Conflating Viability with Risk
August 30th, 2016
A critique against utilizing DNA based methods for detecting food born illnesses is that DNA based methods will fail more material and leave the grower hopeless to employ sterilization techniques. This is known as the Live-Dead problem. Sterilization kills the microbes before the DNA and therefore dead organisms will always conflate your read out.
First, this is simply a false statement. There are multiple techniques published in peer reviewed journals that showcase how to perform Live-Dead PCR. A simple Google search picks many of these up with “Viability-PCR”.
Before we criticize PCR we should be aware that this is scientific hypocrisy considering 99% of microbes cannot be cultured and the real problem is an “everything appears dead” problem in the world of working with culture. That being said, there are ways to use genomics to measure viability of something placed into the right “universal” culture medium. The major caveat being that there is no such universal culture medium and all results will only be a reflection of what can grow in a given medium. All petri-film and culture based platforms are plagued with this problem. qPCR provides the ability to measure culture based viability and culture independent surveys making it a far more sensitive and specific tool.
Live-Dead PCR assays fall into two general categories.
1)Intercalating dyes (or membrane impermeable dyes) that cannot traverse a cell membrane but if the cell membrane is broken they crosslink DNA making it non-amplifiable with PCR. As a result dead DNA cannot contribute to a qPCR signal. PMA is a common dye used for this and it requires UV light to cross link the DNA.
2)ITS qPCR with and without a short incubation. If the organisms are living they should replicate their genomes in 20 minutes to 4 hours. Since the ITS regions are high copy number in the genomes, this signal is easy to pick up even if the cells are partially replicating. This is often referred to as Before-After qPCR. Some labs have increased the sensitivity of this approach by going after the ITS preRNA sequences as those are at higher copy number than gDNA copies of ITS regions.
In addition to there being many published methods on Live-Dead PCR, the EPA recommends qPCR with and without a few hours of growth to capture Live-Dead PCR results.
There are downsides to Live-Dead qPCR. In method #1, the dyes capacity to traverse the cell wall is very organism dependent as cell membranes are often loaded with cellular transport receptors. As a result, there has yet to be a paper demonstrating how to perform this technique on an entire genera like “Total Yeast and Mold”. Most examples in the literature appear to be highly optimized for a single organism. These dyes are expensive, usually teratogens and hazardous. This does not fit a green chemistry philosophy.
There are downsides to Before-After culture qPCR (method #2). First, only 1% of organisms are believed to be culturable. Considering the industry is anchored in 3M plating, they are already pregnant with this blind spot. The other complaint is that it doubles the qPCR reagents and qPCR is already more expensive than 3M.
This is not entirely true. Most labs that employ this technique, only perform the 2nd PCR on samples that give a “Before” qPCR result beyond a concerning threshold. Since qPCR can be done in a few hours, there is a before culture qPCR answer delivered before the after samples are done incubating. As a result, the after culture qPCR only needs to be performed on 10-20% of the samples that have a before qPCR signal of any significance.
We have opted to take the ITS gDNA approach with PathogINDICAtor. The preRNA ITS approach is very clever but it requires working with RNA and Reverse Transcriptases to deliver 5-10X more sensitivity. This is great idea for low to single CFU sensitive assays for sepsis but in Cannabis testing, all single CFU assays are incubated for 24 hours anyway due to sub sampling.
Whats is sub sampling? Since one cannot fit a whole gram of cannabis into a qPCR well or a petri dish without diluting it first often only a sub sample is graduated to a detector. This 10X subsampling of the sample requires an pre-incubation before the subsampling to make up for the limited amount of material that will make it into a testing vial. These dynamics may not be in play in other markets like Sepsis. All other tests in Cannabis microbial testing are 10,000 or 100,000 CFU/gram tests that don’t require enhanced sensitivity at the expense of handling RNA and more expensive enzyme systems.
We have published our methods in open source journals for cannabis microbial testing. To date PathogINDICAtor is the only microbial testing assay in the cannabis space that is peer reviewed and validated on a cannabis matrix. Here is a good review of membrane impermeable dyes and PreRNA ITS based methods: Dead or Alive MVT.
Given the peer reviewed counter factual to this Live-Dead claim, lets stop and ask if this makes sense to pursue? Here is a video presenting many challenges to this simplistic position.
Most microbes are not a risk due to their viability. They are a risk because they synthesize pasteurization resistant toxins that are difficult to detect with ELISA’s alone. ELISA’s can be tuned to work in a given environment but struggle when the background compounds change as they do with the chemotypic diversity seen in cannabis. ELISA’s are notorious for finding THC in the water supplies and THC false positives triggered by baby soap.
In addition to the unreliable results for ELISA’s in complex backgrounds, there are some toxin producing genes we have found on cannabis in which there are no available ELISA’s (paxilline). In addition to the toxins, some microorganisms produce spores that are pasteurization resistant and these spores may require very long culturing times to form a detectable colony. C.botulism and Aspergillus make pasteurization resistant spores that don’t incubate in 70 hours.
Below is an a warning from the Denver Department of Environmental Health regarding C.botulism in Cannabis oils. If you want to HIDE your risks, you may sterilize cannabis so it passes a 3M plate but this is advised against by the DEH as it masks the very risks they are trying to measure with culture based systems undetectable. qPCR systems on the other hand have both a before and after culture measurement and can augment this picture with faster and more specific data.
Specificity is more important than viability
We have also conducted research demonstrating that most Cannabis in Massachusetts in 2016 had Live DNA on it. This was presented at ICRS2016 where we performed Next Generation Sequencing of TYM and TAC primers on before and after DNA from 3M and Biomeriuex Culture platforms (in preparation). This 32 sample study demonstrated that in all cases Cqs gained signal (lower Cq) with culture.
The largest discrepancy in the study was revealed to be due to Total Yeast and Mold culture based systems (3M and Biomerieux) also culturing bacteria. In many cases these were beneficial bacteria that are commonly used to fight fungal infections (serenade). Bacillus and Tricoderma are the most frequently found benign organisms and are the core ingredients for most beneficial microbe products.
We cannot have a TYM testing system that falsely fails samples for mold as it will induce more fungicide use like Eagle 20. Eagle 20 converts to hydrogen cyanide when smoked. This is particularly problematic if the best organic practices for thwarting off fungal infections (using beneficial microbes like bacillus) are being selected against due to the testing systems faith in a 100 year old non-specific technology. qPCR, on the other hand, has no problems discerning bacteria from fungi and will afford the use of IPM (Integrative Pest Management) and fungicide free growth practices.
Culture based systems have no way to adapt to this idea and believe its imperative in terms of specificity. qPCR can address it with intelligent primer design.
In summary, most microbes are beneficial. The key to accurate microbial safety testing is specificity not viability. qPCR is the only platform that can deliver both requirements. Failure to to drive for specificity can result on 15% TYM failures in California with over 50% Pesticide/Fungicide failures (Wurzer et al). The presented 15% TYM failure was performed on 3M TYM Petri-Films. Its unclear what percent of these fails were due to beneficial microbes. The 50% pesticidal fails are shocking and one of the most popular fail is the fungicide known as myclobutanil or Eagle-20. This has also been surveyed in WA/OR with equally concerning results due the fact that pesticides concentrate in most cannabinoid extraction procedures. To what degree low specificity 3M testing is driving the industry toward fungicide use is unknown but current testing regulations are favoring this use over beneficial microbes as microbial testing is mandated in more states than pesticide/fungicide testing and the lowest cost 3M Petri-Film cannot discern bacteria from mold.