Abstract: Basal-like breast cancer is very invasive and highly metastatic with poor prognosis. The molecular mechanisms underlying its tumorigenesis and tumor progression remain unclear, and ideal animal models are lacking. The present study aims to identify the features of spontaneous breast cancer in TA2 mice, to determine whether the mouse model is useful and relevant for human basal-like breast cancer, and to investigate the relationship among pregnancy, mouse mammary tumor virus ( MMTV ), and TA2 breast cancer. Methods: Spontaneous breast cancers were collected from 84 female TA2 mice. The metastasis, morphology, and immunophenotype of the TA2 mice bearing breast cancer tumors were determined using hematoxylin, eosin staining, and immunohistochemistry, respectively. The similarities between the basal-like breast cancers of the animals and that of the human patients were compared. Electrochemiluminescence immunoassay was used to detect the serum levels of estradiol  ( E2 ) and progesterone in the TA2 mice ( n = 10 ), the mice at d15 during the gestation period ( n = 6 ), and the normal mice ( n = 6 ). The MMTV in the TA2 breast cancer cells were checked under electron microscopy ( EM ). The MMTV LTR expression in the mammary glands of the normal and the pregnant mice and the TA2 mice were determined using reverse transcriptase polymerase chain reaction. Results: The spontaneous breast cancer tissues in the TA2 mice were similar to human basal-like breast cancer in terms of biology, morphology, and phenotype. The mean age of the TA2 mice at tumorigenesis was ( 329.81 ± 95.32 ) d, and the mean frequency of delivery was ( 2.70 ± 1.8 ), with rapid tumor growth within the gestation period. Metastasis in the visceral organs frequently occurred in TA2 tumors. However, nodal metastasis was not found in the current research. Pulmonary metastasis was seen in 80% ( 64/80 ) of the tumors. The TA2 cancers were homologous in histomorphology. The cancer nest consisted of roundlet cells with rare cytoplasm, surrounded by well-developed cell matrices. There were undifferentiated or poorly differentiated cancer cells at the center of the solid cancer nest. Caryomitotic figures were seen, with the necrosis at their centers. In the cancer cells at the edge of the cancer nest, aciniform, chrysanthemum-like, or papillar cell differentiation occurred occasionally. Around the aciniform nest, a monostratified or multilamellar myoepithelium and a pyknotic matrix were observed. Accumulated pink excretions were seen in the uniformly differentiated cancer nest. Immunohistochemistry revealed the absence of ER, PR, and HER2 expression, but the overexpression of cyclin D1, PCNA, p53, and CK5/6 were observed in the TA2 cancer. The EM observation showed MMTV viral particles in the tumor cells. The serum E2 and progesterone concentrations in the TA2 mice and the intratumoral expression of MMTV LTR mRNA all increased. The change was similar to that in the pregnant mice. Conclusion: The histomorphology, immunophenotype, and biological features of breast cancer in TA2 mice are similar to those in human basal-like breast cancer, which reflects the full process of tumorigenesis and progression of basal-like cancer. Therefore, the mouse model is feasible for use in the research of basal-like breast cancer. MMTV activation induced by the increases in E2 and progesterone levels plays a key role in the carcinogenesis of basal-like breast cancer.