If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
* This work was supported by Cancer Research Campaign (United Kingdom) funds (to A. H.), a Natural Sciences and Engineering Research Council of Canada fellowship (to M. F. O.), National Institutes of Health Grants GM47458 (to R. A. C.) and R29CA62029 (to D. T.), and an American Cancer Society Faculty Award (to D. T.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ∥ Current address: Dept. of Physiology, Tufts University, Boston, MA 02111.
The lbc oncogene is tumorigenic in nude mice, transforms NIH 3T3 fibroblasts, and encodes a Dbl homology domain found in several transforming gene products including the dbl oncogene product. While both lbc- and dbl-transformed NIH 3T3 foci exhibited a comparable gross appearance, lbc-transformed cell morphology was clearly distinct from that of dbl-transformed cells. Given these differences, we investigated the biochemical activity and target specificity of the Lbc oncoprotein both in vivo and in vitro. Here we show that Lbc associates specifically with the GTP-binding protein Rho in vivo, but not with the Ras, Rac, or Cdc42Hs GTP-binding proteins, and that recombinant, affinity-purified Lbc specifically catalyzes the guanine-nucleotide exchange activity of Rho in vitro. Consistent with an in vivo role for Lbc in Rho regulation, we further demonstrate that micro-injected onco- lbc potently induces actin stress fiber formation in quiescent Swiss 3T3 fibroblasts indistinguishable from that induced by Rho. Finally, lbc-induced NIH 3T3 focus formation is inhibited by co-transfection with a rho dominant-negative mutant. These results strongly indicate that the lbc oncogene encodes a specific guanine nucleotide exchange factor for Rho and causes cellular transformation through activation of the Rho signaling pathway.
The Rho-related GTP-binding proteins function as molecular switches in a diversity of cellular signaling pathways, many of which influence the cellular cytoskeleton, as well as cell polarity and motility (
). Members of this family of GTP-binding proteins include the Rho proteins (RhoA, RhoB, RhoC, and RhoG) and the Rac1, Rac2, Cdc42, and TC10 proteins. As is the case for all members of the Ras superfamily of GTP-binding proteins, the GTP-binding/GTPase cycles of the Rho-related proteins are tightly controlled, with guanine nucleotide exchange factors (GEFs)
). There now appear to be a number of proteins, suspected to be involved in cell growth regulation, that share significant sequence similarity with oncogenic Dbl within an ~250-amino acid region termed the Dbl homology (DH) domain. These include the breakpoint-cluster region (Bcr) protein (
), acts as a highly specific GEF for the Rho proteins. In addition, we show that Lbc specifically associates with Rho in vivo and potently induces actin stress fiber formation in fibroblasts, similar to activated forms of Rho. Finally, we demonstrate that Lbc-induced cellular transformation can be blocked by a dominant-negative Rho mutant that would be predicted to bind with high affinity to GEFs.
Cell Culture and Western Blot Analysis
Stable lbc-transformed NIH 3T3 transfectants were derived as previously detailed (
). Flag: lbc-transfected NIH 3T3 cells (confluent) were washed once with ice-cold buffer A containing 20 m M Tris-HCl, pH 7.4, 100 m M NaCl, 1 m M EDTA, 1 m M phenylmethylsulfonyl fluoride, 10 μg/ml leupeptin and aprotinin before lysing with 0.4% Triton X-100 in buffer A for 30 min at 4°C. The clarified supernatant was either directly subjected to SDS-polyacrylamide gel electrophoresis (lysate) or incubated with 3 μg of M5 anti-Flag monoclonal antibody (IBI) for 1 h, followed by 2 h of additional incubation, after addition of 3 μg of rabbit anti-mouse antibody precoupled to anti-rabbit IgG-agarose beads (Sigma). Duplicate samples in which the M5 antibody was omitted were used as controls. The immunocomplexes were collected by centrifugation and washed three times with buffer A. The immunoprecipitates were blotted either with M5 antibody or with anti-Ras, -RhoA, -Rac1, and -Cdc42Hs antibodies (Santa Cruz Biotech), respectively. The primary antibodies were detected with anti-mouse, anti-rabbit, or anti-rat immunoglobulins coupled to horseradish peroxidase. The bands were visualized by the ECL system (Amersham Corp.).
Expression of Glutathione S-Transferase Fusion Proteins
The GST-RhoA, GST-Cdc42Hs, GST-Rac1, GST-TC10, and GST-RhoG were expressed in E. coli as described previously (
). The GST-Ha-Ras construct and RhoB were kind gifts from Drs. H. Maruta (University of Melbourne) and J. Settleman (Harvard Medical School). RhoC was expressed as a GST fusion protein by inserting the cDNA encoding RhoC (using the polymerase chain reaction) into the pGEX-2T vector (Pharmacia Biotech Inc.). The GST fusion proteins were purified as described (
) and used without removal of the GST for the in vitro GDP/GTP exchange assays. A 1.3-kilobase BamHI- EcoRI fragment of the cDNA containing the entire coding sequence for lbc was ligated with a 600-base pair XbaI- BamHI fragment of the cDNA encoding the GST protein and inserted into the baculovirus transfer vector pVL1392. Recombinant virus was generated according to Ref. 9. Sf9 cells were harvested 70 h after infection and the cell pellets were resuspended in ice-cold buffer A supplemented with 0.4% Triton X-100, after a quick freeze-thaw. The resuspended cells were homogenized with a Dounce homogenizer and then centrifuged for 15 min at 4°C in a microcentrifuge. The supernatants were further incubated with glutathione-agarose beads (Sigma) for 2 h, and the beads were washed four times before the GST fusion proteins were eluted with 10 m M glutathione in buffer A. Excess glutathione was removed from the GST-Lbc and GST-Dbl proteins by dialysis.
GDP/GTP Exchange Assays
The GDP dissociation and GTP binding assays were carried out at 24°C as described (
) was microinjected at 100 ng/μl into the nuclei of serum-starved quiescent Swiss 3T3 cells. After 40 h, cells were fixed in 4% paraformaldehyde, permeabilized in 0.2% Triton X-100, treated with 1 mg/ml sodium borohydride, and blocked with 10% fetal calf serum, 0.1% bovine serum albumin. Actin filaments were detected with TRITC-labeled phalloidin as described previously (
). Lbc expression was detected by incubating with a 1:250 dilution of rabbit anti-Lbc antibody followed by a 1:300 dilution of fluorescein isothiocyanate-labeled goat anti-rabbit IgG antibody. C3 transferase at 0.1 mg/ml and rat IgG at 0.5 mg/ml were co-microinjected into cells that had been microinjected 40 h previously with lbc cDNA. Cells were then incubated for an additional 10 min. C3- and rat IgG-injected cells were identified using a 1:100 dilution of Cascade Blue-labeled goat anti-rat IgG antibody. Cells were photographed using a Zeiss Axiophot microscope with Kodak TMY-400 film.
Focus Formation Assay
NIH 3T3 cells were transfected using calcium phosphate precipitation according to standard procedures detailed previously (
) with 15 ng of pSRα lbc 9a cDNA either alone or together with 300 ng of the rhoA mutant encoding asparagine at position 19 (instead of threonine) or with the wild type rhoA cDNA in pZipneo in the presence of 20 μg of genomic NIH 3T3 carrier DNA/dish. After day 14, the plates were stained with crystal violet and foci were counted.
RESULTS AND DISCUSSION
The lbc cDNA predicts a 424-amino acid protein and encodes a DH (Dbl homology) domain that is present in oncogenic Dbl and a number of other potential growth regulatory proteins (
). Since the Lbc DH domain shares the highest degree of sequence similarity (~24% identity, ~42% homology) with the Dbl oncoprotein, we compared the morphology of lbc-transfected cells with that of dbl-transfected cells. lbc transfectants exhibited an obvious transformed phenotype of an intersecting bundle- and stretching spindle-form (Fig. 1, compare A and B), and subconfluent cells had a higher cytoplasm to nucleus ratio than dbl-transfected cells and displayed striking “feet-like” projections (see arrows in Fig. 1C) similar to that described for cells overexpressing the Rho GTP-binding protein (
). dbl-overexpressing cells showed a clearly distinctive morphology, having a more rounded shape and growing to a higher density in a more detached manner; the characteristic giant multinuclear cells (
) were observed (Fig. 1D). In contrast, while multinuclear lbc-transformed cells were also noted, giant syncytia were not as readily apparent. These subtle yet distinct morphological differences suggested that the mechanism of lbc transformation may be different from that caused by dbl.
Given that Dbl and Cdc24 have been recently shown to serve as guanine nucleotide exchange factors (GEFs) (
), we set out to determine whether the biological activity of the Lbc oncoprotein is manifested by its ability to interact with low molecular weight GTP-binding proteins. To test whether Lbc can directly bind to Ras or Rho family GTP-binding proteins in vivo, we utilized a stable NIH 3T3 transfectant cell line transformed by carboxyl-terminal Flag-tagged lbc cDNA in pSRneo plasmid. Western blotting of transfectant cell lysates detected the presence of a 49-kDa Lbc-Flag band and various small GTP-binding proteins such as Ha-Ras, Rho, Rac, and Cdc42Hs (Fig. 2; first lane in all panels). When the anti-Flag antibody was used toimmunoprecipitate Lbc from the lysates, and the precipitates were Western blotted with anti-Ras, -Rho, -Rac, and -Cdc42Hs antibodies, only Rho was observed to co-precipitate with Lbc (Fig. 2, second lane in all panels). Typically, ~10% of the total Rho and ~60% of the Lbc-Flag protein detected in the whole cell lysates were present in the immunoprecipitates. Thus, despite the high degree of sequence similarity (50-70%) among the different members of the Rho subfamily that were examined, Lbc appears to exclusively associate with the Rho GTP-binding protein in vivo.
We next examined whether Lbc acts as a GEF toward Rho and other members of the Rho subfamily. To do this, the transforming Lbc protein was expressed as a glutathione S-transferase (GST) fusion product in Sf9 insect cells using a baculovirus expression system, and purified by glutathione-agarose affinity chromatography (Fig. 3 A, lane 2). The ability of purified GST-Lbc to stimulate guanine-nucleotide exchange on Rho was measured with the recombinant RhoA protein. The rate of dissociation of [3H]GDP from RhoA was stimulated ~10-fold by the GST-Lbc protein (Fig. 3B). Similarly, the rate of [35S]GTPγS binding, which directly reflects the GDP-GTPγS exchange activity of RhoA, was also stimulated ~10-fold by GST-Lbc (Fig. 3C). Therefore, Lbc does act as a GEF. Lbc also showed comparable effects on the guanine nucleotide exchange activities of the recombinant RhoB and RhoC proteins (data not shown). Fig. 3D compares the abilities of purified GST-Lbc to stimulate GDP dissociation from recombinant Ha-Ras, RhoA, and its close homologs Rac1, Cdc42Hs, TC10, and RhoG. Unlike Dbl, which stimulates the rate of GDP dissociation from both Cdc42Hs and RhoA (
) without affecting Ras, Rac1, and TC10, Lbc specifically accelerated the rate of GDP dissociation from RhoA while showing no detectable effects on GDP dissociation from Cdc42Hs or from Ras, Rac1, or TC10 (Fig. 3 D). The Ras-GRF, which specifically catalyzes Ras GDP/GTP exchange (
), was used as a positive control for stimulated GDP dissociation from Ras, while oncogenic Dbl served as a positive control for a Cdc42Hs-GEF. We also tested for possible regulatory effects of Lbc on the GTPase activities of the above GTP-binding proteins, and found that Lbc had no effect on their GTP hydrolytic activities (data not shown). Taken together, these results indicate that oncogenic Lbc functions as a specific GEF for Rho by stimulating the release of its tightly bound GDP.
In mammalian cells, RhoA regulates actin stress fiber formation and focal adhesion assembly following growth factor stimulation (
). Consequently, the above results would suggest that Lbc may induce the same cytoskeletal changes as Rho. In order to test this, the nuclei of serum-starved quiescent Swiss 3T3 cells were microinjected with onco- lbc cDNA in the same mammalian expression vector used for focus formation. lbc expression (Fig. 4 A) was found to be associated with the formation of actin stress fibers (Fig. 4B), a response typical of Rho stimulation (
), we further predicted that the dominant-negative RhoA mutant, RhoA-T19N, where the threonine at position 19 is changed to an asparagine, would bind to a Rho-GEF (i.e. Lbc), and thereby prevent Lbc from activating endogenous Rho proteins and inhibit Lbc-mediated transformation of NIH 3T3 cells. The results presented in Fig. 4E support this prediction. Co-expression of the lbc oncogene with RhoA-T19N caused up to 80% reduction in the number of foci induced by lbc, whereas no effect was observed when lbc was co-expressed with the cDNA encoding wild type RhoA (Fig. 4 E). Thus, these results are consistent with the findings that the Lbc and Rho proteins interact functionally in vivo and that Lbc-induced transformation proceeds through its ability to stimulate the guanine nucleotide exchange activity of Rho.
Members of the Dbl family share the common structural feature of the DH domain, a region of ~250 amino acids, and there is considerable evidence that the DH domain serves as a functional unit for GEF activity for Rho subfamily GTP-binding proteins. For example, the dbl oncogene product has in vitro GEF activity for the Cdc42Hs and RhoA proteins (
), and is associated with their encoded CDC25 domains, which are clearly distinct from the DH domain. At present, the DH domains within the Sos, Ras-GRF, and Bcr proteins have not been associated with any specific activity. While RhoA appears to be the target for both Lbc and Dbl GEF activity, their mutually exclusive tissue expression (lbc is expressed in hematopoietic cells, muscle, heart and lung (Ref. 14); dbl is expressed in adrenal gland, testes, and fetal brain (Ref.
)) preclude an overlapping function of these two oncogenes in vivo. Thus, the ability of the Lbc oncoprotein to bind to Rho and stimulate its guanine nucleotide exchange activity provides the first demonstration of an interaction between a DH domain-containing protein and a low molecular mass GTP-binding protein that is specific, can occur in vivo, and has an impact on the transformation capability of an oncogene, as illustrated by the results of co-transfection with the dominant-negative rho mutant.
The observation that Lbc potently induces actin re-organization also provides the first direct demonstration of a DH domain-encoding protein which regulates cytoskeletal changes in vivo, and provides a physiological role for Lbc that is corroborated by the biochemical data presented here. Members of the Rho subgroup of Ras-related GTP-binding proteins are known to be essential in yeast (
). The current studies showing that the potent oncogene lbc encodes a specific GEF for Rho suggest a mechanism by which a growth regulatory protein can directly activate Rho and contribute to the full complement of effects associated with cellular transformation, namely cytoskeletal changes that result in the loss of contact-growth inhibition.
We thank Dr. Alessandra Eva for dbl-transfected cells, Drs. Roya Khosravi-Far and Channing J. Der for pZipneo/ rho and pZipneo/N19 rho plasmids, Peggy Atkinson for guidance in preparing insect cell-expressed recombinant GST-Lbc, and Cindy Westmiller for expert technical assistance.