HYBRIDOMA Volume 9, Number Mary Ann Leibert,
6, 1990 Inc., Publishers
Production and Characterization of Anû-RAS p21 Monoclonal Antibodies PETER J. HAMER,1 KEVIN L. TRIMPE,1 THADDEUS PULLANO,1 SIMON NG,1 JOYCE A. LaVECCHIO,1 DEBRA A. PETIT,' RONALD DeLELLIS,2 HUBERT WOLFE,2 and WALTER P. CARNEY' 'Biotechnology Systems, Medical Products Department, E.I. Du Pont, 331 Treble Cove Road, North Billerica, 2Departmenl of Pathology, Tufts New England Medical Center, 175 Washington Street, Boston, MA
ABSTRACT Monoclonal antibodies (MAb) Ras 10 and Ras 11 were raised to an activated human Harvey-ras p21 and shown to react with recombinant p21 as well as p21 derived from human and rodent cells. Characterization studies by ELISA, immunoprecipitation and Western blot procedures demonstrated that MAb Ras 10 (lgG2a) and Ras 11 (lgG2b) react with normal p21, activated p21, and p21 from each of the Harvey, Kirsten and N-ras families. Studies Illustrated that MAb Ras 10 and Ras 11 can also be used in flow cytometry and ¡mmunohistochemistry to specifically detect cellular p21. ELISA, immunoprecipitation and Western blot studies comparing rat anti-p21 MAb Y13-259 with Ras 10 and Ras 11 demonstrated that Ras 10 and Ras 11 had a greater sensitivity for ras protein detection than Y13-259. Collectively, these studies illustrate that MAb Ras 10 and Ras 11 can be applied to a variety of assay formats to detect ras proteins and, therefore, may be valuable tools in detecting and measuring of ras protein expression in normal, neoplastic and pre-neoplastlc cells.
INTRODUCTION Mammalian ras proto-oncogenes encode a series of structurally and functionally related proteins with an approximate molecular weight of 21,000 daltons which are collectively referred to as p21. The ras proteins are localized to the cytoplasmic side of the plasma membrane, bind guanine nucleotides and exhibit GTPase activity. These proteins are part of a larger class of G proteins which have been implicated in signal transduction. The ras gene family consists of at least three known members. The Harvey (Ha) and Kirsten (K¡) ras genes are homologous to the transforming sequences found in the Harvey and Kirsten murine sarcoma viruses, respectively. The third member, designated N-ras, was isolated from a human neuroblastoma cell line and as yet has not been associated with a retroviral counterpart (1). 573
The ras genes have been implicated in neoplastic progression by two mechanisms. The first involves an amino acid substitution at position 12, 13 or 61 of the p21 (2). These alterations have been found in a variety of human cancers including 40%-50% of colon cancers (3), 90% of pancreatic carcinomas (4) and 30% of lung carcinomas (5). The second mechanism proposed for participation of the ras oncogene in neoplasia is a quantitative change in ras protein expression. Overexpression of ras proteins in cancer cells has been reported using Western blot (6), competitive radioimmunoassays (7) and immunohistochemical procedures (8). Several antibodies, both polyclonal and monoclonal, have been used to evaluate ras protein expression. However, the specificity and sensitivity of some antibodies has been questioned and remain a point of controversy in the literature.. In the case of MAb RAP-5, the MAb was raised to a synthetic peptide representing an activated Ha-ras p21 (9) but subsequently shown by several studies not to be specific for ras proteins (10-13). In the case of the widely used rat MAb Y13-259, which was generated to a viral Harvey ras p21 (14), limitations on the ability of the antibody to react with p21 have also been raised (15). The objective of this study was to use recombinant human Ha-ras p21 to produce anti-p21 monoclonal antibodies. MAb Ras 10 and Ras 11 specifically react with both normal and activated ras proteins as well as with p21 from the Ha, Ki and N-ras families. Studies also demonstrated that MAb Ras 10 and Ras 11 can be used in a variety of formats such as ELISA, immunoprecipitation, Western blotting, ¡mmunohistochemistry and flow cytometry. Comparative studies with MAb Y13-259 by ELISA, immunoprecipitation and Western blotting indicated that Ras 10 and Ras 11 were more sensitive in detecting ras p21 than Y13-259. MATERIALS AND METHODS Bacteriallv Produced ras Proteins Ras proteins were expressed in E. coll by insertion of a plasmid containing the coding sequence for the human cellular Harvey ras p21. E. coll cells were harvested, pelleted by centrifugatlon, lysed with 1% Triton X-100 buffer and the protein isolated as previously described (16). The purity of the p21 lysate was determined by SDS-PAGE followed by densitometry of the Coomassie Blue stained proteins. Bacterial lysates used for immunizations Recombinant Ha-ras proteins used in this were greater than 60% ras p21. with included the ¡mmunogen arginine at position 12 (Arg-12 p21), study normal p21 with glycine at position 12 (Gly-12 p21) and recombinant proteins activated by position 61 substitutions of lysine, leucine, arginine or histidine. Recombinant ras proteins were a gift from G. Cooper, Dana Färber Cancer Institute, Boston, MA.
Monoclonal Antibody Production Balb/c x C57BL/6 mice were immunized by intraperitoneal injection of 100 ug of the Arg-12 p21. The first inoculation consisted of p21 mixed with Complete Freund's Adjuvant. Subsequent inoculations consisted of the ras proteins mixed with phosphate buffered saline (PBS). Inoculations were The final given on five occasions, approximately two weeks apart. 574
inoculation was given three days prior to fusion. On the day of fusion, spleens were removed, made into single cell suspensions and fused with Sp2/0 cells as described (17). Ten days after fusion, hybridoma supernatants were screened by ELISA for reactivity with the normal Gly-12 p21 and the ¡mmunogen Arg-12 p21 coated onto microtiter plates. Hybridoma cell lines secreting MAb Ras 10 or Ras 11 were cloned by limiting dilution and inoculated into pristane primed mice to produce ascites fluid. MAb Ras 10 and Ras 11 were purified from ascites fluid by affinity chromatography on Affigel-Protein A (BioRad, Richmond, CA).
Cell Lines and Lysate Preparation All cell lines were maintained in DMEM high glucose medium supplemented with 5% fetal bovine serum, 1% L-glutamine, non-essential amino acids and antibiotics. NIH3T3 cells transformed by a human Harvey ras gene that was genetically engineered to overexpress the normal ras protein were designated 3T3-Ha-ras, whereas NIH3T3 cells transformed by overexpression of the normal Ki-ras were designated 3T3-Ki-ras (18). NIH3T3 cells transformed by an activated K\-ras gene (19) were designated 3T3-Act-Kiras. NIH3T3 cells transformed by an activated N-ras gene were also genetically engineered to have elevated levels of the activated p21 (20) and were designated 3T3-Act-N-ras cells. All cell lines were kindly provided by G. Cooper, Dana Färber Cancer Institute, Boston, MA. For immunoprecipitation studies, cells were labeled with 35S methionine (NEN Products, Boston, MA). Cells (1x106) were plated in 100 mm petri dishes, allowed to grow overnight, washed three times with PBS and incubated with methionine free media (Selectamine, GIBCO, Grand Island NY) containing 0.25 mCi/ml of 35S-methlonine. After an overnight incubation, cells were washed, harvested and lysed by Dounce homogenization in Triton X-100 buffer. Cell lysates were centrifuged and the supernatants used for immunoprecipitation studies (19). For Western blot studies, cells and human tumors were extracted with Triton X-100 lysis buffer as previously described (21). Human tumor specimens were provided by the Department of Pathology, Tufts New England Medical Center, Boston, MA. -
Electroelution of p21 Proteins Lysates from transformed NIH3T3 cells overexpressing the Ha, Ki or tiras p21 were resolved by SDS-PAGE using a 12.5% gel. After electrophoresis, location of p21 in the gel was determined using molecular weight markers. This area was excised from the gel and electroeluted overnight at 100 volts using the ELUTRAP apparatus (Schleicher & Schuell, Keene, NH) (22). ELISA
Immulon II (Nunc, Denmark) immunoassay plates were coated with 63 ng/well of the various recombinant ras proteins. After overnight incubation, plates were washed and blocked. The plates were incubated with either hybridoma supernatants, purified MAb Ras 10, Ras 11, rat MAb Y13-259 or an isotype-matched negative control antibody. After a one hour incubation, plates were washed and developed as previously described (21). 575
35S-methionine labeled cell lysates were incubated with mouse MAb Ras 10, Ras 11, rat MAb Y13-259 or an isotype-matched negative control antibody. Immune complexes were pelleted by the addition of Protein A Sepharose linked to goat anti-mouse or rabbit anti-rat ¡mmunoglobulin. Immunoprecipitated proteins were resolved by SDS-PAGE using a 12.5% gel and visualized by autoradiography as previously described (19). Western Blot
Proteins from cell lysates were separated by electrophoresis on a 12.5% SDS-PAGE gel and transferred overnight to nitrocellulose membranes at 4°C. The membranes were blocked with 5% bovine serum albumin in PBS for one hour and then incubated with Ras 10, Ras 11 or Y13-259. All studies included an isotype-matched myeloma ¡mmunoglobulin as a negative control. The membranes were washed and then incubated with goat anti-mouse or rabbit anti-rat conjugated to horseradish peroxidase (HRP). The reaction was completed as previously described (21). Flow
For flow cytometric analysis, cells were harvested from culture, washed in RPMI medium containing 5% fetal bovine serum and fixed in 1% formaldehyde and 0.25% saponin for one hour at 4° C. After fixation, 2 x 106 cells per sample were washed and resuspended in PBS and incubated with MAb Ras 10 or control antibody for one hour at 4° C. These cells were washed and incubated with a goat anti-mouse fluorescein isothiocyanate conjugate (GAM-FITC)(Tago, Burlingame, CA) for one hour at 4° C. After three additional washes, samples were evaluated on an EPICS-V flow cytometer equipped with an argon laser tuned to 488 nm. Windows were set such that fewer than 5% of the cells were stained with an isotype matched control
I mmunohistochemistry In order to perform
immunoperoxidase studies, cytocentrifuged the high ras expressing cells 3T3-Ha-ras and expressing cells 3T3-Act-Ki-ras. Prior to immunoperoxidase
cells were fixed in 10% neutral buffered formalin at 4° C for ten minutes. The slides were rinsed in PBS and treated with 1% normal horse serum to block non-specific binding of immunoglobulins. The slides were then incubated with MAb Ras 10 or MAb Ras 11 at a concentration of 0.2 mg/ml for 16 hours. After rinsing with PBS, the slides were treated with biotin labeled horse anti-mouse IgG and a preformed avidin-biotin peroxidase
complex (Vector Laboratories, Burlingame, CA) (21). Sites of peroxidase activity were demonstrated with diaminobenzidine tetrahydrochloride. For control studies, an equivalent concentration of an isotype matched MAb was used in place of MAb Ras 10 or MAb Ras 11. For blocking experiments, 200 ug/ml of MAb Ras 10 or MAb Ras 11 was preincubated with 25 mg/ml of bacterial Arg-12 p21 prior to addition to the slides.
RESULTS Monoclonal Antibody Selection To generate monoclonal antibodies reactive with ras proteins, we immunized mice with the Arg-12 recombinant Ha-ras p21. During the immunization period, mouse sera were collected and evaluated by ELISA for reactivity with the p21 ¡mmunogen. Mice that exhibited a serum reactivity of greater than 1:50,000 dilution were selected for fusion. To select hybridomas secreting anti-p21 antibodies, several hundred supernatants were analyzed by ELISA and immunoprecipitation procedures for reactivity with the Arg-12 and Gly-12 p21. By ELISA, the majority of hybridoma supernatants reacted with both recombinant ras proteins whereas only a few were able to immunoprecipitate the ras p21. Based on the intensity of the immunoprecipitation reaction, two MAbs, designated Ras 10 and Ras 11 were selected for additional study. These hybridomas were subcloned and the immunoglobulins purified from ascites fluid by Protein A chromatography. MAb Ras 10 was determined to be an lgG2a while Ras 11 was shown to be an lgG2b using goat anti-class and subclass antibodies
(Boehringer Mannheim, Indianapolis, IN).
Reactivity of MAbs Ras 10 and Ras 11 with Recombinant and Cellular Ras Proteins Various concentrations of MAb Ras 10 or Ras 11 were incubated overnight in microwells previously coated with the recombinant Arg-12 p21 (63 ng/well). The ELISA reaction was completed by the addition of goat antiResults mouse HRP and the OPD substrate as previously described (19). demonstrated strong reactivity of Ras 10 and Ras 11 at all five concentrations tested, whereas the negative control antibody MOPC 141 was unreactive with the ras protein (Figure 1). Studies in Figure 1 also show that MAb Y13-259 was only weakly reactive with the Arg-12 recombinant p21. Experiments with the normal Gly-12 recombinant p21 or recombinant ras proteins activated by position 61 mutations (Lys, Leu, Arg or His) also demonstrated that Ras 10 was strongly reactive with these ras proteins whereas MAb Y13-259 was not (data not shown). Additional studies also showed that Ras 10 and Ras 11 were able to detect as little as 10 ng of p21 coated onto microtiter plates whereas the minimum amount of p21 detected by Y13-259 was 250 ng. These results indicate that Ras 10 and Ras 11 are approximately 25 times more sensitive than Y13-259 in detecting recombinant ras proteins by ELISA. We next wished to determine whether MAb Ras 10 and Ras 11 could react with the three families (Ha, Ki and N) of ras p21. To do this, we immobilized a polyclonal anti-p21 rabbit antibody onto microtiter wells to capture the ras proteins. Following immobilization of the capture reagent, wells received 250 pg of either the Ha, Ki or N-ras p21 electroeluted (as described in Materials and Methods) from 3T3-Ha-ras, 3T3-Ki-ras or 3T3Act-N-ras cells. After an overnight incubation, wells were washed and blocked and then incubated with biotin labeled Ras 10 or Ras 11. Following a one hour incubation, wells were washed and the amount of Ras 10 or Ras 11
-•— RAS 10 -*— RAS 11 -V- Y13-289 -O— MOPC 141
NG OF ANTIBODY
ELISA Reactivity of MAb Ras 10, Ras 11 and Y13-259 with Recombinant Ras p21. MAb Ras 10, Ras 11 or rat MAb Y13-259 were tested for reactivity with recombinant Ha-ras proteins (63 ng/well) by ELISA. Ras 10, Ras 11 or the control myeloma protein MOPC 141 were detected using a goat anti-mouse HRP conjugate. Rat MAb Y13-259 was detected using a rabbit anti-rat HRP
bound to the captured p21 was determined using a Streptavidin system as described (23, 24). Figure 2 illustrates a typical experimental result. These data show that MAb Ras 10 reacted with each of the Ha, Ki and N-ras p21. Similar results were obtained with MAb Ras 11 (data not shown). Collectively, these ELISA studies demonstrate that the anti-p21 MAb Ras 10 and Ras 11 recognize both normal and activated p21 (bacterial and cellular) as well as the Ha, Ki and N-ras p21 derived from mammalian cells.
Immunoprecipitation Studies To evaluate the ability of Ras 10 to specifically immunoprecipitate cellular p21 proteins, we examined cell lysates labeled with 35S-methionine. Figure 3 illustrates that MAb Ras 10 (Panel A) immunoprecipitated the activated Ki-ras p21 (lane 1), the normal Ha-ras (lane 2) and the activated N-ras protein (lane 3). In lane 4, a negative control antibody was substituted Similar results were for Ras 10 in the immunoprecipitation reaction. obtained using Ras 11 as the immunoprecipitating reagent (data not shown). Panel B illustrates that MAb Y13-259 immunoprecipitated the Ki-ras (lane 1) and Ha-ras p21 (lane 2) to a similar degree as Ras 10. However, panel B also shows that Y13-259 is less reactive with the N-ras protein than Ras 10 (lane 3). Our results consistently demonstrated that MAb Y13-259 was less reactive with N-ras proteins than with the Ha and Ki-ras proteins. 578
ELISA OF CELLULAR
WITH MAB RAS 10
2. Detection of Ha, Ki and N-ras p21 using Ras 10 Nunc Maxisorb immunoassay plates were coated with an anti-p21 rabbit polyclonal antibody at 500 ng per well. 250 pg of electroeluted p21 from 3T3-Ha-ras, 3T3-Ki-ras and 3T3-Act-N-ras cells were incubated overnight After wells were washed, biotinylated Ras 10 was at room temperature. added for 1 hour. Bound biotinylated Ras 10 was detected with a Streptavidin horseradish peroxidase conjugate and the O-phenylenediamine substrate as described.
3. Immunoprecipitation of Cellular Ha, Ki and N-ras Proteins 35S Methionine labeled cell lysates were incubated with either 500 ng Ras 10 (Panel A) or 500 ng Y13-259 (Panel B). Lysates tested were 3T3-ActKi-ras lysate (lane 1) (50 x 106cpm), 3T3-Ha-ras (lane 2) (25 x106cpm) or
3T3-Act-N-ras lysate (lane 3) (100 x 106 cpm). Panel A, lane 4 shows the results using a negative control antibody incubated with the 3T3-Ki-ras
Western Blot Studies MAb Ras 10 and Ras 11 were analyzed for their ability to react with cellular ras proteins in the Western blot procedure. Non-radioactive cell were to transferred nitrocellulose and lysates separated by SDS-PAGE, with As Ras 10. shown in Ras detected the Ha-ras p21 10 probed Figure 4, Ki-ras and the N-ras (lane 1), p21 (lane 2) p21 (lane 3). Comparable results were obtained with Ras 11 (data not shown).
4. Western Blot Analysis of Cellular Ha, Ki and N-f?as Proteins Cell lysates (0.3 mg protein/lane) were separated by SDS-PAGE, transferred to nitrocellulose filters and incubated with 5 ug/ml of Ras 10 as described in Materials and Methods. Lysates were derived from 3T3-Ha-ras cells (lane 1), 3T3-Act-Ki-ras cells (lane 2) or 3T3-Act-N-ras cells (lane 3).
We next compared the ability of monoclonal antibodies Ras 10 and Y13Three colon 259 to detect ras proteins in primary human tumors. carcinomas and one breast carcinoma were analyzed for the presence of p21 using Ras 10 or Y13-259. To perform these experiments, tumor lysates were separated by SDS-PAGE and transferred to nitrocellulose. Nitrocellulose strips were incubated with MAb Ras 10 (lanes 2-5) or MAb Y13-259 (lanes 69). Figure 5 illustrates that Ras 10 was able to detect ras p21 in all four In this human tumors whereas no p21 was detectable with Y13-259. that times than was concentration of Y13-259 five the experiment, greater of Ras 10. The majority of experiments employing Y13-259 with mouse, rat and human tumors, as well as a variety of cell lines from all three species, demonstrated that Y13-259 either did not detect a p21 band or showed only weak reactivity with p21 compared to Ras 10.
of Human Carcinomas
9 Ras 10 and Y13-
Lysates of human carcinomas (0.5 mg protein/lane) were separated by SDS-PAGE, transferred to nitrocellulose and incubated with 10 ug/ml of Ras 10 (lanes 2-5) or 50 ug/ml of Y13-259 (lanes 6-9). Molecular weight markers (lane 1); colon carcinoma 1 (lanes 2 & 6), breast carcinoma (lanes 3 & 7), colon carcinoma 2 (lanes 4 & 8) and colon carcinoma 3 (lanes 5 & 9).
We next evaluated the ability of Ras 10 to function as an immunohistochemical reagent. For this analysis we selected the high ras expressing cell line 3T3-Ha-ras and the low ras expressing cell line 3T3Act-Ki-ras. Prior to staining, cells were cytocentrifuged onto glass microscope slides and fixed with formalin. The slides were then stained with Ras 10, followed by the Vectastain avidin-biotin system as described. As seen in Figure 6, Ras 10 showed a much more intense immunoperoxidase staining of cells that overexpressed the Ha-ras p21 (Panel a) than transformed NIH3T3 cells that expressed low levels of the activated Ki-ras p21 (Panel b). The majority of staining in both cell types was located at the periphery of the cell, as would be expected for a protein localized to the cell membrane. Additional experiments demonstrated that the immunoperoxidase staining of cells could be inhibited by incubating 200 ug/ml of Ras 10 with 25 mg/ml of recombinant p21 prior to the addition of Ras 10 to slides (data not shown). These results demonstrate the ability of Ras 10 to specifically stain ras proteins by the immunoperoxidase procedure as well as illustrating the ability of Ras 10 to discriminate the different levels of ras p21 expressed in these cells. The 3T3-Ha-ras and 3T3-Act-Ki-ras cell lines used for immunohistochemistry were also evaluated by immunoprecipitation with Ras 10. Cells were labeled with 35S-methionine and the immunoprecipitation
6. Immunohistochemical Reactivity of MAb Ras 10 3T3-Ha-ras cells overexpressing the normal Ha-ras protein (Panel a) or 3T3-Act-Ki-ras cells expressing low levels of the activated Ki-ras protein (Panel b) were stained with Ras 10 followed by the Vectastain avidin-biotin
also labeled with 3T3-Ha-ras cell lysate (Panel c, lane 1) or 3T3-Act-Ki-ras cell lysate (Panel c, lane 2) was incubated with 500 ng Ras 10. Lane 3 represents results using a ¡sotypematched negative control antibody for immunoprecipitation. Cells used for
35S-methionine for immunoprecipitation studies.
reaction completed as previously described. Figure 6, Panel c illustrates that the 3T3-Ha-ras cells overexpressing the ras protein had a more intense p21 band (lane 1) than the 3T3-Act-Ki-ras cells expressing low levels of ras p21 (Lane 2). A negative control antibody was substituted for Ras 10 (lane 3) with the 3T3-Ha-ras cell lysate.
Flow Cvtometrv The
of Ras 10 to detect
After fixation and permeabilization with 1% formaldehyde and 0.25% saponin, cells were incubated with either Ras 10 or a negative control antibody followed by GAM-FITC. The histograms presented in Figure 7a show that greater than 95% of the cells from each line were stained specifically with Ras 10. In addition, Figure 7b indicates that the different levels of ras p21 (3T3-Ha-ras vs. 3T3-Act-Ki-ras), as seen with immunohistochemistry, could also be distinguished by flow cytometric analysis with Ras 10.
cytometry using 3T3-Ha-ras, 3T3-Ki-ras and 3T3-Act-N-ras cells.
i 200 \
Relative Fluorescence Intensity
Figure 7. Flow Cytometric Analysis with MAb Ras 10 3T3-Ha-ras cells ( ),3T3-Act-Ki-ras cells ( examined for reactivity with Ras 10 ras cells ( ) were
3T3-Ha-ras cells ( ), using Ras 10 by flow cytometry. isotype matched negative control
Background staining antibody ( ) (Panel b).
), and 3T3-Act-Nby flow cytometry ),
specificity of the flow with recombinant Arg-12p21 prior to staining the cells. Preincubation of the immunogen with MAb Ras 10 prevented the MAb from staining the cells and confirmed the specificity of the Ras 10 reaction observed by flow cytometry. In
cytometric results, MAb Ras 10
report, we describe the production and characterization of two monoclonal antibodies designated Ras 10 and Ras 11. These MAb
generated from mice immunized with an activated Ha-ras protein produced in E. coli. Characterization of these antibodies employed a variety of assay formats and a series of genetically engineered cell lines. These studies demonstrated that MAb Ras 10 and Ras 11 can specifically detect both normal and activated recombinant and cellular ras proteins as well as The flow cytometric and all three families of ras proteins. immunohistochemistry studies also illustrated that Ras 10 could be used to discriminate the different levels of ras protein found in cells. Numerous studies have suggested that the ras protein participates in the initiation and progression of neoplasia by two mechanisms (1). The first mechanism, which is most clearly documented, involves point mutations at the DNA level. These genetic changes encode proteins altered at amino acid position 12, 13 or 61 (1, 2). The second mechanism by which ras has been implicated in neoplastic progression is by overexpression of the normal protein. In vitro studies have shown that the ras proto-oncogene can be transforming when the gene is placed under the control of a retroviral LTR, thereby resulting in elevated levels of the ras protein (25). Studies by Spandidos et al. measured ras mRNA and concluded that ras transcripts were elevated in human cancers and that ras overexpression may have a role in Western blot (6), RIA (7) and neoplastic progression (26). ras immunohistochemical (8) studies of expresssion in human tissues have also suggested that ras proteins may be important markers for malignancy. However, in many of these studies, the anti-p21 MAb used to evaluate ras expression has been the rat MAb Y13-259. MAb Y13-259 was raised to viral Ha-ras p21 (14) but was shown to immunoprecipitate only the Ha and Ki ras proteins (14). were
Since our ELISA data demonstrated that Y13-259 was less sensitive with recombinant p21 than Ras 10 or Ras 11, it suggested that Y13-259 might also be less reactive with ras proteins by Western blot. To test this hypothesis, we analyzed a series of transformed NIH3T3 cells, human tumor cell lines and primary tumors with both MAb Ras 10 and MAb Y13-259. In the majority of samples tested, our Western blot studies demonstrated that MAb Y13-259 either did not react with cellular p21 detected by Ras 10 or that the intensity of the reaction with Y13-259 was much less than that of Ras 10. Variations in the concentration of Y13-259 as well as the reagents used to detect Y13-259 did not significantly alter the poor results with Y13-259. In Y13-259 illustrated that biochemical studies our addition, to than Ras a N-ras lesser 10, suggesting degree immunoprecipitated p21 that Y13-259 might not be immunoprecipitating all the available p21 in cells. Studies by Sorrentino et al. (15) have also shown that Y13-259 had less affinity for recombinant ras proteins by ELISA and was less quantitative by immunoprecipitation studies than antibodies that they had generated and Previous reports have also characterized to recombinant ras proteins. indicated that ras may be diminished or absent in certain clinical specimens and from these studies have come conclusions about the significance of the ras oncogene protein in the initiation and progression of cancer (27). However, in light of our results and the reports of others, it is possible that previous studies with Y13-259 may be inaccurate with regard to the assessment of ras protein levels in cells.
REFERENCES genes. Annu. Rev. Biochem.
Bos, J.L. ras oncogenes in human 4682-4689, (1989).
review. Cancer Research. 49:
Bos, J.L., Fearon, E.R., Hamilton, S.R., Verlaan-deVries, M., van Boom, J.H., der Eb, A. & Vogelstein, B. Prevalence of ras gene mutations in human colorectal cancers. Nature. 327: 293-297, (1987).
Almoguera, C, Shibata, D., Forrester, K., Martin, J., Arnheim,
Perucho, M. Most human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell. 53: 549-554, (1988).
Rodenhuis, S., Slebos, R.J.C., Boot, A.J.M., Evers, S.G., Mooi, W.J., Wagenaar, Bodegom, P.C. & Bos, J.L. Incidence and possible clinical
significance of K-ras oncogene activation in adenocarcinoma human lung. Cancer Research. 48: 5738-5741, (1985) 6.
Clair, T., Miller, W.R. & Cho-Chung, Y.S. Prognostic significance of the of a ras protein with a molecular weight of 21,000 by human breast cancer. Cancer Research. 47: 5290-5293, (1987).
Caruso, A., Schlom, J., Vilasi, V., Weeks, M.O. & Hand, P.H. Development of quantitative liquid competition radioimmunoassays for the ras oncogene and proto-oncogene p21 products. Int. J. Cancer. 38: 587-595, (1986). Furth, M.E., Aldrich, T.H. & Cordon-Cardo, C. Expression of oncogene
in normal human tissues.
Horan-Hand, P., Thor, A., Wunderlich, D., Murano, R., Caruso, A. & Schlom, J. Monoclonal antibodies of predefined specificity detect activated ras gene expression in human mammary and colon carcinomas. Proc. Nati. Acad. Sei (USA). 81: 5227-5231, (1984).
Ghosh, A.K., Moore, M. & Harris, M. Immunohistochemical detection of oncogene p21 product in benign and Clin. Path. 39: 428-434, (1986).
malignant mammary tissue
Samowitz, W.S., Pauli, G. & Hamilton, S.R. Reported binding of
monoclonal antibody RAP-5 for formalin fixed tissue sections is not indicative of ras p21 expression. Human Pathol. 19: 127-132, (1988).
Piris, J., Spandidos, D.A. & Nyllie, A.H. Evaluation of monoclonal antibody to ras peptide, RAP-5 claimed to bind preferentially to cells of infiltrating carcinomas. Br. J. Cancer. 54:
12. Robinson, A., Williams, A.R.N.,
13. Ward, Ü.M., Perantoni, A.D. & Santos, E. Comparative immunohistochemical reactivity of monoclonal and polyclonal antibodies to H-ras p21 in normal and neoplastic tissues of rodents and humans. Oncogene. 4:
Furth, M. E., Davis, L.J., Fleurdelys, B. & Scolnick, E. Monoclonal antibodies to the p21 product of the transforming gene of Harvey murine sarcoma virus and the cellular ras gene family. J. Virol. 43 (1): 294304, (1982).
Sorrentino, V., Nebreda, A.R., Alonso, T. & Santos, E. Preparation,
characterization and H-ras p21 proteins. 16.
properties of monoclonal antibodies against intact Oncogene. 4: 215-221, (1989).
Feig, L.A., Pan, B.T., Roberts, T.M. & Cooper, G.M. Isolation of GTP binding using an in situ colony binding assay. Proc. Nati. Acad. Sei. (USA). 83: 4607-4611, (1986). mutants
Galfre, G., Milstein, G. & Wright, B. Rat monoclonal anti-Fd
133, (1979). 18. Der, C.J., of human 19.
hybrid myelomas and a (London). 277: 131-
Finkel, T. & Cooper, G.M. Biological and biochemical properties rasH genes mutated at codon 61. Cell. 44: 167-187, (1986).
Der, C.J. & Cooper, G.M. Altered gene products are associated with activation of cellular ras^ genes in human lung and colon carcinomas. Cell. 32: 201-208, (1983).
20. Guerrero, I., Villasante, A., Cores, V. & Pellicer, A. Loss of the normal tiras alíele in a mouse thymic lymphoma induced by a chemical carcinogen. Proc. Nati. Acad. Sei. (USA). 82: 7810-7814, (1985). 21.
Carney, W.P., Petit, D., Hamer, P., Der, C.J., Finkel, T., Cooper, G.M., LeFebvre, M., Mobtaker, H., DeLellis, R., Tischler, A.S., Dayal, Y., Wolfe, H. & Rabin, H. Monoclonal antibody specific for an activated ras protein. Proc. Nati. Acad. Sei.(USA). 83: 7485-7489, (1986).
Jacobs, E. & Clad, A. Electroelution of fixed and stained membrane proteins from preparative sodium dodecyl sulfate-polyacrylamide gels into a membrane trap. Analytical Biochemistry. 154: 583-589, (1986).
23. Lerner, E.A., Matis, L., Janeway, G.A. Jr., Jones, P.P., Murphy, D.B. Monoclonal antibody against an Ir gene 152: 1085-1101, (1980). 24.
Schwartz, R.H. &
product? J. Exp. Med.
Updyke, T.V. & Nicodson, G.L. Immunoaffinity isolation of membrane antigens with biotinylated monoclonal antibodies and immobilized streptavidin matrices. J. Imm. Methods. 73: 83-95, (1984).
Chang, E.H., Furth, M.E., Scolnick,
transformation of mammalian cells induced by homologous to the oncogene of Harvey murine 297: 479-483, (1982).
D.R. Tumorigenic a normal human gene sarcoma virus. Nature.
Spandidos, D.A. & Agnantis, N.J. Human malignant tumors of the breast, as compared to their respective normal tissue, have elevated expression of the Harvey ras oncogene. Anticancer Res. 4: 269-276, (1984).
Gallick, G.E., Kurzrock, R., Kloetzer, W.S., Arlinghaus, R.B. & Gutterman, J.U. Expression of p21 ras in fresh primary and metastasis human colorectal tumors. Proc. Nati. Acad. Sei. (USA) 82: 1795-1799, (1985).
Walter P. Carney, Ph.D. E.I. du Pont de Nemours and Co. Medical Products Department 331 Treble Cove Road North Billerica, Massachusetts 01862 Received for publication May 9, 1990 Accepted after revision September 12, 1990