HPLC-UV Method Validation for Amobarbital and Pharmaceutical Stability Evaluation when Dispersed in a Hyaluronic Acid Hydrogel: A New Concept for Post-traumatic Osteoarthritis Prevention

Abstract

A mitochondrial electron transport chain member complex I inhibitor, amobarbital, can reduce oxidative damage and chondrocyte death, eventually preventing post-traumatic osteoarthritis (PTOA). Viscosupplementation using a crosslinked hyaluronic acid (HA) hydrogel is currently applied clinically for knee OA pain relief. In this work, we utilized the HA hydrogel as a drug delivery vehicle to improve the long-term efficacy of amobarbital. Here we evaluated the pharmaceutic stability of amobarbital when dispersed in a crosslinked HA hydrogel formulated in proportions intended for clinical use. We validated a high-performance liquid chromatography with an ultraviolet detector (HPLC-UV) method following International Conference for Harmonization Q2(R1) guidelines to ensure its suitability for amobarbital detection. The feasibility of this formulation’s drug delivery capability was proven by measuring the release, solubility, and drug uniformity. The amobarbital/HA hydrogel showed comparable amobarbital stability in different biological fluids compared to amobarbital solution. In addition, the amobarbital/HA hydrogel imparted significantly greater drug stability when stored at 70°C for 24 hours. In conclusion, we confirmed the pharmaceutical stability of the amobarbital/HA hydrogel in various conditions and biological fluids using a validated HPLC-UV method. This data provides essential evidence in support of the use of this amobarbital/HA formulation in future clinical trials for PTOA treatment.

1. Introduction

The articular cartilage that makes up a joint provides near frictionless movement as well as alleviates shear stress, however this tissue type is prone to gradually progressive wear and tear damage as well as physical trauma such as that caused by sports activities or military related injuries [1–3]. Unfortunately, research has shown that damage to the cartilage will not heal independently, owing to the tissue’s diminished metabolic activity and avascularity. Without treatment, the damaged cartilage becomes increasingly worse and induces negative effects on surrounding healthy tissue, ultimately descending into osteoarthritis (OA) [1]. OA is a chronic and progressively degenerative disease which affects not only the articular cartilage and subjacent bone of a joint but also the synovium, ligaments, and muscles [4–8]. In the US alone, there are approximately 30.8 million adults who have been diagnosed with OA and the number of adults affected globally is expected to reach 78.4 million by 2040 [9, 10]. In 2015 alone, the annual average healthcare costs directly to US patients for the treatment and management of OA were estimated to reach $21,335 [11].

OA progression has been classified microscopically by the loss of proteoglycans and collagen, both of which lead to the disruption of the extracellular matrix (ECM) and subsequent impairment of the biomechanical attributes of the joint [1, 3]. OA resulting from an injury to the joint, whether it be an intra-articular fracture (IAF), sprain, or damage to the cartilage, account for 12% of cases and are classified as post-traumatic osteoarthritis (PTOA). IAFs that result from an impact injury to a joint account for a significant proportion of knee PTOA [4]. It has been estimated that anywhere from 23–44% of IAFs in the knee will lead to PTOA and there are 14 million people in the US who experience symptomatic OA of the knee [4, 12]. This disease is the primary reason for loss of mobility and pain in patients [13, 14]. As OA progresses in the knee joint, this has a detrimental effect on the health of chondrocytes [5]. Though normal chondrocytes exhibit limited proliferation, rendering it difficult to determine a reduction in their proliferation rates, it has been shown that certain chondrocyte subpopulations have a direct link to OA progression when the cells experience senescence [15]. Chondrocytes in this diseased state synthesize degradative matrix metalloproteases that destroy ECM components and further aggravate the degeneration of the articular cartilage [16, 17].

Current management options for OA are mainly focused on controlling pain, improving joint functionality, and giving patients a better quality of life [7, 18]. Management options that have been used to treat knee OA include physical therapy exercises, weight control, nerve ablation or modulation, transcatheter arterial embolization, the use of pharmaceutical grade therapeutics such as oral and topical nonsteroidal anti-inflammatory drugs or, in more extreme cases, opioids, and less commonly used methods such as injections of platelet rich plasma, mesenchymal stem cells, or ozone gas [7, 14, 19–22]. Other than fracture fixation, there are no available treatments prior to end-stage joint replacement or fusion, and most treatments are aimed at pain management [1, 23]. For these reasons, there is a need to find procedures able to prevent or impede PTOA, either preventatively or in conjunction with standard procedures, capable of undoing the articular cartilage damage [21, 23]. One route of administration shown to be efficacious for PTOA treatment is the intra-articular injection which offers the additional benefits of being a local, direct delivery to the joint which avoids systemic side effects [5, 7, 10, 19]. The substances used in these intra-articular injections include corticosteroids, such as methylprednisolone and dexamethasone; regenerative medicines aimed at halting the degenerative effects of OA; or viscosupplementation therapies [7, 14, 20]. There is no consensus among researchers in the field as to whether viscosupplementation offers a significant improvement for knee OA treatment over other treatment options [24, 25]. Bowman et al. found that the success rate of viscosupplementation alone in alleviating OA-related pain was only 57% [26] and no studies have shown that viscosupplementation is disease modifying, demonstrating the need for improvements. Hyaluronic acid (HA) is a commonly used species for intra-articular injection to the knee and, in some cases, has been shown to restore the viscoelasticity of synovial fluid, reduce pain, increase mobility, and re-establish the protective effects of endogenous HA [13, 14]. HA is a glycosaminoglycan and essential part of the ECM composition synthesized by chondrocytes, fibroblasts, and type B synovial cells and discharged into the synovial fluid [7, 14, 27]. HA is an essential substrate for chondrocyte growth, metabolism, and maintenance of the chondrocyte phenotype and promotes ECM formation [27]. Yoshioka et al. showed that the crosslinked HA hydrogel Gel-One^® remained at the intra-articular injection site in a rat model for 91 days after administration [28], demonstrating its effectiveness as a depot style drug delivery vehicle.

Another avenue to be considered when looking at strategies for mitigating PTOA after an IAF is to focus on the chondrocyte mechanobiology [29–31]. One of the hallmarks of the progression of this disease is the reduction in chondrocyte cell proliferation, a direct result of the imbalance between the use and production of adenosine triphosphate (ATP) [32, 33]. The traumatic impact event and residual increased contact stress experienced in the injured joint leads to atypical activity by the chondrocyte’s mitochondria [33–35]. The death of chondrocytes and excessive production of reactive oxygen species (ROS) by their mitochondria causes an accumulation of oxidative damage and eventual mitochondrial dysfunction in the surviving cell populations all of which are common occurrences that presage PTOA development [15, 33, 36–38]. It has been demonstrated that inhibition of the mitochondrial electron transport chain (ETC) was not only able to conserve chondrocyte viability, but also impede ROS release and thus oxidative stress [30, 36, 39]. The mitochondrial ETC member complex I is an established target for PTOA treatment because it has been shown that direct oxidative damage to the complex results in the increased production of free radicals which is an attribute commonly seen in osteoarthritic cartilage [30, 33, 40, 41]. Amobarbital was first used in 1930 to make catatonic patients responsive, but it has also been used in research as a mitochondrial ETC member complex I inhibitor [23, 42]. Coleman et al. showed that by delivering amobarbital in this capacity, the treatment was efficacious at preserving the composition and structure of the articular cartilage for up to six months after an IAF [23].

Here, we have created a novel formulation for the treatment of PTOA by combining the commercially available crosslinked HA hydrogel, Gel-One^®, with amobarbital to provide a depot style drug delivery system meant to inhibit mitochondrial ROS production [23]. Since the intended route of administration for this formulation would be as an intra-articular injection to the joint, it was necessary to determine the effect that relevant biological fluids, namely human synovial fluid and human plasma, had on amobarbital stability. In this work, we have presented a simple, fast, and isocratic reverse phase high-performance liquid chromatography (HPLC) method using an ultraviolet (UV) light detector able to reliably measure amobarbital concentrations in aqueous methanol, human synovial fluid, and human plasma. Due to the limited clinical application of amobarbital in recent years, the availability of an accurate and simple method to quantify the drug is also limited [43–45]. Our method is able to detect the drug without the need for more complicated systems such as mass spectroscopy, gradient elution, electrospray ion source or buffers [42–49]. Following the International Conference for Harmonization (ICH) Q2(R1) guidelines, we first demonstrated the validity and selectivity of our HPLC-UV method to quantitatively measure amobarbital in aqueous methanol, human synovial fluid and human plasma [46]. We then demonstrated the release profile achieved when testing two amobarbital/HA hydrogel formulations compared to amobarbital solution. The purpose of this release study was to determine how the release kinetics of the drug were altered when dispersed within the HA hydrogel matrix compared to free drug solution. We next characterized the uniformity of the drug distribution in the amobarbital/HA hydrogel formulation. Our next goal with this project was to evaluate the comparative stability of amobarbital when encapsulated in the HA hydrogel. Since previous research has already described the typical mechanism by which amobarbital and other barbiturates are metabolized [47–49], we were focused on whether the amobarbital concentration was affected differently when the drug was dispersed in the HA hydrogel versus in free drug solution at various storage conditions. The purpose of this stability study was to evaluate the short-term stability of the formulation immediately after injection into the joint capsule compared to samples combined with water using an in vitro model. This was done by measuring the drug content of free drug solution versus the amobarbital/HA hydrogel solution in the presence of water, human synovial fluid, and human plasma at biologically relevant conditions. The end goal of this formulation is to be used in the clinic where the physician will mix the amobarbital/HA hydrogel before injection into the patient. Therefore, we needed to know what the ideal storage conditions would be for this novel formulation. Although the individual components of the amobarbital/HA hydrogel formulation are commercially available and therefore have their own ideal storage conditions, we needed to evaluate the novel formulation to determine its ideal storage conditions. To do this, we performed an accelerated stability test using a forced degradation model and following United States Pharmacopeia (USP) guidelines [50, 51].

2. Materials and Methods

2.1. Materials and Chemicals

All of the chemicals that were used were HPLC grade. Methanol (MeOH) and acetonitrile were purchased from Thermo Fisher Scientific (Waltham, MA). Trifluoroacetic acid was purchased from Sigma-Aldrich (St. Louis, MO). Nanopure water was obtained from a Barnstead™ Nanopure™ Diamond purification system (Thermo Fisher Scientific). Amobarbital (Amytal^® sodium) was obtained from Bausch Health (Bridgewater, NJ) and barbital was purchased from Cayman Chemical (Ann Arbor, MI). The HA hydrogel was purchased from Zimmer Biomet (Gel-One^®; Warsaw, IN). Hank’s Balanced Salt Solution (HBSS) was purchased from Thermo Fisher Scientific. Invitrogen™ UltraPure™ Distilled Water was purchased from Life Technologies (Carlsbad, CA) and is denoted as “ultrapure distilled water” in the manuscript. Pooled human complement plasma was purchased from Innovative Research (Novi, MI) and human synovial fluid was obtained from patients undergoing ACL replacement surgery with approval from the Institutional Review Board of the University of Iowa.

2.2. Instrumentation

HPLC sample measurements were made using an Agilent 1100 Series HPLC station coupled with a UV-Vis diode array detector (Agilent Technologies, Santa Clara, CA). Analytes were weighed using a Mettler Toledo XS104 analytical balance (Mettler-Toledo, Columbus, OH). Extracted drug from human synovial fluid and human plasma was evaporated using a TurboVap LV Evaporator (Caliper Life Sciences, Hopkinton, MA) under a stream of nitrogen gas. Samples that required heat were stored in either a 625D Isotemp Incubator Oven (Fisher Scientific, Pittsburgh, PA) or a Precision™ Compact Oven (Thermo Fisher Scientific) for 37°C and 70°C temperature storage, respectively.

2.3. Analytical and chromatographic conditions

The chromatographic separation for the in vitro drug release study was performed using a Waters Symmetry RP-C18 column (250 × 4.6 mm, 5 μm, 100Å; Waters, Milford, MA). The mobile phase was pumped at an isocratic flow rate of 1 mL/minute (min), injection volume of 50 μL, and wavelength set to 220 nm. The sample run time was set for 12 min. For all other samples, chromatographic separation was performed using a Synergi™ C18 polar reversed phase column (150 × 2 mm, 4 μm, 80 Å) equipped with a SecurityGuard™ Guard Cartridge (4 × 2 mm, 4 μm, 80 Å; Phenomenex, Torrance, CA). The mobile phase was pumped at an isocratic flow rate of 0.3 mL/min, injection volume of 10 μL, and wavelength set to 220 nm. The sample run time was set for 10 min. Chromatographic separation was performed at room temperature (RT) for all samples injected.

The reason for the use of two HPLC-UV methods was because initially the method validation was performed using the Waters Symmetry RP-C18 column (250 × 4.6 mm, 5 μm, 100Å) and following a modification to a previously published protocol [48]. After the release study samples were analyzed, the column was rendered unusable, so we decided that the hydrophilic Synergi™ C18 polar reversed phase column (150 × 2 mm, 4 μm, 80 Å), which was already available to us, would be a more suitable column for our purposes. Since this column had never been used in the literature previously for the analysis of amobarbital, we performed the entirety of the method validation experiments using this column. Also, since the release study samples did not involve any additional contaminants (such as plasma or synovial fluid) which could produce interfering peaks, we felt that the hydrophilic column was not necessary to successfully analyze those samples.

2.4. Mobile phase preparation

The mobile phase was composed of MeOH and water (MeOH/H₂O) at a ratio of 50:50 (v/v) to which trifluoroacetic acid (0.1% v/v) was added. The mobile phase was then degassed using a Branson 5200 ultrasonic bath (Branson Ultrasonics, Danbury, CT).

2.5. Sample preparation

2.5.1. Stock solutions, working solutions, calibration standards and quality control samples

Amobarbital stock solutions were made in nanopure water to give a final concentration of 100 mg/mL with subsequent dilutions made using MeOH/H₂O at a ratio of 4:5 (v/v). Barbital stock solution was used as an internal standard (IS), similar to previously published protocols [52–58], and was dissolved in MeOH to a final concentration of 500 μg/mL.

Our studies began with the analysis of amobarbital diluted in MeOH/H₂O (4:5 v/v) to validate the HPLC-UV method. Quality control samples of amobarbital were made within a concentration range of 0.5 to 100 μg/mL. These samples were analyzed without the addition of the IS. Method validation in human synovial fluid and human plasma was performed using amobarbital diluted to concentrations ranging from 100–8000 μg/mL. These samples were spiked with equal amounts of the IS stock solution.

2.5.2. Preparation of amobarbital in human synovial fluid and human plasma samples

The HPLC-UV method was validated using amobarbital in human synovial fluid and human plasma samples by combining 250 μL of either human synovial fluid or human plasma with 250 μL of a known concentration of amobarbital to give final amobarbital concentrations ranging from 100–8000 μg/mL. These samples were prepared for HPLC-UV injection by first transferring 25 μL of the sample and mixing this with 225 μL of MeOH/H₂O (4:5 v/v) using a mechanical pipette then vortexing for several seconds using a Vortex-Genie^® 2 mixer (Thermo Fisher Scientific). Next, 100 μL of this MeOH/H₂O sample was spiked with 13.5 μL of the IS (500 μg/mL), mixed with 100 μL of acetonitrile, and left in 4°C fridge for 15 min. The samples were then centrifuged at 21,000 ×g for 5 min and the supernatant transferred to clean microcentrifuge tubes. The solvent was evaporated under a stream of nitrogen using the TurboVap LV. The dried residue was reconstituted in 50 μL of water and 40 μL of methanol, and centrifuged at 21,000 ×g for 5 min. The supernatant from this step was used to measure the drug content.

2.5.3. Preparation of amobarbital/HA hydrogel formulations and free drug samples

Two amobarbital/HA hydrogel formulations were prepared using either 0.5% or 0.75% (w/v) HA hydrogel. Each formulation contained 744 μg of amobarbital to remain consistent with the “free drug” control sample which consisted of 7.44 μL of a 100 mg/mL aqueous amobarbital solution diluted in 1 mL of phosphate-buffered saline (PBS). All samples were made in triplicate under sterile conditions. Drug content for these samples was measured following the same procedure as previously described for amobarbital in human synovial fluid and human plasma samples.

2.5.4. Preparation of stability and forced degradation amobarbital/HA hydrogel and free drug samples

Preparation of the amobarbital/HA hydrogel 0.6% (w/v) formulation and aqueous amobarbital (“free drug”) solutions were carried out in a SterilGard^® III Advance biosafety cabinet (The Baker Company, Sanford, ME) to ensure sterility. Sterilized crimp top compatible 10 mL glass vials were filled with 31 μL of aqueous amobarbital solution (100 mg/mL), 3 mL of HA hydrogel, and 2 mL of HBSS. The gel solution was mixed by drawing up and down using a sterile 10 mL syringe fitted with a hypodermic needle (16 gauge and 1 inch long). The vial was sealed using a crimper and sonicated for 5 min to remove air bubbles. Free drug amobarbital solution was prepared in one of two ways to keep the dilution factor equivalent to the HA hydrogel samples. For the stability studies, free drug solution was prepared by combining 31 μL of the stock amobarbital solution (100 mg/mL) with 3 mL of ultrapure distilled water and 2 mL of HBSS. For the forced degradation studies, the free drug solution was prepared by combining 31 μL of the stock solution with 5 mL of ultrapure distilled water. Drug content for these samples was measured following the same procedure as previously described for amobarbital in human synovial fluid and human plasma samples.

2.6. HPLC-UV method validation

The first step for the method validation process was to measure the amobarbital concentration in MeOH/H₂O (4:5 v/v) solutions. Next, we validated the method in human plasma. In both instances, the method was validated for specificity, linearity, limit of detection, limit of quantification, intra- and interday precision, and accuracy, while the short-term stability was assessed only for the MeOH/H₂O samples. Method validation for human synovial fluid samples was limited to specificity, linearity, limit of detection, limit of quantification, and accuracy. Method validation criteria was based on ICH Q2(R1) guidelines [46].

2.6.1. Specificity

The specificity of the HPLC-UV method was evaluated for MeOH/H₂O, human plasma and human synovial fluid samples in the presence and absence of amobarbital and the IS, barbital.

2.6.2. Linearity

The calibration curve was prepared using triplicate samples of nine MeOH/H₂O (4:5 v/v) amobarbital solutions with concentrations ranging from 0.25 to 100 μg/mL. For human synovial fluid and human plasma samples, the calibration curve was prepared using triplicate samples of five amobarbital concentrations ranging from 100–1000 μg/mL or 200–800 μg/mL, respectively. The drug peak area under the curve (AUC) was plotted against the known sample concentration for the MeOH/H₂O solutions. For the human plasma and human synovial fluid samples, the ratio of amobarbital:IS AUC was plotted against the known amobarbital concentration. The calibration curve data was plotted using GraphPad Prism version 9.0.0 (GraphPad Software, San Diego, CA), and the linearity was measured using the least square linear regression model.

2.6.3. Limit of detection (LOD) and limit of quantification (LOQ)

LOD and LOQ were calculated based on ICH guidelines [46] and using the following formulae:

$$LOD=\frac{3.3\times Standarddeviationoftheresponse}{Slopeofthecalibrationcurve}$$ Formula 1:

$$LOQ=\frac{10\times Standarddeviationoftheresponse}{Slopeofthecalibrationcurve}$$ Formula 2:

2.6.4. Precision

The intraday precision was measured by twice injecting five triplicate samples of the amobarbital in MeOH/H₂O (4:5 v/v) calibration curve samples with concentrations ranging from 5–100 μg/mL on the same day. Triplicate human plasma samples were combined with six amobarbital concentrations to give final sample concentrations ranging from 300–8000 μg/mL and injected twice in the same day. The interday precision was evaluated using five triplicate samples of amobarbital in MeOH/H₂O (4:5 v/v) with concentrations ranging from 5–100 μg/mL and four human plasma samples with amobarbital concentrations ranging from 300–2000 μg/mL which were injected over three consecutive days. The precision, denoted as percent relative standard deviation (%RSD), was measured using the mean AUC and standard deviation (SD) for the MeOH/H₂O samples and the ratio of amobarbital:IS AUC for the human plasma samples.

2.6.5. Accuracy

The accuracy of the HPLC-UV method was determined by running triplicate samples of six concentrations of amobarbital ranging from 100–2000 μg/mL or 200–2000 μg/mL in the presence human synovial fluid or human plasma, respectively. The value for accuracy was denoted as recovery (%) and calculated using the following formula:

$$Recovery(\%)=\frac{Meancalculatedconcentration}{Nominalconcentration}\times 100$$ Formula 3:

2.6.6. Short-term stability

The short-term stability of triplicate samples of eight quality control samples of amobarbital in MeOH/H₂O (4:5 v/v) with concentrations ranging from 0.5–100 μg/mL was evaluated. This was done by comparing the experimentally calculated concentrations of freshly prepared samples to those stored in a −80°C freezer (New Brunswick Scientific™ Innova U 725-G Upright Freezer; Eppendorf AG, Hamburg, Germany) after one week and one month. The recovery (%), %RSD, and SD for the mean calculated concentration of the samples was recorded.

2.7. Formulation and characterization of the amobarbital/HA hydrogel formulation

2.7.1. In vitro cumulative release of amobarbital from HA hydrogel formulations

Two amobarbital formulations, 0.5 and 0.75% (w/v) HA, were compared to free drug and prepared as previously described. Afterwards, 1.2 mL of each formulation was transferred to a Float-A-Lyzer^® dialysis tube (10,000 molecular weight cut-off, Spectrum Chemical Manufacturing, New Brunswick, NJ) which was then submerged in 12 mL of release medium (PBS pH 7.4). The samples were stored in an orbital shaker kept at 37°C with rotation speed of 300 rpm (Classic C24 Incubator Shaker, New Brunswick Scientific, Edison, NJ). At predetermined time intervals (0.25, 0.5, 1, 2, 4, 6, 24, 48, and 72 hours (h)), the entirety of the release medium was withdrawn and replaced with fresh medium. Amobarbital content was determined by combining 100 μL of the withdrawn media samples with 80 μL of MeOH and injecting this directly into the HPLC-UV. The data was plotted using GraphPad Prism version 9.0.0.

2.7.2. Amobarbital solubility in the HA hydrogel formulation

Amobarbital solubility was evaluated by combining approximately 165 mg of the lyophilized amobarbital powder to 300 μL of the HA hydrogel (1% w/v) and diluting with 200 μL of normal saline to give a final HA concentration of 0.6% w/v. Triplicate samples were stored in an orbital shaker kept at 37°C with rotation speed of 300 rpm for 48 h. After this time, the samples were left stationary without shaking for 24 h at 37°C to allow for complete equilibration. Samples were taken at time zero and compared to samples taken at the end of the experiment. Samples were prepared for injection into the HPLC-UV system by taking 10 μL of the sample and diluting with 990 μL of MeOH/H₂0 (4:5 v/v), then taking 100 μL of this diluted sample and combining it with 900 μL of MeOH/H₂0 (4:5 v/v). This twice diluted sample was then injected directly into the HPLC-UV system.

2.7.3. Uniformity of amobarbital distribution in the HA hydrogel

The amobarbital/HA hydrogel (0.6% w/v) formulation was prepared as previously described, and the efficiency of the mixing process evaluated by taking triplicate 25 μL samples from the top, middle, and bottom regions of the hydrogel. To measure the uniformity of the drug distribution, these samples were prepared for HPLC-UV injection following the previously described procedure for human synovial fluid and human plasma method validation samples. The data was plotted using GraphPad Prism version 9.0.0.

2.7.4. Stability of the amobarbital/HA hydrogel formulation versus free drug in water, human synovial fluid and human plasma

The stability of the amobarbital/HA hydrogel (0.6% w/v) formulation was tested and compared to that of free drug amobarbital solution when combined with either sterile water, human synovial fluid, or human plasma. Briefly, the HA hydrogel formulation and free drug solution were prepared following the previously described methods. Triplicate stability samples were mixed in sterilized crimp top 2 mL glass vials by combining 250 μL of either the amobarbital/HA hydrogel formulation or free drug solution with 250 μL of either sterile water, human synovial fluid, or human plasma. These vials were then stored in a 37°C oven and 100 μL samples were taken at 0, 0.5 and 24 h time points. Samples were kept at −80°C until HPLC-UV sample preparation was performed as previously described to measure the amobarbital content. The data was plotted using GraphPad Prism version 9.0.0.

2.7.5. Forced degradation studies of the amobarbital/HA hydrogel formulation versus free drug

Following USP guidelines [50, 51], forced degradation studies were performed on the amobarbital/HA hydrogel formulation and compared to free drug solution when triplicate samples were prepared as previously described and kept in one of four forced degradation storage conditions. The forced degradation storage conditions were as follows: 1) RT (22°C) in dark which was the control; 2) RT with continuous light from a SafeGuard™ Bench Fume Hood (Mott Manufacturing, Brantford, ON, Canada) equipped with a 32-Watt, 4100 Kelvin cool white fluorescent light bulb (Slyvania, Danvers, MA); 3) 37°C oven; and 4) 70°C oven. At the 0, 0.5, and 24 h time points, 100 μL samples were taken and kept at −80°C until they were analyzed for the amobarbital content following the previously described protocol. The data was plotted using GraphPad Prism version 9.0.0.

2.8. Statistical Analysis

The stability measurements were evaluated for statistically significant differences between groups and time points using two-way analysis of variance (ANOVA), followed by Tukey’s honestly significant difference test using GraphPad Prism version 9.0.0.

3. Results and Discussion

3.1. HPLC-UV method validation

The HPLC-UV method was validated following ICH guidelines [46] and was used to analyze amobarbital samples in MeOH/H₂O (4:5 v/v), human synovial fluid, or human plasma for specificity, linearity, LOD, LOQ, intra- and interday precision, accuracy, and short-term stability.

3.1.1. Specificity

The specificity of the HPLC-UV method for amobarbital was evaluated in human plasma and human synovial fluid in either the presence or absence of the drug. Several demonstrative chromatograms are shown in Figure 1 where the y-axis denotes milli-absorbance units (mAU) and the x-axis denotes time (min). Confirmation of the specificity was seen by the absence of any overlapping peaks between the amobarbital, the IS, or biological fluids. The retention time of the amobarbital peak came at 4.70 min and the peak for the IS (barbital) was at 2.27 min.

Figure 1.

HPLC-UV chromatograms of (A) MeOH/H₂O mobile phase alone, (B) 100 μg/mL amobarbital in MeOH/H₂O, (C) human plasma alone, (D) human plasma with 300 μg/mL amobarbital, (E) human synovial fluid alone, (F) human synovial fluid with 300 μg/mL amobarbital, (G) free drug amobarbital combined with human synovial fluid stored at 37°C for 24 hours, and (H) amobarbital/HA hydrogel combined with water at time 0 h.

3.1.2. Linearity

The MeOH/H₂O (4:5 v/v) amobarbital calibration curve was prepared using triplicate samples of nine concentrations ranging from 0.5 to 100 μg/mL and resulted in an R² value of 0.9897 (Figure 2A). The calibration curve for the amobarbital in human plasma was performed in triplicate and using five concentrations between 200–800 μg/mL and had an R² value of 0.9673 (Figure 2B). The calibration curve for the human synovial fluid samples was made using five concentrations ranging from 100–1000 μg/mL and had an R² value of 0.9874 (Figure 2B). All of the calibration plots were linear and had R² values greater than 0.97 demonstrating the linearity of the HPLC-UV method.

Figure 2.

Calibration curves depicting: (A) the mean AUC versus concentration for amobarbital solutions in MeOH/H₂O; (B) amobarbital:IS AUC ratio versus amobarbital concentration for human plasma and human synovial fluid samples. The data is represented as mean ± SD and the equation of the line and R² values are inlaid.

3.1.3. LOD and LOQ

The LOD and LOQ values for amobarbital in MeOH/H₂O, human plasma and human synovial fluid are summarized in Table 1. Both the LOD and LOQ values were lowest for the MeOH/H₂O samples compared to the human plasma and human synovial fluid samples. The LOD and LOQ values were the highest for the human plasma samples. The HPLC-UV method was able to distinguish amobarbital in the MeOH/H₂O samples from the baseline at lower concentrations compared to samples that required extraction of the drug from biological fluids. The fact that the LOD and LOQ values for human plasma were higher than human synovial fluid indicates that the HPLC-UV method required higher concentrations of the drug compared to human synovial fluid samples to distinguish the drug from the baseline.

Table 1.

The LOD and LOQ values for amobarbital in MeOH/H₂O, human plasma, and human synovial fluid.

	MeOH/H2O	Plasma	Synovial Fluid
LOD (μg/mL)	0.1982	56.7575	22.5489
LOQ (μg/mL)	0.6006	171.9924	68.3300

3.1.4. Precision

The intraday precision for MeOH/H₂O (4:5 v/v) amobarbital samples was measured for five triplicate samples of concentrations ranging from 5–100 μg/mL (Table 2). The %RSD for these samples remained within the 0.84–7.12% range. The intraday precision for human plasma samples with six amobarbital concentrations ranging from 300–8000 μg/mL had %RSD values which ranged from 1.04–6.92% (Table 3). The interday precision for MeOH/H₂O (4:5 v/v) amobarbital samples was measured for five triplicate samples of concentrations ranging from 5–100 μg/mL (Table 4). The %RSD for these samples ranged from 9.40–11.23%. The interday precision for human plasma samples with four amobarbital concentrations ranging from 300–2000 μg/mL had %RSD values which ranged from 1.83–5.15% (Table 5).

Table 2.

Intraday precision results for the MeOH/H₂O amobarbital samples.

	Injection #1			Injection #2
Concentration (μg/mL)	Mean AUC	SD	%RSD	Mean AUC	SD	%RSD
5	204.6667	8.8512	4.3247	202.2333	9.9042	4.8974
10	400.5667	12.4275	3.1025	412.0667	14.5590	3.5332
25	1082.9333	77.1296	7.1223	1068.6667	48.4102	4.5300
50	2109.9000	18.1849	0.8619	2130.7000	36.6004	1.7178
100	4159.8667	101.2138	2.4331	4286.3000	35.7911	0.8350

Table 3.

Intraday precision results for human plasma combined with amobarbital samples.

	Injection #1			Injection #2
Concentration (μg/mL)	Mean Ratio of Amobarbital/ IS	SD	%RSD	Mean Ratio of Amobarbital/ IS	SD	%RSD
300	0.3264	0.0146	4.4613	0.3261	0.0145	4.4519
600	0.8325	0.0576	6.9223	0.8265	0.0562	6.8027
1000	1.3986	0.0340	2.4331	1.3966	0.0280	2.0027
2000	2.7421	0.0630	2.2981	2.7424	0.0634	2.3130
4000	5.2862	0.0670	1.2678	5.2509	0.0545	1.0384
8000	10.1382	0.2602	2.5669	10.0542	0.2557	2.5432

Table 4.

Interday precision results for the MeOH/H₂O amobarbital samples.

Day One
Concentration (μg/mL)	Mean AUC	SD	%RSD
5	184.8667	20.5052	11.0919
10	360.9000	34.2694	9.4955
25	919.8667	91.6015	9.9581
50	1848.9667	174.9467	9.4619
100	3715.9000	393.6750	10.5943
Day Two
Concentration (μg/mL)	Mean AUC	SD	%RSD
5	184.3000	20.0716	10.8907
10	367.3333	40.1623	10.9335
25	939.0667	96.7674	10.3046
50	1871.7667	179.0299	9.5648
100	3740.2667	405.7313	10.8477
Day Three
Concentration (μg/mL)	Mean AUC	SD	%RSD
5	188.9000	21.2141	11.2304
10	348.5333	33.2031	9.5265
25	900.4667	90.1794	10.0147
50	1848.8333	187.9240	10.1645
100	3735.4333	351.3055	9.4047

Table 5.

Interday precision results for human plasma combined with amobarbital samples.

Day One
Concentration (μg/mL)	Mean Ratio of Amobarbital/IS	SD	%RSD
300	0.3261	0.0145	4.4519
600	0.8349	0.0153	1.8306
1000	1.3894	0.0668	4.8110
2000	2.7211	0.0632	2.3217
Day Two
Concentration (μg/mL)	Mean Ratio of Amobarbital/IS	SD	%RSD
300	0.3249	0.0145	4.4773
600	0.8215	0.0152	1.8501
1000	1.3645	0.0703	5.1525
2000	2.6851	0.0716	2.6652
Day Three
Concentration (μg/mL)	Mean Ratio of Amobarbital/IS	SD	%RSD
300	0.3237	0.0144	4.4605
600	0.8180	0.0240	2.9353
1000	1.3617	0.0579	4.2518
2000	2.6161	0.0705	2.6930

3.1.5. Accuracy

The HPLC-UV method was evaluated for accuracy in the presence of either human synovial fluid or human plasma and denoted as recovery (%). The range for the recovery (%) for the human synovial fluid and human plasma samples was 82.03–102.30% and 95.96–118.70%, respectively (Table 6). We associate the lowered recovery (%) values of the human synovial fluid samples compared to human plasma with a complexation of the drug to HA which we had observed in ultraperformance liquid chromatography-UV (UPLC-UV) experiments (Figure S1). Since human synovial fluid is rich in HA compared to human plasma [59], the drug is bound more strongly and thus not as easily recoverable at the lower amobarbital concentrations.

Table 6.

Accuracy values for amobarbital in human synovial fluid and human plasma.

Synovial Fluid			Plasma
Nominal Concentration (μg/mL)	Mean Calculated Concentration (μg/mL)	Recovery (%)	Nominal Concentration (μg/mL)	Mean Calculated Concentration (μg/mL)	Recovery (%)
100	82.0332	82.0332	200	191.9232	95.9616
600	560.0989	93.3498	600	604.9358	100.8226
1000	970.9850	97.0985	800	847.2316	105.9040
2000	2045.8740	102.2937	2000	2374.0110	118.7006

3.1.6. Short-term stability

The short-term stability after one week and one month of storing MeOH/H₂O amobarbital quality control samples at −80°C was evaluated for triplicate samples and the results are summarized in Table 7. After one week, the ranges for the %RSD and recovery (%) for these stability samples was between 0.38–0.49% and 95.86–104.34%, respectively. The samples displayed comparable stability with %RSD values ranging from 0.08–0.46% and 95.9–102.5 recovery (%) after one month of storage at −80°C. These results indicate that amobarbital remains relatively stable in MeOH/H₂O solution for up to one month of storage at −80°C.

Table 7.

Short-term stability results for amobarbital in MeOH/H₂O samples after storage at −80°C for one week and one month.

One Week
Nominal Concentration (μg/mL)	Mean Calculated Concentration (μg/mL)	SD	%RSD	Recovery (%)
0.5	0.5217	0.0750	0.4862	104.3431
0.75	0.7420	0.0409	0.1749	98.9336
1	0.9789	0.0580	0.1815	97.8879
5	5.0113	0.1338	0.0752	100.2265
10	9.9581	0.3283	0.0920	99.5809
25	24.8334	0.9060	0.1013	99.3337
50	49.7922	1.7902	0.0996	99.5844
100	98.9760	3.4675	0.0970	98.9760
One Month
Nominal Concentration (μg/mL)	Mean Calculated Concentration (μg/mL)	SD	%RSD	Recovery (%)
0.5	0.5125	0.0688	0.4556	102.4997
0.75	0.7190	0.0848	0.3757	95.8613
1	1.0102	0.0308	0.0930	101.0217
5	5.1201	0.1835	0.1010	102.4017
10	10.0779	0.3226	0.0894	100.7792
25	25.0205	0.7632	0.0847	100.0822
50	50.1599	1.8966	0.1048	100.3199
100	99.6562	3.6211	0.1006	99.6562

3.2. Formulation and characterization of the amobarbital/HA hydrogel formulation

3.2.1. In vitro cumulative release

The in vitro cumulative release profile of amobarbital alone (Free Drug) or dispersed in the HA hydrogel (0.5% or 0.75%) was evaluated to find an ideal formulation for our proposed PTOA treatment (Figure 3A). The amobarbital/HA hydrogel formulations displayed an initial burst release of amobarbital within the first 5 h which accounted for 40% and 52% cumulative release of amobarbital in 0.75% and 0.5% (w/v) HA, respectively. At 72 h, the percent cumulative release reached 93% and 106% for the 0.75% and 0.5% (w/v) HA formulations, respectively. There was no statistically significant difference between the two formulations’ release profiles. In contrast, the free drug solution displayed complete dissolution within 5 h. Based on this release profile data, 0.6% (w/v) HA was chosen for further testing because this intermediate concentration would have an optimal release profile for the purposes of our studies.

Figure 3.

(A) In vitro cumulative release profiles of free drug compared to two amobarbital/HA hydrogel formulations (0.5 and 0.75% (w/v) HA). (B) Drug distribution uniformity results for the amobarbital/HA hydrogel (0.6% w/v) formulation. Data is represented as mean ± SD.

3.2.2. Amobarbital solubility

Amobarbital was found to be soluble in the final amobarbital/HA hydrogel at a concentration exceeding 152 mg/mL. Since this concentration was much higher than the 0.6 mg/mL intended concentration for the purpose of our PTOA treatment in a clinical model, no higher concentrations were tested.

3.2.3. Uniformity of amobarbital

The efficiency of the mixing process for the amobarbital/HA hydrogel (0.6% w/v) formulation was determined using triplicate samples (Figure 3B). The samples had mean ± SD amobarbital concentration values of 472.49 ± 37.60, 479.11 ± 33.39, and 450.28 ± 27.63 for the top, middle, and bottom sections of the hydrogel, respectively. The distribution of amobarbital in the HA hydrogel formulation was homogenous with no statistically significant differences between the sections.

3.2.4. Stability of the amobarbital/HA hydrogel formulation versus free drug in water, human synovial fluid and human plasma

The stability of amobarbital in the amobarbital/HA hydrogel (0.6% w/v) formulation was compared to that of free drug solution in the presence of water, human synovial fluid, or human plasma after storage at 37°C for 24 h. A graphical representation of the stability data showing the amobarbital concentration relative to the 0 h time point is shown in Figure 4. The numerical data values of the study are summarized in Table S2. For the amobarbital/HA hydrogel formulation, the %RSD values ranged from 1.60 – 30.56% with the highest value coming from the human synovial fluid samples. The %RSD values for the free drug samples ranged from 1.05–10.77%. The free drug samples remained comparably stable over the 24 h period regardless of whether it was combined with water, human synovial fluid, or human plasma. Though the difference between the water, human synovial fluid, and human plasma samples was not statistically significant from one another at the two time points, we did notice that the absolute value for the difference in the mean amobarbital concentration from the 0 h to 24 h timepoint was greatest for the human synovial fluid (70.2 μg/ml and 37.3 μg/ml for hydrogel and free drug, respectively) and human plasma samples (36.2 μg/ml and 16.8 μg/ml for hydrogel and free drug, respectively) compared to water (5.3 μg/ml and 12.9 μg/ml for hydrogel and free drug, respectively) as shown in Table S2. We attribute the increased variability in the biological fluid samples to protein binding occurring between the proteins present in the biological fluids and amobarbital [60]. The fact that the mean amobarbital concentration in the human synovial fluid and human plasma samples for both the hydrogel and free drug samples increased over time indicated that storage of the sample at 37°C for 24 h allowed for more drug to be released. This phenomenon of drug binding to the proteins present in plasma and synovial fluid has been previously investigated for other drug compounds, however, there has been no definitive study looking specifically at amobarbital [60–63]. We have also provided strong evidence that amobarbital forms a complex with HA over time using a validated UPLC-UV method (Figure S1). Since the aim of this study was to test whether amobarbital would stay intact at the injection site for up to 24 hours after injection, these results indicate that the drug is comparably stable at physiologically relevant temperature compared to the free drug solution and suitable for our intended purposes.

Figure 4.

Graphical representation of the amobarbital concentration relative to the 0 h time point stability study results for amobarbital in the amobarbital/HA hydrogel (0.6% w/v) formulation and free drug solution in the presence of water, human synovial fluid or human plasma and stored at 37°C for 24 h. The data is represented as mean ± SD with the individual data points superimposed.

3.2.5. Forced degradation studies of the amobarbital/HA hydrogel formulation versus free drug

Forced degradation studies were performed on triplicate amobarbital/HA hydrogel (0.6% w/v) formulation and free drug samples kept at one of four forced degradation conditions; 1) RT in dark, 2) RT with continuous light, 3) 37°C oven, and 4) 70°C oven. A graphical representation of the forced degradation data showing the amobarbital concentration relative to the 0 h time point is shown in Figure 5. The numerical data values of the study are summarized in Table S3. For the amobarbital/HA hydrogel formulation, the %RSD values ranged from 1.31 – 33.71%. The %RSD values for the free drug samples ranged from 0.23 – 9.78%. The most notable observation from this study was that the amobarbital/HA hydrogel formulation displayed no significant degradation even after being kept at 70°C for 24 h compared to free drug solution (p = 0.0205). When compared to the free drug solution kept at the same conditions, the HA hydrogel formulation provided protection of the drug from degradation and maintained amobarbital stability. Since the plan for this formulation would be that the clinician would mix the amobarbital/HA hydrogel immediately before injection, these results suggest that the formulation has a suitable shelf life over 24 hours.

Figure 5.

Graphical representation of the amobarbital concentration relative to the 0 h time point results for amobarbital in the amobarbital/HA hydrogel (0.6% w/v) formulation and free drug when stored in four forced degradation conditions for 24 h, where * means p < 0.05. The data is represented as mean ± SD with the individual data points superimposed.

4. Conclusion

In conclusion, we demonstrated the effectiveness of an HPLC-UV method to analyze amobarbital in MeOH/H₂O, human synovial fluid, and human plasma samples. First, samples in MeOH/H₂O were used to demonstrate the validity of the method and a detailed protocol was described which can be used to analyze the amobarbital drug content in an amobarbital/HA hydrogel formulation as well as extract the drug from biological samples. According to ICH guidelines, this method was validated for specificity, linearity, precision, accuracy, and short-term stability all of which indicated the suitability of the HPLC-UV method for the detection of amobarbital. We demonstrated the potential for this amobarbital/HA hydrogel formulation to be a drug delivery system by achieving a sustained release profile not seen for amobarbital solution. We found that the solubility of amobarbital in a 0.6% HA hydrogel well exceeded our purposes and that the drug was homogeneously distributed throughout the hydrogel matrix. We showed that the amobarbital/HA hydrogel and free drug solutions were comparably stable when combined with water, human synovial fluid, and human plasma at biologically relevant temperature. These results give some indication about the short-term stability of the drug immediately after injection into the joint capsule using an in vitro model. Future work will need to be done in order to determine the metabolism of the drug in vivo and the degradation kinetics of the drug after injection. In a forced degradation study, the amobarbital/HA hydrogel formulation displayed no significant degradation even when kept at 70°C for 24 h which indicated that this novel formulation maintained the stability of amobarbital compared to the free drug solution. Since amobarbital is a compound that has been in use clinically since the 1930’s, there have already been stability studies performed and the degradation products for the drug have been well studied [48]. In our current studies, we wanted to see how the hydrogel matrix affected the amobarbital concentration compared to a free drug solution kept in the same conditions. Since this is a novel formulation, we expect that the hydrogel matrix will have some effect on the degradation kinetics of the drug compared to the free drug solution. Future studies will need to be done to determine exactly how the HA hydrogel affects the drug’s degradation kinetics using a more precise system such as liquid chromatography-mass spectroscopy. Here we have demonstrated the extended-release kinetics and improved stability provided by the HA hydrogel compared to the free drug solution as well as the effectiveness of the HPLC-UV protocol to determine the drug content of samples made in MeOH/H₂O, human synovial fluid, and human plasma. These results as well as previously published research [23] are essential data supporting the testing of this formulation as a PTOA treatment in a future clinical trial. The data reported here shows that the drug concentration remains stable in the formulation over a 24 hour period at room temperature.