The role of tumor-initiating cells (TICs) in carcinogenesis and oncological treatment has become a major focus of cancer research. The cancer stem cell theory posits that only a small, specific population of cells within a tumor has the capacity to recapitulate all of the heterogeneous cell types in a metastatic or locally recurrent tumor. These cells are regulated by developmental pathways critical to normal stem cell survival that are attractive targets to potentially eliminate this population. Using approaches developed in the hematopoietic malignancies, Al-Hajj and colleagues were the first to purify a subpopulation of human breast cancer cells that display tumor initiating properties. As few as 100 of these cells, characterized by the expression of cell surface markers Lin⁻CD44⁺CD24^−/low, re-generated tumors when orthotopically implanted in immunocompromised mice, and these cells were then considered TICs. In contrast, 20,000 cells with alternate non-TIC phenotypes failed to form tumors [1]. Dontu and colleagues further refined this phenotype demonstrating that as few as 20 cells from the population positive for both Lin⁻CD44⁺CD24^−/low phenotype and aldehyde dehydrogenase activity can generate a tumor in immunocompromized mice [2]. Furthermore, Dontu and colleagues adapted the stem cell enriched sphere culture system described in the central nervous system literature to the normal mammary gland and was able to demonstrate in vitro self-renewal (the ability to go through numerous cycles of cell division while maintaining the undifferentiated state) and maintenance of multi-lineage differentiation potential in mammosphere culture of normal human breast epithelial cells [3]. In this method, mammary gland cells are cultured in suspension in serum-free, growth factor enriched media. We and others have demonstrated that this approach can be adapted for breast cancer cell lines [4-6]. This approach provides an in vitro model to study TICs and develop therapies targeting TICs.

Tetrandrine is a bisbenzylisoquinoline alkaloid found in Stephania tetrandra, a Chinese medicine which has been used to treat hypertension and inflammation for thousands of years. Tetrandrine has been shown to have multiple pharmacological activities including immunosupression [7], anti-hypertensive [8] and anti-tumor activity [9]. At the molecular level, tetrandrine was found to block calcium channels [10] and multiple drug resistant proteins (MDR) [11]. Recently, it was reported tetrandrine inhibited β-catenin signaling in a colon cancer xenograft model [12],which is of interest given that β-catenin plays an important role in breast cancer TICs [13,14].

Herein, we screened tetrandrine for efficacy against in vitro surrogates of breast cancer TICs. We demonstrate that tetrandrine can inhibit the growth of breast cancer cells. We also found that tetrandrine can inhibit primary mammosphere formation, a surrogate of breast cancer TICs in vitro in IC₅₀ 10 times lower than the IC₅₀ of cell proliferation inhibition. Tetrandrine inhibits the mammosphere formation of cells isolated from fresh patient sample with similar IC₅₀. Moreover, we found that tetrandrine can decrease ALDH positive population, which is a putative marker for breast cancer TICs. In summary, tetrandrine can effectively targets in vitro breast cancer TIC surrogates.

More and more evidence suggests that TICs sustain cancer growth and cause tumor metastasis and recurrence after therapies. From a clinical point of view, therapies targeting TICs may have the potential to cure the cancer because they target the root of tumor. However, more and more evidence shows that TICs are more resistant than non-TICs to multiple cancer therapies including chemotherapy and radiation [6,17-20]. Herein we demonstrate that Tetrandrine, a well-tolerated oral agent found in a common Chinese medical compound is active against in vitro surrogates for TICs in cell lines derived from the most aggressive breast cancers and cells purified from fresh patient samples.

In several solid tumor types, radiation has been shown to enrich for TICs by selectively targeting the non-TICs [6,17,18]. In clonogenic assays comparing first and second generation tumorspheres from MCF-7 breast cancer cells, the TIC enriched second generation spheres are more resistant to radiation [19] and in our laboratory, similar to the work reported by Phillips et al. [6], we find increased radiation resistance in TIC enriched tumorspheres compared to more differentiated colonies formed from monolayer culture in all lines tested independent of the molecular phenotype (MCF-7, SUM-190, SUM-149, MDA-MB-231 and short term primary human breast cancer cells) [6,19]. Numerous studies across other tumor sites have also reported intrinsic radiation resistance in TICs [20]. TICs are also resistant to chemotherapies. Li et al found that chemotherapy can increase the percentage of a cell population with CD44^highCD24^low cell surface mark, the putative mark for breast cancer TICs in pre and post treatment patient samples [21,22]. Tanei et al. reported that ALDH positivity better distinguishes chemo-resistance compared to CD44^highCD24^low [23].

Much effort has been made to identify compounds that specifically target TICs. Overall, few compounds have been identified to target breast cancer TICs. In one paper, salinomycin, a potassium channel blocker used for treatment of Coccidia parasites in chicken, was identified to target breast cancer stem cells [24]. In another paper, dietary supplement sulforaphane was shown to target breast cancer TICs through inhibition of β-catenin signaling [25]. Because β-catenin signaling has been shown to play an important role in breast cancer TICs self-renewal [13,14] and tetrandrine Tetrandrine, a component of the Chinese traditional medicine compound Stephania tetrandra, was found to inhibit β-catenin signaling [12], we used in vitro systems to test if tetrandrine can target breast cancer TICs.

Tetrandrine was shown to cause human leukemia cells U937 to go through apoptosis in a mechanism unrelated to calcium channel blockage [26]. Tetrandrine was also reported to target colon cancer. In one case, tetrandrine caused early G-1 arrest in colon cancer cells by induction of G1-S-specific cyclin-dependent kinases [27]. In another case, tetrandrine inhibited the proliferation of colon cancer cells in vitro and caused tumor growth delay of colon cancer cell xenografts in vivo [12]. Other cancers reported to be sensitive to tetrandrine including liver cancer [28,29] and glioma [30,31]. Tetrandrine has been reported to sensitize various cancers, including neuroblastoma [32] and glioblastoma [33], to radiation therapy. Moreover, because tetrandrine is a MDR protein inhibitor [11], it has been combined with chemotherapeutic agents such as doxorubicin or vincristine to increase the efficacy [34]. MDR1 is the first member of MDR family which is non-specific ion pumps on membranes used by cancer cells to efflux chemotherapeutic compounds and cause chemotherapy resistance. Beyond its original function, over-expression of MDR1 was found to increase hematopoetic stem cells in vitro and in vivo [35]. Interestingly, salinomycin, the most efficacious drug found to target breast cancer TICs in one chemical library screen [24], was also a MDR inhibitor [36].

Several methods have been developed to isolate and characterize breast cancer TICs in vitro. One is mammosphere formation described by Dontu et al. [3] and adapted for breast cancer cell lines. Another method is to use cell surface marker expression to distinguish breast cancer TICs from differentiated cells based on tumor outgrowth after transplantation into the cleared mammary fatpads of immunocompromised mice. Both Lin⁻CD24^lowCD44^high and ALDH positive have been explored as markers for human breast cancers. Herein, we chose ALDH positive as mark for breast cancer TICs and compared the effect of tetrandrine on proliferation versus TIC surrogates, mammosphere forming efficiency and ALDH activity. We found that tetrandrine inhibits proliferation and TIC surrogates and that the inhibition of breast cancer TICs occurs at an IC₅₀ about 10 times lower than the dose needed to inhibit the cell proliferation, suggesting specificity of tetrandrine against the breast cancer TICs. Of course the gold standard for TIC studies remains in vivo limiting dilution assays and our data are limited by the caveats inherent in all in vitro studies. Nevertheless, in vitro screening assays are important in selecting agents for further testing, and prior work suggests correlation between in vitro TIC surrogates and in vivo tumor initiation [15,24]. In addition, activity against primary human breast cancer mammospheres is similar to results obtained from mammosphere assays using immortalized cell lines, which highlights the potential clinical relevance.

Like many chemotherapy agents, tetrandrine has numerous biologic effects, and no specific mechanism has been identified which accounts for the anti-TIC activity. Others have demonstrated inhibition of an important breast cancer TIC mediator, β-catenin, however, tetrandrine failed to decrease the expression of β-catenin in breast cancer cells (data not shown). Further functional signaling analysis, using TOP-GFP a fluorescent β-catenin signaling readout [37], is necessary to rigorously examine this question.

In conclusion, we have shown that tetrandrine can target breast cancer TICs in vitro by the mammosphere formation assay and ALDH assay. Further in vivo and mechanistic studies assays are needed to fully determine the significance of this finding. Nevertheless, tetrandrine has been used for treatment of hypertension and inflammation in China for thousands of years. Recently, tetrandrine has been tried in clinic to treat refractory and relapsed acute myelogenous leukemia in China [38] and in studies in the United States, tetrandrine has been tested for toxicity in primates [39]. If further studies remain promising, clinical testing alone or in combination with standard chemotherapy may be feasible for this non-toxic orally available drug.