JBC Origene Your Gene Company

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M312697200 on June 17, 2004

J. Biol. Chem., Vol. 279, Issue 33, 34922-34930, August 13, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/33/34922    most recent
M312697200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Elad-Sfadia, G.
Right arrow Articles by Kloog, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Elad-Sfadia, G.
Right arrow Articles by Kloog, Y.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Galectin-3 Augments K-Ras Activation and Triggers a Ras Signal That Attenuates ERK but Not Phosphoinositide 3-Kinase Activity*

Galit Elad-Sfadia, Roni Haklai, Eyal Balan, and Yoel Kloog, Incumbent of The Jack H. Skirball Chair in Applied Neurobiology{ddagger}

From the Department of Neurobiochemistry, The George S. Wise Faculty of Life Sciences, Tel-Aviv University, 69978 Tel-Aviv, Israel

Received for publication, November 20, 2003 , and in revised form, May 24, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Depending on the cellular context, Ras can activate characteristic effectors by mechanisms still poorly understood. Promotion by galectin-1 of Ras activation of Raf-1 but not of phosphoinositide 3-kinase (PI3-K) is one such mechanism. In this report, we describe a mechanism controlling selectivity of K-Ras4B (K-Ras), the most important Ras oncoprotein. We show that galectin-3 acts as a selective binding partner of activated K-Ras. Galectin-3 co-immunoprecipitated significantly better with K-Ras-GTP than with K-Ras-GDP, H-Ras, or N-Ras and colocalized with green fluorescent protein-K-Ras(G12V), not with green fluorescent protein-H-Ras(G12V), in the cell membrane. Co-transfectants of K-Ras/galectin-3, but not of H-Ras/galectin-3, exhibited enhanced and prolonged epidermal growth factor-stimulated increases in Ras-GTP, Raf-1 activity, and PI3-K activity. Extracellular signal-regulated kinase (ERK) activity, however, was attenuated in K-Ras/galectin-3 and in K-Ras(G12V)/galectin-3 co-transfectants. Galectin-3 antisense RNA inhibited the epidermal growth factor-stimulated increase in K-Ras-GTP but enhanced ERK activation and augmented K-Ras(G12V) transformation activity. Thus, unlike galectin-1, which prolongs Ras activation of ERK and inhibits PI3-K, K-Ras-GTP/galectin-3 interactions promote, in addition to PI3-K and Raf-1 activation, a third inhibitory signal that attenuates active ERK. These experiments established a novel and specific mechanism controlling the duration and selectivity of signals of active K-Ras, which is extremely important in many human tumors.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The small GTPases H-Ras, K-Ras4B (referred to here as K-Ras), and N-Ras participate critically in the control of complex networks of signaling cascades that regulate cell proliferation, differentiation, survival, and death (1–4). These highly homologous proteins, which are regulated by guanine nucleotide exchange factors (GEFs)1 and by GTPase-activating proteins, differ from each other only in the hypervariable carboxyterminal region, which contributes to differences in trafficking (5) and membrane-microdomain localization (3, 6, 7). Ras proteins can control diverse cellular behaviors because of the ability of all Ras isoforms to activate a common set of effectors, including Raf-1, PI3-K RalGEFs, phospholipase C{epsilon}, RIN, and Tiam-1 (1–4). Our knowledge of how Ras output is coordinated and how each Ras isoform selects its set of effectors remains limited. Nonetheless, a number of experiments have demonstrated that the cellular context and the type of Ras isoform are important determinants of the Ras signal output. First, it was observed that in mice, knockout of K-Ras, but not of H-Ras or N-Ras, is lethal to embryos, suggesting that the Ras isoforms might have distinct biological functions (8, 9). In addition, recent studies showed that Ras regulation of matrix metaloproteinase-2 transcription was abolished in K-Ras but not in N-Ras knockout fibroblasts (10). Second, earlier experiments showed that activated Ras elicits transformation in immortal rodent cells but induces cell senescence in primary rodent cells (11). Third, the types of effectors required for Ras transformation of rodent cells were found to differ from those required in human cells (12). An analogous distinct effector-type requirement was documented in PC-12 cells, in which Raf and PI3-K signaling promote cell cycle arrest, whereas RalGEF signaling promotes cell proliferation (13). Last is the observation that activated K-Ras in human pancreatic tumor cell lines does not correlate with persistent activation ERK (14). These experiments highlight the importance of understanding the mechanisms that allow Ras proteins to select their effectors and to elicit diverse biological effects.

Recent experiments have provided new insights into mechanisms that can promote selectivity of Ras signaling. These studies show that H-Ras and K-Ras proteins are localized in distinct microdomains in the cell membrane that could confer specificity to the Ras signal (3, 6, 7, 15). Reports have also documented the localization of Ras in intracellular compartments (16–18), where Ras targeted to the endoplasmic reticulum and the Golgi activated distinct effectors (17). Recent studies have demonstrated that specificity of Ras signaling can be achieved by regulation of the activity of Ras effectors (19). Another important sign that Ras signaling selectivity can be achieved at the level of Ras-effector interactions comes from the observation that galectin-1, discovered originally as a {beta}-galactoside-binding protein (20), interacts with activated H-Ras and K-Ras proteins and diverts the Ras signal toward Raf-1 and away from PI3-K (21, 22). In addition, we observed in co-immunoprecipitation experiments that H-Ras(G12V) pulls down significantly more galectin-1 than K-Ras(G12V) (22). These experiments indicate that interactions of galectin-1 with H-Ras-GTP and with K-Ras-GTP are not identical. We wondered whether K-Ras or H-Ras isoforms might find another intracellular galectin partner that would elicit a different pattern of Ras membrane interactions, effector utilization, or both. Herein, we demonstrate that galectin-3 acts as a specific binding partner of activated K-Ras and that this interaction promotes strong K-Ras activation of PI3-K and Raf-1 as well as attenuation of ERK activation.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Plasmids and Constructs—We cloned the entire coding sequence of human galectin-3 cDNA from total RNA of macrophage 4437 cells using the forward primer 5'-TATAGAATTCAAATGGCAGACAATTTTTCGC-3' and the reverse primer 5'-TATAGAATTCGATTATATCATGGTATATGAAGC-3' in an reverse transcription-PCR amplification system (ACCESS RT-PCR; Promega). The 756-bp PCR product was subcloned into pGEM-T easy vector (Promega) and then cloned into the NotI site of pCDNA3 (pc-Gal-3). We prepared galectin-3 130-bp antisense RNA corresponding to the 5' end of galectin-3 by PCR using the pcDNA3-galectin-3 vector, the forward primer described above, and the reverse primer 5'-TAT AGG ATC CAT GGC CCC AGG ATA GGA AGC C-3'. The 130-bp fragment was cloned into the NotI site of pCDNA3 (galectin-3 AS) as described above. The orientation and sequence of the galectin-3 sense and antisense vectors were confirmed by DNA sequencing. The expression vectors pEGFP-C3-H-Ras, pEGFP-C3-H-Ras(G12V), pEGFP-C3-K-Ras, pEGFP-C3-K-Ras(G12V), pDCR-H-Ras(G12V), pCGN-K-Ras(G12V), pCGN-K-Ras(wt), pcDNA3-H-Ras(wt), pCMV-N-Ras(wt), pCMV-N-Ras(G13V), pc-EXV-3-Ras GTPase activating protein, pcDNA3-galectin-1, and glutathione S-transferase fused to the A85K mutant of the Ras-binding domain of c-Raf-1 in pGEX-2TH-p85 have been described previously (15, 21, 22).

Cell Culture—H-Ras(G12V)-transformed Rat-1 cells (23), N-Ras(G13V)-transformed Rat-1 cells (24), K-Ras(G12V)-transformed NIH3T3 cells (25), and NIH3T3 cells expressing K-Ras(G12V/C185S) (25), HEK-293, and COS-7 were maintained in Dulbecco's modified Eagle's medium/10% serum. HTB-54 and panc-1 cells (American Type Culture Collection) were grown, homogenized, and fractionated (100,000 x g pellet and supernatant) as detailed previously (23). Cellular fractions were then subjected to Western immunoblotting as detailed below. NIH3T3/K-Ras(G12V) (25) were transfected with 6 µg of galectin-3 AS or pcDNA3 using the calcium phosphate method as detailed elsewhere (22). Ten days later, the cells were fixed, stained with 0.1% crystal violet as detailed previously (21), and the number of foci were evaluated using the ImagePro Plus software (Media Cybernetics, Inc.).

Transiently transfected HEK-293 cells (calcium phosphate method) or COS-7 cells (dextran method) were used 48 h after transfection as detailed previously (22). Cells needed for biochemical and immunoblotting assays were harvested and lysed in a lysis buffer containing protease and phosphatase inhibitors (21, 22). Lysates were normalized for protein and subjected to the specified assays and quantification as described below. Data were expressed as means ± S.D., and differences between means were examined by a Student's t test.

Cross-linking/Two-step Gel Procedure—Galectin-3 was identified as a binding partner of Ras by the cross-linking/two-step gel procedure, performed essentially as described previously (21, 22). Immunoreactivity of Ras, galectin-1, and galectin-3 were detected by immunoblotting with mouse pan anti-Ras Ab (Ab-3; Calbiochem), rabbit anti-galectin-1 Ab (21, 22) and with rat anti galectin-3 Ab (Roche Molecular Biochemicals, Mannheim, Germany), respectively.

Co-immunoprecipitation and Western Immunoblotting—We co-transfected 1.5 x 106 COS-7 cells with 1.5 µg of DNA. The cells were transfected with plasmid DNA coding for galectin-3 (0.75 µg) along with plasmid DNAs coding for H-Ras(G12V) (0.75 µg), H-Ras(wt) (0.09 µg), K-Ras(G12V) (0.375 µg), K-Ras(wt) (0.375 µg), N-Ras(wt) (0.21 µg), or N-Ras(G13V) (0.21 µg). The different amounts of plasmid DNAs coding for the Ras isoforms were chosen to obtain equal levels of Ras expression. Cells were lysed 48 h after transfection (22). Lysates containing 7 mg of protein were precleared with 3 µg of naive mouse IgG and protein G-Sepharose and were then subjected to immunoprecipitation with 3 µg of mouse pan-Ras Ab and protein G-Sepharose followed by SDS-PAGE and Western immunoblotting with either 1:2000 pan anti-Ras Ab and 1:7500 peroxidase-goat anti-mouse IgG (The Jackson Laboratory, Bar Harbor, ME) or 1:500 rat anti-galectin-3 Ab and 1:5000 peroxidase-goat anti-rat IgG (Jackson Laboratory). Ras proteins and the 34-kDa galectin-3 protein were visualized by enhanced chemiluminescence (ECL) and the bands were quantified by densitometry (arbitrary units) with Image Master VDS-CL (Amersham Pharmacia Biotech) using TINA 2.0 software (Ray Tests) (21, 22). All experiments were repeated three times, unless otherwise indicated. Data obtained in the three separate experiments, after subtraction of background values, were averaged using the values recorded at zero time (no stimulation) as a reference point (normalized arbitrary units).

Ras-GTP, Phospho-ERK, Raf-1, and PI3K Assays—These assays were performed in HEK-293 cells (8 x 105) that were transfected with total amount of 6 µg of DNA (21, 22). Cells were transfected with the following combinations: 0.15 µg of plasmid DNA coding for GFP-H-Ras or for GFP-K-Ras and 5.85 µg of empty pcDNA, 0.15 µg of plasmid DNA coding for a GFP-Ras protein, and 2.85 µg of plasmid DNA coding for galectin-3, and 3 µg of pcDNA3, or of 2.85 µg of galectin-3 antisense RNA plasmid (galectin-3 AS) and 3 µg of pcDNA3. We also transected cells with 0.15 µg of DNA encoding K-Ras(G12V) alone or together with 2.85 µg of DNA encoding galectin-3 complemented with 3 µg of pcDNA3. Otherwise, cells were transfected with 6 µg of galectin-3 AS or DNA encoding galectin-3 for the determination of their impact on endogenous Ras-GTP, PI3-K activity, or Raf-1 activity. In addition, cells were transfected with 3.0 µg of DNA encoding p120RasGAP and 3.0 µg of pcDNA3, or 3.0 µg of DNA encoding galectin-3 and 3.0 µg of pcDNA3, or 3.0 µg of DNA encoding p120RasGAP and 3.0 µg of DNA encoding galectin-3 or 6.0 µg of pcDNA3 (vector control). Twenty-four hours after transfection, the cells were starved of serum for 24 h, then stimulated with 100 ng/ml EGF for the indicated times and lysed and subjected to determination of total GFP-Ras (~54 kDa), GFP-Ras-GTP, ERK, or phospho-ERK essentially as detailed earlier (21, 22) and Western immunoblotting as described above. In cells that were not transfected with vectors encoding Ras isoforms, we determined levels of endogenous Ras and Ras-GTP using either pan anti-Ras Ab or Ras isoform-specific Abs. We used 1:100 anti-H-Ras and 1:100 anti-K-Ras Abs (Calbiochem). We determined Raf-kinase activity using a Raf-1 immunoprecipitation kinase cascade assay kit (Upstate Biotechnology) essentially as detailed previously (21, 22) in HEK-293 cells co-transfected with plasmids coding for K-Ras, H-Ras, or K-Ras(G12V) along with a plasmid coding for galectin-3 or with empty vector, as described above. We determined PI3-K activity (26) essentially as detailed earlier (21, 22) in HEK-293 cells that were transfected and treated with EGF as described for the Raf-1 kinase assay.

Confocal Microscopy—In these experiments, cells were co-transfected with GFP-Ras and pcDNA3-gal-3, fixed in PBS containing 4% paraformaldehyde, and permeabilized with 0.5% Triton X-100. Samples were blocked (30 min) with 2% bovine serum albumin and 200 µg/ml goat {gamma}-globulin. Cells were labeled with galectin-3 Ab by successive incubations (each 45 min, 22 °C) with 1 µg/ml anti-galectin-3 Ab, 5 µg/ml biotin-goat anti rat IgG (Jackson ImmunoResearch, West Grove, PA), and 0.5 µg/ml Cy3-streptavidin (Jackson ImmunoResearch). Each incubation was followed by three extensive washes. Background staining was obtained on samples from which galectin-3 Ab was omitted. Dual fluorescence digital images were collected on a Zeiss LSM 510 confocal microscope fitted with fluorescein and rhodamine filters (21). ImagePro Plus software (Media Cybernetics) was used to analyze and quantify the extent of galectin-3 translocation to the cell membrane.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Galectin-3 Is a Specific Binding Partner of Activated K-Ras— Galectin-1 and -3 seem to be the most abundant galectins in human tumors (27–30). We therefore suspected that galectin-3 might be a reasonable candidate as a Ras-binding partner. We first used a previously designed cross-linking/two-step gel procedure in an attempt to detect Ras/galectin-3 complexes (21). This procedure uses a reducible cross-linker, and the cross-linked Ras adducts are first separated by SDS-PAGE under non-reducing conditions. To release Ras and its putative binding partner from the complexes, slowly migrating Ras adducts are extracted from the gel in the presence of the reducing reagent dithiothreitol (DTT). The individual proteins are then resolved by a second SDS-PAGE step (21). We used fibroblasts that stably express the activated GTP-bound H-Ras(G12V), K-Ras(G12V), or N-Ras(G13V) (21) and exhibit significant amounts of endogenous galectin-3 (Fig. 1A). Membranes of these Ras-transformed cells were subjected to the cross-linking/two-step gel procedure. The first gel was sliced, and from each slice proteins corresponding to gel migration of 34–43 kDa, 43–57 kDa, or 57–90 kDa were extracted under reducing conditions using DTT (immunoblots of the first gels are not shown). Three equal aliquots of the extracts were then separated by a second SDS-PAGE step followed by immunoblotting with either pan anti-Ras Ab or anti-galectin-3 Ab. The third sample was immunoblotted with anti-galectin-1 Ab (21) for comparison. This procedure enabled us to determine whether the 21-kDa Ras proteins and the 34-kDa galectin-3 protein were present in complexes of higher Mr and released from them by DTT. Consistent with previous experiments, we found that H-, K-, and N-Ras were present in the more slowly migrating protein complexes of the first gel (Ref. 21 and Fig. 1A). Galectin-3 and galectin-1 were detected in slowly migrating H-Ras(G12V) and K-Ras(G12V) complexes as well (Fig. 1A). In addition, larger amounts of galectin-1 were released from H-Ras(G12V) complexes than from K-Ras(G12V) complexes (Ref. 21 and Fig. 1A). On the other hand, significantly larger amounts of galectin-3 were released from slowly migrating (43–57 kDa and 57–90 kDa) adducts of K-Ras(G12V) than from H-Ras(G12V) adducts (Fig. 1). The galectin-3 immunoreactive band at 34–43 kDa (Fig. 1A) is too small to account for Ras/galectin-3 complexes and seems to correspond to the free 34-kDa galectin-3 protein. However, the 43–57-kDa complexes, from which both the 21-kDa Ras and the 34-kDa galectin-3 were released by DTT, could account for Ras/galectin-3 complexes as their apparent Mr values correspond to the expected sum of their molecular masses (55 kDa). This latter finding pointed to the possible existence of a heterocomplex of K-Ras(G12V) and galectin-3. The Ras/galectin-3 adducts at the highest range (57–90 kDa) might represent high order complexes. It is noteworthy that only very small amounts of galectin-1 or -3 were released from 43–57-kDa or 57–90-kDa adducts of N-Ras(G13V) (Fig. 1A). This suggests that N-Ras(G13V) may not interact with galectin-1 or -3. On the other hand, it is possible that high order galectin-3 complexes that did not enter the first gel were formed in the N-Ras(G13V) cells, reducing the levels of free galectin-3 available for interactions with N-Ras(G13V). Consistent with the second possibility is the observation that no 34–43-kDa galectin-3 immunoreactive band was detected in the N-Ras(G13V) expressing cells (Fig. 1A). Together, these findings suggest that galectin-3 might interact directly with activated K-Ras. Preferential galetin-3/K-Ras interactions, however, may not be conclusively deduced from these experiments because of the the significant excess of galectin-3 in the K-Ras-expressing cells. This possibility was further examined by co-immunoprecipitation experiments using cells that express comparable amounts of galectin-3.



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 1.
Galectin-3 is a specific binding partner of activated K-Ras. A, using the cross-linking/two-step gel procedure, we examined Ras/galectin-3 and Ras/galectin-1 complexes in equal numbers of fibroblasts overexpressing oncogenic H-, K-, and N-Ras isoforms as detailed under "Experimental Procedures." Levels of Ras proteins and galectins in the starting material are shown in the upper (K-Ras(G12V)) has a leader sequence resulting in its slower gel migration). Bands corresponding to Ras adducts of 34–43 kDa, 43–57 kDa, and 57–90 kDa of the first gel were excised, and the proteins were resolved on a second gel in the presence of DTT. The apparent amounts of Ras, galectin-1, and galectin-3 released from these Ras complexes were detected by immunoblotting (IB) with the corresponding Abs and visualization by ECL. Typical immunoblots of one of three experiments with similar results are shown (lower). B, preferential co-immunoprecipitation of galectin-3 with active K-Ras. COS-7 cells were co-transfected with plasmid coding for galectin-3 along with plasmids coding for H-Ras (G12V), K-Ras(wt), and K-Ras(G12V), each with a leader sequence, or H-Ras(wt), N-Ras(wt), or N-Ras(G13V), and the cells were lysed 48 h later. Lysates containing comparable levels of Ras and galectin-3 (left) were subjected to immunoprecipitation (IP) with pan anti-Ras Ab, and this was followed by SDS-PAGE and immunoblotting (IB) with either anti-Ras Ab or anti-galectin-3 (Gal-3) Ab (right). Results of a representative experiment are shown. Similar results were obtained in two additional experiments. C, galectin-3 co-immunoprecipitates with K-Ras from lysates of HTB-54 and panc-1 cells. Levels of expression of endogenous Ras and galectin-3 (left) were determined in the total cell lysates as detailed in B. Loading control (tubulin) is shown as well. Lysates were subjected to immunoprecipitation (IP) with pan anti-Ras Ab (middle)or with naive mouse IgG (right), and this was followed by immunoblotting (IB) with either anti-K-Ras Ab or anti-galectin-3 (Gal-3) Ab. Results of a representative experiment are shown. D, Unfarnesylated K-Ras (G12V/C185S) interacts weakly with galectin-3. Lysates of fibroblasts cells stably expressing K-Ras(G12V) or K-Ras(G12V/C185S) were subjected to immunoprecipitation with pan anti-Ras Ab, and this was followed by SDS-PAGE and immunoblotting with anti-Ras Ab or anti-galectin-3 (Gal-3) Ab, as described above. Left, level of protein expression. Right, Co-immunoprecipitation. Results of a representative experiment are shown.

 
Preferential Co-immunoprecipitation of Galectin-3 with Activated K-Ras—To further examine the possibility that K-Ras interacts with galectin-3, we used co-immunoprecipitation assays, which are independent of the cross-linking procedure. COS-7 cells were co-transfected with galectin-3 together with K-Ras(wt), K-Ras(G12V), H-Ras(wt), H-Ras(G12V), N-Ras(wt), and N-Ras(G13V). The wild-type Ras isoforms were included in these experiments to examine whether interaction of galectin-3 with K- or H-Ras depends on their GDP/GTP states.

Lysates of the transfected cells were subjected to immunoprecipitation with pan anti-Ras Ab followed by immunoblotting with anti-galectin-3 or anti-Ras Ab. Results of a typical experiment (Fig. 1B) demonstrate that galectin-3 co-immunoprecipitated with K-Ras(G12V) from lysates of K-Ras(G12V)/galectin-3 co-transfectants. At comparable levels of expression of Ras and galectin-3 (Fig. 1B, left), only small amounts of galectin-3 co-immunoprecipitated with K-Ras(wt), H-Ras(wt), H-Ras(G12V), N-Ras(wt), or N-Ras(G13V) (Fig. 1B). Thus, galectin-3 co-immunoprecipitated with K-Ras(G12V) much more efficiently than with K-Ras(wt), H-Ras(wt), H-Ras (G12V), N-Ras(wt), or N-Ras(G13V); binding of galectin-3 to these last five proteins (mean ± S.D., n = 3) was, respectively, 23 ± 4% (p < 0.05), 35 ± 6% (p < 0.05), 31 ± 5% (p < 0.05), 14 ± 4% (p < 0.05), and 25 ± 6% (p < 0.05) that of binding to K-Ras(G12V). In keeping with the results of the cross-linking experiments, these results strongly suggest that galectin-3 is a specific binding partner of K-Ras(G12V). In support of this suggestion, Fig. 1C demonstrates that galectin-3 co-immunoprecipitated with K-Ras from lysates of HTB-54 and panc-1 cells that harbor oncogenic K-Ras.

Previous studies showed that unfarnesylated H-Ras interacts weakly with galectin-1, pointing to participation of the farnesyl group in H-Ras-galectin-1 interactions (21, 22). It was therefore of interest to examine whether the farnesyl group of K-Ras(G12V) participates in K-Ras(G12V)/galectin-3 interactions, using fibroblasts stably expressing comparable levels of the normally processed K-Ras(G12V) or the unfarnesylated K-Ras(G12V/C185S) (Fig. 1D). Lysates of these cells were subjected to co-immunoprecipitation with pan anti-Ras Ab and immunoblotting with anti-galectin-3 Ab or anti-Ras Ab. Results of a typical experiment show that significantly less galectin-3 co-immunoprecipitated with the unfarnesylated K-Ras(G12V/C185S) than with K-Ras(G12V) (Fig. 1D), suggesting that the farnesyl group of K-Ras(G12V) is important for its interaction with galectin-3.

Activated K-Ras Increases Translocation of Galectin-3 to the Cell Membrane—To gain further support for the observed interactions of galectin-3 and K-Ras(G12V), we performed confocal fluorescence microscopy analysis of whole cells co-transfected with galectin-3 and GFP-Ras. The assumption in this experiment was that the interactions between Ras and galectin-3 in the whole cell are mediated by recruitment of galectin-3, a cytosolic protein (31), to the cell membrane by the membrane-bound Ras protein. Because a large fraction of galectin-3 is cytosolic, it is difficult to detect its translocation to the plasma membrane by standard fluorescence microscopy. This obstacle can be overcome by imaging based on co-localization of galectin-3 with GFP-Ras, which is largely associated with the cell membrane.

COS-7 cells were co-transfected with galectin-3 and GFP-K-Ras(G12V), GFP-K-Ras(wt), GFP-H-Ras(G12V), or GFP-H-Ras(wt). The cells were fixed and permeabilized, and galectin-3 was labeled with anti-galectin-3 Ab followed by biotin-goat anti-rat IgG and streptavidin-Cy3. Dual fluorescence images (GFP, green; Cy3, red) were collected, overlaid, and examined for co-localization using the co-localization function of the confocal imaging program (Zeiss LSM 510). Results of a typical experiment are shown in Fig. 2. Upon co-expression with GFP-K-Ras(G12V), a significant fraction of the galectin-3 population became co-localized with GFP-K-Ras(G12V) at the rim of the cells, exhibiting typical plasma membrane labeling (Fig. 2A, iii). This effect was significantly less pronounced after co-expression with GFP-K-Ras(wt) (Fig. 2B, iii). The effect was mediated specifically by GFP-K-Ras(G12V) and not by GFP-H-Ras(G12V) or GFP-H-Ras(wt) (Fig. 2, C, iii, and D, iii); this was indicated by the failure of co-expression with these two latter Ras isoforms (judged by immunofluorescence to be expressed at levels comparable with those of GFP-K-Ras(G12V)) to translocate galectin-3 to the plasma membrane. The extent of galectin-3 translocation to the cell membrane in GFP-K-Ras(G12V) cells was 32% higher than that in the cell membrane of GFP-K-Ras(wt) cells (11.5 ± 1.9% and 8.4 ± 1.3% of the total galectin-3, respectively, was localized to the cell membrane, n = 6, p < 0.01). Taken together, these data demonstrate that activated GFP-K-Ras(G12V) specifically mediates translocation of galectin-3 to the plasma membrane in whole cells. In addition, independent experiments showed that K-Ras(G12V) caused a significant shift of endogenous galectin-3 from the cytosolic fraction (S100) toward the particulate fraction (P100) of K-Ras(G12V) transfectants (Fig. 2E).



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 2.
Confocal fluorescence microscopy demonstrates that galectin-3 is translocated to the plasma membrane specifically by GFP-K-Ras(G12V). COS-7 cells were co-transfected with a vector encoding galectin-3 and vectors encoding GFP-K-Ras(G12V), GFP-K-Ras(wt), GFP-H-Ras(G12V), or GFP-H(wt). After fixation and permeabilization, galectin-3 was labeled with anti-galectin-3 Ab and then with biotin-goat anti-rat IgG and steptavidin-Cy3. Dual images (green fluorescence for GFP, red for Cy3) were collected under a Zeiss LSM 510 confocal microscope fitted with non-leaking green and red fluorescence filters. The green and red images were superimposed and analyzed by the co-localization function of the LSM 510 software. Bars, 10 µm. Left, Ai–Di, red images (galectin-3). Middle, Aii–Dii, green images (GFP-Ras). Right, Aiii–Diii, level of co-localization. Marked co-localization of galectin-3 and GFP-K-Ras(G12V) occurs at the rim of the cell (at the plasma membrane; Aiii). No such strong co-localization is observed for galectin-3 or the other GFP-tagged Ras proteins (Biii–Diii). Typical images are shown. Similar images were collected in two independent experiments. E, subcellular distribution of K-Ras(12V) and galectin-3 in K-Ras(12V)-transfected HEK-293 cells. HEK-293 cells were transfected with a vector encoding K-Ras(12V) or with empty vector. Levels of expression of K-Ras(12V), galectin-3, and tubulin in total cell homogenates (expression levels) and in the particulate (P100) and cytosolic fractions (S100) were determined by immunoblotting as detailed under "Experimental Procedures." Results of a representative experiment are shown. The amount of galectin-3 in P100 of the K-Ras(12V) cells (mean ± S.D., n = 3) was 3.0 ± 0.6-fold higher than that in P100 of the empty vector cells.

 
Galectin-3 Augments the Basal and EGF-induced Increase in GFP-K-Ras-GTP but Not in GFP-H-Ras-GTP and Prolongs the EGF Response—Next, we examined the influence of Ras/galectin-3 interactions on the stability of the bound nucleotide using Ras(wt) isoforms that alternate between GDP- and GTP-bound states. After transfections in HEK-293 cells and stimulation with EGF, we pulled down GFP-Ras-GTP with glutathione S-transferase fused to the Ras-binding domain of Raf-1, which binds Ras-GTP but not Ras-GDP (32). In cells transfected with GFP-H-Ras or GFP-K-Ras, we found that the typical EGF-stimulated transient increase in Ras-GTP peaked at 5 min and returned to basal levels after 10–15 min (Fig. 3, A and B). In GFP-K-Ras/galectin-3 co-transfectants, we found higher levels of basal and EGF-stimulated Ras-GTP and prolongation of the EGF response relative to GFP-K-Ras transfectants (Fig. 3A). The apparent levels of GFP-K-Ras-GTP were about 1.6- to 2.0-fold higher in the co-transfectants than in the GFP-K-Ras transfectants and were still relatively high 1 h after stimulation. In contrast, in GFP-H-Ras/galectin-3 co-transfectants, the EGF-induced increase in Ras-GTP was not higher than that observed in H-Ras transfectants (Fig. 3B).



View larger version (34K):
[in this window]
[in a new window]
  		FIG. 3.
Galectin-3 increases basal and EGF-stimulated levels of GFP-K-Ras-GTP but not of GFP-H-Ras-GTP. HEK-293 cells were co-transfected with plasmids encoding GFP-Ras/pcDNA3, GFP-Ras/galectin-3, GFP-Ras/galectin-3 antisense (A and B), or galectin-3 antisense only (C). Twenty-four hours after transfection the cells (A and B) were starved of serum overnight then stimulated with EGF for the indicated periods and lysed. Lysates were subjected to SDS-PAGE followed by immunoblotting with pan anti-Ras Ab for assay of total GFP-Ras or subjected to the glutathione S-transferase-Ras-binding domain pull-down assay followed by SDS-PAGE and immunoblotting with pan anti-Ras Ab for assay of GFP-Ras-GTP (A and B). The experiments were repeated three times each for each set of GFP-K-Ras (A) and the set of GFP-H-Ras co-transfectants (B). The immunoblots, visualized by ECL, were then subjected to densitometry as detailed under "Experimental Procedures." Top, immunoblots of a representative experiment with each set. Bottom, results of densitometry (means ± S.D.) in normalized arbitrary units (AU). C, cells transfected with vector control or with galectin-3 AS were lysed and subjected to SDS-PAGE, and this was followed by immunoblotting with galectin-3 Ab.

 
The above results, in line with the preferential co-immunoprecipitation of galectin-3 with K-Ras-GTP over K-Ras-GDP (Fig. 1B), suggest that galectin-3 binds preferentially to K-Ras-GTP and may render Ras less sensitive to Ras-GTPase–activating proteins (GAPs). Therefore, galectin-3 might have been expected to raise basal Ras-GTP levels by stabilizing the small percentage of basal Ras-GTP that existed before EGF stimulation. This indeed seemed to be the case, in that basal levels of GFP-K-Ras-GTP were 2.2-fold higher in GFP-K-Ras(wt)/galectin-3 co-transfectants than in GFP-K-Ras transfectants (Fig. 3A). The suggestion mentioned above gained additional support. First, we knocked down galectin-3 expression in HEK-293 cells (Fig. 3C) using galectin-3 antisense RNA (galectin-3 AS). Previous experiments showed that galectin-3 AS specifically down-regulates the expression of galectin-3 (33) but not of galectin-1 (21). We co-transfected the cells with galectin-3 AS and GFP-K-Ras or GFP-H-Ras, then stimulated the co-transfectants with EGF and measured the time-dependent change in Ras-GTP levels. Both basal and EGF-induced increases in Ras-GTP were inhibited in the GFP-K-Ras/galectin-3 AS co-transfectants (Fig. 3A), but not in the GFP-H-Ras/galectin-3 AS co-transfectants (Fig. 3B). Second, we examined the impact of galectin-3 and galectin-3 AS on EGF-induced increase in endogenous Ras-GTP levels without overexpressing Ras. In galectin-3 transfectants, we found higher levels of EGF-stimulated increase in endogenous Ras-GTP and prolongation of the EGF response relative to pcDNA3 vector controls (Fig. 4A). In contrast, the EGF-stimulated increase in endogenous Ras-GTP was strongly inhibited in galectin-3 AS transfectants (Fig. 4A). Use of Ras-isoform-specific Abs showed that the effects of galectin-3 and of galectin-3 AS on EGF-induced increase in Ras-GTP were specific to the endogenously expressed K-Ras (Fig. 4B), not H-Ras (Fig. 4C). Third, we examined the effect of Ras/galectin-3 interactions on p120RasGAP activity. To avoid triple transfections, and because our experiments showed that, in HEK-293 cells, galectin-3 specifically enhanced the EGF-stimulated increase in endogenous K-Ras-GTP (Fig. 4), the experiments were performed without Ras overexpression. We thus examined the time-dependent changes in EGF-stimulated Ras-GTP in vector controls, in p120RasGAP transfectants, in galectin-3 transfectants, and in galectin-3/p120RasGAP co-transfectants. As expected, galectin-3 enhanced and prolonged the EGF-stimulated increase in endogenous Ras-GTP; conversely, p120RasGAP decreased the EGF-stimulated increase in Ras-GTP (Fig. 5). The decrease in Ras-GTP levels induced by p120RasGAP was markedly attenuated by galectin-3 (Fig. 5). These results strongly suggest that galectin-3 reduces the efficiency of p120RasGAP-mediated facilitation of GTP hydrolysis by K-Ras. Taken together, these experiments establish a selective functional link between galectin-3 and active K-Ras.



View larger version (59K):
[in this window]
[in a new window]
  		FIG. 4.
Galectin-3 increases EGF-stimulated levels of the endogenous K-Ras-GTP but not of H-Ras-GTP. HEK-293 cells were transfected with pcDNA3 vector control or with a plasmid coding for galectin-3 or with galectin-3 antisense. Twenty-four hours after transfection, the cells were starved of serum overnight and then were subjected to EGF stimulation and determination of Ras and Ras-GTP as detailed in Fig. 3. Three separate samples were then used for SDS-PAGE and immmunoblotting with pan anti-Ras Ab (A), specific anti-K-Ras Ab (B), or specific anti-H-Ras Ab (C). Results of one of two experiments with similar results are shown.

 



View larger version (57K):
[in this window]
[in a new window]
  		FIG. 5.
Galectin-3 reduces the efficiency of p120RasGAP-mediated facilitation of GTP hydrolysis by Ras. HEK-293 cells were transfected with pcDNA3 vector control, with a plasmid encoding galectin-3, or with plasmid encoding p120 RasGAP. Cells were also co-transfected with galectin-3 and p120 RasGAP vectors. Twenty-four hours later, the cells were subjected to EGF stimulation and determination of endogenous Ras and Ras-GTP using pan anti-Ras Ab as detailed in Fig. 3. The immunoblots, visualized by ECL, were then subjected to densitometry. Top, immunoblots of a representative experiment with each set. Bottom, results of densitometry (means ± S.D., n = 3) in normalized arbitrary units (AU).

 
Galectin-3 Enhances the K-Ras-mediated Increase but Not the H-Ras-mediated Increase in PI3-K Activity—To determine whether the functional linkage between galectin-3 and Ras-GTP noted above might influence Ras activation of effectors, we first examined the effects of Ras/galectin-3 interactions on PI3-K (34). HEK-293 cells transfected with H-Ras or K-Ras, or cells co-transfected with galectin-3 and either of these two Ras isoforms, were stimulated with EGF for 5 min (by which time peak Ras-GTP levels are observed (Fig. 3, A and B). Extracts of the cells expressing comparable levels of Ras and galectin-3 (Fig. 6A) were then subjected to a PI3-K assay. We found that EGF stimulated PI3-K activity, increasing it by 1.5 ± 0.1-fold and 1.6 ± 0.3-fold (means ± S.D., n = 3) in H-Ras and K-Ras transfectants, respectively (Fig. 6A). The EGF-stimulated increase in PI3-K activity in K-Ras/galectin-3 co-transfectants was enhanced compared with the response observed in K-Ras transfectants (2.4 ± 0.1-fold and 1.6 ± 0.3-fold increases, respectively; Fig. 6A). Galectin-3 expression had no effect on the EGF-induced response in H-Ras/galectin-3 co-transfectants (Fig. 6A). This finding and the enhanced EGF-induced increase in PI3-K activity observed in K-Ras/galectin-3 co-transfectants stand in marked contrast to the earlier observation that interactions of galectin-1 with K-Ras-GTP or with H-Ras-GTP cancel the EGF-induced activation of PI3K activation (22). The enhanced PI3-K activity induced by galectin-3 in K-Ras-GTP/galectin-3 co-transfectants seems to reflect the galectin-3 induced stabilization of K-Ras-GTP. In cells co-transfected with galectin-3 and K-Ras(G12V), the Ras-induced increase in PI3-K indeed was not enhanced (Fig. 6B). Nonetheless, as expected, interactions of K-Ras(G12V) with galectin-3, but not with galectin-1, promoted PI3-K activation (Fig. 6B). Our finding that the EGF-stimulated increase in PI3-K activity mediated by endogenous Ras was strongly inhibited in galectin-3 AS transfectants is noteworthy (Fig. 6C).



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 6.
EGF induces increase in PI3K activity and in Raf-1 activity in K-Ras/galectin-3 co-tranfectants. H-Ras(wt) or K-Ras(wt) and H-Ras(wt)/galectin-3 or K-Ras(wt)/galectin-3 co-transfectants, as well as galectin-3 AS or pcDNA transfectants, were stimulated 5 min with EGF (A, C, and D) then lysed as described under "Experimental Procedures." Otherwise, K-Ras(G12V), K-Ras(G12V)/galectin-3, or K-Ras(G12V)/galectin-1 co-transfectants were grown without EGF stimulation, then lysed (B). Immunoblot analysis verified that expression patterns of Ras isoforms under the different co-transfection conditions were similar (left, A–C; top, D). Lysates were immunoprecipitated with anti PI3-K p85 Ab, and PI3-K activity was then determined (A–C, middle). 32P-labeled lipid products were separated by TLC and visualized by overnight exposure to x-ray film. The phosphatidylinositol-3-phosphate (PIP) spots were then analyzed by quantitative densitometry (see text). Results of representative experiments are shown; - denotes empty-vector transfections or no EGF stimulation. Differences between mean values (n = 3) in the K-Ras and in the K-Ras/galectin-3 co-transfectants (A) and between values (n = 3) in vector control and galectin-3 AS transfectants (C) were statistically significant (p < 0.01). Lysates were also immunoprecipitated with anti-Raf-1 Ab for the determination of Raf-1 activity as detailed under "Experimental Procedures" (C, right, and D). The amounts of 32P (dpm) incorporated into MBP were then determined. The mean values (± S.D.) recorded in four separate experiments are shown. Differences between mean values in the vector control and galectin-3 AS (C, right) and between the K-Ras and the K-Ras/galectin-3 co-transfectants (D) were statistically significant (p < 0.05).

 
Galectin-3 Promotes the EGF-stimulated Increase in Raf-1 Activity and Attenuates ERK Activation—Next, we examined the influence of Ras/galectin-3 interactions on Raf-1 using HEK-293 cells transfected with galectin-3 or with Ras(wt) alone or co-transfected with Ras(wt) and galectin-3. We measured Raf-1 activity in the cell lysates 5 min after EGF stimulation (21, 22). We found that EGF induced a significantly greater increase in Raf-1 activity in H-Ras or K-Ras transfectants than in vector controls (Fig. 6D). Enhancement of the EGF response in galectin-3/K-Ras(wt) co-transfectants, but not in galectin-3/H-Ras(wt) co-transfectants, although small, was significant compared with the response observed in the cells transfected with K-Ras(wt) or H-Ras(wt) alone (Fig. 6D). In addition, we found that the EGF-stimulated increase in Raf-1 activity mediated by endogenous Ras was inhibited in galectin-3 AS transfectants (Fig. 6C). These experiments thus demonstrate that the association of K-Ras-GTP with galectin-3 promotes activation of Raf-1. In separate experiments, we found that galectin-3 expression also promoted activation of Raf-1 by K-Ras(G12V) (Fig. 6D).

Because Raf-1 activates mitogen-activated protein kinase kinase, which in turn activates ERK, we next examined whether the enhancement in EGF-induced Raf-1 activity was propagated to ERK. Using GFP-K-Ras or GFP-H-Ras transfectants and GFP-Ras/galectin-3 co-transfectants, we measured the EGF-induced increase in active phospho-ERK protein in lysates of HEK-293 cells. In GFP-K-Ras and GFP-H-Ras transfectants, EGF induced a transient increase in phospho-ERK that peaked at 5 min and faded within 15–30 min (Fig. 7). Galectin-3 expression had no effect on the EGF-induced increase in phospho-ERK in H-Ras/galectin-3 co-transfectants (Fig. 7B), but in K-Ras/galectin-3 co-transfectants, the phospho-ERK peak was lower and the duration of the signal was shorter than in K-Ras transfectants, indicating that the EGF response in K-Ras/galectin-3 co-transfectants was attenuated by expression of galectin-3 (Fig. 7A). Likewise, we observed significantly lower levels of phospho-ERK in K-Ras(G12V)/galectin-3 co-transfectants than in K-Ras(G12V) transfectants (Fig. 7C). The lack of effect of galectin-3 on the H-Ras signal was not surprising in light of our results showing that galectin-3 is not a binding partner of H-Ras. On the other hand, attenuation of the EGF-induced signal to ERK was not anticipated, in light of the activation of Raf-1 clearly observed in K-Ras-GTP/galectin-3 co-transfectants (Fig. 6D). A possible interpretation of this finding is that K-Ras-GTP/galectin-3 complexes might promote the activation not only of PI3-K and Raf-1 but also of a third effector that leads to attenuation of ERK activation. The possibility that galectin-3 itself, independently of its impact on Ras-GTP, activated such an inhibitory signal could be ruled out, because in such a case, ERK activation in H-Ras/galectin-3 co-transfectants would have been attenuated, and this did not happen (Fig. 7B). The possibility that an inhibitory signal is indeed promoted by K-Ras-GTP/galectin-3 gains strong support from experiments in which galectin-3 AS was used to down-regulate galectin-3. In those experiments, we examined the effects of GFP-K-Ras or GFP-H-Ras transfectants and GFP-Ras/galectin-3 AS co-transfectants on the EGF-induced increase in phospho-ERK. Galectin-3-AS did not affect the EGF-induced increase in phospho-ERK in GFP-H-Ras/galectin-3 AS co-transfectants (Fig. 7B), although it enhanced and prolonged the EGF response in GFP-K-Ras/galectin-3 AS co-transfectants (Fig. 7A). It is noteworthy that, under the same conditions, the levels of GFP-K-Ras-GTP in the GFP-K-Ras/galectin-3 AS co-transfectants were very low, representing only 20% of those observed in the K-Ras-GTP transfectants (Fig. 3A). Thus, down-regulation of galectin-3 not only canceled the K-Ras/galectin-3 inhibitory signal to phospho-ERK but actually enhanced ERK activation.



View larger version (34K):
[in this window]
[in a new window]
  		FIG. 7.
Galectin-3 attenuates EGF-stimulated ERK activation of K-Ras but not of H-Ras. GFP-Ras, GFP-Ras/galectin-3, or GFP-Ras/galectin-3 antisense co-transfectants were stimulated with EGF for the indicated periods of time. K-Ras(12V) and K-Ras(12V)/galectin-3 co-transfectants were used without EGF stimulation. ERK and active phospho-ERK were then assayed by immunoblotting with anti-ERK and anti-phospho-ERK Abs followed by ECL and densitometry. Top, representative immunoblots. Bottom, results of the analysis (means ± S.D., n = 3) in normalize arbitrary units (AU). Results obtained with the GFP-K-Ras set of transfectants (A), the GFP-H-Ras set of transfectants (B), and K-Ras(12V)/galectin-3 co-transfectants (C) are shown.

 
Next, we examined the impact of galectin-3 AS on K-Ras transforming activity in K-Ras(G12V)-transformed fibroblasts. To this end, we transfected NIH3T3/K-Ras(G12V) cells (25) with galectin-3 AS or with the vector control and measured foci formation 10 days after the transfection. In agreement with previous reports (25, 35), the vector control NIH3T3/K-Ras(G12V) cells presented a transformed phenotype forming multilayer clumps of highly refractile spindle-like cells (Fig. 8). The mean number of transformed foci recorded in these cultures was 70 ± 9 (mean ± S.D., n = 4). Galectin-3-AS transfection strongly enhanced K-Ras(G12V) transformation as evidenced by the increase in the number of the foci (Fig. 8). The number of foci recorded in the galectin-3 AS transfectants (172 ± 8, n = 4) was significantly higher than that observed in the vector controls (p < 0.05). These results are in accord with the observed increase in active ERK in galectin-3 AS transfectants (Fig. 7A).



View larger version (109K):
[in this window]
[in a new window]
  		FIG. 8.
Galectin-3 AS enhances k-Ras transforming activity. K-Ras(G12V)-transformed NIH3T3 cells were transfected with pcDNA3 vector control or with galectin-3 AS, maintained in culture for 10 days, and then fixed and stained with crystal violet to better visualize the transformed foci. Typical dishes of control and galectin-3 AS transfectants are shown. Representative foci are shown in the insets. Quantitation of the dishes (n = 4) indicated that galectin-3 AS induced a marked increase in the number of foci (see text).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Until recently, galectins were viewed as bona fide lectins that operate extracellularly by interacting with proteoglycans (36, 37). There is also significant evidence for the functional importance of intracellular galectin-1 and galectin-3 (21, 22, 38, 39). The present study discloses a hitherto unknown intracellular function of galectin-3. Galectin-3 interacts preferentially with activated K-Ras rather than with activated H-Ras or N-Ras proteins, shifts the basal and the EGF-stimulated K-Ras-GTP/K-Ras-GDP steady state toward K-Ras-GTP, thereby prolonging the EGF-induced response, and reduces Ras-GTP hydrolysis facilitated by p120RasGAP. Galectin-3-AS RNA down-regulates galectin-3 levels and strongly reduces the basal and EGF-induced increase in K-Ras-GTP without affecting H-Ras-GTP levels. Overexpression of galectin-3 augments activation of K-Ras (but not H-Ras) and K-Ras activation of PI3-K and Raf-1 and simultaneously attenuates and curtails K-Ras-mediated activation of ERK. On the other hand, galectin-3 AS RNA enhances and prolongs the EGF-induced K-Ras but not EGF-induced H-Ras activation of ERK. Taken together, these data suggest that galectin-3 serves as a selective binding partner of activated K-Ras and that this partnership determines the strength, duration, and selectivity of the K-Ras signal; galectin-3 seems to increase K-Ras signaling to PI3-K Raf-1 as well as to a third effector pathway that attenuates ERK activation. The nature of this last inhibitory pathway is not yet known. In light of the well documented regulation of ERK by ERK phosphatases, including protein phosphatase 2A and mitogen-activated protein kinase phosphatases (40–42), it is tempting to speculate that K-Ras-GTP/galectin-3 complexes attenuate ERK activation via a pathway that up-regulates ERK phosphatases. Nonetheless, relief of the inhibitory signal by galectin-3 AS (Fig. 7A) clearly enhanced K-Ras(G12V) transformation in NIH3T3 cells (Fig. 8). This demonstrated the biological relevance of the K-Ras/galectin-3-inhibitory signal and of K-Ras galectin-3 interactions. It seems that the enhancement of Ras transforming activity by galectin-3 AS in NIH3T3 cells is in accord with the expected outcome of enhanced Ras/Raf/mitogen-activated protein kinase kinase/ERK signaling operating together with other signals such as Ras/PI3-K. We showed that K-Ras(G12V)-GTP signals to PI3-K well also without galectin-3 overexpression (Fig. 6, A and B), yet without galectin-3 expression, ERK activation is enhanced (Fig. 7). Hence, the augmented transformation in NIH3T3/galectin-3 AS transfectants seems to be, at least in part, a result of enhanced ERK activity combined with significant PI3-K activity. This suggestion is in accord with earlier studies demonstrating that in NIH3T3 cells, active Raf synergistically enhances transformation by H-Ras (G12V/Y40C), a Ras effector-domain mutant that activates PI3-K only (35).

In line with earlier experiments demonstrating that galectin-1 is a Ras-binding partner (21, 22), the present experiments show that galectin-3 is a Ras escort protein with a significant impact on K-Ras signaling. Like galectin-1 (22), which interacts and stabilizes H-Ras-GTP or K-Ras-GTP, galectin-3 interacts preferentially with K-Ras-GTP and stabilizes it in its active state. In addition, like H-Ras/galectin-1 interactions (21), K-Ras/galectin-3 interactions seem to depend, at least in part, on the farnesyl group of the Ras proteins (Fig. 1D). This is in accord with the recent identification of a hydrophobic cavity in galectin-1 that can accommodate a farnesyl group (43) analogous to cavities found in RhoGDI that bind the geranylgeranyl group of Cdc42 (44) and in phosphodiesterase {delta} that bind the prenylated phosphodiesterase {alpha}{beta} (45–47) and the farnesylated Ras protein (48, 49). However, unlike galectin-1, galectin-3 seems to be a selective binding partner of K-Ras. This suggests that the interactions of galectin-1 with Ras are less discriminatory than those of galectin-3 and Ras. The apparent selectivity of galectin-3 toward K-Ras might be related to structural differences between galectin-1 and galectin-3 and differences between the hypervariable C-terminal domains of H-Ras and K-Ras proteins. Although galectin-1 and galectin-3 share a highly homologous carbohydrate-binding domain, galectin-3 also possesses a large N-terminal domain of 110–130 amino acid residues, depending on species (36). This marked difference between the two galectins suggests that the N-terminal domain of galectin-3 may contribute to the preferential interaction of this protein with activated K-Ras. On the other hand, it is possible that the N-terminal domain of galectin-3 hinders the binding of galectin-3 to activated H-Ras but not to activated K-Ras. In any case, the Ras C-terminal hypervariable domain that differentiates H-Ras from K-Ras seems to be a reasonable candidate as a region determining the preferential interaction of active K-Ras with galectin-3 and lack of its interaction with H-Ras.

The discriminating interactions of active H-Ras with galectin-1 and of active K-Ras with galectin-3 might affect the membrane microdomain localization of the GDP- and GTP-bound forms of these Ras isoforms and the segregation of Ras isoforms from each other. There is evidence to support these possibilities. Previous studies showed that membrane anchorage of the activated GFP-H-Ras(G12V) is substantially diminished in the absence of galectin-1 expression (21). However, membrane anchorage of GFP-H-Ras(wt) is not dependent on galectin-1 (21). Because H-Ras-GDP is enriched within membrane rafts and H-Ras-GTP is enriched outside of rafts, we suggested that galectin-1 might stabilize the membrane association of the activated Ras outside of rafts (22). Recent studies indeed showed that H-Ras-GTP is co-localized with galectin-1 in nonraft domains in the inner leaflet of the cell membrane (7). Moreover, it was shown in that study that activated H-Ras and K-Ras occupy distinct microdomains (7). Those observations, together with the present finding that galectin-3 is selectively associated with activated K-Ras, suggest that galectin-3 might stabilize the activated K-Ras in a unique membrane microdomain that differs from that of H-Ras and galectin-1.

Our results also demonstrate that selectivity of Ras signaling might be achieved under physiologically relevant conditions through interactions of Ras proteins with natural constituents of a normal cell. Association of galectin-1 with activated H-Ras or K-Ras confers on these Ras isoforms a conformation that mimics the Ras(T35S) effector loop mutant (22) that binds to and activates Raf-1 but not PI3-K (50, 51). Therefore, high levels of galectin-1 expression result in H-Ras or K-Ras activation of the Raf-1-mitogen-activated protein kinase kinase-ERK pathway, while Ras activation of PI3-K is diminished (22). Galectin-3/K-Ras interactions provide a distinct mode of selectivity for the Ras signal. Galectin-3 associated with activated K-Ras promotes strong activation of both Raf-1 and PI3-K and at the same time endorses a third Ras signal that attenuates active ERK. It is possible that levels of galectin-1 and -3, which are known to vary among normal and cancer cells (36, 52), define outputs of oncogenic K-Ras, the most important Ras oncoprotein in human tumors (2, 4).


   FOOTNOTES
 
* This research was supported by The Israel Science Foundation Grant 339/02-1. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} To whom correspondence should be addressed. Tel.: 972-3-640-9699; Fax: 972-3-640-7643; E-mail: kloog{at}post.tau.ac.il.

1 The abbreviations used are: GEF, guanine nucleotide exchange factor; PI3-K, phosphoinositide 3-kinase; HEK, human embryonic kidney; Ab, antibody; DTT, dithiothreitol; wt, wild type; EGF, epidermal growth factor; GFP, green fluorescent protein; ERK, extracellular signal-regulated kinase; AS, antisense. Back


   ACKNOWLEDGMENTS
 
We thank S. Smith for editorial assistance.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Bar-Sagi, D., and Hall, A. (2000) Cell 103, 227-238[CrossRef][Medline] [Order article via Infotrieve]
  2. Downward, J. (2003) Nat. Rev. Cancer 3, 11-22[CrossRef][Medline] [Order article via Infotrieve]
  3. Hancock, J. F. (2003) Nat. Rev. Mol. Cell. Biol. 4, 373-385[CrossRef][Medline] [Order article via Infotrieve]
  4. Shields, J. M., Pruitt, K., McFall, A., Shaub, A., and Der, C. J. (2000) Trends Cell Biol. 10, 147-154[CrossRef][Medline] [Order article via Infotrieve]
  5. Choy, E., Chiu, V. K., Silletti, J., Feoktistov, M., Morimoto, T., Michaelson, D., Ivanov, I. E., and Philips, M. R. (1999) Cell 98, 69-80[CrossRef][Medline] [Order article via Infotrieve]
  6. Prior, I. A., Harding, A., Yan, J., Sluimer, J., Parton, R. J., and Hancock, J. F. (2001) Nat. Cell Biol. 3, 368-375[CrossRef][Medline] [Order article via Infotrieve]
  7. Prior, I. A., Muncke, C., Parton, R. G., and Hancock, J. F. (2003) J. Cell Biol. 160, 165-170[Abstract/Free Full Text]
  8. Umanoff, H., Edelmann, W., Pellicer, A., and Kucherlapati, R. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 1709-1713[Abstract/Free Full Text]
  9. Johnson, L., Greenbaum, D., Cichowski, K., Mercer, K., Murphy, E., Schmitt, E., Bronson, R. T., Umanoff, H., Edelmann, W., Kucherlapati, R., and Jacks, T. (1997) Genes Dev. 11, 2468-2481[Abstract/Free Full Text]
  10. Liao, J., Wolfman, J. C., and Wolfman, A. (2003) J. Biol. Chem. 278, 31871-31878[Abstract/Free Full Text]
  11. Serrano, M., Lin, A. W., McCurrach, M. E., Beach, D., and Lowe, S. W. (1997) Cell 88, 593-602[CrossRef][Medline] [Order article via Infotrieve]
  12. Hamad, N. M., Elconin, J. H., Karnoub, A. E., Bai, W., Rich, J. N., Abraham, R. T., Der, C. J., and Counter, C. M. (2002) Genes Dev. 16, 2045-2057[Abstract/Free Full Text]
  13. Goi, T., Rusanescu, G., Urano, T., and Feig, L. A. (1999) Mol. Cell. Biol. 19, 1731-1741[Abstract/Free Full Text]
  14. Yip-Schneider, M. T., Lin, A., Barnard, D., Sweeney, C. J., and Marshall, M. S. (1999) Int. J. Oncol. 15, 271-279[Medline] [Order article via Infotrieve]
  15. Niv, H., Gutman, O., Kloog, Y., and Henis, Y. I. (2002) J. Cell Biol. 157, 865-872[Abstract/Free Full Text]
  16. Jiang, X., and Sorkin, A. (2002) Mol. Biol. Cell 13, 1522-1535[Abstract/Free Full Text]
  17. Chiu, V. K., Bivona, T., Hach, A., Sajous, J. B., Silletti, J., Wiener, H., Johnson, R. L., 2nd, Cox, A. D., and Philips, M. R. (2002) Nat. Cell Biol. 4, 343-350[Medline] [Order article via Infotrieve]
  18. Denis, G. V., Yu, Q., Ma, P., Deeds, L., Faller, D. V., and Chen, C. Y. (2003) J. Biol. Chem. 278, 5775-5785[Abstract/Free Full Text]
  19. Rusanescu, G., Gotoh, T., Tian, X., and Feig, L. A. (2001) Mol. Cell. Biol. 21, 2650-2658[Abstract/Free Full Text]
  20. Lahm, H., Andre, S., Hoeflich, A., Fischer, J. R., Sordat, B., Kaltner, H., Wolf, E., and Gabius, H. J. (2001) J. Cancer Res. Clin. Oncol. 127, 375-386[CrossRef][Medline] [Order article via Infotrieve]
  21. Paz, A., Haklai, R., Elad, G., Ballan, E., and Kloog, Y. (2000) Oncogene 20, 7486-7493
  22. Elad-Sfadia, G., Haklai, R., Ballan, E., Gabius, H. J., and Kloog, Y. (2002) J. Biol. Chem. 277, 37169-37175[Abstract/Free Full Text]
  23. Haklai, R., Gana-Weisz, G., Elad, G., Paz, A., Marciano, D., Egozi, Y., Ben Baruch, G., and Kloog, Y. (1998) Biochemistry 37, 1306-1314[CrossRef][Medline] [Order article via Infotrieve]
  24. Jansen, B., Schlagbauer-Wadl, H., Kahr, H., Ress, E., Mayer, B., Eichler, H. G., H., P., Gana-Weisz, M., Ben-David, E., Kloog, Y., and Wolff, K. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 14019-14024[Abstract/Free Full Text]
  25. Elad, G., Paz, A., Haklai, R., Marciano, D., Cox, A., and Kloog, Y. (1999) Biochim. Biophys. Acta 1452, 228-242[Medline] [Order article via Infotrieve]
  26. Pardo, O. E., Arcaro, A., Salerno, G., Raguz, S., Downward, J., and Seckl, M. J. (2002) J. Biol. Chem. 277, 12040-12046[Abstract/Free Full Text]
  27. Camby, I., Belot, N., Rorive, S., Lefranc, F., Maurage, C. A., Lahm, H., Kaltner, H., Hadari, Y., Ruchoux, M. M., Brotchi, J., Zick, Y., Salmon, I., Gabius, H. J., and Kiss, R. (2001) Brain Pathol. 11, 12-26[Medline] [Order article via Infotrieve]
  28. Danguy, A., Camby, I., and Kiss, R. (2002) Biochim. Biophys. Acta 1572, 285-293[Medline] [Order article via Infotrieve]
  29. Berberat, P. O., Friess, H., Wang, L., Zhu, Z., Bley, T., Frigeri, L., Zimmermann, A., and Buchler, M. W. (2001) J. Histochem. Cytochem. 49, 539-549[Abstract/Free Full Text]
  30. van den Brule, F. A., Waltregny, D., and Castronovo, V. (2001) J. Pathol. 193, 80-87[CrossRef][Medline] [Order article via Infotrieve]
  31. Hubert, M., Wang, S. Y., Wang, J. L., Seve, A. P., and Hubert, J. (1995) Exp. Cell Res. 220, 397-406[CrossRef][Medline] [Order article via Infotrieve]
  32. Fridman, M., Maruta, H., Gonez, J., Walker, F., Treutlein, H., Zeng, J., and Burgess, A. (2000) J. Biol. Chem. 275, 30363-30371[Abstract/Free Full Text]
  33. Bresalier, R. S., Yan, P. S., Byrd, J. C., Lotan, R., and Raz, A. (1997) Cancer 80, 776-787[CrossRef][Medline] [Order article via Infotrieve]
  34. Rodriguez Viciana, P., Warne, P. H., Vanhaesebroeck, B., Waterfield, M. D., and Downward, J. (1996) EMBO J. 15, 2442-2451[Medline] [Order article via Infotrieve]
  35. Khosravi Far, R., White, M. A., Westwick, J. K., Solski, P. A., Chrzanowska Wodnicka, M., Van Aelst, L., Wigler, M. H., and Der, C. J. (1996) Mol. Cell. Biol. 16, 3923-3933[Abstract]
  36. Hughes, R. C. (1999) Biochim. Biophys. Acta 1473, 172-185[Medline] [Order article via Infotrieve]
  37. Perillo, N. L., Marcus, M. E., and Baum, L. G. (1998) J. Mol. Med. 76, 402-412[CrossRef][Medline] [Order article via Infotrieve]
  38. Liu, F. T., Patterson, R. J., and Wang, J. L. (2002) Biochim. Biophys. Acta 1572, 263-273[Medline] [Order article via Infotrieve]
  39. Akahani, S., Nangia-Makker, P., Inohara, H., Kim, H. R., and Raz, A. (1997) Cancer Res. 57, 5272-5276[Abstract/Free Full Text]
  40. Alessi, D. R., Gomez, N., Moorhead, G., Lewis, T., Keyse, S. M., and Cohen, P. (1995) Curr. Biol. 5, 283-295[CrossRef][Medline] [Order article via Infotrieve]
  41. Camps, M., Nichols, A., and Arkinstall, S. (2000) FASEB J 14, 6-16[Abstract/Free Full Text]
  42. Pulido, R., Zuniga, A., and Ullrich, A. (1998) EMBO J. 17, 7337-7350[CrossRef][Medline] [Order article via Infotrieve]
  43. Rotblat, B., Niv, H., Andre, S., Kaltner, H., Gabius, H. J., and Kloog, Y. (2004) Cancer Res. 64, 3112-3118[Abstract/Free Full Text]
  44. Hoffman, G. R., Nassar, N., and Cerione, R. A. (2000) Cell 100, 345-356[CrossRef][Medline] [Order article via Infotrieve]
  45. Cook, T. A., Ghomashchi, F., Gelb, M. H., Florio, S. K., and Beavo, J. A. (2000) Biochemistry 39, 13516-13523[CrossRef][Medline] [Order article via Infotrieve]
  46. Gillespie, P. G., Prusti, R. K., Apel, E. D., and Beavo, J. A. (1989) J. Biol. Chem. 264, 12187-12193[Abstract/Free Full Text]
  47. Florio, S. K., Prusti, R. K., and Beavo, J. A. (1996) J. Biol. Chem. 271, 24036-24047[Abstract/Free Full Text]
  48. de Gunzburg, J., Riehl, R., and Weinberg, R. A. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 4007-4011[Abstract/Free Full Text]
  49. Hanzal-Bayer, M., Renault, L., Roversi, P., Wittinghofer, A., and Hillig, R. C. (2002) EMBO J. 21, 2095-2106[CrossRef][Medline] [Order article via Infotrieve]
  50. Joneson, T., White, M. A., Wigler, M. H., and Bar Sagi, D. (1996) Science 271, 810-812[Abstract]
  51. White, M. A., Nicolette, C., Minden, A., Polverino, A., Van Ael