Growth Media Used in Total Yeast and Mold Testing Can Significantly Affect Microbial Diversity, Study Finds

The growth media used in plating-based total yeast and mold count (TYMC) tests can dramatically impact the diversity of microbes that grow, according to a new investigation from scientists at Medicinal Genomics, a specialist cannabis testing and genetics screening firm.

Carried out as part of an AOAC International Emergency Response Validation (ERV) of cannabis testing methods in the state of Michigan, the results of the Medicinal Genomics investigation suggest that regulators should consider complimenting tests done in certain growth media with additional species-specific testing to ensure no microbial risks are going undetected.

PDA with chloramphenicol media demonstrated highest diversity

Three different common growth media were used in Medicinal Genomics investigation:

• Potato dextrose agar (PDA)
• PDA with chloramphenicol
• Dichloran Rose Bengal with chloramphenicol (DRBC)

Samples of dried cannabis flower were homogenized, prepared, and plated on each of the growth media at a range of different dilutions. Whole genome sequencing on the microbes collected from a sample from the same homogenized cannabis flower was also done as a comparison.

The scientists observed that PDA with chloramphenicol demonstrated the highest microbial diversity and the highest concordance with the results seen from whole genome sequencing. The DRBC plating method showed the least diversity. The Medicinal Genomics team say that this was down to DRBC containing three different antibiotic selection agents – dichloran, rose bengal, and chloramphenicol – which limits yeast and mold growth alongside its intended purpose of limiting bacterial contamination.

“There’s several forms of [antibiotic] selection going on there. And that selection, while it’s designed to knock out bacteria, it is also known that some of those antibiotics inhibit the growth of fungi like Aspergillus. In fact, it’s documented in the literature that Aspergillus is sensitive to chloramphenicol,” Kevin McKernan, founder and CSO of Medicinal Genomics, told Analytical Cannabis.

“So we raised the alarm bells that this is not a good idea. If you’re going to put selection in here, you are going to select against the most dangerous pathogen in cannabis. So we should pick a different selection – if you’re going to use selection at all. PCR [polymerase chain reaction testing] doesn’t have any selection, we just amplify all DNA. So this is really a problem of culture [-based methods].”

To further demonstrate the differential effects of each growth media, the Medicinal Genomics team tested colonies from each plate using TYMC and total aerobic count (TAC) quantitative polymerase chain reaction (qPCR) assays, in order to correlate these colony forming units (CFUs) with different fungi and bacterial species.

This qPCR testing works by amplifying identifying segments of DNA present in a testing sample that are characteristic of known microbiological species, which can then be identified and quantified. PCR testing techniques are also sometimes used in place of traditional culture plating methods in cannabis testing. Debate between proponents of the two techniques over which is best suited for such use has been ongoing for many years.

The Medicinal Genomics team found that the DRBC media limited growth of fungal species and undercounted the number of fungi species present by five-fold, compared to the qPCR results. Conversely, PDA without any antibiotic selection overcounted the CFUs due to bacterial contamination on the plate.

Other issues with plate-count methods

The Medicinal Genomics team say that their study has also highlighted two more general limitations of culture plating methods.

Since it is difficult to count the number of colonies when there is an excess of around 100 colonies on a plate, testing laboratory technicians normally need to perform multiple dilutions of the sample. This means that analysts might be multiplying their figures by 10, 100, or even 1000-fold to back estimate the total CFU count. This poor dynamic range introduces a significant source for error given that even a single missed colony could significantly impact the final CFU count and potentially cause a sample with a high CFU count to appear as though it passes under regulatory limits.

With so many colonies on these plates, there are also problems with these colonies overlapping and becoming difficult for even the most skilled laboratory technicians to identify.

“It’s very hard to get speciation with the colony forming units [using plate methods], because you just don’t know what you’re looking at unless you sequence them or use PCR. So we did that study, and we picked about 45 colonies off of all those plates and we did whole genome sequencing on them. So there’s absolutely no question as to what they are. And that can be important to use because you notice in some of the plates there that the density of the colonies is so tight that they’re overlapping a little bit on the images,” McKernan explained.

“That’s one of the challenges with plating in general is that you can’t plate more than really 100 colonies on a lawn, otherwise they start colliding with one another and it gets harder and harder to call them accurately. That’s something that PCR doesn’t struggle with,” he added.

What does this mean for regulation?

For colony visualization, some states preferentially use DRBC growth media for their microbiological testing of cannabis programs. But in light of their new investigation results, which showed that DRBC can greatly inhibit fungal growth, Medicinal Genomics recommends that regulators in these states should look to bring in additional species-specific testing that can complement this plate testing method.

“I think where the cannabis industry has struggled the most on this is that we’ve got a lot of regulators that are, thankfully, showing up and trying to bring some order to the chaos. And we welcome that. I think the challenge is that it’s an unfunded mandate for many of them,” McKernan said.

“Suddenly, there’s a legal market dropped on their lap, they have no funding and have to figure out how to regulate it. And so they’re in a scramble and they’re trying to do the best they can and so they lift what they know from the food industry into cannabis. And there’s a huge difference here; your lettuce and your spinach doesn’t make 20 percent antibiotics by weight,” he adds, referring to the antibiotic effects discovered in a number of important cannabinoids, including CBD and CBG.

“When you use culture [-based methods], you’re actually running a cannabinoid antimicrobial experiment on top of measuring the safety of the plant. And that’s where these things go off the rails trying to get the PCR systems to correlate with the plating systems.”

The new investigation also provides some broader context to the current debates being had over cannabis testing methodologies. Since 2020, AOAC International has been working on an accelerated program to evaluate test kits used in TYMC testing for cannabis flower. Earlier this year, regulators in Michigan announced that they would be pausing the sole use of qPCR methods for the quantification of total yeast and mold, in light of preliminary results from the AOAC emergency response validation (ERV) that had been seen by the regulators. As of July 1, all qPCR assays on yeast and mold must be accompanied by either a traditional culture plating method or a method that has been validated by the AOAC, according to the Michigan Marijuana Regulatory Agency (MRA).

However, given the issues with DRBC undercounting and the plating methods’ general poorer diversity and dynamic range, as compared to qPCR in this Medicinal Genomics investigation, perhaps questions around the sole use of these plating methods without additional species-specific testing or duplicate qPCR testing might be soon to follow.

“What’s happening right now is a lot of folks will do qPCR on day one, so they can get answers and quarantine things that they’re concerned about and then maybe confirm with plating. So it doesn’t have to be either-or; there are places for both of them where they can help and can confirm some concerns,” added McKernan.

Update: This story was updated on July 16, 2021, to include up-to-date guidance on yeast and mold testing from the Michigan Marijuana Regulatory Agency (MRA).