Continued from:
Emerging Technologies
Genetic Engineering and Biological Weapons
The Sunshine Project: Background Paper #12
November 2003http://www.sunshine-project.org/publications/bk/bk12.html
V Emerging Technologies III: New types of weaponsMany other new weapons may become possible in the decades to come. The deciphering of the human genome, synthetic genes and organisms, new approaches to gene therapy and drug delivery, and the sheer volume of genetic engineering experiments with potentially pathogenic microorganisms will increase the availability of much more sophisticated biological agents with a potential for hostile use, not only in classical warfare scenarios, but also for "peacekeeping", "military operations other than war", "low intensity conflict", and covert operations. To illustrate the possibilities, examples of future weapons based on current technologies follow:
Food Weapons
So called "edible vaccines" and "biopharming" (i.e. the production of vaccines or other bioactive substances in edible crops) can be put to hostile use. In the past decade, genetically engineered plants have been investigated as a means to produce and deliver vaccines. There are already a variety of research reports demonstrating that engineered plants can elicit an immune response in humans (Haq et al. 1995, for review see Streatfield/Howard 2003), and clinical trials on humans are currently underway to test vaccines produced in edible crops. [31] These vaccines may be isolated from the plant for further processing or directly delivered to the patients by consumption of the engineered plant.
Vaccines are only one type of bioactive substances being produced in edible crops. Several US companies are using genetically engineered crops to produce industrial enzymes, growth hormones, and other potent pharmaceutical compounds. These techniques pose a serious risk to human health and the environment, especially when the highly active pharmaceuticals are introduced into edible crops. [32]
The possibility of abuse of these crops and/or the underlying technology for hostile purposes is serious. In long term conflicts, it may be tempting to weaponize engineered crops, spiking them with, for example, disease-inducing (e.g. cancer) or debilitating compounds (e.g. affecting human or animal fertility) or built-in deficiencies that could lead to crop failure. Such "weaponized" germplasm may thereafter be introduced in the target country’s seed supply and consequently its food supply through covert actions or simply by means of seed sales or humanitarian aid. This may not be possible with crops that are exported by the target country, as, given today’s global market, the spiked food/feed could end up in the aggressor’s food supply. But for most countries it will be possible to identify food or feed crops in the target country that are not exported.
There are routes to possibly achieve similar effects without sophisticated knowledge to engineer a specific crop with a specific compound. Theft of a few corn kernels from one of the many trials with edible plants producing bioactive substances may be enough. Pharmaceuticals such as blood clotters or blood thinners may not be a weapon of choice, but introduction into the food supply would not be technically difficult. Profusion of such artificial traits would likely produce panic and could be very difficult and expensive to eradicate. Public concern would be amplified if the trait in question was a potent growth hormone, which has been field trialed in the US, or a drug called trichosanthin, which has also been tested. Trichosanthin, considered to be a potential anti-cancer agent, has the same mode of action as the biowarfare agent ricin [33] and is a strong abortion-inducing compound. In the US, trichosanthin production in tobacco plants was induced by a genetically engineered plant virus. That same virus also easily infects crops such as tomatoes and peppers.
A 'contraceptive corn' developed by the US company Epicyte is unlikely to be usable for hostile purposes; but illustrates the potential abuse of pharming. Epicyte genetically engineered corn to produce an antibody against human sperm. The company wants to produce large amounts of the antibody in order to extract it for use in a contraceptive gel. Consumption of the engineered corn or the extracted antibodies is unlikely to confer sterility – but a similar approach would yield dramatically different results. Introduction of a gene for human sperm cell antigens into a crop could create (an easily abused) contraceptive vaccine, preventing women who eat the engineered corn from reproducing.
Edible weapons pose a serious problem for BW non-proliferation efforts. No biological arms control effort could stop a person from stealing a handful of kernels, growing more, and introducing them into a country’s food supply. The technology and especially its products are inherently difficult to control – the past years witnessed a variety of cases where specific genetically engineered crop varieties showed up in unexpected places. In one case, a corn variety that was not permitted for human consumption by US regulatory agencies showed up in a broad variety of human food supplies – despite it being approved for animal feed only. [34]
Considering how easy and effective the hostile abuse of these genetically engineered crops is once they are developed, a complete ban on the production of hazardous compounds in edible crops appears to be justified. This may not stop a criminal from willfully creating an ‘edible weapon’, but it would tremendously raise the threshold compared to wandering into a corn field and grabbing some cobs. In addition, it will be technologically more challenging for a future biowarfare program to develop its own ‘food weapon’ if the technology is not further developed. With each experiment and each field trial, more knowledge on how to turn food crops into dangerous weapons will be accumulated, simultaneously creating pathways to weapons.
A complete ban on this particular technology will not cause severe scientific or industrial setbacks. All bioactive compounds that are currently produced in edible crops may as well be produced through other means that are less prone to hostile use. Some small biotech companies that specialize in biopharming may face problems, but others that focus on different technologies well benefit from such a move.
Fertility ControlCurrently, a variety of new methods for fertility control are under development, for use as contraceptives in humans but also for the biological control of pest animals. Some of them – such as the Australian mousepox experiment – pursue strategies that are based on vaccines, i.e. they try to direct an animal or human immune response against egg or sperm cells to prevent pregnancy and reproduction. It is too early to conclude that these experiments will be successful, but if so, “fertility vaccines“ present opportunities for abuse. If live vaccines are used (as in the mousepox example) that can be transmitted from individual to individual, a large population (of animals or people) may easily be prevented from reproducing, with enormous long term social and economic consequences.
Applied to ‘invasive alien’ or introduced species, such vaccines pose serious ecological threats (if the vaccine spreads to the target's geographic origin); but also significant risk of abuse to cause deliberate harm. This is particularly the case if such vaccines are developed to eradicate species of food or economic importance – for example, a ‘vaccine’ to control feral pigs, goats, rabbits, or other mammals that pose an ecological problem where they have been introduced might be transported to deliberately damage agriculture in other areas
Terminator TechnologySo-called 'terminator technology' renders seed infertile to guarantee a seed corporation's yearly sales. It may eventually be abused for economic warfare. If terminator crops become widespread, it would be easy for a country or a company that controls the technique to stop sales to a specific country or region for political or economic purposes. After some years of planting such seeds, only limited quantities of other seed would be available, thus agriculture could be paralyzed, leading to serious economic crisis and/or famine.
Current Projects in the USThe Sunshine Project has previously documented a series of recent offensive projects in the United States that draw on new developments in biotechnology. Military exploitation of new biotechnological possibilities, most notably with so-called “non-lethal” weapons, have fueled new weapons desires, even in countries that have renounced the use of biological weapons such as the US (and, in the case of “non-lethal” chemical weapons, Russia). The following three cases have been researched and previously published by the Sunshine Project, hence here we present only short summaries. Further reading is available on our website.
Material degrading microorganisms: [35] Natural microorganisms are capable of degrading nearly every kind of material. These organisms are sometimes used for environmental cleanup purposes (‘bioremediation’); but are generally too slow and unreliable for weapons purposes. Genetic engineering, however, is enabling development of organisms effective enough for use as biological weapons. The British government recently warned: "Bioremediation technologies clearly have the potential for development of a means of warfare or for hostile use against materiel crucial for normal civilian life or military operations, such as oils, rubbers and plastic." [36] This potential has raised the interest of several US government research institutions, including the US Naval Research Laboratory, where microorganisms that degrade a variety of materials (plastics, rubber, metals, etc..) were genetically engineered to make them more powerful and focused for bioweapons purposes.
Fungi against drug producing plants: [37] About a decade ago, the United States increased efforts to identify microorganisms that kill drug-producing crops. In the late 1990s, this research focused largely on two fungi. Testing of Pleospora papaveracea to kill opium poppy, conducted in Tashkent, Uzbekistan with US financing and scientific support, was completed in 2001. Pathogenic Fusarium oxysporum strains developed in the United States to kill coca plants were scheduled for field testing in Colombia in 2000, but international protests led to a halt to this project.
Military use of psychoactive substances: So called “non-lethal” chemical weapons were developed by the US military in the 1950s, especially a hallucinogenic substance called “BZ“ . But BZ was considered to be unreliable, leading to its removal from the US chemical arsenal in the late 1960s. Today, modern neurobiology is developing an increasingly comprehensive knowledge of a broad range of specific neuroreceptors and psychoactive substances that trigger (or inhibit) them. Military temptation to exploit these discoveries have made “non-lethal” chemical weapons again attractive for the military. A case in point was the use of a gas in the Moscow theatre hostage situation in 2002. Projects at the US Army’s Aberdeen Proving Ground and at the US Marine Corps Research University have recently investigated the military utility of a variety of incapacitating agents, including calmatives, seizure inducing agents and other psychoactive substances. The US and Russia are also developing delivery devices for chemicals with a range of more than 2.5 kilometers – a distance that makes only sense for warfare scenarios, and not for domestic law enforcement purposes.
Insect fightersThe idea to use insects to deliver biological warfare agents is not new. Insects were systematically explored as a mechanism to spread a variety of diseases (e.g. plague) in the World War II Japanese BW program and the postwar US program. In many cases, such insect vector BW was dismissed as too complicated and unreliable. But genetic engineering may open a new way to use insects as weapons. In the same way as genetically engineered plants may be misused as ‘food weapons’, insects may be engineered to produce toxic compounds and deliver them through their natural feeding habit – e.g. in the saliva of mosquitoes. Again, these compounds may exert a broad range of possible effect, from non-life-threatening illness to sterility to widespread fatal illness in a target population.
Techniques to use insects to deliver vaccines have already been developed and patented. [38] The idea to develop what one company calls ‘flying syringes’ is based on the hope of circumventing costly vaccination programmes in which every individual must be inoculated by trained medical personnel. Genetically engineered mosquitoes or other biting insects could instead deliver minute quantities of vaccine through the saliva every time they bite. The relevant techniques are still in their infancy. In comparison to genetic engineering of crops, for example, insects lag behind; but within several years, development of insect combatants may become a real possibility.
It is, however, questionable, whether genetically engineered insects may really become a weapon of choice. It will be nearly impossible to control these insects and limit their activity to the target country. Even if insects are choosen that are thought to be restricted to certain climate conditions, natural evolution and/or global climate change may rapidly overcome this restriction. State sponsored biowarfare programs tend to be very concerned about restricting unintended distribution of the biowarfare agent – most typical bacterial biowarfare agents are not contagious – and will thus hardly engage in the flying syringe concept.
VI. Ethnic specific biological weaponsCurrent wisdom holds that population specific biological weapons are practically and theoretically impossible. Practically, many consider it impossibly difficult to use genetic variability to kill or otherwise affect populations. Others, including geneticists, argue that no suitable ethnic specific genes exist in the first place. Both notions are wrong. New technologies are indeed available to translate specific genetic sequences into markers or triggers for biological activity. And a recent analysis of human genome data in public databases revealed that hundreds, possibly thousands, of target sequences for ethnic specific weapons do exist. It appears that ethnic specific biological weapons may indeed become possible in the near future.
Weapons targeting specific population groups do not need to be deadly. They could cause temporary incapacitation, illness, sterility, permanent fatigue, or any other condition that may not be fatal but desirable from an aggressors perspective. They may be used in an all out war, in the battlefield or against civilian population, or they may be used in covert operations in conflict situations and with long-term effects, in order to destabilise, harm economically or weaken an enemy society.
Techniques to translate genetic sequence into a weapons effectThe development of ethnic weapons with very specific effects would be easiest with techniques that use a genomic marker as a trigger for an activity that is unrelated to the location of the marker, i.e. the effect would be triggered even if the sequence is in a non-coding or non-translated region of the genome. As far as we are aware, this kind of technology does not yet exist.
There are, however, techniques available that can inhibit genes with a specific sequence. They target mRNA, the molecule that transmits information from the DNA to the place of protein synthesis within a cell. One of these techniques, called RNA interference (RNAi), uses a mechanism by which a specific RNA sequence is degraded by the cell if an externally applied RNA molecule of the same sequence is entering the cell (for review, see Cerutti 2003). A similar approach called antisense technology inhibits further mRNA processing by binding endogenously produced mRNA to an externally applied DNA molecule with the corresponding sequence. The latter technology is currently under development by the US company Ibis Therapeutics. [39]
Both technologies lead to the inhibition of a specific target gene with a specific sequence. If the sequence of the target gene varies from one population to another, this can be used to interrupt the gene in one population and not in the other. Military abuse of this technology would require the identification of population specific sequences in genes that are active and vital for the body function.
Ethnic specific genetic markersDo such genetic markers exist? Markers that are present in one population (at least to a certain percentage) but not in another? Many human geneticists are eager to emphasize that genetic diversity within a population is far greater than between populations. This view is also reflected in a 2001 background paper prepared by the British Government for the last Review Conference of the BWC. It states that "there is as yet no indication of differences that could be used as the basis for ‘genetic weapons’ which would target particular ethnic groups." [40]
99.9% of the genetic sequence of any two human individuals is said to be identical – but the remaining 0.1% accounts for a total of 3 million “letters” of the human genome. There are thought to be several tens of thousands coding genes in the human genome, thus it is possible that every single gene between one individual and another could be slightly (or greatly) different, even if there is 99.9% homology in overall genetic sequence. Some of this huge genetic diversity breaks out in differences between populations. These genetic populations (using the term in its biological sense) appear to often correspond with (culturally-determined) ethnic groups (for a detailed discussion on human genetics and the pitfalls of racial genetic profiling in general see Sankar & Cho 2002, Aldhous 2002, Schwartz 2001, Wood 2001).
From a biological weapons perspective, population specificity would mean more than just a small variation in allele frequencies in different ethnic groups – no effective weapon could be designed that targets a genetic constitution that is also present to any significant extent in the population of the aggressor. From a military perspective, population specificity would mean that these genetic sequences are not or only to a very limited extent present in one (the aggressor’s) population while the same sequences are present in a significant percentage of an opposing population. [41]
While it would certainly be desirable to have a very high percentage – up to 100% – of the target population bearing the target genetic marker, this is by no means a prerequisite for a militarily useful weapon. If as litte as 10% or 20% of a target population would be affected, this would wreak havoc among enemy soldiers on a battlefield or in an enemy society as a whole. Thus, in discussing genetic markers for ethnically-specific weapons, sequences would be needed that have a frequency close to 0% in one population while having a significant frequency in another. For the purpose of this paper, we assume that a frequency of 20% or higher may be enough from a military perspective.
Cytochrome P450 genes The many genes in the cytochrome P450 system have been suggested as possible targets for ethnic specific weapons, for two reasons. They show high ethnic diversity, and they are involved in the detoxification of toxic substances. The notion is that ethnic groups with specific polymorphisms in a cytochrome P450 gene may be less able to detoxify a specifically designed biological or chemical weapon and thus be more susceptible to its action.
In our view, these genes are probably useless as a basis for ethnic weapons, as diversity in most of these cases relates to different percentages of certain alleles in different population, not situations in which one population has a certain allele while the other does not. Hence, a significant part of the aggressor’s population would be potentially vulnerable. In addition, the P450 system comprises many dozens of enzymes with overlapping activities. Targeting a chemical or biological compound to one specific P450 enzyme would be very challenging.
A systematic search in two databases revealed that genetic sequences that fulfill these specifications not only exist, but they do so in unexpectedly high numbers. Our analysis focussed on so called single nucleotide polymorphisms – SNPs – that are by far the most common source of genetic variation. SNPs are basically single-letter variations in the human DNA sequence. In the past years, several million SNPs have been identified by private and public entities. The SNP Consortium (TSC), representing a group of large pharmaceutical companies and not-for-profit organisations, keeps a public database on a many SNPs. Another SNP database, the SNP500Cancer database, is maintained by the Cancer Genome Anatomy Project of the US National Institutes of Health. [42]
Both databases provide data on allelic frequencies in different populations. We analysed a total of nearly 300 SNPs, all in coding regions or genes,[43] from both databases. An unexpectedly high number of these SNPs are indeed population specific: 6.7% of the SNPs in one database (see table 1 below) and 1.6% of the SNPs in the other include one allele that is not present at all in one population while it has a frequency of more than 20% in another population.
From the database of The SNP Consortium (TSC),[44] SNPs were analysed for an ethnic specific allele distribution. The TSC database distinguishes between Caucasian, Asian and African-American samples.[45] From 105 randomly selected SNPs [46] in coding regions of the human genome, 21 had an allele frequency of 0% in one population but were present in at least one other population, 14 of these with a frequency ≥ 10% and 7 of these with a frequency ≥ 20%.
pop – population; A – Asian; C – Caucasian, AA – African American (e.g. A:C means that the minor allele is not present in the Asian population and has its highest frequency in Caucasians).
This finding is consistent with results from Stephens et al. (2001) who identified a total of 1,452 SNPs out of 3899 SNPs (37.2%) to be population specific, although the majority of these were rare SNPs. However, Stephens et al. (2001) also noted that "not all population-specific alleles were observed at a low frequency. In the African-American and Asian samples, some population-specific alleles were found at frequencies >25%."
In some cases, the frequency differences can be very high. For example, in our analysis of 105 SNPs from the TSC database, one SNP (TSC0493622) has a 0% : 94% ratio between major populations (see diagram 1 below). The G-allele of this SNP was present in 94% of the African-Americans and in 0% of the Asians sampled. The nature and function of the gene encoded by this genetic region is still unknown. Another example for a relatively high frequency difference is a polymorphism at the human melanocortin 1 receptor locus (MC1R), an enzyme involved in skin color formation. In a study by Rana et al. (1999) one allele was not identifiable in any Africans, but showed a frequency of 70% in East and Southeast Asians.
Diagram 1: Frequency of the minor allele of the 21 ethnic specific SNPs in the TSC database The majority of the population specific SNPs had a rather low frequency for one allele of less than 20%, but some SNPs with higher frequencies were also identified. 14 SNPs had an allele frequency of 19% and less, while only 7 SNPs had an allele frequency of 20% and higher. For SNPs with ethnic specific alleles in 2 populations, the higher frequency value was choosen for this diagram.
Some caution should be applied not to overestimate or interpolate our results. Both datasets as well as the work of Stephens et al. (2001) are based on a limited number of individuals for each population group. [47] Hence, alleles with a very low frequency in any one population may have been missed. Therefore it is possible and likely that some of the alleles that where not identified in one population group may well be present at low frequencies in these groups, so that many of the SNPs that were included in our analysis as they showed a 0% frequency for the minor allele would have to be excluded as their real frequency may be higher than 0%.
On the other hand, it is safe to assume that a certain percentage of the SNPs included in our analysis will prove to be population specific even if larger numbers of individuals were screened. There are examples of unsuccessful searches for alleles in large populations: The gene for thiopurine methyl transferase (TPMT) is an enzyme involved in metabolism of certain pharmaceuticals. Allele *3A, which is the predominant mutant TPMT allele in individuals of European heritage, has not been identified in East Asian populations despite the analysis of a total of 1068 individuals in 5 independent studies (see van Aken et al. 2003 for review).
To summarize, in can be estimated that a considerable number of ethnic specific SNPs do exist. Recent numbers suggest that SNPs occur with a frequency of about every 200 base pairs in the coding sequences of human genes (Schneider et al. 2003). Given the total number of about 3 billion base pairs, some 15 million SNPs may exist in the human genome. If in a conservative estimate only 0.1% (as compared to the 6.7% and 1.6% determined in our analysis of the two datasets) of these do occur population specific frequencies (here defined as 0% in one population and > 20% in another), some 15,000 possible target sequences may exist for future bioweaponeers.
It should be noted that some of the ethnic specific SNPs we identified in our analysis have a known function and are indeed readily expressed in human tissue. For example, the SNP rs2894804 from the SNP500Cancer database is located in a gene called GSTA1, coding for glutathione S-transferase. This enzyme functions in the detoxification of xenobiotics, including carcinogens, therapeutic drugs and environmental toxins. It was present in the African-American population with a frequency of 23% while it was not identified in any of the other three populations.
ConclusionsIt must be stressed that ethnic or population specific weapons are still a future threat and may not be accomplished within the coming decade. However, the notion that they are impossible and would violate the laws of nature is wrong and outdated. Practical steps can and must be undertaken today to prevent the future development of these kind of weapons. A key step would be to restrict the amount of ethnic specific genomic data to an absolute minimum. We are, however, currently witnessing a scientific development that is actually doing the opposite: creating vast amount of genetic data for different populations and ethnic groups. This happens in a variety of contextes:
Pharmacogenetics and pharmacogenomics: In order to elucidate genetic influence on drug safety and efficacy, an increasing number of studies on pharmacogenetically relevant genes are being undertaken. These include studies on genes for enzymes involved in drug metabolism such as the cytochrome P450 system and many others, but also genes coding for drug transporters or drug target proteins. For the safe implementation of pharmacogenetics on a global basis or in multicultural societies, reliable data on allele frequencies relevant to all populations is needed. Hence many pharmacogenetic studies investigate ethnic specific genetic differences relevant to drug action and are thereby generating large data sets that genetically profile on an ethnic basis. This problem may be circumvented by using pooled samples from a representative cross-section of all relevant population for the analysis of SNPs. Techniques are available today to calculate allele frequencies in pooled samples from up to several 100 individuals. Through this method, all relevant alleles in a pooled sample of all relevant populations could be determined without generating ethnic specific genetic data. The field of pharmacogenetics is specifically risk-prone, as the relevant genes are directly involved in drug metabolism or drug action and may thus be much more easily converted into triggers/markers for the action of biological or chemical agents than other genetic markers.
The HapMap Project: In October 2002, an international project to create a map of haplotypes [48] in the human genome was launched. [49] In this US$ 100 million public-private undertaking, genetic variations in four populations will be investigated: US residents with European ancestry, Han Chinese, Japanese and Yorubas in Nigeria. The HapMap project will provide vast amounts of genetic markers specific for any of the four populations. In the light of the possibility of hostile abuse of these genetic markers the HapMap project should be reconsidered.
Forensic genetics: Genetic fingerprinting enables to match a suspect’s DNA with that found at a crime scene. However, law enforcement is striving to get more information out of crime scene DNA, including the “race“ or ethnicity of the culprit. First steps have been taken in the direction of ethnic affiliation estimation by use of population-specific DNA markers (Shriver et al. 1997). The US National Institute of Justice recently issued a $496,000 grant to the University of Arizona to predict skin colour from DNA samples, [50] and the US-based company DNAPrint Genomics Inc. is offering to determine "race proportions" from crime scene DNA, although the technique is still prone with difficulties (Brenner 1998). It appears that these applications – if successful at all – could be of less concern from a bioweapons perspective, as they do not necessarily rely on markers that show a 0 : x percent distribution in different populations. In the course of the development of more sophisticated approaches for forensic ethnic affiliation estimation, however. if a systematic search for ethnic specific markers is undertaken it may reveal markers abusable for bioweapons purposes.
Others: Some human genetic studies touch on critical genetic data in politically tense areas, such as work on ethnic (Bhattacharyya et al. 1999) or even caste (Bamshad et al. 2001) associated genes in India, or genetic differences between the Basque and non-Basque population in Spain (Arrieta et al. 1997). A thorough assessment of benefits – if any – of this research and the associated risks of abuse appears to be necessary.
VII Conclusions and recommendationsTo summarize, genetic engineering can clearly contribute to make classical biowarfare agents more effective, it can ease access to them, enable the construction of novel BW agents and opens the avenue for a broad array of new types of weapons. It is of crucial importance for scientists and policymakers around the world to address the increasing threat and redouble efforts to strengthen the ban on biological weapons and to control critical technologies.
While the science behind the examples given in this paper is a reality, in most cases the hostile utilization of it (hopefully) has not occurred, so far. For example, terminator technology or fertility control technology do not appear to have been exploited for hostile applications, but it is obvious that once such technologies are more broadly exploited (particularly in commerce), they may become easily acquired and used with malign intentions.
Molecular biology and genetic engineering are still in their infancy. More technical possibilities will arise in the years to come that can be abused for hostile purposes. More efficient classical biowarfare agents will most likely play only a marginal role, even if the genetically engineered superbug is still routinely featured in newspaper reports. More likely and more alarming are the new types of weapons for newly-prevalent types of conflicts and warfare scenarios, for example, low intensity warfare and covert operations, for economic warfare or for sabotage. To prevent the hostile exploitation of biology now and in the future, a bundle of measures must be taken. First and foremost, the Biological Weapons Convention needs to be strengthened through multilaterally agreed, legally binding verification measures. In addition, three immediate steps are of specific importance:
All projects that violate the Chemical and Biological Weapons Conventions must be immediately abandoned. In the United States, such projects include the development of material degrading microbes, development of so-called “non-lethal” (bio)chemical weapons (including delivery devices), and continued development of biological agents to eradicate narcotic crops. Other countries that are engaged in similar projects – such as Russia, which maintains stockpiles of incapacitating chemical weapons and, likely, an R & D program on them – must also halt such research. These agents undermine the Chemical and Biological Weapons Conventions, are lowering the political threshold for use of biological weapons, and are likely to have tremendous environmental and health impacts. Pursuit of these agents as weapons would be a step down a slippery slope, that, following the same logic, could easily lead to the use of other biochemical and biological warfare agents in conflict. Failure to stop these projects will encourage other countries to follow suit with R & D projects on biotechnological weapons, leading to an unravelling of key disarmament treaties.
There is an urgent need to ensure that governments restrict themselves and ensure maximum transparency in their biodefense programs, to prevent a race for offensive capabilities under cover of defense. We call on all governments to adopt the ‘Government Undertaking on Biodefense Program’, which has recently been brought forward by the Sunshine Project. It contains, among others, a provision that “biodefense programs will not, for any purpose, utilize or construct, including single-gene changes, novel biological agents with an enhanced offensive potential” such as treatment resistance, environmental stability, or enhanced pathogenicity.
Research restrictions are necessary in certain situations, for example, in cases where a military abuse appears to be imminent, where no effective multilateral arms control or non-proliferation efforts are presently feasible, and where other technical avenues to reach the same scientific goal are (potentially) available. These criteria apply specifically to the production of bioactive compounds (pharmaceuticals, vaccines) in edible crops, but may also be relevant for some aspects of pharmacogenetics, were the generation of huge amounts of ethnic specific genetic data may be avoided by choosing other techniques that serve the same research purpose. The current ‘bioterrorism’ discussion in the scientific community focuses entirely on restricting the publication of certain research results. This is shortsighted, and may easily be abused to conceal illicit research, particularly since it may be better not to generate dangerous information in the first place. Full transparency in all aspects of biomedical research and development should be guaranteed.
VII ReferencesAldhous P (2002) Geneticist fears ‘race neutral’ studies will fail ethnic groups. Nature 418: 355-356.
Arrieta MI, Martinez B, Millan JM, Gil A, Monros E, Nunez T, Telez M, Martinez F (1997) Study of trimeric tandem repeat locus (SBMA) in the Basque population: comparison with other populations. Gene Geogr. 11:61-72
Bamshad M, Kivisild T, Watkins WS, Dixon ME, Ricker CE, Rao BB, Naidu JM, Prasad BV, Reddy PG, Rasanayagam A, Papiha SS, Villems R, Redd AJ, Hammer MF, Nguyen SV, Carroll ML, Batzer MA, Jorde LB (2001) Genetic evidence on the origins of Indian caste populations. Genome Res 11:994-1004
Basler CF, Reid AH, Dybing JK, Janczewski TA, Fanning TG, Zheng HY, Salvatore M, Perdue ML, Swayne DE, García-Sastre A, Palese P, Taubenberger JK (2001) Sequence of the 1918 pandemic influenza virus nonstructural gene (NS) segment and characterization of recombinant viruses bearing the 1918 NS genes. PNAS 98:2746-2751
Bhattacharyya NP, Basu P, Das M, Pramanik S, Banerjee R, Roy B, Roychoudhoury S, Majumder PP (1999) Negligible male gene flow across ethnic boundaries in India, revealed by analysis of Y-chromosomal DNA polymorphisms. Genome Res 9:711-719
Brenner CH (1998) Difficulties in the estimation of ethnic affiliation. Am J Hum Genet 62:1558-1560
Cello J, Paul AV, Wimmer E (2002) Chemical synthesis of poliovirus cDNA: generation of infectious virus in the absence of natural template. Science 297:1016-1018
Cerutti H (2003) RNA interference: traveling in the cell and gaining functions? Trends Genet 19:39-46
Haq TA, Mason HS, Clements JD, Arntzen CJ (1995) Oral immunization with a recombinant bacterial antigen produced in transgenic plants. Science 268:714-716.
Jackson RJ, Ramsay AJ, Christensen CD, Beaton S, Hall DF, Ramshaw IA (2001) Expression of mouse interleukin-4 by a recombinant ectromelia virus suppresses cytolytic lymphocyte responses and overcomes genetic resistance to mousepox. J Virol 75:1205-1210
Madjid M, Lillibridge S, Mirhaji P, Casscells W (2003) Influenza as a bioweapon. J Roy Soc Med 96:345-346
Manasherob R, Ben-Dov E, Xiaoqiang W, Boussiba S, Zaritsky A (2002) Protection from UV-B damage of mosquito larvicidal toxins from Bacillus thuringiensis subsp. israelensis expressed in Anabaena PCC 7120. Curr Microbiol 45:217-220
Rana BK, Hewett-Emmett D, Jin L, Chang BH, Sambuughin N, Lin M, Watkins S, Bamshad M, Jorde LB, Ramsay M, Jenkins T, Li WH (1999) High polymorphism at the human melanocortin 1 receptor locus. Genetics 151:1547-1557
Reid A, Fanning TG, Janczewski TA, McCall S, Taubenberger JK (2002) Characterization of the 1918 "Spanish" Influenza Virus Matrix Gene Segment. J Virol 76:10717-10723
Rosengard AM, Liu Y, Nie Z, Jimenez R (2002) Variola virus immune evasion design: Expression of a highly efficient inhibitor of human complement PNAS 99: 8808-8813
Sandmann, G., Kuhn, S., Böger, P. (1998) Evaluation of structurally different carotenoids in Escherichia coli transformants as protectants against UV-B radiation. Applied and Environmental Microbiology 64:1972-1974
Sankar P, Cho MK (2002) Toward a new vocabulary of human genetic variation. Science 298: 1337-1338.
Schneider JA, Pungliya MS, Choi JY, Jiang R, Sun XJ, Salisbury BA, Stephens JC (2003) DNA variability of human genes. Mechanisms of Ageing and Development 124:17-25
Schwartz RS (2001) Racial profiling in medical research. NEJM 344: 1392-1393.
Shriver MD, Smith MW, Jin L, Marcini A, Akey JM, Deka R, Ferrell RE (1997) Ethnic-affiliation estimation by use of population-specific DNA markers. Am J Hum Genet 60:957-964
Steinbrunner JD, Harris ED (2003) Controlling dangerous pathogens. Issues in Science and Technology, Spring 2003, pp. 47-54
Stephens JC, Schneider JA, Tanguay DA et al. (2001) Haplotype variation and linkage disequilibrium in 313 human genes. Science 293:489-493
Streatfield SJ, Howard JA (2003) Plant-based vaccines. Int J Parasit 33:479-493Taubenberger JK, Reid AH, Krafft AE, Bijwaard KE, Fanning TG (1997) Initial genetic characterization of the 1918 ‘Spanish’ influenza virus. Science 275:1793-1796
Tumpey TM, Garcia-Sastre A, Mikulasova A, Taubenberger JK, Swayne DE, Palese P, Basler CF (2002) Existing antivirals are effective against influenza viruses with genes from the 1918 pandemic virus. PNAS 99:13849-13854
Upton C, Slack S, Hunter AL, Ehlers A, Roper RL (2003) Poxvirus orthologous clusters: toward defining the minimum essential poxvirus genome. J Virol 77:7590-7600
van Aken JP, Schmedders M, Feuerstein G, Kollek R (2003) Prospects and Limits of Pharmacogenetics: the TPMT Experience. Am J Pharmacogenomics 3:149-155
Wheelis M, Dando M (2002) On the brink: biodefence, biotechnology and the future of weapons control. Chemical & Biological Weapons Convention Bulletin 58:3-7
Wheelis M, Dando M (2003) Back to bioweapons? Bulletin of the Atomic Scientist 59:40-46
Wood AJJ (2001) Racial differences in the response to drugs – pointers to genetic differences. NEJM 344: 1393-1395.
FOOTNOTES1. In 1918, a particularly aggressive influenza virus spread around the globe and killed 20 – 40 million people. This influenza pandemia was dubbed the ‘Spanish flu’.
2. See
www.sunshine-project.org. In Germany, a petition supported by a variety of organisations including the Sunshine Project is currently underway to encourage the government to formally adopt high transparency and strict limits for its biodefense program.
3. See
http://fas.org/bwc/index.html for further reading on recent BWC developments.
4. Specifically, it has become public in the past months that the US is pursuing development of so called non-lethal chemical weapons, material degrading microorganisms and an array of questionable ‘biodefense’ activities. See
www.sunshine-project.org, Wheelis & Dando (2002, 2003) and Steinbrunner & Harris (2003) for further reading.
5. Petro JB, Plasse TR, McNulty JA (2003) Biotechnology: Impact on biological warfare and biodefense. Biosecurity and Bioterrorism Volume 1, Number 3
6. Appeal of the International Committee of the Red Cross on Biotechnology, Weapons and Humanity. September 2002 (online at
www.icrc.org).
7. Before Pasteur and Koch discovered bacteria as disease causing agents in the late 19th century, biological weapons were used. For example, in the 14th century, Mongol invaders catapulted plague victims into besieged cities. In the 18th century Britain distributed smallpox-infected blankets to native Americans.
8. Borzenkov VM, Pomerantsev AP, Ashmarin IP (1993) The additive synthesis of a regulatory peptide in vivo: the administration of a vaccinal Francisella tularensis strain that produces beta-endorphin Biull Eksp Biol Med 116(8):151-3 (Article in Russian)
9. Jane’s Defence Weekly, 13. August 1997, page 6: US DoD reveals horrific future of biological wars
10. Pomerantsev AP, Staritsin NA, Mockov YV, Marinin LI (1997) Expression of cereolysine ab genes in Bacillus anthracis vaccine strain ensures protection against experimental hemolytic anthrax infection. Vaccine 15:1846-1850
11. New York Times, 4. September 2001
12. A. Hay, quoted in ‘The bugs of war’, news feature in Nature 411:232-235
13. An exception may be sophisticated non-state actors which may seek to apply modern genetics for their own hostile interests, especially for low level or private conflicts. This refers less to non-state actors such as Al Qaeda but rather to companies and/or single individuals which due to their professional background have the capability to do so.
14. Background paper on new scientific and technological developments relevant to the convention on the prohibition of the development, production and stockpiling of bacteriological (biological) and toxin weapons and on their destruction. BWC/CONF.V/4/Add.1, 26 October 2001.
15. US develops lethal new viruses. New Scientist, 29 October 2003.
16. Arthur C “Scientists made virus ‘more lethal than HIV’, The Independent, 24 July 2001.
17. For review, see the complete volume 264 of Curr Top Microbiol Immunol (2002), edited by Hacker J & Kaper JB, which focuses on ‘Pathogenicity Islands and the Evolution of Pathogenic Microbes’.
18.
http://www.science.doe.gov/sbir/Solicitations/FY%202003/NN.htm#T119. US Patent 5662908 from 2 Sept. 1997, assigned to Stanford University in Palo Alto, California.
20. In more than 95% of infected persons, only mild flu-like symptoms – if any – are caused by the virus. With only about 1% of the infected having the risk of severe illness, polio does not rank high on a bioweaponeer’s wish list.
21. One sequence of smallpox (Variola virus) with the GenBank code X69198 (identical with NC_001611) was published by a team from Russia’s former offensive biowarfare program, and a second sequence (Variola major virus strain Bangladesh 1975) with the GenBank code L22579 was published by an American team.
22. Personal communication on 26 June 2003 by Dr D. Wheeler, NCBI, to Jan van Aken, Sunshine Project
23. Stanford University News Release 17 September 2003, online at
http://mednews.stanford.edu/news_releases_html/2003/septrelease/bioterror%20flu.htm24. Spanish flu keeps its secrets. Nature science update at
www.nature.com/nsu/990304/990304-5.html25. Profile: Jeffery Taubenberger at
www.microbeworld.org/htm/aboutmicro/what_m_do/profiles/taubenberger.htm26. AFIP scientists discover clues to 1918 Spanish flu,
www.dcmilitary.com/army/stripe/archives/mar28/str_flu032897.html27. The so called ‘nonstructural’ gene (NS)
28. It should be noted that for this experiments, a standard influenza strain was used that was specifically adapted to mice and that was lethal to mice. The scientists reasoned that the 1918 gene probably weakened the lethality for the mice as it stemmed from a human-adapted strain.
29. This time, the 1918 genes for hemagglutinin (HA), neuraminidase (NA) and matrix (M) were used, single and in combination. Only the combination of the 1918 HA and NA genes caused a dramatic increase in lethality if compared to constructs containing genes from a more recent human influenza virus. The scientists concluded: “These data suggest that the 1918 HA and NA genes might possess intrinsic high-virulence properties.”(Tumpey et al. 2002:13853)
30. Letter (4 February 2003) from Robert G. Webster, Professor of Virology at St. Jude Children’s Research Hospital to Stanley Lemon, Dean, School of Medicine, University of Texas Medical Branch (UTMB), in support of the UTMB application to construct a National Biosafety Laboratory.
31. See, for example, ProdiGene press release, 12 August 2002: ProdiGene and NIH beginning phase I study on oral vaccine derived from transgenic corn. At
www.prodigene.com.
32. For a detailed discussion of possible effects on the environment and human health see the background paper “Manufacturing drugs and chemicals in crops” published by Friends of the Earth;
http://www.foe.org/camps/comm/safefood/biopharm/BIOPHARM_REPORT.pdf33. Both, ricin and trichosanthin, are ribosomal inhibitor proteins.
34. For an overview on the escape and potential risks of StarLink corn see Washington Post, 19. March 2001, ‘Biotech Corn Is Test Case For Industry’,
http://www.washingtonpost.com/ac2/wp-dyn/A23092-2001Mar18?language=printer35. See Sunshine Project Backgrounder #9 (
http://www.sunshine-project.org/publications/bk/bk9en.html) for further reading.
36. See footnote 14
37. For extensive reading on Agent Green see the Sunshine Project Backgrounder No. 4 and additional materials at
www.sunshine-project.org.
38. See European patent PCT/GB95/02639 and US patent application 20020124274 (September 5, 2002) by Imperial College of Science Technology and Medicine (London) for a ‘delivery system”.
39.
www.ibisrna.com40. See footnote 14
41. It must, however, be questioned how good the ‘zero’ frequency of the target allele on the aggressor’s side has to be. This may depend heavily on the effect of the ethnic weapon and on the political system of the aggressor. Dictatorships may well accept more ‘collateral damage’ in their own society than others. And if the effects are non-lethal and long-term – such as sterility – it may be more acceptable for an aggressor to have some victims on its own side. If used in a battlefield, an aggressor could also screen and select its soldiers according to this specific sequence, or could apply specific countermeasures.
42.
http://snp500cancer.nci.nih.gov/snplist.cfm. This program studied the genome of 102 individuals of self-described heritage: 24 of African/African American heritage, 31 of Caucasian heritage, 23 of Hispanic heritage, and 24 of Pacific Rim heritage. In this database, we analysed 193 randomly selected SNPs (all validated SNPs in chromosomes 6 and 10). A total of 24 SNPs (12%) showed an allelic freqency of ≥ 10% in at least one population with a 0% frequency in at least one other population. 3 of these (1.6%) had a frequency of 20% or higher in one population.
43. As discussed above, it appears to be a prerequisite for militarily useful genetic markers to have them appear in coding sequences or genes that are active in the human body, rather than in apparently silent parts of the human genome. If new technologies are developed that can use even apparently inactive genomic sequences as a trigger for the desired effect, this would make it easier to translate these genetic differences into weapons.
44.
http://snp.cshl.org/ as of June 24, 2003
45. See
http://snp.cshl.org/allele_frequency_project/panels.shtml for a description of the panels.
46. All SNPs in coding (both synonymous and non-synonymous) regions with a TSC-ID number and with allele frequencies provided for at least two different populations from the first 100MB of chromosomes 1-10 were included in the analysis.
47. The SNP500Cancer Database is based on 23-31 individuals per population group; the TSC-database is based on different panels, most of which included 12-42 individuals per population group; Stephens et al. included 18-21 individuals per population group.
48. Haplotypes are blocks of closely linked SNPs in a genome and are currently viewed as the best tool to study human genetic variation.
49. See
http://hapmap.cshl.org/ for details.
50. NIJ grant number 2002IJCXK010.
