Studies of Lipoxygenase Function
Noel Patrick McCabe
Medical College of Ohio
2004
DEDICATION
This dissertation is dedicated to my loving wife Cari and my son Colin.
Without Cari’s support, none of this would have been possible. Around a year and 8 months ago, we were blessed with a beautiful baby boy, Colin, and let me tell you life can become pretty complicated when you have to completely change your views on life and your career goals to make room for someone you never knew you could love so much. Having a little one can also lead to much unforeseen stress in one’s personal life, and I couldn’t mentally cope without
Cari’s assistance.
I also would like to thank my parents, James and Sharon McCabe, for everything they have done for me throughout my life. Without their guidance growing up and their openness with regards to what interested me, I couldn’t have become the man that I am today. You never realize how important it is to let a child do their own thing and discover the error of their ways until you are grown up and faced with a child of your own.
ii ACKNOWLEDGEMENTS
I would like to thank Dr. Jankun and Dr. Skrzyczak-Jankun. Without them,
I never would have gotten to work on two very exciting, clinically relevant projects. Dr. Skrzyczak-Jankun provided the knowledge and know how of the X- ray crystallography work contained herein. X-ray analysis and data interpretation from such analysis was completed by her, first at the University of Toledo in the
Instrumentation Center then at the Medical College of Ohio in the Department of
Urology. Her knowledge of the field is remarkable. I also would like to thank
Kanjing Zhou, not only was he instrumental in structural refinement, but he also provided some comedic relief by falling asleep periodically while wearing stereoscopic glasses.
I would like to express my deepest gratitude to Dr. Steven Selman. Your enthusiasm for research for the sake of knowledge is refreshing. I have enjoyed our discussions on life, science, and career choices. Please look me up when you come to Cleveland. I would also like to thank you, as well as the other members of the lab, both permanent (Rick) and transient (Ranko and Kurt), for your friendship.
iii TABLE OF CONTENTS
List Page Title Page i
Dedication ii
Acknowledgements iii
Table of Contents iv
Introduction 1
Literature 9
Manuscript 1: “Platelet 12-Lipoxygenase Overexpression 64
by PC-3 Prostate Cancer Cells is Associated with
Enhanced Production of Vascular Endothelial
Growth Factor”
Manuscript 2: “Curcumin inhibits lipoxygenase by binding 96
to its central cavity: theoretical and X-ray evidence”
Manuscript 3: “Structure of curcumin in complex with 123
lipoxygenase and its relation to cancer”
Discussion/Summary 152
Bibliography 158
Abstract 208
iv INTRODUCTION
Prostate cancer is the second most common cause of cancer-related death among American men. Figures published by the American Cancer Society show the gravity and increasing incidence of this disease among North American males. Prostate cancer is the most common malignancy in males and the incidence of disease increases dramatically with age. Current methods of disease detection include the digital rectal exam and quantitation of serum prostate specific antigen (PSA). Prostate specific antigen has proven to be a very valuable indicator of prostate cancer. However, PSA testing lacks specificity, i.e., increased circulating PSA can result from multiple conditions such as benign
prostate hyperplasia and prostatitis. While there is an inherent risk for all men to
develop prostate cancer, chances of developing this disease can be influenced
by multiple factors including, but not limited to, age, race and diet. It is believed that copious consumption of red meat and high-fat dairy foods, thus increasing the amount of animal fats ingested, can increase the chances of developing prostate cancer.
Early animal studies on tumorigenesis provide the first link that specific
dietary polyunsaturated fatty acids promote cancer development (Broitman
1977). Later work determined that these polyunsaturated fatty acids must be
metabolized to their oxygenated products to augment tumorigenesis (Bull et al.
1984; Setty et al. 1987; Glasgow et al. 1992). One such fatty acid, arachidonic
acid, is an essential fatty acid obtained either through diet or by enzymatic
1 conversion of other fatty acids. Enzymatic oxidation of arachidonic acid occurs
through several families of enzymes: cycloxygenases, lipoxygenases, and
cytochrome P450 oxygenases. Metabolism of arachidonic acid by any of these
groups of enzymes results in the generation of eicosanoids. Eicosanoids derived
from arachidonic acid metabolism are known to play significant roles in a
multitude of physiological and pathological conditions. The majority of research
effort is dedicated to conditions related to the production of prostaglandins and other cyclooxygenase metabolites. However, considerable data have accumulated implicating the importance of lipoxygenase products of arachidonic acid in the initiation and progression of numerous pathological disorders, most notably cancer.
Hamberg and Samuelson (1974) were the first to show that a
lipoxygenase was present in human platelets and was capable of metabolizing
arachidonic acid, resulting in the production of 12(S)-hydroxy-5,8,10,14-
eicosatetraenoic acid (12(S)-HETE). This lipoxygenase was capable of inserting
oxygen at carbon 12 in arachidonic acid, therefore, it was dubbed platelet-type
12-lipoxygenase (P12-LOX). This was the first documented case concerning the
existence of lipoxygenases in animals. It wasn’t until 1990 that P12-LOX
expression was shown in human erythrolukemia cells (HEL) (Funk et al. 1990)
disproving the belief that this enzyme was specific to platelets. Shortly
thereafter, it was shown that P12-LOX was expressed in various tumor tissues
and cells including human epidermoid carcinoma A431 (Chang, Liu et al. 1993),
2 Lewis lung 3LL, B16a melanoma (Chen et al. 1994) and prostate (Gao et al.
1995).
Gao et al. (1995) found that P12-LOX mRNA expression was significantly higher in prostate adenocarcinoma tissue compared to matched normal prostate epithelium, and that this increased expression correlated with advanced stage and grade adenocarcinoma. In this pivotal study, tissue from over 130 patients were examined with 38% demonstrating elevated P12-LOX mRNA in cancerous tissue compared to normal matched tissue. The level of elevation of P12-LOX gene expression among high grade prostatic adenocarcinomas compared to that of low and intermediate proved to be statistically significant. This study suggested an association between prostate cancer progression and elevated expression of P12-LOX. Because of this early work, the significance of P12-LOX in prostate cancer became of utmost interest.
Prostate cancer progression, as with other forms of cancer, is dependent upon a process called angiogenesis (Folkman et al. 1989). Angiogenesis, or extension of existing vasculature, is a complex process which depends upon the production and release, in this case from tumor cells, of angiogenic proteins such as basic fibroblast growth factor (bFGF) and vascular endothelial growth factor
(VEGF), as well as specific arachidonate derived eicosanoids (Nie et al. 2000b).
The expression of VEGF in prostate cancer has been shown to be dependent, at least initially, upon the presence of androgen (Joseph and Isaacs 1997; Stewart et al. 2001). Treatment of prostate carcinoma by androgen deprivation, thus effectively reducing VEGF production, will ultimately lead to tumor regression
3 (Moon et al. 1997). This effect is due at least in part by limiting the expansion of
tumor induced vascularization. Unfortunately, prostate tumor resurgence can
occur as a result of androgen independent tumor cell propagation and
recommencement of the angiogenic process.
Platelet 12-LOX is believed to play multiple roles in the progression of
prostate cancer, in part by promotion of angiogenesis. Expression of P12-LOX
has been confirmed in multiple prostate cancer cell lines including PC-3, DU145,
and LnCAP (Timar et al. 2000). Stable overexpression of P12-LOX in PC-3 cells
was able to promote accelerated growth following subcutaneous injection of nude mice, a condition resulting from increased tumor vascularization (Nie et al. 1998).
Overexpression of P12-LOX also leads to constitutive activation of the transcription factor NF-κB (Kandouz et al. 2003), a factor previously linked to induction of the angiogenic protein VEGF (Huang et al. 2000). Lipoxygenase activity can also lead to the production of reactive oxygen species (Roy et al.
1994), which can in turn lead to an increase in VEGF gene expression (Kuroki et al. 1996). More definitively, augmentation of VEGF expression has been shown to occur in human smooth vascular muscle cells following exposure to 12(S)-
HETE (Natarajan et al. 1997a). Vascular endothelial growth factor mRNA is increased in prostate tumors compared to normal prostate tissue, and VEGF inhibition suppresses primary prostatic tumor growth and metastasis
(Kirschenbaum et al. 1997; Melnyk et al. 1999). In addition, VEGF expression levels are increased in prostate cancer cells of higher metastatic potential (Ferrer et al. 1997; Jackson et al. 1997) and this can be attributed to an increased ability
4 of these cells to stimulate angiogenesis, thus providing access to the circulatory
system for tumor dissemination.
The metastatic cascade (reviewed in Dailey and Imming 1999) consists of
intricate interdependent steps which rely on the presence of local vasculature, as discussed above. By a process dubbed intravasation, tumor cells gain access to the blood stream and traverse to secondary locations within the body. Upon arriving at a distant site, tumor cells exit the vasculature, a process known as extravasation, by inducing vascular endothelial cell retraction, promote dissolution of the basement membrane, and invade local tissue. Each step in the metastatic process is regulated by various growth factors, cytokines and bioactive lipid derivatives of arachidonic acid. Multiple steps of the metastatic process are modulated by 12(S)-HETE and a review of 12(S)-HETE’s involvement has recently been written (Honn et al. 1994a). In fact, 12(S)-HETE levels have been shown to correlate with the metastatic potential of several tumor cell lines (Tang and Honn 1994). Significant data have accumulated that display the pleiotropic effects of P12-LOX and 12(S)-HETE in processes associated with the pathology of tumor progression and dissemination.
The focus of the following work is to increase our understanding of the roles that lipoxygenase plays in prostate cancer progression. Given the significant amount of existing data regarding P12-LOX and 12(S)-HETE in tumor related processes, it is evident that P12-LOX inhibition is of clinical relevance.
The research contained herein addresses two specific aims:
5 1) Determine a link in prostate cancer between arachidonic acid
metabolism via the P12-LOX pathway, VEGF expression, and angiogenesis.
Defining a link between P12-LOX and VEGF in prostate carcinoma will serve to
further our understanding of the arachidonic acid cascade and the molecular
mechanisms of prostate cancer related angiogenesis.
2) Characterization of the interaction of the non-specific lipoxygenase
inhibitor, curcumin, with soybean lipoxygenase-3 using three-dimensional
modeling and X-ray crystallography. By characterizing inhibitors of lipoxygenase
and determining the mechanism of such inhibition, it may be possible to design
novel inhibitors that spatially fill the active site of lipoxygenase more efficiently.
This thesis encompasses work which has been completed in the last 5
years regarding lipoxygenase inhibition and the characterization of a novel
function of human platelet-type 12-lipoxygenase in prostate cancer related
angiogenesis.
Manuscript 1 describes work pertaining to the function of human platelet-
type 12-lipoxygenase in prostate carcinoma. Specifically, the effects of P12-LOX
overexpression in PC-3 prostate cancer cells are examined. Signal transduction
stimulated in PC-3 and DU145 prostate cancer cells following exposure to 12(S)-
HETE is also addressed. Additionally, the role this signaling plays in modulating the expression of the proangiogenic factor VEGF is investigated.
6 Manuscript 2 presents preliminary data regarding the interactions of
curcumin with soybean lipoxygenase-3 (soy LOX-3). This compound has been
found to act as a general inhibitor of lipoxygenase activity. We chose to
investigate, by means of theoretical modeling and x-ray crystallography, the associations of curcumin with lipoxygenase in an attempt to better understand curcumin/lipoxygenase interactions.
It is generally considered that all lipoxygenases are structurally similar despite differences in protein size. Soy LOX-3, in lieu of highly purified human
P12-LOX protein and P12-LOX structural data, was chosen as a general model of lipoxygenase structure for investigating the mechanism of curcumin as an inhibitor of lipoxygenases. By using theoretical as well as X-ray analysis of the interaction of lipoxygenase and curcumin, the spatial localization of curcumin, or more specifically a degradation product thereof, in the active site of lipoxygenase was determined, providing insight as to why curcumin can inhibit lipoxygenase activity.
Manuscript 3 is a continuation of manuscript 2, containing structural refinement and mechanistic characterization of the interaction between soy LOX-
3 and curcumin as well as an in depth examination of results drawn from the entire study. As it turns out, inhibition of lipoxygenases by curcumin occurs in a non-competitive manner. Curcumin occupies the catalytic site of lipoxygenase and blocks further access. As soy LOX-3 is a good example of general lipoxygenase structure and function, it can help in the realization of spatial
7 constraints of lipoxygenase active sites and what types of substrate moieties are
best for filling such space. Understanding characteristics for optimal inhibition will
make it possible to generate derivatives of curcumin that possess enhanced
specificity for individual lipoxygenase isoforms. Novel curcumin derivatives
displaying isoform specificity may have therapeutic relevance in the treatment of diseases such as prostate cancer where lipoxygenases play a role.
8 LITERATURE
Arachidonic Acid Cascade
The metabolism of arachidonic acid into biolipid mediators is the focus of
intense investigation. Arachidonic acid (C20:4) is an essential fatty acid which
can be obtained through diet or by conversion of specific C18 fatty acids through
processes of desaturation and elongation. As a major mammalian cell
membrane component, arachidonic acid can be enzymatically cleaved from the
membrane glycerophospholipid pool through the action of phospholipase A2 or
the combined action of various other phosholipases (Needleman et al. 1986).
This response can be the result of exposure to cytokine, growth factor, and/or
hormonal stimuli. Once released, arachidonic acid can be oxidized by cyclooxygenase, lipoxygenase, or P450 epoxygenase family members to form a host of eicosanoids (Figure 1). Eicosanoid formation is generally considered to be regulated by the availability of substrate and the activation state of the respective oxygenase. Translocation of 5-lipoxygenase (5-LOX) to the nuclear membrane is a calcium dependent process (Rouzer and Samuelsson 1987) and enzyme activity is dependent upon an activating protein, 5-LOX activating protein
(FLAP) (Rouzer et al. 1985). Calcium and the hydroperoxide tone of the cell are key components for regulation of 5-LOX activity (Ochi et al. 1983; Weitzel and
Wendel 1993). The 12-LOX family of proteins exhibit calcium dependent translocation to membranes; however, activity is calcium independent but
9 FIGURE 1. ARACHIDONIC ACID METABOLISM
Stimulus Phospholipids
Phospholipase activation Glycerol
Cyclooxygenase Cytochrome P450 Arachidonic acid
Lipoxygenase
Prostaglandins Hydroperoxyeicosatetraenoic Epoxyeicosatrienoic Thromboxanes acids (HpETEs) acids Prostacyclins Hydroxyeicosatetraenoic acids dihydroxyacids Prostamides (HETEs) Leukotrienes Lipoxins Hepoxilins
10 requires a substantial increase in the overall hydroperoxide tone (priming effect)
for maximal activity (Vanderhoek et al. 1982; Walstra et al. 1987). Activation of
15-lipoxygenase (15-LOX) is dependent upon the hydroperoxide tone as well
(Vanderhoek et al. 1982).
Enzymatic oxidation of arachidonic acid by cyclooxygenases results in the
formation of an endoperoxide precursor, prostaglandin H2 (PGH2), a necessary building block for the production of prostacyclins, thromboxanes, and prostaglandins. Cyclooxygenases require the presence of a heme moiety for catalytic activity (Roth et al. 1981) in contrast to the non-heme active site containing a single Fe ion found in lipoxygenases.
Insertion of molecular oxygen into arachidonic acid catalyzed by lipoxygenases results in the production of regioisomeric cis/trans conjugated hydroxyeicosatetraenoic acids (HETEs), lipoxins, hepoxilins, and leukotrienes.
Of the lipoxygenases found in humans, 5-LOX has received the most attention.
Arachidonic acid is metabolized by 5-LOX to form 5(S)-HpETE which is converted into leukotrienes or 5(S)-HETE. Leukotrienes are generated via 5- lipoxygenase activity in eosinophils, neutrophils, mast cells, etc. and play a key role in airway inflammation and vasoconstriction. Several cancer cell lines including lung (Avis et al. 1996) and prostate (Ghosh and Myers 1997) have been shown to express 5-LOX mRNA and protein overexpression has been confirmed in prostate tissue (Gupta et al. 2001). In prostate cancer, 15- lipoxygenase (15-LOX), depending on the isoform, can have increased expression (15-LOX-1) (Kelavkar et al. 2001) or decreased expression (15-LOX-
11 2) (Shappell et al. 1999) compared to normal tissue. The platelet isoform of 12-
lipoxygenase, which will be covered in depth in the following pages, is known to be a key mediator in numerous pathological processes.
Arachidonic acid oxygenation by P450 epoxygenases results in the
formation of various HETEs and epoxyeicosatrienoic acids (EETs). However,
there is a lack of data regarding the role of P450 generated eicosanoids in the
process of tumor biology.
Lipoxygenases
Lipoxygenase enzymes can be found in a wide variety of plant and animal
tissues and exist as a multigene family of dioxygenase enzymes. Lipoxygenases
contain a non-heme iron in the catalytic center that is responsible for the
positional and stereo-specific dioxygenation of select carbon atoms in
polyunsaturated fatty acids containing a 1,4-pentadiene motif, resulting in the
production of corresponding hydroperoxy derivatives (Brash 1999). Depending
upon the substrate, either 18C fatty acids or 20C fatty acids, enzymatic
metabolism by lipoxygenases results in the production of
hydroperoxydecadienoic acids (HpODEs) or hydroperoxyeicosatetraenoic acids
(HpETEs), respectively. These products then can be converted into various
other eicosanoids.
12 Classification
Many species of lipoxygenase exist and expression of these proteins is
known to encompass a large contingent of plant (Shibata and Axelrod 1995),
fungi (Gerwick 1994), and animal species (Brash 1999). Recently, a
lipoxygenase was found to be of bacterial origin (Porta and Rocha-Sosa 2001).
Classification of plant isoforms of lipoxygenases (linoleate: oxygen
oxidoreductase) is based upon positional specificity of oxygen insertion into
linoleic acid (C18). Accordingly, a plant lipoxygenase that inserts molecular
oxygen into linoleic acid at C13 is a 13-lipoxygenase. For soybean
lipoxygenases, nomenclature is numerical, i.e., soybean lipoxygenase -1, -2, or -
3, according to their chronologic discovery. However, they are classified also by
their positional specificity. In contrast, mammalian lipoxygenases are named according to the position of molecular oxygen insertion along the carbon
backbone of arachidonic acid (C20), yielding 5-, 8-, 12-, and 15-lipoxygenases.
Thus, a plant 13-LOX, such as soybean lipoxygenase-1, is comparable to a
mammalian 15-LOX. The arachidonate based classification; however, fails to
account for the fact that most mammalian lipoxygenase isoforms can metabolize
18 carbon fatty acids such as linoleic acid in addition to the 20 carbon
arachidonic acid. Interestingly, some isoforms of 12- and 15-lipoxygenases can
produce a mixture of 12- and 15-lipoxygenation products of arachidonic acid
(Brash 1999). Further, lipoxygenases may also be classified based on their
stereospecificity (e.g., 12(S)-LOX or 12(R)-LOX). Thus, it becomes evident that
13 the positional specificity of lipoxygenases is not an absolute property, but is rather dependent upon the presence of solvent, pH, and substrate itself. Kuhn and Thiele (1999) have assembled a phylogenic tree of mammalian lipoxygenases (Figure 2) based upon substrate specificity, which categorizes mammalian lipoxygenases into four general groups: 1) the dual specificity
12S/15S-lipoxygenases, 2) platelet 12-lipoxygenases, 3) 5-lipoxygenases, and 4) epidermal lipoxygenases.
Structure
Lipoxygenases exist as single chain polypeptide proteins with molecular
masses ranging from ~72-81 kDa in animals and ~94-103 kDa in plants (Brash
1999). As part of a multigene family, lipoxygenases can have considerable
sequence homology. Overall sequence identity of known isoforms ranges from
25-40% homology, with close functional homologues exhibiting 70-95% similarity.
Species have been purified from various plant and animal tissues; however,
isoforms of soybean origin have been investigated to the greatest extent due to
natural abundance. Each mole of lipoxygenase protein contains one gram-atom
of non-heme iron (Draheim et al. 1989), which is essential for oxygenation of substrate (de Groot et al. 1975). Sequences have been reported for approximately 50 different lipoxygenases that range in length from 923 residues
14 FIGURE 2. CLASSIFICATION OF MAMMALIAN LIPOXYGENASES
(From Kuhn and Thiele 1999)
15 (rice lipoxygenase-2) to 661 residues (rabbit reticulocyte 15-lipoxygenase).
Mammalian sequences average 150-200 residues shorter than their plant
counterparts (Prigge et al. 1997).
Although the first lipoxygenase was described over 60 yr ago in dry
soybeans, it was not until the late 1980s that the emergence of molecular biology
techniques permitted determination of the primary structure of lipoxygenases
from both plants and animals. However, in the early to mid 1990s, it became
evident that although the primary structure of many lipoxygenases across
species had been determined and cloned, little was known regarding the three-
dimensional structure of these proteins and the role they played in physiological
and pathological processes. Theorell et al. (1947) reported the first crystalline lipoxygenase in 1947, but the three-dimensional crystal structure of that lipoxygenase (soybean lipoxygenase-1) was not reported until 1993 (Boyington et al. 1993). Thus soybean lipoxygenase-1 (soy LOX-1) remained the sole X-ray crystallographic representation of lipoxygenase structure until 1997 when
Skrzypczak-Jankun et al. (1997) and Gillmor et al. (1997) reported structures for
another soybean lipoxygenase, soy LOX-3, and a mammalian lipoxygenase,
rabbit reticulocyte 15-lipoxygenase, respectively. The soybean lipoxygenases
soy LOX-1 and soy LOX-3 are classical model systems for understanding
structure and functional properties of all lipoxygenases. For the purposes of this
introduction to lipoxygenase structure, soy LOX-1 will be used unless otherwise
noted, as its structure has been studied more extensively with relation to
structure and function. Due to the similarity of lipoxygenase sequences across
16 species, it is thought that many of the characteristics observed for soy LOX-1 will
hold true for other lipoxygenases.
Topology and Domain Structures
Crystallographic data has led to the realization that lipoxygenase enzymes
consist of two domains: 1) a small β-barrel domain at the N-terminus and 2) a
large, predominantly helical C-terminal catalytic domain. Two separate soy LOX-
1 crystal structures (PDB entries 2SBL at 2.6Å and 1YGE at 1.4Å) and one soy
LOX-3 structure (PDB entry 1LNH), when compared, have almost identical three-
dimensional topologies despite significant differences in amino acid sequence
between the two isoforms. Soy LOX-1 contains 839 residues whereas soy LOX-
3 has 857 residues with 7 amino acid deletions, 25 positional insertions, and 224
substitutions resulting in 72% identity (Skrzypczak-Jankun et al. 1997). A
considerably large quantity of these deletions and insertions reside on the
surface of the protein or in segments with no discernable secondary structure. A comparison between soy LOX-3 and rabbit reticulocyte 15-lipoxygenase shows an overall similar topology between plant and mammalian species as well (Figure
3).
The β-barrel domain consists of 146 residues forming eight anti-parallel β- strands, the interior or which contains tightly packed, mostly aromatic hydrophobic side chains. The amino terminal β-barrel domain of soybean
17 FIGURE 3. TOPOLOGY OF SOY LOX-3 AND RABBIT 15-LOX
Blue- soy LOX-3; Yellow- rabbit 15-LOX (From Manuscript 2)
18 lipoxygenases is significantly larger than that of rabbit 15-lipoxygenase. In fact, the β-barrel domain of rabbit 15-lipoxygenase has more in common, i.e., sequence homology, size and structure, to the carboxyl terminal β-barrel domain of mammalian lipases than to the equivalent structure of soybean lipoxygenases
(Gillmor et al. 1997). This β-barrel domain remains separate from the catalytic domain and makes only loose contacts. The large carboxy terminal domain is composed of 693 amino acids consisting of 20 (1YGE) or 23 helices (2SBL) (22 for soy LOX-3) and two anti-parallel β-sheets (one for rabbit).
Iron Coordination and Internal Cavities
Known lipoxygenase sequences, when aligned, reveal the conservation of a histidine motif consisting of His-(X)4-His-(X)4-His-(X)17-His-(X)8-His (Honn et al.
1994a). Two of these His residues have been shown to act directly as ligands for
Fe2+ by X-ray crystallographic studies (Boyington et al. 1993; Minor et al. 1996).
In fact, mutations of proposed iron ligand histidine residues of porcine leukocyte
12-lipoxygenase resulted in a complete loss of iron binding and enzymatic
activity, demonstrating that the specific histidine residues are essential for
enzymatic activity (Suzuki et al. 1994). The isoleucine at the carboxyl terminus is
conserved among all lipoxygenases, except rat leukocyte 5-LOX (Ile→Val) (Sigal
et al. 1988), and acts as a Fe2+ ligand as well. The iron atom is believed to reside at the center of a distorted bipyramidal octahedron comprised of three histidine residues, an Asn residue (Asn→His in rabbit 15-lipoxygenase), the C-
19 terminal isoleucine and a solvent water molecule or some other small molecule
such as a substrate or inhibitor. The description of the shape of the iron
coordination group varies depending on the ionization state of iron or the
presence of some sort of ligand.
Soybean lipoxygenase-1 has been shown by crystallography to contain
two cavities, each along side the iron center, which have the ability to pass
substrate from the surface of the protein to the catalytic center. The first cavity is
funnel shaped and connects to the surface by a 18Å long tunnel composed of 29,
predominantly hydrophobic residues. The second cavity is presumed to be the
site of enzymatic catalysis, although proposed fatty acid entry and exit sites are
variable depending on the structure used (2SBL or 1YGE). A bend in the cavity
of more than 90° occurs close to the iron atom, allowing the substrate to come
into close proximity to the iron. This cavity is encapsulated by mostly
hydrophobic and neutral residues and connects to the surface by way of a long
(~40Å), narrow channel (Boyington et al. 1993). Soybean lipoxygenase-3
contains a third cavity connecting the surface of the protein at the β- barrel/catalytic domain interface to the iron center and may act as passage for substrate/product entry or exit (Skrzypczak-Jankun et al. 1997). This may explain the lack of rigid regiospecificity of soy LOX-3 compared to soy LOX-1 resulting in a mixture of products. The structure for rabbit 15-lipoxygenase shows only one boot shaped cavity (Borngraber et al. 1999).
20 Mechanism
The non-heme iron cofactor present in lipoxygenases is known to be
absolutely essential for enzymatic activity and has been shown to be high spin in
soybean lipoxygenases (Gaffney 1996). As shown by electron magnetic
resonance, two oxidation states are possible with Fe3+ designating the active form of the enzyme and Fe2+ the inactive state (Gaffney 1996). The generalized
mechanism of lipoxygenases can be summed up by two steps, substrate
activation and oxygen addition. There are two proposed mechanisms for
lipoxygenase catalyzed fatty acid oxygenation (summarized in Nelson et al.
1990).
Both schemes (Figure 4) begin with iron in the active form (Fe3+). In scheme I, the general reaction can be expanded into a more precise three step radical dependant process as shown in Figure 4a: 1) The 1,4-diene component of the fatty acid is oxidized by Fe3+, resulting in hydrogen abstraction, to form a
pentadienyl radical. Along with this regioselective step there is also
enantiospecificity, or specificity for the pro-S or pro-R hydrogen of the diene
component. Step one appears to be the rate limiting factor of the lipoxygenase
reaction (Rickert and Klinman 1999). 2) The pentadienyl radical will undergo
rearrangement towards either the methyl end or the carboxyl end (+2 or -2
rearrangement). 3) This radical then reacts with molecular oxygen to form a
peroxy radical which subsequently reoxidizes Fe2+ to produce the resultant
hydroperoxy product. This mechanism is supported by electron paramagnetic
21 FIGURE 4. PROPOSED MECHANISMS OF LIPOXYGENASE ACTION
2+ 2+ Fe Fe • O OO 2
a. +H -H 3+ 3+ Fe Fe HOO
3+ 3+ Fe Fe HOO
b. -H +H
3+ 3+ Fe Fe OO
O 2
(From Prigge et al. 1997)
22 resonance (EPR) spectroscopy (Nelson et al. 1990; Nelson et al. 1994), where solutions of lipoxygenase and substrate produced detectable amounts of a peroxy radical signal. In the crystallographic structures determined thus far, there is no apparent specific binding site for oxygen leading to an assumption that the process is diffusion dependent. Scheme II (Figure 4b), as proposed by
Corey and Nagata (1987), entails 1) Fe3+ assisted diene deprotonation by direct bond formation with the substrate, followed by 2) insertion of O2 into the Fe-C bond, and 3) cleavage of the same bond. There is, however, a lack of proof regarding the formation of an organoiron intermediate. Absolute details about the mechanism of lipoxygenases remain a matter of discussion.
Positional Specificity
At a time when there was no structural data, designer fatty acids were used to investigate the positional specificity of lipoxygenases. Isomers of arachidonic acid containing differential placement of the double bond system suggested that oxygen consumption and positional specificity of lipoxygenases depended upon the location of the bisallylic methylene in relation to the methyl end of the fatty acid (Bryant et al. 1982; Kuhn et al. 1990). Later, as structural data began to emerge, molecular modeling was used to assist in such work.
Modeling of several lipoxygenases based on the known structures of soybean lipoxygenase-1 and rabbit 15-lipoxygenase as well as multiple sequence
23 alignments have helped shed some light upon the question of positional specificity.
Modeling and sequence comparisons of the human 12- and 15- lipoxygenase enzymes has led to suggestions as to which residues are required for determination of the positional specificity. Using the crystalline structure of rabbit 15-lipoxygenase, it was noted that the base of the boot shaped substrate binding pocket was lined by the side chains of Phe353, Ile418, Met419, and
Ile593 and the volume of the pocket could be important for positional specificity.
It is predicted that 12-lipoxygenase has a slightly larger substrate binding pocket compared to 15-lipoxygenase, whereas the pocket of 5-lipoxygenase is predicted to be 20% larger.
To test whether 12/15-positional specificity was due to conserved differences in residues between 12- and 15-lipoxygenases, mutagenesis studies were carried out. By replacing the Ile418 and Met419 amino acid residues at the bottom of the substrate binding pocket of rabbit 15-lipoxygenase with the smaller counterparts found in 12-lipoxygenase, the pocket size was increased resulting in amplification of 12-lipoxygenase product formation (Sloane et al. 1995). In addition, by mutating these residues in 12-lipoxygenaseto those found in 15- lipoxygenase, it was noted that an increase in 15-lipoxygenase product formation occurred. Further, an Ala-scan mutagenesis of the four amino acids lining the pocket of rabbit 15-LOX revealed that Ile418 and Ile593 are the most important residues of 15-LOX for determining positional specificity. Isoleucine593 was shown to be most important in 15-LOX regiospecificity with Ile418 as a
24 secondary positional specificity determinant and Met419 of minimal importance
(Borngraber et al. 1999). However, in the same study, mutational changes resulting in maximal cavity enlargement for the purpose of engineering 5- lipoxygenation resulted in severe catalytic impairment, possibly due to over enlargement and reduced proper alignment of the substrate. A converse study involving mutation of 5-LOX into a 15-LOX was, however, successful (Schwarz et al. 2001). In this study, Schwarz et al. were able to gradually change a 5-LOX into a 15-LOX with intermediate multiple mutations resulting in 8-LOX product formation. One study has investigated the importance of several amino acids in positioning of fatty acid substrate in addition to those listed above (Gan et al.
1996). In this study, structural modeling of enzyme/substrate interaction of 15-
LOX showed the possibility of salt bridge formation between a positively charged residue, Arg403, and the carboxylate group of the substrate. Mutation of Arg403 to Leu resulted in a reduced reaction rate and reduced substrate affinity. These results, however, were different for 5-LOX, where a similar mutation resulted in negligible alteration of enzymatic properties. This suggests that different polar residues may be responsible for substrate alignment in different lipoxygenase species.
Despite the lack of structural data regarding lipoxygenase/substrate interaction, mutagenesis studies suggest that the methyl end of arachidonic acid enters the cavity first where 12- and 15-lipoxygenases are concerned (Prigge et al. 1996). The 15-LOX has a shallow substrate binding cavity compared to that of 12-lipoxygenase. The methyl group of arachidonic acid, upon entering 12-
25 LOXs, is permitted to slide deeper into the pocket, believed to 5% larger according to molecular modeling (Gillmor et al. 1997), thus moving the methyl end further from the catalytic iron atom. Altered alignment of a ‘methyl end first’ fatty acid in relation to the catalytic iron will not produce desired products from 5- and 8-lipoxygenases. For this to occur, arachidonic acid must enter in the reverse orientation, with the carboxyl group leading the way. This will result in spatially equivalent reactions, i.e., 5-:15-LOX and 8-:12-LOX. It has also been suggested that for 5- and 8-lipoxygenases, despite mutational studies on 15-LOX to maximize the substrate binding cavity, that these lipoxygenases can also enter the cavity methyl end first and properly align with the iron for 5- and 8- lipoxygenation.
The positioning of the fatty acid substrate in the catalytic pocket as well as alignment with the iron cofactor appear to be crucial for the positional specificity of lipoxygenase catalyzed oxygenation.
12-Lipoxygenase Enzymes
Arachidonic acid metabolism by 12-lipoxygenase results in the production of 12(S)-Hydroperoxy-5Z,8Z,10E,14Z-EicosaTetraEnoic acid (12(S)-HpETE) which then can be reduced by a glutathione dependent peroxidase to 12(S)-
HETE or undergo isomerization to form hepoxilins A3 and B3 (Pace-Asciak
1993). Three separate isoforms of 12-lipoxygenase have been characterized
26 thus far, each named according to the type of cell in which they were first identified: leukocyte, platelet, and epidermal type. Each of these proteins originates from different transcripts and differs in substrate specificity, regioselectivity, and product profile.
The substrate specificity of the leukocyte 12-lipoxygenase is rather broad, as shown in Table І when compared to that of the platelet-type and epidermal- type 12-lipoxygenases. The leukocyte-type lipoxygenases can oxygenate C18 and C20 fatty acids as well as complex substrates such as low density lipoproteins (LDLs) and phospholipids residing in biomembranes (Takahashi et al. 1993a; Funk 1996; Yamamoto et al. 1997). Human P12-LOX, in contrast, prefers arachidonic acid (C20) as a substrate and is almost inactive with other fatty acids (Yamamoto et al. 1997). Mouse P12-LOX is not as selective as human P12-LOX when it comes to substrates, as linoleic acid oxygenation by murine P12-LOX has been noted recently (Burger et al. 2000). Epidermal 12- lipoxygenase is active with C18 methyl esters and exhibits very low activity towards other free C18 and C20 fatty acids (Siebert et al. 2001). These enzymes have been found in tissues such as porcine (Yoshimoto et al. 1982, 1990a) and human leukocytes (Yoshimoto et al. 1990b), rat brain (Watanabe et al. 1993), human platelets (Hamberg and Samuelsson 1974; Nugteren 1975), murine epidermis (Chen et al. 1994), and bovine epithelium (Hansbrough et al. 1990).
When using arachidonic acid as a substrate, 12-LOXs
27 TABLE І. SUBSTRATE SPECIFICITY OF 12(S)-LIPOXYGENASE ISOFORMS
Isoform Platelet Leukocyte Epidermis
Arachidonic acid Active Active Low
Linoleic/linolenic Low Active Low
Linoleic acid methyl ester – – Active
Phospholipids Low Active –
–; Not determined
(From Yoshimoto and Takahashi, 2002)
28 can catalyze the insertion of molecular oxygen at position C12. The leukocyte isoform however, is also capable of 15-lipoxygenation to the extent of approximately 10% of total product formed (Takahashi et al. 1988; Kishimoto et al. 1996).
Inhibitors
The majority of known 12-LOX inhibitors are non-specific in nature. The predominant amount of 12-LOX inhibitors are classified as antioxidants. The efficacy of antioxidants lies in the compounds’ ability to occupy the active site of lipoxygenase and reduce the Fe3+ (active) to Fe2+ (inactive).
Nordihydroguaiaretic acid (NDGA), a catechol derivative, is the most widely used lipoxygenase inhibitor despite a very low specificity. Several other catechol derivatives including quercetin, esculetin, and baicalein possess antioxidant properties and are known inhibitors of lipoxygenase. Quercetin inhibits eicosanoid formation in platelets (Pace-Asciak et al. 1995). At 20 µM, esculetin was able to completely inhibit the production of 12(S)-HETE in the breast cancer cell line MDA-MB-435 (Liu, X. et al. 1996). Baicalein has been extensively used as a P12-LOX inhibitor (Pidgeon et al. 2003) due to its relative selectivity for this enzyme over cycloxygenase. Baicalein also inhibits angiogenesis and tumor growth of prostate cancer cells following subcutaneous injection in nude mice
(Miocinovic R et al., submitted manuscript). Another antioxidant, N-benzyl-N- hydroxy-5-phenylpentamide (BHPP), is a selective inhibitor of P12-LOX. This
29 compound has been used in numerous studies to inhibit the formation of 12(S)-
HETE (Chen et al. 1994). Flavanoids derived from cocoa have been shown to inhibit P12-LOX as well as 15-LOX (Schewe et al. 2001).
Curcumin, a flavanoid compound derived from the dried roots of the plant
Curcuma longa, has long been known to possess anti-inflammatory properties and is a commonly used spice in Asia. Curcumin was shown to inhibit both 12- lipoxygenase and cyclooxygenase in human platelets (Ammon et al. 1993). In addition, 12(S)-HETE production in the rat lens is inhibited by treatment with curcumin (Lysz et al. 1991). Curcumin possesses many properties in addition to those of lipoxygenase and cyclooxygenase inhibition. As an antioxidant, curcumin’s potency can be as much as five times that of vitamin E. Curcumin has been shown to act as a tyrosine kinase inhibitor (Korutla et al. 1995), particularly that of the epidermal growth factor receptor, and can inhibit activation of the transcription factors AP-1 and NF-κB (Xu et al. 1997). Several recent studies in mice have effectively shown the potential of this compound as an inhibitor of tumorigenesis (Huang et al. 1997) and general eicosanoid formation
(Huang et al. 1991). Unfortunately, in humans curcumin undergoes rapid metabolism in the liver and intestinal wall reducing the overall bioavailability.
Interestingly, concomitant administration of piperine and curcumin increases the bioavailability of curcumin by 2000% (Shoba et al. 1998). Despite low endogenous bioavailability, curcumin, when taken as a dietary supplement, can prevent colon cancer formation (Rao et al. 1995), although the exact mechanism of chemoprevention has yet to be determined.
30 There are other specific and non-specific inhibitors of 12-lipoxygenases with only a select few listed here. A fine review of lipoxygenase inhibitors including structural characteristics and mechanisms of action was written by
Dailey and Imming (1999).
Platelet 12-Lipoxygenase
Originally characterized in human platelets, arachidonate 12- lipoxyygenase (arachidonate: oxygen 12-oxidoreductase, EC 1.13.11.31) was the first lipoxygenase discovered in mammals (Hamberg and Samuelsson 1974).
This protein has since been found in various tissues and cells including keratinocytes (Hussain et al. 1994), human erythroluekemia cells (Funk et al.
1990; Izumi et al. 1990), epidermis (Takahashi et al. 1993b), epidermoid carcinoma A431 cells (Chang et al. 1993), and some mouse blood cells such as megakaryocytes and eosinophils (Nakamura et al. 1995). Human P12-LOX consists of 663 amino acids and has a molecular mass of ~75kDa. Oxygenation of arachidonic acid catalyzed by P12-LOX proceeds for 30 minutes in a linear fashion significantly contrasting that of leukocyte 12-LOX, which undergoes a type of suicide inactivation within minutes (Yamamoto et al. 1997; Yoshimoto and
Takahashi 2002). This suicide inactivation presumably does not occur in the platelet-type 12-lipoxygenase, because there is no observed covalent binding of the hydroperoxy product to the enzyme (Kishimoto et al. 1996).
31 Gene Structure
The P12-LOX gene, designated Alox12p, consists of 14 exons, 13 introns and spans 13-17kb of chromosome 11 (murine) or 17 (human) and is almost twice the size of the leukocyte and epidermal genes (Funk 1996). Using fluorescence in situ hybridization, the human P12-LOX gene has been conclusively mapped to chromosome 17p13.1 (Yoshimoto et al. 1992). Using
Northern blot analysis, the mRNA for P12-LOX was shown to be a single species of 3.1 kb in human erythroleukemia cells (HEL) (Yoshimoto et al. 1990b).
The promoter region of the P12-LOX gene has been described as being similar to that of a housekeeping gene because of a lack of TATA or CCAAT boxes. This is, however, unlikely as protein activity is inducible as covered below. The 5’ flanking regions from human (Funk et al. 1992; Yoshimoto et al.
1992) and murine (Chen et al. 1994) P12-LOX genes have been reported.
Within a 200 base pair stretch in the 5’ untranslated region of the P12-LOX gene, several potential cis-acting regulatory elements are found: three GC boxes for binding of the transcription factor Sp-1 and a TATA-like box. Further upstream, the P12-LOX gene has a CACCC box, several AP-2 binding sites, NF-κB sites and a glucocorticoid-response element (Chen et al. 1994). There is, however, little evidence as to what role, if any, these distant sites play in the regulation of
P12-LOX gene transcription. The mRNA for P12-LOX is not believed to undergo any sort of post-translational modifications.
32 Modulation of Transcription
Two Sp-1 binding sites located at -158 and -123 bases upstream of the transcription initiation codon (Liu, Y. et al. 1997) were shown to be essential for
P12-LOX gene expression by site directed mutagenesis and luciferase promoter assays (Chen et al. 1999). Using human epidermal A431 cells, Chen and Chang showed that overexpression of c-Fos caused an increase in P12-LOX gene transcription and enzyme activity in a dose dependent manner. Co-transfection of c-Fos with c-Jun resulted in a synergistic increase in P12-LOX mRNA expression. By site directed mutagenesis and 5’ deletions, it was shown that the two Sp-1 sites at -158 and -123 are critical for the c-Fos induced response of
P12-LOX gene transcription (Chen and Chang 1999). In the same cells, overexpression of Ha-ras was shown to potentiate Sp-1 binding by c-Jun, facilitating a dose and time dependent increase in P12-LOX promoter activity
(Chen et al. 1997; Chen and Chang 2000).
Several other factors have been shown to modulate the expression of the
P12-LOX gene or alter the protein’s activity. In human epidermoid carcinoma
A431 cells, epidermal growth factor (EGF) was shown to increase P12-LOX mRNA expression by approximately two-fold following a 10 h lag period (Chang et al. 1993). This effect is negated by use of EGF receptor specific tyrosine kinase inhibitors, indicating the requirement for these receptor tyrosine kinases
(Hagmann et al. 1993). In addition, the induction of gene expression in response to EGF was shown to be dependent upon the action of protein kinase C (Liu, Y.
33 et al. 1994). Interestingly, it has been noted that for growth of primary human lens epithelial cells in response to EGF or insulin, 12(S)-HETE is required
(Haque et al. 1999), leading to the assumption that 12-lipoxygenases may play important roles in EGF induced cellular proliferation. Further studies using A431 cells have shown that transforming growth factor-α (TGF-α), a natural ligand for the EGF receptor, and phorbol 12-myristate 13-acetate (PMA) can increase P12-
LOX mRNA and enhance the activity of the protein (Liaw et al. 1998; Chen et al.
1999). In Chinese hamster ovary cells overexpressing the angiotensin II receptor, angiotensin II treatment resulted in enhanced 12(S)-HETE production
(Wen et al. 1996).
Down regulation of EGF induced P12-LOX expression can be attained by pretreatment of A431 cells with the glucocorticoid dexamethasone (Chang et al.
1995). This effect could be completely blocked by treatment with the glucocorticoid antagonist RU486, suggesting that dexamethasone inhibition of
P12-LOX gene expression is mediated through glucocorticoid receptors.
Therefore, the presence of a glucocorticoid-response element in the promoter region seems to down regulate the gene. Another study has shown that negative modulation of P12-LOX gene expression can be potentiated via NF-κB heterodimers (Arakawa et al. 1995). In this particular study, it was noted that in
HEL cells the Sp-1 binding sites located at -158 and -123 had little to no regulatory effects on P12-LOX gene expression. The differential regulation of the
P12-LOX gene in these two cell types, A431 and HEL, is believed to be due to differences among the transcription initiation sites in the respective cells.
34 Modulation of Activity
Epidermal growth factor, in addition to mRNA modulation, can stimulate the activity of P12-LOX in epidermoid carcinoma A431 cells (Hagmann et al.
1996). Other factors such as glucose, angiotensin II (Natarajan et al. 1993), the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (Hagmann et al. 1993;
Mahmud et al. 1993), and fibronectin (Kuchinke and Funk 1994) can modulate the activity of 12-LOX as well. Glutathione depletion, resulting in accumulation of hydroperoxides, significantly increased the activity of P12-LOX in human epidermoid carcinoma A431 cells (Chen et al. 2000). This leads to the assumption that modulation of lipoxygenase activity depends upon the ratio of phospholipid hydroperoxide to glutathione peroxidase. In fact, Sutherland has proposed that in human platelets the hydroperoxide tone of the cell determines if
12-HpETE undergoes reduction to 12(S)-HETE or isomerization resulting in hepoxilins (Sutherland et al. 2001). Dexamethasone, in addition to reducing
P12-LOX gene expression, can decrease the activity level of the enzyme (Chang et al. 1995). Albumin is postulated to affect P12-LOX activity by sequestering arachidonic acid, and in fact can prolong the metabolism of 12(S)-HETE thus promoting a long circulating 12(S)-HETE life span (Dadaian and Westlund 1999).
By using the yeast two-hybrid system in human epidermoid carcinoma A431 cells, type II keratin 5 and lamin A were identified as specific interacting proteins with P12-LOX (Tang et al. 2000). Whether these proteins modulate the activity of P12-LOX has yet to be determined.
35 Tissue Distribution and Subcellular Localization
The platelet-type 12-LOX has been shown to be expressed in various tissues, most notably platelets and epidermis (Yoshimoto and Takahashi 2002), and multiple tumor types, which will be discussed later. Northern blot analysis has indicated that bovine corneal epithelium contains mRNA for P12-LOX
(Liminga and Oliw 1999). Using immunohistochemical staining, it was demonstrated that P12-LOX can be found in open-type enteroendocrine cells of stomach and intestinal epithelium (Nakamura et al. 1997).
Regarding the intracellular localization of P12-LOX, differential centrifugation and immunoelectron microscopy revealed that P12-LOX is predominantly found in the cytosol of cells as well as in the microsomal fraction of human skin epidermal cells (Hussain et al. 1994). However, P12-LOX is translocated to the membrane through a calcium dependant process (Baba et al.
1989; Hagmann et al. 1993) and this translocation can be inhibited in platelets using the anti-platelet agent OPC-29030, resulting in a suppression of 12(S)-
HETE production (Ozeki et al. 1999). Other studies have shown that P12-LOX activity depends upon cell type and varies from predominantly cytosolic (Lagarde et al. 1984) to preferential membrane association (Hagmann et al. 1993;
Mahmud et al. 1993). According to Hagmann et al. (1996), in A431 cells the
P12-LOX protein predominantly resides in the cytosol, however, the activity of the protein is localized in the membrane compartment of the cell.
36 Role in Non-Cancerous States
Multiple normal cell types such as platelets, macrophages, neutrophils
(Kim et al. 1995), smooth muscle cells (Natarajan et al. 1993) and endothelial cells (Nie et al. 2000b) can produce 12(S)-HETE. A well characterized physiological function of P12-LOX regards the modulation of thrombocyte activation (Sekiya et al. 1991). Thrombin induced platelet aggregation is stimulated by 12(S)-HETE (Nyby et al. 1996). 12(S)-HETE is also capable of increasing the surface expression of the αIIbβ3 integrin by promoting integrin translocation from cytoplasmic stores (Steinert et al. 1993). This can, in turn, result in the recruitment of more platelets and tighter packing of cells.
Although there have been no reported defects in the P12-LOX locus, there are several studies indicating that several myeloproliferative bleeding disorders are linked to P12-LOX deficiencies (Schafer 1982; Matsuda et al. 1993).
Germinal layer keratinocytes selectively overexpress P12-LOX during the process of psoriatic inflammation (Hussain et al. 1994). In fact, lesions associated with psoriasis were shown to possess a significant increase in 12(S)-
HETE production (Hammarstrom et al. 1975). As mentioned above, the glucocorticoid dexamethasone can down regulate the expression of the P12-LOX gene and protein activity, thus providing a pharmacological basis for the use of glucocorticoids as anti-inflammatory agents in the treatment of inflammation associated with psoriasis. In the process of psoriatic inflammation, 12(S)-HETE is known to promote chemotaxis of epidermal cells and activate neutrophils (Hein
37 et al. 1991). However, neutrophil activation by 12(S)-HETE is reduced in scale when compared to that of the eicosanoid leukotriene B4. Further, 12(S)-HETE was shown in guinea pig skin to promote epidermal growth as well as psoriatic associated inflammation (Chan et al. 1985). Thus, it can be stated that elevated tissue concentrations of 12(S)-HETE have a profound impact on the growth of psoriatic plaques through effects associated with inflammation and uncharacteristic epidermal hyperproliferation.
Elevated 12(S)-HETE production has been noted in cases of inflammatory bowel disease as well (Shannon et al. 1993). Platelet 12-LOX deficient mice exhibited heightened sensitivity of platelets to adenosine diphosphate (ADP), suggesting that a P12-LOX arachidonate metabolite is able to suppress ADP promotion of platelet activation (Johnson et al. 1998). It has been suggested that high density lipoproteins play a protective role in the development of arteriosclerosis by selective inhibition of P12-LOX and the generation of 12(S)-
HETE (Fujimoto et al. 1994) although the validity of this claim has yet to be determined.
Products of 12-LOX activity are also implicated in the development and progression of diabetes mellitus by regulating the secretion of insulin and/or glucagons from pancreatic islets. Production of 12(S)-HETE was found to be increased in early stages of diabetes mellitus in clinical studies (Fujimoto et al.
1994). Insulin release into the circulation was noted after intra-arterial administration of hepoxilins to the anesthetized rat (Pace-Asciak et al. 1999). An increase in glucagon release stimulated by 12(S)-HETE was noted in pancreatic
38 islets in vitro (Falck et al. 1983). Conversely, it was found that patients with non- insulin dependent diabetes mellitus had reduced platelet associated P12-LOX activity (Tohjima et al. 1998).
Role in Cancer and Related Processes
Arachidonic acid metabolism by P12-LOX results in the production of
12(S)-HETE, which has been shown to play a significant role in cancer progression, angiogenesis, and portions of the metastatic cascade. The expression of 12-LOX has been shown in various tumor tissues including breast
(Natarajan et al. 1997b), colorectal (Kamitani et al. 1999), lung (Soriano et al.
1999), pancreatic (Ding et al. 1999a), renal (Yoshimura et al. 2004), and prostate (Nie et al. 1998). Multiple cancer cell lines were shown to express 12-
LOX and produce 12(S)-HETE including lung (Soriano et al. 1999), melanoma
(Timar et al. 1999), HEL (Funk et al. 1990), gastric (Wong et al. 2001), pancreatic
(Ding et al. 2001), bladder (Yoshimura et al. 2003), renal (Yoshimura et al. 2004), prostate (Gao et al. 1995) and epidermoid carcinoma (Hagmann et al. 1996).
Accumulating evidence has implicated P12-LOX as key player in prostate cancer related angiogenesis and the ensuing metastatic cascade (Honn et al.
1994a). Numerous prostate cancer cell lines express P12-LOX including
DU145, PC-3, LnCAP, TSU, ML2, and PPC-1 (Timar et al. 2000). Any roles that
P12-LOX plays in the processes of angiogenesis and metastasis are dependent
39 upon the production of 12(S)-HETE. Solid tumor growth and metastasis rely on the tumor’s ability to induce angiogenesis (Folkman and Shing 1992).
Angiogenesis
Angiogenesis, the process of new blood vessel formation from preexisting vasculature, is a complex process mediated by endothelial cells of the vasculature. Under normal circumstances, endothelial cells are quiescent and rarely divide. It is well established that tumor cells, in an effort to stimulate angiogenesis, secrete numerous proangiogenic factors including bFGF (New and
Yeoman 1992) and VEGF (Folkman and Shing 1992). Tumors can also regulate the rate of angiogenesis by producing anti-angiogenic factors as well, resulting in a balance of pro- and anti-angiogenic factors that determine the overall angiogenicity of a tumor (Folkman and D'Amore 1996). Endothelial cells, upon being stimulated by the aforementioned growth factors, will then secrete enzymes for the purpose of matrix degradation. Endothelial cells migrate into this newly created space and proliferate, ultimately resulting in blood vessel formation. This newly created blood vessel supplies nutrients to portions of the tumor that cannot obtain them via simple diffusion. Angiogenesis is critical for tumor elaboration in excess of several millimeters in size (Gimbrone et al. 1972).
Overexpression of P12-LOX in human prostate PC-3 cells can induce significant tumor vascularization after subcutaneous injection into nude mice (Nie et al. 1998). This effect is attributed to increased production of 12(S)-HETE.
40 Along those lines, 12(S)-HETE has been shown to be a mitogenic factor for microvascular endothelial cells and promote wound healing in scratch injured cell monolayers (Tang et al. 1995). Inhibition of P12-LOX with BHPP resulted in a reduction of growth factor stimulated endothelial cell proliferation, migration, and tube differention. Conversely, overexpression of P12-LOX in CD4 endothelial cells resulted in a stimulation of cell migration and tube differentiation (Nie et al.
2000b). Vascularization of a tumor not only supplies nutrients and removes waste, but also acts as a portal for tumor cell entrance to the blood stream and thus tumor cell migration to distant sites, a process known as metastasis.
Metastasis
Considerable data has accumulated concerning the role of 12(S)-HETE in various steps of the metastatic cascade (Figure 5). In fact, several studies link the metastatic potential of a tumor cell to the ability to produce 12(S)-HETE.
Different subpopulations of metastatic B16 amelanotic melanoma (B16a) cells were found to have altered ratios of 12(S)-HETE:5-HETE. Low metastatic cells produced equal amounts of the respective eicosanoids, while high metastatic cells produced almost exclusively 12(S)-HETE (Liu, B. et al. 1994b). In addition, after treatment with arachidonic acid, the high metastatic B16a cells produced approximately four times more 12(S)-HETE to equally treated low metastatic cells. This corresponds well with the increase in 12-LOX mRNA found in these cells (Tang and Honn 1994). Other cell lines from Dunning rat prostate
41 carcinoma (Liu, B. et al. 1994a) and murine K-1735 melanoma (Silletti et al.
1994) produce similar results; more highly metastatic cells produced increased amounts of 12(S)-HETE (Tang and Honn 1994). In fact, overexpression of P12-
LOX in PC-3 cells has previously been linked to increased metastatic potential through augmentation of cellular adhesion, spreading, mobility, and invasiveness
(Nie et al. 2003). Interestingly, this study showed metastasis to human bone was increased when compared to non-overexpressors as determined by the SCID-hu bone model (Nemeth et al. 1999). The multitude of roles that 12(S)-HETE plays in the development of the metastatic phenotype are delineated below.
In order for metastasis to occur, metastatic cells must first be released from the primary tumor and migrate towards local vasculature. This is possible through structural rearrangement of the cytoskeleton induced by cytokines such as autocrine motility factor (AMF) (Stoker and Gherardi 1991). Binding of AMF to membrane receptors triggers a signal transduction cascade resulting in rearrangement of filamentous elements of the cytoskeleton. In murine melanoma
K1735 cells, cytoskeletal rearrangement induced by AMF was shown to be dependent upon the production of 12(S)-HETE, which as a secondary messenger is able to promote protein kinase C (PKC) activation (Silletti et al.
1994). In a calcium dependent manner, PKC can regulate the rearrangement of various cytoskeletal elements (Timar et al. 1993).
Once a tumor cell has reached the vasculature, it must then overcome structural components of the basement membrane and the endothelial cells lining the vasculature. Integrins expressed on the surface of tumors cells will bind to
42 FIGURE 5. ROLE OF 12(S)-HETE IN THE METASTATIC CASCADE
proteolytic degradation of
Tumor cells: autocrine enzymes basement membrane Tumor cells: 1.) migration motility 2.) intravasation and motility factor
activation aVb3 receptor 5.) NewTumor of Protein Kinase C upregulation cell colony Endothelial cells with aVb3 integrine receptors (vitronectin)
binding
Endothelial cells upregulates receptor 12(S)-HETE with fibronectin 12(S)-HETE binding integrine upregulation receptors Platelets translocation with integrine of aIIbb3 to receptor aIIbb3 retraction of degradation of membrane surface (fibronectin) endothelial cells basement membrane activation of Protein Kinase C
Tumor cells: proteolytic Tumor cells: 4.) extravasation enzymes 3.) adhesion
(From Daily and Imming 1999) 43 components of the basement membrane, which has been shown in murine melanoma cells to increase 12(S)-HETE production leading to positive modulation of integrin surface expression. Increased surface integrin expression and binding results in the formation of focal adhesion plaques, promoting a tight cell-matrix adhesion. Binding of basement membrane constituents by integrins results in a spike of 12(S)-HETE production and release from the tumor cell.
This 12(S)-HETE production can in turn induce the release, by a PKC dependent pathway, of the proteolytic enzyme cathepsin B from highly metastatic murine
B16a melanoma and human MCF10AneoT mammary carcinoma cells (Honn et al. 1994b). Proteolytic enzymes such as this are then able to degrade the basement membrane enabling access to the bloodstream.
In order to survive the stress forces associated with the blood stream, the tumor cell must protect itself by attaching to endothelial cells lining the vasculature via integrin binding (Tang et al. 1993b). Host platelets then attach themselves to the tumor-endothelial cell complex, resulting in platelet activation, which in turn recruits other platelets to the site and aggregation ensues (Honn et al. 1992b). Enveloped in platelets, the tumor cell/s can traverse the circulation and avoid hemodynamic stress forces. This thrombus may then lodge in a first pass capillary bed, allowing tumor cells to adhere to local endothelial cells, presumably utilizing the “docking and locking” method proposed by Honn (Honn and Tang 1992).
The process of extravasation begins by 12(S)-HETE promoted endothelial cell retraction, a process noted to occur in microvessel endothelial cells (Tang et
44 al. 1993a,b), and continues with binding of tumor cell integrins to the exposed subendothelial matrix (Honn et al. 1989). The processes of extravasation and intravasation are highly similar, in that binding to the basement membrane will increase surface integrin expression, promote the formation of focal plaques, and induce the release of proteolytic enzymes. Once a hole is opened, the tumor cell may exit the vasculature and begin the formation of a metastatic growth. The importance of the 12(S)-HETE in the processes of angiogenesis and metastasis is evident.
12(S)-HETE Receptors
Several studies have shown that receptors for 12(S)-HETE can be found in Lewis lung carcinoma cells (Herbertsson et al. 1998) and in human epidermal
Langerhans cells (Arenberger et al. 1992). In Lewis lung carcinoma cells, cytosolic/nuclear binding sites were identified to be constituents of a high molecular weight cytosolic binding complex containing several heat shock proteins (Herbertsson et al. 1999). The actual binding protein has since been identified as a 50-kDa protein which can further interact as a homodimer, in the presence of 12(S)-HETE, with steroid receptor coactivator-1 in Lewis lung carcinoma cells (Kurahashi et al. 2000). Recently, 12(S)-HETE has been shown to act as a ligand for the peroxisome proliferator-activated receptor γ (PPARγ) (Li et al. 2003). In this study, 12(S)-HETE was shown to act as an activating ligand of PPARγ as indicated in a cell based luciferase reporter system. Treatment with
45 12(S)-HETE promoted a 20-fold increase in reporter activity compared to unstimulated control.
Another line of evidence indicates that specific subsets of G-protein coupled receptors (GPCRs) can act as binding sites for 12(S)-HETE (Hampson and Grimaldi 2002). These receptors are composed of a single polypeptide chain that spans the cell membrane seven times and couples to heterotrimeric
(α, β, and γ) G-proteins, which can activate several intracellular signaling pathways (Neer 1995). The G-proteins are comprised of four subfamilies (Gq/11,
Gi/o, Gs and G412/13). Classification depends upon the type of α subunit, of which
23 have been characterized so far (Gudermann et al. 1996). The diversity of the
β and γ subunits is also considerable, with at least 5 β and 11 γ recognized thus far (Exton 1996). The α subunit, through post translational acylations, is responsible for localizing the protein to the membrane compartment (Casey
1995). Additionally, the α subunit is the GDP-ligated (guanosine diphosphate) subunit of the inactive heterotrimer. Upon ligand binding to the GPCR, a conformational change occurs in the receptor and G-protein, reducing the affinity of the α subunit for GDP, which is displaced by GTP resulting in activation of the
G-protein. Activation of the G-protein results in dissociation of the G-protein heterotrimer from the transmembrane receptor. Interestingly, both the α-GTP subunit and the βγ dimer can elicit effector functions. A recent review of GPCRs and their interactions examines these steps in detail (Sugden and Clerk 1997).
Although each of the G-protein subfamilies can influence multiple signaling pathways, we are solely interested in G-proteins which can be inhibited
46 by pertussis toxin (PTx) pretreatment. This subfamily is composed of Gi/o G proteins (Sugden and Clerk 1997). Ligand binding to Gi/o GPCRs can stimulate the activation of numerous signaling pathways including the mitogen activated protein kinases (MAPKs). Pertussis toxin sensitive GPCRs have been shown to be activated by 12(S)-HETE in several cell types. For example, PTx pretreatment of epidermoid carcinoma A431 cells was able to block 12(S)-HETE stimulation of the MAPK signal transduction pathway (Szekeres et al. 2000a). A
PTx sensitive GPCR was also shown to be responsible for 12(S)-HETE elicited effects in cortical neuronal cultures (Hampson and Grimaldi 2002). Figure 6 summarizes the possible signal transduction events that can occur when 12(S)-
HETE binds to a PTx sensitive GPCR leading to the activation of ERK1/2.
MAPK Pathway
Cell signaling is initiated when a external signaling molecule, i.e., hormones, cytokines, neurotransmitters, drugs, etc. binds to a cell surface receptor protein. Receptor activation results in signal internalization and triggers a cascade of signaling events. Upon crossing the nuclear membrane, the signal modulates DNA transcription and translation events.
The MAPK pathway is one such signal transduction cascade. Numerous receptor classes including tyrosine kinase, cytokine, and G-protein coupled are known to stimulate signaling via the MAPK pathway. The MAPKs are a family of
Serine/Threonine kinases with important roles in intracellular signaling, and this
47 is emphasized by the fact that these proteins are relatively conserved across species (Marshall 1995). Mammalian cells are known to contain at least three subfamilies of MAPKs, including: 1) the extracellular signal-related kinases
(ERKs); 2) the p38 MAPKs; and 3) the c-jun N-terminal kinases (JNKs) or stress activated protein kinases (SAPKs). The ERKs are closely connected to cellular growth and differentiation, while the JNKs/SAPKs and p38 MAPKs are predominantly involved in stress responses (Johnson 2002). For our purposes, we will focus on the ERK subset of MAPKs.
The basic sequence of proteins associated with the MAPK/ERK cascade consists of the monomeric G-protein Ras acting upstream of three protein kinases (shown in Figure 6). Mitogen activated protein kinase kinase kinases, or
Raf subfamily members, phosphorylate MAPK kinases, composed of
MEK1/MEK2, which in turn phosphorylate the MAPKs or ERK1/2. This sequential arrangement allows for signal amplification and regulation (Robinson and Cobb 1997).
Activation of Ras is mediated by the guanine nucleotide exchange factor son of sevenless (SOS), which exchanges Ras bound GDP for GTP (Hilger et al.
2002). Ras then binds to one of several Raf kinases (MAPK kinase kinases) and induces translocation to the cell membrane, where activation of Raf takes place
(Moodie and Wolfman 1994). The serine/threonine kinase activity of Raf phosphorylates MEK1/MEK2 (MAPK kinases) (Schaeffer and Weber 1999) and
MEK in turn, phosphorylates specific threonine and tyrosine residues of ERK1 and ERK2 (Anderson et al. 1990).
48 FIGURE 6. PROPOSED SIGNALING MECHANISMS OF 12(S)-HETE LEADING TO ACTIVATION OF ERK1/2
12(S)-HETE
Sos Ras G-protein Shc Grb2
Src Raf
PI3-K PKCα MEK PLCγ1
IP3 ERK1/2
DAG
PKCζ
49 Two ERKs, the 44 kDa ERK1 (p44) and the 42 kDa ERK2 (p42) are the only identified downstream substrates of MEKs (Seger et al. 1992). The ERK
MAPKs are implicated to play roles in multiple cellular processes including regulation of cell growth, differentiation, and protein synthesis (Seger and Krebs
1995). For ERK regulated transcription to occur, the ERKs must be activated then undergo translocation to the nuclear compartment (Marshall 1995). Over 50 different substrates of ERK1/2 have been identified thus far, including factors involved in transcription stimulation, such as the transcription factors TAL1
(Wadman et al. 1994), NF-κB (Zhou et al. 2003), and ELK-1 (Gille et al. 1992;
Marais et al. 1993). ELK-1 has been shown to induce transcription of c-fos, a component of the AP-1 (activator protein-1) transcription factor (Gille et al. 1995).
12(S)-HETE as a Signaling Molecule
A host of effects in multiple cell types can be elicited by 12(S)-HETE.
Many of these effects involve components of multiple signal transduction pathways. The predominant effects elicited by treatment with 12(S)-HETE depend upon the stimulation of a number of PCK isoforms. Numerous steps of the metastatic cascade, which have shown to be affected by 12(S) -HETE, are done so in a PKC dependent manner. For example, increased tumor cell motility
(Silletti et al. 1994), augmented surface expression of various integrins (Grossi et al. 1989; Chopra et al. 1991), and proteolytic enzyme release in response to
12(S)-HETE (Honn et al. 1994b) require the action of PKC, as inhibition of PKC
50 is able to block the effects of 12(S)-HETE in each of these processes. Also,
12(S)-HETE is able to promote PKC translocation to the cell membrane in a Ca2+ mediated fashion, resulting in an increase of activated PKC in lens epithelial cells
(Zhou et al. 2003). In these cells, 12(S)-HETE was able to promote the release of stored Ca2+. Several studies have implicated that an increase in intracellular
Ca2+ concentrations can lead to ERK activation (Enslen et al. 1996; Lee et al.
2000). This effect may however, be indirect as PKC itself is capable of activating
ERK1/2 (Kolch 2000).
Two of the three pathways of the MAPK cascade (ERK1/2 and p38) have thus far been shown to be stimulated by 12(S)-HETE. Treatment of vascular smooth muscle cells with 12(S)-HETE leads to activation of p38 MAPK (Reddy et al. 2002). As a mediator of angiotensin II effects in vascular smooth muscle cells, 12(S)-HETE stimulates p38 phosphorylation as well as phoshorylation,
DNA-binding and transactivation of the transcription factor cyclic adenosine mono phosphate (cAMP) response element (Reddy et al. 2002). In the same study, overexpression of the murine leukocyte 12/15-lipoxygenase led to an increase in basal ERK activity.
Treatment of pancreatic β cells with 12(S)-HETE results in a dose dependent, transient increase in the phosphorylation of ERK1/2 (Ding et al.
2001). Similar results were found using human epidermoid carcinoma A431 cells
(Szekeres et al. 2000a). Interestingly, in A431 cells, tyrosine phosphorylation of phospholipase Cγ1 (PLCγ1) and SH2-domain-containing a2-collagen-related
(SHC) was found following treatment with 12(S)-HETE, effects shown to
51 stimulate phosphoinositide 3-kinase (PI3-K) activity (Szekeres et al. 2000b). In turn, PI3-K stimulated the ERK cascade at the MAPKK level (MEK). Further, inhibition of PI3-K with LY294002 reduced 12(S)-HETE stimulated MEK activity but had no effect on Raf stimulation. Wen et al. reported that 12(S)-HETE could activate p21-activated kinase (PAK1) in CHO-AT1a cells. Additionally, it was found that PI3-K acted upstream of PAK1 in 12(S)-HETE induced phosphorylation of ERK1/2 (Wen et al. 2000). Taken together, these results indicate that 12(S)-HETE can modulate ERK function by several different mechanisms.
Overexpression of P12-LOX in human prostate carcinoma PC-3 cells leads to constitutive activation of the transcription factor NF-κB (Kandouz et al.
2003). This effect was found to be due to 12(S)-HETE induced degradation of the NF-κB inhibitor IκB via the S6 proteasomal pathway. Once free from IκB, NF-
κB was able to undergo nuclear translocation resulting in κB-regulated transcription. Treatment of untransfected cells with increasing concentrations of
12(S)-HETE resulted in a dose dependent increase in DNA binding activity of
NF-κB as shown by electrophoretic mobility shift assay (EMSA). The expression of multiple genes associated with cancer progression are induced by NF-κB, genes including cell adhesion molecules and numerous inflammation mediators such as inducible nitric oxide synthase, TNFα (tumor necrosis factor α), COX-2, and several other chemokines (Garg and Aggarwal 2002). In addition, NF-κB can mediate the expression of proangiogenic factors such as interleukin-8 (IL-8),
52 IL-6, and VEGF (Kunsch and Rosen 1993; Mukaida et al. 1994; Blackwell et al.
1997; Makarov et al. 1997).
Signaling in Androgen Independent Prostate Cancer
Initially, prostate cancer growth and progression is dependent upon the presence of androgens. For this reason, the standard modality of treatment is androgen deprivation. This treatment will lead to tumor shrinkage, but the tumor will ultimately progress to an androgen independent state. Evidence to support this observation comes from the Shionogi mouse model (Bruchovsky et al. 1990).
According to this model, androgen deprivation can result in significant tumor mass regression. Unfortunately, this model selects for tumor cells displaying an androgen independent phenotype. The androgen independent phenotype has been shown to be linked to an increase in the production of autocrine growth factors. For example, androgen independent DU145 prostate carcinoma cells secrete abnormally high levels of EGF and TGF-α (Connolly and Rose 1989) leading to high basal EGFR activation. Activation of the EGFR can lead to stimulation of the MAPK pathway through Ras, the expresson of which can be correlated to tumor grade in prostate cancer (Viola et al. 1986). Under androgen deprived conditions, oncogenic Ras expression can potentiate the activation of
ERK1/2 (Voeller et al. 1991). It is well established that the MAPK pathway plays a critical role in the processes of proliferation and malignant transformation
(Mansour et al. 1994; Gioeli et al. 1999). The prostate cancer cell line DU145
53 has high basal activation of the ERK/MAPK pathway, a process resulting from abnormal EGFR autophosphorylation as mentioned above (Putz et al. 1999).
Disregulated signaling pathways are also involved in the metastatic potential of prostate cancer cells. High levels of EGFR in DU145 cells lead to an in vivo increase in metastatic potential (Turner et al. 1996), most likely due to increased activation of MAPKs. By overexpressing the EGFR in nontumorigenic rat prostate cells, Marengo and colleagues (1997) showed that these cells were now capable of metastasis to rib skeletal muscle. Further, MAPK pathway activation has been shown to be associated with metastatic prostate carcinoma
(Krueger et al. 2001).
Interestingly, numerous studies have shown a cross-talk network between androgen receptor (AR) and MAPK signaling, where MAPK signaling can activate the AR in the absence of androgen and consequently enhance AR transcriptional activity (Culig et al. 1994; Abreu-Martin et al. 1999). This data explains a possible mechanism of androgen independent prostate tumor progression.
Vascular Endothelial Growth Factor
Vascular endothelial growth factor encompasses a family of secreted growth factors which are structurally similar to platelet derived growth factor
(PDGF) (Risau 1997). Vascular endothelial growth factor (VEGF-A) was originally characterized as a vascular permeability factor in that it could promote
54 protein extravasation from blood vessels associated with tumors (Risau 1997).
Additional members of the VEGF family have been characterized and include
VEGF-B, -C, and -D. These related growth factors, however, will not be covered herein. Alternative splicing of the VEGF (VEGF-A) transcript produces five different known isoforms of VEGF containing 121, 145, 165, 189, and 206 amino acids (VEGF121, VEGF145, VEGF165, VEGF189 and VEGF206, respectively)
(Tischer et al. 1991). Aside from differences in mass, these isoforms possess different biological properties such as cell surface heparin-sulfate proteoglycan binding capabilities and receptor recognition (Neufeld et al. 1999). The most abundant VEGF, VEGF165, is a homodimeric, heparin-binding, 45,000 kDa secreted protein that can bind to cell surface or extracellular matrix heparin- sulfate (Ferrara and Henzel 1989). Secretion of VEGF165 has been noted in a variety of normal and transformed cells, with parallel expression of
VEGF121(non-heparin binding) and VEGF189 (heparin binding) noted in the majority of cases (Houck et al. 1991). Expression of VEGF206 (heparin binding) appears to be rare (Houck et al. 1991), and only recently has VEGF145 been discovered (Poltorak et al. 1997) with little known concerning its expression profile. From here on VEGF will refer to the VEGF165 isoform.
Using binding and cross-linking studies, two receptor tyrosine kinases were identified as specific VEGF receptors (Vaisman et al. 1990): 1) VEGFR-1
(flt-1) and 2) VEGFR-2 (KDR/flk-1), which both possess seven extracellular immunoglobulin-like domains distinguishing them from other tyrosine kinase receptors (de Vries et al. 1992; Terman et al. 1992).
55 For vascular endothelial cells of arterial, venous, or lymphatic origin,
VEGF is a potent mitogen (Ferrara and Davis-Smyth 1997), and withdrawal of
VEGF leads to vascular regression in both physiological and pathological conditions (Ferrara 2001).
Gene Structure
The gene for VEGF, as mentioned previously, encodes five different
VEGF isoforms by alternative exon splicing (Figure 7a). The promoter region of the VEGF gene contains binding sites for multiple transcription factors such as
AP-1, AP-2, and Sp-1 (Figure 7b). Hypoxia regulated VEGF transcription is mediated by hypoxia-inducible factor 1 (HIF-1) binding to a hypoxia responsive element (HRE) (Liu, Y. et al. 1995). Interestingly, the oncogene V-src can itself, under normoxic conditions, induce HIF-1 expression, which in turn leads to an increase in VEGF expression (Jiang et al. 1997).
56 FIGURE 7. THE VEGF GENE: ALTERNATIVE SPLICING AND REGULATORY ELEMENTS WITHIN THE PROMOTER REGION
a.
Exons: 1-5 8 VEGF121
Exons: 1-5 6 8 VEGF145
Exons: 1-5 7 8 VEGF165
Exons: 1-5 6 7 8 VEGF189
Exons: 1-5 6 6 7 8 VEGF206
115 24 17 44 6 Amino acids
(From Poltorak et al. 1997) b.
HRE AP-1 Sp-1/AP-2/Sp-1 VEGF
HRE: Hypoxia Response Element AP-1: AP-1 Binding Site Sp-1 : Sp1 Binding Site AP-2: AP-2 Binding Site
57 VEGF and Vasculogenesis
De novo development of new blood vessels, vasculogenesis, is a crucial process for embryonic development. Targeted gene disruption of VEGF in mice has proven VEGF necessary for vasculogenesis, as even a single allele knock- out results in embryonic lethality due to improper cardiovascular development
(Carmeliet et al. 1996; Ferrara et al. 1996). Similarly, disruption of the genes encoding VEGFR-1 and VEGFR-2 yield animals with severe blood vessel abnormalities. Mice lacking VEGFRR-1 had an impaired ability to produce functional blood vessels (Fong et al. 1995). Loss of functional VEGFR-2 resulted in an absence of endothelial cell differentiation and malformation of blood vessels
(Shalaby et al. 1995). These data indicate that cardiovascular development relies on the presence of functional VEGF and VEGF receptors.
VEGF and Angiogenesis
Vascular endothelial growth factor is a primary regulator of angiogenesis in both physiological (embryonic development, female reproductive cycle) and pathological (tumor growth) processes (Folkman and Shing 1992). For this body of work, we will concentrate on tumor related angiogenesis. Early studies demonstrated the significance of VEGF in tumor angiogenesis by showing that inhibition of VEGF signaling results in impaired tumor angiogenesis and consequently tumor growth (Kim et al. 1993; Millauer et al. 1994). Angiogenesis
58 promoted by VEGF also results in blood vessel fenestrations (Roberts and
Palade 1997; Esser et al. 1998), through which extravasation of blood-bourne proteins can occur (Dvorak et al. 1996). These proteins, for example, can facilitate formation of an extravascular fibrin matrix that can act as a support for vascular endothelial and tumor cells alike (Dvorak et al. 1992).
Hypoxia and VEGF Expression
As tumors grow in size, the innermost cells will become deprived of oxygen (hypoxic) due to an increasing distance from local vasculature. Cells within this hypoxic tumor region begin to produce and secrete VEGF to promote angiogenesis. High levels of tumor cell VEGF expression are associated with areas surrounding hypoxia induced necrotic tumor regions (Shweiki et al. 1995).
Hypoxia is considered to be a major stimulator of VEGF expression (Shweiki et al. 1992) and induction of VEGF mRNA expression by hypoxia is rapid and reversible in vitro and in vivo (Minchenko et al. 1994). In addition to potentiation of VEGF gene expression, hypoxia can facilitate VEGF mRNA stabilization by enhancing the binding of proteins to elements located in the VEGF mRNA 3’ untranslated region (UTR) (Claffey et al. 1998).
59 Non-Hypoxic Regulation of VEGF Expression
Hypoxia independent VEGF production by tumor cells can be a consequence of tumor suppressor inactivation, oncogene activation, or stimulation from hormones and other growth factors. Inactivation of the von
Hippel-Landau (Siemeister et al. 1996) and p53 (Kieser et al. 1994) tumor suppressor genes result in increase tumor VEGF production. Activation of the oncogenes ras (Rak et al. 1995), raf (Gille et al. 1992), and src (Mukhopadhyay et al. 1995) can also promote VEGF expression. Numerous exogenous factors can potentiate VEGF expression including PDGF (Finkenzeller et al. 1997), FGF-
4 (Deroanne et al. 1997), TNF-α (Ryuto et al. 1996), TGF-β (Pertovaara et al.
1994), interleukins-1β (IL-1β) (Li et al. 1995) and -6 (IL-6) (Cohen et al. 1996), and keratinocyte growth factor (KGF) (Frank et al. 1995). In addition, several studies have shown that eicosaniods such as PGE2 (Eibl et al. 2003), 5(S)-HETE
(Romano et al. 2001), 12(R)-HETE (Mezentsev et al. 2002) and 12(S)-HETE
(Natarajan et al. 1997a) can positively modulate VEGF expression, although the mechanisms remain unclear.
MAPK and VEGF Expression
Numerous recent studies have implicated components of various MAPK pathways in the regulation of VEGF expression. When cultured under acid conditions (pH 6.6), human glioblastoma cells express increased VEGF mRNA
60 and protein. After 10 min under acidic conditions, ERK1/2 phosphorylation was significantly augmented, an effect that could be blocked by treatment with the
MEK inhibitors PD98059 and U0126. Using truncations of the VEGF promoter and a luciferase reporter, the region responsible for increased transcription was mapped between -961 and -683, a region containing an AP-1 binding site (Xu et al. 2002).
Transfection of CCL39 hamster fibroblasts with constitutively active Ras and MEK1 mutants significantly increased VEGF gene expression. This increase could be blocked by treating cells with the MEK1 inhibitor PD98059, showing the importance of ERK1/2 in Ras mediated VEGF regulation. Further, this study investigated the transcriptional elements involved in MAPK stimulation of VEGF gene transcription. Positive regulatory elements of the VEGF promoter were mapped to -88 to -66bp upstream from the transcription start site. This region contains two Sp-1 binding sites and one AP-2 site. It was noted that cooperation between both Sp-1 sites and the AP-2 site appears to be necessary for maximal transcriptional activation of the VEGF promoter. Interestingly, deletion of the AP-
1 binding site had little effect on VEGF transcription (Milanini et al. 1998), indicating alternate mechanisms of transcriptional regulation between these cell types. Regulation of VEGF transcription through AP-2 or Sp-1 has also been shown to occur following stimulation by TGF-α (Gille et al. 1997), IL-1β (Tanaka et al. 2000), hepatocyte growth factor (HGF) (Reisinger et al. 2003), and PDGF
(Finkenzeller et al. 1997). In osteoblast-like MC3T3-E1 cells, TGF-β activates
TGF-β activated kinase (TAK1), a MAPK kinase kinase family member and
61 upstream kinase of ERK1/2 (Yamaguchi et al. 1995). Activation of ERK1/2 in
MC3T3-E1 cells by TGF-β treatment promotes VEGF synthesis, again an effect that can be blocked by treatment with the MEK inhibitor U0126 (Tokuda et al.
2003). The transcription factors responsible for TGF-β/MAPK regulated VEGF gene expression were not investigated.
VEGF and Prostate Cancer
Various tumor types express high levels of VEGF, in contrast to the generally low expression levels found in normal tissues (Dvorak et al. 1991).
Inhibition of VEGF has been shown to inhibit tumor induced angiogenesis and limit tumor growth in vivo (Kim et al. 1993). Similar effects were noted by expression of a dominant negative VEGFR (Millauer et al. 1994).
Both malignant and benign prostate cells express VEGF, with malignant cells expressing significantly more VEGF than benign counterparts (Harper et al.
1996). Expression of VEGF in human prostate cancer appears to be, at least initially, dependent on the presence of androgen (Stewart et al. 2001) as castration (Joseph and Isaacs 1997) or hormonal therapy (Moon et al. 1997) results in decreased VEGF production. Several studies have indicated that the extent of tumor vascularization in prostate cancer correlates to the stage of disease (Brawer et al. 1994) as well as invasive (Weidner et al. 1993) and metastatic (Bigler et al. 1993) potentials. In nude mice injected sub-cutaneously with DU145 cells, administration of an anti-VEGF antibody completely inhibited
62 tumor neovascularization, confirming the role of VEGF in tumor related angiogenesis (Borgstrom et al. 1998). Further, mice with human prostate tumor xenografts undergoing radiation treatment were observed to have rapid tumor progression following VEGF administration (Gridley et al. 1997). Recently, it was shown that patients inflicted with metastatic prostate cancer had significantly higher levels of serum VEGF than patients with disease confined to the prostate
(Duque et al. 1999).
63
Manuscript 1 (Submitted)
Platelet 12-Lipoxygenase Overexpression by PC-3 Prostate Cancer Cells is Associated with Enhanced Production of Vascular Endothelial Growth Factor.
N. Patrick McCabe1,2, Steven H. Selman1-3, and Jerzy Jankun1-3*
1Urology Research Center, Departments of 2Urology, 3Physiology & Molecular
Medicine, Medical College of Ohio, Toledo, Ohio 43614-2589
* Correspondence to: Dr. Jerzy Jankun, Urology Research Center, Department of
Urology, Medical College of Ohio, 3000 Arlington Avenue, Toledo, OH 43699-
0008, U.S.A. Phone: (419) 383-3691; Fax: (419) 383-3168; E-mail:
64 ABSTRACT
BACKGROUND. The platelet 12-lipoxygenase (P12-LOX) product 12(S)-
hydroxyeicosatetraenoic acid (12(S)-HETE) promotes vascular endothelial
growth factor (VEGF) expression in human vascular smooth muscle cells. Here we study the effects of P12-LOX overexpression on VEGF production in PC-3
prostate cancer cells and investigate a potential underlying mechanism of this
effect.
METHODS. PC-3 cells were stably transfected to overexpress human P12-LOX.
12(S)-HETE and VEGF production were examined by ELISA of conditioned
media. In addition, 12(S)-HETE induced stimulation of the extracellular-signal
related kinase1/2 (ERK1/2) mitogen-activated protein (MAP) kinase pathway was
examined by western blotting.
RESULTS. P12-LOX overexpression promotes increased accumulation of 12(S)-
HETE and VEGF in culture media. P12-LOX overexpression results in
constitutive ERK1/2 phosphorylation, an effect dependent upon 12(S)-HETE
signaling through a pertussis toxin (PTx) sensitive G-protein coupled receptor
(GPCR) and upon MEK activation. Furthermore, inhibition of MEK and PTx
sensitive GPCRs reduced VEGF accumulation in conditioned media by 70% and
36% respectively.
CONCLUSIONS. Our data provide insight into a possible autocrine mechanism
by which prostate cancer cells with elevated expression of P12-LOX stimulate
the production of VEGF and promote increased local angiogenesis. Thus, P12-
65 LOX inhibition as an adjunctive therapy for prostate cancer treatment may prove beneficial.
66 INTRODUCTION
Prostate cancer is diagnosed in approximately 240,000 men in the United
States per year, making it the most common malignancy in men. While there is
an inherent risk for men to develop prostate cancer, chances can be influenced
by multiple factors including, but not limited to, age, race and diet. Arachidonic
acid, a polyunsaturated fatty acid found in high levels in red meats and high-fat dairy foods, is a major membrane component in mammalian cells. Mobilization of arachidonic acid from membrane phospholipids pools, as a result of stimulation, can lead to the production of eicosanoids via cyclooxygenases (COXs), lipoxygenases (LOXs), or cytochrome P450 epoxygenases. Eicosanoids are known to play significant roles in numerous physiological and pathological
conditions. Recent studies have shown that certain metabolites of arachidonate,
mainly the COX-2 product prostaglandin E2 (PGE2) and the 12-lipoxygenase
product 12(S)-hydroxyeicosatetraenoic acid (12(S)-HETE), play significant roles
in the angiogenic and metastatic potentials of tumors 1.
Lipoxygenase enzymes are found in a wide variety of plant and animal
species and are designated 5-, 8-, 12-, and 15-LOXs. In animals, there are three
12-LOX isoforms (platelet-, leukocyte-, and epidermal type-) that differ in tissue
distribution, substrate specificity, regioselectivity, and product profile. Platelet-
type 12-lipoxygenase was originally characterized in and thought to be exclusive
to human platelets 2, but has more recently been linked to several cancers
including prostate 3,4. P12-LOX protein expression has been confirmed in multiple prostate cancer cell lines including PC-3, DU145, and LNCaP 5.
67 Additionally, Goa et al. show a correlation between P12-LOX expression and
clinical stage and grade in tissue from adenocarcinoma of the prostate 6. The role of P12-LOX and 12(S)-HETE in prostate cancer has, thus far, been shown to be protumorigenic in nature. Overexpression of P12-LOX in PC-3 cells leads to
enhanced tumor growth and angiogenesis in the mouse subcutaneous tumor
growth model 7, but the mechanism of this influence remains elusive.
Angiogenesis, or the process of new blood vessel growth from extension
of pre-existing vasculature, is necessary for sustained tumor growth and
metastasis 8. For vascular endothelial cells of arterial, venous, or lymphatic origin, vascular endothelial growth factor (VEGF) is a potent mitogen 9, and withdrawal of VEGF leads to vascular regression in both physiological and
pathological conditions 10. Immunohistochemical analysis of human prostate
cancer tissue indicates that cancer cells stain positive for VEGF and this staining
correlates with microvascular density 11. Solidifying the role of VEGF in prostate cancer, Borgstrom et al. showed that prostate cancer induced angiogenesis is suppressed by blocking VEGF function with anti-VEGF monoclonal antibodies, thus limiting tumor growth to the pre-vascular phase 12.
Studies indicate that 12(S)-HETE has direct stimulatory effects on multiple processes associated with angiogenesis. For example, 12(S)-HETE has been shown to be a mitogenic factor for microvascular endothelial cells 13 and stimulates endothelial cell migration 14. However, the potential of 12(S)-HETE as
a significant stimulator of pathological angiogenesis may lie elsewhere. Indeed,
12(S)-HETE was shown to induce the expression of VEGF in human vascular
68 smooth muscle cells 15. The present studies were undertaken to determine whether overexpression of P12-LOX by prostate cancer cells leads to augmented production of VEGF. Our data demonstrates that elevated expression of P12-
LOX leads to enhanced VEGF production in vitro. Additionally, we provide a potential mechanism for this phenomenon by reporting that 12(S)-HETE is capable of inducing signaling via the extracellular-signal related kinase1/2
(ERK1/2) mitogen-activated protein (MAP) kinase pathway in PC-3 and DU145 prostate cancer cell lines. These findings suggest that 12(S)-HETE production by
P12-LOX may, in an autocrine/paracrine fashion, stimulate VEGF production and promote prostate cancer induced angiogenesis in vivo.
MATERIALS AND METHODS
Cell Culture and Reagents. The human prostate carcinoma cell lines PC-3
and DU145 were purchased from American Type Culture Collection (Manassas,
VA). Both cell lines were routinely cultured in RPMI 1640 (Sigma, St. Louis, MO)
with 5% FBS, penicillin (100 units/ml) and streptomycin (100µg/ml). Human P12-
LOX cDNA was kindly provided by Dr. Colin Funk (University of Pennsylvania,
Philadelphia, PA) as pcDNA/6HisP12-LOX. Arachidonic acid and 12(S)-HETE
were purchased from Cayman Chemical (Ann Arbor, Michigan). PD98059 and pertussis toxin were from Biomol (Plymouth Meeting, PA), U0126 was from LC
Laboratories (Woburn, MA). PD158780 was a gift from Dr. James Elder
69 (University of Michigan, Ann Arbor, MI). All other chemicals were obtained from
Sigma unless otherwise stated.
Cell Transfection. The cDNA for 6HisP12-LOX was cloned into pTrex-
dest30 using the Gateway system (Invitrogen, Carlsbad, CA). PC-3 cells at 40-
50% confluence in 60mm dishes were transfected with 2µg pTrex/6HisP12-LOX
or empty control vector pcDNA3.1 (Invitrogen) using the Lipofectin reagent
(Invitrogen) according to the manufacturer instructions. Stable transfectants were
selected using 500µg/ml G418 (Invitrogen) and individual colonies were picked
using cloning cylinders. Clones were expanded and maintained in RPMI 1640,
5% FBS and 250µg/mL G418.
Cell Propagation. Cell growth of PC-3 mock and PC-3 pTrex/6HisP12-
LOX transfected cells (P12LOX-4) were monitored using the Aqueous MTS
assay (Promega, Madison, WI). Briefly, 4x103 PC-3 mock and P12-LOX4 cells
were plated per well of a 96 well plate in RPMI 1640 with 5% FBS and Pen/Strep.
After 24 hours, a baseline measurement was taken for normalization. For growth analysis, the media was replaced with serum reduced RPMI (1% FBS) with or without 100nM 12(S)-HETE. Cell growth was monitored at 24 hour intervals following treatment.
RNA Isolation and RT-PCR. PC-3 cells (5x105) were plated in 60mm
dishes in RPMI 1640 with 5% FBS and Pen/Strep then serum starved for 24
hours. Total RNA was extracted using TRIZOL Reagent (Invitrogen). First strand
synthesis was completed using oligo-dT and Superscript III reverse transcriptase
(Invitrogen). This cDNA mixture was then subjected to PCR using the following
70 primers: P12-LOX 16 sense 5` gccaggtatgtggaggggatc 3` and antisense 5`
ggcaccatgtctggctggcg 3` yielding a 404-bp fragment. Reactions conditions were
as follows: 30 cycles of 94°C for 30 sec., 70°C for 30 sec., and 72°C for 45 sec.
β-actin was used as a control with primers from Invitrogen. PCR products were
separated by electrophoresis in a 2% agarose gel and visualized by staining with
ethidium bromide.
Treatments and Western Blot Analysis. PC-3 or DU145 cells were plated in 60mm dishes at a density of 4x105 cells/dish. Cells were grown overnight in
RPMI 1640 with 5% FBS and Pen/Strep. Cells were then serum starved for 24
hours in serum free RPMI with or without 0.5µg/µL pertussis toxin. For recombinant human 6His tagged P12-LOX, expression was detected at this point. Otherwise, media was replaced again for one hour with serum free RPMI
1640 containing vehicle (0.1% DMSO) or with serum free RPMI 1640 containing
PD98059 (50µM for 1 hour), U0126 (10µM for 30 min.), or PD158780 (10µM for
20 min.). Following this treatment, cells were washed twice in serum free RPMI, then treated with vehicle (0.1% EtOH) or 100, 300, or 500nM 12(S)-HETE for various times. Cells were then placed in lysis buffer (20mM Tris-HCl, 100mM
NaCl, 1% triton X-100, 2.0mM EDTA, 1mM sodium orthovanadate, 0.05g sodium deoxycholate, 50µg/mL pepstatin A, 50µg/mL aprotinin, and 250µM leupeptin), scraped and sonicated on ice. Protein concentration was determined using the
BCA method (Pierce, Rockford, IL). Total protein (10µg) was loaded and separated by electrophoresis in 4-12% Tris-Glycine gels (Invitrogen) then transferred to nitrocellulose. For 6HisP12-LOX detection, blots were incubated
71 for 1 hour with anti-HisG (Invitrogen) followed by 30 min. with anti-mouse
horseradish peroxidase (HRP)-conjugated secondary antibodies (Sigma). For
ERK studies, duplicate blots were probed in parallel with either anti-ERK or
phospho-ERK primary antibodies followed by anti-rabbit HRP-conjugated
antibodies (Cell Signal Technologies, Beverly, MA). Immunoreactivity was
determined using enhanced chemiluminescense (Supersignal West Pico, Pierce)
and recorded on CL-XPosure film (Pierce). Blots were stripped in Restore
Stripping Buffer (Pierce) and re-probed with anti-β-actin antibodies (Sigma) for
band intensity normalization.
12(S)-HETE and VEGF Quantitation. PC-3 control and P12-LOX
overexpressing cells were seeded in 60mm dishes at a density of 4x105 cells/dish. Cells were grown overnight in RPMI 1640 with 5% FBS and
Pen/Strep. Media was then replaced with serum reduced media (1% FBS) for 24 hours. The amount of 12(S)-HETE present in conditioned culture media was determined using the Correlate-EIA for 12(S)-HETE Enzyme Immunoassay Kit
(Assay Designs, Ann Arbor, MI) according to the manufacturer instructions.
Similarly, the concentration of VEGF in conditioned media was determined by a
Human VEGF Accucyte EIA (Oncogene, Boston, MA).
Statistics. Data are expressed as mean ± SD. Data were statistically analyzed with Student’s unpaired t test.
72 RESULTS
P12-LOX Overexpression by PC-3 Cells Confers No In Vitro Growth
Advantage. To determine the effects of P12-LOX overexpression in prostate cancer PC-3 cells, we generated stable PC-3 cells overexpressing P12-LOX.
Figure 1A shows, as determined by RT-PCR, the increase in P12-LOX mRNA expression by PC-3 cells following stable transfection. Protein expression of
6HisP12-LOX, as determined by immunoblotting with an anti-6HisGly antibody, by several PC-3 transfected clones (P12-LOX1, P12-LOX4, and P12-LOX5) is shown in Figure 1B. PC-3 clone P12-LOX4 was chosen for further studies. The
MTS proliferation assay indicated that overexpression of P12-LOX confers no in vitro growth advantage to PC-3 cells (Fig. 1C). A lack of growth enhancement induced by the overexpression of P12-LOX in PC-3 cells has previously been noted 7. Furthermore, treatment of PC-3 cells with escalating doses of 12(S)-
HETE (100, 300, and 500 nM) had no effect on the growth of PC-3 cells (data not
shown). These results indicate that 6HisP12-LOX is expressed stably in
transfected PC-3 clones and that P12-LOX overexpression confers no in vitro
growth advantage to these cells.
P12-LOX Overexpression promotes increased production of 12(S)-HETE and VEGF. To investigate the effects of P12-LOX overexpression on production of the bioactive lipid 12(S)-HETE, we analyzed the conditioned media from P12-
LOX4 cells by ELISA. As indicated in Table 1, P12-LOX overexpression leads to
increased production of 12(S)-HETE in serum free conditions. P12-LOX4 cells
73 secrete >3-fold 12(S)-HETE compared to PC-3 controls after 30 and 60 min. in
serum free media. Exposure to 10 ng/mL arachidonic acid increased the amount of 12(S)-HETE secreted by both cell lines with a more pronounced effect noticeable for P12-LOX4 cells.
12(S)-HETE can induce the expression of VEGF in human vascular smooth muscle cells 15 and promotes a angiogenic response by vascular endothelial cells 14, an effect that may involve the induction of VEGF. ELISA of
conditioned media indicated that VEGF protein levels were increased as a result
of P12-LOX overexpression (Table 1). In fact, P12-LOX4 cells secrete
approximately 13-fold more VEGF than that of untransfected controls indicating
that increased expression of P12-LOX can be linked to augmented VEGF
expression by prostate cancer cells.
12(S)-HETE Induces Phosphorylation of ERK1/2 in Prostate Cancer Cells.
Regulation of VEGF protein expression has previously been associated with
ERK1/2 MAPK activation 17-19 and 12(S)-HETE has been shown to induce phosphorylation of ERK1/2 in several cell lines 20, 21. To determine if exposure to
12(S)-HETE can lead to ERK1/2 phosphorylation in PC-3 prostate cancer cells,
we treated these cells with escalating doses of 12(S)-HETE. As shown in figure
2A, 12(S)-HETE induces a transient, concentration dependent increase in
ERK1/2 phosphorylation in PC-3 cells with a peak at 10 min. post exposure. In
serum free conditions, P12-LOX4 cells have high basal levels of ERK1/2
phosphorylation (Figure 2B). Hence, it is reasonable to speculate that abundant
74 12(S)-HETE production by P12-LOX overexpressing cells, thus facilitating
chronic exposure to high concentrations of 12(S)-HETE, can result in enhanced
basal activity of the ERK1/2 MAPK pathway.
The transient increase in ERK1/2 phosphorylation induced in PC-3 cells
by 12(S)-HETE is mirrored in DU145 cells (not shown); however, due to high
basal levels of ERK1/2 phosphorylation in DU145 cells, the effect was less
pronounced. High basal levels of ERK1/2 phosphorylation were previously noted
in DU145 cells, an effect related to increased production of epidermal growth
factor receptor (EGFR) and secretion of the EGFR ligands epidermal growth factor (EGF) and transforming growth factor- α (TGF-α) 22. Indeed, pretreatment
with the EGFR inhibitor PD158780 almost completely blocked basal ERK1/2
phosphorylation but had no effect on 12(S)-HETE induced ERK1/2
phosphorylation (Fig. 2C) indicating that 12(S)-HETE does not induce ERK1/2
phosphorylation through any EGFR dependent pathway in these cells.
12(S)-HETE Induces ERK1/2 Phosphorylation through a Pertussis Toxin
Sensitive G-Protein Coupled Receptor and MEK. 12(S)-HETE was shown to initiate MAPK signaling through a pertussis toxin sensitive GPCR (Gi/o) in human epidermoid carcinoma A431 cells 20. To delineate the mechanism of 12(S)-HETE induced phosphorylation of ERK1/2 in PC-3 and DU145 cells, we treated cells with PTx and inhibitors of MEK, an upstream MAPK of ERKs. Treatment of PC-3 and DU145 cells with 0.5µg/uL pertussis toxin for 18 hours blocked ERK1/2 phosphorylation induction by 12(S)-HETE (Figure 3A), indicating that exogenous
75 12(S)-HETE induction of ERK1/2 phosphorylation is dependent upon 12(S)-
HETE stimulation of pertussis toxin sensitive GPCRs. To determine if 12(S)-
HETE stimulated ERK1/2 phosphorylation is dependent upon MEK, we pretreated PC-3 and DU145 cells with two structurally different inhibitors of MEK;
PD98059 and U0126. As seen in Figure 3B, both PD98059 and U0126 can inhibit 12(S)-HETE induced phosphorylation of ERK1/2 in PC-3 and DU145 cells.
U0126 appeared to be a much more potent MEK inhibitor as it eliminated both basal and inducible ERK1/2 phosphorylation. This may be due to U0126’s ability to inhibit both MEK1 and MEK2, while PD98059 inhibition is specific for MEK1 only.
VEGF Expression is Reduced by Inhibition of PTx Sensitive G-Protein
Coupled Receptors and MEK. In this study, we have shown that P12-LOX overexpression leads to increased VEGF production. To determine if the observed increase in secreted VEGF was due to 12(S)-HETE activation of the
ERK1/2 MAPK pathway, we treated P12-LOX4 cells with PTx to block signaling induced by 12(S)-HETE binding to PTx sensitive GPCRs and with the MEK1/2 inhibitor U0126 and quantified, by ELISA, secreted VEGF 24 hours post initiation of treatment. As shown in Figure 4, treatment of P12LOX-4 cells with PTx reduced the level of VEGF in conditioned media by 36%. Interestingly, U0126 reduced VEGF accumulation by 70% for P12-LOX4 cells and by 56% in control
PC-3 cells indicating MEK as a central player in the production of VEGF.
76 DISCUSSION
Platelet-type 12-lipoxygenase metabolizes arachidonic acid into the
eicosanoid 12(S)-HETE. P12-LOX is expressed by platelets, several endothelial
cell types 14, 23, as well as in several types of animal and human tumors including
prostate 3. In 1995, Gao et al. found that P12-LOX mRNA expression was
significantly higher in prostate adenocarcinoma tissue compared to matched
normal prostate epithelium, and that this increased expression correlated with
advanced stage and grade adenocarcinoma 6. This suggested an association
between prostate cancer progression and elevated expression of P12-LOX.
Stable overexpression of P12-LOX by PC-3 cells was shown to promote enhanced tumor growth and angiogenesis in the subcutaneous mouse model
(Nie et al. 1998); however, the underlying mechanism behind this phenomenon was not reported. Interestingly, in vitro studies show that P12-LOX
overexpression by PC-3 cells confers no growth advantage (herein and 7). It
appears that in vivo, overexpression of P12-LOX results in some sort of alteration
within the tumor microenvironment that promotes tumor growth, possibly by
augmenting the tumor’s angiogenic inducing potential.
Angiogenesis, the process of new blood vessel formation from preexisting
vasculature, is a complex process mediated by endothelial cells of the
vasculature in response to external stimuli. Significant data alludes to a link
between the production of 12(S)-HETE and the processes of tumor induced
angiogenesis. For example, 12(S)-HETE possesses mitogenic properties for
77 microvascular endothelial cells, promotes endothelial cell migration 3, and facilitates wound healing in scratch injured cell monolayers 13. In addition, overexpression of P12-LOX in CD4 endothelial cells promotes cell migration and tube differentiation 14. We found that overexpression of P12-LOX by PC-3 cells led to increased 12(S)-HETE production and accumulation within the culture media in as little as 30 min. under serum free conditions. Addition of the P12-
LOX substrate, arachidonic acid, further enhanced 12(S)-HETE production and accumulation. It is easy to speculate that cells overexpressing P12-LOX possess enhanced angiogenic inducing capacity as a direct result of elevated 12(S)-HETE production. Moreover, 12(S)-HETE was shown recently to stimulate the production of VEGF in human vascular smooth muscle cells 15. Early studies
demonstrated the significance of VEGF in tumor angiogenesis by showing that
inhibition of VEGF signaling results in impaired tumor angiogenesis and
consequently tumor growth 24, 25. We found that P12-LOX overexpression
resulted in a 13-fold increase in VEGF accumulation in culture media under
serum free conditions. These results led us to inquire about a possible
mechanism by which P12-LOX, presumably acting through 12(S)-HETE, could
modulate the expression of VEGF.
Hypoxia is considered to be a major stimulator of VEGF expression 26. In addition to hypoxia, various growth factor stimulated receptor tyrosine kinases can initiate signal transduction cascades leading to the transcription of VEGF 27-
29. One such signal transduction cascade is the MAPK pathway, particularly the
ERK1/2 cascade 18, 19. We found that exposure of PC-3 and DU145 cells to
78 12(S)-HETE stimulates phosphorylation of ERK1/2 in a dose and time dependant
manner and that blockade of MEK, the MAP kinase immediately upstream of
ERK1/2, by two structurally unrelated pharmacological inhibitors, PD98059 and
U0129, inhibited ERK1/2 phosphorylation in PC-3 and DU145 cells. LNCaP cells
have been indicated to be affected similarly upon exposure to 12(S)-HETE 30.
Similar effects were noted previously in pancreatic β cells 21 and human
epidermoid carcinoma A431 cells 20, indicating that this effect may be mirrored in
a broad spectrum of cell types.
Signaling induced by 12(S)-HETE was previously shown to be mediated
by a subset of transmembrane GPCRs that are sensitive to PTx 20, 31. Indeed, we found that 12(S)-HETE signals through a PTx sensitive GPCR (Gi/o) as treatment with PTx blocks ERK1/2 phosphorylation by 12(S)-HETE. Interestingly, blockade of 12(S)-HETE signaling through this GPCR resulted in only a 36% decrease in
VEGF present in the conditioned media of cells overexpressing P12-LOX, whereas inhibition of MEK resulted in a 70% decrease in VEGF present in the conditioned media. These results indicate that 12(S)-HETE signaling through a
PTx sensitive GPCR is only partially responsible for this increased VEGF presence and that other mechanisms of ERK1/2 stimulation are at work, such as
12(S)-HETE interaction with some intracellular, as yet uncharacterized, receptor of 12(S)-HETE. In fact 12(S)-HETE can bind to a 50-kDa protein 32 that exists as part of a high molecular weight cytosolic binding complex 33; however, the functional implications of binding to this receptor remain to be determined. In addition, it is possible 12(S)-HETE may promote VEGF transcription by binding
79 to the nuclear receptor peroxisome proliferated activating receptor gamma
(PPARγ) 34, thus bypassing signaling through MAPKs completely. The role of
PPARγ as an initiator of VEGF transcription is well known 35, 36. Also, we cannot rule out the possibility that P12-LOX overexpression promotes the expression of some other protein, such as COX-2, which in turn enhances VEGF expression 37.
Mechanisms of stimulating VEGF transcription such as these, as well as whether
12(S)-HETE stimulates VEGF transcription or promotes VEGF transcript stability
in prostate cancer cells are currently being addressed in our laboratory.
CONCLUSIONS
Our results indicate that increased expression of P12-LOX in prostate
cancer can result in enhanced production and secretion of the bioactive lipid
12(S)-HETE. Acting in an autocrine/paracrine fashion, 12(S)-HETE can stimulate
ERK MAPK signaling through a pertussis toxin sensitive G-protein coupled
receptor. Activation of ERK1/2 by 12(S)-HETE in prostate cancer cells can
mediate the expression of at least one angiogenic factor, shown herein to be
VEGF. Increased expression of P12-LOX by prostate cancer cells results in
augmented production of 12(S)-HETE and VEGF, two factors known to promote
blood vessel formation, which can lead to enhanced tumor growth and
metastasis by inducing localized vasculature extension. We believe that use of
P12-LOX inhibitors as part of a combined anti-angiogenesis therapy regimen
may prove beneficial in the treatment of prostate cancer.
80 ACKNOWLEDGEMENTS
This work was supported in part by grants from the National Institutes of Health
(CA90524 and CA79450) and The Frank D. Stranahan Endowment Fund for
Oncological Research.
81 REFERENCES
1. Nie, D., Honn, K. V.: Eicosanoid regulation of angiogenesis in tumors.
Semin Thromb Hemost, 30: 119, 2004
2. Hamberg, M., Samuelsson, B.: Prostaglandin endoperoxides. Novel
transformations of arachidonic acid in human platelets. Proc Natl Acad Sci
U S A, 71: 3400, 1974
3. Honn, K. V., Tang, D. G., Gao, X. et al.: 12-lipoxygenases and 12(S)-
HETE: role in cancer metastasis. Cancer Metastasis Rev, 13: 365, 1994
4. Nie, D., Che, M., Grignon, D. et al.: Role of eicosanoids in prostate cancer
progression. Cancer Metastasis Rev, 20: 195, 2001
5. Timar, J., Raso, E., Dome, B. et al.: Expression, subcellular localization
and putative function of platelet-type 12-lipoxygenase in human prostate
cancer cell lines of different metastatic potential. Int J Cancer, 87: 37,
2000
6. Gao, X., Grignon, D. J., Chbihi, T. et al.: Elevated 12-lipoxygenase mRNA
expression correlates with advanced stage and poor differentiation of
human prostate cancer. Urology, 46: 227, 1995
7. Nie, D., Hillman, G. G., Geddes, T. et al.: Platelet-type 12-lipoxygenase in
a human prostate carcinoma stimulates angiogenesis and tumor growth.
Cancer Res, 58: 4047, 1998
8. Folkman, J.: Angiogenesis in cancer, vascular, rheumatoid and other
disease. Nat Med, 1: 27, 1995
82 9. Ferrara, N., Davis-Smyth, T.: The biology of vascular endothelial growth
factor. Endocr Rev, 18: 4, 1997
10. Ferrara, N.: Role of vascular endothelial growth factor in regulation of
physiological angiogenesis. Am J Physiol Cell Physiol, 280: C1358, 2001
11. Ferrer, F. A., Miller, L. J., Andrawis, R. I. et al.: Vascular endothelial
growth factor (VEGF) expression in human prostate cancer: in situ and in
vitro expression of VEGF by human prostate cancer cells. J Urol, 157:
2329, 1997
12. Borgstrom, P., Bourdon, M. A., Hillan, K. J. et al.: Neutralizing anti-
vascular endothelial growth factor antibody completely inhibits
angiogenesis and growth of human prostate carcinoma micro tumors in
vivo. Prostate, 35: 1, 1998
13. Tang, D. G., Renaud, C., Stojakovic, S. et al.: 12(S)-HETE is a mitogenic
factor for microvascular endothelial cells: its potential role in angiogenesis.
Biochem Biophys Res Commun, 211: 462, 1995
14. Nie, D., Lamberti, M., Zacharek, A. et al.: Thromboxane A(2) regulation of
endothelial cell migration, angiogenesis, and tumor metastasis. Biochem
Biophys Res Commun, 267: 245, 2000
15. Natarajan, R., Bai, W., Lanting, L. et al.: Effects of high glucose on
vascular endothelial growth factor expression in vascular smooth muscle
cells. Am J Physiol, 273: H2224, 1997
16. Hagmann, W., Borgers, S.: Requirement for epidermal growth factor
receptor tyrosine kinase and for 12-lipoxygenase activity in the expression
83 of 12-lipoxygenase in human epidermoid carcinoma cells. Biochem
Pharmacol, 53: 937, 1997
17. Milanini, J., Vinals, F., Pouyssegur, J. et al.: p42/p44 MAP kinase module
plays a key role in the transcriptional regulation of the vascular endothelial
growth factor gene in fibroblasts. J Biol Chem, 273: 18165, 1998
18. Xu, L., Fukumura, D., Jain, R. K.: Acidic extracellular pH induces vascular
endothelial growth factor (VEGF) in human glioblastoma cells via ERK1/2
MAPK signaling pathway: mechanism of low pH-induced VEGF. J Biol
Chem, 277: 11368, 2002
19. Tokuda, H., Hatakeyama, D., Akamatsu, S. et al.: Involvement of MAP
kinases in TGF-beta-stimulated vascular endothelial growth factor
synthesis in osteoblasts. Arch Biochem Biophys, 415: 117, 2003
20. Szekeres, C. K., Trikha, M., Nie, D. et al.: Eicosanoid 12(S)-HETE
activates phosphatidylinositol 3-kinase. Biochem Biophys Res Commun,
275: 690, 2000
21. Ding, X. Z., Tong, W. G., Adrian, T. E.: 12-lipoxygenase metabolite 12(S)-
HETE stimulates human pancreatic cancer cell proliferation via protein
tyrosine phosphorylation and ERK activation. Int J Cancer, 94: 630, 2001
22. Connolly, J. M., Rose, D. P.: Secretion of epidermal growth factor and
related polypeptides by the DU 145 human prostate cancer cell line.
Prostate, 15: 177, 1989
23. Funk, C. D., Funk, L. B., FitzGerald, G. A. et al.: Characterization of
human 12-lipoxygenase genes. Proc Natl Acad Sci U S A, 89: 3962, 1992
84 24. Kim, K. J., Li, B., Winer, J. et al.: Inhibition of vascular endothelial growth
factor-induced angiogenesis suppresses tumour growth in vivo. Nature,
362: 841, 1993
25. Millauer, B., Shawver, L. K., Plate, K. H. et al.: Glioblastoma growth
inhibited in vivo by a dominant-negative Flk-1 mutant. Nature, 367: 576,
1994
26. Shweiki, D., Itin, A., Soffer, D. et al.: Vascular endothelial growth factor
induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature,
359: 843, 1992
27. Finkenzeller, G., Sparacio, A., Technau, A. et al.: Sp1 recognition sites in
the proximal promoter of the human vascular endothelial growth factor
gene are essential for platelet-derived growth factor-induced gene
expression. Oncogene, 15: 669, 1997
28. Pertovaara, L., Kaipainen, A., Mustonen, T. et al.: Vascular endothelial
growth factor is induced in response to transforming growth factor-beta in
fibroblastic and epithelial cells. J Biol Chem, 269: 6271, 1994
29. Frank, S., Hubner, G., Breier, G. et al.: Regulation of vascular endothelial
growth factor expression in cultured keratinocytes. Implications for normal
and impaired wound healing. J Biol Chem, 270: 12607, 1995
30. Szekeres, C. K., Trikha, M., Honn, K. V.: 12(S)-HETE, pleiotropic
functions, multiple signaling pathways. Adv Exp Med Biol, 507: 509, 2002
85 31. Hampson, A. J., Grimaldi, M.: 12-hydroxyeicosatetrenoate (12-HETE)
attenuates AMPA receptor-mediated neurotoxicity: evidence for a G-
protein-coupled HETE receptor. J Neurosci, 22: 257, 2002
32. Kurahashi, Y., Herbertsson, H., Soderstrom, M. et al.: A 12(S)-
hydroxyeicosatetraenoic acid receptor interacts with steroid receptor
coactivator-1. Proc Natl Acad Sci U S A, 97: 5779, 2000
33. Herbertsson, H., Kuhme, T., Hammarstrom, S.: The 650-kDa 12(S)-
hydroxyeicosatetraenoic acid binding complex: occurrence in human
platelets, identification of hsp90 as a constituent, and binding properties of
its 50-kDa subunit. Arch Biochem Biophys, 367: 33, 1999
34. Li, Q., Cheon, Y. P., Kannan, A. et al.: A novel pathway involving
progesterone receptor, 12/15-lipoxygenase-derived Eicosanoids, and
peroxisome proliferator-activated receptor gamma regulates implantation
in mice. J Biol Chem, 2003
35. Fauconnet, S., Lascombe, I., Chabannes, E. et al.: Differential regulation
of vascular endothelial growth factor expression by peroxisome
proliferator-activated receptors in bladder cancer cells. J Biol Chem, 277:
23534, 2002
36. Haslmayer, P., Thalhammer, T., Jager, W. et al.: The peroxisome
proliferator-activated receptor gamma ligand 15-deoxy-Delta12,14-
prostaglandin J2 induces vascular endothelial growth factor in the
hormone-independent prostate cancer cell line PC 3 and the urinary
bladder carcinoma cell line 5637. Int J Oncol, 21: 915, 2002
86 37. Gately, S., Li, W. W.: Multiple roles of COX-2 in tumor angiogenesis: a
target for antiangiogenic therapy. Semin Oncol, 31: 2, 2004
87 Figure Legends.
Figure 1. Platelet-type 12-lipoxygenase expression of several stably transfected
human PC-3 prostate cancer cell clones. (A) 5x105 cells were plated in 60mm dishes and grown to ~90% confluence. Total RNA was extracted and reverse transcribed. The P12-LOX gene was amplified using PCR and gene specific
primers. PCR products were separated using 2.0% agarose gel electorphoresis
and visualized by ethidium bromide staining. Lanes: (+), positive control
(pTrex/6HisP12-LOX plasmid); (–), negative control; PC-3 mock (transfected with
empty vector); P12-LOX1,4,5, individual clones. (B) Expression of 6HisP12-LOX
protein was analyzed by western blotting. 10µg of whole cell lysates were
separated by polyacrylamide gel electrophoresis and transferred to nitrocellulose.
6HisP12-LOX protein was detected using an anti-6HisGly antibody; β-actin was detected using anti-β-actin antibody. Lane assignments are similar to those used in A. (C) In vitro growth effects of P12-LOX overexpression by PC-3 cells. Cell
growth rates for PC-3 mock (n=3) and P12-LOX4 cells (n=3) were analyzed by
MTS assay. A baseline measurement was taken 24 h after plating (0 time point).
Subsequent time points (24, 48, and 72 h) were measured as a percent change
in cell number relative to time point 0.
Figure 2. 5x105 cells were plated in 60mm dishes in RPMI containing 5% FBS.
Cells were grown to 60-70% confluence and serum deprived for 24 h in RPMI.
(A) PC-3 cells were exposed to 100, 300, or 500nM 12(S)-HETE for 5, 10, or 30
88 minutes. Cells in lysis buffer were sonicated on ice. 10µg total protein per well
was separated by polyacrylamide gel electrophoresis (in duplicate) and
transferred to nitrocellulose. Phosphorylated and total ERK1/2 were detected
with antibodies specific for each respective form of ERK1/2. (B) Lysates from
PC-3 mock and P12-LOX1, 4, 5 clones were lysed, separated, and analyzed as
in A. (C) Effects of EGFR inhibitor treatment on the 12(S)-HETE induced
phosphorylation state of ERK1/2 in DU145 cells. Cells were plated at a density
of 5x105 cells per 60mm dish in 5% FBS and grown till 60-70% confluent. Cells were serum deprived overnight in RPMI then pretreated with PD158780 (10µM) for 20 min. followed by treatment with either vehicle (0.1% EtOH) or 12(S)-HETE
(100nM) for 10 min. Protein was harvested and detected by western blotting as described in “Materials and Methods”.
Figure 3. Western blot analysis of the phosphorylation state of ERK1/2 following treatment with inhibitors of signal transduction pathways. (A) Effects of pertussis toxin treatment on 12(S)-HETE induced phosphorylation of ERK1/2 in PC-3 and
DU145 cells. Cells were plated at a density of 5x105 cells per 60mm dish in 5%
FBS and grown till 60-70% confluent. Media was replaced with serum free media (with or without 0.5µg/µL PTx) and cells were incubated for 18 h. Cells were then either treated with vehicle (0.1% EtOH) or 100nM 12(S)-HETE for 10 min. (B) Effects of MEK inhibition on 12(S)-HETE induced phosphorylation of
ERK1/2 in PC-3 and DU145 cells. Cells were plated and grown to 60-70% confluence as in A. Following 24 h serum of deprivation, cells were pretreated
89 with vehicle (0.1% DMSO for 30 min), PD98059 (50µM for 1 h) or U0126 (10µM
for 30 min) followed by treatment with vehicle (0.1% EtOH) or 100nM 12(S)-
HETE for 10 min. Protein was harvested and detected by western blotting as
described in “Materials and Methods”.
Figure 4. VEGF protein from conditioned media of P12-LOX4 cells. PC-3 control
and P12-LOX4 cells were seeded in RPMI with 5% FBS at a density of 1x105 cells per well in a 12 well dish and grown to ~60% confluence. Cells were incubated for 24 h in serum reduced media (1% FBS) with or without PTx
(0.5µg/µL). Media was then replaced with fresh serum reduced media containing vehicle (0.1% DMSO), U0126 (10µM), or PTx (0.5µg/µL) for 24 h. Conditioned media was collected and analyzed for VEGF content by ELISA. Data are shown as mean ± SD (untreated, n=6 each group; U0126, n=3 each group; PTx, n=3).
**, statistically significant (P<0.01). *, statistically significant (P<0.05).
90 Figure 1
A
B
C
91 Figure 2
A
B
C
92 Figure 3
A
B
93 Figure 4
94 Table 1
95
Manuscript 2
(Published in the International Journal of Molecular Medicine, Volume 6, Number 5, November 2000)
Curcumin inhibits lipoxygenase by binding to its central cavity: theoretical and X-ray evidence.
EWA SKRZYPCZAK-JANKUN1,2, N. PATRICK McCABE2,3, STEVEN H. SELMAN2-4 and JERZY JANKUN2-4
1Instrumentation Center, College of Arts and Sciences, The University of Toledo, Toledo, Ohio 43606; 2Urology Research Center, Departments of 3Urology, 4Physiology & Molecular Medicine, Medical College of Ohio, Toledo, Ohio 43614- 2589
Running title: evidence of lipoxygenase inhibition by curcumin Key words: curcumin, lipoxygenase, X-ray analysis, molecular modeling, cancer.
Footnotes:
Correspondence to: Dr. Jerzy Jankun, Urology Research Center, Department of Urology, Medical College of Ohio, 3000 Arlington Avenue, Toledo, OH 43699- 0008, U.S.A. Phone: (419) 383-3691; Fax: (419) 383-3168; E-mail: [email protected]; Web address: http://golemxiv.dh.mco.edu/~jerzy/ or Dr. Ewa Skrzypczak-Jankun, Instrumentation Center, College of Art and Sciences, The University of Toledo, Toledo, OH 43606; E-mail: [email protected]; Web address http://golemxiv.dh.mco.edu/~ewa/
96 Abstract
Many lipoxygenase inhibitors including curcumin are currently being studied for their anti-carcinogenic properties. Curcumin is a naturally occurring polyphenolic phytochemical isolated from the powdered rhizome of the plant
Curcuma longa that possesses anti-inflammatory properties and inhibits cancer formation in mice. Recently it was shown that the soybean lipoxygenase L1 catalyzed the oxygenation of curcumin and that curcumin can act as a lipoxygenase substrate. In the current study, we investigate the fate of curcumin when used as a soybean lipoxygenase L3 substrate. By use of X-ray diffraction and mass spectrometry, we have found an unoccupied electron mass that appears to be a unusual degradation product of curcumin (4-hydroxyperoxy-2- methoxyphenol) located near the soybean L3 catalytic site. Understanding how curcumin inhibits lipoxygenase may help in the development of novel anti-cancer drugs used for treatment where lipoxygenases are involved.
97 Introduction
Lipoxygenase enzymes can be found in a wide variety of plant and animal tissues. Lipoxygenases are enzymes that possess a non-heme iron serving as a catalytic center for the stereo- and regio-specific dioxygenation of select carbon atoms in polyunsaturated fatty acids containing a 1,4-pentadiene motif. Eighteen carbon chain fatty acids (eg. linoleate) are the primary substrates of the plant lipoxygenases while the mammalian isozymes mainly catalyze the metabolism of fatty acids of carbon length 20 (eg. arachidonate). The soybean lipoxygenases were the first to be characterized and are named sequentially beginning with soybean lipoxygenase 1 (L1), which was also the first lipoxygenase isozyme to have its three dimensional X-ray crystal structure solved [1, 2]. Nomenclature for the lipoxygenases found in mammals arises from the positional oxygenation along the carbon chain of arachidonic acid (AA). Mammalian lipoxygenases
(LOXs) characterized thus far include 5-, 8-, 12-, and 15- type LOXs. Modeling of human lipoxygenases using pair-wise sequence identity has been performed previously [3] and several theoretical models of substrate mechanisms of action were discussed. To date, rabbit reticulocyte 15-LOX is the only mammalian LOX for which the three-dimensional X-ray structure has been obtained [4].
The amino acid sequences between plant and mammalian LOX enzymes show considerable homology. The soybean lipoxygenases, L1 [1] and L3 [5], are
72% identical in their amino acid sequences, but share only 25% sequence homology to any mammalian 15-LOX. Overall, sequence identity between plant
98 and mammalian pairs of lipoxygenase isozymes is 21-27%, while plant pair
sequence identity ranges from 43-86%, with mammalian pair sequences at 39-
93% identity [3]. The highest level of sequence identity between lipoxygenases
from plants and mammals lies in the area of the catalytic domain containing the non-heme iron atom. Mammalian lipoxygenases are 165-261 residues shorter than the plant lipoxygenases and were believed to lack a N-terminal β-barrel due to the fact that similarities in the sequence identity of the first 200 residues between pairs of plants and animals never exceeds 15%. Comparisons between various mammalian lipoxygenase cDNAs have recently been reviewed [6]. The similarities in sequence data across species lead to the assumption of similar 3 dimensional structures and the comparison of soybean L3 with rabbit 15-LOX confirms that plant and mammalian enzymes share the same topology and overall architecture despite differences in size (Fig. 1)
Soybean lipoxygenases play physiological roles in processes such as growth, development, wound healing and senescence. As the main substrate for soybean lipoxygenases, linoleic acid is metabolized into one of various hydroperoxyoctadecadienoic acids (HPODEs). The biological significance of the
HPODEs has yet to be fully characterized. Mammalian LOXs use arachidonic acid as the primary substrate, which once released from the mammalian membrane through the action of PLA2 or a combination of other phospholipases
[7], can be metabolized into leukotrienes (LT), lipoxins, or other eicosanoids.
Leukotrienes and 5-HETE (hydroxyeicosatetraenoic acid) produced via the 5-
LOX pathway have been shown to be active in promoting asthma and allergic
99 airway inflammation [8]. Inhibitors of the 5-LOX pathway have chemopreventive
abilities in animal lung carcinogenisis [9, 10] and block the oxidation of several
potent carcinogens [11]. 5-HETE has the ability to stimulate growth of lung
cancer cells [12], prostate cancer cells [13], and 5-LOX inhibitors have the ability
to decrease cell proliferation and trigger apoptosis [14]. 12(S)-HETE, the major
metabolite of the 12-LOX pathway, has been shown to correlate with metastatic
potential [15] and stimulate the expression of integrin receptors leading to
increased tumor cell adhesion [16]. Also, 12(S)-HETE can activate PKC, which
mediates the secretion of cathepsin B, a cysteine protease that has been shown
to be involved in tumor metastasis and invasion of colon cancer cells [17]. In
prostate cancer patients, elevated 12-LOX mRNA levels were shown to correlate
with poor differentiation and cancer cell invasiveness [18]. Additionally, 12-LOX
in human prostate carcinoma stimulates angiogenesis and tumor growth [19].
Recently it was shown that 5-HETE and 12(S)-HETE directly stimulate pancreatic cell proliferation and that LOX inhibitors can induce apoptosis and cell differention [20]. 15-LOX and its products have been linked to cell maturation and differentiation [21] as well as implicated in several aspects of atherosclerosis
[22, 23]. As a whole, mammalian LOXs and products produced by substrate metabolism play significant roles in cancer cell growth, metastasis, invasiveness, and cell survival.
Many lipoxygenase inhibitors are currently being studied for their anti- carcinogenic properties. Curcumin is a naturally occurring polyphenolic phytochemical isolated from the powdered rhizome of the plant Curcuma longa.
100 Curcumin has long been known to possess anti-inflammatory properties and is a
commonly used spice in Asia. It has more recently been reported to inhibit
tumorigenesis in mice [24]. Further, curcumin has the ability to decrease the
formation of 5(S)-, 8(S)-, 12(S)-, and 15(S)-HETE in mouse epidermis [25].
Here we investigate the fate of curcumin with soybean lipoxygenase L3.
Another group has studied the soybean L1 catalyzed oxygenation of curcumin
[26] and shown that curcumin can act as a lipoxygenase substrate. Their data and the data presented herein, using soybean L3, lead to the assumption that curcumin can inhibit lipoxygenase activity by blocking the active site. By use of
X-ray diffraction and mass spectrometry, we have found an electron mass located near the soybean L3 catalytic site. This mass appears to be an unusual degradation product of curcumin. Understanding how curcumin interacts with soybean L3 may explain how curcumin inhibits lipoxygenases and HETE formation. Due to the lack of structural data for human LOXs, researchers are still modeling human LOXs using soybean enzymes because of their availability and highly characterized structures. Use of plant lipoxygenases to model mammalian LOXs will prove highly beneficial and aid in structural characterization, mechanism elucidation, and possibly the discovery of novel inhibitors of LOXs.
101
Materials and Methods
Obtaining the Starting Structure. All molecular modeling and structure
visualizations were done on a SGI workstation using the InsightII program package from MSI [27]. Atomic coordinates of soybean L3 were as deposited in the Protein Data Bank (PDB entry lLNH). Hydrogen atoms were added with appropriate charges assigned throughout the molecule of lipoxygenase assuming physiological pH 7.4. Partial and formal charges were assigned accordingly to the extensible systemic force field (esff).
Docking. The docking module enables the calculation of non-bonded energy between molecules assuming that fragments of the molecules are flexible. This program uses a score (Ludi) to quantify ligand-receptor binding for a fully energy minimized structure. The following parameters were used: radius of subset from Fe atom = 12 Å, maximum R change = 6 Å, maximum number of structures minimized = 7 (this number was usually higher since many different initial structures produced the same minimized structure), minimum steps = 100 or less if the maximum derivative was smaller than 0.01 kcal/mol/Å, MC temperature = 20, energy tolerance = 5000. Following this step, all structures were subject to simulated annealing where the temperature was raised to 500K
then gradually reduced to 300K, after which 5 structures with a minimum
102 potential energy were subjected to molecular dynamics for 1000 picoseconds.
The structure with the minimum potential energy was accepted as the most
probable one.
The Ludi scoring method of interactions between a protein and its ligand
was used to quantify the binding characteristics of curcumin to lipoxygenase.
The Ludi method for de novo design of ligands for proteins (i. e. enzyme
inhibitors) is a method for screening a large number of compounds by analyzing
the geometrical fit of given chemicals in the designated protein binding site.
Other determinants of good binding are also calculated and include hydrogen
bond formation, lipophilic interactions, ionic interactions, and acylic interactions.
However, Ludi can also score protein ligand interactions by statistically
evaluating the fit of all potential ligands determined by the Docking module.
Ludi Score = -73.33mol/kcal ∆G
where: ∆G = ∆Go + ∆Ghbf(∆R)f(∆α) + ∆Gionf(∆R)f(∆α) + ∆GlipoAlipo + ∆GrotNR ∆G;
∆Go represents the contribution to the binding energy that does not directly
depend on any specific interactions with the receptor (i. e. the contribution to
binding energy due to loss of transitional and rotational entropy of the fragment),
∆Ghb and ∆Gion represent the contribution from an ideal hydrogen bond and unperturbed ionic interactions respectively, ∆Glipo represents the contribution from
103 lipophilic interactions which is proportional to the lipophilic surface Alipo , ∆Grot
represents the contribution due to freezing of internal degrees of freedom in the
fragment, NR is the number of acylic bonds, ∆R is the deviation of the hydrogen bond length from the ideal value of 1.9 Å, ∆α is the deviation of the hydrogen bond angle from the ideal value of 180o. In general, a higher Ludi score (0-1100
in range) represents higher affinity and stronger binding of a ligand to the
receptor
In addition, the Ludi score can be related to the dissociation constant Ki.
Ludi Score = -100 log Ki
Isolation of soybean lipoxygenase (cutlivare Beeson 80) was done as
previously described [5]. Fractions from a chromatofocusing column were
collected using a Gradi-Frac machine (Pharmacia Biotech; Piscataway, NJ) and
the appropriate peek was concentrated using Centricon concentrators (Millipore;
Bedford, MA). The concentrated protein solution was dialyzed against Tris buffer
(pH 7.0) to remove the histidine buffer and was purified to a single band by SDS-
PAGE.
Crystallization and data collection. Protein crystals were grown using the
“sitting drop” method as described before [5]. Curcumin was dissolved in ethanol
and added to the crystallization dishes so the final concentration of protein to
104 curcumin was approximately 1:1 with ethanol <2% (v/v). The crystals became pale yellow and data was collected at room temperature using a RAXIS IV imaging plate detector with a Cu rotating anode and focusing mirrors. Crystal-to- detector distance was set at 140 mm with an exposure time of 12 minutes per frame and 2° oscillation. To avoid as much bias as possible, crystals of approximately the same size (~0.5 mm) and shape were used to test varying soaking times and total exposure. After 45-90 min (depending on crystal size), the crystal exposed to X-rays had changed to a purple color. This change of color was especially easy to notice on crystals bigger than the diameter of the collimator (0.5 mm), where only the portion of the crystal irradiated by the X-ray beam turned purple with the rest of the crystal remaining yellow. This phenomenon was not observed in the curcumin solution when exposed to X-ray or in the crystal (no curcumin) when exposed to ambient light. The crystal soaked in curcumin remained purple for several hours following X-ray exposure and was still a light pink 24 hours following. The reflections corresponding to the
“purple” phase were processed and integrated using Denzo and Scalepack [28] resulting in three data sets corresponding to 15, 48, and 70 hours of soaking time with not more than 6 hours of total X-ray exposure per crystal. The data extended to a 2.1–2.2 Å resolution with 92-95% completeness and 6-8% of
Rmerge per set.
Mass spectroscopy. L3 crystals soaked with curcumin not exposed to X- ray and crystals after exposure to X-ray were dissolved in deionized water and
105 analyzed by mass spectrometry at the Protein Structure Facility, University of
Michigan, Ann Arbor, Michigan. Electrospray Ionization (ESI) mass spectroscopy was done using the VG Fisons "Platform" single quadrupole mass spectrometer
(m/z limit 0-3,000). Samples were introduced into the mass spectrometer as H2O solutions by flow injection at 5 microliters/min. Samples were examined in positive ion mode to look for protonated ions in the 0-3000 m/z region, which was later electronically refined to give an expanded region of 0-500 m/z. It is important to emphasize that in some cases molecules carry positive charges even without obvious chargeable sites, possibly because the charging can be affected by gas phase proton affinities [29-32].
Results and Discussion
Molecular simulations. The size of the soybean L3 molecule exceeds the dimensions of the program for molecular simulations; therefore, the protein was restricted to radius of 20 Å around the proteins non-heme iron. This volume contains several channels enabling movement of the curcumin into the central cavity. Our calculations show that curcumin can bind to lipoxygenase in the central cavity close to the iron (Fig 2). The calculated affinity of curcumin for L3 reaches the high value of 1.06 x 10-10M. This finding supports the hypothesis that the anticancer activity of curcumin could be linked to lipoxygenase inhibition.
Dietary curcumin consumption levels can be high with no known toxic effects, but due to low water solubility only a fraction can enter the circulation. The high
106 affinity of curcumin for L3, as calculated in the Ludi module, indicates that it
possible to inhibit lipoxygenase with a low concentration of curcumin. To verify
our theoretical findings we used X-ray crystallographic analyses.
Soybean lipoxygenase purification. Soybeans can be purchased in large
quantities, thus providing a steady supply of protein and easily reproducible
crystals for X-ray analysis. Two fractions from the top of the soybean L3 peak
were collected and purified (Fig 3) as described in Materials and Methods.
Crystallography of the lipoxygenase-curcumin complex. A change of color
of lipoxygenase solutions has been previously observed and described [33]. It
has been observed in lipoxygenase crystals in the presence of peroxides, such
as 13-hydroperoxy-9,11-octadecadienoic acid (13HPOD) and cumene
hydroperoxide [34]. The same phenomenon was noticed (i.e. the color change to
blue/purple) with the iron complex [Fe(bppa)(t-BuCOO)]2+ when its t-butyl carboxylic acid ligand is substituted with t-butyl- or cumene hydroperoxide [35].
This color change is associated with the formation of an unstable complex, wherein the peroxy ligand is bound to the iron atom. Our experiments provide evidence that such complexes can exist longer, from several hours to a few days, when “trapped” in a crystal. In the case of the curcumin-L3 complex, the change of color was not observed in the crystals not illuminated by X-rays. This leads to the assumption that a photodynamic reaction is necessary for the change of color to occur. It is known; however, that curcumin can be photobleached, especially
107 by shorter wavelengths of the light [36, 37]. The radical change in the crystals
color, from yellow to purple, occurs only under X-ray illumination in both ambient
light and in the dark. This leaves no doubt about the photodynamic
characteristics of the observed phenomenon. Also, since the reaction takes
place in the crystal, it is obvious that the curcumin molecule would have to be
near the iron atom and in such an orientation that allows binding between the
iron and the created peroxide.
The crystal unit cell was isomorphous monoclinic C2, (a=112.8, b=137.3,
c=61.9Å, β=95.5o) with that of the wild enzyme (a=112.8, b=137.4, c=61.9Å,
β=95.6o). The structure of this complex was solved by molecular replacement
using native L3 as the starting model (PDB entry lLNH with subtracted H2O
molecules). The electron density maps (resolution 8-2.2 Å, 36325 reflections,
R=24% with only protein atoms included) clearly show an unoccupied electron
density near the iron atom. This mass is much too large to be a solvent molecule
but smaller than curcumin. The shape and location of the mass do not agree
with the forcefield calculated position of curcumin that underwent
photodegradation during exposure to X-rays (Fig 4). The shape and volume of
the unoccupied electron density in the immediate vicinity of the iron suggests the
presence of a peroxide (Fig.5, molecule a). Our map does not show any
evidence of the “prostaglandin-like” molecule (Fig. 5, molecule f) which was
found as a product of curcumin oxidation catalyzed by soybean L1 [26].
Mass spectroscopy. As expected, the mass spectrogram revealed
numerous peaks that can be attributed to chemicals present in the media used in
108 the crystallization of L3. However, at least four fragments with masses of 140,
157, 176, and 198 appeared on the spectrum from X-ray irradiated crystals but
are absent on the spectrogram obtained by crystals not exposed to X-rays.
These peaks could be identified as potential products of the photodegradation of
curcumin and are presented in Fig 5. The strongest candidate is molecule 5a, a
peroxide that, while unstable in complex with L3, can easily convert to molecule
5b. Signals of 140 and 157 present on the mass spectrogram strongly support
this assumption. If curcumin were oxygenated leading to the “prostaglandin-like”
product (Fig. 5, molecule f), it would be present in both spectra, due to the fact
that the experiment was conducted in the presence of oxygen and not in
anaerobic conditions. Lack of any outstanding peak in this region (~400)
indicates that either L3 does not react as L1 or, more likely, that such a molecule
cannot be formed under the current experimental conditions without
supplemental oxygen. The identity of all other compounds (Fig. 5, molecules c, d, and e) cannot be determined with absolute certainty and are outside scope of this paper.
The structural details of the complex between the photoproduct of curcumin degradation and L3 will be published upon completion of the crystallographic refinement. Several conclusions; however, are quite clear and can be included herein: 1) curcumin can penetrate L3 and bind in the central cavity near the iron atom, 2) force field calculations predict a very low dissociation constant of 10-10 M suggesting an very high affinity of curcumin as a lipoxygenase ligand making it a good candidate for an agonist/antagonist in drug
109 design, 3) curcumin does not undergo L3 catalyzed oxidation and the complex is stabile under normal conditions; although, 4) X-ray irradiation elicits a photodynamic reaction leading to the degradation of curcumin and the formation of a metastabile complex consisting of L3 and a peroxy photoproduct, and 5) a certain period of time is necessary for the photodynamic reaction to produce enough peroxide molecules to react with L3 and for the color change characteristic of the complex to occur.
We hope that these observations may prove beneficial in the treatment and/or prevention of ailments where lipoxygenases are involved.
Acknowledgments
This paper was supported in part by grants from American Diagnostica,
Inc., Greenwich, CT and the OBR Research Challenge Grant.
110 References
1. Boyington, J.C., B.J. Gaffney, and L.M. Amzel, Structure of soybean
lipoxygenase-I. Biochemical Society Transactions, 1993. 21(Pt 3)(3): p.
744-8.
2. Minor, W., et al., Crystal structure of soybean lipoxygenase L1 at 1.4 A
resolution. Biochemistry, 1996. 35(33): p. 10687-701.
3. Prigge, S.T., et al., Structure conservation in lipoxygenases: structural
analysis of soybean lipoxygenase-1 and modeling of human
lipoxygenases. Proteins, 1996. 24(3): p. 275-91.
4. Gillmor, S.A., et al., The structure of mammalian 15-lipoxygenase reveals
similarity to the lipases and the determinants of substrate specificity
[published erratum appears in Nat Struct Biol 1998 Mar;5(3):242]. Nature
Structural Biology, 1997. 4(12): p. 1003-9.
5. Skrzypczak-Jankun, E., et al., Structure of soybean lipoxygenase L3 and a
comparison with its L1 isoenzyme. Proteins, 1997. 29(1): p. 15-31.
6. Kuhn, H. and B.J. Thiele, The diversity of the lipoxygenase family. Many
sequence data but little information on biological significance. FEBS
Letters, 1999. 449(1): p. 7-11.
7. Needleman, P., et al., Arachidonic acid metabolism. Annual Review of
Biochemistry, 1986. 55: p. 69-102.
111 8. Irvin, C.G., et al., 5-Lipoxygenase products are necessary for ovalbumin-
induced airway responsiveness in mice. American Journal of Physiology,
1997. 272(6 Pt 1): p. L1053-8.
9. Moody, T.W., et al., Lipoxygenase inhibitors prevent lung carcinogenesis
and inhibit non-small cell lung cancer growth. Experimental Lung
Research, 1998. 24(4): p. 617-28.
10. Rioux, N. and A. Castonguay, Inhibitors of lipoxygenase: a new class of
cancer chemopreventive agents. Carcinogenesis, 1998. 19(8): p. 1393-
400.
11. Kulkarni, A.P., Y. Cai, and I.S. Richards, Rat pulmonary lipoxygenase:
dioxygenase activity and role in xenobiotic metabolism. International
Journal of Biochemistry, 1992. 24(2): p. 255-61.
12. Avis, I.M., et al., Growth control of lung cancer by interruption of 5-
lipoxygenase-mediated growth factor signaling. Journal of Clinical
Investigation, 1996. 97(3): p. 806-13.
13. Anderson, K.M., et al., The selective 5-lipoxygenase inhibitor A63162
reduces PC3 proliferation and initiates morphologic changes consistent
with secretion. Anticancer Research, 1994. 14(5A): p. 1951-60.
14. Ghosh, J. and C.E. Myers, Inhibition of arachidonate 5-lipoxygenase
triggers massive apoptosis in human prostate cancer cells. Proceedings of
the National Academy of Sciences of the United States of America, 1998.
95(22): p. 13182-7.
112 15. Tang, D.G. and K.V. Honn, 12-Lipoxygenase, 12(S)-HETE, and cancer
metastasis. Annals of the New York Academy of Sciences, 1994. 744: p.
199-215.
16. Tang, D.G., et al., 12(S)-HETE promotes tumor-cell adhesion by
increasing surface expression of alpha V beta 3 integrins on endothelial
cells. International Journal of Cancer, 1993. 54(1): p. 102-11.
17. Honn, K.V., et al., A lipoxygenase metabolite, 12-(S)-HETE, stimulates
protein kinase C-mediated release of cathepsin B from malignant cells.
Experimental Cell Research, 1994. 214(1): p. 120-30.
18. Gao, X., et al., Elevated 12-lipoxygenase mRNA expression correlates
with advanced stage and poor differentiation of human prostate cancer.
Urology, 1995. 46(2): p. 227-37.
19. Nie, D., et al., Platelet-type 12-lipoxygenase in a human prostate
carcinoma stimulates angiogenesis and tumor growth. Cancer Research,
1998. 58(18): p. 4047-51.
20. Ding, X.Z., et al., Lipoxygenase inhibitors abolish proliferation of human
pancreatic cancer cells. Biochemical & Biophysical Research
Communications, 1999. 261(1): p. 218-23.
21. van Leyen, K., et al., A function for lipoxygenase in programmed organelle
degradation. Nature, 1998. 395(6700): p. 392-5.
22. Yla-Herttuala, S., et al., Colocalization of 15-lipoxygenase mRNA and
protein with epitopes of oxidized low density lipoprotein in macrophage-
113 rich areas of atherosclerotic lesions. Proceedings of the National Academy
of Sciences of the United States of America, 1990. 87(18): p. 6959-63.
23. Kuhn, H. and L. Chan, The role of 15-lipoxygenase in atherogenesis: pro-
and antiatherogenic actions. Current Opinion in Lipidology, 1997. 8(2): p.
111-7.
24. Huang, M.T., H.L. Newmark, and K. Frenkel, Inhibitory effects of curcumin
on tumorigenesis in mice. Journal of Cellular Biochemistry - Supplement,
1997. 27: p. 26-34.
25. Huang, M.T., et al., Inhibitory effects of curcumin on in vitro lipoxygenase
and cyclooxygenase activities in mouse epidermis. Cancer Research,
1991. 51(3): p. 813-9.
26. Schneider, C., et al., 2-[(4"-Hydroxy-3'-methoxy)-phenol]-4-(4"-hydroxy-3"-
methoxyphenyl)-8-hydroxy-6-oxo-3-oxabicyclo[3.3.0]-7-octene: Unusual
Product of the Soybean Lipoxygenase-catalyzed oxygenation of curcumin.
Journal of Molecular Catalysis B: Enzymatic, 1998. 4: p. 219-227.
27. InsightII, ver. 95.0/3.0.0, User's guide. 1995, San Diego, CA, USA:
Molecular Simulations, Inc.
28. Otwinowski, Z. and W. Minor, Methods of Enzymology, 1997(276): p. 307-
326.
29. Whitehouse, C.M., et al., Electrospray interface for liquid chromatographs
and mass spectrometers. Analytical Chemistry, 1985. 57(3): p. 675-9.
114 30. Covey, T.R., et al., The determination of protein, oligonucleotide and
peptide molecular weights by ion-spray mass spectrometry. Rapid
Communications in Mass Spectrometry, 1988. 2(11): p. 249-56.
31. Chowdhury, S.K., V. Katta, and B.T. Chait, An electrospray-ionization
mass spectrometer with new features. Rapid Communications in Mass
Spectrometry, 1990. 4(3): p. 81-7.
32. Smith, R.D., et al., New developments in biochemical mass spectrometry:
electrospray ionization. Analytical Chemistry, 1990. 62(9): p. 882-99.
33. Nelson, M.J., D.B. Chase, and S.P. Seitz, Photolysis of "purple"
lipoxygenase: implications for the structure of the chromophore.
Biochemistry, 1995. 34(18): p. 6159-63.
34. Skrzypczak-Jankun, E., et al. "Purple" Lipoxygenase - X-Ray Analysis of
Complexes with Three Different Peroxides. in American Crystallographic
Association Meeting. 2000. St. Paul, MN.
35. Wada, A., et al., Synthesis and Characterization of Novel Alkylperoxo
Mononuclear Iron(III) Complexes with Tripodal Pyridylamine Ligand: A
Model for Peroxo Intermediates in Reactions Catalyzed by Non-Heme Iron
Enzymes. Inorganic Chemistry, 1999. 38: p. 3592-3593.
36. Gorman, A.A., et al., Curcumin-derived transients: a pulsed laser and
pulse radiolysis study. Photochemistry & Photobiology, 1994. 59(4): p.
389-98.
37. Chignell, C.F., et al., Spectral and photochemical properties of curcumin.
Photochemistry & Photobiology, 1994. 59(3): p. 295-302.
115 Figure Legends
Fig 1. A Comparison of soybean L3 (blue) and rabbit 15-LOX (yellow) 3
dimensional structures. Iron is shown as a brown sphere. Please note the close
similarities in topology despite the differences in size (L3: 857 residues; rabbit
15-LOX: 663 residues).
Fig. 2. Curcumin shown in the central cavity of soybean L3. Iron is shown in
brown color, carbon in green, oxygen in red, hydrogen in white. L3 is shown as a
ribbon model with the front part removed from the figure for clarity (theoretical
predictions based on force field calculations).
Fig. 3. The absorbance of different fractions during the purification of soybean
L3. Only fractions indicated by squares on the graph were used for
crystallization. The insert to the left of the peek is PAGE of this fraction, purified
to >95%.
Fig. 4. L3 molecule with the observed unoccupied electron density (mesh) near
the iron atom and a curcumin molecule (bold) in the position predicted by the
force field calculations. The size, volume and position of the unoccupied (Fo-Fc) difference map indicates the presence of the photodegradation product of curcumin corresponding to molecule ‘a’ in Fig.5.
Fig. 5. Possible products of curcumin photodegradation (a-e) and a suggested product of curcumin oxidation catalyzed by soybean L1 (f).
Fig 6. Mass spectra of soybean L3 crystals soaked with curcumin, not
illuminated (N) and illuminated (X) by X-rays during data collection (expanded
region from 100-500m/z).
116 Fig. 1
117 Fig. 2
118 Fig. 3
119 Figure 4
120 Fig. 5 a b CH3 CH3 O O O O OH
HO O 4-hydroperoxy-2- methoxyphenol 2-methoxycyclohexa-2,5-diene- 156.13 1,4-dione 138.1 c d CH3 O CH3 OH O O
OH HO HO
4-((1E)buta-1,3-dienyl)-2- 3-(4-hydroxy-3-methoxyphenyl) methoxyphenol propane-1,2-diol
178.17 198.21 e f H C CH O 3 O 3 OH O OH H O HO O 3-(4-hydroxy-3-methoxyphenyl) H O propanoic acid
196.20 O HO
H3C HO (4S,6S)-3-hydroxy-6-(4-hydroxy-3- methoxyphenyl)-4-(4-hydroxy-3- methoxyphenoxy)-4,5,6,3a,6a- pentahydro-5-oxopentalen-1-one 400.39
121 Fig. 6
122 Manuscript 3
(Published in the International Journal of Molecular Medicine, Volume 12, Number 1, July 2003)
Structure of curcumin in complex with lipoxygenase and its
significance in cancer.
EWA SKRZYPCZAK-JANKUN1, KANGJING ZHOU1, 3, N. PATRICK MCCABE1, STEVEN H. SELMAN1, 2 AND JERZY JANKUN1, 2.
Medical College of Ohio, 1Urology Research Center, Department of Urology, 2Physiology and Molecular Medicine, Toledo, OH 43614-5807, 3Present address: Institute of Biotechnology, Fozhou University, Fozhou, Fujian 350002, China.
Key Words: lipoxygenase, curcumin, X-ray structure, cancer, angiogenesis, chemoprevention.
Correspondence to: Dr Jerzy Jankun, Urology Research Center, Medical College of Ohio, Toledo, OH 43614-5807, USA E-mail: [email protected] Dr Ewa Skrzypczak-Jankun, Urology Research Center, Medical College of Ohio, Toledo, OH 43614-5807, USA E-mail: [email protected]
123 Abstract
Scientific research provides documented evidence that fatty acid metabolites have profound impacts on carcinogenesis. Intervention into dioxygeneses pathways might therefore effect development, metastasis and progression of many types of cancers. This work delivers the first 3D structural data of the interaction between curcumin and the fatty acid metabolizing enzyme, soybean lipoxygenase-3. Curcumin binds to lipoxygenase in a noncompetitive manner. Trapped in complex with soybean lipoxygenase-3, curcumin undergoes
X-ray induced photodegradation, but utilizes the enzyme’s catalytic ability to form the peroxy complex Enz-Fe-O-O-R as 4-hydroperoxy-2-methoxyphenol, that later transforms into 2-methoxycyclohexa-2,5-diene-1,4-dione. Our observations concerning this radiation and time-dependent inhibition add new information about the role that curcumin might play in cancer prevention and treatment.
124 Introduction
Fatty acid (FA) metabolites impact every aspect of cancer. They are
produced when fatty acids, released from the mammalian membrane through the
action of phospholipase A2 (PLA2) or a combination of other phospholipases, are oxidized by cyclooxygenases (COX, which has an iron containing heme moiety) or lipoxygenases (LOX with a free iron cofactor) (1). It is clear that beneficial, therapeutic effects could be achieved through drug-mediated modulation of these metabolic pathways (2). Literature shows that regulating FA metabolism can effect tumor growth, metastasis, invasiveness, cell apoptosis and induction of tumor necrosis factor (TNF) (3, 4). Dioxygenases catalyse oxygenation of FAs utilizing Fe+2/Fe+3 redox potential and molecular oxygen, swapping electrons between intermediate radicals. Natural phytochemicals are very potent antioxidants, free radical scavengers and known inhibitors of dioxygenases, which have been extensively studied to explore their potential utilization in chemoprevention. All flavonoids have one common feature, aromatic moiety(ies) with unprotected hydroxyl groups. Among them, curcumin seems to be in the forefront in the race to the drug market (5). It is a naturally occurring polyphenolic phytochemical isolated from the powdered rhizome of the perennial plant Curcuma longa, cultivated throughout Asia (especially in India, China and
Indonesia). Curcumin (bis(4-hydroxy-3-methoxyphenyl)-1,6-heptadiene-3,5- dione), a major component of the spice ”turmeric” (usually in a mixture with demethoxycurcumin and bisdemethoxycurcumin) produces a characteristic
125 yellow color when added to food. Its use dates back more than 6,000 years.
Today, scientific research provides strong evidence of curcumin’s topical, anti- inflammatory, analgesic, anti-mutagenic, anti-oxidant, immune stimulating and anticancer properties. Its anticancer effectiveness has been tested and proven in animals, cell cultures of skin, colon, small intestine, esophagus, stomach, breast, mouth, bladder, prostate cancers and leukemia (6-11).
Curcumin`s role in controlling the activities of COXs and LOXs, has also been firmly established on a molecular level (2). In the drug development process, attention has been paid to COX inhibition. In particular, COX-2 specific inhibitors, while blocking the production of metabolites via the LOX pathway has been investigated considerably less, and targeting them both simultaneously has not yet been explored.
There is no structural data providing experimental evidence of the three dimensional molecular structure of curcumin in complex with the enzymes of interest. This work provides a three dimensional structure based on X-ray analysis, supported by theoretical modeling and experimental kinetic studies of curcumin interactions with lipoxygenase. In addition, it provides information about radiation induced and enzymatically catalyzed oxidation of curcumin, leading to a catechol-like photoproduct. Curcumin’s photosensitivity is considered very useful in prevention of skin cancer (2). Blocking the LOX pathway helps to diminish inflammatory side effects in radiation treated patients
(3). Therefore, our findings might be also of value for that aspect of cancer research.
126 Materials and Methods
Obtaining the Starting Structure. All molecular modeling and structure
visualizations were done on a SGI workstation using the InsightII program package from Accelrys, Inc. or Sybyl 6.8 from Tripos (12). Atomic coordinates of soybean LOX-3 were as deposited in the Protein Data Bank (PDB entry lLNH).
Hydrogen atoms were added with appropriate charges assigned throughout the molecule of lipoxygenase assuming physiological pH 7.4. Partial and formal charges were assigned accordingly to the extensible systemic force field (esff).
Docking. The docking module enables the calculation of non-bonded energy between molecules assuming that fragments of the molecules are flexible. The following parameters were used: radius of subset from Fe atom =
12 Å, maximum R change = 6 Å, maximum number of structures minimized = 7
(this number was usually higher since many different initial structures produced the same minimized structure), minimum steps = 100 or less if the maximum derivative was smaller than 0.01 kcal/mol/Å, MC temperature = 20, energy tolerance = 5000. Following this step, all structures underwent simulated annealing, where the temperature was raised to 500K then gradually reduced to
300K. After that, 5 structures with a minimum potential energy were subjected to molecular dynamics for 1000 picoseconds, and the structure with minimum potential energy was accepted as the most probable one.
127 Protein purification. Proteins were extracted from soybean seeds using
the Beeson 80 cultivar, with purification, crystallization, data collection,
preliminary maps and results from mass spectroscopy as described in the
preliminary publication (13). Briefly, fractions from a chromatofocusing column were collected using a Gradi-Frac machine (Pharmacia Biotech; Piscataway, NJ) and the appropriate peak was concentrated using Centricon concentrators
(MWCO 10000, Millipore, Bedford, MA). The concentrated protein solution was dialyzed with Tris buffer (pH 7.0) resulting in a single band by SDS-PAGE.
X-ray analysis and refinement. Curcumin in ethanol was added to the
crystallization media (storing solution) in a 1:1 molar ratio relative to protein, but in an amount not to exceed 2% (v/v) in the storage container. Crystals soaked for different time periods (15, 48, and 72 hours) were analyzed at room temperature on a Raxis dual plate detector with a rotating copper (Cu) anode and
focusing mirrors [resolution 2.2 Å (last shell 2.28-2.20 Å), total number of
independent output reflections: 45106, 43643, 44610, completeness: 95.5, 92.6,
94.7 (last shell 90.9, 87.0, 86.0)%, Rmerge: 7.0, 7.2, 7.5%, reflections with I>σ: 88,
86, 86% respectively] and did not show any difference in the electron density maps. Cryogenic conditions were explored but not pursued due to the obvious disadvantages in a reduced data/variable ratio (14). The data collected from crystals soaked for approximately 15 h, which had the best statistics, were then used for all further refinement. All crystals placed in the X-ray beam changed color from yellow to purple after a lag phase of 45-90 min. depending on crystal
128 size. The “purple” phase lasted at least 6 h without any visible change, and slowly bleached away with time, to a slightly pink color after ~12 h, eventually becoming colorless. Any measurements corresponding to the lag phase were rejected. Data considered for processing were arbitrarily restricted to the 6 h
“purple” phase exposure and two crystals from the same batch were used to achieve completeness and a desirable intensity statistic. Coordinates of the native enzyme (PDB code 1LNH) were used as a model for molecular replacement. The maps calculated after rigid body refinement clearly revealed that the molecule near the active site was not curcumin but a photodegradation product. Different molecules were tested in refinement against X-ray data and 4- hydroperoxy-2-methoxyphenol was assigned as the iron ligand during the
“purple” phase. The structure has been refined, using overall anisotropic B-factor and bulk solvent corrections, followed by positional and Bj crystallographic refinement intertwined with map examination after each round, along with manual corrections to the model for a better fit into the electron density. Water molecules were added at the final stage and the structure refined to convergence with the results presented in Table 1. All calculations and graphical evaluations were done using X-plor ver.3.85, Chain ver.7 and InsightII ver.98.0 (12) on a SGI
Indygo2 Extreme workstation running under Irix ver. 6.5. The atomic coordinates and structure factors have been deposited in the Protein Data Bank, code 1HU9.
129 Results and Discussion
For our studies (in the lack of a highly purified human enzyme) we used
LOX-3 from soybeans. Despite the difference in the number of amino acids between plant and mammalian LOXs, these proteins are similar in topology with high similarities in the active site of these enzymes. Comparison of soybean enzymes with rabbit 15-LOX (15) (PDB entry 1LOX), allows us to point out which residues might be of similar interest for binding in mammalian enzymes (Table
2). Although soybean and rabbit cannot be exactly superimposed, the best alignment accordingly to their 3D models from X-ray analyses show which residues occupy the same space in the respective LOX structure (despite the fact that they do not align in sequence). It is believed that all LOXs follow the same catalytic mechanism, however; it is probably the vicinity of the iron site that determines the regio- and stereospecificity of the particular enzyme.
The positioning of curcumin within the enzyme molecule was deduced by
force field calculations. Accordingly, curcumin goes into the central cavity with its
diene-dione part near iron, in the enol or keto form, with the keto being ~1 kcal
mol-1 energetically less favorable. The calculated affinity was in the range of 10-
10M for both forms of curcumin. The best two theoretical results show the
aromatic moieties in the same location and orientation, with their methoxy and
hydroxy groups reaching out into hydrophilic pockets with Tyr207, Arg562,
Asn558 and Glu527 on one side, and Asp766, Thr575 and water molecules on
the other (Fig. 1). The central part of the molecule can have C=O and C-OH
130 pointing in the same direction (enol form) or two carbonyl groups looking in the
opposite direction. In the first case, the molecule is bent, =CH-(C=O)-(-148°)-
CH=(C-OH)-CH=, with the hydroxy group less than 3.0 Å from Fe. In the other case, the torsion angle is 0°, indicating a strong possibility for the different tautomeric forms and easy delocalization of electrons within double bond system.
In the”‘keto” model, one of the central oxygens points toward Trp519 and can be within hydrogen bonding distance to NE1 from this residue of known flexibility in the structure (16). However, the enol form can provide bidentate participation in the coordination, it is energetically more stable and observed in the X-ray structure of curcumin itself (17, 18).
The kinetic studies on soybean LOX-3 and curcumin have shown a non- competitive mechanism of inhibition (see Table 3) (19). Also, it clearly shows that micromolar quantities are required for inhibition. Due to limitations of the programs used, the “esff” calculations were done using a sphere of 12 Å from Fe, with an incomplete molecule (LOX-3 molecule resembles an ellipsoid with axes roughly 90, 60, and 50 Å), thus only the central cavity with the active site and the channels leading there were examined. Also, the central cavity, where the inhibitor binds, is roughly 12x10x9 Å in dimension and its contents cannot be described even at 100K (PDB entry 1YGE). The discrepancy between experimental and theoretical values result from the inadequacy of the theoretical model, which may be due to inappropriate solvation and charge distribution.
Hence, kinetic data are more reliable in evaluating affinity than theoretical modeling. 131 X-ray analysis was meant to provide the first experimental 3D structural
data about a LOX enzyme in complex with curcumin. However, curcumin
undergoes a photochemical reaction in the X-ray beam when trapped within
LOX. What we observed is soy LOX-3 complexed with an X-ray induced
oxidation and degradation product. The crystallization, data collection and
electrospray mass spectroscopy (EMS) were published previously (13). An
electron density distribution shown by X-ray analysis and EMS provided evidence of the presence of a molecule(s) smaller than curcumin, following exposure to X- rays. The photochemical reaction does not occur under normal illumination, but is initiated by X-ray radiation, causing a pronounced change in the crystal color from yellow to purple. It has been found in other experiments ((16) and other references therein) that the purple form of lipoxygenase corresponds to the complex Enz-Fe-O-O-R. A typical yellow crystal with dimensions 0.50x0.20x0.05 mm (0.005 mm3), requires at least 45 min. irradiation (at 100 mA, 50 kV level, Φ
0.3 mm collimator) for the purple color to be visible in the detectors optical microscope. This color can be observed for at least 6 h after irradiation, after which the color bleaches. The crystals often retain a slightly pinkish color for another several hours, but do not revert to the yellow color observed before X-ray exposure.
Curcumin undergoes photolysis in the laser beam and cytotoxicity is greatly enhanced by light (20). The photodegradation products vanillin and ferrulic acid are not responsible for such toxicity, thus, it could be due to the radiation induced active oxygen species such as singlet oxygen, hydroxyl
132 radicals, superoxides or hydrogen peroxide (20-22). Singlet oxygen was hardly
noticeable when curcumin was solubilized in an aqueous environment (22).
Nevertheless, it has been suggested that in biological systems singlet oxygen,
superoxide and other photodegradation products may all participate in the
photodynamic action of curcumin depending on the environment of this
compound. Curcumin itself is stable in the X-ray beam and its crystal structure
has been determined at two different temperatures (295K (17), and 121K (18)).
In both cases, the molecule shows that the perfectly delocalized central bonds are coplanar with one trans >C=C< system and the adjacent aromatic moiety.
The other trans >C=C< plane is twisted in relation to the first and not coplanar with its nearby phenyl ring. The corresponding torsion angles are given in Fig. 2.
Soybean lipoxygenase-catalyzed dioxygenation of curcumin, in the presence of oxygen but in the absence of X-ray or laser light, results in the formation of 2-[4’- hydroxy-3’-methoxy)phenoxy]-4-(4”-hydroxy-3”-methoxyphenyl)-8-hydroxy-6-oxo-
3-oxabicyclo[3.3.0]-7-octene (Fig. 2, molecules 2, 3, 4), with four chiral carbons at its central tetrahydrofuran ring (23). This experiment, which did not involve irradiation, resulted in undetectable hydroperoxide production. Therefore, the photoreaction observed herein must be induced by X-rays, which in the presence of lipoxygenase stimulate a redox reaction leading to the formation of peroxide and the metastable “purple” LOX. This reaction has an obvious lag phase at the beginning, which must correspond to the breaking and oxidation steps of curcumin’s 1,6-heptadiene-3,5-dione system. Although it is unclear how this occurs and there is no experimental evidence for these transition steps, structural
133 results show the molecule, C7H8O4, as a peroxide bound to Fe. This formula can correspond to two possible structures (Fig. 2, molecules 5 and 6): 4-hydroperoxy-
2-methoxyphenol, or 4-hydroperoxy-2-methoxycyclohexa-2,5-diene-1-one. Both peroxides can convert into the stable compound (molecule 7) 2- methoxycyclohexa-2,5-diene-1,4-dione. Examination and refinement of both models (molecules 5 and 6) against X-ray data were inconclusive (in terms of R- factors). These compounds are very different in nature, but in the electron density map and at this resolution, the only noticeable difference is C4, which in molecule 6 is chiral. The shape of the electron density suggests the C4-O bond to be coplanar with the ring, therefore molecule 5 was assigned as the molecule present in this “purple” complex. Bleaching of the purple color signals the transformation of the peroxy compound Enz-Fe-O-O-R into a molecular complex
Enz-Fe with 2-methoxycyclohexa-2,5-diene-1,4-dione, the presence of which has been confirmed by mass spectroscopy as well. Figure 3 shows the electron density map (2Fo-Fc) with the molecular structure resulting from X-ray analysis.
Fig. 4 presents its superposition with LOX-3 in complex with its natural metabolite 13-(S)-hydroperoxy-9,11-(cis,trans)-octadecadienoic acid 13(S)-
HPODE ((16), PDB entry 1IK3), and with inhibitor, 4-nitrocatechol ((24), PDB entry 1BYT). In the native soybean enzyme, iron cation (Fe+2) has 3 histidines and the enzyme carboxylic terminus as ligands, with asparagine side chain amide group and a water molecule occupying two places in the iron octahedral coordination, but at non-binding distances. Rabbit LOX-15 has a fourth His residue instead of Asn with the same type of coordination ((15), PDB entry
134 1LOX). Since these residues are conserved in other mammalian species, it may
be assumed that human lipoxygenases have the same type of coordination.
Comparison with the native enzyme ((25), PDB entry 1LNH), the other “purple”
form of LOX-3 in complex with 13(S)-HPODE, and the complex with 4-
nitrocatechol, brings the following observations:
a) The photoproduct of curcumin binds as a peroxide in the central cavity of lipoxygenase in a Enz-Fe-O-O-R fashion (Fe-O 2.0 Å), with minor changes in
the positioning of side chains of the nearest residues: Gln514, Val566 and Ile572
(when compared to native enzyme). The hydrogen bonds between residues
surrounding the iron site are preserved. The electron density between Gln514
and inhibitor (Fig. 3) indicates π…π electron interaction between C=OE1 and the aromatic moiety of molecule 5 or inhibitor’s C=O (molecules 6 or 7), rather than a weak hydrogen bond (3.1 Å) between Gln side chain and hydroxyl. This might be especially useful in stabilizing the inhibitor in its later - 1,4-dione – form upon
“bleaching”.
b) In comparison, the “purple” complex containing a 13(S)-HPODE ligand
(Fig. 4, purple molecule), one can see that the peroxy group binds in the same manner, with the positioning of the aromatic moiety of the curcumin photoproduct corresponding to the trans, cis double bond system in the fatty acid metabolite.
c) The 4-nitrocatechol structure (Fig.4, yellow molecule) may be the best approximation to visualize what happens in the “bleached” structure.
d) Overall, rmsd from the native enzyme is 0.78 Å, 0.86 Å from the
“purple” 13HPODE complex, and 0.75 Å from 4-nitrocatechol structure. In all
135 three cases, the noticeable differences are the side chains and the flexible loops at the outskirts of the enzyme molecule. However, the core shows no dramatic changes except for the obvious swing of Gln514 in the 13(S)-HPODE complex, and the adjustment of the hydrophobic residues (Leu773, Ile572, Leu565,
Leu560, Ile557, Leu227) lining this cavity. In addition to Gln514, the most visible changes are in the side chains of Asn713 and His518. Asn713 retains its place in the Fe coordination sphere but with a distance varying from 2.3 Å in the 13(S)-
HPODE, 2.7 Å in this complex, to 3.0 Å in the native enzyme and 3.8 Å in the catechol complex, where it makes strong hydrogen bonds to water attached to
Fe and to His518NE2. In all known structures (soybean, both LOX-1 and its
mutants (26), and LOX-3, deposited in the Protein Data Bank), His518 shows a
much higher mobility than other iron ligands (in terms of Bj). Its imidazole ring
adopts different orientations affecting the distance to iron (2.2 – 2.9 Å) and
participation in hydrogen bonding. This is quite pronounced in the 4-
nitrocatechol complex where both Asn713 and His518 turn to make a network of
H-bonds: 713OD1 …518NE2|518ND1…O10 (4-nitrocatechol) O7…857O. We
hypothesize that the same may take place in the presently reported complex
upon the inhibitor’s transformation from curcumin via a peroxide intermediate to
its quinone form.
Our conclusions are supported by observations of another catecholic
antioxidant and effective LOX inhibitor, nordihydroguaiaretic acid (NDGA). Like
curcuminoids, NDGA derivatives have two catecholic moieties connected by the
aliphatic chain -CH2-CHCH3-CHCH3-CH2-. It has been found that NDGA can be
136 oxidized to a quinone derivative (27, 28). Also, it has been suggested that the
primary potency of NDGA derivatives is associated with their ability to modulate the Fe ion redox potential rather than tight binding to the active side such as in competitive inhibition. Indeed, curcumin itself does not bind to iron and exhibits a non-competitive behavior in LOX-3 inhibition (19).
The unexpected result, peroxidation of curcumin triggered by X-rays and catalyzed by LOX, adds yet another dimension to the possible applications of curcumin or its derivatives in therapy. It has been found that suppressing the production of fatty acid metabolites formed via LOX, but not the COX pathway, could be beneficial in preventing serious inflammatory side affects in patients undergoing radiation therapy (3). In contrast, LOX has a non-heme iron factor, whereas COX has a heme moiety. Studies on porphyrins with iron suggest that a key role of an electronegative substituents to the heme system is to increase the Fe(+3)/(+2) redox potential and thereby accelerate oxidation, making the metal complex a better catalyst for the aerobic oxidation of hydrocarbons by dioxygen (29). It is unclear how curcumin interacts with COX and where it binds, but we know that under normal conditions it prevents the production of thromboxane(s) but not prostacyclin (30). The explanation for this at the atomic level is not known, nor is it known whether this behavior would be the same upon irradiation. In LOX, curcumin not only blocks the fatty acid binding, but binds to iron as a result of a photoreaction, utilizing the catalytic properties of the enzyme on itself rather than on the FA. In time, the peroxide changes to 2-
137 methoxycyclohexa-2,5-diene-1,4-dione, which may inhibit LOX similarly to catechol, and a water molecule.
Curcumin is an excellent example of a natural, nutritional compound that has been used for generations, and rediscovered by modern science. Curcumin is not a selective LOX inhibitor and has been shown to affect several other enzymes. Understanding how curcumin interacts with LOX might allow us to design rational modifications based on its structure in hope of producing selective inhibitors targeting LOX, which might be utilized in cancer therapy.
Acknowledgements
This work was supported in part by grants from: American Diagnostica Inc.,
Greenwich, CT., U.S. Army Medical Research and Materiel Command
(DAMD17-01-1-0553) and Frank D. Stranahan Endowment Fund for Oncological
Research. ESJ thanks Dr. Clarke Slemon from Québépharma Recherche Inc. for helpful discussion concerning curcumin oxidation, and Robert Kernstock for performing kinetic studies.
138 References
1. Needleman P, Turk J, Jakschik BA, Morrison AR, et al: Arachidonic acid
metabolism. Annu Rev Biochem 55: 69-102, 1986.
2. Cuendet M, and Pezzuto JM: The role of cyclooxygenase and
lipoxygenase in cancer chemoprevention. Drug Metabol Drug Interact 17:
109-57, 2000.
3. Hallahan DE, Virudachalam S, Kufe DW, et al: Ketoconazole attenuates
radiation-induction of tumor necrosis factor. Int J Radiat Oncol Biol Phys
29: 777-80, 1994.
4. Chan MM: Inhibition of tumor necrosis factor by curcumin, a
phytochemical. Biochem Pharmacol 49: 1551-6, 1995.
5. Kumar AP, Rajnarayanan R, Garcia GE, et al: Novel curcumin derivatives
induce apoptosis through AKT-NFKB signaling in human prostate cancer
cells. In: Proc 93rd Annual Meeting AACR, pS66, 2002.
6. Sindhwani P, Hampton JA, Baig MM, et al: Curcumin prevents intravesical
tumor implantation of the MBT-2 tumor cell line in C3H mice. J Urol 166:
1498-501, 2001.
7. Dorai T, Cao YC, Dorai B, et al: Therapeutic potential of curcumin in
human prostate cancer. III. Curcumin inhibits proliferation, induces
apoptosis, and inhibits angiogenesis of LNCaP prostate cancer cells in
vivo. Prostate 47: 293-303, 2001.
139 8. Avis IM, Jett M, Boyle T, et al: Growth control of lung cancer by
interruption of 5-lipoxygenase-mediated growth factor signaling. J Clin
Invest 97: 806-13, 1996.
9. Ding XZ, Kuszynski CA, El-Metwally TH, et al: Lipoxygenase inhibition
induced apoptosis, morphological changes, and carbonic anhydrase
expression in human pancreatic cancer cells. Biochem Biophys Res
Commun 266: 392-9, 1999.
10. Ghosh J, and Myers CE: Inhibition of arachidonate 5-lipoxygenase triggers
massive apoptosis in human prostate cancer cells. Proc Natl Acad Sci U S
A 95: 13182-7, 1998.
11. Huang MT, Newmark HL, and Frenkel K: Inhibitory effects of curcumin on
tumorigenesis in mice. J Cell Biochem Suppl 27: 26-34, 1997.
12. Software. Brunger AT. X-plor ver. 3.1. A system for X-ray crystallography
and NMR NH, CT. Sack JS, Chain. A Crystallographic modeling program,
ver. 7, InsightII, ver. 98, Accelrys, Inc., San Diego, CA, Sybyl, ver. 6.8,
Tripos, Inc. St. Louis, MO.
13. Skrzypczak-Jankun E, McCabe NP, Selman SH, et al: Curcumin inhibits
lipoxygenase by binding to its central cavity: theoretical and X-ray
evidence. Int J Mol Med 6: 521-6, 2000.
14. Skrzypczak-Jankun E: Flash-freezing causes a stress-induced modulation
in a crystal structure of soybean lipoxygenase L3. Acta Crystallogr D Biol
Crystallogr 52: 959-65, 1996.
140 15. Gillmor SA, Villasenor A, Fletterick R, et al: The structure of mammalian
15-lipoxygenase reveals similarity to the lipases and the determinants of
substrate specificity. Nat Struct Biol 4: 1003-9, 1997.
16. Skrzypczak-Jankun E, Bross RA, Carroll RT, et al: Three-dimensional
structure of a purple lipoxygenase. J Am Chem Soc 123: 10814-20, 2001.
17. Ishigami YG, Masud M, Takizawa T, et al: The crystal structure and the
fluorescent properties of curcumin. Shikizai Kyokaishi 72: 71-7, 1999.
18. Tonneson HHK and Mostad J: The crystal structure of curcumin. Acta
Chem Scand B 36: 475-80, 1982.
19. Skrzypczak-Jankun E, Zhou K, McCabe NP, et al: Natural polyphenols in
comples with lipoxygenase - structural studies and their relevance to
cancer. In: Proc 93rd Annual Meeting AACR, p205, 2002.
20. Dahl TA, Bilski P, Reszka KJ, et al: Photocytotoxicity of curcumin.
Photochem Photobiol 59: 290-4, 1994.
21. Gorman AA, Hamblett I, Srinivasan VS, et al: Curcumin-derived transients:
a pulsed laser and pulse radiolysis study. Photochem Photobiol 59: 389-
98, 1994.
22. Chignell CF, Bilski P, Reszka KJ, et al: Spectral and photochemical
properties of curcumin. Photochem Photobiol 59: 295-302, 1994.
23. Toth GR, Weckerle M, and Schreier B: Structural elucidation of two novel
products from the soybean lipoxygenase-catalyzed dioxygenation of
curcumin. Mang Reson Chem 38: 51-4, 2000.
141 24. Pham C, Jankun J, Skrzypczak-Jankun E, et al: Structural and
thermochemical characterization of lipoxygenase-catechol complexes.
Biochemistry 37: 17952-7, 1998.
25. Skrzypczak-Jankun E, Amzel LM, Kroa BA, et al: Structure of soybean
lipoxygenase L3 and a comparison with its L1 isoenzyme. Proteins 29: 15-
31, 1997.
26. Tomchick DR, Phan P, Cymborowski M, et al: Structural and functional
characterization of second-coordination sphere mutants of soybean
lipoxygenase-1. Biochemistry 40: 7509-17, 2001.
27. Whitman S, Gezginci M, Timmermann BN, et al: Structure-activity
relationship studies of nordihydroguaiaretic acid inhibitors toward
soybean, 12-human, and 15-human lipoxygenase. J Med Chem 45: 2659-
61, 2002.
28. Kelman C, Krupinski-Oslen R, and Shorter A: Reductive inactivation of
soybean lipoxygenase 1 by catechols: a possible mechanism for
regulation of lipoxygenase activity. Biochemistry 26: 7064-72, 1987.
29. Bottcher AB, Day ER, Gray MW, et al: How do electronegative
substituents make metal complexes better catalysts for the oxidaion of
hydrocarbons by dioxygen? Chem J Mol Catalysis 117: 229-42, 1997.
30. Srivastava KC, Bordia A, and Verma SK: Curcumin, a major component of
food spice turmeric (Curcuma longa) inhibits aggregation and alters
eicosanoid metabolism in human blood platelets. Prostaglandins Leukot
Essent Fatty Acids 52: 223-7, 1995.
142 31. Rao CV, Simi B, and Reddy BS: Inhibition by dietary curcumin of
azoxymethane-induced ornithine decarboxylase, tyrosine protein kinase,
arachidonic acid metabolism and aberrant crypt foci formation in the rat
colon. Carcinogenesis 14: 2219-25, 1993.
32. Huang MT, Lysz T, Ferraro T, et al: Inhibitory effects of curcumin on in
vitro lipoxygenase and cyclooxygenase activities in mouse epidermis.
Cancer Res 51: 813-9, 1991.
143 Table 1. Data collection and refinement statistics
Data: Space group and unit cell dimensions C2, a=112.8, b=137.3, c=61.9 Å, β=95.6° Resolution limits (Å) 40.00 - 2.20 Unique reflections 45106 Completeness overall, 95.5 2.28-2.20Å last shell (%) 90.9 Rmerge overall, 2.28-2.20Å last shell 0.07, 0.33
Refinement: Protein, inhibitor, water - no. of atoms (non-hydrogen atoms only) 6779,11,471 R, Rfree (39461 F >2 σ(F) ) 0.185, 0.271 rms deviation from ideal geometry Bond lengths (Å) 0.008 Bond angles (º) 1.59 Dihedral angles (º) 24.28 Improper angles (º) 1.36
144
Table 2. Selected residues in soy LOX-1, soy LOX-3 and rabbit 15-LOX that are structurally at the same location in space.
Soy LOX-1 H499 N694 Q495 Q697 W500 L546 L754 F557 S491 Soy LOX-3 H518 N713 Q514 Q716 W519 L565 L773 F576 S510 Rabbit 15-LOX H361 H545 E357 Q548 L362 L408 L597 F415 F353
145
Table 3. Kinetic data for lipoxygenases and curcumin.
Lipoxygenase Inhibitor Concentration Activity Reference Colonic mucosal LOX from male Curcumin 50 µM 35-41% (31) F344 rats Liver LOX from Curcumin 50 µM 28-40 % (31) male F344 rats Epidermal LOX Curcumin 10 µM 40 % (32) from mouse Soybean LOX-3 Curcumin 1.2 µM 52 % (19)
146 Figures:
Fig.1. Curcumin bound to soybean LOX-3 lipoxygnase: gray – keto form,
yellow – enol form (theoretical modeling). Surface colored according to
the lipophilic properties of the residues lining the cavity, where brown is
the most hydrophobic; blue is the most hydrophilic. Iron ligands and
selected residues shown as stick models to illustrate curcumin’s
positioning.
Fig.2. Curcumin structural features and reactions with lipoxygenase.
Fig.3. Electron density map 2Fo-Fc contoured at 1σ level, showing the
molecular structure with a radius of 11Å around the inhibitor.
Fig.4. Overlay of the reported structure with another “purple” complex
containing 13-(S)-hydroperoxy-9,11-(cis,trans)-octadecadienoic acid
(13(S)-HPODE in purple), and the 4-nitrocatechol complex (inhibitor and a
nearby water shown in yellow).
147 Figure 1.
148 Figure 2.
149
Figure 3.
150
Figure 4.
151 DISCUSSION/SUMMARY
The work contained within this dissertation comprises two separate projects concerning the structure and function of lipoxygenases. Manuscript 1 covers functional aspects of human P12-LOX as it relates to prostate cancer and angiogenesis. Manuscripts 2 and 3 are comprised of structural data characterizing a complex consisting of the compound curcumin and soy LOX-3.
The basis of what follows will be a discussion of the two individual studies and how they relate to each other.
The importance of P12-LOX in prostate cancer was first shown by Gao et al. (1995). It was found that expression levels of P12-LOX were significantly higher in prostate adenocarcinoma tissue compared to matched normal prostate epithelium, and that this increased expression could be correlated to a more advanced stage and grade of prostate adenocarcinoma. In fact, a recent study by Nie et al. (1998) confirmed the significance of P12-LOX in prostate cancer biology. They noted that subcutaneously injected human prostate PC-3 cells overexpressing P12-LOX in nude mice resulted in rapid tumor growth as a result of augmented tumor vascularization (Nie et al. 1998). A significant amount of data alludes to a link between 12(S)-HETE production and the processes of tumor induced angiogenesis and metastasis (Honn et al. 1994a). For example,
12(S)-HETE is known to be a potent mitogen for microvascular endothelial cells and promotes wound healing in scratch injured cell monolayers (Tang et al.
1995). Cumulatively, these data show the significant contribution that P12-LOX
152 and the eicosanoid 12(S)-HETE play in angiogenic process associated with
prostate cancer progression.
In order to further characterize the role of P12-LOX in prostate cancer,
PC-3 cancer cells stably overexpressing P12-LOX were generated. These cells
produced a significantly increased amount of 12(S)-HETE and supplemental
arachidonic acid further increased this effect. Interestingly, studies conducted by
Nie et al. (1998) and Connolly and Rose (1998) indicate that cells overexpressing
P12-LOX have an enhanced angiogenic phenotype, presumably as a result of increased 12(S)-HETE production. Although 12(S)-HETE has been characterized as proangiogenic, it is possible that the role of 12(S)-HETE in the
angiogenic process lies in its ability to stimulate the expression of a more potent
angiogenic factor such as VEGF. Indeed, Natarajan et al. (1997a) demonstrated
that 12(S)-HETE could increase VEGF production in smooth vascular muscle
cells. As illustrated in manuscript 1, VEGF accumulation in conditioned culture
media of serum starved PC-3 cells overexpressing P12-LOX was elevated
approximately 13-fold compared to control cells. Thus, elevated 12(S)-HETE
production is associated with enhanced VEGF production. These results led us
to inquire about a possible mechanism by which P12-LOX, presumably acting
through 12(S)-HETE, could modulate the expression of VEGF.
Several recent studies have implicated components of the MAPK
pathway, specifically ERK1/2, in the regulation of VEGF expression (Yamaguchi
et al. 1995; Xu et al. 2002). Knowing this, we postulated that 12(S)-HETE may
mediate VEGF expression by activating ERK1/2 MAPKs. Our results showed
153 that PC-3 cells overexpressing P12-LOX posses enhanced basal levels of
ERK1/2 phosphorylation. As 12(S)-HETE production is enhanced in PC-3 cells overexpressing P12-LOX, it is plausible that 12(S)-HETE promotes the observed enhanced basal phosphorylation of ERK1/2 in an autocrine fashion. Manuscript
1 shows this to be a reasonable assumption in that 12(S)-HETE stimulated a transient, dose dependent increase in ERK1/2 phosphorylation. In agreement with these results, Ding et al. (2001) and Szekeres et al. (2000a) reported that
12(S)-HETE stimulated ERK1/2 phosphorylation in pancreatic β cells and human epidermoid carcinoma A431 cells, respectively. The role of ERK1/2 in the stimulation of VEGF protein expression was confirmed by inhibition of MEK, a kinase upstream of ERK1/2 in the MAPK pathway. The MEK inhibitor U0126 reduced VEGF accumulation in culture by 70%.
As indicated by Gao et al. (1995), P12-LOX overexpression correlated
with a more advanced prostate cancer. Manuscript 1 thus provides substantial
insight into a potential role of elevated P12-LOX expression in prostate
carcinoma, in that elevated levels of P12-LOX leads to enhanced 12(S)-HETE
production and a concomitant increase in VEGF production. This increase in
VEGF production can, in turn, lead to enhanced tumor angiogenesis, a process
considered to be a prerequisite for tumor dissemination. In fact, the high
mortality figures associated with prostate cancer are a result of tumor metastasis.
Knowing this and taking into consideration the extensive data implicating P12-
LOX and its product, 12(S)-HETE, in the processes of angiogenesis and
154 metastasis, it is clear that P12-LOX should be a target of anti-carcinogenic and
anti-metastatic treatments.
Manuscripts 2 and 3, in an effort to characterize important structural
components of an effective lipoxygenase inhibitor, provide 3D structural data of the interaction between the nonspecific lipoxygenase inhibitor, curcumin, and a lipoxygenase. In lieu of highly purified human P12-LOX, soy LOX-3 was used.
Soybean lipoxygenases are considered to be useful as general models
lipoxygenase structure and function. As a starting point, molecular simulations
were used to illustrate the interaction of curcumin with soy LOX-3 and indicated
that curcumin exhibited a high affinity for soy LOX-3. This finding is compatible
with those reported by Schneider et al. (1998), who showed that curcumin can
act as a substrate for soy LOX-1. To confirm the results of the molecular
simulation data, X-ray crystallographic analyses of the actual interaction of
curcumin with soy LOX-3 was performed. Using purified soy LOX-3 protein
obtained by salt exclusion and chromatofocusing, crystals were grown and then
soaked in a curcumin/ethanol solution. Upon X-ray irradiation of soy LOX-3
crystals in complex with curcumin, a color change of yellow to purple occurred.
This color change has been previously shown to occur when a peroxide
compound acts as a ligand for lipoxygenase (Skrzypczak-Jankun et al. 2001),
thus the complex consisted of LOX-Fe-O-O-R.
The structure of the complex was elucidated by molecular replacement using previously determined soy LOX-3 molecular determinants (PDB entry
1LNH). Upon examination of the electron density maps, the presence of a
155 molecule smaller than curcumin near the active site of soy LOX-3 in crystals
soaked in the curcumin solution was noted. This small molecule was the product
of a photolytic reaction and this compound could indeed act as a ligand to soy
LOX-3, as indicated by the color change. Using electrospray mass
spectroscopy, it was shown that several compounds smaller than curcumin were
potential products of the photolytic reaction. Interestingly, crystals of curcumin
alone are stable under X-ray irradiation (Tonneson and Mostad 1982; Ishigami et
al. 1999), so these compounds must have been a direct result of X-ray
illumination of the curcumin/lipoxygenase complex.
Structural examination revealed that the degredation product of curcumin
in complex with soy LOX-3 is a peroxide, C7H8O4, bound to Fe. This corresponds to two structurally similar compounds: 4-hydroperoxy-2- methoxyphenol and 4-hydroperoxy-2-methoxycyclohexa-2,5-diene-1-one. By careful examination of the electron density map of the degradation product/soy
LOX-3 complex, the compound 4-hydroperoxy-2-methoxyphenol was determined to be the photolytic curcumin product present in the “purple” complex.
Structurally curcumin contains two catecholic moieties and many catecholic compounds have been shown to inhibit lipoxygenases. The most widely used non-specific lipoxygenase inhibitor, NDGA, also contains two catacholic moieties. Like curcumin, NDGA is a widely used antioxidant. Other catechol related compounds, such as the flavonoids quercetin, escelutin, and baicalein, also possess lipoxygenase inhibitory properties. The mechanism of lipoxygenase inhibition by these inhibitors is due to reduction of the ferric iron
156 (active) of lipoxygenase to the ferrous form (inactive). This reaction denotes a
non-competitive mechanism of inhibition as no bond formation between the
compound and the Fe of lipoxygenase occurs. Curcumin appears to inhibit
lipoxygenase in a similar fashion.
In light the data reported in manuscript 1 and the complete body of work
regarding P12-LOX in prostate cancer and the processes of angiogenesis and metastasis, it seems that the development of inhibitors of P12-LOX that effectively reduce the production of 12(S)-HETE by prostate cancer cells, would prove beneficial in the treatment of prostate cancer by inhibiting metastasis both directly, due to the role of 12(S)-HETE in multiple aspects of the metastatic cascade, as well as indirectly by inhibiting tumor elated angiogenesis as a result of 12(S)-HETE, VEGF, and 12(S)-HETE amplification of VEGF expression.
Therefore, P12-LOX and the P12-LOX metabolite 12(S)-HETE, which modulates metastasis, should be extensively investigated as targets for anti-carcinogenesis and anti-metastasis treatments.
157 BIBLIOGRAPHY
Abreu-Martin, M. T.; Chari, A.; Palladino, A. A.; Craft, N. A.; and Sawyers, C. L.
1999 Mitogen-activated protein kinase kinase kinase 1 activates
androgen receptor-dependent transcription and apoptosis in prostate
cancer. Mol. Cell Biol., 19: 5143-54.
Ammon, H. P.; Safayhi, H.; Mack, T.; and Sabieraj, J. 1993 Mechanism of
antiinflammatory actions of curcumine and boswellic acids. J.
Ethnopharmacol., 38: 113-9.
Anderson, K. M.; Seed, T.; Ondrey, F.; and Harris, J. E. 1994 The selective 5-
lipoxygenase inhibitor A63162 reduces PC3 proliferation and initiates
morphologic changes consistent with secretion. Anticancer Res., 14:
1951-60.
Anderson, N. G.; Maller, J. L.; Tonks, N. K.; and Sturgill, T. W. 1990
Requirement for integration of signals from two distinct
phosphorylation pathways for activation of MAP kinase. Nature, 343:
651-3.
Arakawa, T.; Nakamura, M.; Yoshimoto, T.; and Yamamoto, S. 1995 The
transcriptional regulation of human arachidonate 12-lipoxygenase gene by
NF kappa B/Rel. FEBS Lett., 363: 105-10.
Arenberger, P.; Kemeny, L.; Rupec, R.;Bieber, T.; and Ruzicka, T. 1992
Langerhans cells of the human skin possess high-affinity 12(S)-
hydroxyeicosa tetraenoic acid receptors. Eur. J. Immunol., 22: 2469-72.
158 Avis, I. M.; Jett, M.; Boyle, T.; Vos, M. D.; Moody, T.; Treston, A. M.; Martinez, A.;
and Mulshine, J. L. 1996 Growth control of lung cancer by interruption of
5-lipoxygenase-mediated growth factor signaling. J. Clin. Invest., 97: 806-
13.
Baba, A.; Sakuma, S.; Okamoto, H.; Inoue, T.; and Iwata, H. 1989 Calcium
induces membrane translocation of 12-lipoxygenase in rat platelets. J.
Biol. Chem., 264: 15790-5.
Bigler, S. A.; Deering, R. E.; and Brawer, M. K. 1993 Comparison of
microscopic vascularity in benign and malignant prostate tissue. Hum.
Pathol., 24: 220-6.
Blackwell, T. S.; Blackwell, T. R.; and Christman, J. W. 1997 Impaired
activation of nuclear factor-kappaB in endotoxin-tolerant rats is associated
with down-regulation of chemokine gene expression and inhibition of
neutrophilic lung inflammation. J. Immunol., 158: 5934-40.
Borgstrom, P.; Bourdon, M. A.; Hillan, K. J.; Sriramarao, P.; and Ferrara, N.
1998 Neutralizing anti-vascular endothelial growth factor antibody
completely inhibits angiogenesis and growth of human prostate carcinoma
micro tumors in vivo. Prostate, 35: 1-10.
Borngraber, S.; Browner, M.; Gillmor, S.; Gerth, C.; Anton, M.; Fletterick, R.; and
Kuhn, H. 1999 Shape and specificity in mammalian 15-lipoxygenase
active site. The functional interplay of sequence determinants for the
reaction specificity. J. Biol. Chem., 274: 37345-50.
159 Bottcher, A. B.; Day, E. R.; Gray, M. W.; Grinstaff, H. B.; and Labinger, M. W.
1997 How do electronegative substituents make metal complexes better
catalysts for the oxidaion of hydrocarbons by dioxygen? Chem. J. Mol.
Catalysis, 117: 229-242.
Boyington, J. C.; Gaffney, B. J.; and Amzel, L. M. 1993 The three-dimensional
structure of an arachidonic acid 15-lipoxygenase. Science, 260: 1482-6.
Brash, A. R. 1999 Lipoxygenases: occurrence, functions, catalysis, and
acquisition of substrate. J. Biol. Chem., 274: 23679-82.
Brawer, M. K.; Deering, R. E.; Brown, M.; Preston, S. D.; and Bigler, S. A. 1994
Predictors of pathologic stage in prostatic carcinoma. The role of
neovascularity. Cancer, 73: 678-87.
Broitman, S. A.; Vitale, J.J.; Vavrousek-Jakuba, E.; and Gottlieb, L.S. 1977
Polyunsaturated fat, choesterol and large bowel tumorigenesis. Cancer
(Phila.), 40: 2455-2463.
Bruchovsky, N.; Rennie, P. S.; Coldman, A. J.; Goldenberg, S. L.; To, M.; and
Lawson, D. 1990 Effects of androgen withdrawal on the stem cell
composition of the Shionogi carcinoma. Cancer Res., 50: 2275-82.
Bryant, R. W.; Bailey, J. M.; Schewe, T.; and Rapoport, S. M. 1982 Positional
specificity of a reticulocyte lipoxygenase. Conversion of arachidonic acid
to 15-S-hydroperoxy-eicosatetraenoic acid. J. Biol. Chem., 257: 6050-5.
Bull, A. W.; Nigro, N. D.; Golembieski, W. A.; Crissman, J. D.; and Marnett, L. J.
1984 In vivo stimulation of DNA synthesis and induction of ornithine
160 decarboxylase in rat colon by fatty acid hydroperoxides, autoxidation
products of unsaturated fatty acids. Cancer Res., 44: 4924-8.
Burger, F.; Krieg, P.; Marks, F.; and Furstenberger, G. 2000 Positional- and
stereo-selectivity of fatty acid oxygenation catalysed by mouse (12S)-
lipoxygenase isoenzymes. Biochem. J., 348 Pt 2: 329-35.
Carmeliet, P.; Ferreira, V.; Breier, G.; Pollefeyt, S.; Kieckens, L.; Gertsenstein,
M.; Fahrig, M.; Vandenhoeck, A.; Harpal, K.; Eberhardt, C.; Declercq,
C.; Pawling, J.; Moons, L.; Collen, D.; Risau, W.; and Nagy, A. 1996
Abnormal blood vessel development and lethality in embryos lacking a
single VEGF allele. Nature, 380: 435-9.
Casey, P. J. 1995 Protein lipidation in cell signaling. Science, 268: 221-5.
Chan, C. C.; Duhamel, L.; and Ford-Hutchison, A. 1985 Leukotriene B4 and 12-
hydroxyeicosatetraenoic acid stimulate epidermal proliferation in vivo in
the guinea pig. J. Invest. Dermatol., 85: 333-4.
Chan, M. M. 1995 Inhibition of tumor necrosis factor by curcumin, a
phytochemical. Biochem. Pharmacol., 49: 1551-6.
Chang, W. C.; Kao, H. C.; and Liu, Y. W. 1995 Down-regulation of epidermal
growth factor-induced 12-lipoxygenase expression by glucocorticoids in
human epidermoid carcinoma A431 cells. Biochem. Pharmacol., 50: 947-
52.
Chang, W. C.; Liu, Y. W.; Ning, C. C.; Suzuki, H.; Yoshimoto, T.; and Yamamoto,
S. 1993 Induction of arachidonate 12-lipoxygenase mRNA by epidermal
growth factor in A431 cells. J. Biol. Chem., 268: 18734-9.
161 Chen, B. K.; and Chang, W. C. 1999 Overexpression of c-Fos enhances the
transcription of human arachidonate 12-lipoxygenase in A431 cells.
Biochem. Biophys. Res. Commun., 261: 848-52.
Chen, B. K.; and Chang, W. C. 2000 Functional interaction between c-
Jun and promoter factor Sp1 in epidermal growth factor-induced gene
expression of human 12(S)-lipoxygenase. Proc. Natl. Acad. Sci. U. S. A.,
97: 10406-11.
Chen, B. K.; Liu, Y. W.; Yamamoto, S.; and Chang, W. C. 1997
Overexpression of Ha-ras enhances the transcription of human
arachidonate 12-lipoxygenase promoter in A431 cells. Biochim. Biophys.
Acta, 1344: 270-7.
Chen, C. J.; Huang, H. S.; Lin, S. B.; and Chang, W. C. 2000 Regulation of
cyclooxygenase and 12-lipoxygenase catalysis by phospholipid
hydroperoxide glutathione peroxidase in A431 cells. Prostaglandins
Leukot. Essent. Fatty Acids, 62: 261-8.
Chen, L. C.; Chen, B. K.; Liu, Y. W.; and Chang, W. C. 1999 Induction of 12-
lipoxygenase expression by transforming growth factor-alpha in human
epidermoid carcinoma A431 cells. FEBS Lett., 455: 105-10.
Chen, X. S.; Kurre, U.; Jenkins, N. A.; Copeland, N. G.; and Funk, C. D. 1994
cDNA cloning, expression, mutagenesis of C-terminal isoleucine, genomic
structure, and chromosomal localizations of murine 12-lipoxygenases. J.
Biol. Chem., 269: 13979-87.
162 Chen, Y. Q.; Duniec, Z. M.; Liu, B.; Hagmann, W.; Gao, X.; Shimoji, K.; Marnett,
L. J.; Johnson, C. R.; and Honn, K. V. 1994 Endogenous 12(S)-HETE
production by tumor cells and its role in metastasis. Cancer Res., 54:
1574-9.
Chignell, C. F.; Bilski, P.; Reszka, K. J.; Motten, A. G.; Sik, R. H.; and Dahl, T. A.
1994 Spectral and photochemical properties of curcumin. Photochem.
Photobiol., 59: 295-302.
Chopra, H.; Timar, J.; Chen, Y. Q.; Rong, X. H.; Grossi, I. M.; Fitzgerald, L. A.;
Taylor, J. D.; and Honn, K. V. 1991 The lipoxygenase metabolite 12(S)-
HETE induces a cytoskeleton-dependent increase in surface expression
of integrin alpha IIb beta 3 on melanoma cells. Int. J. Cancer, 49: 774-86.
Chowdhury, S. K.; Katta, V.; and Chait, B. T. 1990 An electrospray-ionization
mass spectrometer with new features. Rapid Commun. Mass Spectrom.,
4: 81-7.
Claffey, K. P.; Shih, S. C.; Mullen, A.; Dziennis, S.; Cusick, J. L.; Abrams,
K. R.; Lee, S. W.; and Detmar, M. 1998 Identification of a human
VPF/VEGF 3' untranslated region mediating hypoxia-induced mRNA
stability. Mol. Biol. Cell, 9: 469-81.
Cohen, T.; Nahari, D.; Cerem, L. W.; Neufeld, G.; and Levi, B. Z. 1996
Interleukin 6 induces the expression of vascular endothelial growth factor.
J. Biol. Chem., 271: 736-41.
163 Connolly, J. M.; and Rose, D. P. 1989 Secretion of epidermal growth factor and
related polypeptides by the DU 145 human prostate cancer cell line.
Prostate, 15: 177-86.
Connolly, J. M.; and Rose, D. P. 1998 Enhanced angiogenesis and growth of
12-lipoxygenase gene-transfected MCF-7 human breast cancer cells in
athymic nude mice. Cancer Lett., 132: 107-12.
Corey, E. J.; and Nataga, R. 1987 Evidence in favor of an organoiron-mediated
pathway for lipoxygenation of fatty acids by soybean lipoxygenase. J. Am.
Chem. Soc., 109: 8107-8108.
Covey, T. R.; Bonner, R. F.; Shushan, B. I.; and Henion, J. 1988 The
determination of protein, oligonucleotide and peptide molecular weights by
ion-spray mass spectrometry. Rapid Commun. Mass Spectrom., 2: 249-
56.
Cuendet, M.; and Pezzuto, J. M. 2000 The role of cyclooxygenase and
lipoxygenase in cancer chemoprevention. Drug Metabol. Drug Interact.,
17: 109-57.
Culig, Z.; Hobisch, A.; Cronauer, M. V.; Radmayr, C.; Trapman, J.; Hittmair, A.;
Bartsch, G.; and Klocker, H. 1994 Androgen receptor activation in
prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth
factor, and epidermal growth factor. Cancer Res., 54: 5474-8.
Dadaian, M.; and Westlund, P. 1999 Albumin modifies the metabolism of
hydroxyeicosatetraenoic acids via 12-lipoxygenase in human platelets. J.
Lipid Res., 40: 940-7.
164 Dahl, T. A.; Bilski, P.; Reszka, K. J.; and Chignell, C. F. 1994 Photocytotoxicity
of curcumin. Photochem. Photobiol., 59: 290-4.
Dailey, L. A.; and Imming, P. 1999 12-Lipoxygenase: classification, possible
therapeutic benefits from inhibition, and inhibitors. Curr. Med. Chem. 6:
389-98.
de Groot, J. J.; Veldink, G. A.; Vliegenthart, J. F.; Boldingh, J.; Wever, R.; and
van Gelder, B. F. 1975 Demonstration by EPR spectroscopy of the
functional role of iron in soybean lipoxygenase-1. Biochim. Biophys. Acta,
377: 71-9.
de Vries, C.; Escobedo, J. A.; Ueno, H.; Houck, K.; Ferrara, N.; and Williams, L.
T. 1992 The fms-like tyrosine kinase, a receptor for vascular endothelial
growth factor. Science, 255: 989-91.
Deroanne, C. F.; Hajitou, A.; Calberg-Bacq, C. M.; Nusgens, B. V.; and Lapiere,
C. M. 1997 Angiogenesis by fibroblast growth factor 4 is mediated
through an autocrine up-regulation of vascular endothelial growth factor
expression. Cancer Res., 57: 5590-7.
Ding, X. Z.; Iversen, P.; Cluck, M. W.; Knezetic, J. A.; and Adrian, T. E. 1999a
Lipoxygenase inhibitors abolish proliferation of human pancreatic cancer
cells. Biochem. Biophys. Res. Commun., 261: 218-23.
Ding, X. Z.; Kuszynski, C. A.; El-Metwally, T. H.; and Adrian, T. E. 1999b
Lipoxygenase inhibition induced apoptosis, morphological changes, and
carbonic anhydrase expression in human pancreatic cancer cells.
Biochem. Biophys. Res. Commun., 266: 392-9.
165 Ding, X. Z.; Tong, W. G.; and Adrian, T. E. 2001 12-lipoxygenase metabolite
12(S)-HETE stimulates human pancreatic cancer cell proliferation via
protein tyrosine phosphorylation and ERK activation. Int. J. Cancer, 94:
630-6.
Dorai, T.; Cao, Y. C.; Dorai, B.; Buttyan, R.; and Katz, A. E. 2001 Therapeutic
potential of curcumin in human prostate cancer. III. Curcumin inhibits
proliferation, induces apoptosis, and inhibits angiogenesis of LNCaP
prostate cancer cells in vivo. Prostate, 47: 293-303.
Draheim, J. E.; Carroll, R. T.; McNemar, T. B.; Dunham, W. R.; Sands, R. H.; and
Funk, M. O., Jr. 1989 Lipoxygenase isoenzymes: a spectroscopic and
structural characterization of soybean seed enzymes. Arch. Biochem.
Biophys., 269: 208-18.
Duque, J. L.; Loughlin, K. R.; Adam, R. M.; Kantoff, P. W.; Zurakowski, D.; and
Freeman, M. R. 1999 Plasma levels of vascular endothelial growth factor
are increased in patients with metastatic prostate cancer. Urology, 54:
523-7.
Dvorak, A. M.; Kohn, S.; Morgan, E. S.; Fox, P.; Nagy, J. A.; and Dvorak, H. F.
1996 The vesiculo-vacuolar organelle (VVO): a distinct endothelial cell
structure that provides a transcellular pathway for macromolecular
extravasation. J. Leukoc. Biol., 59: 100-15.
Dvorak, H. F.; Nagy, J. A.; Berse, B.; Brown, L. F.; Yeo, K. T.; Yeo, T. K.; Dvorak,
A. M.; van de Water, L.; Sioussat, T. M.; and Senger, D. R. 1992
166 Vascular permeability factor, fibrin, and the pathogenesis of tumor stroma
formation. Ann. N .Y. Acad. Sci., 667: 101-11.
Dvorak, H. F.; Sioussat, T. M.; Brown, L. F.; Berse, B.; Nagy, J. A.; Sotrel, A.;
Manseau, E. J.; Van de Water, L.; and Senger, D. R. 1991 Distribution of
vascular permeability factor (vascular endothelial growth factor) in tumors:
concentration in tumor blood vessels. J. Exp. Med., 174: 1275-8.
Eibl, G.; Bruemmer, D.; Okada, Y.; Duffy, J. P.; Law, R. E.; Reber, H. A.; and
Hines, O. J. 2003 PGE(2) is generated by specific COX-2 activity and
increases VEGF production in COX-2-expressing human pancreatic
cancer cells. Biochem. Biophys. Res. Commun., 306: 887-97.
Enslen, H.; Tokumitsu, H.; Stork, P. J.; Davis, R. J.; and Soderling, T. R. 1996
Regulation of mitogen-activated protein kinases by a calcium/calmodulin-
dependent protein kinase cascade. Proc. Natl. Acad. Sci. U. S. A., 93:
10803-8.
Esser, S.; Wolburg, K.; Wolburg, H.; Breier, G.; Kurzchalia, T.; and Risau, W.
1998 Vascular endothelial growth factor induces endothelial fenestrations
in vitro. J. Cell Biol., 140: 947-59.
Exton, J. H. 1996 Regulation of phosphoinositide phospholipases by hormones,
neurotransmitters, and other agonists linked to G proteins. Annu. Rev.
Pharmacol. Toxicol., 36: 481-509.
Falck, J. R.; Manna, S.; Moltz, J.; Chacos, N.; and Capdevila, J. 1983
Epoxyeicosatrienoic acids stimulate glucagon and insulin release from
167 isolated rat pancreatic islets. Biochem. Biophys. Res. Commun., 114: 743-
9.
Fauconnet, S.; Lascombe, I.; Chabannes, E.; Adessi, G. L.; Desvergne, B.;
Wahli, W.; and Bittard, H. 2002 Differential regulation of vascular
endothelial growth factor expression by peroxisome proliferator-activated
receptors in bladder cancer cells. J. Biol. Chem., 277: 23534-43.
Ferrara, N. 2001 Role of vascular endothelial growth factor in regulation of
physiological angiogenesis. Am. J. Physiol. Cell Physiol., 280: C1358-66.
Ferrara, N.; Carver-Moore, K.; Chen, H.; Dowd, M.; Lu, L.; O'Shea, K. S.; Powell-
Braxton, L.; Hillan, K. J.; and Moore, M. W. 1996 Heterozygous
embryonic lethality induced by targeted inactivation of the VEGF gene.
Nature, 380: 439-42.
Ferrara, N.; and Davis-Smyth, T. 1997 The biology of vascular endothelial
growth factor. Endocr. Rev., 18: 4-25.
Ferrara, N.; and Henzel, W. J. 1989 Pituitary follicular cells secrete a novel
heparin-binding growth factor specific for vascular endothelial cells.
Biochem. Biophys. Res. Commun., 161: 851-8.
Ferrer, F. A.; Miller, L. J.; Andrawis, R. I.; Kurtzman, S. H.; Albertsen, P. C.;
Laudone, V. P.; and Kreutzer, D. L. 1997 Vascular endothelial growth
factor (VEGF) expression in human prostate cancer: in situ and in vitro
expression of VEGF by human prostate cancer cells. J. Urol., 157: 2329-
33.
168 Finkenzeller, G.; Sparacio, A.; Technau, A.; Marme, D.; and Siemeister, G. 1997
Sp1 recognition sites in the proximal promoter of the human vascular
endothelial growth factor gene are essential for platelet-derived growth
factor-induced gene expression. Oncogene, 15: 669-76.
Folkman, J. 1995 Angiogenesis in cancer, vascular, rheumatoid and other
disease. Nat. Med., 1: 27-31.
Folkman, J.; and D'Amore, P. A. 1996 Blood vessel formation: what is its
molecular basis? Cell, 87: 1153-5.
Folkman, J.; and Shing, Y. 1992 Angiogenesis. J. Biol. Chem., 267: 10931-4.
Folkman, J.; Watson, K.; Ingber, D.; and Hanahan, D. 1989 Induction of
angiogenesis during the transition from hyperplasia to neoplasia. Nature,
339: 58-61.
Fong, G. H.; Rossant, J. ;Gertsenstein, M.; and Breitman, M. L. 1995 Role of
the Flt-1 receptor tyrosine kinase in regulating the assembly of vascular
endothelium. Nature, 376: 66-70.
Frank, S.; Hubner, G.; Breier, G.; Longaker, M. T.; Greenhalgh, D. G.; and
Werner, S. 1995 Regulation of vascular endothelial growth factor
expression in cultured keratinocytes. Implications for normal and impaired
wound healing. J. Biol. Chem., 270: 12607-13.
Fujimoto, Y.; Tsunomori, M.; Muta, E.; Yamamoto, T.; Nishida, H.; Sakuma, S.;
and Fujita, T. 1994 High density lipoprotein inhibits platelet 12-
lipoxygenase activity. Res. Commun. Mol. Pathol. Pharmacol., 85: 355-8.
169 Funk, C. D. 1996 The molecular biology of mammalian lipoxygenases and the
quest for eicosanoid functions using lipoxygenase-deficient mice. Biochim.
Biophys. Acta, 1304: 65-84.
Funk, C. D.; Funk, L. B.; FitzGerald, G. A.; and Samuelsson, B. 1992
Characterization of human 12-lipoxygenase genes. Proc. Natl. Acad. Sci.
U .S .A., 89: 3962-6.
Funk, C. D.; Furci, L.; and FitzGerald, G. A. 1990 Molecular cloning, primary
structure, and expression of the human platelet/erythroleukemia cell 12-
lipoxygenase. Proc. Natl. Acad. Sci. U. S. A., 87: 5638-42.
Gaffney, B. J. 1996 Lipoxygenases: structural principles and spectroscopy.
Annu. Rev. Biophys. Biomol. Struct., 25: 431-59.
Gan, Q. F.; Browner, M. F.; Sloane, D. L.; and Sigal, E. 1996 Defining the
arachidonic acid binding site of human 15-lipoxygenase. Molecular
modeling and mutagenesis. J. Biol. Chem., 271: 25412-8.
Gao, X.; Grignon, D. J.; Chbihi, T.; Zacharek, A.; Chen, Y. Q.; Sakr, W.; Porter,
A. T.; Crissman, J. D.; Pontes, J. E.; Powell, I. J.; and et al. 1995
Elevated 12-lipoxygenase mRNA expression correlates with advanced
stage and poor differentiation of human prostate cancer. Urology, 46: 227-
37.
Garg, A.; and Aggarwal, B. B. 2002 Nuclear transcription factor-kappaB as a
target for cancer drug development. Leukemia, 16: 1053-68.
Gately, S.; and Li, W. W. 2004 Multiple roles of COX-2 in tumor angiogenesis: a
target for antiangiogenic therapy. Semin. Oncol., 31: 2-11.
170 Gerwick, W. H. 1994 Structure and biosynthesis of marine algal oxylipins.
Biochim. Biophys. Acta, 1211: 243-55.
Ghosh, J.; and Myers, C. E. 1997 Arachidonic acid stimulates prostate cancer
cell growth: critical role of 5-lipoxygenase. Biochem. Biophys. Res.
Commun., 235: 418-23.
Ghosh, J.; and Myers, C. E. 1998 Inhibition of arachidonate 5-lipoxygenase
triggers massive apoptosis in human prostate cancer cells. Proc. Natl.
Acad. Sci. U. S. A., 95: 13182-7.
Gille, H.; Kortenjann, M.; Thomae, O.; Moomaw, C.; Slaughter, C.; Cobb, M. H.;
and Shaw, P. E. 1995 ERK phosphorylation potentiates Elk-1-mediated
ternary complex formation and transactivation. Embo J., 14: 951-62.
Gille, H.; Sharrocks, A. D.; and Shaw, P. E. 1992 Phosphorylation of
transcription factor p62TCF by MAP kinase stimulates ternary complex
formation at c-fos promoter. Nature, 358: 414-7.
Gille, J.; Swerlick, R. A.; and Caughman, S. W. 1997 Transforming growth
factor-alpha-induced transcriptional activation of the vascular permeability
factor (VPF/VEGF) gene requires AP-2-dependent DNA binding and
transactivation. Embo J., 16: 750-9.
Gillmor, S. A.; Villasenor, A.; Fletterick, R.; Sigal, E.; and Browner, M. F. 1997
The structure of mammalian 15-lipoxygenase reveals similarity to the
lipases and the determinants of substrate specificity. Nat. Struct. Biol., 4:
1003-9.
171 Gimbrone, M. A., Jr.; Leapman, S. B.; Cotran, R. S.; and Folkman, J. 1972
Tumor dormancy in vivo by prevention of neovascularization. J. Exp.
Med., 136: 261-76.
Gioeli, D.; Mandell, J. W.; Petroni, G. R.; Frierson, H. F., Jr.; and Weber, M. J.
1999 Activation of mitogen-activated protein kinase associated with
prostate cancer progression. Cancer Res., 59: 279-84.
Glasgow, W. C.; Afshari, C. A.; Barrett, J. C.; and Eling, T. E. 1992 Modulation
of the epidermal growth factor mitogenic response by metabolites of
linoleic and arachidonic acid in Syrian hamster embryo fibroblasts.
Differential effects in tumor suppressor gene (+) and (-) phenotypes. J.
Biol. Chem., 267: 10771-9.
Gorman, A. A.; Hamblett, I.; Srinivasan, V. S.; and Wood, P. D. 1994 Curcumin-
derived transients: a pulsed laser and pulse radiolysis study. Photochem.
Photobiol., 59: 389-98.
Gridley, D. S.; Andres, M. L.; and Slater, J. M. 1997 Enhancement of prostate
cancer xenograft growth with whole-body radiation and vascular
endothelial growth factor. Anticancer Res., 17: 923-8.
Grossi, I. M.; Fitzgerald, L. A.; Umbarger, L. A.; Nelson, K. K.; Diglio, C. A.;
Taylor, J. D.; and Honn, K. V. 1989 Bidirectional control of membrane
expression and/or activation of the tumor cell IRGpIIb/IIIa receptor and
tumor cell adhesion by lipoxygenase products of arachidonic acid and
linoleic acid. Cancer Res., 49: 1029-37.
172 Gudermann, T.; Kalkbrenner, F.; and Schultz, G. 1996 Diversity and selectivity
of receptor-G protein interaction. Annu. Rev. Pharmacol. Toxicol., 36: 429-
59.
Gupta, S.; Srivastava, M.; Ahmad, N.; Sakamoto, K.; Bostwick, D. G.; and
Mukhtar, H. 2001 Lipoxygenase-5 is overexpressed in prostate
adenocarcinoma. Cancer, 91: 737-43.
Hagmann, W.; and Borgers, S. 1997 Requirement for epidermal growth factor
receptor tyrosine kinase and for 12-lipoxygenase activity in the expression
of 12-lipoxygenase in human epidermoid carcinoma cells. Biochem.
Pharmacol., 53: 937-42.
Hagmann, W.; Gao, X.; Timar, J.; Chen, Y. Q.; Strohmaier, A. R.; Fahrenkopf, C.;
Kagawa, D.; Lee, M.; Zacharek, A.; and Honn, K. V. 1996 12-
Lipoxygenase in A431 cells: genetic identity, modulation of expression,
and intracellular localization. Exp. Cell Res., 228: 197-205.
Hagmann, W.; Kagawa, D.; Renaud, C.; and Honn, K. V. 1993 Activity and
protein distribution of 12-lipoxygenase in HEL cells: induction of
membrane-association by phorbol ester TPA, modulation of activity by
glutathione and 13-HPODE, and Ca(2+)-dependent translocation to
membranes. Prostaglandins, 46: 471-7.
Hallahan, D. E.; Virudachalam, S.; Kufe, D. W.; and Weichselbaum, R. R. 1994
Ketoconazole attenuates radiation-induction of tumor necrosis factor. Int.
J. Radiat. Oncol. Biol. Phys., 29: 777-80.
173 Hamberg, M.; and Samuelsson, B. 1974 Prostaglandin endoperoxides. Novel
transformations of arachidonic acid in human platelets. Proc. Natl. Acad.
Sci. U. S .A., 71: 3400-4.
Hammarstrom, S.; Hamberg, M.; Samuelsson, B.; Duell, E. A.; Stawiski, M.; and
Voorhees, J. J. 1975 Increased concentrations of nonesterified
arachidonic acid, 12L-hydroxy-5,8,10,14-eicosatetraenoic acid,
prostaglandin E2, and prostaglandin F2alpha in epidermis of psoriasis.
Proc. Natl. Acad. Sci. U .S. A., 72: 5130-4.
Hampson, A. J.; and Grimaldi, M. 2002 12-hydroxyeicosatetrenoate (12-HETE)
attenuates AMPA receptor-mediated neurotoxicity: evidence for a G-
protein-coupled HETE receptor. J. Neurosci., 22: 257-64.
Hansbrough, J. R.; Takahashi, Y.; Ueda, N.; Yamamoto, S.; and Holtzman, M. J.
1990 Identification of a novel arachidonate 12-lipoxygenase in bovine
tracheal epithelial cells distinct from leukocyte and platelet forms of the
enzyme. J. Biol. Chem., 265: 1771-6.
Haque, M. S.; Arora, J. K.; Dikdan, G.; Lysz, T. W.; and Zelenka, P. S. 1999 The
rabbit lens epithelial cell line N/N1003A requires 12-lipoxygenase activity
for DNA synthesis in response to EGF. Mol. Vis., 5: 8.
Harper, M. E.; Glynne-Jones, E.; Goddard, L.; Thurston, V. J.; and Griffiths, K.
1996 Vascular endothelial growth factor (VEGF) expression in prostatic
tumours and its relationship to neuroendocrine cells. Br. J. Cancer, 74:
910-6.
174 Haslmayer, P.; Thalhammer, T.; Jager, W.; Aust, S.; Steiner, G.; Ensinger, C.;
and Obrist, P. 2002 The peroxisome proliferator-activated receptor
gamma ligand 15-deoxy-Delta12,14-prostaglandin J2 induces vascular
endothelial growth factor in the hormone-independent prostate cancer cell
line PC 3 and the urinary bladder carcinoma cell line 5637. Int. J. Oncol.,
21: 915-20.
Hein, R.; Gross, E.; Ruzicka, T.; and Krieg, T. 1991 12-Hydroxyeicosatetraenoic
acid (12-HETE) is a chemotactic stimulus for epidermal cells. Arch.
Dermatol. Res., 283: 135-7.
Herbertsson, H.; Kuhme, T.; Evertsson, U.; Wigren, J.; and Hammarstrom, S.
1998 Identification of subunits of the 650 kDa 12(S)-HETE binding
complex in carcinoma cells. J. Lipid Res., 39: 237-44.
Herbertsson, H.; Kuhme, T.; and Hammarstrom, S. 1999 The 650-kDa 12(S)-
hydroxyeicosatetraenoic acid binding complex: occurrence in human
platelets, identification of hsp90 as a constituent, and binding properties of
its 50-kDa subunit. Arch. Biochem. Biophys., 367: 33-8.
Hilger, R. A.; Scheulen, M. E.; and Strumberg, D. 2002 The Ras-Raf-MEK-ERK
pathway in the treatment of cancer. Onkologie, 25: 511-8.
Honn, K. V.; Grossi, I. M.; Diglio, C. A.; Wojtukiewicz, M.; and Taylor, J. D. 1989
Enhanced tumor cell adhesion to the subendothelial matrix resulting from
12(S)-HETE-induced endothelial cell retraction. Faseb J., 3: 2285-93.
175 Honn, K. V.; and Tang, D. G. 1992a Adhesion molecules and tumor cell
interaction with endothelium and subendothelial matrix. Cancer Metastasis
Rev., 11: 353-75.
Honn, K. V.; Tang, D. G.; and Crissman, J. D. 1992b Platelets and cancer
metastasis: a causal relationship? Cancer Metastasis Rev., 11: 325-51.
Honn, K. V.; Tang, D. G.; Gao, X.; Butovich, I. A.; Liu, B.; Timar, J.; and
Hagmann, W. 1994a 12-lipoxygenases and 12(S)-HETE: role in cancer
metastasis. Cancer Metastasis Rev., 13: 365-96.
Honn, K. V.; Timar, J.; Rozhin, J.; Bazaz, R.; Sameni, M.; Ziegler, G.; and
Sloane, B. F. 1994b A lipoxygenase metabolite, 12-(S)-HETE, stimulates
protein kinase C-mediated release of cathepsin B from malignant cells.
Exp. Cell Res., 214: 120-30.
Houck, K. A.; Ferrara, N.; Winer, J.; Cachianes, G.; Li, B.; and Leung, D. W.
1991 The vascular endothelial growth factor family: identification of a
fourth molecular species and characterization of alternative splicing of
RNA. Mol. Endocrinol., 5: 1806-14.
Huang, M. T.; Lysz, T.; Ferraro, T.; Abidi, T. F.; Laskin, J. D.; and Conney, A. H.
1991 Inhibitory effects of curcumin on in vitro lipoxygenase and
cyclooxygenase activities in mouse epidermis. Cancer Res., 51: 813-9.
Huang, M. T.; Newmark, H. L.; and Frenkel, K. 1997 Inhibitory effects of
curcumin on tumorigenesis in mice. J. Cell Biochem. Suppl., 27: 26-34.
Hussain, H.; Shornick, L. P.; Shannon, V. R.; Wilson, J. D.; Funk, C. D.;
Pentland, A. P.; and Holtzman, M. J. 1994 Epidermis contains platelet-
176 type 12-lipoxygenase that is overexpressed in germinal layer
keratinocytes in psoriasis. Am. J. Physiol., 266: C243-53.
InsightII, ver. 95.0/3.0.0, User's guide. 1995, San Diego, CA, USA: Molecular
Simulations, Inc.
Irvin, C. G.; Tu, Y. P.; Sheller, J. R.; and Funk, C. D. 1997 5-Lipoxygenase
products are necessary for ovalbumin-induced airway responsiveness in
mice. Am. J. Physiol., 272: L1053-8.
Ishigami, Y. G.; Masud, M.; Takizawa, T.; and Suzuki, Y. 1999 The crystal
structure and the fluorescent properties of curcumin. Shikizai Kyokaishi,
72: 71-77.
Izumi, T.; Hoshiko, S.; Radmark, O.; and Samuelsson, B. 1990 Cloning of the
cDNA for human 12-lipoxygenase. Proc. Natl. Acad. Sci. U .S. A., 87:
7477-81.
Jackson, M. W.; Bentel, J. M.; and Tilley, W. D. 1997 Vascular endothelial
growth factor (VEGF) expression in prostate cancer and benign prostatic
hyperplasia. J. Urol., 157: 2323-8.
Jiang, B. H.; Agani, F.; Passaniti, A.; and Semenza, G. L. 1997 V-SRC induces
expression of hypoxia-inducible factor 1 (HIF-1) and transcription of genes
encoding vascular endothelial growth factor and enolase 1: involvement of
HIF-1 in tumor progression. Cancer Res., 57: 5328-35.
Johnson, E. N.; Brass, L. F.; and Funk, C. D. 1998 Increased platelet sensitivity
to ADP in mice lacking platelet-type 12-lipoxygenase. Proc. Natl. Acad.
Sci. U .S. A., 95: 3100-5.
177 Johnson, G. 2002 Signal transduction. Scaffolding proteins--more than meets
the eye. Science, 295: 1249-50.
Joseph, I. B.; and Isaacs, J. T. 1997 Potentiation of the antiangiogenic ability of
linomide by androgen ablation involves down-regulation of vascular
endothelial growth factor in human androgen-responsive prostatic
cancers. Cancer Res., 57: 1054-7.
Kamitani, H.; Geller, M.; and Eling, T. 1999 The possible involvement of 15-
lipoxygenase/leukocyte type 12-lipoxygenase in colorectal carcinogenesis.
Adv. Exp. Med. Biol., 469: 593-8.
Kandouz, M.; Nie, D.; Pidgeon, G. P.; Krishnamoorthy, S.; Maddipati, K. R.; and
Honn, K. V. 2003 Platelet-type 12-lipoxygenase activates NF-kappaB in
prostate cancer cells. Prostaglandins Other Lipid Mediat., 71: 189-204.
Kelavkar, U. P.; Nixon, J. B.; Cohen, C.; Dillehay, D.; Eling, T. E.; and Badr, K. F.
2001 Overexpression of 15-lipoxygenase-1 in PC-3 human prostate
cancer cells increases tumorigenesis. Carcinogenesis, 22: 1765-73.
Kelman, C.; Krupinski-Oslen, R.; and Shorter, A. 1987 Reductive inactivation of
soybean lipoxygenase 1 by catechols: a possible mechanism for
regulation of lipoxygenase activity. Biochemistry, 26: 7064-7072.
Kieser, A.; Weich, H. A.; Brandner, G.; Marme, D.; and Kolch, W. 1994 Mutant
p53 potentiates protein kinase C induction of vascular endothelial growth
factor expression. Oncogene, 9: 963-9.
Kim, J. A.; Gu, J. L.; Natarajan, R.; Berliner, J. A.; and Nadler, J. L. 1995 A
leukocyte type of 12-lipoxygenase is expressed in human vascular and
178 mononuclear cells. Evidence for upregulation by angiotensin II.
Arterioscler. Thromb. Vasc. Biol., 15: 942-8.
Kim, K. J.; Li, B.; Winer, J.; Armanini, M.; Gillett, N.; Phillips, H. S.; and Ferrara,
N. 1993 Inhibition of vascular endothelial growth factor-induced
angiogenesis suppresses tumour growth in vivo. Nature, 362: 841-4.
Kirschenbaum, A.; Wang, J. P.; Ren, M.; Schiff, J. D.; Aaronson, S. A.; Droller,
M. J.; Ferrara, N.; Holland, J. F.; and Levine, A. C. 1997 Inhibition of
vascular endothelial cell growth factor suppresses the in vivo growth of
human prostate tumors. Urol. Oncol., 3: 3-10.
Kishimoto, K.; Nakamura, M.; Suzuki, H.; Yoshimoto, T.; Yamamoto, S.; Takao,
T.; Shimonishi, Y.; and Tanabe, T. 1996 Suicide inactivation of porcine
leukocyte 12-lipoxygenase associated with its incorporation of 15-
hydroperoxy-5,8,11,13-eicosatetraenoic acid derivative. Biochim. Biophys.
Acta, 1300: 56-62.
Kolch, W. 2000 Meaningful relationships: the regulation of the
Ras/Raf/MEK/ERK pathway by protein interactions. Biochem. J., 351 Pt 2:
289-305.
Korutla, L.; Cheung, J. Y.; Mendelsohn, J.; and Kumar, R. 1995 Inhibition of
ligand-induced activation of epidermal growth factor receptor tyrosine
phosphorylation by curcumin. Carcinogenesis, 16: 1741-5.
Krueger, J. S.; Keshamouni, V. G.; Atanaskova, N.; and Reddy, K. B. 2001
Temporal and quantitative regulation of mitogen-activated protein kinase
(MAPK) modulates cell motility and invasion. Oncogene, 20: 4209-18.
179 Kuchinke, W.; and Funk, C. D. 1994 Fibronectin-induced cell spreading and
down-regulation of 12-lipoxygenase expression in megakaryocytic DAMI
cells. Biochem. Biophys. Res. Commun., 204: 606-12.
Kuhn, H.; and Chan, L. 1997 The role of 15-lipoxygenase in atherogenesis:
pro- and antiatherogenic actions. Curr. Opin. Lipidol., 8: 111-7.
Kuhn, H.; Sprecher, H.; and Brash, A. R. 1990 On singular or dual positional
specificity of lipoxygenases. The number of chiral products varies with
alignment of methylene groups at the active site of the enzyme. J. Biol.
Chem., 265: 16300-5.
Kuhn, H.; and Thiele, B. J. 1999 The diversity of the lipoxygenase family. Many
sequence data but little information on biological significance. FEBS Lett.,
449: 7-11.
Kulkarni, A. P.; Cai, Y.; and Richards, I. S. 1992 Rat pulmonary lipoxygenase:
dioxygenase activity and role in xenobiotic metabolism. Int. J. Biochem.,
24: 255-61.
Kumar, A. P.; Rajnarayanan, R.; Garcia, G. E.; and et al. 2002. Novel curcumin
derivatives induce apoptosis through AKT-NFKB signaling in human
prostate cancer cells. Proc. 93rd Annual Meeting AACR.
Kunsch, C.; and Rosen, C. A. 1993 NF-kappa B subunit-specific regulation of
the interleukin-8 promoter. Mol. Cell Biol., 13: 6137-46.
Kurahashi, Y.; Herbertsson, H.; Soderstrom, M.; Rosenfeld, M. G.; and
Hammarstrom, S. 2000 A 12(S)-hydroxyeicosatetraenoic acid receptor
180 interacts with steroid receptor coactivator-1. Proc. Natl. Acad. Sci. U .S
.A., 97: 5779-83.
Kuroki, M.; Voest, E. E.; Amano, S.; Beerepoot, L. V.; Takashima, S.; Tolentino,
M.; Kim, R. Y.; Rohan, R. M.; Colby, K. A.; Yeo, K. T.; and Adamis, A. P.
1996 Reactive oxygen intermediates increase vascular endothelial growth
factor expression in vitro and in vivo. J. Clin. Invest., 98: 1667-75.
Lagarde, M.; Croset, M.; Authi, K. S.; and Crawford, N. 1984 Subcellular
localization and some properties of lipoxygenase activity in human blood
platelets. Biochem. J., 222: 495-500.
Lee, S. A.; Park, J. K.; Kang, E. K.; Bae, H. R.; Bae, K. W.; and Park, H. T. 2000
Calmodulin-dependent activation of p38 and p42/44 mitogen-activated
protein kinases contributes to c-fos expression by calcium in PC12 cells:
modulation by nitric oxide. Brain Res. Mol. Brain Res., 75: 16-24.
Li, J.; Perrella, M. A.; Tsai, J. C.; Yet, S. F.; Hsieh, C. M.; Yoshizumi, M.;
Patterson, C.; Endege, W. O.; Zhou, F.; and Lee, M. E. 1995 Induction of
vascular endothelial growth factor gene expression by interleukin-1 beta in
rat aortic smooth muscle cells. J. Biol. Chem., 270: 308-12.
Li, Q.; Cheon, Y. P.; Kannan, A.; Shankar, S.; Bagchi, I. C.; and Bagchi, M. K.
2004 A novel pathway involving progesterone receptor, 12/15-
lipoxygenase-derived Eicosanoids, and peroxisome proliferator-activated
receptor gamma regulates implantation in mice. J. Biol. Chem., 279:
11570-81.
181 Liaw, Y. W.; Liu, Y. W.; Chen, B. K.; and Chang, W. C. 1998 Induction of 12-
lipoxygenase expression by phorbol 12-myristate 13-acetate in human
epidermoid carcinoma A431 cells. Biochim. Biophys. Acta, 1389: 23-33.
Liminga, M.; and Oliw, E. 1999 cDNA cloning of 15-lipoxygenase type 2 and 12-
lipoxygenases of bovine corneal epithelium. Biochim. Biophys. Acta, 1437:
124-35.
Liu, B.; Maher, R. J.; Hannun, Y. A.; Porter, A. T.; and Honn, K. V. 1994a 12(S)-
HETE enhancement of prostate tumor cell invasion: selective role of PKC
alpha. J. Natl. Cancer. Inst., 86: 1145-51.
Liu, B.; Marnett, L. J.; Chaudhary, A.; Ji, C.; Blair, I. A.; Johnson, C. R.; Diglio, C.
A.; and Honn, K. V. 1994b Biosynthesis of 12(S)-
hydroxyeicosatetraenoic acid by B16 amelanotic melanoma cells is a
determinant of their metastatic potential. Lab. Invest., 70: 314-23.
Liu, X. H.; Connolly, J. M.; and Rose, D. P. 1996 Eicosanoids as mediators of
linoleic acid-stimulated invasion and type IV collagenase production by a
metastatic human breast cancer cell line. Clin. Exp. Metastasis, 14: 145-
52.
Liu, Y.; Cox, S. R.; Morita, T.; and Kourembanas, S. 1995 Hypoxia regulates
vascular endothelial growth factor gene expression in endothelial cells.
Identification of a 5' enhancer. Circ. Res., 77: 638-43.
Liu, Y. W.; Arakawa, T.; Yamamoto, S.; and Chang, W. C. 1997 Transcriptional
activation of human 12-lipoxygenase gene promoter is mediated through
Sp1 consensus sites in A431 cells. Biochem. J., 324 ( Pt 1): 133-40.
182 Liu, Y. W.; Asaoka, Y.; Suzuki, H.; Yoshimoto, T.; Yamamoto, S.; and Chang, W.
C. 1994 Induction of 12-lipoxygenase expression by epidermal growth
factor is mediated by protein kinase C in A431 cells. J. Pharmacol. Exp.
Ther., 271: 567-73.
Lysz, T. W.; Wu, Y.; Brash, A.; Keeting, P. E.; Lin, C.; and Fu, S. C. 1991
Identification of 12(S)-hydroxyeicosatetraenoic acid in the young rat lens.
Curr. Eye Res., 10: 331-7.
Mahmud, I.; Suzuki, T.; Yamamoto, Y.; Suzuki, H.; Takahashi, Y.; Yoshimoto, T.;
and Yamamoto, S. 1993 Induction of cyclooxygenase and suppression of
12-lipoxygenase in human erythroleukemia cells upon phorbol ester-
induced differentiation. Biochim. Biophys. Acta, 1166: 211-6.
Makarov, S. S.; Johnston, W. N.; Olsen, J. C.; Watson, J. M.; Mondal, K.;
Rinehart, C.; and Haskill, J. S. 1997 NF-kappa B as a target for anti-
inflammatory gene therapy: suppression of inflammatory responses in
monocytic and stromal cells by stable gene transfer of I kappa B alpha
cDNA. Gene Ther., 4: 846-52.
Mansour, S. J.; Matten, W. T.; Hermann, A. S.; Candia, J. M.; Rong, S.;
Fukasawa, K.; Vande Woude, G. F.; and Ahn, N. G. 1994 Transformation
of mammalian cells by constitutively active MAP kinase kinase. Science,
265: 966-70.
Marais, R.; Wynne, J.; and Treisman, R. 1993 The SRF accessory protein Elk-1
contains a growth factor-regulated transcriptional activation domain. Cell,
73: 381-93.
183 Marengo, S. R.; Sikes, R. A.; Anezinis, P.; Chang, S. M.; and Chung, L. W. 1997
Metastasis induced by overexpression of p185neu-T after orthotopic
injection into a prostatic epithelial cell line (NbE). Mol. Carcinog., 19: 165-
75.
Marshall, C. J. 1995 Specificity of receptor tyrosine kinase signaling: transient
versus sustained extracellular signal-regulated kinase activation. Cell, 80:
179-85.
Matsuda, S.; Murakami, J.; Yamamoto, Y.; Konishi, Y.; Yokoyama, C.;
Yoshimoto, T.; Yamamoto, S.; Mimura, Y.; and Okuma, M. 1993
Decreased messenger RNA of arachidonate 12-lipoxygenase in platelets
of patients with myeloproliferative disorders. Biochim. Biophys. Acta,
1180: 243-9.
Melnyk, O.; Zimmerman, M.; Kim, K. J.; and Shuman, M. 1999 Neutralizing anti-
vascular endothelial growth factor antibody inhibits further growth of
established prostate cancer and metastases in a pre-clinical model. J.
Urol., 161: 960-3.
Mezentsev, A.; Seta, F.; Dunn, M. W.; Ono, N.; Falck, J. R.; and Laniado-
Schwartzman, M. 2002 Eicosanoid regulation of vascular endothelial
growth factor expression and angiogenesis in microvessel endothelial
cells. J. Biol. Chem., 277: 18670-6.
Milanini, J.; Vinals, F.; Pouyssegur, J.; and Pages, G. 1998 p42/p44 MAP
kinase module plays a key role in the transcriptional regulation of the
184 vascular endothelial growth factor gene in fibroblasts. J. Biol. Chem., 273:
18165-72.
Millauer, B.; Shawver, L. K.; Plate, K. H.; Risau, W.; and Ullrich, A. 1994
Glioblastoma growth inhibited in vivo by a dominant-negative Flk-1
mutant. Nature, 367: 576-9.
Minchenko, A.; Bauer, T.; Salceda, S.; and Caro, J. 1994 Hypoxic stimulation of
vascular endothelial growth factor expression in vitro and in vivo. Lab.
Invest., 71: 374-9.
Minor, W.; Steczko, J.; Stec, B.; Otwinowski, Z.; Bolin, J. T.; Walter, R.; and
Axelrod, B. 1996 Crystal structure of soybean lipoxygenase L-1 at 1.4 A
resolution. Biochemistry, 35: 10687-701.
Moodie, S. A.; and Wolfman, A. 1994 The 3Rs of life: Ras, Raf and growth
regulation. Trends Genet., 10: 44-8.
Moody, T. W.; Leyton, J.; Martinez, A.; Hong, S.; Malkinson, A.; and Mulshine, J.
L. 1998 Lipoxygenase inhibitors prevent lung carcinogenesis and inhibit
non-small cell lung cancer growth. Exp. Lung. Res., 24: 617-28.
Moon, W. C.; Choi, H. R.; and Moon, S. Y. 1997 Endocrine therapy inhibits
expression of vascular endothelial growth factor and angiogenesis in
prostate cancer. J. Urol., 157: 223.
Mukaida, N.; Morita, M.; Ishikawa, Y.; Rice, N.; Okamoto, S.; Kasahara, T.; and
Matsushima, K. 1994 Novel mechanism of glucocorticoid-mediated gene
repression. Nuclear factor-kappa B is target for glucocorticoid-mediated
interleukin 8 gene repression. J. Biol. Chem., 269: 13289-95.
185 Mukhopadhyay, D.; Tsiokas, L.; Zhou, X. M.; Foster, D.; Brugge, J. S.; and
Sukhatme, V. P. 1995 Hypoxic induction of human vascular endothelial
growth factor expression through c-Src activation. Nature, 375: 577-81.
Nakamura, M.; Ueda, N.; Kishimoto, K.; Yoshimoto, T.; Yamamoto, S.; and
Ishimura, K. 1995 Immunocytochemical localization of platelet-type
arachidonate 12-lipoxygenase in mouse blood cells. J. Histochem.
Cytochem., 43: 237-44.
Nakamura, M.; Ueda, N.; Yamamoto, S.; Ishimura, K.; Uchida, N.; and Arase, S.
1997 Tissue distribution and subcellular localization of platelet-type
arachidonate 12-lipoxygenase. Adv. Exp. Med. Biol., 407: 15-20.
Natarajan, R.; Bai, W.; Lanting, L.; Gonzales, N.; and Nadler, J. 1997a Effects
of high glucose on vascular endothelial growth factor expression in
vascular smooth muscle cells. Am. J. Physiol., 273: H2224-31.
Natarajan, R.; Esworthy, R.; Bai, W.; Gu, J. L.; Wilczynski, S.; and Nadler, J.
1997b Increased 12-lipoxygenase expression in breast cancer tissues
and cells. Regulation by epidermal growth factor. J. Clin. Endocrinol.
Metab., 82: 1790-8.
Natarajan, R.; Gu, J. L.; Rossi, J.; Gonzales, N.; Lanting, L.; Xu, L.; and Nadler,
J. 1993 Elevated glucose and angiotensin II increase 12-lipoxygenase
activity and expression in porcine aortic smooth muscle cells. Proc. Natl.
Acad. Sci. U. S. A., 90: 4947-51.
186 Natarajan, R.; Rosdahl, J.; Gonzales, N.; and Bai, W. 1997c Regulation of 12-
lipoxygenase by cytokines in vascular smooth muscle cells. Hypertension,
30: 873-9.
Needleman, P.; Turk, J.; Jakschik, B. A.; Morrison, A. R.; and Lefkowith, J. B.
1986 Arachidonic acid metabolism. Annu. Rev. Biochem., 55: 69-102.
Neer, E. J. 1995 Heterotrimeric G proteins: organizers of transmembrane
signals. Cell, 80: 249-57.
Nelson, M. J.; Chase, D. B.; and Seitz, S. P. 1995 Photolysis of "purple"
lipoxygenase: implications for the structure of the chromophore.
Biochemistry, 34: 6159-63.
Nelson, M. J.; Cowling, R. A.; and Seitz, S. P. 1994 Structural characterization
of alkyl and peroxyl radicals in solutions of purple lipoxygenase.
Biochemistry, 33: 4966-73.
Nelson, M. J.; Seitz, S. P.; and Cowling, R. A. 1990 Enzyme-bound pentadienyl
and peroxyl radicals in purple lipoxygenase. Biochemistry, 29: 6897-903.
Nemeth, J. A.; Harb, J. F.; Barroso, U., Jr.; He, Z.; Grignon, D. J.; and Cher, M.
L. 1999 Severe combined immunodeficient-hu model of human prostate
cancer metastasis to human bone. Cancer Res., 59: 1987-93.
Neufeld, G.; Cohen, T.; Gengrinovitch, S.; and Poltorak, Z. 1999 Vascular
endothelial growth factor (VEGF) and its receptors. Faseb J., 13: 9-22.
New, B. A.; and Yeoman, L. C. 1992 Identification of basic fibroblast growth
factor sensitivity and receptor and ligand expression in human colon tumor
cell lines. J. Cell Physiol., 150: 320-6.
187 Nie, D.; Che, M.; Grignon, D.; Tang, K.; and Honn, K. V. 2001 Role of
eicosanoids in prostate cancer progression. Cancer Metastasis Rev., 20:
195-206.
Nie, D.; Hillman, G. G.; Geddes, T.; Tang, K.; Pierson, C.; Grignon, D. J.; and
Honn, K. V. 1998 Platelet-type 12-lipoxygenase in a human prostate
carcinoma stimulates angiogenesis and tumor growth. Cancer Res., 58:
4047-51.
Nie, D.; and Honn, K. V. 2004 Eicosanoid regulation of angiogenesis in tumors.
Semin. Thromb. Hemost., 30: 119-25.
Nie, D.; Lamberti, M.; Zacharek, A.; Li, L.; Szekeres, K.; Tang, K.; Chen, Y.; and
Honn, K. V. 2000a Thromboxane A(2) regulation of endothelial cell
migration, angiogenesis, and tumor metastasis. Biochem. Biophys. Res.
Commun., 267: 245-51.
Nie, D.; Nemeth, J.; Qiao, Y.; Zacharek, A.; Li, L.; Hanna, K.; Tang, K.; Hillman,
G. G.; Cher, M. L.; Grignon, D. J.; and Honn, K. V. 2003 Increased
metastatic potential in human prostate carcinoma cells by overexpression
of arachidonate 12-lipoxygenase. Clin. Exp. Metastasis, 20: 657-63.
Nie, D.; Tang, K.; Diglio, C.; and Honn, K. V. 2000b Eicosanoid regulation of
angiogenesis: role of endothelial arachidonate 12-lipoxygenase. Blood,
95: 2304-11.
Nugteren, D. H. 1975 Arachidonate lipoxygenase in blood platelets. Biochim.
Biophys. Acta, 380: 299-307.
188 Nyby, M. D.; Sasaki, M.; Ideguchi, Y.; Wynne, H. E.; Hori, M. T.; Berger, M. E.;
Golub, M. S.; Brickman, A. S.; and Tuck, M. L. 1996 Platelet
lipoxygenase inhibitors attenuate thrombin- and thromboxane mimetic-
induced intracellular calcium mobilization and platelet aggregation. J.
Pharmacol. Exp. Ther., 278: 503-9.
Ochi, K.; Yoshimoto, T.; Yamamoto, S.; Taniguchi, K.; and Miyamoto, T. 1983
Arachidonate 5-lipoxygenase of guinea pig peritoneal polymorphonuclear
leukocytes. Activation by adenosine 5'-triphosphate. J. Biol. Chem., 258:
5754-8.
Otwinowski, Z. and Minor, W. 1997 Methods of Enzymology, 276: 307-326.
Ozeki, Y.; Nagamura, Y.; Ito, H.; Unemi, F.; Kimura, Y.; Igawa, T.; Kambayashi,
J.; Takahashi, Y.; and Yoshimoto, T. 1999 An anti-platelet agent, OPC-
29030, inhibits translocation of 12-lipoxygenase and 12-
hydroxyeicosatetraenoic acid production in human platelets. Br. J.
Pharmacol., 128: 1699-704.
Pace-Asciak, C. R. 1993 Hepoxilins. Gen. Pharmacol., 24: 805-10.
Pace-Asciak, C. R.; Demin, P. M.; Estrada, M.; and Liu, G. 1999 Hepoxilins
raise circulating insulin levels in vivo. FEBS Lett., 461: 165-8.
Pace-Asciak, C. R.; Hahn, S.; Diamandis, E. P.; Soleas, G.; and Goldberg, D. M.
1995 The red wine phenolics trans-resveratrol and quercetin block human
platelet aggregation and eicosanoid synthesis: implications for protection
against coronary heart disease. Clin. Chim. Acta, 235: 207-19.
189 Pertovaara, L.; Kaipainen, A.; Mustonen, T.; Orpana, A.; Ferrara, N.; Saksela, O.;
and Alitalo, K. 1994 Vascular endothelial growth factor is induced in
response to transforming growth factor-beta in fibroblastic and epithelial
cells. J. Biol. Chem., 269: 6271-4.
Pham, C.; Jankun, J.; Skrzypczak-Jankun, E.; Flowers, R. A., 2nd; and Funk, M.
O., Jr. 1998 Structural and thermochemical characterization of
lipoxygenase-catechol complexes. Biochemistry, 37: 17952-7.
Pidgeon, G. P.; Tang, K.; Cai, Y. L.; Piasentin, E.; and Honn, K. V. 2003
Overexpression of platelet-type 12-lipoxygenase promotes tumor cell
survival by enhancing alpha(v)beta(3) and alpha(v)beta(5) integrin
expression. Cancer Res., 63: 4258-67.
Poltorak, Z.; Cohen, T.; Sivan, R.; Kandelis, Y.; Spira, G.; Vlodavsky, I.; Keshet,
E.; and Neufeld, G. 1997 VEGF145, a secreted vascular endothelial
growth factor isoform that binds to extracellular matrix. J. Biol. Chem.,
272: 7151-8.
Porta, H.; and Rocha-Sosa, M. 2001 Lipoxygenase in bacteria: a horizontal
transfer event? Microbiology, 147: 3199-200.
Pourplanche, C.; Lambert, C.; Berjot, M.; Marx, J.; Chopard, C.; Alix, A. J.; and
Larreta-Garde, V. 1994 Conformational changes of lipoxygenase (LOX)
in modified environments. Contribution to the variation in specificity of
soybean LOX type 1. J. Biol. Chem., 269: 31585-91.
190 Prigge, S. T.; Boyington, J. C.; Faig, M.; Doctor, K. S.; Gaffney, B. J.; and Amzel,
L. M. 1997 Structure and mechanism of lipoxygenases. Biochimie., 79:
629-36.
Prigge, S. T.; Boyington, J. C.; Gaffney, B. J.; and Amzel, L. M. 1996 Structure
conservation in lipoxygenases: structural analysis of soybean lipoxygenase-
1 and modeling of human lipoxygenases. Proteins, 24: 275-91.
Poltorak, Z.; Cohen, T.; Sivan, R.; Kandelis, Y.; Spira, G.; Vlodavsky, I.; Keshet,
E.; and Neufeld, G. 1997 VEGF145, a secreted vascular endothelial
growth factor isoform that binds to extracellular matrix. J. Biol. Chem.,
272: 7151-7158.
Putz, T.; Culig, Z.; Eder, I. E.; Nessler-Menardi, C.; Bartsch, G.; Grunicke, H.;
Uberall, F.; and Klocker, H. 1999 Epidermal growth factor (EGF) receptor
blockade inhibits the action of EGF, insulin-like growth factor I, and a
protein kinase A activator on the mitogen-activated protein kinase pathway
in prostate cancer cell lines. Cancer Res., 59: 227-33.
Rak, J.;Mitsuhashi, Y.; Bayko, L.; Filmus, J.; Shirasawa, S.; Sasazuki, T.; and
Kerbel, R. S. 1995 Mutant ras oncogenes upregulate VEGF/VPF
expression: implications for induction and inhibition of tumor angiogenesis.
Cancer Res., 55: 4575-80.
Rao, C. V.; Rivenson, A.; Simi, B.; and Reddy, B. S. 1995 Chemoprevention of
colon cancer by dietary curcumin. Ann. N. Y. Acad. Sci., 768: 201-4.
Rao, C. V.; Simi, B.; and Reddy, B. S. 1993 Inhibition by dietary curcumin of
azoxymethane-induced ornithine decarboxylase, tyrosine protein kinase,
191 arachidonic acid metabolism and aberrant crypt foci formation in the rat
colon. Carcinogenesis, 14: 2219-25.
Reddy, M. A.; Thimmalapura, P. R.; Lanting, L.; Nadler, J. L.; Fatima, S.; and
Natarajan, R. 2002 The oxidized lipid and lipoxygenase product 12(S)-
hydroxyeicosatetraenoic acid induces hypertrophy and fibronectin
transcription in vascular smooth muscle cells via p38 MAPK and cAMP
response element-binding protein activation. Mediation of angiotensin II
effects. J. Biol. Chem., 277: 9920-8.
Reisinger, K.; Kaufmann, R.; and Gille, J. 2003 Increased Sp1 phosphorylation
as a mechanism of hepatocyte growth factor (HGF/SF)-induced vascular
endothelial growth factor (VEGF/VPF) transcription. J. Cell Sci., 116: 225-
38.
Rickert, K. W.; and Klinman, J. P. 1999 Nature of hydrogen transfer in soybean
lipoxygenase 1: separation of primary and secondary isotope effects.
Biochemistry, 38: 12218-28.
Rioux, N.; and Castonguay, A. 1998 Inhibitors of lipoxygenase: a new class of
cancer chemopreventive agents. Carcinogenesis, 19: 1393-400.
Risau, W. 1997 Mechanisms of angiogenesis. Nature, 386: 671-4.
Roberts, W. G.; and Palade, G. E. 1997 Neovasculature induced by vascular
endothelial growth factor is fenestrated. Cancer Res., 57: 765-72.
Robinson, M. J.; and Cobb, M. H. 1997 Mitogen-activated protein kinase
pathways. Curr. Opin. Cell Biol., 9: 180-6.
192 Romano, M.; Catalano, A.; Nutini, M.; D'Urbano, E.; Crescenzi, C.; Claria, J.;
Libner, R.; Davi, G.; and Procopio, A. 2001 5-lipoxygenase regulates
malignant mesothelial cell survival: involvement of vascular endothelial
growth factor. Faseb J., 15: 2326-36.
Roth, G. J.; Machuga, E. T.; and Strittmatter, P. 1981 The heme-binding
properties of prostaglandin synthetase from sheep vesicular gland. J. Biol.
Chem., 256: 10018-22.
Rouzer, C. A.; and Samuelsson, B. 1987 5-Lipoxygenase from human
leukocytes associates with membrane in the presence of calcium. Adv.
Prostaglandin Thromboxane Leukot. Res., 17A: 60-3.
Rouzer, C. A.; Shimizu, T.; and Samuelsson, B. 1985 On the nature of the 5-
lipoxygenase reaction in human leukocytes: characterization of a
membrane-associated stimulatory factor. Proc. Natl. Acad. Sci. U. S. A.,
82: 7505-9.
Roy, P.; Roy, S. K.; Mitra, A.; and Kulkarni, A. P. 1994 Superoxide generation
by lipoxygenase in the presence of NADH and NADPH. Biochim. Biophys.
Acta, 1214: 171-9.
Ryuto, M.; Ono, M.; Izumi, H.; Yoshida, S.; Weich, H. A.; Kohno, K.; and
Kuwano, M. 1996 Induction of vascular endothelial growth factor by
tumor necrosis factor alpha in human glioma cells. Possible roles of SP-1.
J. Biol. Chem., 271: 28220-8.
193 Schaeffer, H. J.; and Weber, M. J. 1999 Mitogen-activated protein kinases:
specific messages from ubiquitous messengers. Mol. Cell Biol., 19: 2435-
44.
Schafer, A. I. 1982 Deficiency of platelet lipoxygenase activity in
myeloproliferative disorders. N. Engl. J. Med., 306: 381-6.
Schewe, T.; Sadik, C.; Klotz, L. O.; Yoshimoto, T.; Kuhn, H.; and Sies, H. 2001
Polyphenols of cocoa: inhibition of mammalian 15-lipoxygenase. Biol.
Chem., 382: 1687-96.
Schneider, C.; Amberg, A.; Feurle, J.; Roß, A.; Roth, M.; Tóth, G.; and Schreier,
P. 1998 2-[(4"-Hydroxy-3'-methoxy)-phenoxy]-4-(4"-hydroxy-3"-methoxy-
phenyl)-8-hydroxy-6-oxo-3-oxabicylo[3.3.0]-7-octene: unusual product of
the soybean lipoxygenase-catalyzed oxygenation of curcumin. Journal of
Molecular Catalysis B: Enzymatic, 4: 219-227.
Schwarz, K.; Walther, M.; Anton, M.; Gerth, C.; Feussner, I.; and Kuhn, H. 2001
Structural basis for lipoxygenase specificity. Conversion of the human
leukocyte 5-lipoxygenase to a 15-lipoxygenating enzyme species by site-
directed mutagenesis. J. Biol. Chem., 276: 773-9.
Seger, R.; and Krebs, E. G. 1995 The MAPK signaling cascade. Faseb J., 9:
726-35.
Seger, R.; Seger, D.; Lozeman, F. J.; Ahn, N. G.; Graves, L. M.; Campbell, J. S.;
Ericsson, L.; Harrylock, M.; Jensen, A. M.; and Krebs, E. G. 1992 Human
T-cell mitogen-activated protein kinase kinases are related to yeast signal
transduction kinases. J. Biol. Chem., 267: 25628-31.
194 Sekiya, F.; Takagi, J.; Usui, T.; Kawajiri, K.; Kobayashi, Y.; Sato, F.; and Saito, Y.
1991 12S-hydroxyeicosatetraenoic acid plays a central role in the
regulation of platelet activation. Biochem. Biophys. Res. Commun., 179:
345-51.
Setty, B. N.; Dubowy, R. L.; and Stuart, M. J. 1987 Endothelial cell proliferation
may be mediated via the production of endogenous lipoxygenase
metabolites. Biochem. Biophys. Res. Commun., 144: 345-51.
Shalaby, F.; Rossant, J.; Yamaguchi, T. P.; Gertsenstein, M.; Wu, X. F.;
Breitman, M. L.; and Schuh, A. C. 1995 Failure of blood-island formation
and vasculogenesis in Flk-1-deficient mice. Nature, 376: 62-6.
Shannon, V. R.; Stenson, W. F.; and Holtzman, M. J. 1993 Induction of
epithelial arachidonate 12-lipoxygenase at active sites of inflammatory
bowel disease. Am. J. Physiol., 264: G104-11.
Shappell, S. B.; Boeglin, W. E.; Olson, S. J.; Kasper, S.; and Brash, A. R. 1999
15-lipoxygenase-2 (15-LOX-2) is expressed in benign prostatic epithelium
and reduced in prostate adenocarcinoma. Am. J. Pathol., 155: 235-45.
Shibata, D.; and Axelrod, B. 1995 Plant lipoxygenases. J. Lipid. Mediat. Cell
Signal., 12: 213-28.
Shoba, G.; Joy, D.; Joseph, T.; Majeed, M.; Rajendran, R.; and Srinivas, P. S.
1998 Influence of piperine on the pharmacokinetics of curcumin in
animals and human volunteers. Planta. Med., 64: 353-6.
195 Shweiki, D.; Itin, A.; Soffer, D.; and Keshet, E. 1992 Vascular endothelial growth
factor induced by hypoxia may mediate hypoxia-initiated angiogenesis.
Nature, 359: 843-5.
Shweiki, D.; Neeman, M.; Itin, A.; and Keshet, E. 1995 Induction of vascular
endothelial growth factor expression by hypoxia and by glucose deficiency
in multicell spheroids: implications for tumor angiogenesis. Proc. Natl.
Acad. Sci. U. S .A., 92: 768-72.
Siebert, M.; Krieg, P.; Lehmann, W. D.; Marks, F.; and Furstenberger, G. 2001
Enzymic characterization of epidermis-derived 12-lipoxygenase
isoenzymes. Biochem. J., 355: 97-104.
Siemeister, G.; Weindel, K.; Mohrs, K.; Barleon, B.; Martiny-Baron, G.; and
Marme, D. 1996 Reversion of deregulated expression of vascular
endothelial growth factor in human renal carcinoma cells by von Hippel-
Lindau tumor suppressor protein. Cancer Res., 56: 2299-301.
Sigal, E.; Craik, C. S.; Highland, E.; Grunberger, D.; Costello, L. L.; Dixon, R. A.;
and Nadel, J. A. 1988 Molecular cloning and primary structure of human
15-lipoxygenase. Biochem. Biophys. Res. Commun., 157: 457-64.
Silletti, S.; Timar, J.; Honn, K. V.; and Raz, A. 1994 Autocrine motility factor
induces differential 12-lipoxygenase expression and activity in high- and
low-metastatic K1735 melanoma cell variants. Cancer Res., 54: 5752-6.
Sindhwani, P.; Hampton, J. A.; Baig, M. M.; Keck, R.; and Selman, S. H. 2001
Curcumin prevents intravesical tumor implantation of the MBT-2 tumor cell
line in C3H mice. J. Urol., 166: 1498-501.
196 Skrzypczak-Jankun, E. 1996 Flash-freezing causes a stress-induced
modulation in a crystal structure of soybean lipoxygenase l3. Acta
Crystallogr. D. Biol. Crystallogr., 52: 959-65.
Skrzypczak-Jankun, E., et al. 2000a “Purple” Lipoxygenase - X-Ray Analysis of
Complexes with Three Different Peroxides. American Crystallographic
Association Meeting., St. Paul, MN.
Skrzypczak-Jankun, E.; Amzel, L. M.; Kroa, B. A.; and Funk, M. O., Jr. 1997
Structure of soybean lipoxygenase L3 and a comparison with its L1
isoenzyme. Proteins, 29: 15-31.
Skrzypczak-Jankun, E.; Bross, R. A.; Carroll, R. T.; Dunham, W. R.; and Funk,
M. O., Jr. 2001 Three-dimensional structure of a purple lipoxygenase. J.
Am. Chem. Soc., 123: 10814-20.
Skrzypczak-Jankun, E.; McCabe, N. P.; Selman, S. H.; and Jankun, J. 2000b
Curcumin inhibits lipoxygenase by binding to its central cavity: theoretical
and X-ray evidence. Int. J. Mol. Med., 6: 521-6.
Skrzypczak-Jankun, E.; Zhou, K.; McCabe, N. P.; et al. 2002 Natural
polyphenols in comples with lipoxygenase - structural studies and their
relevance to cancer. Proc 93rd Annual Meeting AACR.
Sloane, D. L.; Leung, R.; Barnett, J.; Craik, C. S.; and Sigal, E. 1995
Conversion of human 15-lipoxygenase to an efficient 12-lipoxygenase: the
side-chain geometry of amino acids 417 and 418 determine positional
specificity. Protein Eng., 8: 275-82.
197 Smith, R. D.; Loo, J. A.; Edmonds, C. G.; Barinaga, C. J.; and Udseth, H. R.
1990 New developments in biochemical mass spectrometry: electrospray
ionization. Anal. Chem., 62: 882-99.
Software. Brunger AT. X-plor ver. 3.1. A system for X-ray crystallography and
NMR, N. H., CT. Sack JS, Chain. A Crystallographic modeling program,
ver. 7,InsightII, ver. 98, Accelrys, Inc., San Diego, CA, Sybyl, ver. 6.8,
Tripos, Inc. St. Louis, MO.
Soriano, A. F.; Helfrich, B.; Chan, D. C.; Heasley, L. E.; Bunn, P. A., Jr.; and
Chou, T. C. 1999 Synergistic effects of new chemopreventive agents and
conventional cytotoxic agents against human lung cancer cell lines.
Cancer Res., 59: 6178-84.
Srivastava, K. C.; Bordia, A.; and Verma, S. K. 1995 Curcumin, a major
component of food spice turmeric (Curcuma longa) inhibits aggregation
and alters eicosanoid metabolism in human blood platelets.
Prostaglandins Leukot. Essent. Fatty Acids., 52: 223-7.
Steinert, B. W.; Tang, D. G.; Grossi, I. M.; Umbarger, L. A.; and Honn, K. V.
1993 Studies on the role of platelet eicosanoid metabolism and integrin
alpha IIb beta 3 in tumor-cell-induced platelet aggregation. Int. J. Cancer,
54: 92-101.
Stewart, R. J.; Panigrahy, D.; Flynn, E.; and Folkman, J. 2001 Vascular
endothelial growth factor expression and tumor angiogenesis are
regulated by androgens in hormone responsive human prostate
198 carcinoma: evidence for androgen dependent destabilization of vascular
endothelial growth factor transcripts. J. Urol., 165: 688-93.
Stoker, M.; and Gherardi, E. 1991 Regulation of cell movement: the motogenic
cytokines. Biochim. Biophys. Acta, 1072: 81-102.
Sugden, P. H.; and Clerk, A. 1997 Regulation of the ERK subgroup of MAP
kinase cascades through G protein-coupled receptors. Cell Signal., 9:
337-51.
Sutherland, M.; Shankaranarayanan, P.; Schewe, T.; and Nigam, S. 2001
Evidence for the presence of phospholipid hydroperoxide glutathione
peroxidase in human platelets: implications for its involvement in the
regulatory network of the 12-lipoxygenase pathway of arachidonic acid
metabolism. Biochem. J., 353: 91-100.
Suzuki, H.; Kishimoto, K.; Yoshimoto, T.; Yamamoto, S.; Kanai, F.; Ebina, Y.;
Miyatake, A.; and Tanabe, T. 1994 Site-directed mutagenesis studies on
the iron-binding domain and the determinant for the substrate oxygenation
site of porcine leukocyte arachidonate 12-lipoxygenase. Biochim. Biophys.
Acta, 1210: 308-16.
Szekeres, C. K.; Tang, K.; Trikha, M.; and Honn, K. V. 2000a Eicosanoid
activation of extracellular signal-regulated kinase1/2 in human epidermoid
carcinoma cells. J. Biol. Chem., 275: 38831-41.
Szekeres, C. K.; Trikha, M.; and Honn, K. V. 2002 12(S)-HETE, pleiotropic
functions, multiple signaling pathways. Adv. Exp. Med. Biol., 507: 509-15.
199 Szekeres, C. K.; Trikha, M.; Nie, D.; and Honn, K. V. 2000b Eicosanoid 12(S)-
HETE activates phosphatidylinositol 3-kinase. Biochem. Biophys. Res.
Commun., 275: 690-5.
Takahashi, Y.; Glasgow, W. C.; Suzuki, H.; Taketani, Y.; Yamamoto, S.; Anton,
M.; Kuhn, H.; and Brash, A. R. 1993a Investigation of the oxygenation of
phospholipids by the porcine leukocyte and human platelet arachidonate
12-lipoxygenases. Eur. J. Biochem., 218: 165-71.
Takahashi, Y.; Reddy, G. R.; Ueda, N.; Yamamoto, S.; and Arase, S. 1993b
Arachidonate 12-lipoxygenase of platelet-type in human epidermal cells. J.
Biol. Chem., 268: 16443-8.
Takahashi, Y.; Ueda, N.; and Yamamoto, S. 1988 Two immunologically and
catalytically distinct arachidonate 12-lipoxygenases of bovine platelets and
leukocytes. Arch. Biochem. Biophys., 266: 613-21.
Tanaka, T.; Kanai, H.; Sekiguchi, K.; Aihara, Y.; Yokoyama, T.; Arai, M.; Kanda,
T.; Nagai, R.; and Kurabayashi, M. 2000 Induction of VEGF gene
transcription by IL-1 beta is mediated through stress-activated MAP
kinases and Sp1 sites in cardiac myocytes. J. Mol. Cell. Cardiol., 32:
1955-67.
Tang, D. G.; Chen, Y. Q.; Diglio, C. A.; and Honn, K. V. 1993a Protein kinase C-
dependent effects of 12(S)-HETE on endothelial cell vitronectin receptor
and fibronectin receptor. J. Cell. Biol., 121: 689-704.
Tang, D. G.; Grossi, I. M.; Chen, Y. Q.; Diglio, C. A.; and Honn, K. V. 1993b
12(S)-HETE promotes tumor-cell adhesion by increasing surface
200 expression of alpha V beta 3 integrins on endothelial cells. Int. J. Cancer,
54: 102-11.
Tang, D. G.; and Honn, K. V. 1994 12-Lipoxygenase, 12(S)-HETE, and cancer
metastasis. Ann. N .Y. Acad. Sci., 744: 199-215.
Tang, D. G.; Renaud, C.; Stojakovic, S.; Diglio, C. A.; Porter, A.; and Honn, K. V.
1995 12(S)-HETE is a mitogenic factor for microvascular endothelial cells:
its potential role in angiogenesis. Biochem. Biophys. Res. Commun., 211:
462-8.
Tang, D. G.; Timar, J.; Grossi, I. M.; Renaud, C.; Kimler, V. A.; Diglio, C. A.;
Taylor, J. D.; and Honn, K. V. 1993c The lipoxygenase metabolite, 12(S)-
HETE, induces a protein kinase C-dependent cytoskeletal rearrangement
and retraction of microvascular endothelial cells. Exp. Cell. Res., 207:
361-75.
Tang, K.; Finley, R. L., Jr.; Nie, D.; and Honn, K. V. 2000 Identification of 12-
lipoxygenase interaction with cellular proteins by yeast two-hybrid
screening. Biochemistry, 39: 3185-91.
Terman, B. I.; Dougher-Vermazen, M.; Carrion, M. E.; Dimitrov, D.; Armellino, D.
C.; Gospodarowicz, D.; and Bohlen, P. 1992 Identification of the KDR
tyrosine kinase as a receptor for vascular endothelial cell growth factor.
Biochem. Biophys. Res. Commun., 187: 1579-86.
Theorell, H.; Holman, R. T.; and Akeson, A. 1947 Acta Chem. Scand., 1: 571-
576.
201 Timar, J.; Raso, E.; Dome, B.; Li, L.; Grignon, D.; Nie, D.; Honn, K. V.; and
Hagmann, W. 2000 Expression, subcellular localization and putative
function of platelet-type 12-lipoxygenase in human prostate cancer cell
lines of different metastatic potential. Int. J. Cancer, 87: 37-43.
Timar, J.; Raso, E.; Honn, K. V.; and Hagmann, W. 1999 12-lipoxygenase
expression in human melanoma cell lines. Adv. Exp. Med. Biol., 469: 617-
22.
Timar, J.; Tang, D.; Bazaz, R.; Haddad, M. M.; Kimler, V. A.; Taylor, J. D.; and
Honn, K. V. 1993 PKC mediates 12(S)-HETE-induced cytoskeletal
rearrangement in B16a melanoma cells. Cell. Motil. Cytoskeleton, 26: 49-
65.
Tischer, E.; Mitchell, R.; Hartman, T.; Silva, M.; Gospodarowicz, D.; Fiddes, J. C.;
and Abraham, J. A. 1991 The human gene for vascular endothelial
growth factor. Multiple protein forms are encoded through alternative exon
splicing. J. Biol. Chem., 266: 11947-54.
Tohjima, T.; Honda, N.; Mochizuki, K.; Kinoshita, J.; Watanabe, K.; Arisaka, T.;
Kawamori, R.; Nakamura, M.; Kurahashi, Y.; Yoshimoto, T.; and
Yamamoto, S. 1998 Decreased activity of arachidonate 12-lipoxygenase
in platelets of Japanese patients with non-insulin-dependent diabetes
mellitus. Metabolism, 47: 257-63.
Tokuda, H.; Hatakeyama, D.; Akamatsu, S.; Tanabe, K.; Yoshida, M.; Shibata,
T.; and Kozawa, O. 2003 Involvement of MAP kinases in TGF-beta-
202 stimulated vascular endothelial growth factor synthesis in osteoblasts.
Arch. Biochem. Biophys., 415: 117-25.
Tomchick, D. R.; Phan, P.; Cymborowski, M.; Minor, W.; and Holman, T. R.
2001 Structural and functional characterization of second-coordination
sphere mutants of soybean lipoxygenase-1. Biochemistry, 40: 7509-17.
Tonneson, H. H. K.; and Mostad, J. 1982 The crystal structure of curcumin.
Acta Chem. Scand. B, 36: 475-480.
Toth, G. R.; Weckerle, M.; and Schreier, B. 2000 Structural elucidation of two
novel products from the soybean lipoxygenase-catalyzed dioxygenation of
curcumin. Mang. Reson. Chem., 38: 51-54.
Turner, T.; Chen, P.; Goodly, L. J.; and Wells, A. 1996 EGF receptor signaling
enhances in vivo invasiveness of DU-145 human prostate carcinoma cells.
Clin. Exp. Metastasis, 14: 409-18.
Vaisman, N.; Gospodarowicz, D.; and Neufeld, G. 1990 Characterization of the
receptors for vascular endothelial growth factor. J. Biol. Chem., 265:
19461-6.
van Leyen, K.; Duvoisin, R. M.; Engelhardt, H.; and Wiedmann, M. 1998 A
function for lipoxygenase in programmed organelle degradation. Nature,
395: 392-5.
Vanderhoek, J. Y.; Bryant, R. W.; and Bailey, J. M. 1982 Regulation of
leukocyte and platelet lipoxygenases by hydroxyeicosanoids. Biochem.
Pharmacol., 31: 3463-7.
203 Viola, M. V.; Fromowitz, F.; Oravez, S.; Deb, S.; Finkel, G.; Lundy, J.; Hand, P.;
Thor, A.; and Schlom, J. 1986 Expression of ras oncogene p21 in
prostate cancer. N. Engl. J. Med., 314: 133-7.
Voeller, H. J.; Wilding, G.; and Gelmann, E. P. 1991 v-rasH expression confers
hormone-independent in vitro growth to LNCaP prostate carcinoma cells.
Mol. Endocrinol., 5: 209-16.
Wada, A.; Ogo, S.; Watanabe, Y.; Mukai, M.; Kitagawa, T.; Jitsukawa, K.;
Masuda, H.; and Einaga, H. 1999 Synthesis and Characterization of
Novel Alkylperoxo Mononuclear Iron(III) Complexes with a Tripodal
Pyridylamine Ligand: A Model for Peroxo Intermediates in Reactions
Catalyzed by Non-Heme Iron Enzymes. Inorg. Chem., 38: 3592-3593.
Wadman, I. A.; Hsu, H. L.; Cobb, M. H.; and Baer, R. 1994 The MAP kinase
phosphorylation site of TAL1 occurs within a transcriptional activation
domain. Oncogene, 9: 3713-6.
Walstra, P.; Verhagen, J.; Vermeer, M. A.; Veldink, G. A.; and Vliegenthart, J. F.
1987 Demonstration of a 12-lipoxygenase activity in bovine
polymorphonuclear leukocytes. Biochim. Biophys. Acta, 921: 312-9.
Watanabe, T.; Medina, J. F.; Haeggstrom, J. Z.; Radmark, O.; and Samuelsson,
B. 1993 Molecular cloning of a 12-lipoxygenase cDNA from rat brain.
Eur. J. Biochem., 212: 605-12.
Weidner, N.; Carroll, P. R.; Flax, J.; Blumenfeld, W.; and Folkman, J. 1993
Tumor angiogenesis correlates with metastasis in invasive prostate
carcinoma. Am. J. Pathol., 143: 401-9.
204 Weitzel, F.; and Wendel, A. 1993 Selenoenzymes regulate the activity of
leukocyte 5-lipoxygenase via the peroxide tone. J. Biol. Chem., 268: 6288-
92.
Wen, Y.; Gu, J.; Knaus, U. G.; Thomas, L.; Gonzales, N.; and Nadler, J. L. 2000
Evidence that 12-lipoxygenase product 12-hydroxyeicosatetraenoic acid
activates p21-activated kinase. Biochem. J., 349: 481-7.
Wen, Y.; Nadler, J. L.; Gonzales, N.; Scott, S.; Clauser, E.; and Natarajan, R.
1996 Mechanisms of ANG II-induced mitogenic responses: role of 12-
lipoxygenase and biphasic MAP kinase. Am. J. Physiol., 271: C1212-20.
Whitehouse, C. M.; Dreyer, R. N.; Yamashita, M.; and Fenn, J. B. 1985
Electrospray interface for liquid chromatographs and mass spectrometers.
Anal. Chem., 57: 675-9.
Whitman, S.; Gezginci, M.; Timmermann, B. N.; and Holman, T. R. 2002
Structure-activity relationship studies of nordihydroguaiaretic acid
inhibitors toward soybean, 12-human, and 15-human lipoxygenase. J.
Med. Chem., 45: 2659-61.
Wong, B. C.; Wang, W. P.; Cho, C. H.; Fan, X. M.; Lin, M. C.; Kung, H. F.; and
Lam, S. K. 2001 12-Lipoxygenase inhibition induced apoptosis in human
gastric cancer cells. Carcinogenesis, 22: 1349-54.
Xu, L.; Fukumura, D.; and Jain, R. K. 2002 Acidic extracellular pH induces
vascular endothelial growth factor (VEGF) in human glioblastoma cells via
ERK1/2 MAPK signaling pathway: mechanism of low pH-induced VEGF.
J. Biol. Chem., 277: 11368-74.
205 Xu, Y. X.; Pindolia, K. R.; Janakiraman, N.; Chapman, R. A.; and Gautam, S. C.
1997 Curcumin inhibits IL1 alpha and TNF-alpha induction of AP-1 and
NF-kB DNA-binding activity in bone marrow stromal cells. Hematopathol.
Mol. Hematol., 11: 49-62.
Yamaguchi, K.; Shirakabe, K.; Shibuya, H.; Irie, K.; Oishi, I.; Ueno, N.; Taniguchi,
T.; Nishida, E.; and Matsumoto, K. 1995 Identification of a member of the
MAPKKK family as a potential mediator of TGF-beta signal transduction.
Science, 270: 2008-11.
Yamamoto, S.; Suzuki, H.; and Ueda, N. 1997 Arachidonate 12-lipoxygenases.
Prog. Lipid Res., 36: 23-41.
Yla-Herttuala, S.; Rosenfeld, M. E.; Parthasarathy, S.; Glass, C. K.; Sigal, E.;
Witztum, J. L.; and Steinberg, D. 1990 Colocalization of 15-lipoxygenase
mRNA and protein with epitopes of oxidized low density lipoprotein in
macrophage-rich areas of atherosclerotic lesions. Proc. Natl. Acad. Sci. U.
S. A., 87: 6959-63.
Yoshimoto, T.; Arakawa, T.; Hada, T.; Yamamoto, S.; and Takahashi, E. 1992
Structure and chromosomal localization of human arachidonate 12-
lipoxygenase gene. J. Biol. Chem., 267: 24805-9.
Yoshimoto, T.; Miyamoto, Y.; Ochi, K.; and Yamamoto, S. 1982 Arachidonate
12-lipoxygenase of porcine leukocyte with activity for 5-
hydroxyeicosatetraenoic acid. Biochim. Biophys. Acta ,713: 638-46.
Yoshimoto, T.; Suzuki, H.; Yamamoto, S.; Takai, T.; Yokoyama, C.; and Tanabe,
T. 1990a Cloning and sequence analysis of the cDNA for arachidonate
206 12-lipoxygenase of porcine leukocytes. Proc. Natl. Acad. Sci. U. S .A., 87:
2142-6.
Yoshimoto, T.; and Takahashi, Y. 2002 Arachidonate 12-lipoxygenases.
Prostaglandins Other Lipid Mediat., 68-69: 245-62.
Yoshimoto, T.; Yamamoto, Y.; Arakawa, T.; Suzuki, H.; Yamamoto, S.;
Yokoyama, C.; Tanabe, T.; and Toh, H. 1990b Molecular cloning and
expression of human arachidonate 12-lipoxygenase. Biochem. Biophys.
Res. Commun., 172: 1230-5.
Yoshimura, R.; Inoue, K.; Kawahito, Y.; Mitsuhashi, M.; Tsuchida, K.;
Matsuyama, M.; Sano, H.; and Nakatani, T. 2004 Expression of 12-
lipoxygenase in human renal cell carcinoma and growth prevention by its
inhibitor. Int. J. Mol. Med., 13: 41-6.
Yoshimura, R.; Matsuyama, M.; Tsuchida, K.; Kawahito, Y.; Sano, H.; and
Nakatani, T. 2003 Expression of lipoxygenase in human bladder
carcinoma and growth inhibition by its inhibitors. J. Urol., 170: 1994-9.
Zhou, J.; Fariss, R. N.; and Zelenka, P. S. 2003 Synergy of epidermal growth
factor and 12(S)-hydroxyeicosatetraenoate on protein kinase C activation
in lens epithelial cells. J. Biol. Chem., 278: 5388-98.
Zhou, L.; Tan, A.; Iasvovskaia, S.; Li, J.; Lin, A.; and Hershenson, M. B. 2003
Ras and mitogen-activated protein kinase kinase kinase-1 coregulate
activator protein-1- and nuclear factor-kappaB-mediated gene expression
in airway epithelial cells. Am. J. Respir. Cell. Mol. Biol., 28: 762-9.
207 ABSTRACT
Arachidonic acid metabolism leads to the production of biolipid mediatiors
with important roles in multiple pathological processes including prostate cancer.
Expression of platelet-type 12-lipoxygenase (P12-LOX) correlates with clinical
stage and grade of prostate adenocarcinoma and overexpression in prostate
cancer cells leads to large, highly vascularized tumors in a mouse model. The
product of arachidonic acid metabolism via P12-LOX is 12(S)-
hydroxyeicosatetraenoic acid (12(S)-HETE), a lipid previously linked to angiogenesis. Although P12-LOX and its product, 12(S)-HETE, have been shown to play a role in prostate cancer biology, an underlying mechanism has yet to be characterized. Thus, PC-3 prostate cancer cells overexpressing P12-
LOX were employed to define the function of 12(S)-HETE as an angiogenic factor. Overexpression of P12-LOX in prostate cancer cells resulted in increased
12(S)-HETE production and enhanced accumulation of the potent angiogenic protein vascular endothelial growth factor (VEGF). Elevated basal phosphorylation of the extracellular signal-regulated kinase 1/2 (ERK1/2) mitogen activated protein kinase (MAPK) was found in cells overexpressing P12-LOX.
Incubation of PC-3 and DU145 prostate cancer cells with 12(S)-HETE resulted in a significant increase in phosphorylated ERK1/2, whereas preincubation with pharmacological inhibitor/s of MEK and pertussis toxin (PTx) sensitive G-protein coupled receptors (GPCR) blocked 12(S)-HETE induced ERK1/2 phosphorylation. Additionally, inhibition of both MEK and PTx sensitive GPCRs reduced VEGF accumulation in culture media of cells overexpressing P12-LOX.
208 These results indicate that P12-LOX overexpression by prostate cancer cells, a condition mirrored in prostate cancer patients, may have the untoward effect of increasing VEGF production and that this effect is due to the stimulation of
ERK1/2 by 12(S)-HETE in a PTx sensitive GPCR mediated manner. Inhibition of P12-LOX in the treatment of prostate cancer is currently not a viable therapeutic strategy due to a lack of lipoxygenase inhibitor specificity. Using X- ray crystallographic analysis, the 3D interaction of curcumin with soybean lipoxygenase-3 was examined. These pilot studies may help in the development of novel, highly specific inhibitors of P12-LOX. Taken together, these data show that P12-LOX plays a significant role in prostate cancer induced angiogenesis by promoting VEGF production and that X-ray crystallography may prove beneficial in the development of specific inhibitors of P12-LOX.
209