Caveolin-1 Dependent Endocytosis Enhances the Chemosensitivity of HER-2 Positive Breast Cancer Cells to Trastuzumab Emtansine (T-DM1)

The humanized monoclonal antibody-drug conjugate trastuzumab emtansine (T-DM1, Kadcyla) has been approved by the U.S. FDA to treat human epidermal growth factor receptor 2 (HER-2)-positive metastatic breast cancer. Despite its effectiveness in most patients, some are initially resistant or develop resistance. No biomarker of drug resistance to T-DM1 has been identified. Antibody-drug efficacy is associated with antibody internalization in the cell; therefore, cellular sensitivity of cells to the drug may be linked to cellular vesicle trafficking systems. Caveolin-1 is a 22 KD protein required for caveolae formation and endocytic membrane transport. In this study, the relationship between caveolin-1 expression and the chemosensitivity of HER-2-positive breast cancer cells to T-DM1 was investigated. Samples from 32 human breast cancer biopsy and normal tissue specimens were evaluated immunohistochemically for caveolin-1 expression. Caveolin-1 was shown to be expressed in 68% (22/32) of the breast cancer specimens. In addition, eight (72.7%, 8/11) HER-2 positive breast cancer specimens had a higher caveolin-1 expression than normal tissues. HER-2-positive BT-474 and SKBR-3 breast cancer cells that express low and moderate levels of caveolin-1, respectively, were treated with trastuzumab or its conjugate T-DM1. Cell viability and molecular localizations of caveolin-1, antibody and its conjugate were examined. Confocal microscopy showed that T-DM1 and caveolin-1 colocalized in SKBR-3 cells, which also were five times more sensitive to the conjugate in terms of cell survival than BT-474 cells, although T-DM1 also showed improved drug efficacy in BT-474 cells than trastuzumab treatment. Caveolin-1 expression in these lines was manipulated by transfection of GFP-tagged caveolin-1 or caveolin-1 siRNA. BT-474 cells overexpressing caveolin-1 were more sensitive to T-DM1 treatment than mock-transfected cells, whereas the siRNA-transfected SKBR-3 cells had decreased sensitivity to T-DM1 than mock-transfected SKBR-3 cells. The expression of caveolin-1 could mediate endocytosis and promote the internalization of T-DM1 into HER-2 positive cancer cells. Thus, caveolin-1 protein may be an effective predictor for determining the outcome of T-DM1 treatment in breast cancer patients.


Introduction
Internalization of antibody with targeted growth factor receptor through vesicle trafficking systems has been proposed to be associated with drug sensitivity [9,17]. Cellular vesicle traffic systems function in protein transport, membrane receptor recycling, and protein degradation. Vesicle transport mediated through the endocytic system, including endocytosis and recycling, is controlled primarily by membrane endocytotic vesicles and small GTPases of the Rab family [18]. Clathrin is recognized as a membrane endocytic protein for the transport of molecules from the membrane into the cell [19]. Receptor tyrosine kinases (RTKs) such as epidermal growth factor receptor (EGFR) have been shown to be involved in internalization through the clathrin-dependent endocytosis pathway; however, a clathrin-independent mechanism, such as caveolin-1 for RTKs internalization in breast cancer, has also been demonstrated [20,21]. Caveolin-1 is the 22 KD membrane protein that is required for caveolae structure formation and intracellular caveosome transport [22,23]. The vesicle traffic route for membrane receptors may vary depending on cell types and genetic variation. Previous reports have suggested a negative regulatory role for caveolin-1 in tumor development in a number of human cancers [24,25], and recent studies proved that caveolin-1 inhibits breast cancer cell migration and metastasis [26][27][28][29]. Trastuzumab is internalized via endocytosis, but more than 50% is recycled back to the membrane [30]. In a recent report, trastuzumab was demonstrated to colocalize with caveolin-1 in breast cancer cells [31]. However, whether T-DM1 uses the same internalization mechanism as trastuzumab is still unknown. Poor internalization has been considered an important factor leading to primary or acquired resistance in T-DM1 treatment of metastatic breast cancer [32].
The particular trafficking route used affects the half-life of an ADC inside the cell and further influences the drug efficacy. Therefore, deciphering the mechanism of T-DM1 internalization is critical for its clinical application in advanced precision therapy. This study investigated the mechanism of T-DM1 internalization and the effect of this mechanism on drug efficacy. The results will help to identify vesicle trafficking markers for clinical T-DM1 application.

Patients and ethics statements
The study group consisted of 32 female patients (age range 35-85 years old, median age 48.5 years old) who had undergone resection for localized breast cancer between April 2011 and December 2013 at Cheng-Ching General Hospital, Taiwan. The protocol was reviewed and approved by the Cheng-Ching General Hospital Institutional Review Board (HT110018). Written informed consent was obtained from all patients. Three patients were in stage I, 13 patients were in stage II, and 16 patients were in stage III of breast cancer. Tumor histology grade, specific hormone and HER-2 receptor expressions, tumor size, tumor presence or absence in lymph nodes, and pathological stage were recorded. All assessments in this study were carried out without knowledge of pathological or surgical data.

Immunohistochemistry
The specimens with breast epithelial tissue (tumor site and non-tumor site) of patients with breast cancer were proceeded for immunohistochemistry. Five-μm-thick paraffin sections were incubated with anti-caveolin-1 (1:100) or anti-HER-2 (1:1000) antibodies (Dako, Carpinteria, CA, USA) at room temperature for 1 hr. A histostain-SP diaminobenzidine tetrahydrochloride (DAB) kit (Invitrogen, ThermoFisher, Waltham, MA, USA) was then used to visualize the primary antibody. For HER-2, the strength of the membranous staining was graded on 4-point scale from 0 to 3 and only 3+ was considered positive according to HercepTest guideline from Dako. The scoring used for caveolin-1 by immunohistochemistry was the "I" index [33]; the equation for which is I = 0×f0 + 1×f1 + 2×f2 + 3×f3, where f0-f3 are the fractions of the cells showing a defined level of staining intensity (from 0-3). The numbers 0-3 represent the following: "0" negative, no detectable staining, "1" weak, but still detectable staining, "2" moderate, clearly positive but still weak; and "3" heavy and intense staining (Fig 1). Scores ranged from 0 to 3. A score greater than the mean value of 1.5 was considered as high expression, and a score less than 1.5, as low expression of caveolin-1 [34,35].

Cell lines and transfection
To investigate the internalization of HER-2-based antibodies, cultured breast cancer cells with or without HER-2 expression were used. In this study, the triple negative breast cancer MDA-MB-231 cell line, and MCF-7 cell line were used as HER-2 negative cells, and SKBR-3, BT-474, and ZR-75-1 were used as HER2 positive cells, whereas ER (estrogen receptor) and PR (progesterone receptor) expression by Allred scores were (0, 0) in MDA-MB-231, (6,6) in MCF-7, (0, 0) in SKBR-3, (0, 8) in BT-474 and (3,4) in ZR-75-1 cells [36]. Cell lines were provided as gift by Dr. Wang LH in Institute of Molecular and Genomic Medicine of National Health Research Institutes, Miaoli, Taiwan. Cells were grown in Dulbecco's modified Eagle's Medium (DMEM) supplemented with 10% calf serum, penicillin, and streptomycin (GIB-CO-BRL, Gaithersberg, MD, USA) and were kept in an incubator with 5% CO 2 at 37°C. For transfection, BT-474 cells were grown on 24-well plates in normal growth medium without antibiotics, and Lipofectamine 2000 transfection reagent (Invitrogen, ThermoFisher, Waltham, MA, USA) was used to induce GFP-tagged caveolin-1 (Addgene, Cambridge, MA, USA) overexpression. Caveolin-1siRNA (Dharmacon, GE health care, Lafayette, CO, USA) was used in the knockdown experiment in SKBR-3 cells. Cells were analyzed 24 hr post-transfection, and the efficacy of transfection was confirmed by immunoblot analysis of cell lysates using rabbit

Western blot
Cells were scraped in lysis buffer (1% NP-40, 50 mM Tris pH 7.4, 150 mM NaCl, 2 mM MgCl 2 , 1 mM EGTA, and protease and phosphatase inhibitors) on an ice tray, and the protein concentration was determined by bicinchoninic acid (BCA) assay. Protein samples were mixed with sample buffer, boiled for 5 min and separated by SDS-PAGE. Proteins on the gel were then transferred onto a PVDF membrane, blocked in blocking buffer containing 5% bovine serum albumin (BSA), and then probed with primary antibodies against caveolin-1, clathrin, Bcl-2, and GAPDH (Santa Cruz, Dallas, TX, USA), followed by incubation with appropriate HRPconjugated secondary antibodies. Blots were developed using an enhanced chemiluminescence system.

Confocal microscopy for the detection of antibody internalization
Cells grown on glass coverslips were incubated with trastuzumab (10 μg/ml) or T-DM1 (1 μg/ ml) at 37°C for 30 min. After washing, cells were fixed with 3.7% formaldehyde and permeabilized with 0.1% Triton-X 100. The fixed cells were incubated with rabbit anti-caveolin-1 antibody (1:100 dilution in PBS/0.1% Triton-100/3% BSA) at room temperature for 1 hr and then incubated with Cy3-conjugated anti-rabbit secondary antibodies (1:200 dilution in PBS/0.1% Triton-100/3% BSA) at room temperature for 1 hr. The presence of trastuzumab and T-DM1 was revealed by a Cy2-conjugated anti-human secondary antibody. Coverslips were mounted with Gel Mount aqueous mounting medium (Sigma, St. Louis, MO, USA). Images were acquired using a Zeiss LSM 510 META confocal microscope with a 63× objective (1.4 oil).

Cell viability using a trypan blue exclusion assay
Cells were seeded into 24 well plates at a density of 0.1 x 10 6 cells/ml and incubated alone or in 10 μg/ml trastuzumab (Herceptin, Roche Ltd., Basel, Swiss), 1 μg/ml T-DM1 (Kadcyla, Roche Ltd., Basel, Swiss), or 1μg/ml paclitaxel (Phyxol, Sinphar Ltd., I-Lan, Taiwan) for 72 hr [37]. At the end of this period, 0.1 ml of 0.4% trypan blue and deionized water (1:1) was added to the control and treated tubes to estimate the number of dead cells. Cell viability was estimated with a hemocytometer. Dead cells were stained blue, while live cells remained unstained. The cell mortality was expressed as the percentage of trypan blue-positive cells compared to the total number of cells. The viability percentage was presented by the number of live cells divided by the number of untreated control cells ×100.

Cell apoptosis analysis
Cell apoptosis was analyzed at 48 hours after drug treatment [11]. Pro-survival protein Bcl-2 was determined by Western blot. A measurement of the extent of apoptosis was carried out using an annexin V detection kit (Invitrogen, ThermoFisher, Waltham, MA, USA) to label cell surface phosphatidylserine of apoptotic cells. Briefly, treated cells were trypsinized and washed twice with phosphate-buffered saline, then suspended in binding buffer (10 mM HEPES, pH 7.4, 140 mM NaCl and 2.5 mM CaCl 2 ). To each 100 μl cell suspension was added 5 μl annexin V conjugated to fluorescein isothiocyanate (FITC), cells were incubated at room temperature for 30 min. Stained cells were analyzed immediately by flow cytometry. To further monitor the status of cell apoptosis, T-DM1 treated cells were live stained with annexin V and propodium iodide (PI) together, co-staining images were taken by inverted fluorescence microscope.
Annexin V stained cells with PI positive staining were excluded as necrotic cells, apoptotic cells with only annexin V staining were counted and divided by total cells as the percentage of apoptosis. Pan-caspase inhibitor Z-VAD-FMK (Enzo, Farmingdale, NY, USA) with working concentration 20 μM was introduced together with T-DM1 for rescue experiment, cell viability was determined as described in cell viability assay.

Statistics
Results are expressed as mean ± standard deviation. A Student t-test was used to compare continuous variables. Differences at the Ã P <0.05, ÃÃ P <0.01 and ÃÃÃ P <0.001 were considered statistically significant. Tukey Kramer test was used for multiple comparisons, different letters represent significant difference at the P <0.01 level. For cell culture experiments, at least three independent experiments were performed.

Expression of caveolin-1 in breast cancer specimens
In this study, the caveolin-1 expression pattern was examined in 32 breast cancer patients. Imnunohistochemical staining showed that caveolin-1 was concentrated at the membrane, and a diffuse cytoplasmic staining pattern was observed also, in ductal epithelial cells with different expression levels (Fig 1). Twenty-two cases (68.75%) of invasive ductal carcinoma showed high levels (score!1.5) of caveolin-1 staining, whereas low expression of caveolin-1 (score <1.5) was noted in 10 cases (31.25%) (S1 Table). The expression of caveolin-1 did not differ between HER-2 (+) and HER-2 (-) tissue (1.97 ± 0.70 vs 1.77 ± 0.77, P = 0.45). Similarly, the scores in the ER (+) and PR (+) groups did not differ to those of the ER(-) and PR(-) groups (1.80 ± 0.77 vs 1.93 ± 0.69, P = 0.63 and 1.77 ± 0.82 vs 1.91 ± 0.66, P = 0.59) (S2 Table). In addition, in the study of Western blot protein evaluation, 8 (72.72%) of the 11 HER-2 positive breast cancer patients were found to have a high expression of caveolin-1 in tumor tissue than in the nontumor part of the patient (S3 Table). Sample of non-tumor site specimens all showed high expression of caveolin-1 with heterogeneous distribution in ductal epithelial.

Endocytic protein expression in breast cancer cell lines
The expression of the two major endocytic proteins clathrin and caveolin-1 was examined in the breast cancer cell lines, MDA-MB-231, MCF-7, SKBR-3, BT-474, and ZR-75-1. Western blotting demonstrated that clathrin and caveolin-1 expression levels differed in these breast cancer cell lines: caveolin-1 was highly expressed in MDA-MB-231 cells. For HER-2 positive cells, caveolin-1 expression was higher in SKBR-3 than in BT-474 cells (Fig 2A, 2B and 2C,  S1A Fig). Clathrin was more abundant in BT-474 cells than in SKBR-3 cells (Fig 2A, 2D (Fig 3A and 3B, left lane). The intensity of caveolin-1 staining was higher in SKBR-3 than BT-474 cells, and showed a clear cytosolic distribution (Fig 3A and 3B, middle lane). Trastuzumab and T-DM1 were colocalized with caveolin-1 in the SKBR-3 cell membrane and cytosol; however, in BT-474 cells, most of the trastuzumab and T-DM1 was localized only in the cell membrane (Fig 3A and 3B, right lane). These results suggest that trastuzumab and T-DM1 were bound to HER-2 positive cells, and that the internalization of trastuzumab and T-DM1 was associated with caveolin-1 expression.

Caveolin-1 expressing HER2 positive cells were sensitive to T-DM1 treatment
Phase contrast micrographs of MDA-MB-231, SKBR-3, and BT-474 cells treated with trastuzumab, T-DM1, and paclitaxel demonstrated that paclitaxel treatment caused the greatest cell toxicity in each of the three cell types. Trastuzumab treatment also had a cytotoxic effect on HER-2 positive SKBR-3 and BT-474 cells; however, T-DM1 treatment caused more cytotoxicity in SKBR-3 cells than in BT-474 cells (Fig 4A). The results of the cell viability assay showed that SKBR-3 was significantly more sensitive to T-DM1 when compared to trastuzumab  (Fig 4B, P <0.01). The cytotoxic efficacy of T-DM1 was similar to paclitaxel treatment ( Fig 4B).
It was demonstrated that T-DM1 induced cell toxicity through apoptosis at 48 hours [11,32]. In this study, the pro-survival protein Bcl-2 which functions in negative regulation of many caspases activity during apoptosis was examined. In the result, lower level of Bcl-2 was detected in T-DM1 treated SKBR-3 cells via Western blot (Fig 4C and 4D). The percentage of apoptosis, represented by annexin V staining (Fig 4E), was significantly increased in SKBR-3 cells treated with T-DM1 (P <0.01), whereas, fewer necrotic cells were found with Propodium iodide (PI) staining (S2 Fig). Furthermore, drug sensitivity to T-DM1 was decreased when SKBR-3 was treated with pan-caspase inhibitor, Z-VAD-FMK (S3 Fig), this result is based on the 1μg/ml concentration of T-DM1 treatment, and however, the level of T-DM1 induced cell toxicity may be different dependent on T-DM1 concentration. These results indicate that HER-2 positive SKBR-3 cells, which have a higher caveolin-1 expression, are sensitive to T-DM1 treatment.

Caveolin-1 expression is required for T-DM1 Internalization
To determine whether caveolin-1 plays role in T-DM1 internalization and influences drug sensitivity, a molecular strategy was applied to regulate caveolin-1 expression in breast cancer cells. BT-474 and SKBR-3 cells, which express relatively lower and higher caveolin-1 levels, respectively, were transfected with GFP-tagged caveolin-1 for caveolin-1 overexpression, or caveolin-1 siRNA for caveolin-1 knockdown, respectively. Western blotting demonstrated that caveolin-1 was overexpressed in GFP-caveolin-1 transfected BT-474 cells, and was suppressed in siRNA-treated SKBR-3 cells (Fig 5A, S4 Fig). The overexpression of caveolin-1 in BT-474 increased T-DM1 cytotoxicity, however, caveolin-1 knockdown in SKBR-3 cells significantly reduced T-DM1 efficacy (P <0.05) (Fig 5B). This T-DM1-induced cytotoxicity was associated with the level of Bcl-2 expression (Fig 5C and 5D), as shown by the ratio of apoptosis (Fig 5E). The cytotoxic effect of T-DM1 was negatively associated with Bc1-2 levels in both SKBR-3 and BT-474 cells, which express different levels of caveolin-1. The percentage of apoptosis measured by flow cytometry was associated with the chemosensitivity of the cells to T-DM1 induced by the overexpression of caveolin-1. Taken together, these data suggest that caveolin-1 is required for T-DM1 internalization and that this internalized conjugate further induces cell apoptosis.

Discussion
Caveolin-1 regulates the formation of caveolae; the invaginated membrane structure with multiple transport and sorting functions in membrane endocytic system [38]. Caveolin-1 has been demonstrated to have an oncogenic or tumor suppressor role in breast cancer [29,39]. It has Caveolin-1 Regulates T-DM1 Internalization in Breast Cancer Cells also been suggested that the high expression of caveolin-1 in stromal cells has a protective effect against tumor progression in breast cancer and could be a good prognostic indicator [35,40]. In our study, a high level of caveolin-1 expression was noted in 68.8% of breast cancer specimens. This incidence is similar to that of a study by Park et al with 130 breast cancer patients [35]. However, their caveolin-1 expression was correlated inversely with HER-2 expression status. This phenomenon was not observed in our series; possibly due to small sample size. Metaanalysis from 19 eligible studies included a total of 5,926 patients with a median number of 312 patients per study failed to recognize that caveolin-1 expression in tumor cells as a predictive marker for breast cancer prognosis. [41].
However, caveolin-1 may participate in different signaling pathways depending on cell type and the genetic background of the breast cancer [42,43]. In our study, caveolin-1 expression varied in different breast cancer cell lines: triple negative MDA-MB-231 cells showed the greatest caveolin-1 expression, while HER-2 positive SKBR-3 and BT-474 cells showed relatively moderate and lower caveolin-1 expressions, respectively. This observation agreed with the results of a previous study in which the levels of caveolin-1 mRNA level were shown to be associated with DNA methylation [44]. Furthermore, our results demonstrated that the expression of caveolin-1 can improve T-DM1 internalization and drug efficacy, although tumor stage and the expression of molecular effectors may affect the role of caveolin-1 during tumor Caveolin-1 Regulates T-DM1 Internalization in Breast Cancer Cells progression [45]. The efficacies of trastuzumab and T-DM1 in breast cancer cell lines have been demonstrated previously [11,37]. According to Barok's protocol [37], our data also showed trastuzumab treatment had a similar efficacy on 72 hours treatment in SKBR-3 and BT-474 cell lines, based on HER-2 expression, while SKBR-3 cells, which have a higher caveolin-1 expression, were more sensitive to T-DM1. The sensitivity of T-DM1 was associated with caveolin-1 expression. Caveolin-1 has been shown to improve trastuzumab internalization [31]. In this study, the role of caveolin-1 in T-DM1 internalization and its dramatic cytotoxicity was demonstrated for the first time.
Molecular mechanisms have been suggested for the antitumor activity of T-DM1, including disruption of the microtubule network of target cells, the induction of apoptosis, mitotic catastrophe, and defective intracellular trafficking of HER-2 [32,37]. Our results revealed effective cytotoxicity of paclitaxel in the treatment of different breast cancer cells which were correlated with cell apoptosis. Although the Tukey statistic showed no significant difference of Bcl-2 expression in MDA-MB-231 and BT-474 cells when trastuzumab, T-DM1 and paclitaxel treatment were compared. However, paclitaxel treatment showed significant reduction of Bcl-2 expression in MDA-MB-231 (P <0.01) and BT-474 cells (P <0.01) when compared to trastuzumab treatment by t-test. Since the chemotherapeutic drugs of taxane group have high incidence of side effect in clinical application [46], our study focused on the mechanism of caveolin-1 induced chemosensitivity of T-DM1 that also associated with a reduction in Bcl-2 expression and an increase in the percentage of apoptosis. The expression of Bcl-2 protein was demonstrated to be associated with its transcription level [47].
Endocytosis has been recognized as the most significant pathway in the downregulation of EGFR by removing the receptor from cell surface for degradation in lysosomes [48]. Recently, the dominant opinion is that trastuzumab causes the endocytic downregulation of HER-2 [49][50][51]. Although the endocytic downregulation of EGFR has been largely attributed to clathrindependent endocytosis [52], other endocytic pathways have also been proposed in recent years, especially following stimulation with high concentration of EGF. In human mammary glands, the EGF concentration is greater than 100 ng/ml [53]. In a study by Sigismud et al. [54], the EGFR was found to become ubiquitinated and was internalized by caveolae at high concentration of EGF (20 ng/ml). The dynamics of membrane receptors are regulated by vesicle trafficking systems [18], therefore, the trafficking route of receptor-bound antibodies affects the antibody half-life in the cell and determines the drug efficacy.
The action of trastuzumab is dependent on HER-2 signaling through Src hyperactivation, which further inhibits PTEN and its downstream PI3K/AKT signaling in cell proliferation. Patients with a lack of PTEN, or compensatory HER-3 or IGF1-R activation, may be resistant to trastuzumab treatment [55]. Our results, indeed, were consistent with the theory that the efficacy of T-DM1 is mainly through the action of DM1 derivative that was released by the degradation of T-DM1, but that HER-2 expression is also required for T-DM1 treatment [11]. In addition, we confirmed caveolin-1 overexpression could be used to enhance T-DM1 uptake. However, a follow-up study on a second independent dataset is required to validate the correlation between the expression of caveolion-1, as well as drug sensitivity, with the response to T-DM1 clinically.

Conclusion
This study demonstrated the role of endocytic protein caveolin-1 in T-DM1 internalization and improved cytotoxicity in breast cancer cells. The expression of caveolin-1 in HER-2 positive breast cancer cells could be a potential biomarker to predict the efficacy.
Supporting Information S1