![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Centre de Recherche en Rhumatologie et Immunologie (G.M., F.M.) and Biotechnology and Bioengineering Unit, Hôpital St-François d'Assise (E.M., R.C.-G.), Centre Hospitalier Universitaire de Québec, Québec, Québec, Canada
Received July 6, 2005; accepted September 23, 2005
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
B inhibition, apoptosis) and display anti-inflammatory properties. However, these drugs induce massive cell vacuolization at similar concentrations, a response initiated by vacuolar (V-) ATPase-dependent ion trapping into and osmotic swelling of acidic organelles. We have examined whether this overlooked response might be related to the effects on cell proliferation and viability using cultured vascular smooth muscle cells and tumor-derived cell lines (Morris 7777 hepatoma, HT-1080 fibrosarcoma). Giant vacuole formation, of confirmed trans-Golgi origin (labeled with C5-ceramide, p230, golgin-97), is a cellular response to all tested amines in the series (
2.5 mM), including triethylamine. These drugs and the V-ATPase inhibitor bafilomycin A1 inhibited smooth muscle cell proliferation, suggesting that acidification of a cellular compartment is essential to cell division. The cytotoxicity was maximal with metoclopramide, and this effect was minimally influenced by bafilomycin A1; furthermore, metoclopramide (2.5 mM) induced apoptosis in tumor cells as judged by poly(ADP-ribose)polymerase (PARP) cleavage. Triethylamine and procainamide exhibit a low level of cytotoxicity variably reduced by bafilomycin cotreatment. In Morris cells, the secretion of 
-fetoprotein is inhibited by amines, consistent with the impairment of the secretory pathway. The most highly substituted 4-aminobenzamides are significant NF-B inhibitors in smooth muscle cells. Although some effects of 4-aminobenzamides are independent of V-ATPase-driven ion trapping (inhibition of NF-B nuclear translocation, agent-specific cytotoxicity, PARP cleavage), other effects are dependent on this phenomenon (vacuolization, a component of the cytotoxicity, inhibition of secretion).
), but the effect was shared by structurally simpler tertiary amines, such as triethylamine (Fig. 1) or neutral red, a pigmented amine. This reversible drug response consists of the osmotic swelling of the trans-Golgi and is prevented by treatment with bafilomycin A1, a V-ATPase inhibitor necessary for the acidification of some intracellular vacuoles (ion trapping of the amines). The subcellular origin of the giant vacuoles was inferred from their perinuclear origin and staining with fluorophore-labeled C5-ceramide. Procainamide induced other effects on cultured smooth muscle cells, including an inhibition of proliferation and migration, but no frank cytotoxicity at 2.5 mM (Morissette et al., 2004a). The vacuolization of the trans-Golgi and secretory vesicles has been reported in other cell types in response to chemically disparate amines such as chloroquine and Tris (Peterlik and Kerjaschki, 1979; Back and Soinila, 1996).
|
). Procaine and procainamide (0.5–1 mM) reportedly demethylate DNA in cultured cells, presumably after direct binding to DNA (Villar-Garea et al., 2003). A nonapoptotic mitotic arrest has been attributed to this drug effect on cultured cells. Metoclopramide and declopramide (3-chloroprocainamide), N-substituted 4-aminobenzamides that can be considered substituted procainamide derivatives (Fig. 1), exerted further cellular effects, such as apoptosis induction in cultured cells (Liberg et al., 1999; Lindgren et al., 2003), G2/M cell cycle block (Olsson et al., 2002), and anti-inflammatory properties in vivo and in vitro (Pero et al., 1999). Phase II N-acetyl metabolites of procainamide-related drug are occurring in vivo (Roden, 2001); N-acetyl-procainamide (NAPA) is nearly as active as procainamide to induce cell vacuolization in smooth muscle cells (Morissette et al., 2004a), but this drug as well as N-acetyl-declopramide or N-acetyl-metoclopramide (NAMA; Fig. 1) does not induce apoptosis in other cells sensitive to declopramide or metoclopramide. On the other hand, all these drugs except procainamide variably inhibit nuclear factor-B (NF-B) activation in an apparently selective manner in various cell types by preventing IB destruction (Liberg et al., 1999; Lindgren et al., 2001, 2003; Olsson et al., 2002). This inhibition could explain, in part, the radiosensitizing and antitumoral effects of these drugs (Baldwin, 2001) and the decreased cytokine secretion in systems where benzamides are used at millimolar or submillimolar concentrations (Pero et al., 1999).
The present experiments aim at documenting whether ion trapping and cell vacuolization have been overlooked and occur at the high concentrations of 4-aminobenzamides that exert cell cycle and cytotoxic effects and whether that form of cell stress determines these effects. A possible mechanism for agent-selective toxicity of trapped amines has been proposed previously (Firestone et al., 1979): the more lipophilic molecules may become detergents upon protonation and concentration in acidic vacuoles, rupturing them and killing the cells rapidly (necrosis). On the other hand, mitochondria are also a site of ion trapping for basic drugs (Siebert et al., 2004; a V-ATPase-independent process), and mitochondrial dysfunction may mediate the cytotoxic response to some of them (Irwin et al., 2002; Enomoto et al., 2004; Menor et al., 2004; Montiel-Duarte et al., 2004). Finally, toxicity, presumably independent from ion trapping, can be produced by some N-substituted 4-aminobenzamides that cause poorly repaired DNA strand breaks (Lybak and Pero, 1991).
Primary cultures of rabbit smooth muscle cells, a previously exploited primary cell model (Morissette et al., 2004a), have been used to address the cellular effects of procainamide analogs. In addition, the studies were extended to two tumor-derived cell lines to assess the effect of a transformed state on the response to drugs. It was found that massive cell vacuolization and mitotic arrest are predictable and relatively nontoxic effects of 4-aminobenzamides in the millimolar range, but that agent-specific cell death was also recorded via additional mechanisms.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
was purchased from R&D Systems (Minneapolis, MN). The remaining drugs were obtained from Sigma-Aldrich (St. Louis, MO).
Cells and Transfection. Several primary cultures of rabbit pulmonary artery were initiated and maintained as described previously (Morissette et al., 2004a). The red fluorescent protein HcRed (transiently transfected as the pHcRed1-N1 vector; BD Biosciences Clontech, Palo Alto, CA), distributed in all cytosolic water, has also been used as a visualization aid in smooth muscle cells, as described previously (Morissette et al., 2004a). The rat Morris 7777 hepatocarcinoma cell line, subclone 7.6 (Guertin et al., 1988; Morissette et al., 2004b), was a gift from Prof. L. Bélanger, [Centre Hospitalier Universitaire de Québec (CHUQ), Québec, QC, Canada]. This cell line is tumorigenic in syngeneic Buffalo rats and secretes the hepatoma marker rat -fetoprotein. The human fibrosarcoma-derived HT-1080 cell line was a gift from Dr. Eric Petitclerc (Centre de Recherche Hôpital Saint-François d'Assise, Québec, QC, Canada); it is tumorigenic in the chorioallantoic membrane assay (Xu et al., 2001). Both Morris 7777 and HT-1080 were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, fresh glutamine, and antibiotics. Both sublines of Morris 7777 and HT-1080 cells that express green fluorescent protein (GFP) were obtained by transfecting cells with the pEGFP-N3 vector (BD Biosciences Clontech) using the X-Gen 500 transfection reagent (MBI Fermentas Inc., Flamborough, ON, Canada) as directed, and selecting stable transfectants after growing the cells for one month in the presence of geneticin (500 µg/ml; Invitrogen, Carlsbad, CA).
Other Cell-Based Assays. A proliferation assay for smooth muscle cells has been applied as described previously (Morissette et al., 2004a). Poly(ADP-ribose)polymerase I (PARP) is a caspase substrate cleaved in apoptotic cells; immunoblot detection of PARP (116 kDa) and cleaved PARP (89 kDa) was performed in extracts of cells as described previously (Morissette et al., 2004b; based on a polyclonal anti-PARP antibody from Cell Signaling Technologies, Inc.). Immunoblots of -fetoproteins (based on anti-rat -fetoprotein polyclonal antibodies, a gift from Prof. L. Bélanger, CHUQ, Québec, QC, Canada) were performed as described previously (Morissette et al., 2004b) and applied to supernatant (secreted) -fetoprotein (10 µl/lane) or total cell extracts (25 µg of protein/lane).
Confocal and Epifluorescence Microscopy. Mitochondrial staining was applied as 0.2 µM MitoTracker Red CMXRos (Invitrogen) added to the culture medium 30 min before microscopic observation. The Golgi network has been visualized using C5-ceramide-BODIPY FL (Invitrogen) as described previously (Morissette et al., 2004a) by indirect immunofluorescence applied to fixed and permeabilized cells [anti-p230 primary monoclonal antibody; dilution 1:100 (BD Biosciences); secondary antibody goat-anti-mouse IgG-Alexa 488 conjugate, 1:400 (Invitrogen); epifluorescence] or using GFP-golgin-97 [coded by vector GRIP12 (Munro and Nichols, 1999); gift from Dr. Sean Munro, Medical Research Council Laboratory of Molecular Biology, Cambridge, UK]. The endoplasmic reticulum (ER) has been also investigated using an ER-targeted RFP vector coding for a COOH-terminal amino acid sequence SEKDEL ligated downstream to the leader sequence of the surface molecule CD5 (Klee and Pimentel-Muiños, 2005; gift from Dr. Felipe X. Pimentel-Muiños, Centro de Investigación del Cáncer, Universidad de Salamanca-CSIS, Salamanca, Spain). The subcellular fluorescence distribution in live cells was observed without fixation using a Bio-Rad (Hercules, CA) 1024 laser beam confocal microscope (40x objective with oil immersion; HcRed and MitoTracker: emission, 568 nm; detection, above 585 nm; GFP and C5-ceramide-BODIPY FL: emission, 488 nm; detection, above 510 nm) or epifluorescence microscopy. The vacuolization of fluorescent cells (expressing either HcRed or GFP proteins that are excluded from the vacuoles that appear as dark disks) was quantified as the percentage of individual cell pixels below a set threshold, after increasing the contrast of the picture in a standardized manner for each cell line (Photoshop; Adobe Systems, Mountain View, CA). These numerical values were averaged and compared using nonparametric statistical tests. Three-dimensional reconstitutions of cells from series of confocal images were made using the FreeSFP software (http://www.svi.nl).
Cytotoxicity Assay. Subconfluent smooth muscle cells plated in 35-mm Petri dishes were maintained in their regular culture medium; test drugs or combinations were added for 24 h. At the end, the medium was removed and centrifuged at 1000 rpm for 5 min. The pellet containing some detached cells was resuspended in the following reaction medium and reintroduced into the Petri dishes. This medium was composed of 1 ml of fresh fluorescein diacetate 12.5 µg/ml (Sigma-Aldrich) in phosphate-buffered saline diluted from a 5 mg/ml stock in filtered acetone. The cells were incubated for 30 min at 37°C in the dark. Propidium iodide (100 µl of 0.5 mg/ml) was then added, the dishes were further incubated for 3 to 5 min at room temperature and then observed (epifluorescence 100x) without rinsing, using two sets of excitation-emission filters (the red fluorescence indicates propidium iodide uptake by nonviable cells, and the green fluorescence shows the hydrolysis of the fluorescein ester by viable cells). The proportion of nonviable cells (adherent or not) was established by counting. A modification of the same assay was performed on amine-treated, trypsin-EDTA detached cells for the Morris and HT-1080 lines; the green and red fluorescence were quantified using the Elite ESP cytofluorometry apparatus (Beckman Coulter, Fullerton, CA); results were analyzed using the Expo software, version 2.0.
Cell Detachment Assay. A simple assay was applied to tumor-derived cell lines to obtain insight into cell damage. 2.5 x 105 cells were seeded in 35-mm Petri dishes at time -48 h, the fetal bovine serum-containing culture medium (2 ml/dish) was replaced and supplemented by an amine drug at time -24 h, and the detached cells were manually counted at time 0.
Assays for NF-B Nuclear Translocation and a Gene under the Control of NF-B. Translocation of NF- B p65 subunit from the cytosol to the nucleus is studied by indirect immunofluorescence as described previously (Sabourin et al., 2002); the staining intensity of each manually outlined nucleus in the photographic record was quantified as the median pixel intensity (0–255 scale; Photoshop). These numerical values were averaged and compared using nonparametric statistical tests.
The binding of [3H]Lys-des-Arg9-bradykinin to intact, adherent smooth muscle cells was used to evaluate the surface expression of the kinin B1 receptor, a highly regulated protein in immunostimulated cells, as described previously (Sabourin et al., 2002).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). Triethylamine and 4-aminobenzamides that can be considered substituted triethylamine (procainamide, NAPA, metoclopramide, NAMA; Fig. 1) all induced the formation of numerous and relatively large dark vacuoles over a 2-h period (Fig. 1). The 0.5 mM level of each drug was inactive, based on cell pixel analysis (Fig. 1 and data not shown), and the threshold for activity was 2.5 mM, except for NAPA (5 mM). For 24-h drug treatments, the 2.5 mM level was active for all these drugs including NAPA, and a more constant vacuole size was obtained (Fig. 1). The intensity of the vacuolization after 4-h treatments with amines was essentially identical to that of 24-h treatments (see control values in subsequent experiments). The compound p-aminohippuric acid is a 4-aminobenzamide that is devoid of the tertiary amine function in its side chain (Fig. 1); this acidic analog of procainamide is inactive to induce cell vacuolization (Fig. 1). Morphologic Analysis of Responses to Amines in Smooth Muscle Cells and Their Inhibition by Bafilomycin A1. Figures 2 and 3 depict the morphological responses to concentrated amine drugs in smooth muscle cells, a large cell type well adapted for this purpose. A three-dimensional reconstitution of smooth muscle cells treated or not with the low-toxicity amine procainamide is shown in Fig. 2 (left column). HcRed filled the space in a control cell, but procainamide (2.5 mM, 4 h) induced the formation of apparently rigid spherical vacuoles that excluded the fluorescent protein and imprinted on the nucleus.
|
|
), the vacuolization induced by this amine, triethylamine, or metoclopramide (2.5 mM, 4 h) was largely prevented by concomitant treatment with the V-ATPase inhibitor bafilomycin A1 (120 nM; Fig. 3; statistical analysis of numerized cell pictures, Fig. 4). At 4 h of observation time, many cells exposed to metoclopramide exhibited morphological signs of toxicity, such as rounding (Fig. 3; this applied to 24-h treatments and to the drug NAMA as well; Fig. 1); bafilomycin variably reduced those changes (Fig. 3). The identity of the amine-induced giant vacuoles has been further investigated in cells treated with the Golgi marker C5-ceramide. Ribbons clustered adjacent to the nucleus were labeled with this fluorescent compound in control cells; specific staining of the giant vacuole membranes is observed in cells treated with triethylamine, procainamide, or metoclopramide (Fig. 3). The relationship of the giant vacuoles with the Golgi network has been further studied using two trans-Golgi markers, p230 (indirect immunofluorescence, Fig. 3) and golgin-97 (GFP fusion protein, epifluorescence, Fig. 2, right column) in smooth muscle cells. The ribbon-like staining for either protein in control cells was transformed by a 4-h treatment with triethylamine, procainamide, or metoclopramide into a pattern in which a fraction of contiguous giant vacuoles was labeled (microscopy data not shown for triethylamine and metoclopramide with golgin-97). Amine-induced vacuolization did not arise from mitochondria, in that the vacuoles are not continuously lined with the fluorophore MitoTracker, but mitochondrial morphology may be affected by amines: triethylamine dilated some, and procainamide or metoclopramide rather condensed them (better seen in the periphery of the cells or in close-up views, Fig. 3). ER-targeted RFP exhibited a granular cytosolic distribution that also remained independent from procainamide-induced giant vacuoles (Fig. 2, right column; this also applied to cells treated with triethylamine or metoclopramide, data not shown).
|
Effects of Amine Drugs on Smooth Muscle Cell Proliferation and Viability. The inhibition of proliferation is a typical response to procainamide-related drugs (see Introduction). A study conducted over 48 h showed a complete inhibition of proliferation applicable to all amine drugs applied at 2.5 mM, including triethylamine (Fig. 5A, as previously observed for procainamide, Morissette et al., 2004a). The number of recovered cells was inferior to the number of seeded cells (5 x 104) for some of the drug treatments (NAPA, metoclopramide, NAMA). The V-ATPase inhibitor did not prevent the mitotic arrest induced by the noncytotoxic amines procainamide or triethylamine (2.5 mM, Fig. 5B); however, bafilomycin A1 itself produced a complete inhibition of proliferation.
|
), determined a cytotoxic response (24 h, sample photographic record, Fig. 6A; statistical analysis, Fig. 6B). Bafilomycin A1 produced a small cytotoxicity that became synergistic with that caused by actinomycin D when the two drugs were combined. The 2.5 mM concentration level of triethylamine, procainamide, or metoclopramide exerted small cytotoxic effects of increasing intensity in this order over 24 h. At a higher level (5 mM), triethylamine and procainamide exhibited more frequent cell death, and this was partially prevented by bafilomycin A1 cotreatment (Fig. 6B). Metoclopramide (5 mM) exerted an even more frequent cytotoxicity that was not relieved by bafilomycin cotreatment.
|
Effects of Amine Drugs on NF-B and an NF-B-Regulated Gene in Smooth Muscle Cell. In serum-starved smooth muscle cells, the p65 subunit of NF-B had a mainly cytosolic localization (Fig. 7A). Interleukin-1 treatment (1 h) strongly concentrated p65 into the nuclei in most cells (Fig. 7A; statistics on median nuclear pixel intensity in Fig. 7B). At 2.5 mM, agents that induce vacuolization produced no effect on the baseline nuclear fluorescence (triethylamine, procainamide, metoclopramide) or a slight significant inhibition (NAPA) or stimulation of this baseline (NAMA) (Fig. 7, A and B; the sample photographic record also shows p65-excluding cytosolic vacuoles). The amine drugs (applied 2 h before staining) exerted variable interactions with the effect of interleukin-1 (applied 1 h before staining): triethylamine and procainamide did not change the strong activating effect of the cytokine, but NAPA, metoclopramide, and its amide NAMA significantly reduced it (Fig. 7B). Although bafilomycin A1 significantly increased the p65 nuclear content in interleukin-1-cotreated smooth muscle cells, metoclopramide was still effective enough to significantly reduce the nuclear staining intensity when combined with bafilomycin (Fig. 7C). Thus, this effect is not V-ATPase dependent; rather, it is consistent with the reported inhibitory effect of highly substituted 4-aminobenzamides on IB kinase (see Introduction). The kinin B1 receptor is expressed under the control of NF-B in rabbit arterial smooth muscle cells (Sabourin et al., 2002). A binding assay based on a saturating concentration of the agonist radioligand [3H]Lys-des-Arg9-bradykinin (1 nM) showed that a 4-h treatment of the smooth muscle cells with procainamide or metoclopramide (2.5 mM) failed to up-regulate the B1 receptors. On the contrary, metoclopramide significantly decreased the baseline expression of the cell surface binding sites. The documented stimulatory effect of interleukin-1 on receptor abundance is reduced by concomitant treatment with either amine (Fig. 7D). Bafilomycin A1 treatment of cells did not inhibit the interleukin-1 stimulation of B1 receptor expression, but partially and significantly reversed the inhibitory effect of procainamide on this process, not that of metoclopramide (Fig. 7D).
|
The investigations have been extended to the tumor-derived Morris hepatoma and HT-1080 fibrosarcoma cells expressing GFP to probe whether a transformed phenotype sensitizes the cells to cytotoxicity induced by 4-aminobenzamides.
Effect of Amines on Morris 7777 Cells. As for the smooth muscle cell, triethylamine, procainamide, NAPA, metoclopramide, and NAMA induced the vacuolization of Morris cells in 2 h (Fig. 8). The threshold active concentration was 2.5 mM (except for triethylamine at 5 mM). However, a 4-h treatment with 2.5 mM triethylamine was significantly active (Fig. 9A, comparatively less than the same amine at 15 mM or than metoclopramide or procainamide at 2.5 mM). Figure 9A also shows that the vacuolization response was prevented by concurrent bafilomycin A1 treatment in Morris cells. Independently of vacuolization, metoclopramide frequently induced "blebbing" and rounding in Morris cells (Figs. 8 and 9). As in the smooth muscle cells, MitoTracker did not label the giant vacuoles induced by triethylamine (15 mM) or metoclopramide (2.5 mM) that might, nevertheless, modify the appearance of mitochondria in Morris cells (Fig. 9B).
|
|
The hepatoma-derived Morris cell line has the capacity to secrete large quantities of -fetoprotein (Morissette et al., 2004b), which is an opportunity to test the effect of amines that induce vacuolization on the secretory pathway (Fig. 10). In both the cell extract and supernatant of control Morris cells, mature 
70 kDa -fetoprotein is abundant (Fig. 10A). The amines triethylamine, procainamide, and metoclopramide at 5 mM reduced the supernatant protein rather than the intracellular -fetoprotein contents, as if the secretion were impaired. This effect compares well with that of brefeldin A, a known Golgi-disrupting agent (Doms et al., 1989), but not with that the protein synthesis inhibitor anisomycin or the glycosylation inhibitor tunicamycin, which reduced both intra- and extracellular -fetoprotein (and lowered the apparent molecular weight of the glycoprotein for tunicamycin). A detailed concentration-effect study for triethylamine and procainamide showed that these agents inhibit -fetoprotein secretion precisely at concentrations that produce vacuolization (compare Fig. 10B with Fig. 8). Concurrent treatment of the cells with bafilomycin A1 partially restored the secretion of -fetoprotein in cells treated with the amines triethylamine or procainamide but not those treated with brefeldin (Fig. 10C), also supporting the idea that the vacuolization induced by the former agents is associated with the impaired secretion process.
|
|
PARP cleavage occurs more intensely in Morris cells treated with metoclopramide (2.5 mM, 24 h) than in those treated with triethylamine, procainamide, or NAPA (Fig. 12, top). Bafilomycin A1 (120 nM) is not active in this respect. Actinomycin D was used as a positive control for apoptosis induction. Bafilomycin A1 cotreatment did not prevent the effect of metoclopramide on PARP cleavage (Fig. 12).
|
|
|
The special sensitivity of HT-1080 cells to metoclopramide was further documented in a simple detachment assay that integrates various forms of cell damage (Fig. 15). Relative to the Morris cells, for which metoclopramide is only 1.3-fold more active than procainamide in terms of cell detachment, metoclopramide is 5.1-fold more potent than procainamide in HT-1080 cells.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
) or than those binding G protein-coupled receptors in the gastroenterology applications of metoclopramide. Thus, the present results may be of toxicological interest. However, experimental oncology applications and anti-inflammatory properties of 4-aminobenzamides rely on drug concentrations 500 µM (Pero et al., 1999; Olsson et al., 2002; Villar-Garea et al., 2003), a range at which the vacuolization response to organic amines apply to many cell types (Henics and Wheatley, 1997; Morissette et al., 2004a; current study). Whether specific transport or extrusion modulate the accessibility of the amine drugs to intracellular targets is not excluded, but the protonated tertiary amines are in equilibrium with an uncharged form for which the simple diffusion into cells is likely to be effective. According to the literature, procainamide-related 4-aminobenzamide drugs (Fig. 1) exhibit pKa values between 9 and 10, and that of triethylamine is estimated at 10.7 to 11.0.
The 4-aminobenzamides induce massive smooth muscle cell vacuolization in V-ATPase-dependent manner (Fig. 3). The structural basis of this effect is the presence of triethylamine, because this amine and the series of 4-aminobenzamides that includes triethylamine (Fig. 1) are nearly equipotent to induce trans-Golgi swelling in smooth muscle cells, based on the formation of giant vacuoles that exclude cytosolic proteins such as HcRed and whose membranes are labeled with C5-ceramide (Fig. 3), a recognized marker of the Golgi network and downstream secretory pathway (Pagano et al., 1991; Ktistakis et al., 1995). Both golgin-97 and p230 are peripheral membrane proteins associated with the cytosolic face of the trans-Golgi network (Kjer-Nielsen et al., 1999; Munro and Nichols, 1999); the corresponding cell labeling in smooth muscle cells concerns a minority of contiguous giant vacuoles in amine-treated cells (Figs. 2 and 3). This may be caused by the distribution of these proteins to a subcompartment of the trans-Golgi network, or by the vacuolization of other organelles that express V-ATPase further down the secretory pathway. Other laboratories have found evidence of trans-Golgi swelling when cells were treated with various amines, such as chloroquine and Tris (Peterlik and Kerjaschki, 1979; Back and Soinila, 1996). However, the bafilomycin-sensitive vacuolization of HeLa cells induced by hydroxychloroquine (a substituted triethylamine) was attributed rather to the swelling of lysosomes (Boya et al., 2003), an interpretation that may be partly applicable to smooth muscle cells, because lysosomes are trans-Golgi–derived organelles that express V-ATPase. In Morris 7777 cells that abundantly secrete -fetoprotein, the amines triethylamine, procainamide, and metoclopramide inhibited the secretion of the protein but not its synthesis, as judged by the intracellular contents (Fig. 10), consistent with an impairment of the secretory pathway distal to the ER. This effect is similar to the Golgi-disrupting drug brefeldin A, but the antisecretory effect of amines is partially reversed by V-ATPase inhibition, unlike that of brefeldin, supporting the ideas that the formation of the giant vacuoles impairs the trafficking in the secretory pathway and that this effect is more important than that of V-ATPase inhibition.
Mitotic arrest is also a response common to triethylamine and 4-aminobenzamides substituted with triethylamine. The concentration-effect relationship for procainamide-induced mitotic arrest in smooth muscle cell is similar to that of procainamide-induced vacuolization (Morissette et al., 2004a). Bafilomycin A1 itself induced a complete mitotic arrest in smooth muscle cells (Fig. 4) and in several tumor-derived cell lines in a recent study (IC50 well under 100 nM; McSheehy et al., 2003). The latter report documents that this effect was not apoptotic and involved a block in S and G2/M phases. An undetermined step of cell division may require the acidification of a cell compartment by the V-ATPase, and the concentrated amine drugs may reproduce the effect of bafilomycin by directly buffering acidity in this compartment. Whether direct DNA binding explains the nonapototic mitotic arrest induced by procainamide-related drugs (Villar-Garea et al., 2003) is unclear, because the response applies to the simple amine triethylamine or to the unrelated molecule bafilomycin A1. It has been reported that the inhibition of the spontaneous mitosis-associated Golgi fragmentation in NRK cells causes a G2 phase arrest (Sütterlin et al., 2002). Another line of explanation for the mitotic arrest involves iron transport restriction in cells that lack the V-ATPase function (Saurin et al., 1996).
The very high concentration of 5 mM triethylamine or procainamide in smooth muscle cells caused some cytotoxicity that was reversed by bafilomycin A1 (Fig. 6), and this also applied to Morris (procainamide, 5 mM, Fig. 11) and HT-1080 cells (procainamide, 2.5 mM, Fig. 14). This suggests that the concentration of simple amines into organelles possessing the V-ATPase leads to catastrophic vacuolar rupture (as postulated by Firestone et al., 1979). The presence of HcRed within the giant vacuoles induced by at least one of the toxic amines, NAMA (2.5 mM), may support this mechanism (Fig. 1). A series of long-chain N-alkylated imino sugars (tertiary amines) are toxic to cells essentially by disrupting membranes in a manner highly dependent on the alkyl chain length (Mellor et al., 2003). These agents, as well as the some of the molecules in the present series (Fig. 1), may be concentrated in acidic cell vacuoles by V-ATPase-driven ion trapping and may precipitate their rupture.
Confirming previous reports based on leukemia-derived cell lines (Liberg et al., 1999; Olsson et al., 2002), we found that the toxic compound metoclopramide leads to apoptotic cell death in two tumor-derived cell lines (Fig. 12, PARP cleavage assay, a distal apoptotic event). Halogen substitution of the drug aromatic cycle may be necessary for this effect, because it is not produced by procainamide or NAPA (Fig. 12) but is reportedly shared with declopramide (Olsson et al., 2002). Primary smooth muscle cells were rather refractory to the cytotoxic action of aminobenzamides, supporting a selectivity for cell types, but metoclopramide had some activity at 5 mM. Furthermore, metoclopramide-induced cytotoxicity is not importantly relieved by bafilomycin in any of the three tested cell types (Figs. 6, 11, and 14). Metoclopramide is known to accumulate in mitochondrial cell fraction after perfusion in the rat lung (Yoshida et al., 1987) and may accordingly alter the morphology of mitochondria in microscopic observations (Figs. 3, 9B, and 13B). The apoptotic response to metoclopramide or declopramide previously documented in other cell types may derive from mitochondrial alteration, as shown by its reversal by Bcl2 overexpression and by the cytochrome c release from the mitochondria to the cytosol (Olsson et al., 2002). In addition, metoclopramide reportedly causes DNA strand breaks and inhibits DNA repair in vitro (Olsson et al., 1995), a possible basis for agent-specific toxicity. After mild damage to DNA, PARP activates the repair machinery, but after severe damage, the enzyme activity of PARP may promote necrosis/apoptosis by depleting NAD+ and ATP and other mechanisms (Jagtap and Szabó, 2005).
We confirm that the inhibition of NF-B nuclear translocation has its own structural basis, being shared by the most highly substituted 4-aminobenzamides (NAPA, metoclopramide, and NAMA) but not procainamide (Fig. 7B, Introduction). The inhibitory effect of metoclopramide on NF-B signaling is not dependent on V-ATPase (Fig. 7C) and may be dependent on a direct effect of the drug on IB kinase isoforms, as described previously (Liberg et al., 1999; Lindgren et al., 2001, 2003). Benzamide analogs have anti-inflammatory properties and reduce the secretion of pro-inflammatory cytokines such as tumor necrosis factor , interleukin-1, interleukin-6, and interleukin-8 (Pero et al., 1999). The study of the expression of the kinin B1 receptor, a highly regulated process dependent on NF-B in rabbit smooth muscle cells (Sabourin et al., 2002), shows that metoclopramide inhibits the process as expected (Fig. 7D). However, procainamide, which exerts no effect on NF-B (Fig. 7B), can also reduce the membrane B1 receptor density at a concentration that promotes cell vacuolization (Fig. 7D). This decrease was partly reversed by bafilomycin A1 cotreatment, unlike that produced by metoclopramide. Similar to -fetoprotein secretion inhibition in the Morris 7777 cells (Fig. 10), procainamide-induced vacuolization may impair the trafficking of the B1 receptor in the secretory pathway independently of NF-B. This nonspecific antisecretory action may apply to some cytokines in a previously cited study (Pero et al., 1999). NF-B activation can also protect cells from oxidative injury (Chen and Cedarbaum, 1997) and induce the transcriptional activation of multiple genes that suppress apoptosis (Baldwin, 2001). The inhibition of NF-B is one plausible basis for the observed radiosensitization and cytotoxicity induced by metoclopramide in tumor cells (Olsson et al., 1997; Pero 1999).
A full understanding of the pharmacological and toxicological actions of 4-aminobenzamide is necessary before their eventual use as antineoplastic agents (Olsson et al., 1995, 2002; Villar-Garea et al., 2003). Although some effects of 4-aminobenzamides are independent of V-ATPase-driven ion trapping (IB inhibition, agent-specific cytotoxicity, PARP cleavage), other effects are dependent on this phenomenon (vacuolization, inhibition of secretion, a small component of the cytotoxicity, possibly mitotic arrest after the buffering of acidic organelle contents). All amine drugs in this series are almost equally capable of producing massive vacuolization in several cell types, a reversible but not necessarily innocuous response that was apparently overlooked in previous studies.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org.
ABBREVIATIONS: V-ATPase, vacuolar ATPase; NAPA, N-acetyl-procainamide; NAMA, N-acetyl-metoclopramide; NF-B, nuclear factor-B; CHUQ, Centre Hospitalier Universitaire de Québec; GFP, green fluorescent protein; PARP, poly(ADP-ribose)polymerase I; ER, endoplasmic reticulum; RFP, red fluorescent protein.
The online version of this article (available at http://molpharm.aspetjournals.org) contains supplemental material. 
Address correspondence to: Dr. François Marceau, Centre Hospitalier Universitaire de Québec, Centre de Recherche en Rhumatologie et Immunologie, 2705 Laurier Blvd., Québec, QC, Canada G1V 4G2. E-mail: francois.marceau{at}crchul.ulaval.ca
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Baldwin AS (2001) Control of oncogenesis and cancer therapy resistance by the transcription factor NF-B. J Clin Investig 107: 241-246.[CrossRef][Medline]
Boya P, Gonzales-Polo RA, Poncet D, Andreau K, La Vieira H Roumier T, Perfettini JL, and Kroemer G (2003) Mitochondrial membrane permeabilization is a critical step of lysosome-initiated apoptosis induced by hydroxychloroquine. Oncogene 22: 3927-3936.[CrossRef][Medline]
Chen Q and Cederbaum AI (1997) Menadione cytotoxicity to HepG2 cells and protection by activation of nuclear factor-B. Mol Pharmacol 52: 648-657.
Doms RW, Russ G, and Yewdell JW (1989) Brefeldin A redistributes resident and itinerant Golgi proteins to the endoplasmic reticulum. J Cell Biol 109: 61-72.
Enomoto R, Komai T, Yoshida Y, Sugahara C, Kawaguchi E, Okazaki K, Kinoshita H, Komatsu H, Konishi Y, and Lee E (2004) Terfenadine induces thymocyte apoptosis via mitochondrial pathway. Eur J Pharmacol 496: 11-21.[CrossRef][Medline]
Firestone RA, Pisano JM, and Bonney RJ (1979) Lysosomotropic agents. 1. Synthesis and cytotoxic action of lysosomotropic detergents. J Med Chem 22: 1130-1133.[CrossRef][Medline]
Guertin M, LaRue H, Bernier D, Wrange O, Chevrette M, Gingras MC, and Bélanger L (1988) Enhancer and promoter elements directing activation and glucocorticoid repression of the 1-fetoprotein gene in hepatocytes. Mol Cell Biol 8: 1398-1407.
Henics T and Wheatley DN (1997) Vacuolar cytoplasmic phase separation in cultured mammalian cells involves the microfilament network and reduces motional properties of intracellular water. Int J Pathol 78: 343-354.
Irwin W, Fontaine E, Agnolucci L, Penzo D, Betto R, Bortolotto S, Reggiani C, Salviati G, and Bernardi P (2002) Bupivacaine myotoxicity is mediated by mitochondria. J Biol Chem 277: 12221-12227.
Jagtap P and Szabó CS (2005) Poly(ASP-ribose)polymerase and the therapeutic effects of its inhibitors. Nat Rev Drug Discov 4: 421-440.[CrossRef][Medline]
Kjer-Nielsen L, van Vliet C, Erlich R, Toh BH, and Gleeson PA (1999) The Golgi-targeting sequence of the peripheral membrane protein p230. J Cell Sci 112: 1645-1654.[Abstract]
Klee M and Pimentel-Muiños FX (2005) Bcl-XL specifically activates Bak to induce swelling and restructuring of the endoplasmic reticulum. J Cell Biol 168: 723-734.
Ktistakis NT, Kao CY, Wang RH, and Roth MG (1995) A fluorescent lipid analogue can be used to monitor secretory activity and for isolation of mammalian secretion mutants. Mol Biol Cell 6: 135-150.[Abstract]
Liberg D, Lazarevic B, Pero RW, and Leanderson T (1999) N-Substituted benzamides inhibit NF-B activation and induce apoptosis by separate mechanisms. Br J Cancer 81: 981-988.[CrossRef][Medline]
Lindgren H, Olsson AR, Pero RW, and Leanderson T (2003) Differential usage of IB and IB in regulation of apoptosis versus gene expression. Biochem Biophys Res Commun 301: 204-211.[CrossRef][Medline]
Lindgren H, Pero RW, Ivars F, and Leanderson T (2001) N-Substituted benzamides inhibit nuclear factor-B and nuclear factor of activated T cells activity while inducing activator protein 1 activity in T lymphocytes. Mol Immunol 38: 267-277.[CrossRef][Medline]
Lybak S and Pero RW (1991) The benzamide derivative metoclopramide causes DNA damage and inhibition of DNA repair in human peripheral mononuclear leukocytes at clinically relevant doses. Carcinogenesis 12: 1613-1617.
McSheehy PMJ, Troy H, Kelland LR, Judson IR, Leach MO, and Griffiths JR (2003) Increased tumour extracellular pH induced by Bafilomycin A1 inhibits tumour growth and mitosis in vivo and alters 5-fluorouracil pharmacokinetics. Eur J Cancer 39: 532-540.
Mellor HR, Platt FM, Dwek RA, and Butters TD (2003) Membrane disruption and cytotoxicity of hydrophobic N-alkylated imino sugars is independent of the inhibition of protein and lipid glycosylation. Biochem J 374: 307-314.[CrossRef][Medline]
Menor C, Fernandez-Moreno MD, Fueyo JA, Escribano O, Olleros T, Arriaza E, Cara C, Lorusso M, Di Paola M, Roman ID, et al. (2004) Azathioprine acts upon rat hepatocyte mitochondria and stress-activated protein kinases leading to necrosis: protective role of N-acetyl-L-cysteine. J Pharmacol Exp Ther 311: 668-676.
Montiel-Duarte C, Ansorena E, Lopez-Zabalza MJ, Cenarruzabeitia E, and Iraburu MJ (2004) Role of reactive oxygen species, glutathione, and NF-B in apoptosis induced by 3,4-methylenedioxymethamphetamine ("Ecstasy") on hepatic stellate cells. Biochem Pharmacol 67: 1025-1033.[CrossRef][Medline]
Morissette G, Moreau E, C -Gaudreault R, and Marceau F (2004a) Massive cell vacuolization induced by organic amines such as procainamide. J Pharmacol Exp Ther 310: 395-406.
Morissette G, Petitclerc E, and Marceau F (2004b) Loss of function of vascular smooth muscle cells by nitric oxide-dependent and -independent interactions with tumorigenic cells. Int J Cancer 112: 830-839.[CrossRef][Medline]
Munro S and Nichols BJ (1999) The GRIP domain—a novel Golgi-targeting domain found in several coiled-coil proteins. Curr Biol 9: 377-380.[CrossRef][Medline]
Olsson AR, Hua J, Sheng Y, and Pero RW (1997) Neutral metoclopramide induces tumor cytotoxicity and sensitizes ionizing radiation of a human lung adenocarcinoma and virus-induced sarcoma in mice. Acta Oncol 36: 323-330.[Medline]
Olsson AR, Lindgren H, Pero RW, and Leanderson T (2002) Mechanism of action for N-substituted benzamide-induced apoptosis. Br J Cancer 86: 971-978.[CrossRef][Medline]
Olsson A, Sheng Y, Kjellen E, and Pero RW (1995) In vivo tumor measurement of DNA damage, DNA repair and NAD pools as indicators of radiosensitization by metoclopramide. Carcinogenesis 16: 1029-1035.
Pagano RE, Martin OC, Kang HC, and Haugland RP (1991) A novel fluorescent ceramide analogue for studying membrane traffic in animal cells: accumulation at the Golgi apparatus results in altered specific properties of the sphingolipid precursor. J Cell Biol 113: 1267-1279.
Pero RW, Axelsson B, Siemann D, Chaplin D, and Dougherty G (1999) Newly discovered anti-inflammatory properties of the benzamides and nicotinamides. Mol Cell Biochem 193: 119-125.[CrossRef][Medline]
Peterlik M and Kerjaschki D (1979) Interference of tris(hydroxymethyl)aminomethane with structure and function of Golgi membranes. Lab Investig 40: 313-316.[Medline]
Roden DM (2001) Antiarrhythmic drugs, in Goodman and Gilman's The Pharmacological Basis of Therapeutics (Hardman JG, Limbird LE, Gilman AG eds) pp. 933-970, McGraw-Hill, New York.
Sabourin T, Morissette G, Bouthillier J, Levesque L, and Marceau F (2002) Expression of kinin B1 receptor in fresh or cultured rabbit aortic smooth muscle cells: role of NF- B. Am J Physiol 283: H227-H237.
Saurin AJ, Hamlett J, Clague MJ, and Pennington SR (1996) Inhibition of mitogen-induced DNA synthesis by bafilomycin A1 in Swiss 3T3 fibroblasts. Biochem J 313: 65-70.
Siebert GA, Hung DY, Chang P, and Roberts MS (2004) Ion-trapping, microsomal binding and unbound drug distribution in the hepatic retention of basic drugs. J Pharmacol Exp Ther 308: 228-235.
Sütterlin C, Hsu P, Mallabiabarrena A, and Malhotra V (2002) Fragmentation and dispersal of the pericentriolar Golgi complex is required for entry into mitosis in mammalian cells. Cell 109: 359-369.[CrossRef][Medline]
Villar-Garea A, Fraga MF, Espada J, and Esteller M (2003) Procaine is a DNA-demethylating agent with growth-in
2 test): *, P < 0.01; **, P < 0.001.
, P < 10-4). C, the inhibitory effect of metoclopramide on interleukin-1-induced nuclear translocation of p65 is not prevented by bafilomycin A1 cotreatment. Values are the means ± S.E.M. of 49 to 176 determinations. Values in the control group were compared with those from those of treatments that included interleukin-1
, P < 10-4). D. effect of amine drugs on the expression of a NF-
