Appendix B
The Current Regulatory Framework
ENGINEERED MICROORGANISMS AND THE CHEMICALS THEY PRODUCE
The Environmental Protection Agency (EPA) regulates entities that produce and commercialize new chemicals under the Toxic Substances Control Act of 1976 (TSCA). Under this Act, the EPA also regulates the commercial research and development, manufacturing, importing, and processing of “inter-generic microorganisms” intended for commercialization. An inter-generic microorganism is one “formed by the deliberate combination of genetic material originally isolated from organisms of different taxonomic genera.” The EPA has reviewed applications for commercialization of approximately 75 microorganisms since 1998, and it construes the regulatory definition to include organisms created through synthetic biology. However, commentators have noted that microorganisms made using synthetic DNA sequences not found in any existing organism might not fall within the EPA’s definition. In the next 5 to 10 years, however, it is unlikely that the DNA sequences used to engineer organisms for industrial chemical production will use completely de novo DNA sequences, and thus the regulatory definition is probably sufficient for this stage of industrial biology.
At least 90 days prior to manufacturing, importing, or processing a new, inter-generic microorganism for commercial purposes, the responsible firm must submit a complete Microbial Commercial Activity Notice (MCAN), or an exemption request, to the EPA for review. The MCAN must contain test data the manufacturer possesses or controls, and information
from the scientific literature, about the microorganism’s effects on humans (including worker exposure), animals, plants, and other microorganisms. In addition, the MCAN must include information about the identity of the organism, genetic manipulations used to construct it, its properties and phenotype, the traits that have been selected for or modified, the byproducts it produces, and its proposed uses and environmental releases.
The EPA’s review of an MCAN should determine whether the microorganism, under the proposed conditions of use, is reasonably safe for humans and the environment. After the 90-day review period has expired, a firm can manufacture, import, or process the microorganism if the EPA has not taken regulatory action to prevent or constrain such activities. If the agency determines that a chemical or microorganism presents an unreasonable risk of injury to health or the environment, then it can prohibit or limit the manufacture, distribution, or processing of the chemical or microorganism by writing a new regulation.
Review of an exemption request will determine whether the firm’s commercial use of the new, inter-generic microorganism is exempt from TSCA requirements. Exemptions are granted for use of microorganisms, genetic manipulations, and production processes with which the agency has long experience and for which there is a record of safety.
If the new chemical will be produced by a microorganism intended for release into the environment, such as engineered algae grown in an open-pond system, then the manufacturer would have to conduct field tests prior to submitting the MCAN. Field tests of microorganisms present more uncertainty about health and environmental impacts than field tests of chemicals, because microorganisms can replicate and might proliferate beyond the immediate test site or might transfer genes to related organisms in the wild. One cannot as easily control the quantity of a microorganism released, or counteract problems should they occur.
To field test a microorganism intended for commercial use, the manufacturer must submit a TSCA Experimental Release Application (TERA) to the EPA at least 60 days before any field test could commence, unless the proposed testing is eligible for an exemption. The TERA must be approved, with or without conditions, before testing begins. Whoever conducts the research (manufacturer or a contractor) must comply with all terms and conditions in the TERA. The EPA only approves a TERA if it determines the proposed research “does not present an unreasonable risk of injury to health or the environment.” The EPA can also revoke or modify a TERA if it receives new information concerning research risks. Data from field tests conducted under the TERA must be included in the MCAN if the tests are completed during the MCAN review period.
There is some question regarding whether the EPA can adequately review TERA applications for release of organisms created through syn-
thetic biology approaches within the 60-day time frame. Synthetic biologists cannot always predict the effects of complex combinations of synthesized DNA on the organisms into which they are engineered, and the agency lacks models for assessing the health and environmental risks of such organisms to determine whether they are likely safe enough to release in a field test. As scientists and the agency become more experienced, some combinations of DNA segments in some organisms will become more predictable and regulatory risk-prediction models will be developed.
A second concern about precommercial testing of microorganisms has to do with whether the EPA has developed adequate insights into and guidance concerning containment for microbial organisms tested “inside a structure” (not intentionally released into the environment). Manufacturers conducting tests inside a structure need not report to the EPA if they meet certain criteria, although they must keep specified records. The EPA does not require that such interior testing meet the National Institutes of Health’s (NIH’s) Guidelines for Research Involving Recombinant or Synthetic Nucleic Acid Molecules, although it gives exemptions from EPA oversight to research conducted with federal funding from another agency and that is required to meet the NIH Guidelines. Researchers and administrators from several agencies have extensive experience applying the NIH Guidelines, and industry might be well served to voluntarily adopt these guidelines for testing of industrial microorganisms in contained spaces.
Another concern about the EPA’s authority under the TSCA and its implementing regulations is that the agency’s ability to require safety testing of engineered organisms (or new chemicals) requires substantial administrative process and might not be used effectively. The submitter of the MCAN must only provide information within her possession or control; the firm generally need not conduct research to generate new information. If the EPA believes information in the MCAN is inadequate for judging safety, then it has some authority under sections 4 (the “test rule”) and 5(e) of TSCA to require that a manufacturer conduct and report safety research (such as generating and reporting toxicity data). Usually, firms and the EPA negotiate an agreement regarding the data to be generated, and the firm voluntarily signs a Consent Order. However, if the manufacturer does not voluntarily agree, then the agency must write a new regulation to require testing. Such a procedure can take several years and can be challenged in court.
A third concern is whether the EPA has sufficient postmarket authority to regulate industrial microorganisms. The EPA can regulate inter-generic microorganisms that produce chemicals even after the MCAN 90-day review period has expired and the product has entered the market;
however, the agency must have evidence of harm to health or the environment before it may take action. The TSCA and its implementing regulations require manufacturers and distributors to keep records of specified negative health or environmental effects, but the regulations do not require that such information be reported to the EPA unless the agency explicitly requested reports, and the regulations do not require that firms conduct studies to identify problems. In the event of an accidental environmental release of an inter-generic microorganism used to manufacture chemicals, the firm responsible would have to keep records of health and environmental effects of the “spill,” but TSCA does not require that a firm limit or mitigate those effects.
The EPA could have jurisdiction over some types of accidental microbial release under other laws (beyond TSCA). For instance, if algae used for industrial production of a chemical, such as a plastic, were to escape containment and get into natural waterways or lakes, the EPA might have jurisdiction under the Clean Water Act. However, in many instances people affected by an accidental release of an industrial microorganism might have to bring a tort action against the manufacturer, distributor, or other relevant party to obtain environmental cleanup and monitoring.
In the event of an accidental or uncontrolled harmful release of industrial microorganisms, the EPA could act under section 6 of TSCA to impose new containment conditions to prevent similar incidents in the future. However, to take such an action the agency must make a finding, based on risk-benefit calculations, that the microorganism as manufactured and distributed prior to the release posed an unreasonable risk, and that the regulatory action proposed was the least burdensome regulation that could provide adequate protection. The courts have ensured that the evidentiary burden for such findings is high, and the EPA had only issued “section 6 rules” for five chemicals as of 2005.
EPA officials believe they have adequate authority under TSCA to protect the environment, public health, and worker safety by imposing conditions on the manufacture and use of engineered microorganisms. The agency can prohibit or limit the use of new chemicals or engineered microorganisms that pose unreasonable risks to health or the environment, it can enjoin the manufacturing or processing of a commercial microorganism, and it can impose criminal or civil penalties on manufacturers who do not comply with TSCA. The EPA has been creative in using its premarket authority under TSCA to protect health and the environment and has developed cooperative relationships with chemical producers. It has used Consent Orders to impose conditions on particular manufacturers of microorganisms, including conditions for containment and worker safety, and it has used Significant New Use Rules (SNURs) to impose industry-wide standards. However, as with other regulations, a
SNUR can take years to promulgate. In 2012, the EPA proposed a SNUR for the fungus Trichoderma reesei, inter-generic versions of which are used to make enzymes for ethanol production. Agency officials were concerned that the microbe could produce toxic peptides under some growth conditions and that it might not be properly contained. As of this report’s writing, the Trichoderma SNUR has not been finalized.
A final concern about the EPA’s regulation of industrial microbes is that the growth of industrial biology could result in a flood of MCANs and TERAs, which might overwhelm the agency. The agency has reviewed approximately 75 engineered microbes since 1998, a very limited number when compared, for instance, to the thousands of decisions the U.S. Department of Agriculture (USDA) has made over that same time period regarding the field testing and release of engineered crops. If the EPA lacks sufficient staff or resources, then the quality of its reviews could suffer or it could become a bottleneck in the pathway to market. At some point, the EPA might have to reallocate personnel and resources toward review of biological production of new chemicals, or it may need additional resources to carry out such reviews.
CROPS AS BIOREACTORS
If crops are used as bioreactors to produce industrial chemicals, then the USDA Animal and Plant Health Inspection Service (APHIS) is usually the lead agency with jurisdiction, which it could share with the EPA or the Food and Drug Administration (FDA), depending on the chemical’s intended uses. Under the Plant Protection Act, APHIS regulates the importation, interstate movement, and environmental release of “plant pests,” including genetically engineered organisms that might pose a risk to plant health. To date, most transgenic plants intended for commercial use have been modified using vectors or genes from plant pests, and thus APHIS has had jurisdiction over nearly all of them.
If a new, genetically engineered organism (including a plant) is known or suspected to cause damage or disease to a plant or plant product, then it cannot be introduced into the environment without a field trial conducted under APHIS’s authorization. To conduct such a trial, the plant’s developer must file either a notification or permit application with APHIS. Prior to approving the field release of regulated material, APHIS will conduct a review of the notification or application to ensure that under the proposed conditions of use—handling, confinement, and disposal—the risks to plant health and the environment have been appropriately minimized. Permits are generally more restrictive than notifications and are used for types of plants that pose heightened risk to plant health or the environment, or for plants with novel modifications whose risks are
uncertain and with which the agency has less regulatory experience. Notification is used for plants that pose low risk and contain modifications with which the agency has had familiarity. APHIS typically uses permits for plants that produce biopharmaceuticals and industrial chemicals. Sites where a plant-produced biopharmaceutical or industrial chemical is being field tested will be inspected several times during the trial.
Once a new, genetically modified plant has been field tested the developer can file a petition for deregulation, which must include information about the plant’s biology (including genetic modifications) and the field test results.
In evaluating such petitions, APHIS considers numerous factors, including “the expression of gene products, new enzymes, or changes to plant metabolism; weediness and impact on sexually compatible plants; agricultural or cultivation practices; effects on non-target organisms; and the potential for gene-transfer to other types of organisms.” In addition to assessing the petition, the agency will prepare an environmental assessment or an environmental impact statement and seek public comment on the plant’s risks. If APHIS reaches the conclusion that the new plant does not pose a plant-pest risk, it will approve the petition. Once the organism has been deregulated, it can be introduced into fields, and into commerce, without APHIS oversight. As of August 2013, APHIS had overseen the deregulation of 95 genetically engineered crops.
Some new, engineered organisms can be deregulated using a “request for extension of non-regulated status.” This process was established in 1997 and assumes that, from a safety standpoint, many regulated organisms will have only negligible differences from previously deregulated ones. In the extension request, the petitioner compares the regulated organism to an antecedent, deregulated organism to show that the molecular manipulation used to make the new organism raises no serious, new risks to plants or the environment. Extensions are used for interventions such as synonymous nucleotide changes (ones that do not change the amino acid sequence of the encoded protein) and may be less often available to plants engineered using synthetic biology approaches.
When both APHIS and the FDA have jurisdiction over a genetically engineered plant, for instance, if the commercial aim is production of a plant-made pharmaceutical, then APHIS takes the lead in regulating premarket field tests of the engineered plant and the FDA would later subject the plant-made chemical to the premarket approval process typically used for drugs. And, APHIS could share jurisdiction with the EPA if, for example, the engineered organism was a known bacterial plant pest.
APHIS only has jurisdiction over genetically modified plants or other organisms under the Plant Protection Act if there is reason to believe that the engineered organism would harm plants. APHIS does not have broad
jurisdiction to consider possible environmental or health hazards. Thus, there is a category of engineered plants whose field tests or commercial use would not fall under APHIS’s jurisdiction, and that the EPA also does not have clear authority or experience to regulate. If such plants were producing industrial chemicals not intended for a FDA-regulated purpose, then their release into the environment and their commercial distribution would be completely unregulated. Switchgrass engineered for optimal use as a feedstock in biofuel production is an example of a genetically modified plant that did not fall within APHIS’s regulatory authority and is otherwise unregulated.