Pergamon PII: S0043-1354(96)00193-5 Wat. Res. Vol. 30, No. 11, pp. 2812-2816, 1996 Copyright © 1996ElsevierScienceLtd Printed in Great Britain.All fights reserved 0043-1354/96$15.00 + 0.00 USEFULNESS OF BACTERIOPHAGES INFECTING BACTEROIDES FRAGILIS AS INDEX MICROORGANISMS OF REMOTE FAECAL POLLUTION F. LUCENA, R. ARAUJO @ and J. JOFRE @* Department of Microbiology, University of Barcelona, Avenida Diagonal 645, 08028 Barcelona, Spain (First received August 1994; accepted in revised form June 1996) Abstract--Levels of faecal bacteria, somatic coliphages and bacteriophages of Bacteroides fragilis were counted in marine sediments and groundwater as representatives of environments with remote faecal pollution. Additionally, enterovirus were measured in the marine sediments. Clear changes of the proportions of the different microorganisms are observed when microbial numbers in groundwater and marine sediments are compared to the numbers found in sewage or water recently polluted by sewage. These changes show that phages of B. fragilis reach sites that receive persistent pollution more successfully than somatic coliphages or indicator bacteria, with the exception of Clostridium perfringens. Multifactorial analysis of marine sediments and groundwater data clustered phages, viruses and C. perfringens in one factor and faecal bacteria in another. These data clearly suggest that Clostridium, phages of B. fragilis and, in some cases enteroviruses, are better index microorganisms of remote faecal pollution than present bacterial indicators. Copyright © 1996 Elsevier Science Ltd Key words--bacteriophages, enterovirus, bacteria, Bacteroides fragilis, coliphages, indicators, faecal, groundwater, sediments INTRODUCTION The sanitary quality of water is currently judged on the basis of indicator bacteria. However these bacteria may not reflect the quality of water regarding other pathogens such as viruses and parasites (Payment et al., 1991; Casemore, 1991). This lack of correlation is especially significant in environmental samples in which contaminant microorganisms had suffered different inactivating processes, as for example treated waters (Berg et al., 1978), ground- water (Marzouk et al., 1980; Walter et al., 1982; Vaughn et al., 1978), marine water (Lucena et al., 1982) and shellfish (Goyal et al., 1979; Wait et al., 1983). As a consequence, groundwater (Craun, 1979, 1991) and shellfish (Rippey, 1991) account for a very large fraction of the documented waterborne diseases, of which a very important fraction are of viral origin (Keswick and Gerba, 1980; Rippey, 1991). Consequently, the need for good surrogate indicators of viral pollution is widely recognised (IAWPRC Study Group, 1991; Pancorbo and Barnhart, 1992). Some bacteriophages infecting enteric bacteria are being studied as potential indicators of human viruses in the environment (Tartera et al., 1987; Tartera et al., 1989; IAWPRC Study Group, 1991). We now present data on microbial loads in marine *Author to whom all correspondence should be addressed [Tel.: 3 93 402 14 83]. sediments and groundwater. In these environments surviving faecal contaminants have undergone a number of processes that reflect the different factors that determine the occurrence of these organisms in environments with remote faecal pollution. However, in both cases we have selected rather polluted sampling sites in order to obtain useful data. The microorganisms studied are bacterial indicators including C. perfringens and Aeromonas, somatic coliphages, phages infecting B. fragilis and en- teroviruses in some samples. Enteroviruses were not counted in those samples that required a concen- tration step. MATERIALS AND METHODS Sampling sites Groundwater samples were collected at different sites from a free alluvial aquifer belonging to the basin of Besos river. Sampling stations were distributed over an area of 8 km2. The main influx of faecal pollution into the aquifer is through infiltrated water from the Besos river, which is heavily loaded with human faecal pollution. Mean microbial loads per 100 ml of Besos fiver water in the area where this study was performed were: total coliforms, 3.2 × 10% faecal coliforms, 9.1 × 105; faecal streptococci, 1.9 x 105; C. perfringens, 1.2 × 104; somatic coliphages, 1.7 × 105; and bacteriophages infecting B. fragilis HSP40, 1.2 x 103. Marine sediments were collected at distances of 300 m to 12 km from the shoreline and at water depths ranging from 6.5 m to 60 m in an area of around 25 km2 which was highly influenced by different sources of pollution such as the Besos river, the Llobregat river, treated domestic wastewater, some minor sewage outfaUs and dumping of sewage sludge. 2812