3M™ Petrifilm Yeast and Mold Count Plate for the Enumeration of Yeasts and Molds in Dried Cannabis Flower: AOAC Official MethodSM 997.02

Abstract Background The 3M™ Petrifilm™ Yeast and Mold (YM) Count Plate is a sample-ready culture medium system that contains nutrients supplemented with antibiotics, a cold-water-soluble gelling agent, and an indicator system that facilitates yeast and mold enumeration. Objective The 3M Petrifilm YM Plate was validated for enumeration of yeast and mold in dried cannabis flower through the AOAC Emergency Response Validation process. Methods The performance of the 3M Petrifilm YM Plate was compared to that of Dichloran Rose Bengal Chloramphenicol (DRBC) agar. Matrix data were normalized by log10 transformation, and performance indicators included repeatability, difference of means, and inclusivity/exclusivity. Results These studies demonstrated that the 3M Petrifilm YM Plate method detects and enumerates yeasts and molds from cannabis flower at low, medium, and high contamination levels, and the average log counts at 20–25°C for 5 days were equivalent to the average log counts of the DRBC reference method. In strain studies, 59 out of 60 yeasts and molds produced typical colony morphology on 3M Petrifilm YM Plates. Of the nontarget bacterial strains tested, 38 out of 39 strains were not detected on 3M Petrifilm YM Plates. Conclusions The 3M Petrifilm YM Plate is a reliable method for the enumeration of live yeast and mold in dried cannabis flower. Highlights The 3M Petrifilm YM Plate allows for detection of yeast and mold within 5 days of incubation. The sample-ready plates can be incubated at 20–25°C and can be stacked up to 20 plates, thus providing user flexibility and saving incubator space.

Yeast and mold are widespread in nature, can be found in the air, water, soil, and vegetation, and can grow in a wide range of environmental conditions. Because cannabis plants are grown in both outdoor and indoor conditions, plants grown outdoors are exposed to wider ranges and larger populations of fungal species. However, improper watering, type of soil and fertilizer, and poor air circulation can increase the chance of mold growth in indoor environments. Human handling during harvest increases the risk of secondary contamination for both indoor and outdoor-grown cannabis. The final product could develop fungi, or their growth byproduct if humidity and temperature levels of drying and curing rooms are not carefully controlled. Total yeast and mold count (TYMC) is used as an indicator of the overall cleanliness of the product's life cycle, from growing, processing, handling, and to storage. Product with high TYMC can be detrimental to both consumers and cultivators.
While the majority of mold and yeast present in the environment are harmless, certain fungi cause spoilage and produce mycotoxins, a byproduct that is toxic to humans and animals. Several yeast and mold have been found to be prevalent in cannabis, including Cryptococcus, Mucor, Aspergillus fumigatus, Aspergillus niger, and Aspergillus flavus (1,2). Aspergillus species niger, flavus, and fumigatus are known for aflatoxin production, a type of dangerous mycotoxin that can be lethal (3). For this reason, regulations exist to limit the allowable TYMC counts for purposes of protecting consumer safety (4).
The 3M Petrifilm Yeast and Mold Count (YM) Plate is a sample-ready culture medium system that contains nutrients supplemented with antibiotics, a cold-water-soluble gelling agent, and an indicator system that facilitates yeast and mold enumeration (5).
The 3M Petrifilm YM Plate was previously shown to be comparable to the US Food and Drug Administration Bacteriological Analytical Manual (BAM) Chapter 18 reference method (6) for enumeration of yeasts and molds in foods [58 food types, naturally contaminated and laboratory spiked (pre-collaborative study, data not published)] and has Final Action status as Official Methods of Analysis SM 997.02 (7). The current matrix extension study compares the performance of the 3M Petrifilm YM Plate to Dichloran Rose Bengal Chloramphenicol (DRBC) agar for the enumeration of yeasts and molds in dried cannabis flower [9-tetrahydrocannabinol (THC) >0.3%].

Inclusivity/Exclusivity Study
An inclusivity/exclusivity study of the 3M Petrifilm YM Plate was previously performed using 31 strains of yeasts and molds and 40 strains of nontarget organisms in a pre-collaborative study (Q Laboratories, 3M data not published) as part of the AOAC Official Methods of Analysis 997.02 method validation (5). The ERV protocol required data for specific inclusivity and exclusivity organisms commonly found in cannabis matrixes. Additional strains to satisfy this requirement (29 inclusivity, 19 exclusivity) were tested by Q Laboratories.
Yeast organisms were propagated from a stock culture stored at -70 C to Potato Dextrose Broth and incubated at temperatures optimal for growth. Following incubation, yeast organisms were diluted to 100Â the LOD of the 3M Petrifilm YM Plates. Mold organisms were propagated from a stock culture stored at -70 C to Sabouraud Dextrose Agar and incubated for 5-7 days at 30 6 1 C. Following incubation, mold spores were harvested for inclusivity testing by washing cultures with Butterfield's Phosphate Buffered Dilution Water. The mold wash was then diluted to 100Â the LOD of the 3M Petrifilm YM Plates.
Exclusivity organisms were propagated from a stock culture stored at À70 C to trypticase soy agar with 5% sheep blood (SBA) and incubated at conditions optimal for growth. Following incubation, exclusivity organisms were transferred to the nonselective Brain Heart Infusion broth and incubated at conditions optimal for growth. Exclusivity cultures were analyzed undiluted.
All organisms were randomized, blind-coded, and plated onto 3M Petrifilm YM Plates as indicated in the instructions for use. Plates were incubated at 20-25 C and examined after 3 and 5 days. Final results were recorded after 5 days. Colonies were determined to be positive or negative based on the product instructions.

Matrix Study
Cannabis test materials were prepared by Steadfast Analytical from an inventory of retained samples from its Michigan licensed grower, patient, and caregiver customers. Samples were combined to produce batch materials of a low level (<1000 cfu/ g), a medium level (1000-10 000 cfu/g), and a high level (10 000-100 000 cfu/g). Batches were manually mixed in an aseptic manner until homogeneous. For each contamination level, five replicate test portions (10 g) were quantified by spread-plating aliquots of diluted test portions onto DRBC agar plates. Table 3 summarizes the average cfu/g of yeast and mold for each contamination level that was provided to laboratories for analysis in the study.
Individual 10 g test portions from each contamination level were placed in sterile filter Whirl-Pak bags. Five bagged test portions from each of the three contamination levels were selected for each candidate method participating in the ERV project. Test portions were assigned an identification tag in Michigan's Marijuana Regulatory Agency seed-to-sale system for distribution and tracking. This served to blind-code the contamination level of the test portions. The test portions were also assigned random sample numbers for reporting results to AOAC.
Personnel from each of the participating independent laboratories were responsible for picking up and transporting the test portions to their laboratories on Monday, December 7, 2020. Participating laboratories were instructed to analyze samples on Tuesday, December 8, 2020 following the user guides provided with the candidate methods. In addition to the candidate methods, all test portions were enumerated using DRBC agar as described in DRBC Reference section. North Coast Testing Laboratories of Michigan conducted the matrix evaluation for the 3M Petrifilm YM Plate method.

Candidate Method
All analyses were performed using paired test portions. Test portions were prepared for analysis as described in the 3M Petrifilm YM Plate method. Ten-gram portions were homogenized in 90 mL 0.1% peptone water (PW). Tenfold dilutions were made by transferring 10 mL into 90 mL PW and shaking 25 times in a one-foot arc within 7 s to ensure homogeneity. A 1 mL aliquot from each dilution (10 À1 , 10 À2 , 10 À3 , and 10 À4 ) was plated in duplicate on 3M Petrifilm YM Plates. Plates were examined at 3 and 5 days, and final colonies were recorded at 5 days. Plates containing counts between 10 and 150 were used to determine the final results. If mainly yeasts are present, plates with 150 colonies are usually countable. When substantial amounts of mold were present and a more accurate count was obtained on the next dilution, the upper countable limit was lowered per the guidance in the 3M Petrifilm YM Plate method and BAM Chapter 18.

DRBC Reference Media
Paired test portions, prepared following the candidate method dilution protocol, were confirmed by spread-plating aliquots of each dilution onto DRBC agar plates per the recommendation in the SMPR (9). From the initial dilution of the sample, 10 g (homogenized in 90 mL PW) 1.0 mL was spread-plated across two DRBC agar plates (0.5 mL on each plate) in triplicate (six total DRBC agar plates). Additionally, 0.1 mL of the 10 À1 , 10 À2 , and 10 À3 dilutions was plated in triplicate on DRBC to obtain the 10 À2 , 10 À3 , and 10 À4 dilutions, respectively. The agar plates were allowed to dry and were then incubated at 25 6 1 C for 5-7 days before enumeration. Mold appeared as flat or fuzzy, spreading colonies with the natural pigmentation of the sporing structures, and yeast appeared as pink, smooth, raised colonies on DRBC agar plates (10). Plates containing counts between 10 and 150 colonies were enumerated as described in BAM Ch. 18.

A. Principle
The method uses culture plates of dry medium supplemented with antibiotics, dye to enhance visualization of growth, and cold-water-soluble gelling agent. Undiluted or diluted suspensions are added to plates at a rate of 1 mL/plate. Suspension is spread over a 30 cm 2 growth area. Gelling agent is allowed to solidify, plates are incubated, and yeasts and molds are counted.  To prevent exposure to moisture, do not refrigerate opened pouches. Store resealed pouches in a cool, dry place for no longer than four weeks. It is recommended that resealed pouches of 3M Petrifilm YM Plates be stored in a freezer (see product instructions) if the laboratory temperature exceeds 25 C (77 F) and/or the laboratory is located in a region where the relative humidity exceeds 50% (with the exception of air-conditioned premises). After use, plates contain viable yeast and/or mold cultures. Autoclave used plates 15 min at 121 C prior to discarding.

D. Preparation of Test Suspension
Aseptically prepare 1:10 or greater dilution of food samples with sterile diluent. Blend or stomach 2 min and plate. Prepare additional dilutions as required. For dried cannabis flower (THC >0.3%), weigh out 10 g of sample from test portion into a sterile stomacher bag and dilute with 90 mL sterile 0.1% PW. Shake 25 times to homogenize. Prepare additional dilutions as required.

E. Analysis
Place Petrifilm Yeast and Mold Count Plate on a flat surface. Lift the top film, hold the pipet perpendicular to plate, and carefully inoculate 1 mL test suspension onto the center of the film base. Place top film down onto inoculum.
Lift the plastic spreader using the circular handle. Align the center of the spreader with the approximate center of the plate. Distribute suspension evenly using gentle downward pressure on the center of the spreader. Do not slide the spreader across the film. Remove the spreader and leave the plate undisturbed 1 min to let the gel solidify.
Place the plates in the incubator in a horizontal position, clear side up, in stacks not exceeding 20 units. Incubate the plates 5 days at 20-25 C.
Count the plates promptly after incubation period. Yeasts appear as blue-green or off-white in color and form small defined colonies. Mold colonies are usually blue but may also assume their natural pigmentation (e.g., black, yellow, green). They tend to be larger and more diffuse than yeast colonies.
To calculate yeast and mold count, multiply total number of yeast and mold colonies/plate (or average number of colonies/ plate, if counting duplicate plates of same dilution) by the appropriate dilution factor. When counting colonies on duplicate plates of consecutive dilutions, calculate the mean number of colonies for each dilution before determining average yeast and mold count.
Estimated counts can be made on plates with >150 colonies and should be reported as estimated counts. In making such counts, determine average count/1 cm 2 and multiply by 30 (circular growth area is 30 cm 2 ).
High numbers of yeast colonies may cause the entire growth area to turn blue. High numbers of mold colonies may cause growth area to turn blue, black, yellow, etc. When this occurs, do not make estimated counts, but further dilute and plate test suspension to obtain a more accurate count.

Inclusivity/Exclusivity
Results from the inclusivity and exclusivity testing conducted at Q Laboratories have been combined with the results from the original 3M Petrifilm YM Plate AOAC Official Methods of Analysis validation and are presented in Table 1 (inclusivity) and Table 2 (exclusivity). All inclusivity yeasts and molds tested with the exception of Aspergillus vitricola American Type Culture Collection (ATCC V R ) 14557 (Manassas, VA) showed typical colony morphology and thus were considered "positive" on 3M Petrifilm YM Plates. Aspergillus vitricola ATCC 14557 is an osmophilic organism that did not recover on Potato Dextrose Agar as well as the 3M Petrifilm YM Plate. This is associated with the growth characteristics of this organism, which requires a very high glucose concentration. Thirty-eight out of the 39 exclusivity strains were not detected on the 3M Petrifilm YM Plates. One exclusivity strain, Morganella morganii ATCC 25829, had partial breakthrough growth on 3M Petrifilm YM Plates (15 colonies) and breakthrough growth on DRBC agar.

Matrix Study
Paired statistical analysis was conducted for each contamination level comparing the candidate method result to cfu/g obtained on the DRBC agar plates. For each test portion, results were logarithmically (log 10 ) transformed using the equation cfu/g þ 0.1, according to the Least Cost Formulations, Ltd (2020) Paired Method Analysis for Micro Testing version 1.2 (Virginia Beach, VA). After transformation, replicate test portion results for each contamination level for each method were averaged, and the difference of means between methods with standard error and 95% and 90% confidence intervals were determined. Repeatability was also calculated for each contamination level for each method. The matrix study data are presented in Table 4.

Discussion
The 3M Petrifilm YM Count Plate method was incubated at 20-25 C and compared to AOAC SMPR for Viable TYMC Enumeration at 5 days for the detection and enumeration of yeast and mold from dried cannabis flower. Naturally contaminated dried cannabis flower samples were tested at low, medium, and high levels. The log counts from the 3M Petrifilm YM Count Plate method were compared with log counts from DRBC agar. The 90% and 95% confidence intervals indicated there were no significant differences in detection or enumeration between the 3M Petrifilm YM Count Plate method and the DRBC agar for dried cannabis flower samples at the low, medium, and high contamination levels.
In the inclusivity strain studies, 59 out of 60 yeast and mold strains were detected and had appropriate colony morphology on 3M Petrifilm YM Plates. In the exclusivity strain studies, 38 out of 39 nontarget strains tested were not detected on 3M Petrifilm YM Count Plate.

Conclusions
These studies have demonstrated that the 3M Petrifilm YM Count Plate method is an accurate, specific, and repeatable method that detects and enumerates yeast and mold in 5 days at 20-25 C from dried cannabis flower. It is recommended that the 3M Petrifilm YM Plate, AOAC 997.02, be granted a matrix extension for the detection and enumeration of yeasts and molds in dried cannabis flower.

Conflict of Interest
3M Company is the method developer. All authors from 3M Company are salaried employees of the company. North Coast Testing Laboratories and Q Laboratories were contracted as independent laboratories to conduct the study per AOAC guidelines and received payment from 3M Company.