Talking Biotech 030 -- Glyphosate in Bread Milk and Wine?

I want to see science serve a useful purpose to improve the standard of living for all people.

Hey, why is anyone fighting food advance?

A very small percentage of the world's population is fortunate enough to have the luxury of turning down food.

We've arranged a society based on science and technology in which Because nobody understands anything about science and technology.

You can't build a peaceful world on empty stomachs and human misery.

You're listening to Talking Biotech, a weekly podcast illuminating issues in agricultural and medical biotechnology. Your questions and concerns are addressed using a science-based approach with the goal of driving discovery to application with communication. Now here's your host, Dr. Kevin Folta.

Borlaug, Sagan, Benjillette.

What a great way to kick off a podcast. I still— people say, you know, you have that beginning every week and it seems to be a little long and, you know, I gotta fast forward through it because I've heard it 30 times now. And, you know, that's great, but, you know, I kind of like it. I think that, you know, when, when, when we drink from the well, we've got to remember the people who schlepped— wait a minute, what, what's that saying? When we— oh man, screwed that up. When you, when you take a drink of water, you got to remember the folks that dug the well, because we all stand on the shoulders of giants in science. And, and it's always a nice little remember to pay a little homage to them. So today we're going to talk about some changes in the anti-genetic engineering movement, and I kind of don't want to depart from the sciences and get into the politics too much. I tend to like to stay away from that, especially recently, but today we're going to venture that direction a touch. Now over the last few years, we've really seen a shift in how we're seeing technology discussed. The folks on the internet who really don't appreciate the technology, they really have shied away from the claims that the crops are somehow damaging and dangerous. You don't hear that as much anymore. The current target is to go after the chemicals that are used in association with the crop strategy. So they're going after glyphosate, and it's one of the herbicides that's used in herbicide-tolerant crops. The central ingredient in Roundup. So the problem that we have is, is that science is extremely good at detecting very little. We know that we can identify small amounts of compounds when we find them, and just because you can find it doesn't mean it's dangerous. However, a cottage industry has emerged to detect these chemicals for a fee, and and allegedly in any matrix you submit. So you can submit wine, beer, whatever, and they'll tell you— breast milk— and they'll tell you how much glyphosate they detect. And it seems really weird to me because I'm familiar with analytical chemistry. I have great reverence for those that do it, and it's not so simple. Every matrix is different, and every way in which you do sample prep is different, and it's very rigorous and requires a tremendous attention to the details of the assay and reproducibility. And I think that a lot of these labs that are doing this are a little bit suspect in their results because they don't demand replication. They, they see someone waving some money and they're glad to take it.

That's my feeling.

So the problem is, is that these labs give reports saying that they've detected something. and doing it in parts per billion or less. Now, to many people, parts per billion means time to freak out. This is dangerous. It's bad stuff. But parts per billion are seconds in many, many decades. It's really, really small amounts that they're looking at. So when you don't have good replication, and lots of replications to measure not only if something is being faithfully measured the same between samples, also within one sample. It's important to have many replicates and appropriate controls. Today we'll talk about that with our second guest in the second part of the show, with Dr. Thomas Calhoun, who is an expert in measuring very little from complex matrices. When we see these reports, we see that okay, well it's in something like urine, and that I can believe. That's where it's supposed to be. That's where these compounds are are supposed to be excreted. We know that from the pharmacogenetics pharmacokinetics that this is how it moves through the body. So when you're finding it in the urine, it's not a big surprise. But we have to think about what does it really mean? What are the amounts tell us? And we know that there's potentially residues on raw crops like raw soybeans. You may find it there. That's not a surprise. We also know that you find a lot of application in residential as well as municipal applications of this herbicide. So it doesn't surprise me that you would detect it in things like urine or potentially in other, in other places, but it's not a big surprise. The big difference is Is that what is risk? What is risk? And risk is the hazard times exposure. So risk of driving, you can say, well, driving— or let's say risk of flying for me personally. Or let's say even better, risk of shark attack. The hazard is the shark, and they're hungry, and they're looking for something, low-hanging fruit in the water. Uh, which might be you. The other side of that is exposure. So right now, sitting in my office in Gainesville, Florida, I'm not really exposing myself to a shark attack. So even though the hazard exists, there's no exposure. That's what you're seeing with the alleged shark of glyphosate. It is something that has a mild hazard. I mean, just like any chemical, just like water, everything has a threshold where it can become dangerous. The question is, how much are we exposed to it? And even when they do say that they detect it, it's in parts per billion, so very small. And I can kind of believe that. I think that's reasonable. That's what we might expect. So I don't think that there's a whole lot to worry about here. What's really interesting is that so many people are willing to trust the results online, even though there's no statistical treatment. If they were real, that's probably, you know, fine. It does— that still wouldn't be something that would alarm me. But I also do have some severe questions, or very strong questions, about the methods and about how they actually did the measurements, because there's no statistics to provide us with an idea of how variant the assay is. It's not good experimental design, and that makes sense because this wasn't designed by experiments. It was designed by activists that are looking for a chemical, and it's easy to find. But today's podcast starts with someone who didn't find it, Dr. Shelley McGuire. She's a specialist in lactation from Washington State University and interested in the good things and the bad things that are found in baby's first food. She uses her expertise to design protocols to detect glyphosate in breast milk, which is really not easy to work with, but she did it. And The big part is standardization of the experiment. How do you set up what they call a method to be able to detect? And it may take many months just to get the method down. And that's where Dr. Thomas Calhoun, part 2— I know Thomas really well. He's a good guy. I was on his PhD committee. Now he's a faculty member in the Horticultural Sciences, or the Environmental Horticulture Department. Environmental Horticulture. At University of Florida. And if I had to ask one person about the limitations of the technology and how well it works, it'd be Thomas. So we talk about the claims, we talk about glyphosate in wine and that whole set of claims. So here we go. What I say is really simple. This is a very big— this is really cool. This is the way to think about this. Now they're claiming to have found about 1 part per billion 1 part per billion in wine, and they're freaked out from what has been described even by the most conservative factions as a probable carcinogen. 1 part per billion of a probable carcinogen, and yet we know there are 15 million parts per billion of a known carcinogen. Ethanol. So the, the alcohol in wine. So 15 million parts per billion of a known carcinogen versus 1 part per billion of a probable carcinogen. And you really see that these folks may be barking up the wrong tree. So here we go with today's podcast, Dr. Shelly McGuire. So today on Talking Biotech, we're going to take on a really important topic. Lately there's been a lot of discussion about the chemical glyphosate, which is being used as a— primarily as an herbicide on genetically engineered crops, but also in many applications in residential use and in municipal use. It's been much maligned in the media and also on the internet, and one of the people who's been brought into this has been brought in to really assess some of the claims that have been made, and we're We're very fortunate to be able to share some time with her today. So today we'll have a discussion with Dr. Shelley McGuire. Shelley is currently a professor at the— at Washington State. Shelley is currently a professor at Washington State University in the School of Biological Sciences. And welcome to the Talking Biotech podcast, Shelley.

Kevin, thanks so much for having me.

Now this is really cool. It's really exciting to be able to talk to someone who does what you do because I kind of have an idea, but I don't know that much about it. What is your research about, and what are your typical projects in lactation studies?

Well, my research, I have to say, is I think some of the coolest research you can do, because I study human milk. And I study the process of making that milk and what goes into how women alter milk composition by their diets and how variation in milk composition can impact both maternal and infant health. So I just think it's the neatest area of research that anybody could possibly be in. And I've been doing this research since I started graduate school in 1986. It has been my entire career, and I've been at Washington State University studying human milk and lactation for nearly 20 years.

Wow, so 30 years of studying, studying human lactation. That's, it's, um, it's pretty cool. So we're actually kind of contemporaries then. Because I have the same timeline, I kind of get a sense of how you feel because I've been in the same science box too in a lot of ways. But what are the biggest things that have changed in that 30 years in the way that you study your subject?

Well, that's a good question. You know, when I started studying human milk back in the 1980s, we were still trying to characterize the basic nutrients in human milk. For example, my master's thesis was focused on the selenium content of human milk and how variation in that content might influence infant health. And from that work, actually, formula companies started adding selenium to the formulas to make it more like human milk. And that's what we were doing in the 1980s. Since then, we've become very interested in a lot of other factors in human milk. My lab has done a lot of work in fatty acid composition, for example, trans fatty acids. And more recently, we've become super interested in the fact that human milk contains microbes. And so, in fact, human milk is Mother Nature's first probiotic food. And we didn't know that until recently because we just developed the technology to be able to look at these microbes in milk. So there's been a sea change in what we know about what's in human milk over the last 30 years.

And in terms of thinking about the environmental impacts, and, you know, we think about the compounds that are in our environment. And we are literally bombarded with chemistry. I mean, between flame retardants and food, things in our food which are naturally occurring as well as artificially installed, you know, there's a lot of worry and a lot of concern about what can be passed from mother to baby because of baby's vulnerable state. And how has that been changing, or the philosophies around that really been changing with your time in the discipline?

Well, I have to say this is something I'm really interested in because I'm interested in the good stuff in milk and potentially the bad stuff in milk because so many times whatever a mother is exposed to, whatever she eats, etc., ends up in her milk. And we should be concerned about that. So, for example, like I had mentioned, we've done studies in the past looking at trans fatty acids in milk. And some of our work actually Mm-hmm. Helped the U.S. government determine that trans fatty acids should be limited in the diet because they do show up in human milk and they can actually impact milk fat. So we're very interested in what's human— what's in human milk. And I have been interested, like I said, in the good, bad, and the ugly. I don't shy away from anything. If it's in human milk, I want to know how much is in there. And if it's something that shouldn't be there, then I think we need to raise that red And are there— We've looked at a variety of things.

And are there examples from either your work or other work where people have identified deleterious compounds or negative compounds present in human milk?

Sure, absolutely. There's a lot of interest, for example, in mercury and other heavy metals. We know that these contaminants can get into human milk. We also know that women who lose more weight when they're breastfeeding have milk with higher composition of these potentially deleterious compounds. And so, yeah, we know these things are in human milk and we're keeping an eye on them. This is a really important topic of conversation.

And it also seems to be one that is, at least in the recent study that you've published, ties in with a very common environmental chemical, this stuff called glyphosate, which is an herbicide used in many different forms, whether it's just to kill weeds or kill grass or whatever. But it's also used in genetically engineered crops and has become kind of the focal point for a lot of controversy lately, deserved or undeserved is still being unraveled. And the allegations that really you're addressing maybe stem from—

well, why don't I let you explain what was really the genesis of Well, interestingly, the genesis of us looking at whether we could even detect glyphosate in human milk came from emails that I got. Now, I get emails, probably like you do, frequently from the public. And the first email that alerted me to the fact that people were even concerned about glyphosate in human milk came from somebody who claimed to be a Welsh farmer. And he had— he was concerned about glyphosate in his cows. Had some concerns about whether glyphosate was in human milk, and he sent me an email because I'm an expert in human milk, asking me if I knew anything about whether glyphosate was in human milk. And I did like any research professor would do. I went to PubMed and I did some searches, and lo and behold, there were no published studies on this topic. And that's what got me interested in this from the very start. It was an email that came from A farmer.

It's interesting to even think about from a scientific standpoint, as somebody from my side, as someone familiar with agriculture and farming, to plausibly find glyphosate in human milk, you would have to be eating massive amounts of soy, because we know what the residue amounts are. And you'd have to either be eating mountains of this stuff and then having it sequester specifically in the tissues that produce lactation, or there would have to be some sort of concentrating mechanism because this stuff we know just runs right through you. So it'd have to be something truly extraordinary that hadn't been previously posited, and, uh, and certainly seemed like a long shot. So it certainly was a good place to establish the opportunity to test.

Yeah, absolutely. It's one of the simplest things we can do in science. We ask the question As to whether we can detect something, and then we do the study to see if we can detect it. Seriously, this is one of the simplest studies I've ever done in my entire career. But I would say that even though the animal literature all suggests that it would not be in human milk, or it would be very low because it's a water-soluble compound, you know, I never take that for granted. The human mammary gland, the breast, does amazing things. And because glyphosate looks very much like an amino acid, as you know, you know, I thought this was a worthy thing to look at. You never know when the mammary gland is going to do something odd like sequester a compound or whatever in the milk. So it was worth looking at.

I agree 100%.

I didn't go into it, you know, I didn't go into it just Assuming it wasn't going to be there. I really went into this study to determine if it was.

Oh yeah, it's testing a hypothesis, right? You know, it's the hypothesis that glyphosate is detectable in human milk from incidental exposure.

It turns out that after I got the email from this Welsh farmer, I found out that there were some online reports that glyphosate was extremely high in milk. And so there was significant public concern, and I like to do research that is related to public interest. I think that's the best kind of research that we can do. And so with that public interest and these reports, which were somewhat questionable because the reports had never been published in peer-reviewed literature, and it was very, very I was very unsure as to the methods that were used to analyze the milk. The folks that were publishing this online were not forthcoming with a lot of details. So you put all that together, and it was sort of the perfect storm that really sort of pushed me into doing this research and really wanting to find the answer for the public, if not just for myself.

Well, that's a really good point, because if I always think that there are people who maybe don't have the purest of intentions that maybe are doing experiments, or maybe they're doing experiments but not doing them correctly. So a lot of the work on glyphosate has been done using a commercially available kit where it's been published, where people have published data that no one has ever reproduced or really do have some curious attributes about them, like improper controls, detection below a standard curve, below the minimal detectable amount, basically reading noise but calling it a positive signal. And it— there is a lot of false positive out there, so it doesn't necessarily have to be deceptive, but people are publishing information that just isn't scientifically vetted, and as you mentioned, not present in peer review. And then also not reproduced by anybody. And so it takes somebody who is an expert in this to kind of step in and really sort it out for us. So what were the ways that— what are the methods that you use to detect this compound?

Well, we— this is where we actually teamed up with the chemists at Monsanto Company. We do a lot of analyses in our lab, but we do not do the sort of analyses that are needed to detect detect this compound at this extremely low level in what we expected to be a low level in human milk, which, by the way, is the most complex matrix that you'll ever analyze. It's an extremely difficult matrix to work in, and even the simplest compounds that you try to measure in human milk, you often have to do it differently. So instead of us When we didn't have the expertise, it's not what we do, trying to develop an analysis that was sensitive and specific to glyphosate in human milk. We did team up with the chemists at Monsanto who are world experts in this. They have some fantastic analytical people, and they worked for nearly a year to validate an assay for human milk that we were confident in, that we knew would not result in false positives or false negatives. We needed to know that it was accurate. So it was what's called an LC tandem mass spec method with very, very good sensitivity and specificity for glyphosate in the particular matrix of interest, which was human milk. And we also worked with an independent lab called Covance, which is located in Wisconsin, and they were able to then take the analysis, all the parameters of the analysis that the Monsanto chemists had worked up, they were able to take that procedure, revalidate it in their independent lab, and then run our samples in parallel with the Monsanto chemists so that we could get independent verification that what Monsanto chemists were finding could be reproduced in an independent lab.

And were all those samples between Covance And Monsanto, were all those double-blinded so people weren't sure what they were testing, or—

Well, they all knew that they had human milk samples, and we were very aware of the potential conflict of interest or the perceived conflict of interest. Obviously Monsanto has a vested interest in glyphosate, and I will say that they were fantastic people to work with, extremely professional. But we understood the public perception that we were dealing with and the potential So we actually sent duplicate samples to both Monsanto and the Covance chemists. We communicated exclusively with the Covance chemists, and they with us. They did not report their findings to Monsanto. They reported them directly to us without any intervening by Monsanto. And both labs found exactly the same thing, which is that they could not detect any glyphosate. glyphosate in any human milk sample. So we have very, very high confidence in our results.

And did Monsanto know that you were sending a sample to Covance?

Yes, they did.

Okay.

Yes, they did. We were very transparent about all of our methods.

And did you also detect or attempt to detect the metabolites of glyphosate?

We did. We looked at AMPA, which is a metabolite of glyphosate. And it was not detectable in milk as well, so it's the same as glyphosate. And we also collected urine samples from these lactating women, and both glyphosate and AMPA were detectable in almost all the urine samples, but at extremely low levels.

And where would it come from typically?

That's an interesting question, because the dogma is that somehow the AMPA is being metabolized Or being produced as a byproduct of glyphosate metabolism in the human body. However, it's also as likely that it's being produced in plants and other— mostly plants, or that it's also found in detergents. So it could be, and we can't tell from our study, but it could be that that AMPA was actually coming from—

Plants.

other sources. It's possible.

Well, I look at your numbers in the paper, and they're all pretty low. I mean, everything is in the neighborhood or below 1 part per billion, which is thousands— well, millions of times below biological thresholds of where there'd be any kinds of consequence. But maybe even in keeping with residential use or other routine exposures, Or even perhaps residual on food. It's possible.

It's possible. It's possible. You know, I would love— we looked in our study because we had a nice split of, let's say, women who lived on farms and ranches. We live in a very agricultural area versus women who lived in town. We didn't see any differences between the two because we thought, you know, maybe women who are living, you know, in the wheat fields and in these fields where maybe, you know, these products are being used might have higher levels, but we didn't see any differences. We also looked at women who consume mostly organic food, and we live in a community where that's a relatively common thing to do, versus women who just purchase conventional foods, and again, we didn't find a difference. Now, what I'd love to do to really look at this issue is to specifically recruit Let's say women who only eat organic and only eat conventional, or women who apply glyphosate versus women who never have any exposure except for maybe the random exposure at the sidewalk level. But those studies haven't been done. And quite frankly, my hypothesis, even if we were to do those studies, is that levels would still be extremely low and not affected by these—

Yeah, I think that's really the way to focus, is just because we could do science or do a series of experiments, does that mean that we should do them? And I think certainly that, you know, with all the, all the concern and all the interest and the fact that breast milk is something that we should be careful with, that your study here was completely warranted. It's just the levels are so low and so biologically irrelevant Which, you know, at the edge of detection, and which means basically likely not even there, that what would you do next? I mean, what's really the next experiment, and is it worth taking on?

Well, what I would do next, if I had funding to do it, I would do a clinical randomized study where I randomized women to consume conventional Versus organic food, I would do a crossover study, so each woman would see both, each treatment. I would have the treatments long enough, several weeks, maybe a month, maybe a month long. And I would look for presence of glyphosate and AMPA in not only milk, but also in blood and in urine, and I would also look in the infant's urine, because that's really the only way to tell. You do a clinical— Intervention study, and that's how you really find— that's good scientific method. That's what I would do next. But having said that, I have a lot of other studies that I'd like to do. And as you said, our levels were— we had undetectable levels. So is it really worth the public dollar to follow up? I don't know. I think there might be other studies that are probably more important to do.

Well, and, you know, and I don't mean to be creating divisiveness because I don't normally like to do that, but I would almost be willing to bet that if you checked levels of copper, copper ion, you know, heavy metal, that you would find meaningful levels that were being transmitted through other production methods such as organic where you do use a lot of it. And of course, if people are washing their food properly, maybe that's not necessarily the case, but I know how much is used on, on those products and what the packing house's runoff looks like. And, you know, there's copper there, and that would be something that I think would have more biological relevance in breast milk because of its ionic form. And I wouldn't worry about that, but it brings more concern than I would glyphosate.

You know, it's interesting you say that, because if I did that study, I would look at a lot of factors, because, you know, there's so many claims out here You hear all the time that organic foods are healthier, that they're more nutritious, that they're this, that they're that, that they're as affordable. The list goes on and on. And I would test all those things. I would actually do a much larger panel than just looking at glyphosate. I would also look at economics. I would look at a lot of other factors and see, can we find any sort of benefit, biological benefit, economic benefit, etc.? from consuming organic versus conventional foods. And quite frankly, that larger study I think would be worth doing.

So, Shelley, you really just popped up on, you know, at least I've been familiar with your work in the last, you know, year or so. And was it really related to this particular line of research that really kind of got you into the public eyeball, launched into a bigger way, and maybe even on a little bit of a hot seat Yeah, great question.

So I have been for years a spokesperson for the American Society for Nutrition. So I've actually— I've been out there. I mean, if you— I talk with the New York Times, and I talk with the LA Times, and I talk with Glamour magazine. But it's usually about nutrition, and about breastfeeding, and about lactation. So yes, this is my first experience in this whole crazy world glyphosate and GMO and organics. And it's been a very, very interesting introduction to the politics and the science of this whole area. And I will tell you, I've been shocked at what I've seen. And I came into this without a huge opinion about this topic, but this experience has really, really opened my eyes to what's going on, how the public is getting information and misinformation. And now I I feel really— I think it's a really important area for scientists to speak up in and to get the word out and get the science out and to explain to the public how we actually test our hypotheses and what this is all about. I think it's critical, and I have no intention of backing down at this point.

Is this your first real collaboration with Monsanto?

Absolutely. It is mine. Now, my husband, Mark McGuire, did his PhD in a lab that studied DST, so he had some connections with Monsanto early on, but I've had absolutely no connection with Monsanto until this study.

And we did— I guess if we don't say it, we sound kind of silly. That, you know, we really didn't talk about a lot of the other issues that are swirling around this and around the communication space because of this study and its results. And, uh, but it's really great for another time. I think that what's really good about this study is that it was done rigorously and peer-reviewed and now sets a benchmark that others can compare to. And what's nice about this work and the work that preceded it is that it does give complete descriptions of how you did it and what your samples were like and who were your samples and how you did the detection. So it really is a rigorous benchmark, and I think that's sometimes one of the most important things in science is to have something to work against.

I agree with you, and I cannot overstate the importance of peer review and of using good science And adequate methods to analyze anything in a human sample, particularly human milk.

Your hypothesis was that you would detect glyphosate in human milk, but that hypothesis was not supported by the data. How hard was it to publish a paper that basically showed nothing's there?

Well, the paper is being published in the American Journal for Clinical Nutrition, which is the top impact factor journal for the field of nutrition. I appreciate that they're publishing what we call in the field negative data, but they're a very good journal and their editorial staff is fantastic. It went through rigorous peer review, and so I guess in the end it wasn't, you know, good, good studies get published.

Well, Shelley McGuire, thank you so much for spending the time with me today on Talking Biotech.

I really appreciate you sharing your story and helping clarify this Thanks for having me, and, and thanks, Kevin, for all you do as well.

Hello, Talking Biotechers. Thank you for listening to the Talking Biotech podcast. We can tell from the numbers that lots of you are downloading this podcast, and that's just super. However, for innovation to finally reach the people that need it, We need your help. We're talking about technologies and how they can serve people, maybe even defusing some fear in the process. So, here's your homework. Tell at least 2 people to download and listen to this podcast. Carve Talking Biotech Podcast into the bathroom stall at Chipotle. Or maybe get a Talking Biotech Podcast tattoo. Someday it'll serve as a reminder of when we were in the dark days, when science was discounted for beliefs, and technology was kept away from the people that really could use it. Help us spread the word that technology has good things to offer, and can serve the issues that we all care about.

Well, in this part of Talking Biotech, I want to re-explore the plethora of new results that are coming out indicating that glyphosate appears to be everywhere. And it's really important that we look at these very carefully because they're making claims that are scaring people in a lot of ways. But at the same time, if these are legitimate claims, they're claims that we should take very seriously. So we need to separate what's real and what's not. It's important if it's real because we should know. If it's not real, we should know that too. So instead of me trying to work through analytical chemistry techniques and talk to you about that in a clear way, I brought on an expert. I'm sitting here with Dr. Thomas Colquhoun, spelled exactly as it sounds, C-O-L-Q-U-H-O-U-N. Thomas is a friend. He's a fellow faculty member at University of Florida, but he's also an expert in analytical chemistry techniques. And we had talked about glyphosate and its detection. And so, Thomas, one of the things that maybe you could— we could start with is how sensitive can you detect molecules, things like glyphosate?

Glyphosate can be easily detected at parts per billion, and with different techniques, one could see getting down to parts per trillion. Okay, so you can be extremely sensitive and probably very reliable, but you're talking about LC-MS and the quadrupole or One of the most sensitive techniques that we have is an LC-MS technique, and using a mass spectrometer called a triple quad.

Triple quad?

Triple quad. It is one of the most sensitive techniques we have known to man to detect soluble compounds.

And you can detect, I mean, very small amounts, like you say parts per trillion. What does that mean in like real person language? If we had to think of parts per trillion, is there a good analogy?

Yeah, a very, very small amount. Trace levels, levels that would be at something like nanogram per gram, nanogram per liter.

Yeah, and so that's the hard part about this is that it's so abstract to talk about nanograms per liter. It's like maybe it makes more sense to think of it like 1 second In the last— since, uh, the birth of Christ, you know. I mean, it— and we're getting into those kind of levels, and I don't know exactly what they are. I should know, but that's maybe something for us to understand. But when we talk about these things showing up in beer and wine and everything else, how should someone who just wants to— who reads that website, who maybe is being asked by friends, by relatives— what's the best way for us to explain to them what it is and what it isn't?

Well, are you referring to the website of Moms Across America?

Yes.

Okay. First and foremost, this is a completely different analytical technique. This is called an ELISA. So here we're using a plate wells. We're using protein features and protein interactions with chemical compounds. What I was previously talking about with a mass spectrometer is a detection device that is specifically designed to ionize and to understand a mass-to-charge ratio of certain chemical compounds. So you can see these are 2 very different techniques. Now with ELISA, ELISA is well known for false positives. ELISA techniques have many working parts and, and different aspects to them to get to an overall feature of concentration. That feature of concentration is then read by a colorimetric test on a— what, usually a plate reader. A mass spectrometer, using the right techniques and with the right standards, can be much more reliable and robust. Now, usually using a triple quad, techniques like LC-MS, we have that compound we're looking for as a standard, whether that compound is a labeled compound or whatnot. But we always have it as a standard. And what we generally do is we confirm what we're seeing is very similar to what that chemical authentic standard is, and we can run that authentic standard in a dilution series. So we can calculate back an absolute quantitative amount. That is very difficult to do in an ELISA, if not impossible. An ELISA will get you a relative quantitation.

Yeah, so let me see if I can clarify this a little bit. This is, this is tough for, I think, for a lot of folks who might be listening who understand science and are interested in science, but maybe not Yeah. Get in the nuts and bolts. The, the MS, the LC-MS, the mass spec readings, is kind of like if you were to pick a person off of the planet, and then you had an ability to tell that that person, based on standards that were well derived from many other standards, that this was a Taiwanese carpenter who specialized in framing work. Okay, and then you— because you had many other Taiwanese carpenters who were experts in framing, and you could tell them from Finnish carpenters very differently. Not Finnish like from Finland Finnish, meaning like finishing, you know, putting on that kind of stuff. Anyway, so you could tell the difference between— because you knew what the standard was, you could compare it to Taiwanese carpenters who were experts in framing. But if you use an ELISA test, it has a tendency to pick up anything with a hammer, any of those, any person who worked with a hammer. So you might get some shoemakers and you might get some hammery guys, you know, who knows what you're gonna get, right? So that's kind of— and because it may pick up on a unique feature of that thing you're trying to detect, that isn't necessarily correct.

As a rough analogy, I will agree with you.

Well, I need, I need more time to get That's a good analogy. You know, these things don't come as quick as they used to. But you've— so, all right, so basically you got these 2 methods, one which is very specific and extremely sensitive based on standards and careful quantitation, LC-MS. On the other side, you have ELISA, which is based upon rough interactions with features of a molecule with some sort of a target and a detection method that can lose linearity all over the place, that has a lot of very careful caveats. Now, what are they using in the Moms Across America analysis?

An ELISA.

An ELISA.

And, you know, an ELISA— so back in about '08, we were using mass spec LC-MSs to actually quantitate glyphosate, and it is a very expensive process. To give you an example, a triple quad mass spec will run about $250,000. An ELISA plate and analysis you can get online for glyphosate for $130. So it becomes impractical if you're going to be looking at so many different samples to be going to a company or going to a lab to use this large-scale equipment, analytical equipment, for all this, this different work. So at that point, you need to have a more rough analysis, a cheaper analysis A beginning analysis for that work, and that's where the ELISA came into play. So an ELISA, usually you can get a sample done for anywhere from $10 to $15 using an ELISA versus anywhere around $300 to $400 per sample using an LC-MS. And that's if all of those methods have already been developed and you have the people there And the solvents and the chemical standards and all of that.

Yeah, and the method development takes a tremendous amount of time and expertise, right?

Correct. It all depends on the chemical that you're trying to develop a method for, but in general, in my lab, we take anywhere from 3 months to 6 months for method development.

And so when we're talking about this issue of detecting compounds, so you saw the issue with— let's talk about the wine study. Where, well, I use the word study loosely.

Yeah.

Let's talk about the samples that were submitted to a company. When you're testing wine and that you give somebody 10 samples and you're able to detect something in those 10 samples, even in the samples which they said were from an organic farm, and this is important. First, we're finding parts per trillion, you know, like, but in the high hundreds of parts per trillion. Which, all right, we— and, and that was with an ELISA again, the wine?

Yep.

Okay.

Yep.

And so if you got that result where you knew that the organic one, which isn't supposed to use glyphosate, and you're detecting it, what would that tell you as a scientist?

It would tell me that I'm at baseline, and I would have to see my chemical standard at that same baseline giving me a robust signal to even believe in that signal coming from the experimental. It becomes rather tricky when you are quantitating chemicals at a baseline level. First thing I would ask for is reproducibility. Can I see that over and over and over again? And in this wine experimentation, it seems to me these are just single data points. I would have a hard time settling on one data point with any technology that I'm using for measuring. Glyphosate is one of these chemicals that can act like an acid or a base, so it becomes rather tricky when you are even at your extraction process from matrices. So we haven't even gotten into the extraction part of this, but extraction is the very first thing that you're going to need to worry about when you are looking at chemicals in different foodstuffs and different liquids and whatever. And so your extraction methodology becomes very important, and you need to be able to clearly communicate that extraction methodology. And in this example of the wine study, there is no communication of of how that extraction proceeded, whether it was just a simple polar water extraction. You know, I, I don't know.

Yeah.

And these things become very important when you have a complex situation such as extraction. You have your experiment, you have your normalization, your validations. You know, you have to follow this, this, this train of science to be able to have reproducibility in, in your experimentation.

And I think that's really important too, because when you only have one single data point, what it says is that somebody extracted it from a plant sample, or from a wine sample. You extract it, and then you test it. What we do as scientists is, we will do 3 or 4 separate extractions, and then we'll measure each one of those extractions 3 or 4 times.

At least.

Yep.

Because we're not only looking at different replications for the extraction leading to some sort of final product, which there could be variation in, but we're also asking, is our measurement reliable? That when we measure the same sample 3 different times, do we get the same answer? Which we usually don't, because with such fine measurement, you're looking at differences that will be over a range.

Yep.

And so we need to have some sort of statistical handle as to how reliable is our assay, how reliable is our extraction, and we don't get that with one data point.

And what is the standard deviation? Standard deviation is, is just fine, but you have to be a standard deviation away from your lowest detection limit or underneath that to be able to even have confidence that you're above your limit of detection. And usually in ELISA a test, you'll run an empirical set of— an empirical experiment to see what your limit of detection is, and on top of that, what your running limit of detection is, because there's variation between plates, there's variation between samples, and you need to understand all of those numbers to have confidence in what you're seeing. And none of that seems to have happened here.

And one of my other concerns with these particular data is the, the fact that you're seeing it in a sample where it shouldn't be. You're saying in an org— wine from an organic field is showing a detection of glyphosate. Now to the folks online, they say, oh, it's contamination, it's clear evidence that glyphosate is blowing in and it's from everywhere. How do we think about that as scientists?

Well, I would like to see a negative control. I would like to see a positive of control. You know, even, even a water control, you know, using just bottled water that had been purified or whatnot in your extraction all the way through your experimentation, and showing that when you expect with what you're handling is going to be negative for your result, is negative for your result. That can give you confidence in the rest of your results.

ELISA is very frequently can cross-react with other compounds. And so when you're looking at a single compound like glyphosate and you're testing it in something like wine, which is a complex matrix of many chemicals and volatiles, there's a high probability that something that could mimic glyphosate to that assay could react just from wine without any glyphosate being there. And without a negative control, how do we know? And I think that's one, one interpretation. The other interpretation I would go to second is whoever made that organic wine is using glyphosate in their field. You know, those are my two, my first two places I go as a scientist. Occam's razor says the most mundane explanation. I don't go to what must be contamination.

And parsimony should exist here and should exist always. Um, but we also have to remember these were samples that were provided by an anonymous donor. Um, there was no control for the samples, and that, that must be addressed. You know, we, we have to have control for our samples. We have to have reproducibility. We have to have replicates. And, and, and none of that, that scientific standard was, was upheld here in this study.

And that's important too, because the, the people who have commissioned this work and funded it and everything else are people who would love to find glyphosate everywhere.

So, and, and this might sound funny, but if you remember a year, year and a half back when there was a paper published on Bigfoot's DNA, where did those samples come from? And when we actually got a hold of those samples, what were those samples actually? They were not Bigfoot's DNA.

No. And yeah, when the, when the Bigfoot Exploratory Club is looking for a new kit, they're Kickstartering gear, they might want to go find some Bigfoot fur. And I think that's what you're looking for here, is a molecular Bigfoot. And maybe there's a— and so here's the next big question: is this something that is, to you as someone with an analytical chemistry background and someone who really understands this, Is this something that you as a scientist would even think about trying to reproduce in the interest of setting the record straight?

I've given a lot of thought to this, and scientists all over have done this. It's amazing to me that we can have a scientific literature body and support a certain standpoint, support a certain viewpoint, and then a study like this comes across which has very little, if none, scientific merit. And then we have to, as scientists, spend our resources and our time to once again show that if you take the proper controlled scientific approach, that you find opposite results. It's a little perplexing, and it's a little counterintuitive. And I'm sure somebody out there with the appropriate analytical techniques will do one of these studies, and we'll see what their data says. Very much like what happened with breast milk and glyphosate concentrations in urine. And we've done the careful studies as scientists, and we find no detectable levels.

And then the reaction to that is, well, they must all be bought and paid for by big agriculture.

You know, we can't believe anything they say. Because it's all big ag paying them off, right?

I mean, that's what we hear.

Yeah, yeah. I constantly hear, you're a shill, you're paid off, you're this, you're that. There's just no reason to even entertain that discussion.

Well, maybe what we need is to start a foundation that does nothing but assess the scientific veracity of claims, and whether that has to do with autism, Climate, glyphosate, whatever. There should be some way for us to take that on. The big problem is, why not spend it on real science that can actually help people? And why not send it out— spend it on science that can cure disease and enrich farming and feed more people? Instead, we have to play cleanup, you know, screwing around with trying to get this activist group in California who wants nothing more than to get rid of transgenic genetic engineering. Now we have to play defense, and it just would be such better policy to be able to do this and do the right science.

I completely agree.

So in the wine study, what about that disclaimer?

Well, this is fairly, fairly interesting. The disclaimer states the laboratory who did the experiment is not a human clinical diagnosis laboratory and makes no warranty to the fitness of this data for such purposes. That's fairly telling. I haven't published a manuscript and put a disclaimer as to, I give no authority to the fitness of my data. It is exactly opposite of that.

Right.

And the data supports it by using replication.

And that's exactly right, that we put our butts on the line and say, this is— these are data we're going to stand by because we understand, we vouch for their veracity. You know, that we have done this right and we are going to stand by this. And that's the big difference between peer-reviewed publication and a flyer or a website. And I think that's really maybe the best way to describe it. for us to communicate that. Because the part that bothers me about this is people see this and I get bombarded with emails and phone calls from friends and relatives. Kevin, can I drink wine? And I go, hell yeah. You know, first of all, the alcohol is going to kill you way before the trace parts per trillion of an herbicide that doesn't have a physiological relevance at those levels.

Correct.

We know alcohol does. Because that's why we drink it. It makes you feel funny. So long story short, we're putting our efforts in the wrong place. So let's spend a few minutes talking about this wine trial. I hate to call it a study, but these data. First, the title alone is maybe a little bit smacking of an agenda. And what can you tell us about that title and the sample size and what they're claiming via this particular data set.

Well, this struck me right off the beginning. The title is Widespread Contamination of Glyphosate Weed Killer in California Wine. The wines that were tested here in this trial, if you will, were 10 individual wines from the North Coast region of California. It is weird to use 10 different wines and call this widespread of anything, being that there are over 4,000 individual wineries in the state of California.

Yeah, that makes sense. And to call it contamination is also a little bit of a stretch, because you're talking— contamination means something is where it shouldn't be, and I would explicitly expect glyphosate to be in a field where you're using it. But the other question here is, are you actually even detecting something, or are you looking at the noise of the assay? And I think that's the critical question.

Agreed.

How cool to get to work with experts. And when you get to know a guy like Thomas or meet someone like Shelley, it really makes you grateful as a scientist to be able to leverage on their expertise, because these are 2 people who I really have grown to respect and trust and lean on as an expert when I need to know a little bit more about an assay or how, how some sort of science works. So this is Talking Biotech number 30. It's done, it's in the can, and future ones are looking pretty exciting. So thank you very much for listening to Talking Biotech podcast. Tell a friend, get more people involved. I think it's really exciting, the momentum we have and how the messages of telling our story of science is gaining considerable traction. That people get excited about the things we're doing, the innovations we have, and the problems we can solve, and how we can solve problems for people. And that scares the pants off of the people who want to stand in the way of the technology. That we may actually be making a difference and at least giving good information that those who are just concerned about their food can use to weigh the information they're exposed to. My name is Kevin Folta. Reach out anytime at talkingbiotech@gmail.com or @TalkingBiotech on Twitter. And thank you very much for listening, and we'll talk to you next week.

Thank you for listening to the Talking Biotech podcast. Please send your suggestions for guests, comments, or questions to talkingbiotech@gmail.com. Please write a review on iTunes and recommend this podcast to a friend. More downloads and reviews raise the visibility of this podcast and help us reach a wider audience with science.