Background: This series contains brief reports on each of the "600 studies" held on the GENERA database, published at Bifortified.org, in the order in which they were listed on 18 Sept 2013. The general purpose is to counter the extraordinary claims that the entries on this list demonstrate the human food safety of GM crops. See the Introduction to this series [link].
Great Big List of Studies: Entry 13 – Anilkumar et al 2010 [full text link]
Anilkumer et al reported feeding a "Bt cotton" fodder to 16 sheep in a trial of 32 over a 90 day period. At 30 days, 60 days and 90 days they took 5 biochemical measurements. By their report two of these measures were indicators of liver health, two were indicators of kidney health, and one of muscular health. The study reported findings of no detrimental effects on these indicators. Before examining applicable findings of this study it is necessary to look at the bigger story behind it.
The introduction of the study gave reference to reports in Indian print media in 2007 that sheep had died in regions of Andhra Pradesh after grazing on leaves and pods of harvested Bt cotton plant residue in fields. However incidences of mortality and morbidity had proceeded over some years.
In a document dated 28 January 2010 Dr Sagari R Ramdas gave a riveting account of the extensive efforts in difficult circumstances to trace the morbidity and mortality of sheep grazing on various Bt cottons, beginning in the 2004-2005 and proceeding through to the 2008-9 season: "Bt Cotton and Livestock: Health Impacts, Bio-safety concerns and the Legitimacy of Public Scientific Research Institutions" [pdf link] [html at GMWatch]
From her ANTHRA biography: Sagari R Ramdas trained as a veterinarian (1986) and has a Masters in Animal Breeding and Genetics (1991), from the University of California, Davis, USA. She has worked with rural and adivasi communities as a field veterinarian, trainer and researcher on livestock and peoples livelihoods. [link]
Ramdas is a co-director of ANTHRA, an organisation started by a team of women veterinary scientists in 1992 to address the myriad constraints that faced rural livestock rearers. The initial aim of the organization was to search for alternative systems for delivering livestock health and management practices to poor people, especially women in rural areas. [link], ANTHRA has research programs that include Bt Cotton and Its Effects on Livestock [link]
ANTHRA was heavily involved in the epidemiological search for answers, which included the de-worming and vaccination of 8,000 sheep against multiple diseases by the 2008-9 season in an effort to eliminate confounding factors. Loosely denoted as the "Bt Cotton Syndrome profile (BCS)" Ramdas documented the history and symptoms:
- "The symptoms typically began in sheep and goats with dullness, after 3-4 days of grazing on Bt Cotton fields, followed by nasal discharge, cold, cough, respiratory distress (in some cases) and red urine in some cases. There was no fever. 2 bullocks also suffered from similar symptoms. "
The conclusion after post mortem of the dead animals:
- "It appears as if some "stress factor" has affected individual animals resulting in reduced immunity, eliciting a possible allergic response in these animals, manifested symptomatically in the form of cold, cough and nasal discharge. Intense stress also probably results in the occurrence of Pasteurella haemolytica, in some of the animals with resultant death."
Ramdas called for "a more substantial study both at the field as well as "ex-situ" simulation level, which cannot be done independently by 1-2 research organisations."
In the second half of the document Ramdas detailed the official regulatory and research response, offering a scathing criticism
- "a majority of our premier national institutions and the larger public research and development system, have lost complete legitimacy, and are morally bankrupt."
which became more direct in a 25th March posting on an Indian human rights law site under the headline "Science Adulterated!" [link]
In both Ramdas referred to a "Bt cotton" feeding study on sheep, referenced in the second as
- Reddy Gopal, A. et. al. 2008. "Studies on the toxicity of Bt cotton plants incorporated in the feed of small ruminants". Project Report. Venkateshwara Veterinary University, Tirupati. Department Of Pharmacology & Toxicology, College Of Veterinary Science; Rajendranagar , Hyderabad-30.
and followed in each case by criticism of the study's inadequacy.
I suspect the Anilkumar study on the Great Big List was a delayed journal publication of some of the results – though published in 2010, the references to sheep morbidity were to 2007. The corresponding author was Prof. A. Gopala Reddy, Department of Pharmacology and Toxicology, College of Veterinary Science, Rajendranagar, Hyderabad, Andhra Pradesh, India.
This study was seemingly a placation response to the media reports of sheep illness and deaths, saying
- "Cultivation of Bt cotton has raised fears among the shepherds and other livestock owners over the safety of Bt cotton plants and their by-products. Therefore, a toxicological study was carried out to evaluate lipid profile, protein profile, and biomarkers of organ damage in sheep by feeding Bt cotton plants."
There is very little detail provided by the study, and at this distance the generally broad lack of rigor in the reporting of the methods limits the interpretation of the study's findings. I have not been able to find a web version of the full Reddy report in which there might be more detail beyond this published study.
In a study seeking to find answers to the epidemiological experience one would expect the trial to follow the observations. For example, one might expect the sheep to be trialled on various standing post-harvest Bt cottons. Yet this does not seem to have occurred. There is quite a lot of essential detail missing from the report of the study. For example...
The study report did not identify the test feed.
There are many GM cottons containing Bt genes, yet the test feed was only described at "Bt cotton". Ramdas alluded to sheep feeding on one and/or two of Monsanto's products named on the CERA GM Crop Database as Bollgard [link] and Bollgard II [link]. She also referred to GM cottons from other companies.
Bollgard seems to be a description for a multiple set of GM lines known as "531", "757" and "1076". Even though a gene used in the GM transformation is often called "Cry1Ac", it was actually a chimeric inspiration of two different bacterial genes, "Cry1Ab" and "Cry1Ac" of which the codons had been substantially altered. The GM transformations had gone substantially awry in each of these lines, though by Monsanto report the crops produced functional GM proteins. According to FSANZ
- Line 531 contained one complete Cry1Ab/c gene with another partial gene in close proximity, but in reverse.
- Line 757 contained one complete Cry1Ab/c gene with a partial gene at separate site in the genome.
- Line 1076 contains one complete Cry1Ab/c gene with another incomplete Cry gene in tandem at the same site.
There may also have been a line known as "1849" which contained a full and partial copy of a Cry2Aa inspired gene. [FSANZ link]
Bollgard II was developed by engineering line 531 with yet another genetic construct containing a toxin gene inspired by the Cry2Ab bacterial gene fused to genetic material for a chloroplast transit peptide, according to FSANZ. [FSANZ link]
Two other GM lines may have been available during this time: A Syngenta GM cotton "Cot102" contained a "Vip3A(a)" inspired Bt gene and a Dow GM cotton "MXB-13" contained a Cry1Ac and Cry1F inspired Bt genes.
The GM industry itself represents the case that GM crops should be approved on a case by case basis – some might be less risky (or more efficacious) than others. Testing requires identification.
The study report did not characterise the test feed.
It is usual for a study to present information that verified the test substance prior to the trial. One typically reads statements that the genetic material characteristic of the GM lines was confirmed. Without this reassurance we may wonder if the GM feed was actually tested. This report provided no information.
The study report did not identify the control feeds nor verify them to be free of confounding GM presence
The "green fodder" used as a control feed was not identified. As such it doesn't give the reassurance that, in the extreme example, the researcher may have unwittingly used green fodder from a GM Bt maize that could mask the results. This is not uncommon in animal GM feeding studies, where the test feed may be supplemented with soy or corn which could be GM but is neither described nor tested. Control feed of non-Bt cotton was also reported to be used but there was no report of it being verified to be free of GM contamination. Concentrate feed was also described but there was no description of contents nor statement on GM status that could mask results.
The study report did not characterise the nutritional contents of, nor toxic residues in, nor details about the raising/sourcing of the feeds.
The study seemingly was not trying to verify the equivalence of different feeds, but a rigorous study that is attempting to analyse a health response to a test feed should ensure that the feeds are equally nutritious in terms of the usual factors such as protein and component amino acids, fat, carbohydrates and important micronutrients. There was no such statement. All animals were reported to receive the concentrated feed but the fodder could have undergone various pesticide/fertilizer treatments or grown on soils with toxic burdens, all of which may have influenced the performance of the animals.
The study report did not describe the health/life outcomes of the animals, nor any veterinary response that may have been made during the study period.
Even though this study was seemingly conducted to assuage concerns of poor farmers about the health and life expectancy of their animals grazing on various Bt cottons, there was no information reported about the health of the animals through the study period. It is usual for animals to suffer ill health and/or be removed from the trial or die in trials. All we know is that 32 animals began the study, and that average results for participating animals were put forward. This report doesn't tell us how many animals finished and in what health, on indicators beyond the five biochemical values tested. We don't know if the symptoms as described by Ramdas emerged and were treated, successfully or unsuccessfully, or not.
- The symptoms of the "Bt Cotton Syndrome" had already been established early in 2007, as documented by Ramdas:
"On 2nd February 2007, veterinary doctors from Anthra and representative from the AP Sheep and Goat Federation visited the villages [...]. Shepherds reported the occurrence of similar symptoms as the previous year, in sheep which had grazed continuously on Bt cotton for 3-4 days. Symptoms in order of appearance included anorexia, cough and nasal discharge from the second day onwards, occasional difficulty in breathing, nasal discharge which was mucopurulant and / or blood tinged , red coloured urine, bloat from the 3rd day after grazing on the fields, eye swelling / face swelling, difficulty in standing and diarrhoea in a few sheep. There was mortality in some of the morbid sheep after grazing on Bt cotton leaves and pods continuously for 3-4 days."
The study report did not report on the animal health conditions at the beginning of the trial, nor report on the housing or feeding method of the animals, nor total consumption.
A trial looking to reassure farmers should have verified the equivalent condition of animals at the beginning of the trial on variables such as weight, health, and the five biochemical values tested, yet these values were not put forward.
Reading a study such as this one would like to read that the animals were securely penned such that they could not access the feed of others. We would like to read that the animals trialled on the GM cottons actually ate the feed, or whether they could get sufficient nutrition by eating just the concentrate and/or non-GM green fodder. This detail was not provided.
I would have liked to have seen typical levels of the biochemical indicators in comparative healthy animals.
By its report the study did not seem to attempt replicate the epidemiologically observed events
I can't be absolutely sure that this study intended to say that sheep feeding exclusively on post-harvest GM cotton would not be adversely affected. But if so the conditions of the trial didn't adequately represent the epidemiological observations. To mimic the epidemiological observations the GM crops on which the sheep had been feeding at the time they fell ill should have been identified and used in trials, including pesticides that may have been used. Further the sheep (of comparable age and feeding history) should have been trialled on an exclusive diet of post-harvest standing GM cotton. The concentration of the GM Bt proteins varies throughout the growing season and vary on other experiences such soil moisture and associated climate, and a quick preliminary search says that anti-nutrients such as gossypol do likewise. There can also be interactions between other plant toxins such as gossypol and GM proteins and their effect on insects which may affect animals also. Measurements should have been relevant to the symptoms observed, such as feed rejection, lung congestion, nasal discharge etc.. Ramdas raised many more issues in her document and article which are well worth reading.
In the absence of further information, and trusting the report offered, and that assuming equivalent measures applied at the outset, one could say that the average results on five biochemical measures taken of sheep with access to a Bt cotton green fodder food source, beginning with two groups of eight sheep, did not show different results to sheep that did not have access to a Bt cotton green fodder source. However in the absence of any report on health or life outcomes of the sheep one could not conclude that there were not any other adverse results of having access to Bt cotton feed. Further this report should not be used to assuage concerns of sheep farmers in India, since the epidemiology matching of observed experience does not seem close.
I could be corrected.