Tp. Chen et al., TECHNICAL APPROACH FOR THE STUDY OF THE GENETIC EVOLUTION OF BREAST-CANCER FROM PARAFFIN-EMBEDDED TISSUE-SECTIONS, Breast cancer research and treatment, 39(2), 1996, pp. 177-185
We have optimized a technique that allows the study of numerous chromo
somal loci (n = 20-50) from single paraffin-embedded tissue sections b
y microsatellite length polymorphism analysis. DNA samples from normal
and breast cancerous tissue can be obtained from the same section by
means of microdissection. This technique was further improved by subje
cting DNA to several cycles of amplification with a degenerate (univer
sal) primer and then with specific microsatellite primers. This amplif
ied DNA was also used to screen for mutations in the p53 gene by means
of PCR-SSCP, In addition adjacent tissue sections were used to assess
specific chromosome copy number by interphase cytogenetic analyses (c
hromosome in situ hybridization) and to analyze expression of specific
genes such as p53 and ERBB2. As an example of the use of our approach
we performed a detailed chromosome 17 allelotypic analysis in 22 brea
st tumors (5 ductal carcinomas in situ, 13 invasive ductal carcinomas,
and 4 invasive lobular carcinomas), We detected mutations in the p53
gene by PCR-SSCP in 36% of the samples. Samples with significant level
s of p53 protein accumulation detected by immunohistochemistry were al
so positive for mobility shifts in the SSCP analysis. We observed that
chromosome 17 allelic losses and imbalance occurred at as early a sta
ge as ductal carcinoma in situ (DCIS). Although in some cases we obser
ved allelic losses or imbalance affecting the 17p13 region, close to t
he p53 locus, several of the tumors showed dissociation between such l
oss or imbalance and p53 mutation. Lobular carcinomas were predominant
ly disomic for chromosome 17 in contrast with ductal tumors, which oft
en showed polysomy for chromosome 17. This comprehensive approach corr
elating the tumor subtype, its allelotype, with specific chromosome co
py number and specific gene mutations and expression in preinvasive or
early invasive breast cancer lesions will potentially provide informa
tion of relevance for a better understanding of the multistep mechanis
ms of breast carcinogenesis.