Introduction

Cannabis sativa L. predominantly produces more than 100 psychoactive metabolites called cannabinoids (Chakravarti et al. 2014; Abrams and Guzman 2015) which are hydrophobic and terpenophenolic molecules (Chakravarti et al. 2014; Sledzinski et al. 2018; Klumpers and Thacker 2018). Major endocannabinoids which are known as anandamide and 2-arachidonoylglycerol exert physiological effects of cannabinoids (Velasco et al. 2016). Endocannabinoid system components are known to have wide expression patterns in human organ systems such as central nervous system (Pisanti et al. 2013; Howlett and Abood 2017), immune system (Jean-Gilles et al. 2015), genital system (Bilgic et al. 2017), digestive system (DiPatrizio 2021) and respiratory system (Boyacıoğlu et al. 2021; Ramer et al. 2014; Gkoumassi et al. 2007) by regulating mood, motor activity and appetite, immune modulation, fertility, food intake and airway functions, respectively. Recently, synthetic and natural cannabinoid derivatives are highly investigated as drug candidates due to their antinociceptive (Brunetti et al. 2020; Good et al. 2019; VanDolah et al. 2019), antiepileptic (Brunetti et al. 2020; VanDolah et al. 2019; Billakota et al. 2019) and anticancer potential (Boyacıoğlu et al. 2021; Milian et al. 2020; Preet et al. 2011; Donadelli et al. 2011; Dando et al. 2013; Brandi et al. 2013). Both endogenous and exogenous cannabinoids prevent proliferation and induce apoptosis in various types of epithelial cancers (Boyacıoğlu et al. 2021; Donadelli et al. 2011; Dando et al. 2013; Brandi et al. 2013; Roberto et al. 2019) that make them chemotherapeutic candidates for epithelial cancers such as lung cancer (Boyacıoğlu et al. 2021; Milian et al. 2020; Preet et al. 2011; Haustein et al. 2014; Ramer et al. 2012; Winkler et al. 2016). Both endogenous and exogenous cannabinoids induce their anticancer effects through G-protein-coupled CB receptors 1 (CB1) and 2 (CB2) (Chakravarti et al. 2014; Jaarsveld et al. 2016). Our group previously published the dose- and time-dependent antiproliferative and apoptotic effect of a synthetic specific CB1 receptor agonist ACPA (N-(Cyclopropyl)-5Z,8Z,11Z,14Z-eicosatetraenamide; C23H37NO; MW:343.555 g/mole) and ACPA-loaded polycaprolactone (PCL) on non-small cell lung cancer (NSCLC) cells with its downstream cascade (Patent pending for Turkish Patent and Trademark Office application no: TR2019/12451 and Patent Cooperation Treaty application no: PCT/TR2020/050618) (Boyacıoğlu et al. 2021). ACPA is also a drug candidate for other epithelial cancers including pancreatic (Donadelli et al. 2011; Dando et al. 2013; Brandi et al. 2013) and endometrial carcinoma (Bilgic et al. 2017). Therefore, determining the levels of synthetic cannabinoids such as ACPA in biological fluids, cells and tissues is crucial in monitoring the effects of various pharmacological, physiological and pathological stimuli on biological systems (Zou and Kumar 2018). However, it is difficult to quantify them accurately due to their short half-lives (Abrams and Guzman 2015).

Several chromatographic methods are available for the analysis of endocannabinoids in a wide variety of biological matrices such as cell culture (Ottria et al. 2014; Gouveia-Figueira and Nording 2014; Ivanov et al. 2015; Bobrich et al. 2020), blood (Lin et al. 2012; Bilgin et al. 2015), plasma (Ottria et al. 2014; Gouveia-Figueira and Nording 2014, 2015; Zoerner et al. 2012; Balvers et al. 2013; Sergi et al. 2013; Thieme et al. 2014; Gachet et al. 2015; Marchioni et al. 2017; Ozdurak et al. 2010), serum (Lam et al. 2010; Kirkwood et al. 2016), urine (Ottria et al. 2014; Lam et al. 2010), milk (Gouveia-Figueira and Nording 2014, 2015; Lam et al. 2010), cerebrospinal fluid (Leweke et al. 2007; Kantae et al. 2017), tissue (Bobrich et al. 2020; Lin et al. 2012; Marczylo et al. 2010; Gong et al. Statistical analysis

The statistical calculations were carried out using Microsoft Excel software. P values of the regression coefficient and regression equation were calculated by ANOVA test. The robustness of analytical methods was evaluated by one-way analysis of variance (ANOVA) test. For ruggedness, Student’s t test was used for the comparison of two different analysts.

Results and discussion

Method optimization

An IS usage in LC–MS/MS based bioanalytical methods is mandatory for repeatable results due to unexpected changes especially in ionization steps. The IS selected for analysis should have a similar chemical structure to the analyte to eliminate errors that may happen during sample preparation or analysis. Therefore, arachidonoyl ethanolamide, as an endocannabinoid, was selected as IS.

The optimization of the method was systematically studied for MS and LC conditions. Firstly, MS detection parameters were optimized by direct injection of ACPA and IS individually at a concentration of 1000 ng/ml to obtain the highest peak intensity. The MRM was operated in positive ionization mode, and the transitions of ACPA and IS were determined as m/z 344 → 203 and m/z 166 → 120, respectively (Fig. 1).

Fig. 1
figure 1

MS/MS fragmentation pattern of ACPA (Quantifier ion: 203 m/z, qualifier ion: 287 m/z, CE:  − 14 eV)

Full size image

After the optimization of MS conditions, chromatographic conditions were optimized to achieve the best possible analyte separation within the shortest time. Analytical column, mobile phase composition and flow rate have been optimized to achieve precise and reproducible results. Different sizes of C18 columns and various gradients of mobile phase compositions were studied, and a C18 column (GL Sciences, 50 × 3.0 mm, 2.1 μm) was chosen as stationary phase because of symmetric peaks and short retention time. The best ionization was achieved when 0.1% FA in water and 0.1% FA in acetonitrile were used as the mobile phase. The gradient elution was applied to the mobile phase as follows: 0.0–0.01 min at 40% B, 0.01–4.0 min from 40 to 80% B, 4.0–7.0 min 80% B, followed by 3 min of equilibration at initial conditions.

The total run time was 10 min, and the retention time of ACPA and IS was 6.4 min and 5.1 min, respectively (Fig. 2).

Fig. 2
figure 2

Representative chromatograms of A ACPA (1000 ng/mL) and B IS (1000 ng/mL) at optimum chromatographic conditions

Full size image

The suitability of the LC–MS/MS method under optimum analysis conditions was evaluated in terms of asymmetry (10%), column efficiency (theoretical plate number, N), capacity factor (k′) and tailing factor parameters. The values obtained (CV = 0.20%, N = 319,230, k′ = 6.77 and tailing factor = 1.25) are within the specified limits (Bioanalytical Method Validation M10 2019), and it has been determined that the system is suitable for the analysis of ACPA.

Method validation

The LC–MS/MS method was validated for selectivity, linearity, sensitivity, matrix effect, carry over, precision, accuracy, robustness and ruggedness following ICH bioanalytical method guideline (Bioanalytical Method Validation M10 2019).

The selectivity of the LC–MS/MS method was investigated by comparing blank chromatograms of the matrix and ACPA spiked chromatograms (Fig. 3). There is no significant interference higher than 20% of the peak area of ACPA at LOQ level and 5% of the peak area of IS. This indicates that the developed method is selective for the analysis of ACPA from cell culture and placebo samples.

Fig. 3
figure 3

Chromatograms obtained under optimum chromatographic conditions: A blank for ACPA (in cell mixture), B blank for ACPA (in cell supernatant mixture), C ACPA spiked matrix at LOQ concentration, D blank for IS (in cell mixture), E blank for IS (in cell supernatant mixture), F IS spiked matrix (1000 ng/mL)

Full size image

The linearity of the calibration curves was determined over the concentration range of 1.8–1000 ng/ml with a correlation coefficient value which is 0.999 ± 0.0002. The values (mean ± SE; n = 6) of the slope and intercept were 0.0021 ± 0.0001 and 0.0102 ± 0.0029, respectively. The LC–MS/MS method showed an acceptable linearity range from 1.8 to 1000 ng/ml for ACPA. The LOD was 0.6 ng/ml, and the LOQ calculated in this work was 1.8 ng/ml with acceptable accuracy and precision (Table 1). The developed method was highly sensitive for estimating ACPA in the cell culture.

Table 1 Intra- and inter-day accuracy, precision and reinjection reproducibility of the method
Full size table

The matrix effect was evaluated based on ACPA analysis in the presence and absence of the sample matrix. The samples were prepared triplicate and analyzed under the optimum analytical condition. The matrix effect was calculated as 98.60% ± 1.67 for ACPA and 100.49% ± 1.60 for IS. These results showed that the developed method was not affected by the sample matrix. The carryover was not observed in any of the blank matrix samples (≤ 20% of the analyte response at the LOQ and ≤ 5% of the mean IS response of the accepted calibration standards) after the highest calibration standard injection (1000 ng/ml).

Dilution integrity was tested for fivefold and tenfold dilution, and the low bias (≤ 1.70%) and CV (≤ 1.40%) values were that the method has the ability to accurately quantify samples containing high concentrations of analytes diluted 1:10 within acceptance criteria (≤ 15.0%) (Table 1).

Intra-day and inter-day precision and accuracy and reinjection reproducibility were estimated by analyzing three or six replicates at four different concentration levels (Table 1). The low CV and bias values showed that the method was precise and accurate. In addition, the accuracy of the developed method was also examined with recovery studies from placebo, cell culture medium and intracellular matrix. To evaluate the recovery, cell culture matrixes were spiked with ACPA at different concentrations and compared with the same amount of spiked water samples. The recoveries of ACPA were found as between 95.70 and 97.25%. The result indicates the method has high accuracy and recovery.

The robustness of analytical methods was evaluated with a thirteen-run fractional factor design with three experiments under optimized conditions. The results of the analysis were statistically compared with ANOVA test, and p values of the regression coefficient were calculated (Table 2). The p values of each variable were higher than 0.05 indicating that small changes do not have a statistically significant effect on the peak area ratio and robustness of the method.

Table 2 The experimental design and results for robustness study
Full size table

The ruggedness of the method was evaluated by comparing two different analyst’s results. It was found that there was no statistically significant difference (p > 0.05) between them (Table 3). Thus, the method was found rugged.

Table 3 Ruggedness data of the developed method
Full size table

The stability of the developed method was found at least 3 days in the refrigerator, two months in the freezer and 8 h in the autosampler.

Cell culture studies

L929 mouse fibroblast cells (CCL-1™, ATCC) as a well-defined and suggested cell line for use per ISO Standard 10,993–5 were spiked with 0.1 mM ACPA solution. The intracellular matrix and cell culture medium were analyzed with the developed method, and ACPA amounts in the intracellular matrix and cell culture medium were found 773 ± 46.17 and 5175 ± 906.85 ng/ml, respectively (Fig. 4). The amount of ACPA in the intracellular matrix was found 15% of cell culture medium after 24 h incubation.

Fig. 4
figure 4

The ACPA concentrations in cell culture medium and intracellular matrix of L929 mouse fibroblast cells line samples (n = 3)

Full size image

Conclusion

The LC–MS/MS method developed is simple, fast, sensitive and reliable for the quantification of ACPA. The method was validated according to the ICH bioanalytical method guideline and was found selective, linear, sensitive, precise, accurate, robust and rugged. The developed and validated method was successfully applied for the quantification of ACPA in the cell culture medium and intracellular matrix. As a result, it can be recommended to use the LC–MS/MS method for monitoring of ACPA in drug development studies and pharmacokinetic studies to be performed in different biological matrices.