Designing Hybrid Onconase Nanocarriers for Mesothelioma Therapy

Sep 1, 2014 - Preclinical Nuclear Imaging Laboratory, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas,. Texas 7539...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/molecularpharmaceutics

Designing Hybrid Onconase Nanocarriers for Mesothelioma Therapy: A Taguchi Orthogonal Array and Multivariate Component Driven Analysis Rakesh K. Tekade,†,‡ Susanne R. Youngren-Ortiz,† Haining Yang,§ Rahul Haware,∥ and Mahavir B. Chougule*,†,§ †

Department of Pharmaceutical Sciences, The Daniel K. Inouye College of Pharmacy, University of Hawaii at Hilo, Hilo, Hawaii 96720, United States ‡ Preclinical Nuclear Imaging Laboratory, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, Texas 75390, United States § University of Hawaii Cancer Center, University of Hawai’i, Honolulu, Hawaii 96813, United States ∥ College of Pharmacy & Health Sciences, Campbell University, Buies Creek, North Carolina 27506, United States ABSTRACT: Onconase (ONC) is a member of a ribonuclease superfamily that has cytostatic activity against malignant mesothelioma (MM). The objective of this investigation was to develop bovine serum albumin (BSA)−chitosan based hybrid nanoformulations for the efficient delivery of ONC to MM while minimizing the exposure to normal tissues. Taguchi orthogonal array L9 type design was used to formulate ONC loaded BSA nanocarriers (ONC-ANC) with a mean particle size of 15.78 ± 0.24 nm (ζ = −21.89 ± 0.11 mV). The ONCANC surface was hybridized using varying chitosan concentrations ranging between 0.100 and 0.175% w/v to form various ONC loaded hybrid nanocarriers (ONC-HNC). The obtained data set was analyzed by principal component analysis (PCA) and principal component regressions (PCR) to decode the effects of investigated design variables. PCA showed positive correlations between investigated design variables like BSA, ethanol dilution, and total ethanol with particle size and entrapment efficiency (EE) of formulated nanocarriers. PCR showed that the particle size depends on BSA, ethanol dilution, and total ethanol content, while EE was only influenced by BSA content. Further analysis of chitosan and TPP effects used for coating of ONC-ANC by PCR confirmed their positive impacts on the particle size, zeta potential, and prolongation of ONC release compared to uncoated ONC-ANC. PCR analysis of preliminary stability studies showed increase in the particle size and zeta potential at lower pH. However, particle size, zeta potential, and EE of developed HNC were below 63 nm, 31 mV, and 96%, respectively, indicating their stability under subjected buffer conditions. Out of the developed formulations, HNC showed enhanced inhibition of cell viability with lower IC50 against human MM-REN cells compared to ONC and ONC-ANC. This might be attributed to the better cell uptake of HNC, which was confirmed in the cell uptake fluorescence studies. These studies indicated that a developed nanotherapeutic approach might aid in reducing the therapeutic dose of ONC, minimizing adverse effects by limiting the exposure of ONC to normal tissues, and help in the development of new therapeutic forms and routes of administration. KEYWORDS: onconase, mesothelioma, hybrid nanocarrier, Taguchi design, multivariate analysis

1. INTRODUCTION Malignant mesothelioma (MM) is an aggressive tumor mainly caused by asbestos exposure which causes over 3,000 deaths per year in the United States and >100,000 deaths per year worldwide.1 Most patients succumb to the disease within a year, even when intensively combined therapeutic strategies were used. The current treatment modalities for MM include surgery, chemotherapy, and radiation therapy.2 For patients with unresectable tumors, surgery and radiotherapy are not practical treatment options. Additionally, chemotherapy is associated with detrimental side effects and results in lower response.1 The chemotherapeutic agent pemetrexed (Alimta, Eli Lilly, Indianapolis, IN, USA) has been approved by the © XXXX American Chemical Society

United States Food and Drug Administration (FDA) in combination with cisplatin to treat patients with a pleural mesothelioma. However, this combination therapy was found to only extend the patients’ life for approximately 11 weeks.3 This scenario has urged the development of a novel and more effective targeted pharmacological intervention for the treatment of MM. Received: June 6, 2014 Revised: August 25, 2014 Accepted: September 1, 2014

A

dx.doi.org/10.1021/mp500403b | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics

Article

Figure 1. Scheme showing formulation of ONC loaded hybrid nanocarriers (ONC-HNC).

Onconase (ONC) is a member of the ribonuclease-A (RNase-A) superfamily and is extracted from the oocytes of the northern leopard frog Rana pipiens, which showed both in vitro and in vivo activity against MM.4 ONC activity is attributed to targeting RNA degradation followed by induction of cell apoptosis.5 Tumor necrosis factor-alpha (TNF-α) activates the nuclear translocation of p65 subunit of nuclear factor kappa (NF-κB), a transcription factor, which leads to the survival of mesothelioma cells and favors the MM pathogenesis.6 It has been reported previously that ONC is a potent inhibitor of TNF-a dependent NF-κB activity in MM cells.5 ONC inhibited NF-κB activity and MMP9 secretion, which induced apoptosis in MM cells and decreased MM invasion.5 Moreover, it was found that ONC inhibited MM tumor growth in a severe combined immunodeficiency (SCID) xenograft mouse model.5 These unique mechanisms of action of ONC together with a lack of immunogenicity make ONC a promising alternative to current chemotherapies for MM therapy.4,7 However, regardless of these anticipations, the recent reports published on ONC phase III clinical trials for the MM treatment did not completely eradicate the tumors.8,9 Various reports have also demonstrated that the cytosolic entry of ONC might be a limiting factor for anticancer activity of ONC.4,7 In addition, the ONC has a short half-life of 5 μm and 1−5 μm are trapped within the capillary beds and are phagocytosed by Kupffer cells in the liver, respectively. NC of 100−200 nm size are rapidly cleared from circulation by the spleen compared to NC of 20−100 nm.14,15

Thus, the blood circulation lifetime of NC depends on the particle size, which impacts the NC fluid dynamics. Spherical NC with 100−150 nm flows through the middle of the vasculature due to van der Waals forces and hemodynamic fluidics. Therefore, NC maintain a further distance from the endothelial lining, resulting in the prolonged circulatory lifetime.16 NC of 0.05) on the ONC EE (%) and EE (%). To investigate the release profile of ONC from developed formulations, the in vitro release was investigated under physiological pH, which confirmed a prolonged ONC release from the developed HNC (Figure 4). The in vitro release of plain ONC solution showed almost complete release within 4 h. In contrast, chitosan-coated and TPP cross-linked ONCHNC showed ONC release in the range of 49% to 80% within 48 h. In the case of a ONC-HNC4 formulation containing a

Figure 4. In vitro release profile of ONC from various nanocarriers under physiological pH (pH 7.4). Lyophilized formulations were resuspended in PBS solution and filled inside a dialysis membrane bag with molecular weight cutoff of 300 kDa (Sigma, USA). The membrane bags were placed in 50 mL of PBS pH 7.4 medium, the system being maintained at 37 ± 2 °C with continuous slow magnetic stirring at 300 rpm. At specific time intervals, 0.5 mL aliquots of dissolution medium were withdrawn and analyzed using the Biospek UV spectrophotometric method. Results are represented as mean ± SD (n = 4). H

dx.doi.org/10.1021/mp500403b | Mol. Pharmaceutics XXXX, XXX, XXX−XXX

Molecular Pharmaceutics

Article

Table 4. Regression Coefficients of the Design Variables and Individual Nanocarrier Formulation Effects on the Response Variables Particle Size (nm) and Zeta Potential (mV) in Different Buffer Systems Obtained in Separate PCR Modelsa particle size (nm)

zeta potential (mV)

X-variables

PBS 7.4

PBS 6.4

PBS 5.4

PBS 7.4−FBS 10%

PBS 7.4

PBS 6.4

PBS 5.4

PBS 7.4−FBS 10%

regression coefficient ONC-ANC ONC-HNC1 ONC-HNC2 ONC-HNC3 ONC-HNC4 chitosan TPP RMSEC RMSEP R2: RMSEC R2: RMSEP no. of PCs X-variance (%) Y-variance (%)

20.0525 −6.2072 −0.5653 0.8464 2.2572 3.6679 45.1431 45.1431 2.90 7.35 0.9030 0.6018 2 64 96

20.2435 −6.9028 −0.6275 0.9413 2.5101 4.0789 50.2019 50.2019 3.18 8.16 0.9057 0.6022 2 64 96

20.7691 −8.9218 −8.111 1.2166 3.2443 5.272 64.886 64.886 3.61 10.28 0.9255 0.6143 2 64 98

19.0325 −10.6184 −0.9653 1.448 3.8612 6.2745 77.2249 77.2249 7.24 14.28 0.8144 0.5376 2 64 86

−10.701 −10.9453 −0.995 1.4925 3.9801 6.4677 79.6025 79.6025 0.99 13.67 0.9960 0.5119 2 64 100

−8.0498 −10.6411 0.9674 1.4511 3.8695 6.2879 77.3901 77.3901 1.25 12.73 0.9933 0.5538 2 64 100

−5.8218 −11.4524 −1.0411 1.5617 4.1645 6.7673 83.2902 83.2902 1.82 13.39 0.9870 0.5760 2 64 100

−14.03458 −10.7404 −0.9764 1.4646 3.9056 6.3466 78.1117 78.1117 2.18 14.58 0.9801 0.4325 2 64 98

a

The significance of regression is determined by full cross-validation and Martens uncertainty tests and corresponds to approximately p < 0.05.22−24 [Negative sign (−) indicates negative significance, positive sign (+) positive significance, RMSEC root-mean-square error of calibration, and RMSEP root-mean-square error of prediction.]

in terms of particle size and zeta potential and quantitatively analyzed using individual PCR models (Table 4). Regression coefficients of calculated PCR models for the particle size of developed HNC showed an increase in particle size of HNC at lower pH. Further, regression coefficients of the calculated PCR model for the zeta potential showed a statistically significant (p < 0.05) increase in the zeta potential at pH 5.4 compared to pH 7.4. A higher positive impact of chitosan and TPP on the particle size of the developed NC was observed at low pH. Similarly, both chitosan and TPP showed a larger positive impact on the zeta potential at pH 5.4 than at pH 7.4. The increase in the particle size and zeta potential of ONC-HNC at low pH conditions might be attributed to the protonation of amine groups of chitosan.65 Cationic NC have been reported to induce a toxic effect on normal tissues due to membrane damage.45,50The positive charges in the developed HNC were well distributed in contrast to highly localized positively charged polymers such as dendrimer based NC.42 Hence, we expect that the HNC developed with albumin and chitosan will exert minimal membrane damage effect. The conjugations of positively charged NC with PEG have been reported to reduce the membrane damage, and this approach may be used to overcome the limitation of cationic NC.51,52 Aggregation of NC formulations in the presence of serum proteins in blood is a key issue in NC based drug delivery.56 Hence, the stability of ONC-ANC and HNC formulations was also evaluated in PBS 7.4 containing 10% v/v FBS to assess the influence of serum protein present in blood.56 This assay was also performed to determine the extent of particle aggregation due to the presence of varying zeta potential on the various ONC-HNC. A separate PCR model was calculated to understand properties of developed ONC-HNC and to quantify the effect of chitosan and TPP on the particle size and the zeta potential under this buffer condition. Regression coefficients of the calculated PCR model indicated an increase in particle size and zeta potential of developed ONC-HNC in the presence of the PBS 7.4−FBS 10% v/v system. PBS 7.4−FBS 10% v/v was used to mimic the

higher concentration of chitosan and TPP, only 49.08% release was observed within 48 h. Release rates of ONC, ONC-ANC, and HNC formulations were calculated from the slope of release profiles of the first 12 h. An individual PCR model was calculated to separate out the release profiles of individual formulations and to quantify the effect of chitosan and TTP on the release rate. The linear regression equation of the ONC release rate in the PCR model is given in eq 4. RMSE at the calibration stage and at the prediction stage was 0.72%/h and 1.92%/h, respectively. The model r2 at the calibration stage and at the prediction stage was 0.9092 and 0.6000 respectively. RR (%/h) = 8.8912 + 1.6120(ONC‐ANC) + 0.1462(ONC‐HNC1) − 0.2198 (ONC‐HNC2) − 0.5862(ONC‐HNC3) − 0.9525(ONC‐HNC4) − 11.7235 (chitosan) − 11.7235(TPP)

(4)

The regression coefficients of eq 4 indicated that coating the ONC-ANC with chitosan followed by TPP cross-linking resulted in the prolonged release of ONC from HNC. These results strongly support that a HNC approach can be used to achieve the controlled release of ONC. HNC with the size of 20−40 nm have been developed in this investigation. In an earlier investigation, Jun et al. observed that ANC of