Inhibition of Breast Tumour Growth with Intravenously Administered PRKCA siRNA- and PTEN Tumour Suppressor Gene-Loaded Carbonate Apatite Nanoparticles

: Gene therapy aims to silence an oncogene through RNA interference, or replace an ab-normal tumour suppressor via gene augmentation. In this study, we intended RNA interference for PRKCA oncogene and gene augmentation for PTEN tumour suppressor with a view to reduce tumour growth in a mouse model of breast cancer. Inorganic carbonate apatite nanoparticles (CA NPs) were utilized to deliver the synthetic siRNA and the puriﬁed gene-carrying plasmid DNA both in vitro and in vivo. Effects of PRKCA siRNA- and PTEN plasmid-loaded NPs on viability of MCF-7, MDA-MB-231 and 4T1 breast cancer cells were assessed by MTT assay. The cell viability data in MCF-7 cell line demonstrated that combined delivery of PRKCA speciﬁc siRNA and PTEN plasmid with CA NPs had an additive effect to signiﬁcantly decrease cellular growth compared to individual treatments. In addition, we observed a similar pattern of cumulative inﬂuence for combined treatment in triple negative MDA-MB-231 breast cancer cell line. Upon treatment with PRKCA siRNA+PTEN plasmid-loaded NPs, a remarkable decrease in the phosphorylated form of AKT protein of PI3K/AKT pathway was observed in Western blot, indicative of diminished proliferative signal. Moreover, in vivo study in MCF-7 xenograft breast cancer mouse model demonstrated that the rate of growth and ﬁnal tumour volume were reduced signiﬁcantly in the mouse group that received intravenous treatment of PRKCA siRNA+NPs, and PTEN plasmid+NPs. Our ﬁndings demonstrated that PRKCA siRNA and PTEN plasmid loaded into CA NPs attenuated breast tumour growth, suggesting their therapeutic potential in the treatment of breast cancer. the correlation between PRKCA and PTEN proteins has been examined towards breast cancer gene therapy for the ﬁrst time through our study. The novelty of our study reﬂected the additive effect upon co-delivery of siRNA against PRKCA and PTEN plasmid gene in reducing breast cancer cell proliferation, as reﬂected in cell viability assessment and Western blot experiments.


Introduction
Breast cancer is the second highest cancer-related illness worldwide, causing over 0.6 million deaths with around three-fold new cases annually [1]. It is caused through numerous genetic changes including gain of function mutations in proto-oncogenes and loss of function mutations in tumour suppressor genes. Among several intracellular signaling proteins, protein kinase C (PRKC) oncoprotein and PTEN tumour suppressor are examples of potential therapeutic targets for blocking the malignancy of breast cancer cells [2].
PRKC is a family of serine/threonine protein kinases involved in a number of signal transduction pathways which regulate several cellular functions including differentiation, proliferation, and apoptosis. Consequently, its inevitable implication in tumorigenesis and metastasis has led to extensive research to date [3]. Human PRKC isoforms encompass 15 members which are classified into three groups [4]. The protein kinase C alpha (PRKCα) isoform, encoded by PRKCA gene, belongs to the 'classical' group which can be activated by phospholipids, Ca 2+ , and diacylglycerol (DAG), a lipid second messen-In order to protect the siRNA/DNA from nuclease-mediated degradation in intracellular and extracellular environment, smart pH-sensitive inorganic CA NPs have been employed for effective nucleic acid transfer onto the cancer cells with positive therapeutic impact. Chowdhury et al. had established development of proficient, biodegradable inorganic NPs of CA with ability of effective intracellular delivery and release of associated siRNA, and plasmid DNA from endosomes to cytosol, leading to specific cleavage of a particular mRNA transcript, or expression of a desirable protein, respectively [26]. At low pH 7.4, in bicarbonate buffered medium, addition of Ca 2+ salt and plasmid DNA or siRNA generates nano-sized carbonate apatite/siRNA/DNA complexes, following incubation at 37 • C for 30 min. Next, FBS is added to stop further nanoparticle complex formation [27]. In Field Emission Scanning Electron Microscope (FE-SEM) images, CA NPs displayed spherical shape, rough texture, and an average diameter of 50-300 nm [28][29][30]. With molecular formula Ca 10 (PO 4 ) 6−x (CO 3 ) x (OH) 2, the small sized nanoparticles have durable binding affinity towards the siRNA/DNA; the positively charged Ca 2+ present in the CA NPs can electrostatically combine with the negatively charged siRNA/DNA molecules and come close to the cell membrane through ionic interaction, ultimately transporting the siRNA/DNA into cytoplasmic acidic compartments (endosomes) by endocytosis [27]. Upon cellular internalization by endocytosis, particles are quickly dissolved into component ions such as Ca 2+ , PO 4 3− and HCO 3 − in endosomal acidic pH, resulting in particle dissolution, swelling and breakage of the endosomes following development of osmotic pressure across the endosomal membrane, and release of the siRNA/DN A in cytosol [27,28]. In the cytoplasm, the released siRNA is merged into a RISC (RNAinduced silencing complex) and relaxed into single-stranded RNAs by Argonaute-2. This single-stranded RNA-associated RISC finally guides and particularly degrades the complementary target mRNA with the help of Argonaute-2 [31,32]. In case of plasmids, negatively charged plasmid DNA immediately interacts with several microtubule proteins in the cytosol, like kinesin and dynein for plasmid translocation to the nucleus through the nuclear pore or during cell division, promoting a high level of transgene expression in cancer cell lines [33]. With low immunogenicity and toxicity effects, these NPs ensure prolonged plasma half-life and enhanced tumour accumulation of the nucleic acids, eventually enabling their efficient cellular uptake via endocytosis and fast intracellular release [27]. This study sheds light on the very first use of biodegradable CA NPs to deliver PRKCA siRNAs and PTEN plasmid into breast cancer cells and investigate the effect of selective gene silencing and gene augmentation on tumour regression.

Formulation of CA NPs
Commercially available DMEM powder (with high glucose and without sodium bicarbonate) was mixed with 44 mM of sodium bicarbonate and pH was adjusted to 7.4. Formation of CA nanoparticle was attained by adding required volume of 1 M exogenous CaCl 2 ·2H 2 O (4 µL for in vitro and 8 µL for in vivo usage) and/or potential therapeutics (siRNA/plasmid) into 1 mL final volume of the fresh media, and incubating at 37 • C for 30 min, followed by addition of 10% FBS to the vial to prevent further generation and consequential aggregation of the particles. FBS was skipped in case of CA NPs for intravenous delivery of siRNA and plasmid.

Isolation of Plasmid DNA for Gene Transfection
The pCMV-PTEN plasmid contained PTEN gene insert (1209 bp) and a tag/fusion protein named Flag (N-terminal on insert). It is a high copy number plasmid grown in E. coli DH5α cells containing ampicillin resistant genes. For plasmid isolation, E. coli DH5α bacteria harbouring PTEN plasmid gene were grown in LB agar plates (ampicillin 100 µg/mL) overnight at 37 • C. After 16-18 h, individual colonies were inoculated in 100 mL of LB broth in presence of ampicillin antibiotic and grown in an incubator (Labtech Company) with shaking at 220 rpm for 16-18 h at 37 • C. The plasmids were isolated and purified from bacteria using plasmid Mega kit (Qiagen, Germany). The concentration of extracted DNA was measured in 'Implen' nanodrop machine. A ratio of absorbance at 260 and 280 with 1.8 reading or more was indicative of highly pure extracted DNA. Furthermore, purified DNA was run on 0.8% agarose gel and the DNA bands were visualized under ultraviolet transilluminator using 'Quantity One' Bio-Rad software.

Reconstitution of PRKCA siRNA
The lyphilized PRKCA siRNA was reconstituted with RNase-free water to obtain a stock solution of 10 µM and stored at −20 • C as recommended by Qiagen.

Size and Zeta Potential Measurement
The size and zeta potential of PRKCA siRNA and PTEN plasmid complexed with CA NPs were measured using Malvern Zeta Sizer Nano (United Kindom). The NPs, PRKCA siRNA+NPs, PTEN+NPs and PRKCA siRNA+PTEN+NP complexes were formed by adding 4 µL of 1 M CaCl 2 , 1 nM siRNA against PRKCA and 1 µg of PTEN plasmid into 1 mL of DMEM (pH 7.4), followed by incubation at 37 • C for 30 min. After that, 10% FBS was added in order to stop further formation of NPs complexes which were subsequently maintained on ice till measurement was done with Zetasizer. The data was analyzed using Zetasizer software 6.20 and all samples were measured in duplicates, with the values presented as mean ± SD.

Cell Culture and Seeding
The three breast cancer cell lines-MCF-7, MDA-MB-231 and 4T1 were grown in culture flask in DMEM supplemented with 10% FBS and 1% penicillin and streptomycin antibiotic in a 37 • C incubator humidified with 5% CO 2 . One day before treatment, exponentially growing cells were trypsinised, centrifuged and resuspended using DMEM. Cells were counted under optical microscope using a hemocytometer and seeded in a 24-well plate with cell density of 50,000 cells per well. Cells were allowed to attach overnight at 37 • C with 5% CO 2 before transfection.

Cell Viability Assessment with MTT Assay
Following incubation of the breast cancer cells in presence of NPs, PRKCA siRNA+NPs, PTEN+NPs and PRKCA siRNA+PTEN+NPs for 48 h, 50 µL of MTT solution (5 mg/mL in PBS) was added to each well and the cells were incubated for 4 h at 37 • C. The medium was removed and 300 µL of DMSO was added to each well to dissolve the purple formazan crystals and absorbance was measured in a micro plate reader at 595 nm with a reference wavelength of 630 nm. Cell viabilities were normalized to the absorbance of untreated cells (control). Each treatment experiment was performed in triplicate and cell viability was expressed as mean ± SD.
The cell viability in the treated wells was expressed as a percentage and was calculated using the absorbance values obtained from MTT assay. The following formulae were used to calculate the cell viability data: Cell viability (%) = Abs(sample) − Abs (negative control) Abs (positive control) − Abs (negative control) × 100

Western Blotting
MCF-7 cells were incubated with respective siRNA-and or plasmid-loaded NPs for a consecutive period of 48 h and washed with a pre-chilled phosphate buffered saline prior to lysis. The cell lysates were then centrifuged at 13,000 rpm for 10 min at 4 • C using a refrigerated centrifuge (Sartorius Stedim Biotech, Göttingen, Germany) and protein concentration of the cell lysate was estimated using the Bradford assay. The protein extracts were subjected to sodium dodecyl sulphate polyacrylamide gel electrophoresis (precast gel) at 60 V for 90 min, and the resolved proteins were electro-transferred for 7 min at 25 V to PVDF membranes (Thermo Scientific, Waltham, MA, USA) using Turbo transfer unit (Bio-Rad, Hercules, CA, USA). The membranes were blocked for 1 h at room temperature with Tris buffered saline with Tween 20 (TBST) containing 5% skimmed milk (Merck, St. Louis, MO, USA). The membranes were then incubated with 1:1000 primary antibody (Cell Signaling Technology, Beverly, MA, USA) in TBS-T with 5% bovine serum albumin for overnight, at 4 • C with slow rocking. The membranes were washed 5 times with TBST and incubated with 1:3000 secondary antibody (anti-rabbit IgG-HRP linked) for 60 min at RT with slow rocking. Next, Clarity Western ECL substrate (Bio-Rad, Hercules, CA, USA) was added following incubation for 5 min. Bands were visualized using Bio-Rad Gel Doc system with Quantity One chemiluminescent software.

Study of Tumor Regression Pattern in Breast Tumor-Induced Mouse Model (In Vivo)
For in vivo tumour regression study, MCF-7 xenograft mouse model was developed using standard procedure and subsequently treated with NPs carrying siRNA against PRKCA and/or PTEN plasmid, and the tumour growth was monitored in a longitudinal way. The Monash University Animal Ethics Committee approved all the procedures involved in this experiment (MUM/2018/12). Female Balb/c mice (6-8 weeks old) of 15-20 gm of body weights were purchased from School of Medicine and Health Science animal facility, Monash University and maintained in 12:12 light:dark condition by providing them ad libitum chow and water. Using standard methodology, approximately 1 × 10 6 MCF-7 cells (in 100 µL PBS) were injected subcutaneously on the mammary pad of mice unilaterally, and the mice were checked between three to five days for the development of tumour by touching the area of injection with index finger. As the tumour volume reached around 25-30 mm 3 , the mice were randomly distributed into four different groups (5 mice per group) for treatments. On day 8, mice were treated intravenously through tail-vein injection with 100 µL of NP (8 M Ca 2+ ), PRKCA siRNA+NP (100 nM siRNA), PTEN+NP (40 µg of plasmid) and PRKCA siRNA+PTEN+NP (100 nM siRNA and 40 µg of plasmid). FBS was skipped in case of NPs for intravenous delivery of siRNA and plasmid; rather, the formulation was kept on ice to prevent aggregation of nanoparticles. The second dose was administered at day 10, i.e., after 2 days from the 1st dose. During the treatment period, the size of the tumour was measured at regular intervals (4 times/week). The gross body weights of mice were monitored and the lengths and widths of the tumours were measured using the vernier caliper in mm scale for 22 days. Moreover, the gross body weights of mice and their activities were monitored for study purpose. The volume of the tumour was calculated using the following formula: The data are presented here as the mean ± SD of tumour volumes from each group. At the end of the study, at day 22, mice were humanly sacrificed by cervical dislocation following 100% CO 2 exposure for few seconds and tumours were excised for further study.

Statistical Analysis
For cell viability study, data were expressed as means ± standard deviation (SD) for (n = 3) biological replicates. Student's t-test was used throughout. For in vivo tumour regression study, LSD post-hoc test for one way ANOVA was used to analyze and compare the significant difference between treatment groups. The differences among groups were considered significant at p < 0.05.

Size and Zeta Potential of PRKCA Specific siRNA-and PTEN Plasmid-Loaded NPs
Prior to cellular transfection, we assessed the size and zeta potential of the siRNA and/or plasmid NPs complexes, and the results were plotted in Figure 1. CA NPs were prepared with 4 mM exogenous CaCl 2 in 1 mL of bicarbonate buffered-DMEM medium with 1 nM PRKCA specific siRNA and/ or 1 µg of PTEN plasmid DNA, followed by incubation for 30 min at 37 • C. Next, 10% FBS was added to stop further siRNA/plasmid-NPs complex formation. Figure 1a,b demonstrated the average sizes and zeta potential of PRKCA+NPs, PTEN+NPs and PRKCA+PTEN+NPs formulations. The sizes of the NPs with loaded PRKCA siRNA, PTEN plasmid and PRKCA siRNA+PTEN plasmid werẽ 112 nm,~196 nm and~168 nm in diameter, respectively (Figure 1a). These nanoscale-sized complexes are advantageous for cellular transfection compared to large sized vector carriers.
As shown in Figure 1b, the zeta potential of NPs control was observed to be −11 mV. Incorporation of negatively charged siRNA and or plasmid DNA molecules increased the negative surface charge of the siRNA/plasmid +NPs complexes. The zeta potential for PRKCA+NPs, PTEN+NPs and PRKCA+PTEN+NPs complexes was recorded as −12 mV, −13 mV and −11.5 mV, respectively. As our NPs contain both positive and negative ions, such as Ca 2+ , PO 4 3− , OH − , the negative charge may come from the protein corona formed around the NPs in serum-containing media. In fact, a recent study revealed that at physiological pH, CA NPs attach to both positive and negative proteins; however, the negative proteins outnumbered the positive ones, which could contribute to the negative zeta potential [28]. Negative zeta potential is advantageous in terms of longer retention in blood stream [34]. Comparison of particle diameter and zeta potential of siRNA, plasmid and NP complex showed that large particles (PTEN+NPs,~196 nm) possess slightly more negative zeta-potential (−13 mV) compared to small particles (PRKCA+NPs are~112 nm and PRKCA+PTEN+NPs are~168 nm).  In MCF-7 cells (Figure 3a), treatment with 1 nM PRKCA siRNA+NPs reduced cell viability to ~50% after 48 h of incubation. Again, 1 μg PTEN+NPs treated cells exhibited ~57% of cell viability. Interestingly, combined treatment with PRKCA siRNA+PTEN+NPs complex curtailed the cell viability to ~37%. In other words, compared to single siRNA and plasmid treated cells, co-delivery of PRKCA siRNA and PTEN gene in this cell line revealed statistically significant decrease (p < 0.05) in cell growth, indicating additive effect of the treatment.
In MDA-MB-231 cell line (Figure 3b), the PRKCA siRNA+NPs treatment declined cell viability to approximately 84%, and intracellular delivery of PTEN+NPs could lower the cell viability to ~85%. Moreover, PRKCA siRNA+PTEN+NPs complex reduced the cell viability to ~75%. Altogether, delivery of both PRKCA siRNA and PTEN genes could induce additive effect compared to the individual siRNA or plasmid treatment in this cell line.
Our observation in 4T1 cells (Figure 3c) showed that treatment with PRKCA siRNA + NPs and PTEN+NPs complexes for a consecutive period of two days led to cell viability of ~64% for both treatments. However, when co-delivered (PRKCA+PTEN+NPs), it reduced the cell viability to ~61% only, indicating no net decline in cell viability was observed compared to the individual siRNA or plasmid gene transfection in 4T1 cell line.
In all of the three cell lines, NPs-only administration reduced cell viability, when compared to untreated cells. The cytotoxicity caused by NPs can be explained by the notion that particles associated with cell membrane could damage cell membrane, while the particles internalized into the cells release Ca 2+ upon dissolution into the endosomes, enhancing cytosolic Ca 2+ level which is associated with necrosis and apoptosis of the cells [35]. In MDA-MB-231 cell line (Figure 3b), the PRKCA siRNA+NPs treatment declined cell viability to approximately 84%, and intracellular delivery of PTEN+NPs could lower the cell viability to~85%. Moreover, PRKCA siRNA+PTEN+NPs complex reduced the cell viability to~75%. Altogether, delivery of both PRKCA siRNA and PTEN genes could induce additive effect compared to the individual siRNA or plasmid treatment in this cell line.
Our observation in 4T1 cells (Figure 3c) showed that treatment with PRKCA siRNA+NPs and PTEN+NPs complexes for a consecutive period of two days led to cell viability of~64% for both treatments. However, when co-delivered (PRKCA+PTEN+NPs), it reduced the cell viability to~61% only, indicating no net decline in cell viability was observed compared to the individual siRNA or plasmid gene transfection in 4T1 cell line.
In all of the three cell lines, NPs-only administration reduced cell viability, when compared to untreated cells. The cytotoxicity caused by NPs can be explained by the notion that particles associated with cell membrane could damage cell membrane, while the particles internalized into the cells release Ca 2+ upon dissolution into the endosomes, enhancing cytosolic Ca 2+ level which is associated with necrosis and apoptosis of the cells [35].

Effect of Co-Delivery of PRKCA siRNA and PTEN Gene on AKT Signaling Pathway: Western Blot Analysis
In order to further verify our cell viability results in MCF-7 cells and to interpret the effect of the co-treatment on cellular proliferation pathway, after treatment of the cancer cells with NPs (control), PRKCA siRNA+NPs, PTEN+NPs and PRKCA siRNA+PTEN+NPs, cell lysates were prepared using a cell lysis buffer. The protein content of the lysates was measured by Bradford protein assay, and 10 µg of proteins for each treatment was loaded onto a precast gel for their separation. Afterwards, the proteins were transferred onto a PVDF membrane using the Trans-Blot Turbo Transfer System (Bio-Rad) and the PVDF membrane was incubated overnight with monoclonal antibodies directed against phospho-AKT (P-AKT), total AKT and GAPDH as housekeeping loading control.
After incubating the PVDF membrane with P-AKT antibody, the same membrane was stripped with a stripping buffer and incubated with another antibody for detection of total AKT in the lysates. As evident from the Western blot image and densitometry analysis (Figure 4a,b), the expression levels of P-AKT protein in the cells treated with PRKCA siRNA+NPs and PTEN+NPs were lower than the NPs control. Notably, the band was almost faded or invisible in the lysate of the cells treated with PRKCA siRNA+PTEN+NPs formulation.   was stripped with a stripping buffer and incubated with another antibody for detecti of total AKT in the lysates. As evident from the Western blot image and densitome analysis (Figure 4a,b), the expression levels of P-AKT protein in the cells treated w PRKCA siRNA+NPs and PTEN+NPs were lower than the NPs control. Notably, the ba was almost faded or invisible in the lysate of the cells treated with PRKC siRNA+PTEN+NPs formulation.  The significantly reduced amount of P-AKT protein was a direct indication of the successful co-delivery effect of the siRNA and the plasmid. Reduced AKT activation refers to subsided survival signal via AKT/mTOR pathway and thus retarded cellular growth [36].

Influence of NPs-Facilitated PRKCA siRNA and PTEN Tumor Suppressor Gene Delivery on Tumor Regression
Based on cell viability and Western blot results, PRKCA has been revealed as a potential target for siRNA-mediated knockdown in reducing proliferation and/or survival of breast cancer cells. In addition, gene augmentation therapy by delivering PTEN tumour suppressor plasmid with the aid of NPs also showed promising cell viability profile. Moreover, co-treatment of the siRNA and plasmid remarkably exhibited significant decrease in breast cancer cell viability as well as diminished level of proliferation signal by downregulating P-AKT. Considering all the results, we further examined the feasibility of the nano-formulations in an in vivo setting. Because our in vitro cell viability results in MCF-7 cell line were very prominent compared to other two cell lines, we established MCF-7 xenograft mouse model to test our nucleic acid-based nanotherapeutics, PRKCA siRNA+NPs, PTEN+NPs and PRKCA siRNA+PTEN+NPs in comparison with the control (NPs) on the tumour-bearing mice randomly assigned to different groups. Within eight days of cell injection, tumour volume reached approximately 25-30 mm 3 (Figure 5a), and mice were treated intravenously through tail-vein injection. As plotted in Figure 6, following intravenous delivery of the formulations, the trend for development of tumour in the mice treated with PRKCA siRNA complexed with CA NPs was drastically reduced compared to the NPs only and the difference was significant at day 12, day 14, day 16, day 18, day 20 and day 22; (* p < 0.05 for PRKCA+NPs vs. NPs).   Moreover, compared to NPs control group, noticeable slow tumour growth was similarly observed in the mouse group treated with PTEN+NPs formulation with a p-value significant at day 12, day 14, day 16, day 18, day 20 and day 22 (* p < 0.05 for PTEN+NPs vs. NPs). This tumour regression profile observed following delivery of wild type PTEN tumour suppressor plasmid in mouse breast tumours using NPs cargo might support the role of PTEN in reducing tumour growth. However, unlike the additive effects both in cell viability and Western blot study, combined delivery of the two components could not significantly reduce the tumour volume in the mouse group treated with PRKCA siRNA+PTEN+NPs formulation. However, the trend for tumour reduction pattern in this group of mice was almost same as the regression pattern followed by the PTEN+NPs treatment group.
Additionally, no mice from any of the treatment groups showed any alteration in the behaviour after the injections while body weight remained almost the same for each group of the mice tested (data not shown).

Quantitation of Tumor Volumes upon Treatments
As shown in Figure 7, at day 22, upon sacrifice of mice, PRKCA siRNA+NPs-treated mose group showed significant reduction in tumour volume compared to the NPs control group (~245 mm 3 vs. 532 mm 3 , p < 0.05). In addition, PTEN+NPs and PRKCA siRNA+PTEN+NPs treated mice had smaller tumour volumes compared to the control (~162 mm 3 vs. 532 mm 3 , p < 0.05) and (~261 mm 3 vs. 532 mm 3 , p < 0.05), respectively. With regards to the control, all three treatment groups had reduced tumour volumes, implying the crucial roles of PRKCA suppression and restoration of PTEN tumour suppressor's function in tumour volume reduction.  Additionally, no mice from any of the treatment groups showed any alteration in the behaviour after the injections while body weight remained almost the same for each group of the mice tested (data not shown).

Quantitation of Tumor Volumes upon Treatments
As shown in Figure 7, at day 22, upon sacrifice of mice, PRKCA siRNA+NPs-treated mose group showed significant reduction in tumour volume compared to the NPs control group (~245 mm 3 vs. 532 mm 3 , p < 0.05). In addition, PTEN+NPs and PRKCA siRNA+PTEN+NPs treated mice had smaller tumour volumes compared to the control (~162 mm 3 vs. 532 mm 3 , p < 0.05) and (~261 mm 3 vs. 532 mm 3 , p < 0.05), respectively. With regards to the control, all three treatment groups had reduced tumour volumes, implying the crucial roles of PRKCA suppression and restoration of PTEN tumour suppressor's function in tumour volume reduction.

Discussion
As a promising cancer therapeutic strategy, siRNA has numerous advantages over chemotherapeutic drugs including high degree of safety and specificity. siRNA inhibits the post-transcriptional stage of gene expression by complementary base pairing with a target mRNA through RNA interference system, without adverse side effects [37]. An- Figure 7. Quantitation of tumour volumes in the mouse groups treated with NPs (control), PRKCA siRNA+NPs, PTEN+NPs and PRKCA siRNA+PTEN+NPs. Statistical analysis was found significant compared to NP control group, * when p < 0.05.

Discussion
As a promising cancer therapeutic strategy, siRNA has numerous advantages over chemotherapeutic drugs including high degree of safety and specificity. siRNA inhibits the post-transcriptional stage of gene expression by complementary base pairing with a target mRNA through RNA interference system, without adverse side effects [37]. Another crucial advantage of siRNA is its high efficacy, restraining the expression of a target gene markedly in a single cancer cell with just several copies [38].
However, 'free' or 'naked' siRNA can hardly penetrate the cell or tissue because of its negatively charged phosphate backbone that could be repelled by anionic extracellular matrix molecules. In addition, its half-life is very short due to the risk of degradation by serum nucleases, demanding for a robust delivery vector to protect it from cleavage and deliver to the target site with high efficacy [39]. In a recent study, Kamaruzzman et al. investigated the application of siRNA nanotherapeutics in breast cancer targeting several oncogenes such as EGFR, ERBB2 and BCL-2 using CA NPs of size ranging 100-300 nm. This siRNA mediated simultaneous silencing of cell-surface growth factor receptors and antiapoptotic genes synergistically killed breast cancer cells through inhibition of MAPK and PI-3 kinase pathways [40]. In addition, Chua et al. reported that CA NP-mediated intracellular delivery of c-ROS1 siRNA sensitized MCF-7 breast cancer cells to chemotherapy drugs cisplatin and paclitaxel [41]. In order to further investigate other oncogene signaling pathways, in this study, CA NPs were employed to deliver the PRKCA siRNA, which showed great implication in minimizing breast cancer cell growth both in vitro and in vivo.
Evidence has emerged from several studies describing importance of PRKCA knockdown using siRNA targeted to PRKCA gene in different types of cancer. Specific and efficient silencing of PRKCA gene was achieved using chemically synthesized siRNAs, resulting in prolonged obstruction of neuroblastoma cell division [42]. Wu et al. demonstrated that PKC-alpha siRNA transfection nearly abrogated PKC-alpha expression and significantly reduced melanoma cell migration compared to siRNA controls [43].
PTEN, another potential target in our study, is considered as a candidate tumour suppressor gene based on the finding that mutation or loss of this gene has been linked to a variety of common human cancers, including breast, prostate, glioblastoma, melanoma and lymphoid cancer [44]. A group of researchers affirmed that using lipofectamine 2000 transfection method, over-expression of PTEN induced cell growth arrest and apoptosis in PTEN-null human breast cancer cells [45]. Similarly, a study by Chen et al. concluded that exogenous PTEN suppressed the growth of human breast cancer cell and prompted apoptosis by phosphatase activity. They also used lipofectamine 2000 for delivery of PTEN gene in breast cancer cells [46]. Unlike lipofectamine 2000, we utilized inorganic pH sensitive CA NPs which are small in size, convenient, cost-effective, biodegradable and biocompatible nano-vectors for transfection of the PTEN plasmid in breast cancer cell line with very noticeable outcomes. Our cell viability and protein expression results showed that over-expression of PTEN in MCF-7 cells caused growth suppression by downregulating P-AKT expression, in agreement with the study by Chen et al. as mentioned above [46].
Delivering double elements to maximize the treatment outcome is an attractive concept in clinical medicine. The additive response of the combined formulation in our study can be explained by their opposite regulatory roles in PI3K-AKT-mTOR pathway. Upon activation, PKCα catalytically phosphorylates a variety of protein substrates (e.g., AKT to P-AKT) [47,48], thus activating the protein of interest towards cellular proliferation, whereas PTEN catalyses the opposite reaction, dephosphorylation (e.g., P-AKT to AKT), and implies deactivation of the proliferative signal [49]. To the best of our knowledge, the correlation between PRKCA and PTEN proteins has been examined towards breast cancer gene therapy for the first time through our study. The novelty of our study reflected the additive effect upon co-delivery of siRNA against PRKCA and PTEN plasmid gene in reducing breast cancer cell proliferation, as reflected in cell viability assessment and Western blot experiments.
However, for the in vitro study, we observed differences in cell viability results in three different breast cancer cell lines. Because breast cancer is a complex and heterogenous disease, gene expression profiling of the different representative cell lines also revealed heterogeneity at molecular level-explaining differential behavior of the treatments in cell lines [50]. A study by Sandra et al. established that different types of proteins interact with NPs, and the composition of protein corona could control uptake into specific cell types [51].
The in vivo testing of the potential nucleic acid-based nanotherapeutics composed of PRKCA siRNA+NPs and PTEN plasmid+NPs in the mice bearing MCF-7-induced xenograft breast tumours revealed quite similar results, as observed in cell viability assessment and Western blot tests. Mouse groups treated with PRKCA siRNA+NPs and PTEN plasmid+NPs formulations exhibited significant reduction in tumour size compared to the control group of mice. As observed from the size and zeta potential results of the siRNA, plasmid and NsP complex, PTEN+NPs possessed slightly more negative zeta-potential compared to other complexes, and the mean tumour volume was smallest (162 mm 3 ) in PTEN+NPs group of mice, implying that more negative zeta potential was advantageous in terms of longer retention of the formulation in blood stream [29]. However, the additive effect of the combined delivery could not be verified with the in vivo tumour regression results which rather followed the tumour reduction pattern similar as the mouse groups treated with individual elements of PRKCA siRNA+NPs and PTEN plasmid+NPs. Principally, the in vivo system is a relatively more complicated biological environment controlled by multiple factors and signaling pathways. A study by Islam et al. demonstrated that upon systemic delivery, NPs come into interaction with serum proteins, forming a protein corona (PC). The composition of PC depends on size, charge and other physicochemical features of NPs, and in in vivo mouse studies, the PC can be remodeled with the change of proteome in different treatment formulations [28,52,53]. Another possibility of epigenetic changes in in vivo environment may affect gene expression and protein interaction in different treatment groups; hence, future study might focus on the in vivo mechanism of the combined therapy.

Conclusions
Inhibiting PRKCA oncogene and restoration of PTEN tumour suppressor activities through intracellular delivery of PRKCA siRNA and PTEN gene, with the help of CA NPs, significantly inhibited breast tumour growth both in vitro and in vivo, supporting the notion that PRKCA as well as PTEN might be the potential targets for gene therapy of breast cancer while shedding light on the potential applications of the pH-responsive NPs as a powerful tool for nucleic acid delivery in clinical setting.