Isolation of ras GTP-binding mutants using an in situ colony-binding assay.

We have developed a strategy to isolate mutant ras genes encoding proteins defective in GTP binding. Random in vitro mutagenesis of a v-Harvey (Ha)-ras expression vector was followed by an in situ GTP-binding assay on lysed bacterial colonies. Single amino acid substitutions at ras codon 83, 119, or 144 decreased the affinity of p21 for GTP by a factor of 25-100 primarily as a consequence of increased rates of dissociation of GTP from p21. Nevertheless, these mutant genes induced transformation of NIH 3T3 cells with efficiencies comparable to wild-type v-Ha-ras. In transformed cells, mutant p21s were phosphorylated to a degree similar to that of wild-type v-Ha-ras p21, suggesting that a decrease in affinity by a factor of 100 did not prevent the mutant ras protein from binding GTP in vivo. These results are discussed with respect to the role of GTP in the regulation of p21 function.     

X-ray crystal structures of transforming p21 ras mutants suggest a transition-state stabilization mechanism for GTP hydrolysis.

RAS genes isolated from human tumors often have mutations at positions corresponding to amino acid 12 or 61 of the encoded protein (p21), while retroviral ras-encoded p21 contains substitutions at both positions 12 and 59. These mutant proteins are deficient in their GTP hydrolysis activity, and this loss of activity is linked to their transforming potential. The crystal structures of the mutant proteins are presented here as either GDP-bound or GTP-analogue-bound complexes. Based on these structures, a mechanism for the p21 GTPase reaction is proposed that is consistent with the observed structural and biochemical data. The central feature of this mechanism is a specific stabilization complex formed between the Gln-61 side-chain and the pentavalent gamma-phosphate of the GTP transition state. Amino acids other than glutamine at position 61 cannot stabilize the transition state, and amino acids larger than glycine at position 12 would interfere with the transition-state complex. Thr-59 disrupts the normal position of residue 61, thus preventing its participation in the transition-state complex.     

Biochemical and crystallographic characterization of a complex of c-Ha-ras p21 and caged GTP with flash photolysis.

The GTP binding domain of the c-Ha-ras protooncogene product (p21'c) and the corresponding region from an oncogenic mutant form of the protein in which glycine at position 12 has been replaced by valine [p21'(G12V)] have been crystallized with P3-1-(2-nitro)phenylethylguanosine 5'-O-triphosphate (caged GTP) at their active sites. The crystals give x-ray diffraction patterns to a resolution of better than 0.3 nm. Photolysis can be achieved in the crystal, after which GTP hydrolysis takes place at the rate expected from solution studies. Complete x-ray data sets have been obtained for the starting caged-GTP state and the final GDP state after photolysis and hydrolysis, demonstrating the feasibility of time-resolved structural investigations of the process of GTP hydrolysis.     

Intrinsic GTPase activity distinguishes normal and oncogenic ras p21 molecules.

The 21-kilodalton protein (p21) encoded by normal cellular Harvey-ras has been expressed in Escherichia coli as a fusion protein by using the pUC8 vector and has been purified to greater than 95% homogeneity by ion-exchange chromatography and gel filtration. The purified protein molecules possess intrinsic GTPase activity on the basis of the following criteria: (i) elution of the GTPase activity with p21 GDP-binding activity in two different chromatography systems, (ii) parallel thermal inactivation of GTPase activity and p21 GTP-binding activity, and (iii) immunoprecipitation of the GTPase activity with monoclonal antibodies to p21. At 37 degrees C, the rate of GTP hydrolysis by the purified normal p21 assayed in solution was 5.3-6.6 mmol/min per mol of p21. The rate of GTP hydrolysis by a form of p21 [Val12] encoded by a human oncogene was significantly lower (1.4-1.9 mmol/min per mol of p21). The presence of a threonine phosphate acceptor site at residue 59 also decreased p21 GTPase activity. For regulatory proteins that use GTP as part of their biochemical mechanism, the hydrolysis of GTP to GDP reverses the biological activity of the respective proteins. The observation that oncogenic forms of p21 lose GTPase activity suggests that GTP hydrolysis may be a biochemical event that inactivates the growth-promoting effects of a p21 X GTP complex.     

Antibodies specific for amino acid 12 of the ras oncogene product inhibit GTP binding.

An antibody (anti-p21ser) was raised against a ras p21-related synthetic peptide and was able to recognize specifically the substitution of serine for glycine at amino acid 12 of p21. This substitution causes oncogenic activation of p21. Anti-p21ser was found to immunoprecipitate v-Ki-ras p21 and to strongly inhibit its ability to autophosphorylate and to bind GTP in an immunoabsorption assay. Furthermore, binding of the antibody to p21 was specifically inhibited by GTP or GDP, suggesting that amino acids around position 12 are part of the GTP/GDP binding site. These results, taken together with the observation that the microinjection of anti-p21ser into cells transformed by v-Ki-ras p21 causes a transient reversion of the cells to a normal phenotype [Feramisco, J. R., Clark, R., Wong, G., Arnheim, N., Milley, R. & McCormick, F. (1985) Nature (London) 314, 639-642], support the idea that interaction of p21 with guanine nucleotides is crucial to the transforming function of this protein.     

Spontaneous activation of a human proto-oncogene.

It has been recently shown that malignant activation of the c-has/bas proto-oncogene in T24 human bladder carcinoma cells was mediated by a single point mutation. A deoxyguanosine located at position 35 of the first exon of this proto-oncogene was substituted by thymidine. These findings predicted that the resulting oncogene would code for a structurally altered p21 protein containing valine instead of glycine as its 12th amino acid residue. We now report the spontaneous activation of the human c-has/bas proto-oncogene during transfection of NIH/3T3 cells. As in T24 cells, this in vitro activated oncogene also acquired malignant properties by a single point mutation. In this case we have detected a G leads to A transition, which occurred at the same position as the mutation responsible for the activation of the T24 oncogene. These results predict that the p21 protein coded for by the spontaneously activated c-has/bas gene will incorporate aspartic acid as its 12th amino acid residue. Computer analysis of the secondary structure of c-has/bas encoded p21 proteins indicates that substitution of the glycine residue located at position 12, not only by aspartic acid or valine but also by any other amino acid, would result in the same structural alteration. These findings indicate that a specific conformational change is sufficient to confer transforming properties to this p21 protein. Moreover, they predict that any mutation affecting the coding properties of the 12th codon of the c-has/bas proto-oncogene will lead to its malignant activation.     

Purification of ras GTPase activating protein from bovine brain.

In cytosolic extracts of bovine brain, we detected ras GTPase activating protein (GAP) activity that stimulated the GTP hydrolytic activity of normal c-Ha-ras p21 but not that of the oncogenic [Val12]p21 variant. GAP was purified 19,500-fold by a five-column procedure involving DEAE-Sephacel, Sepharose 6B, orange dye and green dye matrices, and Mono Q resins. A single major protein band of 125 kDa was observed on NaDodSO4/polyacrylamide gels that correlated with the elution of GAP activity on Mono Q. Purified GAP was devoid of inherent GTP hydrolytic activity, suggesting that it was a regulator of ras intrinsic GTPase activity. Under submaximal velocity conditions, the second-order rate constant of GTP hydrolysis at 24 degrees C for p21-GTP + GAP (4.5 X 10(6) M-1.sec-1) was at least 1000-fold greater than that for [Val12]p21-GTP + GAP (less than 3 X 10(3) M-1.sec-1).     

Ha-ras proteins exhibit GTPase activity: point mutations that activate Ha-ras gene products result in decreased GTPase activity.

Several ras genes have been expressed at high levels in Escherichia coli and the resultant ras proteins were shown to be functional with respect to their well-known specific, high-affinity, GDP/GTP binding. We were able to detect a weak GTPase activity associated with the purified proteins. The normal cellular ras protein (p21N) exhibits approximately equal to 10 times higher GTPase activity than the "activated" proteins. Even though the turnover rate of the reaction is very low (0.02 mol of GTP hydrolyzed per mol of p21N protein per minute), the reaction appears to be catalytic; one molecule of p21N hydrolyzes more than one molecule of GTP. The GTPase and the GDP binding activities both have been recovered from a Mr 23,000 protein eluted following NaDodSO4/polyacrylamide gel electrophoresis, suggesting that these two activities are associated with the same protein. Mg2+ ions and dithiothreitol are required for GTPase activity and the optimal pH is between 7 and 8. Guanidine X HCl, which is required for solubilizing bacterially expressed ras protein, is strongly inhibitory to GTPase activity at concentrations higher than 0.5 M.     

Signal transduction and the ras gene family: Molecular switches of unknown function.

The ras family consists of 20 or more genes that encode small GTP/GDP-binding proteins, of 20-26 kDa, the functions of which are unknown. This article discusses possible roles of the ras proteins in signal transduction and the interaction of p21(ras) and other members of the ras family with GTPase-accelerating proteins (GAPS) that may be regulatory elements of the signaling machinery.     

Guanosine nucleotide binding by highly purified Ha-ras-encoded p21 protein produced in Escherichia coli.

High-level expression of the p21 protein product of the BALB murine sarcoma virus v-ras gene (similar to the product of the Harvey murine sarcoma virus v-Ha-ras gene) has been reported recently, and highly purified preparations of this protein have been obtained. We used a nitrocellulose filter assay for measuring the binding of GDP and GTP to the purified protein. Previously p21 antibodies had been used to precipitate p21-guanosine nucleotide complexes from crude extracts containing the protein. Using the filter assay, we find that the v-Ha-ras gene product binds [3H]GDP stoichiometrically. The binding is time-dependent and is faster at 30 degrees C than at 0 degrees C. Optimum binding is obtained in the presence of dithiothreitol and magnesium ions and at pH 7.4. In terms of its GDP binding activity, p21 is heat stable and pronase sensitive. The dissociation constants (Kd) of p21 for [3H]GDP and [3H]GTP, determined by Scatchard analysis, are 6 X 10(-8) M and 2.5 X 10(-8) M, respectively.     

Both p21ras and pp60v-src are required, but neither alone is sufficient, to activate the Raf-1 kinase.

The raf genes encode a family of cytoplasmic proteins with intrinsic protein-serine/threonine kinase activity. The c-raf gene is the cellular homolog of v-raf, the transforming gene of murine sarcoma virus 3611. The constitutive kinase activity of the v-Raf protein has been implicated in transformation and mitogenesis. The activity of Raf-1, the protein product of the c-raf gene, is normally suppressed by a regulatory N-terminal domain. Activation of various tyrosine-kinase growth factor receptors results in activation of Raf-1 and its hyperphosphorylation. Further, Raf-1 has been shown to act either downstream or independently of the p21ras protein, as indicated by experiments involving microinjection of anti-Ras antibodies. To investigate the potential role of p21ras in the activation of Raf-1 by tyrosine kinases, we have used the baculovirus/Sf9 cell system to overproduce various wild-type and mutant forms of pp60src, p21ras, and Raf-1 proteins. We show that either pp60v-src or p21c-ras can independently activate the autokinase activity of Raf-1, but only to a limited extent. Surprisingly, both pp60v-src and p21c-ras are required to fully activate Raf-1. Analysis of the Raf-1 autokinase activity in vitro shows that Raf-1 autophosphorylation sites are distributed equally on serine and threonine residues. When Raf-1 is analyzed by immunoblotting, as previously reported for mammalian cell experiments, a marked increase in the apparent molecular weight of Raf-1 is seen only when it is coexpressed with both pp60v-src and p21ras.     

Identification of resonances from an oncogenic activating locus of human N-RAS-encoded p21 protein using isotope-edited NMR.

A sample of Escherichia coli-expressed human N-RAS-encoded p21, a 21-kDa protein, was selectively labeled with 15N at each of the 14 glycine amide positions. Two-dimensional proton-observe 15N correlation spectra showed one peak for each glycine residue. Five glycine resonances were identified with residues near the nucleotide binding site and provide useful reporters of several oncogene-activating positions. Three of these resonances were assigned to residues 10, 15, and 115 from the spectrum of a sample that was also labeled with [13C]valine. These resonances showed extra splitting or broadening due to the 13C label, which could be eliminated by 13C decoupling. Two other peaks were unambiguously identified as Gly-12 and Gly-13 using a one-dimensional edited nuclear Overhauser experiment and by spectral comparison with an Asp-12 mutant. These assignments have provided several site-specific probes of critical domains in p21.     

Involvement of ras p21 protein in signal-transduction pathways from interleukin 2, interleukin 3, and granulocyte/macrophage colony-stimulating factor, but not from interleukin 4.

The protooncogene ras acts as a component of signal-transduction networks in many kinds of cells. The ras gene product (p21) is a GTP-binding protein, and the activity of the protein is regulated by bound GDP/GTP. Recent studies have shown that a certain class of growth factors stimulates the formation of active p21-GTP complexes in fibroblasts and that oncogene products with enhanced tyrosine kinase activities have a similar effect on ras p21. We have measured the ratio of active GTP-bound p21 to total p21 in several lymphoid and myeloid cell lines in order to understand the role of ras in the proliferation of these cells. Interleukin 2 (IL-2), IL-3, and granulocyte/macrophage colony-stimulating factor (GM-CSF) enhance the formation of the active p21.GTP, whereas IL-4 has no effect on p21-bound GDP/GTP. These results strongly suggest that ras p21 acts as a transducer of signals from IL-2, IL-3, and GM-CSF, but not from IL-4.     

The cytokine-activated tyrosine kinase JAK2 activates Raf-1 in a p21ras-dependent manner.

JAK2, a member of the Janus kinase superfamily was found to interact functionally with Raf-1, a central component of the ras/mitogen-activated protein kinase signal transduction pathway. Interferon-gamma and several other cytokines that are known to activate JAK2 kinase were also found to stimulate Raf-1 kinase activity toward MEK-1 in mammalian cells. In the baculovirus coexpression system, Raf-1 was activated by JAK2 in the presence of p21ras. Under these conditions, a ternary complex of p21ras, JAK2, and Raf-1 was observed. In contrast, in the absence of p21ras, coexpression of JAK2 and Raf-1 resulted in an overall decrease in the Raf-1 kinase activity. In addition, JAK2 phosphorylated Raf-1 at sites different from those phosphorylated by pp60v-src. In mammalian cells treated with either erythropoietin or interferon-gamma, a small fraction of Raf-1 coimmunoprecipitated with JAK2 in lysates of cells in which JAK2 was activated as judged by its state of tyrosine phosphorylation. Taken together, these data suggest that JAK2 and p21ras cooperate to activate Raf-1.     

Mutations of the ras gene product p21 that abolish guanine nucleotide binding.

We have constructed several point mutations affecting the GTP-binding site of p21, the ras-encoded protein. Both lysine (116K) and tyrosine (116Y) mutations of asparagine-116, which, by analogy with the crystal structure of elongation factor Tu (EF-Tu), has critical interactions with the guanine base, abolish GTP binding and transforming activities of p21. These activities are retained by proteins with a mutation at position 117 or 118. Both 116K and 116Y mutant p21s, when overproduced in Escherichia coli, are apparently devoid of GTP-binding and autokinase activities. Similarly, the mutant DNAs do not transform NIH 3T3 cells in a focus-forming assay. By cotransfection with pSV-neo, cell clones resistant to the neomycin analog G418 have been isolated. Cells transfected with 116K or 116Y mutant DNA are contact inhibited. In contrast to competent clones, the defective mutants have no detectable phosphorylated p21. The present results suggest that the basic structure of the GTP-binding site is conserved between p21 and EF-Tu and that this binding site is crucial for ras gene function.     

Purification of a factor capable of stimulating the guanine nucleotide exchange reaction of ras proteins and its effect on ras-related small molecular mass G proteins.

We have previously identified a membrane factor capable of stimulating guanine nucleotide exchange activity for ras p21 proteins. The ras guanine nucleotide exchange factor (rGEF) was purified from bovine brain to near homogeneity by successive chromatographies on DE52 DEAE-cellulose, Sepharose 6B, hydroxylapatite, and FPLC phenyl-Superose resins. SDS/polyacrylamide gel electrophoresis of the purified rGEF showed a single major protein with a molecular mass of 35 kDa. rGEF increased the exchange rate of GDP in normal [Gly12]p21 or oncogenic [Val12]p21 up to 30- to 40-fold under physiological concentrations of Mg2+. Since the factor was free from GDP/GTP binding activity and nonspecific GDP hydrolytic activity, we propose that rGEF may regulate GDP/GTP exchange reaction of ras proteins in response to external growth signals. Moreover, rGEF enhanced the dissociation of bound GDP from some of ras-like G proteins, R-ras, rap1-A, rab1-B, and rho proteins, raising the possibility that rGEF may affect the activities of these proteins.     

ras p21 deletion mutants and monoclonal antibodies as tools for localization of regions relevant to p21 function.

Deletion mutants of the viral Harvey ras oncogene were generated by removing different lengths of the gene from either the amino or the carboxyl terminus. The deletion mutants, ras p21 expressed in Escherichia coli, yielded proteins of approximately 8, 10, 12, 14, 17, 18, 19, and 20 kDa. These proteins were utilized to identify epitopes recognized by a series of recently generated monoclonal antibodies as well as some previously reported monoclonal antibodies. Monoclonal antibodies that inhibited GTP binding, a major biochemical activity of the p21 protein, recognized two major regions of the protein. These regions were localized from amino acids 5 to 69 and 107 to 164, respectively, and were separated by another stretch from residues 70 to 106, whose antigenic determinants were not directly involved in GTP binding. Thus, the mapping of epitopes within the p21 molecule recognized by monoclonal antibodies has made it possible to localize important functional regions within the ras p21 molecule.     

Mechanism of GTP hydrolysis by G-protein alpha subunits.

Hydrolysis of GTP by a variety of guanine nucleotide-binding proteins is a crucial step for regulation of these biological switches. Mutations that impair the GTPase activity of certain heterotrimeric signal-transducing G proteins or of p21ras cause tumors in man. A conserved glutamic residue in the alpha subunit of G proteins has been hypothesized to serve as a general base, thereby activating a water molecule for nucleophilic attack on GTP. The results of mutagenesis of this residue (Glu-207) in Gi alpha 1 refute this hypothesis. Based on the structure of the complex of Gi alpha 1 with GDP, Mg2+, and AlF-4, which appears to resemble the transition state for GTP hydrolysis, we believe that Gln-204 of Gi alpha 1, rather than Glu-207, supports catalysis of GTP hydrolysis by stabilization of the transition state.     

Identification of a nucleotide exchange-promoting activity for p21ras.

The biological activity of proteins encoded by the ras family of oncogenes is dependent on whether they are bound to GTP or GDP: the type of nucleotide bound is dependent on the rate of GTP hydrolysis (promoted by the GTPase-activating protein, GAP) and the rate of nucleotide exchange with cytosolic pools. A protein that stimulates the rate of exchange of guanine nucleotide on p21ras has been identified and characterized in cytoplasmic extracts of human placenta. The exchange-promoting protein runs on a gel filtration column with an apparent relative molecular weight of about 60,000. It is sensitive to heat and to trypsin. The exchange-promoting protein acts reversibly and does not cause degradation of p21ras. It is inactive towards the alpha subunit of a heterotrimeric GTP-binding protein (Go alpha) but acts on a large number of different mutant ras proteins, including transforming and effector mutants that are insensitive to the action of GAP. This protein, which we have termed REP (ras exchange-promoting), has the characteristics expected of a physiological activator of p21ras in cellular growth-signal-transduction pathways.     

GTPase domains of ras p21 oncogene protein and elongation factor Tu: analysis of three-dimensional structures, sequence families, and functional sites.

GTPase domains are functional and structural units employed as molecular switches in a variety of important cellular functions, such as growth control, protein biosynthesis, and membrane traffic. Amino acid sequences of more than 100 members of different subfamilies are known, but crystal structures of only mammalian ras p21 and bacterial elongation factor Tu have been determined. After optimal superposition of these remarkably similar structures, careful multiple sequence alignment, and calculation of residue-residue interactions, we analyzed the two subfamilies in terms of structural conservation, sequence conservation, and residue contact strength. There are three main results. (i) A structure-based alignment of p21 and elongation factor Tu. (ii) The definition of a common conserved structural core that may be useful as the basis of model building by homology of the three-dimensional structure of any GTPase domain. (iii) Identification of sequence regions, other than the effector loop and the nucleotide binding site, that may be involved in the functional cycle: they are loop L4, known to change conformation after GTP hydrolysis; helix alpha 2, especially Arg-73 and Met-67 in ras p21; loops L8 and L10, including ras p21 Arg-123, Lys-147, and Leu-120; and residues located spatially near the N and C termini. These regions are candidate sites for interaction either with the GTP/GDP exchange factor, with a GTPase-affected function, or with a molecule delivered to a destination site with the aid of the GTPase domain.