MGSA/GRO-mediated melanocyte transformation involves induction of Ras expression

MGSA/GRO-mediated melanocyte transformation involvesinduction of Ras expression

Dingzhi Wang1, Wei Yang1, Jianguo Du1, Madhav N Devalaraja1, Peng Liang1, KenMatsumoto3, Keisuke Tsubakimoto3, Takeshi Endo3, and Ann Richmond1,2,*

1 Department of Cell Biology, Vanderbilt University School of Medicine, Nashville, Tennessee, TN 37232,USA

2 VA Medical Center, Nashville, Tennessee, TN 37232, USA

3 Department of Biology, Chiba University, Japan

AbstractThe MGSA/GRO protein is endogenously expressed in almost 70% of the melanoma cell lines andtumors, but not in normal melanocytes. We have previously demonstrated that over-expression ofhuman MGSA/GROα, β or γ in immortalized murine melanocytes (melan-a cells) enables these cellsto form tumors in SCID and nude mice. To examine the possibility that the MGSA/GRO effect onmelanocyte transformation requires expression of other genes, differential display was performed.One of the mRNA’s identified in the screen as overexpressed in MGSA/GRO transformed melan-aclones was the newly described M-Ras or R-Ras3 gene, a member of the Ras gene superfamily. Over-expression of MGSA/GRO upregulates M-Ras expression at both the mRNA and protein levels, andthis induction requires an intact glutamine-leucine-arginine (ELR)-motif in the MGSA/GRO protein.Western blot examination of Ras expression revealed that K- and N-Ras proteins are also elevatedin MGSA/GRO-expressing melan-a clones, leading to an overall increase in the amount of activatedRas. MGSA/GRO-expressing melan-a clones exhibited enhanced AP-1 activity. The effects ofMGSA/GRO on AP-1 activation could be mimicked by over-expression of wild-type M-Ras or aconstitutively activated M-Ras mutant in control melan-a cells as monitored by an AP-1-luciferasereporter, while expression of a dominant negative M-Ras blocked AP-1-luciferase activity in MGSA/GRO-transformed melan-a clones. In the in vitro transformation assay, over-expression of M-Rasmimicked the effects of MGSA/GRO by inducing cellular transformation in control melan-a cells,while over-expression of dominant negative M-Ras in MGSA/GROα-expressing melan-a-6 cellsblocked transformation. These data suggest that MGSA/GRO-mediated transformation requires Rasactivation in melanocytes.

Keywordschemokine; MGSA/GRO; Ras; AP-1; melanocytes; transformation

IntroductionMelanoma growth stimulatory activity/growth regulated protein (MGSA/GRO) is a memberof the CXC chemokine family. MGSA/GRO plays a fundamental role in recruitment andactivation of neutrophils, lymphocytes and monocytes in host defense. There are four MGSA/GRO genes. Three of them, MGSA/GROα, β and γ, encode closely related proteins (Haskill

*Correspondence: A Richmond, Department of Cell Biology, Vanderbilt School of Medicine, Nashville, Tennessee, TN 37232, USA.

NIH Public AccessAuthor ManuscriptOncogene. Author manuscript; available in PMC 2009 April 9.

Published in final edited form as:Oncogene. 2000 September 21; 19(40): 4647–4659. doi:10.1038/sj.onc.1203820.

NIH

-PA Author Manuscript

NIH


NIH


et al., 1990; Iida and Grotendorst 1990; Kurdowska et al., 1994), and the other, MGSA/GROδ, appears to be a pseudogene (Shattuck-Brandt et al., 1997). All three MGSA/GROproteins bind to the CXC chemokine receptor designated CXCR2, a member of G-proteincoupled receptor family. The order of potency for these three isoforms regarding neutrophiland basophil chemotactic activity, Ca2+ flux, respiratory burst, exocytosis, shape change, andCXCR2 receptor binding is MGSA/GROα>γ>β (Geiser et al., 1993). An ELR (glutamic acid,leucine and arginine)-motif is present near the N-terminus of all the neutrophil-activating CXCchemokines that bind to CXCR1 and CXCR2. Muta-genesis studies on MGSA/GROα haverevealed that the ELR motif is a critical domain for receptor binding and biological activity(Hesselgesser et al., 1995; Arenberg et al., 1997). The order of loss of ligand sensitivity uponmutation of this motif is R>E>L (Hébert et al., 1991).

MGSA/GROα expression is correlated with melanocyte tumor progression (Richmond andThomas, 1986; 1988; Balentien et al., 1991; Mattei et al., 1994; Shattuck et al., 1994). Ourlaboratory has demonstrated that mouse immortalized melanocytes (melan-a cells) stablytransfected with MGSA/GROα, β, or γ exhibit enhanced ability to form large colonies in softagar and tumors in nude mice (Balentien et al., 1991; Owen et al., 1997). In contrast, parentalmelan-a cells (immortalized murine melanocytes), melan-a cells expressing empty vector(V1 controls), or melan-a cells expressing ELR motif mutant forms of MGSA/GROα form few,if any, tumors in nude mice (Bennett et al., 1987; Haghnegahdar et al., 2000). Moreover,neutralizing antibodies to MGSA/GROα or γ proteins slowed or inhibited the growth of theseMGSA/GRO-induced melan-a tumors in SCID mice (Haghnegahdar et al., 2000).

The signaling pathways leading to MGSA/GRO-induced melanocyte transformation areunclear. Activation of the PI3-kinase/Ras/Raf/Soc/MEK/ERK pathway is common for Gprotein-coupled receptors (Hawes et al., 1996; Della Rocca et al., 1997; Lopez-llasaca et al.,1997). Stimulation of CXCR2 with IL-8 revealed that activation of the PI3-kinase/Ras/Rafpathway is required for human neutrophil migration. However, the regulation of cell migrationby IL-8 is independent of ERK activation (Knall et al., 1997), but is dependent upon Gβγactivation (Neptune and Bourne, 1997). For some cell types, MGSA/GRO activation of ERKcan also be through a Ras/Raf1 independent pathway (Shyamala and Khoja, 1998; Yang etal., 1999). In some cell systems, activation of Ras is accompanied by enhanced expression ofc-Fos and enhanced formation of the activation protein-1 (AP-1) complex, and these are crucialin cellular transformation (Johnson et al., 1996; van Darn et al., 1998; Saez et al., 1995).Blocking of AP-1 activity in Ras-transformed cells has been shown to suppress transformation(Lloyd et al., 1991). Enhanced AP-1 activity is frequently associated with in vivo and invitro models of skin cancer (Rutberg et al., 1997; Domann et al., 1994).

We postulated that continuous expression of MGSA/GRO in melanocytes would beaccompanied by an altered mRNA expression profile. It is known that IL-8 induces α4β7αLβ2 and αvβ2-integrin expression in leukocytes (Sadhu et al., 1998), and that MGSA/GROinduces α6 integrin expression in keratinocytes (Rennekampff et al., 1997). In an attempt tofurther explore the mechanism by which MGSA/GRO induces melanocyte transformation, weperformed differential display to identify mRNAs over-represented in MGSA/GROα- or γ-expressing melan-a cells, as compared to vector control cells. One gene identified in this screenis a member of the Ras family, called M-Ras or R-Ras3 (Matsumoto et al., 1997; Kimmelmanet al., 1997). Additional experiments revealed that expression of N- and K-Ras was alsoinduced, with a resultant marked increase in the presence of activated Ras. To furthercharacterize the M-Ras function in melanocytes, we examined the activation of thetranscription factor AP-1, which is a downstream target of Ras. Over-expression of wild-typeM-Ras or a constitutively activated M-Ras mutant in control melan-a cells (V1) increasedactivation of AP-1. Over-expression of a dominant negative M-Ras mutant inhibited MGSA/GRO-induced AP-1 activation. In the in vitro transformation assay, wild-type M-Ras over-

Wang et al.

Oncogene. Author manuscript; available in PMC 2009 April 9.

NIH


NIH


NIH


expression induced cellular transformation, while dominant negative M-Ras blocked theMGSA/GROα-induced transformation in melanocytes. Our results suggest that MGSA/GRO-induced melanocyte transformation is through a Ras dependent mechanism which leads toactivation of specific transcription complexes, and altered gene expression.

ResultsIdentification of M-Ras by differential display

The over-expression of MGSA/GRO plays a key role in melanocyte transformation (Balentienet al., 1991; Owen et al., 1997). However, changes in gene expression that occur as a result ofcontinuous expression of MGSA/GRO in melanocytes have not been characterized. Todetermine whether constitutive MGSA/GRO altered the expression of genes whose proteinproducts may be involved in melanocyte transformation, differential display was performed.Total RNA preparations extracted from the transformed MGSA/GROα-expressing Mel-a-6cells, the MGSA/GROγ-expressing γ3-14 cells, and the non-transformed melan-a cellsexpressing the neomycin selective vector alone (V1 cells) (Table 1) were used to performdifferential display using 24 different primer combinations. The 16 reamplified differentiallyexpressed cDNA fragments originally identified by the differential display analysis (seeMaterials and methods) were sub-cloned into the PCR-TRAP vector and sequenced. Sequencehomology searches on the obtained cDNA sequences indicated that three of the fragmentsmatch known sequences: M-Ras; homeodomain protein Meis (a transcription factor whichregulates cell differentiation); and extracellular matrix associated protein (sc1). Thirteen cDNAfragments did not match any genes in the database, as shown in Table 2. We focus here on M-Ras, which was expressed at a higher level in Mel-a-6 and γ3-14 cells than in the V1 cells inthe differential display assay (Table 2). The amino acid sequence of murine M-Ras shares 49and 46% identity to mouse H-Ras and R-Ras, respectively. M-Ras contains a conserved domainincluding the GTP-binding site and the consensus prenylation signal CAAX known to play animportant role in the plasma membrane localization of Ras (Matsumoto et al., 1997;Kimmelman et al., 1997).

Over-expression of MGSA/GRO up-regulates M-Ras mRNA expressionM-Ras differential expression was verified by Northern blot using the cDNA fragments of M-Ras (Figure 1a, upper panel) and by RT–PCR (Figure 1b). The probe for M-Ras is a specific200 bp-cDNA fragment that only hybridizes M-Ras mRNA, not other members of the Rassuper-family. A cyclophilin (1B15) probe was used as a loading control (Figure 1a, middlepanel). The bar graph data (lower panel) presented in Figure 1a shows the band intensitycorresponding to M-Ras after normalization to cyclophilin (1B15) mRNA. The quantitationof the Northern blot reveals that Mel-a-6 cells have a 6.5-fold higher level of expression of the1.0 kb M-Ras transcript, whereas the γ3-14 cells show a 4.5-fold higher level of expression ascompared to the V1 cells (Figure 1a, lower panel). The RT–PCR results are consistent with theNorthern blot results (Figure 1b). There is a similar level of receptor expression on the surfaceof melan-a clones, based on the observation that V1, γ3-14 and Mel-a-6 cells exhibitequivalent 125I-MGSA/GROα specific binding activity (1477, 1717, 1369 c.p.m.s per 105 cells,respectively). These data suggest that the differences in the expression of M-Ras exhibited byV1, γ3-14 and Mel-a-6 cells are not due to clonal differences in receptor expression.

In parallel, to determine whether the addition of exogenous MGSA/GROα increases M-Rasgene expression in parental melan-a cells, Northern blot analyses were performed on total RNAfrom the parental melan-a cells treated with recombinant human MGSA/GROα (25 ng/ml) for0, 4, 12 or 24 h (Figure 1c, upper panel). A cyclophilin (1B15) probe was used as an internalcontrol to monitor RNA loading (Figure 1c, middle panel). Confirming the above results, a3.0- or 4.5-fold increase in the M-Ras mRNA transcript was noticed after treatment of parental

Wang et al.


NIH


NIH


NIH


melan-a cells with MGSA/GROα for 12 or 24 h, respectively (Figure 1c, lower panel). Theseresults demonstrate that MGSA/GRO up-regulates the 1.0 kb M-Ras transcript in melan-a cells.

Over-expression of MGSA/GRO up-regulates M-Ras protein expressionTo confirm whether this MGSA/GRO-enhanced M-Ras mRNA expression results in anincrease in M-Ras protein, we examined M-Ras expression in V1, γ3-14 and Mel-a-6 cells byWestern blot using an anti-M-Ras antibody. The 24 kDa M-Ras band migrated with the controlrecombinant M-Ras protein expressed in transfected HEK293 cells (Figure 2a, lane 6). M-Rasprotein levels were significantly higher in γ3-14 and Mel-a-6 cells, compared with the V1 cells(Figure 2a lanes 2, 3 versus lane 1). In order to determine whether the higher expression of M-Ras in MGSA/GRO-expressing cells is a result of MGSA/GRO over-expression, stable melan-a clones over-expressing the ELR-motif mutant forms of MGSA/GROα (AAA/ELR) andMGSA/GROα (EAR/ELR) (E6A) were analysed for M-Ras protein expression. The level ofM-Ras protein expression in melan-a clones either expressing mutant MGSA/GROα (ELR/AAA) or mutant MGSA/GROα (ELR/ALR) protein was similar to the level in V1 cells (Figure2a lanes 4 and 5 versus lane 1) and was much lower than the level in Mel-a-6 and γ3-14 cells(Figure 2a lanes 4 and 5 versus lanes 2 and 3). ELISA data for MGSA/GROα revealed equallevels of MGSA/GROα secretion in cells expressing wild type and mutant forms of thischemokine (data not shown). A ligand-binding assay revealed equal levels of 125I-MGSA/GROα binding to melanocyte clones expressing wild type and mutant MGSA/GROα (data notshown). The failure of the MGSA/GROα ELR-mutant to increase M-Ras protein expressiondemonstrates that the elevated expression of M-Ras in Mel-a-6 or γ3-14 cells is a result ofbiological responses induced by MGSA/GRO over-expression and does not result from non-specific clonal differences in gene expression, since the MGSA/GROα ELR-mutant hasreduced affinity for CXCR2 and little biological activity.

To examine the overall expression of Ras (H, N and K-Ras), Western blots were performedusing a pan-Ras antibody (AB-3) (Oncogene Research Products, Cambridge, MA, USA) whichreacts with mammalian H, N and K-Ras proteins. Overall Ras protein levels were significantlyhigher in γ3-14 and Mel-a-6 cells, as compared to the V1 cells (Figure 2b, top panel). Rasexpression from three different clones from each vector, MGSA/GROα- and MGSA/GROγ-expressing melan-a cells was examined and similar results were obtained (data not shown).The higher Ras expression in Mel-a-6 cells is mostly contributed by K-Ras and N-Ras, whilethe higher Ras expression in γ3-14 is contributed by N-Ras. There is no significant differencein H-Ras expression between these cells (Figure 2b). To confirm that equivalent amounts ofprotein were loaded on the gel, the Western blot was probed with antibody to Bcl-2. All theextracts from these clones exhibited the same level of Bcl-2 protein expression (Figure 2b,lower panel). In addition, overall Ras protein expression levels in melan-a clones expressingELR-mutant forms of MGSA/GROα were determined by Western blot using a pan-ras antibody(AB-3). All three melan-a clones expressing ELR-motif mutants failed to show an increasedlevel of Ras protein expression, as compared to the V1 cells (Figure 2c). However, the L7Amutant form of MGSA/GROα exhibited a higher level of Ras protein expression than E6A andR8A mutant forms of MGSA/GROα, a result which is consistent with the higher affinity ofthis ligand for CXCR2 (Hebert et al., 1991). These results indicate that the ELR-motif isrequired for MGSA/GRO-enhanced Ras expression.

Enhanced Ras expression is correlated with Ras activityBased on the current knowledge of Ras effector molecules, activation-specific probes forendogenous activated Ras have recently been developed based upon the ability of activatedRas to bind Raf through a specific activated Ras binding domain on Raf (Raf RBD) (Taylorand Shalloway, 1996). The Raf RBD has a significantly lower affinity for Ras-GDP than forRas-GTP, such that only Ras-GTP is detected under the conditions of this assay (Taylor and

Wang et al.


NIH


NIH


NIH


Shalloway, 1996). This construct preferentially recognizes and precipitates GTP-loaded formsof Ras and allows determination of endogenous Ras activation in cells. To test whether over-expression of MGSA/GRO increases endogenous Ras activation in melanocytes, the level ofendogenous Ras activation in MGSA/GRO-expressing clones was examined by determiningthe amount of Ras-GTP (activated) that would bind to a GST-cRaf-1 RBD through the Rasbinding domain. The level of Ras-GTP was 10-fold higher in γ3-14 and Mel-a-6 cells, comparedwith the V1 cells (Figure 3a, lanes 3 and 4 versus lane 2). These results demonstrate that over-expression of MGSA/GRO constitutively increases a total amount of activated Ras.Furthermore, we examined whether addition of exogenous MGSA/GROα increased theamount of endogenous GTP-bound Ras in the parental melan-a cells treated with recombinanthuman MGSA/GROα (50 ng/ml) for the indicated time (Figure 3b). MGSA/GROα increasesRas activation in 5 min and lasts 24 h. In an early response (from 5 – 120 min), MGSA/GROα enhanced Ras activation without changing Ras expression (data not shown). In a lateresponse (from 6 – 24 h), MGSA/GROα-enhanced Ras activity is contributed by increasingthe level of Ras proteins (data not shown). To further determine whether over-expression ofMGSA/GRO alters the Ras exchange ratio, Ras-GTP/GDP binding and GTPase assays wereperformed. Mel-a-6 and γ3-14 cells exhibit greater Ras-GTP/GDP-binding, as compared toV1 (9.5 and 5.7-fold, respectively) (Figure 3c, left panel). The level of Ras-GTP-binding is inclose agreement with the level of Ras-GTP measured by precipitating the Ras-GTP using theRas binding domain of GST-c-Raf-1 RBD (Figure 3a). The GTP/GDP ratio is not changed inany of the clones (Figure 3c, right panel). To examine the possibility that failure to see a changein the GTP/GDP ratio in these clones was due to the long reaction time resulting in saturationof the GTP/GDP ratio, the GTP/GDP ratio over a time course was measured to follow theGTPase activity over time. No difference in the Ras-GTP/GDP ratio was observed in any ofthese three clones (data not shown). These results demonstrate that overexpression of MGSA/GRO results in an increase in the total amount of GTP and GDP bound to Ras, but does notgreatly alter the GTP/GDP exchange ratio. These data suggest that over-expression of MGSA/GRO does not change the Ras-GTPase activity per molecule, but MGSA/GRO-enhanced Rasexpression results in an increase in total constitutive Ras activation. Consistent with the resultsof Ras protein expression, continuous expression of mutant ELR-motif forms of MGSA/GROα fails to enhance Ras activation. Mutant MGSA/GROα (E6A and R8A) expressingclones have the same levels of Ras-GTP and Ras GDP as the V1 control. The mutant L7A cellsslightly increased Ras-GTP and Ras GDP, as compared to V1 (2.5-fold) (Figure 3d, left panel).The Ras-GTP/GDP ratio is not greatly changed in any clones (Figure 3d, right panel). Thesedata further confirm that the high level of total constitutive Ras activity in γ3-14 and Mel-a-6clones is a specific response to over-expression of MGSA/GROγ and α in these cells.

MGSA/GROα increases the activation of AP-1We hypothesized that MGSA/GRO uses a Ras pathway to activate AP-1, an event that resultsin the modulation of expression of genes whose encoded proteins are crucial for thistransformation event. To evaluate whether melanocytes continuously expressing MGSA/GROexhibit enhanced basal AP-1 activation, two different approaches were used: electrophoreticmobility shift assay (EMSA) and luciferase reporter transactivation assay. EMSA wasperformed to evaluate the DNA binding activity of AP-1 in V1, γ3-14 and Mel-a-6 cells usingthe consensus AP-1 DNA binding sequence as the radiolabeled probe. The results showed thatthe level of basal AP-1 DNA binding activity was higher in γ3-14 cells and Mel-a-6 cells thanV1 cells (Figure 4a). To exclude the possibility that these observations were the result of clonaldifferences, AP-1 EMSAs were performed on nuclear extracts from MGSA/GROα-expressingMel-a-4 and Mel-a-9 clones; from MGSA/GROγ-expressing γ3-12 and γ1-37 clones; and fromtwo additional vector control clones, V4 and V6 (Table 1). These results confirmed that thelevel of AP-1 DNA binding activity was elevated in the MGSA/GROγ-expressing clones(γ3-14, γ3-12 and γ1-37) and MGSA/GROα-expressing clones (Mel-a-6, Mel-a-4 and Mel-a-9)

Wang et al.


NIH


NIH


NIH


(data not shown). To determine whether addition of exogenous MGSA/GROα to parentalmelan-a cells also induces AP-1 DNA binding activity, the AP-1 DNA binding activity innuclear extracts from parental melan-a cells treated with MGSA/GROα was analysed over atime course of 10 min to 24 h. The increased AP-1 binding activity was observed after 10 minand lasted to 24 h (Figure 4b). The results show that addition of exogenous MGSA/GROαinduced the AP-1 DNA binding activity in parental melan-a cells. Super-shift EMSA usingantibodies which recognize most forms of Jun (pan-Jun Ab) or Fos (pan-Fos Ab) identifiedthe subunit composition of the MGSA/GRO-induced AP-1 complex observed by EMSA. Asshown in Figure 4c, the pan-Fos antibody super-shifted the AP-1 complex. In contrast, the pan-Jun antibody had no effect on the AP-1 complex in the EMSA. These results indicate thatMGSA/GROα induces AP-1 DNA binding activity through increased nuclear localization andactivation of Fos in melan-a cells. To examine the effect of MGSA/GRO on AP-1transactivation, cells expressing MGSA/GROα or γ were cotransfected with a luciferasereporter gene driven by the SV40 promoter containing two copies of the consensus AP-1binding site and a pSV-β-Gal expression construct. AP-1 luciferase and β-Gal activities weremeasured 48 h after transfection. The mean of the relative activity of the AP-1 luciferase afternormalization for transfection efficiency based upon β-gal activity was threefold higher in theMel-a-6 clone (P<0.0003) and 2.1-fold higher in the γ3-14 clone (P<0.03), as compared withthe V1 control (Figure 4d). These data demonstrate that over-expression of MGSA/GROαenhances AP-1 transactivation.

M-Ras mediates AP-1 activationIn order to determine whether the Ras farnesyl-transferase inhibitor blocked AP-1 activationin MGSA/GROα-expressing cells, nuclear extracts from the Mel-a-6 cells treated with solventalone or with Ras farnesyl-transferase inhibitor (SCH44342) were analysed by EMSA with theconsensus AP-1 DNA binding sequence as the probe (Figure 5a). The SCH44342 inhibitorblocked about 50% of the AP-1 binding activity at the 0.5 μM concentration (lane 3) and 90%of the AP-1 binding activity at the 10 μM concentration (lane 4). This result suggests that theMGSA/GROα-induced AP-1 activation is through a Ras-dependent pathway. To determinewhether Ras mediates MGSA/GRO-induced AP-1 activation, we examined the effect of M-Ras (w.t.), dominant active M-Ras (G22V) or dominant negative M-Ras (S27N) on MGSA/GRO-induced AP-1 transactivation. V1 cells were cotransfected with the AP-1 luciferasereporter construct and an M-Ras expression plasmid (either w.t., constitutively active, orcontrol pBK-CMV vector). Over-expression of wild-type M-Ras or a constitutively active formof M-Ras (G22V) increased basal AP-1 activity in V1 cells by 2.3- and 6.4-fold, respectively(Figure 5b, left panel). Since over-expression of wild-type M-Ras or dominant active M-Rasin melanocytes did not affect transactivation of the IP-10 CAT reporter gene (Figure 5b, rightpanel), the effects of Ras on AP-1 appear to be specific. The induction of AP-1 transactivationby M-Ras in the absence of any additional stimulus indicates that the M-Ras protein can activateAP-1 similarly to other Ras proteins including H-Ras, K-Ras or N-Ras (Johnson et al., 1996;Vojtek and Der 1998).

In a second series of experiments, dominant negative M-Ras (S27N) expression and AP-1reporter constructs were cotransfected into γ3-14 and Mel-a-6 cells. The dominant negative M-Ras (S27N) blocked the basal AP-1 activity in γ3-14 or Mel-a-6 cells by 53 and 60%,respectively (Figure 5c, left panel). Since over-expression of dominant negative M-Ras inγ3-14 or Mel-a-6 cells did not affect transactivation of the IP-10 CAT reporter gene (Figure5c, right panel), the effects of dominant negative M-Ras on AP-1 are not non-specific. Inaddition, we characterized the effect of the Ras inhibitor on the basal transcription of a minimalMGSA/GROα-350 bp promoter where the NF-kB and IUR elements have been mutated, butthe Sp-1 element is left intact (Wood and Richmond 1995). We saw no effect of co-transfectionwith the dominant negative Ras on the basal transcription of this promoter (data not shown).

Wang et al.


NIH


NIH


NIH


To further confirm whether dominant negative M-Ras can directly block MGSA/GROα-induced AP-1 activation, dominant negative M-Ras (S27N) expression constructs and AP-1reporter constructs were cotransfected into V1 cells, where the AP-1 reporter is stimulated byMGSA/GROα treatment after the transfection. The dominant negative M-Ras (S27N)completely inhibited MGSA/GROα-induced AP-1 activation, but did not affect the basal AP-1activation (Figure 5d). Taken together, these data demonstrate that Ras mediates MGSA/GRO-enhanced AP-1 activation. Since it is likely that the S27N mutant of M-Ras also blocks eventsdownstream of H-, N- and K-Ras, our results demonstrate that a generalized increase in Rasactivity (not limited to M-Ras) is essential for the enhanced AP-1 activity observed in MGSA/GROα- and γ-expressing melan-a cells.

M-Ras induces melanocyte transformationTo test the ability of M-Ras to induce melanocyte transformation, melan-a clones weretransfected with wild-type, dominant active, or dominant negative forms of M-Ras in an invitro transformation assay. V1 cells transfected with M-Ras (w.t.) formed transformed foci,while V1 cells transfected with the vector control alone (pBK-CMV) did not form foci (Figure6a). This result indicates that over-expression of M-Ras can induce melanocyte transformation.V1 cells transfected with dominant active form of M-Ras (G22V) formed four-times moretransformed foci than those transfected with wild-type Ras (Figure 6, lower-left panel). MGSA/GROα-expressing Mel-a-6 cells transfected with the dominant negative form of M-Ras (S27N)showed 85% inhibition of focus formation (Figure 6, lower-right panel). Since the dominantnegative M-Ras would potentially inhibit all forms of Ras activity, we conclude that Ras isinvolved in MGSA/GROα-induced melanocyte transformation.

To examine the effect of loss of each residue in the ELR-motif of MGSA/GROα on cellulartransformation, an in vitro transformation assay was performed. Mel-a-6 cells expressingMGSA/GROα exhibited greater formation of transformed foci, as compared to V1 cells. Incontrast, ELR-motif mutants (E6A/ELR and R8A/ELR) failed to exhibit focus formation andL7A/ELR mutant clones exhibited fewer transformed foci (Figure 6b). This result is consistentwith the observation that the mutant forms of MGSA/GROα fail to enhance Ras expression,further confirming that high Ras expression in Mel-a-6 is functionally correlated to MGSA/GRO-induced transformation.

DiscussionIn an attempt to identify MGSA/GRO-regulated genes which may be involved in MGSA/GRO-induced melanocyte transformation, using differential display we found that continuousexpression of MGSA/GRO up-regulates the expression of M-Ras at both the mRNA andprotein levels, as well as K-Ras and N-Ras expression. MGSA/GRO induction of Ras is throughthe CXCR2 receptor, since melan-a clones expressing the ELR-motif mutant forms of MGSA/GROα failed to exhibit increased Ras protein expression. Our results show that over-expressionof MGSA/GRO does not affect the specific activity of the Ras-GTPase (activity per moleculeof Ras). Rather, over-expression of MGSA/GRO increased the total amount of the activatedRas by increasing the level of Ras protein expression. These data suggest a mechanism bywhich Ras activity can be regulated at both the mRNA and protein levels, in addition to thepost-translational level of GTP binding and GTPase activity.

It is not clear how MGSA/GRO regulates the expression of M-Ras and other Ras familymembers. Recently, the Renauld group reported that Interleukin-9 induces mRNA expressionof M-Ras within 6 h after addition of IL-9, with induction maximal at 24 h. This induction waspostulated to be regulated at the transcriptional level (Louahed et al., 1999). UV-B induces 5– 10-fold higher cHa-Ras gene expression in both papillomas and carcinomas (Husain et al.,1990). Moreover, estrogen induces c-Ha-Ras expression in uterine endometrial fibroblasts and

Wang et al.


NIH


NIH


NIH


cancer cells (Fujimoto et al., 1995). Some of these reagents that induce Ras expression alsocause induction of P38 stress-associated kinase, an event which enhances mRNA stability(Winzen et al., 1999). The increase in M-Ras mRNA in the MGSA/GRO-expressing melan-aclones could be due to the alteration in the transcription and/ or enhanced stability of M-RasmRNA. Since transcription of Ras is thought to be regulated largely by Sp1 and is thereforenot inducible, we postulate that MGSA/GRO induces Ras expression by increasing the stabilityof Ras mRNA, consequent to alterations in signals such as p38 or ERK1, 2.

Human melanoma appears to belong to the group of tumors whose growth and tumorigenicitydepend on autocrine loops (Rodeck et al., 1991). MGSA/GRO has been shown to act throughboth autocrine and paracrine modes to enhance tumor progression (Owen et al., 1997; Luanet al., 1997). The deregulation of MGSA/GRO may also mimic effects of aberrant Ras functioneven in the absence of a Ras gene mutation. Thus, the influence of normal and aberrant Rasfunction on the biology of human melanoma may be even greater than expected from thefrequency of Ras gene mutations. Indeed, overexpression of receptor tyrosine kinases such asErbB2 or epidermal growth factor receptor leads to cell transformation that involves theparticipation of Ras protein expression (Smith et al., 1987; Huang et al., 1997; Kolibaba andDruker 1997).

M-Ras is closely related to R-Ras, Tc21, H-Ras, K-Ras and N-Ras. Ras family members playvery important roles in cell growth, differentiation, transformation and apoptosis (Vojtek andDer, 1998). Over-expression of activated Ras in melanocytes null for p16 induces melanomatumors (Chin et al., 1999) and over-expression of mutant activated M-Ras induces transformingfoci in NIH3T3 cells, although the ability of M-Ras to induce transforming foci is weaker thanthat of Ha-Ras (Quilliam et al., 1999). It has been reported that over-expression of any of threenormal Ras genes, N-Ras (McKay et al., 1986), H-Ras (Westaway et al., 1986; Zhang et al.,1997) or K-Ras (George et al., 1986) leads to in vitro transformation. In vivo, over-expressionof normal N-Ras is associated with development of hyperplasia and tumors in transgenic mice(Mangues et al., 1992). Here, we show for the first time that over-expression of normal M-Rasinduces melanocyte transformation in vitro. The foci that formed grew in the absence of serumor exogenous growth factors and without attachment to the substrate. These results indicatethat over-expression of normal M-Ras protein can induce melanocyte transformation, similarlyto H-Ras, K-Ras or N-Ras in other cell types. Moreover, it is conceivable that the upregulationof Ras expression is a consequence of the MGSA/GROα- and γ-induced transformation event,based upon experiments showing that expression of dominant negative M-Ras, which probablyinhibits other forms of Ras in addition to M-Ras, inhibited MGSA/GRO-induced cellulartransformation. These data suggest that Ras expression enhanced by over-expression ofMGSA/GRO is required for melanocyte transformation. However, it is possible that MGSA/GRO induction of Ras would be essential for both the MGSA/GRO induction of cell growthand the transformation process. Then dominant negative M-Ras may block cell transformationby directly interfering with the transformation process, by indirectly inhibiting cellproliferation or by both. Since it is not possible to monitor transformation in the absence ofcell proliferation, we cannot discern between these three possible mechanisms. Taken together,these data indicate that M-Ras can play a role in MGSA/GRO-induced melanocytetransformation as well as can other members of Ras. It is highly probable that over-expressionof wild-type M-Ras will also contribute to the development of malignancies in vivo.

Interpretation of our data is complicated by the observation that melan-a clones expressingMGSA/GROβ only weakly induce Ras expression and activity (M-Ras plus pan-Ras) (data notshown). There are now several reports that MGSA/GROβ binds the receptor differently andhas different biological activities from the α and γ forms of this chemokine (Geiser et al.,1993; Cao et al., 1995; Rosenkilde and Schwartz 2000). A recent report demonstrated thatMGSA/GROβ activates the sphingomyelinase/ceramide pathway (Limatola et al., 1999). Thus,

Wang et al.


NIH


NIH


NIH


the mechanism by which MGSA/GROβ affects progression of immortalized melanocytes tomelanoma may be different from that for MGSA/GROα and γ. We postulate that otherpathways may be activated in MGSA/GROβ-expressing melan-a clones to effecttransformation. For example, we found that MGSA/GROβ-expressing melan-a clones exhibitenhanced VEGF expression, while MGSA/GROα or γ clones do not (data not shown). SinceV-EGF induces endothelial cell proliferation, migration, and angiogenesis, the additive effectof V-EGF secretion must supplement the weak induction of Ras in MGSA/GROβ-expressingmelan-a clones to facilitate the transformation process.

To our knowledge, this is the first report that chemokines can up-regulate the expression ofRas and enhance AP-1 activation. Over-expression of M-Ras in melanocytes induces cellulartransformation, and continuous expression of MGSA/GRO-induced transformation can beblocked by expression of a dominant negative form of M-Ras.

Materials and methodsCell culture

The non-transformed mouse immortalized melanocyte cell line, parental melan-a (gift of DrDorthea Bennett), was cultured in DMEM supplemented with 50 units/ml penicillin, 50 μg/mlstreptomycin, 3 mM glutamine, 10% heat-inactivated fetal bovine serum (GIBCO – BRL,Rockville, MD, USA) and 200 nM TPA (Sigma, St. Louis, MO, USA). The MGSA/GROexpression plasmids were constructed by inserting human MGSA/GROα, γ or mutant MGSA/GROα into the pRC/CMV vector and stably transfecting into parental melan-a cells (Table 1).These clones were cultured in the same media supplemented with 800 μg/ml G418 (Sigma, St.Louis, MO, USA) as previously described (Owen et al., 1997). The level of chemokineexpression in these clones has been previously characterized (Owen et al., 1997). 125I-MGSA/GROα was purchased from DuPont NEN (Boston, MA, USA) at a specific activity of 272 mCi/mg.

125I-MGSA/GROα binding assayThe binding assay was conducted 48 h after the initial plating as previously described (Yanget al., 1999). Non-specific binding was monitored by determining the binding of 125I-MGSA/GROα in the presence of a 100-fold excess of cold MGSA/GROα. The specific binding activitywas calculated by subtracting non-specific binding from total 125I-MGSA/GROα bindingactivity.

Differential display—Differential display was performed using the RNA image kit(GenHunter, Nashville, TN, USA) following manufacturer’s instruction. 250 ng of total RNAfrom V1, γ3-14 or Mel-a-6 clones was reverse transcribed by using 1 μl MMLV reversetranscriptase and one of three different one-base anchored oligo-dT primers (H-T11G, H-T11Aand H-T11C). The resulting cDNA was amplified by PCR using 0.2 μl Taq DNA polymerase(Qiagen, Valencia, CA, USA) and one of three oligo-dT primers and one of eight arbitrary 13-mers (H-AP1, H-AP2, H-AP3, H-AP4, H-AP5, H-AP6, H-AP7 and H-AP8) in the presenceof [α33P]dATP (DuPont-NEN, Boston, MA, USA). The resulting products were fractionatedon a 6% polyacrylamide sequence gel that was dried and exposed to film. Sixteen differentiallyexpressed bands were recovered from sequencing gels and reamplified using the same primerset and PCR conditions, except the dNTP concentration was at 20 μM. PCR products were runon a 1.5% agarose gel and stained with ethidium bromide. The bands were recovered from theagarose gel. The cDNA fragments were either used as probes for Northern blots or cloned intothe PCR-TRAP vector using the PCR-TRAP Cloning System Kit (GenHunter, Nashville, TN,USA) for sequencing.

Wang et al.


NIH


NIH


NIH


DNA sequencing and DNA sequence homology search—DNA sequencing of cDNAfragments in the PCR-TRAP vector was performed by the dideoxychain termination methodusing the T7 sequenase 2.0 polymerase kit (Amersham Pharmacia Biotech, Piscataway, NJ,USA) with Lgh or Rgh primers that flank the cloning site of the PCR–TRAP vector. Genedatabase searches were performed at the National Center for Biotechnology Information byusing the BLAST network service.

Northern blot—20 μg of total cellular RNA, isolated from cells using the Ultraspec RNAIsolation System (Biotecx Laboratories, Inc., Houston, TX, USA), was resolved underdenaturing conditions on a 1.0% formaldehyde-agarose gel, and transferred to Nytran Plusfilters (Amersham Pharmacia Biotech, Piscataway, NJ, USA). The filter was hybridized at 42°C for 12 – 16 h with gel-purified 32P-labeled differential display cDNA fragments, which werelabeled using the Multiprime DNA labeling system (Amersham Pharmacia Biotech). Afterhybridization, the filter was washed twice with 1× saline sodium citrate with 0.1% SDS for 15min at 25°C, followed by two washes with 0.25× saline sodium citrate with 0.1% SDS for 20min at 45°C, and then exposed to film.

Whole cell extracts and Western blot—Western blots were performed followingprotocols provided by Santa Cruz Biotechnology Inc. The cells were washed at 4°C with 1×PBS and lysed in 0.25 – 0.5 ml of RIPA buffer (1× PBS, 1% NP-40, 0.5% sodium deoxycholate,0.1% SDS) with protease inhibitor cocktail tablets (Boehringer Mannheim Corp., Indianapolis,IN, USA). 50 μg soluble protein was boiled, loaded and separated on a 12% SDS-polyacrylamide gel and then electrophoretically transferred to a 0.45 μm nitrocellulosemembrane (BIO-RAD, Hercules, CA, USA). The membrane was blocked with 5% dry milkin TBS-T buffer (10 mM Tris-HCl, pH 8.0; 150 mM NaCl; 0.05% Tween-20) for 1 h and thenincubated for 12 – 16 h at 4°C in a 1:200 dilution of the anti-M-Ras antibody (gift from DrEndo), in a 1:3000 dilution of the anti-pan-Ras antibody (AB-3) (Oncogene Research Products,Cambridge, MA, USA), in a 1:200 dilution of anti-N-Ras, anti-K-Ras or anti-H-Ras (SantaCruz Biotechnology Inc) or in a 1:1000 dilution of the anti-Bcl2 (Santa Cruz BiotechnologyInc). The membrane was then incubated in a 1:3000 dilution of the anti-mouse or anti-rabbitimmunoglobulin conjugated with horseradish peroxidase (Boehringer Mannheim Corp.) inTBS-T buffer with 5% dry milk for 1 h at room temperature. The protein bands were detectedwith ECL Western blotting detection reagents (Amersham Pharmacia Biotech) according tothe manufacturer’s instructions.

Activation assays for RasActivation assays for Ras were performed as described in detail (Taylor and Shalloway,1996). Cells were washed twice with ice-cold HBS and lysed in Mg2+-containing lysis buffer(MLB: 25 mM HEPES, pH 7.5, 150 mM NaCl, 1% NP-40, 0.25% Na deoxycholate, 10%glycerol, 25 mM NaF, 10 mM MgCl2, 1mM EDTA, 1 mM, protease inhibitor cocktail tablets).After lysis for 15 min, lysates were spun at 16 000 g for 20 min, and supernatants collected.The protein concentrations of the lysates were then determined. Equal amounts of total proteinof cell lysate (400 μg) were immediately affinity-precipitated for 1 h using 20 μg of recombinantGST-c-Raf-1 RBD (1 – 149) (a gift from Dr David Shalloway) fusion proteins, which had beenfreshly coupled to glutathione-Sepharose beads (Amersham Pharmacia Biotech). Theprecipitates were washed three times with MLB and eluted by boiling in SDS –PAGE samplebuffer. The proteins were separated on a 12% SDS-polyacrylamind gel and thenimmunoblotted with pan-Ras antibody (AB-3) (Oncogene Research Products, Cambridge,MA, USA).

Wang et al.


NIH


NIH


NIH


GTP/GDP-binding and GTPase assayImmunoprecipitations were performed by using 400 μg of whole cell lysates of melan-a clonesincubated overnight at 4°C with 2 μg of anti-pan-ras antibody (AB-3). Subsequently, 30 μl ofprotein-A/G-agarose (Sigma, St. Louis, MO, USA) was added and incubated at 4°C for 3 h.The Ras immune complexes were pelleted and washed six times with RIPA buffer and oncewith GTP/GDP-binding buffer (50 mM NaCl, 50 mM Tris-HCl [pH 7.5], 5 mM EDTA, 0.1mM EGTA, 0.1 mM DTT and 10 μM ATP). The GTP/GDP-binding and GTPase assays werecarried out as previously described (Matsumoto et al., 1997). Briefly, each Ras immunecomplex bound to protein-A/G-agarose was incubated with 1 μl of [α-32P]GTP (3000 Ci/mmol,Amersham Pharmacia Biotech) in 100 μl GTP/GDP-binding buffer at 37°C for 10 min. Afterthe pellets were washed six times with wash buffer (20 nM MgCl2, 50 mM Tris-HCl [pH 7.5],1 mM DTT and 1 mg/ml bovine serum albumin), the GTPase reaction was performed inGTPase reaction buffer (50 mM NaCl, 5 mM MgCl2, 50 mM Tris-HCl [pH 7.5], 5 mM DTT)at 37°C for 30 min. The pellets were washed with wash buffer twice, and bound nucleotideswere eluted by incubating the pellets in 1% SDS and 20 mM EDTA at 68°C for 20 min. Thereleased 32P-labeled nucleotides were counted directly by Cerenkov scintillation counting(Beckman LS 3801), and then GTP and GDP were loaded on PEI-cellulose plates (JT BakerInc., Phillipsburg, NJ, USA), resolved by 0.75 M KH2PO4 and then visualized byautoradiography.

Nuclear extracts and mobility shift assay—Cells were lysed with buffer (10 mMHEPES, 10 mM sodium chloride, 1.5 mM magnesium chloride, 0.5 mM DTT, 5 mM β-mercaptoethanol, 100 μM PMSF) containing 0.5% Nonidet P-40. The nuclear proteins wereextracted from the nuclear pellets using high salt extraction (25% glycerol, 0.2 mM EDTA, 20mM HEPES, 1.5 mM magnesium chloride, 0.5 mM DTT, 10 mM potassium chloride, 240 mMsodium chloride) with protease inhibitors and phosphatase inhibitors, as described previously(Shattuck and Richmond, 1997). EMSA was performed by incubating 0.25 ng double strandedoligonucleotide end-labeled with 32P by polynucleotide kinase (New England Biolabs) andequal amounts of nuclear protein (5 μg) for 30 min at room temperature, as described previously(Shattuck and Richmond, 1997). The binding reaction mixtures were resolved on a 6%nondenaturing polyacrylamide gel. Antibody super-shift-EMSA was performed by incubatingnuclear extract proteins with antibodies (1 μg) against pan-Fos, pan-Jun, or control antibodyr-IgG (Santa Cruz Biotechnology Inc) at room temperature for 1 h before adding the labeledprobe.

Transient transfection assay—M-Ras (w.t.), constitutively active M-Ras (G22V),dominant negative M-Ras (S27N) expression constructs and an empty expression vector (pBK-CMV) were generated as described previously (Matsumoto et al., 1997). An AP-1 luciferasereporter gene construct was constructed by inserting two copies of the AP-1 consensus DNAbinding sequence into the pGL-3 promoter vector (Promega, Madison, WI, USA). The IP-10CAT reporter gene construct is a gift from Dr Peng Liang. 5×105 cells (30 mm dishes) weretransiently co-transfected with either 0.5 μg AP-1 luciferase reporter gene or 0.5 μg IP-10 CATreporter gene with either 5 μg pBK-CMV vector or 5 μg M-Ras expression construct and 2.5μg PBL-CAT by the calcium phosphate precipitation technique previously described (Shattucket al., 1994). Three hours later, the cells were glycerol shocked for 3 min. The cells were washedtwice with complete medium, and then incubated with complete medium at 37°C/5% CO2.Two days later, the cells were washed with cold PBS and lysed in 1× reporter lysis buffer(Promega, Madison, WI, USA) for 15 min at room temperature and the lysate was cleared bycentrifugation. The luciferase or CAT activity was measured according to standard protocols(Promega, Madison, WI, USA) using a Monolight 2010 luminometer (AnalyticalLuminescence Laboratory, San Diego, CA, USA) or Beckman LS 3801.

Wang et al.


NIH


NIH


NIH


Transformation assays—Focus formation assays were performed in duplicate in threeindependent experiments. V1, Mel-a-6, E6A/ELR, L7A/ELR or R8A/ELR cells (Table 2) werecultured in DMEM supplemented with 10% FBS, and the foci of transformed cells werecounted 18 days after plating the cells. For M-Ras transformation experiments, V1 or Mel-a-6cells were cotransfected with 0.2 μg of a puromycin resistant vector (pBABE) and either 2 μgof a pBK/CMV, M-Ras (w.t.), dominant active M-Ras (G22V) or dominant negative M-Ras(S27N) by LIPOFECTAMINE PLUS reagent following manufacturer’s protocols (GIBCO –BRL, Rockville, MD, USA). The cells were cultured in 10% FBS DMEM with 0.8 mg/mlG418 and 0.5 mg/ml puromycin, and the foci of transformed cells were counted 18 days aftertransfection.

AcknowledgementsWe are indebted to the NIH for support: CA56704 (A Richmond); Vanderbilt Cancer Center CA 68485, Departmentof Veterans Affairs Merit Award and Career Scientist Award (A Richmond), and to James Owen and Eddy Balentienfor developing the MGSA/GRO expressing melan-a clones. We are also indebted to Ben Johnston and Amy Pruitt forexcellent technical assistance.

AbbreviationsCXC chemokine

chemokine with the first two conserved cysteine residues separated by anintervening amino acid

CMV cytomegalovirus

DMEM Dulbecco’s modified Eagle’s medium

IL-8 interleukin-8

MGSA/GRO melanoma growth-stimulatory activity/growth-regulated protein

AP-1 activation protein-1

RT-PCR reverse transcription-polymerase chain reaction

SV-40 simian virus-40

EMSA electrophoresis mobility shift assay

MMLV moloney murine virus

ELR motif glutamine-leucine-arginine tripeptide sequence near the N-terminus of MGSA/GRO proteins

Wang et al.


NIH


NIH


NIH


ReferencesArenberg DA, Polverini PJ, Kunkel S, Shanafelt A, Hesselgesser J, Horuk R, Strieter M. J Leukocyte

Biol 1997;62:554 – 562. [PubMed: 9365108]Balentien E, Mufson BE, Shattuck RL, Derynck R, Richmond A. Oncogene 1991;6:1115 – 1124.

[PubMed: 1861861]Bennett DC, Cooper PJ, Hart IR. Int J Cancer 1987;39:414 – 418. [PubMed: 3102392]Cao Y, Chen C, Weatherbee JA, Tsang M, Folkman J. J Exp Med 1995;182:2069 – 2077. [PubMed:

7500052]Chin L, Tam A, Pomerantz J, Wong M, Holash J, Bardeesy N, Shen Q, O’Hagan R, Pantginis J, Zhou

H, Horner JW, Cordon-Cardo C, Yancopoulos GD, DePinho RA. Nature 1999;400:468 – 472.[PubMed: 10440378]

Della Rocca GJ, van Biesen T, Daaka Y, Luttrell DK, Luttrell LM, Lefkowitz RJ. J Biol Chem1997;272:19125 – 19132. [PubMed: 9235901]

Domann FE Jr, Levy JP, Finch JS, Bowden GT. Mol Carcinog 1994;9:61 – 66. [PubMed: 8142009]Fujimoto J, Ichigo S, Hori M, Morishita S, Tamaya T. Mol Biol 1995;55:25 – 33.Geiser T, Dewald B, Ehrengruber MU, Clark-Lewis I, Baggiolini M. J Biol Chem 1993;268:15419 –

15424. [PubMed: 8340371]George DL, Glick B, Trusko S, Freeman N. Proc Natl Acad Sci USA 1986;83:1651 – 1655. [PubMed:

3513183]Haghnegahdar H, Du J, Wang D, Strieter RM, Burdick MD, Nanney LB, Cardwell N, Luan J, Shattuck-

Brandt R, Richmond A. J Leuk Biol 2000;67:53 – 62.Haskill S, Peace A, Morris J, Sporn SA, Anisowicz A, Lee SW, Smith T, Martin G, Ralph P, Sager R.

Proc Natl Acad Sci USA 1990;87:7732 – 7736. [PubMed: 2217207]Hawes BE, Luttrell LM, van Biesen T, Lefkowitz RJ. J Biol Chem 1996;271:12133 – 12136. [PubMed:

8647803]Hébert CA, Vitangcol RV, Baker JB. J Biol Chem 1991;266:18989 – 18994. [PubMed: 1918013]Hesselgesser J, Chitnis CE, Miller LH, Yansura DG, Simmons LC, Fairbrother WJ, Kotts C, Wirth C,

Gillece-Castro BL, Horuk R. J Biol Chem 1995;270:11472 – 11476. [PubMed: 7744785]Huang HS, Nagane M, Klingbeil CK, Lin H, Nishikawa R, Ji XD, Huang CM, Gill GN, Wiley HS,

Cavenee WK. J Biol Chem 1997;272:2927 – 2935. [PubMed: 9006938]Husain Z, Yang QM, Biswas D. Arch Dermatol 1990;126:324 – 330. [PubMed: 2178557]Iida N, Grotendorst GR. Mol Cell Biol 1990;10:5596 – 5599. [PubMed: 2078213]Johnson R, Spiegelman BM, Hanahan D, Wisdom R. Mol Cell Biol 1996;18:4504 – 4511. [PubMed:

8754851]Kimmelman A, Tolkacheva T, Lorenzi MV, Osada M, Chan AM. Oncogene 1997;15:2675 – 2685.

[PubMed: 9400994]Knall C, Worthen GS, Johnson GL. Proc Natl Acad Sci USA 1997;94:3052 – 3057. [PubMed: 9096344]Kolibaba KS, Druker BJ. Biochim Biophys Acta 1997;1333:F217 – F248. [PubMed: 9426205]Kurdowska A, Cohen AB, Carr FK, Stevens MD, Miller EJ, Mullenbach G, Tekamp-Olson P. Cytokine

1994;6:124 – 134. [PubMed: 8031995]Limatola C, Mileo AM, Giovannelli A, Vacca F, Ciotti MT, Mercanti D, Santoni A, Eusebi F. J Biol

Chem 1999;274:36537 – 36543. [PubMed: 10593952]Lloyd AC, Yancheva N, Wasylyk B. Nature 1991;352:635 – 638. [PubMed: 1907719]Lopez-Ilasaca M, Crespo P, Pellici PG, Gutkind JS, Wetzker R. Science 1997;275:394 – 397. [PubMed:

8994038]Louahed J, Grasso L, De Smet C, Van Roost E, Wildmann C, Nicolaides NC, Levitt RC, Renauld JC.

Blood 1999;94:1701 – 1710. [PubMed: 10477695]Luan J, Shattuck-Brandt R, Haghnegahdar H, Owen JD, Strieter R, Burdick M, Nirodi C, Beauchamp D,

Johnson KN, Richmond A. J Leuk Biol 1997;62:588 – 597.Mangues R, Seidman I, Gordon JW, Pellicer A. Oncogene 1992;7:2073 – 2076. [PubMed: 1328989]Matsumoto K, Asano T, Endo T. Oncogene 1997;15:2409 – 2417. [PubMed: 9395237]

Wang et al.


NIH


NIH


NIH


Mattei S, Colombo MP, Melani C, Silvani A, Parmiani G, Herlyn M. Int J Cancer 1994;56:853 – 857.[PubMed: 7509778]

McKay IA, Marshall CJ, Cales C, Hall A. EMBO J 1986;5:2617 – 2621. [PubMed: 3023052]Neptune ER, Bourne HR. Proc Natl Acad Sci USA 1997;94:14489 – 14494. [PubMed: 9405640]Owen JD, Strieter R, Burdick M, Haghnegahdar H, Nanney L, Shattuck-Brandt R, Richmond A. Int J

Cancer 1997;73:94 – 103. [PubMed: 9334815]Quilliam LA, Castro AF, Rogers-Graham KS, Martin CB, Der CJ, Bi C. J Biol Chem 1999;274:23850

– 23857. [PubMed: 10446149]Rennekampff HO, Hansbrough JF, Woods V Jr, Dore C, Kiessig V, Schroder JM. Arch Dermatol Res

1997;289:204 – 212. [PubMed: 9143736]Richmond A, Thomas HG. J Cell Physiol 1986;129:375 – 384. [PubMed: 3465735]Richmond A, Thomas HG. J Cell Biochem 1988;36:185 – 198. [PubMed: 3356754]Rodeck U, Becker D, Herlyn M. Cancer Cells 1991;3:308 – 311. [PubMed: 1931440]Rosenkilde MM, Schwartz TW. Mol Pharmacol 2000;57:602 – 609. [PubMed: 10692502]Rutberg SE, Lee EJ, Hansen LH, Glick AB, Yuspa SH. Mol Carcinog 1997;20:88 – 98. [PubMed:

9328439]Sadhu C, Masinovsky B, Staunton DE. J Immunol 1998;160:5622 – 5628. [PubMed: 9605168]Saez E, Rutberg SE, Mueller E. Cell 1995;82:721 – 732. [PubMed: 7545543]Shattuck RL, Wood LD, Jaffe GJ, Richmond A. Mol Cell Biol 1994;14:791 – 802. [PubMed: 8264646]Shattuck-Brandt RL, Wood LD, Richmond A. DNA Sequence 1997;7:379 – 386. [PubMed: 9524820]Shattuck-Brandt RL, Richmond A. Cancer Res 1997;57:3032 – 3039. [PubMed: 9230219]Shyamala V, Khoja H. Biochemistry 1998;37:15918 – 15924. [PubMed: 9843397]Smith JJ, Derynck R, Korc M. Proc Natl Acad Sci USA 1987;84:7567 – 7570. [PubMed: 3499610]Taylor SJ, Shalloway D. Curr Biol 1996;6:1621 – 1627. [PubMed: 8994826]van Dam H, Huguier S, Kooistra K, Baguet J, Vial E, Vandereb AJ, Herrlich P, Angel P, Castellazzi M.

Genes Dev 1998;12:1227 – 1239. [PubMed: 9553051]Vojtek AB, Der CJ. J Biol Chem 1998;273:19925 – 19928. [PubMed: 9685325]Westaway D, Papkoff J, Moscovici C, Varmus HE. EMBO J 1986;5:301 – 309. [PubMed: 3011401]Winzen R, Kracht M, Ritter B, Wilhelm A, Chen C-YA, Shyu A-B, Muller M, Gaestel M, Resch K,

Holtmann H. EMBO J 1999;18:4969 – 4980. [PubMed: 10487749]Wood LD, Richmond A. J Biol Chem 1995;270:30619 – 30626. [PubMed: 8530498]Zhang B, Shih JW, Wear DJ, Tsai S, Lo SC. Proc Sco Exp Biol Med 1997;214:359 – 366.Yang W, Wang D, Richmond A. J Biol Chem 1999;274:11328 – 11333. [PubMed: 10196223]

Wang et al.


NIH


NIH


NIH


Figure 1.MGSA/GRO up-regulates M-Ras expression in mRNA level in melanocytes (a) Northern blotanalysis of M-Ras: Total cellular RNA (20 μg) from V1 (lane 1), γ3-14 (lane 2) or Mel-a-6(lane 3) cells was extracted, separated on a formaldehyde-agarose gel, and transferred to aNytran Plus filter. The filter was hybridized with the 32P-labeled mouse M-Ras differentialdisplay fragment. Hybridizing bands were identified at 1.0 kb. The filter was re-probedwith 32P-labeled cyclophilin (1B15). The relative M-Ras band density is M-Ras band densitynormalized to cyclophilin (1B15) band density quantitated by a densitometer. The bar graphpresents a mean (±s.e.m.) of fold induction of relative M-Ras band intensity from threeindependent experiments (lower panel). (b) RT– PCR: Total cellular RNA (20 μg) from V1(lane 1), γ3-14 (lane 2) or Mel-a-6 (lane 3) cells was extracted and RT–PCR amplification wasperformed as described in Materials and methods, using specific oligonucleotides (5′ primer:ACCAGCGCTGTTCCAAGTGAA; 3′ primer: TCACAAGATGACACA CTGCAG). Thepost-PCR products were analysed in an ethidium bromide-stained 1.5% agarose gel. (c)Northern blot analysis of M-Ras induced by exogenous MGSA/GROα: Total cellular RNA(20 μg) from parental melan-a cells treated with recombinant human MGSA/GROα at 25 ng/ml for the indicated period was used in Northern blot analysis as described above. The relativeM-Ras band density is M-Ras band density normalized to cyclophilin (1B15) band densityquantitated by a densitometer. The bar graph presents a mean (±s.e.m.) of fold-induction ofrelative M-Ras band intensity from three independent experiments (lower panel)

Wang et al.


NIH


NIH


NIH


Figure 2.Over-expressing MGSA/GRO in melanocytes up-regulates M-Ras protein expression. (a)Western blot analyses of M-Ras protein expression: Whole cell extracts (50 μg) from melan-a clones V1 (lane 1), γ3-14 (lane 2), Mel-a-6 (lane 3), AAA-MGSA/GROα (lane 4), or E6A-MGSA/GROα mutant (lane 5) were resolved by 12% SDS-polyacrylamide gel electrophoresis,blotted onto nitrocellulose membranes, and probed with anti-M-Ras antibody as described inMaterials and methods. Lane 6 represents the whole cell extract (20 μg) from HEK293 cellsover-expressing M-Ras protein as a positive control. The band density was quantitated by adensitometer, and the results are reported as a mean of fold-induction (±s.e.m.) from threeindependent experiments (low panel). (b) Western blot analyses of Ras protein expression:Whole cell extracts (30 μg) from V1 (lane 1), γ3-14 (lane 2), or Mel-a-6 (lane 3) were resolvedby 12% SDS-polyacrylamide gel electrophoresis, blotted onto nitrocellulose membranes, andprobed with anti-pan-ras, anti-N-Ras, anti-K-Ras, anti-H-Ras or anti-Bcl2 antibody asdescribed in Materials and methods. The bar graph represents a mean of fold induction (±s.e.m.)of the total Ras band density quantitated by a densitometer from three independent experiments.(c) Western blot analyses of Ras protein expression in ELR-mutant clones: Western blots wereperformed by using whole cell extract (30 μg) from V1 (lane 1), Mel-a-6 (lane 2), E6A (lane3), R8A (lane 4) or L7A (lane 5) and probed with anti-pan-Ras as described above

Wang et al.


NIH


NIH


NIH


Figure 3.Ras activation. (a) Activation of endogenous Ras in MGSA/GRO-expressing clones: GTP-bound Ras was affinity-precipitated from 400 μg of whole cell extract of V1, γ3-14 or Mel-a-6using a GST-c-Raf-1/Ras binding domain (RBD) fusion protein. Ras proteins were detectedwith a pan-Ras antibody. Lane 1 is a negative control of GST alone to precipitate Ras proteinsfrom extract of Mel-a-6 cells. The amount of Bcl2 protein is shown in the lower panel forcontrol of total proteins in extracts from different clones. The bar graph represents the meanof fold-induction (±s.e.m.) of the band density of activated Ras quantitated by a densitometerfrom three independent experiments. (b) Activation of endogenous Ras in parental melan-acells treated with MGSA/GROα: GTP-bound Ras was affinity-precipitated from 400 μg ofwhole cell extract of parental melan-a cells treated with MGSA/GROα for the indicated timesusing the GST-c-Raf-1 RBD fusion protein. Ras protein bands were detected with a pan-Rasantibody. The expression of Bcl2 protein is shown in the lower panel to control for proteincontent in extracts. (c) Ras-GTP/GDP-binding and GTPase activity in MGSA/GRO-expressing clones: Left-panel: Ras proteins were immunoprecipitated from 400 μg of wholecell extract of V1, γ3-14 or Mel-a-6 with 2 μg of anti-pan-ras antibody (AB-3). Ras immunecomplexes were incubated with 1 μl of [α-32P]GTP (3000 Ci/mmol, Amersham) in 100 μl GTP/GDP-binding buffer and processed as described in Materials and methods. The results arereported as a mean of fold-induction (±s.e.m.) from three independent experiments. The foldinduction was calculated as the radioactivity (c.p.m.) of 32P-labeled nucleotides from γ3-14 orMel-a-6 divided by the radioactivity of 32P-labeled nucleotides from V1. Right-panel: Thereleased 32P-labeled GTP and GDP were loaded on PEI-cellulose plates, resolved by 0.75 MKH2PO4 and then visualized by autoradiography. (d) Ras-GTP/GDP-binding and GTPaseactivity in ELR-mutant clones: GTP/GDP-binding and GTPase activity assays were performedusing 400 μg of whole cell extract of V1, Mel-a-6, E6A/ELR, L7A/ELR and R8A/ELR asdescribed above

Wang et al.


NIH


NIH


NIH


Figure 4.MGSA/GRO stimulates AP-1 activity. (a) AP-1 binding activity in MGSA/GRO expressingclones: EMSA was performed on nuclear extracts (5 μg) from V1 (lane 1), γ3-14 (lane 2), andMel-a-6 (lane 3), using radiolabeled double-stranded AP-1 consensus oligonucleotide probe(CGCTTGATGAGTCAGCCGGAA) as described in Materials and methods. (b) AP-1 bindingactivity in parental melan-a cells treated with MGSA/GROα: EMSA was performed on nuclearextracts (5 μg) from parental melan-a cells treated with 50 ng/ml MGSA/GROα for theindicated times (lanes 2 – 7) or with carrier buffer alone (lane 1). (c) Identification of proteincomposition of AP-1 complex by EMSA super-shift analyses: The EMSA experiment wasperformed as in Figure 3a after incubation of nuclear extracts (5 μg) from Mel-a-6 clone with1 μg of antibody to pan-Jun (lane 3), pan-Fos (lane 4) or both (lane 5) at room temperature for30 min. (d) AP-1 luciferase transactivation analyses: Transient co-transfections wereperformed by the calcium phosphate method, using 2.5 μg of the AP-1 luciferase reporter geneand the pSV-β-Gal expression plasmid in V1, γ3-14 and Mel-a-6 cells. The luciferase activityand β-Gal activity were measured 48 h after transfection. The graph shows the average ofrelative luciferase activities (absolute activity of sample was normalized by β-gal activity) fromfour independent transfections. The error bars indicate the s.e.m

Wang et al.


NIH


NIH


NIH


Figure 5.Ras mediated AP-1 activation. (a) Ras inhibitor (SCH44342) blocks AP-1 binding activity:Mel-a-6 cells were pre-treated with the indicated concentrations of SCH44342 (lanes 2 – 4) orthe solvent control (lane 1) overnight. Nuclear proteins were extracted and EMSA’s wereperformed by using 5 μg of protein and 0.4 ng of 32P-labeled AP-1 consensus oligonucleotidein EMSA. (b) M-Ras increases basal activation of AP-1: Left panel: V1 cells were cotransfectedwith 2.5 μg AP-1 luciferase reporter and either 5 μg pBK-CMV vector (empty bar), M-Ras(striped bar) or the dominant active form of M-Ras (G22V) (solid bar), together with 2.5 μgpBL-CAT3 plasmid. Luciferase and CAT activity were measured 48 h later. The relativeluciferase activity represents the luciferase activity normalized by CAT activity. The resultsare reported as a mean (±s.e.m.) of fold-induction (the relative luciferase activity of dominantactive form of M-Ras divided by the relative luciferase activity of pBK-CMV vector) fromthree independent experiments. Right panel: V1 cells were cotransfected with 2.5 μg IP-10CAT reporter and 5 μg pBK-CMV vector (empty bar), M-Ras (striped bar) or the dominantactive form of M-Ras (G22V) (solid bar), together with 2.5 μg pSV-β-Gal expression plasmid.CAT and Luciferase activity were measured 48 h later. (c) Dominant negative M-Ras blocksMGSA/GRO-increased basal AP-1 activation: Left panel: γ3-14 or Mel-a-6 cells werecotransfected with 2.5 μg AP-1 luciferase reporter and 5 μg pBK-CMV vector (empty bar) orthe dominant negative M-Ras (solid bar), together with 2.5 μg pBL-CAT3 plasmid. Luciferaseand CAT activity were measured 48 h later. The relative luciferase activity represents theluciferase activity normalized by CAT activity. The results are reported as a mean (±s.e.m.) ofrelative inhibition (the relative luciferase activity of dominant negative form of M-Ras dividedby the relative luciferase activity of pBK-CMV vector) from three independent experiments.Right panel: γ3-14 or Mel-a-6 cells were cotransfected with 2.5 μg IP-10 CAT reporter and 5μg pBK-CMV vector (empty bar) or the dominant negative M-Ras (solid bar), together with2.5 μg pSV-β-Gal expression plasmid. CAT and Luciferase activity were measured 48 h later.(d) Dominant negative M-Ras blocks MGSA/GRO-induced AP-1 activation in V1 cells: V1cells were cotransfected with 2.5 μg AP-1 luciferase reporter and 5 μg pBK-CMV vector (emptybar) or the dominant negative M-Ras (solid bar), together with 2.5 μg pBL-CAT3 plasmid.After transfection, cells were treated with carrier buffer alone or with 50 ng/ml MGSA/GROα for 48 h. The relative luciferase activity represents the luciferase activity normalized

Wang et al.


NIH


NIH


NIH


by CAT activity. The results are reported as a mean (±s.e.m.) of relative luciferase activityfrom three independent experiments

Wang et al.


NIH


NIH


NIH


Figure 6.M-Ras induces focus formation and dominant negative M-Ras inhibits MGSA/GROα-inducedtransformation. (a) V1 or Mel-a-6 cells were cotransfected with a puromycin resistant vector(pBABE) and either a pBK/CMV, wild-type M-Ras, dominant active M-Ras (G22V) ordominant negative M-Ras (S27N). The cells were cultured in 10% FBS DMEM with G418and 1 μg/ml puromycin. Foci of transformed cells were stained with crystal violet and counted18 days after transfection. Each data point in the bar graph represents the mean (±s.e.m.) offold induction from a total of six different platings in three independent experiments. (b) TheV1, Mel-a-6, E6A, L7A or R8A cells were cultured in 10% FBS DMEM and the foci werestained with crystal violet and counted 18 days after transfection. The data shown representsone of three independent experiments with similar results each time (top panel). Each datapoint represents the mean (±s.e.m.) of colonies of six different platings in three independentexperiments (lower panel)

Wang et al.


NIH


NIH


NIH


NIH


NIH


NIH


Wang et al.

Table 1Name of melan-a clones

Clone name Vector Expressing gene Cells transfected Cellular transformation

V1 pRC/CMV Parental melan-a cells No

V4

V6

γ3-14 pRC/CMV MGSA/GROγ (ELR) Parental melan-a cells Yes

γ3-12

γ3-37

Mel-a-6 pRC/CMV MGSA/GROα (ELR) Parental melan-a cells Yes

Mel-a-4

Mel-a-9

E6A pRC/CMV MGSA/GROα (ALR) Parental melan-a cells No

L7A pRC/CMV MGSA/GROα (EAR) Parental melan-a cells No

R8A pRC/CMV MGSA/GROα (ELA) Parental melan-a cells No


NIH


NIH


NIH


Wang et al.

Table 2Differential display clones with altered expression in transfectant cells

Differential display clone Expression in transfectant cell Proposed identity

1 Mel-a-6 No match

2 γ3-14 No match

3 γ3-14, Mel-a-6 No match

4 γ3-14, Mel-a-6 M-Ras

5 Mel-a-6 No match

6 γ3-14, Mel-a-6 No match

7 β2-19 No match

8 V1, γ3-14 No match

9 V1, β2-19 Meis 2

10 V1, β2-19 SC1

11 V1, γ3-14, Mel-a-6 No match

12 Mel-a-6 No match

13 β2-19 No match

14 V1 No match

15 β2-19 No match

16 β2-19 No match

Differential display assay − 250 ng of total RNA from V1, β2-19, γ3-14 and Mel-a-6 were subjected to the differential display assay to examine M-Rasexpression using the primers (H-T11G and H-AP4) and conditions described in Materials and methods. The M-Ras identified by DNA sequence analysiswas differentially detected


MGSA/GRO-mediated melanocyte transformation involves induction of Ras expression

Documents

Transcript of MGSA/GRO-mediated melanocyte transformation involves induction of Ras expression