CCN5 activation by free or encapsulated EGCG is required to render triple‐negative breast cancer cell viability and tumor progression

Abstract Epigallocatechin‐3‐gallate (EGCG) has been considered an anticancer agent despite conflicting and discrepant bioavailability views. EGCG impairs the viability and self‐renewal capacity of triple‐negative breast cancer (TNBC) cells and makes them sensitive to estrogen via activating ER‐α. Surprisingly, the mechanism of EGCG’s action on TNBC cells remains unclear. CCN5/WISP‐2 is a gatekeeper gene that regulates viability, ER‐α, and stemness in TNBC and other types of cancers. This study aimed to investigate whether EGCG (free or encapsulated in nanoparticles) interacts with the CCN5 protein by emphasizing its bioavailability and enhancing its anticancer effect. We demonstrate that EGCG activates CCN5 to inhibit in vitro cell viability through apoptosis, the sphere‐forming ability via reversing TNBC cells’ stemness, and suppressing tumor growth in vivo. Moreover, we found EGCG‐loaded nanoparticles to be functionally more active and superior in their tumor‐suppressing ability than free‐EGCG. Together, these studies identify EGCG (free or encapsulated) as a novel activator of CCN5 in TNBC cells and hold promise as a future therapeutic option for TNBC with upregulated CCN5 expression.


| INTRODUC TI ON
Breast cancer (BC) is a genetically heterogeneous disease characterized by a mixed bag of cells. 1 Although overall BC patients' mortality has declined significantly, the incidence remains high. It is still the second most common cause of cancer death in women. 2 BC is broadly classified into distinct molecular subtypes, including normal-like, luminal A and B, HER2+, and basal-like. The basal-like subtype is also known as triple-negative breast cancer (TNBC) because they lack estrogen receptor (ER), progesterone receptor (PR), and HER2. 2 TNBC is characterized by resistance to chemotherapy, the acquisition of the stem character, and unfavorable prognoses due to its highly metastatic phenotype. 3 TNBC patients still have minimal treatment options, 3 and chemotherapy is currently the only treatment available for metastatic TNBC. 3 Although checkpoint inhibitors, including programmed cell death protein 1 (PD-1) and programmed death-ligand 1 (PD-L1), were found to elicit a response in TNBC in initial clinical trials, 3 optimistic results have not yet emerged from these trials. Thus, the detection of appropriate targeted therapeutic regimens for TNBC therapy and prevention has remained an elusive challenge to many laboratories. These malignant breast tumors often consist of a small subset of the cell population, known as tumor-initiating cells (BTICs) or cancer stem cells (BCSCs). This subpopulation of BC cells is known to be CD44 positive (CD44 +/+ ) with negligible or no CD24 (CD24 low/− ) and are responsible for the acquisition of chemoresistance properties and tumor recurrence. 4,5 Hence, targeting these residual cells may act as a novel therapeutic approach to prevent tumor recurrence and improve long-term survival in breast cancer patients.
During our program of identifying molecule(s) that could play an inhibitory role against TNBC, we found that cellular communication network factor 5 (CCN5, previously known as WISP-2), a matricellular 29-35 kDa protein and a member of the CCN family of growth factors, can modulate breast cancers by imparting an inhibitory effect on tumor progression. [6][7][8][9] We have demonstrated that ectopic expression of CCN5 or administration of human recombinant CCN5 protein in TNBC cells resulted in suppressing tumorigenic properties and induction of growth arrest. 9 Subsequently, the consistent role of CCN5 in tumor suppression was also reported by others. [10][11][12][13] is also known to inhibit the stemness and reverse the EMT process in BC cells 9,11 and activates estrogen receptor-alpha (ERα) in TNBC cells. 14 These recent advances, exploring the role of CCN5-signaling in breast cancer, strongly suggest that targeting TNBC-BCSC by activating CCN5 would be an ideal strategy to prevent breast tumors' growth and relapse.
Epigallocatechin-3-gallate (EGCG), the most abundant dietary polyphenol in green tea, has been extensively studied in cancer prevention, incidence, or motility. [15][16][17] EGCG has proven effective in delaying tumor incidence and significantly reducing tumor burden. 18 Although the bioavailability of EGCG is a big concern in the clinics, multiple studies found that EGCG can activate cell-death programs and suppress the invasiveness of TNBC both in vitro and in vivo via activation of ERα. [19][20][21] However, the precise molecular mechanism of EGCG action on TNBC growth inhibition and suppressing invasive phenotypes is unclear. Since the recent shreds of evidence suggest that activation of CCN5 could help sensitize TNBC cells to conventional hormonal therapies through activation of ERα, 14   containing 10% FBS (ATCC) at 37°C in a humidified chamber. Cell lines are maintained at ≤18 passage from receipt and were characterized using short tandem repeat analysis before initial culture. Besides, cell lines were routinely checked for mycoplasma contamination.

| Cell viability assay
Antiproliferative effects of EGCG (0-100 µM) on various cancer cells were determined by crystal violet assay following the protocol published previously. 22 Cells were treated with EGCG for 72 hours and, after treatment, subjected to crystal violet staining for 10 min. The resultant crystal violet complex was then dissolved in 10% acetic acid, and the absorbance was measured at 600 nm, using a SpectraMax-340 microplate reader (Molecular Devices, Sunnyvale, CA). Cell viability was calculated using the following mathematical expression: [At and As indicated the absorbance of the test substances and solvent control, respectively 23 ].

| Anchorage-dependent growth (ADG) assay
Anchorage-dependent growth (ADG) or colony-formation assay for EGCG-treated and untreated cells was performed by following the previously published protocol. 22 To determine the colony-forming ability of TNBC cells, MDA-MB-231 cells were treated with EGCG (0-75 µM) for 72 hours and seeded at a density of 2 cells/μl. EGCGtreated or untreated MDA-MB-231 cells were distributed in two treatment groups. In one group, pretreated cells were subjected to colony formation for 7 days (pretreatment group-T1) in the absence of EGCG. In contrast, in another group, pretreated cells were seeded for colony formation and subjected to EGCG treatment for the remaining 6 days (posttreatment group-T2). Colonies were then stained with crystal violet, and the number of colonies was determined using the Colony Doc-It Image station (UVP, Upland, CA).

| Real-time PCR
The qRT-PCR mixture was prepared using SYBR green master mix (Applied Biosystems). Master mix and cDNA were combined in a 48-well plate, and the samples were run in an Applied Biosystem StepOne Real-time PCR machine. The CCN5 and GAPDH primers were used as listed in Table S1. The relative changes of gene expression were calculated using the following formula: fold change in where ΔCt = Ct (CCN5) -Ct (GAPDH) and Ct represents threshold cycle number.

| Immunohistochemistry
All immunohistochemistry experiments were performed using the Histostain-Plus IHC Kit (Life Technologies) following the previously published protocol. 9 Briefly, after microwave treatment of deparaffinized tissue sections in citrate buffer and endogenous peroxide blocking, the sections were incubated in a specific antibody solution overnight at 4°C in a moist chamber. The immune reactivity was detected by DAB (3,3′-Diaminobenzidine), and the sections were counterstained with hematoxylin.

| Western blot analysis
Western blot analysis was performed to determine the expression levels of the apoptotic proteins, EMT-related proteins, and CCN5 in untreated and EGCG-treated BC cell lines using our previous method. 26

| Generation and characterization of EGCG nanoparticles (NPs)
The EGCG-NPs were prepared and characterized as the protocol described earlier. 27 Two types of nanoparticles, viz., FA-NPs-PEG and FA-PEG-NPs, were prepared as described in Figure

| Cellular uptake studies
Cellular uptake of nanoparticles was determined as described previously. 27 Briefly, TNBC cells were cultured with a density of 20,000 cells/well in six-well plates for 2 days to achieve 70% confluent monolayer. Subsequently, Alexa Fluor 647 (AF-647)-labeled nanoparticle suspensions (80 μl or equivalent amount containing 75 μM EGCG) were added to cells. Following the treatment at various time points (1, 5, and 24 hours), fluorescence emission intensity was visualized and quantified using a Nikon Eclipse 90i microscope equipped with imaging software. The fluorescence intensity was quantified using NIS Elements BR Software and verified by the ImageJ 1.45s Software (NIH, Bethesda, MD).

| Tumor xenografts in Athymic Female Nude Mice
Six-to 8-week-old athymic female nude mice were obtained from Jackson Laboratory and were used for tumor development.

| Statistical analysis
The data arrangement, organization, and statistical analysis were performed as per Michel et al. 28 All data are presented as the mean ± SD of "n" independent measurements, as indicated in the corresponding figure legends. Statistical comparisons between treated and untreated control groups were calculated by Student's t tests using GraphPad Prism 6, and multiple groups were determined by ANOVA test. A value of p < 0.05 was considered significant.

| Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries.

| EGCG transcriptionally activates CCN5 in breast cancer cells
Both CCN5 and EGCG restore ERα expression and activity in TNBC cells. 14,19,21 In this study, we investigated whether EGCG can induce or enhance CCN5 expression in TNBC cell lines. We  weakens the steaminess of these aggressive cells 31 . Thus, we tested whether EGCG treatment effectively upregulates CCN5 expression in Panc-1 cells, an aggressive PDAC cell line. We found that CCN5 protein level significantly increased in Panc-1 cells following EGCG treatment for 48 hours ( Figure 2D).
Next, we determined whether EGCG transcriptionally regulates

| EGCG decreases cell viability through apoptosis in BC cells via upregulation of CCN5
The cell viability studies demonstrate a dose-dependent effect of EGCG on cell killing in four BC cell lines ( Figure 3A

| CCN5 recombinant protein therapy synergizes EGCG's sensitivity on cell viability
We have previously shown that human recombinant CCN5 (hrCCN5) protein treatment slightly but significantly suppresses TNBC cell viability. Thus, we investigate whether hrCCN5 protein treatment enhances EGCG sensitivity on TNBC cell lines' viability. MDA-MB-231 and 4T1 cells were exposed to EGCG alone or in a combination of EGCG and hrCCN5 protein in various concentrations for 48 hours.
The synergy (additive or super-additive) effect of hrCCN5 and EGCG was evaluated using the Loewe model, 32,33 and the results are illustrated in Figure 4. Loewe analysis was conducted using the Combenefit® software. 33 We found that the combination treat-

| EGCG inhibits the sphere-forming ability of TNBC cells through upregulation of CCN5
Cancer stem cells, also known as TICs, are responsible for metastasis, tumor relapse, and acquisition of chemoresistance properties. 4 were reseeded for mammosphere formation. We found that hrCCN5 protein-treated cells exhibited greater sensitivity to EGCG than EGCG-alone treated, while CCN5 Ab -treated cells were less sensitive to EGCG (Figure 5F-G). These findings provide a mechanistic basis for EGCG therapy and indicate that CCN5 is the target for activation of EGCG to exert its anti-EMT effect.

| Effect of EGCG-loaded nanoparticles on CCN5 expression in TNBC cells
Although EGCG exhibits numerous promising health-promoting impacts in several in vitro and in vivo studies, weak bioavailability considerably higher than NPs and PEG-NPs and suggested that this could be the presence of FA, which binds with the cancer cells having FA receptor. 27 Therefore, in this study, we first established the targeting ability of nanoparticles, and to do so, we determined the cellular uptake of EGCG-encapsulated nanoparticles. As shown in Figure 6A-B, the cellular uptake of nanoparticles, based on fluorescence intensities, was significantly higher in cells treated with FA-PEG-NPs as compared to FA-NPs-PEG-and PEG-NPs-treated cells.
However, a significant increase in cellular uptake of FA-NPs-PEG was also observed when compared with PEG-NPs-treated cells. The cellular uptake data suggest that FA's spatial arrangement within the chemical constructs is critical to portend the nanoparticles' sensitivity.
Based on the above cellular uptake results, we anticipated that EGCG containing FA-PEG-NPs could be more effective than EGCG containing FA-NPs-PEG in cancer cells' physiology. Thus, we first investigated the effect of FA-NPs-PEG and FA-PEG-NPs on CCN5 expression in TNBC cells. We found that FA-PEG-NPs significantly upregulated CCN5 protein expression in a dose-dependent fashion compared to drug-free NPs ( Figure 6C). On the other hand, the effect of FA-NPs-PEG on CCN5 was weak compared to FA-PEG-NPs ( Figure 6D). The dose-dependent studies revealed that FA-NPs-PEG upregulated CCN5 at the dose of 75 μM, while the effect of FA-PEG-NPs on CCN5 expression was first detected at the dose of 25 μM. In conclusion, EGCG containing FA-PEG-NPs was found to be functionally superior to FA-NPs-PEG in upregulating CCN5 in TNBC cells.

| Effect of EGCG-loaded nanoparticles on in vitro cell viability and sphere-forming ability of TNBC cells
To gain some insight into EGCG-loaded nanoparticles' potential, we next examined the capacity of EGCG containing FA-PEG-NPs to inhibit TNBC cell viability and sphere-forming ability. We   Figure 8B). We found that free-EGCG and FA-PEG-NPs had no impact on tumor growth than the untreated group, while EGCG-loaded FA-PEG-NPs significantly delayed tumor growth ( Figure 8G) with no effect on body weight (data are not shown). Collectively, these studies suggest that targeting TNBC by EGCG-loaded nanoparticles may have substantial therapeutic benefit. However, further studies are warranted.

| DISCUSS ION
Over the last decade, several studies, including our laboratory, have revealed the modulatory roles of CCN5 in breast cancer progression. 6,[8][9][10] In addition to suppressing breast cancer cell viability, 7,8,42 reprograming of MET, a hallmark of reversing cancer stemness mechanism, 43 is also known to be regulated by CCN5 through TGFβ-mediated signaling. 9,13 Further experimental pieces of evidence have revealed that CCN5 suppresses the expression of the microRNA-10b, which plays a critical role in microinvasion and metastasis of breast cancer cells 9 and induces the expression of the tumor suppressor protein p27. 44 In a very recent study, the immu- F I G U R E 7 EGCG-loaded FA-PEG-NPs is equally effective as free-EGCG in inhibiting colony-and sphere-forming ability of TNBC cells. (A-B) The colonyforming ability of free-EGCG and EGCGloaded nanoparticles treated TNBC cells as determined using anchoragedependent growth (ADG) assay. The graph shows the mean ± SD from triplicate measurements, ns, non-significant. (C-D) The sphere-forming ability of free-EGCG and EGCG-loaded nanoparticles treated TNBC cells as determined using the mammosphere assay (C). The graph (D) shows the mean ± SD from triplicate measurements, ns, non-significant It has been reported that the effectiveness of EGCG against several carcinomas is mediated by promoting cell death and reversing the EMT/cancer stemness. 16,17 But how EGCG regulates cell death and EMT/stemness in TNBC cells is quite uncertain and warrants further investigation. Hence in the present study, our goal was to investigate the link between CCN5 and EGCG's pathophysiological roles.
In breast cancers, CCN5 is only detected in non-invasive cells.
In most TNBC cells and pancreatic cancer cells, which are aggressive in nature and mesenchymal type, CCN5 expression is undetected. 7,8,14,31,45,46 We found that EGCG has the potency to upregulate CCN5 in TNBC and pancreatic cancer cells in a dose-dependent manner and at the transcription level ( Figure 2). These findings suggest the possible involvement of CCN5 in reprograming cell death and MET by EGCG.

The cell viability studies in different breast cancer cell lines
showed that EGCG significantly suppresses cell growth in a dosedependent fashion with varying IC 50 values (Figures 3 and 4). We found that the reduced cell viability by EGCG was due to apoptosis via enhancing the CCN5 signal in TNBC cells (Figures 3 and 4).
However, how CCN5 specifically promotes apoptosis is unclear, and thus further studies are warranted.
The BTIC/BCSCs, which survive after conventional chemotherapy treatment, typically express unique signature protein markers Data represent error bar mean ± SD, n = 4 animals/group. P-value determined by one-way ANOVA and Student's t test, data represent error bar mean ± SD, n = 4 animals/group. (G) Quantification of tumor volume in free-EGCG and EGCG-loaded nanoparticles treated (14 days) tumor xenograft mouse model. Data represent error bar mean ± SD, n = 4 animal/group associated with EMT and stemness 5,47 and are also known to associate with enhanced clonal growth ability self-renewal properties. 37,48 We observed that EGCG, independent of the antiproliferative effect, significantly blocked the self-renewal or mammosphere-forming ability of TNBC cells ( Figure 5). Simultaneously, EGCG treatment resulted in the drastic inhibition of the expression of mesenchymal and stemness markers. On the other hand, epithelial markers were found to be significantly upregulated ( Figure 5). Overall, these findings implicate MET reprograming and self-renewal in EGCG target in TNBC and potentially in other malignancies.
Although the mechanism of regulation of self-renewal event by EGCG is unknown, the mechanistic studies found that hrCCN5 protein treatment enhanced the suppressing effect of EGCG on the sphere-forming ability of TNBC cells, while CCN5 antibody treatment impaired EGCG action ( Figure 5). Thus, it is alluring to consider that CCN5 is poised to be a key player in this perhaps currently poorly understood mechanism.
The impact of EGCG in the clinical setup is inconsistent and highly debated, which could be attributed to its very low bioavailability. 49 Thus, we exploited a nanostructure-based drug delivery system, known to be an enhancer of bioavailability even at much lower doses than conventional preparations, as the stability of EGCG in the simulated intestinal fluid is significantly improved by encapsulation methods. 40,49,50 Given the importance of nanoparticle-based drug delivery, we loaded EGCG into two structurally different nanoparticles established previously 27 and determined the efficacy. We found that one of the two EGCG-loaded nanoparticles (FA-PEG-NPs) (Figure 1) was significantly efficient in activating CCN5, reducing cell viability and sphere-forming ability of TNBC cells (Figures 6 and 7). Based on these novel observations, we may conclude that the reactivation of CCN5 in TNBC by EGCG provides a potentially unique therapeutic strategy for eliminating the residual tumor cells and preventing tumor growth and recurrence.

ACK N OWLED G M ENTS
We thank the members of our laboratories for their helpful discussion and VA Core facilities for confocal microscopy. We also thank Melinda Broward for editorial help. The work is supported by a Merit Review grant from the Department of Veterans Affairs

CO N FLI C T O F I NTE R E S T
All authors declare no competing financial interests.

D ECL A R ATI O N O F TR A N S PA R EN C Y A N D SCI ENTI FI C R I G O R
This declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigor of preclinical research as stated in the BJP guidelines for Design and Analysis, cell biology techniques, and Animal studies and as recommended by funding agencies, publishers, and other organizations engaged with supporting research.

DATA AVA I L A B I L I T Y S TAT E M E N T
The authors affirm that all the supporting data of these studies are available within the articles and Supplementary Information or from the corresponding authors on reasonable request.