Mimetic sHDL Nanoparticles: a Novel Drug-Delivery Strategy to Target Triple-Negative Breast Cancer

Abstract

Introduction:

Withanolides are naturally-derived HSP90 inhibitors that are potent in preclinical models of triple negative breast cancers (TNBC). Conjugation to synthetic. high-density lipoprotein (sHDL) nanoparticles improves solubility and targets delivery to the scavenger receptor B1 (SR-B1). TNBCs highly overexpress SR-B1 and we hypothesize that encapsulation of the novel withanolide WGA-TA by sHDL will have enhanced efficacy against TNBC in vivo.

Methods:

Validated human TNBC cell-lines were evaluated for SR-B1 expression by quantitative PCR (qPCR) and Western blot (WB). WGA-TA IC₅₀ values were obtained using CellTiter-Glo assays. SR-B1-mediated drug uptake was evaluated in vitro under fluorescence microscopy, and in vivo with IVIS imaging of mouse xenografts (MD-MBA-468LN). To evaluate drug efficacy, mice were treated with sHDL alone, WGA-TA alone, WGA-TA-sHDL, and chemotherapy or PBS (control).

Results:

TNBC cell lines had greater SR-B1 expression by qPCR and WB vs. controls. Fluorescent-labeled sHDL uptake was SR-B1 mediated in vitro, and in vivo tumor uptake using IVIS imaging demonstrated significantly increased tumor radiant efficiency vs. control. IC₅₀ for WGA-TA treated cells with or without sHDL encapsulation were 70–200 fold more potent than sHDL alone. In TNBC mouse xenografts, treatment with sHDL-WGA-TA resulted in a 54% decrease in tumor volume compared to control or sHDL alone.

Conclusion:

sHDL-WGA-TA nanoconjugates are potent against TNBCs and show improved SR-B1 receptor-mediated targeting. Treatment with sHDL encapsulated WGA-TA is able to significantly decrease growth of tumor in mice compared to control and has better efficacy than the current standard of care, warranting further evaluation as a novel therapeutic agent.

Introduction

There are more than 271,000 new diagnoses of invasive breast cancer predicted for 2019, making breast cancer the most commonly diagnosed cancer and the second leading cause of cancer-related death in women in the US (1). Substantial strides in screening, management of local tumor, adjuvant systemic therapy, and hormone/receptor therapies have resulted in decreasing the 5-year mortality rates over the last three decades (2). Triple negative breast cancers (TNBCs) lack expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER-2/neu), creating a more aggressive subset of tumors that remain a challenge for clinicians to treat. These TNBCs account for ~10–20% of invasive breast cancers (3, 4), and are associated with a more aggressive, basal-like subtype histopathology and BRCA1-related breast cancers (5). The lack of ER, PR, and HER-2/neu renders hormonal and anti-HER-2/neu targeted-therapies essentially ineffective. Although TNBC has demonstrated increased chemosensitivity compared to ER-positive breast cancer, only 30–45% receive a complete pathologic response due to the presence of residual disease after treatment (6). An optimal selective systemic treatment thus far has not been identified, and first-line chemotherapy is based on combination therapy with anthracycline and taxane combinations, followed by capecitabine, cyclophosphamide, or more recently platinum-based compounds at the time of disease progression (6–8).

Limited systemic treatments combined with the more aggressive TNBC phenotype results in decreased overall survival for patients with TNBCs compared to those with other types of breast cancer (4, 5). This problem highlights the need for the development of novel and efficacious therapies targeting these tumors. Although ER, PR, and HER-2/neu are not viable targets in TNBC, continued targeted research and molecular profiling has demonstrated promise in poly-ADP ribose polymerase (PARP) inhibitors, receptor tyrosine kinase inhibitors, mitogen-activated protein kinase inhibitors, phosphoinositide 3-kinase (PI3K)/Akt/mechanistic target of rapamycin (mTOR) inhibitors, heat shock protein 90 (HSP90) inhibitors, and Janus kinase/signal transducers and activators of transcription (JAK/STAT) inhibitors to name a few (6, 8). While each of these holds potential as a future TNBC therapeutic, they remain in phase I/II/III clinical trials.

We demonstrated previously that withanolides are potent, novel inhibitors of HSP90 derived from natural plant sources in the Solanacea family and contain a 28-carbon steroidal lactone structure. While these compounds have been shown in vivo to be highly efficacious in breast cancers (9), challenges remain in their clinical translation due to their hydrophobicity, poor solubility in the plasma, and a circulation half-life of only 2–3 h. Our group previously mitigated these clinical limitations through developing a novel`technique of nanoparticle drug delivery by encapsulating the poorly soluble withanolide, withalongolide A 4,19,27-triacetate (WGA-TA), with synthetic high density lipoprotein (sHDL) (10). These 8–12 nM nanoparticles conferred improved circulation half-life and plasma solubility, and sHDL has been demonstrated as safe and well tolerated with minimal adverse reactions in Phase III clinical trials (11). HDL is naturally metabolized in the liver and binds to the scavenger receptor B1 (SR-B1), which mediates the uptake of selective HDL cholesteryl esters into the liver. Importantly, SR-B1 is increasingly recognized as a biomarker and potential therapeutic target of multiple cancers including TNBC (12–14). BecauseSR-B1 provides a target receptor for sHDL nanoparticles to bind, this facilitates uptake of cholesterol esters and anti-cancer drugs into the cytosol via a non-endocytic pathway, resulting in delivery of the encapsulated payload (13, 15). We hypothesized, therefore, that sHDL encapsulation of WGA-TA would specifically target high SR-B1-expressing TNBC cell lines resulting in enhanced delivery of the WGA-TA payload in vitro and in vivo.

Materials and Methods

Cell Culture/Reagents:

The human normal female fibroblast (FF) and breast cancer cell lines SUM159, MDA-MB-231, MDA-MB-468LN (triple ER/PR and HER-2/neu negative), T47D, and MCF7 (both luminal A, ER⁺) were validated by standard DNA fingerprinting techniques and 2-D cultured in high glucose Dulbecco’s modified Eagle medium (DMEM) (Life Technologies, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich, St. Lewis, MO), 100 U/mL penicillin and 100 µg/mL streptomycin (Life Technologies,) according to our labs previously published methods (16). JURKAT cells were cultured in RPMI 1640 (Thermo Fisher Scientific, Waltham, MA) with 10% FBS and penicillin/streptomycin. NCI-H295R cells were cultured in 1:1 DMEM:F12 nutrient mixture (Thermo Fisher Scientific) supplemented with 10% FBS, penicillin/streptomycin, and ITS (Thermo Fisher Scientific). Drug compounds utilized for these experiments included the novel HSP90/cdc37 inhibitor withalongolide A 4,19,27-triacetate (WGA-TA) which was isolated and purified in the laboratory of Dr. Barbara Timmermann (University of Kansas, Lawrence, KS), and synthetic high density lipoprotein (sHDL) nanoparticles which were created in the laboratory of Dr. Schwendeman (University of Michigan, Ann Arbor, MI). The withanolide WGA-TA, a dialkylcarbocyanine lipophilic tracer dye (DiO), and the fluorescent dye 1,1’-Dioctadecyl-3,3,3’,3’ Tetramethylindotricarbocyanine Iodide (DiR) were loaded into sHDL nanoparticles using methods as described previously (10) with the addition of DiO/DiR dye or WGA-TA during sHDL preparation. Control DiR-loaded liposomes (DiR-L) were created by dissolving 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), and DiR in chloroform, which was removed subsequently by nitrogen flow and a vacuum oven to obtain the lipid film. The lipid film was hydrated with phosphate buffered saline (PBS, pH 7.4) by bath and probe sonication to obtain DiR-L, which were diluted to 20 g/mL by PBS for in vivo IVIS Spectrum imaging.

Cell Proliferation Assay:

Equal numbers of FF, SUM159, MDA-MB-231, and MDA-MB-468LN cells were plated and left to adhere for 24 h in opaque, walled 96-well plates. Cells were then treated with serial concentrations of the study drugs sHDL, WGA-TA, and WGA-TA-sHDL at varying concentrations over 5 orders of magnitude. After 72 h of treatment, cell viability was assessed through ATP quantification using the CellTiter-Glo luminescent cell viability assay (Promega, Madison, WI) according to the manufacturer’s instructions. Briefly, CellTiter-Glo reagent was added to each well, mixed, and incubated at room temperature for 10 min. Luminescence was then read on a BioTek Synergy Neo plate reader using the Gen 52.01 software (BioTek, Winooski, VT). The IC₅₀ values for sHDL, WGA-TA, and WGA-TA-sHDL in each of the cell lines were calculated using dose-response curves in GraphPad Prism (GraphPad Software, Inc., La Jolla, CA).

Quantitative Real Time Polymerase Chain Reaction (qPCR) and Western Blot Analysis for SR-B1 Expression:

RNA from TNBC (SUM159, MDA-MB-231, MDA-MB-468LN), luminal A, ER⁺ (MCF7 and T47D), negative controls (FF and Jurkat), and positive control (NCI-H295R) cell lines were extracted using Qiagen RNA isolation kit (Qiagen Sciences, Valencia, CA). Approximately 500 ng of RNA was reverse transcribed using a SuperScript Reverse Transcriptase kit from Life Technologies (Thermo Fischer Scientific). Quantitative polymerase chain reaction (PCR) was performed in a StepOne real-time PCR (qPCR) system (Thermo Fischer Scientific) using SR-B1 and actin-specific primer sets. Relative gene expression levels were calculated after normalization with internal controls. Expression levels of SR-B1 were confirmed with Western blot analysis per our previously published methods (16). Briefly, equal amounts of protein were loaded and separated using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), transferred to nitrocellulose membrane, and blotted. We obtained SR-B1, donkey anti-rabbit IgG HRP (1:5000) and goat anti-mouse IgG HRP (1:5000) secondary antibodies from Santa Cruz Biotechnology (Santa Cruz, CA) and β-Actin from EMD Millipore (Billerica, MA). Actin levels were assessed to ensure equal loading and transfer of proteins. ImageJ software (National Institutes of Health, Bethesda, MD) was used to obtain density of protein bands and results are described as normalized to actin expression.

siRNA knockdown of SR-B1 in SUM159 TNBC cells:

Validated SR-B1 siRNA as well as non-targeting siRNA were transfected into SUM159 cells using Lipofectamine 2000 as per the manufacturer protocols. Then, 48 h post transfection the cells were either collected for Western blot analysis or treated with sHDL-DiO for 4 h for subsequent uptake studies.

In Vitro Uptake of DiO-Labeled sHDL Nanoparticle:

The TNBC cells lines SUM159, MDA-MB-231, MDA-MB-468LN, the negative control FF cells, and positive control NCI-H295R cells were incubated for 4 h with sHDL nanoparticles labeled with long-chain dialkylcarbocyanine lipophilic tracer DiO (Invitrogen, NY). To assess SR-B1 mediated uptake, cells were pre-treated for 1 h with excess blank sHDL prior to the addition of sHDL-DiO nanoparticles. Cells were fixed in paraformadehyde, nuclei stained with 4’, 6-diamidino-2-phenylindole (DAPI), and fluorescent images obtained using Nikon fluorescent microscope.

In Vivo Distribution of DiR-Labeled sHDL Nanoparticle in MDA-MB468LN Xenograft:

Three million MDA-MB-468LN cells in 100 L PBS were injected into the right posterior flank of athymic nude mice ELSEVIER ASK THE AUTHORS WHERE THE MICE WERE OBTAINED). Once tumors reached 5–8 mm in diameter (approximately 6 weeks), mice were injected intra-peritoneally with 0.6% DiR-labeled sHDL nanoparticle. After 24 h, whole body and organ fluorescent imaging and quantification of organs was performed using the Xenogen IVIS Spectrum imaging system (Caliper Life Sciences, Hopkinton, MA) at the University of Michigan Center for Molecular Imaging core facility.

In Vivo Evaluation of Drug Efficacy:

Three million MDA-MB-468LN cells in 100 L PBS were injected into the left mammary fat pad of athymic nude mice. Once the tumor size reached an approximate volume of 50mm³, the mice were divided randomly into treatment groups of 4 mice each (PBS/control, sHDL, WGA-TA, WGA-TA-sHDL, or standard of care chemotherapy) for 21 days. The treatment groups received daily intraperitoneal injection of the drugs (200 µL PBS, 5 mg/kg WGA-TA, 5 mg/kg WGA-TA-sHDL, or equivalent volume of sHDL). The mice in the chemotherapy group were treated with a combination of 2mg/kg doxorubicin and 100mg/kg cyclophosphamide once weekly. The mice were weighed thrice weekly to assess general toxicity of the treatment, and tumor volumes were measured twice weekly with a digital caliper. Tumor volume was calculated using the formula: length x (width)² /2. Post-treatment survival of the animals was followed for an additional 11 days at which point all mice were killed due to control mice reaching pre-determined endpoints of tumor burden. All of the experiments involving mice were carried out in compliance with the University Committee on Use and Care of Animals at the University of Michigan.

Statistical Analysis:

Significance (set as 95%, p<0.05) was determined using two-way ANOVA and Student’s t-test. Best-fit normalized, dose-response curves for sHDL, WGA-TA, and WGA-TA-sHDL were used to calculate IC₅₀ values with 95% confidence intervals (CI) in GraphPad Prism (GraphPad Software, Inc., La Jolla, CA). Data are presented as mean values with error bars denoting standard deviation. Experiments were replicated in triplicate to ensure accuracy. Bio-statistical data points were determined throughout the in vivo study, and tumor volumes and animal masses were calculated as mean ± standard error. Treatment differences with respect to tumor size were assessed via 2-tailed Student’s t-test. P-values less than 0.05 in t-test were considered statistically significant. The tumor volume progression/regression were plotted using GraphPad Prism software.

Results

Scavenger Receptor B1 (SR-B1) is Highly Expressed on Breast Cancer Cells:

Synthetic HDL nanoparticles are a natural ligand to the SR-B1 receptor. Increased expression of SR-B1 in target cancer cells is important for sHDL-drug conjugates to target these cells effectively and preferentially. We demonstrated that the TNBC cell lines SUM159, MDA-MB-231, and MDA-MB-468LN, as well as several luminal A, ER⁺ cell lines (MCF7 and T47D) have increased expression of SR-B1 mRNA and protein compared to normal FF and Jurkhat cells. Using qPCR compared to normal FF cells (controls), SR-B1 mRNA expression was increased 3.2-fold in SUM159, 1.7-fold in MDA-MB-231, 9-fold in MDA-MB-468LN, 1.7 fold in MCF7, and 26 fold in T47D (p<0.05 each). mRNA expression for NCI-H295R cells, which are known to express high levels of SR-B1, was significantly increased 63.6-fold compared to FF cells and is shown as a positive control (Figure 1A). This increased SR-B1 expression was confirmed by Western blot analysis, which demonstrated statistically significantly increased SR-B1 protein levels in SUM159, MDA-MB-468LN, and MCF7 cell lines with a trend toward significance for MDA-MB-231, and T47D (ELSEVIER HAVE THE AUTHORS PUT IN THE P VALUES IN THIS SENTENCE; IF THE P VALUE IS >0.15 THEN THE SENTENCE WILL NEED TO BE CHANGED TO SAY THERE WAS NO DIFFERENCE IN THESE LAST TWO CELL LINES)) (Figure 1B and 1C).

Figure 1A.

Quantitative polymerase chase reaction (qPCR) levels of scavenger receptor B1 (SR-B1) mRNA in several breast cancer cell lines, including luminal A, ER⁺ cell lines and triple negative breast cancer (TNBC) cells. Compared to normal female fibroblast (FF) cells, there are 1.7 and 26.0 fold increased in mRNA expression for MCF7 and T47D, respectively, and 3.2, 1.7, and 9.0 fold increased mRNA expression in TNBC cells lines SUM159, MDA-MB-231, and MDA-MB-468LN, respectively (p<0.05) B. Western blot analysis of SR-B1 protein expression. Increased protein expression in MCF7, SUM159, and MDA-MB-468LN breast cancer cell lines. * indicates p<0.05. C. Densitometric analysis of SR-B1 protein expression relative to actin is shown.

SR-B1 Receptor Modulates Uptake of DiO-sHDL Conjugated Nanoparticles:

Scavenger receptor B1 functions to internalize HDL during normal metabolism of cholesterol. In order to assess the role of SR-B1 in mediating the uptake of the sHDL nanoparticle cargo, we used DiO dye conjugated to sHDL nanoparticles. Positive control (high SR-B1 expressing NCI-H295R cells), negative control (FF cells), and TNBC cell lines SUM159, MDA-MB-231 and MDA-MB-468LN were incubated with DiO-sHDL for 4 hours, with or without 1 hour of excess sHDL pre-treatment. The uptake of the DiO dye was then captured by fluorescent microscopy (Figure 2A). The amount of DiO uptake into the cells correlated with the levels of SR-B1 expression identified in Figure 1 with minimal uptake of the green DiO dye in FF cells, increased uptake in highly SR-B1 expressing NCI-H295R cells, and uptake in all three TNBC cell lines. Additionally, pre-treatment with excess sHDL in all the cell lines and siRNA against SR-B1 in SUM159 (Figure 2B) completely blocked DiO uptake into the cells, which indicates that the SR-B1 receptor likely plays a key role in modulating the internalization of the sHDL nanoparticles and their cargo. Knock down of SRB1 protein was verified using immune blot analysis (Figure 2C)

Figure 2A.

TNBC cell lines were treated for 4 h with DiO-sHDL conjugated nanoparticles with or without 1 h of excess sHDL pretreatment. There was minimal uptake of DiO dye in FF cells, whereas there was increased uptake in all three TNBC cell lines. This uptake was blocked completely by sHDL pretreatment. Level of DiO uptake corresponds to the level of SR-B1 expression shown in Figure 1. B. Pre-treatment with small interfering RNA (siRNA) to SR-B1 in SUM159 cells also completely abolished DiO uptake C. siRNA against SR-B1. SR-B1 expression is abolished completely with siRNA treatment.

Cell Viability/Proliferation:

The effect of sHDL and WGA-TA alone or after WGA-TA encapsulation with sHDL on cell viability was determined using CellTiter-Glo luminescent cell viability assays. The half maximal inhibitory concentration (IC₅₀) was determined from dose-response curves (Figure 3) and reported in Table I. Treatment with sHDL alone had minor effects on cell viability, with an IC₅₀ value >24 µM for FF cells, and ranging from 7.1–20. 5 µM for TNBC cells. Treatment with WGA-TA was highly potent, with TNBC IC₅₀ values ranging from 10.3–68 nM. This drug effect in the TNBC cells demonstrated a greater than 5-fold increased potency over treatment in FF cells. While encapsulation of WGA-TA with the mimetic sHDL slightly increased the IC₅₀ values across all cells lines, the drug still maintained statistically significant potency in TNBCs compared to FF cells.

Figure 3.

Dose-response curves of normal female fibroblasts (FF) and the three TNBC cell lines. The half maximal inhibitory concentration (IC50) was determined using a best-fit curve. Table I reports the IC₅₀ values, in which WGA-TA treated cells had IC₅₀ concentrations ranging from 10 to 121 nM that were 70–200 fold more potent than cells treated with sHDL alone (IC₅₀ 7 to 24 µM). TNBCs were 5 times more sensitive to WGA-TA tr eatment than normal fibroblasts.

TABLE I:

Antiproliferative Activity of sHDL, WGA-TA and WGA-TA-sHDL Against Human Triple Negative Breast Cancer Cells

Cell Line	sHDL	WGA-TA	WGA-TA-sHDL
FF	>24.4 µM	370 nM (337–406)	438 nM (396–484)
SUM159	7.1 µM (5.4–9.2)	10.3 nM (9.3–11.5)	32 nM (2 9.7–34.6)
MDA-MB-231	12.6 µM (11.8–13.6)	68.0 nM (59.9–77.2)	121 nM (115–128)
MDA-MB-468LN	20.5 µM (17.3–24.3)	27.8 nM (23.1–33.5 )	96 nM (88.5–104)

▪Results shown as the half maximal inhibitory concentration (IC₅₀) with 95% confidence intervals

Increased In Vivo TNBC Xenograft Uptake of DiR-sHDL Nanoparticles:

We next examined the whole body and organ distribution of sHDL nanoparticle after treatment using MDA-MB-468LN xenografts in athymic nude mice. Conjugated DiR-sHDL was injected intra-peritoneally, and the distribution of DiR was determined after 24 h using IVIS Spectrum imaging. Control mice were injected with non-specific DiR-liposome. The whole body distribution (Figure 4A) identified increased uptake of DiR-sHDL in the MDA-MB468LN tumors compared to controls at 24 h. After the animals were killed, a necroscopy performed,, and individual organ imaging of brain, heart, lung, spleen, kidneys, liver and tumor was performed (Figure 4B), there was a significant increase in DiR uptake in the high SR-B1 expressing MDA-MB-468LN tumors compared to controls (p<0.001). While there was also increased uptake in the heart, lung, and kidney, this uoptake was relatively minor. Both control and DiR-sHDL mice had increased uptake in the spleen and liver, which is expected, because the reticuloendothelial system (RES) is a known aggregation site for nanoparticles, and the liver naturally expresses SR-B1 for the metabolism of cholesterol.

Figure 4A.

Whole body and organ distribution of DiR-sHDL nanoparticle in a MDA-MB-468LN mouse xenograft model. Mice were injected intra-peritoneally then imaged 24 h later using IVIS Spectrum. Control mice received injection of DiR-liposome (DiR-L) B. Graph of average radiant efficiencies of brain, heart, lung, spleen, kidneys, liver, and tumor. There were undetectable levels in all control organs except for spleen and liver. There is a statistically significant increase in DiR uptake in the MDA-MB-468LN tumors, as well as a more limited uptake in the heart, lung, and kidney of DiR-sHDL mice.

WGA-TA-sHDL is Effective at Suppressing Breast Cancer Tumor Growth In Vivo:

Approximately 3 million MDA-MB-468LN cells were injected into the mammary fat pad of nude mice. Once the tumors reached 50 mm³, the mice were treated with PBS (control), sHDL, WGA-TA, WGA-TA-sHDL, or standard of care chemotherapy (doxorubicin plus cyclophosphamide) for 21 days. As shown in Figure 5, at the end of the treatment period, tumor volumes for mice treated with sHDL and chemotherapy were equivalent to that of control mice, while tumor volumes for mice treated with WGA-TA and WGA-TA-sHDL were 59% (p=0.02) and 50% (p=0.04), respectively, that of control mice. At the end of the study (11 days after the end of treatment), tumor volumes for mice treated with sHDL were again equivalent to that of control mice, while tumor volumes for mice treated with WGA-TA, WGA-TA-sHDL, and chemotherapy were 80% (p=0.27), 46% (p=0.03), and 69% (p=0.10), respectively, that of control mice. When comparing the different treatment groups, tumor volumes for mice treated with WGA-TA-sHDL were smaller than tumors treated with sHDL alone (p=0.05) and trended toward being s smaller compared to tumors treated with WGA-TA alone (p=0.06). Thetumor volumes were no different with WGA-TA-sHDL treatment compared to chemotherapy (p=0.15).

Figure 5.

Approximately 3 million MDA-MB-468LN cells were injected in the mammary fat pad of nude mice; once the tumors reached approximately 50 mm³, the mice were treated for 21 days with PBS (control), sHDL, WGA-TA, WGA-TA-sHDL, or chemotherapy (doxorubicin plus cyclophosphamide). At the end of the study, tumor volumes for mice treated with sHDL, WGA-TA, WGA-TA-sHDL, and chemotherapy were 95% (p=0.82), 80% (p=0.27), 46% (p=0.03), and 69% (0.10), respectively, that of control mice.

Toxicity was assessed by evaluating average percent change in body weight. As seen in Figure 6, mice treated with WGA-TA and WGA-TA-sHDL experienced weight loss during the course of treatment; at the end of the treatment period, mice treated with WGA-TA and WGA-TA-sHDL experienced approximately 7% and 6% body weight loss, respectively (p<0.05 for WGA-TA compared to control mice).; however, after stopping injections, weights of all mice rebounded quickly, and there were no differences in weight in all treatment groups.

Figure 6.

Toxicity was assessed by evaluating average percent change in body weight. Mice treated with WGA-TA and WGA-TA-sHDL experienced weight loss during treatment; however, after stopping injections, weights of all mice regained weight rapidly no differences in weight in all treatment groups.

Discussion

TNBC remains a difficult malignancy to treat, because only 30–45% of patients treated with conventional systemic chemotherapy achieve complete pathologic responses, leaving the majority of patients with ongoing disease and poor clinical outcomes (4). Given the critical need for better treatments for this disease, there are now a number of promising new targeted therapies aimed at inhibiting one or more oncogenic or proliferative pathways in TNBC that are still in clinical trials,including new PARP inhibitors, HSP90 inhibitors, and checkpoint inhibitors (6).

A recent targeted strategy for TNBC is based on the identification of increased SR-B1 expression in TNBCs (13, 14). While HDL and its mimetic sHDL bind effectively to the SR-B1 receptor, this lipoprotein by itself is not potent against breast cancers. Applying the strategy of using sHDL to carry a drug payload to cells overexpressing SR-B1 is a therapeutic concept that has been shown effective in adrenocortical cancers (10). Therefore, we chose to use this same strategy in TNBCs, because WGA-TA is a potent inhibitor of these cells in vitro. The main difficulties in translating this natural withanolide in vivo have been its relatively hydrophobic nature that limits its solubility, as well as a short, 2–3 hour half-life in serum. To both improve the pharmacokinetic profile of WGA-TA and provide enhanced drug delivery to highly expressing SR-B1 TNBC tumors, our group has developed a method of encapsulating WGA-TA in sHDL nanoparticles (10). In this study, we show that encapsulation of WGA-TA in sHDL does not dramatically alter the drug potency, because the IC₅₀ values remained <121 nM in TNBC, still several fold more selective than normal fibroblast cells. The majority of this effect is likely representative of WGA-TA-sHDL uptake being influenced by levels of SR-B1 expression, and of the imperfect (85%) yield of WGA-TA during encapsulation in sHDL. As such, the WGA-TA concentration is likely 15–20% less in WGA-TA-sHDL than the unencapsulated WGA-TA drug, which would account for most of the effect of slightly greater WGA-TA-sHDL IC₅₀ values we observed.

SR-B1 receptor is not only a cancer biomarker but is also tied to the activation of several pro-survival and oncogenic pathways (14, 17). Expression of SR-B1 has been associated recently with invasive ductal breast cancer, more aggressive tumor behavior, and worse overall survival (18). These findings have set the stage for using synthetically customized HDL nanoparticles to target SR-B1 and deliver cytotoxic payloads in a number of malignancies, including breast, prostate, ovarian, hepatic, neuroblastoma, lymphoma, and adrenal cancers(10, 15, 19, 20). In breast cancer, paclitaxel loaded into discoidal, reconstituted, high density lipoproteins has demonstrated improved targeting, cytotoxicity, and limitation of tumor growth compared to unmodified paclitaxel (20). Loading small interfering RNA molecules into reconstituted HDL has also shown enhanced SR-B1 tumor-selective targeting, cytosolic delivery through a non-endocytotic mechanism to bypass endo-lysomal trapping, with enhanced anti-tumor effects in breast cancer (21). In our study, we demonstrated that TNBC has a 1.7 to 9.0 fold increase in SR-B1 expression compared to normal fibroblast cells, and that several luminal A, ER⁺ cell lines also have increased SR-B1 expression. By exploiting this increased expression over normal cells, sHDL encapsulation increased payload delivery to TNBC cells as demonstrated by increased dye uptake both in vitro and in vivo. This process is clearly SR-B1 mediated, because excess sHDL and siRNA pre-treatment effectively blocked this uptake. In MDA-MB-468LN xenografts, sHDL encapsulation increased tumor targeting compared to a non-specific, liposome drug delivery system.

In light of these findings, our study demonstrated the use of a synthetically developed HDL nanoparticle that was created to improve the pharmacokinetics of WGA-TA and to enhance targeting and delivery of cytotoxic drug payloads via SR-B1 expression in breast cancer. In vivo, we demonstrated that conjugation of WGA-TA with sHDL is able to decrease growth of the breast cancer after treatment for 21 days, an effect that appears durable after discontinuation of drug administration. Notably, without nanoparticle encapsulation of WGA-TA, WGA-TA alone is unable to deliver the same anti-tumor effect seen with WGA-TA-sHDL. Additionally, unlike in other cancer models such as nasopharyngeal cancer, the sHDL particle alone did not demonstrate any anti-tumor effect through SR-B1 targeting (12). WGA-TA-sHDL treatment resulted in superior efficacy compared to the current standard of care, which is combination cytotoxic chemotherapy with doxorubicin and cyclophosphamide. Although mice treated with WGA-TA and WGA-TA-sHDL experienced mild weight loss, they regained weight quickly after cessation of drug administration. Further steps in this investigation would include a large-scale translational animal study with patient-derived xenografts. Additionally, it would be exciting to evaluate possible synergistic effects of these agents in combination with standard of care chemotherapy drugs for treatment of TNBC.

Conclusion

In summary, withanolides have shown a remarkable therapeutic index with high potency in cancer cells. Their challenge in clinical development and application has been their hydrophobicity, poor solubility in serum, and rapid clearance with a half-life of only 2–3 hours. Conjugation of these withanolides with sHDL appears to solve this issue of solubility, while improving the half-life through its sustained-release effect (data not shown) and providing an improved tumor cell targeting delivery strategy through the SR-B1 receptor. The current study demonstrated that WGA-TA-sHDL conjugation is a potent, anti-cancer strategy in vitro and in vivo in TNBCs that targets these tumors selectively and effectively, demonstrating early promise for further translational evaluation.