Studying tissue-based gene expression in different cell populations often requires immunohistochemistry-guided microdissection. However, mRNA degradation occurs during long staining procedures. We combined a novel rapid immunoperoxidase technique with laser capture microdissection (LCM) and real-time quantitative RT-PCR to compare p27 mRNA expression in prostatic basal/secretory cells. Eight frozen prostate sections were immunostained with antibody 34βE12 (high–molecular-weight keratin). Secretory and basal cells were separately collected by LCM. p27 transcripts from each cell group were quantitated by real-time RT-PCR, with GAPDH as standard. Immunostaining took 22 minutes, with RNA extraction from ∼40 dissected cells from each compartment initiated within 40 minutes. Qualitative RT-PCR gave a product of the expected size from each sample. Quantitative RT-PCR gave basal/secretory p27/GAPDH ratios of 0.99–16.24 (mean 5.53 ± 0.643). Immunostaining for keratin 34βE12 can be done on frozen sections in ∼20 minutes, and mRNA from pure cell populations can be quantitated by RT-PCR. We used this technique to show that p27 transcript levels are greater in basal than in secretory prostate cells, suggesting, when combined with prior studies, that regulation of p27 occurs at the protein level in normal cells. This technique may have wide applicability to studies of gene expression in distinct cell populations in heterogeneous tissues.
From the Department of Pathology, Brigham & Women's Hospital (N.L., S.S., M.L.), Boston, MA; and the Department of Adult Oncology, Dana-Farber Cancer Institute (D.W., S.S., M.L.), Boston, MA
Supported by grants from the NIH (R01-CA81755), a Novartis and Barr Investigator grant, a CaP CURE award to M.L., a Hershey Prostate Cancer/Survivor Walk Award, a CaPCURE Award, and a Department of Defense Grant (DAMD17-01-1-0051) to S.S.
Address correspondences and reprint requests Dr. Massimo Loda, M.D., Department of Adult Oncology, Dana Building 740B, Dana-Farber Cancer Institute, 44 Binney Street, Boston, MA 02115 (e-mail: Massimo_Loda@dfci.harvard.edu).