Improved Bioavailability and Antitumor Effect of Docetaxel by TPGS Modified Proniosomes: In Vitro and In Vivo Evaluations

A novel oral drug delivery system, TPGS modified docetaxel proniosomes (DTX-TPGS-PNs), was designed to enhance the oral bioavailability and antitumor efficiency of the poorly water-soluble drug docetaxel. DTX-TPGS-PN niosomes were 93 ± 6.5 nm in size, −18.53 ± 1.65 mV in zeta potential and exhibited spherical morphology, with an encapsulation efficiency of 97.31 ± 0.60%. The system showed sustained release in both simulated gastric and intestinal fluid. The results of caco-2 monolayer, everted gut sac model and improved single-pass intestinal perfusion model transport studies showed that DTX-TPGS-PN niosomes could significantly improve the absorption of DTX. The pharmacokinetics study suggested the absolute bioavailability of DTX-TPGS-PN niosomes were 7.3 times that of DTX solution. In addition, a higher antitumor efficacy than DTX solution was demonstrated in MCF-7 and MDA-MB-231 cells in vitro and in MCF-7 tumor-bearing mice model in vivo. Our results demonstrated DTX-TPGS-PN is promising in enhancing the bioavailability and efficiency of poorly water-soluble drug DTX, and the potential of proniosomes as stable precursors for oral drug delivery.

Scientific RepoRts | 7:43372 | DOI: 10.1038/srep43372 been shown to induce hypotension, tachycardia, a rise in histamine levels, and promote the generation of biologically active complement products 11,12 . Clinically, the Tween-80 formulation is known to cause severe allergic reactions and peripheral neuropathy 13 . Therefore, various formulations were developed to improve the solubility of DTX while avoiding the side effects.
Oral delivery was an attractive route to deliver therapeutics via nanoparticles, for it affording easy handling, high patient compliance, less stringent production conditions and lower costs 14 . As for oral delivery of DTX, the major problem is its low bioavailability, which in part is caused by the excretion effect of P-glycoprotein. Here, we developed a new oral drug delivery system of DTX, the TPGS modified proniosomes (DTX-TPGS-PN), to overcome the low permeability and poor oral bioavailability of DTX and avoid the side effects of Taxotere ® .
Characterization of DTX-TPGS-PN niosomes was performed by dynamic light scattering (DLS) particle sizer and transmission electron microscope (TEM). The in vitro Caco-2 cell model, ex vivo everted gut sac model and in situ improved single-pass intestinal perfusion model were used to investigate the effect of DTX-TPGS-PN niosomes on absorption of DTX. Then, the antitumor efficacy of DTX-TPGS-PN niosomes was studied in vitro by MCF-7 and MDA-MB-231 cells and in vivo by MCF-7 tumor-bearing nude mice.

Results and Discussion
Preparation and characterization of DTX-TPGS-PN. Proniosomes were prepared by the method reported by Mahmoud Mokhtar et al. with some modifications (Fig. 1a) 15 . The prepared DTX-TPGS-PN niosomes had average diameters of around 90 nm (Fig. 1b) and a zeta potential of − 18.53 ± 1.65 mV. TEM data collected for DTX-TPGS-PN niosomes demonstrated that the nano-sized particles are spherical in shape and are uniformly distributed, however, with a smaller diameter of around 40 nm (Fig. 1c). It could be explained by that the nanoparticles measured by DLS were hydrated whereas those by TEM was dehydrated 16 . The DTX-TPGS-PN niosomes displayed a strong negative zeta potential, which is good for the stability of the niosomes against aggregation and fusion 17 . The encapsulation efficiency were 97.31 ± 0.60% for DTX-TPGS-PN niosomes and 91.50 ± 2.17% for DTX-PN niosomes. Indeed, it has been reported that TPGS is an efficient emulsifier that could realize 100% encapsulation efficiency (EE) of paclitaxel in poly(lactic-co-glycolic acid) (PLGA) NPs 18 . As we all known, particle size plays an important role on clearance and tumor uptake of nanoparticles 19 . Smaller particles (50-300 nm) are commonly removed from the circulation relatively slowly 20 . So the size distribution is generally controlled around 100 nm. The in vitro release results showed that both DTX-TPGS-PN and DTX-PN niosomes totally released in 12 h and 8 h in simulated intestinal fluid (SIF) and simulated gastrointestinal fluids (SGF), respectively. Transport of DTX-TPGS-PN niosomes across the Caco-2 cell monolayer. The effect of DTX-TPGS-PN and DTX-PN niosomes on transport of DTX across the Caco-2 monolayer was investigated, and comparing to DTX solution. The Caco-2 cell monolayer has a transepithelial electrical resistance of 488 ± 73 Ω ·cm 2 . Figure 2 showed that compared to DTX solution, the DTX-PN niosomes had higher apparent permeability coefficient (P app ), and once modified with TPGS, the P app of DTX-TPGS-PN niosomes was significantly higher than the DTX-PN niosomes. Although is often used to study intestinal transport, one disadvantage of the Caco-2 monolayer is its impermeability to hydrophilic or paracellular transport chemicals. Therefore it's a better model for colonic tissue permeability study with restricting reliability for the small intestinal tissue permeability study 14 . And because there are only epithelial cells in the intestinal epithelium, in contrast to many other cell types including mucosal cells and microfold cells (M-cells) in the real intestinal, we further used everted gut sac model and in situ improved single-pass intestinal perfusion model to further verify the results. Everted gut sac model study. Lactate dehydrogenase (LDH) is an intracellular enzyme, detected following damage to cell membranes, to be used as a biochemical marker of intestinal wall damage. The LDH release results in everted gut sac model are shown in Fig. 3a. At 0 h, LDH activity in the incubation medium of duodenum was 54.59 U/L/cm 2 . This was probably due to the surgical procedures when excising and everting the gut sac. Compared to the start, no significant differences of LDH activity at 1 h, 2 h, 3 h or 3.5 h were found in duodenum, and neither did in jejunum and ileum. It was concluded that the everted gut sacs maintained their viability during the period of the experiment and consequently this model was suitable for testing the DTX transport.
The concentration ratios of glucose between serosal side and mucosal side are shown in Fig. 3b. It was obvious that there were no significant differences on the ratios during the whole experiment process, indicating the tissue of the gut sac was viable and well-functioning.
The time course of transport of DTX across small intestinal segments is illustrated in Fig. 3c-e. There were no significant regional differences in intestinal absorption of DTX solution. The total drug uptake is: ileum < jejunum < duodenum (0.0015, 0.0024, and 0.0049 ug/cm 2 , respectively), which may probably due to physiological differences in expression of P-gp among them. DTX is a substrate of P-gp and the expression of P-gp in the ileum was 2.31-fold higher than that in the jejunum of rat. These may cause more DTX retained in duodenum than in ileum or jejunum. For each segment, drug amount was accumulated along with the time increasing. And significant improvement of DTX absorption by DTX-TPGS-PN niosomes was observed (p < 0.05) from 1.5 h to 2 h, indicating DTX-TPGS-PN niosomes may have better permeability than DTX. It might be attributed to the combination of the small particle size of DTX-TPGS-PN niosomes which allows efficient uptake in the intestine and the TPGS which could enhance the bioavailability by inhibiting P-gp activity 5,21 . Improved single-pass intestinal perfusion model study. Improved single-pass intestinal perfusion model provides a functional intestinal barrier and conditions that closely mimic the normal physiological state following oral administration in terms of providing an intact blood supply 22 . And a high correlation between rat and human P eff values has also been found 23 . So the improved single-pass intestinal perfusion model was also used to verify the intestinal absorption of the DTX-TPGS-PN niosomes. Figure 4 showed the DTX permeability of DTX solution, DTX-PN and DTX-TPGS-PN niosomes in the improved in situ perfusion model. Compared to the 48.7 × 10 −6 cm/s for DTX solution, the P eff value for DTX-PN and DTX-TPGS-PN niosomes were improved to 57.89 × 10 −6 cm/s and 73.84 × 10 −6 cm/s, which indicated the niosomes improved the bioavailability of DTX. The P eff value of DTX-TPGS-PN group was significantly higher than the DTX group, demonstrated the DTX-TPGS-PN niosomes do have an enhancing effect on the permeability of DTX. The P eff value of DTX-TPGS-PN group was also significantly higher than the DTX-PN group.
While, comparing with single-pass perfusion model, the improved single-pass perfusion model conducted with the whole small intestine instead of a part of intestine was proved feasible and had the advantage of maintaining better physiological conditions of the intestine 24 . Each model has its limitations but an association of them would provide us a better understanding of DTX absorption. By comparing the results of Caco-2 monolayer model, everted gut sac model and single-pass intestinal perfusion model, we can draw a conclusion that DTX-TPGS-PN niosomes have an improved permeability in intestine than DTX solution. And this may be explained by the advantages of niosomes with small size and the inhibition effect of P-glycoprotein by TPGS. It is well known that DTX is a P-gp substrate and the oral absorption is partly affected by P-gp activity. TPGS with P-gp inhibition ability may increase the oral absorption as well as tumor cell toxicity.
Pharmacokinetic studies of DTX-TPGS-NP. The blood concentration-time curves of DTX after p.o.
administration of DTX solution, DTX-PN and DTX-TPGS-PN niosomes in rats were showed in Fig. 5.
As showed in Table 1, the area under the curve (AUC) was significantly increased from 109.04 ± 17.74 ng/mL·h for DTX solution to 184.70 ± 13.05 ng/mL·h for DTX-PN niosomes and 796.44 ± 83.93 ng/mL·h for DTX-TPGS-PN niosomes, which corresponds to a significantly increase of the oral bioavailability from 2.32% to 3.93% and 16.94%, respectively. On the other hand, C max for DTX solution was 19.83 ± 6.30 ng/mL, which is statistically lower than 27.43 ± 1.53 ng/mL for DTX-PN and 155.67 ± 4.73 ng/mL for DTX-TPGS-PN. While the T max for DTX solution and DTX-PN niosomes was reduced from 1 h to 0.5 h for DTX-TPGS-PN niosomes. The enhanced and faster absorption of DTX-TPGS-PN niosomes in intestine was correlated with the in vitro absorption results. A particle size of around 100 nm and incorporation of TPGS, which works as P-gp excretion     . 6c and d). The IC 50 of DTX solution, DTX-PN and DTX-TPGS-PN niosomes were 0.715 μ g/mL, 0.508 μ g/mL, 0.378 μ g/mL in MCF-7 cells and 12.337 μ g/mL, 5.469 μ g/mL, 2.542 μ g/mL in MDA-MB-231 cells, respectively. It has been reported that TPGS could lead to a higher cellular uptake and therapeutic effects in cancer cells 5,6,27 . Our study also proved the DTX-TPGS-PN niosomes enhanced the cytotoxicity of DTX compared with DTX solution and DTX-PN niosomes on MCF-7 and MDA-MB-231cell lines. The cytotoxic effect of blank PN and TPGS-PN niosomes were also evaluated in the two cell lines ( Fig. 6a and b). In a wide range of vehicle concentrations (~625 μ g/ml), the PN and TPGS-PN niosomes did not show significant cytotoxicity toward the cells, indicating that blank PN and TPGS-PN niosomes are nontoxic to tumor cells. NS and vehicle TPGS-NP treatments had no substantial effect on the tumor growth, and the tumor volumes increased rapidly. The treatment with DTX (p.o., 5 mg/kg) and DTX-PN (p.o., 5 mg/kg) was effective in tumor regression to some extent, but neither have advantage when was comparable to the i.v. DTX injection or DTX-TPGS-PN niosomes, which were more efficacious in tumor reduction (Fig. 7d). Figure 7b showed the masses of the tumors were with the same tendency. The data confirmed the anti-tumor efficacy of i.v. DTX injection and p.o. DTX-TPGS-PN niosomes. This may be contributed to the enhanced absorption of DTX-TPGS-PN niosomes. The role of TPGS is necessary which not only enhanced the absorption of DTX to inhibit the excretion of P-gp but also served as an apoptosis inducer 28 . It has been widely demonstrated that polymers with a MW > 40 kDa and nanoparticles between 10 and 200 nm in size display reduced renal clearance (enhanced PK), and are able to migrate through open malignant neovasculature and accumulate in tumors via the enhanced permeability and retention effect 29 . The EPR effect of nanoparticles may also results in the enhanced antitumor effect of DTX-TPGS-PN niosomes. In addition, body weight was monitored through the study to assay the possible toxicity effect caused by DTX-TPGS-PN niosomes (Fig. 7c). Upon statistical analysis, there was no significant change of body weight during the whole experimental period, excepting the DTX injection group. The body weight reduction of the DTX injection group maybe caused by the systemic toxicity cause by the high concentration of DTX in plasma. It has been reported that Taxotere may cause significant stress and neutropenia to the mice at 40 mg/kg DTX 30 . Body weight in mice of TPGS-PN was almost as the same as control group, which may explained as the safety of the TPGS-PN. Although the body weight increase was a little lower compared to the control group, no significant differences was found, indicating that the oral delivery of DTX-TPGS-PN was safe.

Conclusion
A novel oral delivery system of docetaxel (DTX-TPGS-PN) was developed to improve the oral bioavailability of docetaxel. The prepared DTX-TPGS-PN niosomes had a diameter of 93 ± 6.5 nm and a high encapsulation

Materials and Methods
Materials and animals. Docetaxel was purchased from Shanghai Techwell Biopharmaceutical Co. Ltd (Shanghai, China). D-alpha-tocopheryl polyethylene glycol 1000 succinate (TPGS) was purchased from Sigma-Aldrich Co. (MO, USA). LDH commercial test kit was supplied by Jiancheng Biotech Institute (Nanjing, China). Glucose commercial test kit was purchased from Rongsheng Biotech Co. Ltd (Shanghai, China). Cell counting kit-8 was from Dojindo Laboratories (Kumamoto, Japan). Dulbecco's modified Eagle medium, fetal bovine serum, nonessential amino acid, penicillin, and streptomycin were from Invitrogen Life Technologies (Carlsbad, CA). All other chemicals and solvents were of analytical reagent grade.
Male Sprague Dawley rats of 200-220 g weight and female nude mice of 4-6 weeks were from Shanghai Super B&K Laboratory Animal Corporation Ltd (Shanghai, China) and maintained at 22 °C ± 2 °C on a 12-hour lightdark cycle with access to food and water ad libitum. The animals used for the experiment were treated according to the protocols evaluated and approved by the ethical committee of Shanghai University of Traditional Chinese Medicine (Shanghai, China). Caco-2 cells were a kind gift from Professor Jianxin Wang (School of Pharmacy, Fudan University, Shanghai, China). MCF-7 cells were obtained from Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China).

Preparation of proniosomes (PN).
In brief, 0.300 g Span 40, 0.097 g cholesterol and 0.004 g DTX were dispersed in 800 mg ethanol and heated in a water bath at 70 °C for 5 min, then 0.32 mL deionized distilled water of the same temperature was added in and incubated at 70 °C for 3 min. The system was cooled down and white creamy proniosomal gel was obtained at room temperature. All formulations were hydrated to niosomes before use in this study.  10 min, with vortexing for 3 times during the heating. The final volume was adjusted to 20 mL by the same buffer without TPGS. DTX-PN niosomes were prepared by the same method as DTX-TPGS-PN except that 14 mL Sorensen's buffer solution without TPGS was used instead.

Preparation of niosomes.
Particle size and zeta potential. Size and zeta potentials measurements were performed at room temperature by a Nicomp Particle Sizing system (380ZLS, Santa Barbara, California, USA). Particle sizes were measured directly and zeta potentials were analyzed after being diluted 100 times with distilled water.
Morphology. The morphology of DTX-TPGS-PN niosomes was observed by transmission electron microscope (TEM, H-6009IV, Hitachi, Japan). The samples were diluted with distilled water and negatively stained with phosphotungstic acid before observation.

In vitro release.
To investigate the release kinetics of DTX-TPGS-PN niosomes in different body fluids, freshly prepared niosomes were exposed to simulated gastrointestinal fluids (0.1% Tween-80, pH 1.2; SGF: 1 L contains 2 g sodium chloride, 3.2 g pepsin, 7 mL hydrochloric acid) or simulated intestinal fluid (0.1% Tween-80, pH 6.8; SIF:1 L contains 6.8 g monobasic potassium phosphate, 10 g pancreatin, 77 mL 0.2 N sodium hydroxide) 31 . The test tube each with 1 mL DTX-TPGS-PN niosomes solution was immersed in a water bath at 37 °C and gently shaken at 100 rpm. Six tubes were withdrawn at each predetermined time intervals, three of them were centrifuged at 12000 rpm for 20 min immediately and the supernatant was diluted 1:3 (v/v) with acetonitrile and followed by assay of HPLC, which is defined as the amount of released DTX in the media. The other 3 samples were directly diluted 1:3 (v/v) with acetonitrile and followed by assay of HPLC, which is defined as the total amount of DTX in the solution. Transport of DTX in the everted gut sac model. The everted gut sac was used as an in vitro model to study the intestinal absorption of the DTX-TPGS-PN niosomes. To prepare the gut sac, male SD rats fasted overnight (free access to water) were anesthetized and the intestinal segments of interest were identified (duodenum, starting 2 cm below the pylorus; jejunum, starting from 20 cm below the pylorus; ileum, starting 20 cm above ileocecal junction) and isolated 33 . Each intestinal segment was flushed with cold saline and immediately placed in 37 °C oxygenated (O 2 /CO 2 , 95%:5%) Kreb-Ringer's buffer (glucose 7.78 mM, NaCl 133 mM, KCl 4.56 mM, NaH 2 PO 4 1.50 mM, MgCl 2 0.20 mM, NaHCO 3 16 mM, CaCl 2 3.33 mM, pH was adjusted to 7.0-7.2) solution. The underlying mesenterium and fat were carefully removed and the sacs were gently everted with a glass rod. One end of the segment was clamped and tied with a silk thread forming a sac, while the other end was attached to a sampler. After the blank solution (1 mL) was introducing into the everted sac (serosal side), a 10-cm-long everted gut sac was prepared. Each sac was individually placed in a 15 cm glass tube (containing oxygenated Kreb-Ringer's buffer solution), which was incubated in a 37 °C water bath with 50 rpm shaking.
Four everted gut sacs of the duodenum, jejunum and ileum segment, respectively, were made to investigate the permeability of different intestine segments. The prepared sacs were pre-incubated for 5 min in oxygenated blank Kreb-Ringer's buffer 24 . Then, sacs were incubated in 37 °C oxygenated Kreb-Ringer's buffer solution, and DTX solution and DTX-TPGS-PN niosomes were added respectively. At each predetermined time, 0.1 mL of serosal fluid was aspirated from the gut sac and preheated blank buffer of same volume was added. After the experiment, the area of each sac was measured. Samples were kept at − 20 °C before analysis by using a validated HPLC method.
LDH release of the everted gut sac. To investigate the viability of the gut sac during the experiment period, the release of LDH, a cytosolic enzyme, was monitored 34 . Its concentration in the incubation medium of mucosal side was measured from 0 min to 240 min according to the protocols of the LDH kit.
Glucose transport across the everted gut sac. Glucose is actively transported in the small intestine.
Healthy and metabolically active sacs will concentrate glucose in the serosal medium and will maintain this gradient. So glucose concentrations were measured in both mucosal and serosal sides for 4 h period of the experiment, their ratios were calculated and compared to the control group to verify the viability and integrity of the gut sac 35  In situ single-pass intestine perfusion in rats. Single-pass intestine perfusion studies were performed in rats with some modifications 23,34,36 . In brief, rats were fasted for 16-18 h (free access to water) prior to the experiment, and anesthetized by intraperitoneal injection of urethane (30 mg/kg), and supplemental doses of anesthetic were given as necessary 37 . After being placed on a heated pad to maintain normal body temperature, a midline incision was made on the abdomen. The biliary duct was ligated, then whole small intestine was isolated and gently flushed with saline (37 °C). Silicone tubes were inserted into the proximal duodenum and terminal ileum and attached to a peristaltic pump (Shanghai Huxi Analysis Instrument Factory, China). Arranged in a uniform S or multi-S pattern to avoid kinks, the small intestine was returned to the abdominal cavity to maintain its viability without disrupting blood vessels. The exposed area was covered with gauze, and saline (37 °C) was applied to keep it warm and moist during the experiment. The intestine was flushed with blank Kreb-Ringer's buffer (37 °C) until the effluent perfusates were clear.

Effects of DTX-TPGS-PN on DTX transepithelial permeation.
To investigate the effects of DTX-TPGS-PN niosomes on DTX transepithelial permeation, Krebs-Ringer's buffer solution containing DTX, DTX-PN niosomes and DTX-TPGS-PN niosomes (DTX concentration of 2 μ g/mL) were perfused at a constant flow rate of 0.3 mL/min. The intestinal segment was perfused with test solution for about 30 min to achieve absorption equilibrium and stable outflow rates. Subsequently, the effluent perfusates were collected in 10 min interval for 90 min in pre-weighed 5 mL glass vials with a lid. Samples were frozen immediately and stored at − 20 °C until analysis. At the end of the experiments, the segment between two cannulas was excised without dragging and its length was measured using silk thread. The DTX concentrations were measured and the permeability coefficients (P eff ) values were calculated and compared thereafter.
Water flux correction and permeability calculation. The effective permeability P eff was calculated to compare the absorption of DTX-TPGS-PN niosomes. The net water flux (NWF) was determined by a gravimetric method 24 . C cor , the drug concentration of effluent perfusates which was corrected for water flux, was calculated according to the following equation: where C out is the concentration of tested drug in the effluent perfusates (μ g/mL), Q in and Q out are the inlet and outlet flow rate, respectively, which are adjusted for liquid density (mL/min). The effective permeability coefficients (P eff ) were calculated from where C in is the concentration of tested drug in the influent perfusate, 2π rL is the area of the mass transfer surface (cm 2 ) within the intestinal segment which is assumed to be a cylinder area.
Pharmacokinetic studies of DTX-TPGS-NP. Pharmacokinetic study was performed in healthy male SD rats. The rats were randomly assigned into four groups (5 rats per group). The pharmacokinetic properties of DTX-TPGS-NP were compared with DTX-NP and a marketed DTX injection (Jiang Su Heng Rui Medicine Co. LTD). The DTX injection was administered intravenously at a dose of 20 mg/kg. Besides, the DTX injection, DTX-PN and DTX-TPGS-PN niosomes were administered orally at a dose of 20 mg/kg. Whole blood samples were collected by retro-orbital puncture into heparinized tubes at 0.25, 0.5, 1, 2, 4, 8, and 12 hours following oral administration and at 0.033, 0.083, 0.25, 0.5, 1, 2 and 4 hours following intravenous administration. The plasma samples obtained were immediately centrifuged at 10,000 rpm for 3 minutes, and 100 μ l of plasma was mixed with 300 μ l paclitaxel (50 ng/mL) as the internal standard. The supernatant was collected after centrifugation at 12,000 rpm for 10 minutes, and mixed with 100 μ L deionized water for the drug content analysis by HPLC-MS (API-3000, Applied Biosystems, USA). The pharmacokinetic parameters were calculated using a non-compartmental model by the Drug and Statistics (DAS) software (version 2.1.1, Mathematical Pharmacology Professional Committee, China). The absolute bioavailability of DTX after oral administration compared to the intravenous injection was defined as follow: po iv iv po AUC represents the area under the curve, dose iv stands for the dose of intravenous injection, dose po stands for the dose of oral administration. Antitumor effect of DTX-TPGS-NP on MCF-7 tumor-bearing mice. The antitumor effect of DTX-TPGS-NPs was investigated on a MCF-7 breast cancer xenograft model. The female nude mice of 4-6 weeks old were subcutaneously injected with 0.1 mL of cell suspension containing 2 × 10 6 MCF-7 cells in the right flank without estrogen supplement 38 . When the tumors grew to approximately 50-80 mm 3 , the mice were assigned randomly into seven groups (8 mice per group): NS group (control); TPGS-PN group (p.o., 20 mg/kg); DTX group (p.o., 20 mg/kg); marketed DTX injection group (i.v., 10 mg/kg); DTX-PN group (p.o., 20 mg/kg);. DTX-TPGS-PN group (p.o., 10 mg/kg, 20 mg/kg, respectively). The mice were administrated intravenously or orally once every 3 days for 15 days. For tumor growth inhibition study, tumor size and body weight were measured every 3 days during the experimental period. The tumor volume was calculated based on the equation (a × b 2 )/2, where a and b were the length of the major axis and minor axis, respectively. On day 18, the mice were sacrificed and the tumor in each group was collected for further analysis.
The inhibition ratio was calculated by the following formula:

Statistical analysis.
All the data in this study were analyzed by SPSS 19.0 (SPSS Inc., Chicago). Data are expressed as mean ± SD. For values that were normally distributed, direct comparison between two groups was conducted by Independent Sample's T test. P value of < 0.05 was considered statistically significant.