Comparative Gene Signature of (-)-Oleocanthal Formulation Treatments in Heterogeneous Triple Negative Breast Tumor Models: Oncological Therapeutic Target Insights

Abstract

Triple negative breast cancer (TNBC) heterogeneity and limited therapeutic options confer its phenotypic aggressiveness. The discovery of anti-TNBC natural products with valid molecular target(s) and defined pharmacodynamic profile would facilitate their therapeutic nutraceutical use by TNBC patients. The extra-virgin olive oil (EVOO) is a key Mediterranean diet ingredient. S-(-)-Oleocanthal (OC) leads the bioactive anti-tumor EVOO phenolic ingredients. A previous study reported the solid dispersion formulated OC with (+)-xylitol (OC-X) suppressed the in vivo progression and recurrence of the TNBC MDA-MB-231 cells. This study investigates the ability of OC-X formulation to suppress the in vivo heterogeneous BC initiation and progression utilizing advanced preclinical transgenic MMTV-PyVT and TNBC PDX mouse models. Furthermore, the clustering of the gene expression profiles in MMTV-PyVT and PDX mouse tumors treated with OC-X acquired by a Clariom S microarray analysis identified the distinctly affected genes. Several affected novel signature genes identified in response to OC-X treatments and proved overlapped in both mouse and human tumor models, shedding some lights toward understanding the OC anticancer molecular mechanism and assisting in predicting prospective clinical outcomes. This study provides molecular and preclinical evidences of OC-X potential as a nutraceutical suppressing heterogeneous TNBC model and offers preliminary gene-level therapeutic mechanistic insights.

Keywords: (−)-oleocanthal; MMTV-PyVT; PDX; TNBC; extra-virgin olive oil; microarray.

Figure 1

Schematic overview of in vivo studies in MMTV-PyVT and PDX mice. Mice were randomly divided into placebo control and OC-X treated groups. OC-X was dissolved in sterile water and administered by oral gavage 6×/week, at a dose equivalent to 7.5 mg/kg OC. MMTV-PyVT mice began treatments on the fourth week of age and ended at week 12, while the PDX mice began treatments on the eighth week of age and ended at week 11.

Figure 2

OC-X suppressed tumor growth and metastasis in the lungs in MMTV-PyVT transgenic mice. (A) Tumor volumes monitored for treatment and placebo control groups and measured over the 4th to 12th weeks of mice age. A caliper was used to measure individual tumors with the formula V = (W2 × L)/2, where V is the volume, W is the width, and L is the length of the tumor. The results show significant tumor suppression in OC-X treated mice (7.5 mg/kg OC by oral gavage, 6×week, n = 4 mice) versus the control group (n = 3 mice). (B) Comparison of the collective mean tumor weight collected from the glands of treatment and control groups after mice were sacrificed at the end of the experiment. (C) Comparison of the incidence of tumor development in OC-X treated versus control group mouse glands. OC-X treatments reduced the incidence of palpable lesions in most of the treated mouse glands (22 mammary glands out of 40 had tumors; 55%), leading to decreased mice tumor burden in the treatment group versus the control (28 mammary glands out of 30 had tumors; 93%). (D) The mean average body weights for each mice group were monitored over the experiment duration. No significant mice body weight change was observed between treated and control animals. (E) Comparison of representative tumor images of the OC-X treatment versus control group mice. Photographs are shown in two magnifications, 1× and 3×, highlighting the vascularity and morphological changes of mice tumors and lungs in both groups. H&E staining was utilized to compare treated and control mice tumor micromorphology using 40× magnification. p-value indicated by using Student’s t-test, * p < 0.05.

Figure 3

OC-X suppressed tumor growth in PDX mice. (A) Comparison of representative tumor images of the OC-X treatment versus the control group PDX mice (n = 5, each). (B) Tumor volumes were monitored for OC-X treated and control groups and measured over the 21 days of the course of the experiment. Tumor volume (V) was calculated using the formula V = L/2 × W2. (C) Comparison of collective mean tumor weight collected from the glands of treatment and control groups after mice were sacrificed at the experiment end. (D) The average body weights for each mice group were monitored over the experiment duration. No significant mice body weight change was observed between treated and control animals, regardless of the slight weight decrease for treated mice in the initial treatment days. The results are expressed as the mean SEM; p-value indicated by using Student’s t-test, ** p < 0.01.

Figure 4

Comparative microarray analysis of OC-X treatment effects in MMTV-PyVT transgenic and PDX mice tumors using Mice and Human Clariom S arrays, respectively, highlighted the differential gene expression and ontology (GO) results. Transcriptome analysis using the Affymetrix Transcriptome Analysis Console (TAC) filtered and assessed all 1.8-fold expression dysregulations with nominal significance, and then characterized all genes affected by OC-X treatments into up- and downregulated based on their global dataset expression in quantitative schematic representations. (A) MMTV-PyVT mouse tumor array analysis. (B) PDX mouse tumor array analysis. The majority of deregulated genes were categorized as coding and multiple complex genes, the most interrogated category. (C) Ingenuity pathway analysis (IPA) revealed the top downregulated canonical pathways, along with predicted activation scores. (D) OC-X treatments predicted the affected cellular functions and diseases with the lowest z-scores. Analyses focused on the cancer-related and overlapping pathways in both mouse models.

Figure 5

Association of differentially affected genes by OC-X treatments with TNBC progression stages (neoplasia, cell-to-cell adhesion signaling and interaction, invasion, migration, and metastasis) in MMTV-PyVT and PDX mouse tumors. (A) Affected genes in MMTV-PyVT transgenic mouse tumors. (B) Affected genes in PDX mouse tumors. IPA regulator and functional prediction analysis was based on negative (red bars) and positive (green bars) z-scores. Some common genes contribute to multiple TNBC progression stages due to its relevance in TNBC pathogenesis.

Figure 5

Association of differentially affected genes by OC-X treatments with TNBC progression stages (neoplasia, cell-to-cell adhesion signaling and interaction, invasion, migration, and metastasis) in MMTV-PyVT and PDX mouse tumors. (A) Affected genes in MMTV-PyVT transgenic mouse tumors. (B) Affected genes in PDX mouse tumors. IPA regulator and functional prediction analysis was based on negative (red bars) and positive (green bars) z-scores. Some common genes contribute to multiple TNBC progression stages due to its relevance in TNBC pathogenesis.

Figure 6

Comparative ingenuity pathway analysis (IPA) of the differential expression of genes affected in both MMTV-PyVT and PDX mice tumor arrays. (A) The heat map of the topmost downregulated genes overlapped in both MMTV-PyVT and PDX models. (B) The heat map visualization of the topmost upregulated genes overlapped in both MMTV-PyVT and PDX models. The magnified part visualizes the quantitative topmost 15 genes within each dataset. Left magnification panel; topmost PDX mouse tumor genes arranged in ascending order compared with their counterparts in MMTV-PyVT mouse tumors. Right magnification panel; topmost MMTV-PyVT mouse tumor genes arranged in ascending order compared with their counterparts in PDX mouse tumors. The complete overlapped genes list is provided in Supplementary Material.

Figure 7

Predicted IPA-generated network mapping of OC-X treatment effects on the signal transduction and molecular interactions of the five topmost affected genes that overlapped in both MMTV-PyVT and PDX array data. (A) Predicted mapping of affected genes was connected based on direct (solid lines) and indirect (dotted lines) interactions with MYC and its intermediate regulators, the best gene linked to the affected genes. The saturation of color is directly proportional to each gene fold change. (B) IPA-generated predicted correlation of MYC-driven genes and TNBC progression stage in response to OC-X treatment.