Dual-responsive (pH/temperature) Pluronic F-127 hydrogel drug delivery system for textile-based transdermal therapy

A dual-responsive hydrogel (pH/temperature) was developed from a thermos-responsive polymer, pluronic F-127 (PF127), and pH-responsive polymers, N,N,N-trimethyl chitosan (TMC) and polyethylene glycolated hyaluronic acid (PEG-HA). Gallic acid, the principal component of the traditional Chinese drug Cortex Moutan was loaded into the hydrogel (PF127/TMC/PEG-HA) for possible application in textile-based transdermal therapy as Cortex Moutan has been proven to be an effective drug for the treatment of atopic dermatitis (AD). TMC and PEG-HA were synthesized, characterized (1H-NMR and FTIR), and added to the formulations to enhance drug release from the hydrogels, and increase the drug targeting of the carriers. The thermo-responsive properties of the hydrogel were assessed by dynamic viscosity analysis and the tube inversion method, and the pH-responsiveness of the formulation was determined by changing the pH of the external media. Rheology study of the hydrogels showed that complex viscosity and storage/loss moduli for PF127/TMC/PEG-HA hydrogel formulation are higher than PF127 hydrogel. The microstructure analysis by reflection SAXS indicated similar type of frozen inhomogeneity of hydrogel formulations. Various characterizations such as FTIR, SEM, TEM, zeta potential, and degradation of the hydrogel formulation indicated that the PF127/TMC/PEG-HA hydrogel showed better physico-chemical properties and morphology than did the PF127 hydrogel, and drug release was also higher for the PF127/TMC/PEG-HA hydrogel than for PF127. The drug release from hydrogels followed more closely first-order rate model than other rate models.

solution and then, de-ionized water for 3 days. A solid sample of PEG-HA was obtained after lyophilization (Alpha 1-4 LD, Christ). The reaction scheme for the preparation of PEG-HA is given in Fig. 1A.

Synthesis of TMC.
TMC was synthesized using one-step methylation process 48 in which 0.25 g of chitosan and 0.6 g of sodium iodide were first dissolved in 10 ml of 1-methyl-2-pyrrolidinone at 60 °C with stirring for 24 h, and then, reacted with 2 ml of 0.15 (g/ml) aqueous sodium hydroxide solution and 2 ml of methyl iodide at room temperature under stirring for 2 h. The chemical synthesis scheme of preparing TMC from chitosan is given in Fig. 1B. The product was precipitated using ethanol, followed by isolation of the product using centrifugation and the same process was repeated 2-3 times. The product was dissolved in 5 ml of water, to which 30 ml of 1 M HCl in ethanol (96%) was added to exchange iodide for chloride. Water-soluble TMC, as the final product, was obtained in the form of a white solid powder after centrifugation, followed by washing with ethanol and ether, and finally drying in vacuum at 40 °C.
Preparation of drug-loaded hydrogel formulations and assessment of their thermo-responsive behavior. The hydrogel forming polymers PF127, TMC, and PEG-HA were added to the water and ethanol mixture along with the drug (gallic acid) to obtain a drug-to-polymer weight ratio (w/w) of 1:9 in the final hydrogel formulation of PF127/TMC/PEG-HA. In brief, 1.35 g PF127, 0.008 g TMC, and 0.008 g PEG-HA were added to the de-ionized water (7.74 g) and ethanol (0.55 g) mixture along with 0.15 g gallic acid dissolved in 0.2 g de-ionized water to obtain a hydrogel formulation (PF127/TMC/PEG-HA) of 10.006 g (∼10.0 g) and PF127 was maintained at 13.5 (wt%) of the total formulation weight. In the drug loaded hydrogel formulation made of only PF127 (PF127), the hydrogel formulation was composed of 1.35 g PF127 dissolved in 7.75 g deionized water, 0.55 g ethanol mixture, and 0.15 g gallic acid dissolved in 0.2 g de-ionized water. The final weight of the hydrogel formulation was maintained at 10.0 g, the drug:polymer ratio was kept at 1:9 (w/w), and PF127 was also maintained at 13.5 (wt%) of the total formulation weight in the sample. The hydrogel formulations in the sol phase were vortexed followed by incubation at 4 °C to obtain a clear solution. The thermo-responsiveness of gel formulations was investigated using dynamic viscosity measurements and the tube inversion method after applying heat or changing the temperature 1 . The formation of hydrogels with thermo-responsive properties is schematically presented in Fig. 1C.
Thermo-responsiveness of hydrogel formulations evaluated by dynamic viscosity analysis and tube inversion method. The dynamic viscosity measurements of the formulations PF127/TMC/PEG-HA and PF127 loaded with drug (gallic acid) were carried out using a DVE Brookfield viscometer with a spindle (s05) setting shear rate at 85 s −1 (50 rpm). The percent torque (%) displayed on the instrument was maintained within the range of 10-90% during the viscosity measurement of the samples. The temperature was varied from 5 to 50 °C during the measurement to determine the dynamic viscosity values of the hydrogel formulations at different temperatures. The temperature was increased manually at a rate of 4 °C/min over the range of 5-50 °C and the inflection point for the hydrogel formulations was determined from the dynamic viscosity vs temperature plot. The temperature inside the solution was thoroughly measured by dipping the thermometer into the sample system to determine gelation temperature of the samples.
The sol-gel transition of drug loaded formulations PF127/TMC/PEG-HA and PF127 only was determined by the tube inversion method where the flowability of 10 ml formulations in 20 ml capped glass tubes was checked at different temperatures in the range of 5-50 °C with increments of 1 °C/stage inside thermostatic oven (ESCO www.nature.com/scientificreports www.nature.com/scientificreports/ Isotherm, Forced Convection Laboratory Oven) 49 . The temperature and flowability of the solutions inside the tubes were checked at each stage to determine the temperature at which the sample stopped flowing inside the tube after tube inversion 45 . Rheology study. The modular compact rheometer (MCR302, Anton Paar) with TruStrain TM control was used to measure complex viscosity (Pa.s), storage modulus (Pa), and loss modulus (Pa) of drug loaded formulations PF127 and PF127/TMC/PEG-HA. The rheological study was done with 25 mm parallel plates and the values were taken within the temperature range of 5-50 °C at a constant heating rate of 1 °C/min. All the rheological parameters were recorded as the function of temperature and the temperature at which sol-gel conversion started was determined from the inflection point of the graphs. pH-responsiveness of hydrogel materials evaluated by the swelling ratio. The freeze-dried hydrogel formulations of PF127/TMC/PEG-HA and PF127 (0.2 g) were allowed to swell separately in 2 ml of 0.1 M acetate buffer (pH 5.4) and 0.1 M PBS (pH 7.4) for 4 h at 30 °C in capped 5 ml glass vials. At different time intervals, the swelled samples were collected, and the adhered liquid was removed using paper towels, finally, the dried samples were weighed to determine the swelling ratio of the freeze dried samples in response to different pH conditions. The samples collected after the mass measurement were dispersed in the same volume of fresh buffer to maintain uniform sink conditions throughout the experiments. The swelling ratio of the hydrogel formulations was determined from Equation (1) where W t is the final mass of the hydrogels after swelling in buffer and W 0 is the initial mass of the freeze-dried hydrogel samples. All experiments were carried out in triplicate, and the average values are reported with ± standard deviations (SD). To interpret the scattering intensity (I) data of reflection SAXS curve, the scattering wave vector, q is used and given by the following equation: π θ λ = q 4 sin / (2) where 2θ is the scattering angle and λ = 1.54 Å is the used wave length of X-rays.

Scanning electron microscopy (SEM) and transmission electron microscopy (TEM
Zeta potential of the formulations. The zeta potential measurements of the formulations PF127/TMC/ PEG-HA and PF127 in the sol phase were performed using a ZetaPlus zeta potential analyzer (BIC) over a storage period of 14 days at 30 °C. The samples were diluted 100 times in de-ionized water before the measurements.  www.nature.com/scientificreports www.nature.com/scientificreports/ both formulations was 13.5 wt% of the total weight of hydrogel formulations, and both samples showed a gelation temperature of 37 °C. The dynamic viscosity values of the PF127/TMC/PEG-HA and PF127 hydrogels at 37 °C were 7.02 Pa.s and 5.6 Pa.s, respectively. The higher dynamic viscosity value of PF127/TMC/PEG-HA than that of PF127 at 37 °C was due to the addition of TMC and PEG-HA in the PF127-based hydrogel system and that created enhanced inter-micellar interactions by increasing hydrophobicity of the whole system 1 . Moreover, the higher dynamic viscosity value of the PF127/TMC/PEG-HA hydrogel at the point of gelation suggested better resistance to deformation under stress. In the literature, PF127 is reported to act like a non-Newtonian fluid in the gel phase (37 °C) and the dynamic viscosity values varied as a function of shear rate 52 . PF127 in the sol-phase acted like a Newtonian fluid 52 . In the present study, the dynamic viscosity values of formulations in sol phase at 5 °C were 0.40 Pa.s and 0.32 Pa.s for PF127/TMC/PEG-HA and PF127, respectively, which are in good agreement with previously reported values in the literature 52 .

Hydrogel degradation study in vitro.
The Tube inversion method was applied to visualize gelation with temperature change by measuring the flowability of PF127/TMC/PEG-HA [ Fig. 4(II)] and the system showed reversible gelation with a sol-gel transition at 37 °C. PF127 chains along with two other compounds (TMC and PEG-HA) use temperature as trigger and form hydrogels through reversible physical linking of the polymer chains 53 . The hydrogels revert to the solution state after the thermal stimulus is removed. The inter-micellar aggregation of thermo-responsive polymers near the gelling temperature shows a positive entropy change (ΔS) and a negative free energy change (ΔG) of aggregation 54 . The water-water associations cause an increase in entropy known as the hydrophobic effect, which is the guiding force for gel formation at LCST 54 .
Rheology study. The rheological parameters of the formulations namely, complex viscosity (Fig. 5I), storage modulus (Fig. 5II), and loss modulus (Fig. 5III) have been illustrated as the function of temperature. As shown in Figure 5, all the rheological parameters are highly dependent on temperature and the sol-gel conversion of both formulations has been found to be started near 30 °C as obtained from the inflection point of the graphs. After gel formation, all the rheological parameters of both formulation are found to be much higher than their sol phase. The complex viscosity (I), storage modulus (II), and loss modulus (III) values of PF127/TMC/PEG-HA are significantly higher than those of PF127 in their gel state indicating that the modified PF127 hydrogel with TMC and PEG-HA are mechanically stronger than PF127 hydrogels. The improved rheological parameters of PF127/TMC/ PEG-HA hydrogel might occur due to strong inter-micellar interactions and TMC and PEG-HA would possibly enhance the stability of micelles formed by PF127 in the hydrogel.  www.nature.com/scientificreports www.nature.com/scientificreports/ Swelling study (pH-responsiveness of hydrogel formulations). The swelling of freeze-dried hydrogel particles under acidic pH (pH 5.4) using 0.1 M acetate buffer against time (Fig. 6) at 30 °C revealed that the gel particles formed by PF127/TMC/PEG-HA in the swelled state resisted complete dissolution until 30 min and then, it started to be dissolved in the media to degrade completely just after 2 h. In contrast, PF127 in the swelled state resisted degradation until 15 min, and was completely degraded after 1 h. The hydrophilicity of PF127 was increased under acidic conditions due to the polymer-water interaction, which resulted in rapid dissolution of the gel structure 21 . The pH-responsiveness of the freeze-dried hydrogel particles was modified after adding TMC and PEG-HA in the system, and the enhanced inter-micellar interaction of the modified hydrogel system rendered resistance to the acid mediated degradation/dissolution of gel particles for a longer time.
The swelling of gel particles under neutral pH (7.4) using 0.1 M PBS at 30 °C revealed that the hydrogel particles remained swelled for 4 h without any evidence of dissolution. The swelling ratio of PF127/TMC/PEG-HA (5.01) at pH 7.4 after 4 h was higher than that of PF127 (4.23) indicating that the interconnected porous network of PF127/TMC/PEG-HA accumulated more water molecules in their structure in the swelled state (Fig. 6).  www.nature.com/scientificreports www.nature.com/scientificreports/

Hydrogel degradation study under mechanical stirring (mechanical stability test). The hydrogel
degradation study of PF127/TMC/PEG-HA and PF127 under mechanical stirring (70 rpm) for 14 days in neural pH conditions indicated that PF127/TMC/PEG-HA resisted gel degradation to a higher extent than did the PF127 hydrogel (Fig. 7). The hydrogel made of PF127/TMC/PEG-HA showed a remaining mass of 45.4% after mechanical stirring for 14 days, while the PF127 system showed higher mass loss after 14 days (remaining mass of 39.2%). Therefore, the mechanical stability of the hydrogel was increased after the addition of TMC and PEG-HA to PF127 polymer chains in the formulation because a more interconnected hydrogel structure was formed by this modification.

SEM and TEM study of hydrogel formulations.
The SEM images of gallic acid loaded PF127 (Fig. 8A,B) and PF127/TMC/PEG-HA (Fig. 8C,D) hydrogels in the freeze-dried state show agglomerated porous structures with irregular pore sizes after freeze drying, as the removal of water caused all interconnected networks in the hydrogel to stack. Nevertheless, the image of PF127/TMC/PEG-HA (magnified blue box of Fig. 8C) in Fig. 8D exhibits a better interconnected network with more distinctive pores than that of PF127 alone (magnified red box of Fig. 8A) as inter-micellar interactions of PF127 chains were enhanced after modification with TMC and PEG-HA 55 . The loaded drug inside the hydrogel was relatively more favorably distributed inside the gel structure of PF127/TMC/PEG-HA and the sustained and controlled drug release was better in the modified PF127system due to the more interconnected porous structure in the original formulation.
The TEM images of PF127 (Fig. 8E) and PF127/TMC/PEG-HA (Fig. 8F) formulations in the sol state show micellar aggregates/interconnected micelles varying from 100 to 1000 nm in size which appear on the images as granules of varying shapes as indicated by the red dotted circles in Fig. 8E and the blue dotted circles in Fig. 8F. As shown in Fig. 8, the micelles formed in PF127/TMC/PEG-HA (F) were more compact and stable than those made of PF127 only (E). The micelles were mainly formed by PF127 polymer chains, and TMC and PEG-HA in the PF127/TMC/PEG-HA formulation influenced the formation of stable and compact inter-micellar structures or micellar aggregates of PF127 through the hydrophobic interaction. Therefore, the delivery system made of PF127/ TMC/PEG-HA hydrogel could show good drug release due to their stable inter-micellar structures.
SAXS study of hydrogels. The freeze-dried form of the hydrogels coated on the glass plate was placed parallel to the sample holder on the instrument for reflection SAXS. As shown in Fig. 9, the values of I (a. u.) of the hydrogel samples were plotted against q (Å −1 ). SAXS in reflection mode is done when X-ray hits a flat sample almost parallel to the surface and gives the idea of inhomogeneity of the hydrogel network 56 . The plot of I (a. u.) vs q (Å −1 ) in Fig. 9 has shown peak appearance at q = 0.02 Å −1 which is observed in hydrogels indicating the presence of frozen inhomogeneity and this is due to the presence of highly electron density crystallized region originating from the inherent network defects of hydrogels 57 .
Zeta potential of hydrogel formulations. The zeta potential of the formulations in the sol phase was measured for 14 days to monitor the stability of the components in the system during storage. The surface charge of the components in the formulation determines their stability, solubility and clearance 58 , and the surface charge of a component is measured by the zeta potential. Both PF127 and PF127/TMC/PEG-HA formulations with drug (gallic acid) in the sol phase exhibited negative zeta potential values (Table 1). PF127 in the sol phase with loaded gallic acid showed a negative zeta potential of −18.7 mV ± 6.1 at 30 °C and PF127/TMC/PEG-HA with drug (gallic acid) in the sol phase showed a negative zeta potential value of −16.3 mV ± 5.9. As shown in Table 1, the zeta potential values of PF127 and PF127/TMC/PEG-HA were −18.7 mV ± 6.5 and −14.6 mV ± 7.7, respectively, after 14 days of storage at 30 °C. Therefore, all components in the formulations were uniformly distributed in the system without precipitation as no significant change in zeta potential values was found for both formulations over the storage period. Moreover, the surface charges of the micelles are not changed significantly in PF127/TMC/ www.nature.com/scientificreports www.nature.com/scientificreports/ PEG-HA formulation as TMC and PEG-HA are oppositely charged and therefore, the resultant zeta potential of PF127/TMC/PEG-HA is similar to the zeta potential of formulation with PF127 only.
FTIR study of drug loaded hydrogels. Figure    Cumulative release study. The release study was performed to examine the release of gallic acid from PF127/TMC/PEG-HA and PF127 hydrogels in 0.1 M PBS (pH 7.4) and at 37 °C (Fig. 11). Both formulations showed a burst release of the drug (gallic acid) at an initial stage (within 5 h) with 64.60% ± 1.112 and 50.31% ± 0.411 drug release for PF127/TMC/PEG-HA and PF127 hydrogels, respectively. After 5 days, a cumulative drug release of 87.61% ± 1.112 and 75.20% ± 0.850 was registered for PF127/TMC/PEG-HA and PF127 hydrogels, respectively, indicating that the modified hydrogel system made of PF127 with TMC and PEG-HA worked better as a drug delivery system. The morphological changes of PF127/TMC/PEG-HA, such as enhanced inter-micellar interactions and a well-formed porous network structure improved the drug release under neutral pH conditions. Figure 11 The cumulative release values of drug (gallic acid) from the hydrogels are fitted with different kinetic rate models (Fig. 11), and the rate constants of different rate models for all hydrogel varieties are tabulated in Table 2. The fitting of release data to various rate models are expressed in R 2 values (Fig. 11).
The zero-order rate model is given by the equation:  www.nature.com/scientificreports www.nature.com/scientificreports/ where Q t is the cumulative amount of drug release from the hydrogel at time t (h), Q o is the initial amount of drug loaded into the hydrogel, and k 0 is the zero-order rate constant (sec −1 ). The drug release rate by zero-order rate equation is independent of the initial drug amount loaded into the hydrogel.
The first-order rate model is given by the following non-linear form: where Q t is the cumulative amount of drug release from the hydrogel at time t (h), Q o is the initial amount of drug loaded into the hydrogel, and k 1 is the first-order rate constant (sec −1 ). The drug release rate by first-order rate equation is dependent on its concentration (the initial drug amount loaded into the hydrogel).
Higuchi rate equation suggests release of drug from hydrogels by diffusion method and the non-linear form of the Higuchi rate equation is: where Q t is the cumulative amount of drug release from the hydrogel at time t (h), and k H is the Higuchi constant (sec −0.5 ).
The fitting of release data to the different rate models (Fig. 11) indicates that drug release rate of the hydrogel formulations closely follows first-order rate equation as both gallic acid loaded PF127 and PF127/TMC/PEG-HA hydrogels show R 2 values of 0.990 which are higher than those obtained from other rate models used in this study. Therefore, the rate of drug release from the hydrogels is dependent on the initial concentration of drug loaded into the hydrogels. As found in Table 2, k 1 (sec −1 ) is lower than other rate constants for both varieties of hydrogels, and so, the sustained release of drug from the hydrogel follows more closely first-order rate model than any other rate model used here.

conclusions
A dual-responsive (pH/temperature) hydrogel (PF127/TMC/PEG-HA) was loaded with gallic acid for possible application in textile-based transdermal therapy especially for the treatment of AD. The hydrogel system made of thermo-responsive polymer PF127 was added with two pH-responsive chemically synthesized compounds, namely, TMC form chitosan and PEG-HA from HA, and the synthesis of the compounds was confirmed by 1 H-NMR and FTIR. The thermo-responsiveness of the hydrogel system was confirmed by a dynamic viscosity study and the tube inversion method, and the PF127/TMC/PEG-HA formulation underwent a sol-gel  www.nature.com/scientificreports www.nature.com/scientificreports/ transformation at 37 °C. The change of complex viscosity and storage/loss moduli of formulations with temperature indicated that the rheological parameters are temperature dependent and sol-gel conversation was started near 30 °C for both varieties. Furthermore, all the rheological parameters are higher for PF127/TMC/PEG-HA than PF127 in the hydrogel state indicating higher mechanical stability of modified PF127 hydrogel system with TMC and PEG-HA. The pH-responsiveness of the freeze-dried hydrogel material indicated complete dissolution in acidic pH (5.4), and considerable stability at pH 7.4. The microstructure analysis by reflection SAXS indicated similar type of frozen inhomogeneity of hydrogel formulations and the frozen inhomogeneity of hydrogels was found at q = 0.02 Å −1 . The FTIR study of the drug-loaded formulation in the freeze dried state indicated that the drug was properly loaded into the gel structure of the formulation and that the release was much higher for the PF127/TMC/PEG-HA hydrogel system than that for the PF127 hydrogel. SEM and TEM of the formulations indicated that the modified hydrogel system showed a good interconnected network and that the micelles were well formed, respectively, which might cause PF127/TMC/PEG-HA to perform better performance of as a drug carrier for gallic acid. The sustained drug release from the hydrogel formulations followed more closely first-order rate model than other rate models.
The main focus of future publications will be to load traditional Chinese medicine in the dual responsive hydrogel system and apply it to textile-based transdermal therapy. The forthcoming research will comprise drug release studies, other physico-chemical characterizations of the hydrogel system, and, most importantly, clinical trials for successful application in the treatment of AD by textile based transdermal therapy.