SUITABILITY OF MICROBIAL ASSAYS FOR POTABLE WATER AND WASTEWATER APPLIED TO LAND

C. W. Keevil

School of Biological Sciences, University of Southampton, Bassett Crescent East, Southampton SO16 7PX, UK

1 INTRODUCTION

Emerging methods of sample preparation and pathogen detection have been developed in the food, clinical and veterinary industries, involving direct specific detection or post-enrichment culture screening with molecular, immunological and enzyme assay technologies. Some of these have also been used, with some success, by water companies involved in the supply of potable water and have been validated as ISO and CEN standards. Specific immunomagnetic separation technologies, avoiding selective enrichment, appear particularly promising and are already incorporated into several national standards for Cryptosporidium and Giardia detection in raw and treated waters. The human population continues to expand at an ever increasing rate, requiring greater intensification of animal rearing for food production, and there are inevitably greater demands on the environment to deal with the human and animal faecal wastes which are generated. Many instances of foodborne and waterborne disease outbreaks point to the transmission of new and reemerging pathogens in faecal wastes which are recycled to land for economic and environmental reasons. The challenge for analysts, therefore, is to adapt some of the methods described above, or develop new methods, to provide an evaluation of safe storage and treatment practices for pathogen removal before the wastes can be safely recycled to land. This problem has been exacerbated by the realisation that some pathogens may be sub-lethally damaged, but possibly still capable of causing disease, due to the storage and treatment processes and might be missed using conventional culture recovery techniques. The molecular techniques are presently more suited to pathogen detection, not viability, and research is now focusing on appropriate resuscitation techniques involving either presence/absence or the newer quantitative methodologies.

This paper will address some of the current legislation to ensure safe pathogen limits in wastes and the practicality of available or prototype methods for their detection in complex faecal matrices.

2 MICROBIAL ASSAYS FOR POTABLE WATER

The water industry has relied for many years on surrogate general indicators of faecal pollution to maintain a wholesome supply of potable water. Bile salt-tolerant, facultatively anaerobic bacteria isolated on MacConkey agar at 37°C, and capable of producing acid and gas from lactose at this temperature, were described as "coliform organisms". Semi-selective MacConkey medium (supplied commercially by Oxoid, Difco et al.) was devised to reduce the overgrowth when isolating faecal lactose-fermenting microorganisms; the medium incorporates sodium taurocholate (bile salts), to inhibit common environmental contaminants, and lactose is fermented to produce the neutral red coloration. Pale colonies of Salmonella or Shigella spp. are usually seen among red colonies of E. coli. Because coliform organisms can be of non-faecal origin, presumptive faecal coliforms were described as those capable of producing acid at 44°C, the majority subsequently being identified as Escherichia coli. Total coliform detection came to be relied upon because of the relative ease, speed and low cost of detection.

Alternative media include membrane lauryl sulphate broth and agar (MLSB; MLSA), violet red bile agar (VRBA), lauryl tryptose lactose agar (LTLA), desoxycholate agar and CLED agar. MLSA and desoxycholate agar (DCA) are able to differentiate E. coli from Klebsiella oxytoca in environmental water samples by virtue of E. coli' s ability to grow at 44°C as well as 37°C. MLSB is recommended by the PHLS/SCA for the isolation and enumeration of total coliforms and E. coli by membrane filtration. The recommended method for membrane filtration involves incubation of duplicate membranes on MLSB- soaked pads at 30°C for 4 h, followed by incubation of one at 37°C for 14 h (for presumptive coliforms) and the other at 44°C for 14 h (for presumptive E. coli). Minerals Modified Glutamate Medium (MMGM) is recommended for the isolation of coliforms from waters by multiple tube and is slightly superior to LTLA.4 MT7 agar (Difeo) was developed to isolate injured coliforms and is particularly suited to the recovery of chlorine treated samples.

Latterly, enzyme detection methods have been developed, based on the hydrolysis by β-galactosidase of chromogenic and fluorogenic substrates, such as 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal) and β-methylumbelliferyl-β-D-galactopyranoside, respectively. These methods offer a much quicker indication of the presence of "biologically active" coliforms. Furthermore, specificity to detect E. coli has been introduced by incorporation in various media of chromogenic and fluorogenic substrates of β-glucuronidase. These assays can be performed in liquid culture or aid in the differentiation and identification of bacterial colonies recovered on agar media. Incorporation of 4-methylumbelliferyl-β-D-glucuronide (MUG) into any of these media may give some indication of the presence of E. coli due to β-glucuronidase activity producing blue-white fluorescent colonies when viewed under ultraviolet light at 366 nm. Incorporation of MUG at 50 or 100 mg l-1 for liquid or agar media, respectively, can be used to provide presumptive evidence for the presence of E. coli which should be confirmed by further biochemical or immunological tests. Sartory and Howard described an agar medium, membrane lactose glucuronide agar (mLGA), for the simultaneous detection of E. coli and coliforms by membrane filtration without the need for UV light observation. The 5-bromo-4-chloro-3-indolyl-β-glucuronide chromophore in mLGA turned β-glucuronidase-producing colonies green (lac+ glue- colonies are yellow) and gave similar E. coli confirmation rates (92%) as isolation on MLSA. Pyruvate was also added to mLGA to substantially improve recovery of chlorine stressed coliforms. This method has now been successfully adapted to quantify the numbers of E. coli surviving in raw and treated sewage sludge 9; agar plates are incubated at 30°C for 4 [+ or -] 0.5 hours, followed by 44.0°C for 14 hours.

The use of bacterial indicators has demonstrated the effectiveness of physical and chemical treatment barriers at treatment works and in supply to prevent distribution of viable pathogens. There are several caveats, however. Are some chlorine-treated coliforms only sub-lethally damaged and, although viable, incapable of growth in the normal isolation media? As discussed, several chlorine-stress recovery agar media have been developed to help overcome this problem. Are there pathogens more tolerant of chlorine than the coliforms? Occasional waterborne outbreaks of infection caused by pathogens resistant to chlorine disinfectants, particularly Cryptosporidium parvum, have highlighted the need for more specific tests of pathogens in treated water. Because of the potentially low infectious dose of some of these emergent pathogens, the assay methods require a high precision to provide the confidence of being able to detect very low numbers in large volumes of treated water. This volume provides several important examples of the progress made in this area.

3 RECYCLING OF WASTES TO LAND

In 1992 approximately 470,000 dry tonnes of sewage sludge or biosolids were disposed of to soil in the UK. This amount has now doubled, due mainly to the ban on sea dumping of sewage which came into force in 1998 as a result of the EC Urban Waste Water Treatment Directive (1991; incorporated into UK law 1994). Land application is recognised as the Best Practicable Environmental Option for using sewage sludge. Sewage sludge and agricultural wastes are recycled to soil with the aim of improving soil condition and fertility. However, sewage sludge may contain a range of microorganisms pathogenic to man, including bacteria (e.g. Salmonella, Campylobacter, Listeria and various strains of E. coli), virus particles (e.g. Polio and Hepatitis), protozoa (e.g. Cryptosporidium) and other intestinal parasites (e.g. Helminths) (Table 1).

Without suitable treatment, there is potential for pathogens present to wash into adjacent surface waters, contaminate crops (fresh produce is of particular concern), or spread directly to man or farm and domestic animals using the land. Application of human sewage sludge currently represents a small proportion of waste applied to agricultural land; by far the greatest amounts are contributed by a variety of animal wastes including compost, faecal slurries, poultry litter, etc. (Table 2). The availability of suitable detection methods to facilitate the understanding the survival of potential human and animal pathogens in these wastes, before and after application to land, is critical to delivering safe agricultural products to the market place.

EU Council Directive 86/278/EEC regulates sewage sludge applications to agricultural land throughout the EU. The Directive is implemented in the UK by statutory instrument The Sludge (Use in Agriculture) Regulations 1989 (SI 1989, No 1263, as amended SI 880 1990; HMSO), regarding the application of raw and treated sewage sludge to agricultural land. The regulations are supported by the DOE Code of Practice for Agricultural Use of Sewage Sludge 1989 (revised 1996; DOE). These strictly limit how the sewage sludge is to be applied, under what conditions and to which crops, reducing as far as possible occasional contact with animals and man. The Code of Practice was developed in the 1970s from the data available at the time, and before the emergence of highly infectious pathogens such as E. coli 0157.

However, recent articles in the press have indicated the increasing public concern about the dangers associated with the routine dumping of blood, offal and raw sewage onto British farmland (see, for example, the "Fields of filth" article by Day). These articles paint an alarming picture of the perceived risks, largely because of concerns over newly emergent foodborne diseases believed to be caused by variant prion proteins, bacteria such as verocytotoxic E. coli, S. enteritidis and antibiotic-resistant S. typhimurium DT 104, and parasitic protozoa such as C. parvum. The problems these now cause have exacerbated the recurring problems of foodborne Campylobacter, Salmonella and Shigella spp. There are over 100,000 cases of gastro-enteritis reported in the UK each year, with some believing that the actual number infected is close to 1 million.

Pathogens such as toxin-producing Escherichia coli 0157 may indeed be occasionally present in sewage and this strain's very low infectious dose (possibly <10 cells) has heightened public health concern over its transmission and persistence in the environment. Work in strictly controlled laboratory conditions has shown that this pathogen is physiologically versatile (James and Keevil, 1999) and may survive for months in cattle faeces, manure, model soil systems (Figure 1) and river water, highlighting the potential to contaminate growing crops. Nevertheless, caution must be urged in interpretation of the data at this time as there has yet to be an investigation under field conditions of pathogen survival in a representative range of the different untreated and treated sewage or abattoir waste which may be applied to land. Preliminary work in the laboratory indicates that E. coli 0157 may be killed in sewage sludge during lime stabilisation (Maule and Keevil, unpublished data).

Similarly, C. parvum oocysts and C. jejuni may persist in aquatic environments for several months, attached to surfaces, and have been associated with foodborne outbreaks. Salmonellae cause a significant number of foodborne infections associated with fresh produce, including S. enteritidis PT4, and there is concern over the increasing disease incidence of the multiple drug resistant S. typhimurium DT104 which is found in cattle and poultry waste. L. monocytogenes is widely distributed in nature and work has shown it to be a versatile, robust pathogen capable of survival and growth at low temperatures whilst maintaining a virulent phenotype. It is undoubtedly food borne, however it cannot be truly described as a food poisoning organism since its symptoms are rarely associated with the gastrointestinal tract. Infection ranges from symptomless carriage to fatal septicaemia, meningitis or neonatal abortion. In addition to consumption of dairy products, other implicated foods include meat pates, coleslaw, poultry, fin fish and shellfish. Several large foodborne outbreaks occurred in North America in the 1980s which were notable for a case fatality rate of 30%.

In addition to domestic sources, effluents from agricultural and commercial enterprises (i.e. cattle markets, abattoirs etc.) can be responsible for releasing large numbers of Cryptosporidium sp. oocysts into sewage. The first recorded outbreak of waterborne Cryptosporidiosis was associated with contamination of a well water supply with oocysts derived from sewage. Examination of stool specimens from 79 residents who consumed water from the well source indicated that gastrointestinal illness was 12 times greater than in the surrounding communities which consumed water from a different water source. Since then, numerous waterborne outbreaks of human Cryptosporidiosis have been reported. The potential for causing disease in large numbers of individuals was demonstrated in the Milwaukee outbreak which affected an estimated 403,000 individuals. Through the use of molecular techniques, four isolates from this outbreak have been found to infect only humans, indicating that human sewage might have been the source.

Approximately 6000 cases of Cryptosporidiosis in England and Wales are reported to the PHLS each year, and this is believed to be gross underestimate of actual cases. There are more cases of Giardiasis reported, but these are mainly sporadic and the source of infection untraceable. The environmental reservoir for the aetiological agent, Giardia intestinalis (lamblia), is unclear but it has been reported to be present in beavers (hence the name "Beaver fever") and farm animals, and spread in water. Although Giardia cysts are considered less robust than Cryptosporidium oocysts, one report has suggested that 10% of infective cysts can survive several weeks of incubation in swine manure slurry.25 This reinforces the need for proper treatment of organic wastes before disposable to agricultural land.

The term enteroviruses is a general description for viruses that infect and replicate in cells of the gastrointestinal tract. A prominent family of viruses within the large group of viruses that are found in the intestinal tract are the Picornaviridae. Within this family are several genera including polioviruses, coxsackieviruses (two groups, A and B), echoviruses and enteroviruses (a specific name and as opposed to the general description above). Enteroviruses infecting humans are found world-wide and humans are the only known natural hosts. Young children are the most susceptible to infection. Transmission is usually by the faeco-oral or by the respiratory route where there is an associated respiratory illness. The viruses may be excreted in the stool for many weeks. Poliovirus in sewage is decreasing due to vaccination and the increased use of disposable nappies diverting contaminated faeces to solid waste. Enteroviruses have been detected in water, soil, vegetables and shellfish and may possibly be transmitted in the community by contact with contaminated food or water. In practice, nearly all foodborne viral illness is either hepatitis A or viral gastro-enteritis due frequently to small round structured viruses (SRSV). Food borne transmission of other viruses causing gastro-enteritis, such as rotavirus, appears rare. Current detection methods still rely on the "gold standard" of electron microscopy, but newer methods involving cell culture (although this is not possible yet with SRSV) and/or PCR, RT-PCR and antibodies are being developed. Cell culture at least allows the detection and possible quantification of viable organisms.