Transcriptomic signature of the chemopreventive
Bexarotene (Rexinoid LGD1069) on mammary gland
from three transgenic mouse mammary cancer models
Background: The rexinoid bexarotene (LGD1069, Targretin) is a highly selective retinoid X receptor (RXR) agonist that inhibits the growth of pre-malignant and malignant breast cells. Bexarotene was shown to suppress the development of breast cancer in transgenic mice models without side effects. The chemopreventive effects of bexarotene are due to transcriptional modulation of cell proliferation, differentiation and apoptosis. Our goal in the present study was to obtain a profile of the genes modulated by bexarotene on mammary gland from three transgenic mouse mammary cancer models in an effort to elucidate its molecular mechanism of action and for the identification of biomarkers of effectiveness.
Methods: Serial analysis of gene expression (SAGE) was employed to profile the transcriptome of p53-null, MMTV-ErbB2, and C3(1)-SV40 mammary cells obtained from mice treated with bexarotene and their corresponding controls.
Results: This resulted in a dataset of approximately 360,000 transcript tags representing over 20,000 mRNAs from a total of 6 different SAGE libraries. Analysis of gene expression changes induced by bexarotene in mammary gland revealed that 89 genes were dysregulated among the three transgenic mouse mammary models. From these, 9 genes were common to the three models studied.
Conclusion: Analysis of the indicated core of transcripts and protein-protein interactions of this commonly modulated genes indicate two functional modules significantly affected by rexinoid bexarotene related to protein biosynthesis and bioenergetics signatures, in addition to the targeting of cancer-causing genes related with cell proliferation, differentiation and apoptosis.
Supplementary data file 1: Differentially expressed genes in mammary gland as the result of systemic bexarotene treatment versus control. SAGE libraries from p53-Null, MMTV-erbB2 and C3(1)/SV40 transgenic mice mammary cancer models (p< 0.05). (384 KB Excel file, zipped.)
Supplementary data file 2: Co-occurring mammary gland deregulated transcripts as the result of systemic bexarotene treament among transgenics mice mammary cancer models (p< 0.05). (64 KB Excel file, zipped.)
Figure 1. Candidate genes and pathways modulated in normal mammary epithelium by rexinoid bexarotene in three different transgenic mice mammary cell models were identified through three-stage process: A. Identification of differentially expressed genes in mammary gland as a result of treatment with bexareotene comparing with vehicle control, in each of the mammary cancer models B. Inter-model comparison for the identification of overlapping gene expression profiles. C. Identification of associated functional modules and pathways affected by bexarotene treatement.
Figure 2. Deregulated transcripts in mammary gland by systemic treatment with bexarotene in the three transgenic mice mammary cancer models. (A) Scatter-plot representation of differentially expressed genes between bexarotene treated mice and vehicle control SAGE libraries (p< 0.05). (B) Gene ontology (GO) classification of bexarotene induced differentially expressed transcripts on mammary gland from the different transgenic models. Relative representation of the deregulated transcripts with specific GO term annotations related to Biological processes or molecular function.
Figure 3. Co-occurring differentially expressed genes among transgenic mouse mammary models. Eighty-nine genes were identified as modulated in more than one transgenic mice model. (A) Heat map of the 89 deregulated transcripts. Color scale at the bottom depicts the approximate fold change in expression for each transcript and library relative to control mammary gland. Negative fold change (transcripts with decreased expression in bexarotene treated animals) is represented in green, and positive fold change (transcripts with overexpression in bexarotene treated mice) is represented in red. (B) Statistical comparison between MMTV-erbB2 vs. p53-null and MMTV-erbB2 vs. C3(1)/SV40 T-antigen transgenic mice models showing a highly significant number of overlapping genes (p< 0.001). The number of overlapping genes between p53-null and C3(1) SV40 models it is not statistical significant (p> 0.05). (C) Gene ontology classification of the 89 transcripts deregulated by bexarotene treatment across models.
Figure 4. Graph of interactions among the common core of genes modulated by rexinoid bexarotne in the studied mammary cancer genetic models generated using database STRING. Genes without known interactions with other genes are listed in the left of the figure. In the network: links between proteins means the various interactions data supporting the network, colored by evidence type.