6
Virulence-Factor Activity Relationships

INTRODUCTION

The term “virulence-factor activity relationship,” or VFAR (formerly referred to as virulence-activity relationship or VAR; NRC, 1999a), is rooted in a recognition of the utility of using structure-activity relationships (SARs) to compare the structure of newly identified or produced chemicals to known chemical structures to enable prediction of their toxicity and other physical properties. In essence, the committee believes the same principle can be applied to waterborne pathogens. It is important to state that many sections of this chapter necessarily include more extensive use of scientific terms and language than might typically be found in the body of a National Research Council (NRC) report. That is, rather than deleting, simplifying, or relegating such relevant technical language to an appendix, the committee decided to keep all information related to VFARs in one comprehensive chapter. This chapter should be read with that qualification in mind.

For microorganisms, there are many levels of structure, such as the cell or organism itself and the larger internal components that comprise the microorganism (e.g., nucleus, micronemes, flagellae). These morphological components can sometimes be used to identify pathogenic microorganisms. Beyond these relatively large structures, there are smaller, biochemical components of the organism, including proteins, carbohydrates, and lipids. Many of these biochemical building blocks are directly related to how a particular microorganism causes disease. Some examples of these include the outer coat of some bacteria (the lipid polysaccharide coat), attachment and invasion factors, and bacterial toxins. Thus, the central premise of VFARs is to relate the architectural and biochemical components of microorganisms to potential human disease.



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 143
Classifying Drinking Water Contaminants for Regulatory Consideration 6 Virulence-Factor Activity Relationships INTRODUCTION The term “virulence-factor activity relationship,” or VFAR (formerly referred to as virulence-activity relationship or VAR; NRC, 1999a), is rooted in a recognition of the utility of using structure-activity relationships (SARs) to compare the structure of newly identified or produced chemicals to known chemical structures to enable prediction of their toxicity and other physical properties. In essence, the committee believes the same principle can be applied to waterborne pathogens. It is important to state that many sections of this chapter necessarily include more extensive use of scientific terms and language than might typically be found in the body of a National Research Council (NRC) report. That is, rather than deleting, simplifying, or relegating such relevant technical language to an appendix, the committee decided to keep all information related to VFARs in one comprehensive chapter. This chapter should be read with that qualification in mind. For microorganisms, there are many levels of structure, such as the cell or organism itself and the larger internal components that comprise the microorganism (e.g., nucleus, micronemes, flagellae). These morphological components can sometimes be used to identify pathogenic microorganisms. Beyond these relatively large structures, there are smaller, biochemical components of the organism, including proteins, carbohydrates, and lipids. Many of these biochemical building blocks are directly related to how a particular microorganism causes disease. Some examples of these include the outer coat of some bacteria (the lipid polysaccharide coat), attachment and invasion factors, and bacterial toxins. Thus, the central premise of VFARs is to relate the architectural and biochemical components of microorganisms to potential human disease.

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration Virulence can be defined as the quality of being poisonous or injurious to life (i.e., virulent). For an organism to be virulent, it must be able to infect its human host, reproduce, and/or cause a disease. This broad definition of virulence is more inclusive than the narrow definition commonly used by microbiologists (i.e., virulence is solely the severity of the disease produced after exposure and infection). Each of the microbiological attributes that contribute to virulence can in general be linked to specific architectural elements or biochemical compounds within the organism. Together, these elements and compounds can generally be termed “virulence factors,” and the blueprints for them are included in the genetic code of an organism. For this reason, a principal topic of this chapter is the genetic structure of various microorganisms because of its direct relationship to virulence factors. Owing to recent advances in molecular biology, the genetic structures of many thousands of organisms (especially bacteria and viruses) have been identified, reported, and stored in what are called gene banks. Sophisticated computer programs allow for the sorting and matching of genetic structures and specific genes. The discipline that organizes and studies these genes is known as bioinformatics, while the study of genes and their function is known as genomics. In addition, a growing area of related interest is functional genomics, that is, understanding the specific role of genes in terms of the function of the organism. The ability to use these and related tools to address the microbial contamination of drinking water is illustrated by some of the following observations: The genetic structures of most known waterborne pathogens have been characterized at least partially, with the information stored in gene databanks. The complete genome of several important waterborne pathogens, such as Vibrio cholerae (the agent of cholera), is now known, and many more will be characterized in the near future (Heidelberg et al., 2000). Other related information is accumulating that allows the use of these databanks to determine or predict the ability of a microorganism to produce virulence factors, such as toxins, attachment factors, and other surface proteins, and genes that encode bacterial resistance to antibiotics. On a more basic level, these data can be used to characterize similarities and differences between a microorganism of interest and known pathogens.

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration Data of this kind can also be used to identify sources of, and thus exposures to, microorganisms through molecular “fingerprinting.” The functional genomics or bioinformatics expertise needed to establish a nationwide VFAR program already exists in the private and public sectors. Thus, the committee concludes that a VFAR concept, with many parallels to the SAR concept used for chemicals, would be a powerful approach to examining emerging waterborne pathogens, opportunistic microorganisms, and other newly identified microorganisms. STATEMENT OF THE PROBLEM As noted in the committee’s first report (NRC, 1999a), the current approach to identifying and controlling waterborne disease is limited. It has followed a similar path since cholera was first linked to transmission via water (“from the Broad Street pump”) in London, England, nearly a century and an half ago, and since Koch first proposed his famous postulates regarding causation (see Okun, 1999). Typically, a disease outbreak is reported only when a significant portion of a community is recognized to have been affected, the responsible microorganism has been identified, and an epidemiological study is undertaken to determine possible sources of exposure to the agent in the community. If any of these three elements is lacking, the outbreak is generally missed and goes unreported. If the consumption of drinking water is identified as a potential source of exposure, a public health advisory to boil water may be issued. Alternatively, the culpable part of the system may be identified and isolated until the cause of contamination is eliminated. However, for most of the waterborne outbreaks in the United States, the etiology is never determined, the responsible microorganism is never identified, and public water systems are not easily fixed or shut off. The identification of pathogens is thus unnecessarily related to the recognition of an outbreak. Under the amended Safe Drinking Water Act, microbial contamination, regardless of whether it is associated with an outbreak or not, must be addressed. Hundreds if not thousands of microorganisms have the potential to be spread through drinking water supplies and distribution systems. While data on health effects for many of these are described in the medical literature, there are no occurrence databases or even routine methodologies for developing these databases (NRC, 1999a). One of the princi-

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration pal dilemmas to be addressed is that current regulatory practice requires that methods to culture organisms of interest be developed before occurrence data can be gathered. Thus, a microorganism ordinarily must first be identified as a pathogen, and be capable of in vitro culture, before occurrence data are acquired. This long-standing paradigm makes it very difficult or impossible to develop a database of potential or emerging pathogens. There is also no widely accepted approach for prioritization of waterborne pathogens, other than through expert judgment. For example, and as noted earlier, the U.S. Environmental Protection Agency (EPA) and the American Water Works Association Research Foundation have jointly sponsored a series of expert workshops since 1996 for the development of a decision process for prioritizing emerging waterborne pathogens that is nearing completion. These expert judgments must be made, of necessity, by a very small number of researchers in the discipline of health-related environmental microbiology. This approach to the process makes transparency very difficult to achieve, the importance of which is discussed in Chapters 2 and 5 of this report. The committee believes strongly that if EPA continues to rely on exposure and health effects as two primary data categories for screening potential microbial drinking water contaminants, progress will continue to be unacceptably slow. Current efforts are able to address only one or two microorganisms every 5 to 10 years with the current CCL development and implementation approach. To illustrate the dilemma, consider that of the ten microorganisms and groups of related microorganisms on the 1998 CCL, nine are in the “research priorities” category (see Table 1–3) and will go unregulated in the first CCL cycle. Of these nine microbial contaminants, only one, Aeromonas hydrophila, is slated for delayed screening level monitoring during the first cycle of the Unregulated Contaminant Monitoring Regulation (UCMR) (EPA, 1999c) (see Table 1–4). It is clear that a severe bottleneck exists in identifying and addressing important microbial contaminants in drinking water. Thus, a new approach to assessing pathogens could help overcome this ongoing problem. VFAR ANALOGY TO SARS AND QUANTITATIVE SARS A variety of terminology has developed in the literature to identify various classes of correlations useful for predicting the properties of

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration agents in environmental and health sciences. For example, chemical properties are amenable to prediction through use of structure-activity relationships, which can be distinguished from property-activity relationships (PARs) and structure-property relationships (SPRs). Although careful classification along these lines certainly has heuristic value (e.g., Brezonik, 1990), few researchers adhere to these distinctions rigorously. Instead, only a few terms are commonly used and these are often applied to a wider range of correlation types than strict use of each expression would allow (Tratnyek, 1998). One example of this is the term linear free energy relationship (LFER), which originally referred to a specific type of correlation used by physical organic chemists but eventually came to represent the entire field of correlation analysis in organic chemistry (Shorter, 1973). Similarly, the term quantitative structure-activity relationship (QSAR) was originally coined for use in drug design but is now commonly used to refer to many types of correlations employed in the pharmaceutical, toxicological, and environmental sciences. By analogy to the above discussion, the committee has coined the term “virulence-factor activity relationship” and defined it as the known or presumed linkage between the biological characteristics of a microorganism and its real or potential ability to cause harm (pathogenicity). FRAMEWORK The central concept is to use microbial characteristics to predict virulence via what the committee terms a virulence-factor activity relationship. Microbial VFARs would function in much the same way as QSARs do, namely to assist in the early identification of at least several potential elements of virulence. Research increasingly has shown certain common characteristics among virulent pathogens, such as the production of specific toxins, specific surface proteins, and specific repair mechanisms that enhance their ability to infect and inflict damage in a host. Recently some of these “descriptors” (the terminology often used in QSARs) have been tied to specific genes, and it has become evident that the same can be done for others. Identification of these descriptors, either directly or through analysis of genetic databases, could become a powerful tool for estimating the potential virulence of a microorganism. This is particularly true for two important aspects of virulence: potency and persistence in the environment. The committee conceives of VFAR

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration as being the relationship that ties specific descriptors to outcomes of concern (see Figure 6–1). FORMULATING VFARS Conceptually, pathogens of interest must be related in that they exhibit pathogenicity through a common mechanism but are also likely to be distinguished through secondary characteristics that cause virulence to be variable. Since virulence is the target property to be predicted by the VFAR, it is by definition the dependent or “response” variable in a VFAR. Variability in the virulence of pathogens may be characterized by one or more independent variables (i.e., variability in the genetic makeup) —referred to as “descriptor variables” —that can be conven- FIGURE 6–1 Schematic drawing of VFAR predicting outcomes of concern (virulence, potency, persistence) using the presence or quality of “descriptor” variables.

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration iently measured or otherwise determined. When correlation of the response and descriptor variables yields a consistent relationship, the result can be used as a (quantitative) model for comparing and predicting properties of related pathogens. Note that there are many potential ways in which response and descriptor variables may be defined, and this gives flexibility to the VFAR approach, such that it should in principle be able to accommodate many complicating factors. For example, when variability in a pathogen’s virulence is related substantially to host factors (e.g., when the host is in an immuno-compromised state) then an “interaction effect” could generate cases that do not obey a VFAR. However, if the response variable (virulence) is defined in such a way that it is unaffected by the behavior of opportunistic pathogens, or if descriptor variables are used that incorporate opportunistic behavior, then a VFAR can incorporate this effect and outliers can be avoided. Such subtleties suggest that developing and validating robust and reliable VFARs will require considerable research, but the committee believes that the promise of VFARs should make them a high priority for such research. Response (Outcome) Variables As noted previously, the response variable of concern in VFARs is virulence. Narrowly defined, pathogenicity can be characterized as the ability to cause disease and clinical virulence as a measure of the severity of disease. A broader definition is used in this report, where virulence (with respect to VFARs) incorporates both the concept of pathogenicity and the narrower concept of clinical virulence. Viewed in this manner, it may be useful to include attributes of persistence in the environment as contributing to virulence. It is also conceivable that pathogenicity, clinical virulence, and environmental persistence could be considered separate response variables that work together to contribute to the broadly defined “virulence” of a pathogenic organism. There are a number of potential metrics of virulence (broadly defined) that may be used as a quantitative outcome measure. These include the duration of symptomatic illness and the intensity of symptoms (perhaps using a disability-weighted scale).

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration Descriptor Variables Descriptor variables, in this context, are those attributes of a microorganism that may prove useful in predicting their virulence. For example, the presence of toxins, adherence factors, adhesins, invasins, capsular components, fimbria, hemolysins, metabolic pathways, and antibiotic resistance could prove to be effective descriptors of microbial virulence. Alternatively, association with certain families of pathogenic microorganisms may be sufficient as a descriptor (e.g., for viruses), and species and genotype may be all that is necessary for protozoa. As our knowledge of pathogens improves, the definition and calibration of specific descriptors will evolve as well. For many pathogens, the specific mechanisms or virulence descriptors that underlie the range of virulence from one genotype to the next are not well understood. Because of this circumstance, it has already been demonstrated in waterborne pathogens that a genetically based VFAR approach could be particularly powerful. For example, recent studies suggest that various isolates or species of Cryptosporidium are virulent to varying degrees in humans (e.g., Okhuysen et al., 1999; Widmer et al., 1998). The ability to recognize and differentiate the genomic content of these different isolates or species, and thus recognize differences in virulence, is based upon the same intellectual concepts that underlie the recognition of toxin-encoding bacterial genes. The power of a VFAR approach is that it has the ability to genuinely reflect the true biological diversity found in human pathogens, even when the exact mechanisms that shape this diversity are not yet understood (Morgan et al., 1999a; Sulaiman et al., 2000). The committee anticipates that the VFAR paradigm is robust enough to accommodate the reality that sometimes the mere presence of a protozoan in drinking water is not of public health concern. For example, there is now abundant evidence that the species Cryptosporidium parvum is, in fact, made up of a number of genotypes, each with different virulence where the human population is concerned (Xiao et al., 2000). Furthermore, one study (Morgan et al., 1999a) used genetic methods to identify eight different species of Crytposporidium: parvum (many mammals), muris (rodents, cattle), felis (cats), wrairi (guinea pigs), meleagridis and baileyi (birds), serpentis (reptiles), and nasorum (fish). The same study demonstrated seven genotypes of parvum: genotype 1 infects humans only; genotype 2 infects cattle, sheep, goats, and humans; genotype 3 infects mice and bats; genotype 4 infects pigs; genotype 5 infects koalas and kangaroos; genotype 6 infects dogs; and genotype 7

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration infects ferrets. Subsequent studies indicate the existence of an eighth genotype (Sulaiman et al., 2000). Of these, only genotypes 1 and 2 are believed to infect immuno-competent humans, but several genotypes have been found to infect immuno-compromised individuals (Morgan et al., 1999b). In addition, several non-parvum species (C. felis, C. meleagridis) have been found to infect people with AIDS (Morgan et al., 2000). Genomics and Proteomics Proteomics, a discipline within functional genomics, is the study of protein sets made (expressed) when the genomic blueprint of an organism is actually translated into functional molecules. When faced with changing environmental conditions, organisms will respond by making different sets of proteins to help them survive. For example, it has been estimated that Vibrio cholerae is capable of making approximately 3,900 different proteins depending on environmental conditions (Heidelberg et al., 2000). These proteins are the actual molecules that build other important structural molecules, such as lipids, deoxyribonucleic acid (DNA), and ribonucleic acid (RNA), and are capable of having both structural and catalytic or enzymatic functions. It is known that some important bacterial toxins (such as Shiga toxin, discussed later in this chapter) are maximally produced under very specific conditions (Acheson et al., 1991). Faced with a hostile environment, many bacteria will shift production of a protein set that is associated with growth to another set associated with a viable but nonculturable state or to the formation of spores as discussed later in this chapter. Thus, knowledge of the set of proteins being made by an organism can impart information far more revealing that that gained from studying the expression of a single protein. The committee anticipates that because the state of the art of genomics is currently more advanced than that of proteomics, the initial emphasis in VFAR formulation will be genetic. While much is already known about the growing field proteomics, the committee believes it would be premature to discuss or make recommendations about how much research and data will be needed to examine this aspect of developing VFARs, particularly under changing environmental conditions. Nonetheless, the logical extension of identifying and understanding the entire genome of an organism is ascertaining how this is translated into the expression of proteins and other structural building blocks. In this

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration regard, the committee anticipates that the same rationale that exists for using genomics also exists for proteomics. For example, in a subsequent section of this chapter the committee discusses the use of DNA chips that act as sensors for finding the characteristic DNA elements that encode a particular virulence factor. These chips function via a binding interaction between a section of DNA spotted onto the chip and the complementary strand of a target DNA molecule, such as one from a pathogenic organism. Protein chips that bind the actual virulence protein factors could work in an analogous fashion. Under such a scenario, binding molecules known to attach to specific bacterial toxins (e.g., monoclonal antibodies) could be spotted onto a chip and used to sift through the proteins expressed by a novel bacterium to see if a protein of concern is made by the targeted organism. CURRENT LEVEL OF GENETIC CHARACTERIZATION In this section, three existing, major bodies of endeavor that have relevance to the development and implementation of VFARs are discussed. Microbial Genome Projects and Comparative Databases The first major endeavor to be discussed in this section is the set of single-organism genome projects and the large genomic databases that are used for comparing the genes of one organism with those of others. The genome projects are comprehensive attempts to sequence the entire genomes of organisms, such as yeast, pathogenic microorganisms, and humans. Computerized analysis and the growing use of automated polymerase chain reaction (PCR) techniques have allowed for tremendous gains in the study of microbial genomics as well as of whole organisms. The databases that exist to store such information are large and expanding daily. For example, the Institute for Genomic Research (TIGR) maintains a collection of databases containing DNA, protein and gene expression, and taxonomic data for microbes, plants, and even humans (see http://www.tigr.org for further information). TIGR also provides links to worldwide genome sequencing projects. A number of microorganisms are listed in Table 6–1 whose genomes have already been studied; the results of much of this work are available in the published literature. A number of these organisms are associated

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration TABLE 6–1 Examples of Microbial Genome Databases for Waterborne Pathogens Microorganism Size (Million Base Pairs) Campylobacter jejuni 1.7 Encephalitozoon cuniculi 2.9 Enterococcus faecalis 3.0 Escherichia coli 4.6 Giardia lamblia 12 Helicobacter pylori 1.66 Klebsiella pneumoniae   Legionella pneumophila 4.0 Leptospira interrogans serovar icterohaemorrhagiae 4.8 Mycobacterium avium 4.7 Pseudomonas aeruginosa 5.9 Salmonella paratyphi A 4.5–4.8 Salmonella typhi   Salmonella typhimurium   Shigella flexneri 4.7 Vibrio cholerae 4.0   SOURCE: TIGR (see http://www.tigr.org). with waterborne disease. For example, studies on the Giardia genome have recently been published (Adam, 2000), and the complete sequence of Vibrio cholerae was recently announced with great acclaim (Heidelberg et al., 2000). The Cryptosporidium genome is being sequenced by investigators at the University of Minnesota (http://www.cbc.umn.edu/ResearchProjects/AGAC/Cp/index.htm), with other important work being conducted in the United Kingdom (http://www.mrclmb.cam.ac.uk/happy/CRYPTO/Ref.html), California (http://medsfgh.ucsf.edu/id/CpTags/), and elsewhere. Notably, funding for the Vibrio cholerae and Cryptosporidium genome projects was provided by the National Institute of Allergy and Infectious Diseases (NIAID) at the National Institutes of Health (see http://www.niaid.nih.gov/dmid/genomes/genome.htm for a listing of genome projects currently supported by NIAID). On May 30,2000, an important report entitled Interagency Report on the Federal Investment in Microbial Genomics was published by the Biotechnology Research Working Group—a subcommittee of the National Science and Technology Council (BRWG, 2000). The charge for

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration niques in detecting nonculturable bacteria rather than on persistence responses.) In some cases, however, the pathogenicity of the putative viable nonculturable organisms may be low, as reported by Caro et al. (1999) for Salmonella typhimurium. In this dormant state, assessment of microbial levels by plate counts may indicate little occurrence (unless resuscitation has been triggered), while other assays such as total microscopic count or nucleic acid assay may indicate higher levels. Table 6–2 summarizes some published reports of the occurrence of viable nonculturable states in bacteria. Range of Decay Rates Regardless of the mechanism(s) promoting loss of viability of microorganisms in aquatic systems, several researchers have reported the observed rate of disappearance under different conditions as summarized below. TABLE 6–2 Viable Nonculturable States in Bacteria Microorganism References Campylobacter jejuni Buswell et al., 1998 Rollins and Colwell, 1986 Coliforms McFeters et al., 1986 Escherichia coli O157:H7 Rigsbee et al., 1997 Klebsiella pneumoniae Byrd et al., 1991 Enterobacter aerogenes Agrobacterium tumefaciens Streptococcus faecalis Micrococcus flavus   Salmonella enteriditis Chmielewski and Frank, 1995 Salmonella dysenteriae Islam et al., 1993 Enterococcus faecalis Lleo et al., 1998 Legionella pneumophila Steinert et al., 1997 Vibrio vulnificus Weichartand Kjelleberg, 1996

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration Determination of Viability For studies that measure decay rates in microorganisms, the particular assays used for assessment of viability become important. In most cases, but particularly for data obtained before the mid-1990s, the assessment of viability was based on culturing methods such as agar plate growth of bacteria and most-probable-number assays for bacteria or viruses (in the form of the TCID50 determination [i.e., tissue culture infectious dose, or dose required to infect 50 percent of the tissue culture in which a sample is inoculated]). As noted previously, however, bacteria may form viable nonculturable stages that by definition are not readily enumerated using culture techniques. Therefore, reliance on culture techniques may incorrectly estimate the true decay rates. In more recent years, some investigators have used molecular genetic techniques such as PCR (Abbaszadegan et al., 1999; DiGiovanni et al., 1999; Shieh et al., 1997; Sturbaum et al., 1998) to assess occurrence or decay of pathogens in environmental systems. Although PCR and other molecular methods may allow more efficient data collection, they may also overestimate the occurrence or persistence of viable microorganisms (Deere et al., 1996; Dupray et al., 1997). Thus, any reports of survival times (or occurrences) of pathogens in water should be accompanied by a description of the methods used to assess viability. Data for Established Pathogens As part of a mid-1980s reevaluation of the coliform standards for drinking water, a comprehensive review of the decay of indicators and pathogens in water was performed under the sponsorship of EPA. A summary of these decay values is provided in Table 6–3. The original tabulated values of times required for 50, 90, 99, or 99.9 percent reduction are indicated—this is preferable to conversion to a single metric (e.g., half-life), since in many cases the underlying data differed from ideal first-order decay. The information in this table reflects microorganism survival under diverse conditions ranging from raw water (of various sources) to finished drinking water (although in no circumstances was there any disinfectant residual present).

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration Data for More Recent Pathogens There have been studies of additional microorganisms since the efforts summarized below in Table 6–3. Although a comprehensive review of such studies is beyond the objective of this report, a brief synopsis of findings for some emerging pathogens is appropriate. DeRegnier et al. (1989) suspended Giardia muris in river water and lakewater at ambient temperature to monitor viability using propidium iodide and animal infectivity. As measured by infectivity, cysts remained viable at least 40 days. It should be noted, however, that the small number of test animals did not likely permit measurement of inactivation beyond one log. The authors concluded …G. muris cysts suspended in environmental water remained viable for 2 to 3 months, and their survival was enhanced by exposure to low water temperature, despite the fact that the cysts were suspended in the fecal biomass within the sample vial. TABLE 6–3 Survival Times for Pathogens in Raw or Finished Watersa   Temperature (°C) Time for Indicated Die-off (Days)   Microorganism T50 T99 T99.9 Reference Campylobacter jejuni 25     2–3 Blaser et al., 1980 Campylobacter jejuni 4     3–18 Blaser et al., 1980 Coxsackievirus A9 19–25   >21   Herrmann et al., 1974 Coxsackievirus B1 4–8   2.4   O’Brien and Newman, 1977 Coxsackievirus B3 20     6–8 Hurst and Gerba, 1980 Echovirus 7 20     3 Hurst and Gerba, 1980 Entamoeba histolytica 4     55–60 Chang, 1943 Entamoeba histolytica 6–8     38–42 Chang, 1943 Entamoeba histolytica 21–22     7–8 Chang, 1943 Poliovirus 1 19–25   >21   Herrmann et al., 1974 Poliovirus 1 4–8   2   O’Brien and Newman, 1977 Poliovirus 1 and 3 23–27   1,1.8   O’Brien and Newman, 1977 Rotavirus SA-11 20     10–14 Hurst and Gerba, 1980 Salmonella enteritidis 9.5–2.5 0.66–0.79     McFeters et al., 1974 Shigella flexneri 23–25     4–21 Mohadjer and Mehrabian, 1975 Vibrio cholerae 9.5–12.5 0.29     McFeters et al., 1974 a Underlying data obtained using culture techniques. SOURCE: Adapted from Sobsey and Olson, 1983.

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration Robertson et al. (1992) used a differential dye inclusion assay to monitor viability decay of two strains of Cryptosporidium parvum oocysts. One was originally isolated from deer and cultured in sheep, and the other was a bovine isolate. The organisms were held in membrane diffusion chambers in river water under ambient conditions. Their results indicate that 1-log inactivation is estimated to occur at 100 and 180 days for the two strains examined. More recently, Jenkins et al. (1997) determined that the survival of oocysts in fecal material (as measured by vital dyes) correlates well with the ability of the oocysts to excyst. However there is continuing controversy about the suitability of dye incorporation assays versus other techniques with respect to assessing oocyst viability (Belosevic et al., 1997; Bukhari et al., 2000). As noted earlier in this chapter, FISH techniques that use mRNA or other targets may be superior to estimate microbial viability. Enriquez et al. (1995) reported that adenovirus held for 60 days in dechlorinated tap water produced only a 2-log reduction at 23°C as measured through tissue culture assays. Intrinsic Factors Influencing Decay A key question is to what degree the persistence or decay of pathogens in water can be predicted quantitatively and how this information could be used in the construction of VFARs. Based on the preceding information, it is clear that a wide range of variation exists in the removal rates of pathogens in aquatic systems. However, beyond some broad generalizations, a fully quantitative model that incorporates effects of adverse conditions on a range of pathogens has eluded investigators. It is encouraging, however, that in some cases for specific microorganisms, an overall predicted model can be developed (Auer and Niehaus, 1993; Auer et al., 1998; Canale et al., 1993; Chamberlin and Mitchell, 1978; Mancini, 1978). Nonetheless, there are identifiable differences between microorganisms that should allow for a semiquantitative assessment of environmental persistence; these are ability to sorb suspended solids;4 ability to form dormant stages, including viable nonculturable; 4   Assuming that sorption to suspended solids does not result in increased decay rates.

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration ability to survive and/or multiply within aquatic protozoa or other microbial hosts; ability to survive freezing; ability to survive desiccation; ability to survive wastewater treatment and to reenter drinking water; and ability to survive in anaerobic sediment. INTERPRETATION OF ISSUES Data Information and Management Issues Although the technology, methodology, and even the genetic databanks exist, the application of a VFAR approach to assess waterborne pathogens would require considerable effort and expenditure of resources by EPA in conjunction with the Centers for Disease Control and Prevention, National Institutes of Health, and other federal and state health organizations (NRC, 1999b). Such an interagency “Waterborne Microbial Genomics” (WMG) project would also require extensive expertise in bioinformatics, molecular microbiology, environmental microbiology, and infectious disease. Initially, existing gene banks would have to be screened and evaluated for key targets. For example, the National Center for Biotechnology Information jointly established by the National Library of Medicine and the National Institutes of Health maintains GenBank (www.ncbi.nlm.nih.gov). However, the available sequence data are not error free, and greater quality control and quality assurance would have to go into screening the genetic information. Of course, new data on genetic sequences would have to be added to the WMG as they are reported in the clinical literature. As a start, this would require the evaluation of literature for references to microorganisms having the potential to be waterborne, found in feces or urine, and naturally occurring in the water environment. Furthermore, protocols would have to be established and tested regarding data entry, validation, and use in the development of microbial VFARs. Background levels and determination of prevalence, persistence, and quantity of key target genes would have to be gathered for analysis and interpretation of health risk and/or exposure potential. Outbreaks and contamination events would provide useful information to enhance the database. Once key parameters had been established, the

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration possibility of screening hundreds of water samples for thousands of key microbial hazards could be achieved through development of custom chip arrays. The use of appropriate data and sophisticated data management would be crucial to the development and validation of VFARs. Because VFARs have the potential to be very powerful, they will require thorough validation and careful use, with attention to the limits of their validity. Validation can proceed in a variety of ways, such as the statistical measure of how well a VFAR fits into the descriptor or response data, the use of sensitivity analysis to validate VFAR data that are already available, using a VFAR with new data after it has been established with older data, and comparing the predictions of one VFAR to those of another. All of these will require the development of appropriate data sets and data management tools. The committee also notes that the use of prototype classification methods to help select PCCL contaminants for inclusion on a CCL can obviously be applied to VFARs. That is, training sets of descriptor and response variables could be developed and used in conjunction with prototype classification methods to help derive VFARs. This again implies that such training sets would have to be appropriate and robust. The committee fully recognizes that just the initial establishment of such a program (excluding maintenance and expansion) is likely to require at least a five-year commitment and significant cooperation and expenditure of resources by EPA and other participating organizations. However, the opportunities for rapid identification of microbial hazards in water afforded by such a program would greatly improve the ability of EPA to quickly and successfully protect public health and improve water quality. FEASIBILITY For the VFAR concept to be ultimately adopted and used by EPA in its drinking water program, it must be feasible. Several aspects of feasibility are discussed below, including scientific validity and applicability; actual technological feasibility; application of these technologies to studying disease in humans (validation); the degree to which these methodologies are being universally adopted within the scientific community; and the need for their development and use to adhere to the principles of transparency, public participation, and other sociopolitical considerations reviewed in Chapter 2. To one extent or another, each of these elements

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration affects the ability of the VFAR concept to be developed, used, or validated. Since these elements either are present or can reasonably be expected to be available in the near future, the committee strongly concludes that the development and use of VFARs is indeed feasible. Having carefully noted some caveats and limitations in the preceding text, the committee remains enthusiastic about the utility of developing and using VFARs in the protection of the nation’s drinking water. First, the underlying concept must be scientifically valid and robustly applicable. As previously noted, the relationship between specific microbial attributes and human disease has been known for more than a century. This linkage has become increasingly documented and precise as our knowledge of microorganisms and human disease has dramatically improved in the past few decades. While the illustrative examples provided in this chapter are by no means exhaustive, they certainly speak to the power of specific microbial attributes to predict virulence in humans. It is not only possible, but in fact now routine, to associate very specific human disease outcomes with the presence, absence, or variability of specific microbial characteristics (or “descriptors” in the language of SARs). To state that the concept is robustly applicable, the committee means that it is neither limited nor narrow, but is in fact valid across a very large number of microorganisms and remains valid when small variations in a single organism are explored in great depth. The fact that this is indeed the case, and is considered a paradigm of biomedical science, provides a convincing demonstration of the validity and robustness of the concept. Thus, the committee deems that the first necessary condition for feasibility exists. For the VFAR concept to be useful, it must be able to extend a known relationship between a virulence attribute and human disease to a situation in which the attribute is found in a new or unexpected circumstance or in microorganisms that have not heretofore been recognized as potentially pathogenic. The profound scientific revolution associated with the unraveling of DNA’s double helix speaks to this second aspect of feasibility—the disciplines of bioinformatics, proteomics, and genomics. The ability to rapidly, and completely, sequence the genome of entire organisms, and to use bioinformatic techniques to compare gene sequences of different organisms, provides this mechanism for comparison. The committee adds that the time and cost required for sequencing a microorganism have both declined markedly in just the past few years. Independent of any judgment made by this committee, a number of biotechnology and pharmaceutical companies have chosen to aggressively pursue opportunities that rely on comparative genomics, which the com-

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration mittee regards as confirming its judgment of the adequacy, power, and affordability of the methodology. The committee thus warrants that the necessary condition of technological feasibility also exists. Although the committee is unclear as to whether compelling logic currently exists for the development of VFARs based on genomic techniques that can be extended to proteomics, it fully acknowledges the potential for this in the near future. The possibility for proteomics to also play a role in the development of VFARs adds both depth and an additional dimension to ways in which VFARs might operate in the future within EPA’s drinking water program. A third element by which to judge feasibility is the likelihood that adverse human outcomes (e.g., diseases) will continue to be discovered in association with the action or presence of a microbial contaminant, microbial gene, or gene product in the clinical setting. This clinical linkage between diseases and specific microorganisms or their products has been a hallmark of medical sciences for the past two centuries. There is no indication that the accelerating pace of these medical discoveries is abating; one need only consider prions and mad cow disease, Ebola virus, Escherichia coli O157:H7, Helicobacter pylori, and nanobacteria to be reminded of important pathogens that were unknown to science until a few decades ago. As discussed previously, the intent of the VFAR is to characterize, categorize, and make scientific the potential linkage, as outlined in Figure 6–3 below. This external element represents both a validation of the linkage for those organisms already known to cause disease and an element that will be helpful in validating the discovery of emerging waterborne pathogens through the use of VFARs. For “established” waterborne microoorganisms such as Vibrio cholerae, Salmonella, and the pathogenic protozoa, this linkage was, and is, made easily. That is, these microorganisms are already easily cultured or visible in human specimens—no new technological or scientific advances are required for them to be linked to human disease. In the case of emerging microorganisms that are often unsuspected agents of disease, or are difficult to detect using traditional methods of culture or microscopy, it is likely that the novel molecular detection techniques discussed in this chapter (e.g., PCR, gene chips) will continue to lead to new medical associations. Identification of some viruses (e.g., herpes simplex virus) or specific microbial antigens through PCR techniques has now become commercially viable and widespread in the medical setting, replacing earlier methods. Many diagnostic kits are now available to detect antigens shed by pathogenic bacteria and viruses in urine, blood, and spinal fluid. Viruses that were once “too expensive”

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration FIGURE 6–3 Linkage between microbial virulence factors and human disease. to look for on a routine basis (e.g., rotavirus) are now tested routinely for in the clinical microbiology laboratory using simple ELISA (enzyme-linked immunosorbent assay) methods. Indeed, ELISA kits that recognize Giardia and Cryptosporidium antigens are already available (e.g., Alexon-Trend’s ProSpecT® assays). Thus, the committee can foresee the likelihood that the same technologies needed for constructing VFARs will already be used in the clinical setting to detect microorganisms or their products, further strengthening the utility of the VFAR relationship. A fourth element regarding feasibility is the likelihood that EPA’s movement to adopt VFARs for use in its drinking water program will be congruent with the direction that other government, private, and public agencies are taking. In simplest terms, is such an effort likely to be adopted solely by EPA, or is it likely to be a direction that other agencies will follow? Resoundingly, all evidence points to a massive public and private investment in genomics, bioinformatics, and proteomics, which are the key disciplines behind developing and using VFARs. Further-

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration more, the specific needs of EPA are likely to be recognized by others as being of crucial importance to shared problems. For example, the “identification” problem that exists for EPA—identifying potentially pathogenic microbial water contaminants that are difficult or impossible to culture—has already been recognized by other government agencies as a high-priority area for research (BRWG, 2000). There is every reason to believe that the path EPA must follow to develop VFARs will be similar to one already blazed by many other agencies. Indeed, it can be argued persuasively that EPA will benefit very substantially from the synergistic efforts of other agencies and independent outside researchers as recommended in the committee’s second report (NRC, 1999a). In all likelihood, the resources of EPA will have to be directed primarily to (1) focusing its own internal efforts and the attention of other government agencies on waterborne pathogens and (2) integrating a tremendous knowledge base (in large part developed outside the agency) with the purpose of VFAR construction, validation, and use in EPA’s drinking water program. Thus, adoption of the technologies behind VFARs by the wider scientific community considerably improves the feasibility of any related EPA efforts to develop and use VFARs. The high-priority issues identified in this report are well recognized by other agencies that have similar needs but different applications in mind. Lastly, the committee’s (Chapter 2) recommendations that the process for selecting drinking water contaminants for future CCLs be systematic, scientifically sound, transparent, and involve broad public participation should also be met in the development and use of VFARs. CONCLUSIONS AND RECOMMENDATIONS Despite the identification and discussion of some necessary caveats and limitations, the committee concludes that the construction and eventual use of VFARs in EPA’s drinking water program are feasible and merit careful consideration. More specifically, the committee makes the following recommendations: Establish a scientific VFAR Working Group on bioinformatics, genomics, and proteomics, with a charge to study these disciplines on an ongoing basis and periodically inform the agency as to how these disciplines can affect the identification and selection of drinking water contaminants for future regulatory, monitoring, and research activities. The committee acknowledges the importance of several practical considera-

OCR for page 143
Classifying Drinking Water Contaminants for Regulatory Consideration tions related to the formation of such a working group within EPA, including how it should be administered and supported (e.g., logistically and financially) or where it could be located. However, the committee did not have sufficient time in its meetings to address these issues or make any related recommendations. The findings of this report, and especially those of the Biotechnology Research Working Group (BRWG, 2000) should be made available to the VFAR Working Group at its inception. The committee views the activities of such a working group as a continuing process in which developments in the fields of bioinformatics, genomics, and proteomics can be assessed rapidly and adopted for use in EPA’s drinking water program. The working group should be charged with the task of delineating specific steps and related issues and time lines needed to take VFARs beyond the conceptual framework of this report to actual development and implementation by EPA. All such efforts should be made in open cooperation with the public, stakeholders, and the scientific community With the assistance of the VFAR Working Group, EPA should identify and fund pilot bioinformatic projects that use genomics and proteomics to gain practical experience that can be applied to the development of VFARs while it simultaneously dispatches the charges outlined in the two previous recommendations. EPA should employ and work with scientific personnel trained in the fields of bioinformatics, genomics, and proteomics to assist the agency in focusing efforts on identifying and addressing emerging waterborne microorganisms. EPA should participate fully in all ongoing and planned U.S. government efforts in bioinformatics, genomics, and proteomics as potentially related to the identification and selection of waterborne pathogens for regulatory consideration.