Cold Chain-Free Storable Hydrogel for Infant-Friendly Oral Delivery of Amoxicillin for the Treatment of Pneumococcal Pneumonia.

Pneumonia is the major cause of death in children under five, particularly in developing countries. Antibiotics such as amoxicillin greatly help in mitigating this problem. However, there is a lack of an infant/toddler-friendly formulation for countries with limited clean water orr electricity. Here, we report the development of a shear-thinning hydrogel system for the oral delivery of amoxicillin to infant/toddler patients, without the need of clean water and refrigeration. The hydrogel formulation consists of metolose (hydroxypropyl methylcellulose) and amoxicillin. It preserves the structural integrity of antibiotics and their antibacterial activity over 12 weeks at room temperature. Pharmacokinetic profiling of mice reveals that the hydrogel formulation increases the bioavailability of drugs by ∼18% compared to that with aqueous amoxicillin formulation. More importantly, oral gavage of this formulation in a mouse model of secondary pneumococcal pneumonia significantly ameliorates inflammatory infiltration and tissue damage in lungs, with a 10-fold reduction in bacterial counts compared to those in untreated ones. Given the remarkable antibacterial efficacy as well as the use of FDA-regulated ingredients (metolose and amoxicillin), the hydrogel formulation holds great promise for rapid clinical translation.

Introduction

Over 1 million children die annually as a result of pneumonia, far exceeding the combined death toll of malaria, measles, and acquired immunodeficiency syndrome.[1] Although pneumonia affects children worldwide, the problem is particularly concentrated in the top fifty low- and middle-income countries in Africa and Southeast Asia.[2,3] Common antibiotics exhibit great potential in halting this alarming situation. However, the lack of clean water and electricity in developing areas makes this fight highly challenging for humanitarian agencies and local/mobile healthcare workers.[4,5] For example, amoxicillin is an inexpensive and highly effective drug used against a number of pathogens, including Streptococcus pneumoniae, the most common cause of community-acquired pneumonia.[6,7] It is one of the most commonly prescribed antibiotics for children, being on the list of essential medicines recommended by World Health Organization.[3] Amoxicillin is commercially produced as a powder and readily available in the market in a variety of drug formulations including syrups, reconstituted aqueous suspensions, capsules, chewables, and tablets.[8,9] Liquid-based formulations of amoxicillin require stringent storage conditions (i.e., refrigeration) to prevent activity loss due to degradation.[10−13] The use of nonliquid-based forms (i.e., capsules, chewables, and tablets) constitute a viable option to prevent amoxicillin degradation due to hydrolysis.[14,15] However, such options are not quite convenient for infant/toddler patients who cannot safely swallow or chew hard substances.[16,17] Thus, there is a great need for an infant/toddler-friendly formulation for patients in the low-income countries, with limited clean water and electricity.

Hydrogels are three-dimensional and cross-linked networks of water-soluble polymers. They are highly biocompatible due to the high water content (i.e., typically >90%) and physiochemical similarity to the native extracellular matrix.[18−21] Recently, they have been used to deliver antibiotics for applications like wound dressing and local disinfections.[22−24] Unfortunately, to our best of knowledge, the utilization of hydrogels for oral delivery of antibiotics is yet to be explored.[25,26] Here, we report the oral delivery of amoxicillin, with a shear-thinning hydrogel, for the treatment of pediatric pneumonia (Figure ). The hydrogel is made of metolose, a biocompatible derivative of (methyl) cellulose, which is widely utilized as an additive in the food industry.[27−29] Pharmaceutical-grade amoxicillin is physically and homogeneously incorporated within the hydrogel matrix (e.g., 3 wt % metolose, 2.5 wt % drug, 94.5 wt % water). The shear-thinning behavior of hydrogel facilitates the administration of antibiotics with a smart squeezing tube or syringe. More importantly, the metolose matrix preserves the structural integrity of amoxicillin for at least 12 weeks at room temperature, offering a viable option for drug storage in areas lacking access to electricity or refrigeration. Further, hydrogel would mask the unpleasant smell/taste of amoxicillin, improving the compliance of pediatric patients. Finally, the hydrogel components (i.e., metolose and amoxicillin) have been regulated by Food and Drug Administration (FDA) and manufactured according to the pharmaceutical standards (e.g., United States Pharmacopeia, European Pharmacopoeia, and Japanese Pharmacopoeia).[30] There should be less hurdles in clinical translation of the hydrogel formulation.

Figure 1

Schematic diagram of the metolose-based shear-thinning hydrogel formulation for oral delivery of amoxicillin from a dispenser device for pediatric patients.

Experimental Section

Materials

Metolose (90SH-100000SR) was acquired from Shin-Etsu Chemical. Amoxicillin, cefadroxil, monobasic sodium phosphate, dibasic sodium phosphate, methanol, acetonitrile, paraformaldehyde, formalin, hematoxylin, and eosin were purchased from Sigma-Aldrich. The influenza virus A/Puerto Rico/8/34 H1N1 strain (PR8) was obtained from the American Type Culture Collection and propagated in embryonated eggs. Viral titers were determined by plaque assay. Serotype 19F S. pneumoniae, obtained from a clinical isolate from Singapore and cultured in a brain–heart infusion Broth (Sigma-Aldrich), was added with 5% heat-inactivated fetal bovine serum, under anaerobic conditions in an incubator (Thermo Fisher) at 37 °C until the mid-logarithmic phase.

Preparation and Examination of the Amoxicillin Hydrogel Formulation

The powders of metolose and amoxicillin were weighed and mixed with 0.1 mol/L phosphate buffer (pH 5, 6, or 7), followed by being vortexed for 5 min. The final concentration of metolose and amoxicillin in the hydrogel were, 1–5 wt % (∼10–50 mg/mL) and 2.5 wt % (∼25 mg/mL), respectively. The final pH of the hydrogels was monitored with a pH-indicator strip (MColorpHast; Merck KGaA). The pH of hydrogels prepared using phosphate buffers of pH 5, 6, and 7 were around pH 4, 5, and 6, respectively. The hydrogels were then placed at 4 °C in a refrigerator for at least 48 h to remove the air bubbles and reach an equilibrium. To observe the hydrogel networks under microscopy, hydrogels were frozen in liquid nitrogen and lyophilized. Then, the dried metolose–amoxicillin networks were cryo-sectioned and sputter-coated with platinum for scanning electron microscopy (SEM) with a JSM-6700F field-emission SEM system (JEOL).

Viscosity Measurement and Viscoelastic Property of Hydrogel

Viscosity measurement was performed on an AR2000ex rheometer (TA Instruments) under continuous flow mode using a parallel plate (25 mm in diameter), with a truncation height of 500 μm. The viscosity of hydrogels, with varying weight percentages of metolose (1–5 wt %), was recorded, with the change of the shear rate (0.1–100 s–1) at room temperature (25 °C). Then, the zero-shear viscosity was calculated using a Williamson viscosity versus shear rate model available in the software provided by the manufacturer (Rheology Advantage Data Analysis, version 4.7; TA Instruments). The viscosity of metolose hydrogels, with varying amounts of amoxicillin (0–37 mg/mL), was measured using a similar experimental setting, again followed by the calculation of zero-shear viscosity. In a typical shear recovery experiment, the change of viscosity was monitored after hydrogel was first exposed to a low shear rate of 0.1 s–1 for 10 min, followed by exposure to a high shear of 100 s–1 for 10 min. Finally, the applied shear force was reduced to 0.1 s–1, and the recovery of viscosity over time was recorded.

The viscoelastic properties of the hydrogels were measured by performing strain sweep experiments in the oscillation mode, using a similar parallel plate as used in the flow tests (AR2000ex rheometer). The frequency was set at 1 Hz, and the storage modulus (G′) and loss modulus (G″) values were recorded by sweeping tests changing the strain from 0.01 to 10. The measurement parameters were determined to be within the linear viscoelastic region, on the basis of the preliminary experiments made before collecting the actual data in a systematic manner.

Stability Test of Amoxicillin in Hydrogel Matrix

The amoxicillin hydrogel formulation was prepared as described above. Then the glass vial containing hydrogel was placed in the dark at room temperature over 12 weeks. At specific time points (i.e., 0, 2, 4, 8, 10, and 12 weeks), aliquots of hydrogels were collected and dissolved in deionized water at a final concentration of 5 mg/mL (n = 3) of hydrogel. The chemical integrity of amoxicillin was examined with an Agilent 1100 high-performance liquid chromatography (HPLC) setup equipped with a UV detector (Agilent Technologies). The mobile phase, composed of a mixture of phosphate buffer (0.01 mol/L, pH 4.8) and acetonitrile (96:4 v/v), was used under a flow rate of 0.5 mL/min through the column (Poroshell 120 EC-C18, 2.7 μm, 100 × 4.6 mm; Agilent Technologies) at room temperature. Amoxicillin was detected using a UV detector at 254 nm. The percentage of intact drugs was calculated by the normalization of the peak area of intact amoxicillin (∼5 min of elution time), with the areas of all peaks in the HLPC spectra. The repetitiveness of hydrogel formulation was also examined by running HPLC analyses of five batches of separately prepared hydrogels, at a gel concentration of 5 mg/mL in water. The drug concentration was then normalized to the values from theoretical calculation, assuming the drug was homogenously distributed in the gel matrix.

To examine the effect of the ionic strength on amoxicillin stability, metolose hydrogels were prepared in a 0.1 mol/L phosphate buffer (pH = 6), with varying NaCl concentrations (i.e., 0, 0.3, 1, and 3 N). After 6 weeks of storage at room temperature and without shielding from day light (to accelerate degradation of the drug), the chemical integrity of amoxicillin was examined with HPLC, as described above.

Release Profile of Amoxicillin from the Hydrogel Formulation in Vitro

The drug-release profiles of the amoxicillin hydrogel formulation were examined in 0.1 N NaCl buffer under varying pH (i.e., pHs 2.5, 4, 5.5, 7, and 9, adjusted using 0.1 N HCl and 0.1 N NaOH; n = 3). A portion of hydrogels (∼25 mg) was incubated with 5 mL of the release buffer, with gentle stirring. At selected time points (i.e., 5, 15, 30, and 60 min), supernatants were collected, neutralized with 10× PBS buffer, and analyzed by HPLC, as described above. The percentage of cumulative release of amoxicillin was then plotted against the release time. In the drug-release studies under standstill conditions, 0.5 mL of amoxicillin-loaded hydrogel, as prepared above, was incubated without any stirring in 10 mL of 0.1 N NaCl under various pHs (i.e., pHs 1.2, 2.5, 4, 5.5, and 7 and adjusted with 0.1 N HCl and 0.1 N NaOH; n = 3). At 1 and 2 h, the supernatant of the release buffer was aliquoted and analyzed with HPLC. The remaining hydrogels were also weighed after carefully draining the aqueous solution in the bottle at each time point. The gel weight was plotted against time.

In Vitro Antibacterial Assay

Amoxicillin hydrogels were prepared as described above using pH 6 phosphate buffer and stored at room temperature for at least 1 month. To quantitatively measure the antibacterial efficacy of the hydrogel formulation, anaerobic culture of S. pneumoniae was carried out in the brain–heart infusion broth (Sigma #53286) at 37 °C for 2 h. The cultures were then treated with metolose hydrogel without amoxicillin, amoxicillin suspension, or amoxicillin hydrogel (n = 3). Serial dilutions of amoxicillin (either in amoxicillin suspension or amoxicillin hydrogel) were prepared to obtain final amoxicillin concentrations in the culture of 0.16, 0.08, 0.04, 0.02, and 0.01 μg/mL. Another 2 h of anaerobic culture at 37 °C was carried out before the optical density of the bacterial culture was measured at 600 nm (OD600) using a Synergy H4 microplate reader (BioTek Instruments) to compare bacterial growth under different treatment conditions. A plain brain–heart infusion broth served as the control.

Pharmacokinetic Study of Amoxicillin in Mice

BALB/c mice (∼20 g in body weight) were divided into two groups (n = 3). The amoxicillin hydrogel formulation (100 μL, 25 mg/mL) or the aqueous suspension of amoxicillin (100 μL, 25 mg/mL) was fed to the mice through gavage using animal-feeding needles (18G × 1.5 in.; Cadence Science). The dosage of amoxicillin was 125 mg/kg for both the hydrogel and suspension formulations. At selected time points (i. e., 10, 30, 50, 70, 90, and 110 min), 300 μL of blood was collected via cardiac punctures. Then, the blood was mixed with 20 μL of sodium citrate (37 mg/mL) and centrifuged (3000g) at 4 °C for 5 min. The supernatant was taken and stored at 4 °C before the analysis of amoxicillin.

The extraction of amoxicillin from the plasma was performed according to a previous report.[31] Briefly, 80 μL of plasma was mixed with 20 μL of PBS (0.01 M), 7.5 μL of cefadroxil (1 mg/mL), and 200 μL of ice-cold methanol. The mixture was vortexed for 10 s and centrifuged (13 000g) at 4 °C for 15 min. The supernatant was filtered by a 450 nm filter and analyzed by HPLC. The mobile phase contained a mixture of a phosphate buffer (0.01 mol/L, pH 4.8) and acetonitrile (95:5 v/v) under a flow rate of 0.5 mL/min through the column (Poroshell 120 EC-C18, 2.7 μm, 100 × 4.6 mm, with a guard column UHPLC guard Poroshell 120 EC-C18, 2.7 μm, 5 mm × 2.1 mm; Agilent Technologies) at room temperature. Amoxicillin and cefadroxil (as the internal standard) were detected by a UV detector at 229 nm. The amoxicillin was quantified by plotting relative ratios of peak heights between amoxicillin and cefadroxil, as a function of amoxicillin concentration.

The PKSolver program, an add-in macro for Microsoft Excel environment, was used to calculate the pharmacokinetic parameters.[32] Using PKSolver, the relative bioavailability of amoxicillin was calculated according to the following equationwhere AUCgel and AUCsuspension are the areas under the curve values obtained for hydrogel formulation and aqueous suspension, respectively, and Dgel and Dsuspension are the dosage of amoxicillin in the hydrogel formulation and aqueous formulation, respectively. The care and use of laboratory animals were performed according to the approved protocols of the Institutional Animal Care and Use Committee (IACUC) at Nanyang Technological University, Singapore (#ARF-SBS/NIE-0302).

Therapeutic Efficacy of the Hydrogel Formulation with the Mouse Pneumonia Model

BALB/c mice (∼20 g in body weight) were anaesthetized using 80 mg/kg of ketamine and 10 mg/kg of xylazine and infected via intratracheal delivery, with a sublethal dose of 5 plaque-forming units of influenza virus (H1N1 PR8 strain). 7 days after the influenza infection, 3000 colony-forming units (CFU) of S. pneumoniae were delivered intratracheally to the mice.[33,34] Following the development of bacterial infection, the mice were fed with amoxicillin aqueous suspension or amoxicillin hydrogel via oral gavage at doses of 125 mg/kg twice on a daily basis for 2 days (n = 5). The mice were then euthanized, and their lung tissues were harvested at the same time point (e. g. day three) for further analyses. Lung lobes were either fixed in 4% paraformaldehyde for histological analysis or snap-frozen in liquid nitrogen and stored at −80 °C for later assays.

Frozen mouse-lung tissues were homogenized in sterile PBS, and the homogenates were normalized to ensure the same protein concentrations, as measured by the Nanodrop 2000 UV–visible spectrophotometer (Thermo Fisher Scientific). To quantify the number of bacterial CFU in the mouse lungs, colony counts were determined using the spread plate method by plating normalized homogenate aliquots at 50 μL per blood agar plate (BD Diagnostics, #221261) after 24 h of anaerobic culture at 37 °C.

Formalin-fixed lung lobes were dehydrated using ethanol and xylene before being embedded in paraffin. Lung-tissue sections were prepared with 5 μm thickness. Hematoxylin and eosin (H&E) staining was performed for analysis of the lung histopathology. Representative images of the H&E staining were taken at multiple fields per sample, and the quantifications of intact alveoli (%) and infiltrating inflammatory leukocytes (ratio, # was normalized with total number of alveoli) were performed according to the histopathological features of alveoli and leukocytes, respectively.

Statistical Analysis

All data are expressed as mean ± standard deviations. Statistical significance between the two groups was determined by the two-tailed Student’s t-test. P < 0.05 was considered statistically significant.

Results and Discussion

Characterization of Amoxicillin-Loaded Metolose Hydrogels

We first examined the influence of metolose concentration on viscosity of hydrogels. Within a concentration range of 1–5 wt %, all hydrogels exhibited the shear-thinning property (Figure A): the higher the shear, lower the viscosity. And, for all samples, the zero-shear viscosity increased with increasing concentration of the polymer (Figure A,B). Although a higher viscosity may facilitate the control of the drug dosage and administration speed, we observed that a significant amount of air bubbles was trapped inside the hydrogel when the metolose concentration was 4 or 5% (Figure B inset). Thus the hydrogel with 3% metolose was chosen for the encapsulation and delivery of amoxicillin.

Figure 2

Effect of metolose and amoxicillin concentrations on viscosity of hydrogels. (A) Viscosities of hydrogels prepared with various concentrations of metolose (1–5 wt %) under the changing shear rates; (B) zero-shear viscosity of metolose (1–5 wt %); (C) viscosities of 3 wt % metolose hydrogel, with varying amount of amoxicillin (amox); (D) zero-shear viscosity of 3 wt % metolose hydrogel, with varying amounts of amoxicillin (0–37 mg/mL). (Inset figure seen in (B) shows the images of hydrogels with various concentrations of metolose studied.)

Because the concentration of amoxicillin might also influence hydrogel, the shear-thinning properties of hydrogels containing various amounts of amoxicillin were examined as well (Figure C). Interestingly, all of them showed similar zero-shear viscosity and shear-thinning properties (Figure D). Thus, amoxicillin did not show an influence similar to that of the gelator (i.e., metolose) on the viscosity of the hydrogel. Considering that commercially available amoxicillin syrup is packed at 25 mg/mL, the amoxicillin hydrogel we demonstrate here has the same concentration as that of marketed products. To ensure accuracy of the drug dosage in hydrogel formulation, we examined the repetitiveness of drug concentration in hydrogels prepared in different batches through HPLC analyses. Amoxicillin concentration did not show a significant difference in hydrogels prepared in five separate batches (Figure S1, Supporting Information), confirming the reproducibility of the hydrogel formulation.

SEM study revealed that the rod-shaped amoxicillin particles were evenly distributed in the porous network of hydrogels (comprised of amoxicillin 2.5 wt %, metolose 3 wt %, and water 94.5 wt %), without any observable aggregation (Figure A,C). Comparing with that of blank metolose hydrogel (Figure B,D), the amoxicillin-loaded hydrogel showed an increase in matrix porosity that was consistent with previous reports on composite hydrogel[35] and could be attributed to physical adsorption of polymer chains on the microsized amoxicillin particles (Figure C).

Figure 3

Characterization of amoxicillin-loaded hydrogel formulation. (A–D) SEM images of metolose hydrogels with (A,C) and without (B,D) 2.5 wt % amoxicillin. (E,F) Viscoelastic property (E) and viscosity recovery (F) of amoxicillin hydrogel formulation (amoxicillin 2.5 wt %, metolose 3 wt %, and water 94.5 wt %). Note that scale bars seen in (A) and (C) and (B) and (D) indicate 100 and 10 μm, respectively.

After observing the morphology with SEM, the next step was to characterize the viscoelastic properties of the hydrogel formulation (Figure E). The G′ values of amoxicillin hydrogel were greater than G″ values, and tan(δ) values were less than “1” when a strain of 0.01 was applied. These suggest that the gel state was successfully maintained in the formulation. With the increase of strain, the G′ values decreased gradually, whereas G″ increased. When strain was larger than 1, G′ was smaller than G″ and tan(δ) (i.e., loss modulus/storage modulus) was greater than 1, when the gel state was destroyed.[36] Such kind of viscoelastic property is indeed desirable in the drug feeding process, particularly for children, because even a gentle squeeze would facilitate the flow of the loaded hydrogel into the mouth easily. A rapid recovery of viscosity was also observed in a shear recovery experiment (Figure F). With a low shear rate at 0.1 s–1, the hydrogels showed a high viscosity (∼400 Pa·s). When the shear rate increased to 100 s–1, the viscosity decreased dramatically to ∼10 Pa·s. When the shear rate decreased from 100 to 0.1 s–1, the viscosity quickly recovered to the original values. The quick recovery of viscosity would preserve the suspension state of the drug particles in the gel matrix after oral administration of the formulation.

Amoxicillin Release Profile of the Hydrogel Formulation

Metolose hydrogels are formed through physical interactions between the metolose chains and water molecules[37−39] that could facilitate the dissipation of hydrogels under agitation. To investigate the dissipating process, we examined the release profile of amoxicillin from the hydrogel formulation under various pH conditions (i.e., pHs 2.5, 4, 5.5, 7, and 9). During release, stirring was used to simulate the food movement and processing in stomach. As shown in Figure A, more than 90% of amoxicillin rapidly released within 30 min under stirring regardless of the pH conditions tested. Furthermore, HPLC spectra of released amoxicillin confirmed that the drug retained its structural integrity during both incorporation and release processes (Figure B).

Figure 4

Effect of pH and ionic strength on the release profile and stability of amoxicillin in hydrogel formulations. (A) Release of amoxicillin in buffers at various pH values (i.e., pHs 2.5, 4, 5.5, 7, and 9). (B) HPLC spectra of released amoxicillin at 60 min. (C) Stability of amoxicillin in hydrogels during storage at room temperature under dark over 12 weeks. (D) Percentage of intact amoxicillin after 6 weeks of storage in hydrogels prepared with varying ionic strengths (0, 0.3, and 1 N NaCl). *P < 0.05; **P < 0.01.

To better understand the drug-release mechanism, we performed hydrogel swelling and drug-release studies under standstill conditions in buffers, with pH values ranging from 1.2 to 7. As the normal digestion process (from oral to the stomach to the small intestine) takes 1–2 h, the swelling and drug-release studies were limited to 2 h. As shown in Figure S2A, the weight of all hydrogels increased to over 130% after 1 h, indicating the swelling of hydrogels in aqueous buffers. However, we didn’t observe a significant difference among the swelling of hydrogels under different pH values (i.e., pHs 1.2, 2.5, 4, 5.5, and 7), which could be attributed to the fact that metolose does not contain any charged moiety and is neutral under all pH conditions. In comparison, the drug-release profiles were different under varying pH (Figure S2B); amoxicillin release was clearly faster at low pHs (1.2 and 2.5), which was possibly due to the increased solubility of amoxicillin under these conditions.

Stability of Amoxicillin in Hydrogel Formulation at Room Temperature

It is of great importance to highlight that amoxicillin molecules maintained their structural integrity in the metolose hydrogel when stored at room temperature conditions, without the need for refrigeration (i.e., cold chain). As reported previously, the β-lactam ring in amoxicillin could hydrolyze in water.[40] This should cause a shift in the elution peaks and a decrease in the peak intensities in HPLC.[41,42] Thus, we examined the stability of three hydrogel formulations of the same metolose and amoxicillin content prepared under various pH (i.e., pHs 4, 5, and 6), taking the commercially available amoxicillin syrup as a control. All of the formulations were stored in glass vials at room temperature and in darkness for a period of 3 months. As shown in Figure C, amoxicillin was most stable in the hydrogel prepared at pH 5, and over 90% of drugs were still intact over 12 weeks (3 months). In contrast, from week 6 onward, amoxicillin displayed a significant degradation in hydrogels prepared at pHs 4 and 6. Importantly, for the amoxicillin syrup, over 30% of degradation took place when it was stored under the same condition. As the isoelectric point of amoxicillin is 4.7,[43] we should expect the least net charge at pH 5 to lead to the lowest solubility in water. Further, the hydrophobic surface of the amoxicillin particles is available for the adsorption of hydrophobic domains of metolose (type 90SH), that is, the methoxy and hydroxypropoxy groups, with a degree of substitution at 1.4 and 0.2%, respectively.[30]

The adsorption of metolose to amoxicillin particles takes place plausibly due to the intermolecular forces (i.e., van der Waals forces and hydrogen bondings) between metolose and amoxicillin in a pH and ionic strength dependent manner.[44−46] To better interpret the importance of such interactions in our case, we have prepared amoxicillin-loaded metolose hydrogels with varying ionic strengths by incorporating sodium chloride at varying concentrations (0, 0.3, 1, and 3 N) at a fixed pH of 5. We observed that metolose precipitated in 3 N NaCl and failed to form hydrogel, plausibly due to changes of hydrogen bonding and salting-out of hydrophobic domains of metolose. The mixture of amoxicillin and metolose formed hydrogels in the other NaCl solutions (i.e., 0, 0.3, and 1 N). When stored at room temperature, over 90% of amoxicillin showed structural integrity (as measured by HPLC) in metolose hydrogels, with NaCl concentrations of 0 and 0.3 N, respectively (Figure D). In contrast, we observed a significant decline in the percentage of intact amoxicillin (83.6 ± 3.8%) when 1 N NaCl was used in hydrogel. These results suggested that intermolecular forces played an important role in the stabilization of amoxicillin via physical adsorption of the metolose chains on drug particles. This may be related to the drug–metolose interaction, which stabilizes the β-lactam ring of amoxicillin. Further study is being undergone to uncover this mechanism.

In Vitro Antibacterial Activity of the Amoxicillin Hydrogel Formulation

Next, we tested the antibacterial performance of amoxicillin hydrogel using the serotype 19F strain of S. pneumoniae. The inhibitory activities of amoxicillin in both the hydrogel and suspension formulations were quantified at varying drug concentrations (0.01–0.16 μg/mL) using the broth culture method. The blank metolose hydrogel (amoxicillin-free) did not display any antibacterial activity, whereas the amoxicillin-loaded hydrogel exerted antibacterial activity (P > 0.05) similar to that of the amoxicillin suspension at all concentrations that were tested (Figure ). These results confirmed the retention of the antibacterial activity of amoxicillin in the metolose hydrogel.

Figure 5

In vitro antibacterial activities of amoxicillin suspension or hydrogel formulation against the growth of S. pneumoniae.

Pharmacokinetics of Amoxicillin Administered through the Hydrogel Formulation

Through HPLC, we examined the plasma concentration of amoxicillin at different time points after the oral administration of aqueous suspension and hydrogel formulation to BALB/c mice. Both the terminal half-life (t1/2) and maximum concentration of drug in the plasma (Cmax) were comparable for aqueous suspension and hydrogel formulation (Table ). The time when the plasma concentration reached maximum (Tmax) was 30 and 50 min for the suspension and hydrogel groups, respectively. The delay of Tmax from 30 min in aqueous suspension to 50 min in hydrogel formulation could be attributed to the time required for dissipation of the hydrogel matrix in the stomach. Importantly, the plasma concentrations of amoxicillin in the hydrogel-treated mice were significantly higher (P < 0.05) than those observed in the suspension-treated mice at time points ranging from 50 to 110 min after the drug administration (Figure ). Further, the area under the curve (AUC) of the hydrogel-treated group was significantly higher. The significant increase of AUC in the hydrogel group was possibly attributed to the combination effects of two factors: the rather short half-life of amoxicillin in mouse (14–17 min)[47−49] and the longer duration for the drug to accumulate in blood in the hydrogel-treated group (the respective Tmax for hydrogel and suspension groups were 50 and 30 min). Because both suspension and hydrogel formulations shared the same dose during administration, the AUC values allow us to compare the bioavailability of the formulations (eq , Section ). Quantitatively, the relative bioavailability of the amoxicillin hydrogel was ∼18% higher than that of the aqueous suspension.

Table 1

Pharmacokinetic Parameters of Plasma Concentration of Amoxicillin in Micea,b

	amoxicillin suspension	amoxicillin hydrogel
t1/2 (min)	14.2 ± 4.1	16.9 ± 1.8
Tmax (min)	30 ± 0	50 ± 0c
Cmax (μg/mL)	35.2 ± 0.2	34.6 ± 2.7
AUC (μg/mL·min)	1461.9 ± 148.5	1724.1 ± 70.6c

Note: The abbreviation t1/2 refers to the terminal half-life of amoxicillin in the plasma, Tmax refers to the time when plasma concentration of the drug reached the maximum, Cmax refers to the maximum plasma concentration of the drug, and AUC refers to area under the curve for the plasma concentration vs time curve from 10 to 110 min.

Mice received a single oral gavage of aqueous suspension or hydrogel formulation, with a dose of amoxicillin at 125 mg/kg. Results are shown as the mean values ± standard deviation (n = 3).

Tmax and AUC of the hydrogel formulation were significantly larger than those of the aqueous suspension (P < 0.05).

Figure 6

Pharmacokinetic profiles of amoxicillin in mice that were orally administered with amoxicillin suspension or amoxicillin hydrogels. *P < 0.05.

In Vivo Efficacy of Pneumonia Treatment Using the Amoxicillin Hydrogel Formulation

Finally, the therapeutic efficacy of amoxicillin hydrogel was evaluated in a mouse model of secondary pneumococcal pneumonia.[34] BALB/c mice were first infected with influenza virus for 7 days followed by intratracheal challenge with S. pneumoniae. After secondary bacterial infection, amoxicillin hydrogel or aqueous suspension was orally administered to mice twice on a daily basis for 2 days. The lungs of the infected mice displayed obvious pulmonary edema without treatment (Figure Aii). Histopathological analyses revealed that the lung alveoli were severely damaged, with infiltration of abundant inflammatory cells (Figure Bii). In contrast, tissue damage was alleviated in the lungs of mice treated with amoxicillin suspension (Figure Aiii,Biii). Most noteworthy was that the lungs appeared to be the least damaged in hydrogel-treated mice (Figure Aiv,Biv). The percentage of intact alveoli in lungs of hydrogel-treated mice was significantly higher than that in the lungs of untreated and suspension-treated mice, respectively (Figure C). The number of infiltrating inflammatory leukocytes in the lungs of hydrogel-treated mice was considerably diminished compared to that in the lungs of untreated or suspension-treated mice (Figure D). The improved therapeutic efficacy of the hydrogel group compared to that of the suspension group may be attributed to the enhanced bioavailability of the hydrogel formulation (Figure ). To quantify the residual bacterial load in the lungs, we homogenized the infected lungs followed by culture on blood agar, with equal amounts of homogenate protein. The mean CFU counts in the lungs of untreated, suspension-treated, and hydrogel-treated mice were 28.0 ± 2.6, 6.7 ± 1.5, and 2.7 ± 0.6, respectively (Figure E). The 1 order magnitude decrease in the lung bacterial load of the hydrogel-treated group proved that the amoxicillin hydrogel was highly efficacious in eradicating S. pneumoniae in the lungs of mice.

Figure 7

Therapeutic efficacy of amoxicillin hydrogel in a mouse model of secondary pneumococcal pneumonia. (A) Representative images of three mouse lungs, (B) H&E staining of lungs in mice: (i) healthy, (ii) infected, (iii) infected with amoxicillin suspension treatment, and (iv) infected with amoxicillin hydrogel treatment. Quantification of intact alveoli (C), infiltrating inflammatory leukocytes (D) and CFU in lungs (E) of mice subjected to various treatments are shown. Arrows indicate the presence of infiltrating inflammatory leukocytes. *P < 0.05; **P < 0.01.

Conclusions and Future Perspective

This report describes the development of a simple yet highly effective hydrogel formulation for the oral delivery of amoxicillin for pneumococcal pneumonia treatment. The hydrogel consists of metolose and amoxicillin and displays a shear-thinning behavior facilitating drug administration for the infant/toddler patients. The hydrogel formulation demonstrates exceptional stability at room temperature and does not require refrigeration. When stored at room temperature for almost 3 months, it preserves the structural and functional integrity of amoxicillin, due to the physical adsorption of metolose on drug particles. Importantly, the hydrogel formulation demonstrates uncompromised antibacterial activity both in vitro and in vivo. Because all components of hydrogels are FDA-regulated and of the pharmaceutical grade, the hydrogel formulation holds great promise for rapid clinical translation and is especially suitable for pediatric patients in low-income countries that lack access to clean water and electricity. In future, such hydrogel formulations can be further extended to encapsulate other antibiotics such as ampicillin and vancomycin for the treatment of medically important bacterial infections.