Extraordinary claims… require extraordinary evidence.

Within the past few weeks, a letter written by a Dr. Don Huber to Secretary of Agriculture Vilsack has been making the rounds on the ‘net. The letter was allegedly given to the Farm and Ranch Freedom Alliance, and they claim to have confirmed that it was written by Dr. Huber. You can find the full text of the letter on the FRFA site with the ominous title Researcher: Roundup or Roundup-Ready Crops May Be Causing Animal Miscarriages and Infertility.

The story has been picked up by many bloggers, including Jill Richardson, and even made an appearance on Reuters. I haven’t seen any posts dedicated to a critical analysis of the letter, instead there is a rush to assume that it is correct, despite the lack of citations or other evidence provided for the extraordinary claims in the letter. The story is often accompanied with horrific pictures of dead fetal calves and the words “Emergency!” and “Danger!” Are we really all in danger? The claims in the letter bring to mind Carl Sagan’s famous statement: “extraordinary claims require extraordinary evidence.” Let’s investigate the claims and determine whether enough evidence is provided.

“This organism appears NEW to science!”

In the letter, Dr. Huber claims that there is a never-before-seen pathogen that is caused by or exacerbated by either glyphosate containing Roundup herbicide or the widely used glyphosate resistance gene. The letter opens:

A team of senior plant and animal scientists have recently brought to my attention the discovery of an electron microscopic pathogen that appears to significantly impact the health of plants, animals, and probably human beings. Based on a review of the data, it is widespread, very serious, and is in much higher concentrations in Roundup Ready (RR) soybeans and corn—suggesting a link with the RR gene or more likely the presence of Roundup. This organism appears NEW to science!


Right here in the first paragraph is Extraordinary Claim #1. Dr. Huber is claiming that a single pathogen can “significantly impact” the health of corn, soy, and animals. Not impossible, but extraordinary evidence is required to back up the claim because known pathogens are generally very host specific, whether they are bacteria, virus, fungus, or parasite. A corn pathogen will not infect soy. A human pathogen will not infect cows. In cases where a single pathogen will affect multiple species, it affects groups of very similar species, not corn and cows.

What evidence does Dr. Huber provide for this extraordinary claim? None, actually. Just more extraordinary claims that seem to get more and more extraordinary with each paragraph.

Extraordinary Claim #2 is that the “organism is only visible under an electron microscope (36,000X), with an approximate size range equal to a medium size virus. It is able to reproduce and appears to be a micro-fungal-like organism. If so, it would be the first such micro-fungus ever identified.” He leaves us with far more questions than answers. What characteristics, exactly, cause him to compare this claimed pathogen to a fungus? How could it be possible to have a fungus so small? Where are the pictures? How big is the claimed organism and what does it look like? What is the evidence that it is reproducing? What other tests have been done to confirm its existence?

Fungi and viruses – not at all similar

Fungi have some special characteristics that make them easily identifiable. First, fungi are eukaryotes, meaning that they have complex cells with structures enclosed in membranes called organelles, along with plants and animals, but unlike bacteria which lack organelles. Eukaryotic cells range between roughly 10 and 100 micrometers (μm) long. Second, fungi have some characteristics that make them unique compared to other eukaryotes. Like plants, they have cell walls but unlike plants, those cell walls contain chitin instead of cellulose. At minimum, if we want to call something a fungus, it needs to have organelles like other eukaryotes and needs to have those unique cell walls.

Eukaryotic cells are many times larger than viruses. "Scanning electron micrograph of the surface of a mouse cell infected with murine leukemia virus. A large number of virus particles are shown in the process of budding." By R. MacLeod via The Free Dictionary.

Viruses are completely unlike eukaryotes or bacteria. They have a wide range of shapes but all look quite different from eukaryotic or bacterial cells. Viruses are little more than some nucleic acid surrounded by a protein coat, allowing them to be much smaller than cells, at a range of roughly 0.01 to 0.1 micrometers (μm). Even the largest virus is much smaller than the smallest eukaryotic cell. In fact, viruses are smaller than the any of the organelles inside a eukaryotic cell.

Saying that something is a “micro-fungal-like organism” as small as a virus just doesn’t make any sense. Of course, there’s been other strange things discovered, things that defied existing biological knowledge. Maybe this thing is from space, transported on meteorites. Who knows!? If it is true, then Dr. Huber and colleagues would undoubtedly be lauded for their amazing discovery. But this extraordinary claim requires extraordinary evidence and Dr. Huber provides none.

Electron microscopy – it’s not easy

Series of images of a snowflake taken by USDA researchers. Click the pic for larger images.

When I worked for the USDA Animal and Plant Health Inspection Service (APHIS) in Beltsville, MD as an undergraduate, I had the opportunity to use an electron microscope to look for viruses in plant tissue samples. Our goal was to identify plant pathogens before plant material got shipped all over the country. The normal procedure was to wait a pre-determined period of time to see if a plant would show symptoms, but if we could ID viruses before symptoms showed we could save a lot of time. Unfortunately, the technique didn’t pan out, at least while I was working there, because the experts weren’t able to find a technique that allowed them to accurately ID viruses with electron microscopy.

Electron microscopy is very touchy, with many things that could go wrong. Strange artifacts or errors in the images can be introduced by the processing a sample must undergo before viewing, by less than perfect use of the instrument, and by the instrument itself. Consider this series of images of a single snowflake taken at increasing magnification with an electron microscope. As the magnification goes up, the likelihood that meaning could be ascribed to a random bump also goes up.

Paul Vincelli, Professor of Plant Pathology at
University of Kentucky and member of the American Phytopathological Society (APS), has expertise in plant pathogens including viruses and fungi. He has commented on the post Scientists warn of link between dangerous new pathogen and Monsanto’s Roundup by Rady Arnada indicating that he has seen the claimed “micro fungus” research himself. He said he has spoken with another researcher that has seen the electron micrographs, who concluded that the supposed “micro fungus” is actually just artifacts and that “detailed molecular data were needed before concluding that the structures observed were actually organismal.” Hopefully Dr. Huber plans to relase the images soon so additional experts can examine them. You have to wonder why the images haven’t already been released.

Pathogen presence

Extraordinary Claim #3 is that the claimed pathogen “is found in high concentrations in Roundup Ready soybean meal and corn, distillers meal, fermentation feed products, pig stomach contents, and pig and cattle placentas.” Why is this extraordinary? There is no control information provided.

We need to know what are the relative concentrations of the claimed pathogen in corn and soy plants grown in identical conditions, preferably in multiple environments of the following categories so we can isolate the effects of the Roundup Ready gene and of Roundup:

  1. Roundup Ready plants that are treated with Roundup
  2. Roundup Ready plants that are weeded by hand or other non-chemical method
  3. non-Roundup Ready plants that are genetically similar to the Roundup Ready plants that are weeded by hand or other non-chemical method (negative control)

Without these comparisons, saying “high concentrations” is meaningless. We also need to know the relative concentration of the claimed pathogen in animals fed these different plant samples under strictly controlled conditions. We also need to know how the presence of the claimed pathogen was determined and whether it was confirmed with any additional tests, such as nucleic acid or protein analysis.

Similarly, the claim that the “organism is prolific in plants infected with … sudden death syndrome (SDS) in soy, and Goss’ wilt in corn” also requires comparison to uninfected plants with and without Roundup and the RR gene. Dr. Huber continues: “The pathogen is also found in the fungal causative agent of SDS (Fusarium solani fsp glycines).” Found in? As in inside the cells? How do you know? Again, where are the pictures?

Cattle, swine, and horses (oh, my)

French dairy cows. Are these ladies luckier with their calves than American cows? Image by Meg Hourihan via Flickr.

Extraordinary Claim #4 is that there has been “escalating frequency of infertility and spontaneous abortions over the past few years in US cattle, dairy, swine, and horse operations. These include recent reports of infertility rates in dairy heifers of over 20%, and spontaneous abortions in cattle as high as 45%.” For comparison, the expected rate of spontaneous abortion in dairy cattle is about 2-5%, according to Virginia Cooperative Extension’s Abortions in Dairy Cattle and West Virginia University Extension’s Abortion in Dairy Cows and Heifers, and the expected successful insemination rate is 50% or higher with proper technique.

Don’t you think that if the rate of spontaneous abortion in livestock was skyrocketing that we’d have heard about it earlier? We’d see a huge spike in the cost of meat and dairy if farmers had to artificially inseminate their sows and cows an increased number of times to succeed in a pregnancy and if a high rate of those pregnancies resulted in late spontaneous abortions. What about the relative rates of AI success and spontaneous abortions in countries that use glyphosate and RR crops vs those that don’t? Shouldn’t we see major differences?

Dr. Huber claims that the “micro-fungus” has been detected “in a wide variety of livestock that have experienced spontaneous abortions and infertility. Preliminary results from ongoing research have also been able to reproduce abortions in a clinical setting.” How was the claimed pathogen detected? With “laboratory tests”, of course! Unfortunately, zero explanation is provided of what these tests are, how or where they were conducted, etc.

Anecdotes aren’t sufficient evidence to justify policy changes

We are provided with an anecdote: “450 of 1,000 pregnant heifers fed wheatlege experienced spontaneous abortions. Over the same period, another 1,000 heifers from the same herd that were raised on hay had no abortions. High concentrations of the pathogen were confirmed on the wheatlege, which likely had been under weed management using glyphosate.”

Likely? This single word causes me to seriously doubt that a scientist wrote this letter. This anecdote is clearly not a scientific study because there are no controls and there is no confirmation of whether the feed did or did not have Roundup residues or the mysterious claimed pathogen present. To make conclusions based on a single situation we don’t even have details on is irresponsible at best. It is even more irresponsible to call for changes in national policy based on an anecdote.

Let’s consider this anecdote more closely. Glyphosate has been used as a herbicide since the 1970s. The amount of glyphosate use has increased with glyphosate resistant crops, and the amount of other herbicides used has decreased, at least until glyphosate overuse caused weeds to develop resistance (but that’s another story). As the use of Roundup and other glyphosate products has been increasing steadily, and crops that have been grown in fields that were treated with glyphosate have been being fed to livestock more and more over the years. If there is a link between glyphosate use and the rate of spontaneous abortions in livestock, then we should see a linear correlation between the two. In other words, the spontaneous abortion rate should be steadily increasing as glyphosate use has steadily increased.

Now let’s look at the two types of feed. Dr. Huber claims that 0% of heifers fed hay had abortions while 45% of heifers fed wheatlage (not wheatlege) had abortions. The wheat may or may not have been “under weed management using glyphosate”. Since there are zero genetically engineered varieties of wheat (Roundup Ready or otherwise) we know that the wheat itself was not sprayed with glyphosate because without the resistance gene it would die. Instead, glyphosate may have been used before the wheat was planted or along the edges of the field. Is this enough glyphosate to cause spontaneous abortions? If it was, then there would be a lot more abortions in livestock.

Can we think of anything else that may have caused the claimed abortion rates? Yes. Going back to the extension documents Abortions in Dairy Cattle and Abortion in Dairy Cows and Heifers, we learn that there are multiple causes for increased number of spontaneous abortions in cattle, including undiagnosed genetic abnormalities, heat stress and infection by certain types of viruses, bacteria, and parasites. Feed contamination with a variety of types fungi that produce toxins can also cause abortions in cattle, especially when the cattle are otherwise immunocompromised by things like stress or disease.

This anecdote can be easily tested by having two groups of randomly selected cattle fed feeds that are identical and grown under identical conditions except one has been under weed management with glyphosate and the other was weeded by hand or other non-chemical means.

Who is Don Huber?

We need to examine Dr. Huber’s experience and positions so we can determine whether he has relevant expertise to be discussing both the extraordinary claims made in this letter and his more reasonable claims that glyphosate could have an effect on mineral uptake and disease resistance. Unfortunately, the letter doesn’t lend him much credibility, assuming that he did indeed write it.

The letter is signed “COL (Ret.) Don M. Huber, Emeritus Professor, Purdue University, APS Coordinator, USDA National Plant Disease Recovery System (NPDRS)”. Dr. Huber retired in 2006 or 2007. He is listed as a faculty/staff member at Purdue but I wasn’t able to find a bio or CV page on the Purdue website (or indeed a bio or CV elsewhere, either, but that may be due to of all the blog posts re-posting the letter that may be pushing other results back more pages than I’m willing to sort through).

The NPDRS is a program called for in Homeland Security Presidential Directive Number 9 in 2004 “to ensure that the tools, infrastructure, communication networks, and capacity required to mitigate the impact of high consequence plant disease outbreaks are such that a reasonable level of crop production is maintained in the US.” It was “a cooperative effort of university, industry, and government scientists sponsored by The American Phytopathological Society (APS) and the United States Department of Agriculture (USDA).”

As far as I can tell, the last activity of NPDRS was in 2008, and their list of recommendations on the USDA page is a broken link (the correct link is here). Dr. Huber completed work on late wilt of corn for NPDRS and was the chair for that project, but is not listed as the coordinator of NPDRS and I could find no mention of him being the coordinator of the APS side of the partnership. Instead, Kent Smith, a USDA employe, is listed as the contact person for NPDRS. Don Huber is not listed as an employee of the USDA at this time.

Dr. Huber is a member of the Emerging Diseases and Pathogens Committee of the American Phytopathological Society (APS). He served as President of the APS North Central Division in 1988, and has served on other APS committees throughout the years, but does not currently hold any leadership positions with APS that I was able to find.

What work has Dr. Huber done?

Photo of Dr. Huber from a 2010 article in No-Till Magazine.

A search on PubMed for DM Huber results in 11 papers (one of which is not this DM Huber), including these two most recent listings:

  1. Thompson IA, Huber DM, Schulze DG. Evidence of a Multicopper Oxidase in Mn Oxidation by Gaeumannomyces graminis var. tritici. Phytopathology. 2006 Feb;96(2):130-6. PMID: 18943915
  2. Thompson IA, Huber DM, Guest CA, Schulze DG. Fungal manganese oxidation in a reduced soil. Environ Microbiol. 2005 Sep;7(9):1480-7. PMID: 16104870

I don’t know why PubMed has such paltry results. Web of Science provides 115 results for DM Huber in the Life Science category. None of the papers have any mention of a “micro fungus”. The two most recent are probably the most meaningful for this discussion. Each has been cited 9 times (mostly by the authors themselves).

  1. Zobiole LHS, de Oliveira RS, Huber DM, et al. Glyphosate reduces shoot concentrations of mineral nutrients in glyphosate-resistant soybeans. Plant and Soil. 2010 Mar;328(1-2):57-69.
  2. Johal GS, Huber DM. Glyphosate effects on diseases of plants. European Journal of Agronomy. 2009 Oct;31(3 SI):144-152.

Long story short, assuming that at least half of the 115 papers in Web of Science are actually this DM Huber (at least some belong to a DM Huber at the University of Cincinnati), we can say that he is a well published scientist that has published relevant subject matter in some fairly reputable journals for his field, including Phytopathology as recently as 2007 which has an impact factor of 2.2 (out of 5) according to Journal Citation Reports (not great, but not bad, either). Dr. Huber appears to have relevant and recent expertise on the subject of the effects of glyphosate on mineral uptake and disease resistance.

Next steps for “micro fungus”

The claimed “micro fungus” may indeed be a never before seen pathogen, perhaps a virus. At this time, however, there is not enough evidence to require action. More data needs to be collected in well designed experiments that needs to then be subjected to peer review.

Peer review is the “checks and balances” of science. A team of researchers writes up a report of their experimental design and results and submits it to a journal. Before it is published, it is reviewed by a team of scientists who evaluate whether the experimental design is sound, whether the conclusions are supported by the data, whether the statistics were done properly, and so on. Peer review isn’t perfect for multiple reasons, but as of now it is the best form of quality control for scientific research that we have. For a very good discussion of what peer review means to scientists, see Does peer review mean the same to the public as it does to scientists? This is just one part of an excellent discussion of peer review in Nature that should be required reading for every scientist as well as anyone even slightly interested in what scientists do and how to interpret science: Nature’s peer review debate.

Getting a paper through the peer review process is a necessary part of science validation, in part because of its rigid requirements that go above and beyond what one might put in a letter or a blog post. For one scientist’s first person experiences with peer review, see From blog to Science (thanks to Mary M. for the referral). Avoidance of the peer review system indicates that a researcher knows that their work won’t pass muster.

It is through the peer review process that extraordinary claims can begin to accumulate enough evidence to become accepted. There are plenty of examples of researchers who had extraordinary, some would say impossible, claims that have been proven to be true. Here are two of my favorite examples:

Susan Lolle claimed to find some examples of non-Mendelian inheritance in the plants she was studying. It looked like the seeds were “remembering” what type of environment their parents were in, which seems impossible! Other scientists tore her papers up, and pretty much openly laughed at her. She persevered, kept doing more very well designed experiments, and eventually convinced other scientists she had something. Now we understand that epigenetics is a way that DNA can “remember” environmental conditions. It’s a very exciting and still very strange new field of genetics.

Stanley Prusiner claimed to have isolated the cause of mad cow disease, claiming it was a protein that was misfolded that caused other proteins to also misfold. Like Lolle, Prusiner sounded crazy. How could this be possible? Through perseverance and hard scientific evidence, Prusiner proved that he was right and eventually won the Nobel Prize in medicine.

Any scientist who thinks they’ve find something extraordinary can either give up or persevere. If I found something that was unexpected in a preliminary experiment, I’d redo it first. If the same thing resulted, I’d talk to statisticians and experts in the field, make sure my experimental design was top notch. If I still got the strange result then I’d find a well respected scientist in the same field and ask their lab to redo the experiment or at least part of it to make sure it wasn’t just my lab coming up with the weird results. If it then was still happening, it’d be time to publish an impressive paper in Nature or Science with my well respected colleague as a co-author.

Not following this sort of path is a major shortcoming for a lot of scientists who have found unusual things. For whatever reason, there seem to be a lot of examples of scientists finding results about genetic engineering that go against established science that don’t bother going past that initial finding. The example that first comes to mind is Arpad Pusztai. Why didn’t he work on much better experimental designs before going to publish? Why didn’t he talk to some experts in plant studies so he could have had the proper controls? He took his preliminary results from some poorly designed studies and then ran with it and now people wonder why his work isn’t taken seriously. If Dr. Huber wants to be taken seriously with his “micro fungus” claims then he needs to emulate Lolle and Prusiner, not Pusztai.

Conclusions

This letter makes very little sense both in its sheer existence and in its details. Why would a reasonably well published scientist suddenly throw away everything we know about the scientific method to make claims about biologically impossible organisms with no evidence? Why is so little evidence presented and why is the evidence that is presented given as anecdotes instead of hard science? Most importantly, why would he make claims without going through the peer review process to ensure that his claims would be at least vetted by his peers?

Multiple sites have claimed to have spoken with Dr. Huber to confirm that he did indeed write this letter, but I remain skeptical that an experienced scientist would have released something so unscientific. Someone with as much experience as Dr. Huber should know that his fellow scientists (as well as government agencies) would require at least some proof before acting on extraordinary claims. Fred Gerendasy at Cooking Up a Story, wonders if the letter is a fraud. Perhaps the letter is real and he knew that no one with any knowledge of biology would accept the claims, but also knew that many non-scientists would latch on to claims that confirmed their own biases without question.

Dr. Huber’s colleagues at Purdue have responded to his claims about glyphosate use and crop mineral uptake (which I describe in Does glyphosate restrict crop mineral uptake?), but they are conspicuously silent on the “micro fungus”. The absence of analysis of the “micro fungus” claims tells me that his colleagues are politely ignoring this bizarre outburst. I would have done so as well, if it wasn’t for the prolific repetition of the claims on blogs and even news sites. It’s long past time for us to apply the Seven Warning Signs of Bogus Science to Dr. Huber’s claims. Hopefully this post will give some balance to the discussion.

Advertisements