Nanoparticles are well established, most efficient, versatile drug carrier systems. Estradiol was entrapped in cationic PLGA nanoparticles stabilized by poloxamer 188. PLGA-Poloxamer 188, a well-established polymer-stabilizer combination, has proven to yield stable nanoparticles harboring an array of molecules with differing physiochemical properties such as docetaxel (Yan, Zhang et al. 2010). doxorubicin (Gelperina, Maksimenko et al. 2010), 5-fluorouracil (Li, Xu et al. 2008). Poloxamer 188 was evaluated at 0.5% and 1% w/v concentration as a stabilizer for estradiol PNP. Estradiol, a steroidal hormone with a log P value of 4, is expected to be solubilized in the hydrophobic PLGA matrix of the nanoparticle. With a view to attain stable nanosystems with superior estradiol loading, two different PLGA proportions viz. 1%w/v and 2%w/v were assessed. It is imperative to impart positive surface charge for superior hydroxy apatite binding concurrently evading potential cytotoxic effects which are associated with cationic polymers like PEI. Therefore, low molecular weight (5000 Da) linear PEI was used at 0.25%w/v and 0.5%w/v concentration to confer the necessary cationic surface charge.
The method of preparation is governed by the properties of the drug molecule and the polymers employed. In the present work, estradiol loaded cationic nanoparticles were prepared by emulsion solvent evaporation technique. Owing to the differing solubilities of the polymers in organic solvents, PLGA and PEI were solubilized in organic phase comprised of a DCM and methanol.. Poloxamer 188 was dissolved in the aqueous phase. Primary coarse o/w emulsion was obtained on addition of organic phase to aqueous phase in 1:5 ratio under shear at a maximum speed of 30000 rpm, by high speed homogenizer (IKA, Ultra Turrax,Mumbai). Further nanosizing of the oil droplets was attained on probe sonication. Removal of solvents from nanoemulsion under vacuum resulted in nanoprecipitation of estradiol loaded polymers in aqueous phase. The effect of formulation factors was assessed to determine an optimized formula.
4.1. Particle size and polydispersity index
Emulsification and stabilization are predominant factors controlling particle size and stability of nanoparticles. A number of reports in literature suggest stabililisation of PLGA nanoparticles with poloxamer 188. Formula P1 and P2 prepared with 1% and 2% w/v PLGA respectively were stabilized by 0.5% w/v poloxamer 188 (table.1). The mean particle size and PI increased from 135 nm to 204 nm and 0.35 to 0.39 respectively with an increment in PLGA content. On increasing the stabilizer proportion to 1% w/v in P3 and P4, smaller particles were obtained. This implied that the amount of stabilizer played a crucial role. In an emulsion, surfactant molecules align themselves at the interface, lowering the free energy thereby inhibiting coalescence of droplets. Solvent removal from droplets yields stable nanoparticles. Higher content of poloxamer 188 was vital to stabilize large surface area of nanodroplets in emulsion obtained with higher amount of PLGA.
Table 1
Effect of stabilizer with (Nanoparticles with 0.5% PEI content)
Formulation
|
PLGA
% w/v
|
Poloxamer 188 (% w/v)
|
Particle Size (nm)
|
Polydipersity index
|
P1
|
1
|
0.5
|
135 ± 2.3
|
0.35 ± 0.06
|
P2
|
2
|
0.5
|
204 ± 22
|
0.39 ± 0.04
|
P3
|
1
|
1
|
127 ± 1.7
|
0.26 ± 0.07
|
P4
|
2
|
1
|
178 ± 2.1
|
0.24 0 ± .03
|
n=3 |
4.2. Entrapment efficiency
Estradiol, a hydrophobic molecule is assumed to solubilize in amorphous PLGA matrix. The amount of carrier PLGA available for lodging estradiol plays a crucial role in governing entrapment. Estradiol was loaded at a maximum of 10% w/w of PLGA in PNP. Higher PLGA content is expected to increase the solid state solubility of estradiol in matrix. The entrapment efficiency of was 75-80% (table. 2). This was superior to that of values reported by Mittal et al which was 67%. (Mittal, Sahana et al. 2007). This discrepancy could be explained by the difference in grade and source of PLGA, organic to aqueous ratio, shear force applied, rate of solvent evaporation, factors which govern estradiol partitioning to the organic polymeric phase.
Table. 2. Effect of estradiol: PLGA proportion on entrapment efficiency.
(Nanoparticles with 0.5% w/v PEI and 1% w/w poloxamer 188)
Formulation
|
PLGA
% w/v
|
Estradiol
(%polymer content)
|
Entrapment efficiency
|
P5
|
1
|
5
|
75.6 ± 2.4
|
P6
|
1
|
10
|
77.12 ± 3.4
|
P7
|
2
|
5
|
79.4 ± 1.9
|
P8
|
2
|
10
|
78.78 2.5
|
n=3 |
4.3. Surface charge
PEI is the significant component in PNP responsible to bestow positive charge. Surface charge of nanoparticles with and without PEI is as depicted in Fig. 1. Negative zeta potential of -11 mV was observed with plain PLGA nanoparticles. Formulations with 0.25% w/v and 0.5% w/v PEI showed zeta potential values of +28 and +35 mV respectively. This increase in the surface charge confirms the presence of polycationic polymer, PEI on the surface of nanoparticle. The positive charge increased with proportion of PEI. The high zeta potential value observed in our investigation indicates good colloidal stability and is expected to have profound influence in bone binding property.
4.4. In vitro hydroxy apatite (HA) binding efficiency
In vitro HA binding of polymeric nanoparticles with and without PEI was evaluated. Figure 1.2 depicts HA binding affinity of different PNP formulations. As mentioned earlier, an increase in positive charge was observed with increase in PEI concentration. Plain PLGA nanoparticles without PEI showed meager 28% binding efficiency (Fig. 2.). On the contrary nanoparticles bearing PEI on their surface displayed superior binding of 82%. At physiological conditions, bone surface bears negative charge. The electrostatic interactions between negative surface and positive particle is expected to result in strong binding. This result affirms the findings by Zhang et al (Zhang, Wright et al. 2007). Taking into account potential cytotoxicity of PEI at high doses, and remarkable binding results in in vitro experiments, 0.5%w/v PEI concentration was considered optimum.
Based on the results it was construed that at 0.5% w/v of PEI concentration, 1% w/v of poloxamer 188 and 2% w/v PLGA yields positively charged nanoparticles desirable for strong binding of PNP to bone surface.
4.5. Cryo TEM
Morphological evaluation of PEI coated PLGA nanoparticles revealed their dense spherical structure. The solid core represents PLGA matrix with estradiol. The lighter layer on the outer surface surrounding the particle could be attributed to the coating with PEI and poloxamer 188. The particles in images appear smaller with 100-110 nm size (Fig. 3.) compared to the 178 nm obtained on assessment with particle size analyser. Photon correlation spectroscopy analyser works on dynamic light scattering principle which measures hydrodynamic diameter of the dispersed particles additionally accounting for the water layer surrounding nanoparticle surface. Whereas in cryotem, the morphology and particle size of frozen particles per se is determined. Moreover, the number distribution of particles assessed in analyser is larger than in cryo tem thus reflecting polydispersity of the sample. Similar disparity in particle size was also observed in few reports (Venkataraman and Nagarsenkar, 2013 ).
4.6. Thermal analysis
In the present investigation, EST drug sample and EST nanoparticles were subjected to heat cycle at 100 C per min rate. Crystalline nature of pure estradiol was evident from the sharp melting peak at 1780 C (Fig. 4). Absence of this peak in nanoparticle sample indicates estradiol solubilisation in PLGA matrix. It is anticipated that hydrophobic molecule EST remains in an amorphous solubilized state in hydrophobic PLGA matrix similar to the findings reported Musumeci et al (Musumeci, Ventura et al. 2006).
4.7. Freeze drying
Freeze drying is a promising approach for the stabilisation of colloidal dispersions. In the case of PLGA nanoparticles, the hydrolytic instability of the polymer in aqueous dispersion is the main issue. PNP dispersion was freeze dried to improve its stability for long term storage. Trehalose was used as a cryoprotectant at 10% w/v. The visual inspection of the lyophilised products did not show any signs of collapse or shrinkage after the freeze drying process. The freeze dried cakes were brittle and easy and rapid to reconstitute (table. 3).
Table 3
Particle size, surface charge and entrapment efficiency post rehydration of lyophilized product
Formulation
|
Mean particle size (nm)
|
Polydispersity index
|
Surface charge (mV)
|
Entrapment efficiency (%)
|
PNP dispersion
|
178
|
0.24
|
33.32
|
78.78
|
Freeze dried PNP (10% w/v trehalose) (rehydrated)
|
181
|
0.26
|
34.79
|
79.32
|
4.8. In vitro release studies.
PLGA nanoparticles are well established systems for sustained release of drug in vivo. A number of factors such as the ratio of lactide- glycolide, particle size, exposure to release medium (surface area and shape), temperature, degradation rate of PLGA, govern drug release from the matrix. In non targeted conventional PLGA nanoparticles (CNP), estradiol is entrapped in polymeric matrix composed of PLGA (50:50) Resomer rg 504H with equal portions of fast degrading, hydrophilic PGA and slow degrading hydrophobic PLA. EST loaded PLGA nanoparticles released EST only to an extent of 40% in 48 hours (Fig. 5.) compared to a faster release of 85% observed from solution. (P˂0.05). This sustained effect is owing to the slow erosion and subsequent dissolution of EST from PLGA matrix In case of bone targeted cationic PEI coated nanoparticles (PNP), 33% EST released in 48 h which was slower than plain PLGA nanoparticles (CNP). This could be attributed to the presence of additional PEI polymeric barrier on nanoparticle surface, hampering EST release from nanoparticles. EST release from PLGA nanoparticles reported for oral delivery also exhibited sustained release (Mittal, Sahana et al. 2007). Sustained EST release from PLGA nanoparticles can potentially reduce systemic drug levels, thereby releasing and maintaining therapeutic EST levels in the bone mileau reducing the dosing frequency in osteoporosis therapy.
Drug release from nanoparticles can follow a number of mechanisms. The drug release could be due to PLGA degradation which leads to erosion of matrix alongwith, diffusion mediated dissolution of drug from nanoparticles. Though these mechanisms work in harmony, their rates vary. With a view to identify the predominant mechanism, release data was fitted into various kinetic models. Graphs of cumulative percent drug release vs time (zero order), log percent drug remaining vs time (first order), cumulative percent drug release vs square root of time(Higuchi), cube rootof percent drug remaining vs time (Hixson crowell) and log percent drug release vs log time (Korsmeyer Peppas) were plotted. Table 4. summarizes the correlation values for these models. The best fit model was Korsmeyer peppas with highest correlation value of 0.98.
This clearly indicates that EST release from nanoparticle is primarily governed by diffusion mechanism. The magnitude of the release exponent value (n) < 0.43 implies that the release follows fickian diffusion mechanism. As the polymer degrades/erodes with time, the particle size of nanoparticles decrease. This increase in surface area and concentration gradient further stimulates drug release. On the contrary increased membrane thickness attenuates drug release resulting in sustained effect.
Table 4
Summary of correlation values for release kinetic models
|
Zero
|
First
|
Higuchi
|
Hixson Crowell
|
Korsmeyer Peppas
|
|
R2
|
R2
|
R2
|
R2
|
N
|
R2
|
EST CNP
|
0.9760
|
0.9392
|
0.9729
|
0.9689
|
0.0983
|
0.9773
|
EST PNP
|
0.9489
|
0.921
|
0.9489
|
0.9333
|
0.1143
|
0.9646
|
4.9. Pharmacokinetic studies
Pharmacokinetic profiles of EST solution, conventional and bone targeted polymeric systems obtained are as shown in Fig. 6 depicts drug plasma levels. EST solution followed one compartmental kinetics whereas both nanoparticles followed 2 compartmental kinetics with good relation between predicted and observed values. The r2 value was more than 0.97 for all systems. Two compartmental kinetics implies biexponential decline in plasma levels with nanosystems as opposed to single exponential plasma level reduction seen with solution. Post intravenous administration of solution, drug levels dropped drastically and were undetectable after 6 hours. Plasma levels of EST from conventional (CNP) and bone targeted polymeric nanoparticles (PNP) were 689 ng/mL and 824.15 ng/mL respectively at 30 minutes as shown in Fig. 6 Interestingly, there was no statistically significant difference in drug plasma levels observed with conventional and bone targeted nanosystems. The distribution of targeted systems to less perfused organ bone, and subsequent binding is not expected to reflect significantly in plasma drug levels.
EST was rapidly cleared from solution. Clearance and AUC0-inf values obtained for EST solution were 0.54 (µg)/ (ng/ml)/h and 458.87 ng/ml*h respectively. In case of conventional (CNP) and bone targeted nanoparticles (PNP) the clearance values significantly reduced to 0.061 and 0.057 (µg)/ (ng/ml)/h whereas AUC0-inf values substantially increased to 7022.61
and 14311.71 ng/mL*h respectively (Table 5). This trend is similar to the pharmacokinetic data of sustained release nanosystems reported in literature (Patere, Pathak et al, 2016). Based on invitro release studies and in vivo pharmacokinetic data, it can be construed that polymeric nanosystems exhibit sustained release of EST which has potential for lowering frequency of administration compared to plain drug solution. This is expected to reduce the adverse side effects associated with the long term exposure of EST to organs in conventional therapy.
Table 5
Pharmacokinetic parameter of EST loaded nanoparticles
Pharmacokinetic
parameter
|
EST solution
|
EST CP
|
EST PNP
|
A1 (1/h)
|
-
|
65719.12
|
16099.1
|
Alpha (1/h)
|
-
|
14.46
|
14.6
|
B (ng/mL)
|
-
|
745.59
|
809.03
|
Beta (1/h)
|
-
|
0.3
|
0.24
|
T1/2 alpha (h)
|
0.509
|
0.0479
|
0.0474
|
T1/2 beta (h)
|
-
|
2.305
|
2.819
|
Vss (µg/(ng/ml)
|
0.4
|
0.1432
|
0.0569
|
Cl ((µg)/(ng/ml)/h)
|
0.544
|
0.0601
|
0.057
|
AUC0-inf (ng/ml*h)
|
458.8714311.71
|
7022.61
|
|
MRT (h)
|
0.73
|
1. 21
|
0.9878
|
A1: concentration in distribution phase,
a: rate constant of drug elimination from the central compartment including both
elimination from the body and distribution to other compartment,
B: concentration in combined phase,
b: rate constant of drug elimination from the body once the
distribution is complete in a 2-compartment model,
t1/2a: distribution half-life,
t1/2b: physiologic half- life and represents combined effects of distribution and elimination in
2-compartment model,
Vss: apparent Volume of distribution at steady state,
AUC0inf: area Under the Curve from time 0 to infinity, and
MRT: mean residence time.
4.10. In vivo toxicity studies
Acute toxicity study was conducted to evaluate the adverse effect on single dosing of polymeric nanoparticle in rodents. The animals were observed for any critical, early toxic effects for over a period of 14 days. There was no abnormality or mortality observed in animals in the study period. The organ weights of treated animals were not significantly different (P ˃ 0.05) from that of control group (fig .7). Normal serum levels of alkaline phosphatase, creatinine, bilirubin, albumin and acceptable levels of complete blood counts like WBC, RBC, lymphocytes, monocytes were obtained. There was no evidence of any artefacts or abnormality in histological sections of vital organs which further reaffirmed the non toxic nature of polymeric nanoparticles (Fig. 8). It was imperative to determine the safety of polymeric nanoparticles with PEI due to the cytotoxicity reportedly associated with this cationic polyamine. Molecular weight, degree of branching and zeta potential govern the toxicity of PEI. In the present work, the concentration of PEI in formulations is within the safety levels, below the LD 50 values reported in animals. Based on the hematological and histological findings it is concluded that the concentration of PEI present in novel polymeric system is safe in vivo.
4.11 Biodistribution of polymeric nanoparticles
4.11.1 Radiolabelling of nanoparticles.
Radiolabelling of nanoparticles involved two steps. Firstly, sodium pertechnetate was converted to technitium using stannous chloride reducing agent. This was followed by addition of formulation and incubation in order to facilitate interaction with technitium for efficient radiolabelling.
Since the reaction conditions have a paramount influence on the radiolabeling efficiency, pH, radioactive dose and stannous chloride concentration were optimised. The radiolabelling efficiency was found to be 98-99% which could be attributed to increased binding of negatively charged technetium to a positively charged polymeric surface.
4.11.2 Invasive biodistribution study
Spatial and temporal in vivo behavior of radiolabelled polymeric nanoparticles was studied. Polymeric nanoparticles were radiolabelled and administered intravenously to mice. Rapid substantial uptake by primary site, RES organs such as liver and spleen, compared to other organs was evident in Fig. 9. Surface hydrophilicity bestowed upon polymeric nanoparticles by poloxamer stabilizer is expected to aid in evasion of opsonisation and further macrophage uptake. However, interplay of other factors such as surface charge and particle size also govern RES uptake, phagocytosis and thereby in vivo biodistribution. In the present investigation, polymeric particles with an average particle size of 178 nm showed significant uptake in liver and spleen. This could be because of the retention of 150-300 nm sized polymeric nanoparticles via filtration in these highly perfused organs which have 150 nm fenestration cut off size (Gaumet, Vargas et al. 2008). Hence larger polymeric nanoparticles with an average particle size of 178 nm are not expected to escape vasculature, instead get entrapped within liver and spleen. Whereas particles smaller than 150 nm escape via filtration. This is reflected in high radioactive RES organ uptake. This observation is in agreement with a number of reports (Vargas et al). There are reports stating that cationic particles have higher propensity to sequestration by RES organs compared to negative or neutral nanoparticles. This could be attributed to higher interaction of cationic nanoparticles with anionic cell membranes thereby leading to phagocytosis and RES uptake. This was distinctly evident from radioactive uptake data in spleen as shown in Fig. 9. However cationic nanoparticles displayed lower hepatic disposition vis-à-vis negatively charged conventional polymeric PLGA nanoparticles. This discrepancy in biodistribution could be attributed to the nature of opsonin proteins bound to the nanoparticle surface. The composition of nanoparticles and their surface characteristics influence the opsonisation and biodistribution process. The presence of splenic phagocyte specific opsonins on cationic nanoparticles would exhibit more affinity to spleen thereby preferentially sequestered in that organ. Whereas kupffer cell specific opsonins bound to negatively charged PLGA nanoparticles is expected to show predominant uptake in liver. However these observations are contrary to the reports which suggest higher propensity of cationic nanoparticles to accumulate in liver compared to neutral or negative ones (Abraham and Walubo 2005).
Positively charged nanoparticles showed profound increase in radioactive levels in kidney relative to non targeted ones. The endothelial cell lining of glomerular capillaries have fenestrations between 30-100 nm. Moreover, the glomerular basement membrane is anionic which is conducive to bind cationic nanoparticles in circulation (Bennett, Zhou et al. 2008).
In 1 h nearly 4% targeted and non targeted nanoparticles were accumulated in bones. There was gradual reduction in the uptake of non targeted nanoparticles with time. At 24h, 1.5% of non targeted nanoparticles and 6% cationic nanoparticles were retained in bones (Fig. 10). This superior bone binding observed with targeted nanoparticles could be endowed by the interaction of cationic PEI with negative functional groups of HA ensemble in physiological pH. The initial accumulation observed with non targeted nanoparticles could be attributed to the preferential uptake of 30-150 nm particles by bone marrow. Progressive clearance of accumulated non targeted nanoparticles from bone marrow explains the reduction in bone levels with time. It can be deduced that lack of cationic surface molecules to impart bone binding potential led to low bone retention of non targeted nanoparticles in 24h. The polymeric nanoparticles surface modified with cationic PEI exhibited superior bone binding potential both in vitro and in vivo, corroborating the proposed hypothesis.
4.11.3 Non invasive biodistribution study
In addition to determining and computing radioactive counts in individual organs, the in vivo fate of developed formulations was evaluated non invasively using SPECT imaging. The superimposed images of CT and SPECT clearly displayed the differences in the intensity of radioactive uptake by organs. These observations commensurate the organ uptake values calculated in invasive study. The images Fig. 11 and 12 depict the uptake of conventional (plain PLGA) nanoparticles and cationic nanoparticles. It was distinctly seen from images that positively charged nanoparticles displayed lower hepatic uptake compared to plain PLGA nanoparticles. This is in agreement with the findings of invasive uptake study.
It is to be noted that the invasive study indicates nearly 5-8 fold higher uptake by liver compared to bones, a less perfused organ. Moreover the distribution of radioactivity calculated in invasive method accounts for the entire skeleton. Hence the SPECT images fail to visually depict the difference in the uptake by bone or different formulations.
4.11.4 In vivo efficacy study
Surgically ovariectomised rat model is one of the most reliable preclinical evaluation tools to study the effect of hormonal intervention in post menopausal condition. Ovariectomy causes decline in estrogen levels further leading to bone loss. Estrogen therapy reportedly causes improvement in bone density and mechanical properties in ovariectomised rats. In the present work, animals were treated with EST solution and nanoparticles. There was no mortality was observed in any groups during the study period. There was no abnormality in activity and their food intake.
4.12. 1. Mechanical properties
It is reported that the rat OVX model in the proximal tibia, distal femur, and lumbar vertebrae
mimics conditions in the postmenopausal woman and is suitable for the evaluation of potential
therapeutic agents for the prevention of osteoporosis (Thompson. Simmons et al, 1995). Therefore in the present work, tibia and femur have been isolated and studied for the influence of EST on bone properties. Bone density was determined using Archimedes principle. Comparable hardness and density values of cationic nanoparticles with SHAM operated rats indicate osteoprotective nature of these hormone replacement therapy formulations (figure 13, 14). It implies that there is no significant difference in osteoprotective activity of natural estrogen (in sham operated rats) and EST in PEI coated nanoparticles. As expected bone density of animals treated with cationic nanoparticles was significantly higher than the OVX control group (P<0.001). Bone strength was evaluated by subjecting a force longitudinally to cause breakage. Bones with hard matrix would require higher force to break the bone. Higher the hardness value, stronger the bone. Bones with low mass and density are more susceptible to fracture. Higher density and hardness values indicate higher bone strength and resistance to bone breakage under force. Solutions were administered intravenously twice / week. The prolonged sustained drug exposure exhibited from targeted nanoparticles vis a vis poor drug bioavailability in solutions. Therefore solutions did not have significant influence in the bone strength.
4.12. 2. Biochemical parameters
Biochemical parameters such as alkaline phosphatase (ALP), calcium, phosphorous
are biomarkers which reportedly govern changes occurring in bone (Gupta, Goyal et al, 2014). ALP is widely found in bone cells and liver. ALP levels increase in skeletal disorders. In the present study, serum levels were significantly higher in ovx control animals compared to sham control and the rats treated with nanoparticles (Fig. 15.). These results are in agreement with the observations made in ovx females treated with estrogen (Pedrazzoni, Alfano et al, 1995). Alk P levels in formulation treated animals were similar to that of sham control group. However they were significantly lower compared to solution (P<0.001) implying inadequate estrogen levels from solution to aid bone remodeling.
Serum calcium and phosphorus levels were significantly higher in sham control vis a vis ovx control animals (P<0.001) (Fig. 16, 17). Formulations and sham control animals exhibited similar serum calcium and phosphorus levels. Gupta et al reported a similar observation in ovx animals treated with plant extracts. Sustained release of EST from formulations ensured desired therapeutic levels of EST for pronounced osteoprotective activity. Biochemical and mechanical evaluation of EST bone targeted nanoparticles confirm the superior osteotropic and protective action in ovorectomised rats.