Determination of pyrrolizidine alkaloids (PA) in plant material by SPE-LC-MS/MS Method Protocol BfR-PA-Tea-2.0/2014 Determination of PA in plant material by SPE-LC-MS/MS CONTENT 1 Scope ............................................................................................................................ 1 2 Principle ........................................................................................................................ 1 3 Reagents ....................................................................................................................... 1 3.1 General ................................................................................................................. 1 3.2 Chemicals ............................................................................................................. 2 3.3 Solutions ............................................................................................................... 3 3.3.1 Extraction solution ............................................................................................. 3 3.3.2 Aqueous ammoniacal solution for neutralisation ................................................ 3 3.3.3 2.5 % ammonia in methanol for SPE elution (black tea and green tea) ............. 3 3.3.4 HPLC mobile phase .......................................................................................... 3 3.3.5 Standard solution for calibration ........................................................................ 3 4 Apparatus ...................................................................................................................... 4 5 Procedure ...................................................................................................................... 6 5.1 Sample preparation (grinding of plant material) ..................................................... 6 5.2 Extraction .............................................................................................................. 6 5.3 SPE-procedure ..................................................................................................... 7 5.4 Reconstitution of the sample ................................................................................. 7 6 HPLC-MS/MS analysis .................................................................................................. 7 6.1 Liquid chromatographic separation ....................................................................... 7 6.2 Mass spectrometric operation conditions .............................................................. 7 6.3 Measurement ........................................................................................................ 8 7 Calculation .................................................................................................................... 8 7.1 Calibration function ............................................................................................... 8 7.2 Quantification ........................................................................................................ 8 7.3 Reporting of results ............................................................................................... 9 8 Annex ...........................................................................................................................10 8.1 LC-MS/MS measurement .....................................................................................10 8.2 Typical Chromatogram .........................................................................................13 8.3 Provider of PA-Standards.....................................................................................14 8.4 Flow chart of the sample preparation procedure ..................................................17 Determination of PA in plant material by SPE-LC-MS/MS 1 Scope Pyrrolizidine alkaloids (PA) are secondary plant metabolites with carcinogenic and genotoxic properties. Currently, more than 600 PA are known. They occur in plants of families of Bo- raginaceae, Asteraceae and Fabaceae. The worldwide spread of these plants may lead to a contamination of herbal foodstuff, herbal medicines and animal feed (EFSA Panel on Contaminants in the Food Chain (CONTAM) 2011). This method describes the determination of the following PA in plant material: echimidine (Em), echimidine-N-oxide (EmN), erucifoline (Er), erucifoline-N-oxide (ErN), europine (Eu), europine-N-oxide (EuN), heliotrine (Hn), heliotrine-N-oxide (HnN), intermedine (Im), inter- medine-N-oxide (ImN), jacobine (Jb), jacobine-N-oxide (JbN), lasiocarpine (Lc), lasiocar- pine-N-oxide (LcN), lycopsamine (La), lycopsamine-N-oxide (LaN), monocrotaline (Mc), monocrotaline-N-oxide (McN), retrorsine (Re), retrorsine-N-oxide (ReN), senecionine (Sc), senecionine-N-oxide (ScN), seneciphylline (Sp), seneciphylline-N-oxide (SpN), sene- civernine (Sv), senecivernine-N-oxide (SvN), senkirkine (Sk), trichodesmine (Td). The limits of determination and quantification for toxins determined during in-house valida- tion are listed in the Annex 8.1. 2 Principle A test portion of plant material is sonicated twofold in aqueous sulphuric acid solution for PA extraction. After centrifugation an aliquot of the supernatant is purified by solid-phase extrac- tion (SPE) using reversed phase C18 material. PAs are released from the cartridge using methanol. Subsequently, the eluate is evaporated to dryness and reconstituted in metha- nol/water (initial HPLC conditions). For chromatographic separation, an RP-HPLC column is used with a binary gradient. Ana- lytes are detected by triple stage quadrupole mass spectrometry. Quantification of pyrroliz- idine alkaloids is accomplished by means of matrix matched calibration. 3 Reagents 3.1 General Please note: Since the use of this method involves reagents harmful to health, appropriate precautionary and protective measures such as avoiding skin contact and us- ing an extractor hood must be taken. If not specified elsewise, reagents of analytical grade and solvents suitable for HPLC-MS/MS must be used. Water must be distilled in glass vessels or de- mineralised before use, or must be of equivalent purity. 1 of 17 Determination of PA in plant material by SPE-LC-MS/MS 3.2 Chemicals 3.2.1 echimidine (Em) 3.2.2 echimidine-N-oxide (EmN) 3.2.3 erucifoline (Er) 3.2.4 erucifoline-N-oxide (ErN) 3.2.5 europine (Eu) 3.2.6 europine-N-oxide (EuN) 3.2.7 heliotrine (Hn) 3.2.8 heliotrine-N-oxide (HnN) 3.2.9 intermedine (Im) 3.2.10 intermedine-N-oxide (ImN) 3.2.11 jacobine (Jb) 3.2.12 jacobine-N-oxide (JbN) 3.2.13 lasiocarpine (Lc) 3.2.14 lasiocarpine-N-oxide (LcN) 3.2.15 lycopsamine (La) 3.2.16 lycopsamine-N-oxide (LaN) 3.2.17 monocrotaline (Mc) 3.2.18 monocrotaline-N-oxide (McN) 3.2.19 retrorsine (Re) 3.2.20 retrorsine-N-oxide (ReN) 3.2.21 senecionine (Sc) 3.2.22 senecionine-N-oxide (ScN) 3.2.23 seneciphylline (Sp) 3.2.24 seneciphylline-N-oxide (SpN) 3.2.25 senecivernine (Sv) 3.2.26 senecivernine-N-oxide (SvN) 3.2.27 senkirkine (Sk) 3.2.28 trichodesmine (Td) 3.2.29 formic acid 98 – 100%, e.g. Sigma-Aldrich 3.2.30 methanol (MeOH) in LC-MS quality, e.g. Merck LiChrosolv® 3.2.31 sulphuric acid 98%, e.g. Merck 2 of 17 Determination of PA in plant material by SPE-LC-MS/MS 3.2.32 ammonia 32%, e.g. Merck 3.2.33 ammonium formate in LC-MS quality, e.g. Fluka 3.2.34 acetonitrile, e.g. Merck LiChrosolv® 3.3 Solutions 3.3.1 Extraction solution 2.665 mL of sulphuric acid (H SO ) (3.2.31) are filled up to 1000 mL with water. The 2 4 final concentration is 0.05 M. 3.3.2 Aqueous ammoniacal solution for neutralisation To prepare the ammonical solution for neutralisation of sample extracts before SPE, 5 mL of ammonia (3.2.32) are filled up to 25 mL with water. 3.3.3 2.5 % ammonia in methanol for SPE elution (SPE elution for black tea and green tea) To prepare the 2.5 % ammonia solution in methanol 7.8 mL of ammonia (3.2.32) are filled up to 100 mL with methanol (3.2.30). The solution has to be freshly prepared per working day. 3.3.4 HPLC mobile phase Eluent A: 315 mg ammonium formate (3.2.33) are dissolved in 5 mL of water, 1 mL of formic acid (3.2.29) is added and filled up to 1000 mL with water. Eluent B: 315 mg ammonium formate (3.2.33) are dissolved in 5 mL of water, 1 mL of formic acid (3.2.29) is added and filled up to 1000 mL with methanol (3.2.30). 3.3.5 Standard solution for calibration Stock solution (0.1 mg/mL): To create a stock solution, 1 mg of a pyrrolizidine alkaloid standards are weighed us- ing analytical balance (4.4) and filled up with acetonitrile (3.2.34) in a volumetric flask to make 10 mL. The concentration of the stock solution is 0.1 mg/mL. Standard working solution (PA mixture, 1 µg/mL) For preparation of the standard working solution, respective volumes of each PA stock solution (0.1 mg/mL) are pipetted into a volumetric flask and is filled up with acetonitrile (3.2.34), to obtain a concentration of 1 µg/mL. Preparation of matrix matched standards (MMS) For a correction of matrix effects a matrix matched calibration is used. In order to ob- tain the same matrix effect for MMS and for the samples the blank plant material has to be processed as described in section 5. Afterwards, MMS levels have to be pre- pared according to table 1. 3 of 17 Determination of PA in plant material by SPE-LC-MS/MS Please note: In order to prepare of a sufficient amount of blank-extract, two times 10 mL of neutralised blank-extract (5.2) from one blank sample can be purified using two SPE-cartridges. Thereby, 2 mL of reconstituted blank-extract for the MMS preparation can be received without extrac- tion of an additional blank sample. table 1: Matrix matched standards Final PA mass Final PA Aliquot taken Aliquot Aliquot taken concentration in mass from Volume from blank calibration concentration plant material solution extract ng/mL µg/kg µL μL MMS_1 5,0 10,0 MMS_5 20 280 MMS_2 10,0 20,0 MMS_8 20 280 MMS_3 25,0 50,0 MMS_8 20 100 MMS_4 50,0 100,0 PA-Mix 10 100 MMS_5 75,0 150,0 PA-Mix 15 185 MMS_6 100.0 200,0 PA-Mix 20 180 MMS_7 125,0 250,0 PA-Mix 25 175 MMS_8 150,0 300,0 PA-Mix 60 340 4 Apparatus 4.1 General Usual laboratory glassware and equipment should be used and, in particular, the following: 4 of 17 Determination of PA in plant material by SPE-LC-MS/MS 4.2 centrifugal mill with 0.5 mm sieve, e.g. Retsch 4.3 various piston pipettes and multiple dispensers, e.g. Brand 4.4 analytical balance, capable of weighing to 0,0001 g 4.5 centrifuge for 50 mL centrifuge tubes, capable of at least 5 000 x g 4.6 ultrasonic bath 4.7 overhead shaker, e.g. Heidolph 4.8 laboratory shaker, e.g. Vortex 4.9 evaporation station, e.g. TurboVap 4.10 centrifuge tube 50 mL 4.11 test tubes 15 mL 4.12 volumetric flasks, 10 and 20 mL 4.13 folded filters, e.g. Munktell 4.14 SPE cartridges: DSC-C18 SPE (Supelco), 500 mg sorbent material, 6mL 4.15 SPE vacuum chamber 4.16 Membrane filter 0.2 µm, e.g. VWR 0,5 mL centrifugal filters, modified nylon membrane 4.17 HPLC vials 2 mL 4.18 Glass inserts, 250 µL conic for HPLC vials 4.19 chromatographic column, e.g. Thermo, Hypersil Gold®; 150 x 2.1 mm; 1,9 µm 4.20 LC-MS/MS system 5 of 17 Determination of PA in plant material by SPE-LC-MS/MS 5 Procedure 5.1 Sample preparation (grinding of plant material) To determine the PA-content which is representative for the entire sample, the plant material should meet the following characteristics: uniform particle size and a homogenous distribu- tion of PA or PA-containing material, respectively. Therefore, the entire sample material is mixed with dry ice (ratio 2:1), ground to a particle size of 0.5 mm (4.2) and homogenized for example by shaking over head (4.7). As an alternative to grinding with dry ice (excellent grinding results due to shear forces and porosity of the frozen sample material), the sample material may also be ground to a particle size of 0.25 mm if there is no considerably genera- tion of heat. If the test material can neither be ground with dry ice nor ground to a particle size of 0.25 mm, it is also possible to increase the weighed sample amount in contrast to 5.2 to at least 10 g (particle size 0.5 mm). In order to keep a constant ratio of sample amount to ex- traction volume, the used volume of extraction solution (3.3.1) needs to be increased by the same factor as the weighed sample amount. 5.2 Extraction For the extraction of PA 2.0 g ± 0.1 g of plant are weighed into a centrifuge tube (4.10). Extraction step 1 For the first extraction step 20 mL of the extraction solution (3.3.1) are added to the sample. The sample material has to be wetted completely before extraction in an ultrasonic bath (4.6) for 15 min at ambient tem- perature. Centrifugation The sample is centrifuged for 10 min ± 2 min at 3800 x g (4.5). The supernatant (extract 1) is transferred into a clean test tube. The sedi- ment is used for the second extraction step. Extraction step 2 Next, 20 mL of extraction solution (3.3.1) are added to the already ex- tracted sample. The centrifuge tube is shaken vigorously to distribute the sample (the sample can also be stirred if necessary). The sample is again extracted in the ultrasonic bath for 15 min at ambient tempera- ture. Centrifugation The sample is centrifuged applying the above mentioned conditions. The supernatant is added to the first extract. Neutralisation The combined extracts are set to pH 7 using the neutralisation solution (3.3.2). Control of the pH value is accomplished using indicator strips. Usually, about 500 µL to 1000 µL of the solution are needed. The complete neutralised extract is passed through a folded filter (4.13). An aliquot of the filtrate is used for SPE. The filtration step prior to SPE can be repeated in case of larger quantities of remaining particles in the solution. Thereby, blockage of SPE cartridges can be avoided. 6 of 17 Determination of PA in plant material by SPE-LC-MS/MS 5.3 SPE-procedure The Solid Phase Extraction (4.14) is carried out using a vacuum chamber (4.15). Conditioning step 1 5 mL of methanol (3.2.30) Conditioning step 2 5 mL of water Sample load 10 mL sample (filtrate of neutralised extract) Washing step 2 x 5 mL of water Drying of cartridges 5 - 10 min (use the vacuum chamber (4.15)) Elution of PA 2 x 5 mL methanol (3.2.30) or in case of green and black tea 2 x 5 mL 2.5 % ammonia in methanol (3.3.3) The eluate is dried under a nitrogen stream at 50 °C ± 5 °C. 5.4 Reconstitution of the sample The residue is dissolved in 1 mL of methanol/water (5/95, v/v) by shaking (4.8). The reconstituted sample extracts are filtered through 0.2 µm membrane filters (4.16). When using centrifugal filters, 500 µL of the sample are centrifuged at 20 000 x g for 10 min ± 3 min. 200 µL of the filtrate are transferred into an HPLC vial (4.17) with a glass insert (4.18). 6 HPLC-MS/MS analysis 6.1 Liquid chromatographic separation The measurements can be carried out with different high-performance liquid chromato- graphs (HPLC) and separation columns. The chromatographic conditions can be chosen freely. The acceptable minimum retention time is twice the retention time for the dead vol- ume of the column. Analytes which cannot be distinguished by means of mass spectrometry must be separated chromatographically. The conditions listed in the annex 8.1 using a C18- column (4.19) and the mobile phase described in 3.3.4 have shown to be suitable in pre- trials. However, they are to be seen as examples only. 6.2 Mass spectrometric operation conditions The measurements can be carried out with MS/MS devices of different manufacturers. In the annex 8.1, the device-specific settings of one measuring system are given as an example. These conditions have shown to be suitable in pre-trials. Please note: For the qualitative detection and for quantification, it is necessary to detect and report two substance-specific transitions per analyte. 7 of 17 Determination of PA in plant material by SPE-LC-MS/MS 6.3 Measurement For a quantitative analysis, the following criteria are defined. Injection: Samples and standards are injected in duplicate in order to assess repeatability of MS detection and to check for possible response drift during the sequence. Sequence To determine pyrrolizidine alkaloids, the following array of analysis is defined in a sequence. 1. Matrix matched standards (5 – 150 ng/mL) 2. Solvent blank 3. Samples (first injection) 4. Solvent blank 5. Matrix matched standards (5 – 150 ng/mL) 6. Solvent blank 7. Samples (second injection) 7 Calculation The quantitative determination is performed according to the method of the matrix matched standard by integration of the peak areas in relation to the calibration line. 7.1 Calibration function Equation1: Calibration function f  y axb (x) where y is the peak area of the target analyte a is the slope of the calibration function x is the concentration of the target analyte [ng/mL] in the MMS b is the intercept of the calibration function 7.2 Quantification Equation2: Calculation of the PA content (analysis equation)  1 V 1 PAconcentration x DF  ybx x Extract x xV  a m V sample weight Application 8 of 17