び病理組織学的評価ではLyoA/Tでの胃粘膜障害は明らかに他の3群よりも軽度であった。in situ TUNELアッセイおよびcleaved capase-3の免疫組織学的検索でもLyoA/Tにおけるアポトーシスは他の3群よりも有意に減少していた。一方、胃粘膜細胞のPGE2濃度はA、MixA/T、LyoA/Tの3群すべてで、対照群に対して有意に抑制されており、LyoA/TでもアスピリンのCOX抑制効果は保たれているものと考えられた。しかしながら4種の炎症性サイトカイン (IL-1β,IL-6,TNF-α,ICAM-1)のmRNA発現量は、すべての群で有意な差は認められず、トレハロースが炎症性サイトカインを抑制している所見は得られなかった。

以上の結果より、ラットin vivoモデルにおいても、LyoA/Tはアスピリン作用を維持しながら、胃粘膜障害を有意に減少させうることが示され、またその理由としては、炎症性サイトカイン産生抑制よりもアポトーシス抑制が主たる経路であると推測された。

第3章ではより大型動物である犬を用い、4週間にわたる長期投与を行った。健康なビーグル犬を対照群、A群、LyoA/T群の3つにわけ、アスピリン25mg/kg相当量を1日2回、28日間連続で投与した。投与前、投与5、14、28日目に内視鏡検査を行い、胃粘膜障害の程度をLanzaスケールにて評価した。A群の胃粘膜障害スコアは、5日後には急激に増加し14日後も有意な増加を維持したのに対し、LyoA/T群では実験期間を通じてほとんど上昇は認められなかった。また薬剤投与後24時間にわたる経時的な採血を行い、血中サリチル酸濃度を液体クロマトグラフィーで測定したところ、A群、LyoA/T群ともにほぼ同様のカーブを示し、有効血中濃度に到達した。以上の結果より、犬を用いたin vivoの検討でも、LyoA/Tはアスピリンとしての薬効を維持しながら、胃粘膜障害を大幅に抑制する可能性が示された。

これらの結果を総合すると、アスピリンとトレハロースの凍結乾燥品は、胃粘膜障害を大幅に減少させた新しいNSAIDsとして、有用である可能性が示唆された。よって審査委員一同は、本論文が博士(獣医学)論文として価値あるものと認めた。