Metastasis of breast carcinoma cells from the primary tumor to secondary organ sites such as lung, liver, brain, and bone is the leading cause of mortality in patients with breast cancer [1
]. Bone is the most frequent site of metastasis in breast cancer patients, observed in approximately 65 to 75% of patients with metastatic breast cancer [2
]. Bone metastases often cause significant pain and morbidity in these patients due to osteolysis and bone resorption, and the median survival time after detection these metastases is approximately two years [3
Researchers studying breast tumorigenesis and metastatic progression in vivo
utilize several types of mouse models including transgenic, xenograft, and syngeneic mouse models. Transgenic mouse models that generate spontaneous mammary tumors have been developed using promoters such as the mouse mammary tumor virus (MMTV) promoter to drive oncogenes, including polyoma middle T antigen (MMTV-PyMT) and ErbB2/Neu (MMTV-Neu) (for review see [5
]). Transgenic mice lacking the p53 tumor suppressor gene (p53 −/−
), which is mutated in 40-50% of human breast cancers, have also been utilized extensively in cancer studies but do not reproducibly form mammary tumors [6
]. A major drawback to these transgenic models, along with the commonly used C3(1)-SV40 T-antigen transgenic mice that also develop mammary tumors independent of steroid supplementation, is that bone metastases cannot be detected (for review see [7
]). Given that these tumors occur spontaneously via transformation of the host’s normal cells that do not have specific, imageable cellular tags, it is very difficult to track these cells in vivo
during tumor progression and metastasis.
In the xenograft mouse model, human breast cancer cells are injected into immunocompromised, athymic mice. This model is useful because it allows breast cancer cells of human origin to be studied. However, when studying the metastatic cascade in these models, in which the host organism and the implanted tumor cells are not of the same species, important tumor and host stromal interactions may be disrupted due to interspecies signaling incompatibilities [8
]. Additionally, due to the compromised immune system of the athymic mouse, immune-mediated tumor and host stromal interactions important for metastasis may be lost in this model (for review see [9
Syngeneic models in which the tumor cells are placed into the same species they originated from, and therefore the tumor microenvironment is of the same species, are able to overcome limitations of both xenograft and transgenic models in studying metastasis. Commonly used syngeneic mouse models of breast cancer utilize cells isolated from mammary tumors that occurred spontaneously in wild-type Balb/c mice. These cells can then be injected orthtopically into Balb/c mice and will reproducibly grow a mammary tumor. Currently, there is a series of mouse mammary cancer cell lines derived from the spontaneous breast tumors of Balb/c mice including non-metastatic 67NR cells, 66c14 cells that metastasize to lung, and 4 T1 cells that metastasize to lung and liver [10
]. 4 T1.2 cells are a sub-clone of the original osteolytic 4 T1 mammary tumor cell line that have been selected for their increased propensity to metastasize to bone [10
]. While 66c14 cell metastasis is restricted to the lymph nodes and lung, 4 T1.2 cells closely resemble the metastatic profile in humans with metastasis to the bone, lymph nodes, and lung [10
]. Additionally, the 4 T1.2 orthotopic model results in an increase in hypercalcemia due to osteolysis, an important characteristic that resembles human bone metastasis [12
]. Engineering these mouse cells to be traceable and imageable in a live animal make them even more valuable for studying metastasis to bone in vivo
in a syngeneic mouse model.
Currently, there are several human breast cancer cell lines engineered to express bioluminescence imaging (BLI) tags such as luciferase (luc) that can be used to track the metastases of these cells in vivo
. These cell lines include MDA-MB-231-luc2 and MCF-7-luc2, both available commercially (Caliper Life Sciences, Hopkinton, MA). In general, human mammary cancer cells utilized in xenograft mouse models of breast cancer demonstrate a low propensity to metastasize to bone unless introduced via intra-cardiac injection, which bypasses critical initial steps in the metastatic cascade (for review see [13
]). To date, there are only two mouse mammary cancer cell lines commercially available that stably express the luc gene, 4 T1-luc2 and 4 T1-luc2-GFP cells, available from Caliper Life Sciences (Hopkinton, MA, USA).
In this study we generated by transfection of 66c14 and 4 T1.2 mouse mammary carcinoma cells with the pGL4 vector two new cell lines that stably express the luciferase gene. These two cell lines, referred to as 66c14 luc2 and 4 T1.2 luc3, produce high levels of bioluminescence in vitro and in vivo. Additionally, tumor progression in vivo was consistent with the parental 66c14 and 4 T1.2 cell lines. Importantly, the overall metastatic burden, quantified as the number and photon density of BLI areas, was higher in mice injected orthotopically with the 4 T1.2 luc3 cells as compared to both 66c14 luc2 cells and the commercially available 4 T1 luc2 cells (Caliper Life Sciences, Hopkinton, MA). In vivo and ex vivo imaging of metastasis showed that the 4 T1.2 luc3 cells had a statistically significant higher level of metastases to the spine and lung as compared to the 66c14 luc2 cells, which nonetheless showed detectable levels of spine and lung metastases. Therefore, the newly established 4 T1.2 luc3 and 66c14 luc2 cell lines are reliable tools for monitoring and quantifying breast cancer metastasis to bone and other organs as well as for studying the signaling pathways associated with these processes.