Calcitriol stimulates gene expression of cathelicidin antimicrobial peptide in breast cancer cells with different phenotype

In normal and neoplastic cells, growth-promoting, proangiogenic, cytotoxic and pro-apoptotic effects have all been attributed to cathelicidin antimicrobial peptide (CAMP). Nevertheless, little is known about the factors regulating this peptide expression in breast cancer. Herein we asked if the well-known antineoplastic hormone calcitriol could differentially modulate CAMP gene expression in human breast cancer cells depending on the cell phenotype in terms of efficacy and potency. The established breast cancer cell lines MCF7, BT-474, HCC1806, HCC1937, SUM-229PE and a primary cell culture generated from invasive ductal breast carcinoma were used in this study. Calcitriol regulation of cathelicidin gene expression in vitro and in human breast cancer xenografts was studied by real time PCR. Tumorigenicity was evaluated for each cell line in athymic mice. Estrogen receptor (ER)α + breast cancer cells showed the highest basal CAMP gene expression. When incubated with calcitriol, CAMP gene expression was stimulated in a dose-dependent and cell phenotype-independent manner. Efficacy of calcitriol was lower in ERα + cells when compared to ERα- cells (<10 vs. >70 folds over control, respectively). Conversely, calcitriol lowest potency upon CAMP gene expression was observed in the ERα-/EGFR+ SUM-229PE cell line (EC50 = 70.8 nM), while the highest was in the basal-type/triple-negative cells HCC1806 (EC50 = 2.13 nM) followed by ERα + cells MCF7 and BT-474 (EC50 = 4.42 nM and 14.6 nM, respectively). In vivo, lower basal CAMP gene expression was related to increased tumorigenicity and lack of ERα expression. Xenografted triple-negative breast tumors of calcitriol-treated mice showed increased CAMP gene expression compared to vehicle-treated animals. Independently of the cell phenotype, calcitriol provoked a concentration-dependent stimulation on CAMP gene expression, showing greater potency in the triple negative HCC1806 cell line. Efficacy of calcitriol was lower in ERα + cells when compared to ERα- cells in terms of stimulating CAMP gene expression. Lower basal CAMP and lack of ERα gene expression was related to increased tumorigenicity. Our results suggest that calcitriol anti-cancer therapy is more likely to induce higher levels of CAMP in ERα- breast cancer cells, when compared to ERα + breast cancer cells.


Background
Paradoxical effects have been described for cathelicidin (CAMP) in cancer biology. Some studies have shown cytotoxic, antiproliferative and pro-apoptotic effects [1][2][3][4], whereas others reported growth-promoting, proangiogenic, prometastatic and invasive-inductive effects of CAMP in different malignant-type cells [5][6][7][8]. These effects are the result of tissue-specific signaling pathways triggered by CAMP in an intra-or extra-cellular manner [9] involving several growth factor receptors [10][11][12] and/or toll-like receptors [13]. In fact, CAMP overexpression has been shown to suppress tumorigenesis in colon and gastric cancer but also to promote development and progression of ovarian, lung and breast cancer [9]. Of note, CAMP signalization may activate signaling cascades potentially involved in carcinogenesis, such as those involving mitogen activated kinases, protein kinase C or nuclear factor kappa B. Therefore, overexpression of CAMP is generally associated with tumor promotion activity, in a concentration and/or tissue specific fashion. Particularly in the breast, CAMP is abundantly produced in both normal and malignant conditions, while its maximum expression has been found among high-grade breast tumors [5]. Interestingly, CAMP expression is closely correlated with that of epidermal growth factor receptor 2 (HER2) and with the presence of lymph node metastases in estrogen receptor (ER) + breast tumors, suggesting a prometastatic role for CAMP in breast malignancy [14]. The regulatory factors acting upon CAMP are not yet completely understood. Normally, the expression of CAMP is induced in response to injury or bacterial challenge, resulting in its accumulation in the site of distress. Indeed, inflammatory mediators may induce CAMP in some settings [15]; however, this is not always the case, as seen in tumor necrosis factor-α treated trophoblasts and other cell types, where CAMP was either not regulated or downregulated by inflammatory cytokines [16,17]. In humans, there is evidence that the most robust CAMP inducer is calcitriol, the vitamin D more active metabolite. This hormone, acting through its nuclear receptor (VDR), transcriptionally induces robust expression of CAMP by acting through a vitamin D response element located in its promoter [18]. Calcitriol is well known for its anticancer properties, which are being studied in preclinical and clinical settings. Given the potential pharmacological use of calcitriol for therapeutic purposes in breast cancer patients, herein we thought of importance to investigate the regulatory actions of this hormone upon CAMP gene expression under in vitro and in vivo conditions using different phenotypes of breast cancer cells.

Breast cancer cell cultures
The established human breast cancer cell lines MCF7, BT-474, HCC1806, HCC1937 (ATCC, Manassas, VA), SUM-229PE (Asterand, San Francisco, CA), and a primary cell culture generated from invasive ductal breast carcinoma (IDC) [19], were maintained under standard cell culture conditions. For experiments, cells were incubated in the presence of different calcitriol concentrations (0.1-1000 nM, Sigma-Aldrich, St Louis, MO) or its vehicle (0.1 % ethanol) during 24 h. Afterwards, cells were used for RNA isolation. Characterization of the cells was performed by immunocytochemistry in order to analyze the expression of particular molecular markers.

PCR amplifications
Calcitriol effects upon CAMP gene expression were studied by extracting total RNA from treated cells and resected tumors using Trizol reagent (Life Technologies, CA, USA). The concentration of RNA was estimated spectrophotometrically at 260/280 nm and a constant amount of RNA (2 μg) was reverse transcribed using a commercial assay (Roche Applied Science, IN, USA). Gene expression of the housekeeping gene β-actin (ACTB) was used as internal control. Primers sequences were as follows: CAMP [GenBank:NM_004345.3]: forward: tcg gat gct aac ctc tac cg, reverse: gtc tgg gtc ccc atc cat and ACTB [GenBank:NM_001101.3]: forward: cca aac cgc gag aag atg a, reverse: cca gag gcg tac agg gat ag. Corresponding probe numbers from the universal probe library (Roche) were: 85 and 64 for CAMP and ACTB, respectively. Real time PCR amplifications were carried on a LightCycler® 480 Instrument (Roche), according to the following protocol: activation of Taq DNA polymerase and DNA denaturation at 95°C for 10 min, proceeded by 45 amplification cycles of 10 s at 95°C, 30 s at 60°C, and 1 s at 72°C.
The calcitriol concentration producing 50 % CAMP gene expression stimulation (EC 50 ) was calculated by non-linear regression analysis using sigmoidal fitting with a sigmoidal dose-response curve by means of the scientific graphing software Origin (OriginLab Corporation, Northampton, MA, USA).

Induction of tumors in athymic mice
Athymic female BALB/c homozygous, inbred Crl:NU (NCr)-Foxn1nu nude mice (~6 weeks of age) were kept in ventilated cages with bedding of aspen wood-shavings, controlled temperature, humidity and 12:12 light:dark periods. Sterile water and feed (standard PMI 5053 feed) were given ad libitum. Endpoints compatible with the scientific objectives of this work were cautiously observed preserving strict animal welfare standards. To evaluate the physical status of each mouse, a scoring method was used which included the following categories: 1) dehydration/ loss of appetite, 2) body weight, 3) natural behavior, 4) provoked behavior and 5) inflammation/ulceration in injection site. A value of 1-2 was assigned to the first category while 1-3 was used for the last 4 categories. A total score of 14 indicated wellbeing, while lower scores indicated progressive health deterioration. A score < 9 was an automatic endpoint. Tumorigenicity was evaluated for each cell line used in this study by subcutaneous injection of 2.0 × 10 6 cells in 0.1 mL of sterile saline solution into the upper part of the posterior limb of each mouse.

Therapeutic protocol
When the tumors reached a palpable mass (~3 mm), mice were separated in two groups: control and calcitrioltreated (calcitriol Geldex, GELpharma, México, 12.5 μg/kg of body weight i.p. in 100 μL once a week during 3 weeks). IDC and HCC1806 cells were used to xenograft mice (total mice = 22; 11 for each cell line). Body weights and tumor sizes were measured thrice weekly throughout the experiment. Tumor volume was calculated using the standard formula (length x width 2 )/2, where length is the largest dimension and width the smallest dimension perpendicular to the length. Tumors were measured with a caliper always by the same person. Relative tumor volume was calculated for each tumor by dividing the tumor volume on day 21 by that on day 0 (which corresponded to the tumor volume in the first day of treatment, and was set to one). After sacrifice, tumors were excised and processed for RNA extraction.

Statistical analysis
Statistical differences for in vitro dose-response assays were determined by one-way ANOVA followed by appropriate post-hoc tests using a specialized software package (SigmaStat, Jandel Scientific). For in vivo comparisons between control and calcitriol-treated groups Student's t-test was used. Differences were considered statistically significant at P < 0.05.

Characterization of the cells used in this study
Expression of VDR, ERα, HER2 and EGFR for each cell line is depicted in Table 1.
Also, the functionality of the VDR was corroborated by the calcitriol-dependent induction of CYP24A1 gene expression ( Table 2).
Calcitriol induces CAMP gene expression in cultured breast cancer cells of different phenotype, but more strongly in ERα-cells Differential basal CAMP gene expression was observed depending on the cell line. In particular, ERα + breast cancer cells showed the highest basal CAMP gene expression, while the lowest was obtained in ERα-cells ( Fig. 1). On the other hand, in all cell lines tested, a calcitriol dose-dependent stimulation of CAMP gene expression was observed (Fig. 2). In particular, in the ERα-/EGFR+ cell line SUM-229PE, calcitriol, at the highest concentration tested, showed the greatest efficacy in terms of stimulating CAMP gene expression (>200 folds over control). Meanwhile, in the basal-type/ triple-negative cell lines HCC1937 and HCC1806 calcitriol increased CAMP gene expression by approximately 70-100 folds over the control. In contrast, ERα + cells MCF7 and BT-474 responded more moderately to calcitriol (<10 folds over control, Fig. 2). Based on the EC 50 values, the potency/sensibility of calcitriol upon CAMP gene expression was: HCC1806 > MCF7 > BT-474 > HCC1937 > IDC > SUM-229PE (Table 3).

In a xenograft model of breast cancer, calcitriol induced CAMP gene expression
We first tested tumorigenicity of all cancer cell lines in a murine model. Under the conditions of this study, only HCC1806 and IDC readily formed tumors. We observed that lower basal CAMP gene expression and lack of ERα positivity were cell features related to increased tumorigenicity. Therefore, CAMP gene expression in vivo studies were carried out in HCC1806 and IDC tumors. Considering the in vitro calculated potency of calcitriol upon CAMP gene expression, which was higher in HCC1806 compared to IDC (2.13 nM vs. 17.1 nM, respectively), and the greatest efficacy of calcitriol to induce its canonic  (Fig. 3). In both xenotransplanted mouse models calcitriol reduced, although not significantly, the relative tumor volume (Fig. 4).

Serum levels of total calcium and body weight in calcitriol-treated xenotransplanted mice
Serum samples from each experimental group were pooled. As expected, serum total calcium was higher in calcitriol treated mice compared to controls (10.6 vs. 9.9 mg/dL); however, no signs of hypercalcemia were detected (e.g. dehydration, weight loss). Final body weights were not significantly different among the treated and control groups.

Discussion
Cathelicidin is produced by the human mammary gland and is found in human milk exerting antimicrobial activity [20], which highlights the important physiological role of this antimicrobial peptide in the newborn innate immune defense during lactation. Nevertheless, in a pathological scenario of the breast, cathelicidin effects are controversial since it has been implicated in tumor-suppressive activities, but also in promoting tumor growth and vascularization [5,8,10,21]. Given that calcitriol, a recognized antineoplastic hormone, is the most known robust inducer of CAMP expression in humans, herein we studied the regulatory actions of this compound upon CAMP expression in vitro and in vivo in different types of breast cancer cells.  Depicted cell lines were incubated with 10 nM calcitriol during 24 h and afterwards RNA was extracted and qPCR performed. Control was normalized to one, results are expressed as fold induction over control Fig. 1 Basal CAMP gene expression. Basal CAMP gene expression was evaluated in several breast cancer cell lines with different phenotype. Data are depicted as the mean ± SD. N = 3. Results were normalized against ACTB mRNA expression Probably, the time and dose used herein to treat mice with calcitriol might account on the lack of statistical significance found upon tumor volume in this study.
Regarding CAMP biological actions in tumoral cells, it is noteworthy to mention that CAMP expression has been closely correlated to HER2 [14] and has been shown to transactivate EGFR [24,25], which may explain why cancer cells exposed to the CAMP active peptide LL-37 show increased cell proliferation and invasion [8]. Similarly, in animal models CAMP treatment promoted tumor growth and metastasis [14]. Nevertheless, the fact that the highest CAMP levels have been found in breast tumors of greater malignancy grade [5], together with the observation of increased CAMP expression in blood of breast cancer patients compared to healthy women [26], strongly encourages to explore the biological actions of CAMP in breast tumor progression. In this regard, binding of LL-37 to type I insulin-like growth factor receptor in different types of breast cancer cells has resulted in intra-cellular signaling activation and increased migratory and invasive potential of malignant cells [10]. While additional studies of CAMP effects on breast cancer biology await to be undertaken, the calcitriol-mediated induction of CAMP   [27][28][29][30][31]. Of particular interest is the observation that in early premalignant and fully malignant breast cells a similar stimulatory effect of a calcitriol analogue upon CAMP gene expression has been observed [32]. However, to our knowledge this is the first study to show a differential CAMP gene expression profile after calcitriol stimulation depending of the cell type phenotype. Given that calcitriol is an antineoplastic drug under intense investigation for therapeutic purposes, the results in this study may help to translate calcitriol therapy into the clinic. Since CAMP may regulate tumorigenesis and/or cell proliferation, more studies are needed in order to clarify exogenous calcitrioldependent CAMP synthesis and biological actions in breast tumors with different phenotype.

Conclusions
Breast cancer cells showed differential basal CAMP gene expression depending on the cell phenotype: ERα + breast cancer cells showed the highest while ERα-the lowest. Availability of data and materials All data generated or analyzed during this study are included in this published article, with the exception of mice final body weights, which are available from the corresponding author on reasonable request.
Authors' contributions JGQ and LD were involved in the conception, design and coordination of the study, data analysis/interpretation and experimental procedures. RGB participated in experimental procedures, analysis and interpretation of data and critically revised the manuscript. NSM was involved in immunocharacterization of cells and critically revised the content of the manuscript. EA contributed in the interpretation of data and critically revised the manuscript for important intellectual content. FL participated in the interpretation of data, made substantive intellectual contribution to the study and helped to draft the manuscript. LD and JGQ drafted the manuscript. All authors read and approved the final manuscript.

Competing interests
The authors declare that they have no competing interests.

Consent for publication
Not applicable.

Ethics approval
All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. Studies involving mice were performed according to the rules and regulations of the Official Mexican Norm 062-ZOO-1999. The study was approved by the Institutional Committee for the care and use of laboratory animals (protocol number BRE-1291-14/17-1, CINVA 1291) of