Autocrine growth hormone (hGH) and chemotherapeutic drug resistance in mammary carcinoma cells 3

Thus, understanding this protective effect of autocrine hGH in mammary carcinoma cells may contribute to novel adjunct therapeuticapproaches to the treatment of mammary cancer 1.3 Possib

Chapter 1 Introduction

Breast cancer is by far the most frequent cancer in women and the leading cause of cancer related death in women Breast cancer has been a major fatal disease in westernized countries, but it is becoming a prominent one also in several developing countries Numerous factors have been shown to be associated with increased breast cancer risks Understanding those factors might contribute to novel adjunct therapeuticapproaches in the treatment of breast cancer

A large body of evidence has indicated an association between hormones and breast cancer risk In addition to estrogen, growth hormone has recently been emerging as one hormone involved in breast cancer

1.1 Hormones and breast cancer risk

That hormones influence risk of breast cancer has been known for decades Women who have had bilateral oophorectomy early in life are at markedly reduced risk of subsequently developing breast cancer; the earlier oophorectomy is done, the greater the reduction Early age at menarche (11 years or less) and late age at natural menopause (55 or older) are associated with increased risk Late age at full-term

pregnancy (30 or more) increases risk compared to an early age (less than 20) Nulliparity increases risk and high parity decreases risk, at least after age 50 (Kelsey

et al 1993; Rosner et al 1994) Obesity, which increases risk in postmenopausal women, is thought to operate through a hormonal mechanism (Sellers et al 1992) Obese women metabolize androstenedione, produced in the adrenal cortex, to oestrogen in adipose cells The higher levels of serum oestrone in obese as compared

to nonobese postmenopausal women may account for the excess breast cancer risk (Cauley et al 1989) In premenopausal women, breast feeding, if continued for many months or years, appears to reduce risk A multinational study conducted by the World Health Organization, suggested that oral contraceptives are associated with increased risk for breast cancer incidence before the age of 40-45 (Thomas and Noonan 1990)

Many studies have been done to elucidate the association of hormones and breast cancer risk Associations have been demonstrated between estrogen, prolactin, growth hormone (GH) and insulin-like growth factor-1 (IGF-1) with breast cancer In this study, we focused on the role of GH in breast cancer

GH is one of the mammotrophic hormones involved in the development of the breast hGH mRNA identical to pituitary hGH mRNA is also expressed bynormal mammary tissue and by benign and malignant human mammary tumors and the immunoreactive

hGH is restricted to epithelial cells (Mol et al 1995a) A recent study showed increased expression of the hGH gene in proliferative disorders of the mammary gland (Raccurt et al 2002) The pituitary and mammary gland GH gene transcripts originatefrom the same transcription start site but are regulated differentially, since mammary gland GH gene transcription does not require Pit-1 (Lantinga-van Leeuwen

et al 1999) GH receptor mRNA and protein have also been detected in human mammary gland epithelia (Sobrier et al 1993) Both endocrine GH and autocrine/paracrine-producedGH possess the capacity to exert a direct effect onthe development and differentiation of mammary epithelia in vitro (Plaut et al 1993) and

in vivo (Feldman et al 1993) All these findings indicate the possible involvement of hGH in mammary carcinoma

To further elucidate the effect of autocrine hGH in mammary carcinoma cell, a new

in vitro cell culture model was developed (Kaulsay et al 1999) The hGH gene or a

translation-deficient hGH gene were stably transfected into MCF-7 cells, a human mammary carcinoma cell line (Kaulsay et al 1999) The autocrine hGH producing cells display a marked insulin-like growth factor-1-independent increase in cell number in both serum-free and serum-containing conditions as well as a specific increase in STAT5-mediated transcription (Kaulsay et al 1999) Also, autocrine hGH productionresulted in enhancement of the rate of mammary carcinoma cell spreading

on a collagen substrate In a later study, Kaulsay demonstrated that the increase in mammary carcinoma cell number was not only a result of increased mitogenesis but also a decrease in apoptosis by autocrine hGH (Kaulsay et al 2000) This antiapoptotic effect of autocrine hGH might explain the phenomenon that metastatic

mammary carcinoma, which show the highest expression of hGH, is more resistant to chemotherapy Thus, understanding this protective effect of autocrine hGH in mammary carcinoma cells may contribute to novel adjunct therapeuticapproaches to the treatment of mammary cancer

1.3 Possible mechanisms involved in resistance to breast cancer treatment

It has been frequently noticed that chemotherapy is efficient in early stages of breast cancer, whereas advanced tumors are usually resistant to the same treatments The molecular basis for this resistance is not understood Many chemotherapeutic drugs are DNA or cytoskeleton damaging drugs that show some specificity towards dividing cells In recent studies, increased cellular oxidative level was shown to be a key effector of cell death induced by chemotherapeutic drugs (Wen et al 2002) However, in advanced tumors the cellular oxidative level is down regulated and, consequently, death signaling is suppressed This difference in the level of antioxidant enzymes in tumors may underlie the difference in responsiveness to chemotherapy (Akman et al 1990) Increased expression of antioxidant enzymes has been documented in a wide variety of malignant tumors including breast cancer (di Ilio et al 1985; Iscan et al 2002) and higher expression level of antioxidant enzymes was shown to be associated with large, poorly differentiated breast tumors (Thomas et

al 1997)

Understanding these mechanisms of chemoresistance may help us find new effective therapeutic approaches for the treatment of chemoresistant breast cancer patients

1.4 Direction of the study

As stated above, both the clinical and cell model data suggest an association between autocrine hGH and breastcancer Autocrine hGH production by mammary carcinoma cells could impact on breast cancer cell behavior and consequently cancer prognosis

An increased antioxidant enzyme level has been shown to contribute to chemoresistance of advanced cancer cells and this increased level been documented

in a wide variety of malignant tumors including breast cancer (di Ilio et al 1985; Iscan et al 2002) and is associated with large, poorly differentiated breast tumors (Thomas et al 1997) It is clear that more studies need to be done to further elucidate the possible relationship between autocrine hGH and antioxidant enzymes level in breast cancer cells

The first line defense antioxidant enzymes are superoxide dismutase (SOD), glutathione peroxidase (GPX), glutamylcystein synthetase (GCS) and catalase Among them, catalase is the most potent enzyme regulating oxidative stress level by converting H2O2 to H2O and O2 in peroxisomes A higher expression level of catalase has been found in tumor tissues compared with normal tissues (Ripple et al 1997), and this has been correlated with mammary tumor malignancy grade and prognosis (Thomas et al 1997) However, to date the mechanisms responsible for the regulation

of catalase are largely unknown Therefore, further work is necessary to elucidate the pathways and molecules involved in the regulation of catalase and explore the

possible signaling pathways used by autocrine hGH to promote chemoresistance and influence the outcome of cancer treatment

1.5 Objectives

The purpose of the study was to analyze the impact of autocrine hGH on chemotherapy resistance, concentrating especially in further characterization of the role of autocrine hGH on antioxidant enzymes, especially catalase, in mammary carcinoma cells, but also exploring other mechanisms of chemoresistance in relation

to autocrine hGH My work is therefore mainly focused on exploring the signaling pathways by which autocrine hGH regulates catalase level and its consequent influence on the responsiveness of mammary neoplasia to chemotherapy A detailed description of the purpose of my study is listed below:

I Analysis of the impact of autocrine hGH on chemotherapy resistance using the MCF-7 cell model

a Analyzing the effect of autocrine hGH on cell death induced by chemotherapeutic drugs in mammary carcinoma cells

b Analyzing the effect of autocrine hGH on cell death induced by other oxidative stress inducers in mammary carcinoma cells

II Characterization of the role of autocrine hGH on antioxidant enzymes in mammary carcinoma cells might contribute to the chemoresistance effect of autocrine hGH in mammary carcinoma cells

a Analyzing the effect of autocrine hGH on antioxidant enzymes: catalase, SOD, GPX, GCS at mRNA and protein level by RT-PCR and Western Blot in mammary carcinoma cells respectively

b Analyzing the contribution of induced catalase expression by autocrine hGH to the protective effect of autocrine hGH in mammary carcinoma cells using catalase-specific inhibitor and forced expression of catalase

by its expression vector

c Exploration of the signaling pathways by which autocrine hGH regulate catalase level

• Analyzing effect of autocrine hGH on a 4.5Kb catalase gene promoter in mammary carcinoma cells

• Using different pathway inhibitors to identify the main pathway through which catalase level is regulated by autocrine hGH in mammary carcinoma cells

• Confirming the importance of this pathway involved in the protective effect of autocrine hGH

III Characterization of the role of autocrine hGH on apoptotic proteins in mammary carcinoma cells might contribute to the chemoresistance effect of autocrine hGH in mammary carcinoma cells

a Analyzing the effect of autocrine hGH on apoptotic proteins: 2,

Bcl-xl, Bak, Bax, p53 at protein level by Western Blot in mammary carcinoma cells

b Exploring the effect of autocrine hGH on Bcl-xl promoter activity

c Exploring the effect of autocrine hGH on Bcl-2 phosphorylation status

IV Characterization of the role of autocrine hGH on telomerase activity in mammary carcinoma cells which might contribute to the chemoresistance effect

of autocrine hGH in mammary carcinoma cells

a Analyzing the effect of autocrine hGH on telomerase activity

b Analyzing the effect of autocrine hGH on telomerase associated proteins

at mRNA and protein level by RT-PCR and Western Blot in mammary carcinoma cells respectively

c Exploring the regulation mechanism of autocrine hGH on hTERT mRNA

by promoter reporter assay

Ongoing research of growth hormone receptor-specific antagonists holds promiseat blocking the actions of hGH at the endocrine and autocrinelevels within the breast Analyzing the relationship and signalling cascades between autocrine hGH and catalase in mammary carcinoma cells may answer the question of differences in chemotherapy response of breast cancer patients Thus, it might yield a novel adjuvant therapeutic approach for the treatment of chemoresistant breast cancer patients

Chapter 2 Literature review

2.1 Growth hormone

2.1.1 Growth hormone structure

Human growth hormone (hGH) is a 191 amino acid single chain polypeptide with a molecular weight of 22kDa It consists of four antiparallel alpha helices (Abdel-Meguid et al 1987)with two interchain disulfide bonds (Li and Dixon 1971; Niall 1971) and an unsubstituted amino terminus (Wallis 1992) Two human GH genes, hGH-N and hGH-V have been identified (MacLeod et al 1991) In man, 22kDa hGH

is expressed in the pituitary from the hGH-N gene (located on chromosome 10), which contains 5 exons The 22kDa from of GH is the most abundant one and contributes to 70-75% of circulating GH (Wallis 1992) The hGH-N undergoes alternate splicing at the 3’ acceptor site in exon 3 (DeNoto et al 1981) to produce a 20kDa GH variant by a deletion of amino acids 32-46 (Lewis et al 1978), which constitutes up to 15% of secreted GH (DeNoto et al 1981) Western blot analysis of human serum has demonstrated 4 different molecular weight forms of GH at 27 KDa, 22kDa, 20kDa and 17kDa (Sinha and Jacobsen 1994; Warner et al 1993) The 17kDa hGH consists of amino acids 44-191 (reportedly diabetogenic fragment of hGH) and

is altered by the physiologic state of the individual (Sinha and Jacobsen 1994) The hGH-V variant GH (Wallis 1992) is synthesized by the placenta and is encoded by a

separate gene (Frankenne et al 1987) Human GH and hGH-V differ by only 13 of a total of 191 amino acids

Growth hormone (GH) belongs to a family of polypeptide hormones, which includes prolactin (PRL), placental lactogen (PL; somatomammotropin), and proliferin (Linzer and Nathans 1985; Nicoll et al 1986; Wallis 1992) Human GH shares the greatest sequence homology with human PL, more than with porcine GH (73%), bovine GH, equine GH and rat GH (each approximately 65%) (Wallis et al 1978), or the PRL family (approximately 27%) (Paladini et al 1983) Only primate GHs possess growth promoting activity in humans (Knobil 1959) due to the fact that only primate GHs have the capacity to bind the hGH receptor (hGHR) (Carr and Friesen 1976) Primate

GH is also the only GH type that possesses non-species-specific lactogenic activity by binding to the PRL receptor in non-primates (Nicoll et al 1986; Ranke et al 1976) due to specific domains of the hGH molecule, which are homologous to PRL (Lowman et al 1991; Nicoll et al 1986) Amino acid residues responsible for binding

of GH to the GHR (Cunningham et al 1989; Cunningham and Wells 1989) and PRL receptor (PRLR) (Lowman et al 1991) have been identified

2.2 Pituitary regulation and extrapituitary sites of GH expression

2.2.1 Pituitary regulation of GH expression

Somatotroph-specific GH expression is controlled mainly by the pituitary-specific factor Pit-1 (also known as GHF-1 and PUF-1) Pit-1 has been implicated in GH promoter activation in response to increases in intracellular cAMP levels (Cohen et

al 1999a; Sekkali et al 1999) Multiple binding sites for the pituitary POU domain of the transcription factor Pit-1 have been identified within the promoter region of the hGH-N gene, expressed exclusively in somatotrophs and lactotrophs of the anterior pituitary, and have been shown to be required for high level, somatotroph-specific expression of a linked hGH-N gene (Shewchuk et al 1999) In response to phorbol esters or cAMP treatment, the phosphorylation of Pit-1 has been detected at a highly conserved residue (220 Thr) in the POU homeodomain and shown to modify Pit-1 recognition and the transcriptional activation of specific target genes (Kapiloff et al 1991) However, phosphorylated Pit-1 is not required for either basal or induced GH promoter activity (Okimura et al 1994) Cotransfection assays with GH-reporter genes and in vitro mutated or native pit-1 have demonstrated that phosphorylation-deficient Pit-1 proteins are required for, and able to mediate, cAMP response at the

GH promoter Thus, it has been suggested that Pit-1 is not the functionally relevant target of PKA and PKC-mediated signal transduction, but that an additional, Pit-1 dependent factor must be responding to hormonal/environmental cues that increase intracellular cAMP levels In analyses of the hGH flanking region, two CREs and a

Pit-1 recognition element were found to be necessary for the response of the GH gene

to forskolin addition to GC cells (Shepard et al 1994) CREB/ATF-1 was further identified as the proteins bound at the CREs (Shepard et al 1994), and, therefore, may directly interact with Pit-1 to mediate the hGH activation by cAMP in the pituitary Further, mutations in a second pituitary transcription factor, known as Prophet of Pit-1 (Prop-1), have been detected in dwarf mice (Pfaffle et al 1999), and the patients with Prop-1 mutations exhibit combined pituitary hormone deficiency (Pfaffle et al 1999) Prop-1 gene expression has also been documented in human pituitary tumors and normal human adult pituitary tissues, suggesting that Prop-1 may serve as an essential transcriptional factor for pituitary specific gene expression in human (Nakamura et al 1999)

Retinoic acid (RA) induction of the Pit-1 gene, containing the protein domains specific and POU homeo both necessary for DNA binding and activation of the hGH gene, can be impaired by a Pit-1 gene mutation (Cohen et al 1999b) Activin, a member of the transforming growth factor β family, negatively regulates GH expression, and decreases both basal and growth hormone-releasing hormone (GHRH)-induced GH synthesis, primarily acting to decrease the level Pit-1, which in turn decreases GH mRNA synthesis (Struthers et al 1992) Pit-1 mRNA expression

POU-in GH-secretPOU-ing pituitary adenomas has further been demonstrated to be associated with the activin beta A subunit, suggesting a role for activin in the expression of Pit-1

in these adenomas (Sanno et al 1998)

Glucocorticoid stimulation of cultured rat anterior pituitary cells has been reported to increase GH mRNA levels (Isaacs et al 1987; Matsubara et al 1997; Nogami et al 1997), and is consistent with the transient increases in plasma GH concentrations seen

in humans after acute corticoteroid administration (Dieguez et al 1996) Additionally, intracellular cAMP levels affect the proliferation of somatotrophs as well as modulate

GH synthesis (Sekkali et al 1999; Sirotkin and Makarevich 1999) Thyroid hormone also plays a major role in the control of GH synthesis (Matsubara et al 1997) In somatotropic cell lines, the addition of triiodothyonine increased GH transcription,

GH protein levels, and promoted marked ultrastructural changes in these cells compatible with a stimulated synthesis of GH (Bonaterra et al 1998; Giustina and Wehrenberg 1995)

2.2.2 Extrapituitary sites of GH expression

Extrapituitary sites of GH synthesis have been demonstrated, including neuronal groups within the central nervous system (Gossard et al 1987; Hojvat et al 1982), circulating lymphocytes (Weigent and Blalock 1989), bone marrow and thymus (Binder et al 1994) These extrapituitary sites of GH synthesis exhibit both different ontogenesis and non-coordinate regulation compared to pituitary-derived GH (Hojvat

et al 1982; Hojvat et al 1986) GH expression has also been demonstrated in fetal serum of various species (Werther et al 1993) In addition, GH mRNA has been demonstrated to be expressed as early as 72 hours after fertilization (Pantaleon et al 1997), and GH immunoreactivity as early as day 4 (Urbanek et al 1992), while the

placenta has been shown to produce GH variants in some species Furthermore, GH mRNA-synthesizing cells have been shown to be present in the endothelial cells and lymphocytes of the human thymus (Wu et al 1996), thymocytes and thymic epithelial cells of the human thymus (de Mello-Coelho et al 1998; van Buul-Offers and Kooijman 1998), human immune system (Yang et al 1999), human spleen (Pankov 1999), and T and B cell lymphocytes (Hooghe et al 1998) In the human leukemic HL-60 cell line, major immunoreactive GH bands of molecular weights 22, 20 and 40kDa were detected by western blot analysis (Costoya et al 1996), while normal GH gene (GH-N) expression, whose expression was dependent upon the proliferative state of the cells, was concordantly detected While GH transcripts were easily detectable in actively proliferating cells, only minute amounts were observed when cells were induced to differentiate with DMSO (Costoya et al 1996)

GH has also been reported to be produced in normal dog mammary gland (Selman et

al 1994; van Garderen et al 1997), in transformed canine mammary tissue (van Garderen et al 1997) and in feline mammary gland fibroadenomas (Mol et al 1995b)

In the canine mammary gland, immunoreactive GH was observed to be localized to the mammary epithelial cells and correlated with the presence of GH mRNA (van Garderen et al 1997), and local GH synthesis was also confirmed microscopically via the observation of GH-containing secretary granules (van Garderen et al 1997) Further, GH was also associated with areas of hyperplastic mammary epithelium, which may indicate that locally produced GH enhances cell proliferation within the canine mammary gland, acting in an autocrine/paracrine manner In the canine

mammary gland, a Pit-1-independent mechanism for progestin/progesterone induced mammary GH expression has been documented with the finding of little or no detectable Pit-1 mRNA in progestin-treated dogs with high mammary GH gene expression (Lantinga-van Leeuwen et al 1999) However, the possibility that Pit-1 may still play a role in mammary GH expression after malignant transformation remains, as suggested by GH expression in Canine malignant mammary tumors despite the fact that no progesterone receptors were detectable by ligand analysis (Mol et al 1995b; van Garderen et al 1997) Consistent with these findings, Pit-1 is not required for GH gene expression in bone marrow cells in mice (Kooijman et al 1997) Further, expression of GH-related prolactin gene also appears to be independent of Pit-1 in non-pituitary tissues (Gellersen et al 1994), and no Pit-1 mRNA was detected in a human prolactin-producing tumor cell line (Gellersen et al 1995)

2.3 Cellular and transcriptional regulation by GH

2.3.1 Cellular effects of GH

In vitro, GH has been demonstrated to stimulate cellular proliferation and DNA synthesis in a variety of cell types, including the epiphyseal growth plate, adipose tissue (Wabitsch et al 1996), skeletal muscle, liver, kidney, heart, and gastric mucosa (Lobie et al 1992) GH also stimulates proliferation and DNA synthesis in undifferentiated rat primary adipocyte precursor cells (Wabitsch et al 1996) In vitro,

GH has been shown to stimulate the proliferation of insulinoma cells (Billestrup and Martin 1985), and is also known to stimulate proliferation of pancreatic β cells (Nielsen et al 1999; Sekine et al 1998) along with a corresponding increase in DNA synthesis within these cells (Sekine et al 1998) Further, GH-stimulated mitogenesis

in islet β cells has been shown to be dependent upon GTP-binding proteins and dependent mechanisms (Sjoholm et al 2000) GH also stimulates the proliferation of theca and granulose cells of preantral follicles in vitro (Kobayashi et al 2000) In particular, theca cells proliferated to the point of formation of complete layers after incubation with GH (Kobayashi et al 2000) Additionally, in vitro, GH stimulated proliferation of human breast cancer cells (Fujikawa et al 1998), and further synergized with estradiol to promote proliferation in these cells (Fujikawa et al 1998) Similarly, an increased mitogenic response of mammary epithelial cells concordant with increased DNA synthesis in response to GH stimulation has also been observed (Weber et al 2000) Treatment of aged female rhesus monkeys with

PKC-GH has further been associated with a 3 to 4-fold increase in mammary glandular size and epithelial proliferation index (Ng et al 1997)

Several other cell types have also been reported to respond mitogenically to GH stimulation in vitro, including GH receptor-transfected Chinese hamster ovary (CHO) cells (Moller et al 1992) in a p44/42 and p38 MAO kinase-dependent manner (Zhu and Lobie 2000), human leukemic HL-60 cells (Costoya et al 1996), fetal liver and brain cells (Botero-Ruiz et al 1997), erythroid and myeloid progenitor and peripheral blood cells via increased IGF-1 concentrations (Kotzmann et al 1996; Merchav 1998), and osteoblastic cells upon stimulation with IGFBP-5-mediated upregulation

of GHR mRNA levels and GH binding (Slootweg et al 1996) Further, in a separate study, bovine GH (bGH) was demonstrated to stimulate the proliferation of activated murine T cells in vitro treated with either Con A or anti-CD3 antibody (Postel-Vinay

et al 1997) Further, bGH was able to increase T cell lymphocyte proliferation by 2.5-fold over the effect of anti-CD3 alone (Postel-Vinay et al 1997), in a direct manner and not via locally-produced IGF-1, as addition of IGF-1 did not affect cell proliferation when added alone at concentrations ranging form 10-9 to 10-7 M (Postel-Vinay et al 1997) In GH-transgenic mice, increased hepatocellular proliferation and increased DNA turnover was evident (Snibson et al 1999) Additionally, incubation

of human enterocytes with GH increased cell proliferation by 85% and 3H-thymidine incorporation by 64%, compared to control specimens (Canani et al 1999) Crypt cells of the human duodenal mucosa (Challacombe and Wheeler 1995) and colon

epithelial cells have also been reported to undergo increased proliferation in response

of STAT5b (Gebert et al 1999b) Further, IGF-1 gene activation by GH is possible only via synergism with STAT5b and hepatocyte nuclear factor (HNF)-1α (Meton et

al 1999) In addition, both STAT5a and STAT5b have been demonstrated to mediate GH-dependent regulation of CYP3A10/6β-hydroxylase promoter activity (Subramanian et al 1998) GH has also been demonstrated to upregulate the expression pattern of the liver gene CYP2C7 by enhancing the expression of the CYP2C7 protein specifically in the periportal liver region (Oinonen et al 2000) Additionally, NF-Y, a transcription factor involved in the maintenance of high transcriptional levels of many genes and known to functionally interact with other factors (Schwartzbauer et al 1998), modulates the binding activity of STAT5, thereby modulating GH-mediated activation of its transcription (Subramanian et al 1998)

Similarly, GH stimulated transcription of HNF6 gene by a mechanism involving activation of STAT5 and HNF4 (Lahuna et al 2000) Consistent with these findings,

GH treatment of hypophysectomized rat enhanced hepatic HNF4 and HNF3β mRNA concentrations (Lahuna et al 2000)

2.3.3 Other effects of GH

GH has further been reported to stimulate the expression of various transcription factors, including c-fos, egr-1, and Elk-1, a transcription factor associated with the c-fos serum response element (SRE) (Gong et al 1998) in CHO cells expressing the GHR, and the Spi 2.1 gene (Gong et al 1998) In contrast to Spi 2.1 gene activation, however, the N-terminal half of the GHR cytoplasmic domain is sufficient to mediate stimulation of c-fos and egr-1 expression, and Elk-1 activation (Gong et al 1998) The role of the CCAAT/enhancer binding proteins (C/EBPs) in GH regulated transcription has also been investigated C/EBP has been demonstrated to constrain GH-stimulated expression of c-fos in vitro (Liao et al 1999) Specifically, GH stimulates the binding of C/EBP beta and delta to the c-fos promoter region and enhances the dephosphorylation of two isoforms of C/EBP-β, liver activating protein (LAP) and liver inhibitory protein (LIP) (Liao et al 1999), events that may result in regulation of GH-stimulated c-fos expression by C/EBP beta and delta

GH has also been demonstrated to utilize the p38 MAP kinase for regulation of differentiation or differentiated cell function (Zhu and Lobie 2000) p38 MAP kinase

has, further, been shown to be required for hGH stimulation of ATF-2 and mediated transcriptional activation (Zhu and Lobie 2000), hGH-stimulated mitogenesis (Zhu and Lobie 2000) The trans-activation of ATF-2 may provide one mechanism for the p38 MAPK-dependent portion of GH-stimulated cell proliferation since ATF-2 has been demonstrated to cooperated with v-Jun to promote cell proliferation (Huguier et al 1998) Previous reports have demonstrated that modulation of actin dynamics by p38 MAP kinase requires the phosphorylation of HSP27 downstream of MAPKAP-2 and MAPKAP-3 (Hedges et al 1999; Larsen et

CHOP-al 1997) MAPKAP-2 has previously been reported to be activated by GH (Hodge et

al 1998) Furthermore, GH-stimulated reorganization of the actin cytoskeleton has been demonstrated to require phosphatidylinositol-3 (PI-3) kinase activity (Goh et al 1997), suggesting that PI-3 kinase may also be upstream of the GH-dependent increase in p38 MAPK activity It is, therefore, apparent that GH activation of p38 MAPK plays a pivotal role in the mediation of pleiotropic cellular effects of GH

2.4 GH dependent intracellular signaling

2.4.1 GH dependent intracellular signaling

Binding of GH to both of its receptor binding domains results in receptor dimerization at the cell surface, a precursor for the association of JAK2 at the membrane proximal proline rich Box 1 region and subsequent autophosphorylation of JAK2 (Carter-Su and Smit 1998) An influx of calcium ions intracellularly may be an additionally required event at this stage of GHR dimerization although the nature of this is undefined (Carter-Su and Smit 1998) Phosphorylated JAK2 phosphorylates the GHR and several other proteins involved in GH signal transduction, including non-receptor (c-Src, c-Fyn, FAK) and receptor kinases (EGF) (Carter-Su and Smit 1998) and other proteins including the IRS family (IRS-1, 2 and 3), the MAPK family (p44/42, p38 and SAPK), and the STAT family (STATs 1, 3 and 5) Additionally, proteins downstream of the MAP kinases such as Elk-1, CHOP and ATF-2 are also activated (Hodge et al 1998; Zhu and Lobie 2000) GH has also been demonstrated to stimulate the formation of a multiprotein complex centered around CrkII and p130Cas which also comprises of c-Src, c-Fyn, RAK, tensin, paxillin, IRS-1, the p85 subunit of phosphatidylinositol 3-kinase, c-Cbl, Nck, C3G, Shc, Grb-2 and Sos-1 (Zhu et al 1998) Such complex formation would allow for initiation of various pleiotropic cellular effects of GH including activation of the SAP kinases (Zhu et al 1998) and PI-3 kinase (Goh et al 2000) Signal transduction of GH is abolished via SHP-1 action (Carter-Su and Smit 1998) and by removal of GH signal transduction

pathway components through the proteosome complex (Gebert et al 1999a) In contrast, the phosphatase SHP-2 positively regulates GH signal transduction (Stofega

et al 1998) while short-term downregulation of the cellular response to GH is achieved by the SOCS family of proteins (Adams et al 1998)

2.4.2 GH activation of the mitogen-activated protein kinase (MAPK) pathway

MAP kinases play an important role in mammalian cell regulation of gene transcription, cellular proliferation and prevention of apoptosis One of the initial findings leading to the hypothesis that GH activates a tyrosine kinase was the observation that GH stimulates the tyrosine phosphorylation of a large number of cellular proteins, both nuclear and cytoplasmic (Campbell et al 1993; Gronowski and Rotwein 1994; Silva et al 1993) This suggests that JAK2 phosphorylates proteins in addition to JAK2 and GHR and/or that JAK2 activation initiates signaling pathways that involve activation of other tyrosine kinases or inactivation of tyrosine phosphatases Two such proteins that are rapidly phosphorylated and activated in response to GH are the p44/42 MAP kinases (Campbell et al 1992; Winston and Bertics 1992) The MAP kinases are serine/threonine kinases that require phosphorylation on both a tyrosine and threonine for activation (Cobb and Goldsmith 1995)

Ten mammalian MAP kinase family members have been identified to date (Treisman 1996) The MAP kinases p44/42 are predominantly activated by phorbol esters and

polypeptide growth factor including GH (Campbell et al 1992; Moller et al 1992; Winston and Bertics 1992) The other 8 members of the MAP kinase family could be activated predominantly by adverse stimuli such as heat, UV irradiation, osmotic stress and proinflammatory cytokines and are, therefore, termed stress-activated protein kinases (SAP kinases) GH has been reported to activate the JNK/SAP kinase pathway (Zhu et al 1998) One pathway by which the membrane receptor tyrosine kinases activate MAP kinases involves Shc, Grb2, Son-of-sevenless (Sos), Ras, Raf and p44.42 MAP kinase (ERKs) (Crews and Erikson 1993) Experiments using the SH2 domain of the Shc proteins fused to glutathione S-transferase suggest that GH stimulates association of Shc proteins with JAK2-GHR complexes via the SH2 domain of Shc proteins Studies using truncated and mutated GHR suggest that Shc proteins may bind to phosphorylated residues in both GHR and JAK2 (Yi et al 1996) Furthermore, GHR mutants that fail to bind and activate JAK2 also fail to phosphorylate Shc and activate ERK1 and ERK2 (Yamauchi et al 1997; Yi et al 1996) Interestingly, PI-3 kinase activity has been shown to be required for full GH activation of the MAP kinases (Kilgour et al 1996), thus lending credence to the possibility of multiple alternate mechanisms of GH activation of different signaling pathways

Several studies provide evidence that GH initiates the MAPK reaction cascade (Fig 2.1.), including the MAPK designated ERKs (extracellular signal-regulated kinases)-

1 and -2 CHO cells transfected with the full length GH receptor cDNA and then treated with GH resulted in both phosphorylation of the MAPK substrate, myelin

basic protein, as well as in increased mitogenecity of these cells by 30-60% (Moller et

al 1992) CHO clones, expressing the wild type GH receptor with tyrosine residues at position 333 and 338 substituted by phenylalanine, treated with GH resulted in consequent activation of MAPK to the same extent in both cell lines (Lobie et al 1995) GH has also been demonstrated to induce the rapid and transient association of Shc with growth factor receptor bound 2-Sos (Shc-Grb2-Sos) correlating with the time course of Ras, Raf and MEK activation (Vanderkuur et al 1997), indicating that

GH stimulation promotes the association of the Shc-Grb2-Sos complex that then activates Ras, thereby participating in the Raf-MEK-p44/42 MAPK pathway Inactivation of Ras, Raf and MEK correlated directly with serine/threonine phosphorylation of Sos (which was blocked by MEK inhibitor PD098059) and dissociation of Sos from Grb2, but not Grb2 from tyrosine phosphorylated Shc (Vanderkuur et al 1997) It has been further demonstrated that GH activation of p44/42 MAPK is required for elk-1, c-fos, egr-1 and junB mediated transcription (Hodge et al 1998) Hunan GH activation of the p44/42 MAPK pathway has further been demonstrated in the enhancement of mitogenesis in the CHO-GHR1-638 cell line (Zhu and Lobie 2000)

Fig 2.1 GH signaling through the p44/42 MAP kinase pathway.

2.5 GH and mammary gland

2.5.1 GH regulation of mammary gland development

Non-lactogenic and lactogenic growth hormones have been shown to be potent stimulators of mammary development while GHR has been shown to mediate the action of GH within the mammary gland (Kleinberg 1998) A rat animal model was initially employed by Kleinberg and colleagues (1993; 1995) to address whether the effects of GH within the mammary gland were achieved locally In this rat model system (Ruan et al 1995), E2 was administered systemically to hypophysectomized, gonadectomized, and sexually immature rats by subcutaneous implantation of a silastic capsule Different forms of GH in the form of pellets were also implanted into the lumbar mammary gland on one side of the rat Internal controls of bovine serum albumin (BSA) pellets were also implanted into the contralateral mammary gland Mammary gland development was observed within the GH, but not the BSA-containing gland (Ruan et al 1995) Further, a variety of wild-type or mutant GHs were also tested in combination with E2 within the model, and mammary development was apparent only after treatment with hormones expected to bind the rat GHR, specifically hGH, rGH, and bGH (Feldman et al 1993) Little or no mammary development was evident when E2 was administered either alone or in combination with PRL or other lactogenic substances that do not bind appreciably to GHR (Feldman et al 1993) Treatment of rats with either bGH or hGH significantly increased estrogen receptor (ER) mRNA expression, which occurred primarily in the

mammary stoma (Feldman et al 1999) Further, analysis of mammary gland ER by immunocytochemistry revealed that while ER was present in the epithelial cells of non-treated animals, only GH treated animals had ER clearly visible in both glandular and fat cells of the tissue (Feldman et al 1999), demonstrating a specific effect of GH

on the ER in the mammary fat cell

In addition, the expression of GHR mRNA has been documented in the mouse mammary gland, with expression levels of GHR decreasing gradually throughout pregnancy starting on day 8 of gestation and declining further during lactation (Ilkbahar et al 1995) GH receptor mRNA expression has also been documented in the mouse mammary epithelium and stoma (Ilkbahar et al 1999), with expression significantly higher in the stromal versus the epithelial compartment, and highest in virgins, declining during late pregnancy, and lowest in the lactating gland (Ilkbahar et

al 1999) GHR localization has also been observed in ductal and alveolar epithelial cells from rat and rabbit mammary glands (Lincoln et al 1995) In the bovine mammary gland, GHR mRNA is observed from the third to the ninth month of pregnancy localized in the glandular epithelium, surrounding mesenchymal cells, endothelial cells of vessels and in the stratum basale of the epidermis of fetal mammary glands (Knabel et al 1998) Similarly, bGH mRNA expression in the transgenic mouse model has been demonstrated to be specifically confined to the mammary gland at lactation (Oh et al 1999), and in cows, bGH treatment has been reported to result in significant increases in milk yield, but not of milk fat percentage (Beswick and Kennelly 2000) In a transgenic mouse model secreting high levels of

bGH into milk, the mammary gland was found to be the major organ expressing bGH mRNA throughout gestation and during lactation (Thepot et al 1995) Concordantly, the expression of hGH in normal and neoplastic human mammary glands identical to pituitary hGH has been observed (Mol et al 1995a) Similarly, normal, hyperplastic, and neoplastic canine mammary tissue and have also been demonstrated to express

GH mRNA transcripts identical to canine pituitary GH mRNA (Mol et al 1995b)

GH immunoreactivity was further observed to be localized to the mammary epithelial cell compartment and hyperplastic epithelium of the canine mammary gland (Mol et

al 1999), suggesting that locally produced GH enhanced proliferation, acting in an autocrine and/or paracrine manner In agreement with these findings, treatment of aged female rhesus monkeys with GH resulted in a 3 to 4 –fold mitogenic increase in mammary glandular size and epithelial proliferation index (Ng et al 1997) Similarly,

GH has also been implicated in the mammogenesis and early lactogenesis of ewes (Kann 1997)

2.5.2 Effect of GH on the stromal-epithelial compartment

Stromal-epithelial interactions are known to play a critical role in the embryonic to adult stages of mammary development (Cunha 1994; Sakakura 1991; Sakakura et al 1987) Mammary glandular development from ductal branching to terminal end bud (TEB) formation, subsequent alveolar genesis, and milk protein production is dependent on the mammary fat pad (Sakakura 1991; Sakakura et al 1987) and various forms of mesenchyme affect epithelial differentiation in different ways

(Sakakura 1991; Sakakura et al 1987) Further, despite the fact that adult mammary epithelia respond to heterotypic neonatal and embryonic mesenchymes, branching morphogenesis, alveoli formation, and milk production are variable depending on the type of mesenchyme (Cunha 1994)

Although both the stromal and epithelial compartments of the mammary gland possess estrogen and GH receptors, suggesting action of these hormones at either location, several pieces of evidence demonstrate that the action of GH is more predominant in fat tissue, rather than on glandular tissue These include the following observations: 1) GH is known to stimulate IGF-1 production in fat and IGF-1 mRNA was found to be present in white adipose tissue of the rat in concentrations equal to those in liver (Peter et al 1993); 2) GH regulated not only IGF-1 mRNA but also IGF-1 protein synthesis and IGF binding proteins (IGFBPs)-2, -3, -4 and –5 in rat white adipose tissue (Peter et al 1993); 3) Injection of bGH into sexually immature, hypophysectomized, and oophorectomized female rats resulted in an equal effect of bGH on IGF-1 mRNA production in both gland-free and gland-rich mammary fat pads (Walden et al 1998) If the major effect of bGH was on the glands themselves, one might expect greater production of IGF-1 mRNA in the gland-rich section of the mammary gland; 4) In the female rat, the glandular epithelial compartment in the

“gland-rich” areas comprises a relatively small fraction of the whole glands, with stromal tissues accounting for the majority Therefore, all or some of the observed bGH-induced increase in IGF-1 mRNA might be taking place in the connective tissue (Walden et al 1998); 5) The effect of bGH treatment alone or in combination with

estrogen in vivo in these oophorectomized and hypophysectomized female rats resulted in increased IGF-1 mRNA production in the subscapular fat pad region versus the gland-free mammary fat pad region (Walden et al 1998) Further, in contrast to the mammary fat pad region, the subscapular fat pad region demonstrated

no synergism between bGH and E2

In contrast, treatment with either bGH or hGH increased ER activity at both the transcriptional and translational levels, whereas PRL or E2 treatments alone exerted

no effects (Feldman et al 1999) Further, the effect of GH occurred primarily in the mammary stromal compartment, as no differences in GH stimulation of ER between gland-free and gland-containing mammary fat pads were observed Further, analysis

of mammary gland ER demonstrated that while ER was present in the epithelial cells

of non-treated rats, only GH-treated animals possessed ER clearly visible in both glandular and fat cells of the tissue, and treatment of rats with IGF-1 did not result in increases in ER, nor in the staining of the fat cell nuclei for ER (Feldman et al 1999)

2.6 GH and mammary carcinoma

2.6.1 GH/IGF-1 axis and mammary carcinoma

The GH/IGF-1 axis plays an important role in regulating mammary gland during postnatal development (Ruan et al 1992) and lactation period (Goffin et al 1996) However, it has been shown that GH/IGF-1 axis is also involved in regulating epithelial cell proliferation and cell turnover in the mature breast

In rodents with energy-restricted diets, known to suppress IGF-1 (Thissen et al 1994), cancer incidence decreases (Ruggeri et al 1989) Also, there is a positive correlation between birth weight and breast cancer risk (Michels et al 1996), knowing that low IGF-1 level has been correlated with low birth weight (Osorio et al 1996) And height, positively associated with breast cancer risk in most epidemiologic studies (Hunter and Willett 1993; Vatten and Kvinnsland 1992), is positively correlated with IGF-1 levels (Juul et al 1995) Furthermore, transgenic mice overexpressing GH and with consequently high levels of circulating IGF-1 exhibit high frequencies of breast cancer (Tornell et al 1992; Tornell et al 1991) Primates exposed to exogenously applied GH also show histological evidence of mammary gland epithelial cell hyperplasia (Ng et al 1997), changes that have also been associated with increased breast cancer risk in human (Marshall et al 1997)

Breast cancer growth is also influenced by the host GH/IGF-1 axis A comparison was made between the growth of human MCF-7 in control mice to that in homozygous mice for a missense GH-releasing hormone receptor, which resulting in suppressed level of GH and IGF-1 (Cao et al 1995) Breast cancer growth was significantly reduced (by 50%) comparing to control mice (Yang et al 1996) It strongly suggests the pivotal role of host GH/IGF-1 axis on breast cancer behavior

Furthermore, tamoxifen is shown to be able to reduce circulating IGF-1 level (Friedl

et al 1993; Pollak et al 1990) The mechanism by which tamoxifen reduced IGF-1 levels by suppression of pulsatile GH secretion by antiestrogens and has been

documented in both in vivo and in vitro models (Malaab et al 1992; Tannenbaum et

al 1992) Estrogen receptor levels are positively correlated with IGF-1 receptor levels

in human breast cancers (Pekonen et al 1988), therefore neoplasms predicted to be most responsive to antiestrogens on the basis of high ER level might also be most dependent on stimulation by exogenous IGF-1

2.6.2 IGF-independent effect of GH in mammary carcinoma cells

hGH mRNA identical to pituitary hGH is also expressed bynormal mammary tissue and by benign and malignant human mammary tumor, immunoreactive hGH being restricted to epithelial cells (Mol et al 1995a) A recent study showed increased expression of the hGH gene in proliferative disorders of the mammary gland (Raccurt

et al 2002) The pituitary and mammary gland GH gene transcripts originatefrom the

same transcription start site but are regulated differentially,since mammary gland GH gene transcription does not require Pit-1 (Lantinga-van Leeuwen et al 1999) GH receptor mRNA and protein have also been detected inthe mammary gland epithelia

of murine and rabbit (Ilkbahar et al 1995; Jammes et al 1991; Lincoln et al 1990), bovine (Glimm et al 1990), and human species (Sobrier et al 1993) Both endocrine

GH and autocrine/paracrine-producedGH therefore possess the capacity to exert a direct effect onthe development and differentiation of mammary epithelia in vitro (Plaut et al 1993) and in vivo (Feldman et al 1993)

Kaulsay (Kaulsay et al 1999) generated a model system to study the role of autocrine-produced hGH in mammarycarcinoma by stable transfection of either the hGH gene or a translation-deficient hGH gene into a human mammary carcinoma (MCF-7) cell line (Kaulsay et al 1999) The autocrinehGH-producing cells display a marked insulin-like growth factor-1 (IGF-1)-independent increase in cell number in both serum-free and serum-containing conditions The increase in mammary carcinoma cell number as a consequenceof autocrine production of hGH was a result

of both increased mitogenesis and decreased apoptosis and was dependent on the activities ofboth p44/42 and p38 MAP kinases (Kaulsay et al 2001) Also, autocrine hGH production resulted in enhancement of the rate of mammary carcinoma cell spreadingon a collagen substrate (Kaulsay et al 2000) All of the studied effects ofautocrine hGH on mammary carcinoma cell behavior are mediated via the hGH receptor (Kaulsay et al 2001) Thus, autocrine production of hGH by mammary

carcinoma cells may direct mammary carcinoma cell behavior in an independent pathway

IGF-1-Several genes mediating the pleiotropic effects of hGH on mammary carcinoma cell function have been identified It has been demonstrated that autocrine hGH increases the expression and activity of HOXA1, a powerful oncogene for human mammary epithelial cells (Zhang et al 2003) Forced expression of HOXA1 protected human mammary carcinoma cells from apoptotic cell death in a Bcl-2-dependent manner (Zhang et al 2003) CHOP mRNA and protein levels was increased by autocrine hGH in human mammary carcinoma cells (Mertani et al 2001) Increased expression

of CHOP offers dramatic protection from apoptotic cell death in mammary carcinoma cells stimulated with hGH (Mertani et al 2001) Autocrine hGH also increases cell proliferation by repression of PTGF-beta gene expression, which could promote cell cycle arrest and apoptosis (Graichen et al 2002)

Thus it is clear that autocrine hGH in mammary carcinoma cells produces a hyperproliferative state in an IGF-1 independent manner This might contribute negatively to the final clinical prognosis of breast cancer Further studies should be done to elucidate the role of autocrine hGH in mammary cell function

2.7 Balance between oxidants and antioxidants

2.7.1 Oxidative stress

Reactive oxygen species (ROS), including hydroxyl radicals (•OH), superoxide anions (O2•−) and hydrogen peroxide (H2O2), are produced as a consequence of aerobic respiration and substrate oxidation Small amounts of ROS are constantly generated in aerobic organisms in response to both external and internal stimuli (Hurst et al 1997; Jornot et al 1998; Mills et al 1998)

ROS may immediately react with cellular macromolecules and may thereby directly cause cell damage In the presence of transition metals, H2O2 can be reduced to generate •OH via the Fenton reaction (Halliwell and Gutteridge 1989):

H2O2 + Fe++ (Cu+) → Fe+++ (Cu++) + OH− + •OH

•OH can attack cell membranes by setting off free radical chain reactions in which free radicals are passed from one macromolecule to another Such a reaction cascade results in extensive damage of cell membranes and other cellular structures (Halliwell and Gutteridge 1990) Thereby, reactive oxygen intermediates may react with the polyunsaturated fatty acids and cholesterol present in cell membranes (Halliwell 1991; Halliwell and Gutteridge 1990) and initiate the formation of oxidized lipids, which finally leads to apoptosis (Christ et al 1993)

Reactive oxygen intermediates may also cause the accumulation of p53 and DNA damage which activates poly-ADP-ribosyle transferase (Clarke et al 1993) Both

processes are associated with apoptosis (Christ et al 1993) In the later case, the polymerization of ADP-ribose to proteins results in a rapid depletion of cellular NAD/NADPH pools, leading to depletion of ATP stores and cell death Thus redox modulation may play a prominent role in the regulation of cell death (Hockenbery et

al 1993; Kane et al 1993), and increased levels of reactive oxygen intermediates may result in apoptosis (Buttke and Sandstrom 1994; Greenspan and Aruoma 1994; Meyer

et al 1993) It appears that the concentration of the reactive compounds is important with respect to the induced biochemical effects, since it is well known that high concentrations of H2O2 may induce necrosis rather than apoptosis (Lennon et al 1991) Thus, the concentration of reactive compounds seems to determine the form of cell death that occurs (Wyllie 1987) Alternatively, an oxidative shift in the cellular redox state could alter the outcome of a stimulatory signal, resulting in cell death as opposed to proliferation (Zimmerman et al 1989)

As part of the activation sequence and as a way of self-protection, cells initiate various antioxidant mechanisms; for example, secreting antioxidant proteins such as catalase (Lipsky 1984), releasing cysteine for glutathione synthesis (Gmunder et al 1990), or inducing superoxide dismutase expression in T cells via the secretion of cytokines like interleukin-1 or tumor necrosis factor α (TNF-α) (Hirose et al 1993; Wong and Goeddel 1988) If antioxidant mechanisms are insufficient, antigenic activation may lead to T-cell anergy or apoptosis Thus, the reductive capacity of cells appears to be important as the type and the degree of oxidative stress in determining whether cells enter apoptotic pathways or overcome oxidative challenge

Antioxidant pathways are common in cells to establish a redox equilibrium which is necessary to maintain cell survival and health Besides antioxidant defense mechanisms of cells such as the reactive-oxygen scavenger enzymes superoxide dismutase and catalase, several cellular genes regulating apoptosis have been identified to be involved in antioxidant pathways (Buttke and Sandstrom 1994; Greenspan and Aruoma 1994) For example, the proto-oncogene Bcl-2 suppresses apoptosis induced by growth factor deprivation or viruses (Nunez et al 1990; Steller 1995; Thompson 1995; Vaux et al 1988) Although the mechanisms by which such a control is achieved are not clear, the available data indicate a role for Bcl-2 in an antioxidant pathway (Hockenbery et al 1993; Kane et al 1993) Bcl-2 interferes with the conversion of O2•− to peroxides and acts as a free radical scavenger which traps nonreactive free radicals Bcl-2 was found to inhibit lipid peroxidation and

accumulation of various reactive oxygen species in cells triggered to undergo apoptosis

In summary, the presence of reactive species activates several antioxidant pathways that are necessary to establish a balance between oxidants and antioxidants An oxidative shift may thus result not only from the excess production of reactive oxygen intermediates but also from a suppression of cellular antioxidants

2.7.3 Reactive oxygen species (ROS) and human diseases

Low levels of ROS are indispensable in many biochemical processes, including intracellular messaging in the cell differentiation and cell progression or the arrest of growth, apoptosis (Ghosh and Myers 1998), immunity (Yin et al 1995) , and defense against micro-organisms (Bae et al 1997; Lee et al 1998) In contrast, high doses and inadequate removal of ROS result in oxidative stress, which may cause severe metabolic malfunction and damage to biological macromolecules (Chopra and Wallace 1998; Czene et al 1997; Wojtaszek 1997)

An imbalanced production of ROS plays a role in the pathogenesis of a number of human diseases such as ischemia/reperfusion injury, atherosclerosis, neurodegenerative diseases, cancer, and allergy When antioxidants, free radical scavenging systems are overwhelmed, inflammation, hypersensitivity, and autoimmune conditions may result Inflammatory cells may also increase DNA

damage by activating pro-carcinogens to DNA-damaging species, e.g., neutrophils

can activate aromatic amines, aflatoxins, estrogens, phenols and polycyclic aromatic hydrocarbons by ROS-dependent mechanisms Cancer is also related to the effects of continuous damage over a life span by toxic oxygen (Chinery et al 1998; Torzewski

et al 1998; Wiseman and Halliwell 1996) ROS have been also implicated in many lung diseases, including asthma (Beasley et al 1991; Kurosawa et al 1993) and pneumonia (Dandyshev 1998) In addition, oxidative stress, superoxide production and an imbalance in antioxidant enzymes has been related with many other specific pathologies, as Alzeimer’s disease (Yan et al 1994), chronic granulomatous disease (Chang et al 1998), Downs syndrome (Odetti et al 1998), diabetic complications (Aguirre et al 1998; Hannon et al 1998), hepatitis (Cuzzocrea et al 1998),

rheumatoid arthritis (Aaseth et al 1998), Influenza virus (Peterhans 1997), cataract

(Spector et al 1997) and etc

The enzymatic and non-enzymatic antioxidant defenses include superoxide dismutase (SOD), glutathione peroxidase (GPX), catalase (CAT), glutamylcysteine synthetase (GCS), ascorbic acid (vitamin C), α-tocopherol (vitamin E), glutathione (GSH), β-carotene, and vitamin A The balance between both the activities and the intracellular levels of these antioxidants are essential for the survival of organisms and their health (Grazioli et al 1998; Lapenna et al 1998; Mao et al 1993; Sohal et al 1995)

2.8 Antioxidant enzymes

The ability to adapt to changing environmental stresses is a fundamental aspect of homeostatic responses in living organisms Adaptive responses to oxidative stress have been demonstrated in prokaryotic and eukaryotic systems, reflecting the ubiquitous nature of oxidative stress in aerobic organisms To minimize the detrimental effects of ROS, cells have developed antioxidant defense mechanisms that include various small molecule scavengers as well as the antioxidant enzymes (Halliwell 1999) In eukaryotes, in order to adapt to changes in levels of ROS under physiologic and pathologic conditions, cells are able to regulate and coordinate the levels of antioxidant enzymes gene expression as part of the defense system (Harris 1992; Ichikawa et al 1994; Quinlan et al 1994) Many extracellular stimuli, including various cytokines, growth factors, and hormones, can affect expression levels of antioxidant enzymes TNF-α, IL-1, lymphotoxin, leptin, estrogen, and dexamethasone (Brown-Borg et al 1999; Clerch et al 1991; Masuda et al 1988; McCormick et al 1991; Watson et al 1999; Wong et al 1996) have been reported to induce antioxidant enzymes gene expression in various tissues, and some of these stimuli may cause oxidative stress to cells (Schulze-Osthoff et al 1993) Direct oxidative conditions, however, also significantly increase expression and activity of the antioxidant enzymes The specific pattern of changes depends on both the nature

of the oxidant challenge and the tissue under investigation (Harris 1992; Ichikawa et

al 1994; Quinlan et al 1994; Rohrdanz and Kahl 1998; Shull et al 1991) Studies have shown that antioxidant enzyme genes are transcriptionally activated following oxidative stress (Cowan et al 1993; Ho et al 1991)