Considerations for the Analysis of Pesticides and Mycotoxins in Hemp, Part 1

Using various dispersive solid phase extraction (dSPE) mixes to clean-up hemp and cannabis extracts for the analysis of mycotoxins

Undoubtedly, the analysis of low-level contaminants, such as pesticides and mycotoxins, in hemp and cannabis matrices (plant material and any derived product) is a challenging undertaking.  Cannabis and hemp are particularly complex matrices that contain high concentrations of metabolites, such as cannabinoids, terpenes, and flavonoids, which interfere with the detection and quantitation of analytes of interest at concentrations in the order of parts per billion. We frequently receive inquiries from customers at cannabis testing labs who are interested in improving their analytical workflows by adjusting their sample prep approaches. The main challenge of sample prep in multiresidue analysis is that analytes display a broad range of physical chemical properties, and in many cases some of them are very similar to the interferences analysts want to remove. This means that selective extraction and/or cleaning is not always feasible or easily attainable. In this blog series, we would like to share with our readers some insights on what to expect when trying different strategies to improve the detection of contaminants in hemp and cannabis. In this first blog, we will discuss the effect of various dispersive solid phase extraction (dSPE) mixes on the recovery and detection of five mycotoxins (aflatoxins G1, G2, B1, B2 (AG1, AG2, AB1, and AB2) and ochratoxin A (OA)) in hemp extracts.

Mycotoxins are metabolites produced by organisms of the kingdom fungi (e.g. mold) that can cause severe illness to humans and animals. Cannabis and hemp can be affected by mold growth or contamination, depending on cultivation, storage, and processing conditions. Quantitative analysis of mycotoxins in cannabis is currently required in various states. For instance, in California, Michigan, and Nevada, the total concentration of AG1, AG2, AB1 and AB2 should not exceed 20 ppb, and the concentration of OA should not be higher than 20 ppb. This requires very low limits of quantitation (LOQ) which means finding approaches to obtain cleaner samples are always desired.

Figure 1. Mycotoxin recoveries after cleaning hemp extracts with various dSPE mixes (Restek catalog numbers displayed). Sample preparation: pulverized hemp was spiked at 20 ppb by adding 20 µL of a 1 ppm solution containing all the mycotoxins. Subsequently, 5 mL of acetonitrile acidified with 0.2% formic acid was added. Samples were vigorously vortexed for 15 min at 2500 rpm. After centrifugation, 1 mL of supernatant was transferred to 2 mL vials with pre-weighed dSPE mixes. Samples were vortexed and centrifuged once again. 450 µL of extract was mixed with 150 µL of water. Recoveries were estimated by comparing analyte responses measured in extracts from samples spiked before extraction vs. responses corresponding to extracts from blank hemp samples spiked after extraction.

Table 1. dSPE mixes evaluated in the clean-up of hemp extracts for mycotoxins analysis.

Catalogue #	mg of sorbent
	MgSO4	PSA	GCB	C18
26123	150	50	50
26124	150	50
26125	150	50		50
26215	150	25
26216	150	25		25
26217	150	25	2.5
26218	150	25	7.5
26219	150	50	50	50
26242	150			50
26243	150	50	7.5	50

Figure 1 shows the recoveries obtained for the five targeted mycotoxins after using 10 different dSPE mixes, and Table 1 lists the composition of each of the dSPE mixes evaluated. As can be seen, OA exhibited recoveries below 20% in extracts cleaned with mixes containing 50 mg of PSA and recoveries between 20 and 60% in cases where 25 mg of PSA were used. As expected, the highest recovery value for OA was found when samples were cleaned with the dSPE mix 26242, which does not contain PSA. For AB2 and AB1, recoveries were below 40% when high amounts of GCB (50 mg) were employed as clean-up sorbents, and for AG2 and AG1, recoveries were above 60% in all the cases. Highest recoveries for all the aflatoxins were found when mixes 26124, 26218, and 26242 were used.

In addition to the assessment of mycotoxin recoveries, we evaluated whether there were differences in the ionization effects (also known as absolute matrix effects) for each analyte when different dSPE mixes were used. A full explanation on what matrix effects are and how to assess them can be found in one of my previous blogs (here). Briefly, the assessment of absolute matrix effects can give us relevant information on whether the presence of co-extracted components in a sample extract is affecting the ionization of our analytes of interest. As can be seen in Figure 2, there is some level of suppression for the majority of aflatoxins (values were below 100%), whereas OA displays some ionization enhancement in all the treatments tested. Although values within 70 and 130% are considered acceptable, ionization effects can be corrected by using appropriate internal standards and/or matrix matched calibrations. It is important to emphasize that ionization effects significantly depend on the LC gradient conditions as they determine the resolution among analytes and interferences. For this experiment, we used a 20 min gradient which led to the elution of AG2, AG1, AB2, AB1, and OA at 3.4, 3.5, 3.9, 4.2, and 6.0 min, respectively. Column, mobile phases and other LC conditions are listed in Table 2.

Figure 2. Absolute matrix effects assessed in hemp extracts cleaned with 10 dSPE mixes (Restek catalog numbers displayed). Matrix effects were estimated using the equation: Matrix effects = Analyte response (post-spiked in blank extract)/Analyte response (spiked in neat solvent))*100. Blank hemp samples were prepared as described in Figure 1 caption. 414 µL of blank hemp extract was spiked with 18 µL of a 100 ppb mycotoxins standard, and then mixed with 150 µL of water.

Table 2.  LC parameters used for this set of experiments.

Column	Raptor ARC-18 2.7 µm, 100 mm x 2.1 mm (Restek Cat.# 9314A12)
Guard Column	Raptor ARC-18 EXP Guard Column Cartridge 2.7 µm, 5 x 2.1 mm (cat.# 9314A0252)
Mobile Phase A	Water, 2 mM ammonium formate, 0.1% formic acid
Mobile Phase B	Methanol, 2 mM ammonium formate, 0.1% formic acid
Flow	0.4 mL/min
Column T	40 °C
Autosampler T	10 °C
Inj. Volume	3 μL

Summary

Overall, the use of amounts higher than 25 mg of PSA and 50 mg of GCB may lead to significant losses of some of the mycotoxins currently regulated by several states in cannabis products. However, results may vary depending on the sample type and extraction conditions, so it is always advisable to test sample prep conditions in-house.  At the experimental parameters chosen for this evaluation, aflatoxins showed a certain level of suppression and OA displayed ionization enhancement. Alternative sample prep strategies and method adjustments may be needed to get optimum analytical conditions for this test.

Please stay tuned for upcoming blogs on this challenging, yet interesting topic!