Proteasome inhibitors: Their effects on arachidonic acid release from cells in culture and arachidonic acid metabolism in rat liver cells

Background I have postulated that arachidonic acid release from rat liver cells is associated with cancer chemoprevention. Since it has been reported that inhibition of proteasome activities may prevent cancer, the effects of proteasome inhibitors on arachidonic acid release from cells and on prostaglandin I2 production in rat liver cells were studied. Results The proteasome inhibitors, epoxomicin, lactacystin and carbobenzoxy-leucyl-leucyl-leucinal, stimulate the release of arachidonic acid from rat glial, human colon carcinoma, human breast carcinoma and the rat liver cells. They also stimulate basal and induced prostacycin production in the rat liver cells. The stimulated arachidonic acid release and basal prostaglandin I2 production in rat liver cells is inhibited by actinomycin D. Conclusions Stimulation of arachidonic acid release and arachidonic acid metabolism may be associated with some of the biologic effects observed after proteasome inhibition, e.g. prevention of tumor growth, induction of apoptosis, stimulation of bone formation.


Background
The proteasome degrades many cellular proteins, several with regulatory functions. It is not surprising that proteasome inhibitors affect many biologic processes [1] including prevention of cancer [2]. The effect of proteasome inhibition on cell growth and possible cancer chemoprevention has been reviewed by Adams [3].
In this report, evidence is presented that proteasome inhibitors stimulate PGI 2 production by rat liver cells as well as the release of AA from rat liver, rat glial, human colon carcinoma and human breast carcinoma cells in culture. The stimulation of AA release from rat liver cells is partially inhibited by preincubation of the cells with actinomycin D.

Results and Discussion
Of the cells examined (C-9 rat liver, C-6 rat glial, HT-29 human colon carcinoma and BT-20 human breast carcinoma) the prostanoid metabolic profile has been described only for C-9 rat liver cells (95% is PGI 2 and less than 5% is PGE 2 and PGF 2α ) [13]. At the low cell densities used in this study, only PGI 2 , the main product of COXmediated synthesis, can be quantitatively estimated. The rat liver cells express COX-2 both constitutively and after induction [14]. The effect of time on basal and 12-0-tetradecanoylphorbol-13-acetate (TPA) induced PGI 2 synthesis during incubation of cells with epoxomicin is shown in Fig. 1.
The stimulation of basal PGI 2 production by epoxomicin and TPA-induced PGI 2 production by epoxomicin and lactacystin as a function of dose is shown in Fig. 2. As little as 0.3 µM epoxomicin stimulates TPA-induced PGI 2 production significantly (Fig. 2-B). It is 10 to 15 times more effective than lactacystin (compare Fig. 2-B and 2-C). Using purified bovine erythrocyte proteasomes, epoxomicin inhibits the chymotrypsin-like activity, about 4 to 5 times more effectively than does clasto-lactacystin β-lactone, the derivative of lactacystin [5]. They are almost equally effective on inhibiting the trypsin-like and PGPHlike activities [5]. Assuming that epoxomicin and lactacystin have equal access to the proteasome and that proteasome activity is regulating COX-2 in rat liver cells similarly to neuronal cells [12] then COX-2 may be degraded in the proteasome by cleavage after large hydrophobic residues.
The amplification of PGI 2 production (Figs. 1 and 2) after inhibition by epoxomicin could reflect not only stabilization of COX-2 but also an intracellular increase in the concentration of the substrate i.e. the AA that is produced during hydrolysis of the membrane phopholipids by PLase activity [15]. Extracellular and/or intracellular release of AA will depend, in part, on the localization of the phospholipids in the membrane, e.g. in its inner or outer leaflet [16]. Release of AA in response to several agonists has been described [17][18][19][20].
The effect of a 2, 4 or 6-h incubation on AA release from rat liver and rat glial cells by 1.0 µM epoxomicin was determined. Only after the 6-h incubation were the differences significant statistically. Regulation of PLase activity by the proteasome pathway appears to be a relatively slow Dose-response of epoxomicin, lactacystin and ZLLL on AA release from rat liver cells Figure 3 Dose-response of epoxomicin, lactacystin and ZLLL on AA release from rat liver cells. After incubation for 6 hours. The analyses were performed with triplicate dishes. *-Statistically significant vs MEM/BSA. process. After a 6-h incubation, epoxomicin, lactacystin and ZLLL stimulate the release of extracellular AA from rat liver cells (Fig. 3) and AA release after TPA-induction (3.7% vs 13.5% in the presence of 1.0 µM epoxomicin). Epoxomicin also stimulates the release of AA from rat glial, human colon carcinoma and human breast carcinoma cells ( Table 1). The stimulation of AA release from the rat liver cells after incubation with epoxomicin is partially inhibited by pre-incubation of the cells for 2-h with actinomycin ( Fig. 4) suggesting that a fraction of the PLase is induced. As expected, the inhibition of TPA-induced PGI 2 production by actinomycin D is complete (Fig. 5). Thus, some mechanisms leading to maximum AA release appear to be genomic. The induced PLase activity, probably PLA 2 , could reflect expression of either a secretory or cytosolic PLA 2 or some combination of both enzymes [21].
The release of AA from rat liver cells, most likely resulting from PLase activation, is associated with cancer chemoprevention [14,[17][18][19], [22][23][24]. In addition to its intrinsic biologic activities, AA regulates production of lipoxygen-  ase, cytochrome P-450, and epoxygenase products as well as COX activities. Prostanoid profiles differ with cell type and individual AA metabolites have different pharmacological properties [15]. COX-2 activity, as measured by PGI 2 production, is stimulated by proteasome inhibition (Fig. 1 and 2). Thus, some biologic effects of proteasome inhibition, e.g. stimulation of bone formation [25], may reflect the metabolism of the intracellular AA.
Inhibition of COX-2 activity is one possible mechanism that has been proposed to prevent colon cancer [26]. However, rather than inhibiting, tamoxifen and raloxifene, statins and epoxomicin stimulate COX-2 activity and AA release from rat liver cells [14,[17][18][19]. As shown in Table 1, epoxomicin stimulates AA release from human colon carcinoma, breast carcinoma and rat glial cells. Tamoxifen and simvastatin also stimulate AA release from the human colon carcinoma and human breast carcinoma cells (unpublished data). These drugs have been reported to prevent cancer [27,28]. At least as measured by the COX activity of rat liver cells, tamoxifen, raloxifene, statins and proteasome inhibitors could be preventing cancer by a COX independent mechanism.
AA resulting from proteasome inhibition has many intrinsic biologic properties [reviewed in [29]]. Some of these activities may trigger PLase activity. The causal relationship of AA to cancer prevention (if any) is unclear. Production of AA by the tumor-suppressive type-II phospholipase A 2 (PLA 2 G 2 A) may be related to the cancer prevention [22][23][24]. It is not surprising that control of PLase activities present an attractive area for cancer prevention studies [30].

Methods
The rat liver (C-9 cell line) and human breast carcinoma (the BT-20 cell line) were purchased from the American Type Culture Collection (Manassas, VA, USA). The rat liver glial cells (C-6 cell line) was obtained from Dr.