Who is regulating the release of the GM mosquito?

By Jason Major

TechNyou

A burst of research involving modified mosquitos to control diseases such as malaria and dengue has prompted scrutiny of the research itself and the regulatory oversight attached to it.

For previous news on this research see

here and here

Even though the research has been happening in the lab for some time, there are concerns we don’t have sufficient scientific knowledge about potential effects on the environment and human health and the regulatory authorities in the respective countries haven’t the legislation to deal with this novel technology. Others of course dispute this, namely one of the companies involved in the development of a genetically-modified mosquito.

PLOS has dedicated their January 2012 edition to tropical and neglected diseases and sought the opinion of the various players in this area for their views on the science surrounding genetically modified insets and their regulation. The focus is on the US and Europe, but Australia gets a mention and generally we fair pretty well

Issues raised

The main issues were raised by Reeves et al in their article, Scientific Standards and the Regulation of Genetically Modified Insects.

Their key gripes were a lack of transparency in the regulatory system, a lack of data to support their decisions and an over reliance on unpublished data, as suggested by the following:

“The selective use of unpublished or non-peer reviewed evidence to support contentious conclusions has been repeatedly questioned”

“Additionally, the failure of US regulators to publish data (in this case their own data) prior to their inclusion in regulatory documents represents a worrying precedent for the scientific quality and transparency of future environmental impact assessments.”

“However, despite these advantages, the flow of information to the public and scientific domain over the last 9 years has been limited. In fact, it can be argued that dissemination of relevant information is so restricted that it undermines the value of public consultation exercises, as comments will almost by definition be ill-informed and readily dismissed as such.”

For the defence

The defence being the people from Oxitec who is developing genetics-based control solutions for pest insects. One of these is the GM mosquito modified to control dengue fever, which they have had limited trial releases of in the Caymans and Malaysia.

These guys actually agree that greater transparency in permit applications, reports and the science is in most, if not all, cases better. If for no other reason, one would hope it would help build confidence in the regulatory system. But the Oxitec guys defend this position only so far claiming a right to protect their intellectual property.

The question is can we achieve public accountability and still allow companies to protect their IP? This ain’t my area of expertise, but I would have thought that it was possible.

They also defend the regulators use of unpublished data calling their assumptions naive. Their following barbs pretty sum up their feelings here:

“Reeves et al. are also on weak ground when they assert that regulators should consider only information published in peer-reviewed journals. This assertion depends on three assumptions: that journal peer-review is a superior guarantee of quality than any other method, that no data from any other source can be of adequate quality to warrant consideration, and that regulators themselves are incapable of adequately assessing the quality and significance of data provided to them. Each of these assumptions is naïve at best.

…Furthermore, journals select papers for publication using criteria that are not fully congruent with the needs of regulators. In particular, it is difficult to publish negative data. Studies showing lack of difference between a transgenic strain and its unmodified wild-type counterpart may be of great interest to regulators but not to journal editors.”

Suffice to say that there is healthy debate. The difference with these opposing groups is that there is no outright rejection of the science or its potential uses simply because it involves GMOs. The debate lies with the regulatory response and how the scientific and regulatory process is communicated to society to ensure adequate oversight.

Oz regulations

One of the problems, of course, is that regulations are not consistent across countries. So how does it work in Australia if a company or research group wants to release genetically modified mosquitos, or any other insect?  Who regulates and monitors this and do we have oversight that is robust enough to minimise risk to the environment and human health? Of course, what is an appropriate risk will differ for each person, but the following is how it works in Australia. Whether it is adequate is for you to judge.

For anything to do with a genetically modified organism (GMO) the bucks starts and stops with the  Australian Gene Technology Regulator who comes with the Arnie-inspired nickname, The Regulator. He (Joe Smith) gets support and advice from the Australian Office of the Gene Technology Regulator (OGTR). The legislation for regulating gene technology is comprised of the Commonwealth Gene Technology Act 2000, the Gene Technology Regulations 2001 and corresponding State and Territory legislation.

Although Australia has yet to release into the environment any genetically modified animal there has been considerable research on using the technology to control pest animals such as the European Carp, mouse, rabbit and fox. None of this left the laboratory.

Process to release GM animals in Australia

The following relates to groups wanting to intentionally release a GMO into the environment. This falls under the criteria called Dealings involving and Intentional Release (DIR).

They can be a limited controlled release (ie small numbers in a limited area usually a field trial), or a commercial release such as what Oxitec want to eventually do with their GM dengue mozzie. Each are treated differently under OGTR regulations

A limited release DIR has a 150 day assessment process where they assess the application, confer with the states and territories and put together a risk management plan. This is then made public (online, etc) for comment. The OGRT make any decisions regarding the granting of a licence or if there are any risk management criteria that accompany such a licence based on these consultations and any other associated scientific literature and reports available.

If the request is for a commercial release then the assessment process is 255 days and there are two rounds of public consultation allowed for assessment of the risk management strategy.

Some context

As pointed out in the last PLOS article by Mumford, billions of mass-reared insects, which are not GM, are released into the wild each week around the world. Many are non-native pest species reared, irradiated, and distributed in sterile insect release programs.

In Australia we have released loads of fertile insects and other organisms such as fungi as biological control agents. Thousands of other similar biological control agents have been released worldwide. The cane toad is the obvious example that went wrong, but today there is years of science done and intense regulatory oversight for any proposed biological control organism.

The Wolbachia example

Wolbachia-infected mosquitos have had a limited trial release in Australia in the hope they will help control dengue fever. For the story on the Wolbacia research see here

Because the Wolbachia-infected mosquito is not a genetically modified organism, the scientists involved in the research and the mosquito’s release report at a conference they struggled to find the appropriate regulatory procedures that applied to their technology to ensure responsible oversight and approval. The risk assessors in that case have called for the regulatory process to be reviewed to ensure that it is appropriate for both proponents and regulators.

From Mumford, PLOS: “Although it is not a GM release as neither of the organisms involved have foreign genes inserted, the recent Australian release of Ae. aegypti transformed with Wolbachia (which reduces the capacity of the mosquito to act as a vector of dengue) is an interesting example of the state of regulation in this general area. The authors state “Approval for the release of Aedes aegypti containing Wolbachia was provided by the Australian Pesticides and Veterinary Medicines Authority.”

Bulletproof skin stops a speeding gunshot

A futuristic tissue that reinforces human skin cells with spider silk can stop a whizzing projectile without being pierced.

New Scientist: 30 January 2012

Image: Yuji Sakai/Digital Vision/Getty

GMO pig feeding trials – the data

By Jason Major

TechNyou

In reference to the news story I posted this morning about the feeding trials of pigs with Bt corn, the kind people from Teagasc has supplied me with some proceedings and slide presentations from this research – see below.

I will appreciate anyone’s thoughts on the study, their analysis, it implications and any shortcomings in the research methodology and initial results, etc,

I mentioned that I thought the data was unpublished. It turns out that part of the research – the short-term feeding stuff – has been published in the British Journal of Nutrition and is now available for review (Walsh et al., 2011).

Links to

Proceeding paper

Pig_Conference_Proceedings2011

Slide presentation

8_GMSAFOODSlidePresentation

Further Teagasc research 

Hanging questions and caveats

The proceedings paper doesn’t explicitly say if the GM and non GM Bt corn used in the trials were exactly the same cultivars – see following extract:

The Bt maize was grown in a neighbouring plot to the non-GM maize counterpart under identical environmental conditions in Valtierra, Navarra, Spain and the only difference between the two maize types was the Bt toxin.

This is important as two cultivars of corn can have radically different nutrient profiles, and levels of other compounds such as naturally occurring allergens, toxins and anti-nutrients. They were at least grown under identical conditions – something else that can significantly change these components.

These results only apply to the short and medium-term studies because the laboratory analysis for the long-term data is still happening. There a few lines of, “further research will need to be done to find out what is happening…” in the proceedings so we will have to wait and see. It is possible the long-term feeding analysis will provide some answers and clarification.

Couple of findings

I have yet to analyse the paper in detail and will probably wait until the long-term trial analysis is completed, but here a couple of the findings that jumped out.

Results from the medium-term study found that feeding Bt maize to pigs for 110 days had no effect on body weight, their average daily weight gain or their average daily feed intake.  But average daily feed intake (ADFI) was higher in pigs fed Bt maize during days 14-31 and overall from day 0-31. However, there was no difference in body weight between the two groups of pigs on day 31. Overall, feed efficiency was not affected by feeding Bt maize even though pigs fed Bt maize were less efficient on days 14-31 which was a consequence of higher feed intake by these pigs at that time. In other words weanling pigs were eating more during days 14-31 but their weight gain was the same as the conrol pigs.

If this difference in ADFI at days 14-31 is real (ie, it would need to be repeated and probably with a larger sample) then it would be interesting to know why this is happening.

In the slide presentation they present the finding that Bt maize may offer protection against an inflammatory response observed following non-GM maize consumption. I wonder if this is because maize is notoriously affected by a fungal toxin that is prevalent in corn damaged by insects?  Has anyone got any knowledge on this and if this may be a cause?

Update (30 Jan 2012)

Peadar Lawlor the head of the Irish component of the project got back to me re: the research and plans for publication. His response, “… we will have ~12 peer review papers from this work by the middle of this year. We have published in the British Journal of Nutrition, Animal, PLoS ONE etc. We will publish all our work from the project.” All links to papers and research updates will be posted on their website see Teagasc link above

And here is the link to their PLoS One paper

Home page image: impactlab.net

Gene silencing to help treat Huntington’s Disease

Researchers have successfully delivered inhibitory RNA sequences from stem cells directly into neurons, significantly decreasing the synthesis of the abnormal huntingtin protein.

University California Davis: 19 January 2012

New microbe turns sugary seaweed into fuel

Scientists have built a microbe capable of digesting seaweed and converting it into ethanol or other fuels or chemicals.

RSC 19 January 2012

The less technical version

Scientific American 19 January 2012

Image: © Bio Architecture Lab

One mistake away from a worldwide flu pandemic

Two labs have made lethal, highly transmissible versions of bird flu virus – do the risks of it escaping outweigh the benefits of the research?

New Scientist: 12 January 2012

And a previous story from Nature News December 2011

Two tribes at war. What The Atlantic’s very real dangers of GM food reveals about us.

By Jason Major

TechNyou

The Atlantic which ran the article, “The very real dangers of genetically modified foods”, by Ari Levaux has unleashed an unusually massive online response. The reasons seem to be many, but I would suggest it starts with the journalist’s complete mangling and distortion of the science that produced a misleading, even deceptive article. The result was 334 comments, at last count, some from TechNyou. An analysis of this commentary reveals the obvious two distinct tribes fighting trench warfare at a stalemate, but more revealing was the difference in their tactics, which might be indicative of why this debate never moves forward..

The ample commentary tearing apart Levaux’s faulty analysis and interpretation has forced him to rewrite his article and I won’t repeat the criticism of the article from others more qualified than myself. Nevertheless because Levaux’s claims the main point of his article was to point out the inadequacy of substantial equivalence as a tool to assess the safety of GM foods, calling it an outdated and an unproven hypothesis, at the end of this post I have included some of his points and the key responses to it. I have done this largely because Australia’s regulators also rely on substantial equivalence as their key assessment tool for GM foods.

Two tribes

I thought I would take a different tack with a dissection of the commentary rather than the article itself.

There are two distinct and obvious tribes: those in favour (or at least not vehemently opposed) to GM Foods, and those against it. The interesting this is the difference in tactics. Each fight with different weapons that pitted against each other appear to neutralize themselves, so that neither side gains ground.

The following is my quick and subjective analysis of the commentary. I sorted the comments into those obviously arguing in favour of GM foods and those arguing against them. Within the ‘in favour’ and ‘against’ I split them again into additional categories: those making commentary using scientific evidence to support their arguments and those using value-based judgements to support their arguments. I also had a category for those that appeared neutral. I didn’t review all the comments because basically I got bored, but I got through close to two-thirds of them. There were a few threads I ignored because they were the same two people going tit-for-tat: you’re wrong; no you’re wrong. There were also other ignored comments that were simply along the lines of, “you are a nutter”.

In the ‘in favour’ camp under scientific arguments I made a third category: ‘science neutral, which is explained below.

The categories

In favour of GM food

Scientific argument

Examples:

see comment from Karl Haro Von Mogel toward end of post

Or, “We know where the proteins in a genome are by sequencing messenger RNA that is in the cell and matching it up to the genome. We know where the miRNAs are by sequencing short RNAs that are in the cell (those that are about 21 bases) and matching them up with the genome… In humans, the miRNAs are generally in between genes, or within genes in the part that gets spliced out (the introns). In plants the miRNAs are generally not within genes at all…”

Values argument

Example: “I understand why Monsanto tried to use the Terminator gene system, because I understand intellectual property. A company can’t give something away for perpetual use for a one time fee.”

Or, “Yes, I suppose I could be worried that every funny looking potato in my garden is a dangerously toxic mutation, but part of being rational is sticking your head in the ground and not letting your life be consumed by worry over every trivial risk.”

Science neutral

This category was included because there were a number of comments from people who in previous posts appeared to, in principle, support GM food, though taken in isolation such comments would have appeared neutral because in this case they simply stated the science without appearing to use them to support or denigrate GM foods.

Example: “The word *can* is important here. Just as you say the substantial equivalence concept is a hypothesis so is the idea that GMO foods can cause metabolic disorder. This idea has never been demonstrated. I’m not saying GMO is great but if you are going to use science to demolish an opponents argument you should apply the same standards to your own.”

Against GM food

Scientific argument

I could have split this category again because there were about five comments using false or misunderstood assumptions about the science, such as the 1st example below. But I decided to lump them all together.

Examples: “But whereas organic apples and non-GMO are selected over a period of decades, GM food is spliced with viruses including Ebola. Not sure I like the idea of consuming that.”

Or, “genetic engineering can cause changes in metabolism that are much more extreme than mere cross breeding. This is because genetic engineering depends on DNA expression promoters and the introduction of completely novel proteins into organisms that have never produced them before.”

Values argument

Examples: “I disagree*. Evolution isn’t motivated by profit.”

Or, “Sadly, every decision made by big business is based upon the bottom line rather than the common good. Until this mentality changes, we are doomed to endure that which will benefit commercial interests.”

Neutral

This covers a lot of commentary, and some may not necessarily be neutral, but they appear to be so relative to the GM issues. For example there a long thread discussing heirloom seeds and hybrid seeds and how this related to their yields and ability to save seeds. Unless they had posted an earlier comment that revealed whether they were for or against, it was impossible to know whether these people were for or against GM and if they were relating their comments to the GM discussion. That is, just because you may wish to have everyone grow heirloom seeds, doesn’t necessarily mean you are anti-GM. That, at least, was the assumption I chose to make. Other examples included arguments about monoculture V polyculture.

The results

Table 1. Number of comments for each of the two tribes for each category.

The analysis

It is no wonder this debate never moves forward and we keep plodding over the same old ground.  From the Table 1, the difference in how the two sides approach the debate is immediately obvious.

The “In favour” camp use science and values equally to argue their case whereas the “Against” camp rely almost exclusively on values. Science barely rates a mention in the “against” camp and if it does it is often based on a flawed understanding of the science, in this case about 50% of the time. Note, however, there were incorrect assumptions from a handful of those arguing in favour of GM crops, though as with the following example, he or she was simply misinformed rather than having a flawed scientific understanding: “Africa would be decimated right now due to massive wheat shortages from the crops being susceptible to bacteria. GM wheat helped revive Africa’s wheat production, staving off a massive food shortage that could have been catastrophic.” I have no idea where this guy would have got this info from.

The majority of the values of those against GM foods alluded to the dislike, mistrust, hate, undesirable ethics of the handful of companies who own the GM crops and the respective patents. This aligns with TechNyou’s experience with the Australian public, as does the breakdown of the comments in general. Other against-based values comments aligned with the “Mother Nature” philosophy and that we are tinkering with things we don’t understand, or that natural is best. The labelling issue (a right to know) and the occasional dose of the yuck factor – ie that it is inherently wrong to put genes from one species into another – were also present.

Risk

It is difficult to extrapolate from my analysis to the general population, but if you took the second values example used above for those in favour of GM foods as a reflection of society as a whole re: this issue, it exemplifies another difference between the tribes and that is our perception of risk. Those in favour of GM foods are less risk averse than those against it. Again. however, this aligns well with the more robust studies out there.

For example, compare the above view with the following comment linked to the Levaux article:

“The insertion of powerful DNA expression promoters and proteins that are very different from those occurring naturally in plants can have dangerous, unintended and unanticipated consequences. Some plants produce toxic chemicals and may even produce harmful proteins or peptides. It is arrogant in the extreme to assume, as you do, that all possible combinations of DNA, proteins and biochemistry is safe for everybody to consume for a lifetime.”

One side is, benefit dependant, prepared to accept some risk, the other side accepts only zero, risk at least when it comes to food. The story is very different if you talk about topics such as medicine.

Venus and Mars

If you want to use a scientific argument, it will more than likely be countered with a values argument, regardless of whether the values are rational or not. Values define a person and they are not about to change them easily. Certainly, attempting to do so with scientific evidence is unlikely to succeed. Conversely, the person who can only see the scientific facts can have trouble understanding the nature of their opponent’s values, though it is notable that the “in favour” camp has an even split between those that used science and values to back their arguments. This is not a criticism of either camp. From where I sit it is human nature and if we are to have such debates move forward then we need to better understand where each group comes from and what makes them tick.

Caveat

Note, as mentioned above this is not a robust analysis. It is my quick, dirty and SUBJECTIVE analysis, so I am sure any worthy social scientist will be able to pick holes in it. However, this analysis does align reasonably well with my personal experience (again anecdotal) and some of the more robust public opinion surveys and peer-reviewed literature.

Substantial equivalence

OK, here is the basics of the arguments put forward about substantial equivalence (SE).

Response from Ari Levaux to critical comments about his position on SE

“I think substantial equivalence is outdated, and Monsanto’s stance against toxicity testing is transparent, arrogant, and reckless. microRNA shows just how much more complicated things are than the assumptions built into substantial equivalence. That said, microRNA provides nothing more than a possible path by which a problem might occur.”

And again

“… Anyway, none of the nitpickers – except the impressively level-headed mirna, have provided any useful response to my points that substantial equivalence is a transparent joke of an unproven hypothesis, and that Monsanto’s stance against toxicity testing is based in business objectives, not scientific objectives.”

My initial thought to Levaux’s response was that his inability to understand and interpret the science (sounds like he didn’t even try) severely dented any credibility to the point that it make it hard to respect his thoughts on substantial equivalence.  Nevertheless, the questions of substantial equivalence and the precautionary principle, which he also refers to, are legitimate ones. Whether you agree with Levaux or not is another matter. Levaux claims the objectives are based on business rather than science.

The precautionary principle is equally controversial. I am no expert, but the conversations I have had with people on this topic are steeped more in values and personal ethics than anything to do with science. Much of it comes down to your definition of safe, though that is a bit simplistic.  Interpretations of the precautionary principle are as many and varied as those of the bible. This has been the case even as far back as my third year undergrad class 20 years ago when it was part of an assignment and tutorial topic.

Becasue Australia uses substantial equivalence as the basis to assess GM foods, I thought it pertinent to at least present some more details on the issue. Levaux’s arguments can be read in his articles and his responses to the comments. The following details are largely from the dudes at Biofortified

Karl Haro Von Mogel – part of a posted reply to Levaux

…Your treatment of Substantial Equivalence is simplistic and you don’t indicate that you understand precisely what is meant by it, and why it was developed. There is a perception amongst anti-GE individuals that it is some sort of way of avoiding testing, when in fact it is a way to determine if there is a biochemical or nutritional change that falls outside the natural range of variation, which would suggest further testing. It is not an assumption of equivalency, but a determination that is reached only after testing has occurred. It doesn’t preclude further testing, nor is it a claim that there is no possible side effects from genetic engineering. Genes exist in a connected network of activity, and any change, including by mere breeding, can have effects on that system. Substantial equivalence is in fact a recognition of this fact that a genetically engineered trait can significantly affect this system, and if it does, further scrutiny is advised.

Anastasis Bodnar

Bodnar had previously written a reasonably extensive article for Biofortified, though it is written from a US/FDA perspective.

What substantial equivalence can do is give us a starting point.

We know that there is variation in amounts and types of proteins and metabolites, gene expression, and other parameters from variety to variety, from environment to environment, and from plant to plant. For example, if I use a microarray to find similarly and differently expressed genes in two genetically identical plants grown in slightly different environments, such as different temperatures, I will find some genes that have significantly different expression. Similarly, plants of different varieties grown in the same environment will have different gene expression profiles and even two identical plants in the same environment will have some differences.

The first step in a comparative assessment is to test and compare the genetically engineered variety to a genetically similar variety that doesn’t have the trans- or cis-gene. Tests can include gene expression, metabolic profiles, feeding studies, and more. If differences aren’t found in a reasonably wide panel of tests, then the genetically engineered variety can be called substantially equivalent to the genetically similar variety.  More here…

Finally, the perspective of the Australian Regulator, Food Standards Autralia New Zealand (FSANZ)

Silkworms spin spider webs

Silkworms have been genetically modified with spider genes to produce a composite silk that is stronger than spider silk.

PhysOrg 3 January 2012

PNAS Jan2012 Early Edition

Taking the random out of genome editing

Nature mag has hailed techniques that allow scientists to introduce precise, targeted and tailored changes into the genomes of any species their research method of the year for 2011.

Nature Methods: 28 December 2011

Assessment of long-term, multigen feeding studies says GM foods are safe

Jason Major

TechNyou

A review of long-term and multi-generational animal feeding studies declares GM foods are safe, though there are many criticisms of the studies.

The review appears in the journal Food and Chemical Toxicology (in press). They analysed 12 long-term (greater than 90 days) feeding trials and 12 multi-generational feeding trial on a range of GM crops and traits. They analysed these alongside the many 90 day or less feeding studies.

Their overall conclusion is that GM plants are nutritionally equivalent to their non-GM counterparts and can be safely used in food and feed.  The caveat here is that the reviewers had serious concerns with the study design and consequent strength of the statistical analysis for a lot of the studies reviewed. Topping the concern list was the lack of a comparable isogenic lines in the feeding trials. This means that the GM cultivar the researchers used was different to the non-GM one they used as a control. For example, if you go to the nursery to buy sweet corn, you will have loads of different varieties to choose from – all different cultivars. Each of these varieties (or cultivars) will have different nutritional and metabolomic/proteomic, etc profiles and consequently your body will produce different metabolites, etc in response to them. Hence if you have a specific GM cultivar, you should use the same cultivar without the transgenic trait for an accurate comparison.

There were also concerns with many trial having insufficient animals to provide sufficient statistical strength – the more you have the more certain you can be about your results.

Statistics V Biology

One thing that needs to be pointed out is that there is a big difference between something that is statistically significant and something that is biologically significant. In a few of the studies some parameters showed small differences that were statistically significant but not biologically significant. In the case of toxicology, it is the biologically significant that is important. For example, if 10 units of a particular toxin found naturally in a food is declared harmless for humans, and the normal range in a specific cultivar or plant is between 2 and 5 units. If your GM plant is found to have 6 units that may be declared statistically significant, but it is still going to be harmless from a biological perspective so has no biological significance and would be considered safe to eat.

It will be interesting to see if there is any controversy of this review. I have yet to see any which is surprising given the amount of uncertainty regarding the scientific methodology on some of the public papers and the criticism from the review authors. I would have thought this would have provided loads of ammunition for the opposing camps in this debate.  Maybe it is too close to Christmas?