ISTITUTO ZOOPROFILATTICO SPERIMENTALE DELL’UMBRIA E DELLE MARCHE-PERUGIA Comparison of three preparation procedures for determination of Organochlorine pesticides in food of animal origin (fat) Angela Paoloni, Sabrina Alunni, Alessandro Pelliccia, Ivan Pecorelli Laboratorio Contaminanti Ambientali - Istituto Zooprofilattico Sperimentale dell’Umbria e delle Marche, via Salvemini 1, Perugia Italy. i.pecorelli@izsum.it Introduction A number of sample preparation techniques and methods of analyses have been developed for the determination of pesticide residues in a wide range of food of animal origin. Analyses of pesticides from food of animal origin were generally performed with an accelerated solvent extraction (ASE) step by a solvent mixture of n-hexane and acetone with a suitable ratio (example 3:2 v/v), fat separation was then carried out with gel permeation chromatography (GPC) and a clean-up step by SPE silica columns was finally performed. Most of these procedures are time and solvent consuming with consequent high cost of analyses. An alternative method may be QuEChERS, developed for the purification of pesticide residues in food of plant origin, is also now being used for other matrices such as milk, eggs or food of animal origin in general. The aim of this work was to evaluate the application of QuEChERS method, coupled with GC-MS/MS (QqQ) detection in Multiple Reaction Monitoring (MRM) for the determination of Organochlorine pesticide residues in fatty animal matrices, alternatively, to classic preparation. The performance of two different types of dispersive solid phase extraction methods (d-SPE) with different concentration of MgSO4 (900 mg and 1200 mg); PSA (150 mg and 400 mg) and C18 (150 mg and 400 mg) respectively, with the presence of ceramic homogenizer in the second type of d-SPE, valuated and compared to validation data obtained by classic ASE/GPC method. GC-MS/MS conditions Agilent 7890 A GC-MS/MS equipped with PTV injector Analytes were separated by a Restek GC column Rtx-Pesticides2 (30 m x 0,25 mm, 0,20 µm) with in-line Siltek filter (5 m x 0,25 mm ID). Total Run Time: 40 min Temperature Program (see Table 1); Carrier gas: Helium 5.5 Collision gas: Nitrogen 6.0 Injection volume: 2 μL Table 1:GC oven parameter Sample preparation Add 4 g of anidrous Magnesium Sulfate (MgSO4) 1 g of Sodium Chloride (NaCl) 1 g of Sodium Citrate Tribasic Dihydrate (C6H5Na3O72H2O) 0.5 g of Sodium Hydrogencitrate Sesquihydrate (C6H6Na2O71.5H2O). 1 g of fat Extraction with (ACN/H2O) The mixture is shaken by vortex for 1 min and then centrifuged at 3000 rpm for 15 min. The supernatant is transferred into a specific d-SPE for purification. Extract is evaporated to dryness and the residue is re-dissolved in 200 μL of isooctane and injected into GC-MS/MS. Figure 1: Cromatogram of matrix matched calibration standard at 0.10 µg/mL. Time (min.) Temp. (°C) Flow (°C/min) 80 20 160 5 300   3 (Post run) Method Comparison The method performances were evaluated in terms of linearity in solvent and in matrix, sensitivity, mean recovery (as measure of trueness or bias) and precision by analysis of 5 replicates, at 0.010 mg/Kg (LOQ) and 0.100 mg/Kg (10xLOQ) in fat. In table 2 - 4 the data for the LOQ (0,010 mg/Kg) level for the three methods are reported: ASE+GPC, QuEChERS with classic d-SPE (150 mg) and with high concentration (400 mg) of C18. Table 2: Validation data at LOQ for ASE + GPC procedure Table 3: Validation data at LOQ for QuEChERS method (150 mg C18) Table 4: Validation data at LOQ for QuEChERS method (400 mg C18) ASE +GPC Pesticide spike 1 spike 2 spike 3 spike 4 spike 5 Average R% SD RSDr alpha-HCH 94 76 80 102 81 87 11 13 Aldrin 95 82 105 101 92 12 14 alpha-Endosulfan 77 66 88 75 8 10 beta-Endosulfan 117 86 85 96 91 16 17 beta-HCH 109 106 83 90 98 cis-Chlordane 100 84 99 Chlorpyrifos(-ethyl) 97 120 107 Chlorpirifos-methyl 119 19 Dieldrin 118 103 Endosulfan sulfate 89 112 115 Endrin 67 20 cis-Heptachlorepoxid 79 9 trans-Heptachlorepoxid Heptachlor 29 72 73 26 36 gamma-HCH (Lindane) 5 6 Hexachlorobenzene (HCB) 64 63 54 15 4,4'-Methoxychlor 65 -- 18 o,p'-DDD 161 116 30 o,p'-DDE o,p'-DDT p,p'-DDD 114 174 35 31 p,p'-DDE 71 93 p,p'-DDT trans-Chlordane 108 57 42 74 D-SPE 150 mg C18 Pesticide spike 1 spike 2 spike 3 spike 4 spike 5 Average R% DS RDSr alpha-HCH 75 100 89 126 80 94 21 22 Aldrin 69 88 65 107 53 76 28 alpha-Endosulfan 74 117 93 129 77 98 24 25 beta-Endosulfan 51 84 81 20 beta-HCH 78 110 102 115 16 cis-Chlordane 72 85 97 91 12 13 Chlorpyrifos(-ethyl) 127 95 18 19 Chlorpirifos-methyl 105 96 15 Dieldrin 79 86 108 87 Endosulfan sulfate 121 113 17 Endrin 103 10 11 cis-Heptachlorepoxid 256 101 60 118 67 trans-Heptachlorepoxid 64 27 68 32 46 Heptachlor -- 90 gamma-HCH (Lindane) 92 120 Hexachlorobenzene (HCB) 47 62 43 40 4,4'-Methoxychlor 83 109 o,p'-DDD 112 70 o,p'-DDE 42 29 o,p'-DDT 66 p,p'-DDD 114 p,p'-DDE 59 99 23 p,p'-DDT trans-Chlordane 71 D-SPE 400 mg C18 Pesticide spike 1 spike 2 spike 3 spike 4 spike 5 Average R% DS RDSr alpha-HCH 86 85 95 94 89 5 6 Aldrin 66 70 65 73 68 3 alpha-Endosulfan 75 82 78 92 88 83 7 9 beta-Endosulfan 81 80 84 97 87 8 beta-HCH 96 93 101 cis-Chlordane 77 71 Chlorpyrifos(-ethyl) 90 100 91 Chlorpirifos-methyl 102 Dieldrin 72 Endosulfan sulfate Endrin cis-Heptachlorepoxid trans-Heptachlorepoxid Heptachlor 10 gamma-HCH (Lindane) Hexachlorobenzene (HCB) 53 57 62 59 11 4,4'-Methoxychlor 116 105 103 o,p'-DDD o,p'-DDE 76 69 o,p'-DDT 74 4 p,p'-DDD 98 p,p'-DDE p,p'-DDT trans-Chlordane Conclusions In accordance with performance parameters described in point G6 of Document N° SANTE/11945/2015, high C18 (400 mg) QuEChERS method validation results, obtained from 5 replicates, showed mean recovery in the range of 70-120 % and particularly good RSDs values below 12% for all the 24 molecules analyzed. Only Hexachlorobenzene (HCB) and Aldrin has a mean recovery value below 70% (with good precision). The presence of high PSA and C18 sorbent with ceramic homogenizers seem to be essential to achieve good performances and more clear purified extracts .The proposed method is now applied in our laboratory for routine samples and will be tested in the next EUPT-AO on pig lard. Reference: [1] Document N° SANTE/11945/2015.