**1. Introduction**

Arthritis and diabetes mellitus are highly prevalent diseases with a total of over 350 million patients worldwide [1,2]. The most common types of arthritis and diabetes mellitus are osteoarthritis (OA) and type 2 diabetes mellitus (T2DM), respectively [3]. OA affects 14% of adults aged ≥25 years, and 34% of these patients are aged >65 years [4]; similarly, T2DM affects 12% of adults aged ≥20 years, and 26% of these are aged >65 years [5]. A recent survey estimated that the prevalence of OA was higher in individuals with T2DM than in those without T2DM [6]. Thus, T2DM is generally recognized as a comorbidity of arthritis [7], while some previous studies have focused on diabetes as a risk factor of arthritis [8,9]. Anyway, it is evident that T2DM is closely associated with an increased incidence and prevalence of OA, though the reasons remain unclear.

Repaglinide (REP), as shown in Figure 1, is a short-acting oral antidiabetic drug belonging to the class of meglitinides and is used to lower postprandial blood glucose levels in T2DM patients [10]. It stimulates insulin release from the pancreas, depending on the residual function of β-cells in the pancreatic islets [11]. REP is eliminated primarily by CYP2C8- and CYP3A4-mediated oxidative metabolism [12]. Systemic exposure to REP has been reported to be altered by co-administration of trimethoprim (inhibitor of CYP2C8) [13], itraconazole (inhibitor of CYP3A4) [14], or rifampicin (inducer of CYP3A4) [11]. Celecoxib (CEL), as shown in Figure 1, is a cyclooxygenase-2 (COX-2) selective non-steroidal anti-inflammatory drug (NSAID) and is used to treat pain and inflammation associated with OA and rheumatoid arthritis (RA) [15]. Because gastrointestinal mucosal integrity is compromised by COX-1 inhibition, traditional nonselective NSAIDs such as aspirin, ibuprofen, and indomethacin that inhibit both COX-1 and COX-2 may cause serious side effects in the gastrointestinal tract [16]. Thus, CEL exhibits distinctly reduced gastrointestinal toxicity as compared to conventional NSAIDs, thereby becoming a blockbuster drug for treating OA and RA [17]. CEL is eliminated primarily by extensive metabolism through methyl hydroxylation to form hydroxycelecoxib, which is catalyzed by CYP2C9 and CYP3A4 [18]. β

**Ketoconazole**

**Figure 1.** Chemical structures of repaglinide, celecoxib, and ketoconazole (internal standard, IS).

μ Because T2DM frequently co-exists with OA, there is a possibility of concurrent administration of REP and CEL. Hence, a bioanalytical method of simultaneous determination of REP and CEL could be useful and efficient for further pharmaceutical development and therapeutic optimization. To date, several bioanalytical methods have been developed and validated for quantitative determination of REP or CEL individually using HPLC with UV/Vis detection [19–23] or using liquid chromatography with tandem mass spectrometry (LC-MS/MS) systems [24–26]. However, these methods are associated with a few limitations, such as an insufficient sensitivity, relatively large sample volume, and/or time-consuming liquid–liquid extraction procedures with volatile solvents that are potentially hazardous to health. Moreover, LC-MS/MS methods require relatively complex and/or expensive instrumentation, which may not be affordable for small-sized laboratories and companies in resource-limited settings. To the best of our knowledge, there have been no reported methods of simultaneous quantification of REP and CEL in biological samples using HPLC coupled with a fluorescence detector (HPLC-FL). Furthermore, a previous in vitro study reported that CEL inhibited REP metabolism in pooled human liver microsomes (HLM) with a K<sup>i</sup> of 3.1 µM [27]. This suggests the possibility of pharmacokinetic drug interaction between REP and CEL, but no information is currently available regarding this issue. Therefore, further investigation of the pharmacokinetic drug interaction between REP and CEL is necessary to prevent adverse effects in the use of these drugs.

In the current study, a sensitive and simple HPLC-FL method was developed and fully validated for simultaneous quantification of REP and CEL in rat plasma. The linearity, sensitivity, precision, accuracy, recovery, matrix effect, and stability of this HPLC-FL method were determined. Next, the potential for the pharmacokinetic drug interactions between REP and CEL was investigated in vivo using Sprague-Dawley rats and in vitro using rat liver microsomes (RLM) and HLM.

#### **2. Materials and Methods**

#### *2.1. Materials*

CEL (purity ≥98%), ketoconazole (as internal standard [IS]; purity ≥98%), as shown in Figure 1, and REP (purity >98%) were purchased from Tokyo Chemical Industry Co. (Tokyo, Japan). DMSO, ethanol, potassium phosphate monobasic/dibasic, and polyethylene glycol 400 were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA). β-nicotinamide adenine dinucleotide phosphate (NADPH), HLM, and RLM were purchased from BD-Genetech (Woburn, MA, USA). ACN and methanol of HPLC grade were purchased from Thermo Fisher Scientific, Inc. (Waltham, MA, USA).

#### *2.2. Animals*

Male Sprague-Dawley rats (nine-week-old; approximately 300 g) were purchased from Samtako Bio Korea Co. (Gyeonggi-do, Korea). They were kept in a clean room of the Laboratory Animal Center of Pusan National University (Busan, Korea) at a relative humidity of 50 ± 5% and temperature of 20–23 ◦C with 12 h dark (19:00–07:00) and light (07:00–19:00) cycles. They were housed in metabolic cages (Tecniplast USA Inc., West Chester, PA, USA) with tap water and standard chow diet (Agribrands Purina Canada Inc., Levis, QC, Canada) provided ad libitum. The present animal study protocols were approved by the Pusan National University-Institutional Animal Care and Use Committee (PNU-IACUC, Busan, South Korea) for ethical procedures and scientific care (approval number: PNU-2018-1848; approval date: 01/05/2018).

#### *2.3. Calibration Standards and Quality Control Samples*

Stock solutions of REP, CEL, and IS (1000 µg/mL in DMSO) were prepared. The stock solutions of the mixture of REP and CEL were diluted with mobile phase for the preparation of working standard solutions with concentrations ranging from 1 to 200 µg/mL. The working solution of IS (final concentration of 5 µg/mL in ACN) was prepared by diluting the stock solution of IS with ACN. Calibration standard samples were prepared by spiking blank rat plasma with each working solution, yielding final plasma concentrations of 10, 20, 50, 100, 200, 500, 1000, and 2000 ng/mL. Quality control (QC) samples were prepared from separate stocks of REP and CEL in an identical manner to the preparation of calibration standards. The concentration levels of QC samples were 10 (lower limit of quantification; LLOQ), 30 (low; LQC), 120 (middle; MQC), and 1200 ng/mL (high; HQC).
