A cDNA encoding the complete normal human p53 protein was expressed in Escherichia coli using an expression system based on the bacteriophage T7 promoter. The cDNA was adapted so that the full-length protein was produced without fusion to any other sequence. Large amounts of the protein were isolated and the purified protein used to produce very high titre polyclonal antibodies to p53. These new antibodies permit the sensitive detection of p53 and p53 complexes in ELISA and immunoblotting assays. Most importantly, they also permit the detection of p53 in archival tumour material that has been conventionally fixed in formalin and embedded in paraffin wax. Using this reagent we have found that aberrant expression of p53 is a frequent feature of human breast cancer. We are able to recognise six different classes of p53 expression pattern that may be of help in the subclassification of breast tumours.

Tumour suppressor genes (Marshall, 1991) were first recognised because of the dominance of the normal phenotype in cell fusion experiments where malignant cells were fused to normal cells. The molecular identification of these genes has been derived from the study of rare inherited cancers, and more recently from the study of proteins that interact with the oncogene products of the DNA tumour viruses. The two best characterised tumour suppressor genes are the retinoblastoma gene Rb and the p53 gene (Levine and Momand, 1990). Germline mutation in the Rb gene results in hereditary retinoblastoma, but somatic mutations in Rb have also been found in other tumour types. Germline mutations in p53 are the basis of the Li-Fraumeni cancer family syndrome (Malkin et al. 1990). Somatic mutation in p53 is the most common molecular change identified in human cancer, and over 300 mutations have been analysed (Hollstein et al. 1991). The mutations are unusual both in the vast range of malignancies in which they are found (Nigro et al. 1989; Bártek et al. 1991) and in the fact that most of them are point mutations. The point mutations are associated with a dramatic increase in the stability of the protein and many seem not only to neutralise the suppressor activity of p53 but also lead to a gain in function, so that the mutant protein actually acts as a dominant oncogene (Lane and Benchimol, 1990; Levine et al. 1991). These properties of the mutant protein make it an ideal target for new diagnostic and therapeutic approaches and focus attention on the biochemical properties of mutant and normal p53. To this end we have used a sophisticated Escherichia coμ-based expression system (Studier and Moffatt, 1986; Tabor and Richardson, 1985) to overproduce large amounts of p53. This has permitted the production of very high titre antibodies to p53 and the establishment of sensitive and quantitative assays for human p53 protein expression in tumour cell lines and in tumour resection samples.

Expression of human p53 protein in E. coli

Human p53 was expressed in E. coli under the control of the bacteriophage T7 RNA polymerase promotor, <M0 (Tabor and Richardson, 1985) (Fig. 1). The host E. coli strain BL21(DE3)pLysS (Studier et al. 1990) is a ÂDE3 bacteriophage lysogen of BL21 (F-OmpT rBmB) that carries the gene for T7 RNA polymerase under the control of the isopropyl-β-D-thiogalactopyranoside (IPTG)-inducible promoter ZocUV5. The strain also harbours a plasmid, pLysS, expressing T7 lysozyme, which confers chloramphenicol resistance and further contributes to plasmid stability by reducing the basal level of T7 RNA polymerase activity in induced cells.

Fig. 1.

Map showing pT7-7Hup53. The Hup53 gene is expressed under the control of the inducible bacteriophage T7 RNA polymerase promoter, Φ10. The original pT7-7 vector was modified to introduce an SphI site at the start codon ATG by replacement of the Xbal-EcoRl fragment with a synthetic linker sequence:

Fig. 1.

Map showing pT7-7Hup53. The Hup53 gene is expressed under the control of the inducible bacteriophage T7 RNA polymerase promoter, Φ10. The original pT7-7 vector was modified to introduce an SphI site at the start codon ATG by replacement of the Xbal-EcoRl fragment with a synthetic linker sequence:

Cells containing pT7-7Hup53 were grown at 37°C in Luria broth with 50 jtg ml-1 ampicillin, 25 μg ml-1 chloramphenicol. At an Asco of 0.4 expression was induced by addition of IPTG to 0.5 mM, and the culture was incubated for a further 2 h before harvesting. Under these conditions most of the p53 protein recovered was insoluble (Fig. 2).

Fig. 2.

IPTG induction of the expression of Hup53. Cell lysates were fractionated by 10% SDS-PAGE and visualised by Coomassie blue staining. Lane 1, Mr standards (×10−3); lane 2, uninduced BL21(DE3)plysS cells containing pT7-7Hup53; lane3, the same cells after induction with IPTG; lane 4, soluble protein fraction in the supernatant; lane 5, insoluble protein fraction obtained after pelleting proteins from induced cell lysates at 10,000 g. The position of p53 is indicated.

Fig. 2.

IPTG induction of the expression of Hup53. Cell lysates were fractionated by 10% SDS-PAGE and visualised by Coomassie blue staining. Lane 1, Mr standards (×10−3); lane 2, uninduced BL21(DE3)plysS cells containing pT7-7Hup53; lane3, the same cells after induction with IPTG; lane 4, soluble protein fraction in the supernatant; lane 5, insoluble protein fraction obtained after pelleting proteins from induced cell lysates at 10,000 g. The position of p53 is indicated.

The 477 base-pair 5’ Main fragment of human p53 was inserted into the Sphl site and the remaining sequence subsequently introduced as a 1109 base-pair PvuII-EcoRI fragment.

Purification of human p53

The cell pellet from a 11 culture was resuspended in 30 ml of lysis buffer (50 mM Tris-HCl, pH 8.0, 2 mM EDTA, 100 mM NaCl, 1 mM phenylmethylsulfonylfluoride (PMSF)) and lysed by two or three cycles of freeze-thawing, which activates the internally expressed lysozyme. Viscosity was reduced by gentle sonication and the insoluble proteins, including p53, were pelleted by centrifugation at 10,000 g. The pellet was resuspended in lysis buffer containing 0.5% Triton X-100 and re-pelleted as described above. This washing procedure was repeated twice to obtain a firm pellet. The insoluble protein was solubilised in 10 ml of 5 M guanidine hydrochloride, 50 mM Tris-HCl, pH 8.0, 0.005% Tween 80 at 4°C for 5 h, mixing gently The remaining insoluble material was pelleted as described above and the supernatant was diluted to a final concentration of 1 M guanidine hydrochloride, 50 mM Tris-HCl, pH 8.0, 0.005% Tween 80, 2 mM reduced glutathione, 0.02 mM oxidised glutathione. Refolding of the protein was facilitated by gentle mixing at 4°C for a further 12–18 h, followed by extensive dialysis against 50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.005% Tween 80. Insoluble material was pelleted as described above and the supematent was applied to a DEAE Biogel A ion exchange column (Bio-Rad). The protein was eluted with a 0.15 M to 0.5 M NaCl gradient in 50 mM Tris-HCl, pH 8.0.

Production of the anti-p53 antibody CM-1

The polyclonal anti-p53 antibody was produced by immunising a rabbit with the purified p53 protein. The first injection of 1 mg of protein was given in Freund’s complete adjuvant subcutaneously. Three subsequent booster injections of the same dose were given at intervals of one month in incomplete Freund’s adjuvant. A strongly positive titre against p53 was detected after the third injection.

Monoclonal antibodies

Monoclonal antibodies used in these studies were the anti-p53 antibodies PAb421 (Harlow et al. 1981), PAbl801 (Banks et al. 1986), PAb240 (Gannon et al. 1990) and Bp53-ll (Bártková et al. 1991). The anti-T antibody was PAb419 (Harlow et al. 1981). The anti-galactosidase monoclonal antibody BG-2 was used as a control.

Cell culture

The human breast carcinoma-derived line BT-20 was grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% foetal calf serum (FCS) and 10 μg ml-1 bovine insulin (Sigma). All other cells including human breast cancer cell lines PMC-42, MDA-468, T47D and BT-549, human sarcoma cell lines RD and Hs913T, human vulval carcinoma cell line A431, Burkitt’s lymphoma line Namalwa, colon carcinoma line HT 29 and human SV40-transformed kératinocyte line SVK14 were cultured in DMEM with 10% FCS.

Immunoprecipitation and immunoblotting

The cultured cells were lysed in 150 mM NaCl, 50 mM Tris-HCl, pH 8.0, 5 mM EDTA, 1% NP40,1 mM PMSF for 30 min on ice and the extracts centrifuged at 30,000g for 30 min. The supernatants were pre-absorbed with Protein G-Sepharose (Pharmacia), followed by overnight incubation with purified monoclonal antibody PAb421 or the CM-1 antiserum. The immunocomplexes were then collected on protein G beads, the beads were washed 4 times in lysis buffer and boiled briefly in 2×concentrated Laemmli sample buffer.

Direct immunoblotting was performed with whole-cell lysates prepared by harvesting confluent cells in 90°C Laemmli sample buffer. Both the immunoprecipitates and whole-cell lysates were separated by polyacrylamide gel electrophoresis in the presence of SDS. Pre-stained molecular weight standards (Bio-Rad) were run in parallel. The transfer of proteins separated on 12.5% gels onto nitrocellulose membrane, and visualization by indirect immunoperoxidase procedure were performed as described previously (Bártek et al. 1985).

Plate immunoassays

Falcon 96-well microtitre plates were incubated overnight at room temperature with 50 μl per well of 20 μg ml-1 purified mouse monoclonal antibody in 0.1 M sodium bicarbonate buffer, pH 9.5, rinsed in phosphate-buffered saline (PBS) and blocked for 2 h in 3% bovine serum albumin (BSA, Sigma) in PBS. Cell extracts were prepared as described above (see Immunoprecipitation) and protein concentration was estimated using the Bio-Rad Protein Assay. Plates were washed once in PBS, twice in 0.1% NP40 in PBS, once in PBS again and 50 μl of extract or serial dilutions in lysis buffer were added to each well and incubated overnight at 4°C. The plates were washed 4 times as above, CM-1 antiserum (diluted 1:1000 in 1% BSA in PBS with 0.1% NP40) was added (50 μl per well) and the incubation continued for 3 h. The plates were washed 4 times again and horseradish peroxidase (HRP)-conjugated swine antiserum to rabbit immunoglobulins (Dako, diluted 1:1000) was added for an additional 2μ3 h. After the final 4 washes, HRP activity was visualized with tetramethylbenzidine over a 30 min incubation in the dark, the reaction was stopped by adding 50 μl of 1 M H2SO4 to each well and the absorbance at 450 nm was recorded in a Molecular Devices ELISA plate reader. All assay points were in duplicate and the specificity controls included wells coated with irrelevant monoclonal antibodies (to replace PAb421 or PAb419), or normal rabbit serum used instead of CM-1.

Immunofluorescence

For immunofluorescence examination, the cells were cultured on sterile glass coverslips, washed twice in PBS, fixed in a cold (–15°C) mixture of methanol and acetone (1:1, v/v) for 7 min, rehydrated in PBS and incubated with primary antibodies. Biotinylated secondary antibodies (Vector Laboratories) were diluted 1:100 in PBS with 10% FCS, all washes were done with PBS and the bound antibodies were detected with Texas red-conjugated streptavidin (Vector Laboratories) at 1:100 dilution.

Immunohistochemistry on paraffin sections

Antibody CM-1 has been found to work well on both frozen material and routinely fixed paraffin-embedded material. Frozen sections were cut, fixed and stained immediately. Sections were placed on poly-L-lysine (Sigma)-coated slides and allowed to dry at room temperature for an hour before being fixed. A 50:50 (v/v) solution of methanol:acetone, previously kept at —20°C, was used for fixation. The slides were placed in the staining rack and the cold fixative was applied with a Pasteur pipette. Sections were left for 10 min, during which time the organic solution evaporated and the sections were allowed to dry and then covered with 0.01 M phosphate-buffered saline, pH 7.6 (PBS), for 10 min. When it was not possible to fix and stain immediately the cut sections were wrapped in foil and stored frozen at-70°C until used.

Care has to be taken when selecting the fixative for routine processed material that is to be used for investigation of the presence of p53 protein. In our experience good results can be obtained when methacam (methanol:chloroform:acetic acid, 6:3:1, by vol.) is used for 2 h at room temperature, formol saline (4% formaldehyde, 0.9% sodium choride) is used overnight at 4°C, or phenol formaldehyde (2% phenol in formol saline) is used for 4 h at room temperature. This latter method is a modification of that described by Hopwood et al. (1989). Heat should be kept to a minimum during the entire fixation/processing procedure, although embedding in wax at 60°C cannot be avoided. The morphology of the nuclei in tissue treated in any of these fixatives is excellent and the staining patterns are very easy to study microscopically. Further work needs to be done by comparing staining of optimally fixed tissue with that obtained in the frozen tissue. Early studies in our laboratory indicate that similar results are obtained.

Fixed sections were cut and floated onto poly-L-lysine-coated slides and then allowed to dry overnight at room temperature. Heat must not be used to aid the sticking of the sections to the slides. Sections were dewaxed in two changes of xylene and two of absolute ethanol. Endogenous peroxidase was blocked by immersing the sections for 10 min in methanol:PBS (5:1, v/v) containing 0.2% hydrogen peroxide. The sections were then washed in tap water and covered with PBS. Three changes of PBS were used for all washings between applications of the staining reagents. Non-specific binding was blocked with either 20% foetal calf serum in PBS or 20% normal goat serum in PBS, depending on the choice of detection method for the antibody/antigen reaction. Primary antibody CM-1, diluted in 1:1500 in PBS 1, was applied to sections, which were then incubated either at room temperature for an hour or at 4°C overnight. A peroxidase-linked or alkaline phosphatase-linked avidin/biotin or streptavidin/biotin method was used for detecting the antibody/antigen reaction. The ABC or LSAB kits from Dako, or the Vectastain ABC kit, are all suitable. If a biotinylated second antibody is required, this should be diluted in PBS containing 3% human serum and 15% foetal calf serum. The chosen enzyme activity was then demonstrated with suitable reagents. If it is considered necessary to try to improve the sensitivity of the method, heavy metal enhancement of staining may be used. Light counterstaining may be performed by placing the sections in haematoxylin solution for 10 s. Sections are then washed in tap water, differentiated in acid-ethanol and allowed to turn blue in tap water for 10 min before mounting.

Expression of human p53 in E. coli

The pT7–7 plasmid (Tabor and Richardson, 1985) was modified by cloning a short synthetic double-stranded DNA fragment between the Xbal and EcoRI sites (Fig. 1). This introduced a novel and unique Sphl site at the initiating ATG creating the plasmid pT7–7 Sphl. A fulllength human p53 was cloned in two steps into this vector. First, a 477 bp AValll fragment from the 5’ end of the gene was inserted at the Sphl site and then the rest of the p53 gene was inserted as a 1109 bp PvuII to EcoRI fragment. This resulted in a full-length p53 cDNA open reading frame initiating at the correct endogenous ATG under the control of the Φ 10 T7 polymerase binding site and Φ 10 ribosome binding site (Fig. 1). This construct was transformed into the host E. coli strain BL21(DE3)pLysS and the integrity of the construct confirmed by restriction mapping and sequencing. When expression was induced by addition of isopropyl-μ-D-thiogalactopyranoside (IPTG) to the growth media, a prominent new band of 53 kDa (1 kDa=103Mr) was apparent in the SDS-polyacrylamide gel analysis of bacterial extracts. This new band reacted in immunoblots with a panel of anti-p53 monoclonal antibodies including those (PAbl801) that recognise N-terminal (Banks et al. 1986) as well as C-terminal (PAb421) (Harlow et al. 1981; Wade-Evans and Jenkins, 1985) areas of the p53 protein. Cell fractionation studies showed that the majority of the p53 protein produced was insoluble (Fig. 2). The p53 protein was purified from the insoluble fraction by dissolving the washed pellet in 5 M guanidine hydrochloride, refolding it by selective dilution of the guanidine in the presence of glutathione, and then separation by ion-exchange chromatography. The final preparation yielded approximately 10 mg of purified refolded human p53 for each litre of bacterial culture. We noted that the plasmid construct was rather unstable in the expression host strain. Before each protein preparation, therefore, a fresh transformation was carried out using a stock of pure plasmid and fresh competent cells.

Production and characterisation of new anti-p53 antibodies

The pure soluble recombinant human p53 was a potent immunogen and high titre anti-p53 antibodies (called CM-1) were produced after a short immunisation protocol. This was probably due at least in part to the relatively high doses of antigen available for immunisation. Three separate animals all produced equivalent responses. The specificity of the CM-1 serum was confirmed in an extensive immunochemical analysis. Initial results suggested that the CM-1 antibody unlike any other anti-p53 reagent was able to detect the protein in archival histology specimens fixed in formalin and embedded in paraffin (see below). Because of the potential of this observation we were anxious to confirm rigorously the specificity of this serum. The antibody stained the nuclei of SV40-transformed human kératinocytes and of the human breast carcinoma cell fine BT-20 (Fig. 3). This line contains a homozygous mutant p53 gene that results in the production of a stable p53 protein. The staining pattern was equivalent to that seen with other anti-p53 antibodies on these cell lines. The nuclei showed an intense granular immunofluorescence with exclusion of stain from the nucleoli. As with other anti-p53 antibodies there was no staining of the chromosomes of metaphase cells, and instead the reactivity was dispersed throughout the metaphase cell (Fig. 3). The antibody failed to stain the Hs913T cell fine that does not express p53.

Fig. 3.

Immunofluorescence visualization of p53 with the CM-1 antiserum. (A) Strong nuclear positivity excluding nucleoli as seen in cultured human SV40-transformed kératinocyte line SVK14. ×260. (B) Nuclear accumulation of mutant p53 protein in BT-20 human breast cancer cell line. ×800.

Fig. 3.

Immunofluorescence visualization of p53 with the CM-1 antiserum. (A) Strong nuclear positivity excluding nucleoli as seen in cultured human SV40-transformed kératinocyte line SVK14. ×260. (B) Nuclear accumulation of mutant p53 protein in BT-20 human breast cancer cell line. ×800.

Immunoblots of whole cell extracts (Fig. 4) showed that the CM-1 antibody recognised a single species of 53 kDa present in extracts from p53 mutant-expressing cell lines Namalwa, RD and PMC-42 (lanes 2,4,5). The band precisely comigrated with the p53 band detected with the monoclonal antibody PAb421 (lane 1). The CM-1 antibody did not react with any band in an equivalent immunoblot of the p53-negative Hs913T cells. As final confirmation of the specificity of the antibody we immunoprecipitated the BT-20 cell lysates with CM-1, an anti-p53 monoclonal antibody (PAb421) and two control antibodies. Immunoblots of these immunoprecipitates were then probed with PAb421 and with CM-1. The results showed that PAb421 and CM-1 recognised the same protein species and that both reactions were specific.

Fig. 4.

Characterization of the CM-1 antiserum by immunoblotting and immunoprecipitation. In immunoblots of the whole-cell lysates of Namalwa cells (lanes 1, 2, 3) both the control PAbl801 (lane 1) and CM-1 (lane 2) show a positive 53 kDa band that is not seen by the negative control antibody (lane 3). Similar 53 kDa bands are specifically recognized by CM-1 in immunoblots of wholecell lysates of RD (lane 4) and PMC-42 (lane 5) cell lines expressing high levels of p53, while no reactivity is seen on Hs913T (lane 6) cells that lack any p53 transcripts. Immunoblots of immunoprecipitates from BT-20 cell lysate (lanes 7, 8, 9, 10) show that CM-1 (lane 7) brings down the p53 protein recognized by PAb421 and vice versa: PAb 421 immunoprecipitates a protein recognized by CM-1 (lane 9). Corresponding control immunoprecipitates with normal rabbit serum (lane 8) and PAb419 (lane 10) are negative when immunoblotted with PAb421 and CM-1, respectively. Air values of the markers are given (×10−3).

Fig. 4.

Characterization of the CM-1 antiserum by immunoblotting and immunoprecipitation. In immunoblots of the whole-cell lysates of Namalwa cells (lanes 1, 2, 3) both the control PAbl801 (lane 1) and CM-1 (lane 2) show a positive 53 kDa band that is not seen by the negative control antibody (lane 3). Similar 53 kDa bands are specifically recognized by CM-1 in immunoblots of wholecell lysates of RD (lane 4) and PMC-42 (lane 5) cell lines expressing high levels of p53, while no reactivity is seen on Hs913T (lane 6) cells that lack any p53 transcripts. Immunoblots of immunoprecipitates from BT-20 cell lysate (lanes 7, 8, 9, 10) show that CM-1 (lane 7) brings down the p53 protein recognized by PAb421 and vice versa: PAb 421 immunoprecipitates a protein recognized by CM-1 (lane 9). Corresponding control immunoprecipitates with normal rabbit serum (lane 8) and PAb419 (lane 10) are negative when immunoblotted with PAb421 and CM-1, respectively. Air values of the markers are given (×10−3).

Sensitive two-site ELISA assay for p53

The availability of a rabbit polyclonal antibody to p53 permitted the establishment of a sensitive two-site ELISA assay for p53 without requiring the specific labelling of either of the anti-p53 reagents. In this assay design plates are coated with mouse monoclonal antip53 antibody, antigen is added, and then bound antigen is detected using unlabelled CM-1 serum, and finally a commercial enzyme labelled anti-rabbit immunoglobulin. Provided that the anti-rabbit immunoglobulin does not react with mouse immunoglobulin, this provides a very simple and accurate quantitative test for p53. The results are very satisfactory (Fig. 5). Using a mono-clonal antibody to SV40 T antigen as a solid-phase capture antibody we are also able to use the assay to detect T-p53 complexes. The assay shows exceptionally low’ backgrounds when p53 negative cell extracts or control capture antibodies are used. The assay is sensitive to sub ng/ml amounts of p53.

Fig. 5.

Application of the CM-1 antiserum in two-site immunoassays. The ELISA plates were coated with PAb421 (▼—— ▼), PAb419 (•–•) or a negative control antibody (×–×) and the CM-1 antiserum was used to detect the captured p53 protein. (A) High levels of both the p53 protein (▼—— ▼) and T-p53 complexes (•–•) are detected in SV40-transformed human kératinocyte (SVK14) cell extracts. (B) Accumulation of p53 (▼—— ▼) but no T-p53 complexes (•–•) are detectable in extracts of A431 cells known to express the mutant p53 protein. (C) Neither p53 (▼—— ▼) nor T-p53 complexes (•–•) are present in the Hs913T cell line, reported to lack p53 gene transcripts.

Fig. 5.

Application of the CM-1 antiserum in two-site immunoassays. The ELISA plates were coated with PAb421 (▼—— ▼), PAb419 (•–•) or a negative control antibody (×–×) and the CM-1 antiserum was used to detect the captured p53 protein. (A) High levels of both the p53 protein (▼—— ▼) and T-p53 complexes (•–•) are detected in SV40-transformed human kératinocyte (SVK14) cell extracts. (B) Accumulation of p53 (▼—— ▼) but no T-p53 complexes (•–•) are detectable in extracts of A431 cells known to express the mutant p53 protein. (C) Neither p53 (▼—— ▼) nor T-p53 complexes (•–•) are present in the Hs913T cell line, reported to lack p53 gene transcripts.

Expression of p53 protein in breast cancer

Having validated the specificity of the CM-1 serum we then carried out an extensive investigation of its reactivity with a wide range of human breast tumours. We compared the effects of fixation, with respect to both time and type. We also examined the effects of how the tissue was processed and mounted for immunohistochemistry.

Excellent results were obtained with material fixed in methacam, formol saline or phenol-formaldehyde as well as frozen sections. A number of factors were found to reduce the intensity of the p53 reaction. Prolonged fixation periods (over 48 h) in formol saline and the presence of calcium in the fixative were both deleterious. Heating of the sections normally used to aid their adherance to the slide was also harmful. We developed a protocol using poly-L-lysine-coated slides and overnight incubation, which gave very good results.

After examination of over two hundred correctly processed examples we were able to summarise the characteristics of staining into six basic patterns. The most striking pattern was an intense nuclear positive staining confined to the tumour cells and present in the majority (>65%) of the tumour cells, this was seen in about 15% of cases (Fig. 6A). The stain was excluded from condensed chromosomes and showed some variation in intensity from nucleus to nucleus. Strong positive staining patterns of this type were most common in advanced grade III tumours. In 35% of the tumours the staining reaction was slightly less intense and fewer of the nuclei were stained. The positive cells were evenly distributed throughout the lesion. A further 20% of tumours showed a very weak positive stain, while 20% showed a complete absence of reaction. When staining occcurred in cases of carcinoma in situ it was often noted that the cells at the outer margin of the lesion stained most intensely (Fig. 6B). Finally, two unusual but nevertheless quite characteristic patterns were also noted. In one of these, p53 staining was seen in the cytoplasm (Fig. 6C) rather than the nucleus. The reaction could be intense and the staining was usually focal and perinuclear. In the second pattern very rare, strongly positive nuclei were seen scattered throughout the tumour. The positive cells could not be distinguished morphologically from the negative cells and always occurred as single cells not as pairs or clusters (Fig. 6D).

Fig. 6.

Immunohistochemical staining of p53 in formalin-fixed breast carcinoma tissue. (A) Strong nuclear staining pattern. (B) Strong staining at the margin of an in situ lesion. (C) Strong cytoplasmic staining pattern. (D) Strong staining of scattered nuclei in the lesion.

Fig. 6.

Immunohistochemical staining of p53 in formalin-fixed breast carcinoma tissue. (A) Strong nuclear staining pattern. (B) Strong staining at the margin of an in situ lesion. (C) Strong cytoplasmic staining pattern. (D) Strong staining of scattered nuclei in the lesion.

Mutation in the p53 gene and accumulation of the p53 protein are common changes associated with the development of malignant disease. In order to understand p53 function we need to learn more about the basic biochemical properties of the protein. To that end we set out to produce large quantities of human p53. The expression system we employed is very efficient as it exploits the use of the highly active and specific T7 RNA polymerase, which allowed us to produce fulllength intact p53 in E. coli. Unfortunately the protein produced was not soluble. While careful refolding yielded a soluble preparation, functional studies indicate that this refolding was not authentic, as the soluble protein is defective in T antigen binding (J.V. Gannon, personal communication). The availability of large amounts of p53 has, however, allowed us to produce very high titre polyclonal antibodies and a panel of new monoclonal antibodies (Bártek et al. unpublished results) to p53. These new antibodies are of great significance, since they are able to recognise p53 in archival material. Indeed, we have recently obtained excellent results from 35-year-old blocks. Because of the importance of reagents of this kind it is essential to verify their specificity. Our data indicate that the serum is monospecific, consistent with the purity of the antigen preparation.

The variety of staining patterns obtained may then be interpreted with some confidence. Work from our own laboratory and more recent studies from other groups have shown that there is a simple relationship between mutation of the p53 gene and the high level of expression of the p53 protein in tumours (Bártek et al. 1990b; Bennett et al. 1991; Iggo et al. 1990; Rodrigues et al. 1990; Wright et al. 1991; Davidoff et al. 1991; Marks et al. 1991). This phenotype is present in about 15% of breast tumours but occurs at a much higher rate in lung and colon cancers. It appears to be caused by the increased stability of mutant p53 proteins in the environment of a tumour cell (Reihsaus et al. 1990).

The biochemical basis of the other phenotypes is not completely resolved, though they are all, except the null phenotype, specific to the area of the tumour and therefore must represent some tumour-specific alteration in the control of p53 expression. The pattern of staining in which fewer (approximately 30%) of the cells are positive has been seen in teratocarcinoma cells in culture that have been reported to contain a normal p53 gene (Gannon et al. 1990). In this case the high level of p53 may represent transcriptional activation (Reich et al. 1983). Cytoplasmic staining may be found in cultured cells when the nuclear transport signal is absent or its activity is counteracted by binding to the adenovirus Elb oncoprotein (Zantema et al. 1985) or perhaps by heat shock protein 70 binding (Gannon and Lane, 1991). The pattern in which rare positive cells are seen is most difficult to explain but again it has been encountered in tumour cell lines (Bártek et al. 1990a; Bártek et al. 1990b). It may represent an epigenetic event or a lethal phenotype, since it does not seem to be found in daughter cells. Finally, the absence of any detectable staining in a tumour as seen in normal tissue may just reflect the normal very low level of p53 or in some case deletion of both alleles of the p53 gene (Bennett et al. 1991). If all of these changes in p53 play a causal role in tumour growth then they represent a prime target for the development of new diagnostic criteria and therapeutic approaches for up to 60% of human breast cancers and many other tumour types.

This research was supported by the ICRF and the CRC. B.V. is a long-term fellow of EMBO.

Banks
,
L.
,
Matlashewski
,
G.
and
Crawford
,
L.
(
1986
).
Isolation of human p53 specific monoclonal antibodies and their use in the studies of human p53 expression
.
Eur. J. Biochem
.
159
,
529
534
.
Bártek
,
J.
,
Bártková
,
J.
,
VojtfJek
,
B.
,
StaJková
,
Z.
,
Lukáí
,
J.
,
Rejthar
,
A.
,
Kovaffk
,
J.
,
Midgley
,
C. A.
,
Gannon
,
J. V.
and
Lane
,
D. P.
(
1991
).
Aberrant expression of the p53 oncoprotein is a common feature of a wide spectrum of human malignancies
.
Oncogene
6
(in press).
Bártek
,
J.
,
Bártková
,
J.
,
VojtéJek
,
B.
,
Staáková
,
Z.
,
Rejthar
,
A.
,
Kovaffk
,
J.
and
Lane
,
D. P.
(
1990a
).
Patterns of expression of the p53 tumour suppressor in human breast tissues and tumours in situ and in vitro
.
Int. J. Cancer
46
,
839
844
.
Bártek
,
J.
,
Durban
,
E. M.
,
Hallowes
,
R. C.
and
Taylor-Papadimitriou
,
J.
(
1985
).
A subclass of luminal epithelial cells in the human mammary gland defined by antibodies to cytokeratins
.
J. Cell Sci
.
75
,
1
17
.
Bártek
,
J.
,
Iggo
,
R.
,
Gannon
,
J.
and
Lane
,
D. P.
(
1990b
).
Genetic and immunochemical analysis of mutant p53 in human breast cancer cell lines
.
Oncogene
5
,
893
899
.
Bártková
,
J.
,
Bártek
,
J.
,
LukáJ
,
J.
,
VojtKek
,
B.
,
StaJková
,
Z.
,
Rejthar
,
A.
,
Kovaffk Midgley
,
C. A.
and
Lane
,
D. P.
(
1991
).
p53 protein alterations in human testicular cancer including preinvasive intratubular germ-cell neoplasia
.
Int. J. Cancer
49
(in press).
Bennett
,
W. P.
,
Hollstein
,
M. C.
,
He
,
A.
,
Zhu
,
S. M.
,
Resau
,
J.
,
Trump
,
B. F.
,
Metcalf
,
R. A.
,
Welsh
,
J. A.
,
Gannon
,
J. V.
,
Lane
,
D. P.
and
Harris
,
C. C.
(
1991
).
Archival analysis of p53 genetic and protein alterations in Chinese esophageal cancer
.
Oncogene
6
(in press).
Davidoff
,
A. M.
,
Humphrey
,
P. A.
,
Iglehart
,
J. D.
and
Marks
,
J. R.
(
1991
).
Genetic basis for p53 overexpression in human breast cancer
.
Proc. Nat. Acad. Sci. U.S.A
.
88
,
5006
5010
.
Gannon
,
J. V.
,
Greaves
,
R.
,
Iggo
,
R.
and
Lane
,
D. P.
(
1990
).
Activating mutations in p53 produce a common conformational effect. A monoclonal antibody specific for the mutant form
.
EMBO J
.
9
,
1595
1602
.
Gannon
,
J. V.
and
Lane
,
D. P.
(
1991
).
Protein synthesis required to anchor a mutant p53 protein which is temperature-sensitive for nuclear transport
.
Nature
349
,
802
806
.
Harlow
,
E.
,
Crawford
,
L. V.
,
Pim
,
D. C.
and
Williamson
,
N. M.
(
1981
).
Monoclonal antibodies specific for simian virus 40 tumour antigen
.
J. Virol
.
39
,
861
869
.
Hollstein
,
M.
,
Sidransky
,
D.
,
Vogelstein
,
B.
and
Harris
,
C.
(
1991
).
p53 mutations in human cancer
.
Science
253
,
49
53
.
Hopwood
,
P.
,
Slidders
,
W.
and
Wayman
,
G. R.
(
1989
).
Tissue fixation with phenol formaldehyde for routine histopathology
.
Histochem. J
.
21
,
228
234
.
Iggo
,
R.
,
Gatter
,
K.
,
Bártek
,
J.
,
Lane
,
D.
and
Harris
,
A. L.
(
1990
).
Increased expression of mutant forms of p53 oncogene in primary lung cancer
.
Lancet
335
,
675
679
.
Lane
,
D.
and
Benchimol
,
S.
(
1990
).
p53: Oncogene or anti-oncogene
.
Genes Dev
.
4
,
1
8
.
Levine
,
A. J.
and
Momand
,
J.
(
1990
).
Tumor suppressor genes: the p53 and retinoblastoma sensitivity genes and gene products
.
Biochim. Biophys. Acta
1032
,
119
136
.
Levine
,
A. J.
,
Momand
,
J.
and
Finlay
,
C. A.
(
1991
).
The p53 tumor suppressor gene
.
Nature
351
,
453
456
.
Malkin
,
D.
,
Li
,
F. P
,,
Strong
,
L. C.
,
Fraumeni
,
J. F
,,
Nelson
,
C. M.
,
Kim
,
D. H.
,
Kassel
,
J.
,
Gryka
,
M. A.
,
Bischoff
,
F. Z.
,
Tainsky
,
M. A.
and
Friend
,
S. H.
(
1990
).
Germline p53 mutations in a familial syndrome of breast cancer’sarcomas and other neoplasias
.
Science
250
,
1233
1238
.
Marks
,
J. R.
,
Davidoff
,
A. M.
,
Kerns
,
B. J.
,
Humphrey
,
P. A.
,
Pence
,
J. C.
,
Dodge
,
R. K.
,
Clarke-Pearson
,
D. L.
,
Iglehart
,
J. D.
,
Bats
,
R. C.
and
Berchuck
,
A.
(
1991
).
Overexpression and mutation of p53 in epithelial ovarian cancer
.
Cancer Res
.
51
,
2979
2984
.
Marshall
,
C. J.
(
1991
).
Tumor suppressor genes
.
Cell
64
,
313
326
.
Nigro
,
J. M.
,
Baker
,
S. J.
,
Preisinger
,
A. C.
,
Jessup
,
J. M.
,
Hostetter
,
R.
,
Cleary
,
K.
,
Bigner
,
S. H.
,
Davidson
,
N.
,
Baylin
,
S.
,
Devilee
,
P.
,
Glover
,
T.
,
Collins
,
F. S.
,
Weston
,
A.
,
Modali
,
R.
Harris
,
C. C.
and
Vogelstein
,
B.
(
1989
).
Mutations in the p53 gene occur in diverse human tumour types
.
Nature
342
,
705
708
.
Reich
,
N. C.
,
Oren
,
M.
and
Levine
,
A. J.
(
1983
).
Two distinct mechanisms regulate the level of a cellular tumor antigen, p53
.
Mol. Cell. Biol
.
3
,
2143
2150
.
Reihsaus
,
E.
,
Kohler
,
M.
,
Kraiss
,
S.
,
Oren
,
M.
and
Montenarh
,
M.
(
1990
).
Regulation of the level of the oncoprotein p53 in nontransformed and transformed cells
.
Oncogene
5
,
137
145
.
Rodrigues
,
N. R.
,
Rowan
,
A.
,
Smith
,
M. E. F.
,
Kerr
,
I. B.
,
Bodmer
,
W. F.
,
Gannon
,
J.
and
Lane
,
D. P.
(
1990
).
p53 mutations in colorectal cancer
.
Proc. Nat. Acad. Sci. U.S.A
.
87
,
7555
7559
.
Studier
,
F. W.
and
Moffatt
,
B. A.
(
1986
).
Use of bacteriophage T7 RNA polymerase to direct selective high level expression of cloned genes
.
J. Mol. Biol
.
189
,
113
130
.
Studier
,
F. W.
,
Rosenberg
,
A. H.
,
Dunn
,
J. J.
and
Dubendorff
,
J. W.
(
1990
).
Use of T7 RNA polymerase to direct expression of cloned genes
.
Meth. Enzym
.
185
,
60
89
.
Tabor
,
S.
and
Richardson
,
C. C.
(
1985
).
A bacteriophage T7 polymerase/promoter system for controlled exclusive expression of specific genes
.
Proc. Nat. Acad. Sci. U.S.A
.
82
,
1074
1078
.
Wade-Evans
,
A.
and
Jenkins
,
J. R
, (
1985
).
Precise epitope mapping of the murine transformation-associated protein, p53
.
EMBO J
.
4
,
699
706
.
Wright
,
C.
,
Mellon
,
K.
,
Johnston
,
P.
,
Lane
,
D. P.
,
Harris
,
A. L.
,
Horne
,
C. H. W.
and
Neal
,
D. E.
(
1991
).
Expression of mutant p53, c-erbB-2 and the epidermal growth factor receptor in transitional cell carcinoma of the human urinary bladder
.
Br. J. Cancer
63
,
967970
.
Zantema
,
A.
,
Schrier
,
P. L
,
Davies-Olivier
,
A.
,
Van Laar
,
T.
,
Vaessen
,
R. T. M. J.
and
Van der Eb
,
A. J.
(
1985
).
Adenovirus serotype determines association and localisation of the large Elb tumor antigen with cellular tumor antigen p53 in transformed cells
.
Mol. Cell. Biol
.
5
,
3084
3091
.