摘要 Abstract

我們先前曾顯示，大麻素和他的新陳代謝地穩(wěn)定類型，甲酰胺基，通過一氧化氮產(chǎn)生血管舒張，對聯(lián)接的升糖指數(shù)、內(nèi)皮的非-CB1/CB2公認(rèn)的“大麻素接收器”有著顯著作用(Mukhopadhyay et al .,2002)。最近,我們已經(jīng)表明,大麻素通過胃腸道蛋白質(zhì)和PI3K-Akt通路的激活能夠激活內(nèi)皮一氧化氮合酶(以挪士)。
 

額外的調(diào)查中我們發(fā)現(xiàn),大麻素通過一種蛋白激酶通路激活內(nèi)皮細(xì)胞產(chǎn)生血管生成反應(yīng)，這些內(nèi)皮細(xì)胞無法受到大麻素受體激動(dòng)劑WIN55212或CP55940的模仿。CB1受體拮抗劑SR141716A或LY320135未能阻止大麻素的調(diào)整，血管生成建表明大麻素受體參與這個(gè)反應(yīng)。在本研究中,我們發(fā)現(xiàn)甲酰胺基造成MMP-2的增加，CB1受體中的MMP-9活動(dòng)阻止LNCaP前列腺癌細(xì)胞。
 

在“大麻素受體”中，MMP-2和MMP-9活動(dòng)下調(diào)了，阻礙了LNCaPWIN55212和甲酰胺基細(xì)胞的存在。這些數(shù)據(jù)顯示,細(xì)胞遷移下的基質(zhì)金屬蛋白酶活性的分子機(jī)制是由大麻素、CB1受體和大麻素受體活性調(diào)整的。數(shù)據(jù)描述了大麻素調(diào)整的作用，沒有生產(chǎn)前列腺癌細(xì)胞MMP-2，MMP-9也未有定論。
 

We have previously shown that anandamide and its metabolically stable analog, methanandamide, produces vasorelaxation via the production of nitric oxide in rabbit aortic ring preparations, acting on a Gi-coupled, endothelial non-CB1/CB2 putative “anandamide receptor” (Mukhopadhyay et al., 2002). Recently we have demonstrated that anandamide activates endothelial nitric oxide synthase (eNOS) via the activation of Gi protein and PI3K-Akt pathway.
 

In the additional investigation we have found that anandamide produced angiogenic responses via the activation of Akt pathway in endothelial cells which could not be mimicked by the cannabinoid receptor agonists WIN55212 or CP 55940. CB1 receptor antagonist SR141716A or LY320135 failed to block anandamide –mediated angiogenesis suggesting the involvement of anandamide receptor in this response. Within this study we found that methanandamide produced an increase in MMP-2 and MMP-9 activity in CB1 receptor blocked LNCaP prostate cancer cells.
 

In addition MMP-2 and MMP-9 activity was down-regulated in “anandamide receptor” blocked LNCaP cells in the presence of both WIN55212 and methanandamide. These data sets suggest that the molecular mechanism of matrix metalloproteinase activity in cell migration is regulated by the anandamide-mediated activity of the CB1 receptor and the anandamide receptor. Data characterizing the role anandamide-mediated NO production in prostate cancer cells MMP-2 and MMP-9 remains inconclusive.
 

CHAPTER I

INTRODUCTION

Marijuana’s psychoactive ingredient is known as -9tetrahydrocannabinol (THC). Studies of THC have uncovered a family of rhodopsin-like G-protein coupled receptors as the target of this compound. Researched conducted in the rat brain and the spleen lead to the identification and cloning of CB1R and CB2R (Matsuda et al., 1990; Gerard et al., 1991). The identification of the cannabinoid receptors fueled the hypothesis that the endogenous ligands to these receptors would bear structural similarities to the THC (Mechoulam et al., 1995). The isolation of the first member this family, now called endogenous cannabinoids or endocannabinoids, arachidonoylethanolamide (anandamide) supported this hypothesis (Devane et al., 1992). Studies into the pharmalogical profile of anandamide suggested an additional an additional binding site for this compound.#p#分頁標(biāo)題#e#
 

The investigations into the differential actions of anandamide in endothelial cells, within our research group, suggested the presence of another G protein-coupled receptor, the putative an anandamide receptor (Mukhopadhyay et al., 2002; McCollum et al., 2007). These data indicated an anandamide receptor mediated vasorelaxation response in rabbit aortic endothelial cells (RAEC) and human umbilical vein endothelial cells (HUVEC) (Mukhopadhyay et al., 2002; McCollum et al., 2007). The findings in our laboratory were supported by similar observations in various another microvascular structures (Kunos et al., 2000a, Kunos et al., 2000b, Kunos 2002). Further experimentation within our laboratory revealed the attenuation of proangiogenic response with si-CB1HUVEC cells and CB1R knockout mice with the administration of O-1918. These data series lend support to the role of anandamide receptor in the initiation of angiogenesis.
 

Angiogenesis is the process of constructing new blood vessels. In order for angiogenesis to occur the extracellular matrix (ECM) must be removed and remodeled. This portion of angiogenesis possesses a similarity to the process of metastasis of cancer cells. A family of enzymes that are documented as play a crucial role in both of these processes is the matrix metalloproteinase (MMP) family (Rundhuang 2005; Deryuqina, Quigley, 2006). MMPs are enzymes that breakdown several substrates in the ECM. Two enzymes in this family have been correlated with angiogenesis and may have involvement in cancer cell metastasis MMP-2 (gelatinase A) and MMP-9 (gelatinase B). The active forms of these two enzymes digest type IV collagen, which is a unique component in the basement membrane (BM) [4].
 

The structure of MMP enzymes is what indicates its substrate thus MMP-2 and 9 have the same basic structure. Both enzymes have a pre domain this is removed after it directs the enzyme through endoplasmic reticulum [4]. The pre domain is followed by the pro domain this region stays attached to the enzyme until it is activated. The next portion is a catalytic domain, which contains a zinc-binding site; this is also the region that binds to the specific cleavage site on the substrate [4]. The last region is a hemopexin domain, which is, connected to the catalytic domain by a hinge [4]. As explained MMP-2, 9 are especially structured to assist in the removal of collagen. In order to form a new microvascular structure or to establish a metastatic node within secondary tissuses the ECM must be degraded indicating that MMP-2 and MMP-9 may involved in tumorgensis as well as angioenesis.
 

A tumorgenic cancer cells exhibits a large amount of MMP activity in order for the newly generated cells cluster the cells must break down the ECM. LNCaP cells posses both known cannabinoid receptors therefore they are affected by endocannabinoids. Yet there is a discrepancy in how these drugs affect the metastasis of the LNCaP cells (Sarfaraz er al 2006; Sarfaraz et al., 2005; Mimeault et al., 2003). It is theorized that the endocannabinoids down-regulate the production of MMPs and up-regulation the production of tissue inhibitors of metalloproteinase (TIMP) [5]. If correct this hypothesis would imply that endocannabinoids impede the tumorgenic activity of the LNCaP cells. Still other theories suggest that methanandamide and WIN55212 up-regulate the production of MMP-2 and MMP-9 and therefore promote metastasis in LNCaP cells. The inclusiveness of these findings fostered the hypothesis of this work, anandamide stimulates a non-CB1/non-CB2, “anandamide receptor” activation of matrix metalloproteinase 2 (MMP-2) and MMP-9 in the LNCaP human prostate cancer cell line.#p#分頁標(biāo)題#e#
 

In order to adequately address the hypothesis concerning the role of the anandamide receptor three specific aims were identified. The first challenge was to determine the role of the CB1 receptor in relation to the anandamide receptor in MMP-2 and MMP-9 activity within LNCaP cell. Next the elucidation the role of the CB2 receptor was identified to further validate the stimulatory role of the anandamide receptor. The final aim was to determine the role of nitric oxide in the regulation of MMP activity to indicate the signal transduction pathway of these receptors.
 

CHAPTER II

LITERATURE REVIEW

Endogenous Cannabinoids

Investigation into the presence of endogenous cannabinoids began with the discovery of the mechanism of action attributed to the psychoactive component in marijuana -9 tertahydrocannabinol (-9 THC). In 1984, Howlett and Fleming reported that levonantradol (a structural analog of THC) inhibited the production of adenylyl cyclase activity in neuroblastoma cells (Howlett A et al., 1984). Following this discovery, the same group later identified a specific high affinity binding site for another analog [3H] CP55940 in areas of the CNS which were attributed with mediating the psychotropic properties of cannabinoids (Devane et al., 1988) The identification and localization of this binding sites lead to the cloning of the first cannabinoid receptor (CB1) (Matsuda et al., 1990) . The second cannabinoid receptor (CB2) was cloned several years later in spleen and considered an immunological cannabinoid receptor
 

The classification and cloning of CB1 and CB2 fueled the search for the class of compounds that acted as the endogenous ligands for these receptors. Mechoulam and colleagues hypothesized that the endogenous compounds would be lipid soluble and share structural similarities to the Cannabis plant extracts that had been previously purified (Mechoulam et al., 1995). This hypothesis was supported in 1992 when Devane extracted organic solvents from the porcine brain and later from the canine guts (Devane et al., 1992a). The extracts were chromatographed to separate the lipids and two major lipid endogenous cannabinoids (endocannabinoids) were isolated. The first compound characterized was N-archidonylethanolamide (Anandamide) (Devane et al., 1992a). A second major component was characterized from the porcine extract, 2-arachidonoyl glycerol (2-AG) (Suguira et al., 1995). Later three more endocannabinoid candidates were identified arachidonyol ether (noladin), O-arachidonoyl-ethanolamide (virhodamine) and N-arachidonoyl-dopamine (NADA) (Hanus et al., 2001; Bisogno et al., 2000; Porter et al., 2002). The structural properties of each the endogenous cannabinoids are depicted in figure 1 (Bisogno et al., 2005).
 

Synthesis of Endogenous Cannabinoids

Prior to the identification of the endocannabinoids Schmid and colleagues delineated and provided supporting experimentation for a common synthesis pathway of the N-acyl ethanolamine (NAE) family (Di Marzo et al., 1994). According to the schematic (Fig.2) generated by this group a minor phospholipid group N-acyl phosphatidyletahanolamines (NAPEs) are the immediate precursors to the formation of NAEs. NAPEs are thought to be generated by N-acyl transferase which catalyzes the exchange of the acyl group in sn-1 position from donor phospholipid to primary amine group of phosphatidylethanolamine (PE). The hypothesis indicates that NAPEs give rise to the NAEs through the catalytic action of a phosphodiesterase, phosopholipase D (PLD). The support for this schematic in the specific production of Anandamide has been difficult to generate. Experimental data has identified two pathways of production for anandamide, one including the production NAPEs and another independent of this derivative.#p#分頁標(biāo)題#e#
 

Experimental data from several scientific groups suggest that Anandamide is generated from a phosopholipid-derived precursor NAPEs. This chemical process is catalyzed by an incompletely characterized NAPE-PLD. NAPE-PLD facilitates the hydrolysis of the bond between phosophate and ethanolamide producing Anandamide and a byproduct, phosphatidic acid (Fig 2). This reaction was supported by data in neuron primary culture (Di Marzo et al., 1994), rat brain microsomes (Sugiura et al., 1996; Saski et al., 1997) and rat testes (Sugiura et al., 1996; Kondo et al., 1998). These cell groups were observed as they converted radiolabeled N-arachidonyl PE to radiolabeled Anandamide. Schmid and colleagues detected NAPE-PLD activity in microsomal and mitochondrial fraction of the rat heart (Schmid et al., 1983), and later reported additional activity synaptosomes and myelin in the dog brain. Studies on the enzyme have identified calcium or magnesium independent activity in in-vitro assays where the enzyme was stimulated by Triton X-100 ( Natarajan et al., 1984), and calcium dependent activation in J774 cells (Di Marzo et al., 1996).
 

NAPE-PLD biochemical characteristics have not been completely identified. Preliminary studies have identified NAPE-PLD as a novel phospholipase because this enzyme does not catalyze transphosphatidylation and it is not activated by phosphatidylinositol bisphosphate (PIP2), which are two characteristics commonly attributed to phospholipase (Petersen et al., 1999). Saski group reports that the activity of NAPE-PLD is enhanced by oleate (Saski et al., 1997), while evidence from other laboratories indicated that oleate inhibits NAPE-PLD within the rat heart (Petersen et al., 1999). These findings are preliminary and leave several steps in Schmid schematic undefined as it relates to Anandamide. In addition, findings from other groups indicate that Anandamide can be generated through a NAPE-independent pathway.
 

Evidence produced by several groups supports the concept that Anandamide and other NAEs were synthesized through enzymatic condensation of ethanolamine and free-fatty acid (Colodzin et al., 1963; Bachur and Udenfriend, 1965; Deustch and Chin, 1993; Kruszka and Gross, 1994; Devane and Axelrod, 1994). These experiments to substantiate this hypothesis were performed using the homogenates and membrane preparations from the liver, brain, and kidney. The synthase activity describe in cells lines has been correlated to the activity of fatty acid amide hydrolyase (FAAH). Further experimentation by separate investigators uncovered the commintant expression of the action contributed to FAAH, and the synthase activities observed in the NAPE-independent synthesis of Anandamide (Arreaza et al., 1997; Kurahashi et al., 1997).
 

The abundance of 2-AG in tissues and cells indicates to there is a probable de novo production of this compound when it is employed in the endocannabinoid system. Just as in the pathways that characterize the production anandamide, 2-AG synthesis is most likely calcium dependent. With these stipulations in mind two production pathways have been identified for 2-arachidonylglycerol (2-AG) through the use of hippocamal neurons (Stella et al., 1997) and N18TG2 neuroblastoma cells (Bisogno et al., 1997).#p#分頁標(biāo)題#e#
 

Studies of the production pathway in hippocamal neurons revealed that under conditions of high frequency stimulation 2-AG production is tetrodotoxin sensitive (Stella et al., 1997). The data also suggested stimulation of the 2-AG synthesis in hippocamal neurons involved the sequential activation of phosopholipase C and diacylglycerol (DAG) lipase (Stella et al., 1997). Further investigations indicated that 2-AG is synthesized from 2-arachidonate-containing diacylglycerols (DAGs) precursor. This reaction is catalyzed by a sn-1-selective DAG lipase (sn-1-DAGL) (Berdyshev et al., 2001). The production of the precursor within this cell type is hypothesized to be meditated through phosphoinositides (PI) and catalyzed by PI-selective phosopholipase C. Stimulation of neuroblastoma cells with high levels of ionomycin revealed a calcium-dependent production of 2-AG. This biosynthetic pathway did support the involvement of DAG as an intermediate in the synthesis of 2-AG but indication no involvement of PLC. Later groups theorized that in this pathway the DAG precursors are produced through the hydrolysis of 2-arachidonate-containing phosphatidic acid (PA) phosphohydrolyase (Berdyshev et al., 2001). These data sets provide a preliminary evidence of the possible synthesis pathway for 2-AG but further experimentation must be preformed using compounds that correlate to physiological stimulation in order to confirm these findings.
 

Cellular Intake and Degradation

The elucidation of the mechanism for the synthesis of endocannabinoids leads to questions concerning the reuptake of these ligands and their inactivation pathway. Preliminary data indicated that the reuptake of endocannabinoids is mediated through a specific protein carrier. The bulk of the data suggested that this carrier protein began the process of inactivation. The key enzyme of this process was identified as a unique hydrolyase Fatty acid amide hydrolyase (FAAH).
 

Several studies support the hypothesis that endocannabinoids are transported for inactivation through a specific protein carrier. Data indicating the presence of this carrier was collected in neurons (Di Marzo et al.1994; Hillard et al., 1997); mouse J774 macrophages and RBL-2H3 cells (Bisogno et al., 1997); and in human neuroblastoma (CHP100) and lymphoma cells (U937) (Maccarrone et al.,1998). These studies demonstrate that the transport of Anandamide is dependent on several key characteristics of a protein carrier including temperature dependence, saturability, and substrate selectivity (Di Marzo et al.1994; Hillard et al., 1997; Piomelli et al, 1999). AM404, an uptake carrier inhibitor, was also employed as an experimental. This compound inhibitor provided supportive data of a role of the Anandamide carrier protein in the inactivation of exogenous Anandamide. The regulation of this carrier has not been completely uncovered but work performed by Maccarrone and colleagues indicate that its activity is enhanced by the presence of nitric oxide donors (Maccarrone et al.,1998) . An additional hypothesis has been proposed for the putative Anandamide carrier protein, that it also plays a role in the release of Anandamide following production. Within cerebellar cells a temperature-sensitive accumulation of radiolabeled-anandamide was observed (Hillard et al., 1997).#p#分頁標(biāo)題#e#
 

2-AG uptake was been reported both a competitive and non-competitive process. In-vitro studies conducted with astroctyoma cells indicate that 2-AG does compete with Anandamide for uptake with and IC50 of 18.5um (Piomelli et al, 1999). Contrary research performed in macrophages (Di Marzo et al., 1999) and RBL-2H3 (Di Marzo et al., 1998) indicated a noncompetitive uptake of these compounds. Additional experimentation with macrophages suggest that uptake in these cells may not be carrier-mediated. The resolutions of these findings and the answers to the questions concerning and Anandamide uptake may lay in the molecular characterization and cloning of this transporter protein.
 

Anandamide is hydrolyzed by membrane associated enzyme, known as Fatty acid amide hydrolyase (FAAH), into two inactive substrates free fatty acid and ethanolamine. FAAH and a second hydrolyase identified in a macrophage cell line (Di Marzo et al., 1999) and in the brain are responsible for the breakdown of 2-AG to arachdonic acid and glycerol. FAAH has also been attributed with the breakdown arachidonyl methyl ester, which indicates that this enzyme may hydrolyze virodhamine. FAAH was first identified within the brain (Natarajan et al., 1984) and the liver (Schmid et al., 1985) through the work of Schmid and colleagues. Further research uncovered FAAH enzymatic activity in human tissues and cells. The highest specific activity of this enzyme occurs in brain microsomal/myelin preparations (Hillard et al., 1995; Desarnaud et al., 1995), liver microsomes and liver mitochondria.
 

FAAH has been sequenced and cloned from samples purified from rat liver microsomes (Patricelli et al., 1998). The sequencing of this enzyme revealed an active transmembrane portion near the N-terminus. The identification of this sequence supported the findings of activity in membrane preparations but not within samples of the cytosol. Experimentation with COS cell mutants that lack this N-terminus domain demonstrated a continued prevalence of membrane associate and activation, which suggested a different functionality for this region (Omeir et al., 1998). Further investigation with COS cells that over express the wild-type of FAAH enzyme suggest that the function of the N-terminus domain is self-association. The oligmer formation of this enzyme has not been substantiated in cells expressing normal levels of protein. In addition to the transmembrane domain, a proline-enriched segment was identified, which is proposed to bind the src homology 3 (SH-3). Studies propose that SH-3 binding regulates the activity and membrane association of FAAH (Egertova et al., 1998).
 

FAAH action of has been colocalized with the expression the cannabinoid CB1 receptors in rat brain (Egertova et al., 1998; Tsou et al., 1998). Support of this interaction is exhibited by the immunoreactivity of Purkinje cells and high expression of the CB1 in granule cells axons, which impinge on Purkinje cells (Tsou et al., 1998). The pattern of localization regarding catabolic enzyme identifies cells producing FAAH as possible production cells for endocannabinoids. These hypotheses are the framework for the continuing investigation of the endocannabinoid system.#p#分頁標(biāo)題#e#
 

Nonclassical Cannabinoids

Investigations into the analgesic properties of THC lead to development of a series of AC-bicyclid and ACD-tricyclic cannabinoid analogs (Melvin et al., 1984; Melvin et al., 1993). Further experimentation with the structure of this series lead to the development of one of the most commonly used synthetic cannabinoids, CP55940 (Melvin et al., 1993) CP55940 is a bicyclic THC analog (fig.3) which is less lipophilic than the endogenous and exogenous cannabinoids. This property and the radiolabeled, [H3] CP55940, version identified to compound as the best research tool for locating and characterizing CB1 receptor (Devane et al., 1988). In vivo studies and in vitro studies CP55940 has been established as a full agonist of CB1R and CB2R and display maximal effects far exceeding those of other cannabinoid receptor agonist (Johnson and Melvin 1986). In additional this compound exhibited an activity rate 10 to 50 times higher than that in THC in the mouse tetrad model (Little et al., 1988). The potency, affinity, and its radiolabeled version have secured this compound as a useful investigative tool in areas of cannabinoid response and endocannabinoid system research.
 

Aminoalkylindoles

Cannabimimetics such as CP55940 were designed according to the structure THC but R-(+)-WIN55212 represents a departure from this type of experimental tool. R-(+)-WIN55212 is an aminoalkylindoles (fig. 3) which posses reduce anti-inflammatory. This aminoalkyindoles has a high affinity for cannabinoid receptors but is moderately selective toward the CB2 receptor (D’Ambra et al., 1992; Eissenstat et al., 1995). WIN55212 has the ability to elicit high levels of the intrinsic activity attributed to both cannabinoid receptors. Data from several groups suggest that the binding of this compound differs from that of classical and non-classical cannabinoids but can be displaced by either. The difference in reaction of this analog when judged against its counterparts may provide a manner in which to observe the importance of each bond within the binding sites of the cannabinoid receptors.
 

Eicosanoids

Anandamide is an endogenous cannabinoid and is readily degraded by FAAH, as described previously. This instability makes anandamide an unsuitable research tool therefore a more stable analog was synthesized methanandamide depicted in figure 3. Methanandamide possesses an additional methyl group on the 1’carbon which reduces its susceptibility to FAAH, this structure ensures that it has considerable bioavailability in in vivo studies. In addition to reducing the degradation, the methyl group confers on methanandamide a higher binding affinity to the CB1 receptor (Khanolkar et al., 1996; Lin et al., 1998).
 

Cannabinoid Receptor Antagonists

Diarylpyrazoles is category of compounds that incorporate two cannabinoid receptor inhibitors, SR141716A and SR144528 (fig. 4). SR141716A or rimonabant is a potent inhibitor of CB1 receptor activation. This compound is attributed with antagonist and inverse agonist characteristics. The antagonist action of rimonabant was confirmed thru data of its inhibition of CB1 receptor signal transduction including the attenuation of reduction of cAMP production and adenylyl cyclase activation (Rinaldi-Carmaona et al., 1994; Mato et al., 2002). The inverse agonist action of rimonabant is attributed to the interaction of this compound with an allosteric binding site during the constitutive activation of the CB1 receptor in the absence of exogenous and endogenous cannabinoids (Gessa et al., 1997, 1998a; Schlicker et al., 1997; Acquas et al., 2000; Sim-Selley et al., 2001). rather to the competitive binding site to which it has a high affinity.#p#分頁標(biāo)題#e#
 

SR144528 is identified as an antagonist an inverse agonist of the CB2 receptor. The competitive binding this compound to the CB2 receptor produces converse reactions to that of agonist (Pertwee, 1999). Similar to rimonabant, SR144528 elicits inverse agoinist properties through an allosteric site which it also has a low affinity. SR144528 provides a pathway to understand the mechanism of CB2 receptor and how it differs from that of CB1 receptor
 

O-1918 or abnormal cannabidiol (fig. 4) is an inhibitor of a yet to be categorized non-CB1/non-CB2 receptor, the putative anandamide receptor. The inhibitory characteristics of O-1918 were observed its ability to attenuate non-CB1/non-CB2- mediated nitric oxide production and vasorelaxation in rabbit aortic endothelial cells (RAEC) (Mukhopadhyay 2002 et al.) Preferential binding of this compound to either of the known cannabinoid receptors was not observed at physiological levels. Low affinity to the cannabinoid receptors was observed at 30uM concentration (Offertaler et al., 2003).
 

Cannabinoid Receptors

The mysteries of how -9- tertahydrocannabinol (-9-THC) elicited its effects stimulated the search for a specific group of proteins that recognized this ligand. In the beginning of this endeavor, several scientific groups hypothesized that the effects of this and other cannabinoid drugs were attributed to their lipophilic properties and therefore the compounds’ ability to diffuse across the plasma membrane [1, 2]. Later observations indicated that the effects of cannabinoid drugs on ATPase and monoamine oxidase activities, hormone and neurotransmitter binding, and synaptosomal uptake of neurotransmitters were due to their ability to intercalate into plasma membranes and increase the fluidity of the membrane [3, 4]. Drs. Howlett and Fleming determined that a sub-micromolar concentration of cannabinoid drugs inhibited adenylyl cycylase activity and the accumulation of cyclic AMP in neuronal cells, which were stimulations attributed to receptor function [5, 6, 7].
 

These data sets suggested that the cannabinoid drugs induced their responses through a specific receptor and its signal transduction pathway. Measurements of Mg2+ ion and GTP concentrations during the inhibition of adenylyl cyclase activity alluded to the involvement of a member of the G-protein family as a mediator of secondary messenger activation [7]. The identity of the G-protein as a Gi/o subtype was supported by the attenuation of the cannabinoid-mediated inhibitory response following the administration of pertussin toxin, a specific Gi/o protein inhibitor [8]. These studies and additional pharmacological data lead to the identification and cloning of the CB1and CB2 receptors. The seven transmembrane helical structure and conserved portions of the amino acid sequence lead to the categorization of these receptors as members of the rhodopsin-like family [4]. Still further research was required to uncover the signal transduction pathway activated by the binding of anandamide.#p#分頁標(biāo)題#e#
 

The activation of CB1R initiated by the binding anandamide stimulates the exchange of GDP (guanine diphosphate) nucleotide on heterotrimeric G-protein for GTP (guanine triphosphate) at the nucleotide binding site. This action activates the alpha and gamma subunits of the heterotrimeric Gi/o-protein [81, 82]. This hypothesis was supported through the measurement of [35S]GTPγS binding in rat cerebellar membranes pretreated with phenylmethylsulfonyl fluoride (PMSF), an inhibitor of FAAH, followed by the administration of anandamide or (R)-methanandamide, a metabolically stable analog of anandamide [12,13]. The efficacy percentage reported from these groups indicated that maximal stimulation of anandamide was 70-80% of WIN55212-2, a full CB1R/CB2R synthetic cannabinoid agonist. The lack of reproducibility of this result in the laboratory of Griffin and colleagues was attributed to PMSF sequestering anandamide and reducing its availability in the system. The use of fluoromethanandamide, a non-reactive analog, furnished an efficacy percentage of 50% of the maximal activity observed for WIN55212-2 [14]. This information was used to substantiate the categorization of anandamide as a partial agonist of the CB1 receptor.
 

Observations of CHO cells that express the CB2 receptor reveal that the signal transduction pathway of this receptor is mediated by a Gi/o-protein. Data evaluating the efficacy of anandamide to stimulation GTP binding on this receptor is contradictory. Gonsiorek and colleagues reported that anandamide acts as a partial agonist stimulating 35% of maximal effect of HU210, a full agonist of the CB2 receptor [15]. While data from another group indicated that [35S] GTPγS binding was poor in the presence of anandamide [16]. This group was unable to elicit half maximal effects with an anandamide concentration of 10uM. These data sets make it difficult to determine the efficacy or potency of anandamide.
 

Additional investigations into anandamide-mediated activation of the Gi/o-protein coupled to the CB1 receptor revealed that this activation inhibits the adenylyl cyclase activity and cAMP production. The inhibition of cAMP production was observed in intact N18TG2 neuroblastoma cells that had been stimulated by forskolin [17]. The inhibition of adenylyl cyclase activity was observed in N18TG2 and rat cerebellar membranes [18, 19]. Data from each of these groups indicated that anandamide has low potency but high efficacy with a percentage of 80% of WIN55212-2 with regard to the inhibition of adenylyl cyclase [19]. The involvement of Gi/o-protein was supported by the attenuation of both inhibition activities in presences of pertussis toxin [16, 17, 20,].
 

To evaluate whether anandamide-mediated CB2R stimulation would inhibit the production of cAMP or the activity of adenylyl cyclase CHO cells expressing a human recombinant CB2 receptor were assayed. Initial reports indicated that the activation of CB2R by anandamide did inhibit cAMP accumulation in these cells [20]. While additional reports revealed that no measurable inhibition by anandamide even at higher doses then previously examined [16]. This difference in activity could be attributed to differences in experimental design, procedures, or reagent selection. Evaluations of this assay preformed in the same indicate that anandamide exhibit low potency and low efficacy.#p#分頁標(biāo)題#e#
 

Additional information relating to the signal transduction pathways of both cannabinoid receptors was provided during investigation into the role of these receptors in cell proliferation and death (Guzman et al., 2001). The downstream the mechanism of action during these processes was proposed to activate the mitogen-activated protein kinase (MAPK), a member of the extracellular signal regulated kinase (ERK) family. Additional data indicated that ERK activation was mediated through the G-protein coupled to the cannabinoid receptor in several cell types (Bouaboula et al., 1995; Bouaboula et al., 1996; Wartmann et al., 1995; Liu et al., 2000). The immediate steps between the G-protein activation and the phosphorylation of ERK were hypothesized to differ for each cannabinoid receptor. The activation of ERK when mediated through CB2R demonstrates a dependency on the activation protein kinase C (PKC), while CB1R activation of ERK is blocked by a phosphotidylinsitol 3-kinase inhibitior (Bouaboula et al., 1995).
 

The phosphorylation of ERK through activation of the cannabinoid is reported to induce Krox 24, an immediate early gene [90, 91], activate NHE1, a sodium hydrogen exchanger [93], and stimulate the oxidation of glucose in astrocytes [92]. The evidence of several signal transduction pathways indicates that anandamide stimulation of the known cannabinoid receptors elicits systemic responses. Research into the physiological effects of endocannabinoid system has uncovered responses in the stomach, the reproduction system, and cardiovascular system. Additional investigation has suggested that the cannabinoid receptors play an inhibitory role in the metastasis in several types of cancer cells [28, 29, 30,]
 

The Role of Cannabinoid Receptors in Metastasis

Recent studies exploring additional treatments for cancer have identified the CB1R as a potential therapeutic target. Previously researcher have suggested that anandamide-meditated CB1R stimulation inhibits apex ras activity and blocks proliferation of v-K-ras transformed rat thyroid cells in vivo and in vitro [10]. This initial data lead to investigations into how the proliferation was arrested. K-ras has been shown to contribute to the malignancy of transformed cells through the support of angiogenesis and metastasis. Through experiments employing 2-methyl-2’-f anandamide (Met-F-AEA), a metabolically stable analogue for anandamide, a hypothesis has been developed for the mechanism by which CB1R inhibits angiogenesis. Data from this series of experiments indicated that anandamide increased the levels of the cyclin-dependent kinase inhihibitor p27 (kip1). This kinase inhibitor is known to reduce the expression of the proangiogenic factor vascular endothelial growth factor (VEGF) and its receptor flt-1/VEGFR-1 by the down-regulation of the kinase p21ras [10].
 

Further investigations employing Met-F-AEA as a ligand in a metastizing animal model revealed a reduction in size and number of metastatic nodes [11]. The involvement of CB1R in this reduction was supported by the attenuation of these results upon administration of SR141716A. These findings were supported by reports of the role of the endocannabinoid system in impaired non-solid tumor cell migration [20]. The molecular mechanism of how migratory inhibition occurs has become a focal point of endocannabinoid system research. Grimaldi and colleagues hypothesize that the phosphorylation of the focal adhesion kinase (FAK) and Src phosphorylation play an important role in the activation of proteins necessary for migration such as integrins, a family of trans-membrane adhesion receptors, and matrix metalloproteinases (MMP) [34]. Unfortunately, the data generated from the investigations fostered by this group demonstrated no causal relationship between the phosphorylation of FAK and the activation of migratory proteins. Still, the data sets did confirm correlation between the anandamide-mediated activation of CB1 and an inhibition of migration. These results provide a possible new therapeutic target and a new direction for further cancer research.#p#分頁標(biāo)題#e#
 

The role of CB2R in the cancer model is currently being investigated by several research groups. Recent developments have categorized this receptor as possessing an inhibitor role in cancer metastasis. Still, the role of this receptor has been elucidated outside of evaluations that included the stimulation of the CB1R [35, 36]. Further research may uncover the specific role of the CB2R and the mechanism thru which it operates in order to use its signal transduction pathway as a therapeutic mechanism.
 

Anandamide Receptor

In addition to the two known cannabinoid receptors there is pharmacological and biochemcal evidence of a non-CB1/ non-CB2 receptor, the anandamide receptor. The research into additional receptors for cannabinoid ligands began with pharmalogical data indicating differences in the activation profile of anandamide versus the synthetic cannabinoid analogues (Mechoulam et al., 1995; Pertwee 1999; Khanolkar et al., 2000). Preliminary studies of this phenomenon suggested that the differences seen in the anandamide pharmacological profile were due to the action of its metabolites or its behavior as a partial agonist of CB1R (Randall et al., 1997; Kunos 2002; Chaytor et al., 1999). Additional studies suggested that the alternatives effects were mediated through a non-CB1R/ non-CB2R (anandamide receptor) and direct connect with various ion channels.
 

The support for the presence of a novel receptor was first observed in mouse astrocytes. The inhibition of isoproterenol-induced cAMP accumulation was unable to be attenuated by the administration of either of the known cannabinoid receptor antagonist in these cells. Further research into endogenous cannabinoid mechanism revealed several vascular responses that were not characteristic of the known cannabinoid receptor. The responses included intiation of hypotension and bradycardia in animal models, and produce of vasorelaxation in various types of microvasculature (Kunos et al., 2000a; Kunos et al., 2000b; Kunos 2002; Mukhopadhyay et al., 2002). Within our laboratory data recordings of proangiogenic responses in si-CB1R HUVEC and CB1 knockout mouse model have attenuated through the administration of O-1918, the reported antagonist for the anandamide receptor. These data lend support to the hypothesis that the anandamide receptor plays a role in endocannabinoid-mediated neovascularization.
 

Over twenty angiogenic stimuli have been identified; as a mediator of angiogenic responses anandamide is theorized, by our laboratory, to novel member of this group when mediated through the anandamide receptor [5]. Currently the vascular endothelial growth factor (VEGF), angiogenic response model is the most highly characterized and is thus used to demonstrate the signal transduction pathway of angiogenic stimuli and their activation of promigratory enzymes such as MMP-2 and MMP-9.
 

To begin angiogenesis VEGF binds to its receptor, VEGFR, and uses an influx of calcium, heat shock protein 90, and the (phosphatidylinositol-3-OH kinase) PI3k-Akt pathway to phosphorylate endothelial nitric oxide synthase (eNOS) [18,19]. Once phosphorylated the activated eNOS catalyzes the conversion of L-arginine to L-citrulline and manufactures nitric oxide. These responses bear similarity to confirmed non-CB1/ non-CB2 anandamide-mediated activation of ion channels and production of nitric oxide .#p#分頁標(biāo)題#e#
 

The newly manufactured nitric oxide gas is proposed to stimulate the synthesis guanylyl cyclase, which generates cyclic GMP (cGMP). This compound activates cGMP-dependent protein kinase (PKG) that phosphorylates mitogen activated protein kinase (MAPK). After the activation of MAPK the mechanism that leads to MMP-2 and MMP-9 secretion is not well characterized. Studies employing normal and reverse zymography, Western blot, and immunogold analysis indicate that preformed MMP-2, MMP-9, and membrane-type (MT1-MMP) enzymes are shed from secretory vesicles following pro-angiogenic stimulation [5]. These proteolyetic enzymes are then activated. The active form of gelatinase A generated from the thrombin/APC activation disrupts the existing capillary bed. Gelatinase B is secreted in short bursts and is proposed to degrade the basement membrane.
 

Following the breakdown of the basement membrane the digest of the interstitial stroma begins. The type I collagen of the interstitial stroma upregulates the production of MT1-MMP, this active transmembrane enzyme then catalyzes the activation of MMP-2 [6]. This MMP-2 activation is sustained until the new blood vessel secretes its basement membrane [2]. Once the interstitial stroma is removed the newly formed endothelial cells, derived from the cell division of cells of a pre-existing capillary, migrate down the provided space and form the new blood vessel through bipolar interaction [8,9]. MMP-2 and MMP-9 are quickly inhibited to avoid excessive removal of the matrix, which would destabilize the newly formed blood vessel.
 

Signal Transduction Inhibitors

Pertussis toxin is an AB5-exotoxin manufactured by the bacterium Bordetella pertussis (Ryan and Ray 2004). The B subunit binds to the receptor on the cell membrane which activates the A subunit. The active enzyme catalyzes the ADP-ribosylation the α subunit of Gi, Go and Gt heterotrimeric proteins. The blockade of α-subunit activation prevents the coupling of these G-proteins to perspective receptors (Burns 1988).
 

Nitric oxide inhibitors are used to prevent the phosphorylation of nitric oxide synthase. L-NG-Nitroarginine methyl ester (hydrochloride) is a nonselective nitric oxide inhibitor. L-NAME requires hydrolysis in order to be activated (Griffith and Kilbourn 1996). This compound once active inhibits endothelial, neuronal, and inducible nitric oxide synthases. NG-nitro-L-Arginine, L-NG-Nitroarginine (L-NNA) is a competitive inhibitor of the neuronal and endothelial isoforms.
 

Nitric oxide donors generate differing levels of nitric oxide reactive species. Linsidimine or 3-Morpholino-syndonimine (SIN-1 chloride) is a potent vasorelaxtant which produces both nitric oxide and superoxide. Sodium nitroprusside dehydrate (SNP) is a secondary potent vasorelaxant, which liberates nitric oxide from the conversion of haemoglobin to cyanomethaemaglobin (Butler and Megson 2002). In addition to the release of nitric oxide cyanide ions are produced of the reactions catalyzed by this compound.#p#分頁標(biāo)題#e#
 

Matrix Metalloproteinase

The primary goals of a cancer cell are to proliferate, establish nutrient source, and acquire the ability to metastasize. The first two goals allow the cells to survive for a limited amount of time but is metastasis that is responsible for the pathology and fatality attributed to cancer. In order for metastasis to occur a cancer cell or cells must complete four essential steps; detachment of the cell from a primary tumor into the circulation, attachment and invasion of a target site (i.e. an organ, or tissue), growth of the cell in cluster, and neovascularization for nutriment of the new tumor [1]. The first step, detachment involves the cells overcoming the two primary forms of adhesion; cell-cell adhesion, and cell-extracellular matrix. Cell-cell to adhesion, which involves the functioning the E-cadherins, and cell-extracellular matrix adhesion, which is facilitated by integrins, are diminished in malignant cells [2]. Once the malignant cells overcome cell-cell adhesion they must degrade the ECM in order to break the cell-ECM. A special family of protease are produced and secreted in order to accomplish this goal matrix metalloproteinases (MMP).
 

The matrix metalloproteinase (MMP) family comprises over twenty-five proteolytic enzymes, which breakdown the extracellular matrix and cell surface proteins [3]. Two members of the MMP family have been shown to facilitate the degradation of the extracellular matrix (ECM) prior to cell movement instances such as angiogenesis and metastasis, MMP-2 and MMP-9 or gelatinase A and gelatinase B respectively [5]. Both of these enzymes breakdown collagen, elastin and gelatin types I, IV, V. These three protein fibers are the principle components of the extracellular matrix [2]. This matrix provides a barrier against primary cancer cell motility. The massive releases of MMP-2 and MMP-9 due to elevated cellular response are theorized to provide the initiate mechanism for movement of new cells in the tumor mass.
 

To aid in the disclosure of the matrix metalloproteinase’s purpose in metastasis, many researchers have investigated the structure of the entire matrix metalloproteinase (MMP) family [2]. From the examination of each MMP it has been revealed that all enzymes in this family have at least three functional domains, the signal peptide (pre-peptide), the propeptide, and the catalytic domain. The signal peptide directs the uptake of the MMP enzymes by the endoplasmic reticulum. Within the endoplasmic reticulum the enzymes are modified and sent to the Golgi apparatus to be packaged for secretion. The propeptide is domain removed when the enzyme is activated. This domain consists of approximately 80-90 amino acids [2]. This highly conserved sequence, PRCGVPD, of amino acids has a cysteine residue that is involved in activation of MMP enzymes [2]. The catalytic domain is ellipsoid shaped with a small active site cleft. In this cleft there are two zinc ions, and a calcium ion, which are important in the catalytic action of this enzyme [3, 5, 6]. In addition to the three domains previously described, most MMPs posses another common region in the C-terminal domain. This domain is composed of a sequence homologous to hemopexin [4]. Additional domains present in MMP structure are used to divide these endopeptidases into subdivisions.#p#分頁標(biāo)題#e#
 

Each member of a sub-group shares at least one structural difference from the other MMPs. The two matrix metalloproteinases that play an important role in angiogenesis are in the gelatinase sub-group, MMP-2 (gelatinase A) and MMP-9 (gelatinase B). The structure that differentiates these two enzymes from the other members of the MMP family is the three repeats of fibronectin-type II module in the catalytic domain [6]. MMP-9 has an additional domain that is not present in the MMP- 2 structure, a collagen-like domain [4]. The fibronectin-type II module and the collagen domain are regions that allow MMP-2 and MMP-9 to bind tissue inhibitors of metalloproteinases and breakdown collagen. An additional MMP, which plays an auxiliary role in angiogenesis, is membrane type 1 matrix metalloproteinase (MT1-MMP). This proteolytic enzyme has a transmembrane domain that anchors it to the plasma membrane and allows the enzyme to process membrane proteins such as an integrins and tissue transglutaminase [8].
 

Identification of two MMPs as possible initiation steps for metastasis would seem to suggest that the regulation and performance of this process would be straightforward and a simple therapeutic target. Unfortunately, this is not the case matrix metalloproteinase (MMP) activity is highly regulated from transcription through secretion. Consequently, it is important to fully comprehend the activation and regulation of MMP-2, MT1-MMP (as an activator of gelatinase A), and MMP-9. The following sections will address gene construction, activation, and regulation of each of these matrix metalloproteinases.
 

Gelatinase A (MMP-2)

MMP-2 is a constitutively produced enzyme that digests elastin, collagen, and gelatin types [9]. Gelatinase A is considered a constitutive enzyme because it does not have a well-defined regulatory element. Gelatinase A promoter lacks a TPA responsive element, activation protein-1 (AP-1), and PEA-3. AP-1 and PEA-3 are known transactivator sequences [2]. This proteolytic enzyme also has an unusual TATA box, whose sequence differs from the sequence seen in most promoters. This sequence is suspected to be the element that controls the basal secretion of this endopeptidase. The final variation in the gelatinase a promoter is the absence of the transforming growth factor b (TGF-b) inhibitory region [2]. Consequently, the TGF-b transcription factor is not an inhibitor of MMP-2 production. From this promoter the basic transcription of MMP-2 is regulated and the translation of this sequence gives rise to the 72kD latent form, pro-MMP-2. This zymogen form is not active; the propeptide region must be cleaved in order for MMP-2 to perform its proteolytic functions.
 

Two pathways regulate the cleavage of the propeptide and subsequent activation of gelatinase A. Each pathway manufactures a MMP-2 with a different molecular weight hence the 64kD and 62kD active forms of this enzyme. The first activation pathway employs the previously mentioned MT1-MMP protease. MT1-MMP must be activated and correctly position in the cell membrane before this enzyme can cleave pro-MMP-2. MT1-MMP is transformed into an active form intracellularly and transported to the cell membrane where it is localized on the cell surface by caveolae [8]. Caveolae are cholesterol rich microdomains involved in clathrin-independent internalization of receptors, recycling of molecules to the Golgi complex and signal transduction [8]. This lipid complex concentrates MT1-MMP in the area of the endothelial cell membrane adjacent to where neovascularization stimulants are localized. The function of the caveolae is to help ensure that matrix degradation is confined to a specific area of the extracellular matrix.#p#分頁標(biāo)題#e#
 

Once MT1-MMP is localized on the cell surface the second phase of this activation pathway can begin. Tissue inhibitor of metalloproteinase 2 (TIMP-2) uses its N-terminal region to bind to the catalytic domain of the MT1-MMP enzyme. This binding primes the membrane bound endopeptidase, which then binds progelatinase A through its C-terminal domain. The interaction orientates the pro-MMP-2 enzyme so that a free MT1-MMP can cleave its propeptide domain. The cleavage of the propeptide region gives rise to the partially active 62kD form of MMP-2 [2, 8]. The 62kD MMP-2 enzyme is later auto catalyzed to a fully active form at a weight of 59kD [2]. This pathway is not inhibited as long as the concentration of TIMP-2 is proportional to the available MT1-MMP enzymes [11].
 

The second pathway of MMP-2 activation is independent of MT1-MMP and is activated during the anti-coagulation process [2]. The catalysis for this pathway, thrombin, may directly activate MMP-2 or mediate the stimulation of this proteolytic enzyme. For thrombin to cleavage pro-MMP-2 directly the cell membrane must be presence but the incorporate of the thrombin receptor in this membrane is not necessary [12]. This division of thrombin activation is rapidly induced. The alternate division of this pathway does employ thrombin but it does not directly interact with MMP-2. Thrombin interacts with a constitutively produced cell surface protein thrombomodulin [2]. The thrombin-thrombomodulin complex is the catalysis for the conversion of protein C, an endothelial cell surface protein to, the anticoagulant serine protease, activated protein C (APC). APC directly cleaves the propeptide domain of pro-MMP-2 and generates active MMP-2. APC can also catalyze the activation of the partially active intermediate produced in the MT1-MMP pathway.
 

Gelatinase B (MMP-9)

MMP-2 and MMP-9 have four identical domains and the same substrates, but these two endopeptidases differ greatly. Their differences are evident in their structure, promoter composition, and regulation. The latent form of MMP-9 has a weight of 92kD, much larger than the 72kD reported for MMP-2. This difference in weight is partially due to the glycosylation noted on the secreted structure at three N-linked sites and several O-linked sites [7].
 

The promoter region composition of MMP-9 demonstrates that enzyme is highly regulated thus is not constitutively produced. The degree of regulation is evident by the presence of two TPA responsive elements (TRE), and a transforming growth factor-β (TGF-b) inhibitory sequence [7]. The two TRE sites may serve as a binding site for activation protein-1 (AP-1) transcription factor, while as the name implies the inhibitory sequence binds TGF-b [7]. Unlike MMP-2 little is known about the stimulation of MMP-9 production or how this enzyme is catalyzed to generate its active form.
 

The regulation of MMP-9 is not well characterized. The only production stimulant that has been identified is the activated CD40 T cell. Active CD40 cells stimulate gelatinase production when they interact with endothelial nerve growth factor receptors on saphenous vein endothelial cells [15]. There is very little information available about how this proteolytic enzyme is cleavaged to give rise to its active forms. Drs. Sternlicht and Werb hypothesize that the active form of MMP-2 cleaves the propeptide of MMP-9 [16]. No other group has confirmed this hypothesis, nor has the specific mechanism by which this activation occurs been elucidated. The packaging of MMP-9 for secretion does lend support to this hypothesis.#p#分頁標(biāo)題#e#
 

Active and latent forms of MMP-9 are packaged in secretory vesicles with active MMP-2, in the cytoplasm [5]. These vesicles are proposed to serve as a catalytic environment for MMP-9 [5]. The present secretion model for gelatinase B indicates that its secretion into the extracellular matrix is mediated by the presence of phorbol myristate acetate, a tumor promoting chemical and enhanced by tumor necrosis factor alpha (TNF-α) in endothelial cells. Once secreted it is suggested that MMP-9 can complex with either TIMP-1 in the matrix or bound to the hyaluronan receptor (in CD 44 cells) [17].
 

Inhibition of MMP-2 and MMP-9

In the description of both MMP-2 and MMP-9 tissue inhibitors of metalloproteinases (TIMPs) were mentioned as enzymes that aided in the activation of these two proteolyetic enzymes. As eluted to in the MMP-2 regulation description TIMP enzymes are also responsible for the inhibition of matrix metalloproteinase enzymes. In the case of MMP-2 and MMP-9 these enzymes are inhibited by TIMP-2 and TIMP-1 respectively [10]. This family of inhibitors neutralizes the proteolytic action of MMPs by binding to the catalytic site of these enzymes and sequestering the zinc ion in a cysteine residue. This cysteine residue is located on the N-terminus of the TIMP inhibitor [6]. The TIMP-mediated inhibition of MMP activity is rapid but this response is not long lasting. Two more potent inhibition pathways with considerably more longevity have been identified [6, 18].
 

Dr. Seo and his colleagues have demonstrated that TIMP-2 is involved is the central enzyme in one of these pathways. This inhibition pathway is independent of MMP deactivation [6]. Dr. Seo’s group has determined that this mechanism is the major mode of matrix degradation inhibition. In this pathway, with hMVEC cell model, TIMP-2, and the a3b1 extracellular receptors interact with a receptor tyrosine kinase, such as fibroblast growth factor receptor [6]. This interaction disrupts the mitogenic effect necessary for metastasis; the halt in proliferation of the cells inhibits the signal transduction cascade and therefore the downregulates the activation of MMP-2 and MMP-9.
 

The potent inhibition of MMP activity and metastasis is not limited to the TIMP enzymes. Research conduct by Dr. Portella and colleagues, using cell gliomas or human astrocytomas, indicates that cannabinoid analogs when mediated through the CB1 receptor inhibit cell proliferation [18]. This proliferation inhibition suppresses the expression of MMP-2 and MMP-9. With the completion of the description of the known MMP-2 and MMP-9 inhibitors we have covered the mechanism of both of these proteases. Additional research is required to determine the correlation between the signal transduction pathway of metastasis and the actions of the endocannabinoid system.
 

CHAPTER III

ANALYSIS OF DATA

#p#分頁標(biāo)題#e#

Cell culture

The LNCaP cells, a gift from Dr. Delores Grant (Julius L. Chambers Biomedical/Biotechnology Research Institute - North Carolina Central University, Durham, NC, USA), were seeded into T-175 flasks obtained from Fisher Scientific/Naglene (Pittsburgh, PA, USA). The cells were maintained on RPMI 1640 media containing 10% fetal bovine serum, 5% streptomycin and 5% MEM nonessential amino acid solution purchased from Invitrogen/ Gibco division (San Jose, CA, USA). The media volume was maintained at17mL per flask, and the media were changed every two days or as needed depending on the growth rate of the cells. Each T-175 flask was split at 1: 6 ratio once approximately 90% confluent. In order to harvest the cells the media was removed and the cells were washed with 6ml 1X PBS (Phosphate Buffered Saline), which diluted from the 10X stock purchased through Invitrogen/ Gibco division (San Jose, CA, USA) and sterilized using and autoclave system, to eliminate any remaining media. The PBS/media solution was removed and 6mL of versene (1:5000 PBS-EDTA) was used to round the cell and reduce the cell-cell adhesion. After this the versene was pipetted off and 6mL of complete media, or media containing all the supplements, was added. The media was used to detach the cells from the surface of the flask. The media cell suspension was placed in a 15ml sterile conical tube.
 

One milliliter of the cells suspended in media was reseeded into the T-175 flask to maintain the cycle. The other 5ml of the solution were seeded on five 100mm3 culture plates. In addition to the 1mL cell suspension 9mL of complete media were added to the five culture plates. Within 24 hours the initial 10mL of complete media was removed and fresh media was added. Once the culture plates were 90% confluent the complete media was removed from each plate and 5ml serum-free RPMI media were added for approximately 18 hours before the time point at which the cells were dosed. The serum-free was used to synchronize the cell cycles of the samples and eliminate any growth factors contained in the complete media
 

Treatment of Cells

After the 18 hour wash period, fresh serum-free media was placed on the culture plates to prepare the cells for the dosing procedure. The optimal dose concentration obtained through experimentation for each drug (R(+)-Methanandamide (R(+)-Arachindonyl-1’-hydroxypropyl-2’-propylamide), CP55940 (5-(1,1-Dimethylheptyl)-2-[5-hydroxy-2-(3-hydroxypropyl) cyclohexyl] phenol], WIN55212-2 (R)-(1) [2,3-dihydro-5-methyl-3-[4-morphlino)methyl]pyrrolo-[1,2,3-de]-1,4-benzixaun-6-yl](1-naphthyl)methanone, SR141716A(N-piperidin-1-yl)-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide hydrochloride), SR144528 (N-[(1S)-Endo-1,3,3,-trimethylbicyclo[2.2.1.] heptan-2-yl]-5-(4-chloro-3-methylphenyl)-1-(4-methylbenzyl)-pyrazole-3-carbo) and O-1918) was determined to be 1mM in a vehicle of FBSA (Fatty acid free). The vehicle was prepared prior to reach dosing series from a 50mg/mL stock solution. The stock solution of FBSA was diluted to a concentration of 0.001mg/mL when used as a vehicle in these experiments. The control samples were treated with representative amounts of FBSA for each treatment criteria.#p#分頁標(biāo)題#e#
 

The dosing procedure began with the samples that were to be treatment with a receptor antagonist such as SR141716A (a CB1 receptor antagonist), SR144528 (a CB2 receptor antagonist) obtained as gifts from Research Triangle Institute (Research Triangle Park, NC, USA)/National Institute of Drug Abuse contract (Bethesda, MD, USA), or O-1918 (the putative Anandamide receptor antagonist) a gift from Dr. George Kunos (National Institute on Alcohol Abuse and Alcoholism of the National Institutes of Health, Bethesda, MD, USA). These compounds were applied and given a period of thirty minutes to preferentially bind to the appropriate receptor. Following the administration of the receptor antagonist the cell samples were treated with the synthetic cannabinoid (methanandamide purchased from Sigma (St. Louis, MO, USA), WIN55212-2, or CP95540). The actually treatment time began once the cells were treated with the agonist compound. Once properly dosed the plates were placed back in the incubator and keep at a constant temperature of 37°C for the duration of their treatment period.
 

Protein Assay

At the end of the treatment period the plates were collected and placed directly on ice. The media was removed and pipetted into microphage tubes. Each of the samples was kept on ice through out the duration of the preparation. The samples were then spun in a refrigerated clinical centrifuge at a speed of 13,000 rpm for thirty minutes. Sixty microliters of each sample was then extracted from the lowest portion of the microfuge tube and placed in a fresh microfuge tube. Thirty microliters of these samples were placed in a second set of microfuge tubes and diluted with deionized water to obtain a volume of 470ul. Thirty microliters of HME was added to diluted protein sample. A series of control protein samples with known globulin protein concentrations were generated in a like manner to provide a standard curve. Two milliliters of Bradford reagent [1X] was made for the stock solution (125mL Coomassie blue-EtOH, 250 mL Phosphoric acid and diluted to a volume of 500 mL). The samples and the standards were read at a visible light wavelength of 495 nm. The standards or vials with know protein concentrations were used to generate a standard curve. The absorbance of the samples was compared against the slope of the standard curve. These calculations were used to determine the protein concentration of each sample.
 

Zymography

The remaining 30uL of each sample were prepared for Zymography using the protocol developed by Drs Leber and Blakwill. The samples were combined with 30uL of the Zymogram sample buffer [2X] (2.5ml of Tris, 2ml of Glycerol, 4ml of 10% SDS, 0.5ml of 0.1% Bromophenol blue solution, and 1ml of H2O)[1]. In addition to preparing the samples an active and inactive enzyme mixture of MMP-2 (human) and MMP-9 (mouse) (obtained from Sigma) were prepared at a concentration of 12.5ng in 30uL of 1X Zymogram sample buffer to be used as positive controls.#p#分頁標(biāo)題#e#
 

Once the protein concentration was determined to be constant for all the samples 30mL of each sample was loaded into a lane of the 10% Zymogram (Gelatin) gel (Invitrogen pre cast SDS-PAGE gel with 0.1% gelatin). Prior the insertion of the gel the 800ml of 1X Tris–Glycine SDS running buffer (100ml of 10X Tris–Glycine SDS Running buffer to 900ml of deionized water) was placed the gel running apparatus (Invitrogen X Cell Sure Lock) [2]. The upper buffer chamber was filled with 200ml of the running buffer, while the lower chamber was filled with 600ml. Once the apparatus and the gel were properly prepared, the device was run at a constant voltage of 125V for 120 minutes or until the dye front was run off the gel, [2].
 

After two hours the gels were removed from the casting plates and placed in a sterile dishes containing 100ml 1X renaturing Buffer (100ml of 10X renaturing Buffer (Invitrogen) in 900ml of deionizied water) for 60 minutes with gently agitation at room temperature [2]. Next the renaturing buffer was decanted, and 100ml of 1X developing buffer was added (100ml of 10X developing Buffer (Invitrogen) in 900ml of deionizied water) for 5 minutes at room temperature to equilibrate the gel [2]. This solution was decanted and 100 mL fresh Developing buffer was placed on the each gel [2]. The gels were incubated in the within an incubator at 37°C for 16 hours.
 

Prior the removal of the gels from the incubator, staining (40% methanol, 10% acetic acid, 50% water, 0.5% (w/v) Coomassie Brilliant Blue R250) and destaining solutions (10% acetic acid solution) were prepared. The gels were then removed from the incubator and developing buffer was decanted. Each gel was washed with deionizied water to refer any residues left by the buffer and 100ml of the staining solution was immediately added. The staining solution remained on the gels for 1 hour. The staining solution was decanted and immediately replaced by 50ml of the destaining solution. The gels were allowed to destain until the positive controls became apparent in the each gel. The gels were then placed in deionized water to neutralize the destaining process and maintain the gels through the data recoding procedure.
 

The destained gels placed on a light box and, exposed to transillumating white light in the Alpha Innotech. The images of gels were viewed on the FluroChem software. The focus and aperture size on the camera within the alpha intact was adjusted until the image of gel displayed the sharpest resolution. Once a sufficient image was obtained the gel was photographed at an exposure time of one-tenth of a second. The background of the gel was then adjusted was the contrast functions of the program and image was printed. The digital image was stored as a TIF file and employed to quantify the results using the Image-Pro software.
 

Evaluation of Data

Raw data obtained from the Image-Pro protocol were normalized using the basal reading of each gel as a standard and dividing each treatment group by this figure. Next the percentages of each group including basal were decreased by 100, which placed all basal reading at zero. The data groups with an N=2 and above were first evaluated using a one-way ANOVA test. Data groups posses n values of three and above were analyzed by Student t test as a post evaluation tool.#p#分頁標(biāo)題#e#
 

Results

The optimazition of the zymogram procedure was accomplished by employing human umbilical vein endothelial cells (HUVEC) with documented responses to the cannabinoid receptor agonists and the cannabinoid and anandamide receptor antagonist. With appropriate conditions in place to conduct the zymogram, time courses were conducted to determine a time point of anandamide mediated activation of MMP activity in LNCaP prostate cancer cells. In order to determine the appropriated time course for the treatment of the LNCaP prostate cancer cells, the cells where treated with a physiological relevant dose of methanandamide [1uM] at time points of 5, 10, 30, 60, and 120 minutes (figure not shown). No data was retrieved from the initial experiment therefore the experiment was repeated to ensure that this experiment was not an artifact. The use of purified MMP-2 and MMP-9 enzymes as controls ensured that the reagents were functioning and the protocol was appropriate. A second time course experimental series was then preformed employing the same concentration of methanandamide at time points including 4, 6, 8, 12, 16, and 24 hours. This data set revealed no activation of the MMP enzymes as well.
 

To additional time course where preformed with the same time line as the prior time course experiments. During this series of experiments the LNCaP cells were pretreated for 30 minutes with SR141716A [1uM] prior to treatment with methanandamide, according to the treatment team protocol. MMP activity was observed at the 24 hr time point only, following the administration of the CB1R antagonist (figure not shown).
 

Dose Response Curve

The determination of the optimal time of treatment allowed the group to begin to evaluate the concentration of Met-AEA needed to elicit stimulation of MMP activity in LNCaP cells. A dose response curve was first conducted with varying concentrations Met-AEA alone to obtain support for the findings of the initial time course experiments. The LNCaP cells were treated for 24 hours at concentrations of 0.001uM, 0.01uM, 0.1uM, 1uM, and 10uM. No data above basal expression was retrieved from this experiment during its first or second trial (fig. 1A). The dose response curve was preformed again this time incorporating Met-AEA at varying concentrations in the presence of SR141716A [1uM] according to the methods section. The Met-AEA concentrations corresponded to the concentrations listed for the first dose response curve. The highest responses were seen at methanandamide concentrations of 1uM, and 10uM.
 

Met-AEA treatment in the presence of SR141716A stimulated MMP activity in LNCaP prostate cancer cells and was attenuated by O-1918

Met-AEA [1uM] in the presence of SR141716A stimulated MMP activity in LNCaP cells as indicated by an increase in the MMP optical density readings (fig. 2A & 2B). The stimulation of the MMP activity was attenuated by the addition of O-1918 to the Met-AEA/SR141716A treatment (fig. 2B). The treatment of LNCaP cells with O-1918 revealed a statically significant increase in MMP activity which was attenuated by the administration of Met-AEA in the presence of O-1918.#p#分頁標(biāo)題#e#
 

Met-AEA mediated dose response curves. (A) Dose response curve conducted with varying concentration of Met-AEA[1uM] alone for 24 hours exhibited no significant response above that of the basal response. (B) LNCaP prostate cancer cells within the second response curve were pretreated with SR141716A [1uM] according to the treatment protocol described in materials and method section. An increased enzyme activity was detected in LNCaP cells treated with Met-AEA [1uM] and SR141716A [1uM] for 24 hours.
 

The effect of O-1918, the anandamide receptor antagonist, on Met-AEA/SR141716A mediated stimulation of MMP activity. (A) LNCaP cells treated with Met-AEA [1uM] and SR141716A [1uM] expressed a significant increase in MMP activation (* P< 0.01with respect to basal reading or # P< 0.05 with respect to Met-AEA alone) N=5. (B) The significantly increased MMP activation mediated through Met-AEA/SR141716A treatment was attenuated by the addition of O-1918 [1uM] N=3. (C) MMP activation of LNCaP cells treated with O-1918 [1uM] alone was significantly higher than basal levels (* P< 0.05) and Met-AEA (# P<0.05) N=6. A return to basal levels was observed when were cells were treated with Met-AEA [1uM] and O-1918 [1uM] according to the cell treatment protocol.
 

SR144528 in the presence of Met-AEA elicited a stimulatory response, but in the presence of CP55940

The treatment of LNCaP cancer cells with Met-AEA in the presence of SR144528 stimulated the activation of MMP. This result is exhibited by the increased optical density recorded (fig. 3A). This reading was considerably lower than the response recorded for Met-AEA treatment in the presence of SR141716A. CP55940, a specific cannabinoid receptor agonist, did not elicit a response above basal even in the presence of SR144528.
 

WIN55212-2 and CP55940 in the presence of SR141716A caused no stimulation of MMP activity in LNCaP cells

CP55940 administration in the presence and absence of SR141716A did not display the increase of MMP activity observed with Met-AEA/SR141716A treatment of these cells (fig. 4A). The optical densities observed during these experiments were comparable to the basal levels. Although not statically significant there is a detectable inhibition seen in the CP55940/SR141716A measurement versus those of SR141716A alone (fig. 4A). The treatment of LNCaP cells with WIN55212, a mixed CB1/CB2 receptor agonist, did not demonstrate the ability to increase MMP activation alone or in the presence of SR141716A (fig. 4B). The optical densities recorded bear no statically significant difference from basal or Met-AEA levels.
 

Effect of cannabinoid agonists and CB2R antagonist on MMP activity in LNCaP cells. (A) A statistically significant difference was detected between basal levels of MMP activity and levels recorded in cells treated with Met-AEA and SR144528, the CB2R antagonist (* P< 0.05) N=6. (B) Administration of CP55940 (CP), a cannabinoid receptor agonist, did not elicit the same response in the presence of SR144528 (N=2).#p#分頁標(biāo)題#e#
 

The effects of cannabinoid specific agonist in the presence of SR141716A. (A) LNCaP cells treated with CP55940 [1uM] alone and in the presence of SR141716A displayed no significant difference from basal levels (N=2). (B) Cannabinoid receptor agonist WIN55212-2 [1uM] administered alone and in the presences of SR141716A showed no statistical difference between basal readings (N=4).
 

Cannabinoid specific agonist demonstrate no stimulatory response in the presence of O-1918

The CB1/CB2 receptor antagonists, WIN55212 and CP55940, were no able to elicit a stimulatory response in the presence of O-1918 (fig 5A and 5B). The treatment of LNCaP cells with CP55940 [1uM] in the presence and absence of O-1918 [1uM] displayed an optical density reading with no detectable difference from the readings produced by O-1918 alone (fig. 5A). The measurements gathered from treatment of LNCaP cells with WIN55212 [1uM] in the presence of O-1918 displayed no statistically significant difference from the measurements obtained for basal conditions or conditions triggered by the administration of WIN55212 alone (fig 5B).
 

The effect of cannabinoid specific receptor antagonists in the presence of O-1918. (A) CP55940 [1uM] treated LNCaP cells exhibited no significant change in MMP activity in the presence and absence of O-1918 [1uM] (N=2). (B) LNCaP cells expressed no activity above basal in the presence of WIN55212-2 [1uM] alone and in the presence of O-1918[1uM] (N=4).
 

Nitric oxide synthase inhibitors do not effect the expression of MMP activity

The administration of L-NNA in the presence of the Meth-AEA/SR141716A treatment with the 24 hour time point did not inhibit or stimulate MMP activation when compared against Meth-AEA/SR141716A treatments. Additional studies into the mechanism employing a non-specific nitric oxide synthase inhibitor L-NAME demonstrated a small increase in MMP activity above Meth-AEA/SR141716A, but this increase was not deemed statiscally significant.
 

Nitric oxide donor SNP demonstrates an observable increase in MMP activity while Sin-1 displays a slight inhibitory response

The brief investigations into the effects of nitric oxide donors revealed contrary results. SNP demonstrated a slight stimulatory effect (fig.7), which was attributed to proliferation effects of this compound documented within our laboratory through cell viability experimentation. Sin-1 displayed a slight inhibitory response which requires further documentation in order to confirm this response.
 

The effects of nitric oxide inhibitors on Met-AEA/SR141716A mediated stimulation of MMP activity. (A and B) L-NAME [1uM] and L-NNA [1uM] demonstrated no inhibition or stimulation of MMP activity to any significant level (N=2).
 

The response mediated by nitric oxide donors in the presence of Met-AEA and SR141716A. (A) SNP initial data indicates a possible stimulatory role of this nitric oxide donor, but these data are obscured by the increased cell death observed in the samples. (B)The preliminary experience gathered from Sin-1 and do not suggest a significant difference between Met-AEA/SR141716A reading and the readings with this nitric oxide donor.#p#分頁標(biāo)題#e#
 

CHAPTER IV

DISCUSSION, CONCLUSION, AND, RECOMMENDATIONS

The determining optimal conditions for the zymogram process in involved the use of the human umbilical vein endothelial cell (HUVEC) as a model. In previous documentation our laboratory demonstrated that the treatment of HUVEC cells with anandamide and its stable analog methanandamide (Met-AEA) elicited the vasorelaxation response which is identified as a primary step in the process known as angiogenesis (Mukhopadhyay et al 2002). It was therefore proposed by our group that this same cell model might exhibit anandamide-mediated activation of the two matrix metalloproteinases considered to play a principle role in angiogenesis MMP-2 and MMP-9. This hypothesis was confirmed through the detection of activation within our lab (Johnson et al 2004). With this data and experimental established optimum time point and dose concentration, our group was able to evaluate the accuracy of the initial zymogram experiments. The initial experiments aided in the identification of an appropriated run time, incubation length for the activation of the enzyme and an appropriate dye concentration for the visualization of the gel. In addition the initial experiments helped determine the appropriate settings for recording the data to eliminate variations caused by several settings.
 

Preliminary Investigations

The first series of time courses attempted demonstrated that MMP activity could not be evaluated in the presence of Met-AEA alone. This finding suggests an inhibitory mechanism in one of the receptors attributed with mediating Met-AEA signal transduction. This hypothesis was supported by the detection of MMP activity at the 24 hour time point during the second series of time courses which employed the treatment of LNCaP cells with Met-AEA in the presence of SR141716A. The attenuation of the inhibitory responses by SR141716A suggested to that CB1R mediated this response through Met-AEA activation. The optimal time point of 24 hours also suggests that the activity of MMP attributed to the Met-AEA/SR141716A treatment was not a mechanism facilitated by simple secretion.
 

The optimal time point was then used to determine that the physiological concentration of Met-AEA suggested by the experimentation within our group and literature searches (Grimaldi et al., 2006; Portella et al., 2003; Joseph et al., 2004; Offer taler) was appropriate for the LNCaP prostate cancer cell line. The dose response curve was conducted with Met-AEA alone to confirm the observations seen during the time course at several concentrations. The confirmation that no activation above basal activity was produced at any of the range of concentrations tested [0.001uM- 10uM] supported an inhibitory role of a receptor within the endocannabinoid system. The results from the second dose response.
 

With appropriate conditions in place to conduct the zymogram, time courses were conducted to determine a time point of anandamide mediated activation of MMP activity in LNCaP prostate cancer cells. In order to determine the appropriated time course for the treatment of the LNCaP prostate cancer cells, the cells where treated with a physiological relevant dose of methanandamide [1uM] at time points of 5, 10, 30, 60, and 120 minutes (figure not shown). No data was retrieved from the initial experiment therefore the experiment was repeated to ensure that this experiment was not an artifact. The use of purified MMP-2 and MMP-9 enzymes as controls ensured that the reagents were functioning and the protocol was appropriate. A second time course experimental series was then preformed employing the same concentration of methanandamide at time points including 4, 6, 8, 12, 16, and 24 hours. This data set revealed no activation of the MMP enzymes as well.#p#分頁標(biāo)題#e#
 

To additional time course where preformed with the same time line as the prior time course experiments. During this series of experiments the LNCaP cells were pretreated for 30 minutes with SR141716A [1uM] prior to treatment with methanandamide, according to the treatment team protocol. MMP activity was observed at the 24 hr time point only, following the administration of the CB1R antagonist (figure not shown).
 

Dose Response Curve

The determination of the optimal time of treatment allowed the group to begin to evaluate the concentration of Met-AEA needed to elicit stimulation of MMP activity in LNCaP cells. A dose response curve was first conducted with varying concentrations Met-AEA alone to obtain support for the findings of the initial time course experiments. The LNCaP cells were treated for 24 hours at concentrations of 10nM, 100nM, 1uM, and 10uM. No data above basal expression was retrieved from this experiment during its first or second trial (fig. 1A). The dose response curve was preformed again this time incorporating Met-AEA at varying concentrations in the presence of SR141716A [1uM] according to the methods section. The Met-AEA concentrations corresponded to the concentrations listed for the first dose response curve. The highest responses were seen at methanandamide concentrations of 1uM, and 10uM.
 

Met-AEA treatment in the presence of SR141716A stimulated MMP activity in LNCaP prostate cancer cells and was attenuated by O-1918

Met-AEA [1uM] in the presence of SR141716A stimulated MMP activity in LNCaP cells as indicated by an increase in the MMP optical density readings (fig. 2A & 2B). The stimulation of the MMP activity was attenuated by the addition of O-1918 to the Met-AEA/SR141716A treatment (fig. 2B). The treatment of LNCaP cells with O-1918 revealed a statically significant increase in MMP activity which was attenuated by the administration of Met-AEA in the presence of O-1918.