Acta agriculturae Slovenica, 117/2, 1–11, Ljubljana 2021 doi:10.14720/aas.2021.117.2.1822 Original research article / izvirni znanstveni članek Pesticide residues in bee pollen - validation of the gas chromatography- mass spectrometry multiresidual method and a survey of bee pollens from Slovenia Helena BAŠA ČESNIK 1* Received August 11, 2020; accepted April 09, 2021. Delo je prispelo 11. avgusta 2020, sprejeto 9. aprila 2021. Pesticide residues in bee pollen - validation of the gas chro- matography-mass spectrometry multiresidual method and a survey of bee pollens from Slovenia Abstract: A new analytical method for determining en- vironmental pesticide residues in pollen was introduced and validated. The extraction was conducted using acetonitrile, the clean-up using Supelclean Ultra 2400 solid phase extrac- tion cartridges, which contain Grapsphere, anion exchanger, C18 and zirconia-based sorbent, and the determination was conducted using gas chromatography coupled with mass spectrometry. The method was applied in practice. A total of 49 active substances (pesticides) were sought in 30 bee pollen samples gathered from Slovenian beekeepers from all 12 sta- tistical regions of Slovenia. The fungicide azoxystrobin was the only active substance found and was found only in one sample with a concentration of < 0.05 mg kg -1 . The active substances sought were not detected in 96.7 % of the samples analysed. The risk assessment revealed that the analysed pollen samples do not represent an unacceptable risk for consumers. The results were compared with those from the literature and the outcome was that bee pollen from Slovenia contained a lower number of active substances at mainly lower contents as compared pollen from some other European countries. Key words: bee pollen; GC-MS; pesticide residues; mul- tiresidual method 1* Agricultural Institute of Slovenia, Hacquetova ulica 17, SI-1000 Ljubljana, Slovenia, PhD. e-mail: helena.basa@kis.si, corresponding author Ostanki fitofarmacevtskih sredstev v cvetnem prahu - valida- cija multirezidualne metode s plinsko kromatografijo sklo- pljeno z masno spektrometrijo in preiskava cvetnega prahu iz Slovenije Izvleček: Uvedli in validirali smo novo analizno metodo za določanje ostankov fitofarmacevtskih sredstev iz okolja. Ekstrakcijo smo izvedli z acetonitrilom, čiščenje z Supelclean Ultra 2400 koloncami za ekstrakcijo na trdni fazi, ki vsebu- jejo Grapsphere, anionski izmenjevalnik, C18 in sorbent na osnovi cirkonija, in določitev s plinsko kromatografijo sklo- pljeno z masno spektrometrijo. Metodo smo uporabili v praksi. V 30 vzorcih cvetnega prahu slovenskih čebelarjev iz vseh 12 statističnih regij Slovenije smo določali skupno 49 aktivnih spojin (pesticidov). Edina najdena aktivna snov je bil fungicid azoksistrobin in sicer le v enem vzorcu, pri koncentraciji < 0,05 mg kg -1 Iskanih aktivnih snovi nismo detektirali v 96,7 % ana- liziranih vzorcev. Z oceno tveganja smo ugotovili, da analizirani vzorci cvetnega prahu ne predstavljajo tveganja za potrošnika. Rezultate smo primerjali z literaturnimi podatki in ugotovili, da je cvetni prah v Sloveniji vseboval manjše število aktivnih spo- jin pri v glavnem nižjih vsebnostih fitofarmacevtskih ostankov kot cvetni prah iz nekaterih Evropskih dtžav. Ključne besede: cvetni prah; GC-MS; ostanki fitofar- macevtskih sredstev; multirezidualna metoda Acta agriculturae Slovenica, 117/2 – 2021 2 H. BAŠA ČESNIK 1 INTRODUCTION Bee pollen is a dietary supplement. It contains car- bohydrates (mainly fructose, glucose and sucrose (13-55 %), proteins (10-40 %), lipids (1-13 %), and crude fibre (0.3-20 %)), minerals (mainly potassium, phosphorus, calcium, magnesium, zinc, manganese, iron and copper (2-6 %)), vitamins (0.005-5.6 mg kg -1 ), and polyphenols (0.69-213.2 mg GAE g -1 ) (Thakur and Nanda, 2020). Bee pollen has antioxidant activity, antimicrobial activity, an- ti-inflammatory activity, anticarcinogenic activity, cardi- oprotective effects, hepatoprotective effects, antiallergic activity and it boosts the immune system (Li et al., 2018). A diet supplemented with bee pollen strengthens mus- cles and improves the physical health of humans (Salles et al., 2014). Bee pollen also benefits those undertaking strenuous mental/physical work (Nakajima et al., 2009). Honeybees fly up to 4.8 km from their apiary (Eck- ert, 1933) to collect pollen. When hives are located near agricultural fields, plants treated with plant protection products (PPP) are a possible source of contamination for bee pollen (Tosi et al., 2018). Honeybees may come into contact with PPP residues through the nectar, pollen or plant leaves of treated plants, or through air, soil and water where PPPs have drifted (Crenna et al., 2020). Bee pollen is usually harvested by means of a trap fixed at the entrance of beehives (Thakur and Nanda, 2020). This type of pollen is called corbicular pollen. Some beekeepers also collect pollen from hives deposited in combs by bees. This type of pollen is called beebread. Numerous analytical methods have been developed to analyse PPP residues in pollen. The more recent ones are based on the QuEChERS method, which has been introduced to analyse a wide range of PPP residues in fruit and vegetables (Anastassiades et al., 2003, Lehotay, 2007). In this method, acetonitrile is used as an organic solvent for the extraction. The advantage of acetonitrile is that it minimizes the co-extraction of lipids and proteins by precipitating the proteins (Wang et al., 2012) and lim- iting the lipid solubility (Lozano et al., 2014). This makes acetonitrile a suitable solvent for extracting PPP residues from pollen. In some cases (Tosi et al., 2018; Wiest et al., 2011), n-hexane was added to remove fatty acids and fatty acid esters. The clean-up in the original QuEChERS method was conducted using primary secondary amine (PSA) sorbent and C18 (Anastassiades et al., 2003; Lehotay, 2007). In the case of pollen some authors used either PSA (Cabrera de Oliveria, 2016; Kasiotis et al., 2014), PSA and C18 sorbent (Mullin et al., 2010), PSA, C18 and gra- phitized carbon black (GCB) sorbent (David et al., 2016), or PSA, C18 and zirconia-based sorbents such as Z-Sep, which consists of a mixture of C18 and silica coated with zirconium dioxide sorbents (Hakme et al., 2017, Vázquez et al., 2015). In our laboratory we used Supelclean Ul- tra 2400 solid phase extraction (SPE) cartridges, which contain Grapsphere (graphitized spherical carbon), PSA, C18 and Z-Sep. PSA retains acidic interferences such as fatty acids. Grapsphere removes planar molecules such as pigments and at the same time enables better recovery of planar pesticides than GCB. The bottom layer of the car- tridge contains Z-Sep, which removes oily residues and provides additional retention of some pigments (Stener- son, 2018). C18 retains lipids (Lehotay,2007). Thus, these SPE cartridges combine all the common clean-up proce- dures from the literature. Determination of PPP residues can be performed using gas chromatography coupled with mass spectrom- etry (GC-MS) (Li et al.; 2015; Mullin et al.,2010; Raimets et al., 2020), GC coupled with tandem mass spectrom- etry (GC-MS/MS) (Cabrera de Oliveria, 2016), GC cou- pled with time-of flight mass spectrometry (GC-TOF) (Hakme et al., 2017), and/or liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) (David et al., 2016; Kasiotis et al., 2014; Raimets et al., 2020). Multiresidual methods are fit-for-purpose when their limits of quantification are lower or equal to the Maximum Residue Limits (MRLs) established for pol- len in Regulation (EC) 396/2005. When the MRLs in this Regulation are set at the LOQ determined by the analyti- cal method (this LOQ was gathered by different labora- tories), an * is added to mark this fact. Many pesticides have an MRL and LOQ of 0.05 mg kg -1 , meaning that GC-MS is still suitable despite its smaller sensitivity than tandem mass detectors. Numerous authors have analysed pesticide residues in pollen. García-Valcárcel et al. (2019) analysed 10 ac- tive substances in pollen samples in Spain. Vázquez et al. (2015) analysed 253 active substances in pollen samples in Spain. Hakme et al. (2017) tested pollen samples from Spain for 100 active substances. Wiest et al. (2011) intro- duced a method for determining 80 active substances in French pollen. Tosi et al. (2018) analysed 66 active sub- stances in Italian pollen. Kasiotis et al. (2014) analysed 115 active substances in Greek pollen. Raimets et al. (2020) analysed 47 active substances in Estonian pollen. David et al. (2016) analysed 20 active substances in pol- len from the United Kingdom. Many of active substances sought in these studies were introduced in our study as well. Our selection of active substances was based on both those authorised for use in Slovenia and those not authorised for use in Slovenia, the latter to cover misuse of PPP. Of those selected, 59 % were acaricides and/or insecticides, which may be the main reason for the death of bees. The purpose of this paper is to present the multire- Acta agriculturae Slovenica, 117/2 – 2021 3 Pesticide residues in bee pollen - validation of the gas chromatography-mass spectrometry multiresidual ... of bee pollens from Slovenia sidual GC-MS method introduced for identifying 49 ac- tive substances in pollen using acetonitrile as the extrac- tion solvent and Supelclean Ultra 2400 SPE cartridges for the clean-up. The validation parameters are summarised, as well as the practical use of the method on 30 samples of bee pollen gathered from Slovenian beekeepers. The contents of pesticide residues were compared with those from the literature. Finally, a risk assessment for consum- ers was conducted. 2 MATERIAL AND METHODS 2.1. MATERIALS 2.1.1 Chemicals The certified standards were supplied by Dr. Ehren- storfer (Augsburg, Germany). The acetonitrile HPLC- grade (used for the extraction procedure) and acetone HPLC-grade (used for preparation of standards) were supplied by J.T.Baker (Deventer, Netherlands). All other chemicals used were supplied by Sigma-Aldrich (Stein- heim, Germany). The water used was MilliQ deionised water. The Ultra 2400 3 ml SPE columns were supplied by Supelco (Bellefonte, USA). 2.1.2. Preparation of the solutions Stock solutions in acetone of individual active sub- stances were prepared with the concentrations of 625 μg pesticide ml -1 . From 49 stock solutions, two mixed solu- tions of all 49 active substances were prepared: one with a concentration of 5 μg ml -1 and the second at the LOQ of active substances. All solutions used to determine the linearity and the LOQs and to perform calibration dur- ing sample analysis were prepared from a mixed solution of 5 μg ml -1 with proper dilutions. For other validation parameters, a mixed solution with a concentration at the LOQ was used. 2.2. EXTRACTION PROCEDURE The samples were analysed within a maximum pe- riod of 27 days after arrival at the laboratory. During that time, they were stored at -20 °C. To 10 g of pollen in the beaker, 50 ml of acetonitrile was added. The mixture was homogenised for 2 minutes with a mixer. The mixture was left for 30 minutes so that the sediment settled on the bottom of the beaker. The liq- uid part was transferred to a 50 ml centrifuge tube and centrifuged for 10 minutes at 7000 rpm. The supernatant was filtered through 15 g anhydrous Na 2 SO 4 and black strip filter paper into a 100 ml Soxhlet flask. Then 30 ml of acetonitrile was added to the sediment in the beaker. The mixture was homogenised for 2 minutes with a mix- er and transferred to a 50 ml centrifuge tube. Centrifuga- tion followed for 10 minutes at 7000 rpm. This superna- tant was combined with the first one after it was filtered through 15 g of anhydrous Na 2 SO 4 and black strip filter paper into a 100 ml Soxhlet flask. The Na 2 SO 4 was rinsed with 15 ml of acetonitrile. Then acetonitrile in Soxhlet flask was evaporated to approximately 2 ml on a rota- vapor and dried with nitrogen flow. The dry eluate was dissolved in 1 ml of acetonitrile using ultrasound. The extract was transferred onto a column of Ultra 2400 3 ml, preconditioned with 3 ml of acetonitrile. The SPE col- umn was rinsed with 16 ml of acetonitrile. The flow rate was 3-4 ml min -1 under vacuum. The whole eluate (partly taken from the SPE column after the sample was applied to the SPE column and partly eluate created during rins- ing of the SPE column) was combined in a beaker. The content of the beaker was transferred to a 100 ml Soxhlet flask. The beaker was rinsed twice with 5 mL of acetoni- trile and the content was transferred to the Soxhlet flask. The acetonitrile was then evaporated to approximately 2 ml on a rotavapor and dried with nitrogen flow. The dry eluate was dissolved in 1 ml of acetone using ultrasound in order to prepare a sample. For matrix match standards, 1 ml of the working solutions with proper concentrations was added and dissolved using ultrasound. 2. 3. DETERMINATION The samples were analysed using a gas chromato- graph (Agilent Technologies 7890A, Shanghai, China) equipped with a Gerstel MPS2 multipurpose sampler (Gerstel, Mülheim an der Ruhr, Germany) and a HP-5 MS UI column (Agilent Technologies, 30 m, 0.25 mm i.d., 0.25 μm film thickness) with a constant flow of he- lium at 1.2 ml min -1 . The GC oven was programmed as follows: 55 °C for 2 min, from 55 °C to 130 °C at 25 °C min -1 , held at 130 °C for 1 min, from 130 °C to 180 °C at 5 °C min -1 , held at 180 °C for 30 min, from 180 °C to 230 °C at 20 °C min -1 , held at 230 °C for 16 min, from 230 °C to 250 °C at 20 °C min -1 , held at 250 °C for 13 min, from 250 °C to 280 °C at 20 °C min -1 , held at 280 °C for 20 min. In order to determine the analytes, a mass spectrometer (Agilent T echnologies 5975C, upgraded with a triple-axis detector, Palo Alto, CA, USA) was used. The temperature of the ion source was 230 °C, the auxiliary temperature was 280 °C and the quadrupole temperature was 150 °C. For qualitative determination, the retention time and Acta agriculturae Slovenica, 117/2 – 2021 4 H. BAŠA ČESNIK mass spectrum in the SIM were used. For each active substance, one target and two qualifier ions, presented in Table 2, were used. The calibration was performed to matrix match standards. 2. 4. V ALIDATION OF METHODS LOQ and linearity The linearity was verified using the matrix match standards (two repetitions for one concentration level, four to six concentration levels for the calibration curve). The linearity and range were determined by linear regres- sion, using the F test. LOQs were estimated from the chromatograms of matrix match standards. LOQs were chosen at a mini- mum of S/N = 10. MRLs for environmental pesticide residues are set in Regulation (EC) 396/2005. Where the MRLs are set at the LOQ determined using the analytical method (this LOQ was gathered by different laboratories) in the Regu- lation an * is added to mark this fact. Therefore, in cases where MRLs were marked with an *, our LOQs were set at those MRLs. Precision Blank pollen was bought in store and analysed to prove that it contains no pesticide residues. For the de- termination of precision (ISO 5725), i.e. repeatability and reproducibility, the extracts of spiked blank pollen were analysed at LOQ. Within a period of 10 days, two parallel extracts were prepared each day for each concentration level. Each one was injected once. Then the standard de- viation of the repeatability of the level and the standard deviation of reproducibility of the level were both calcu- lated. Uncertainty of repeatability and uncertainty of re- producibility The uncertainty of repeatability and the uncer- tainty of reproducibility were calculated by multiplying the standard deviation of repeatability and the standard deviation of reproducibility by the Student’s t factor, for nine degrees of freedom and a 95 % confidence level (t 95;9 = 2.262). U r = t 95; 9 x s r ; U R = t 95; 9 x s R The measurement uncertainty for PPP residues should be 50 %, as proposed in SANTE/11813/2017. When validating, analysts must prove that their meas- urement uncertainty is below or equal to the proposed measurement uncertainty. Accuracy The accuracy was verified by checking the recover - ies. The average of the recoveries from the tests for preci- sion (10 days, 2 parallel samples each day) was calculat- ed. According to the requirements for method validation procedures (SANTE/11813/2017), acceptable mean re- coveries are those within the range of 70 % to 120 %, with an associated repeatability of RSDr ≤ 20 %. According to the guidelines for single-laboratory validation (Alder et al. 2000), acceptable mean recoveries are as follows: - at level > 0.01 mg kg -1 ≤ 0.1 mg kg -1 , acceptable mean recoveries are those within the range of 70 % to 120 %, with an associated repeatability RSDr ≤ 20 % and - at level > 0.001 mg kg -1 ≤ 0.01 mg kg -1 , acceptable mean recoveries are those within the range of 60 % to 120 %, with an associated repeatability RSDr ≤ 30 %. 2. 5. CONSUMER RISK ASSESSMENT Long-term exposure was calculated using the EFSA PRIMo model revision 3.1, accessible online at https:// www.efsa.europa.eu/en/applications/pesticides/tools. Chronic consumer exposure was expressed in % of the ADI. The acceptable limit for long-term exposure is 100 % of the ADI. 2. 6. SAMPLING A total of 30 bee pollen samples (none of them beebread) were collected in May and June 2020 from Slovenian beekeepers that produce apiculture products sold on the market. Samples were gathered from all 12 statistical regions in Slovenia. The sampling distribution is presented in Table 1. All samples originated from con- ventional production. 3 RESULTS AND DISCUSSION 3. 1. COMPARISON OF QUECHERS METHOD WITH OUR METHOD In the original QuEChERS method, 10 ml of ace- tonitrile was added to 10 g of the sample (Anastassiades et al.; 2003). In our method this ratio was different: 80 ml of acetonitrile was added to 10 g of the sample. The reason for increasing the solvent volume was that when we tested the addition of 10 ml of acetonitrile to 10 g of pollen, the recoveries were 20-30 % lower. Acta agriculturae Slovenica, 117/2 – 2021 5 Pesticide residues in bee pollen - validation of the gas chromatography-mass spectrometry multiresidual ... of bee pollens from Slovenia Table 1: Number of pollen samples collected from different statistical regions in Slovenia in 2020 Statistical region Number of samples Gorenjska 2 Goriška 2 Jugovzhodna Slovenija 1 Koroška 2 Notranje kraška 3 Obalno kraška 2 Osrednja Slovenija 7 Podravska 3 Pomurska 2 Savinjska 3 Spodnje posavska 1 Zasavska 2 Sum 30 The clean-up in our method was not conducted with dispersive SPE as in the original QuEChERS meth- od (Anastassiades et al.; 2003, Lehotay; 2007), but with Ultra 2400 3 ml SPE columns. In the QuEChERS method, aliquots of extracts were cleaned-up, while in our method, the transference of the extracts was quantitative. 3. 2. V ALIDATION OF METHOD LOQ and linearity The linear model is valid for all active substances presented in Table 2. Linearity was proven in the range of 0.01 mg kg -1 to 0.15 mg kg -1 for six active substances, in the range of 0.05 mg kg -1 to 0.12 mg kg -1 for one active substance and in the range of 0.05 mg kg -1 to 0.15 mg kg -1 for 42 active substances. R 2 ranged from 0.974 to 0.996. The LOQs are presented in Table 2. Six active sub- stances have an LOQ of 0.01 mg kg -1 and 43 of them 0.05 mg kg -1 . The LOQs are equal to MRLs set in Regulation (EC) 396/2005. Accuracy The results for the recoveries are given in Table 2. The recoveries at LOQs for the active substances scanned with GC-MS are in the range of 73.0 % to 93.4 %, with RSDs of 5.6 % to 17.7 %. More precisely, the recoveries at LOQs of 0.01 mg kg -1 are within the range of 75.4 % to 93.4 % with RSDs of 9.3 % to 17.7 % and the recoveries at LOQs of 0.05 mg kg -1 are within the range of 73.0 % to 88.2 % with RSDs of 5.6 % to 15.9 %. All recoveries and RSDs are within the re- quired ranges from the literature (Alder et al., 2000; SANTE/11813/2017). Uncertainty of repeatability and uncertainty of re- producibility The uncertainty of repeatability and uncertainty of reproducibility were determined at contents equal to the LOQs. The results are presented in Table 2. Uncertainty of repeatability ranged from 0.001 mg kg -1 to 0.013 mg kg -1 , which is 10.0 % to 30.0 % of LOQ and uncertainty of reproducibility ranged from 0.002 mg kg -1 to 0.015 mg kg -1 , which is 12.0 % to 30.0 % of LOQ. Acta agriculturae Slovenica, 117/2 – 2021 6 H. BAŠA ČESNIK Active substance Activity type a MRL b (mg kg -1 ) Ions scanned c (m/z) T, Q 1 , Q 2 Linearity range (mg kg -1 ) R 2 LD (mg kg -1 ) LOQ (mg kg -1 ) Recovery (%) RSD d (%) U r e (mg kg -1 ) U r f (%) U R g (mg kg -1 ) U R h (%) acrinathrin A, I 0.05* 181, 208, 289 0.05-0.15 0.991 0.015 0.05 87.8 13.8 0.010 20 0.014 28 azinphos-methyl A, I / 160, 132, 105 0.05-0.15 0.987 0.015 0.05 84.2 9.9 0.009 18 0.010 20 azoxystrobin F 0.05* 344, 388, 345 0.05-0.15 0.987 0.015 0.05 85.3 11.8 0.011 22 0.011 22 bifenthrin A, I 0.05* 181, 165, 166 0.05-0.15 0.992 0.015 0.05 76.6 8.0 0.010 20 0.010 20 boscalid F 0.05* 140, 342, 142 0.05-0.15 0.984 0.015 0.05 85.0 11.9 0.009 18 0.012 24 carbaryl A, I 0.05* 144, 115, 116 0.05-0.15 0.988 0.015 0.05 81.9 7.0 0.010 20 0.010 20 carbofuran A, I 0.05* 164, 149, 131 0.05-0.15 0.992 0.015 0.05 80.6 8.2 0.007 14 0.008 16 chlorpropham H 0.05* 213, 127, 154 0.05-0.15 0.983 0.015 0.05 74.9 8.0 0.007 14 0.007 14 chlorpyriphos A, I 0.05* 314, 316, 197 0.05-0.15 0.982 0.015 0.05 76.0 8.5 0.010 20 0.010 20 chlorpyriphos-methyl I 0.05* 286, 288, 125 0.05-0.15 0.996 0.015 0.05 77.6 7.0 0.010 20 0.010 20 clomazone H 0.05* 125, 204, 127 0.05-0.15 0.988 0.015 0.05 75.2 10.9 0.006 12 0.009 18 cyhalotrin-lambda I 0.05* 181, 197, 208 0.05-0.15 0.987 0.015 0.05 88.2 12.3 0.011 22 0.012 24 deltamethrin I 0.05* 181, 251, 255 0.05-0.15 0.990 0.015 0.05 80.1 14.9 0.010 20 0.014 28 diazinon A, I 0.01* 179, 304, 199 0.01-0.15 0.994 0.003 0.01 81.3 12.9 0.002 20 0.002 20 dichlofluanid A, F / 226, 123, 167 0.05-0.15 0.990 0.015 0.05 79.1 7.9 0.010 20 0.010 20 dimethachlor H 0.05* 134, 197, 210 0.05-0.15 0.995 0.015 0.05 81.2 6.5 0.006 12 0.006 12 dimethoate A, I / 87, 229, 143 0.05-0.15 0.994 0.015 0.05 81.7 5.6 0.010 20 0.010 20 diphenylamine F 0.05* 169, 167, 168 0.05-0.12 0.982 0.015 0.05 73.2 9.1 0.008 16 0.008 16 endosulfan-sulphate A, I 0.01* 272, 274, 387 0.01-0.15 0.995 0.003 0.01 82.4 9.3 0.001 10 0.002 20 fenbuconazole F 0.05* 198, 129, 125 0.05-0.15 0.989 0.015 0.05 78.9 15.9 0.013 26 0.014 28 fenitrothion I 0.01* 277, 260, 109 0.01-0.15 0.981 0.003 0.01 84.7 14.7 0.002 20 0.003 30 flonicamid I 0.05* 174, 146, 229 0.05-0.15 0.992 0.015 0.05 79.8 7.7 0.007 14 0.007 14 fludioxonil F 0.05* 248, 154, 127 0.05-0.15 0.990 0.015 0.05 85.5 9.8 0.009 18 0.010 20 fluquinconazole F 0.05* 340, 342, 108 0.05-0.15 0.989 0.015 0.05 82.6 10.7 0.010 20 0.010 20 HCH-alpha I 0.01* 219, 181, 183 0.01-0.15 0.996 0.003 0.01 75.4 12.3 0.002 20 0.002 20 Table 2: Validation parameters, ions scanned, MRLs and activity type of active substances Acta agriculturae Slovenica, 117/2 – 2021 7 Pesticide residues in bee pollen - validation of the gas chromatography-mass spectrometry multiresidual ... of bee pollens from Slovenia Active substance Activity type a MRL b (mg kg -1 ) Ions scanned c (m/z) T, Q 1 , Q 2 Linearity range (mg kg -1 ) R 2 LD (mg kg -1 ) LOQ (mg kg -1 ) Recovery (%) RSD d (%) U r e (mg kg -1 ) U r f (%) U R g (mg kg -1 ) U R h (%) HCH-deltha I / 219, 181, 183 0.05-0.15 0.979 0.015 0.05 82.8 12.7 0.007 14 0.012 24 iprodione F 0.05* 314, 316, 187 0.05-0.15 0.986 0.015 0.05 81.6 9.8 0.010 20 0.010 20 kresoxim-methyl F 0.05* 116, 206, 131 0.05-0.15 0.992 0.015 0.05 84.5 9.3 0.008 16 0.009 18 mecarbam A, I 0.05* 131, 159, 329 0.05-0.15 0.974 0.015 0.05 79.7 10.8 0.010 20 0.010 20 methacrifos A, I 0.05* 208, 180, 240 0.05-0.15 0.993 0.015 0.05 73.0 8.4 0.006 12 0.007 14 metrafenone F 0.05* 393, 408, 379 0.05-0.15 0.992 0.015 0.05 80.5 9.2 0.010 20 0.010 20 parathion A, I / 291, 292, 235 0.05-0.15 0.993 0.015 0.05 82.3 15.7 0.009 18 0.015 30 permethrin A, I / 183, 163, 165 0.05-0.15 0.989 0.015 0.05 78.3 10.2 0.008 16 0.008 16 phorate A, I 0.01* 231, 260, 97 0.01-0.15 0.996 0.003 0.01 82.4 17.7 0.002 20 0.003 30 phosalone A, I 0.01* 182, 367, 121 0.01-0.15 0.995 0.003 0.01 93.4 13.9 0.003 30 0.003 30 pirimicarb I 0.05* 166, 238, 167 0.05-0.15 0.983 0.015 0.05 79.5 7.2 0.006 12 0.007 14 pirimiphos-methyl A, I 0.05* 290, 305, 276 0.05-0.15 0.992 0.015 0.05 78.4 7.0 0.010 20 0.010 20 procymidone F 0.05* 283, 285, 96 0.05-0.15 0.981 0.015 0.05 80.1 9.6 0.010 20 0.010 20 propyzamide H 0.05* 173, 175, 145 0.05-0.15 0.993 0.015 0.05 78.3 9.8 0.007 14 0.009 18 pyridaphenthion I / 199, 340, 188 0.05-0.15 0.992 0.015 0.05 83.9 9.2 0.008 16 0.009 18 quinalphos A, I 0.05* 146, 298, 157 0.05-0.15 0.992 0.015 0.05 81.2 7.8 0.010 20 0.010 20 quinoclamine H 0.05* 207, 172, 209 0.05-0.15 0.990 0.015 0.05 76.9 12.8 0.010 20 0.010 20 tetradifon A 0.05* 159, 229, 356 0.05-0.15 0.990 0.015 0.05 82.1 10.7 0.009 18 0.010 20 tolclofos-methyl F 0.05* 265, 267, 250 0.05-0.15 0.994 0.015 0.05 78.4 7.9 0.010 20 0.010 20 tolylfluanid F 0.05* 238, 137, 240 0.05-0.15 0.985 0.015 0.05 80.8 7.7 0.007 14 0.007 14 triadimefon F 0.05* 208, 210, 181 0.05-0.15 0.993 0.015 0.05 80.4 9.0 0.007 14 0.008 16 triazophos A, I 0.05* 161, 162, 285 0.05-0.15 0.982 0.015 0.05 84.1 10.4 0.010 20 0.010 20 trifloxystrobin F 0.05* 116, 222, 186 0.05-0.15 0.992 0.015 0.05 83.7 8.7 0.008 16 0.008 16 vinclozolin F 0.05* 285, 124, 187 0.05-0.15 0.995 0.015 0.05 83.9 8.1 0.008 16 0.008 16 a A = acaricide, I = insecticide, F = fungicide, H = herbicide b Regulation (EC) 396/2005, * means that MRL is set at LOQ of analytical method c T = target ion, Q = qualifier ion d RSD was obtained during recovery analyses e,f U r = uncertainty of repeatability g,h U R = uncertainty of reproducibility Acta agriculturae Slovenica, 117/2 – 2021 8 H. BAŠA ČESNIK 3. 3. SURVEY OF PESTICIDE RESIDUES IN BEE POLLEN SAMPLES The Ministry of Agriculture, Forestry and Food reported that in Slovenia in 2020, 582 PPPs, containing 239 active substances, are authorised for use on different agricultural products. The Statistical Office announced that in 2018, 1,172 tons of active substances were sold in Slovenia, where we have 476,000 hectares of cultivated agricultural area. This suggests broad use of PPPs among farmers. Since bees collect pollen not only on flowers, acacia, spruce, sage, lime and chestnut but also on agri- cultural products treated with PPPs, such as oilseed rape, fruits, etc., we wanted to research if these kinds of pesti- cide residues are found in bee pollen. We were search- ing for authorised (33 % of active substances sought) and non-authorised active substances in Slovenia, to cover the possible misuse of PPPs. Of the 30 bee pollen samples analysed, only one contained one active substance: azoxystrobin, with a concentration of < 0.05 mg kg -1 . This means that in 96.7 % of all samples analysed, no pesticide residues were de- tected. The MRL for azoxystrobin in pollen is 0.05 mg kg -1 and it was not exceeded. In Slovenia, azoxystrobin is authorised as a fungicide for use on oilseed rape, vine and ornamentals (among others) in 14 different PPPs. These are the plants on which bees collect pollen. A consumer risk assessment was performed using the EFSA PRIMo model rev. 3.1, in which 36 national di- ets from EU countries are included. This model was used since Slovenia has not created a model of its own. The same model is used in the process of registration of PPPs in Slovenia. Since azoxystrobin was the only substance found and it was only found in one sample at a concen- tration of < LOQ, the LOQ for this substance was used as the input value in PRIMo model. It was compared to the Acceptable Daily Intake (ADI) of azoxystrobin (0.2 mg (kg bw) -1 d -1 ). The calculations of chronic exposure for azoxystrobin showed that the highest was observed in the German diet for children. It represented 0.003 % of ADI. Since no Acute Reference Dose was set for azoxystrobin, no acute exposure was calculated. Based on these calcu- lations, the conclusion was that the analysed bee pollen samples are of no cause for concern for consumers. Our results were compared with the results from other scientific papers. Azoxystrobin was found in the Estonian pollen by Raimets et al. (2020) in 3.4 % of all samples analysed up to a concentration of 0.04 mg kg - 1 . Tosi et al. (2018) wrote that azoxystrobin was found in 2.9 % of the Italian pollen samples analysed, with a maximum concentration of 0.054 mg kg -1 . Vázquez et al. (2015) reported that azoxystrobin was found in a con- centration of up to 0.235 mg kg -1 in the Spanish pollen. Azoxystrobin was found in 3.3 % of the Slovenian pollen samples analysed, which is comparable to Estonia and Italy. The concentration of azoxystrobin found in Slove- nia is comparable to that found in Estonia and Italy, but much lower than in Spain. Other active substances analysed in our labora- tory, namely acrinathrin, bifenthrin, boscalid, carbaryl, carbofuran, chlorpyrifos, clomazone, dimethoate, feni- trothion, fludioxonil, iprodione, lambda-cyhalothrin, permethrin, trifloxystrobin and vinclozoline, were not detected in Slovenian pollen, but were found in samples analysed in Estonia, France, Greece, Italy, Spain and the United Kingdom. All active substances sought by our laboratory and positively identified in Europe were measured up to con- centrations higher than our LDs. The exception is car- bofuran, which was found at a concentration 10-times lower than our LD. Literature results for these active sub- stances are presented in Table 3. 4 CONCLUSIONS In our research, a method for determining pesti- cide residues originating from the environment in pollen was introduced and validated. The limit of detection was 0.003 mg kg -1 for 6 active substances and 0.01 mg kg -1 for 43 active substances. The limit of quantification was 0.01 mg kg -1 for 6 active substances and 0.05 mg kg -1 for 43 active substances. The calibration curves gave a linear response with R 2 0.974 to 0.996. The recoveries ranged from 73.0 % to 93.4 % with RSDs from 5.6 % to 17.7 %. The measurement uncertainty of repeatability ranged from 10 to 30 % and the measurement uncertainty of re- producibility from 12 to 30 %. The method was found to be fit for purpose of measuring possible breaches of MRL for 49 active substances. The method was used to analyse 30 bee pollen sam- ples gathered from Slovenian beekeepers, all from con- ventional production. A total of 49 active substances were sought, but only the fungicide azoxystrobin was found in only one of these samples. In 96.7 % of the sam- ples analysed, the active substances sought were not de- tected. A risk assessment revealed that the Slovenian bee pollen samples are no cause for concern for consumers. 5 ACKNOWLEDGEMENTS The author expresses her thanks to Mojca Trček for her help with the preparation of the extracts. The author acknowledges the financial support of the Slovenian Re- search Agency (research core funding No. P4-0133). Acta agriculturae Slovenica, 117/2 – 2021 9 Pesticide residues in bee pollen - validation of the gas chromatography-mass spectrometry multiresidual ... of bee pollens from Slovenia Table 3: Literature results for active substances sought, but not found in our laboratory Active substance   Limit of detection (mg kg -1 ) Max content (mg kg -1 ) Ratio of positive samples (%) Country of origin  Reference  acrinathrin not reported 0.458 20.0 Spain Calatayud-Vernich et al., 2018 acrinathrin 0.015 0.055 not reported Spain Vázquez et al., 2015 bifenthrin 0.015 0.015 not reported Spain Vázquez et al., 2015 boscalid 0.0025 0.058 0.7 Italy Tosi et al., 2018 boscalid 0.00012 0.021 52.0 United Kingdom David et al., 2016 boscalid 0.0015 0.03 not reported Spain Vázquez et al., 2015 carbaryl 0.0007 0.015 8.0 France Wiest et al., 2011 carbaryl 0.00025 0.001 0.2 Italy Tosi et al., 2018 carbofuran 0.0004 0.002 2.0 France Wiest et al., 2011 chlorpyrifos not reported 0.05 14.0 Spain Hakme et al., 2017 chlorpyrifos 0.001 0.3982 not reported Spain García-Valcárcel et al., 2019 chlorpyrifos not reported 0.1 31.1 Spain Calatayud-Vernich et al., 2018 chlorpyrifos 0.0015 0.07 not reported Spain Vázquez et al., 2015 chlorpyrifos 0.008 0.14 4.0 France Wiest et al., 2011 chlorpyrifos 0.0032 0.046 not reported Greece Kasiotis et al., 2014 chlorpyrifos 0.001 0.179 30.3 Italy Tosi et al., 2018 clomazone not reported 0.02 5.0 Spain Hakme et al., 2017 dimethoate not reported 0.042 20.7 Estonia Raimets et al., 2020 dimethoate 0.0015 0.015 not reported Spain Vázquez et al., 2015 dimethoate not reported 0.022 8.9 Spain Calatayud-Vernich et al., 2018 dimethoate 0.00025 0.163 7.9 Italy Tosi et al., 2018 dimethoate 0.0028 0.1445 not reported Greece Kasiotis et al., 2014 dimethoate 0.0091 0.0182 1.0 France Wiest et al., 2011 fenithrothion not reported 0.014 2.2 Spain Calatayud-Vernich et al., 2018 fludioxonil 0.0015 0.033 not reported Spain Vázquez et al., 2015 iprodione 0.0156 0.0195 1.0 France Wiest et al., 2011 lambda- cyhalothrin not reported 0.077 17.2 Estonia Raimets et al., 2020 permethrin not reported 0.034 5.0 Spain Hakme et al., 2017 permethrin 0.0015 0.0035 not reported Spain Vázquez et al., 2015 trifloxystrobin 0.0086 0.058 not reported Greece Kasiotis et al., 2014 trifloxystrobin 0.00024 0.01 40.0 United Kingdom David et al., 2016 trifloxystrobin 0.00025 0.046 5.6 Italy Tosi et al., 2018 trifloxystrobin 0.0015 0.0154 not reported Spain Vázquez et al., 2015 vinclozoline 0.0015 0.07 2.0 France Wiest et al., 2011 Acta agriculturae Slovenica, 117/2 – 2021 10 H. BAŠA ČESNIK 6 REFERENCES Alder L., Hill A., Holland P .T., Lantos J., Lee S.M., MacNeil J.D., O'Rangers J., van Zoonen P ., Ambrus A. (2000). Guidelines for single-laboratory validation of analytical methods for trace-level concentrations of organic chemicals, Principles and practices of method validation (ed.: A. Fajgelj, A. Am- brus). The Royal Society of Chemistry, pp. 179 – 252. Anastassiades M., Lehotay S. J., Štajnbaher D., Schenck F. J. (2003). Fast and easy multiresidue method employing acetonitrile extraction/partitioning and »dispersive solid- phase extraction« for the determination of pesticide resi- dues in produce. Journal of AOAC International, 86, 412- 431. https://doi.org/10.1093/jaoac/86.2.412 Cabrera de Oliveira R. C., Queiroz S. C. do N., da Luz C. F. P., Porto R. S., Rath S. (2016). Bee pollen as a bioindica- tor of environmental pesticide contamination. Chemos- phere, 163,525-534. https://doi.org/10.1016/j.chemos- phere.2016.08.022 Calatayud-Vernich P., Calatayud F., Simó E., Picó Y. (2018). Pesticide residues in honey bees, pollen and beeswax: As- sessing beehive exposure. Environmental Pollution, 241, 106-114. https://doi.org/10.1016/j.envpol.2018.05.062 Crenna E., Jolliet O., Collina E., Sala S., Fantke P. (2020). Characterizing honey bee exposure and effects from pes- ticides for chemical prioritization and life cycle assess- ment. Environment International, 138, 105642. https://doi. org/10.1016/j.envint.2020.105642 David A., Botías C., Abdul-sada A., Nicholls E., Rotheray E. L., Hill E. M., Goulson D. (2016). Widespread contamination of wildflower and bee-collected pollen with complex mix- tures of neonicotinoids and fungicides commonly applied to crops. Environment International, 88, 169-178. https:// doi.org/10.1016/j.envint.2015.12.011 Eckert J.E. (1933). The flight range of the honeybee. J o u r n al of Agricultural Research, 47, 257-285. García-Valcárcel A. I., Martínez-Ferrer M. T., Campos-Rivela J. M., Guil M. D. H. (2019). Analysis of pesticide residues in honeybee Ž(Apis mellifera L.) and in corbicular pol- len. Exposure in citrus orchard with an integrated pest management system. Talanta, 204, 153-162. https://doi. org/10.1016/j.talanta.2019.05.106 Hakme E., Lozano A., Gómez-Ramos M. M., Hernando M. D., Fernández-Alba A. R. (2017). Non-target evaluation of contaminants in honey bees and pollen samples by gas chromatography time-of-flight mass spectrometry. Chem- osphere, 184, 1310-1319. https://doi.org/10.1016/j.chemos- phere.2017.06.089. ISO 5725. (1994). Accuracy (trueness and precision) of meas- urement methods and results - Part2: Basic method for the determination of repeatability and reproducibility of a stand- ard measurement method, pp. 1-42. Kasiotis K. M., Anagnostopoulos C., Anastasiadou P ., Machera K. (2014). Pesticide residues in honeybees, honey and bee pollen by LC-MS/MS screening: Reported death incidents in honeybees. Sciience of the Total Environment, 485-486, 633-642. https://doi.org/10.1016/j.scitotenv.2014.03.042 Lehotay S. J. (2007). Determination of pesticide residues in foods by acetonitrile extraction and partitioning with magnesium sulfate: collaborative study. Journal of AOAC International, 90, 485-520. https://doi.org/10.1093/jaoac/90.2.485 Li Q.-Q., Wang K., Marcucci M. C., Sawaya A. C. H. F., Hu L., Xue X.-F., Wu L.-M. (2018). Nutrient-rich bee pollen: A treasure trove of active natural metabolites. Journal of Functional Foods, 49, 472-484. https://doi.org/10.1016/j. jff.2018.09.008 Li Y., Kelley R. A., Anderson T. D., Lydy M. J. (2015). Devel- opment and comparison of two multi-residue methods for the analysis of select pesticides in honey bees, pollen, and wax by gas chromatography-quadropole mass spectrom- etry. Talanta, 140, 81-87. https://doi.org/10.1016/j.talan- ta.2015.03.031 Lozano A., Rajski Ł., Uclés S., Belmonte-Valles N., Mezcua M., Fernández-Alba A. R. (2014). Evaluation of zirconium dioxide-based sorbents to decrease the matrix effect in avo- cado and almond multiresidue pesticide analysis followed by gas chromatography tandem mass spectrometry. Talan- ta, 118, 68-83. https://doi.org/10.1016/j.talanta.2013.09.053 Mullin C. A., Frazier M., Frazier J. L., Ashcraft S., Simonds R., vanEngelsdorp D., Pettis J. S. (2010). High levels of miti- cides and agrochemicals in North American Apiaries: im- plications for honey bee health. PLOS one, 5, e9754. https:// doi.org/10.1371/journal.pone.0009754 Nakajima Y., Tsuruma K., Shimazawa M., Mishima S., Hara H. (2009). Comparison of bee products based on assays of antioxidant capacities. BioMed Central, 9, 4. https://doi. org/10.1186/1472-6882-9-4 Raimets R., Bontšutšnaja A., Bartkevics V., Pugajeva I., Kaart T., Puusepp L., Pihlik P., Keres I., Viinalass H., Mänd M., Karise R. (2020). Pesticide residues in beehive matrices are dependent on collection time and matrix type but inde- pendent of proportion of foraged oilseed rape and agricul- tural land in foraging territory. Chemosphere, 238, 124555. https://doi.org/10.1016/j.chemosphere.2019.124555 Salles J., Cardinault N., Patrae V., Berry A., Giraudet C., Col- lin M.-L., Chanet A., Tagliaferri C., Denis P., Pouyet C., Boirie Y., Walrand S. (2014). Bee pollen improves muscle protein and energy metabolism in malnourished old rats through interfering with the Mtor signaling pathway and mitochondrial activity. Nutrients, 6, 5500-5516. https://doi. org/10.3390/nu6125500 SANTE/11813/2017. Guidance document on analytical qual- ity control and method validation procedures for pesticide residues analysis in food and feed. DG SANTE, European Comission, 2017. Stenerson K. K, (2018). Analysis of pesticides in turmeric pow- der by LC-MS/MS and GC-MS/MS after cleanup with a novel dual-layer SPE cartridge. Supelco Analytical Prod- ucts, Analytix reporter, 1, 2018. Thakur M., Nanda V . (2020). Composition and functionality of bee pollen: A review. Trends in Food Science & Technology, 98, 82-106. https://doi.org/10.1016/j.tifs.2020.02.001 Tosi S., Costa C., Vesco U., Quaglia G., Guido G. (2018). A 3-year survey of Italian honey bee-collected pollen reveals widespread contamination by agricultural pesticides. Sci- ence of the Total Environment, 615, 208-218. https://doi. org/10.1016/j.scitotenv.2017.09.226 Vázquez P . P ., Lozano A., Uclés S., Ramos M. M. G., Fernández- Acta agriculturae Slovenica, 117/2 – 2021 11 Pesticide residues in bee pollen - validation of the gas chromatography-mass spectrometry multiresidual ... of bee pollens from Slovenia Alba A. R. (2015). A sensitive and efficient method for rou- tine pesticide multiresidue analysis in bee pollen samples using gas and liquid chromatography coupled to tandem mass spectrometry. Journal of Chromatography A, 1426, 161-173. https://doi.org/10.1016/j.chroma.2015.11.081 Wang P.-C., Lee R.-J., Chen C.-Y., Chou C.-C., Lee M.-R. (2012). Determination of cyromazine and melamine in chiken eggs using quick, easy, cheap, effective, rugged and safe (QuEChERS) extraction coupled with liquid chroma- tography-tandem mass spectrometry. Analytica Chimica Acta, 752, 78-86. https://doi.org/10.1016/j.aca.2012.09.029 Wiest L., Buleté A., Giroud B., Fratta C., Amic S., Lambert O., Pouliquen H., Arnaudguilhem C. (2011). Multi-residue analysis of 80 environmental contaminants in honeys, hon- eybees and pollens by one extractuion procedure followed by liquid and gas chromatography coupled with mass spec- trometric detection. Journal of Chromatography A, 1218, 5743-5756. https://doi.org/10.1016/j.chroma.2011.06.079