Key words
Casearia sylvestris
- Salicaceae - clerodane diterpene - hepatic metabolism - IVIVE
Abbreviations
[S]:
substrate concentration
cas B:
casearin B
casg F:
caseargrewiin F
CL:
clearance
CLH
:
hepatic total clearance
CLint
:
intrinsic clearance
CLint,H
:
hepatic intrinsic clearance
CYP:
cytochrome P-450
EH
:
hepatic extraction ratio
ESI:
electrospray ionization
fu,mic
:
unbound fraction in microsome medium
fu,p
:
unbound fraction in plasma
HLMs:
human liver microsomes
HQC:
high quality control concentration
IS:
internal standard
IVIVE:
In vitro-in vivo extrapolation
KM
:
Michaelis-Menten constant
LLOQ:
lower limit of quantification
log k:
log-transformed retention factors
log P:
octanol/water partition coefficient
LQC:
low quality control concentration
m/z
:
mass-to-charge ratio
MM:
Michaelis-Menten
MPPGL:
microsomal protein per gram of liver
MQC:
medium quality control concentration
Peff,man
:
human effective permeability
pKa:
acid dissociation constant
PTFE:
polytetrafluoroethylene
QC:
quality control
QH
:
hepatic blood flow
SGF:
simulated gastric fluid
v:
velocity
V0
:
initial velocity of the reaction
Vmax
:
maximum reaction velocity
Introduction
Casearia sylvestris Swartz, Salicaceae, is a native plant from Brazilian flora known as guaçatonga [1]. Herbal preparations of C. sylvestris leaves
have been used as antiseptic, anti-inflammatory, antiulcer, febrifuge, depurative, antidiarrheal, and antivenom drugs in Brazilian folk medicine. Its leaves contain different bioactive
secondary metabolites such as monoterpenes, sesquiterpenes, diterpenes, gallic acid derivatives, and glycosylated flavonoids [2]. Oxygenated tricyclic
cis-clerodane diterpenes (casearin-type diterpenes) are chemotaxonomic markers for Casearia. Over 40 diterpenes were identified in C. sylvestris ([Fig. 1]) [3], [4], most of them casearin-type diterpenes, some with antitumor, antiulcerogenic, anti-Helicobacter pylori,
and anti-inflammatory activity [2], [5], [6], [7], [8], [9], [10].
Fig. 1 Molecular structure of caseargrewiin F (casg F) and casearin B (cas B) (ChemBioDraw Ultra 14.0) [14], [45].
The antitumor action of casearin-type diterpenes has been evaluated since the 1980 s [11]. Besides the cytotoxicity against human tumor cell lines [2], [3], [5], [11], [12], [13], [14], [15], [16], casearin X and a pool of casearins inhibited tumor growth in mice when
orally administered [5]. Casearins promoted cell death via apoptotic pathways in vitro
[16] and were selective to tumor cell
lines [15]. C. sylvestris leaf extracts and casearin-type diterpenes showed antiulcer activity after oral administration in rats [10], [17]. Interestingly, cas B and casg F exhibited anti-inflammatory activity without gastric side effects when orally administered to rats [7]. However, the oral absorption and metabolism processes of the clerodane diterpenes cas B and casg F were never explored. In vitro assays simulating oral
administration of casearins J and O generated the respective dialdehydes as main degradation products and metabolites [18], as previously described for casearin X
[15].
The selection of drug candidates for oral administration is based on the desired physicochemical properties of the active compounds. The rule-of-five for drug-like compounds includes (1) less
than 5 H-bond donors, (2) molecular weight lower than 500, (3) log P values lower than 5, and (4) less than 10 H-bond acceptors [19]. The complementary assessment
of drug permeability, protein binding, in vitro metabolism, and preclinical pharmacokinetics reduces the risk of failure of drug canditates in later and more expensive clinical research
[20]. Due to the potential therapeutic applications of casearin-type diterpenes, this study aimed to evaluate cas B and casg F in physiological conditions,
including stability in plasma and SGF. Metabolism assays were conducted in HLMs in the presence and absence of a NADPH-regenerating system. The in vitro metabolism parameters were then
extrapolated to predict human hepatic clearance.
Results and Discussion
Complementary assays were conducted to characterize log P values, evaluate the stability of cas B and casg F in physiological conditions, and their metabolism in HLMs. The clerodane
diterpenes casg F and cas B were identified based on 1H and 13C NMR, UV, and UPLC-MS/MS data [14]. The chromatographic purities of casg F and
cas B were 98.6 and 95.7%, respectively, as confirmed by the normalization area method using HPLC-PDA/UV at 235 nm. Analytical and bioanalytical methods showed linearity for casg F and cas B
in the range of 0.039 to 1.25 µg/mL by applying the 1/y2 weighting factor (Table 1S-4S, Supporting Information).
Log P is a critical physicochemical property to predict drug permeability and oral bioavailability. The log P values assessed by the chromatographic method were 3.8 and 4.0 for casg F and cas
B, respectively, and were similar to predicted values ([Table 1] and Fig. 1S, Supporting Information). The chromatographic approach for log P determination
is accurate for log P values within 0 – 6 [21]. The moderate lipophilicity (2.0 < log P < 4.0) indicates that cas B and casg F have high cell permeability.
The predicted Peff,man values in the jejunum based on compound type, pKa, molecular weight, and log P was 15.8 × 10−4 and 16.7 × 10−4 cm/s (SimCYP V21
simulator). A similar Peff,man value was estimated for the analog casearin X (16.4 × 10−4 cm/s). Casearin X showed an apparent permeability of 66.9 × 10−6 cm/s
in Caco-2 cells, with an efflux ratio of 0.1, which suggests that active transport contributes to cell accumulation [21], [22], [23]. Up to now, there has been no information on the role of drug transporters in the oral bioavailability of casg F and cas B. In general, cas B and casg F have shown
favorable physicochemical properties for oral administration but may require formulation strategies to enhance aqueous drug solubility.
Table 1 Estimated log P values for the diterpenes caseargrewiin F (casg F) and casearin B (cas B) and their proposed primary metabolites through predictive fragmentation
and chromatographic methods.
|
|
log P*
|
clog P*
|
Observed** log P
|
|
*Predicted using ChemBioDraw Ultra 14.0; **experimentally determined by the chromatographic method
|
|
casg F
|
4.1
|
4.4
|
3.8
|
|
casg F dialdehyde
|
2.9
|
4.7
|
3.6
|
|
cas B
|
3.9
|
4.7
|
4.0
|
|
cas B dialdehyde
|
2.7
|
4.6
|
3.9
|
Stability assays in aqueous solutions at different physiological pH values are used to induce compound degradation in the gastrointestinal tract and body fluids after oral administration.
Both casg F and cas B rapidly degraded in SGF, indicating extensive degradation through the stomach pH. Casg F was more susceptible to acid degradation than cas B ([Fig. 2 a, b]). Degradation is probably related to the hydrolysis in the ester groups. Although casg F, cas B, and the ethanolic extract of C. sylvestris have shown oral
anti-inflammatory activity in animal models, the mechanism of action has not yet been elucidated. In vitro immunoassays for cas J and O showed no inhibition activity in nitric oxide in
macrophages or COX-I and II enzymes. Given the traditional oral use of C. sylvestris and its derivatives containing casearins, in vivo biological activity seems to be related to
the acid degradation products of casearins [18].
Fig. 2 Stability of caseargrewiin F (casg F) and casearin B (cas B) in simulated gastric fluid (a and b; pH 1.2), Hankʼs buffer (c and d; pH 7.4), and
TRIS buffer (e and f; pH 8.8). Data is presented as mean values ± SD of three independent experiments in two different concentrations (LQC = 0.078 µg/mL and HQC = 1.0 µg/mL).
*Statistical difference in relation to basal values (ANOVA followed by Tukeyʼs test, p < 0.05).
At the physiological plasma pH of 7.4, chemical degradation may reduce casg F and cas B plasma concentration levels ([Fig. 2 c, d]). Chemical degradation of both
diterpenes was observed at pH 8.8 but in a lower intensity when compared to pH 1.2 ([Fig. 2 e, f]). These results suggest that casg F and cas B undergo chemical
degradation as they pass through the intestinal lumen, probably due to the hydrolysis of their esters groups. The stability of casg F and cas B in human plasma was assessed up to 4.5 h, a time
equivalent to the plasma protein binding assay ([Fig. 3]). The decrease in diterpene concentrations was attributed to plasma esterases since plasma incubation with
the esterase inhibitor NaF (200 mM) resulted in no significant change in drug concentration over the incubation period. These results are consistent with the stability casg F and cas B in
aqueous buffer at pH 7.4 observed for up to 960 min [24]. The most common esterases in human plasma are carboxylesterases. Previous studies have shown that
casearin X is a substrate for carboxylesterases in rat and human liver microsomes and in Caco-2 cells [24], [25], [26].
Fig. 3 Stability of caseargrewiin F (a, casg F, 1.0 µg/mL) and casearin B (b, cas B, 1.0 µg/mL) in human plasma in the presence (blue) and absence (red) of the
esterase inhibitor NaF (pH 8.8). Data is presented as the mean values ± SD of three independent experiments. *Statistical difference in relation to basal values (ANOVA followed by Tukeyʼs
test, p < 0.05).
The unbound fractions of casg F and cas B in human plasma (fu,p) and microsomal medium (fu,mic) were determined by ultracentrifugation (Table 5S, Supporting
Information) [25], [27]. Mean fu,p values of casg F and cas B were 0.26 and 0.13, respectively. The low fu,p may
be attributed to the high lipophilicity of the diterpenes [28], [29]. In the incubation media, the fu,mic was 0.23 for casg
F and 0.13 for cas B. Fu,mic values are pivotal to accurately extrapolating in vitro metabolism parameters into in vivo parameters, and both drug lipophilicity and
protein concentration are determinants of fu,mic
[30]. Predictive tools are less accurate for highly lipophilic drugs [30],
highlighting the importance of the experimental determination of fu,mic. The fu,p and fu,mic of casg F and cas B suggested no binding saturation in the range
of 1.0 to 10 µg/mL [25], [28].
The stability of the diterpenes was evaluated in HLMs. Diclofenac depletion was assessed as a positive control for enzyme activity in the presence of a NADPH-regenerating system. The
substrate concentration of 1.0 µg/mL (1.98 µM for casg F and 1.73 µM for cas B) was selected according to previous studies with casearin X, to assure steady-state conditions with acmicrosomal
protein concentration of 0.2 mg/mL [25]. The depletion of casg F was observed within 10 min, while for cas B, the linear decline occurred over 15 min, with only 5
and 30% of remaining casearin, respectively. Both diterpenes were stable in PBS up to 30 min incubation, which was used as a negative control (Fig. 2S, Supporting Information). These
results show that casg F and cas B are substrates for human microsomal enzymes, and the in vitro metabolism conditions were suitable to describe depletion concentration-time profiles.
The incubation time of 5 min was selected for in vitro metabolism assays to ensure linear substrate depletion and steady-state conditions.
Similar depletion profiles observed without a NADPH-regenerating system indicate that the metabolism of casg F and cas B is not mediated by cytochrome P-450 (CYP) enzymes (Fig. 2S,
Supporting Information). NADPH is crucial for transferring electrons in CYP-mediated oxidative reactions [31], [32]. The metabolism of
Casg F and cas B in HLMs is likely associated with hepatic esterases. The depletion of both diterpenes in HLMs was inhibited by the esterase inhibitor NaF. Also, casg F and cas B showed a
similar metabolic profile as casearin X, which has been described as a substrate of hepatic microsomal esterases [25].
The classical MM model described the in vitro metabolism kinetics of casg F and cas B. This model assumes only one enzymatic binding site operating independently. However, some enzymes
may have more than one binding site, and substrate binding may increase or decrease the affinity to other binding sites. This unusual in vitro kinetic behavior results in a
sigmoid-shaped initial velocity of reaction versus the substrate concentration (V0 × S) curve. In some cases, it may be difficult to distinguish these curve profiles, leading to
erroneous in vivo predictions, with under or overestimating the maximum reaction velocity (Vmax) or clearance (CL) [33]. Eadie-Hofstee plots show
the velocity (v) plotted versus the ratio of velocity to the substrate concentration (v/[S]) to diagnose the atypical enzyme models. In MM kinetics, the hyperbolic curves are transformed into
a linear function in Eadie-Hofstee plots. Sigmoid plots indicate self-activation, while convex shapes indicate substrate inhibition [33], [34]. The enzyme kinetics of casg F and cas B in HLMs were described by a hyperbolic V0 × S function and linear V0 × V0/S function, confirming MM
kinetics ([Fig. 4]).
Fig. 4 In vitro metabolism of caseargrewiin F (casg F, upper panels) and casearin B (cas B, bottom panels) in human liver microsomes (n = 3). Substrate concentration
(S)-velocity (v) curves (a and d), Eadie-Hofstee plots of v versus v/S (b and e), and clearance plots of V0/S versus S (c and f).
Points were experimentally determined and is presented as the mean values ± SD of three independent experiments. In graphs a and d, solid lines are fit to the
Michaelis-Menten equation. In plots b and e, the solid line represents the linear regression of Eadie-Hofstee transformed data.
The Vmax for casg F was 2-fold higher than that of cas B. Similar KM values were observed for casg F and casg B in HLMs, suggesting that their apparent affinity to
microsomal enzymes is similar. This parameter allows for scaling the in vitro kinetic data to predict the in vivo clearance of substances (CLint) [35], [36]. The CLint of casg F was close to that previously reported for casearin X and approximately 2-fold higher than cas B
([Table 2]). On the other hand, the KM of casearin X in HLMs was 2-fold lower, indicating a higher apparent affinity for HLMs [25].
Table 2 Enzyme kinetics for caseargrewiin F (casg F) and casearin B (cas B) in human liver microsomes and in vitro-in vivo extrapolation (IVIVE) to predict
human hepatic clearance.
|
Compound
|
Vmax
(nmol/min/mg of protein)
|
KM
(µM)
|
CLint
(mL/min/mg of protein)
|
CLH (mL/min/kg)
|
QH
(mL/min/kg)
|
EH
(%)
|
|
Data is presented as an average (standard deviation, n = 3). Vmax: maximum reaction velocity; KM: Michaelis-Menten constant; CLint: intrinsic
clearance; CLH: hepatic total clearance; QH: hepatic blood flow; EH: hepatic extraction ratio
|
|
casg F
|
648 (14)
|
66.4 (6.5)
|
0.977 (0.05)
|
19.6
|
20
|
98.0
|
|
cas B
|
327 (6)
|
61.4 (5.1)
|
0.534 (0.08)
|
19.2
|
20
|
96.0
|
In vitro metabolism data were applied in the well-stirred liver model to predict hepatic metabolic clearance. The well-stirred model is often applied for its simplicity, considering
that the liver is a single isolated compartment with a homogeneous distribution of metabolic enzymes. The model assumes rapid distribution, and equilibrium between tissue and blood
concentrations [35], [37]. Predicted CLH values of casg F and cas B were high and corresponded to 98.0 and 96.0% of total
QH. This IVIVE suggests that both diterpenes are almost entirely cleared in the liver. These findings are consistent with the ones reported for casearin X [25].
Our data reveal that the oral bioavailability of casg F and cas B is low due to chemical instability in the gastrointestinal pHs and pre-systemic hepatic elimination. The in vivo
activity of cas B and casg F might be associated with active degradation products or metabolites. The current study has limitations. First, phase II metabolism was not assessed, and metabolism
was not assessed in tissues other than the liver. Secondly, intestinal apparent permeability was not experimentally determined in vitro. This information is relevant for estimating the
human effective permeability. However, due to the high lipophilicity, intestinal permeability does not seem to limit oral bioavailability. Up to now, there is no information regarding the
effect of drug transporters in the kinetic disposition of cas B and casg F. These results break new ground to support the investigation of active degradation products of oral preparations of
C. sylvestris leaf extracts.
Materials and Methods
Plant material
Leaves were collected from 10 specimens of C. sylvestris (guacatonga) (25 g each specimen) at the School of Pharmaceutical Sciences, São Paulo State University, Araraquara, SP,
Brazil (21°48′83″3 to 21°48′98″9 S and 48°11′86″1 to 48°12′13″3 W), in June 2017. The specimens were identified by Prof. Dr. Luis Vítor Silva do Sacramento. Exsiccates of the specimens (FAC
101 – 110) were deposited in the Scientific Herbarium of the State Maria Eneyda P. Kauffman Fidalgo of the Botanic Institute of State of São Paulo Government. Leaves were dried at
40 °C for 3 days and powdered using a knife mill [38]. This work was registered in the Genetic Heritage Management System and Associated Traditional Knowledge
(SisGen certificate n° AC1CEE6).
Diterpenes extraction and purification
Casg F and cas B were isolated from the ethanolic extract of the leaves of C. sylvestris. The dried and powdered leaves (250 g) were extracted by maceration with ethanol (1 : 15 w/v,
120 h) at 40 °C with occasional stirring. The ethanol was evaporated using an IKA DEST KV 05S3 evaporator to yield the dry extract (29.5 g, 11.8% w/w). The extract was fractionated as
described by Spósito et al. [6] through normal-phase solid-phase extraction and column chromatography. Casg F and cas B were obtained from subfractions SF 20 – 23
after HPLC-UV semipreparative purification (t
R: 16.8 and 19.4 min, respectively) using an Agilent Eclipse XDB C18 column (250 × 21.20 mm, 7 µm) and
methanol : water (73 : 27% v/v) as the mobile phase in the isocratic mode for 90 min. UV detection was monitored at 235 nm. The flow rate was 8.0 mL/min, and 1000 µL of the samples were
injected into the chromatographic system. The amounts of purified casg F and cas B were 158.0 and 75.0 mg, respectively.
Compound identification
Casg F and cas B were analyzed by UPLC-QTOF-MS/MS. The Acquity UPLC Waters chromatograph coupled to a quadrupole/time of flight system (XEVO-QToF; Waters) was equipped with a BEH C18 Waters
column (150 × 2.1 mm, 1.7 µm). A linear gradient with acetonitrile : water (both containing 0.1% formic acid) 30 : 70 to 90 : 10 (%, v/v) was performed for 30 min. The flow rate was
0.25 mL/min, and the injection volume was 5 µL (0.05 – 0.1 mg/mL). The capillary voltage of ESI was 2.6 kV, and the cone voltage was set at 0.5 V in the positive ion mode. The source and
desolvation temperatures were set at 120 and 350 °C respectively, and the desolvation gas flow rate was 500 L/h. Full-scan analysis was performed in the mass range of 300 to 700 mass units
(m/z). The data acquisition and analysis were carried out using MassLynx version 4.1 (Waters) [18], [39].
The NMR spectra of casearins were obtained using a Bruker 7.0 T spectrometer in pulsed-gradient mode at 600 MHz for 1H and 150 MHz for 13C in
pyridine-d5
, and chemical shifts are expressed in δ (ppm) units. Spectroscopic data of casg F and cas B are provided as Supporting Information.
Octanol/water partition coefficient determination
Log P was determined using the chromatographic method [40]. Different combinations of methanol : water, varying from 60 to 80% of methanol (v/v), and flow rates
(0.6 to 1.0 mL/min) were tested in a Thermo chromatograph equipped with a Nucleosil C18 Macherey-Nagel column (250 × 4.6 mm, 5 µm). The mobile phase was methanol : water (75 : 25%, v/v) in
the isocratic mode for 50 min at a flow rate of 0.8 mL/min. The injection volume was 40 µL. UV detection was performed at 210 nm. Seven reference compounds with known log P values
(acetanilide, benzene, toluene, naphthalene, biphenyl, phenanthrene, and triphenylamine) were analyzed. Log-transformed retention factors (log k) of each reference compound were plotted
against log P values, and the linear regression equation was used to estimate log P values for casg F and cas B. Log P estimated values were compared to those predicted by ChemBioDraw Ultra
14.0.
Chemical stability
Chemical stability was investigated in SGF (200 mg NaCl + 0.7 mL 37% HCl, pH 1.2) at pH 8.8 using Tris buffer [18.17 mg of tris(hydroxymethyl)aminomethane (Tris) + 2 M NaOH in 1 L of
ultrapure water), and at pH 7.4 using Hankʼs buffer [9.83 g of Hankʼs Balanced Salt solution Sigma H8264 in 1 L ultrapure water, 0.35 g of NaHCO3 and pH adjusted with 1 N HCl].
Quality control samples at low (LQC = 0.039 µg/mL) and high (HQC = 1.0 µg/mL) concentration levels of cas B and casg F were evaluated. The solutions were kept at 37 °C with constant
agitation at 100 rpm. Samples were collected in triplicate at times 0, 2.5, 5, 10, 15, 20, 30, 45, 60, 90, 120, and 240 min for pH 1.2; at times 0, 20, 40, 60, 90, 120, 240, 360, 480, 960,
and 1440 min for pH 7.4; and at times 0, 5, 10, 15, 30, 45, 60, 120, 240, 480, 960, and 1440 min for pH 8.8. After filtration using a PTFE syringe filter (0.22 µm, 13 mm), the remaining
concentration of the diterpenes in the samples was quantified by LC-MS. Results are expressed as the remaining compound in percentage (%).
Stability in human plasma
The stability of casg F and cas B was assessed in human plasma obtained after blood donations from 6 healthy volunteers. The local institutional review board approved the clinical protocol
for blood donation (CAAE nº. 96 952 918.3.0000.5426). All assays were carried out for at least 4 h, which is equivalent to the duration of the plasma protein binding assay. Working solutions
(100 µg/mL) were used to prepare plasma samples with the initial concentration of casg F and cas B of 1.0 µg/mL in plasma, equivalent to 1.98 µM for casg F and 1.73 µM for cas B. The final
percentage of organic solvent in plasma was less than 1.0%. All samples were kept at 37 °C in 12-well plates with a final volume of 1.5 mL and shaken horizontally at 100 rpm. The effect of
esterase activity was assessed by preincubating the plasma with the esterase inhibitor NaF (200 mM) for 10 min [41]. Samples were collected at times 0, 5, 10,
15, 30, 60, 120, and 270 min. The remaining drug concentration in plasma was quantified by a bioanalytical LC-MS method validated in this study.
Metabolism in human liver microsomes
HLMs (MOLTOX) were diluted in 0.1 M PBS saline phosphate buffer pH 7.4 (Sigma). The NADPH-regenerating system was composed of a mixture of 2.89 mM NADP+, 0.89 U/mL
glucose-6-phosphate dehydrogenase, 7.34 mM glucose-6-phosphate, and 7.34 mM MgCl2
[25]. The incubation medium including HLMs and the
NADPH-regenerating system were preincubated at 37 °C (100 rpm) for 10 min in 12-well plates. Reactions were initiated by adding casg F or cas B dissolved in PBS + 1.0% glycerin to each well.
All reactions were conducted in triplicate with varying substrate concentrations (0.5 – 100 mg/mL; 0.1% max organic content). Positive control samples were prepared by adding HLMs, a NADPH
regenerator system, and 2 µM diclofenac. Negative control samples were prepared by adding substrate solutions to 0.1 M PBS only. The NADPH effect was assessed by adding or not the
NADPH-regenerating system. The effect of esterase metabolism was evaluated by preincubation with 200 mM NaF. All assays were carried out in a final volume of 1.5 mL. Enzyme concentration in
the incubation medium was 0.2 mg/mL, corresponding to half of the linear range of diterpene depletion (0 – 0.4 mg/mL) [25]. The depletion of casg F and cas B was
evaluated at times 0, 1.0, 2.5, 5.0, 7.5, 10, 15, and 30 min and all reactions were terminated by adding 150 µL iof ce-cold acetonitrile.
Metabolism data analysis
The kinetic parameters (KM and Vmax) were determined by nonlinear regression using GraphPad Prism 9.4.1. Casg F and cas B were monitored in the linear range of
depletion. The CLint was determined as the ratio Vmax/KM. The estimated kinetic parameters and fu,p and fu,mic were extrapolated to the
CLH and the EH
[42].
In vitro-in vivo extrapolation
The CLint in HLMs was scaled up to the CLint,H using the equation below. The scaling factors applied were the MPPGL (45 mg/g of liver) and the liver weight per kg of
body weight (20 g/kg of body weight) [28], [35].
The well-stirred model, represented in the following equation, was used to predict the CLH, where the QH was set to 20 mL/min/kg [28], [35].
LC-MS analysis
Stock solutions of casg F (98.6% purity) and cas B (95.7% purity) were prepared at 5000 µg/mL in acetonitrile and kept at − 20 °C. A Waters HPLC ACQUITY QDa chromatograph equipped with a
Nucleosil C18 Macherey-Nagel column (250 × 4.6 mm, 5 µm) was used. The mobile phase was a linear gradient of 80% acetonitrile (0.1% formic acid) (v/v) to 100% acetonitrile (0.1% formic acid)
(v/v) in 8 min; 100% acetonitrile (0.1% formic acid) (v/v) for 5 min, and re-equilibration with 80% acetonitrile (0.1% formic acid) (v/v) for 4 min. The flow rate was 0.7 mL/min, and the
injection volume was 10 µL. The mass range used in the full-scan analysis mode was from 300 to 700 mass units (m/z) and casg F and cas B were monitored by the masses of molecules
cationized with sodium [M + Na]+ at m/z 527 and 599, respectively. In the bioanalytical method, thymol (IS) was monitored at m/z of 151 [M + H]+. The
LC-MS method was applied for different matrices and validated according to official guidelines [43], [44]. Further information on
the method is provided in the Supporting Information.
Analytical method
The calibration standards were prepared by serial dilution at the levels 0.039, 0.078, 0.156, 0.3125, 0.625, 1.0, and 1.25 µg/mL in acetonitrile or Hankʼs buffer. All solutions were
filtered through a PTFE syringe filter (0.22 µm, 13 mm) and placed in vials for the LC-MS analysis. QCs were prepared at the LLOQ (0.039 µg/mL), LQC (0.078 µg/mL), MQC (0.625 µg/mL), and
HQC (1.0 µg/mL) levels of concentration.
Bioanalytical method
Stock solutions of casg F and cas B (5000 µg/mL, in acetonitrile) were diluted in plasma containing 1% glycerin or microsomal medium (without cofactors) to prepare the calibration
standards at 0.039, 0.078, 0.156, 0.3125, 0.625, 1.0, and 1.25 µg/mL in each matrix. NaF (200 mM) was added to avoid esterase enzymatic degradation. QC samples were prepared at the same
levels for the analytical method. A solution of 50 µg/mL of thymol (purity ≥ 99.9% GC area; Sigma-Aldrich) in acetonitrile was used as an IS.
Protein precipitation using organic solvent yielded a recovery > 95% for the diterpenes and IS [22]. Briefly, plasma samples (150 µL) were added to the
same volume of acetonitrile with the IS at 50 µg/mL, mixed in a shaker for 10 s, and centrifuged at 13 000 × g for 15 min. The supernatants were collected, filtered through a PTFE
syringe filter (0.22 µm), and transferred into vials for LC-MS analysis. Post-processing stability (up to 24 h at 25 °C) was evaluated for LQC and HQC in quintuplicate. Deviations of up to
± 15% from the nominal value were considered acceptable.
Statistics
Statistical analyses were performed on GraphPad Prism 9.4.1 using analysis of variance (ANOVA) with Tukeyʼs post hoc test assuming significance for p < 0.05. The results are presented
means ± the standard error. RStudio v4.0.3 with dplyr, ggplot2, patchwork, plyr, and stats R packages were used for the plots.
Contributorsʼ Statement
Design of the study: F. B. Oda, R. G. Peccinini, N. V. Moraes, A. G. Santos; data collection: F. B. Oda, F. A. Carvalho, J. A. Oliveira, G. J. Zocolo, P. R. V. Ribeiro; statistical analysis:
F. B. Oda, P. A. Yamamoto, N. V. Moraes; analysis and interpretation of the data: F. B. Oda, F. A. Carvalho, P. A. Yamamoto, J. A. Oliveira, R. G. Peccinini, N. V. Moraes, A. G. Santos;
drafting the manuscript: F. B. Oda, N. V. Moraes, A. G. Santos; F. A. Carvalho; revision of the manuscript: F. B. Oda, P. A. Yamamoto, N. V. Moraes, A. G. Santos.
