Science in Society Archive

New Hazards in GMOs from Synonymous Mutations

Change in base sequence that does not alter amino acid sequence of proteins encoded nevertheless may result in alterations of the protein that make it unsafe. Prof Joe Cummins

Single nucleotide mutations form the majority of genetic polymorphisms (single nucleotide polymorphisms, SNPs) in populations. When found in gene protein coding regions, SNPs can be synonymous, i.e., causing no change in the amino acid encoded, or non-synonymous, when the amino acid is altered.  Until the current decade, synonymous mutations were assumed to be neutral, with no effect on the protein or any other functions of the organism.  Sequencing a vast array of genomes has revealed surprisingly, that many synonymous mutations were causing dysfunctions and illnesses in plants and animals. Synonymous mutations may lead to changes in protein folding related to translation pausing, RNA splicing, and alterations in enzyme specificity [1].

All current GM crops use protein coding genes from bacteria that have been  altered by introducing synonymous  codons, replacing plant-preferred codons for the  bacterial codons in order to enhance the production of protein from the transgenes. However, the toxicity of the transgenic proteins in animals and humans was not studied. Instead the original proteins produced in bacteria were used as surrogates in feeding trials.  The synonymous codons were assumed to be neutral and to have no effect on the transgenic proteins, and presumed to be safe [2] (Bt Toxins in Genetically Modified Crops: Regulation by Deceit, SiS 22).

The genetic code is made up of 64 three letter codons (code words) for twenty amino acids (61 codons) plus words for translation start and stop. The number of code words for amino acid varies from one for methionine and tryptophan to six for arginine, leucine, and serine. The first two positions of the codon are fixed for a particular amino acid while the third position is said to ‘wobble’, allowing for alternate code letters, hence two or more codons for most amino acids. The frequency with which different codons are used varies between groups of organisms; so for example, genes from bacteria are poorly read in higher plants and vice versa.  For optimum expression, the code for a transgene frequently needs to be rewritten. The codon bias characteristic of each group of organisms is believed to be caused by the presence of distinct transfer RNA families in the different groups of organisms. In synthesizing transgenes for GM crops, say, a Bacillus thuringiensis (Bt) cry toxin gene, a table of plant preferred codons is used to substitute for the bacterial codons [3]. Sometimes, it is necessary to substitute one or more of the amino acids (non-synonomously) so that the final cry toxin can function in the plant cell environment [4]. As plant genetic engineering has “advanced” the crucial active domains of toxins, and enzyme are “improved” to such an extent that the gene for the original source protein is hardly recognizable.

There are many examples of synonymous mutations that are not neutral.  Synonymous mutations may lead to ribosome stalling, thereby changing protein folding pathways affecting enzyme activity or antigenicity [5].  Synonymous mutations  in  the HIV gene Rev enhance HIV-1 replication,  providing resistance to the drug enfuvirtide [6].  Synonymous codons in the oncogenes of the rabbit papillomavirus increased oncogenicity and immunogenicity [7]. A single synonymous mutation was sufficient to alter the substrate specificity of a multidrug resistance phenotype in mammalian cells [8].Synonymous mutations affect the stability of mRNA secondary structure in mammals [9].

To Conclude

It is now clear that synonymous mutations are not neutral but instead frequently adversely affect the proteins encoded.  Essentially all commercial GM crops contain bacterial genes with codons altered to enhance the expression of transgenes in plants. The products of such transgenes carrying synonymous mutations were risk assessed using surrogates produced originally in bacteria and free from synonymous mutation.  Thus the regulatory approval of GM crops is based on the false assumption that synonymous mutations are neutral, and hence illegitimate. It cannot be assumed that the crops are free of defect just because millions have consumed GM food. GM food has not been labelled in the market, and no effort has been made to trace the impact of consuming such food. Nevertheless, there has been an apparent “marked deterioration of health” reported with the introduction of GM crops based on inspection of official data (see [10] Scientific American Disinformation on GMOs, SiS 60).   It is reasonable to assume that the synonymous mutations may have created proteins with adverse phenotypes and contributed to the deterioration in public health, and this must now be thoroughly investigated together with mandatory labelling of GM products.

Article first published 25/11/13


  1. Parmley JL, Hurst LD. How do synonymous mutations affect fitness? Bioessays 2007, 29(6), 515-9.
  2. Cummins J.  Bt toxins in genetically modified crops: regulation by deceit.  Science in Society 22, 32-3, 2004.
  3. Fischoff D. and Perlak F. Synthetic plant genes 1996 United States Patent 5,500,365 pp1-59
  4. Payne J, Cummings D, Cannon R, Narva K. and Stelman S. Bacillus thuringiensis genes encoding lepidopteran-active toxins 1998 United States Patent 5,723,758 pp 1-32
  5. Tsai CJ, Sauna ZE, Kimchi-Sarfaty C, Ambudkar SV, Gottesman MM, Nussinov R. Synonymous mutations and ribosome stalling can lead to altered folding pathways and distinct minima J Mol Biol. 2008 Nov 7;383(2):281-91. doi: 10.1016/j.jmb.2008.08.012.
  6. Ueno M, Kodama EN, Shimura K, Sakurai Y, Kajiwara K, Sakagami Y, Oishi S, Fujii N, Matsuoka M. Synonymous mutations in stem-loop III of Rev responsive elements enhance HIV-1 replication impaired by primary mutations for resistance to enfuvirtide. Antiviral Res 2009, 82(1), 67-72. doi: 10.1016/j.antiviral.2009.02.002
  7. Cladel NM, Budgeon LR, Hu J, Balogh KK, Christensen ND. Synonymous codon changes in the oncogenes of the cottontail rabbit papillomavirus lead to increased oncogenicity and immunogenicity of the virus. Virology 2013, 438(2), 70-83. doi: 10.1016/j.virol.2013.01.005.
  8. Kimchi-Sarfaty C, Oh JM, Kim IW, Sauna ZE, Calcagno AM, Ambudkar SV, Gottesman MM. A "silent" polymorphism in the MDR1 gene changes substrate specificity. Science 2007 315(5811), 525-8.
  9. Chamary JV, Hurst LD. Evidence for selection on synonymous mutations affecting stability of mRNA secondary structure in mammals. Genome Biol 2005, 6(9), R75.
  10. Ho MW, Sirinathsinghji E and Saunders PT. Scientific American disinformation on GMOs. Science in Society 60, 2-4, 2013.

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a tripod?

There are 6 comments on this article so far. Add your comment above.

Douglas Comment left 26th November 2013 09:09:42
Professor Cummins has gathered and presented us with irrefutable proof that the "Substantially Equivalent" classification awarded to crops genetically modified by recombinant biotechnology was unjustified and constitutes a real and present threat to the environment, the food supply and public health. Anyone choosing to ignore these discoveries does so at his or her own peril, physically, scientifically, agriculturally and morally. The case is closed. GM crops have been demonstrated to be a gigantic hoax perpetrated on mankind. The evidence is undeniable and steadily mounting. For how long will the Biotech Industry (and their representatives posing as regulators) be able to ignore these findings? FOR AS LONG AS THE REST OF US LET THEM!

Rory Short Comment left 27th November 2013 03:03:15
My gut feeling when GM foods first started to be promoted as quite okay by companies like Monsanto was, no they are not okay, Monsanto just wants to make money out of them.

Ben Davidson Comment left 4th January 2014 05:05:30
Bacteria have anionic membrane proteins (LPS and LTA) that act like gateways that might block things like Roundup, but which also act as endotoxins in the human body, stimulating subtle - lasting immune stress on every system in our bodies. Whereas bacteria have wholly anionic membranes which allow our cationic antimicrobial peptides to find the pathogens, slipping the endotoxin in among net neutral membranes (zwitterionic) makes it nearly impossible to find. The same thing that makes your body respond/get sick/get stressed is the thing they put in the food. This is so plain I am disgusted… I wish I wasn't 6 years old in 1992.

Beckman, K. Comment left 16th January 2014 08:08:24
I find your interpretation of the articles cited as evidence misleading. [7] States that one hypothesis for explaining why Papillomaviruses use rare codons is to help limit protein expression (such as the E6/E7 tumor suppressor proteins). By changing rare codons in these oncogenes, the researchers suggest that the levels of protein is increased. An increase in the already present proteins is what causes increased immunogenicity and oncogenesis, not a change in proteins structure, and only applies because the protein is “bad” to begin with and is attempting to avoid detection. [6] The protein structure of the RPE is not altered. Synonymous mutations increase the stability of a stem loop in the viral RNA (not the protein) which improves replication. The author also notes that this is considered a secondary mutation that contributes little to the resistance of HIV-1 to enfuvirtide. Again, the main point here is that the protein is not altered. [9] From the title alone, you can see that this doesn’t support your argument of introducing synonymous mutations as being harmful to humans. We're talking about the mRNA here, not the protein produced. [5] is quite interesting. This is not my area, so I don’t feel it’s fair to comment on. Hopefully someone with more of a kinetics background can chime in. Introducing synonymous mutations is a GOOD thing. Using an organism’s preferred codons helps to achieve faster translation rates and higher accuracy. It would be worse to use, for example, an E. Coli gene in a higher-order plant in it’s unmodified form as it is would not be optimized for the host environment and could result in increased protein misfolding and reduced production efficiency (if expression occurred at all) In short, the journal articles you cited as evidence only shows that the rate of protein production (in plants, not humans) can possibly be altered due to synonymous codons. But to put it simply, the only way this is a bad thing is if the evil GMO corps went to the FDA and said “please only test this little amount of protein from the bacteria, we’re sure our plants only produce this much”, which is not the case. The protein produced is identical.

joe cummins Comment left 16th January 2014 08:08:30
Thank you K. Beckman for your provocative comment. I will not here go into a lengthy list of references showing that synonymous mutations are not neutral as you imply. The problem with synonymous mutations in genetically modified (GM) crops is that the transgenes obtained from bacteria have been tuned in those crops using synonymous codons so that the crops will produce quantities of a desired protein. Those altered transgenic proteins are loaded with synonymous mutations. The transgenic proteins were not directly tested for their toxicity but instead surrogate proteins from bacteria were tested on the assumption that the synonymous mutations were neutral. For over a decade the assumption that synonymous mutations are neutral has been proven to be wrong. Simply google synonymous mutations and human disease in pubmed, and you will see a vast array of human pathologies brought on by synonymous mutations. It is frankly insane as you suggest to assume introducing synonymous mutations to be a ‘GOOD’ thing. Using an organism’s preferred codons helps to achieve faster translation rates and higher accuracy, as well as correct folding. ‘The protein produced is identical’ is not true as proteins can be folded and/or assembled differently due to synonymous mutations. What should be done is to test the actual transgenic proteins. The fact that millions of people have been exposed to GM crops loaded with untested synonymous mutations does not imply safety of the crops because the GM crops have not been labeled and that failure to label has prevented those injured by the crops to be identified and to obtain redress for their injury.

Debbie Morton Comment left 2nd May 2014 00:12:37
I am not a scientist; I am a consumer. I have definite negative reactions to GMO corn and soy. Unfortunately, there have not been any scientists at my door when these reactions happened. I do not have negative reactions to non-GMO corn or soy. I am not willing to get IBS, cancer or otherwise sick waiting for 'official' confirmation of harm from GMOs. Get those rat and pig studies done beyond criticism please! Hurry; my friends and family members are dying.