Liposomes Targeting P21 Activated Kinase-1 (PAK-1) and Selective for Secretory Phospholipase A2 (sPLA2) Decrease Cell Viability and Induce Apoptosis in Metastatic Triple-Negative Breast Cancer Cells

P21 activated kinases (or group I PAKs) are serine/threonine kinases whose expression is altered in prostate and breast cancers. PAK-1 activity is inhibited by the small molecule “Inhibitor targeting PAK-1 activation-3” (IPA-3), which has selectivity for PAK-1 but is metabolically unstable. Secretory Group IIA phospholipase A2 (sPLA2) expression correlates to increased metastasis and decreased survival in many cancers. We previously designed novel liposomal formulations targeting both PAK-1 and sPLA2, called Secretory Phospholipase Responsive liposomes or SPRL-IPA-3, and demonstrated their ability to alter prostate cancer growth. The efficacy of SPRL against other types of cancers is not well understood. We addressed this limitation by determining the ability of SPRL to induce cell death in a diverse panel of cells representing different stages of breast cancer, including the invasive but non-metastatic MCF-7 cells, and metastatic triple-negative breast cancer (TNBC) cells such as MDA-MB-231, MDA-MB-468, and MDA-MB-435. We investigated the role of sPLA2 in the disposition of these liposomes by comparing the efficacy of SPRL-IPA-3 to IPA-3 encapsulated in sterically stabilized liposomes (SSL-IPA-3), a formulation shown to be less sensitive to sPLA2. Both SSL-IPA-3 and SPRL-IPA-3 induced time- and dose-dependent decreases in MTT staining in all cell lines tested, but SPRL-IPA-3-induced effects in metastatic TNBC cell lines were superior over SSL-IPA-3. The reduction in MTT staining induced by SPRL-IPA-3 correlated to the expression of Group IIA sPLA2. sPLA2 expression also correlated to increased induction of apoptosis in TNBC cell lines by SPRL-IPA-3. These data suggest that SPRL-IPA-3 is selective for metastatic TNBC cells and that the efficacy of SPRL-IPA-3 is mediated, in part, by the expression of Group IIA sPLA2.


Introduction
Breast cancer is one of the most common invasive cancers affecting women worldwide and is a leading cause of cancer-related deaths [1]. The median age at the time of breast cancer diagnosis is 61 years, and about 20% of breast cancer patients are diagnosed before the age of 50. In general, 60% of breast cancers are diagnosed at a localized stage that is treated with breast-conserving surgery to have high levels of expression in both primary tumors and metastasis [37]. Thus, this study tested the hypothesis that SPRL-IPA-3 displays an enhanced ability to inhibit breast cancer cell growth and induce cell death as compared to free IPA-3 or IPA-3 encapsulated in SSL.  The above studies suggest the clinical utility of SSL-and SPLR-IPA-3 for prostate cancer. Our studies also suggest that SSL-IPA-3 can inhibit breast cancer cell growth. However, SPRL has never been validated with a different drug in non-prostate cancer cell lines. Studies are also limited with regards to the molecular determinants of SPRL efficacy. This prospect is exciting as we have previously shown that SPRL containing doxorubicin, a standard treatment for breast cancer, demonstrated improved efficacy over SSL in xenograft models of prostate cancer [35]. Besides, a recent study assessing the expression of Group IIA PLA2 in a cohort of advanced breast cancer patients reported a large proportion to have high levels of expression in both primary tumors and metastasis [37]. Thus, this study tested the hypothesis that SPRL-IPA-3 displays an enhanced ability to inhibit breast cancer cell growth and induce cell death as compared to free IPA-3 or IPA-3 encapsulated in SSL.

Effect of IPA-3 Encapsulated Liposomes (SSL-IPA-3) on Breast Cancer Cell Viability
We previously showed that TNBC cells had a higher sensitivity to the activity of free unencapsulated IPA-3 compared to non-metastatic cancer cells (MCF-7) [38]. Treatment of metastatic TNBC cells with SSL-IPA-3 decreased MTT staining in a dose-dependent manner ( Figure 2). SSL-IPA-3 did not significantly decrease MTT staining in MCF-7 cells, which was expected as these cells appear to be resistant to the activity of free IPA-3 [38].   Table summarizing the hydrodynamic diameter, polydispersity index (PDI), and zeta potential of SSL-IPA-3 and SPRL-IPA-3 as determined using dynamic light scattering (DLS). (B) The size distribution of the liposomal suspensions (SSL-IPA-3 in black and SPRL-IPA-3 in red) as determined using dynamic light scattering (DLS). Data are representative of at least three different measurements.

Effect of IPA-3 Encapsulated Liposomes (SSL-IPA-3) on Breast Cancer Cell Viability
We previously showed that TNBC cells had a higher sensitivity to the activity of free unencapsulated IPA-3 compared to non-metastatic cancer cells (MCF-7) [38]. Treatment of metastatic TNBC cells with SSL-IPA-3 decreased MTT staining in a dose-dependent manner ( Figure 2). SSL-IPA-3 did not significantly decrease MTT staining in MCF-7 cells, which was expected as these cells appear to be resistant to the activity of free IPA-3 [38]. Annexin V and PI staining were measured to determine if decreases in MTT staining were a result of cell death. We focused these studies on MDA-MB-468 cells, which showed the highest sensitivity to both free IPA-3 and SSL-IPA-3. Treatment of these cells with empty liposomes did not cause significant changes in the number of cells staining positive for annexin V and/or PI when compared to control cells. Treatment of cells with free IPA-3 increased the number of cells staining positive for annexin V (Figure 3). Increases were also seen in cells staining positive for both annexin V and PI; however, free IPA-3 did not result in an increase in cells positive for PI, even at the highest Annexin V and PI staining were measured to determine if decreases in MTT staining were a result of cell death. We focused these studies on MDA-MB-468 cells, which showed the highest sensitivity to both free IPA-3 and SSL-IPA-3. Treatment of these cells with empty liposomes did not cause significant changes in the number of cells staining positive for annexin V and/or PI when compared to control cells. Treatment of cells with free IPA-3 increased the number of cells staining positive for annexin V (Figure 3). Increases were also seen in cells staining positive for both annexin V and PI; however, free IPA-3 did not result in an increase in cells positive for PI, even at the highest concentrations tested. Similarly, SSL-IPA-3 also increased the percentage of cells staining positive for annexin V alone, as well as those staining positive for both annexin V and PI. Interestingly, and unlike free IPA-3, SSL-IPA-3 treatment increased the number of cells staining positive for PI alone, suggesting that these liposomes may also induce cell death via necrosis. concentrations tested. Similarly, SSL-IPA-3 also increased the percentage of cells staining positive for annexin V alone, as well as those staining positive for both annexin V and PI. Interestingly, and unlike free IPA-3, SSL-IPA-3 treatment increased the number of cells staining positive for PI alone, suggesting that these liposomes may also induce cell death via necrosis.

Effect of Secretory Responsive Liposomes Containing IPA-3 (SPRL-IPA-3) on Breast Cancer Cell Viability
Only a few studies exist determining the expression of Group IIA sPLA2 in breast cancer [37,39,40]. Those studies that do exist measuring Group IIA sPLA2 protein expression in clinical tissues report similar levels in both primary and metastatic tumors [37]. We validated that this type of expression was also seen in the cell line panel tested ( Figure 4). Further, similar to prostate cancer cells and patients, expression of Group IIA sPLA2 was higher in cell lines derived from highly metastatic and aggressive TNBC cells, as compared to non-cancerous (MCF-10A) and non-metastatic cells (BT-474, MCF-7). We also assessed the location of Group IIA sPLA2 in select cells by treating them with pronase, which can cleave and release membrane proteins. As predicted for a membranelocalized protein, pronase treatment decreased Group IIA sPLA2 levels in all cell lines tested (Supplementary Figure S2). Bright-field microscopy was used to verify the viability of these cells after pronase treatment to ensure that any decreases were not due to cell death from membrane rupture.

Effect of Secretory Responsive Liposomes Containing IPA-3 (SPRL-IPA-3) on Breast Cancer Cell Viability
Only a few studies exist determining the expression of Group IIA sPLA 2 in breast cancer [37,39,40]. Those studies that do exist measuring Group IIA sPLA 2 protein expression in clinical tissues report similar levels in both primary and metastatic tumors [37]. We validated that this type of expression was also seen in the cell line panel tested ( Figure 4). Further, similar to prostate cancer cells and patients, expression of Group IIA sPLA 2 was higher in cell lines derived from highly metastatic and aggressive TNBC cells, as compared to non-cancerous (MCF-10A) and non-metastatic cells (BT-474, MCF-7). We also assessed the location of Group IIA sPLA 2 in select cells by treating them with pronase, which can cleave and release membrane proteins. As predicted for a membrane-localized protein, pronase treatment decreased Group IIA sPLA 2 levels in all cell lines tested (Supplementary Figure S2). Bright-field microscopy was used to verify the viability of these cells after pronase treatment to ensure that any decreases were not due to cell death from membrane rupture.
Having verified the expression and membrane location of Group IIA sPLA 2, we next compared the efficacy of SPRL-IPA-3 to SSL-IPA-3 using MTT staining. Similar to the effect of free IPA-3 and SSL-IPA-3, SPRL-IPA-3 did not decrease MTT staining in the less metastatic MCF-7 cells ( Figure 5  Having verified the expression and membrane location of Group IIA sPLA2, we next compared the efficacy of SPRL-IPA-3 to SSL-IPA-3 using MTT staining. Similar to the effect of free IPA-3 and SSL-IPA-3, SPRL-IPA-3 did not decrease MTT staining in the less metastatic MCF-7 cells ( Figure 5). In contrast, SPRL-IPA-3 appeared to induce greater decreases in MTT staining at the highest dose used in MDA-MB-468 and MDA-MB-435 cells.
We further compared the ability of SPRL-IPA-3 to induce cytotoxicity using annexin V and PI staining ( Figure 6). We focused these studies on MDA-MB-468 cells as these appeared to be the most sensitive cell line tested based on decreases in MTT staining ( Figure 5). Similar to free IPA-3 and SSL-IPA-3, SPRL-IPA-3 treatment increased the percentage of cells staining positive for both annexin V and PI ( Figure 6). Further, SPRL-IPA-3 treatment appeared to have a significantly higher effect in inducing apoptosis as evidenced by the increased number of cells staining positive for annexin V and PI, as compared to SSL-IPA-3. An increase in the number of annexin V-or PI-positive cells was similar between SPRL-IPA-3 and SSL-IPA-3 ( Figure 6C). These data suggest that SPRL-IPA-3 can induce cell death in breast cancer cell lines and the mechanism of cell death is similar to SSL-IPA-3. These data also suggest that SPRLs induce cell death at a superior efficacy in MDA-MB-468 cells.
Our data suggest that the efficacy of SPRL-IPA-3 in cells with high Group IIA sPLA2 is higher than that seen in cells with lower expression. To investigate whether this observation is drugdependent, we tested the effect of SPRL loaded with doxorubicin (SPRL-Dox) in BT-474 cells, which had relatively lower levels of Group IIA sPLA2 expression, and MDA-MB-468 cells, which had relatively higher levels of Group IIA sPLA2 (Figure 4). Treatment of BT-474 cells with SPRL-Dox slightly decreased MTT staining as compared to controls. In contrast, treatment of MDA-MB-468 cells resulted in significant decreases in MTT staining, even at the lowest concentration tested (Supplementary Figure S3A). Such a finding suggests that the efficacy of both SSL and SPRL liposomes is dependent on both the encapsulated drug and the cell type. We further compared the ability of SPRL-IPA-3 to induce cytotoxicity using annexin V and PI staining ( Figure 6). We focused these studies on MDA-MB-468 cells as these appeared to be the most sensitive cell line tested based on decreases in MTT staining ( Figure 5). Similar to free IPA-3 and SSL-IPA-3, SPRL-IPA-3 treatment increased the percentage of cells staining positive for both annexin V and PI ( Figure 6). Further, SPRL-IPA-3 treatment appeared to have a significantly higher effect in inducing apoptosis as evidenced by the increased number of cells staining positive for annexin V and PI, as compared to SSL-IPA-3. An increase in the number of annexin V-or PI-positive cells was similar between SPRL-IPA-3 and SSL-IPA-3 ( Figure 6C). These data suggest that SPRL-IPA-3 can induce cell death in breast cancer cell lines and the mechanism of cell death is similar to SSL-IPA-3. These data also suggest that SPRLs induce cell death at a superior efficacy in MDA-MB-468 cells.
Our data suggest that the efficacy of SPRL-IPA-3 in cells with high Group IIA sPLA 2 is higher than that seen in cells with lower expression. To investigate whether this observation is drug-dependent, we tested the effect of SPRL loaded with doxorubicin (SPRL-Dox) in BT-474 cells, which had relatively lower levels of Group IIA sPLA 2 expression, and MDA-MB-468 cells, which had relatively higher levels of Group IIA sPLA 2 (Figure 4). Treatment of BT-474 cells with SPRL-Dox slightly decreased MTT staining as compared to controls. In contrast, treatment of MDA-MB-468 cells resulted in significant decreases in MTT staining, even at the lowest concentration tested (Supplementary Figure S3A). Such a finding suggests that the efficacy of both SSL and SPRL liposomes is dependent on both the encapsulated drug and the cell type.
in prostate cancer cells. Thus, we tested the ability of the sPLA2 inhibitor varespladib to alter the ability of SPRL-Dox to decrease MTT staining in breast cancer cells. Varespladib is more potent and selective for Group IIA sPLA2 than LY311727. Similar to our previous studies using the LY311727 inhibitor, varespladib pretreatment did not alter the activity of SPRL-Dox in any cell line tested, which included those with the highest expression of Group IIA sPLA2 (Supplementary Figure S3B   It should be noted that SPRLs are designed to interact with sPLA 2 and our previous studies showed that the sPLA 2 inhibitor LY311727 did not alter the activity of SPRL-Dox [35]. This suggested that the efficacy of SPRL is independent of sPLA 2 enzymatic activity. This result had only been shown in prostate cancer cells. Thus, we tested the ability of the sPLA 2 inhibitor varespladib to alter the ability of SPRL-Dox to decrease MTT staining in breast cancer cells. Varespladib is more potent and selective for Group IIA sPLA 2 than LY311727. Similar to our previous studies using the LY311727 inhibitor, varespladib pretreatment did not alter the activity of SPRL-Dox in any cell line tested, which included those with the highest expression of Group IIA sPLA 2 (Supplementary Figure S3B-D).

Discussion
PAK-1 is an attractive therapeutic target in various cancers including breast cancer [7]. This protein is overexpressed in the early stages of breast cancer during the conversion of the normal epithelium to ductal carcinoma in situ (DCIS) [41]. Further, overexpression of PAK-1 induced malignant transformation of mammary cells in transgenic mouse models, in addition to other breast lesions such as ductal hyperplasia and formation of solid nodules [42].
There is a strong correlation between increased nuclear localization of PAK-1 and resistance to the anti-estrogen tamoxifen in breast cancer [43]. PAK-1 phosphorylates the estrogen receptor alpha (ERα), correlating to increased ER receptor expression and resistance to tamoxifen [43][44][45]. This suggests that PAK-1 inhibition may be an effective strategy to overcome tamoxifen resistance in breast cancer [43,46].
Alteration of PAK-1 expression has also been associated with hormone receptor-positive breast cancer cells and increased PAK-1 expression was associated with lymph node metastasis [47,48]. Studies using RNA interference targeting PAK-1 in breast cancer cells revealed major roles of PAK-1 in cell survival and transformation [49]. Consistent with these findings, studies using transgenic mice models showed a role for overexpressed PAK-1 in the paraneoplastic and breast carcinoma

Discussion
PAK-1 is an attractive therapeutic target in various cancers including breast cancer [7]. This protein is overexpressed in the early stages of breast cancer during the conversion of the normal epithelium to ductal carcinoma in situ (DCIS) [41]. Further, overexpression of PAK-1 induced malignant transformation of mammary cells in transgenic mouse models, in addition to other breast lesions such as ductal hyperplasia and formation of solid nodules [42].
There is a strong correlation between increased nuclear localization of PAK-1 and resistance to the anti-estrogen tamoxifen in breast cancer [43]. PAK-1 phosphorylates the estrogen receptor alpha (ERα), correlating to increased ER receptor expression and resistance to tamoxifen [43][44][45]. This suggests that PAK-1 inhibition may be an effective strategy to overcome tamoxifen resistance in breast cancer [43,46].
Alteration of PAK-1 expression has also been associated with hormone receptor-positive breast cancer cells and increased PAK-1 expression was associated with lymph node metastasis [47,48]. Studies using RNA interference targeting PAK-1 in breast cancer cells revealed major roles of PAK-1 in cell survival and transformation [49]. Consistent with these findings, studies using transgenic mice models showed a role for overexpressed PAK-1 in the paraneoplastic and breast carcinoma transformations [42]. In addition, PAK-1 has been suggested to be involved in the regulation of apoptosis, and it is suggested that activation and overexpression of PAK-1 protect against chemotherapeutically induced cell death [19,[50][51][52][53].
Even though studies have suggested that IPA-3 may be a viable therapeutic option for cancers with altered PAK-1 expression, IPA-3 has limitations due to its poor stability and efficacy in vivo [24]. We previously addressed this limitation in prostate cancer cells using SSL-IPA-3 [32] but did not test this formulation in other cancer cell models. Our data demonstrate that both SSL-IPA-3 and SPRL-IPA-3 induce cell death in metastatic TNBC cells, which extends the possible utility of these liposomes. Previous studies from our laboratories have already demonstrated that IPA-3 exposure increased the expression and activity of apoptotic proteins such as caspase-3 and -9 [24]. We have also demonstrated the ability of both free IPA-3 and liposome-encapsulated IPA-3 to induce apoptosis as shown by TUNEL staining [32]. It is worthy to note that MDA-MB-435 has been used for many years as a metastatic breast cancer model; however, some recent genetic profiling of this cell line identified it as a melanoma cell instead [54].
The expression of sPLA 2 and, in particular, Group IIA sPLA 2 has been suggested as an independent prognostic factor for disease recurrence and death in human breast cancer [55]. The only study that could be found investigating the expression of Group IIA sPLA 2 in breast cancer patients did show expression in both primary and metastatic tumors, but no correlation between clinical stages. Data in this study not only demonstrated that Group IIA was expressed in all breast cancer cells tested but also revealed increased expression in metastatic TNBC cells.
We previously reported that SPRLs containing doxorubicin (SPRL-Dox) were more effective at inhibiting prostate tumor growth as compared to SSLs containing doxorubicin [35,36]. This finding suggested that SPRLs, which were engineered to take advantage of higher expression of Group IIA sPLA 2 in prostate cancer cells, are more efficacious than their SSL counterpart. However, SPRLs have never been validated in non-prostate cancer cell lines. As such, we tested the ability of SPRL-IPA-3 to alter breast cancer cell viability and compared their efficacy to SSL-IPA-3.
SPRL-IPA-3 demonstrated increased efficacy in cells derived from TNBC as compared to SSL-IPA-3. The only difference between these two formulations was an increase in DSPE in SPRL-IPA-3. This suggests that the increased anionic lipids allow for increased association with Group IIA sPLA 2 . It is unlikely that the increased efficacy of SPRL-IPA-3 was mediated by Group IIA sPLA 2 activity, as varespladib, a selective Group IIA sPLA 2 inhibitor, did not alter the efficacy of SPRL encapsulated doxorubicin. This is similar to data derived in prostate cancer cells in which LY311727, a non-selective sPLA 2 inhibitor, also had no effect [35].
Our previous data showed that SSL and SPRL containing doxorubicin demonstrated equal potency in vitro against a panel of prostate cancer cells that had a differential expression of Group IIA sPLA 2 ; however, SPRL were 2-fold more effective at inhibiting growth in xenograft mouse models of prostate cancer [35]. One reason for the differences between these two studies was that previous formulations contained doxorubicin. This hypothesis is supported by data in Supplementary Figure S3 demonstrating increased efficacy of SPRL-Dox in MDA-MB-468 (high sPLA 2 expression) cells as compared to BT-474 (low sPLA 2 expression). A direct comparison of SPRL-IPA-3 and SPRL-Dox is not practical given the different molecular targets of these drugs; however, these data suggest that both formulations may be viable therapeutics for the treatment of triple-negative breast cancer. Further, these studies are the first to test the efficacy of either formulation in breast cancer cells. Phospholipids used in liposomes (Table 1) were purchased from Avanti Polar Lipids, Inc. (Alabaster, AL, USA). IPA-3 was obtained from Tocris Bioscience (Bristol, UK). Cholesterol, MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide], and annexin V/PI kit were purchased from ThermoFisher Scientific (Waltham, MA, USA). All other chemicals were obtained from Fisher Scientific (Pittsburgh, PA, USA). Liposomes were prepared as described previously [32,35,36]. Liposome compositions are shown in Table 1. Briefly, Cholesterol (5 µmol/mL), phospholipids including DSPC (8 or 9 µmol/mL), DSPE-PEG (1 µmol/mL) in chloroform, DSPE (1 µmol/mL) in chloroform, and IPA-3 (4 µmol/mL in ethanol) were mixed, and organic solvents were evaporated under vacuum using a rotary evaporator (Buchi Labortechnik AG, Postlfach, Switzerland). The formed thin films were hydrated and suspended in phosphate saline buffer (PBS) followed by five cycles of freeze-thaw cycles and then high-pressure extrusion through a Lipex extruder repeated at least five times (Northern Lipids, Inc., Burnaby, BC, Canada) using double-stacked polycarbonate membranes (80 nm, GE Osmonics, Trevose, PA, USA). Dialysis in 10% (w/v) sucrose overnight was conducted to eliminate unencapsulated IPA-3 and lipids. A dynamic light scattering particle size analyzer was used to determine the size of liposomes (Zetasizer Nano ZS, Malvern Instruments, Enigma Business Park, Grovewood Road, Malvern, Worcestershire, UK). The size of the liposomes was further confirmed using tandem electron microscopy (TEM) imaging (Georgia Electron Microscopy, Athens, GA USA).

In Vitro Activity of Liposomal IPA-3 on MTT Staining
The efficacy of free and liposomal IPA-3 was determined in four cell lines representing different stages of breast cancer (MCF-7, MDA-MB-231, MDA-MB-468, and MDA-MB-435) using the cellular metabolic activity MTT assay [56]. Cells were seeded and then treated with free IPA-3, SSL-IPA-3 for 24 h. DMSO (vehicle for IPA-3) and empty liposomes (vehicles for encapsulated IPA-3) were used as controls to treat the different cell lines. The treated cells were incubated for 24, 48, and 72 h. MTT was added at each time point at a final concentration of 0.25 mg/mL and plates were kept in a 5% CO 2 incubator at 37 • C for 2 h. Plates were then shaken for 15 min and absorbance of each well including control and blank wells was measured at 590 nm using a Spectra Max M2 plate reader (BMG Lab Technologies, Inc., Durham, NC, USA).

In Vitro Cytotoxicity of Liposomal IPA-3 as Assessed by Annexin V/PI Staining
Annexin V and propidium iodide (PI) staining was assessed using flow cytometry to confirm MTT staining and to assess the mechanism of cell death. The method used was as previously described [32,57]. Briefly, cells were seeded and allowed to grow for 24 h prior to treatment with free IPA-3, SSL-IPA-3, SPRL-IPA-3, or empty liposomes. After 48 h, cells were collected, washed with PBS, and then stained with Alexa Fluor ® annexin V-FITC and PI (100 µg/mL) for 15 min according to the manufacturer's protocol. Annexin V and PI staining were quantified using a Dako Cyan ADP 9 color flow cytometer (Beckman Coulter, Inc., Miami, FL, USA). For each measurement 20,000 events (cells) were counted.

Immunoblot Analysis
Cell lysates from different cell lines were collected in RIPA buffer, which contained a protease inhibitor cocktail (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). The concentration of proteins in different samples was determined using the BCA assay. Cell lysates were first separated using gel electrophoresis and then transferred to nitrocellulose membranes and blocked for 2 h. The nitrocellulose membranes were incubated with a rabbit sPLA 2 IIA antibody (Cell Signaling Technology, Danvers, MA, USA) at a dilution of 1:500 in TBS-T with 1% (w/v) BSA overnight. The antibody against GAPDH (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) was used at a dilution of 1:4000 in 1% (w/v) BSA in TBS-T for 1 h. Membranes were then incubated with the appropriate peroxidase-conjugated secondary antibody (Promega, Madison, WI, USA) used at a dilution of 1:2500. Membranes were then washed with TBS-T three times for 10 min. Bands were developed using chemiluminescent substrates for horseradish peroxidase (Thermo Scientific, Waltham, MA, USA) and visualized using a Fluorchem SP digital imager (Alpha Innotech, San Leandro, CA, USA). Densitometry to quantify immunoblot bands was performed using the National Institutes of Health Image J software (ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA).

Proteolytic Digestion of Cell Surface Proteins
Metastatic TNBC cells (MDA-MB-231, MDA-MB-468, and MDA-MB-435) were treated with pronase, which is reported to induce the cleavage of membrane proteins [58]. Briefly, subconfluent cell monolayers were incubated with 0.1% pronase in serum-free media for 20 min at 37 • C. Cells were then collected as single-cell suspensions and washed in PBS by centrifugation at 800× g for 5 min at 4 • C. Cell viability following pronase treatment was checked under a bright field microscope. Cells were then prepared for immunoblot analysis. Cells not treated with pronase were used as control.

Statistical Analysis
All experiments were repeated at least three times (n = 3) and cells were isolated from three different passages. The average of all replicates ± SEM are shown. Data with confirmed Gaussian distribution were compared using unpaired Student's t-test. A nonparametric test such as the Mann-Whitney test was used if data did not have Gaussian distribution using GraphPad Prism software (La Jolla, CA, USA). The significance level (alpha) was set at 0.05 (marked with symbols (*)).

Conclusions
These data demonstrate the novel finding that a liposome formulation designed to target PAK-1 and Group IIA sPLA 2 demonstrated increased efficacy against TNBC cells as compared to cells derived from non-metastatic breast cancer tissues. These data also suggest that the efficacy of the Group IIA sPLA 2 responsive liposomes (SPRL-IPA-3) is independent of the enzyme activity and that the efficacy of these liposomes is dependent on multiple factors, including the encapsulated drug and the expression of PAK-1 (the target of IPA-3) and/or Group IIA sPLA 2 (the target of SPRL). These data provide additional information regarding the molecular determinants of liposome efficacy and suggest that developing liposomes that have dual-target proteins known to be expressed in metastatic cancer can increase efficacy and identify novel therapeutics.

Funding:
The work was primarily funded by the Department of Defense Prostate Cancer Research Program Idea Development Award (PC150431 GRANT11996600) to P.R.S. and B.S.C. Partial financial support was also provided by the NHLBI grant R01HL103952, NCATS grant UL1TR002378, Wilson Pharmacy Foundation (intramural), and Translational Research Initiative grant (intramural) to P.R.S. The funders had no role in the study design, data collection, analysis, and decision to publish the data. The contents of the manuscript do not represent the views of the Department of Veteran Affairs or the United States Government.

Conflicts of Interest:
The authors declare no conflict of interest.