Identification of conserved gene expression features between murine mammary carcinoma models and human breast tumors


BACKGROUND: Although numerous mouse models of breast carcinomas have been developed, we do not know the extent to which any faithfully represent clinically significant human phenotypes. To address this need, we characterized mammary tumor gene expression profiles from 13 different murine models using DNA microarrays and compared the resulting data to those from human breast tumors.

RESULTS: Unsupervised hierarchical clustering analysis showed that six models (TgWAP-Myc, TgMMTV-Neu, TgMMTV-PyMT, TgWAP-Int3, TgWAP-Tag, and TgC3(1)-Tag) yielded tumors with distinctive and homogeneous expression patterns within each strain. However, in each of four other models (TgWAP-T121, TgMMTV-Wnt1, Brca1Co/Co;TgMMTV-Cre;p53+/- and DMBA-induced), tumors with a variety of histologies and expression profiles developed. In many models, similarities to human breast tumors were recognized, including proliferation and human breast tumor subtype signatures. Significantly, tumors of several models displayed characteristics of human basal-like breast tumors, including two models with induced Brca1 deficiencies. Tumors of other murine models shared features and trended towards significance of gene enrichment with human luminal tumors; however, these murine tumors lacked expression of estrogen receptor (ER) and ER-regulated genes. TgMMTV-Neu tumors did not have a significant gene overlap with the human HER2+/ER- subtype and were more similar to human luminal tumors.

CONCLUSION: Many of the defining characteristics of human subtypes were conserved among the mouse models. Although no single mouse model recapitulated all the expression features of a given human subtype, these shared expression features provide a common framework for an improved integration of murine mammary tumor models with human breast tumors.


Identification of conserved gene expression features between murine mammary carcinoma models and human breast tumors
Herschkowitz JI, Simin K, Weigman VJ, Mikaelian I, Usary J, Hu Z, Rasmussen KE, Jones LP, Assefnia S, Chandrasekharan S, Backlund MG, Yin Y, Khramtsov AI, Bastein R, Quackenbush J, Glazer RI, Brown PH, Green JE, Kopelovich L, Furth PA, Palazzo JP, Olopade OI, Bernard PS, Churchill GA, Van Dyke T, Perou CM.
Genome Biol. 2007;8(5):R76. PMCID: PMC1929138 [ Full Text ]


GEO DataSets: GSE3165, GSE1992, GSE2740, GSE2741

Additional Data Files:

  1. Additional data file 1 (54K, .xls)
    Mouse tumor and normal sample associated data including source, transgene and promoter information

  2. Additional data file 2 (4.0M, PDF )
    Samples are colored according to mouse model from which they were derived, and the genes were selected using a variation filter of three-fold or more on three or more samples

  3. Additional data file 3 (932K, PDF)
    Complete mouse models cluster diagram using the 866 gene murine intrinsic gene list

  4. Additional data file 4 (4.6M, PDF)
    (a) CC matrices generated using the 866 gene mouse intrinsic list, by cluster numbers K = 2 through K = 15. (b) Empirical cumulative distribution (CDF) plot corresponding to the consensus matrices in the range K = 2 to 15. (c) CC directly compared to the hierarchical clustering-based results. The dendrogram from?Figure1 (using the intrinsic gene set) is shown and immediately below is a colored matrix showing sample assignments based upon the various number of K clusters from the CC. By comparison, the analysis performed on the mouse dataset using all genes (bottom matrix) is presented

  5. Additional data file 5 (5.4M, PDF)
    This unsupervised cluster analysis is based upon the orthologous gene overlap between the human and mouse microarrays, and then we selected for the subset of genes that varied three-fold or more on three or more arrays.

  6. Additional data file 6 (1.3M, PDF)
    This analysis was used to determine a human samples subtype (basal-like, luminal, HER2+/ER-, and so on), which was then used the various SAM and GSEA analyses. Samples are colored according to their subtype: red = basal-like, blue = luminal, pink = HER2+/ER-, yellow = claudin-low and green = normal breast-like.

  7. Additional data file 7 (6.8M, TIFF)
    Histological characterization of six different human 'claudin-low' tumors using hematoxylin and eosin sections.

  8. Additional data file 8 (49K, DOC)
    GSEA of murine pathway models versus five human subtypes.

  9. Additional data file 9 (51K, DOC)
    GSEA of ten murine classes versus clinical ER status and HER2 status in ER negative patients.

  10. Additional data file 10 (50K, DOC)
    GSEA of murine pathway models versus clinical ER status and HER2 status in ER negative patients.