Dynamic Existence of Waterborne Pathogens ... - ACS Publications

Feb 10, 2009 - Department of Chemistry & Biology, Ryerson University,. Toronto, ON, M5B 2K3 Canada. Received August 26, 2008. Revised manuscript ...
1 downloads 0 Views 3MB Size
Environ. Sci. Technol. 2009, 43, 1737–1743

Dynamic Existence of Waterborne Pathogens within River Sediment Compartments. Implications for Water Quality Regulatory Affairs I A N G . D R O P P O , * ,†,§ S T E V E N N . L I S S , ‡,§ DECLAN WILLIAMS,§ TARA NELSON,† CHRIS JASKOT,† AND BRIAN TRAPP† Environment Canada, P.O. Box 5050, Burlington, ON, L7R 4A6 Canada, Department of Environmental Biology, University of Guelph, Guelph, ON, N1G 2W1 Canada, and Department of Chemistry & Biology, Ryerson University, Toronto, ON, M5B 2K3 Canada

Received August 26, 2008. Revised manuscript received December 16, 2008. Accepted December 23, 2008.

The transport and fate of indicator E. coli and Salmonella are shown to be highly influenced by their relationship with flocculated suspended and bed sediment particles. Flocs were found to dominate the suspended sediment load and have the effect of increasing the downward flux of the sediment to the river bed. Bacteria counts were consistently higher within sediment compartments (suspended and bed) than for the water alone, with the bed sediment found to represent a possible reservoir of pathogens for subsequent remobilization and transport to potentially high risk areas. The mechanism of microbial attachment and entrapment within the sediment was strongly linked to the EPS fibrils secreted by the biological consortium of the aquatic system. It is suggested that the sediment/pathogen relationship should be of concern to public health officials because of its potential effects on pathogen source fate and effect with implications on public health risk assessment. Current standard sampling strategies, however, are based on an assumption that bacteria are entirely planktonic and do not account for the potentially significant concentration of bacteria from the sediment compartments. The lack of understanding around pathogen/sediment associations may lead to an inaccurate estimate of public health risk, and, as such, possible modification of sampling strategies to reflect this association may be warranted.

Introduction Drinking water warnings and beach closures are frequently reported in the news and are reactive measures to unsafe levels of indictor organisms (usually E. coli) particularly during summer and wet weather months. Pathogens (including opportunistic pathogens) are introduced into the aquatic environment typically by surface and subsurface runoff, bird/ wildlife excrement, septic systems, and urban runoff including combined storm and sanitary sewer overflows (1, 2). In an epidemiological study, Schutze et al. (3) found that the sources of Salmonella infections in children were more likely * Corresponding author e-mail: [email protected]. † Environment Canada. § Ryerson University. ‡ University of Guelph. 10.1021/es802321w CCC: $40.75

Published on Web 02/10/2009

 2009 American Chemical Society

a result of environmental contamination than from food sources. While water-borne pathogenic organisms can represent a significant health risk if exposure is above an infectious dose, the degree of risk is uncertain with the knowledge that pathogens can be associated with the sediment compartments (suspended and bed sediment) of aquatic systems (4-6). This risk uncertainty is exacerbated by current microbial monitoring policies which do not consider the phases of pathogen association (i.e., free floating and sediment associated). Furthermore, there is a general lack of understanding of the nature and relevance of pathogen/sediment associations to human infection, pathogen viability, transport, and fate within aquatic systems. Bacteria often show an affinity for sediment attachment (4, 6, 7), as it represents a beneficial environment for nutrient/ food (dissolved organic carbon, DOC) assimilation and protection from environmental stress such as contaminants and predation (8). Within the water column, this association is complicated by the knowledge that sediment exists in the form of flocculated particles (9). These are conglomerate particles made up of inorganic (e.g., clays) and bioorganic (e.g., detritus, bacteria, fungus, etc.) particles, water, and pores. The association of various organisms (e.g., bacteria, algae, fungi) with flocs dictates their transport behavior, as flocs are significantly different in their hydrodynamic behavior from their constituent particles (including the organisms) because of a difference in size, shape, porosity, density, and composition (9). Flocculation increases the downward flux of sediments and, by association, also increases the downward migration of pathogens within the water column (6, 10, 11). The incorporation of pathogens within the surficial bed sediment (SBS) is actively facilitated by the existing biological population’s (bacteria, alga, diatoms, etc.) secretion of extracellular polymeric substances (EPS) during the production of a Biofilm (12). Individual EPS fibrils are colloidal in nature (2-20 nm in diameter (13)) and are the structural matrix of biofilms (13, 14). Biofilm formation on/within the SBS has been shown to increase the stability of the sediment (i.e., resistance to erosion) (15, 16). Erosion of the surface active layer and further transport of bacteria will occur, however, once the critical bed shear stress for erosion is surpassed and even with simple decay and sloughing of biofilms (11, 16). The sediment/bacteria association is further complicated by the potentially transient nature of the floc/ bacteria relationship itself due to the dynamic nature of the flocculation process. Flocs are continuously undergoing a breaking and building process resulting in the sediment bacteria association being in a continual state of flux. As such, the strength of the floc will dictate the potential for pathogen dissociation with attendant increased human health risk. Further, flocculation can have an influence on the numerical assessment of indicator bacteria used for management decisions. Traditional plate counts are often used as the standard method to assess indicator bacteria counts. This method makes the assumption that bacteria are primarily planktonic in nature and ignore the significant number of bacteria that may exist within suspended sediment (SS) flocs (17). As such, numeration of indicator bacteria without including those incorporated within flocs may result in an underestimation of counts with potential erroneous management decisions and significant health risk implications. Demonstrating that pathogenic bacteria should be investigated in relation to the SS and SBS compartments of rivers would contribute to a better understanding and VOL. 43, NO. 6, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1737

evaluation of the source, fate, and effect of these organisms. To that end, this paper focuses on the following objectives: (1) examine the association of indicator bacteria with SS floc and SBS, (2) examine the structural characteristics of flocs with associated indicator bacteria, (3) assess the degree to which this association may influence pathogen transport and fate, and (4) discuss the policy implications of such association for safe water practices.

Materials and Methods Study Site. Samples of water, bulk SS, and SBS were collected from the South Nation River (SNR) near Ottawa, Ontario, Canada (drainage area 3700 km2; see Figure S1, Supporting Information). A 10 km reach of the main stem of the SNR was investigated from St. Albert to Cassleman including two tributaries: Butternut Creek and Little Castor River. The river transports primarily fine-grained cohesive sediment with a d50 of 10 µm and has a bed composition ranging from gravel to cohesive glacial marine clay. Further details on the geology and soils of the SNR can be found in Chapman and Putnam (18). Water and Sediment Particle Sampling for Floc Characterization and Microbial Assessment. Plankton chambers (19) were used to collect suspended floc samples for morphological and size analysis using conventional optical microscopy. Whole water samples (i.e., water with ambient SS included) for microbial assessment were collected by dipping 500 mL autoclaved polyethylene bottles into the centroid of flow. SBS was collected for bulk grain size analysis (Sedigraph) and surficial microbial assessment with a PONAR bed sediment sampler. Only the top few millimeters of sediment was scraped of the surface of the grab and placed in sterile polyethylene containers. SBS was also manually resuspended from the bed by gently agitating the water above the sediment-water interface and collecting the suspension within 4 L polyethylene jugs (used to determine floc settling velocity, density, and porosity of the recently deposited sediment). Bulk SS was collected using a continuous flow centrifuge (Westfalia Model KA 2-06-075) for microbial assessment [water (>2000 L) was pumped into a stainless steel bowl with a rotational speed of 9470 rpm where >90% of the SS is recovered and placed in autoclaved polyethylene containers (the total volume of SS collected is dependent on the SS concentration of the river at the time of sampling)]. All samples were kept on ice until analysis (all microbial analysis was performed within 24 h of sampling). Conventional Optical Microscopy (COM). Flocs collected/settled within the plankton chambers were imaged and sized using a Zeiss Axiovert 100 (Carl Zeiss Inc., Toronto, Canada) microscope interfaced with a digital camera (Hamamatsu; Quorum Technologies Inc., Guelph, Canada) and Open Lab image analysis system (Improvision, Coventry, UK). This method allows for the nondestructive direct sampling and observation/measurement of flocculated material (19). Floc samples were also sonicated (50 W; 20 kHz for 2 min) to break up the composite particles (no deflocculating chemicals added) and resized with optical validation of primary particles. Particle structure is imaged down to a lower resolution of approximately 2 µm (10× objective). Resuspended SBS floc settling velocity was determined by allowing drops of sediment from a wide-mouth pipet to settle through an insulated 2.5 L capacity settling column. Settling flocs pass through the field of view of a stereoscopic microscope (Nikon SMZ-2T - Nikon Canada Inc., Mississauga, Canada) interfaced with the digital camera and imaged using the Open Lab system. Settling velocity was derived by digitally overlaying two captured frames containing identified flocs separated by a known time interval. Further, the density and porosity of the floc was determined using Stokes equation and a mass balance of densities respectively (19). 1738

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 6, 2009

Confocal Laser Scanning Microscopy (CSLM). CLSM and fluorescence in situ hybridization (FISH) was employed to view the association of bacteria within the flocculated material. Molecular probes were used to visualize Eubacteria [EUB338I, labeled with the fluorophore BODIPY 493/510 (Sigma Genosys, The Woodlands, TX) and E. coli [ECOII labeled with CY3 (Sigma Genosys). The DNA sequences of these probes were obtained from probeBase (20). Hybridization of the probes to the target bacteria was performed using the method of Amann et al. (21). Fluorescence-labeled lectins were used to visualize carbohydrates associated with the sediment particles. Sediment samples were treated with tetramethylrhodamine-labeled wheat germ agglutinin, Alexafluor 633-labeled concanavalin A, and Alexafluor 488-labeled soy bean agglutinin. These lectins bind N-acetylglucosaminyl residues, R mannopyranosyl and R glucopyranosyl residues, and to oligosaccharide structure with terminal R- or β-linked N-acetylgalactosaminyl, and, to a lesser extent, galactopyranosyl residues, respectively. All probes were from Molecular Probes, Eugene, OR. Transmission Electron Microscopy (TEM). More detailed ultrastructural observations of sediment samples were made by preparing the samples for TEM following the 4-fold multipreparatory technique (7). This technique allows for the enhanced observation of specific components of the floc such as cells and polymeric material. Ultrathin sections were imaged in transmission mode (TEM) at an accelerated voltage of 80 kV using a JEOL 1200 Ex II TEMSCAN scanning transmission electron microscope. Enumeration of Live/Dead Bacteria. The microbial population (bulk average cell counts) of bed and bulk suspended sediment samples were assessed using the LIVE/ DEAD BacLight nucleic acid staining technique. Depending on the suspended solid concentration, 0.5 mL to 5 mL of raw sample was filtered through a 0.2 µm black polycarbonate filter. One milliliter of 0.085% NaCl solution was placed on the filtered sample, and 30 µL of a 10% dilution of LIVE/ DEAD BacLight staining solution (in 0.085% NaCl) was added to the mixture. The apparatus was incubated for 15 min in the dark, and the dye mixture was filtered. Filters were mounted on a slide with BacLight mounting oil and examined using a Zeiss Axiovert 100 fluorescent microscope equipped with a 520 nm barrier and 470-490 nm excitation filter. Enumeration of E. coli and Salmonella spp. E. coli and Salmonella spp. within water were isolated from 100 and 500 mL of surface water, respectively, by initially filtering the samples through a sterile 0.45 µm membrane. Membranes were placed face up on MacConkey and bismuth sulfite agar (Becton Dickinson and Company, Sparks, MD) for E. coli and Salmonella spp. and incubated for 12 and 48 h, respectively, at 37 °C (22). For sediment samples (100 µL; both SBS and SS), a dilution series of 10-1, 10-2, 10-3, and 10-4 in 1 × PBS was prepared. Each dilution (100 µL aliquot) was spread plated onto MacConkey and bismuth sulfite agar, respectively. All plating was done in triplicate. Presumptive identification of E. coli was based on the number of lactose-fermenting colonies (colonies appearing pink on MacConkey agar). Three representative colonies were selected from each plate, and these were used to inoculate tubes of EC medium containing 4-methylumbelliferyl-β-Dglucuronide (MUG) (Becton, Dickinson and Company, Sparks, MD). Replicate sets of tubes were incubated at 37 and 44 °C for 24 h and examined for gas production and fluorescence under UV light using a LKB 2011 Macrovue transilluminator (LKB, Bromma, Sweden). Colonies that induced fluorescence in this medium were identified as E. coli. For Salmonella, all colonies were counted following 48 h incubation on bismuth sulfite agar. Three representative colonies from each plate were selected and subsequently inoculated onto Oxoid Salmonella Chromogenic Agar (Oxoid,

FIGURE 1. (a) Representative micrograph of floc particles and (b) grain size distributions, based on equivalent spherical volume, for both the flocculated and primary (sonicated) particles. Nepean, ON, Canada) in triplicate and incubated at 37 °C for 18 h. Purple colonies were presumptively identified as Salmonella. EPS Characterization. EPS were extracted from sediment samples using the cation exchange method (23) and the EPS constituents quantified. The neutral hexose and acidic sugar content of polysaccharides were measured using the anthrone method (24) and the m-hydoxydiphenyl sulfuric acid method (25), respectively. Protein and humic acids were measured using the Lowry method (26). A Spectronic 20 (Thermo Electron Corporation, Madison WI) was used to measure absorbance of the samples for all of these colorimetric methods. DNA was quantified using a BioRad Fluorescent DNA Quantitation Kit (BioRad Laboratories, Hercules, CA) and a Shimadzu RF-Mini 150 Recording Fluorometer (Shimadzu, Columbia, MD).

Results and Discussion Floc Gross Scale Characteristics. Figure 1a illustrates the typical structure of flocs from the SNR. While generally small (d50 of all distributions ranged from 11 to 63 µm) relative to other rivers (27), the flocs exhibit the porous matrix which is common among all rivers transporting cohesive sediments. Such a matrix allows for the exchange of water, nutrients, microbes, and microbial byproduct within the floc. The small size is likely related to the very slow moving water (