- Email: [email protected]
A validated bioanalytical method in mouse, rat, rabbit and human plasma for the quantification of one of the steroid glycosides found in Hoodia gordonii extract
Food and Chemical Toxicology 50 (2012) S14–S19
Contents lists available at ScienceDirect
Food and Chemical Toxicology journal homepage: www.elsevier.com/locate/foodchemtox
A validated bioanalytical method in mouse, rat, rabbit and human plasma for the quantification of one of the steroid glycosides found in Hoodia gordonii extract Weijun Wang a,⇑, Paul J. Russell b, Graeme T. Clark c, Derek Lewis c, Kun Nang Cheng c a
Phytopharm plc, Lakeview House, 2 Lakeview Court, Ermine Business Park, Huntingdon, Cambridgeshire PE29 6UA, UK Safety and Environmental Assurance Centre, Unilever, Colworth Science Park, Sharnbrook, Bedfordshire MK44 1LQ, UK c Quotient Bioresearch Ltd., Rushden, Northamptonshire NN10 6ER, UK b
a r t i c l e
i n f o
Article history: Available online 25 August 2010 Keywords: Bioanalysis Steroid glycoside Plasma LC–MS/MS Validation Hoodia gordonii extract
a b s t r a c t Hoodia gordonii extract contains steroid glycosides, fatty acids, plant sterols and polar organic material. Certain steroid glycosides show appetite suppressant activities following oral ingestion. This study describes the validation of a bioanalytical method for the quantification of one of the steroid glycosides, H.g.-12 (10% (w/w) of the extract), in mouse, rat, rabbit and human plasma. The method utilises a liquid–liquid extraction with methyl-tert-butyl ether followed by chromatographic separation on a 2.1 50 mm C18 Genesis high performance liquid chromatography (HPLC) column and detection on a triple quadrupole mass spectrometer. Detection of H.g.-12 and its stable isotope internal standards is performed using positive TurboIonspray™ ionisation in multiple reaction monitoring mode. The validation procedure demonstrated assay sensitivity, linearity, accuracy, precision and selectivity over the calibration range of 0.5–150 ng/mL in human plasma (500 lL sample volume), 1.0–100 ng/mL in rat and rabbit plasma (150 lL sample volume) and 1.0–250 ng/mL in mouse plasma (150 lL sample volume) with good recoveries (P77%). H.g.-12 was stable in plasma for P6 months at 20 °C, for up to 4 h at ambient temperature (ca 22 °C) and after 3 freeze–thaw cycles. Plasma extracts were stable for up to 24 h at ambient temperature. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction Hoodia gordonii extract is obtained from H. gordonii, a succulent plant originally found in the Kalahari Desert of Southern Africa (Knight et al., 2012). The potential of Hoodia extracts for weight management was first identified in several animal species by the National Food Research Institute of the CSIR in South Africa (van Heerden et al., 1999, 2002; Tulip et al., 2001) which conducted a nationwide survey to determine the nutritional value of food from the wild. The genus Hoodia was studied because certain species of this succulent plant were known to be consumed by the San people as a substitute for food and water. A CSIR scientist who studied the biological effects of Hoodia extracts on small laboratory animals, observed that the animals lost their appetite, accompanied by a loss of weight. In order to determine the systemic exposure, folAbbreviations: HPLC, high performance liquid chromatography; FDA, Food and Drug Administration; LLE, liquid–liquid extraction; LC–MS/MS, liquid chromatography tandem mass spectrometry; MRM, multiple reaction monitoring; MTBE, methyl-tert-butyl ether; QC, quality control; CAD, collision activated dissociation; RE, relative error; CV, co-efficient of variance; ANOVA, analysis of variance; LLOQ, lower limit of quantification. ⇑ Corresponding author. Tel.: +44 (0) 1480 437 697; fax: +44 (0) 1480 417 090. E-mail address: [email protected] (W. Wang). 0278-6915/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.fct.2010.08.014
lowing administration of H. gordonii extracts within pre-clinical and clinical studies (Dent et al., 2012; Scott et al., 2012; Blom et al., 2011), an appropriate bioanalytical method required development and validation to the standard criteria outlined in the regulatory guidance document (Food and Drug Administration (FDA) document1). Previous investigations (van Heerden et al., 1999) determined that one major component of Hoodia extract (10% (w/w) of H. gordonii extract) responsible for eliciting an efficacy response is the steroid glycoside H.g.-12 (Fig. 1). This component was isolated and purified from H. gordonii for use in the method development and formal validation. In conjunction, deuterated H.g.-12 (produced via high pressure deuterium exchange) was used as an internal standard for all species validation. This was subsequently replaced by an isotopically substituted H.g.-12 (which was chemically synthesised from isolated H.g.-12) via partial validation for all reported species except rabbit. Liquid–liquid extraction (LLE) is a useful procedure to isolate particularly lipophilic compounds from biofluids. The extraction properties of the organic solvent tend not to differentiate between 1 Guidance for Industry: Bioanalytical Method Validation. May 2001, US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER), Center for Veterinary Medicine (CVM).
S15
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
O O
O
OH HO MeO
O
O
O OH
O
O OMe
O
OMe
Fig. 1. Chemical structure of steroid glycoside reference standard H.g.-12.
subtle structural differences and as Hoodia extracts contain a number of structurally related moieties, LLE is deemed appropriate for sample preparation. The hyphenated technique of liquid chromatography tandem mass spectrometry (LC–MS/MS) has become the industry standard for quantitative measurements of both parent compound and metabolically related material present in biological matrices. When combined with reversed phase chromatography (utilising the multiple reaction monitoring (MRM) capability of triple quadrupole mass spectrometers), highly specific measurements of diagnostic precursor and product ions can produce LC–MS/MS methods capable of sensitive and selective quantification in a relatively short chromatographic cycle time (when compared with other chromatography based detection techniques). A method for the separation and analysis of extracted plasma samples was developed on C18 based reverse phase high performance liquid chromatography (HPLC) column with MRM detection on a triple quadrupole mass spectrometer. In this paper, we present the validated methodology for quantification of H.g.-12 in human, rat, mouse and rabbit plasma samples. The validations were based on the single H.g.-12 chemical entity, and then the method was applied to the measurement of co-occurring epimers (referred to as H.g.-12 equivalents) in preclinical and clinical study samples. This article is one of six published as a supplement providing information on various aspects of H. gordonii. 2. Materials and methods 2.1. Chemicals and reagents Acetonitrile (HPLC grade), ammonium acetate (HPLC grade), formic acid (specified reagent 98/100%) and methyl-tert-butyl ether (MTBE – HPLC grade) were all purchased from Fisher-Scientific (Loughborough, UK). HPLC grade water was obtained from an in-house Milli-Q purification system. Blank human plasma (stabilised with lithium heparin) was obtained from Richard Pharmacology, Atkinson Morley’s Hospital (London, UK), whilst blank rat, mouse and rabbit plasma was obtained from B&K Universal Limited (Grimston, UK). All blank plasma was stored at ca 20 °C when not in use. H.g.-12 (P92.4%) reference standard and H.g.-12 internal standards (deuterated H.g.-12 and isotopically substituted H.g.-12) were all supplied by Phytopharm plc. 2.2. Preparation of stock solutions, spiking solutions, quality control (QC) samples and spiking internal standard solution Stock solutions of the standard (H.g.-12), deuterated H.g.-12 (D-H.g.-12) and isotopically substituted H.g.-12 (IS-H.g.-12) were prepared in acetonitrile:water (1:1, v/v) at a final concentration of ca 100 lg/mL. These stock solutions were all stored at ca 20 °C when not in use. Calibration standards 0.5–150 ng/mL for H.g.-12 in human plasma (500 lL sample volume), 1.0–100 ng/mL in rat and rabbit plasma (150 lL sample volume) and 1.0–250 ng/mL in mouse plasma (150 lL sample volume) were prepared daily. Quality control samples were prepared from independent stock solutions (separate weighing) at four nominal concentrations of 0.5, 1.5, 60 and 120 ng/mL for H.g.-12 (human plasma); 1.0, 3.0, 125 and 200 ng/mL (mouse plasma); 1.0, 3.0, 50 and 80 ng/mL (rat and rabbit plasma). Quality control samples were prepared in bulk at the start of the validation study and stored at ca 20 °C when not in use.
D-H.g.-12 was prepared at either 500 or 1000 ng/mL for human plasma analysis (depending on batch of material) and 250 ng/mL for rat, mouse and rabbit plasma by appropriate dilution of the stock solution with acetonitrile:water (1:1, v/v). ISH.g.-12 spiking solution for human, rat and mouse plasma was prepared at 100 ng/mL by appropriate dilution with acetonitrile:water (1:1, v/v). 2.3. Sample preparation (extraction) The QC samples and spiking solutions were allowed to reach room temperature, vortex mixed and the following procedures followed: Blank plasma (500 lL for human, 150 lL for all other species) for blank, zero and calibration standards were aliquotted into glass tubes and each calibration standard was spiked with its respective spiking solutions at appropriate volumes (e.g., 50 lL). Each bulk prepared QC (500 lL for human, 150 lL for all other species) was aliquotted into glass tubes (n = 6 each per level) and each was spiked with appropriate volumes (e.g., 50 lL) of acetonitrile:water (1:1, v/v). Each zero and blank sample was spiked with appropriate volumes (e.g., 50 lL) of acetonitrile:water (1:1, v/v). All samples (except blanks) were spiked with 50 lL D-H.g.12 internal standard (except human (125 lL)) or 20 lL IS-H.g.-12 (except human (50 lL) and rabbit (not done)). Blanks were spiked with an equivalent volume of acetonitrile:water (1:1, v/v). The samples were then diluted with water (400 lL) and briefly vortex mixed (ca 2–5 s). MTBE (5 mL) was added to each sample prior to mixing by rotation (ca 15 min), centrifuged (3000 rpm, 10 min) and then stored at ca 80 °C for P30 min at which point the MTBE layer was then carefully removed, taking care not to disturb or sample any of the precipitated material immediately above the frozen lower layer. The organic layer was concentrated to dryness and reconstituted in acetonitrile:water (1:1, v/v, 120 lL). 2.4. HPLC conditions (separation) Chromatographic retention and elution of H.g.-12 and internal standard was performed on a Genesis C18, 2.1 50 mm, 120 Å, 4 lm HPLC column (Kinesis, Bedfordshire, UK). Elution was performed with an isocratic mobile phase of acetonitrile (10 mM ammonium acetate + 0.05% (v/v) formic acid), 80:20 (v/v) at a flow rate of 200 lL/min. A Security Guard C18 2.0 4 mm pre-column (Phenomenex, Cheshire, UK) was used during the course of the validation, which was replaced prior to each analytical batch. 2.5. MS conditions (detection) Detection of H.g.-12 and internal standard was obtained on an API 3000 mass spectrometer (Applied Biosystems-MDS Sciex., Warrington, UK) operating under positive TurboIonspray™. All analytes were observed as the [M+NH4]+ ammoniated precursor ions which provided diagnostic product ions under MRM conditions with nitrogen as both the nebuliser and collision activated dissociation (CAD) gas. Full MS parameters can be found in Table 1.
3. Results 3.1. Chromatography and linearity Typical LC–MS/MS chromatograms (MRM) for a human plasma blank, extracted zero (internal standard only), extracted S1 (lower limit of quantification – 0.5 ng/mL) and extracted S8 (upper limit of quantification – 150 ng/mL) are presented in Fig. 4. These demonstrate that there is no endogenous material or analytical interference in either the H.g.-12 or internal standard ion channels. Comparable chromatography was observed for all other species. A linear calibration response was observed across the calibration range of 0.5–150 ng/mL for H.g.-12 in human (500 lL sample
S16
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
Table 1 API 3000 mass spectrometric parameters for the detection of H.g.-12, and D-H.g.-12 or IS-H.g.-12. Analyte
NEB
CUR
IS (V)
TEM (°C)
CAD
Mass transition
CE
H.g.-12 D-H.g.-12 IS-H.g.-12
7–8
6–9
4000–4500
200–300
4
m/z 896 ? 295 m/z 907 ? 298 m/z 906 ? 295
33 39 33
NEB, Nebuliser gas; CUR, curtain gas; IS, ionspray voltage; TEM, temperature; CAD, collision activated dissociation; CE, collision energy.
volume), 1.0–250 ng/mL in mouse (150 lL sample volume) and 1.0–100 ng/mL in rat and rabbit plasma (150 lL sample volume). Linear regression with 1/x2 weighting was applied. Correlation co-efficients for H.g.-12 during the course of the human validation with D-H.g.-12 as the internal standard were >0.99. Back-calculated concentrations for the calibration standards provided relative errors of the mean of 6 7.2% and precision (CV) of 69.9%. Example calibration lines for H.g.-12 from this validation can be seen in Fig. 5. Comparable statistics were observed for rat, mouse and rabbit validations with D-H.g.-12 and for all three species (human, rat and mouse) which were partially validated with IS-H.g.-12.
three species (Table 3). The accuracy (RE) and precision (CV) values of H.g.-12 were 6 4.7% and 612.7% in human plasma, 618.3% and 66.5% in rat plasma and 6 10.4% and 65.1% in mouse plasma. 3.3. Recovery Recovery of H.g.-12 and the internal standard (D-H.g.-12) were good (P77%) and reproducible from all species investigated (human, rat and mouse). There is a balance between higher recoveries and the avoidance of possible interference by sampling the precipitated material in the MTBE layer after freezing.
3.2. Accuracy and precision
3.4. Selectivity
3.2.1. Human, rat and mouse plasma with D-H.g.-12 internal standard Accuracy (RE) and precision (CV) were determined on three separate occasions, at the four QC levels in replicate (n = 6 per level) in human, rat and mouse plasma. Inter-batch analysis was performed using the analysis of variance (ANOVA) method with the full details being presented in Table 2. Results for H.g.-12 in these three species were well within the acceptance criteria of ±15% (±20% at lower limit of quantification – LLOQ) for both accuracy (in terms of relative error (RE)) and precision (as measured by co-efficient of variation (CV)). The accuracy (RE) and precision (CV) values of H.g.-12 were 6 13.1% and 612.5% for human plasma, 66.1% and 68.3% for rat plasma, and 6 7.6% and 613.8% for mouse plasma.
Analysis of different plasma batches (n = 6 per species) demonstrated no interference for the analysis of H.g.-12 at its respective retention time. The addition of either internal standard (referred to as zero samples) did not introduce any additional chromatographic peaks in either ion channel. Quantification of the six different plasma batches from each species at an appropriate level within the calibration range (S6) demonstrated accuracy and precision within ±15% for H.g.-12 in human, rat and rabbit plasma when using D-H.g.-12 as the internal standard. Quantification of H.g.-12 in mouse plasma with D-H.g.12 provided results outside the recommended ±15% highlighting a selectivity issue unique to this species. However, quantitative selectivity re-assessment with IS-H.g.-12 produced accuracies well within ±15% for all species investigated (human, rat and mouse), implying that the issues observed with selective quantification in mouse plasma (using D-H.g.-12) were due to a specific ion suppression effect on the structural analogue, which varied between individuals.
3.2.2. Rabbit plasma with D-H.g.-12 internal standard Accuracy (RE) and precision (CV) were determined on a single occasion, at the four QC levels in replicate (n = 6 per level) with the full details being presented in Table 2. Results for H.g.-12 were well within the acceptance criteria of ±15% for both accuracy and precision (±20% at LLOQ), and the accuracy (RE) and precision (CV) values of H.g.-12 were 66.9% and 612.1%. 3.2.3. Human, rat and mouse plasma with IS-H.g.-12 internal standard Accuracy (RE) and precision (CV) were determined on a single occasion, at the four QC levels in replicate (n = 6 per level) for each species. Results for H.g.-12 were well within the acceptance criteria of ±15% for both accuracy and precision (±20% at LLOQ) for all
3.5. Matrix effects (ion suppression and enhancement) A consequence of LC–MS/MS bioanalysis is the potential for coextracted material to co-elute under the relatively short chromatographic cycles times and interfere with the processes controlling the ionisation of analyte and internal standards. These matrix effects can either suppress or enhance the ion signal resulting in an under estimation of the analyte (if the analyte is suppressed
Table 2 Accuracy (relative error = RE) and precision (co-efficient of variance = CV) for H.g.-12 in human, rat, mouse and rabbit plasma QC samples from three batches with D-H.g.-12 as internal standard. H.g.-12 Human plasma QCLLOQ RE range (CV) intra-batch QCLOW RE range (CV) intra-batch QCMED RE range (CV) intra-batch QCHIGH RE range (CV) intra-batch QCLLOQ RE range (CV) inter-batch QCLOW RE range (CV) inter-batch QCMED RE range (CV) inter-batch QCHIGH RE range (CV) inter-batch
4.5% to +7.3% (6.1%) 5.5% to +2.4% (12.5%) 13.1% to +3.1% (9.6%) 11.8% to 2.5% (6.4%) 1.0% (7.0%) 2.7% (3.0%) 3.1% (8.6%) 5.6% (5.4%)
Rat plasma 5.2% 3.7% 2.0% 6.0% 0.1% 2.0% 1.7% 2.7%
to +6.1% (6.7%) to 0.3% (6.7%) to 1.2% (6.5%) to 0.0% (8.3%) (5.1%) (NC) (NC) (2.3%)
Mouse plasma
Rabbit plasma
7.6% to +5.2% (13.8%) 6.4% to +2.3% (5.7%) 1.3% to +2.4% (2.6%) 3.0% to 1.7% (3.5%) 0.8% (5.0%) 0.9% (4.5%) 0.2% (1.7%) 2.6% (NC)
+6.9% (12.1%) 1.1% (12.1%) 1.7% (6.8%) 2.3% (8.0%) NA NA NA NA
NA, Not applicable, only single batch analysed; NC, not calculable by ANOVA method due to the intra-batch variability being greater than the inter-batch variability.
S17
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
3.7. Stability
Table 3 Accuracy (relative error = RE) and precision (co-efficient of variance = CV) for H.g.-12 in human, rat and mouse plasma QC samples from single batch with IS-H.g.-12 as internal standard.
Assessments of stability were performed (24 h extract, 3 freeze– thaw, 4 h at room temperature, storage at ca 20 °C for 1 week and 1 month) on the low and high level QC samples for each species. No significant deterioration of H.g.-12 was observed in any of the assessments with relative error within ±15% of the nominal concentrations being obtained in all cases. Long term storage stability studies (1, 3, 6, 12, 16 and 24 months for human and rat plasma; 1, 3, 6, 9, 12, 18 and 24 for mouse plasma; 1, 3 and 6 months for rabbit plasma) were conducted, and the results indicated that H.g.-12 in human and rat plasma QC samples were stable for 24 months, mouse plasma for 12 months (after 12 months, there was not enough QCHIGH stability sample left for analysis) and rabbit plasma for 6 months (the last time point) when stored at ca 20 °C. Details of human plasma storage stability can be seen in Table 4.
H.g.-12
QCLLOQ RE and (CV) intra-batch QCLOW RE and (CV) intra-batch QCMED RE and (CV) intra-batch QCHIGH RE and (CV) intra-batch
Human plasma
Rat plasma
Mouse plasma
0.7% (12.7%)
18.3% (6.5%)
1.5% (5.1%)
0.1% (2.2%)
0.6% (4.0%)
10.4% (1.7%)
5.1% (1.8%)
4.1% (1.2%)
9.4% (2.9%)
8.7% (4.8%)
1.4% (2.1%) 4.7% (5.8%)
Table 4 Stability of H.g.-12 in human plasma QCLOW and QCHIGH samples expressed at mean accuracy relative to target (%) with D-H.g.-12 or IS-H.g.-12 as internal standard. QC samples storage conditions
The effect of dilution was assessed by preparing a solution of H.g.-12 in human, rat and rabbit plasma at 10 QC medium (QCMED 10), and in mouse plasma at 10 and 50 QC medium (QCMED 10 and 50). The dilution results demonstrated excellent accuracy and precision of the procedure for the analyte in all species. The human plasma dilution results are presented in Table 5.
Mean accuracy relative to target (%) for QCLOW and QCHIGH samples
4 hr in plasma at RT (ca 22 °C) 24 hr extract stability (ca 22 °C) Three freeze thaw cycles (ca 20 °C) Storage (ca 20 °C): 1 week Storage (ca 20 °C): 1 month Storage (ca 20 °C): 3 months Storage (ca 20 °C): 6 months Storage (ca 20 °C): 12 months Storage (ca 20 °C): 16 months Storage (ca 20 °C): 16 months* Storage (ca 20 °C): 24 months* *
3.8. Dilution
H.g.-12 QCLOW
H.g.-12 QCHIGH
96.9 104.9 88.4 89.7 112.3 97.2 101.1 87.7 91.8 100.7 109.8
90.8 90.4 89.9 91.6 109.9 90.9 95.0 82.6 90.6 101.6 97.0
4. Discussion H. gordonii extracts contain a number of chemical entities (>20) which are chemical analogues or epimers of H.g.-12 (Pawar et al., 2007; Abrahamse et al., 2007; Janssen et al., 2008; Russell and Swindells, 2012). Following oral administration of H. gordonii extracts in animals or in humans, the metabolic process may produce more structural analogues. Therefore, it was impractical to develop and validate a bioanalytical method to separate all these isomers with acceptable criteria and reasonable chromatographic cycle time to support pre-clinical and clinical studies. The initial development and validations were based on the single H.g.-12 epimer using deuterated H.g.-12 (D-H.g.-12) as an internal standard. The D-H.g.-12 (Fig. 2), produced by catalytic reduction of H.g.-12 with deuterium, consists of several deuterated species due to the catalytic reduction of one or both double bonds in H.g.-12 along with non-selective deuterium incorporation. By using this compound as the internal standard, all species passed the standard validation criteria outlined in the guidance (FDA document1) with the exception that the mouse plasma selectivity assay results were outside the recommended ±15% (i.e., 6±20.3%), indicating a specific ion suppression effect on this species. Following further research and development, a new batch of internal standard was synthesised by introducing 13C3 and 2H7 into H.g.-12 (Fig. 3). Quantitative selectivity re-assessment with this new isotopically substituted H.g.-12 internal standard (IS-H.g.-12) produced accuracies well within ±15% for all species (human, rat and mouse). When this method was used to analyse study samples, one or two small shoulders were visible on the chromatographic peak of the analyte for all species. Chromatographic separation of the shoulder peaks using a long analytical column (Janssen et al.,
These two time points used IS-H.g.-12 as internal standard.
or the internal standard is enhanced) or over estimation of the analyte (if the analyte is enhanced or the internal standard suppressed). To determine whether any matrix effects are present, calibration lines were prepared, both in the presence of matrix (post-extracted plasma calibration line) and in the absence (spiked on the top of plasma extract calibration line and calibration line prepared in solvents). Comparison of the peak areas demonstrated that H.g.12 in the presence of human plasma matrix ions provided on average 82.5% the peak area of that observed in the absence of matrix ions, respectively. This implies that slight ion suppression is observed with H.g.-12 but the effect is insignificant with the use of an internal standard. Comparable observations were made in rat, mouse and rabbit plasma.
3.6. System carryover Assessment of blank extracted plasma injected directly after a high level QC sample demonstrated a peak area response <20% that of the lowest calibration standard implying no significant carryover occurs between samples for any species.
Table 5 Effect of dilution of H.g.-12 in human plasma at QCMED 10 concentration (600 ng/mL). Test
Analyte
Target conc. prepared (ng/mL)
Concentration measured (ng/mL)
Dilution 1 in 10
H.g.-12
600
531
544
559
598
589
582
Mean
sd
CV (%)
Accuracy relative to target (%)
567
27
4.7
94.6
S18
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
Fig. 2. The structure of deuterated H.g.-12.
Fig. 3. The structure of isotopically substituted H.g.-12.
Fig. 4. LC–MS/MS (MRM chromatograms) of an extracted human plasma (A) blank, (B) zero (internal standard only), (C) S1 (0.5 ng/mL) and (D) S8 (150 ng/mL). Top trace H.g.12 and bottom trace internal standard (D-H.g.-12).
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
9.3 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0
PHY192_MS06_SG_31Jul06_Accand precandselec_Batch5_RUS0603.rdb (PYM50057 IS 298): "Linear" Regression ("1 / (x * x)" weighting) : y = 0.0641 x + 0.00374 (r = 0.9991)
S19
Conflict of Interest The authors declare that there are no conflicts of interest. References
0
10
20
30
40
50
60
70 80 90 100 110 120 130 140 150 Analyte Conc. / IS Conc.
Fig. 5. Example calibration line of H.g.-12 (D-H.g.-12 internal standard) in human plasma (0.5–150 ng/mL – 1/x2 weighting).
2008) by the same MRM detection mode confirmed that they were H.g.-12 equivalents (i.e., epimers of H.g.-12, unpublished results). This study also indicated that although the ratio of the equivalents peaks changed over time, their identities were the same across species (human and mouse) and were present in the dose materials. Based on the above research results, therefore, we reported the analytical results in study samples as H.g.-12 equivalents, and use the equivalents concentration to assess the systemic exposure between species. In conclusion, a liquid chromatographic tandem mass spectrometric (LC–MS/MS) bioanalytical method has been developed and validated for H.g.-12 in human, rat, mouse and rabbit plasma in accordance with FDA guidelines. The validations were based on the single H.g.-12 chemical entity, and the method has been successfully applied to analysing study samples as H.g.-12 equivalents for supporting pre-clinical and clinical studies.
Abrahamse, S.L., Povey, K.J., Rees, D.D., 2007. Appetite suppressant compositions. US Patent Application 2007/0207227. Blom, W.A.M., Abrahamse, S.L., Bradford, R., Duchateau, G.S.M.J.E., Theis, W., Orsi, A., Ward, C.L., Mela, D.J., 2012. Effects of 15-d repeated consumption of Hoodia gordonii purified extract on safety, ad libitum energy intake, and body weight in healthy, overweight women: a randomized controlled trial. The American Journal of Clinical nutrition 94, 1171–1181. Dent, M.P., Wolterbeek, A.P.M., Russell P.J., Bradford, R. Safety profile of Hoodia gordonii extract: Rabbit prenatal developmental toxicity study (2012) Food and Chemical Toxicology 50 supplement 1 S26–S33 . Published online ahead of print doi:10.1016/j.fct.2011.04.008. Janssen, H.G., Swindells, C., Gunning, P., Wang, W., Grun, C., Mahabir, K., Maharaj, V.J., Apps, P.J., 2008. Quantification of appetite suppressing steroid glycosides from Hoodia gordonii in dried plant material, purified extracts and food products using HPLC–UV and HPLC–MS methods. Analytica Chimica Acta 617, 200–207. Knight, T.L., Swindells, C.M., Craddock, A.M., Maharaj, V.J., Buchwald-Werner, S., Ismaili S.A., McWilliam, S.C. (2012) Cultivation practices and manufacturing processes to produce Hoodia gordonii extract for weight management products. Food and Chemical Toxicology 50 supplement 1 S1–S5. Published online ahead of print doi:10.1016/j.fct.2011.09.042. Pawar, R.S., Shukla, Y.J., Khan, S.I., Avula, B., Khan, I.A., 2007. New oxypregnane glycosides from appetite suppressant herbal supplement Hoodia gordonii. Steroids 72 (6–7), 524–534. Russell P.J., Swindells, C.M. (2012) Chemical characterisation of Hoodia gordonii extract. Food and Chemical Toxicology 50 supplement 1 S6–S13. Published electronically doi:10.1016/j.fct.2011.02.020. Scott, A.D., Orsi, A., Ward, C.L., Bradford, R. (2012) Genotoxicity testing of a Hoodia gordonii extract Food and Chemical Toxicology 50 supplement 1 S34–S40. Published online ahead of print doi:10.1016/j.fct.2011.01.022. Tulip, O.L., Harbi, N.A., Mihalov, J., DerMadorosian, A., 2001. Effect of hoodia plant on food intake and body weight in lean and obese LA/Ntul//cp-rats. FASEB Journal 15 (4), A404. van Heerden, F.R., Vleggaar, R., Horak, R.M., Learmouth, R.A., Maharaj, V., Whittal, R.D., 1999. Pharmaceutical compositions having appetite-suppressant activity. Patent Application PCT/GB98/01100. van Heerden, F.R., Vleggaar, R., Morak, R.M., Learmouth, R.A., Maharaj, V., Whittal, R.D., 2002. US Patent 6376657 B1.
Contents lists available at ScienceDirect
Food and Chemical Toxicology journal homepage: www.elsevier.com/locate/foodchemtox
A validated bioanalytical method in mouse, rat, rabbit and human plasma for the quantification of one of the steroid glycosides found in Hoodia gordonii extract Weijun Wang a,⇑, Paul J. Russell b, Graeme T. Clark c, Derek Lewis c, Kun Nang Cheng c a
Phytopharm plc, Lakeview House, 2 Lakeview Court, Ermine Business Park, Huntingdon, Cambridgeshire PE29 6UA, UK Safety and Environmental Assurance Centre, Unilever, Colworth Science Park, Sharnbrook, Bedfordshire MK44 1LQ, UK c Quotient Bioresearch Ltd., Rushden, Northamptonshire NN10 6ER, UK b
a r t i c l e
i n f o
Article history: Available online 25 August 2010 Keywords: Bioanalysis Steroid glycoside Plasma LC–MS/MS Validation Hoodia gordonii extract
a b s t r a c t Hoodia gordonii extract contains steroid glycosides, fatty acids, plant sterols and polar organic material. Certain steroid glycosides show appetite suppressant activities following oral ingestion. This study describes the validation of a bioanalytical method for the quantification of one of the steroid glycosides, H.g.-12 (10% (w/w) of the extract), in mouse, rat, rabbit and human plasma. The method utilises a liquid–liquid extraction with methyl-tert-butyl ether followed by chromatographic separation on a 2.1 50 mm C18 Genesis high performance liquid chromatography (HPLC) column and detection on a triple quadrupole mass spectrometer. Detection of H.g.-12 and its stable isotope internal standards is performed using positive TurboIonspray™ ionisation in multiple reaction monitoring mode. The validation procedure demonstrated assay sensitivity, linearity, accuracy, precision and selectivity over the calibration range of 0.5–150 ng/mL in human plasma (500 lL sample volume), 1.0–100 ng/mL in rat and rabbit plasma (150 lL sample volume) and 1.0–250 ng/mL in mouse plasma (150 lL sample volume) with good recoveries (P77%). H.g.-12 was stable in plasma for P6 months at 20 °C, for up to 4 h at ambient temperature (ca 22 °C) and after 3 freeze–thaw cycles. Plasma extracts were stable for up to 24 h at ambient temperature. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction Hoodia gordonii extract is obtained from H. gordonii, a succulent plant originally found in the Kalahari Desert of Southern Africa (Knight et al., 2012). The potential of Hoodia extracts for weight management was first identified in several animal species by the National Food Research Institute of the CSIR in South Africa (van Heerden et al., 1999, 2002; Tulip et al., 2001) which conducted a nationwide survey to determine the nutritional value of food from the wild. The genus Hoodia was studied because certain species of this succulent plant were known to be consumed by the San people as a substitute for food and water. A CSIR scientist who studied the biological effects of Hoodia extracts on small laboratory animals, observed that the animals lost their appetite, accompanied by a loss of weight. In order to determine the systemic exposure, folAbbreviations: HPLC, high performance liquid chromatography; FDA, Food and Drug Administration; LLE, liquid–liquid extraction; LC–MS/MS, liquid chromatography tandem mass spectrometry; MRM, multiple reaction monitoring; MTBE, methyl-tert-butyl ether; QC, quality control; CAD, collision activated dissociation; RE, relative error; CV, co-efficient of variance; ANOVA, analysis of variance; LLOQ, lower limit of quantification. ⇑ Corresponding author. Tel.: +44 (0) 1480 437 697; fax: +44 (0) 1480 417 090. E-mail address: [email protected] (W. Wang). 0278-6915/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.fct.2010.08.014
lowing administration of H. gordonii extracts within pre-clinical and clinical studies (Dent et al., 2012; Scott et al., 2012; Blom et al., 2011), an appropriate bioanalytical method required development and validation to the standard criteria outlined in the regulatory guidance document (Food and Drug Administration (FDA) document1). Previous investigations (van Heerden et al., 1999) determined that one major component of Hoodia extract (10% (w/w) of H. gordonii extract) responsible for eliciting an efficacy response is the steroid glycoside H.g.-12 (Fig. 1). This component was isolated and purified from H. gordonii for use in the method development and formal validation. In conjunction, deuterated H.g.-12 (produced via high pressure deuterium exchange) was used as an internal standard for all species validation. This was subsequently replaced by an isotopically substituted H.g.-12 (which was chemically synthesised from isolated H.g.-12) via partial validation for all reported species except rabbit. Liquid–liquid extraction (LLE) is a useful procedure to isolate particularly lipophilic compounds from biofluids. The extraction properties of the organic solvent tend not to differentiate between 1 Guidance for Industry: Bioanalytical Method Validation. May 2001, US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER), Center for Veterinary Medicine (CVM).
S15
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
O O
O
OH HO MeO
O
O
O OH
O
O OMe
O
OMe
Fig. 1. Chemical structure of steroid glycoside reference standard H.g.-12.
subtle structural differences and as Hoodia extracts contain a number of structurally related moieties, LLE is deemed appropriate for sample preparation. The hyphenated technique of liquid chromatography tandem mass spectrometry (LC–MS/MS) has become the industry standard for quantitative measurements of both parent compound and metabolically related material present in biological matrices. When combined with reversed phase chromatography (utilising the multiple reaction monitoring (MRM) capability of triple quadrupole mass spectrometers), highly specific measurements of diagnostic precursor and product ions can produce LC–MS/MS methods capable of sensitive and selective quantification in a relatively short chromatographic cycle time (when compared with other chromatography based detection techniques). A method for the separation and analysis of extracted plasma samples was developed on C18 based reverse phase high performance liquid chromatography (HPLC) column with MRM detection on a triple quadrupole mass spectrometer. In this paper, we present the validated methodology for quantification of H.g.-12 in human, rat, mouse and rabbit plasma samples. The validations were based on the single H.g.-12 chemical entity, and then the method was applied to the measurement of co-occurring epimers (referred to as H.g.-12 equivalents) in preclinical and clinical study samples. This article is one of six published as a supplement providing information on various aspects of H. gordonii. 2. Materials and methods 2.1. Chemicals and reagents Acetonitrile (HPLC grade), ammonium acetate (HPLC grade), formic acid (specified reagent 98/100%) and methyl-tert-butyl ether (MTBE – HPLC grade) were all purchased from Fisher-Scientific (Loughborough, UK). HPLC grade water was obtained from an in-house Milli-Q purification system. Blank human plasma (stabilised with lithium heparin) was obtained from Richard Pharmacology, Atkinson Morley’s Hospital (London, UK), whilst blank rat, mouse and rabbit plasma was obtained from B&K Universal Limited (Grimston, UK). All blank plasma was stored at ca 20 °C when not in use. H.g.-12 (P92.4%) reference standard and H.g.-12 internal standards (deuterated H.g.-12 and isotopically substituted H.g.-12) were all supplied by Phytopharm plc. 2.2. Preparation of stock solutions, spiking solutions, quality control (QC) samples and spiking internal standard solution Stock solutions of the standard (H.g.-12), deuterated H.g.-12 (D-H.g.-12) and isotopically substituted H.g.-12 (IS-H.g.-12) were prepared in acetonitrile:water (1:1, v/v) at a final concentration of ca 100 lg/mL. These stock solutions were all stored at ca 20 °C when not in use. Calibration standards 0.5–150 ng/mL for H.g.-12 in human plasma (500 lL sample volume), 1.0–100 ng/mL in rat and rabbit plasma (150 lL sample volume) and 1.0–250 ng/mL in mouse plasma (150 lL sample volume) were prepared daily. Quality control samples were prepared from independent stock solutions (separate weighing) at four nominal concentrations of 0.5, 1.5, 60 and 120 ng/mL for H.g.-12 (human plasma); 1.0, 3.0, 125 and 200 ng/mL (mouse plasma); 1.0, 3.0, 50 and 80 ng/mL (rat and rabbit plasma). Quality control samples were prepared in bulk at the start of the validation study and stored at ca 20 °C when not in use.
D-H.g.-12 was prepared at either 500 or 1000 ng/mL for human plasma analysis (depending on batch of material) and 250 ng/mL for rat, mouse and rabbit plasma by appropriate dilution of the stock solution with acetonitrile:water (1:1, v/v). ISH.g.-12 spiking solution for human, rat and mouse plasma was prepared at 100 ng/mL by appropriate dilution with acetonitrile:water (1:1, v/v). 2.3. Sample preparation (extraction) The QC samples and spiking solutions were allowed to reach room temperature, vortex mixed and the following procedures followed: Blank plasma (500 lL for human, 150 lL for all other species) for blank, zero and calibration standards were aliquotted into glass tubes and each calibration standard was spiked with its respective spiking solutions at appropriate volumes (e.g., 50 lL). Each bulk prepared QC (500 lL for human, 150 lL for all other species) was aliquotted into glass tubes (n = 6 each per level) and each was spiked with appropriate volumes (e.g., 50 lL) of acetonitrile:water (1:1, v/v). Each zero and blank sample was spiked with appropriate volumes (e.g., 50 lL) of acetonitrile:water (1:1, v/v). All samples (except blanks) were spiked with 50 lL D-H.g.12 internal standard (except human (125 lL)) or 20 lL IS-H.g.-12 (except human (50 lL) and rabbit (not done)). Blanks were spiked with an equivalent volume of acetonitrile:water (1:1, v/v). The samples were then diluted with water (400 lL) and briefly vortex mixed (ca 2–5 s). MTBE (5 mL) was added to each sample prior to mixing by rotation (ca 15 min), centrifuged (3000 rpm, 10 min) and then stored at ca 80 °C for P30 min at which point the MTBE layer was then carefully removed, taking care not to disturb or sample any of the precipitated material immediately above the frozen lower layer. The organic layer was concentrated to dryness and reconstituted in acetonitrile:water (1:1, v/v, 120 lL). 2.4. HPLC conditions (separation) Chromatographic retention and elution of H.g.-12 and internal standard was performed on a Genesis C18, 2.1 50 mm, 120 Å, 4 lm HPLC column (Kinesis, Bedfordshire, UK). Elution was performed with an isocratic mobile phase of acetonitrile (10 mM ammonium acetate + 0.05% (v/v) formic acid), 80:20 (v/v) at a flow rate of 200 lL/min. A Security Guard C18 2.0 4 mm pre-column (Phenomenex, Cheshire, UK) was used during the course of the validation, which was replaced prior to each analytical batch. 2.5. MS conditions (detection) Detection of H.g.-12 and internal standard was obtained on an API 3000 mass spectrometer (Applied Biosystems-MDS Sciex., Warrington, UK) operating under positive TurboIonspray™. All analytes were observed as the [M+NH4]+ ammoniated precursor ions which provided diagnostic product ions under MRM conditions with nitrogen as both the nebuliser and collision activated dissociation (CAD) gas. Full MS parameters can be found in Table 1.
3. Results 3.1. Chromatography and linearity Typical LC–MS/MS chromatograms (MRM) for a human plasma blank, extracted zero (internal standard only), extracted S1 (lower limit of quantification – 0.5 ng/mL) and extracted S8 (upper limit of quantification – 150 ng/mL) are presented in Fig. 4. These demonstrate that there is no endogenous material or analytical interference in either the H.g.-12 or internal standard ion channels. Comparable chromatography was observed for all other species. A linear calibration response was observed across the calibration range of 0.5–150 ng/mL for H.g.-12 in human (500 lL sample
S16
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
Table 1 API 3000 mass spectrometric parameters for the detection of H.g.-12, and D-H.g.-12 or IS-H.g.-12. Analyte
NEB
CUR
IS (V)
TEM (°C)
CAD
Mass transition
CE
H.g.-12 D-H.g.-12 IS-H.g.-12
7–8
6–9
4000–4500
200–300
4
m/z 896 ? 295 m/z 907 ? 298 m/z 906 ? 295
33 39 33
NEB, Nebuliser gas; CUR, curtain gas; IS, ionspray voltage; TEM, temperature; CAD, collision activated dissociation; CE, collision energy.
volume), 1.0–250 ng/mL in mouse (150 lL sample volume) and 1.0–100 ng/mL in rat and rabbit plasma (150 lL sample volume). Linear regression with 1/x2 weighting was applied. Correlation co-efficients for H.g.-12 during the course of the human validation with D-H.g.-12 as the internal standard were >0.99. Back-calculated concentrations for the calibration standards provided relative errors of the mean of 6 7.2% and precision (CV) of 69.9%. Example calibration lines for H.g.-12 from this validation can be seen in Fig. 5. Comparable statistics were observed for rat, mouse and rabbit validations with D-H.g.-12 and for all three species (human, rat and mouse) which were partially validated with IS-H.g.-12.
three species (Table 3). The accuracy (RE) and precision (CV) values of H.g.-12 were 6 4.7% and 612.7% in human plasma, 618.3% and 66.5% in rat plasma and 6 10.4% and 65.1% in mouse plasma. 3.3. Recovery Recovery of H.g.-12 and the internal standard (D-H.g.-12) were good (P77%) and reproducible from all species investigated (human, rat and mouse). There is a balance between higher recoveries and the avoidance of possible interference by sampling the precipitated material in the MTBE layer after freezing.
3.2. Accuracy and precision
3.4. Selectivity
3.2.1. Human, rat and mouse plasma with D-H.g.-12 internal standard Accuracy (RE) and precision (CV) were determined on three separate occasions, at the four QC levels in replicate (n = 6 per level) in human, rat and mouse plasma. Inter-batch analysis was performed using the analysis of variance (ANOVA) method with the full details being presented in Table 2. Results for H.g.-12 in these three species were well within the acceptance criteria of ±15% (±20% at lower limit of quantification – LLOQ) for both accuracy (in terms of relative error (RE)) and precision (as measured by co-efficient of variation (CV)). The accuracy (RE) and precision (CV) values of H.g.-12 were 6 13.1% and 612.5% for human plasma, 66.1% and 68.3% for rat plasma, and 6 7.6% and 613.8% for mouse plasma.
Analysis of different plasma batches (n = 6 per species) demonstrated no interference for the analysis of H.g.-12 at its respective retention time. The addition of either internal standard (referred to as zero samples) did not introduce any additional chromatographic peaks in either ion channel. Quantification of the six different plasma batches from each species at an appropriate level within the calibration range (S6) demonstrated accuracy and precision within ±15% for H.g.-12 in human, rat and rabbit plasma when using D-H.g.-12 as the internal standard. Quantification of H.g.-12 in mouse plasma with D-H.g.12 provided results outside the recommended ±15% highlighting a selectivity issue unique to this species. However, quantitative selectivity re-assessment with IS-H.g.-12 produced accuracies well within ±15% for all species investigated (human, rat and mouse), implying that the issues observed with selective quantification in mouse plasma (using D-H.g.-12) were due to a specific ion suppression effect on the structural analogue, which varied between individuals.
3.2.2. Rabbit plasma with D-H.g.-12 internal standard Accuracy (RE) and precision (CV) were determined on a single occasion, at the four QC levels in replicate (n = 6 per level) with the full details being presented in Table 2. Results for H.g.-12 were well within the acceptance criteria of ±15% for both accuracy and precision (±20% at LLOQ), and the accuracy (RE) and precision (CV) values of H.g.-12 were 66.9% and 612.1%. 3.2.3. Human, rat and mouse plasma with IS-H.g.-12 internal standard Accuracy (RE) and precision (CV) were determined on a single occasion, at the four QC levels in replicate (n = 6 per level) for each species. Results for H.g.-12 were well within the acceptance criteria of ±15% for both accuracy and precision (±20% at LLOQ) for all
3.5. Matrix effects (ion suppression and enhancement) A consequence of LC–MS/MS bioanalysis is the potential for coextracted material to co-elute under the relatively short chromatographic cycles times and interfere with the processes controlling the ionisation of analyte and internal standards. These matrix effects can either suppress or enhance the ion signal resulting in an under estimation of the analyte (if the analyte is suppressed
Table 2 Accuracy (relative error = RE) and precision (co-efficient of variance = CV) for H.g.-12 in human, rat, mouse and rabbit plasma QC samples from three batches with D-H.g.-12 as internal standard. H.g.-12 Human plasma QCLLOQ RE range (CV) intra-batch QCLOW RE range (CV) intra-batch QCMED RE range (CV) intra-batch QCHIGH RE range (CV) intra-batch QCLLOQ RE range (CV) inter-batch QCLOW RE range (CV) inter-batch QCMED RE range (CV) inter-batch QCHIGH RE range (CV) inter-batch
4.5% to +7.3% (6.1%) 5.5% to +2.4% (12.5%) 13.1% to +3.1% (9.6%) 11.8% to 2.5% (6.4%) 1.0% (7.0%) 2.7% (3.0%) 3.1% (8.6%) 5.6% (5.4%)
Rat plasma 5.2% 3.7% 2.0% 6.0% 0.1% 2.0% 1.7% 2.7%
to +6.1% (6.7%) to 0.3% (6.7%) to 1.2% (6.5%) to 0.0% (8.3%) (5.1%) (NC) (NC) (2.3%)
Mouse plasma
Rabbit plasma
7.6% to +5.2% (13.8%) 6.4% to +2.3% (5.7%) 1.3% to +2.4% (2.6%) 3.0% to 1.7% (3.5%) 0.8% (5.0%) 0.9% (4.5%) 0.2% (1.7%) 2.6% (NC)
+6.9% (12.1%) 1.1% (12.1%) 1.7% (6.8%) 2.3% (8.0%) NA NA NA NA
NA, Not applicable, only single batch analysed; NC, not calculable by ANOVA method due to the intra-batch variability being greater than the inter-batch variability.
S17
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
3.7. Stability
Table 3 Accuracy (relative error = RE) and precision (co-efficient of variance = CV) for H.g.-12 in human, rat and mouse plasma QC samples from single batch with IS-H.g.-12 as internal standard.
Assessments of stability were performed (24 h extract, 3 freeze– thaw, 4 h at room temperature, storage at ca 20 °C for 1 week and 1 month) on the low and high level QC samples for each species. No significant deterioration of H.g.-12 was observed in any of the assessments with relative error within ±15% of the nominal concentrations being obtained in all cases. Long term storage stability studies (1, 3, 6, 12, 16 and 24 months for human and rat plasma; 1, 3, 6, 9, 12, 18 and 24 for mouse plasma; 1, 3 and 6 months for rabbit plasma) were conducted, and the results indicated that H.g.-12 in human and rat plasma QC samples were stable for 24 months, mouse plasma for 12 months (after 12 months, there was not enough QCHIGH stability sample left for analysis) and rabbit plasma for 6 months (the last time point) when stored at ca 20 °C. Details of human plasma storage stability can be seen in Table 4.
H.g.-12
QCLLOQ RE and (CV) intra-batch QCLOW RE and (CV) intra-batch QCMED RE and (CV) intra-batch QCHIGH RE and (CV) intra-batch
Human plasma
Rat plasma
Mouse plasma
0.7% (12.7%)
18.3% (6.5%)
1.5% (5.1%)
0.1% (2.2%)
0.6% (4.0%)
10.4% (1.7%)
5.1% (1.8%)
4.1% (1.2%)
9.4% (2.9%)
8.7% (4.8%)
1.4% (2.1%) 4.7% (5.8%)
Table 4 Stability of H.g.-12 in human plasma QCLOW and QCHIGH samples expressed at mean accuracy relative to target (%) with D-H.g.-12 or IS-H.g.-12 as internal standard. QC samples storage conditions
The effect of dilution was assessed by preparing a solution of H.g.-12 in human, rat and rabbit plasma at 10 QC medium (QCMED 10), and in mouse plasma at 10 and 50 QC medium (QCMED 10 and 50). The dilution results demonstrated excellent accuracy and precision of the procedure for the analyte in all species. The human plasma dilution results are presented in Table 5.
Mean accuracy relative to target (%) for QCLOW and QCHIGH samples
4 hr in plasma at RT (ca 22 °C) 24 hr extract stability (ca 22 °C) Three freeze thaw cycles (ca 20 °C) Storage (ca 20 °C): 1 week Storage (ca 20 °C): 1 month Storage (ca 20 °C): 3 months Storage (ca 20 °C): 6 months Storage (ca 20 °C): 12 months Storage (ca 20 °C): 16 months Storage (ca 20 °C): 16 months* Storage (ca 20 °C): 24 months* *
3.8. Dilution
H.g.-12 QCLOW
H.g.-12 QCHIGH
96.9 104.9 88.4 89.7 112.3 97.2 101.1 87.7 91.8 100.7 109.8
90.8 90.4 89.9 91.6 109.9 90.9 95.0 82.6 90.6 101.6 97.0
4. Discussion H. gordonii extracts contain a number of chemical entities (>20) which are chemical analogues or epimers of H.g.-12 (Pawar et al., 2007; Abrahamse et al., 2007; Janssen et al., 2008; Russell and Swindells, 2012). Following oral administration of H. gordonii extracts in animals or in humans, the metabolic process may produce more structural analogues. Therefore, it was impractical to develop and validate a bioanalytical method to separate all these isomers with acceptable criteria and reasonable chromatographic cycle time to support pre-clinical and clinical studies. The initial development and validations were based on the single H.g.-12 epimer using deuterated H.g.-12 (D-H.g.-12) as an internal standard. The D-H.g.-12 (Fig. 2), produced by catalytic reduction of H.g.-12 with deuterium, consists of several deuterated species due to the catalytic reduction of one or both double bonds in H.g.-12 along with non-selective deuterium incorporation. By using this compound as the internal standard, all species passed the standard validation criteria outlined in the guidance (FDA document1) with the exception that the mouse plasma selectivity assay results were outside the recommended ±15% (i.e., 6±20.3%), indicating a specific ion suppression effect on this species. Following further research and development, a new batch of internal standard was synthesised by introducing 13C3 and 2H7 into H.g.-12 (Fig. 3). Quantitative selectivity re-assessment with this new isotopically substituted H.g.-12 internal standard (IS-H.g.-12) produced accuracies well within ±15% for all species (human, rat and mouse). When this method was used to analyse study samples, one or two small shoulders were visible on the chromatographic peak of the analyte for all species. Chromatographic separation of the shoulder peaks using a long analytical column (Janssen et al.,
These two time points used IS-H.g.-12 as internal standard.
or the internal standard is enhanced) or over estimation of the analyte (if the analyte is enhanced or the internal standard suppressed). To determine whether any matrix effects are present, calibration lines were prepared, both in the presence of matrix (post-extracted plasma calibration line) and in the absence (spiked on the top of plasma extract calibration line and calibration line prepared in solvents). Comparison of the peak areas demonstrated that H.g.12 in the presence of human plasma matrix ions provided on average 82.5% the peak area of that observed in the absence of matrix ions, respectively. This implies that slight ion suppression is observed with H.g.-12 but the effect is insignificant with the use of an internal standard. Comparable observations were made in rat, mouse and rabbit plasma.
3.6. System carryover Assessment of blank extracted plasma injected directly after a high level QC sample demonstrated a peak area response <20% that of the lowest calibration standard implying no significant carryover occurs between samples for any species.
Table 5 Effect of dilution of H.g.-12 in human plasma at QCMED 10 concentration (600 ng/mL). Test
Analyte
Target conc. prepared (ng/mL)
Concentration measured (ng/mL)
Dilution 1 in 10
H.g.-12
600
531
544
559
598
589
582
Mean
sd
CV (%)
Accuracy relative to target (%)
567
27
4.7
94.6
S18
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
Fig. 2. The structure of deuterated H.g.-12.
Fig. 3. The structure of isotopically substituted H.g.-12.
Fig. 4. LC–MS/MS (MRM chromatograms) of an extracted human plasma (A) blank, (B) zero (internal standard only), (C) S1 (0.5 ng/mL) and (D) S8 (150 ng/mL). Top trace H.g.12 and bottom trace internal standard (D-H.g.-12).
W. Wang et al. / Food and Chemical Toxicology 50 (2012) S14–S19
9.3 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0
PHY192_MS06_SG_31Jul06_Accand precandselec_Batch5_RUS0603.rdb (PYM50057 IS 298): "Linear" Regression ("1 / (x * x)" weighting) : y = 0.0641 x + 0.00374 (r = 0.9991)
S19
Conflict of Interest The authors declare that there are no conflicts of interest. References
0
10
20
30
40
50
60
70 80 90 100 110 120 130 140 150 Analyte Conc. / IS Conc.
Fig. 5. Example calibration line of H.g.-12 (D-H.g.-12 internal standard) in human plasma (0.5–150 ng/mL – 1/x2 weighting).
2008) by the same MRM detection mode confirmed that they were H.g.-12 equivalents (i.e., epimers of H.g.-12, unpublished results). This study also indicated that although the ratio of the equivalents peaks changed over time, their identities were the same across species (human and mouse) and were present in the dose materials. Based on the above research results, therefore, we reported the analytical results in study samples as H.g.-12 equivalents, and use the equivalents concentration to assess the systemic exposure between species. In conclusion, a liquid chromatographic tandem mass spectrometric (LC–MS/MS) bioanalytical method has been developed and validated for H.g.-12 in human, rat, mouse and rabbit plasma in accordance with FDA guidelines. The validations were based on the single H.g.-12 chemical entity, and the method has been successfully applied to analysing study samples as H.g.-12 equivalents for supporting pre-clinical and clinical studies.
Abrahamse, S.L., Povey, K.J., Rees, D.D., 2007. Appetite suppressant compositions. US Patent Application 2007/0207227. Blom, W.A.M., Abrahamse, S.L., Bradford, R., Duchateau, G.S.M.J.E., Theis, W., Orsi, A., Ward, C.L., Mela, D.J., 2012. Effects of 15-d repeated consumption of Hoodia gordonii purified extract on safety, ad libitum energy intake, and body weight in healthy, overweight women: a randomized controlled trial. The American Journal of Clinical nutrition 94, 1171–1181. Dent, M.P., Wolterbeek, A.P.M., Russell P.J., Bradford, R. Safety profile of Hoodia gordonii extract: Rabbit prenatal developmental toxicity study (2012) Food and Chemical Toxicology 50 supplement 1 S26–S33 . Published online ahead of print doi:10.1016/j.fct.2011.04.008. Janssen, H.G., Swindells, C., Gunning, P., Wang, W., Grun, C., Mahabir, K., Maharaj, V.J., Apps, P.J., 2008. Quantification of appetite suppressing steroid glycosides from Hoodia gordonii in dried plant material, purified extracts and food products using HPLC–UV and HPLC–MS methods. Analytica Chimica Acta 617, 200–207. Knight, T.L., Swindells, C.M., Craddock, A.M., Maharaj, V.J., Buchwald-Werner, S., Ismaili S.A., McWilliam, S.C. (2012) Cultivation practices and manufacturing processes to produce Hoodia gordonii extract for weight management products. Food and Chemical Toxicology 50 supplement 1 S1–S5. Published online ahead of print doi:10.1016/j.fct.2011.09.042. Pawar, R.S., Shukla, Y.J., Khan, S.I., Avula, B., Khan, I.A., 2007. New oxypregnane glycosides from appetite suppressant herbal supplement Hoodia gordonii. Steroids 72 (6–7), 524–534. Russell P.J., Swindells, C.M. (2012) Chemical characterisation of Hoodia gordonii extract. Food and Chemical Toxicology 50 supplement 1 S6–S13. Published electronically doi:10.1016/j.fct.2011.02.020. Scott, A.D., Orsi, A., Ward, C.L., Bradford, R. (2012) Genotoxicity testing of a Hoodia gordonii extract Food and Chemical Toxicology 50 supplement 1 S34–S40. Published online ahead of print doi:10.1016/j.fct.2011.01.022. Tulip, O.L., Harbi, N.A., Mihalov, J., DerMadorosian, A., 2001. Effect of hoodia plant on food intake and body weight in lean and obese LA/Ntul//cp-rats. FASEB Journal 15 (4), A404. van Heerden, F.R., Vleggaar, R., Horak, R.M., Learmouth, R.A., Maharaj, V., Whittal, R.D., 1999. Pharmaceutical compositions having appetite-suppressant activity. Patent Application PCT/GB98/01100. van Heerden, F.R., Vleggaar, R., Morak, R.M., Learmouth, R.A., Maharaj, V., Whittal, R.D., 2002. US Patent 6376657 B1.