EXPRESSION OF p53 IN ADRENOCORTICAL TUMOURS: CLINICOPATHOLOGICAL CORRELATIONS
A. M. McNICOL, C. E. NOLAN, A. J. STRUTHERS, M. A. FARQUHARSON, J. HERMANS* AND H. R. HAAKŤ
University Department of Pathology, Glasgow Royal Infirmary University NHS Trust, Glasgow, U.K .; * Department of Medical Statistics, University Hospital, Leiden, The Netherlands; } Department of Medicine, Diaconessenhuis, Eindhoven, The Netherlands
SUMMARY
There are limited data to suggest that abnormalities of p53 expression may be a late event in the development of adrenocortical tumours. This has been investigated further by examining a series of adrenocortical adenomas and carcinomas by immunohistochemistry for p53 expression and a subset for evidence of mutation in exons 5-8 of the p53 gene using polymerase chain reaction/single strand conformational polymorphism (PCR/SSCP). In carcinomas, the findings have been correlated with survival data and with tumour ploidy. Immunopositivity for p53 was seen in 4 of 34 adenomas and 22 of 42 carcinomas. Mobility shifts were identified in 1 of 4 adenomas and 10 of 21 carcinomas. There was no correlation between immunostaining pattern or PCR/SSCP evidence of mutation and either survival or disease-free survival in carcinoma. There was also no correlation between p53 status and tumour ploidy. While these findings support a role for p53 in tumour progression, abnormal p53 expression does not appear to have any significant prognostic effect in carcinoma. KEY WORDS-adrenal cortex; neoplasm; p53 gene; immunohistochemistry; PCR/SSCP; ploidy; survival
INTRODUCTION
The molecular genetic basis of adrenocortical tumours is unclear, but loss of heterozygosity on chromosomes 11p, 13q, and 17p in adrenocortical carcinoma suggests that genes located at these sites may be important in tumourigenesis or tumour progression.1 Sporadic adrenocortical carcinoma is rare, but the tumour occurs more frequently in the Li-Fraumeni2 syndrome, associ- ated with germline mutations in the tumour suppressor gene p533 on chromosome 17p13. p53 must therefore also be a strong candidate gene in the pathogenesis of sporadic adrenocortical tumours. The p53 gene encodes a protein which plays significant roles in cell-cycle control, DNA repair, and apoptosis.4 Inactivation of the protein might therefore be expected to result in cells progressing into S phase with damaged DNA, leading to genomic instability5 with the development and progres- sion of neoplastic clones. Inactivation is common in human malignancies, most often due to mutation of the p53 gene,6 but also due to binding of the wild-type protein to a variety of endogenous or viral proteins.7 Most p53 mutations occur in exons 5-8,6 in some tumours as an early event8 and in others late.9,10 In some, accumulation of p53 protein has been reported to correlate with more aggressive clinical behaviour.11,12
Previous studies on small series of adrenocortical tumours13-15 have provided evidence for mutations in the conserved regions of p53 in exons 5-8, most prob- ably as a late event, since they were detected more frequently in carcinomas than in adenomas.13,15 Reported mutations in exon 4 in 12 of 15 adrenocortical adenomas, mainly associated with Conn’s syndrome, 14
are of uncertain significance at present, as less than 5 per cent of mutations occur outside exons 5-8 in other tumours and this region has not been widely investigated.
To investigate more fully the role of p53 in adreno- cortical tumours, we firstly applied immunohistochemis- try to a series of cases to detect p53 protein. Wild-type p53 has a short half-life and is usually not detected by this method, whereas mutated p53, or p53 bound to other proteins, is stabilized16 and accumulates to levels which are detectable. To look for mutations in exons 5-8, we used polymerase chain reaction/single strand conformational polymorphism (PCR/SSCP) analysis, in which PCR products from mutated sequences show a band shift on gel analysis compared with wild type. We compared the findings in adenomas and carcinomas to assess whether changes in p53 expression are an early or late event. We also correlated survival data from carci- noma with p53 status, to assess whether p53 is a prognostic factor in this disease. Finally, we examined the relationship between p53 immunoreactivity and ploidy, since genomic instability associated with p53 inactivation might result in the development and expansion of aneuploid clones.5
MATERIALS AND METHODS
Cases
Thirty-four adrenocortical adenomas and 42 carcino- mas were included, classified on the basis of clinical behaviour and on the histological criteria of van Slooten et al.17 All were fixed in formalin and embedded in paraffin wax. Normal adrenal glands obtained at surgery for non-adrenal disease, or from autopsies where there was not evidence of adrenal dysfunction, were used as controls.
Addressee for correspondence: Dr A. M. McNicol, Department of Pathology, Glasgow Royal Infirmary University NHS Trust, Castle Street, Glasgow G4 0SF, U.K.
Immunocytochemistry
Sections (5 um thick) were cut and mounted on glass slides coated with 3-aminopropyltriethoxysilane (Sigma). They were stained using an avidin-biotin tech- nique with a primary sheep polyclonal antibody to p53, S206-20 (Scottish Antibody Production Unit, SAPU), which was raised to recombinant p53 protein and recog- nizes both wild-type and mutated p53.22 Slides were microwave treated in citrate buffer, pH 6.0, for 20 min prior to staining. A biotinylated polyclonal rabbit anti- sheep immunoglobulin (Vector Laboratories) was used as the secondary antibody, with 3’,3’-diaminobenzidine as chromogen. A breast carcinoma with a proven muta- tion in the p53 gene was used as a positive control. For negative controls, the primary antibody was replaced by non-immune sheep serum.
Cases were classified as positive if there was diffuse nuclear staining throughout the tumour, or where sig- nificant numbers of positive nuclei were seen in most high-power fields. Where only a few positive nuclei were seen, these tumours were classified as negative.
PCR/SSCP
Twenty-one carcinomas and four adenomas were ana- lysed by PCR amplification of sequences in exons 5-8 known to have common mutations. These comprised 13 immunopositive carcinomas, eight immunonegative car- cinomas, and four immunonegative adenomas. Positive controls were breast carcinomas with proven mutations in each exon. Human tonsil was the normal control. The primer sequences were:
Exon 5: - TTCCTCTTCCTACAGTACTC-,
-CCCAGCTGCTCACCATCG-
Exon 6: - CCTCACTGATTGCTCTTAGG- -AGTTGCAAACCAGACCTCAG-
Exon 7: - GTGTTGTCTCCTAGGTTGGCT- -AAGACTCCAGGTCAGGAGCCACT-
Exon 8: - CCTATCCTGAGTAGTGGT- -AAGCGAGGTAAGCAAGCAGGA-
Ten-micrometre sections were dewaxed in xylene fol- lowed by alcohol and then dried in a vacuum desiccator. The samples were digested overnight at 37℃ with 500 u/l proteinase K (Sigma), inactivated by heating at 95℃ for 10 min, and stored at 4℃. Some samples required further phenol/chloroform extraction of DNA before amplification could be achieved.
The PCR mixture consisted of 1-5 ul of genomic DNA, 1 unit of Taq polymerase (Perkin Elmer), oligo- nucleotide primers at 1 µM, aNTPs at 200 UM in 10 mM Tris-HCI, 10 mM potassium chloride, 1-5 mM magne- sium chloride, and 0-001 per cent gelatin (Perkin-Elmer). The samples underwent 50 cycles of amplification (96°℃ × 1 min, 58°℃ × 1 min, 72°℃ × 1 min) in a ther- mal cycler (Hybaid). Amplification was confirmed by the presence of a single band of appropriate size on ethidium bromide-stained agarose gels.
For SSCP analysis, 5 ul of each sample was denatured with 1 ul of 0.5 M sodium hydroxide/10 mM EDTA at
| p53-positive | p53-negative | |
|---|---|---|
| Sex | ||
| Male | 8 | 7 |
| Female | 14 | 13 |
| Hormone profile | ||
| Cushing's syndrome | 3 | 2 |
| Cushing's/virilization | 6 | 10 |
| Virilization | 4 | 3 |
| Feminization | 1 | 0 |
| Conn's syndrome | 0 | 1 |
| Non-hormone secreting | 8 | 4 |
| Age | ||
| ≤40 years | 7 | 15 |
| >40 years | 15 | 5 |
| Metastases | ||
| No | 13 | 15 |
| Yes | 9 | 5 |
| Surgery | ||
| No | 1 | 1 |
| Subtotal resection | 9 | 5 |
| Total resection | 12 | 14 |
42℃ for 5 min. 2.5 ul of loading buffer (95 per cent formamide, 0-05 per cent bromophenol blue, 0.05 per cent xylene cyanol) was added and samples were loaded onto a 0.5 x MDE (Mutation Detection Enhancer, Hoefler Scientific Instruments) gel containing 5 per cent glycerol. Gels were run at 15 W for 3-4 h at room temperature on a Protean II gel apparatus (Bio-Rad). Bands of DNA were detected with a silver staining kit (Bio-Rad) and gels were dried on a gel drier (Bio-Rad).
Ploidy studies
Tumour ploidy was assessed by flow cytometry as described previously.18 Tumours were classified on the basis of the DNA index (DI) as (near) diploid (DI 1.00), hyperdiploid (DI 1.01-1-40), hypotetraploid (DI 1.41- 1.89), or tetraploid (DI 1.90-2.10).
Statistical analysis
Survival curves, calculated from the time of diagnosis to the time of death, and disease-free survival curves, calculated from the time of diagnosis to the time of tumour recurrence, were analysed by the Kaplan-Meier method. The Lee-Desu statistic test was used for com- parison of survival curves.19 Data were available from 41 patients in the immunocytochemical study and 20 in the PCR/SSCP study. These have been the subject of previous studies. 18,20
RESULTS
Immunocytochemistry
There was no staining in normal adrenal glands. The breast carcinoma with a known mutation showed strong
SURVIVAL
100
p=0.15
75
50
p53 -
25
p53+
0
0
12
24
36
48
60
72
84
96
MONTHS
| Near-diploid | Hyperdiploid | Hypotetraploid | Tetraploid | |
|---|---|---|---|---|
| p53+ | 2 | 7 | 11 | 1 |
| p53 - | 0 | 4 | 11 | 2 |
| Mutation | 0 | 1 | 6 | 1 |
| No mutation | 1 | 2 | 6 | 1 |
positive nuclear staining. Four of 34 adenomas (11.8 per cent) and 22 of 42 carcinomas (52.4 per cent) were classified as positive. The clinical features of the carci- nomas are shown in Table I. Immunopositive tumours appear to occur more commonly in older patients and the incidence of metastatic disease at the time of surgery or within 3 months of the first operation was higher in immunopositive (40-9 per cent) than in immunonegative cases (25 per cent).
Survival and disease-free survival curves are shown in Figs 1 and 2. There was no significant difference in either analysis patients with immunopositive or immunonegative tumours.
Correlation of ploidy status with p53 immuno- reactivity is shown in Table II. There was no difference in distribution between immunopositive and immuno- negative cases.
PCR/SSCP analysis
Normal tonsil showed appropriate bands in all runs. The breast carcinomas used for each exon showed changes in banding (Fig. 3). In 3 of 25 tumours, the exon 8 sequence could not be amplified and in one, exon 6. Band shifts were identified in 11 tumours (44 per cent) (Fig. 3 and Table III). Six of 13 (46.1 per cent)
SURVIVAL
100
p=0.48
75
50
25
p53 -
p53+
0
0
12
24
36
48
60
72
84
96
MONTHS
1
2
3
4
V
immunopositive carcinomas, 4 of 8 (50 per cent) im- munonegative carcinomas, and 1 of 4 (25 per cent) immunonegative adenomas showed a band shift.
| p53-positive carcinoma | p53-negative carcinoma | p53-negative adenoma | |
|---|---|---|---|
| Exon 5 | 4 | 1 | 0 |
| Exon 6 | 1 | 1 | 0 |
| Exon 7 | 1 | 0 | 0 |
| Exon 8 | 0 | 2 | 1 |
| No mutation | 7 | 4 | 3 |
Survival curves and disease-free survival are shown in Figs 4 and 5. Again, there was no significant difference between patients with band shifts and those without. Ploidy status of positive and negative tumours is shown in Table II. There was no difference in distribution between the two groups.
DISCUSSION
Two previous studies have examined adrenocortical tumours for p53 mutations in the usual sequences of exons 5-8. In the first,13 PCR/SSCP and sequencing
SURVIVAL
100
p=0.24
75
p53 mutant
50
25
p53 wild type
0
0
12
24
36
48
60
72
84
96
MONTHS
were applied to 18 adenomas, 15 carcinomas, and three metastases. One adenoma showed a non-miscoding mutation at codon 295, and three carcinomas showed mutations, one in exon 8 and the others in exon 5. An additional three carcinomas showed band shifts, but no sequencing abnormality. Immunostaining was not performed in that study. The second study15 applied immunohistochemistry, PCR, and sequencing to five adenomas, 11 carcinomas, and two cell lines. The authors detected no immunoreactivity in adenomas. Two adrenal carcinomas were strongly positive, with a further three showing less than 10 per cent positive cells. These latter cases would have been classified as negative in the present study. Three of 11 carcinomas and both cell lines showed p53 mutations, in three cases point mutations correlating with strong immunoreactivity; in one carcinoma and in the cell line NCI-H-295, large deletions and insertions were associated with negative immunostaining.
The combination of immunocytochemistry and PCR/ SSCP of exons 5-8 without sequencing of the PCR product has been used to identify p53 abnormalities in a variety of other tumour types.21 We did not investigate exon 4.18 Our proportion of immunopositive carci- nomas, at about 50 per cent, was greater than in Reincke et al.’s study.15 This may be related to the different antibody and to the use of microwave antigen retrieval.
Nevertheless, our failure to detect immunoreactivity in normal adrenal and myometrium22 and in the majority of adenomas would indicate that only where increased levels of p53 protein were present were we identifying it.
Our data from PCR/SSCP analysis indicating altered mobility in about 50 per cent of carcinomas and rarely in adenomas are similar to those of Ohgaki et al.13 Because we did not sequence the PCR products, we cannot give an indication as to what proportion of these specifically correlate with mutations. The majority of band shifts were detected in exon 5, where three of six previously reported mutations in sporadic tumours have been sited.13,15 While this may indicate a preferential hot spot in exon 5 in these tumours, each of the previous muta- tions was different. Thus, if there is an adenoma- carcinoma sequence in the adrenal, our data support the theory that p53 abnormalities are a late event. This may reflect a growth advantage and clonal expansion of tumour cells with p53 mutations, as has been suggested in the progression of gliomas.23
In common with other studies,21,22,24 immunohisto- chemical detection of p53 protein and PCR/SSCP analy- sis were not always concordant. Immunopositivity without band shifts most probably reflects stabilization of p53 by some alternative mechanism. Investigation of the expression of p53-binding proteins,7 such as mdm2, in adrenocortical tumours should prove of interest.
SURVIVAL
100
p=0.50
75
50
p53 mutant
25
p53 wild type
0
0
12
24
36
48
60
72
84
96
MONTHS
Immunonegative tumours showing band shifts might have more gross abnormalities of the p53 gene, deletions or translocations causing absence of expression, or the presence of a truncated protein unrecognized by the antibody, as described by Reincke et al.15 In that study, large insertions were identified in exon 8 in two tumours. This might provide an explanation for our inability to amplify exon 8 in three cases, the product being too large to amplify from fixed tissue.
Our analysis showed no correlation between p53 status and either function or outcome in adrenal carci- noma, although there was a suggestion that immuno- positivity tended to occur in older patients and where early metastatic disease was present. Published data concerning the prognostic correlations of abnormal p53 expression are variable11,12,25 and it has been suggested that significant value is found mainly in studies of large series.26 With rare tumours such as adrenocortical carcinoma, it is difficult to obtain good clinical data on large numbers. Our findings, however, add weight to the opinion that p53 is not of major clinical prognostic significance across a broad range of tumours.
Wild-type p53 prevents the progression into S phase of cells with damaged DNA, and therefore in tumours where the protein is inactivated, aneuploid clones might expand.5 This would imply that p53-immunopositive tumours should be more commonly aneuploid than
immunonegative tumours. Our analysis of adrenocorti- cal carcinomas did not confirm this hypothesis. In these tumours, the development of aneuploidy appears to be governed by mechanisms other than p53.
REFERENCES
1. Yano T, Lineham M, Anglard P, et al. Genetic changes in adrenocortical carcinomas. J Natl Cancer Inst 1989; 81: 518-523.
2. Li FP, Fraumeni JF Jr. Soft-tissue sarcomas, breast cancer, and other neoplasms: a familial syndrome? Ann Intern Med 1969; 71: 747-752.
3. Srivastava S, Zou Z, Pirollo K, Blattner W, Chang EH. Germline transmis- sion of a mutated p53 gene in a cancer-prone family with Li-Fraumeni syndrome. Nature 1990; 348: 747-749.
4. Lane DP. p53, guardian of the genome. Nature 1992; 358: 15-16.
5. Hartwell L. Defects in a cell cycle checkpoint may be responsible for the genomic instability of cancer cells. Cel/ 1992; 71: 543-546.
6. Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 mutations in human cancers. Science 1991; 253: 49-53.
7. Pietenpol JA, Vogelstein B. No room at the p53 inn. Nature 1993; 365: 17-18.
8. Schmitt FC, Leal C, Lopes C. p53 protein expression and nuclear DNA content in breast intraductal proliferations. J Pathol 1995; 176: 233-241.
9. Mazars R, Pujol P, Mauadelonde T, Jeanteur P, Theillet C. p53 mutations in ovarian cancer: a late event? Oncogene 1991; 6: 1685-1690.
10. Nakamura T, Yana I, Kobayashi T, et al. p53 gene mutations associated with anaplastic transformation of human thyroid carcinomas. Jpn J Cancer Res 1992; 83: 1293-1298.
11. Andersen TI, Holm R, Nesland JM, Hamdal KR, Ottestad L, Borresen AL. Prognostic significance of TP53 alterations in breast carcinoma. Br J Cancer 1993; 68: 540-548.
12. Silvestrini R, Benini E, Daidone MG, et al. p53 as an independent prognostic marker in lymph node-negative breast cancer patients. J Natl Cancer Inst 1993; 85: 965-970.
13. Ohgaki H, Kleihues P, Heitz PU. p53 mutations in sporadic adrenocortical tumors. Int J Cancer 1993; 54: 408-410.
14. Lin S-R, Lee Y-J, Tsai J-H. Mutations of the p53 gene in human functional adrenal neoplasms. J Clin Endocrinol Metab 1994; 78: 483-491.
15. Reincke M, Karl M, Travis WH, et al. p53 mutations in human adreno- cortical neoplasms: immunohistochemical and molecular studies. J Clin Endocrinol Metab 1994; 78: 790-794.
16. Finlay CA, Hinds PW, Tan TH, Eliyahu D, Oren M, Levine AJ. Activating mutations for transformation by p53 produce a gene product that forms an hsc70-p53 complex with an altered half-life. Mol Cell Biol 1988; 8: 531-539.
17. Van Slooten H, Schaberg A, Smeenk D, Moolenaar AJ. Morphologic characteristics of benign and malignant adrenocortical tumors. Cancer 1985; 55: 766-773.
18. Haak HR, Cornelisse CJ, Hermans J, Cobben L, Fleuren GJ. Nuclear DNA content and morphological characteristics in the prognosis of adrenocortical carcinoma. Br J Cancer 1993; 68: 151-155.
19. Mathews DE, Farewell V. Using and Understanding Medical Statistics. Basel: Karger, 1985.
20. Haak HR, Hermans J, Van de Velde CJH, et al. Optimal treatment of adrenocortical carcinoma with mitotane: results in a consecutive series of 96 patients. Br J Cancer 1994; 69: 947-951.
21. Fujimo M, Dosaka-Akita H, Kato M, Kinoshita I, Akie K, Kawakami Y. Simultaneous use of the PCR-SSCP method and immunohistochemistry for increasing the detection efficacy of p53 abnormalities in human lung cancer. Am J Clin Pathol 1995; 104: 319-324.
22. Jeffers MD, Farquharson MA, Richmond JA, McNicol AM. p53 immuno- reactivity and mutation of the p53 gene in smooth muscle tumours of the uterine corpus. J Pathol 1995; 177: 65-70.
23. Sidransky D, Mikkelsen T, Schwecheimer K, Rosenblum ML, Cavanee W, Vogelstein B. Clonal expansion of p53 mutant cells is associated with brain tumour progression. Nature 1992; 355: 846-847.
24. McManus DT, Yap EPH, Maxwell P, Russell SEH, Toner PG, McGee JO’D. p53 expression mutation, and allelic deletion in ovarian cancer. J Pathol 1994; 174: 159-168.
25. Caleffi M, Teague MW, Jensen RA, et al. p53 gene mutations and steroid receptor status in breast cancer. Cancer 1994; 73: 2147-2156.
26. Dowell SP, Hall PA. The p53 tumour suppressor gene and tumour prognosis: is there a relationship? J Pathol 1995; 177: 221-224.