Engineering lipid nanoparticles to deliver drug combinations to improve therapeutic efficacy in triple negative breast cancer

Abstract

Triple negative breast cancers (TNBCs) represent a heterogenous disease with high patient mortality relative to other breast cancers. Conventional chemotherapy, such as Doxorubicin (DOX), is widely used for TNBC treatment. However, it often accompanies severe side effects that limit the effective therapeutic dose for cancer treatment. The plasticity of cancer epigenetics makes them plausible candidates for therapeutic intervention. DNA methyltransferases (DNMTs) and histone deacetyltransferases (HDACs) are two of most well studied enzymes that regulate transcription and chromatin compaction. Aberrant expressions of DNMTs and HDACs have been implicated in a variety of cancers. A few DNMT inhibitors (DNMTi) and HDAC inhibitors (HDACi) have been approved for treating T-cell lymphoma, multiple myeoloma, and myelodysplastic syndromes. For instance, DNMTi Decitabine (DAC) and HDACi Panobinostat (PAN) were shown to reverse abnormal methylation of DNA and altered chromatin structure, respectively, leading to increased expression of tumor suppressor genes (TSGs) and decreased expression of oncogenes (OGs). However, epigenetic agents as a monotherapy did not show a therapeutic benefit against solid tumors. A targeted drug delivery vehicle may improve efficacy of epigenetic drugs. Increased synthesis of lysophosphatidic acid (LPA) and its receptors (LPAR1-3) are implicated in breast cancers. The LPA/LPAR1 axis has been linked to breast cancer metastasisboth in vitro and in vivo. Herein, we took advantage of elevated expression of LPAR1 in TNBC tissues to direct chemotherapeutics to breast cancer cells. DAC and PAN were delivered using LPAR1-targeted, lipid nanoemulsions (LNEs). Our results demonstrated that the cell uptake and in vivo biodistribution of LNEs were dependent on LPAR1 expression in TNBCs. DAC/PAN-LNEs were effective in inhibiting the growth of mesenchymal breast cancer cells by restoring CDH1 and suppressing FOXM1 expression. Epithelial breast cancer cells that inherently express low FOXM1 and high CDH1 were unaffected by DAC/PAN-LNEs. Overall, we successfully designed LPAR1-targeted LNEs that selectively act on CDH1(low)/FOXM1(high) TNBC cell lines. However, in vivo therapeutic efficacy of DAC/PAN-LNEs did not show significant tumor suppression in the time frame of the experiment. We then hypothesized that the combination of DOX with PAN may improve therapeutic efficacy by attacking abnormal DNA replication and dysregulated transcription landscape simultaneously. Thus, DOX was delivered in combination with PAN using previously designed LPAR1-targeted LNEs. DOX/PAN LNEs had a significantly higher uptake of DOX compared to DOX delivered by non-targeted formulations or free drug. DOX and PAN synergistically inhibited tumor cell growth in vitro and in vivo. Lastly, a PCR microarray was used to screen the change of oncogene and tumor suppressor gene expressions induced by LNEs encapsulating DOX and/or PAN. We identified that Dox/Pan LNE treatment upregulated a tumor suppressor gene that sequesters a ROS (reactive oxygen species) scavenging protein, leading to higher ROS accumulation in TNBCs. Invasiveness biomarkers, also found in papillary thyroid carcinoma, were also expressed in TNBC; the expression of these proteins were reversed upon Dox/Pan LNE treatment. Fine-tuning of the hydrophobic core of LNEs to improve the solubility of chemotherapeutics is challenging due to heterogeneity of fats in oil. To maximize the encapsulation of hydrophobic drugs in a lipid nanoparticle, a careful selection of matrix lipids are required. We developed a strategy to fabricate multi-drug delivery nanocarriers to overcome matrix-limited containment of more than one drug. First, solubility of model drugs, 3, 3'-diindolylmethane (DIM) and PAN, were measured in several lipids. Second, a solid lipid and a liquid lipid that have maximal affinity toward model drugs were screened. Third, dual-phase lipid nanocarriers (DPLNs) were fabricated and physiochemical properties were optimized. Lastly, cytotoxicity of DIM and PAN encapsulating DIM/PAN lipid nanoparticles (LNPs) were evaluated in TNBCs. Overall, we demonstrated the therapeutic efficacy and mechanism of LPAR1-targeted LNEs in vitro and in vivo. A strategy to develop DPLNs was suggested to facilitate the construction and optimization of a multi-drug delivery carrier to overcome matrix-limited solubility. (*) This work was supervised by Professor Debra Auguste