Andy Smith
MIne-action specialist

The truth about "red plants"
or the "purple weed system"


[Please email me if you want to see my original report of the 2003 trials.]

I was the official independent observer at the ARESA Red-Plant detection trial that took place in October and November 2003 in Angola. I was asked to advise in designing and carrying out the trials, and also in ensuring that there was no cheating or false recording of events. I was asked because I had advised the developers for several years (unpaid) - which I had done because I wanted them to develop a valuable QA tool. I had also built and laid the targets in the test-area, so could design trials to make maximum use of what was available. ARESA assured me that they had developed a complete system that was ready for utilisation - and the field-trial was meant to confirm this.

I had found the original ARESA concept interesting and worked (unpaid) to agree ways in which it could be used. As with all methods intended for field use, it is essential to design-in practical deployment methods from early stages of research and development. The seed was to be selected as a temperature hardy weed modified so that it could not flower and produce pollen or seed. The seed was to be mixed in a slurry of nutrients and water that was sprayed out to form a "skin" on the ground - and did not need any further intervention. (The "skin" prevented evaporation and was a variant of something that already exists as an agricultural tool.) In theory, plants would have grown and turned red even on top of an exposed mine. The projected use was entirely for QA of an area already cleared manually so free of undergrowth. No mechanistically objective and repeatable QA system currently exists, so this was to fill a need for a means of QA that was demonstably fair and unbiased, and also cheap. The cost was to be lower than 10c per square metre. Found item markers were to be left until the QA was completed so that red plants around them could be ignored.

I am keen to gather missed-mine data so that some genuine assessment of the relative "efficiency" of the various demining methods/systems would be possible. Everyone talks about efficiency as if it only involved relative cost, safety and speed. In fact, the baseline must be "was the job done well?" - because it is obviously easy to be fast, cheap and "safe" if the job is not done at all. If the plant had worked it could have also been a unique tool for checking the performance of men, dogs and machines in trials. If it had been as low cost as it was supposed to be, it could have provided a means of demonstrably unbiased and objective QA for general use.

Right up to the moment of field deployment, I was assured that the system was ready for realistic trial. In fact, it was not.

The trials did not go as planned and I agreed to keep my report "in-house" as long as the ARESA Biotech crew did not misrepresent events at those trials in order to try to raise money from the general public. They have done so, and accordingly I have made the report available on request. I believe that this is essential for the maintenance of my own integrity. The first journalist article covering the trials and drawing on my report was published by a magazine in Denmark in February 2004. Subsequent press and TV reports have varied in their "angle", some trying to talk up a Danish success, others being somewhat more realistic about what has become an ongoing scandal. The trials were recorded by a Danish TV crew, but most of their footage was taken from them when the trials were aborted. They have subsequently produced a documentary about the plants that is unbalanced and biased towards supporting ARESA. I am featured in that documentary in a manner that shows selective editing and a cynical disregard for the truth.

Events surrounding the trials were complex but they yielded no information about the crucial point - which is the plants' ability to react to High Explosives - TNT or RDX. The professionalism of those involved was shown up badly - as was my own because I was associated with it. If you read the Annexes you will see that the advance planning by ARESA seemed professional - but I accept that I should have had the courage to withdraw entirely when the reality of their ill-prepared efforts became apparent. I did halt the trials for safety reasons, but was persuaded to allow them to continue so that ARESA had a chance to recover from a bad start.

After three years and a lot of money, what ARESA had done with the concept in terms of making a "system" was, sadly, nothing at all. At the same time, their projected field cost of using the system has risen to twice the real cost of current manual clearance. There was no attempt to distribute seed in a slurry of nutrients and water that would coat the ground. Seed was distributed by hand in an uncontrolled manner. They chose to use cress seed because the genome was already in the public domain, and it was patently the wrong plant. Watercress is not very temperature hardy, and is very highly dependent on water. To achieve any growth, huge quantities of water had to be applied daily with the trial lanes protected by shade netting. Both the daily application of large volumes of water and the need for shade netting would have made the system obviously impractical in a mined area. Many other methods would have been cheaper. But critically, even when pampered well beyond what was realistic, the modified watercress seed did not grow to more than a few millimetres in height. The selected plant already grows "red" when "stressed" (ie, without sufficient water or nutrients) so a colour change would not have actually proven anything, but there were no colour changes evident at the end of the trial.

When using genetically modified seeds there are safety concerns because the plant will enter the food chain (birds eat the seedlings) so the benefits must demonstrably outweigh the disadvantages associated with the genetic modifications. In this case, the possible disadvantages to humans (pointed out by ARESA in Annex A of their planning document) included introducing resistance to antibiotics to an area where antibiotic use has been so low that it is probable that no resistance currently exists. During the trial, genetically modified material (seeds) were spread in an uncontrolled way.

My report on the trials was not written for publication. I hope you will be enlightened by it - and agree that the concept of making safe genetic modifications to plants to detect the presence of High Explosive does deserve exploration by professionals.

The main report is only about eight pages. Annex A is interesting because it details what they intended to do - providing quite a contrast with what they actually did.... The rest you can probably skip - bar noticing that tens of thousands of litres of fresh water were applied over the period, (but the quantity was not actually measured). Anyone who wants the full report should email me and ask.

 

FINAL DRAFT

Not for distribution or quotation without the express permission of the author. 23rd November 2003

Independent Observer report

On the ARESA Biotech testing and data acquisition "Development trials" conducted in Ondjiva, Angola (October/November 2003)

Author: Andy Smith (AVS Mine Action Consultants)

Using IMAS definitions, this series of tests, trials and data-acquisition exercises are defined as "development trials" - assessing the current status of the technology rather than its immediate field utility. For this reason the criteria for success vary from "Acceptance trials" and measures of success cannot be taken to indicate the field-readiness of the technology unless specifically so stated.

The largest plant after 21 days growth.

Organisations directly involved:

ARESA Biotech - trial originators and the Client [ARESA]
MgM (Menschen gegen Minen) - Host and site-supplier [MgM]
AVS Mine Action Consultants - Observer and Advisor [AVS]

 

Contents

The independence of the author
Summary
Aim and scope
Authority
Constraints and limitations
Trial format
Criteria
Execution
Conclusions
Recommendations
Annex A: ARESA pre-trial instruction and risk assessment
Annex B: TRIAL INSTRUCTION - ARESA "development trial" - AVS, Sept/Nov 2003
Annex C: Pre-Trial Assessment - Andy Smith 30/09/03
Annex D: AVS/ ARESA Biotech contract
Annex E: The MgM detector test-area
Annex F: Angola ARESA development Test diary - AVS 2003

 

The independence of the author

The role of Andy Smith (AVS) was described in his contract with ARESA Biotech as:

Observer: AVS shall document the experiment carried out in Angola by writing up a report that describes the test results achieved. It is of high priority that the report emphasises on the sensitivity of the plant line(s) grown in respect to the different type of mines detected/not detected. Furthermore, the report shall contain reflections upon observations made in respect to the practical use of the technology (seeding, watering techniques and systems, water requirements, plant growth, etc.) in light of the complexity when operating in a landmine infected area.

Technical advisor: AVS shall act as an advisor on the technology in development on behalf of his in-depth experience within de-mining. Hence, AVS shall challenge existing ways of thinking, and discuss new ideas generated with the employees of the company.

Safety advisor: AVS shall act as an advisor on safety related issues due to his experience from operating in landmine infected areas. Hence, AVS should guide the employees of Aresa and employees of Bastard [the film company] to behave as safely as possible in Angola. In case of any doubt, AVS should consult the MGM safety Officer Ken O'Connell for advice on safety related matters.

The full contract is reproduced at Annex D.

 

Disclaimer
Andy Smith (AVS) was retained to produce an independent report on all the ARESA tests, trials and data-gathering exercises conducted at the MgM camp in Ondjiva, Angola during October and November 2003. As an "Observer" and advisor, AVS had no control over the content and conduct of the trials. Further, AVS declares that he has no "interest" in the outcome and so was able to apply disinterested and objective observation overlaid with opinion and advice. The fact that AVS has responded to requests to offer advice and opinion (recorded in this document) does not imply any acceptance of responsibility for the conduct of the tests/trials, the materials used, or any consequences (legal or otherwise) arising for the activities of ARESA Biotech or MgM in Angola or elsewhere.

AVS reserved the right to withdraw from his contract if activities that he considered to be physically or ecologically irresponsible or unsafe occurred.

 

Summary

Responsibility for the detailed aim and scope of the ARESA "development trial" rested entirely with ARESA Biotech. Although they were advised on some issues by AVS and MgM, there was no obligation for them to heed advice. I prepared a formal "Trial Instruction" based on ARESA documents, reproduced at Annex B. However, ARESA had not described the trial aims until shortly before my departure, and so did not receive the Trial Instruction in time to take notice of it before they deployed.

MgM provided a dedicated detector test-area with appropriate facilities (detailed at Annex E), including a clean potable water supply and some staff assistance. MgM provided further site facilities.

The conduct of the trial was such that I resigned from my contract after 24 days (14 days at the site). At that time, I advised MgM that it was not possible to safely halt the trials without allowing the seeded plants to grow so that they could be seen. [No means of destroying seed that not not yet germinated was available.} MgM made my continued presence a condition of the Trials' continuation, so I felt obliged to remain as unpaid observer. I did so with enhanced authority to prevent further spreading of GM material outside the test area.

Ambient conditions combined with the methods of seed dispersal combined to limit plant growth and the Trial failed to realise any of its stated aims. It was halted prematurely by the decision of ARESA staff to leave when the BASTARD television crew were unexpectedly obliged to leave by the apparent hostility of the authorities. ARESA staff left the area in haste, leaving GM material. AVS remained for a further seven days, watering the test area and tending it as required by ARESA. No further significant growth having occurred, I then (by agreement with ARESA) destroyed the stunted growth and made the site as "safe" as possible (within the constraints of my limited knowledge of biotech safety). [In addition to the inadequate weedkillers left by ARESA, I used a gas-torch to burn off the area and its surrounds, including dry areas with no growth where ungerminated seed was likely to rest.]

 

1 Aim and scope

The aim of the ARESA development trial can be summarised as follows.

A) To establish whether the genetically modified plants (known as A line: Genotype PAP1/tt4/nii-CHS/ga1-3/ExenA (designated LA); B line: Genotype PAP1/tt4/nii-CHS/ga1-3/ExenB (designated LB), change colour when grown above hidden landmines and UXO in the test area, and to assess the reliability of any colour change that does occur.

B) To establish whether the genetically modified plants produce flowers.

C) To determine the efficacy of the proposed seed distribution system and reach a view on whether it and its ancillary parts have the potential to achieve field utility.

D) To determine whether either of two flowering genetic variants have the same or better potential than the main (LA and LB) variants by reference to their growth in the prevailing conditions.

E) To devise and implement a watering regime on unmodified plants to determine the minimum acceptable water requirements and also the requirement that is optimal for fast growth.

F) To gather data to assist in further ARESA research.

G) To gain some field experience of the way in which Humanitarian Demining is controlled and conducted.

H) To facilitate the recording of a video record of the testing.

The Trials took place over a period of 25 days at the dedicated MgM detector test area in Ondjiva, Angola.

 

1.1 Authority

The relevant Authority for the trial was the provincial Governor (Cunene) and the Department of Agriculture in Angola. A further authority was MgM, the host who assisted with the requesting of permission from the relevant authorities and imposed some of their own constraints on ARESA activities.

Those involved in the trial were:

ARESA Biotech - trial originators and the Client

AVS Mine Action Consultants - Observer and Advisor

MgM (Menschen gegen Minen) - Host and site-supplier

ARESA Biotech assumed full responsibility for the design of the trials and the day to day conduct of data-gathering and plant care. A TV documentary crew from the Danish company BASTARD accompanied ARESA personnel to the site and made a record of what occurred. Filmed interviews about the testing were made with the permission of the senior ARESA staff member (Carsten Meier) who had ultimate authority over what went on the public record. Carsten Meier for ARESA placed no restrictions on the interview content. MgM had editorial rights over the footage eventually used, so I responded frankly to questions from the TV crew. [The TV crew subsequently reneged on its agreement to give MgM editorial rights. Ed.]

 

1.2 Constraints and limitations

Constraints intended to be appropriate for the testing of genetically modified material were adopted by ARESA in their pre-trial documentation (see Annex A).

Prior permission for testing GM plants was sought from the Provincial and National Authorities. The request to the National Authorities was made less than two weeks prior to the trial's start date. It was known that Angola is a signatory to a SADEC treaty limiting GM crops and produce, so it was considered unlikely that the Government would approve the trial. Permission was granted at Provincial level and no direct denial of permission was received from the National Authorities. No formal response to the request to hold the trials had been received from the Angolan National Authorities at the time the trial finished.

ARESA personnel, myself and associated visitors to the MgM site were obliged to obey MgM camp rules and while staying in the area were requested only to travel in company with MgM representatives or after seeking prior advice.

The greatest predictable constraint placed on the trials was the perceived risk of the escape of biologically modified material into the environment. A second major constraint was the need to carry out experiments in areas where landmines had been concealed for a long period of time.

 

1.2.1 Unanticipated constraints

The trial was constrained by the pre-prototype design of ARESA equipment brought to the site and by the fact that it had not been fully assembled and/or tested prior to its arrival. Examples were the seed-distribution method, insect-proof netting cloches, area-marking signs and the watering systems.

The fact that the netting was not "pollen proof" was a surprise because ARESA management had previously used that expression to describe it. This may have been a simple communication difficulty caused by their use of English as a second language. The pollen is far smaller than the anticipated insect threat and could be wind-borne through the netting used. This increased concern over the escape of pollen if the plants were allowed to flower.

 

1.3 Trial format

The trails took place over a 25 day period. By request, I was on site for one week after the ARESA staff left, and had been engaged for 10 days prior to the formal start of the trial in the test area.

By agreement, the positions and identity of targets concealed in the test-area was made known to the data-gathering team.

Details of the prior preparation of the lanes containing test-targets, their sifting, drainage and depth are in Annex E along with details of the targets, their depth and their positioning.

Although an ARESA staff member had visited the site once before, most aspects of the site conditions required variations to be made to the pre-trial planning. Perhaps most obvious of these was the fact that all tests involving GM and mutant seeds occurred within the double-fenced detector-test area (the dog-training area was not used).

ARESA took sole responsibility for the conduct of the trial, variations to its objectives and the methods of gathering data. I advised constantly, as detailed in Annex F.

MgM retained sole responsibility for safety aspects, excluding those pertaining to any biological hazard resulting from the test materials and with the proviso that all visitors acted at their own risk. No MgM staff member was assigned to the team, so that safety relied heavily on heeding advice and applying self-discipline.

 

1.4 Criteria

I was not asked to develop trial criteria. To develop detailed judgement criteria without greater knowledge of biotechnology would not be appropriate. However, I was asked to comment on the outcome and will do so with reference to the following:

1. A perceived change of colour occurring above buried ERW.
2. The consistency of any colour change with reference to the specific ERW beneath.
3. Whether or not the G3 modification flowered.
4. Whether any random colour changes occurred without the presence of ERW.
5. The practicality of the deployment method in the field.
6. The apparent professionalism with which the trial was conducted and records were kept.
7. A subjective judgement about the potential utility of the technology.

Methodology, evaluation and acceptance criteria were determined and judged internally by ARESA, but are commented on by me when appropriate. I advised on all aspects on which I felt qualified to comment and some of that advice is recorded in this report.

 

2. Execution of the Trials

The following is a simplified Trial schedule:

AVS left for South Africa in pursuit of a visa on Sunday 28th September
AVS arrived in Namibia on Saturday 4th October
AVS arrived in Ondjiva on Sunday 5th October and began pre-trial preparations.
ARESA and BASTARD staff arrived at the site on Wednesday 8th October
Seeds were first spread on Sunday 12th October
AVS resigned from his "Observer" contract with ARESA and became site "policeman" and QA on Monday 20th October
ARESA and BASTARD left the site on Monday 27th October
AVS left the site after "cleaning up" on Tuesday 4th November

A record of the events during the trial is available.

The trial was brought to an end by the decision of ARESA staff to leave when it became clear that the BASTARD film crew were no longer welcome. The ARESA staff decided to leave with the film crew. By that time the failure of the plants to grow was obvious, leading me to conclude that ARESA took advantage of an "excuse" to leave so that the trials would be "unfinished" rather than a failure. The ARESA staff said that they left because they also felt under some threat from the local authorities. The reported threat was only to the film crew. The Provincial authorities showed no apparent interest in the activity at the test-area after the ARESA and BASTARD personnel had departed.

 

2.1 Preparation

ARESA preparation was generally inadequate. Examples are:

The varied seed brought by ARESA was not prepared in the way that was documented in their pre-trial papers. The GM seeds were mixed together and had not been coated or "pelleted". Coating the seeds made them large enough to be seen and so relatively easy to disperse evenly. Uncoated seeds were a fine dust that was readily wind-borne and so were widely dispersed unintentionally. The entire conduct of the trials was affected adversely by ARESA's failure to bring appropriately prepared seed.

The schedule was constrained by the late arrival of a box of equipment that was only available to be unpacked on Sunday 12th October. It did not contain everything that AVS had been told it did, which implies that the ARESA staff did not know what had been packed.

Tools and equipment used by ARESA were not appropriately "scientific" to allow them to achieve the aims stated in their pre-trial documentation.

 

2.2 Achieving the aims

It was not possible to achieve the stated aims for a variety of reasons. The aims and the results are summarised below.

A) It was not established whether the genetically modified plants changed colour when grown above hidden landmines and UXO in the test area. At the end of the Trials, AVS was told that the root needed to reach a length of 20cm in order for a colour change to occur. The longest root was under 20mm in length at that time. The seed from which that plant has grown had been planted on 12th October (it was re-seeded on 21st but the largest plants presumably came from the first seeding). The Trial ended on 2nd November so that the plants had 21 days to grow. The scientists had expected the plants to have matured within that time-frame, achieving a height above ground of around 15cm. At the end of the Trial, the above ground height of the largest plant was not more than 3mm.

B) It was not established whether the genetically modified plants would produce flowers. Pre-trial assurances that the genetic coding had been modified with what was nicknamed a "suicide-gene" were varied when ARESA was on site, and their scientist announced that the length of the day might over-ride the "suicide-gene". The day length was well under 14 hours from dawn to dusk (and so quite predictable).

C) It was determined that the seed distribution "systems" tried with the GM seeds were either unsafe and/or failed to achieve an adequate seed distribution. They also failed to achieve distribution of seeds that had been suitably affected by the growth hormone with which they were supposed to be "activated".
The spray distribution of pelleted seeds (using a compressor and water/seed spray mix imported by ARESA) did distribute seeds fairly evenly over a ten metre box (with some overspray of seeds which flew further than the water). Despite being unmodified and pelleted seed, none of this seed grew. It is not possible to determine whether the distribution system played any part in that failure.
The use of a DriWater slurry to aid dispersal was tried without success. DriWater is a powder that, added to water, forms a transparent slurry that is physically akin to wallpaper-paste. It is intended to provide a "skin" over the ground and limit evaporation. Failure may have been because the growth hormone was added after the dispersal of seed in DriWater, and so the hormone did not reach the seed.
Although the DriWater was still wet on application of the hormone, its gelatinous nature seems likely to have prevented contact. Alternatively, the method of applying the hormone may have limited its distribution. It was applied mixed with water and using a bucket with holes in the bottom. A later, second application of hormone appeared entirely ineffective.

D) It was not possible to determine whether either of two flowering genetic variants had the same or better potential than the main (LA and LB) variants by reference to their growth in the prevailing conditions because they were not grown under the same conditions. The flowering variants were grown under white "insect-proof" netting which provided (unmeasured) shade and UV protection. The seed was dispersed over a much smaller area that had been "ponded", preventing seed dispersal. The plants appeared more quickly, but did not appear to grow any larger than the GM variety. The flowering plants were hoed out of the ground by ARESA staff when hurrying to make their premature departure, so AVS was unable to measure root length.

E) It was not possible to devise and implement a watering regime on unmodified plants to determine the minimum acceptable water requirements and also the requirement that is optimal for fast growth. This was because, no matter how much water was applied, the plants appeared reluctant to grow. The ARESA presumption that water availability would be the most significant growth factor appears to have been erroneous.

F) It is not possible for AVS to determine how much data of assistance to further ARESA research was gathered. However, the absence of any scientific instruments to record the context of the trials, and the failure to constrain and record variables, must mean that any data gathered is of questionable utility.

G) ARESA staff did gain some field experience of the way in which Humanitarian Demining is controlled and conducted, but very little.

H) The aim to facilitate the recording of a video record of the testing was not achieved despite the presence of the BASTARD television crew because all film and associated records were confiscated from them as they left the country (or so they subsequently reported).

The more scientific aims could not be achieved because appropriate instrumentation to record the context of the trial was not brought by the ARESA team. The more obvious failings in this regard are listed below:

1) No means of measuring temperature or humidity was on site (an AVS thermometer measuring up to 50C was borrowed, but ground temperature regularly exceeded 50C).

2) The means of measuring soil-moisture utilised two cheap gardening prods that measured electric resistance between probes and used a scale that was not understood. Without an analysis of other soil properties, they gave only the crudest indication of moisture content. The probes were left behind at the site, so preventing any later comparison of their results with another instrument.

3) No means of measuring wind-speed was on site. Blustery wind affected dry seed dispersal significantly.

4) No means of measuring UV intensity was used. This meant that the influence of the shade-netting brought (and second type purchased locally) could not be measured.

5) The compressor, water tank and spray lances brought by ARESA were not much used, apparently because varying the pressure was not understood, Near the end of the trial ARESA informed AVS that pressure could (after all) be varied. Similar equipment had not been assembled and tested prior to dispatch, which meant that its strengths and limitations were not known. This equipment was not necessary unless a mixture of pelleted seed, Driwater and growth hormone were to be dispersed. The failure of ARESA to bring pelleted GM seed made the equipment largely redundant.

6) No soil samples were taken to allow later analysis of its properties.

 

2.3 Safety

The trials were conducted in a manner that probably dispersed dry-dust seed outside the test area. The trials were also conducted in a way that allowed insects and birds access to the plant heads as they emerged from the soil, and the presence of both insects and birds around the plants (apparently eating) was observed. It is not certain that the birds ate the plants, but the fact that they were attracted to the new growth makes this likely. Bird scarers (ineffective) were only erected at my insistence (after resigning from the ARESA "Observer" contract). The shade-netting (potentially bird-proof) was not kept sealed by ARESA personnel.

ARESA disposed of utensil "washout" water in two pits - so mixing varied seeds and growth hormone in the same small areas. AVS marked the first pit (that had been concealed) and later burnt both pits out using a gas-torch. The effectiveness of this could not be measured.

ARESA did not provide a "pollen-proof" netting for the "flowering" plants as described in their pre-trial documentation. The netting was nominally "bug-proof" but could be watered through and was rapidly torn in places (in any case it was placed on top of bugs in an area where bugs lived below ground). The failure of the plants to grow to any significant size means that there were no obviously adverse safety consequences as a result of ARESA's failure to provide the protection planned.

The "weed-killer" provided by ARESA and used by AVS during the clean-up of the site was not of a "known" safety. The edge of the site is within 50 metres of the MgM camp's potable water borehole. I repeatedly asked ARESA to check the safety of using their locally purchased weed-killer (branded "KOMBAT Wipe Out") with regard to its long-term ground contamination. This request for assurance was repeated to ARESA management in Denmark (seen Annex F). No response was ever received. The weed-killer is "a water soluble concentrate non-selective, foliar, selective herbicide for the control of a wide range of annual and perennial grasses, broadleaf weeds and certain woody perennials". Its active ingredient is "Glyphosate timesium (suffosate) at 480g/l".

The area in which I used the weed-killer was more than 120 metres from the borehole, which reportedly draws water at a depth of 40 metres. Without suitable advice from ARESA, I was obliged to make an uninformed risk-assessment and decided to use the weed-killer. Any seed that may still germinate will not be affected by the weed-killer.

I used a gas-torch, a hoe and weed-killer to destroy all visible plants and burn the top of the soil in areas where seed was known to have been planted. [All damage limitation of this kind would not not been done without my independent decision to do so.]

 

3 AVS Comment

I was asked to comment on the outcome with reference to each of the following topics. The topic is repeated and a comment appended:

1. A perceived change of colour occurring above buried ERW.
This did not occur - possibly because the plants did not reach maturity. However, an 18mm root length did mean that the roots were in proximity to (beneath) at least one concealed device.

2. The consistency of any colour change with reference to the specific ERW beneath.
There was no colour change to be consistent.

3. Whether or not the G3 modification flowers.
No plants approached a stage of maturity at which they could have flowered.

4. Whether any random colour changes occur without the presence of ERW.
No colour changes occurred, random or otherwise.

5. The practicality of the deployment method in the field.
The deployment methods used (including those contrived on site) were entirely impractical for use in or near a genuinely hazardous mined area.

6. The apparent professionalism with which the trial is conducted and records are kept.
The conduct and recording of the trials did not appear professional and did not lead to the acquisition of any accurate scientific data.

7. A subjective judgement about the potential utility of the technology.
I believe that the concept has value, but the plant type used was adopted for expediency (its genome was already known and in the public domain) and the particular plant is not appropriate for the exploitation of the concept. Successful exploitation would require the development of an appropriate slurry dispersal system (including all water, nutrients and seed) ensuring that the plants grew anywhere. Another prerequisite is that the type of plant be hardy enough to grow in most areas of the world.

I had reserved the right to withdraw if activities that I considered to be physically or ecologically irresponsible or unsafe occurred, and I was obliged to withdraw part way through the Trial. By negotiation, I remained as Observer and QA with enhanced authority to ensure that ARESA gave appropriate weight to safety concerns.

In the contract between AVS and ARESA, the following confidentiality constraint is detailed:

In order to control and protect the patent strategy and the communication strategy of the company, AS agrees to treat all knowledge and insight material regarding the use and the development of the technology platform, which may come to his possession, as strictly confidential. Hence, AS will only communicate any information about the company and its technology to a third party, after getting accept from the management of the company.

I (AVS or AS) have acquired no "knowledge and insight material" regarding the use or supposed development of the "technology platform" (GM cress), so I am not in any position where I could either wilfully or inadvertently breach this confidentiality agreement.

The events and conduct of the trial are not covered by this agreement - and I was expressly required to record them with objectivity and honesty. Accordingly, I reserve the right to comment on the events and conduct of the trial as documented herein whenever appropriate. I would consider it (professionally) appropriate to do so if the events of the trial were publicly misrepresented by any other party.

 

4 Conclusions

A "development trial" is defined as a trial seeking to establish whether the concept under assessment has the potential to achieve its stated aims and whether any materials involved are fundamentally flawed in a manner that would render it unusable in the field. It is not possible to state definitively whether the concept under assessment has potential, but it seems evident that the means of applying the concept (European water-cress) has no potential in the prevailing conditions in Ondjiva (those conditions were not recorded).

The trials were conducted in a manner that did not guarantee that genetically modified seed and material failed to enter the wider environment and the food chain.

The Trials did not fully achieve any of the stated aims. There were a number of identifiable reasons for this:

1) Inadequate preparation of seed.
2) Inadequate lab-preparation under simulated conditions.
3) Inadequate pre-trial planning (including the planned use of "controls").
4) Inadequate pre-trial testing of both seed and equipment.
5) Failure to deploy appropriate means of recording local soil and weather conditions.
6) Failure to react in a timely way to local conditions (shade-netting and plant-feed).
7) Failure to ensure the legality of the testing prior to deployment.
8) Failure to deploy appropriate clean-up equipment.

Looking at the details of the results anticipated in the ARESA pre-trial instruction and risk assessment (See Annex A) it is inexplicable that ARESA should have failed to bring as much instrumentation as a thermometer with them. The cost of the trial must have been high, and much of that cost could have been avoided if sensible preparation had been made. That preparation would undoubtedly have led to a delay, which would have had two advantages. The Angolan government authorities would have had time to reply to the request for the trial to take place. And the seed used could have been appropriately prepared (as ARESA had always intended it should be).

The only real result from the trials is the conclusion that the use of a Genetically Modified European watercress for the purpose of identifying concealed TNT and RDX in Southern Angola did not work at all, because the plant would not grow in the (unrecorded) conditions prevailing at the test centre. Lessons to be derived from this are severely limited by the failure to record local conditions and the uncontrolled (and so unrepeatable) methods of seed mixing and dispersal. It simply is not known how the (apparently) infertile soil, its rapid drainage or the UV intensity, ambient temperature, or random seed dispersal and inefficient application of the growth hormone affected these trials.

 

5. Recommendations

I have no recommendations to make to ARESA, who have expressed no interest in receiving this report. MgM should check the test-area for plant growth every month for six months, and destroy by burning any of the plants that appear.

MgM should have the camp's water supply subjected to chemical analysis (checking for "suffosate" or derivative traces) at three month intervals until confident that no contamination has occurred. If contamination is present, another potable water source should be found.

MgM should not consider allowing similar trials to take place without first gaining the permission of the appropriate National Authorities..

 

Annex A: ARESA pre-trial instruction and risk assessment

Reproduced from an original *.pdf file received on 26/09/03.

Comments in square brackets [ ] are additions made by the author of this report.

A R E S A B I O D E T E C T I O N A p s
S Ø L V G A D E 1 4 A ~ 1 3 0 7 C O P E N H A G E N K

T E L : ( 4 5 ) 7 0 2 2 7 7 4 7 ~ F A X : ( 4 5 ) 7 0 2 2 7 7 5 7
M A I L : i n f o @ a r e s a . d k ~ W E B : w w w . a r e s a . d k

Description and risk assessment of the landmine detection test carried out by Aresa Biodetection at the test site of MGM in Ondjiva, Angola

This document shall provide information and insight to the testing performed at the test side of MGM in Ondjiva, Angola for a period of five weeks beginning on the 8th of October, 2003.

Aresa Biodetection (hereafter: Aresa) is a new player within the de-mining community. The company has several years of experience within molecular and physiological studies of plant development. The ideas behind the technology developed by Aresa grew out of The Department of Plant Molecular Biology at the University of Copenhagen. This research group, and thus the employees, is famous on an international scale for its scientific findings. In order to combine the biotechnological technology of Aresa, the company has signed up with Menschen Gegen Minen (MGM) to carry out the first testing of the plant lines developed when growing these in the presence of real landmines at the test site in Ondjiva, Angola.

This test is of utmost importance for the further development of the plant technology of Aresa, which eventually shall contribute to the removal of landmines in Angola and other landmine infected areas. Therefore, Aresa is grateful to MGM for contributing to the research by making it possible to test the developed plant lines in the test lanes containing landmines, and to perform other relevant tests related to the application of the technology. Furthermore, Aresa is also grateful to the Provincial Governor Mr. Mutinde, who has approved the use of the test site for growing the genetically engineered plants at the test side of MGM.

The Chief Technology Officer Carsten Meier will be in charge of any decisions to be made at the test site in Ondjiva, Angola. Carsten Meier will act as the point of contact in Angola regarding the testing, and Simon Østergaard (Managing Director, Aresa) will act as the point of contact in Denmark.

This document will describe the experiments designed, as well as the purpose with these experiments. Furthermore, a thorough risk assessment on the use and the release of the genetically engineered plants in the test site is carried out in a scientifically sound manner with the conclusion that Aresa is confident in the performance of the outlined test - with full attention on reducing and controlling any risks identified before hand and during the testing. Hence, it is ensured that development, handling, transport, and the release of the genetically modified plants, as well as the clean-up procedure after the testing, are undertaken in a manner that prevents or reduces the risks to biological diversity. All types of seeds (genetically engineered seeds and non-engineered seeds (mutant seeds and different ecotypes)) transferred from Denmark to Angola via Namibia, will be packaged and marked separately at the University of Copenhagen, in order to achieve the safest handling of the seeds as possible.

 

Content

The technology platform in brief

The test in Angola:
1. Detection of landmines in the 7 test lanes
2. Seed spreading and maintenance of growth using the water pump system
3. Seed density test
4. Background physiological measurements
5. Examination of the herbicide concentration on local plant growth
6. Growth experiments with different ecotypes
7. Growth experiments with different mutants plant lines
8. Growth control experiments with different genetically engineered plant lines

Risk assessment of the field trial:
1. Information about Thale cress (Arabidopsis thaliana)
2. Why is it necessary to follow up the trials under closed conditions in the greenhouse with trials in open conditions?
3. Description of the inserts (transgenes)

Environmental risk assessment:
1. Which non-target organisms could possibly be effected?
2. Prevention of the spread of genetically modified material in the course of this study
3. To which plants could the transgenic Arabidopsis conceivably transfer its genetic information?
4. Additional genetic markers were introduced to the plant lines along with the genetic origins of interest
5. Assessment of antibiotic resistance
6. Assessment of herbicide resistance
7. Deficiency in production of the growth hormone gibberellin (ga1-3 mutation)
8. Scientific background for the ga1-3 mutation

References

Declaration of the content of this document

 

The technology platform in brief

The technology platform of Aresa is based on genetic engineering of the plant Arabidopsis thaliana. This plant is a widely used genetic model system, and the plant is part of the natural flora all around the world. The plant has been engineered in order to respond to specific outer stimuli present in the environment resulting in a highly visible colour change of the plant within 3-4 weeks if the system is triggered. Thus, the genetically engineered plant serve as a bio-detection system able to monitor specific compounds of interest. Using this technology platform for de-mining, the plant line will react on NO2-groups, which are cleaved off from the explosive molecules that leak out of the buried landmine. Consequently the triggered colour change of the plants will occur in the vegetative stage of plant growth, which enables identification of the specific plants growing in the near by presence of landmines within 3-4 weeks.

The genotypes of the specific transgenic plant lines are shown in appendix A. [Appendix A was not made available.] The different functions of the genetic components that are engineered or introduced by traditional crossing in comparison with wildtype plant lines (such as different ecotypes) are briefly explained below:

CHS: The gene that is responsible for production of Chalcone Synthetase, which enables biosynthesis of the flavonoid derived pigment anthocyanin (red colour).

Nii: The regulative element (the promoter) that is fused with the CHS gene. Thus, the regulation of the Nii-promoter is responsible for expression of the CHS gene in the presence of NO2-molecules.

PAP1: Transcription factor involved in overproduction of Chalcone synthetase.

tt4: Mutation in the original gene encoding Chalcone Synthetase. This mutation serves as a null-genetic background where no Chalcone Synthetase is produced, unless the production is triggered via the Nii-CHS construct.

ga1-3: Mutation in the GA biosynthetic pathway responsible for gibberellin production, which makes the plant unable to germinate and to set seeds making the plant male sterile.

ExenA/ExenB: Bacterial reductases that cleave the NO2-group of explosives, and by then, enhance the sensitivity of the genetic system.

 

The test in Angola

The test that will be carried out in Ondjiva will include growth of the genetically engineered plant line in the test lanes containing landmines, as well as a number of growth experiments on different ecotypes or mutants plants. It is obviously very important to investigate how suitable are the current plants for detection of landmines, but it is also of great value to examine a variety of physiological parameters. The latter is important since the application of the technology should be based on an ecotype that is able to grow optimally under the growth conditions of Africa. The goals and the test procedures are described below:

 

1. Detection of landmines in the 7 test lanes

Goal: To test the genetically engineered plant lines. It is of utmost importance to test the mixture of seeds from the two specific plant lines that contain all the several traits introduced (the genetic cassette responsible for the colour change, reductases that cleave explosives, and gibberellin deficiency). The sensitivity of these two specific plant lines will be explored by growing these homogenously in the test lanes and then observing the expected colour change caused by explosives of the buried landmines.

Procedure: The 7 test lanes are grown with a mixture of pelleted seeds of plant lines LA and LB (the two plant lines only differ in the specific reductases introduced (ExenA/ExenB)).

A line: Genotype PAP1/tt4/nii-CHS/ga1-3/ExenA (designated LA)

B line: Genotype PAP1/tt4/nii-CHS/ga1-3/ExenB (designated LB)

The seed concentration is chosen to be 3000 seeds pr m2. These plant lines are maintained on high water levels. Visual inspection and photos are taken on a daily basis.

 

2. Seed spreading and maintenance of growth using the water pump system

Goal: To establish a method of dispensation of the seeds and maintain growth in a typical grid used in actual de-mining projects.

Procedure: Firstly, pelleted wildtype seeds (not genetically engineered) will be sprayed onto an area using a pump that works together with the water supply. The size of the specific area is estimated to 125-250 square meter, which is dependent on the actual amount of pelleted wild type seeds available, and the seed concentration per square meter. Secondly, non-pelleted wild type seeds in a mixeture of Driwater® will be sprayed onto a minor area to test whether the pelleting process may be substituted with the use of the Driwater®. The growth of these plants will be maintained using the pump and the watering system throughout the test.

 

3. Seed density test

Goal: To determine the minimum seed concentration and the following density of plants on a square meter.

Procedure: The concentration of seeds (# seeds/sqm) and the following density of plants are tested. Wild type (non-genetically engineered) plant lines are used for this experiment. The wild type plant line Col-0 is grown in 5 squares (of 1 sqm each).

The 5 different seed concentrations within these squares are given as follows:
" 1000 naked Col-0 seeds on a sqm (not pelleted)
" 1000 pelleted Col-0 seeds on a sqm
" 2000 pelleted Col-0 seeds on a sqm
" 3000 pelleted Col-0 seeds on a sqm
" 4000 pelleted Col-0 seeds on a sqm

 

4. Background physiological measurements

Goal: To determine the background physiological parameters for the testing Procedure: The following parameters are measured at 09.00 am and at 09.00 pm every day):

" Air temperature
" Soil temperature
" Air humidity
" Soil humidity
" PAR (solar radiation without UV)

 

5. Examination of the herbicide concentration on local plant growth

Goal: To determined the effect of the herbicide concentration on the local plant population.

Procedure: The local plant vegetation is observed when arriving at the test site, where after the specific procedure for this test is designed. Dependent on the observations made at the site, small squares containing local plant vegetation will be marked on the site. These squares will be treated with different concentrations of Basta (10 mg/L, 20 mg/L, and 30 mg/L). Furthermore, the plant lines growing in the test lanes (LA or LB) are tested for the ability to grow in such areas treated with Basta. Thus, LA or LB will be grown in some of these specific squares after the Basta treatment, in order to observe growth of these plant lines in competition with potential local plant vegetation naturally resistant to Basta.

 

6. Growth experiments with different ecotypes

Goal: To examine the growth ability of several ecotypes of Arabidopsis thaliana in order to identify superior ecotypes when growing in the African soil. This knowledge is very important in order to determine which ecotype that is most suitable for this technology (most adaptable to the African growth conditions, growth rates, etc.).

Procedure: Growth of the following ecotypes is examined:
Ecotype: Designated:
" H. Tsukaya (NASC stock # N3963) Jap
" Ita-0 (NASC stock # N1244) Ita-0
" India (NASC stock # N9030) Ind
" Santa C. (NASC stock # N8069) S.C.
" Ler Ler " Col-0 Col

The germination rates and biomass measurements are determined, as well as the water requirements by varying the amount of water added per week in the following concentrations (0, 1, 2, 3.5, 7 l water per week, respectively). Hence, this test requires 30 squares (6 ecotypes x 5 different water concentrations) of 1 x 1 meter, where 81 L of water is added per week in total.

 

7. Growth experiments with different mutant plant lines

Goal: To examine the growth ability of these two naturally selected mutant plant lines of Arabidopsis thaliana in order to examine the effect of the specific mutation on the overall growth performance. The germination rates, the biomass measurements, and the water requirement serve as scientifically controls, which are very important if the plant lines in the test lines (see 1.) are not growing properly.

Procedure: Growth of the following mutant plant lines is examined (in addition to the Col-0 laboratory wild type plant line serving as a control):

Mutant: Designated:
" tt4-58 stop codon in CHS tt4
" ga1-3 GA deficient mutant ga1

The germination rates and biomass measurements are determined, as well as the water requirements by varying the amount of water added per week in the following concentrations (0, 1, 2, 3.5, 7 l water per week, respectively). Hence, this test requires 15 squares (2 mutant plant lines and the Col-0 plant line x 5 different water concentrations) of 1 x 1 meter, where 40,5 L of water is added per week in total.

 

8. Growth control experiments with different genetically engineered plant lines

Goal: To examine the growth ability of the genetically engineered plant lines of Arabidopsis thaliana in order to examine the effect of the introduced traits on the overall growth performance. The germination rates, the biomass measurements, and the water requirement serve as scientifically controls, which are very important if the plant lines in the test lines (see 1.) are not growing properly.

Procedure: Growth of the following genetically engineered plant lines is examined:
GMO lines: Designated:
" F3 genotype pap1/tt4/ F3
" Line 5 genotype pap1/tt4/nii-CHS L5
" A line genotype pap1/tt4/nii-CHS/ga1/ExenA LA
" B line genotype pap1/tt4/nii-CHS/ga1/ExenB LB

The germination rates and biomass measurements are determined, as well as the water requirements by varying the amount of water added per week in the following concentrations (0, 1, 2, 3.5, 7 l water per week, respectively). Hence, this test requires 20 squares (4 plant lines x 5 different water concentrations) of 1 x 1 meter, where 54 L of water is added per week in total. Since this experiment is carried out with the plant lines F3 and L5 that are not deficient in ga1, it is decided to let these two specific plant lines grow under net protection in order to eliminate the risk of spreading genetically engineered seeds to the environment.

 

Risk assessment of the field trial

The initial risk assessment of the Angola field trial experiment should be based on what is expected to occur and what could reasonably be expected during the trial. If a result occurs that is not expected, then it would be necessary to review the risk assessment. The risk assessment is written such that it is understandable to persons skilled in the art of molecular biology, such as an HSE inspector or individual laboratory worker.

Generic risk assessments (such as we have for some plant species including Arabidopsis thaliana - see information below about Thale cress) are acceptable for the wild type plants included in the Angola test. Relevant issues are assessed in relation to the GMO (Gene Modified Organisms) plants used in the Angola test.

Information about Thale cress (Arabidopsis thaliana)

Arabidopsis thaliana is part of the natural environment in Angola, so the testing in Ondjiva does not involve introduction of a foreign plant specie. Arabidopsis thaliana is a cruciferous plant and probably contains several glucosinolates that are deleterious if consumed in large volumes under certain conditions. However, other cruciferous plants such as cabbage and mustards are eaten in large amounts without deleterious effects. These compounds are products of complex secondary metabolic plant biosynthetic pathways and it is therefore extremely unlikely that such products could arise as a result of cDNA cloning in bacterial hosts (Keller et al. 1978).

Why is it necessary to follow up the trials under closed conditions in the greenhouse with trials in open conditions?

The greenhouse in effect provides a protected environment. The plants react differently when exposed to wind, rain and to natural pathogens. Thus, closed greenhouse conditions actually favoured the growth of a colonizing fungus, while the vegetation hall completely prevented infection with the fungus even in the non transgenic controls. Therefore field conditions cannot be replicated by any means at our disposal.

Description of the inserts (transgenes)

See Appendix A for a detailed description of the genetically engineered plant lines.

 

Environmental risk assessment

The environmental risk assessment is carried out collectively, since all plants included in the Angola test are Arabidopsis thaliana. Hence, containment of the testing is explained by giving answers to a number of questions, and providing background material related to these questions.

Which non-target organisms could possibly be effected?

Non-target organisms include various ground bacteria, other useful fungi (mychorrhiza), as well as a types of flea, flea beetle, springtail, lice and leaf beetles. To date we have been unable to establish a difference between transgenic and non transgenic controls. Insofar as we can judge, these plants do not therefore pose an ecological risk.

Prevention of the spread of genetically modified material in the course of this study From a purely scientific point of view we can assume that genetic material will not be released from the transgenic plants since Arabidopsis is strictly self-pollinating, the ecotype chosen (labaratory strain which may not be as adaptable to the growth conditions in Ondjiva as the native ecotype), and the ga1-3 mutation carried by the plants. However, in view of the public's reservations on this issue, we intend to coverthe transgenic plants not carrying the ga1-3 mutation in the 1x1 m 2 transgenic control squares with small pollen-proof tents to prevent the escape of pollen and cross-pollination by insects in the test period.

Even though the majority of these experiments are carried out with non-genetically engineered plant lines (ecotypes and mutant plant lines), Aresa will make sure that a complete and thorough clean-up procedure will be followed. The clean-up procedure will include removal of all plants by several methods such as manual harvesting of living plant material (also needed for biomass determination), the use of herbicides, and the use of a weed burner. Hence, no plants are left when leaving the test site.

All plant residues that are harvested during the clean-up will be burned on a fire for complete destruction.

To which plants could the transgenic Arabidopsis conceivably transfer its genetic information?

The transgenic Arabidopsis could transfer its genetic information to other Arabidopsis plants within a radius of two metres of the plant. In Angola this could occur likewise with Arabidopsis and a number of wild species of the family Brassicaceae. The genus Arabidopsis contains about 10 diploid species, such as Arabidopsis lyrata and A. halleri. More distantly related is a diverse group of North American N = 7 "Arabis" (now re-classified as genus Boechera), centered in western North America. Koch et al (2000) estimated that the genus Arabidopsis diversified about 5 million years ago, and Boechera has been separated from the Arabidopsis lineage for about 10 million years. Finally, there are agriculturally important Brassicas, and the Arabis alpine group across Eurasia.

[Figure 1 was not made available.]

Figure 1: Strict consensus tree of 10 most parsimonious trees from Fitch parsimony analysis based upon Chs sequences. Bootstrap support (above branches) and decay indices (below branches) are indicated. From Koch et al, 2000.

Plants resulting from a theroretical outcross are, however, not necessarily capable of survival. Partly due to the ecotype (the laboratory strain Col-0) and the ga1-3 mutation, it is almost impossible that the transfer of genetic information to normal diploid wild plants would produce plants capable of survival. Furthermore, the trial area is located at a considerable distance from other Brassicaceae fields (rape and mustard).

Additional genetic markers were introduced to the plant lines along with the genetic origins of interest.

Since the advent of genetic engineering, researchers have used antibiotic resistance genes as selective markers for the process of genetic modification. The concern has been raised that the widespread use of such genes in plants could increase the antibiotic resistance of human pathogens. As is common in research, the transgenic Arabidopsis plants are prototypes. As well as the Pap1, CHS, ExenA and ExenB gene, which is of interest to us, the plants also contain three markers for technical reasons.

Hygromycin, kanamycin are antibiotics and phosphinotricine (Basta) an herbicide used in selecting the transgenic plant lines. This gives rise to the general public's fear that a ground bacterium could acquire the antibiotic selection marker (gene) from the plant, and subsequently transfer it horizontally to a bacterium that poses a risk for humans. This bacterium could then no longer be treated with this particular antibiotic, which is of course undesirable. The bacterium could however be treated with a different antibiotic.

Assessment of antibiotic resistance

The original and most widely used selectable marker is a bacterial gene for neomycin phosphotransferase (NPTII), an enzyme that inactivates a number of related antibiotics including kanamycin. After introduction of constructs containing the NPTll gene into plant cells, kanamycin is applied to kill untransformed tissue. Transformed cells expressing NPTII are protected from the effects of the antibiotic and, using appropriate cell culture media, can regenerate into whole transgenic plants. A number of other systems employing antibiotic resistance as a selectable marker are employed in the production of genetically modified plants. These include the aad gene for streptomycin and spectinomycin resistance and the hpt gene, which confers resistance to hygromycin. Numerous studies have suggested that the presence of any of these antibiotic resistance genes in any crop or crop products will have negligible impact on food safety. The scientific evidence for this conclusion is summarised briefly below.

A possible concern about the use of antibiotic resistance as a selectable marker is the potential to compromise the use of antibiotic used in human (and animal) therapy. The possible transfer of resistance to gut microorganisms and the potential for transfer of resistance to potentially hazardous microorganisms has been considered, as has the presence of the gene product in food or feed. It has been concluded from extensive experimentation that the potential for compromising the efficacy of kanamycin or neomycin for therapeutic use in humans or animals by consuming the food and feed products are derived from genetically modified plants containing them is effectively zero because:

The transfer of the NPTll gene from plant material to gut microflora is extremely unlikely because there is no evidence that such transfer can occur. This conclusion is supported by studies that demonstrated that horizontal gene transfer from plants to microbes did not occur under a variety of test conditions (Nap et al., 1992; Redenbaugh et al., 1994; Schloter et al., 1995; Prins and Zadoks, 1994). Moreover, there are numerous barriers in the gut, which make this event extremely unlikely to occur and the event, if it were ever to occur, would be unlikely to be maintained in the absence of constant selection pressure for resistance due to the extremely limited use of these particular antibiotics.

In the unlikely event of a transfer occurring by an unknown mechanism from the genome of genetically modified plants or products derived from them to gut microflora and this event being maintained, this would not add significantly to the inherently large microbial population of kanamycin and neomycin resistant microbes in the gut of either humans (Shaw et al., 1993; Levy et al., 1988; Nap et al., 1992; Kelch and Lee, 1978) or animals (McAllan et al., 1973; Nap et al., 1992). The expression of the NPTll gene in genetically modified plants is controlled by a plant specific promoter, which is not expected to function in bacteria.

In the unlikely event of transfer of the NPTll gene and stable propagation of the intact gene fragment in bacteria, the gene is unlikely to be expressed and even less likely that a DNA rearrangement event occurs that places the functional NPTll -encoding open reading frame in front of a bacterial promoter. Even if expressed in intestinal bacteria, antibiotic therapy would not be compromised as the co-factors necessary for the enzyme to inactivate kanamycin and neomycin are not present at the required concentration range in the gut. Moreover, the NPTII protein would be rapidly degraded in the gut.

There is limited therapeutic/veterinary usage of kanamycin and neomycin. Thus, it can be concluded that the potential for the transfer of the NPTll gene from the genome of genetically modified plant material to microbes in human or animal gastrointestinal systems is insignificant. There is no evidence that horizontal gene transfer from plant cells to gut microflora can occur. In the unlikely event of NPTll being transferred to a gut microbe, no protein would be produced from the gene as the plant promoter would not be expected to function in bacteria. For the antibiotic resistance-conferring protein to be expressed, a rearrangement of the bacterial DNA would have to occur that places the functional NPTll-encoding open reading frame proximal to a bacterial promoter. As this event is very unlikely to occur, potential concerns of transformants gaining a selective advantage over other gut microbes through the expression of an acquired antibiotic resistance trait are further reduced.

The safety of the NPTII protein was established as it has been shown that nucleic acids are rapidly degraded in simulated gastric and intestinal fluids (McAllan et al., 1973; Nap et al., 1992).

Furthermore, the NPTll gene occurs ubiquitously in nature and resistance to this class of antibiotics is already widespread. Therefore, in the highly unlikely situation that a transfer event did occur in gut microflora in humans or animals, this would not significantly impact the overall frequency of kanamycin or neomycin resistant bacteria in the gut or rumen. Therefore, the overall risk is considered to be effectively zero and the therapeutic use of antibiotics in humans or animals will not be impacted by the commercialization of transgenic crops containing antibiotic resistant selectable marker genes.

Assessment of herbicide resistance

While antibiotic resistance genes continue to be used as a selective agent for the generation of a wide range of transgenic plants, resistance to specific herbicides may also provide an effective means for plant selection. In many cases herbicide resistance genes have in fact provided a more effective system for plant transformation (Chandler, 1995). For example, in legume species such as peas, luoomo and lupina, herbicide resistance has proved to be a useful selectable marker for plant regeneration (Schroeder et al., 1993).

Herbicide resistance marker genes may provide considerable advantage over antibiotic resistance genes in cases where either high levels of the antibiotic may interfere in the plant regeneration process or where plant tissues may exhibit a high level of intrinsic resistance. In addition, cell death in the presence of herbicide is generally more rapid and complete, thus providing more efficient selection. The presence of the herbicide resistant trait in transformed plants also provides a convenient and easily assayable system whereby the transgenic material can readily be identified/screened using simple techniques such as leaf painting.

Genes conferring resistance to a number of herbicide groups including the triazines, the sulfonylureas, bromoxynil, glyphosate and phosphinothricin are readily available. Of these, the bar gene isolated from Streptomyces hygroscopicus (Thompson et al., 1987) has been widely used as an effective selectable marker in the presence of the herbicide phosphinothricin (PPT). The enzyme inactivates phosphinothricin by the addition of an acetyl group from acetyl coenzyme A. This gene is freely available for research purposes and has proved particularly useful in cereals and grasses (Vasil et al., 1993; Wilmink and Dons, 1993), as well as in legumes.

While the introduction of herbicide resistant genes into specific target crops is often a major objective of plant biotechnology programs, the extensive use of such genes as selectable markers does, however, have some limitations. A number of potential problems can be identified, in particular, the presence of herbicide residues in the crop or food products derived from it. Other issues associated with herbicide resistance may also prove to be significant (Huppatz et al., 1995). Of these, the possibility of the herbicide resistant biotype itself becoming a weed and the chance of gene transfer between the herbicide resistant crop and weedy relatives are often cited as potential problems. However, herbicide resistant plants growing as volunteers within a subsequent rotational crop could generally be eliminated by treatment with a herbicide from a different chemical family. While gene transfer between a crop and a closely related weed with which it can interbreed is possible, astute management should eliminate any such risk.

Of perhaps greater significance are regulatory issues, particularly the use of herbicide resistant genes as selectable markers in crops where there has been no registration for use of that particular herbicide. Although the presence of herbicide resistance as a selection tool may not be the primary purpose for the generation of the transgenic crop, it nevertheless may be tempting to exploit the presence of a herbicide resistance gene as part of a weed control strategy. Herbicide manufacturers may well be reluctant to allow commercialization in such circumstances, particularly with the risk of litigation should the herbicide be used outside the regulatory framework.

In our opinion the use of antibiotic resistance in prototypes in the context of small, strictly supervised experiments poses no additional risk. This type of decision always represents a political compromise between the various parties involved. These parties represent differing interests which naturally are coloured by ideological differences. Therefore such decisions do not so much reflect scientifically based concerns as the power relations in the various decision making bodies. Such decisions do not therefore necessarily imply an actual risk as the public may imagine.

As the lifting of what was to all intents and purposes a moratorium by the EU demonstrates these decisions are subject to change.

Deficiency in production of the growth hormone gibberellin (ga1-3 mutation)

The plant has been modified by eliminating the growth hormone, gibberellin, encoded by the GA gene (Koornneef M, van der Veen JH, 1980). Since the plant line for detection of landmines does not contain a functional GA biosynthetic pathway, the plant is not able to germinate and to set seeds making the plant male sterile.

Hence growth of the plant lines requires addition of gibberellin in the initial phase of vegetative growth, which exhibits a practical control mechanism. Without addition of gibberellin, the plant will not grow, and hence, undesired spreading of the plant is avoided.

Scientific background for the ga1-3 mutation

The most dramatic phase change that flowering plants undergo is the transition from vegetative to reproductive growth. For this transition to be successful, plants must integrate a variety of environmental signals with endogenous cues, such as plant age (Bernier 1988).

In the facultative long-day plant Arabidopsis, the transition to reproductive growth occurs rapidly in long days but much more slowly in short days. Several flowering-time mutants, in which the timing of this transition is changed, have been isolated.

Analysis of the responses of different mutants to the environment together with studies of their genetic interactions have resulted in a two-pathway model showing how the transition to flowering is regulated (Martinez-Zapater et al., 1994; Weigel, 1995; Peeters and Koornneef, 1996). According to this model, long days induce flowering via a facultative and fast pathway, whereas under noninductive photoperiods, an autonomous and much slower pathway is rate-limiting. The latter pathway is thought to be related to plant age.

The gibberellin (GA) class of plant hormones plays a role in many processes during plant development, including seed germination, cell elongation, and flowering (Finkelstein and Zeevaart, 1994). In Arabidopsis, physiological and genetic experiments have implicated GAs specifically in the autonomous pathway of flowering. Exogenous application of GAs accelerates flowering in wild-type Arabidopsis, particularly in short days (Langridge, 1957).

That there is a causal connection between endogenous GA levels and flowering in Arabidopsis has been confirmed with several GA biosynthesis and signaling mutants. Mutants in which GA levels are severely reduced, such as ga1-3, are unable to flower in short days (Wilson et al., 1992). ga1-3 mutants carry a deletion of the gene encoding ent-copalyl diphosphate synthase (formerly ent-kaurene synthetase A), which controls a key step in early GA biosynthesis (Koornneef and Van der Veen, 1980; Zeevaart and Talon, 1992; Sun and Kamiya, 1994). In long days, flowering of ga1-3 mutants is only slightly delayed when compared with the wild type, indicating partial redundancy of the pathway involving GA1 under these conditions (Wilson et al., 1992; Silverstone et al., 1997). The sequence of the ga1-3 allele suggests that this is likely to be a null mutant (Sun and Kamiya, 1994). However, the ga1-3 mutant contains a small amount of GA, despite the complete absence of the ent-kaurene synthetase gene that is required for an early step in GA biosynthesis (Zeevaart and Talon, 1992; Sun and Kamiya, 1994)

[Page 14 in original was blank.]

References

Bernier, G. (1988) The control of floral evocation and morphogenesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 39:175-219.

Chandler, S.F. 1995. The use of herbicide resistance genes as selectable markers for producing transformed plants in Herbicide-resistant crops and pastures in Australian farming systems, eds. McLean, G.D. and Evans, G. (Bureau of Resource Sciences, Canberra) pp. 29- 239.

Finkelstein, R.R., and Zeevaart, J.A.D. (1994). Gibberellin and abscisic acid biosynthesis and response. In Arabidopsis, E.M. Meyerowitz and C.R. Somerville, eds (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press), pp. 523-553.

Flavell, R.B" Dart, E., Fuchs, R.L. and Fraley, R.T. 1992. Selectable marker genes are safe for plants? Bio/Technology 10:141-144.

Fuchs, R.L., Ream, J.E., Hammond, B.G., Naylor, M.W., Leimgruber, R.M. and Berberich,S.A. 1993. Safety Assessment of the Neomycin Phosphotransferase H (NPTH) protein. Bio/Technology 11:1543-1547.

Huppatz, J.L., Llewellyn, D.J., Last, D.L, Higgins, T.J. and Peacock, W.J. 1995. Development of herbicide-resistant crops - strategies, benefits and risks m Herbicide-resistant crops and pastures in Australian fanning systems, eds. McLean, G.D. and Evans, G. (Bureau of Resource Sciences, Canberra) pp 15-24.

Keeler, Richard F., K.R. VanKampen and L.F. James, eds. 1978. Effects of poisonous plants on livestock. Academic Press, New York, NY. xvii, 600 p.

Kelch, W.J. and Lee, J.S. 1978. Antibiotic resistance patterns of Gram-negative bacteria isolated from environmental sources. Appl. Environ. Microbiol. 36:450-456.

Koch, M., Haubold, B., Mithchell-Olds, T. 2000. Comparative evolutionary analysis of chalcone synthase and alcohol dehydrogenase loci in Arabidopsis, Arabis and related genera. Mol. Biol. Evol. 17 (10): 1483-1498.

Koornneef, M., and Van der Veen, J.H. (1980) Induction and analysis of gibberellin sensitive mutants in Arabidopsis thaliana.. Int. Z. Theor. Angew. Genet. 58:257-263.

Langridge, J. (1957) Effect of day-length and gibberellic acid on the flowering of Arabidopsis.. Nature 180:36-37.

Levy, S.B., Marshall, B., Schluederberg, S" Rowse, D. and Davis, J. 1988. High frequency of antimicrobial resistance in human fecal flora. Antimicrobial Agents and Chemotherapy 32:1801-1806.

Martínez-Zapater, J.M., Coupland, G., Dean, C., and Koornneef, M. (1994). The transition to flowering in Arabidopsis. In Arabidopsis, E.M. Meyerowitz and C.R. Somerville, eds (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press), pp. 403-433.

McAllan, A.B. and Smith, R. H. 1973. Degradation of nucleic acids in the rumen. British J. Nutrition 29:467-474.

Nap, J.P., Bijvoet, J. and Strikema, W.J. 1992. Biosafety of kanamycin-resistant transgenic plants: an overview. Transgenic Crops 1:239-249.

Peeters, A.J.M., and Koornneef, M. (1996) Genetic variation of flowering time in Arabidopsis thaliana.. Semin. Cell Dev. Biol. 7:381-389.

Prins, T.W. and Zadoks, J.C. 1994. Horizontal gene transfer in plants, a biohazard? Outcome of a literature review. Euphytica 76:133-138.

Redenbaugh, K., Hiatt, W., Martineau, B., Lindemann, J. and Ernlay, D. 1994. Aminoglycoside 3'-phosphotransferase II: review of its safety and use in the production of genetically engineered plants. Food Biotechnology 8:137-165.

Schloter, K., Futterer, J. and Potrykus, 1. 1995. "Horizontal" gene transfer from a transgenic potato line to a bacterial pathogen (Erwinia chryanthemi) occurs -if at all-at an extremely low frequency. Bio/technology 13:1094-1098.

Schroeder, H.E., Schotz, A.H., Wardley-Richardson, T., Spencer, D. and Higgins, T.J.V. 1993. Transformation and regeneration of two cultivars of pea (Pisum sativum L). Plant Physiology 101:751-757.

Shaw, K.J., Rather, P.N., Hare, R.S. and Miller, G.H. 1993. Molecular genetics of aminoglycoside resistance genes and familiar relationships of the aminoglycoside-modifying enzymes. Microbiol. Reviews 57:138-163.

Silverstone, A.L., Mak, P.Y., Martinez, E.C., and Sun, T.-p. (1997) The new RGA locus encodes a negative regulator of gibberellin response in Arabidopsis thaliana.. Genetics 146:1087-1099.

Sun T, Kamiya Y (1994). The Arabidopsis GA1 locus encodes the cyclase ent-kaurene synthetase A of gibberellin biosynthesis. Plant Cell 6: 1509 1518

Thompson, C.J., Mowa, N.R., Tizard, R., Crameri, R., Davies, J.E., Lauwereys, M. and Botterman, J. 1987. Characterisation of the herbicide-resistance gene bar from

Streptomyces hygroscopicus. EMBO Journal 6:2519-2523. Vasil, V., Srivastava, V., Castillo, A.M., Fromm, M.E. and Vasil, I.K. 1993. Rapid production of transgenic wheat plants by direct bombardment of cultured immature embryos.

Biotechnology 11:1553-1558. Weigel, D. (1995) The genetics of flower development: From floral induction to ovule morphogenesis. Annu. Rev. Genet. 29:19-39.

Wilmink, A. and Dons, J.J.M. 1993. Selective agents and marker genes for use in transformation of monocotyledonous plants. Plant Molecular Biology Reporter 11:165-185.

Wilson, R.N., Heckman, J.W., and Somerville, C.R. (1992) Gibberellin is required for flowering in Arabidopsis thaliana under short days. Plant Physiol. 100:403-408.

Zeevaart, J.A.D., and Talón, M. (1992). Gibberellin mutants in Arabidopsis thaliana. In Progress in Plant Growth Regulation, C.M. Karssen, L.C. Van Loon, and D. Vreugdenhil, eds (Dordrecht, The Netherlands: Kluwer Academic Publishers), pp. 34-42.

 

Declaration of the content of this document

It is hereby declared and confirmed that the information provided is correct in respect to the designed testing in Ondjiva. Furthermore, the scientifically judgement of the use of the genetically engineered plants is to our knowledge in accordance with safety guidelines in biotechnology. We declare that to the best of our knowledge, having made reasonable inquiries, the information herein is true and correct.

Notifying Organisation's Representative:

Signature: ________________________Date: ______________________________

Printed Name: Simon Østergaard

Project Supervisor:

Signature: ________________________Date: ______________________________

Printed Name: Carsten Meier

 

ANNEX B:

TRIAL INSTRUCTION - ARESA "development trial" - AVS, Sept/Nov 2003

For these purposes a "development trial" is defined as a trial seeking to establish whether the concept under assessment has the potential to achieve its stated aims and whether any materials involved are fundamentally flawed in a manner that would render it unusable in the field. In this case it also involves gathering field data required for further technology development. A "development trial" may involve the assessment of parts or sub-systems that are believed to be ready for field deployment and any such parts or sub-systems must be identified prior to the trial and assessed by "Acceptance Trial" criteria rather than those used for "development trials".

This particular trials are being conducted in co-operation with Menschen gegen Minen (MgM), an international demining NGO with appropriate facilities in Southern Angola. Conditions for conducting trials at the MgM test-site at Ondjiva in Angola do not restrict the scope and conduct of well-planned development trials. Technology developers can observe field-conditions while retaining access to the logistic support they need - and while living in relative comfort.

Aim and scope

The aim of the ARESA development trial can be summarised as follows.

AIMS

A) To establish whether the genetically modified plants (known as A line: Genotype PAP1/tt4/nii-CHS/ga1-3/ExenA (designated LA);B line: Genotype PAP1/tt4/nii-CHS/ga1-3/ExenB (designated LB), change colour when grown above hidden landmines and UXO in the test area.

B) To establish whether the genetically modified plants produce flowers.

C) To determine the efficacy of the proposed seed distribution system and reach a view on whether it and its ancillary parts have the potential to achieve field utility.

D) To determine whether either of two flowering genetic variants have the same of better potential than the main (LA and LB) variants.

E) To devise and implement a watering regime on unmodified plants to determine the minimum acceptable water requirements and also the requirement that is optimal for fast growth.

F) To gather data to assist in further ARESA research.

G) To gain some field experience of the way in which Humanitarian Demining is controlled and conducted.

H) To facilitate the recording of a video record of the testing.

The trial will take place over a period of six weeks (5-7 weeks) at the dedicated MgM detector and dog test areas in Ondjiva, Angola. Assisted by ARESA personnel and within the limitations of restricted biotechnical knowledge, AVS will record the detailed activities of the trial as it progresses in order to produce an "Independent end-of-trial" report.

 

Constraints and limitations

ARESA personnel, AVS and associated visitors to the MgM site must obey MgM camp rules and while staying in the area are requested only to travel in company with MgM representatives.

 

Trial format

The dedicated detector test-area is already fenced. Additional mined-area marking tape may be used to mark observation lanes throughout the area.

The dedicated dog-testing area is already fenced. Additional mined-area marking tape will be used to mark observation lanes throughout the area.

The seeds will be broadcast using equipment brought by ARESA to the area and by hand. Some seed in will be pelleted in one of two growth mediums, others will not.

Two one metre square areas will be sown with variant modified seed and grown under pollen-proof netting.

The observation lanes will be used to provide access for daily watering using regimes designed to give ARESA the data required to determine optimum water needs and also to allow close inspection without damaging the plants.

The observation lanes will be used to record the plant progress three times a day throughout the trial. A photographic record will be kept by ARESA personnel.

By agreement, and because MgM is not assessing the technology, the positions and identity of targets (ERW) buried in the detector-test area has already been made known to ARESA. (This is a "Development trial", not an "Acceptance Trial".) Details of the lane preparation, sifting, drainage and depth are documented separately. Similarly, details of the targets, their depth and their positioning are documented separately.

 

Criteria

AVS has not been asked to develop trial criteria. To develop detailed judgment criteria without greater knowledge of biotechnology would not be appropriate. However, AVS has been asked to comment on the outcome and will do so with reference to the following:

1. A perceived change of colour occurring above buried ERW.
2. The consistency of any colour change with reference to the specific ERW beneath.
3. Whether or not the G3 modification flowers.
4. Whether any random colour changes occur without the presence of ERW.
5. The practicality of the deployment method in the field.
6. The apparent professionalism with which the trial is conducted and records are kept.

Methodology, evaluation and acceptance criteria are to be determined and judged internally by ARESA, but may be commented on by AVS when deemed appropriate. AVS will advise on aspects where he feels qualified to comment and that advice may be recorded in his report for ARESA.

AVS will also comment on any safety aspects that arise and reserves the right to withdraw if activities that he considers to be physically or ecologically irresponsible or unsafe occur.

 

Execution

The trial schedule is documented by ARESA. The AVS representative at the test site is Andy Smith. MgM staff may assist if asked, but no such arrangement has been made in advance. The ARESA staff present will be Anders Søndergaard and Carsten Meier. The person with overall responsibility for safety will be the MgM camp manager, Ken O'Connell or his deputy when he is absent.

It is anticipated that two personnel from the film Company Bastard will also be present throughout the trial.

The trial will take place in dedicated test areas prepared for testing detection technologies and Explosive Detection Dogs (EDDs) on the outskirts of the town of Ondjiva in Southern Angola. The trial personnel will be accommodated in a well-appointed hotel 200 meters away from the MgM camp.

There is no anticipated explosive risk during the trials.

 

Annex C: Pre-Trial Assessment - Andy Smith 30/09/03

C1. Introduction

ARESA plan to carry out a number of plant-growth tests and data-gathering exercises in the MgM campsite and test-area at Ondjiva, Angola. The plants to be used are all variants of Arabidopsis thaliana, which is also commonly known as "Thale cress". Plants of this type are known to occur naturally in the area, which has given rise to concerns about the escape of genetically modified material - see Section C2, Pre-Trial Risk Assessment.

The author of this report is not a biologist, so cannot comment in detail on the accuracy of scientific material presented to MgM and the Angolan authorities by ARESA.

Testing will be of the growth and colour-change of two genetically engineered plant lines in the "detector-test area". This will occur alongside a number of growth experiments on different ecotypes or mutant plants in the "dog-training area". The later should allow the identification of an ecotype best suited to the growth conditions in that geographical location.

C1.1. Detection of ERW in the detector test-area

The stated goal of this test is to determine the efficacy of the genetically engineered plant lines in terms of both growth and reaction to the presence of buried ERW.

The seven test lanes will be sown with a mixture of pelleted seeds of plant lines LA and LB, which have a minor genetic variation.

A line: Genotype PAP1/tt4/nii-CHS/ga1-3/ExenA (designated LA)

B line: Genotype PAP1/tt4/nii-CHS/ga1-3/ExenB (designated LB)

The seeds will be spread as a concentration of 3000 seeds per square metre and they will be watered regularly. Visual inspection and a photographic record will be made by ARESA daily.

Two small areas within the detector test area will also be used to determine the growth of the LA and LB seed types in the presence of a specific herbicide that they are modified to resist.

AVS comment: care must be taken not to overspread the seeds and any subsequent growth medium because the area backs onto a public path, separated by a chain-link fence. No overspray into a public area should occur.

AVS recommends marking access paths between the test-lanes to facilitate observation and maintenance access without damaging plants.

 

C1.2. Seed and water broadcasting using a water pump system

The stated goal of this test is to evaluate a proposed method of spreading seeds and water that is intended for practical field use. The pump will not be used to broadcast modified seed, so allowing assessment without risk of uncontrolled dispersal of GM seed.

AVS comment: ARESA should be aware that manual demining does not normally make use of a "grid system". Some Humanitarian Demining organisations use EDDs in this way.

 

C1.3. Seed density (data gathering exercise)

The stated goal of this test is to determine optimum seed concentrations to achieve good coverage and the variation in plants per square metre that results from varying the seed concentration. The seeds used for this experiment will not be genetically modified, and the test will occur in discrete sections of the dog-test area. Five separate square metre areas will be used. The five variations of seed deployed will differ only in whether they are pelleted in a growth medium and in the density with which they are sown.

AVS Comment: While this test does not involve GM seed, it does involve sowing seeds from varied geographical locations around the world. AVS recommends that these plants be destroyed as soon as flowers start to appear (first buds form).

 

C1.4. Measuring context variables (background physiological measurements)

The stated goal of this data gathering exercise is to record the following variables at the site, so providing a context within which to view the test results. The exercise involves recording the following data at 09.00 am and at 09.00 pm each day:

" Air temperature
" Soil temperature
" Air humidity
" Soil humidity
" PAR (solar radiation without UV)

 

C1.5. Applying herbicide concentrations to local plant growth

The stated goal of this data-gathering exercise is to determine the effect of the herbicide concentration on the local plant population. Small areas (single square meters) of local plants and the plants in the detector test-area will be treated with varied concentrations of the herbicide "Basta".

AVS comment: the purpose of this test is not outlined in the ARESA documentation but is believed to be to indicate whether spraying Basta in a particular concentration could kill local plants while allowing the genetically modified plants (LA and LB) to grow. This could allow the seeds to be disseminated over a suspect area without cutting the undergrowth, so saving time and cost.

 

C1.6. Growth experiments with different naturally occuring types of "Thale cress"

The stated goal of this data-gathering exercise is to examine the growth of several types of the ""Thale cress" plant with the aim of identifying whether any one grows better in the local soil.

The types or "Ecotypes" to be used are:

" H. Tsukaya (NASC stock # N3963) Jap
" Ita-0 (NASC stock # N1244) Ita-0
" India (NASC stock # N9030) Ind
" Santa C. (NASC stock # N8069) S.C.
" Ler Ler
" Col-0 Col

The germination rates and biomass achieved within a timeframe will be measured, and the water requirements of each type assessed. These experiments will require 30 x 1 metre squares (6 ecotypes x 5 different water concentrations), with a total of 81 litres of water added each week.

AVS Comment: While this test does not involve GM seed, it does involve sowing seeds from varied geographical locations around the world. AVS recommends that these plants be destroyed as soon as flowers start to appear (first buds form).

 

C1.7. Growth experiments with genetically modified variants

The stated goal of this data-gathering exercise is to examine the growth ability of two "mutant" plant lines of Arabidopsis thaliana ("Thale cress") in order to examine the effect of the mutations on the overall growth performance. Germination rates and biomass achieved within a time frame will be measured, and a record of the water applied will be made. These experiments could prove crucial if the genetically modified plants in the detector test-area do not grow as anticipated.

Details of the growth of the following mutant plant lines will be gathered along with a Col-0 laboratory wild plant line that serves as a "control". This data-gathering will take place over 15 discrete square metres. The two GM plant types and the Col-0 plant line will each be in five separate square metre plots receiving five 5 different concentrations of water. In total, 40.5 litres of water will be added each week.

The genetically modified plants to be used are designated as:

" tt4-58 stop codon in CHS tt4

" ga1-3 GA deficient mutant ga1

AVS Comment: it is hard for me to see how this data-gathering exercise varies from C1.8 below. AVS recommends that these plants be destroyed as soon as flowers start to appear (first buds form).

 

C1.8. Growth control experiments with other genetically modified plant lines

The stated goal of this data-gathering exercise is to examine the growth ability of four other "genetically engineered" plant lines in order to examine the effect of the modifications on their overall growth performance. Germination rates and biomass achieved within a time frame will be measured, and a record of the water applied will be made. These experiments could prove crucial if the genetically modified plants in the detector test-area do not grow as anticipated.

Details of the growth of the following modified plant lines will be gathered. This data-gathering will take place over 20 discrete square metres. Each of the four plants types will have five separate plots, each receiving a different concentration of water each week.

The genetically modified plants are designated as:

" F3 genotype pap1/tt4/ F3
" Line 5 genotype pap1/tt4/nii-CHS L5
" A line genotype pap1/tt4/nii-CHS/ga1/ExenA LA
" B line genotype pap1/tt4/nii-CHS/ga1/ExenB LB

The F3 and L5 variants are not modified in a way that will stop them flowering and producing pollen, so their plots will each be covered by "pollen-proof" net protection.

AVS Comment: it is hard for me to see how this data-gathering exercise varies from C1.7 above. There appears to be no research advantage in allowing these plants to flower. AVS recommends that these plants be destroyed as soon as flowers start to appear (first buds form)

 

C1.9 AVS recommendations

If the plants "will react on NO2-groups, which are cleaved off from the explosive molecules that leak out of the buried landmine", it would seem to a layman to be appropriate to test whether the plant also reacted to the nitrate residues left when common nitrate fertilisers had been applied to the land in previous years.

It should be noted that the presence of nitrate fertilisers could falsify the results - and so some kind of pre-test assessment of their presence in the detector test-area would seem desirable.

 

C2. Pre-Trial Risk Assessment (AVS)

AVS (Andy Smith) has no background in biotechnology and cannot assess the ARESA Risk Assessment for completeness or consistency. That said, the planning documentation has given rise to the observations and recommendations below. Risk Assessments are intended to predict things that could go wrong, and take measures to minimise the chance of that happening. This is what I have tried to do in what follows.

I identify three kinds of risk in what follows. These are:
1. The risk of failing to collect the required test-data.
2. The risk of releasing GM material into the environment.
3. The risk of an unintended detonation or fire during burn-off.

The first of these can be minimised procedurally following the advice under C2.1 and C2.4 below. The fear of the second can be minimised using the recommendations under C2.2 and C2.3 below, but AVS is not qualified to assess how significant any residual risks may be. The third of these perceived risks is negligible and can be reduced further if the recommended measures under C2.3 below are followed.

 

C2.1 Complexity and human error

The series of experiments, tests and data-gathering exercises planned by ARESA are complex, and that complexity introduces risks of human and "nuisance" error. Human error is self explanatory. "Nuisance" error in this instance can be defined as the intended or unintended disturbance/removal of some or all marking systems (by strong winds, animals, birds, etc).

By my calculation, eight variant seed types that are genetically modified or "mutant" (I am not sure of the difference) and five naturally occurring seed types are to be used over a total of more than 74 discrete areas.

The 74 areas are:

Detector test area: two GM seed variants plus two areas sprayed with herbicide = 4 discrete areas.

Dog-training area: six GM/mutant seed types and one control, each in five discrete areas = 35 discrete areas.

Plus six naturally occurring seed variants each in five discrete areas = 30 discrete areas.

Plus five discrete square metres of seed-density variation exercises = 5 discrete areas.

Plus an uncertain number of herbicide exercises on naturally occurring growth.

There is a very real potential for confusion over labelling and mapping which area is which. Not only could this render the results of all data-gathering exercises uncertain, I believe that it could also be perceived as hazardous if the genetically modified plants are allowed to mature and flower.

AVS Recommendation
A complete map of the testing area should be drawn prior to spreading any seed. When the map has been approved by MgM (giving permission to use that ground) the areas must be clearly marked out using stakes and marking tape. The stakes used in those areas with genetically modified/mutant seed must be painted a different colour to those with naturally occurring seeds. Each discrete area must be labelled with a durable label BEFORE seeds are spread (so helping to ensure that the right seeds are spread in the right places). The stakes used to mark the areas where the two GM seed-types that are expected to produce flowers will be sown should be a third colour - so making it easy to immediately identify these areas if the pollen-proof netting is disturbed for any reason.

 

C2.2 Continued concern about pollen escapes

The most important reason for ARESA's presence in Ondjiva is the performance testing of the two genetically modified plants that will be used in the "detector test-area" where there are some items of ERW with their original HE content.

ARESA state that "the triggered colour change of the plants will occur in the vegetative stage of plant growth, which enables identification of the specific plants growing in the near by presence of landmines within 3-4 weeks".

My understanding is that these plants are also effectively "neutered" and cannot produce seed that would survive. The GA-13 modification is intended to prevent flowering, but in the ARESA assessment it is clear that day length can over-ride this restraint and so the plants in Ondjiva (where sun intensity and day length have not been determined) may flower (and may even be expected to flower).

If they produce flowers (and so pollen) not everyone would be convinced that allowing the pollen to be transferred to similar plant types in the area by insect activity is safe.

If the plant flowers, it is not possible for MgM or myself to determine whether the pollen presents a threat of successful cross-fertilisation with local plants. ARESA identify a wide range of plants that could theoretically be affected in their Risk Assessment. They also state that "Plants resulting from a theoretical outcross are… not necessarily capable of survival". This means that ARESA recognise that they may survive.

AVS Recommendation
Because flowering should not occur until after the "vegetative stage" of plant growth and any colour change should have become apparent before then, I recommend that the GM plants types in the detector test-area be destroyed as soon as any flower buds appear.

Given the declared purposes of the data-gathering exercises, I further recommend that all plant types, genetically modified, mutant or otherwise, in the dog training-area should be destroyed when the first buds appear in any discrete test area.

The declared purposes of these tests do not require the plants to flower. I include the non-modified plants in this recommendation because we have no right to release pollen from a plant that only naturally occurs in Japan or India with no knowledge of how that might effect local strains - and also because the flowering is not necessary for the successful completion of the declared research aims. Even the most minimal risk should be avoided if there is no advantage in taking that risk.

 

C2.3 Antibiotic and herbicide resistance

AVS is not qualified to comment in detail on the accuracy of ARESA's view that the genetic modifications and/or markers present no risk to human or animal population if they enter the wider environment and so the food chain. AVS does observe that ARESA's confidence that the antibiotic resistance could not be transferred is qualified by the declaration that any transfer would "have negligible impact on food safety". No impact at all is publicly acceptable, so every effort must be made to avoid it, and must be seen to be made.

AVS recommendation
I strongly recommend that all measures be taken to ensure that no animals enter the test areas and that birds are discouraged (using bird-scarers) from visiting the areas where ALL plants are grown. Given the small potential for ground-bacteria transferring genetic material, I recommend that the ground in the test and training areas be processed by burning at the end of the tests, so destroying all roots and making an effort to kill any affected bacteria. (I do not know whether this would be effective, but it could be done relatively easily so I recommend it.) It may be necessary to lift some of the mines/devices in the "detector test-area" to prevent fire damage before this is done. No mines/devices should be lifted without express permission from MgM. All mines/devices that contain any explosive element and are within an agreed depth of the ground surface MUST be lifted before burning off the area (suggested depth 10cm).

Any ideas for inhibiting the access of small mammals should also be pursued.

 

C2.4 Objectivity of the detector test area testing

There is a risk that ARESA will not be able to convince the world of its results when the testing and data-gathering has been completed.

AVS recommendation
In order to ensure that the "independent observations" that AVS makes of the results from the "detector test-area" are considered trustworthy by the outside world, AVS recommends that the "detector test-area" be kept locked at all times.

 

Annex D: AVS/ ARESA Biotech contract

Contract between: Mr. Andy Smith and Aresa Biodetection Aps

Plant technology testing in Ondjiva, Angola, 2003

Aresa Biodetection (the company) is about to test its technology platform at the landmine test area of Menschen Gegen Minen (MGM) in Ondjiva located in the southern part of Angola.

The company will employ Andy Smith (AS) on a contract basis for this specific trip in order to fulfil the roles described below.

The role of AS in Angola:
Observer: AS shall document the experiment carried out in Angola by writing up a report that describes the test results achieved. It is of high priority that the report emphasises on the sensitivity of the plant line(s) grown in respect to the different type of mines detected/not detected. Furthermore, the report shall contain reflections upon observations made in respect to the practical use of the technology (seeding, watering techniques and systems, water requirements, plant growth, etc.) in light of the complexity when operating in a landmine infected area.

Technical advisor: AS shall act as an advisor on the technology in development on behalf of his in-depth experience within de-mining. Hence, AS shall challenge existing ways of thinking, and discuss new ideas generated with the employees of the company.

Safety advisor: AS shall act as an advisor on safety related issues due to his experience from operating in landmine infected areas. Hence, AS should guide the employees of Aresa and employees of Bastard to behave as safely as possible in Angola. In case of any doubt, AS should consult the MGM safety Officer Ken O'Connell for advice on safety related matters.

Terms and conditions:
Period:
The plant technology testing is scheduled for 6 weeks with estimated time of departure from Europe in the beginning of October. Hence, AS will be offered employment on a contract basis for the company in the period of October 6 to November 16, 2003.

Working schedule:
As a general rule, AS is hired for 5 days a week during the period. Nevertheless, the working schedule of AS is flexible, but AS shall aim at choosing a schedule that allows him to achieve the best possible documentation of the testing.

Compensation:
AS will receive [excised] per week, which equals [excised] in total for the period. Furthermore, to cover food and drinks, AS will receive a daily allowance of DKK [excised] per day (approximately equal to [excised] according to the exchange rate (on July 22) of DKK 10,49 per £). If AS and the company agree to extend the period for more than 6 weeks, AS will get paid accordingly with a weekly rate of [excised] plus daily allowance of DKK [excised].

Payment will occur in two separate, but equal, transactions. Half of the compensation [excised] plus daily allowances for three weeks will get transferred to the account of AS on October 1, 2003. The same amount of money will be transferred to the account of AS at the end of the period. Payment will be transferred to the following bank account of AS: AVS Consultants Ltd., Business Account, Midland Bank…[details excised].

Expenses:
Expenses covering visa, airfares to Namibia and back to United Kingdom, transportation between Windhoek and Ondjiva (and back), and hotels will be covered by the company.

Compensation if cancellation occurs:
The testing of the technology is dependent on getting appropriate approvals (to test the technology platform by MGM; to transfer the genetically engineered seeds via Namibia to Angola by the authorities; and to grow these seeds). Due to the engagement of AS to the project, and booking of the trip, AS will receive compensation for the full period ([excised] in total) if the testing is cancelled before departure. AS will not receive any kind of compensation if he cancels his engagement, or fails to participate on the trip for personal reasons (sickness, etc.).

Confidentiality agreement:
In order to control and protect the patent strategy and the communication strategy of the company, AS agrees to treat all knowledge and insight material regarding the use and the development of the technology platform, which may come to his possession, as strictly confidential. Hence, AS will only communicate any information about the company and its technology to a third party, after getting accept from the management of the company. On behalf of the company:

 

 

Annex E: The MgM Ondjiva Test Area

[The original "map" of the test-area has photographs of each lane and of each item in the lanes. The pictures have been removed from this Annex in order to keep this report a manageable size.]

AVS had previously constructed a small detector test area for MgM near the town of Ondjiva in Southern Angola. It was not constructed for the ARESA tests and so many of the "targets" concealed in the test lanes do not contain any High Explosive. However, shallow mines containing both TNT and RDX are present, along with ordnance and deeply concealed AT mines.

 

Annex E, part 2: Test field details

Removed for web publication.

 

Annex F: Relevant emails

-----Original Message-----

From: Andy Smith [mailto:mgmang10@mail.station12.com]
Sent: Friday, October 31, 2003 2:50 PM
To: 'simon@aresa.dk'
Subject: To Andy Smith

Dear Simon,

The attached show the plants in Lane 7 today. The arrow is pointing to a cluster of 17. They tend to be in small clusters like this. This shows the most advanced plants I can find. The other picture [not reproduced] shows the local plant growth compared to the seeded growth.

If I do not hear from you, I will destroy them in the morning. My phone number is 00 244 92 645083

Regards

Andy

 

-----Original Message-----
From: Simon Østergaard [mailto:simon@aresa.dk]
Sent: Friday, October 31, 2003 11:57 AM
To: 'MgMAng10'
Subject: SV: To Andy Smith

Dear Andy,

Thanks for your mail! I hope it is not too late to comment on the development of the plants. I'm sad to hear that you are planning on giving up lane 7. I would very much appreciate if you could stay for longer.

I believe the plants should be growing as they were in the beginning of this week, and since the second round of seeds sawed is delayed in comparison with the first round. It is impossible to see plant growth just by looking at these form day to day; it is important to measure growth of the rosette by putting sticks in the ground and observing the growth over a couple of days (I'm aware that you may already have done this). Furthermore, I believe that the roots should go down 5-10 cm in the soil in order to induce the red colour change due to the extensive watering carried out.

I hope it is possible to motivate you to give it a further go. Let me know if this is possible by any chance.

Best regards,

Simon.

 

-----Original Message-----
From: Andy Smith [mailto:mgmang10@mail.station12.com]
Sent: Friday, October 31, 2003 3:11 PM
To: 'Simon Østergaard'
Subject: RE: plant success....

Dear Simon,

The attached shows the largest plant I could find in Lane 7 thirty minutes ago. Your suggestion of placing sticks in the ground and measuring daily growth is absurd. I admit that the root is longer than I would have expected, but at less than two centimetres life is just too short to wait for it to grow another eight.

I don't know what Carsten has been telling you, but this is reality. I expect to end the trial tomorrow.

Regards

Andy

 

-----Original Message-----
From: Simon Østergaard [mailto:simon@aresa.dk]
Sent: Saturday, November 01, 2003 1:21 AM
To: 'MgMAng10'
Subject: SV: plant success....

Dear Andy,

No doubt it will take longer to get the roots to the desired depth than the timeframe within we can keep you on site. Thus, we will accept that you will be finishing up the trial tomorrow. We appreciate your effort and that you will clean up lane 7.

Many thanks.

Regards,

Simon

-----Original Message-----
From: Andy Smith [mailto:mgmang10@mail.station12.com]
Sent: Saturday, November 01, 2003 7:12 AM
To: 'Simon Østergaard'; 'im@mgm.namib.com'; 'Hendrik Ehlers'
Subject: RE: Magic mushrooms in Ondjiva....

Dear Simon,

It was agreed with Carsten that if there was no significant increase in the plant's size after one full week under shade netting, the trial would end. At that time, I pointed out that the very sparse coverage meant that even if these plants matured, there was little chance of them doing so on top of a mine. The plants have not grown above ground at anything remotely approaching the rate that Carsten told me he anticipated, and have barely grown at all despite limitless water. Fungi have been having a wild time - with several psylopsybin in evidence, from which I infer that the shade netting is effectively cutting out the sunlight (this was never measured to my knowledge - so the inference is not of much scientific value). The fungi grow more than 5cm in a day and have withered by the following day. Funny but I never anticipated coming to Ondjiva to grow magic mushrooms.

I now have the problem that there have been torrential rains in the last two days. You seem to think that I only have to worry about Lane 7? I do not only have to clean up Lane 7, but the whole area including those places where various seed did not germinate but may be dormant and overspray areas. A little of this was done on Monday and Tuesday but it takes a long time and nothing occurred on Wednesday when I was off for the day (despite my best efforts to ensure that it was - this is how it goes in Africa). It is now impossible to finish the burning off process for two reasons. The gas torch supplied produces a very gentle flame - probably because of low pressure in the bottled gas used for cooking here. And the ground and vegetation is waterlogged. Burning was wanted because of dispersed GM seed that could not possibly be seen and became mixed with dust that has now become a surprisingly sticky mud.

The Roundup weedkiller (restricted in Denmark) works over 10 days and will not work at all if it rains within six hours of application (written in the instructions). It should be sprayed using a plastic bottle, spray and hoses - which I do not have. (The instructions warn that using metal parts risks an explosion of hydrogen gas.) So I would have to apply it by using a plastic watering can. Its time-delay means that I cannot SEE whether it is effective, and it will be ineffective if the rain continues. (I know that the Roundup is not meant to kill the GM plants, but do not really believe that: in any case, if I remove the shade netting, I believe that they will die like all the others.) The other weedkiller bought by Carsten in Namibia is an unknown entity and would probably be banned in Europe. I asked Carsten to find out its ground-life more than two weeks ago, and several times thereafter. This is important because the camp uses a borehole and I do not want to apply a restricted substance that may contaminate a sweet water source. Carsten told me that he had given you the list of ingredients and asked you to find out the safe usage. Please advise whether Carsten did in fact ask you and, if so, what the result of your inquiries was?

Regards

Andy

 

----- Original Message -----
From: "Didactylos"
To: "Simon "
Sent: Friday, November 07, 2003 5:42 PM
Subject: Home

Dear Simon,

I am back in UK now. I burned off the Lane 7 area on Monday this week.

Do you want a final report or not? If so, there will be additional costs involved. If you want copies of the photographs I took (an incomplete record because everyone else was taking pictures professionally) please let me know. There will, of course, be costs involved.

Regards to all,

Andy

 

----- Original Message -----
From: "Simon Østergaard"
To: "'Didactylos'"
Sent: Monday, November 10, 2003 12:07 PM
Subject: SV: Home

Dear Andy,

We don't want a report and photographs for additional costs. Although you were paid for an extra week including per day allowances, after you cancelled the contract with Aresa, we will not ask you to refund anything.

We consider the collaboration as finished in relation to the deliverables and economic transactions. Due to our conversation with Hendrik regarding possibly continuation of the testing in the future, we ask you to respect the confidentially agreement included in your contract.

Best regards,

Simon.

 

----- Original Message -----
From: "Didactylos"
To: "Simon Østergaard"
Sent: Monday, November 10, 2003 1:56 PM
Subject: Re: payments

Dear Simon,

> We don't want a report and photographs for additional costs.

I will finish the report for my own records - also for the MgM directors who had most concern.

> Although you were paid for an extra week including per day allowances,
> after you cancelled the contract with Aresa, we will not ask you to
> refund anything.

May I remind you that - at your request - I stayed on for a full week after the others left. Also, I had continued to monitor the trial before that - because MgM had made my "policing" a condition of the trial's continuation. So I would not have been inclined to refund anything anyway. I do not think that I owe ARESA anything, rather - I believe that the opposite applies.

Please send me an up-to-date email address for the BASTARD TV crew. I have pictures of them at work which they might need. Thank you.

Regards

Andy

See also ARESA "red plants" and the con goes on*