Page 55

D

Workshop Presentations

THE CURRENT THOUGHT ON BIOTERRORISM: THE THREAT, PREPAREDNESS, AND RESPONSE

David R. Franz

Southern Research Institute

I have spent the last 15 years thinking about biological warfare. Unfortunately, over the last few months that topic has become extremely popular. Thus, this morning I am going to offer a broad perspective of biological terrorism, with the understanding that this biology is a subdivision of chemistry.

The last 60 years, as evidenced by the former U.S. and Soviet programs, has been the modern era of biological warfare. The agents studied in both nations' programs were very similar, and the workhorse agents in both programs were the zoonotic agents—agents that are transmissible from animals to humans (see Figure D.1). The physical and infectious properties of biological agents are not all the same: they differ in how they act; whether they cause illness or death; their stability during growth, production, weaponization, storage, and dissemination; the number of organisms that cause illness; and how infectious they are. The only seriously contagious biological agents, smallpox and plague, were actually weaponized by the Soviets and put into refrigerated nosecones of ballistic missiles to blanket the United States.

The fundamental differences between chemical and biological agents are very important as we look to the future. Chemical agents are volatile and dermally active and it is immediately apparent when there is human contact with a chemical agent. Biological agents, however, are not volatile, not dermally active, and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 55
Page 55 D Workshop Presentations THE CURRENT THOUGHT ON BIOTERRORISM: THE THREAT, PREPAREDNESS, AND RESPONSE David R. Franz Southern Research Institute I have spent the last 15 years thinking about biological warfare. Unfortunately, over the last few months that topic has become extremely popular. Thus, this morning I am going to offer a broad perspective of biological terrorism, with the understanding that this biology is a subdivision of chemistry. The last 60 years, as evidenced by the former U.S. and Soviet programs, has been the modern era of biological warfare. The agents studied in both nations' programs were very similar, and the workhorse agents in both programs were the zoonotic agents—agents that are transmissible from animals to humans (see Figure D.1). The physical and infectious properties of biological agents are not all the same: they differ in how they act; whether they cause illness or death; their stability during growth, production, weaponization, storage, and dissemination; the number of organisms that cause illness; and how infectious they are. The only seriously contagious biological agents, smallpox and plague, were actually weaponized by the Soviets and put into refrigerated nosecones of ballistic missiles to blanket the United States. The fundamental differences between chemical and biological agents are very important as we look to the future. Chemical agents are volatile and dermally active and it is immediately apparent when there is human contact with a chemical agent. Biological agents, however, are not volatile, not dermally active, and

OCR for page 55
Page 56 ~ enlarge ~ FIGURE D.1 Of the diseases often associated with biological warfare, zoonotic diseases are the most likely to be used as weapons. have delayed onset of disease. These properties place serious constraints on biological agent delivery. It is easiest to present biological weapons as irrespirable aerosols because they are nonvolatile, but it is a complicated task to prepare agents in that form. Difficulties also arise because distribution of the agent in air is completely dependent on meteorology (outdoors or indoors). Biological agents also present problems to antiterrorist law enforcement officials. Although chemical weapons can cause fairly immediate death, biological weapons yield no sign of exposure. There are also no tools to detect exposure to a biological agent before the onset of sickness. Medical doctors are unfamiliar with exotic agents used in biological weapons, and the prevalence of flu-like symptoms in the beginning stages of agent-induced illness often leads to misdiagnosis. Not only does this yield a higher probability of serious health problems or even death if infected, but these factors feed an enormous psychological fear of biological agents. Other problems include the lack of “universal” vaccines, the lack of antiviral drugs, social or political issues revolving around prophylaxis, and the difficulty of forensics. Biological terrorism is a unique threat because of its dual-use nature, evolving technologies, and political factors. The production of vaccines, for instance, requires the growth and subsequent death of viruses or bacteria. A certain veterinary vaccine facility in Russia is capable of producing 12 metric tons of

OCR for page 55
Page 57 foot and mouth disease virus in one run. It has been admitted that during World War II, the mission of this facility was to produce 240 metric tons of variola virus, which causes smallpox, to be used as a weapon. A similar dual-use situation exists with technology like crop dusters. It is difficult to know if pesticides are being sprayed or biological weapons are being tested. Biotechnology such as genetic modification adds yet another dimension to the prophylaxis/vaccine dilemma. Political issues over the last 10 years have had a profound effect on the threat of biological terrorism. As the value of the Russian ruble fell and the country continued in its decline, between 30,000 and 40,000 Russian scientists and engineers, formerly employed at Ministry of Defense weapons facilities, lost their jobs. Perhaps these highly skilled scientists, who had families to feed and rent to pay, simply switched careers, but some were certainly recruited by Syria, Libya, Iran, and North Korea. It is only in the last 4 or 5 years that the American public has become aware of the perceived terrorist threat. It is actually not significantly different from the threat during the Cold War. To produce an event that causes the death of thousands in this nation, the same agents selected during the Cold War must be used for the same reasons (ease of production, ease of distribution, rate of infection). Most likely, state sponsorship is required to produce the agents in the kind of formulation that would be effective for such a scenario. The exception is the use of a highly contagious agent like smallpox, which would not need to be weaponized. Agricultural terrorism remains a possibility. The threat here is not to the human body; animal disease agents are not human pathogens and we could safely eat infected meat. The threat here is to the economy: in 1997 one infected and contagious piglet in Taiwan decimated the pork production capacity of the island, costing $5 million in initial damages and eventually costing $14 billion in lost revenue. Where do the risks truly lie? The highest concern lies with highly contagious viruses like smallpox. Such a virus could cause the greatest damage; however, a smallpox attack is the least likely scenario to occur because the only legal supplies of the virus are in Atlanta and Novosibirsk. Illegal supplies probably do exist, but are difficult to obtain. Foreign animal disease viruses have the next greatest decimation potential. Classical agents that we normally think of but that are fairly difficult to deliver effectively are next on the list and then the hundreds of other more mundane agents like salmonella that will cause illness but not death. In the future, we may also need to beware of genetically engineered agents. Bioterrorism may occur in the United States today because we cannot be beaten with conventional methods. We do not know how much “brain drain” occurred in the former Soviet Union, and the dual-use nature will never allow the problem to be dealt with through regulation. In the event of an attack, it would be ideal to have a universal detector that would simply indicate “yes” or “no” to the

OCR for page 55
Page 58 presence of any biological agent. The next steps would be to identify the tack, the agent, and the people involved, then to neutralize, decontaminate, and remedy the attack site. All of this would ideally be accomplished in 24 to 48 hours, and without any public panic. Detection is an area that chemists have been involved in for quite a while. Detection of biological agents is much more difficult than originally believed at the beginning of the Gulf War. Although anthrax, with an infectious dose of 100 organisms per liter is relatively easy to detect, many agents like Q fever have an infectious dose of only 10 organisms, requiring sensors to detect 10 organisms in 100 liters. Currently, biological agent detection in an infected person can only be accomplished by measuring the antibody response (unless methods like nasal swabbing are used, which have a high rate of false negatives, especially for agents that require a very small number of organisms for infection). However, it takes a few days for the antigen to circulate in the blood and for the immune system to respond (see Figure D.2). At this point, it is almost too late to treat the infected individual. Chemical and biochemical research needs to focus on tools to allow us to identify exposed individuals before the onset of clinical disease. The question remains, how do current times compare with the Cold War era? It is now clear that there are people in the world willing to bring harm to civilians. Medical doctors must add new diseases to their differentials. The public now has ~ enlarge ~ FIGURE D.2 Waiting to measure the antibody response to an infection allows the disease to progress too far.

OCR for page 55
Page 59 a better understanding of the threat through the experience of the anthrax letters, and funding for counter-bioterrorism will increase. The nation's list of vulnerabilities is undeniably different. However, the technical difficulty of bioterrorism, the importance of meteorology, the difficulty of intelligence, detection and response (public health), deterrence and preparedness (law enforcement), and education on related issues have not changed at all since the end of the Cold War. The solution comes as a two-pronged approach: education and a strong technical base. We have to educate people to be aware of the risk. Any developed nation in the world could produce biological weapons, and we, as scientists who speak the worldwide language of science, have an opportunity to prevent this. Our nation also must have a strong technical base to be ready for both the expected and the unexpected. This will improve our surveillance, diagnostics, and communication (data integration). So many unanswered questions remain regarding biological agents, the immune system, detection, and the like. The time is ripe for basic research. The current bioterrorism situation is about more than just science and gadgets. I think it's reflected by the flags that I see on my street and on your street. It's reflected by the flags I see now on the lapels of business suits that I didn't see before. And I think it's about the American spirit: we're all Americans or scientists and we have a very important job to do.

OCR for page 55
Page 60 WHAT CAN THE INDUSTRIAL CHEMICAL COMMUNITY CONTRIBUTE TO THE NATION'S SECURITY? Scott D. Cunningham DuPont Today I am speaking on behalf of larger chemical companies. While I can only relate what I have seen at DuPont, my experiences are most likely typical. The first response of the chemical industry to the events of September 11 was to donate materials needed in the recovery effort. Then, it began to think about its own vulnerability. Generally, the chemical industry enjoys a good safety record because of a necessarily heightened consciousness of the many hazardous materials handled. Recent events, however, have shown that technically knowledgeable people are willing to die in attacks against U.S. interests, and this is what causes concern in the industry. Companies began to rethink nearly everything, from infrastructure to checking identification at the door. They have changed capital expenditure plans, logistics, operations, and research and development. Industry members are now using the momentum of the September attacks to identify weak points in their systems and to reengineer operations. For example, storage tanks have became a major concern as targets of terrorist attacks, causing companies to redesign processes to minimize the accumulation and storage of hazardous intermediates and products. As the resident HAZMAT authorities in many areas, chemical company employees found themselves responding to numerous community requests. There were so many calls for assistance that there was a shortage of trained responders to handle them. People began to think about communication devices, new systems, and new ways of doing things. Questions were asked like “What did I learn and how do I do better next time? When a call goes out, how do you mitigate the damage? What systems should be blocked off when entering a potentially contaminated building?” The experience gained through helping communities allowed chemical companies to improve their response procedures. Chemical companies also received calls for many perceived antiterrorist materials such as filters, Tyvex, Nomex, Evacuate, and Kevlar, and pharmaceutical companies received calls for vaccines. The industry began to investigate who was buying these products and whether the products would meet the customers' needs. Materials scientists began to contemplate faster and different production methods, to combine products, and to define new uses for older products. Companies asked, “What do these things mean to the world? How can we make them better and protect our people and our food?” People responded creatively; for example, pharmaceutical companies are now working on faster vaccine production and thinking about new production paths through plants or bacteria. In the corporate boardroom, the volition to act runs high, not to make a profit, but to help the national security effort. In fact, the chemical industry has a history

OCR for page 55
Page 61 of responding to national needs, evidenced by their production of black powder in the War of 1812, explosives and synthetic materials during the world wars, and materials for the space program. For this national effort, though, it is not clear whether there are technology solutions, and it is uncertain what the chemical industry can do to help. The chemical industry has a proven track record of safety, and has helped companies to make their production plants safer. The industry also has global presence and experience, providing decision tree analysis and disaster planning around the world, including prevention, detection, damage mitigation, and mediation. The chemical industry is capable of integrating multiple sciences to produce a large amount of product at low cost. The industry has great experience with sensors, and is the largest designer, user, and consumer of sensors for risk avoidance and for quality control. These sensors have already been integrated into production plants, but not yet into office building monitoring. Many chemical companies produce decontamination agents. DuPont, for instance, makes chlorine dioxide, the material used to decontaminate the Hart office building after anthrax contamination. DuPont did not perform the decontamination, however, because they did not know the effect of that product on a building, its infrastructure, computers, papers, or even the carpet. Although DuPont makes both the carpet and the decontaminant, we have never investigated the interaction of the two products. This is the kind of new, integrated thinking that is called for. The chemical industry contacts supply chains, such as those for construction materials, automotive materials, and food, at multiple points. We can clean, disinfect, and genetically engineer seed. We can provide advanced coatings and packaging materials, as well as systems to detect bacterial contamination and spoilage. Can we make supply chains safer? The industry wants to help the nation to get security systems where they are needed, at the right times and the right levels, but to do so industry must have more information about specific needs. The chemical industry brings expertise in product development, management, and integration. It wants to be useful. It wants to help mitigate risk. However, it has been very difficult for companies to get a clear idea of what the needs really are. Confusion about what is necessary does not inspire executives to spend employee time, money, and energy to solve a problem that has not been completely identified. According to the Technical Support Working Group and the General Accounting Office, the chemical industry is focused on the short term. Certainly parts are. But perhaps 20 percent of research and development funding is focused on what the world may need next, things that are interesting but unsure. Yet there needs to be a better interface between what Defense Advanced Research Projects Agency (DARPA) produces and what industry does. There is a desire to spend more research and development money on terror mitigation, but there needs to be a better way to integrate the programs and systems that DARPA, the National

OCR for page 55
Page 62 Science Foundation, and industry are funding, especially as the developer of each small component attempts to protect its own intellectual property. The terrorist attacks have had a significant impact on the chemical industry. After looking to their own internal safety, companies want to contribute to the homeland defense effort. There is a lot to be gained from industry's scientific knowledge, culture of safety, role in society, science technology, and manufacturing. The chemical industry is willing to help, but obvious solutions are beyond any individual group or company. The process of formulating and coordinating contracts, objectives, and partnerships is underway, but it has only just begun. Getting all the groups and pieces lined up and together is going to take national will, a bigger sense of urgency, or some greater force. Perhaps organizations like the National Academies can help in this process.

OCR for page 55
Page 63 THOUGHTS AND QUESTIONS ON COUNTERING THE TERRORIST THREAT Richard L. Garwin Council on Foreign Relations, New York As an resident of IBM's Watson Research Center, and especially as a senior fellow of the Council on Foreign Relations, New York, I'd like to share with you some of the national security and homeland defense related thoughts that I've had since September 11, 2001. Unfortunately, terrorism and counterterrorism are enormous subjects that are not only vitally important, but are also urgent. A terrorist attack can come in many forms; the most worrisome of these is airborne bioterrorism. Anthrax distributed upwind of a city is a serious threat that could kill 100,000 people or more. More frightening, however, is an agent like smallpox that is not only infective, but contagious. Past experience with smallpox has proven its efficiency at causing tens of millions of deaths around the world. Another tremendous threat is nuclear explosives, either stolen or improvised. A modern nuclear weapon could kill many millions of people. Even an improvised weapon in New York, on the level of the first-generation weapon used in World War II, could kill a million people. Additionally, there would be complete destruction of a 10-square-kilometer region and hundreds of thousands of people would be exposed to lethal fallout within the first hour. The devastation, however, is limited in extent, unlike a bioterrorism event, placing nuclear weapons second on the list of worrisome scenarios. Bioattacks on food production need no explanation after seeing the results of the outbreak of foot and mouth disease in England. Since foot and mouth disease does not affect humans, the effects were primarily economic. Attacks on agricultural products would likely be just as damaging. Based on the sarin attacks in the Japanese subway, chemical agent attack on humans has shown to be rather difficult. Biological toxins, chemical agents that are often overlooked, may also be difficult to use, but if used correctly can wreak much more havoc. Inhaling a single microgram of botulinum toxin is deadly, compared to a lethal dose by ingestion of 1 milligram for sarin and other agents. Widespread devastation may occur as a result of explosive attack on chemical plants or chemicals in transit, as seen in the accidents in Italy and Bhopal, India. Also on the list of potential methods of terrorist attack is radiological attack. In the long term, relatively few additional people would die of cancer; psychological damage would occur in the short term. Finally, the threat of a calculated explosive attack on structures still looms large after the three airplanes hit the World Trade Center and the Pentagon. Through the Council on Foreign Relations, other scientists and I met with New York Governor Pataki's public safety officials to discuss specific counter-terrorism

OCR for page 55
Page 64 measures for New York state. We discovered that links between the scientific community and state officials barely exist despite the knowledge that this type of communication is essential. Collective protection using filtered air was one topic of discussion at the New York meetings. In the event of an anthrax attack, spores will drift into a building or house even with tightly closed windows. This problem can be alleviated with high-efficiency particulate air filters (HEPA filters) that have been in use since the Manhattan project. These filters are currently used in a number of places such as hospitals. In some buildings, HEPA filters can be easily interchanged for regular filters and their cost is not prohibitive. A change of inside air for outside air needs to occur once every half hour. If the change occurs every few minutes for the circulating air, it would reduce the amount of biological agent that reaches a person inside a building by a factor of 10. If a HEPA filter is used in place of a normal filter, the agent dose is reduced by another factor of 10. Positive pressure within the building that allows no unfiltered air to leak in from the outside also greatly improves the protection level. Countering explosives was another topic of discussion in New York. We need to be able to detect the presence of hundreds of grams of explosives on aircraft passengers and in bags. The simple act of bag matching is not sufficient, because today's terrorists will take their explosive-loaded bag onto an aircraft for a suicide mission. Explosives in vehicles, which can cause enormous problems if detonated at a choke point in the highway system or near a building, also need to be detected. Luckily, to do damage in this type of scenario, a large amount of explosive material is needed, which is hard to secrete in an ordinary car. These are some of the issues currently facing us and that are being discussed around the nation. Scientists and engineers can provide options and can contribute to a rational decision-making process helping to make our nation safer for our residents.

OCR for page 55
Page 65 VULNERABILITY OF PUBLIC WATER SUPPLIES Rolf A. Deininger The University of Michigan Water supply systems are vulnerable to destruction and contamination. This is nothing new and has its roots in antiquity. A late director of the FBI called this to everyone's attention, 1 and his prescription for defensive measures were gates, guards, and guns. He also recognized the threat by a disgruntled insider and advised a background check on all employees. In 1970, the World Health Organization (WHO) published a booklet on “Health Aspects of Chemical and Biological Weapons.” One appendix of that publication deals with the sabotage of water supplies and discusses agents, scenarios, and expected outcomes of a contamination. There will be a new release of this booklet in 2002. The major elements of a water supply system are the raw water source (lake, river, reservoir, or groundwater aquifer), the water treatment plant, the pumping stations, and the distribution system consisting of pipes and intermediate storage reservoirs. Dams on rivers can be destroyed by explosives, not only leading to a loss of the source, but possibly causing serious damage and loss of life due to flooding. Pumping stations are not easily replaced and may require considerable time for repair since replacement pumps are not usually stored on site and may require a lengthy ordering time. A partial destruction is not too serious since there is usually spare capacity allowing a somewhat reduced service. Destruction of pipelines is easily accomplished, but can be fixed in a short time since such occurrences normally happen in distribution systems due to earthquakes and other natural disasters. Utilities are well prepared for such events. There is a great deal of redundancy in a distribution system, and several key elements may be taken out of service without removing the ability of the system to provide service to the consumers at a reduced rate. The key to a higher level of security is therefore the redundancy of the system. This redundancy should be investigated for each particular system to ensure security compliance. A water supply system may be contaminated at the raw water source, at the water treatment plant, or in the distribution system. Contamination of the raw water source is easily accomplished since it is usually at a location far from the service area. It is not, however, a very effective measure due to the large quantity of water involved. A contamination at the water treatment plant would be more effective, but increases the likelihood of detection since a water treatment plant is staffed around the clock. In addition, the treatment processes will reduce the contaminants by one or two orders of magnitude. Thus the distribution system is 1J. E. Hoover. 1941. Water Supply Facilities and National Defense. Journal of the American Water Works Association 33(11): 1861-1865.

OCR for page 55
Page 88 formulations and technologies to rapidly restore civilian facilities after a terrorist attack. One simple but elusive goal is to create a single formulation to be used by first responders that destroys all chemical AND biological agents. A second objective is to address the complex issues associated with final decontamination and remediation of a facility including “how clean is clean” and convincing the public that a facility is safe. Three types of projects are funded under decontamination and remediation: methodology (deployment and use), new nontoxic formulations and technology, and field verification. In the area of methodology, two projects, one at Oak Ridge National Laboratory and one at Lawrence Livermore National Laboratory, are focusing on how clean is safe. Several nontoxic formulations have been developed, including the Sandia Decon Foam, which is active against chemical and biological agents, and the L-Gel, which travels through air ducts. A plasma jet technology for decontamination of sensitive equipment has also been built. Mock offices at the Dugway Proving Ground were contaminated with anthrax surrogate, and four novel emerging decontamination technologies were tested there as part of the field verification and testing program. Data from this test was used for technology evaluations and selections for the decontamination and remediation of the Capitol Hill office buildings. I have been working on the Sandia Decon Foam formulation (see Figure D.6). Generally, foam is quicker to deploy and react with chemical and biological agents than a water- or fog-based decontaminant, and it has very low logistic support and water demand. The Sandia Decon Foam has very low toxic and corrosive properties and provides a single decontamination solution for both chemical and biological agents. Decontamination of the Capitol Hill office buildings was a learning experience for our nation. The Sandia Decon Foam was one of many decontaminants used in the cleanup efforts. It became clear that no single technology is universally effective and that a suite of decontamination technologies is needed. It is also necessary to develop better methods to decontaminate sensitive equipment and objects such as electronics, computers, copiers, and valuable paintings. Consequently, facilities and standard methods are required for product testing. Decontamination programs exist in other federal agencies as well. For example, the U.S. Army Soldier and Biological Chemical Command in Edgewood, Maryland, is focusing on military needs. They are looking at solution, oxidation, and enzymatic chemistry for liquid-type decontamination formulations, especially decontamination of sensitive equipment. Their developing technologies include supercritical carbon dioxide, solvent wash systems, thermal approaches, and the plasma jet, which is a cooperative project with the Department of Energy. They are also conducting large-scale tests with solid-phase chemistry focused on decontamination of bulk quantities of chemical and biological agents, if an entire military base, for example, needed to be decontaminated. Most of these types of projects are funded by the Department of Defense.

OCR for page 55
Page 89 ~ enlarge ~ FIGURE D.6 Sandia Decon Foam is deployed. In conclusion, I think future work should focus on several areas of decontamination. First, we need noncorrosive decontamination formulations and technologies for use on sensitive equipment and items as well as methods to decontaminate hard-to-reach places like air ducts. Fundamental issues like the number of anthrax spores necessary for infection and how clean is clean should be investigated. We also need more extensive lab and field data, and large-scale testing of decontamination formulations and methodologies, especially coordination and integration with sensors and modeling and simulation. Ideally, if we could tie all these things together, effort could be directed at the problem areas without wasting time on areas that remain clean. Personnel decontamination is also a main issue. With each anthrax incident, real or perceived, a personnel

OCR for page 55
Page 90 decontamination effort goes into effect, putting people through portable showers and spraying them with bleach, for example. This may not be a realistic measure as shown by the Tokyo subway incident: almost everyone who went to a hospital hopped in a cab and went on their own, thus contaminating the hospitals. Sociological issues are important, as well. First responders destroy evidence when they deploy foam; coordination with forensic and criminal investigators must occur. What evidence will be admissible in court? Finally, regulatory issues will also come into play in establishing standard test methods to determine the efficacy of and give approval to sensors and decontamination formulations for chemical and biological agents.

OCR for page 55
Page 91 HOW INTEGRATION WILL MAKE MICROFLUIDICS USEFUL Stephen R. Quake California Institute of Technology You've already heard about the basics of microfluidics and the lab-on-a-chip concept from Andrea Chow. So today, I'm going to talk about the gadgets based on microfluidics that my research group is making. I hope that this talk sparks your imagination and helps you to envision ways that microfluidics can help in the area of national security and homeland defense. Over the last 10 years or so, scientists and engineers have been focusing on miniaturizing specific functions of lab equipment rather than miniaturizing the entire lab. This is because they had no way to integrate all of the pieces; the plumbing was lacking. More specifically, plumbing is controlled by valves, which are extremely difficult to miniaturize. To solve this problem, several years ago my research group developed a soft microvalve system using multilayer soft lithography. If you have two layers of orthogonally arranged pipes and fill the bottom layer with fluid, a pneumatic pressure applied to the top pipe will cause the bottom pipe to pinch closed (see Figure D.7). This creates a valve. The valves are made of a soft, inexpensive polymer that allows the use of low actuation forces on the valves and yields small footprints. Pumping in this system is based on peristalsis, enabling the movement of multiple nanoliters per second through valves 100 µm wide and 10 µm deep. A nanopipette is built into the system, with different size plungers allowing different amounts of fluid to be introduced. A number of devices have been made based on this microvalve system. One of the first devices built was a cell sorter on a chip. By manipulating nanoliters of fluid, different strains of fluorescently activated E. coli were introduced, sorted according to their fluorescent properties, recovered from the chip, and cultured. My research group felt that high-throughput protein crystallization on a chip would be a useful challenge to meet. We first determined that it was possible to grow protein crystals on a chip, but only under the same conditions that allowed crystal growth in bulk experiments. The next experiment screened for protein crystallization, a difficult process. Success seems to be more due to sheer numbers and probability than on rational design, and due to the small amount of protein available for use, mixing and metering during the screening process are difficult to achieve. The microfluidic, microvalve system we have developed is able to overcome these obstacles. The mixing ability of microfluidic systems was tested using a rotary pump, a circular channel with inputs and outputs that can be peristaltically pumped, opened, and closed. It was found that after only a few minutes of active mixing (due to pumping), a uniform mixture of particles is obtained that would have taken hours to achieve by diffusion. This is also useful for accelerating diffusion-

OCR for page 55
Page 92 ~ enlarge ~ FIGURE D.7 Integrated microfluidic devices, which contain valves that can be pinched closed by pressurized channels, are fabricated from rubber by soft microlithography at low cost. limited reactions; a 60-fold enhancement in the kinetics of some assays has been measured. The rotary pump has also been used as a 12 nL polymerase chain reaction system. In the electronics industry, a Pentium computer chip has hundreds of millions of transistors and only a hundred pins in and out. If each transistor had to be addressed individually, it would be impossible to have such a chip. Microfluidic systems are similarly easy to control: n fluid lines can be controlled by 2log n control lines. Additionally, the pressure to actuate a valve depends on the width of the control line, so by choosing our pressure carefully, one thinner fluid line can be closed while the wider fluid lines remain open due to insufficient pressure. This idea has been used to develop microfluidic systems that can screen enzymatic libraries and perform in vitro transcription translation of DNA to protein in approximately 30 minutes. I am sure that this technology is useful in many ways to chemists and chemical engineers and can help in the arena of national security and homeland defense.

OCR for page 55
Page 93 BIOSYNTHETIC ENGINEERING OF POLYKETIDE NATURAL PRODUCTS C. Richard Hutchinson Kosan Biosciences This morning I'd like to address basic medicinal chemistry, and specifically, how to manipulate drugs from natural sources. Biosynthetic engineering has recently been exploited in antibacterial drug discovery, but can also be applied to anticancer drugs, antiviral drugs, and others. Kosan Biosciences was formed almost 6 years ago, founded on an interest in polyketides, microbial metabolite-based drugs. Polyketides have many diverse chemical structures including erythromycin, which will be mentioned again later. These chemicals include fused-ring aromatic compounds, compounds decorated with sugars, and compounds with large stretches of double bonds. Each of these compounds has different biological activities and utilities, but they are all made in nature by very similar biochemistry. This biochemistry resembles how long-chain fatty acids are made. Acyl-coenzyme A substrates can be carboxylated, reduced, dehydrated, or otherwise changed from the original molecule. Each of these biochemical activities is accomplished one at a time, by single-function enzymes that are produced by an organism and collectively organized to make a particular kind of long-chain fatty acid. Polyketides use enzymes like this in much more complicated ways than required to make fatty acids. While biological organisms may use only a handful of activities, with modern genetic methods biochemists can combine genes for any or all of the activities to create a complex array of thousands of natural products. Imagine a set of genes in the chromosome of a bacterium. The protein products of these genes have individual active sites (domains) and are grouped in modules. Each module can have a different number of active sites and hence a different function: joining bonds, selecting and loading substrates, and carrying the substrates, for example. Not every module is used for every synthesis; this characteristic imparts the synthetic flexibility and diversity that creates the eventual polyketides. Modules are strung together in very large proteins to take simple substrates the organism provides, assemble them in a sequential manner, and produce a polyketide. The enzyme then continues to produce polyketides and accumulate the product (see Figure D.8). For Kosan and others in the pharmaceutical industry, the intent is to learn enough about these enzymes from a structure-function viewpoint so they can be manipulated. Because polyketides are built by starting from one point and continuing sequentially along the pathway dictated by protein structure, a biochemist can trace through a molecule and make structure-function predictions for the assembly enzymes with reasonable accuracy. Using recombinant DNA methods,

OCR for page 55
Page 94 ~ enlarge ~ FIGURE D.8 To synthesize polyketides, an enzyme joins protein modules and is recycled to continue molecule production. it is possible to remove a single module or a specific domain within a module, place it somewhere else in the protein, and make a product with a different structure and functionality. This technique has been used in the attempt to create new antibacterial agents that target drug-resistant pathogens. For example, erythromycin is a very well known antibacterial drug that has been used for approximately 50 years for a variety of infections, primarily in the lung. A polyketide synthase produces erythromycin and has six modules, each with a certain number of active sites. By manipulating the active sites or modules of the polyketide synthase, the alkyl groups could be taken out or changed, the oxidation state of the hydroxyls or the ketone could be changed, or the length and size of the lactone ring could be changed. Although many variations of the synthesis were attempted, none developed a better antibacterial drug than the existing erythromycin. The experiment was valuable, though, because it established a paradigm for structure-function relationships. Similarly, we could create a mutant in part of the polyfunctional protein and allow the synthetic substrate to be accepted by the functional parts and carried through to produce a completely different compound. The benefit of this type of manipulation is that synthetic procedures that are very difficult for chemists to do

OCR for page 55
Page 95 can be accomplished easily. These processes have allowed scientists to test a large number of macrolide antibiotics and to develop lead compounds for a new type of drug called ketolides, which has activity against usually drug-resistant pathogens. To summarize, polyketides are simply bacterial products that include erythromycins and chemically derived ketolides. New erythromycins and polyketides can be made by genetic engineering of the polyfunctional giant proteins called polyketide synthases or by taking products of the engineered microbial metabolism and further modifying them. These new antibiotics can be more insightfully designed due to increased knowledge of structure-function relationships for how such antibiotics bind to bacterial ribosomes. This allows treatment of drug resistant organisms that are of importance in community-acquired infections, pathogenic diseases, and especially unique bacterial pathogens that have previously been ignored because they had never been a threat. Although we cannot predict drug efficacy, drug production is more approachable using the technology of microbial metabolism modification and structure-function information for bacterial ribosomes.

OCR for page 55
Page 96 CHALLENGES IN RAPID SCALE-UP OF SYNTHETIC PHARMACEUTICAL PROCESSES Mauricio Futran Bristol-Myers Squibb Even though synthetic organic chemistry is a very old scientific pursuit, I believe that today most medications needed for homeland defense will be created by means of synthetic organic chemistry. This type of research and development is done by industrial organic chemists and chemical engineers, and there are many opportunities to involve academic scientists to increase the speed of development. The pharmaceutical industry can improve by better partnering with academia to focus on solving the most pressing problems. In general, there is a lack of understanding of automation and parallel experimentation among chemistry graduates. Engineering students also need training in modeling and computation, specifically in the areas of predicting physical and transport properties, computational fluid dynamics, specific process operations, and control theory. This facilitates obtaining scale-up parameters from small-scale experiments better known as “micro-piloting,” which is of utmost importance. Further development in the areas of in-line analytical and microchip technology would be useful for rapid drug development. In a national emergency, the pharmaceutical industry could be called upon to bring a medication to market, to a commercial scale, very quickly. Normally, the complex scale-up process does not operate on such a short timeline. Assuming that we can partner with government agencies to ensure compliance in a streamlined manner, assuming we have better predictive toxicology, and assuming that the development of the dosage form (capsule, tablet) is not an issue, there remain a number of barriers to the scale-up process to create the active pharmaceutical ingredients. Often in a medicinal laboratory, synthetic routes are linear instead of convergent and may contain many steps with low yields and operating conditions not favorable to scale-up (for example, strong exotherms, mass transfer limitations, and gums/heavy slurries). Reactions involving chiral molecules may have low selectivity for the medicinally active isomer. In addition, the reagents used may be toxic, and using large amounts of toxic agents for a scale-up is not an acceptable solution. Some laboratory materials are not readily available in large quantities. Scale-up to 100-g quantities for animal testing, which must precede human studies, typically takes 3 to 8 months. Then, production must be scaled up to commercial quantities. Traditionally, the entire process takes 2 to 3 years—too long to be useful in a national emergency. Much of the knowledge needed for manufacturing a pharmaceutical is related to the last step in the process, including understanding the bulk active ingredient, its impurity profile, chiral purity, and crystal form. To cut down the entire scale-up

OCR for page 55
Page 97 process to only 6 months would require some of this knowledge to be investigated from the beginning, in parallel with other information. A 6-month timespan may still seem slow, but that is one-fourth the time normally taken by industry. The scale-up process can be broken into smaller steps. First, the synthetic pathway must be defined. This includes determining how to make the final compound in a safe, environmentally sensitive, and affordable manner. This may require thousands of experiments and an equally large amount of analytical tests. The second step involves developing the pathway into a process. Issues in this stage include ensuring the right chiral and chemical purities, which requires hundreds of experiments. The last stage is performing hundreds more experiments to learn about process parameters and optimization (see Figure D.9). To minimize the time required for these three steps, chemists need to appreciate the process-related issues not only of how this elegant structure can be made, but also how it can be made efficiently. They should also know chemical engineering concepts like kinetics and continuous processing so that these ideas can be incorporated into the synthesis process. Chemists must be comfortable with automation—with the tools they need for the job. Commonly, automated systems are technologies that sit underutilized in chemical laboratories. Chemical engineers are lacking in-depth knowledge of organic chemistry and particular spectroscopic methods, which are essential in modern industry. It is impossible to function in the pharmaceutical industry without understanding what the chemists are saying. It would also be advantageous for engineers to have hands- ~ enlarge ~ FIGURE D.9 Each stage of drug development involves many issues and experiments.

OCR for page 55
Page 98 on industrial experience with taking a product from the concept stage to the commercialization stage. The educational system can address many of these problems. As mentioned earlier, parallel and automated methods are a key part of making the scale-up process faster. Automation technology already exists to provide good agitation, heating, reflux, inert headspace, sample filtration, and sample taking in a variety of vessel sizes. However, there remain challenges in the automated handling of solids and in the use of on-line spectroscopic instrumentation on very small samples, which would aid the quick determination of process optimization, parameters, kinetics, partition coefficients, and the like. It is essential that the automated technology be user friendly, so that chemists can focus on the chemistry and not on how to use the equipment. Continuous systems will also become an important topic for rapid scale-up during a national emergency. First, continuous reactions allow the scale-up to be considerably smaller than that needed for a batch reaction and allow equal or greater productivity. Continuous systems can much more easily handle highly reactive, unstable ingredients. They can also better handle hazardous reagents; since the reagents are made in situ and are used immediately, they never accumulate. Use of continuous systems will require chemists and chemical engineers to work together more closely. It will also need more and improved prior knowledge of the physical properties and transport properties of reagents. Consequently, our predictive and modeling capabilities must improve. Currently, after each step in the synthesis process, a sample is analyzed to approve or reject the operation. Not only is this inefficient for the industry in general, this is also a stumbling block for a rapid scale-up. Controls and measurements of the process need to be accomplished in situ, which is again an issue that requires the chemists and chemical engineers to work together toward a solution. It may, in fact, involve lab-on-a-chip technologies and microreactors that were presented by other speakers in this workshop. To integrate everything into a continuous process, accelerated reaction kinetics are needed. A reaction that takes 20 hours to complete is difficult to use in a continuous process. However, we have microwave chemistry, ultraviolet chemistry, laser stimulation, sonication, and other technologies to accelerate proven but slow reaction chemistries. Research in this area is merely beginning. Another area that is lacking is the chemistry of solids. In pharmaceuticals, an active drug is often made into a tablet or some other solid. Unknowns remain in the areas of solids flow, solids compressibility, and compaction. The physical incompatibility of various inert ingredients also remains a mystery. The engineering community has a great opportunity to help in this area. In conclusion, research is needed for the improvement of automated tools, handling of solids, modeling and computational tools, and in-line analytical technology. To speed the scale-up of pharmaceuticals in a national emergency, these issues must be addressed by both chemists and chemical engineers.