Biosensors & Bioelectronics 7 (1992) 3 17-32 1
EN\IIRONUIENlAlMO#CTOFCIIJO
BIOSENSORS FOR ENVIRONMENTAL MONITORING K. R. Rogers and J. N. Lin In this article we will outline several biosensor applications which may fill existing technology gaps in the area of environmental monitoring. The requirements for these environmental biosensors, as well as diffkulties in commercialization, are also addressed.
ENVIRONMENTAL MONITORING REQUIREMENTS Monitoring the environment for the presence of compounds which may adversely affect human health and local ecosystems is a fundamental part of the regulation, enforcement and remediation processes which will be required to maintain a habitable environment. Because of the diversity of current environmental monitoring methods, the choice of techniques used for sample collection, extraction and analysis depends on the compound or compounds of interest and the matrix they contaminate. Although classical analytical techniques are being continually refined and improved, these laboratory-based methods are, for the most part, relatively expensive and time-consuming. Because biosensor technology lends itself to fast, economical and portable analysis, these devices may provide solutions for some of the problems currently encountered in the measurement of environmental contaminants.
K.R. Rogers is a research chemist at the US Environmental Protection Agency, Environmental Monitoring Systems Laboratory, PO Box 93478, Las Vegas, NV 89193-3478, USA. J.N. Lin is a research investigator at Bcehringer Mannheim Corp, 9 115 Hague Road, PO Box 50457 Indianapolis, IN 46250, USA.
Environmental monitoring in the USA is primarily driven by legislation enacted by the Congress. Examples include the Resource Conservation and Recovery Act (RCRA), the Comprehensive Environmental Response, Compensation and Liability Act of 1980 (CERCLA) a.k.a. Superfund and its 1986 amendment, the Superfund Amendments and Reauthorization Act (SARA), as well as mandates that require exposure assessments for agricultural and industrial chemicals released into the Superfund requires the environment. identification, evaluation and clean-up of sites placed on the National Priorities List (NPL). More than 1200 NPL sites need characterization and remediation and the scope of this monitoring challenge is enormous. Over 600 chemical compounds have been identified at hazardous waste sites alone and there may be thousands of unidentified pollutants. The majority of compounds found on the Agency for Toxic Substances and Disease Registry’s (ATSDR’s) priority list are volatile organic compounds; however, other prominent chemical classes include inorganics, phenols, phenoxy acids, polyaromatic hydrocarbons, nitrosamines and aromatic amines. In many cases, identification and quantitation of specific compounds in mixtures of these chemicals must be made in media such as air,
O!Z6-5663/92/$05.000 1992Elsevier Science Publishers Ltd.
317
EFMROMUENTALMONllORiNo water, soil and sludge as well as biological matrices such as blood, saliva and urine. Although most biosensors must operate in aqueous environments, the development of gas-permeable membranes and appropriate soil extraction procedures can extend the range of media accessible to analysis by these devices.
ENVIRONMENTAL
BIOSENSORS
Biosensors typically consist of a biological sensing element (enzyme, receptor, antibody or microbe) in intimate contact with a chemical or physical transducer (electrochemical, optical, mass or thermal). Because of their selectivity and high binding affinities toward certain environmental pollutants, a variety of biological macromolecules may prove to be useful candidates as sensing elements for environmental biosensors. Examples of enzymes which have been used in biosensors include acetylcholinesterase which binds with high affinity to organophosphate insecticides and Japanese pine-comb fish luciferase which has been used to measure Hg*+. Receptors that may prove useful as sensing elements include the y-aminobutyric acid receptor (GABA) which has been shown to bind cyclodienes, pyrethroids, bicyclophosphates and orthocarboxylates; the muscarinic receptor which binds organophosphate insecticides; and the aryl hydrocarbon receptor which binds with high affinity to dioxins. Other biological macromolecules which have been shown to be versatile and powerful as biological sensing elements are antibodies. Analytes which have been measured using antibodies include dozens of insecticides and herbicides as well as environmental pollutants such as polychlorinated biphenyls (PCBs), dioxins, pentachlorophenol, benzo(a)pyrene and benzene, toluene and xylene (BTX). Microbes have also been used as biological sensing elements. Cyanobacteria have been used to measure herbicides, Trichosporon cells have been used to measure biochemical oxygen demand and a genetically altered
318
Biosensors & Bioelectronics
Pseudomonas has been used to measure naphthalene. The sensitivity and specificity of a biosensor is primarily conferred by its biological sensing element. Although enzymes, receptors and some antibodies show high binding affinities to their respective analytes, they typically bind to a group of closely related molecules. This feature lends itself to applications such as screening for the presence of closely related environmental pollutants (for example, organophosphate insecticides, triazine herbicides and BTX). On the other hand, as some antibodies show little cross-reactivity even among closely related analytes, biosensors using these molecules as sensing elements may be highly specific. For environmental monitoring, there are several general areas in which biosensors may have distinct advantages over current analytical methods. Miniaturization of biosensors has, in the clinical field, led to the development of economical (i.e. disposable or reusable) sensor modules which can be detached from the display module. For example, a blood glucose analysis can typically be made for under $1. This is an attractive feature for environmental monitoring, considering that the average cost of a laboratory analysis for an environmental sample can range from $130 to $200. The development of portable biosensors could also eliminate the time, expense and chain-of-custody considerations required to transport samples to a laboratory. Other possible advantages of biosensor-based assays include the measurement of analytes in complex matrices with a minimum of sample preparation, operation of the biosensor by untrained personnel, completion of the analysis in less than 1 hour and continuous or remote monitoring capabilities.
SPECIFIC AREAS OF APPLICATION There is a critical and growing need for more cost-effective techniques for identification and quantitation of pollutants in complex environmental matrices. For biosensors to become competitive, their characteristics must be
Biosensors & Bioelectronics
exploited to fill existing technology gaps left by classical laboratory analysis and emerging field screening methods such as portable gas chromatography, mass spectrometry, chemical sensors, ion mobility spectrometry and immunoassay test kits. This is not a trivial problem considering that out of hundreds of existing designs, relatively few biosensors have been developed into commercial products. As and chromatographic, immunoassay spectroscopic-based field screening assays improve, biosensors must meet or exceed the quality of data produced by these methods for the specific analyte of interest. Issues which must be addressed in the development of biosensors include detection limits, dynamic range, positive and negative controls, interferences and matrix effects. Some general requirements for biosensors which might be used in field assay applications are listed in Table 1. As problematic toxic waste repositories are identified and remediation procedures are
implemented, “real-time” monitoring assays become crucial for directing appropriate clean-up Existing emerging strategies. and chromatographic and spectroscopic field screening technologies appear well suited to the detection of certain pollutants such as volatile organic compounds and heavy metals. Furthermore, immunoassays have been demonstrated for a wide range of insecticides (e.g. chlorinated hydrocarbons and organophosphates), herbicides (e.g. thiocarbamates, phenoxyaliphatic acids, bipyridiliums and triazines), fungicides (e.g. benzimidazoles, acylalanines and triazoles) and rodenticides (e.g. coumarins). Biosensors can exploit the sensitivity and specificity afforded by antibodies used in these assays and take advantage of technological design improvements in areas of speed, simplicity and ‘user friendliness’. Another area in which biosensors may find a niche involves continuous and/or remote monitoring of industrial or agricultural
TABLE 1
General requirements for biosensors in environmental field applications.
Specification range
Requirement cost
$1-15 per analysis
Portable
Can be carried by one person; no external power requirements
Fieldable
Easily transported in a van or truck; limited external power required
Assay time
l-60 minutes
Personnel training
Can be operated by minimally trained personnel after l-2 hours training period
Matrix :
Minimal preparation for ground water, soil extract, blood, urine or saliva
Sensitivity
Parts per million/parts per billion
Dynamic range
At least two orders of magnitude
Specificity
Enzymes/receptors: specific to one or more groups of related compounds Antibodies: specific to one compound or one group of closely related compounds
Specific requirements may vary depending on the application.
319
Biosensors & Bioelectronics
groundwater or runoff. Chemical sensor technologies have made recent progress in this area. These devices use mass, optical, electrochemical or thermal sensors coupled to chemically selective coatings. Because of the similarities in signal transducers, chemical sensors and biosensors share many of the same characteristics (e.g. size, speed and cost of analysis). Continuous and/or remote monitoring applications, however, have specific requirements in that unattended operation over extended periods is needed. Chemical sensors have the advantages of ruggedness and durability for this application; however, biosensors have the advantages of sensitivity and selectivity. Challenges for biosensors in this area include the stabilization of the biologically active membranes and coatings and the development of methods for continuous delivery of the biological macromolecules into the system. Human exposure assessment is another area in which biosensors could make a significant contribution. Regulations governing the release of chemicals into the environment are typically influenced by the perceived risk to public health and local ecosystems. Determination of human exposure requires measurements of the compound of interest at locations, within the time frames and in the media (e.g, air, soil or water) most likely to result in exposure. This determination is complicated by the fact that an individual typically engages in a variety of activities and moves through a number of microenvironments in which exposure may occur. A useful tool in the determination of exposure to airborne contaminants is a sampling device worn by an individual (e.g. a dosimeter). These dosimeters are typically configured as a sample concentration medium from which the compounds of interest are extracted and analysed by a classical method. Current efforts in this area include work on configurations in which the analyte is trapped by an immunospecific medium and analysed by immunoassay techniques. A cumulative exposure over a given time period can then be calculated. A gas permeable membrane
320
interfaced to an immunosensor might also allow instantaneous measurements, as well as time integrated average exposures. This information would prove invaluable for integrating exposure patterns into risk assessment models. Biosensors may also contribute to the area of risk assessment that attempts to correlate exposure to effective dose, adverse biological effect and ultimately to clinical disease. Because of the diversity of possible routes of entry, however (e.g. skin, lung and gut), exposure is not easily correlated even to effective dose. Extrapolations from animal systems must be used and these approximations are often confounded by differences in pharmacokinetics between species. The measurement of a pollutant or its metabolites in easily sampled biological fluids such as blood, urine or saliva has provided a useful tool for determining dose in humans. Measurement of statistically significant populations using classical analytical methods, however, is expensive and time-consuming. Biosensors may again provide a solution for fast and economical field assays for estimation of human exposure to hazardous pollutants.
PROSPECTS FOR COMMERCIALIZATION
Because of the wide range of their possible applications, biosensors have attracted tremendous interest and attention in science since the late 1970s. After more than a decade of intense research, however, most effort still remains at the level of research curiosities in universities and research institutes. At present, there are only a few biosensor technologies, among hundreds of existing designs, that have been developed into commercial products. None of these biosensors have been commercialized for environmental applications. Although each sensor has its own performance requirements and design specifications, there are several common obstacles to commercialization of biosensors in general.
Biosensors & Bioelectronics
First, the performance characteristics such as sensitivity, dynamic range, reproducibility, and stability, for sensors demonstrated in the research stage have not quite met the needs of users. Furthermore, the performance of the biosensor is further reduced in the development and manufacturing stages as a result of mass production. Therefore, to make a practical device, it is important to keep chemical and physical design characteristics simple, using proven technologies wherever possible. The second obstacle is the lack of effort toward system integration and user-interfacing. The sensor concepts need to be taken one step further by building a prototype model that can be easily operated by untrained users. The third obstacle involves the competitiveness of the biosensors. Commercialization becomes difficult or impossible if no definite advantages over other competing technologies are achieved. Finally, probably the most important factor responsible for the slow commercialization of
biosensors is the market size. Because of the complexity of these devices, millions of dollars must be invested to develop a biosensor. Therefore, until profitable marketplaces are realized, private companies will not be motivated to investigate and transfer the biosensor technologies to products. Issues related to competitiveness and market size for most biosensor technologies are still being defined by current research and market developments. We believe that once the marketplaces and attributes of biosensors are fully realized, they will have a more important role in various sectors of the analytical industry, such as environmental monitoring. Although the research described herein has been funded wholly or in part by the US Environmental Protection Agency, it has not been subject to Agency review and therefore does not necessarily reflect the views of the Agency, and no oficial endorsement should be inferred. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.
FURTHER READING Eldefrawi, M.E., Eldefrawi, A.T., Rogers, K.R., Valdes, J.J. (1992) Pharmacological biosensors, In Immunochemical Assays and Biosensor Technology for the 199Os, 1992, edited by Nakamura, R.M.,
Kashahara, Y. and Rechnitz, G.A. American Society for Microbiology, Washington, DC. National Research Council (1991). Environmental epidemiology, Vol. 1, public health and hazardous wastes, National Academy Press, Washington, DC, USA. Van Emon, J.M. and Lopez-Avila, V. (1992). Immunochemical methods for environmental analysis. Anal. Chem., 64,79A-88A.
321
EN\IIRONUIENlAlMO#CTOFCIIJO
BIOSENSORS FOR ENVIRONMENTAL MONITORING K. R. Rogers and J. N. Lin In this article we will outline several biosensor applications which may fill existing technology gaps in the area of environmental monitoring. The requirements for these environmental biosensors, as well as diffkulties in commercialization, are also addressed.
ENVIRONMENTAL MONITORING REQUIREMENTS Monitoring the environment for the presence of compounds which may adversely affect human health and local ecosystems is a fundamental part of the regulation, enforcement and remediation processes which will be required to maintain a habitable environment. Because of the diversity of current environmental monitoring methods, the choice of techniques used for sample collection, extraction and analysis depends on the compound or compounds of interest and the matrix they contaminate. Although classical analytical techniques are being continually refined and improved, these laboratory-based methods are, for the most part, relatively expensive and time-consuming. Because biosensor technology lends itself to fast, economical and portable analysis, these devices may provide solutions for some of the problems currently encountered in the measurement of environmental contaminants.
K.R. Rogers is a research chemist at the US Environmental Protection Agency, Environmental Monitoring Systems Laboratory, PO Box 93478, Las Vegas, NV 89193-3478, USA. J.N. Lin is a research investigator at Bcehringer Mannheim Corp, 9 115 Hague Road, PO Box 50457 Indianapolis, IN 46250, USA.
Environmental monitoring in the USA is primarily driven by legislation enacted by the Congress. Examples include the Resource Conservation and Recovery Act (RCRA), the Comprehensive Environmental Response, Compensation and Liability Act of 1980 (CERCLA) a.k.a. Superfund and its 1986 amendment, the Superfund Amendments and Reauthorization Act (SARA), as well as mandates that require exposure assessments for agricultural and industrial chemicals released into the Superfund requires the environment. identification, evaluation and clean-up of sites placed on the National Priorities List (NPL). More than 1200 NPL sites need characterization and remediation and the scope of this monitoring challenge is enormous. Over 600 chemical compounds have been identified at hazardous waste sites alone and there may be thousands of unidentified pollutants. The majority of compounds found on the Agency for Toxic Substances and Disease Registry’s (ATSDR’s) priority list are volatile organic compounds; however, other prominent chemical classes include inorganics, phenols, phenoxy acids, polyaromatic hydrocarbons, nitrosamines and aromatic amines. In many cases, identification and quantitation of specific compounds in mixtures of these chemicals must be made in media such as air,
O!Z6-5663/92/$05.000 1992Elsevier Science Publishers Ltd.
317
EFMROMUENTALMONllORiNo water, soil and sludge as well as biological matrices such as blood, saliva and urine. Although most biosensors must operate in aqueous environments, the development of gas-permeable membranes and appropriate soil extraction procedures can extend the range of media accessible to analysis by these devices.
ENVIRONMENTAL
BIOSENSORS
Biosensors typically consist of a biological sensing element (enzyme, receptor, antibody or microbe) in intimate contact with a chemical or physical transducer (electrochemical, optical, mass or thermal). Because of their selectivity and high binding affinities toward certain environmental pollutants, a variety of biological macromolecules may prove to be useful candidates as sensing elements for environmental biosensors. Examples of enzymes which have been used in biosensors include acetylcholinesterase which binds with high affinity to organophosphate insecticides and Japanese pine-comb fish luciferase which has been used to measure Hg*+. Receptors that may prove useful as sensing elements include the y-aminobutyric acid receptor (GABA) which has been shown to bind cyclodienes, pyrethroids, bicyclophosphates and orthocarboxylates; the muscarinic receptor which binds organophosphate insecticides; and the aryl hydrocarbon receptor which binds with high affinity to dioxins. Other biological macromolecules which have been shown to be versatile and powerful as biological sensing elements are antibodies. Analytes which have been measured using antibodies include dozens of insecticides and herbicides as well as environmental pollutants such as polychlorinated biphenyls (PCBs), dioxins, pentachlorophenol, benzo(a)pyrene and benzene, toluene and xylene (BTX). Microbes have also been used as biological sensing elements. Cyanobacteria have been used to measure herbicides, Trichosporon cells have been used to measure biochemical oxygen demand and a genetically altered
318
Biosensors & Bioelectronics
Pseudomonas has been used to measure naphthalene. The sensitivity and specificity of a biosensor is primarily conferred by its biological sensing element. Although enzymes, receptors and some antibodies show high binding affinities to their respective analytes, they typically bind to a group of closely related molecules. This feature lends itself to applications such as screening for the presence of closely related environmental pollutants (for example, organophosphate insecticides, triazine herbicides and BTX). On the other hand, as some antibodies show little cross-reactivity even among closely related analytes, biosensors using these molecules as sensing elements may be highly specific. For environmental monitoring, there are several general areas in which biosensors may have distinct advantages over current analytical methods. Miniaturization of biosensors has, in the clinical field, led to the development of economical (i.e. disposable or reusable) sensor modules which can be detached from the display module. For example, a blood glucose analysis can typically be made for under $1. This is an attractive feature for environmental monitoring, considering that the average cost of a laboratory analysis for an environmental sample can range from $130 to $200. The development of portable biosensors could also eliminate the time, expense and chain-of-custody considerations required to transport samples to a laboratory. Other possible advantages of biosensor-based assays include the measurement of analytes in complex matrices with a minimum of sample preparation, operation of the biosensor by untrained personnel, completion of the analysis in less than 1 hour and continuous or remote monitoring capabilities.
SPECIFIC AREAS OF APPLICATION There is a critical and growing need for more cost-effective techniques for identification and quantitation of pollutants in complex environmental matrices. For biosensors to become competitive, their characteristics must be
Biosensors & Bioelectronics
exploited to fill existing technology gaps left by classical laboratory analysis and emerging field screening methods such as portable gas chromatography, mass spectrometry, chemical sensors, ion mobility spectrometry and immunoassay test kits. This is not a trivial problem considering that out of hundreds of existing designs, relatively few biosensors have been developed into commercial products. As and chromatographic, immunoassay spectroscopic-based field screening assays improve, biosensors must meet or exceed the quality of data produced by these methods for the specific analyte of interest. Issues which must be addressed in the development of biosensors include detection limits, dynamic range, positive and negative controls, interferences and matrix effects. Some general requirements for biosensors which might be used in field assay applications are listed in Table 1. As problematic toxic waste repositories are identified and remediation procedures are
implemented, “real-time” monitoring assays become crucial for directing appropriate clean-up Existing emerging strategies. and chromatographic and spectroscopic field screening technologies appear well suited to the detection of certain pollutants such as volatile organic compounds and heavy metals. Furthermore, immunoassays have been demonstrated for a wide range of insecticides (e.g. chlorinated hydrocarbons and organophosphates), herbicides (e.g. thiocarbamates, phenoxyaliphatic acids, bipyridiliums and triazines), fungicides (e.g. benzimidazoles, acylalanines and triazoles) and rodenticides (e.g. coumarins). Biosensors can exploit the sensitivity and specificity afforded by antibodies used in these assays and take advantage of technological design improvements in areas of speed, simplicity and ‘user friendliness’. Another area in which biosensors may find a niche involves continuous and/or remote monitoring of industrial or agricultural
TABLE 1
General requirements for biosensors in environmental field applications.
Specification range
Requirement cost
$1-15 per analysis
Portable
Can be carried by one person; no external power requirements
Fieldable
Easily transported in a van or truck; limited external power required
Assay time
l-60 minutes
Personnel training
Can be operated by minimally trained personnel after l-2 hours training period
Matrix :
Minimal preparation for ground water, soil extract, blood, urine or saliva
Sensitivity
Parts per million/parts per billion
Dynamic range
At least two orders of magnitude
Specificity
Enzymes/receptors: specific to one or more groups of related compounds Antibodies: specific to one compound or one group of closely related compounds
Specific requirements may vary depending on the application.
319
Biosensors & Bioelectronics
groundwater or runoff. Chemical sensor technologies have made recent progress in this area. These devices use mass, optical, electrochemical or thermal sensors coupled to chemically selective coatings. Because of the similarities in signal transducers, chemical sensors and biosensors share many of the same characteristics (e.g. size, speed and cost of analysis). Continuous and/or remote monitoring applications, however, have specific requirements in that unattended operation over extended periods is needed. Chemical sensors have the advantages of ruggedness and durability for this application; however, biosensors have the advantages of sensitivity and selectivity. Challenges for biosensors in this area include the stabilization of the biologically active membranes and coatings and the development of methods for continuous delivery of the biological macromolecules into the system. Human exposure assessment is another area in which biosensors could make a significant contribution. Regulations governing the release of chemicals into the environment are typically influenced by the perceived risk to public health and local ecosystems. Determination of human exposure requires measurements of the compound of interest at locations, within the time frames and in the media (e.g, air, soil or water) most likely to result in exposure. This determination is complicated by the fact that an individual typically engages in a variety of activities and moves through a number of microenvironments in which exposure may occur. A useful tool in the determination of exposure to airborne contaminants is a sampling device worn by an individual (e.g. a dosimeter). These dosimeters are typically configured as a sample concentration medium from which the compounds of interest are extracted and analysed by a classical method. Current efforts in this area include work on configurations in which the analyte is trapped by an immunospecific medium and analysed by immunoassay techniques. A cumulative exposure over a given time period can then be calculated. A gas permeable membrane
320
interfaced to an immunosensor might also allow instantaneous measurements, as well as time integrated average exposures. This information would prove invaluable for integrating exposure patterns into risk assessment models. Biosensors may also contribute to the area of risk assessment that attempts to correlate exposure to effective dose, adverse biological effect and ultimately to clinical disease. Because of the diversity of possible routes of entry, however (e.g. skin, lung and gut), exposure is not easily correlated even to effective dose. Extrapolations from animal systems must be used and these approximations are often confounded by differences in pharmacokinetics between species. The measurement of a pollutant or its metabolites in easily sampled biological fluids such as blood, urine or saliva has provided a useful tool for determining dose in humans. Measurement of statistically significant populations using classical analytical methods, however, is expensive and time-consuming. Biosensors may again provide a solution for fast and economical field assays for estimation of human exposure to hazardous pollutants.
PROSPECTS FOR COMMERCIALIZATION
Because of the wide range of their possible applications, biosensors have attracted tremendous interest and attention in science since the late 1970s. After more than a decade of intense research, however, most effort still remains at the level of research curiosities in universities and research institutes. At present, there are only a few biosensor technologies, among hundreds of existing designs, that have been developed into commercial products. None of these biosensors have been commercialized for environmental applications. Although each sensor has its own performance requirements and design specifications, there are several common obstacles to commercialization of biosensors in general.
Biosensors & Bioelectronics
First, the performance characteristics such as sensitivity, dynamic range, reproducibility, and stability, for sensors demonstrated in the research stage have not quite met the needs of users. Furthermore, the performance of the biosensor is further reduced in the development and manufacturing stages as a result of mass production. Therefore, to make a practical device, it is important to keep chemical and physical design characteristics simple, using proven technologies wherever possible. The second obstacle is the lack of effort toward system integration and user-interfacing. The sensor concepts need to be taken one step further by building a prototype model that can be easily operated by untrained users. The third obstacle involves the competitiveness of the biosensors. Commercialization becomes difficult or impossible if no definite advantages over other competing technologies are achieved. Finally, probably the most important factor responsible for the slow commercialization of
biosensors is the market size. Because of the complexity of these devices, millions of dollars must be invested to develop a biosensor. Therefore, until profitable marketplaces are realized, private companies will not be motivated to investigate and transfer the biosensor technologies to products. Issues related to competitiveness and market size for most biosensor technologies are still being defined by current research and market developments. We believe that once the marketplaces and attributes of biosensors are fully realized, they will have a more important role in various sectors of the analytical industry, such as environmental monitoring. Although the research described herein has been funded wholly or in part by the US Environmental Protection Agency, it has not been subject to Agency review and therefore does not necessarily reflect the views of the Agency, and no oficial endorsement should be inferred. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.
FURTHER READING Eldefrawi, M.E., Eldefrawi, A.T., Rogers, K.R., Valdes, J.J. (1992) Pharmacological biosensors, In Immunochemical Assays and Biosensor Technology for the 199Os, 1992, edited by Nakamura, R.M.,
Kashahara, Y. and Rechnitz, G.A. American Society for Microbiology, Washington, DC. National Research Council (1991). Environmental epidemiology, Vol. 1, public health and hazardous wastes, National Academy Press, Washington, DC, USA. Van Emon, J.M. and Lopez-Avila, V. (1992). Immunochemical methods for environmental analysis. Anal. Chem., 64,79A-88A.
321
