Hydroponics and aquaponics have been gaining popularity to grow produce. Escherichia coli O157:H7 may internalize within edible produce. This study was to investigate E. coli O157:H7 internalization in red sails lettuce in a custom designed/constructed model hydroponic system under laboratory-controlled conditions. The objective was to determine not only the likelihood and level of internalization within hydroponically-grown lettuce but also potentially facilitated internalization with injured lettuce roots. The hydroponic tank was contaminated with Escherichia coli O157:H7 (E. coi) at 5-log CFU/mL. Roots of randomly selected plants were damaged by cutting either two- or three-times (two experimental treatment groups). Experimental control did not have root damage. Following lettuce harvest and surface decontamination, E. coli was enumerated on 3-M petrifilm (TM) EC. Internalized E. coli was detected at 2.4 +/- 0.7, 4.0 +/- 1.9, and 3.3 +/- 1.3 log CFU/g in control, lettuce with two-, and three-cuts to their roots, respectively. Root injury showed a trend (p > 0.05) towards increasing E. coli internalization. Two-vs. three-cuts to roots did not (p > 0.05) affect E. coli internalization, although it showed 0.9-log CFU/g of a numerical difference. The highest level of internalized E. coli was detected at 4.7 log CFU/g in a lettuce plant that received two cuts to its root system.

Escherichia coli O 157:H7 can internalize in produce. Therefore, a food safety risk exists for lettuce grown using mixed water sources such as aquaponics, which combines aquaculture with hydroponics. Objectives were to determine antimicrobial efficacy of UV-radiation on reduction of E. coli/coliform in a pilot-scale demonstration aquaponic system and assessment for possibility of E. coli/coliform internalization in aquaponically-grown lettuce. Water from aquaculture and hydroponic was sampled for 6 weeks prior to lettuce harvest. Water and lettuce samples were spread-plated on 3-M petrifilm (TM) EC (E. coii/coliform) and on m-Endo agar. The average bacterial counts in aquaculture raceways were approximately 10(2) CFU/ml and 10(4) CFU/L on 3-M petrifilm and m-Endo agar, respectively. Bacterial counts did not (p > 0.05) increase in the aquaculture raceways or hydroponic greenhouse, suggesting that neither fish nor plants contributed to bacterial growth. The coliform bacteria detected in the aquaponic system reflected their normal presence in the environment where the aquaponic system was located. UV treatment reduced (p < 0.05) bacterial counts by approximately 1.5 and 3.0-log on 3-M petrifilm and m-Endo agar, respectively. Internalized coliforms or E. coil were not detected in lettuce. This study demonstrates the effectiveness of coliform inactivation by UV-radiation in a pilot-scale real-life aquaponic system.

Following, is a summary of research projects supported through the National Mine Land Reclamation Center up to June 1992. The Center Has three regional centers: the Eastern Center consists of West Virginia University and Penn State University, the Midwestern Center consists of Southern Illinois University and the Western Center comprises University of North Dakota and North Dakota State University. Each regional center has an advisory committee of industry and regulatory agency staff which directs the activities of the center toward regional problems of greatest impact.

The proposed health benefits, product freshness, and concern about antibiotic-resistant bacteria have bolstered sales of organic poultry products. However, outdoor requirements for organic poultry could increase exposure to bacteria, including Salmonella. The objective of this study was to assess the effects of prebiotics and probiotics in feed (experiment 1) and acidifying water treatments (experiment 2) on organic broiler performance and the presence of preslaughter Salmonella. In experiment 1, a prebiotic (MOS), 2 probiotics, and a control treatment were implemented. In experiment 2, raw apple cider vinegar, an organic acid blend, H(2)O(2), and a control treatment were incorporated into the watering systems. For both experiments, 300 one-day-old Cobb 500 male chicks were randomly assigned to treatment and pen. On d 21, birds were weighed and allocated at 13 per pen within each of the 5 housing locations providing pasture access at the West Virginia University Certified Organic Farm. Data collection occurred from d 21 to 49. On d 21, probiotic 1, the prebiotic, and the organic acid blend yielded the highest BW but demonstrated the same ending bird BW. However, birds provided probiotic 1 and the prebiotic were at least 86 and 72% less likely to be contaminated with Salmonella preslaughter. The lowest bird BW were achieved by birds provided H(2)O(2) throughout the study; however, these birds were at least 7% less likely to harbor Salmonella preslaughter compared with birds in the control treatment.

Seasonal treatment performance of small‐surface flow wetlands was evaluated during their second operational year and compared with community‐level physiological profiles (CLPP) of the heterotrophic bacterial community obtained from sole‐carbon source utilization patterns in BIOLOG GN (Haywood, California) microplates. The CLPP patterns varied significantly by season, indicating reduced functional diversity in the heterotrophic community during warmer months of active plant growth (April through October). Principal component analysis (PCA) suggested that the functional community differed in planted versus unplanted wetlands during the growing season. Wetlands generally improved wastewater quality; however, treatment performance was reduced in the second year. Despite differences in the heterotrophic community suggested by CLPP, treatment efficiency with respect to removal of five‐day biochemical oxygen demand or reduction in fecal indicator organisms generally was not significantly changed as a function of growing season or plant treatment.

In order to evaluate the efficacy of constructed wetlands for treatment of domestic wastewater for small communities located in rural areas, small-scale wetland mesocosms (400L each) containing two treatment designs (a mixture of Typha, Scirpus, and Juncus species; control without vegetation) were planted into two depths (45 or 60cm) with pea gravel. Each mesocosm received 19L/day of primary-treated domestic sewage. Mesocosms were monitored (inflow and outflow samples) on a monthly basis over a 2-year period for pH, total suspended solids (TSS), 5-day biochemical oxygen demand (BOD5), total Kjeldahl nitrogen (TKN), dissolved oxygen (DO), and conductivity. Microbiological analyses included enumeration of fecal coliforms, enterococci, Salmonella, Shigella, Yersinia, and coliphage. Significant differences between influent and effluent water quality for the vegetated wetlands (p<0.05) were observed in TSS, BOD5, and TKN. Increased DO and reduction in fecal coliform, enterococcus, Salmonella, Shigella, Yersinia, and coliphage populations also were observed in vegetated wetlands. Greatest microbial reductions were observed in the planted mesocosms compared to those lacking vegetation. Despite marked reduction of several contaminants, wetland-treated effluents did not consistently meet final discharge limits for receiving bodies of water. Removal efficiencies for bacteria and several chemical parameters were more apparent during the initial year compared to the second year of operation, suggesting concern for long-term efficiency and stability of such wetlands.

In vitro laboratory studies were performed to assess the effects of antecedent growth conditions on the recovery of Escherichia coli ATCC 25922 and Salmonella typhimurium ATCC 14028 following chloramine disinfection. Six- and 18-h cultures of each organism were grown under aerobic, fermentative, and nitrate-reducing conditions prior to disinfection. At predetermined time intervals during a 10-min exposure to chloramine, survivors were surface plated on nonselective recovery media to determine C(n)t values. It was observed that nitrate-reducing growth predisposed the test organisms towards an increased sensitivity to chloramine stress over cells grown under fermentation or aerobic conditions (p < 0.01).

Three common Appalachian plant species ( Juncus effusus L., Scirpus validus L., and Typha latifolia L.) were planted into small-scale constructed wetlands receivingprimary treated wastewater. The experimental design includedtwo wetland gravel depths (45 and 60 cm) and five plantingtreatments (each species in monoculture, an equal mixture of the three species, and controls without vegetation), with two replicates per depth × planting combination. Inflow rates (19 L day -1 ) and frequency (3 times day -1 ) were designed to simulate full-scale constructed wetlands as currently used for domestic wastewater treatmentin West Virginia. Influent wastewater and the effluent from each wetland were sampled monthly for ten physical, chemical and biological parameters, and plant demographic measurements were made. After passing through these trough wetlands, the average of all treatments showed a 70% reduction in total suspended solids (TSS) and biochemical oxygen demand (BOD), 50 to 60% reduction in nitrogen (TKN), ammonia and phosphate, anda reduction of fecal coliforms by three orders of magnitude. Depth of gravel (45 or 60 cm) had little effect on wetland treatment ability, but did influence Typha and Scirpus growth patterns. Gravel alone provided significant wastewater treatment, but vegetation further improved many treatment efficiencies. Typha significantly out-performed Juncus and Scirpus both in growth and in effluent quality improvement. There was also some evidence that the species mixture out-performed species monocultures. Typha was the superior competitor in mixtures, but a decline in Typha growth with distance from the influent pipe suggested that nutrients became limiting or toxicities may have developed.

A skin attachment model was used to determine if ZnCl2 would reverse or inhibit Salmonella attachment to broiler skin. In the reversal experiments, skin samples, treated first with 1 ml of Salmonella Typhimurium suspension (10(8) CFU/ml) for 30 min, were then treated with 25 or 50 mM ZnCl2 for 5 or 15 min. Zinc chloride solutions were applied while the culture was present on the skin. In the inhibition experiments, ZnCl2 solutions were added first; treatment solutions were discarded after 5 or 15 min of application, and then the culture was added. Firmly and loosely attached Salmonella were enumerated on xylose lactose tergitol plates. A duplicate section of skin, subjected concurrently to the above treatments, was observed under a scanning electron microscope to enumerate attached bacteria directly. In the reversal experiments, 25 and 50 mM ZnCl2 reduced (P < 0.01) firmly attached cells by 77 and 89%, respectively, when compared to the control (water). Micrographs indicated that 25 and 50 mM ZnCl2 reduced (P < 0.1) Salmonella attachment by 69 and 99.9%, respectively, in the reversal experiments. In the inhibition experiments, 25 and 50 mM ZnCl2 reduced (P < 0.01) firmly attached cells by 82 and 91%, respectively. Reduction of Salmonella may be attributed, in part, to the bactericidal activity of ZnCl2 in addition to bacterial cell detachment.

Aims: The chemical/physical environment of groundwater may contribute to the existence of a subpopulation of small-sized bacteria (filterable bacteria) that fails to be trapped on conventional 0.45 mum-pore-size membrane filters during routine bacteriological water quality analyses. Efforts were directed to determining an efficient recovery method for detection of such cells.Methods and Results: Individual groundwater supplies in a rural setting were examined by a double membrane filtration procedure to determine the presence of heterotrophic plate count (HPC) bacteria capable of escaping detection on conventional pore size (0.45 mum) membrane filters but retained on 0.22 mum-pore-size filters. Since optimum cultural conditions for recovery of filterable bacteria are not well defined, initial efforts focused on evaluation of various media (R2A, m-HPC and NWRI) and incubation temperatures (15, 20, 28 and 35 degreesC) for specific recovery of filterable bacteria. Maximum recovery of small-sized HPC bacteria occurred on low-nutrient concentration R2A agar incubated for 7 d at 28 degreesC. Similarly, identical cultural conditions gave enhanced detection of the general HPC population on 0.45 mum-pore-size filters. A 17-month survey of 10 well water supplies conducted with the cultural conditions described above resulted in detection of filterable bacteria (ranging in density from 9 to 175 cfu ml(-1)) in six of the groundwater sources. The proportion of filterable bacteria in any single sample never exceeded 10% of the total HPC population. A majority of the colonies appearing on the 0.22 mum membrane filters was pigmented (50-90%), whereas the proportion of colonies demonstrating pigmentation on the larger porosity filters failed to exceed 50% for an!: of the samples (19-49%).Conclusions: A reliable recovery method was developed for the detection of filterable bacteria from groundwater. During a subsequent survey study using this procedure, filterable bacteria were detected in a majority of the groundwater supplies examined; however, the density of filterable bacteria in any single sample never exceeded 10% of the total HPC population. Identification of randomly selected isolates obtained on the 0.22 mum filters indicated that some of these filterable bacteria have been implicated as opportunistic pathogens.Significance and Impact of the Study: We have determined the presence of small-sized HPC bacteria in ground water that may go undetected when using standard porosity membrane filters for water quality analyses. Further study is needed to assess the significance and possible health risk associated with presence of filterable bacteria in drinking water supplies from groundwater sources.

Neutralization of acid mine drainage (AMD) in man-made cattail (Typha) wetlands was mvestlgated over a four-year period utilizing experimental models (480 x 60 x 60 cm) constructed in a greenhouse.A naturally occurring AMD (430 mg/L Fe, 5 mg/L Mn, 2900 mg/L sulfate, pH 2.75) was collected in the field and added to the greenhouse wetlands at 60.5 L/day.Monthly water• samples were collected at the wetland influent and effluent at 35 cm depth.Sediments and sediment pore water samples from four depths (IO, 20, 30, and 40 cm) were obtained from the influent, midpoint, and effluent locations of the wetland.During the first year of AMD treatment, near neutral pH (6.5) and anoxic conditions (-300 mV) were observed in subsurface sediments of wetlands.The wetlands retained au estimated 65% of the total applied iron in the first year, primarily in the exchangeable, organically bound, and oxide form.During later years, 20 to 30% of the influent iron was retained predominantly as precipitated oxides.Iron sulfides resulting from sulfate reduction accounted for less than 5% of the iron retained, and were recovered primarily as monosulfides during the first year and as disulfides in the fourth year.Improvement in effluent pH was primarily attributed to limestone dissolution in the anaerobic subsurface sediments, which decreased with time.Constructed wetlands exhibit finite lives for effective AMD treatment and provisions should be made for their periodic rejuvenation or replacement.

The effect of decreased oxygen concentration during incubation of M-Endo medium on detection of coliforms from rural groundwater supplies was examined. Incubation oxygen concentrations of 0¿(anaerobic GasPak), 4, 8, and 21% (atmospheric) were examined. Our findings point to several advantages of using anaerobic incubation for the isolation of coliforms: (i) higher verification rates with concomitant decreases in occurrence of false-positive coliforms; (ii) overall reduction in growth of nonsheen colonies; and (iii) reduction in colony size for nonsheen organisms, thereby minimizing crowding effects and facilitating enumeration of coliform colonies. However, these advantages were not sufficient to permit increased recovery of total coliforms as compared with aerobic incubation. In addition, the increased frequency of detecting false-negative coliforms during anaerobic incubation is a disadvantage to this method. While detection of total coliforms was reduced under conditions of anaerobiosis,the detection of fecal coliforms and (or) E. coli was not impeded.

The water quality of 24 rural, domestic groundwater supplies treated with point-of-use, powdered activated carbon (PAC) filters was monitored to determine how such treatment might impact the bacteriological quality of private, residential drinking water supplies. Heterotrophic-plate-count (HPC) and total coliform analyses were performed on raw, PAC-treated, and overnight or stagnant (first-draw) PAC-treated water samples. Densities of HPC bacteria were elevated by 0.86 and 0.20 orders of magnitude for spring and well water systems, respectively, in PAC-treated effluents following overnight stagnation compared with levels in untreated treated effluents. Densities of HPC bacteria in PAC-treated effluents were significantly reduced (P < 0.01) below influent levels, however, after the point-of-use device was flushed for 2 min. While PAC significantly reduced the number of coliforms in product waters (P < 0.01), these indicator organisms were still detected in some effluents. Seasonal variations were evident in microbial counts from spring but not well water systems. It appears that aside from periods following stagnant-water use, such as overnight, PAC treatment does not compromise the bacteriological quality of drinking water obtained from underground sources.

More than 250 coliform and noncoliform bacteria, isolated by standard membrane filtration methods from rural, untreated groundwater supplies, were examined for resistance to 16 antibiotics. All of the noncoliforms and 87% of the coliforms were resistant to at least one antibiotic, with resistance most commonly directed toward novobiocin, cephalothin, and ampicillin. The frequency of multiple antibiotic resistance (MAR) within each species was also determined. Approximately 60% of the coliforms were MAR, including 14, 64, and 94% of Escherichia coli, Citrobacter freundii, and Enterobacter cloacae isolates, respectively. In comparison, MAR was demonstrated by more than 95% of the noncoliforms and included isolates of Acinetobacter calcoaceticus, Aeromonas hydrophila, and Serratia marcescens. Ampicillin-resistant environmental isolates were assayed for ability to transfer resistance to ampicillin-sensitive strains of E. coli and Salmonella typhimurium. Environmental E. coli isolates were capable of in vitro transfer to both recipients in a nutrient-rich environment (trypticase soy broth) at frequencies ranging from 1.1 × 10−5 to 1.0 × 10−4. Isolates of C. freundii and Klebsiella pneumoniae exhibited resistance transfer to the E. coli recipient under similar conditions at reduced rates. None of the environmental isolates demonstrated transfer of ampicillin resistance when placed in an environment containing filter-sterilized well water. The presence of antibiotic resistant bacteria, and particularly MAR bacteria, in rural groundwater supplies used as drinking water sources may have important public health implications.

The mobility and biodegradability in soil of a dilute waste oil emulsion generated by an aluminium rolling industry was investigated. Laboratory simulations and field evaluation of waste disposal suggested that the majority of the oil emulsion was retained in surface soil following application. However, potential leaching of waste to the subsurface was demonstrated, particularly at higher loading rates in soils of sandy texture. Strategies to enhance rates of biodegradation in surface soils were investigated, including fertilization and microbial inoculation. A single strain inoculum was obtained from a group of 81 isolates selected for their ability to partially mineralize the waste oil emulsion, and was tentatively characterized as a hydrocarbonoclastic Corynebacterium sp. Inoculation did not effectively stimulate waste removal in soil compared with fertilization, which significantly increased respiration and biodegradation. The maximum loss of the applied oil emulsion from soil was 30% during a 56-day in vitro incubation. Fertilized, aerated liquid waste emulsion was more rapidly degraded, resulting in loss of 65% of the waste emulsion within 18 days.

In vitro pure-culture studies were conducted to assess growth and sheen formation of groundwater bacteria on M-Endo medium incubated under reduced oxygen concentrations (0, 4, 8, 12, and 16%). Coliform and noncoliform bacteria were isolated from 17 untreated, rural groundwater supplies on M-Endo medium. All 16 coliform isolates tested were capable of sheen formation at oxygen concentrations of 4% or greater, yet some of these same isolates (Enterobacter aerogenes, Enterobacter cloacae, and Hafnia alvei) were either unable to grow or failed to produce a metallic sheen when incubated under strict anaerobiosis. Approximately 70% of the 21 noncoliform isolates examined exhibited growth inhibition at oxygen concentrations of 8% or less. The growth of a false-positive coliform isolate of Serratia fonticola was inhibited when incubated under reduced oxygen concentrations of 16% or less. Our findings suggest that the selectivity of M-Endo medium, and resultant inhibition of noncoliforms and false-positive coliforms, is enhanced by incubation in the absence of oxygen. However, the failure of strict anaerobiosis to permit detection of total coliforms such as Hafnia and Enterobacter spp. may compromise the reliability of this technique for evaluating the sanitary quality of some waters. On the other hand, oxygen concentrations of 4, 8, 12, and 16% permitted adequate sheen development of all coliforms tested while inhibiting some noncoliforms.

Four strains of Escherichia coli were examined for response to heat stress (60°C) as a function of physiological age and antecedent oxygen growth conditions. Exponential phase cells were more susceptible to heat than cells grown to the stationary phase. Anaerobically grown, exponential phase cells were more susceptible to thermal stress than were cells grown to a similar physiological state but under aerobic conditions. In the case of stationary phase cells, differences in response to heat stress as related to prior oxygen growth conditions were equivocal. Repair characteristics of thermally injured cells were also examined. Cells grown anaerobically prior to heat stress required 1.5 h longer than their aerobically grown counterparts to complete repair. These findings suggest that antecedent oxygen growth conditions influence the response of E. coli to thermal stress and perhaps, more generally, that persistence of environmentally stressed enteric microorganisms must be considered in relation to prior oxygen growth conditions in vivo.

Rural drinking water systems supplied by untreated groundwater were examined to determine whether coliform or heterotrophic plate count bacteria are capable of escaping entrapment on standard porosity (0.45-micron-pore-size) membrane filters. Filterable bacteria were present in 42% of the 24 groundwater sources examined by using nonselective media (R2A, full strength m-HPC, and 0.1x m-HPC agars). Pseudomonads were the most frequently identified group of filterable bacteria detected. Flavobacterium, Alcaligenes, Acinetobacter, and Achromobacter isolates were also identified. Total coliforms were not recovered from any of the 24 groundwater samples following filtration through 0.45-micron-pore-size membrane filters by using selective M-Endo LES agar or mT7 agar. In addition, none of the isolates identified from nonselective media were coliforms. Similarly, neither total coliforms nor specifically Escherichia coli were detected in these filtrates when Colilert P/A medium was used.