| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 34, 31871-31878, August 22, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Department of Cell Biology, Cleveland Clinic Lerner College of Medicine, Cleveland, Ohio 44195
Received for publication, February 24, 2003 , and in revised form, June 10, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The high specific activity of MMP-2 toward collagen and its pericellular localization imply that it is a key component in the degradation of basement membranes and cell migration (1, 9). Increased expression of MMP-2 in tumor cells correlates with invasiveness and metastatic potential (10–13). Tumor cells expressing oncogenic Ras proteins possess higher potential to metastasize, in part because of the up-regulation of MMP-2 expression (14–17). Ras proteins are a group of small G proteins. The most common forms of oncogenic Ras have mutations at either codon 12, 13, or 61 (18). These mutations result in a higher proportion of Ras proteins with GTP in their binding site. Introduction of oncogenic H-Ras into fibroblasts results in increased expression of MMP-2 (14, 17). Blocking Ras-dependent signaling decreased both expression of MMP-2 and invasion of Src-transformed cells, indicating a role for Ras in regulating MMP-2 expression (19).
Three genes encode the four immediate members of the Ras family: H-, N-, K(A)-, and K(B)-Ras. K(A)- and K(B)-Ras proteins arise from alternative splicing of the K-Ras mRNA (20, 21). We have generally found c-Ha-Ras protein expression to be detectable only in cells of neuronal origin, whereas c-N-Ras and both K-Ras isoforms are expressed in all tested cell lines. Recent data suggest that different Ras isoforms possess distinct cellular functions. The most compelling evidence is that both the H-Ras knockout and the N-Ras knockout mice are viable, whereas the K-Ras knockout is embryonic lethal. The loss of two K-Ras proteins was not mimicked by the Ha-, N-Ras double knockout mice, which, similar to the single knockouts, were viable (22–26). Furthermore, over-expression experiments suggest that the different Ras isoforms might have preferential targets (27, 28). We have shown in mouse fibroblasts that Raf-1 is the preferential binding partner for plasma membrane-associated c-N-Ras (29, 30). Using immortalized fibroblasts derived from the N-Ras knockout mice, we demonstrated that c-N-Ras provides a steady-state antiapoptotic function (31). The N-Ras and K-Ras knockout fibroblasts provide excellent tools to study the role the K- and N-Ras isoforms in regulating the expression of MMP-2.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antibodies—MMP-2 (Ab-3) was obtained from Oncogene. Anti-HSP90 monoclonal antibody was obtained from BD Transduction Laboratories. Anti-AKT1 antibody was obtained from Cell Signaling Technology, Inc. The polyclonal antibody against K(B)-Ras, P85, and the monoclonal antibodies against N-Ras and Ha-Ras were obtained from Santa Cruz Biotchnology, Inc.
Cell Culture—All cell lines were maintained in Dulbecco's
modified Eagle's medium plus 10% fetal calf serum. For the preparation of
conditioned medium for zymography, cells (at 
80% confluence) were washed
three times with serum-free medium. Cells were then cultured for an additional
18 h in serum-free medium before collection of the conditioned medium for MMP
assays. When using pharmacological inhibitors, cells were cultured for 24 h in
the presence of the inhibitor, washed, and then placed in serum-free medium
containing the same concentration of the specific inhibitor for an additional
18 h.
Preparation of Cell Lysates—All lysis buffers contained the
following phosphatase and protease inhibitors: 30 mM
-glycerophosphate, 5 mM p-nitrophenyl phosphate, 1
mM phosphoserine, 1 mM phosphothreonine, 0.2
mM phosphotyrosine, 100 µM sodium orthovanadate, 25
µg/ml each aprotinin, leupeptin, pepstatin A, and 1 mM
phenylmethylsulfonyl fluoride. Cells were scraped into phosphate-buffered
saline and pelleted by centrifugation (500 x g for 5 min). The
cell pellet was resuspended in p21 buffer (20 mM MOPS, 5
mM MgCl2, 0.1 mM EDTA, 200 mM
sucrose, pH 7.4) containing 1% CHAPS (United States Biochemical Corp.). The
cells were lysed for 20 min on ice and centrifuged at 13,000 x
g for 10 min to remove nuclei and cell debris. The supernatant
solution was retained for further experiments. The protein concentration was
determined by the method of Bradford
(39).
Zymography of MMP in Conditioned Medium—Conditioned, serum-free medium was prepared as described above. The medium was centrifuged at 15,000 x g to remove cellular debris. The amount of conditioned medium was normalized to the amount of protein in the corresponding parallel cell lysates (prepared as described above). Non-reducing sample buffer was added to the appropriate amount of conditioned medium and the samples were separated by SDS-PAGE using gels made with 1 mg/ml gelatin in the separating gel. The gels were electrophoresed at 25 mA until the bromphenol blue dye front reached the bottom. The gels were rinsed three times for 10 min each with 50 mM Tris, pH 7.4, and 2% Triton X-100 followed by three washes (5 min each) with 50 mM Tris, pH 7.4. The gels were then incubated overnight at 37 °C in developing buffer (50 mM Tris, pH 7.4, 0.2 M NaCl, 1% Triton X-100, 0.02% NaN3, and 5 mM CaCl2). The next day, the gels were rinsed and stained with 0.2% Coomassie blue in 50% methanol and 10% acetic acid for 1 h. Finally, the gels were destained in 20% methanol and 10% acetic acid. MMP activity was observed by the generation of a negative clear band in the blue-gelatin staining background. The gels were quantitated using a Microtek scanner and NIH Image software version 1.60b7 and normalized with respect to untreated controls.
Reverse Transcription-PCR—1 µg of total RNA from K+C, K+13, K-2B, and K-6 were used for reverse transcription with Roche TaqMan reverse transcription reagents (poly-dT primer). Reverse transcription was performed at 25 °C for 10 min, 48 °C for 30 min, and 95 °C for 5 min. PCR were performed with Roche High-Fi polymerase. The denature temperature was 94 °C, annealing temperature was 60 °C, and elongation temperature was 72 °C. The oligonucleotide primers were synthesized by Sigma. The sequences of primers were: atg gag gca cga gtg gcc tg (MMP-2 sense primer), tca gca gcc cag cca gtc tg (MMP-2 antisense primer), tcg gcg tgaa cgg att tgg ccg ta (glyceraldehyde-3-phosphate dehydrogenase sense primer), and tgg cat gga ccg tgg tca tga gtc (glyceraldehyde-3-phosphate dehydrogenase antisense primer).
Immunoblotting—Lysates containing equal amounts of protein were separated by SDS-PAGE and transferred to polyvinylidene difluoride membrane (Hybond P; Amersham Biosciences). The membrane was blocked with blocker casein in phosphate-buffered saline (Pierce Chemical Co.) containing 10% newborn calf serum (Invitrogen). The washed blots were incubated with primary antibodies (1:1000) for 2 to 3 h at room temperature or overnight at 4 °C. After washing in Tris-buffered saline/0.1% Tween 20, the blots were incubated with secondary antibodies (1:1000) for 1 h at room temperature. After washing, the signals were detected using standard enhanced chemiluminescence techniques. The signals were quantitated using a Microtek scanner and NIH Image software version 1.60b7 and normalized to untreated controls.
Luciferase Reporter Assay—1 x 105 cells per well were seeded into 12-well dishes. The MMP-2 promoter luciferase reporter and pRL-CMV were transiently introduced into fibroblasts in combination with other indicated constructs using LipofectAMINE/Plus Reagent (Invitrogen). The amount of total DNA was kept constant using the appropriate empty vectors. The luciferase activity of the MMP-2 reporter was analyzed with the dual-luciferase reporter assay system (Promega) and normalized to the activity of the Renilla reniformis luciferase. The basal level of luciferase activity was analyzed by transfection with pGL2-Basic plasmid and pRL-CMV. The relative luciferase activity is reported as the fold increase of the MMP-2 reporter luciferase activity compared with the basal level of luciferase activity.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
K-Ras Knockout Fibroblasts Fail to Secrete MMP-2—The constitutive production of MMP-2 by mouse fibroblasts allowed us to use N-Ras and K-Ras negative cell lines to examine the role of each Ras isoform in the maintenance of MMP-2 synthesis and secretion. To avoid potential artifacts arising from cell line divergence, we generated multiple immortalized cell lines from the K-Ras knockout and control MEF populations. These are represented by K-1C, K-2B, K-3, K-4, and K-6 for the K-Ras negative cell lines and K+C, K+2, K+13, and K+14 for the control cell lines. N–/– and N+/+ indicate, for these experiments, a single N-Ras negative and control immortalized fibroblast population. We verified, by immunoblotting, the presence and absence of the different Ras isoforms in this spectrum of cell lines (Fig. 2A). As expected, the K-Ras negative cell lines fail to express K(B)-Ras and K(A)-Ras, whereas the N-Ras negative cells express both K(B) and K(A)-Ras but not N-Ras. There was no significant difference in the level of c-N-Ras expressed by the control K-Ras-expressing cells and the K-Ras negative cells (expressing 9 ng of c-N-Ras per 30 µg of cell lysate). Consistent with our previous experience using mouse fibroblasts, c-Ha-Ras was not detected. In this case, a lysate generated from C3H10T1/2 cells transformed by the minimal expression of an oncogenic G12V-Ha-Ras (11A) and PC12 cell lysate were used as positive controls for Ha-Ras expression. As has been our experience with cultured cell lines, c-Ha-Ras is only detectable in cell lines of neuronal origin (with a 2-ng sensitivity).
|
The five K-Ras negative fibroblast cell lines did not possess detectable MMP-2 activity. In contrast, all the wild-type, control fibroblast cell lines expressed MMP-2 activity comparable with that of other mouse fibroblast cell lines (Fig. 2B, top). Both N-Ras knockout and wild-type, N+/+, control fibroblasts expressed levels of MMP-2 activity consistent with that observed for both the NIH3T3 and C3H10T1/2 cell lines. These data were confirmed by Western analysis of the conditioned media (Fig. 2B, bottom). These data indicate that the steady-state level of MMP-2 in mouse fibroblasts requires the function(s) of a c-K-Ras gene product rather than that of c-N-Ras. The failure to detect c-Ha-Ras protein in any of these cell lines rules out its involvement in the regulation of steady-state MMP-2 expression.
K-Ras Regulates MMP-2 Levels through a Transcriptional Mechanism—We first tested whether the K-Ras negative cells were producing MMP-2 but failing to secrete the protein into the extracellular milieu. Similar to our observations with conditioned medium, whole-cell extracts generated from the K-Ras knockout fibroblasts possessed significantly less (an 85% decrease) MMP-2 protein compared with control cells (Fig. 3A). MMP-2 mRNA, as detected by reverse transcription-PCR analysis, suggested that K-Ras regulated the steady-state transcription of the MMP-2 gene (Fig. 3B). In contrast, there was a lack of MMP-2 mRNA amplification in the K-Ras negative cell lines suggesting the absence of MMP-2 mRNA. The 519-bp glyceraldehyde-3-phosphate dehydrogenase fragment was amplified in all samples as a loading control. Similar results were obtained by Northern analysis in that we did not detect significant levels of MMP-2 mRNA in the K-Ras-negative cell lines. MMP-2 mRNA was detected in both K+/+ cell lines and the positive control, HT1080, a human fibrosarcoma transformed by an activated N-Ras allele (data not shown). We confirmed this observation using a transient, luciferase reporter assay (Fig. 3C). MMP-2 transcription levels from K-Ras knockout and wild-type fibroblasts were measured by transfection of a rat MMP-2 promoter luciferase reporter construct and pRL-CMV. The luciferase activity of the MMP-2 reporter construct was analyzed with the dual-luciferase reporter assay system (Promega) and normalized to the activity of R. reniformis luciferase. The relative luciferase activity indicates the fold increase of MMP-2 reporter luciferase activity related to the basal level luciferase activity. Only minimal reporter activity was detected in the K-Ras negative cell lines, whereas 10–30-fold higher reporter activity was detected in the K-Ras-expressing control fibroblasts. These data are consistent with the hypothesis that K-Ras function is required for the transcription of the MMP-2 gene.
|
Ras Signaling Is Required for Transcription of MMP-2—To
explore the possible role of Ras signaling in the regulation of steady-state
MMP-2 expression, a dominant-negative Ha-Ras-N17 vector was co-transfected
with the previously described Rat MMP-2 promoter luciferase reporter
construct. Ras-N-17 blocks Ras activity be tying up Ras exchange factors into
a non-productive cycle (32).
The luciferase activity of the MMP-2 reporter construct was analyzed as
described in Fig. 3C.
Expression of the dominant-negative Ha-Ras-N17 inhibited MMP-2 promoter
activity by 80% in the K-Ras-expressing control fibroblasts
(Fig. 4A). We next
used expression plasmids encoding the separate K-Ras isoforms in an attempt to
modulate MMP-2 reporter activity (Fig.
4B). Only c-K(B)-Ras restored MMP-2 reporter activity in
the transient transfection assay described in
Fig. 4A (p
< 0.05). In this instance, we used phorbol 12-myristate 13-acetate addition
as a positive control. In separate experiments, we found that phorbol ester
addition increased the basal expression of MMP-2 protein by 2-fold (data not
shown). Therefore, the level of transcriptional activation by K(B)-Ras, though
only 50% above the vector control, is consistent with the level of
transcriptional activation by the positive control. Combining both K(A)- and
K(B)-Ras encoding plasmids did not give rise to a synergistic activation of
MMP-2 luciferase activity. These results suggest that constitutive expression
of MMP-2 is controlled through the function(s) of only c-K(B)-Ras and not
c-K(A) or c-N-Ras.
|
PI-3 Kinase/AKT Signaling Pathway Is Required for the Steady-state MMP-2 Expression in Fibroblasts—Ras proteins are key molecular switches that mediate transmembrane signaling through the activation of multiple downstream pathways, which include the MAPK and PI-3 kinase pathways (33). A series of pharmacological inhibitors were used to test the potential contribution of one or more Ras-dependent signaling pathways that might regulate the steady-state transcription of the MMP-2 gene. Secretion of MMP-2 into the medium of K-Ras-expressing control fibroblasts was analyzed by immunoblotting for MMP-2 after a 48-h treatment with one of the following signaling inhibitors: PD098059 (mitogen-activated protein kinase kinase-specific inhibitor, 37 µM), SB203580 (p38-specific inhibitor, 10 µM), LY294002 (PI-3 kinase-specific inhibitor, 20 µM), genistein (tyrosine kinase inhibitor, 100 µM) or PD153035 (epidermal growth factor receptor tyrosine kinase inhibitor, 10 µM). LY294002 treatment dramatically inhibited the secretion of MMP-2 in wild-type fibroblasts (Fig. 5A). There was also a less pronounced, but reproducible, effect using the general tyrosine kinase inhibitor genistein. Others have also reported the inhibition of MMP-2 expression by genistein (34). The PI-3 kinase-dependent, constitutive expression of MMP-2 was mimicked upon examination of the protein levels in whole-cell lysates (Fig. 5B). The specific PI-3 kinase inhibitor LY294002 was effective in blocking the steady-state accumulation of MMP-2 in whole-cell lysates, again suggesting that the expression of MMP-2 is regulated through a K-Ras-dependent pathway that flows through PI-3 kinase. We used overexpression of the regulatory p85 subunit of PI-3 kinase as a dominant negative to block basal signaling through receptor-mediated tyrosine kinases. Overexpression of the p85 regulatory subunit inhibited MMP-2 promoter activity in both K-Ras-expressing control cell lines (Fig. 5C). The K+C control cell line was more susceptible to this inhibition, which is consistent with this cell line's increased sensitivity of MMP-2 promoter activity to the tyrosine kinase inhibitor, genistein (Fig. 5B). These data all support the hypothesis that constitutive MMP-2 expression is regulated through a steady-state signaling event passing through c-K(B)-Ras and PI-3 kinase.
|
The lipid products of PI-3 kinase activity result in the activation of a downstream serine kinase, AKT. Using a similar strategy, we examined the levels of MMP-2 secretion in cells that were wild-type, heterozygous, or homozygous knockouts for AKT1. In complete agreement with our previous observations and the documented relationship between PI-3 kinase lipid products and AKT activation, there was a significant decrease in constitutive MMP-2 expression in the AKT1-negative MEF populations (Fig. 6A, top). It is interesting that the decreased MMP-2 expression was similar between the AKT1+/– and the AKT1–/– MEFs, suggesting a possible threshold effect for AKT expression. We have observed a similar phenomenon in N-Ras heterozygous MEFs in their heightened response to apoptotic agents; N-Ras+/– cells responded in a manner more similar to the N–/– cells than the N+/+ cells.2 Western analysis of AKT1 verified that the AKT1+/– cells expressed AKT1 at about 50% of the level observed in the AKT1+/+ MEFS. To verify the threshold nature of the MMP-2 expression of AKT levels, we quantitated the amount of AKT in the different MEF populations (Fig. 6A, bottom). To confirm the role of AKT signaling in mediating MMP-2 expression, we used an AKT-specific inhibitor (20 µM) to block AKT-dependent gene expression in control K-Ras-expressing fibroblasts. Similar reductions in the steady-state levels of MMP-2 in whole-cell lysates were observed in the presence of either the PI-3 kinase inhibitor LY294002 or the specific AKT inhibitor (Fig. 6B). These data support the hypothesis that constitutive MMP-2 expression is regulated through the steady-state functions of c-K(B)-Ras, PI-3 kinase and AKT.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Signal transduction has traditionally been examined by short-term stimulation of cells with agents that induce mitogenesis, differentiation, or apoptosis (36–38). We have found, however, that maintenance of basal cell function also requires "steady-state" signal transduction. In this specific instance, the steady-state production of a single matrix metalloproteinase, MMP-2, is regulated through the steady-state functions of c-K(B)-Ras, PI-3 kinase, and AKT. Our previous work using N-Ras-negative cells suggested that short-term signaling pathways readily adapt to the absence of essential signal transduction components (31, 35). In contrast, we have found that cellular steady-state functions do not adapt as rapidly. Using immortalized N-Ras- and K-Ras-negative cell lines, we have identified a novel steady-state function for c-K(B)-Ras. The K-Ras-negative cell lines expressed detectable levels of only c-N-Ras. Consistent with our previous observations, we have not observed detectable levels of c-Ha-Ras in mouse fibroblasts, even as a compensatory mechanism for the loss of either the N-ras or K-ras genes. The K-Ras-negative cells failed to synthesize detectable levels of MMP-2, one of the matrix metalloproteinases involved in cell detachment, migration, and metastasis of invasive tumors. Failure to express the c-N-Ras gene product did not alter the levels of MMP-2 generation, suggesting a direct relationship between one of the K-Ras gene products and the steady-state production of MMP-2. Transient assays confirmed that only one of the K-Ras gene products, c-K(B)-Ras, is responsible for controlling the transcriptional activity of the MMP-2 gene. This observation is supported by the inhibition of basal MMP-2 production by transient expression of a dominant-negative Ras construct, N17-Ha-Ras, which functions through the sequestering of Ras exchange factors.
The identification of steady-state Ras dependent functions opens novel possibilities regarding the role of each oncogenic Ras isoform in specific cancers. The common hypothesis of the involvement of Ras in the development of tumors has always stemmed from their role in short-term signal transduction. Given the data presented in this manuscript, we could now predict that tumors containing an activating K(B)-Ras rather than N-Ras should expressed elevated levels of MMP-2. This should result in a higher metastatic potential and probably a poorer overall prognosis. Therefore, although much of the transforming potential of each Ras isoform might arise from "continuous short-term signaling," there is probably a secondary component that arises from changes in Ras isoform-dependent "steady-state" signal transduction that has been overlooked.
|
FOOTNOTES |
|---|
To whom correspondence should be addressed: 9500 Euclid Avenue, Cleveland, OH
44195; Tel.: 216-444-1228; Fax: 216-444-9404; E-mail:
wolfmaa{at}ccf.org.
1 The abbreviations used are: MMP, matrix metalloproteinase; MOPS,
4-morpholinepropanesulfonic acid; CHAPS,
3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; CMV,
cytomegalovirus; PI-3, phosphoinositide-3; MEF, mouse embryonic
fibroblast.
2 J. C. Wolfman and A. Wolfman, unpublished observations.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. P. Fotiadou, C. Takahashi, H. N. Rajabi, and M. E. Ewen Wild-Type NRas and KRas Perform Distinct Functions during Transformation Mol. Cell. Biol., October 1, 2007; 27(19): 6742 - 6755. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Liao, S. M. Planchon, J. C. Wolfman, and A. Wolfman Growth Factor-dependent AKT Activation and Cell Migration Requires the Function of c-K(B)-Ras Versus Other Cellular Ras Isoforms J. Biol. Chem., October 6, 2006; 281(40): 29730 - 29738. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Risinger Jr., T. S. Hunt, D. L. Updike, E. C. Bullen, and E. W. Howard Matrix Metalloproteinase-2 Expression by Vascular Smooth Muscle Cells Is Mediated by Both Stimulatory and Inhibitory Signals in Response to Growth Factors J. Biol. Chem., September 8, 2006; 281(36): 25915 - 25925. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Goldman and E. Shalev Difference in Progesterone-Receptor Isoforms Ratio Between Early and Late First-Trimester Human Trophoblast Is Associated with Differential Cell Invasion and Matrix Metalloproteinase 2 Expression Biol Reprod, January 1, 2006; 74(1): 13 - 22. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. F. Deroanne, D. Hamelryckx, T. T. G. Ho, C. A. Lambert, P. Catroux, C. M. Lapiere, and B. V. Nusgens Cdc42 downregulates MMP-1 expression by inhibiting the ERK1/2 pathway J. Cell Sci., March 15, 2005; 118(6): 1173 - 1183. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Singh, A. P. Sowjanya, and G. Ramakrishna The wild-type Ras: road ahead FASEB J, February 1, 2005; 19(2): 161 - 169. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ishikawa, K. Takenaka, K. Yanagihara, R. Miyahara, Y. Kawano, Y. Otake, S. Hasegawa, H. Wada, and F. Tanaka Matrix Metalloproteinase-2 Status in Stromal Fibroblasts, Not in Tumor Cells, Is a Significant Prognostic Factor in Non-Small-Cell Lung Cancer Clin. Cancer Res., October 1, 2004; 10(19): 6579 - 6585. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Elad-Sfadia, R. Haklai, E. Balan, and Y. Kloog Galectin-3 Augments K-Ras Activation and Triggers a Ras Signal That Attenuates ERK but Not Phosphoinositide 3-Kinase Activity J. Biol. Chem., August 13, 2004; 279(33): 34922 - 34930. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||
