To achieve optimal resolution and peak shape for DAP and TNL (as shown in Fig. 2), various chromatographic parameters were systematically optimized. These parameters included mobile phase composition, flow rate, detection wavelength, and column temperature. Several trials have been taken with single organic modifier but did not get proper peak shape and also eluted at longer retention time. Therefore, the proportion of organic modifier such as methanol and acetonitrile were optimized in order to reduce tailing effect as well as avoid longer runtime of both drugs. The optimized values are presented in Table 1.

RP-HPLC chromatogram of TNL (Rt: 6.6 min) and DAP (Rt: 12.6 min) at 224 nm

The method validation for the analytical procedure adhered to ICH guidelines. Daily system suitability tests confirmed method performance, with all parameters meeting established criteria for HPLC analysis of DAP and TNL, as summarized in Table 2. Specificity testing ensured no matrix interference at the retention times of DAP and TNL. Figure 3 presents an overlay of the matrices, including diluent, blank, DAP standard, TNL standard, and sample solution, demonstrating the method’s specificity for these analytes.

Overlay chromatograms of standard TNL, DAP, mixture and methanol (near to far)

A linear response for DAP and TNL was observed across the concentration ranges of 12.5–37.5 µg/mL and 50–150 µg/mL, respectively, extendable to 75 µg/mL for DAP and 300 µg/mL for TNL. The correlation coefficients (r) exceeded 0.999 for both drugs. Calibration curves were generated, and the mean slope and intercept values were used for quantitation of the analytes from the samples, detailed in Table 3.

Intra-day and inter-day precision, calculated using the specified equation, showed assay values based on standard and sample peak area comparisons, with relative standard deviations (RSD) within the acceptable limit of ≤ 2%. This confirms the method’s precision for routine application. Accuracy was assessed using the standard addition method for DAP and TNL, with consistent recovery percentages detailed in Table 3. The LODs for DAP and TNL were 0.5 µg/mL and 2 µg/mL, respectively, with a signal-to-noise ratio greater than 3. The LOQs were 1.5 µg/mL for DAP and 6.2 µg/mL for TNL, with a signal-to-noise ratio above 10.

The method underwent evaluation for potential minor deviations arising from instrumental or human error. Intentional, minor adjustments were made to process parameters, including flow rate, column temperature, wavelength detection, and mobile phase pH. System suitability parameters and assay values were then assessed, comparing the peak areas of the standard and sample as per the given equation. RSD values below 2% indicated that the results remained stable despite these slight variations in chromatographic conditions. Table 3 presents a summary of the validation parameters, confirming the method’s specificity, precision, accuracy, and robustness.

The force degradation study was conducted using HPLC under optimized conditions to induce degradation of the drugs within a targeted range of 20–30%. Individual drug solutions underwent exposure to various stress conditions to evaluate their stability. Dapagliflozin (DAP) demonstrated remarkable stability across all tested conditions, with no discernible degradation products detected. Conversely, teneligliptin (TNL) exhibited the formation of degradation products under acidic, alkaline, and oxidative conditions.

When DAP and TNL were combined in solution, the chromatograms revealed degradation products of TNL at a retention time (Rt) of 3.24 min under acidic and alkaline conditions, while oxidative degradation yielded a degradation product at an Rt of 3.91 min (Fig. 4). Notably, photolytic and thermolytic degradation conditions did not yield any detectable degradation products in the chromatograms.

Overlay chromatogram of force degradation of photolytic controlled and uncontrolled, thermal, acidic, alkaline, oxidative conditions of sample solutions of DAP and TNL (near to far)

Consequently, samples subjected to acidic, alkaline, and oxidative degradation were subjected to further analysis using LC-MS/MS for mass-based identification. The HPLC method was precisely developed to ensure compatibility with LC-MS/MS, taking into account factors such as buffer volatility, buffer strength, and optimal LC eluent flow rate to facilitate analyte ionization.

Method parameters were systematically optimized to achieve adequate resolution between degradation products (DPs) and analytes. Furthermore, the LC-MS/MS system was configured to selectively detect and analyze only the degradation products, bypassing the retention times corresponding to DAP and TNL. This approach was implemented to mitigate mass saturation resulting from the injection of higher concentrations of DAP and TNL, thereby enabling accurate identification of degradation products. Table 4 illustrates the summary of forced degradation data along with % degradation.

The HPLC-PDA analysis of force degradation studies confirmed the presence of DPs from both combined and individual drug solutions. DAP remained stable, while TNL exhibited DPs under acidic, alkaline, and oxidative conditions. Subsequent LC-MS/MS analysis facilitated DP mass identification without altering chromatographic conditions, maintaining method compatibility. A divert valve selectively allowed DPs for mass identification while bypassing ± 1 min of TNL and DAP retention times to prevent mass saturation. In acidic and alkaline degradation, DP-1 appeared at Rt 3.25 min (Figs. 5a and 6a), while in oxidative degradation, DP-2 emerged at 3.9 min (Fig. 7a). Mass scanning in positive mode yielded intense peaks for better ionization. Mass spectral data revealed DP-1 mass at m/z 356.1 (Figs. 5b and 6b) and DP-2 mass at m/z 443.0 (Fig. 7b). Fragmentation patterns, analyzed via MS2 scan and then product ion scanning, provided insights into structural information and tentative degradation pathways.

a Total ion chromatogram of acid degradation sample, b mass spectra for acid DP-1 (356.1 m/z), c mass spectra for acid DP-1 product ions (242.9 m/z and 269.1 m/z)

a Total ion chromatogram of alkali degradation sample, b mass spectra for alkali DP-1 (356.1 m/z), c mass spectra for alkali DP-1 product ions (243.0 m/z and 269.1 m/z)

a Total ion chromatogram of oxidative degradation sample, b mass spectra for oxidative DP-1 (443.1 m/z), c mass spectra for oxidative DP-1 product ions (425.1 m/z, 243.0 m/z and 87.8 m/z)

Under acidic conditions, the protonation of the amide nitrogen atom could lead to the formation of an acylium ion intermediate at 355.2 m/z, which may subsequently undergo nucleophilic attack by water molecules, resulting in the cleavage of the amide bond. This cleavage event could generate a fragment with a mass corresponding to the observed m/z 269, potentially representing the protonated form of the resulting fragment (Fig. 5c).

Similarly, under alkaline conditions, hydroxide ions could facilitate nucleophilic attack on the amide carbonyl carbon, leading to the formation of a tetrahedral intermediate at 355.2 m/z, followed by the elimination of the leaving group and cleavage of the amide bond. This hydrolysis reaction could produce a fragment with a mass corresponding to the observed m/z 269 (Fig. 6c).

Furthermore, the observed mass peak at m/z 243 could suggest further structural modifications or degradation of the resulting fragments, possibly involving additional functional group transformations or fragmentations (Fig. 8). Therefore, based on the experimental findings and the proposed degradation mechanisms, it is plausible to infer that the amide bond in TNL is a key site of susceptibility to acidic and alkaline hydrolysis reactions, leading to the formation of characteristic degradation products as observed in the LC-ESI-MS/MS analysis.

Proposed fragmentation pathway of TNL degradation products

Under oxidation conditions, teneligliptin likely undergoes initial oxidation of its primary or secondary amine functional groups, leading to the formation of an N-oxidized intermediate. Subsequent rearrangement or further oxidation of this intermediate results in the formation of a degradation product with a mass of 443 m/z (Fig. 7b). LC-ESI-MS/MS analysis reveals fragmentation patterns, including fragments at 425.1, 243.0, and 87.8 m/z (Fig. 7c), indicating cleavage of specific bonds within the molecule. Figure 8 illustrates the prediction of degradation pathway which suggests the fragment at 425.1 m/z could be due to rearrangement or oxidation reactions, possibly involving the amide or aromatic ring. The 243.0 m/z fragment likely arises from cleavage of the amide bond or aromatic carbon–nitrogen bond. Further fragmentation at 87.8 m/z indicates additional fragmentation of primary fragments, with the structure corresponding to piperazine.