Identification of Id4 as a regulator of BRCA1 expression by using a ribozyme-library-based inverse genomics approach

Abstract

Expression of the breast and ovarian cancer susceptibility gene BRCA1 is down-regulated in sporadic breast and ovarian cancer cases. Therefore, the identification of genes involved in the regulation of BRCA1 expression might lead to new insights into the pathogenesis and treatment of these tumors. In the present study, an "inverse genomics" approach based on a randomized ribozyme gene library was applied to identify cellular genes regulating BRCA1 expression. A ribozyme gene library with randomized target recognition sequences was introduced into human ovarian cancer-derived cells stably expressing a selectable marker [enhanced green fluorescence protein (EGFP)] under the control of the BRCA1 promoter. Cells in which BRCA1 expression was upregulated by particular ribozymes were selected through their concomitant increase in EGFP expression. The cellular target gene of one ribozyme was identified to be the dominant negative transcriptional regulator Id4. Modulation of Id4 expression resulted in inversely regulated expression of BRCA1. In addition, increase in Id4 expression was associated with the ability of cells to exhibit anchorage-independent growth, demonstrating the biological relevance of this gene. Our data suggest that Id4 is a crucial gene regulating BRCA1 expression and might therefore be important for the BRCA1 regulatory pathway involved in the pathogenesis of sporadic breast and ovarian cancer.

Figure 1

The hairpin ribozyme library and gene vector. (A) Sequence and secondary structure of the hairpin ribozyme (capital letters), bound to its target RNA (lowercase letters). Both target-binding arms (helices 1 & 2) are randomized, generating a library with as many as 4¹² different catalytic molecules (ribozymes). (B) Schematic representation of the retroviral vector following insertion of the ribozyme library DNA template. Ribozyme expression is driven by the tRNA^val promoter. (C) Flow cytometry for EGFP of SK-BR-3, PA-1, and T47-D cells. Numbers in parentheses indicate the relative expression levels of endogenous BRCA1. Analysis of untransfected control cells (parental cells) and dilution clones from cells, transfected with pBR-EGFP (BR-EGFP) or with pCMV-EGFP (CMV-EGFP).

Figure 2

Selection of ribozymes that up-regulate the BRCA1 promoter. (A) FACS-selection of P8EGBR3 cells (untransduced, control vector, or ribozyme-library-transduced) for cells with an increase in EGFP expression. Rz library-transduced cells show a peak on sort #4 (*) and a shift of the majority of the population on sort #5 (+), whereas control vector or untransduced cells remain unchanged throughout the experiment. MFI, mean fluorescence intensity of cell population, given in percent of control vector transduced cells. (B) Quantitative real-time RT-PCR analysis of EGFP and endogenous BRCA1 RNA normalized to GAPDH RNA after sort #5 in cells transduced with control vector or ribozyme library. Numbers are given as relative RNA levels in percent compared with control vector transduced cells (mean ± SEM).

Figure 3

BRCA1 expression and morphology of selected cells expressing Rst-RH5. Cells transduced with an enriched ribozyme library were subjected to several rounds of selection for high EGFP expression and ribozyme rescue resulting in a population expressing Rst-RH5 as the sole predominant ribozyme. These cells were examined for BRCA1 expression and cell morphology. (A) Quantitative real-time RT-PCR analysis of EGFP and endogenous BRCA1 RNA normalized to GAPDH in control vector-transduced or Rst-RH5 expressing cells. Numbers are given as relative RNA levels in percent compared with control vector transduced cells (mean ± SEM). (B) Morphology in monolayer culture of Rst-RH5 expressing cells compared with control vector transduced cells.

Figure 4

Identification and validation of Id4 as a negative regulator of the BRCA1 promoter. (A) Schematic illustration of the predicted binding of Rst-RH5 to Id4 mRNA (numbering of Id4 nucleotides according to GenBank accession no. NM 001546). (B) Northern analysis for Id4 RNA in ribozyme-library- (lane 2) or control vector (lane 1)-transduced P8EGBR3 cells after sort #5 (compare with Fig. 2). Different messages result from alternative polyadenylation. The major message is indicated (arrow). (C) Effects of Rst-RH5 and two validation ribozymes (VRz1 or VRz2) on endogenous BRCA1 and Id4 expression measured by quantitative real-time RT-PCR. Expression of BRCA1 or Id4 mRNA was normalized to GAPDH mRNA. Numbers are given as relative RNA levels in percent compared with the control ribozyme (mean ± SEM). (D) The Hygromycin B resistance assay. Quantitative analysis of cell survival and of endogenous BRCA1 expression (measured by quantitative real-time RT-PCR and normalized to GAPDH message) in control vector-, Id4-sense-, or Id4-antisense-transduced cells after phenotypic selection of Hygromycin B reporter cells with Hygromycin B. Numbers are given in percent compared with control vector (mean ± SEM). (E) Anchorage-independent growth of P8EGBR3 cells transduced with control vector, Id4-sense, or Id4-antisense expression vectors.

Figure 5

Hormone stimulation of MCF-7 cells. Cells were grown in phenol-red free media without serum for 36 h before selective stimulation with 10 nM β-estradiol (E2) or 10 nM progesterone (Pg), or no stimulation (controls). Cells grown in media containing 10% FCS were used as controls (with FCS). Following stimulation, cellular RNAs were harvested and analyzed for endogenous BRCA1 and Id4 expression levels by using quantitative real-time RT-PCR. Numbers are given as percent (mean ± SEM) in comparison to cells grown without serum.