血基質氧化酶(Heme oxygenase, HO)

於 1968 年，血基質氧化 酶首度被證實普遍存在大鼠的脾臟、肝、腎和骨髓等組 織，且主要分布在細胞內的微粒體(microsomes)中(Tenhunen et al., 1968)。血基質氧 化酶是血基質(heme)代謝過程中的速率限制酶，藉由催化血基質的分解，產生等莫 耳數的膽綠素(biliverdin)、一氧化碳(carbon monoxide, CO)以及二價鐵離子(Fe^2＋)，

其中膽綠素會進一步經膽綠質還原酶(biliverdin reductase)轉換為膽紅素(bilirubin)，

而鐵離子會誘發攜鐵蛋白(ferritin)的生成(圖 1-13)(Farombi and Surh, 2006)。研究發 現，血基質的代謝產物，具有許多重要的生理功能，其功能分別敘述如下：

（1） 膽綠素和膽紅素：

已被證實能夠減少脂質過氧化反應、清除過氧化氫避免自由基的產生，有助於 降低氧化壓力，減少血管局部缺氧的傷害(Baranano et al., 2002; Stocker and Ames, 1987)。

（2） 一氧化碳：

在體內高濃度的一氧化碳會產生傷害，但有研究指出低濃度一氧化碳卻具有類 似一氧化氮(Nitrite oxide, NO)的功能，可作為訊息傳遞分子，活化血管平滑肌胞內 的 soluble guanylate cyclase (sGC)，增加環狀鳥嘌呤單磷酸(cyclic guanosine

monophosphate, cGMP)的生成，同時具有促使血管舒張、防止血小板凝集及抑制血 管平滑肌增生等作用(Morita et al., 1997; Piantadosi, 2008)。

（3） 鐵離子：

游離鐵離子(free iron)對細胞是具有氧化毒性的，而血基質氧化 酶分解血基質而

產生的二價鐵離子(Fe^2＋)會誘發攜鐵蛋白(ferritin)的產生，攜鐵蛋白除了可捕捉游離

鐵以減少氧化傷害之外，並可透過與細胞內鐵離子結合，增加細胞鐵儲存的效率，

維持細胞鐵離子濃度的恆定(Balla et al., 2005; Poss and Tonegawa, 1997)。

圖 1-13. 血基質氧化酶之作用及其代謝產物(Farombi and Surh, 2006) 。

（一）血基質氧化酶之類型

截至目前，哺乳類生物的HO依不同基因序列和蛋白質分布等特性，區分為 HO-1、HO-2和HO-3三種不同亞型(Tenhunen et al., 1968)，分別為第一型血基質氧化 酶(HO-1)，分子量約為32 KDa，為熱休克蛋白(Heat shock proteins, HSPs)的一種；第 二型血基質氧化酶(HO-2)，分子量約為34 KDa；以及第三型血基質氧化酶(HO-3)分 子量為約33 kDa。HO-2為持續表現型(constitutive expression)，存在各組織細胞中，

其中以腦、神經系統、心血管以及睪丸等組織具有較高的濃度(Fan et al., 2011; Maines, 1997)；HO-3首度由老鼠腦中發現，存在於脾臟、肝臟、胸腺、前列腺、心臟、腎 臟、大腦和睾丸中，HO-3活性為三種亞型中最低且催化能力最弱，其生理功能目前 仍未釐清(McCoubrey et al., 1997)。其中HO-1屬於誘發型(inducible form)且為參與血 基質代謝中最主要的亞型，HO-1容易被血基質或其他物質如過氧化氫、ultraviolet (UV)、氧化壓力、發炎的細胞激素及重金屬(Lin et al., 2005; Ryter et al., 2006)所誘 發。許多研究證據顯示，當組織或細胞受到氧化損傷時會誘導HO-1大量表現，達到 保護細胞之目的。HO-1不只能利用其代謝產物達到抗氧化、抗發炎、抗增生及抗細 胞凋亡等作用 (Clark et al., 2000; Lee et al., 2009a; Petrache et al., 2000)，最近研究發

現，DHA可透過活化PI-3 kinase/AKT和MEK/ERK訊號傳遞路徑，誘發HO-1基因的 轉錄作用，進一步降低BV-2腦微膠細胞(microglia)神經發炎反應(neuroinflammatory responses)，達到抗憂鬱之作用(Lu et al., 2010a)；也有文獻證實sulforaphane及quercetin 等植物化學素(phytochemical)可藉由誘導HO-1的增加進而抑制肝癌及乳癌細胞的生 長及遷移的能力(Cornblatt et al., 2007; Keum et al., 2006; Lin et al., 2008b)；更有文獻 證明骨型態發生蛋白-6 (bone morphogenetic protein, BMP-6)顯著抑制MCF-7乳癌細 胞移行及侵襲作用與其誘發HO-1基因表現進而抑制MMP-9活性有關，其中，BMP-6 主要透過磷酸化活化Smad1/5蛋白質，並促使其結合至HO-1基因promoter上的 Smad-responsive element，進而增加HO-1基因轉錄活性(Wang et al., 2011)。由上述文 獻可得知，HO-1可經由多種訊號傳遞路徑所活化，進而達到抗發炎、抗氧化以及抑 制腫瘤細胞生長等作用。

研究目的

根據中華民國行政院衛生署於100年所公布統計資料顯示，台灣女性乳癌發生率 為所有癌症之冠，且乳癌的死亡率位居女性癌症第四名，乳癌致死原因主要來自於 乳癌細胞具有高度轉移能力，使癌細胞易擴散並侵犯鄰近組織及器官所致。不少研 究指出乳癌之發生與飲食攝取的油脂有密切關係，而脂肪酸除了作為生物能量的來 源外，也是細胞膜上磷脂質組成之重要成分；飲食中的n-3 及n-6 脂肪酸比例會影 響類二十碳烯酸(Eicosanoids)的代謝，對於在維持人體正常生長、發展、調控細胞生 理、生化及代謝作用上扮演著重要的角色。在目前已有許多研究支持n-3 脂肪酸具 有減少腫瘤之進展；然而，對於n-6 脂肪酸調控腫瘤生長及代謝作用仍具有爭議；

因此本研究將探討二十二碳六烯酸(Docosahexaenoic acid, DHA)和亞麻油酸(Linoleic acid, LA)是否具有降低乳癌細胞轉移(metastasis)和侵襲(invasion)作用的能力，並進 一步瞭解其可能調控的訊息傳遞路徑，以期透過飲食來降低乳癌的發生或惡化的進 程。

第二部分

Effect of Docosahexaenoic Acid and Linoleic Acid on TPA-Mediated MMP-9 Expression in MCF-7 Human Breast Cells

Introduction

Breast cancer is the most common female cancer and is the second leading cause of cancer deaths in Western women. About 30% to 40% of women with this form of cancer will develop metastases and eventually die of this disease (Weigelt et al., 2005).

According to the statistical data of Department of Health of Taiwan, the incidence of breast cancer has increased 4.5 fold in the past twenty years, and is the fourth leading cause of cancer death in Taiwanese women (Chang, 2006).

Metastatic spread of cancer cells is the main cause of death of breast cancer patients (Weigelt et al., 2005). Breakdown of the extracellular matrix (ECM) by proteinases is an essential step in cancer metastasis (Werb, 1997). Matrix metalloproteinases (MMPs), a family of ECM degrading proteinases, are divided into four subclasses based on the substrate including collagenases, gelatinases, stromelysin, and elastases (Nelson et al., 2000; Yan and Boyd, 2007). Activation of MMP-2 (gelatinase-A) and MMP-9

(gelatinase-B) is intensely correlated with the tumor invasion and metastasis in different types of cancer cell, including human breast (Blanckaert et al., 2010; Hanemaaijer et al., 2000), hepatoma (Zhao et al., 2011), prostate (Wegiel et al., 2008) and lung cancer cells (Kamaraj et al., 2010). In general, MMP-2 is constitutively expressed in highly metastatic tumors, whereas MMP-9 can be stimulated by the growth factor, such as epidermal growth factor and transforming growth factor beta (TGF-β) (Ramirez et al., 2011), the inflammatory cytokine such as tumor necrosis factor-α (TNF-α) (Youn et al., 2011), ultraviolet radiation (Kimura and Sumiyoshi, 2011), or phorbol ester (Lin et al., 2008a).

The phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA), a potent tumor promoter, stimulates renal tumor cell proliferation through activation of protein kinase C (PKC)(Kolb and Davis, 2004). TPA-induced MMPs activation was mediated by

modulating the activation of transcription factors such as NF-κB and AP-1 through PKC, PI3K and mitogen-activated protein kinase (MAPK) signaling pathways (Blumberg, 1988;

Jang et al., 2007). Recent studies showed that the dietary factors such as α-lipoic acid,

capsaicin, andconjugated linoleic acid (CLA) are protective against cancer migration, invasion and angiogenesis by suppressing MMP-9 expression or enzyme activity (Hwang et al., 2011b; Kunigal et al., 2007)). In our previous study, phenobarbital-induced JNK1/2 and ERK2 activation was down-regulated by DHA (Lu et al., 2009) which suggests DHA may possess the ability to suppress the MMP-2 or MMP-9 activation. In other words, DHA can be the potential candidate for antitumor.

Dietary lipids are important to human beings because of their role in energy and essential fatty acids supplies. Linoleic acid (18:2 n-6) and α-linolenic acid (18:3 n-3) are essential fatty acids that must be obtained from diets. These polyunsaturated fatty acids (PUFAs) and their metabolic products play critical roles in a variety of physiological processes, such as regulation of inflammation (Masson and Mensink, 2011), insulin resistance (Perez-Martinez et al., 2011), blood pressure (Sagara et al., 2011) and lipid metabolism (Neff et al., 2011). Epidemiologic studies showed that high consumption of n-3 PUFAs such as eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) from fatty fish is associated with a reduced risk for breast cancer (Kim et al., 2009).

Experimental animal and cell culture studies provided evidences that dietary n-3 and n-6 PUFAs inhibit the promotion and progression stages of carcinogenesis (Lee et al., 2009b;

Lu et al., 2010a; Sun et al., 2008).

Heme oxygenase 1 (HO-1) is one of the members of HO system. HO-1 is also known as HSP32 (heat shock protein of 32 kDa), and it is an inducible enzyme and expressed relatively low in most tissues under basal conditions. HO-1 is induced by a wide variety of stimuli such as ultraviolet A radiation, endotoxin and cytokines (Chung et al., 2011; Luo et al., 2011; Ronco et al., 2011; Xu et al., 2011; Zhong et al., 2010). In addition to anti-oxidant and anti-inflammatory activities of HO-1 (Seo et al., 2010), HO-1 has also been shown to possess anti-tumorigenic action in breast cancer cells (Li et al., 2011; Pae et al., 2010; Wang et al., 2011). It is also shown that HO-1 is induced by a wide array of phytochemicals through Nrf2 (Velmurugan et al., 2009). In addition to the above mentioned stimuli, induction of HO-1 by DHA in BV-2 microglia (Lu et al., 2010a) and mouse peritoneal macrophages (Wang et al., 2010) was reported. However, the effect of n-3 and n-6 PUFAs on HO-1 induction in human cancer cells lacks.

Because of the HO-1 induction capability of DHA, it is possible that DHA can exert antitumor activity. According to previous studies describing the antitumor activity of n-3

and n-6 PUFAs, we investigated the metastasis and invasion inhibition effects of n-3 and n-6 PUFAs in TPA-induced MCF-7 human breast cancer cell and the possible mechanism involved.

Materials and Methods

Chemicals

Dulbecco's Modified Eagle Medium (DMEM), OPTI-MEM, 25% trypsin-EDTA, and penicillin-streptomycin solution were from GIBCO-BRL (Grand Island, NY); fetal bovine serum (FBS) was from HyClone (Logan, UT);

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), albumin, bovine serum essentially fatty acid free (BSA), sodium bicarbonate, calcium chloride, Triton X-100, 12-O-tetradecanoylphorbol 13-acetate (TPA), GF109203X (PKC kinase inhibitor), wortmannin, and LY294002 (PI3K kinase inhibitor) were from Sigma-Aldrich, Inc. (St.

Louis, MO); SP600125 (JNK inhibitor), PD98059 (ERK inhibitor), SB203580 (p38 inhibitor) were from TOCRIS (Ellisville, MO); docosahexaenoic acid and linoleic acid were from Cayman Chemical (Ann Arbor, MI); collagen was from Collaborative Biomedical Products (Bedford, MA); TRIzol reagent was from Molecular Research Center, Inc (Cincinnati, OH); antibodies against Akt, phospho-Akt (T308 and S473), ERK1/2, phospho-ERK1/2, p38, and phospho-p38were from Cell Signaling Technology (Danvers, MA); antibodies against JNK1 and phospho-JNK1/2 were from Santa Cruz Biotechnology (Santa Cruz, CA); antibody against HO-1was from Calbiochem

(Darmstadt, Germany); and DharmaFECT 1 Transfection Reagent was from Dharmacon (Lafayette, CO).

Cell culture

The human breast cancer cell line MCF-7 was a kindly gift from Dr. Yi-Hsien Hsieh, Chung Shan Medical University, Taichung, Taiwan, and was cultured on collagen-coated cell culture dishes in DMEM (pH 7.2) supplemented with 1.5 g/L NaHCO₃, 10% FBS,

100 units/mL penicillin, and 100 μg/mL streptomycin at 37℃ in a 5% CO2 humidified incubator.

Cell viability assay

Cell viability was assessed by the MTT assay. The MTT assay measures the ability of viable cells to reduce a yellow 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide to a purple formazan by mitochondrial succinate dehydrogenase. MCF-7 cells were grown to 70-80% confluence and were then treated with different concentrations of docosahexaenoic acid or linoleic acid (0-200 μM) for 20 h followed by incubation with TPA (100 ng/mL) for another 24 h. Finally, the DMEM was removed, and the cells were washed with PBS. The cells were then incubated with MTT (0.5 mg/mL) in DMEM at 37℃ for an additional 3 h. The medium was removed, and isopropanol was added to dissolve the formazan. After centrifugation at 20,000g for 5 min, the supernatant of each sample was transferred to 96-well plates, and absorbance was read at 570 nm in an ELISA reader. The absorbance in cultures treated with 0.005% ethanol was regarded as 100% cell viability.

Western blot analysis

After each experiment, cells were washed twice with cold PBS and were harvested with 150 μL of lysis buffer (10 mM Tris-HCl, pH 8.0, 0.1% Triton X-100, 320 mM sucrose, 5 mM EDTA, 1 mM PMSF, 1 mg/L leupeptin, 1 mg/L aprotinin, and 2 mM dithiothreitol).

Cell homogenates were centrifuged at 14,000g for 20 min at 4℃.The resulting

supernatant was used as a cellular protein for Western blot analysis. The total protein was analyzed by use of the Coomassie Plus protein assay reagent kit (Pierce Biotechnology

Inc., Rockford, IL). Equal amounts of proteins were electrophoresed in a sodium dodecyl sulfate-polyacrylamide gel, and proteins were then transferred to polyvinylidene fluoride membranes (Millipore Corp., Bedford, MA). Nonspecific binding sites on the membranes were blocked with 5% nonfat milk in 15 mM Tris/150 mM NaCl buffer (pH 7.4) at 4℃

overnight. After blocking, the membranes were incubated with anti-phospho-Akt (T308 and S473), anti-phospho-JNK1/2, anti-phospho-ERK1/2, anti-phospho-p38, anti-Akt, anti-JNK1, anti-ERK1/2, anti-p38, anti-MMP-9, anti-HO-1, and anti-β-actin antibodies at 4℃ overnight. Thereafter, the membranes were incubated with the secondary

peroxidase-conjugated anti-rabbit or anti-mouse IgG antibodies at room temperature for 1 h, and the immunoreactive bands were developed by use of the Western Lightning™

Plus-ECL kit (PerkinElmer, Waltham, MA) and were scanned by a luminescent image analyzer (Fujifilm LAS-4000, Japan). The bands were quantified with an ImageGauge (Fujifilm).

RNA isolation and RT-PCR

Total RNA of MCF-7 cells was extracted by using TRIzol reagent. Briefly, after treatment, cells were washed twice with cold PBS and scraped with 500 μL of TRIzol reagent. Samples were mixed with 100 μL of chloroform and centrifuged at 11,000g for 15 min. The supernatant was collected and mixed with 250 μL of isopropyl alcohol. After centrifugation at 12,000g for 20 min, the supernatant was discarded and the cell pellet

was stored in 70% ethanol or dissolved in deionized water for quantification.

We used 0.4 μg of total RNA for the synthesis of first-strand cDNA by using Moloney murine leukemia virus reverse transcriptase (Promega Co., Madison, WI) in a 20-μL of final volume containing 250 ng of oligo-dT and 40 U of RNase inhibitor. PCR was

conducted in a thermocycler in a reaction volume of 50 μL which containing 20-μL of cDNA, BioTaq PCR buffer, 50 μM of each deoxyribonucleotide triphosphate, 1.25 mM MgCl₂, and 1 U of BioTaq DNA polymerase (BioLine). Oligonucleotide primers of MMP-9 (forward, 5’-CACTGTCCACCCCTCAGAGC-3’; reverse,

5’-GCCACTTGTCGGCGATAAGG-3’), HO-1 (forward, 5’-CTGAGTTCATGAGGAACTTTCAGAAG-3’; reverse,

5’-TGGTACAGGGAGGCCATCAC-3’), and glyceraldehyde-3-phosphate

dehydrogenase (GAPDH) (forward, 5’-CCATCACCATCTTCCAGGAG- 3’; reverse, 5’-CCTGCTTCACCACCTTCTTG-3’) were designed on the basis of published sequences (Lin et al., 2008a; Sun et al., 2009). Amplification of MMP-9 was achieved when samples were heated to 95℃ for 5 min and then immediately cycling 30 times through 30 sec denaturing step at 94℃, 30 sec annealing step at 56℃, and a 1min

elongation step at 72℃. Amplification of HO-1 was achieved when samples were heated to 95℃ for 5 min and then immediately cycling 39 times through a 1 min denaturing step at 95℃, a 1 min annealing step at 55℃, and a 2 min elongation step at 72℃, respectively.

The GAPDH cDNA level was used as the internal standard. PCR products were resolved in a 1% or 2% agarose gel, scanned by using a Digital Image Analyzer (Alpha Innotech) and quantified with an ImageGauge.

RNA interference by small interfering RNA of HO-1

Predesigned small interfering RNA (siRNA) against human HO-1 and nontargeting control pool siRNA were purchased from Dharmacon (Lafayette, CO). MCF-7 cells were transfected with HO-1 siRNA SMARTpool by using DharmaFECT1 transfection reagent

according to the manufacturer’s instructions. Specific silencing was confirmed by at least three independent immunoblotting assays with cellular extracts 24 h after transfection.

Nuclear extract preparation

After each experiment, cells were washed twice with cold PBS and were then scraped from the dishes with 1000 μL of PBS. Cell homogenates were centrifuged at 2,000g for 5 min. The supernatant was discarded, and the cell pellet was allowed to swell on ice for 15 min after the addition of 200 μL of hypotonic buffer containing 10 mM HEPES, 1 mM MgCl₂, 1 mM EDTA, 10 mM KCl, 0.5 mM DTT, 0.5% Nonidet P-40, 4 μg/mL leupeptin, 20 μg/mL aprotinin, and 0.2 mM PMSF. After centrifugation at 7,000g for 15 min, pellets containing crude nuclei were resuspended in 50 μL of hypertonic buffer containing 10 mM HEPES, 400 mM KCl, 1 mM MgCl2, 0.25 mM EDTA, 0.5 mM DTT, 4 μg/mL leupeptin, 20 μg/mL aprotinin, 0.2 mM PMSF, and 10% glycerol at 4℃ for 30 min. The samples were then centrifuged at 20,000g for 15 min. The supernatant containing the nuclear proteins was collected and stored at -80℃ until the Western blot assay and electrophoretic mobility shift assays.

Electrophoretic mobility shift assay (EMSA)

EMSA was performed according to our previous study (Cheng et al., 2004). The LightShift Chemiluminescent EMSA Kit (Pierce Chemical Co., Rockford, IL) and synthetic biotin-labeled double-stranded AP-1 consensus oligonucleotides (forward:

5’-GCCTCAGCTGGTAAATGGATAA-3’; reverse:

5’-AAAGGCCCCAGAGCCAGCC-3’) were used to measure AP-1 nuclear protein-DNA binding activity (Tsai et al., 2007). Ten micrograms of nuclear extract, poly (dI-dC), and

biotin-labeled double stranded AP-1 oligonucleotide were mixed with the binding buffer (LightShift EMSA Kit; Pierce Chemical Co., Rockford, IL) to a final volume of 20 μL, and the mixture was incubated at room temperature for 30 min. Unlabeled

double-stranded AP-1 oligonucleotide and a mutant double-stranded oligonucleotide were used to confirm the protein-binding specificity. The nuclear protein-DNA complex was separated by electrophoresis on a 6% TBE-polyacrylamide gel and was then transferred

double-stranded AP-1 oligonucleotide and a mutant double-stranded oligonucleotide were used to confirm the protein-binding specificity. The nuclear protein-DNA complex was separated by electrophoresis on a 6% TBE-polyacrylamide gel and was then transferred