The Challenges of Microbial Analysis in Cannabis Matrices: Plating vs Molecular Techniques
by Yan Shipelskiy, PhD, lab director at Green Scientific Labs
Complete the form below and we will email you a PDF version of "The Challenges of Microbial Analysis in Cannabis Matrices: Plating vs Molecular Techniques"
As more states push for cannabis legalization and establish or refine cannabis safety testing, there’s an increased focus on the detection and enumeration of microbial content in cannabis flower and derivative products. Microbial analysis is particularly important since it safeguards public health, is a significant cause of harvest batch failures, and cannabis recalls often involve microbial contamination. Generally speaking, there are two approaches for detecting and quantifying microbial contamination in cannabis samples.
The first approach involves identifying the microbes potentially present in samples by culturing them. This culturing methodology has been developed and utilized throughout the 20th century and can take between 24 to over 120 hours to produce results, depending on the organism. The total yeast and mold count (TYMC) analysis can be particularly time-consuming, but chromogenic media that contain dyes can expedite this analysis from ~120 hours to 48-72 hours. Furthermore, certain organisms are particularly challenging to detect utilizing culturing methodology, including Shiga-toxin producing E. coli (STEC) and Aspergillus species. Overall, this approach is widely accepted for quantifying microbial content in samples.
The second approach involves isolating genetic material (either DNA or RNA) from potentially present microbial organisms. This molecular methodology has been developed and utilized since the 1980s and relies on amplification of genetic material utilizing the polymerase chain reaction (PCR). This approach does not detect microbes, per se, but rather the unique genetic material from these organisms. Certain organisms are unculturable and can only be currently identified via molecular techniques. STEC and Aspergillus species, which present a challenge for culturing methodologies, are easily detected via molecular techniques. This technique typically yields results within around 24 hours. Overall, this approach is well suited for detecting microbial content in samples.
Cannabis safety testing action limits are set by individual states and can vary for medicinal or recreational products within a state. Cannabis samples that exceed these limits fail the microbial analysis and need to be remediated to be released for distribution. These limits are typically provided as either “detected or not detected in one gram” or “< X colony forming units per gram (CFU/g)”, with X ranging from 1 to 100,000 CFU/g.
Analyses that fall under the “detected or not detected in one gram” are qualitative and seek to determine if there is any presence of a particular species in a sample, regardless of the amount. Molecular techniques are commonly used and are ideal for this type of evaluation. Analyses that fall under the “< X CFU/g” are quantitative and inherently imply culturing methodologies, in particular plating.
Since regulations are often written with “CFU/g” language, the use of molecular PCR techniques or any other techniques (including non-plating culturing methodology) require a conversion from whatever output the method produces to CFU/g. For molecular techniques, results are provided in quantification cycle (Cq) count. These values indicate that a reaction crossed a particular signal threshold, at which point a result is considered positive. The lower the Cq value, the more genetic material is in the sample, which is correlated to microbial load using Cq to CFU/g conversions. There are different conversion equations for flower, extracts, infused products and other matrices and for every organism evaluated.
Additionally, the different phases of microbial culture growth cycle can affect the amount of genetic material that’s recovered. Molecular techniques often underestimate the microbial concentration during the lag phase of a microbial culture. Unfortunately, proficiency tests and microbial standards are often in this lag phase of growth when they’re evaluated during validations. However, during the decline stage of growth, molecular techniques overestimate the microbial concentration because of the presence of free DNA from dead cells. The presence of free DNA presents an additional challenge following remediation of microbe-contaminated cannabis products, as the treatment kills the cells but the genetic material is still present. While some manufacturers provide products (nucleases) to control for free DNA in samples, this step increases the challenges of validation and the cost/time associated with adding this free DNA removal step to the microbial analysis workflow. Because of these issues, trying to utilize molecular techniques for quantitative (CFU/g) analyses makes validations challenging.
On the other hand, the use of “< 1 CFU/g” action limits for certain organisms creates a unique challenge for culturing methodologies. When using plating methodologies for example, achieving a true result of “< 1 CFU/g” is technically challenging, if not impossible. Since cannabis samples must be diluted in buffer or media for analysis, recovering enough buffer or media with this dilution for quantitative culturing analysis is not realistic. States often interpret these results as a qualitative “detected or not detected in one gram” action limit, which is not ideal from a regulatory compliance standpoint because the “< 1 CFU/g” regulatory limits are explicitly quantitative in nature. This presents both unrealistic technical requirements along with compliance pitfalls for labs operating in states where this is the case. Florida, Hawaii, Maryland (and by extension New Jersey), Oklahoma, and South Dakota unfortunately have these regulatory requirements. Regulators in these states and other states deciding on action limits for microbial testing should avoid setting anything less than a “< 10 CFU/g” action limit, and to convert any “< 1 CFU/g” action limits to “detected or not detected in one gram.”
Microbial analysis is critical for safeguarding public health and instilling trust in cannabis safety testing and the cannabis industry in general. While no one questions the wisdom of producing food and agricultural products despite frequent recalls, the legal cannabis industry is under greater scrutiny to provide products that satisfy sensible regulatory standards. It’s important for regulators to set realistic limits and to ensure that proper techniques are being used for microbial analyses. It’s important for labs to provide accurate and reliable results to growers/processors, regulators, and the public.
Despite significant progress for microbial enumeration utilizing molecular PCR techniques, these approaches often diverge from results provided by traditional, culture-based techniques. Given that “CFU/g” is inherently culture-based regulatory language, safety testing labs and the industry in general would provide more consistent and accurate results for microbial enumeration with culture-based methodologies. In the case of “detected or not detected in one gram” qualitative analyses, molecular methodology is more effective, or sometimes necessary for accurate identification. Some regulators have already moved towards qualitative, and regulators, labs, and the cannabis industry at large would provide more accurate and consistent results following this approach.