SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Aims: Survival of Escherichia coli and enterococci was evaluated in bovine manure incorporated into two Wisconsin soils.

Methods and Results: Silty clay loam (SCL) and loamy sand (LS) were mixed with fresh bovine manure, exposed daily to 10 h at 22°C/14 h at 9°C, and watered weekly for 12 weeks. Escherichia coli numbers increased 1–2 log cfu g–1, then decreased < 1 and about 2 log cfu g–1 in SCL and LS, respectively. Enterococci numbers rose less and then declined faster than those of E. coli. Watering intervals of 3, 7 and 14 days were evaluated in weeks 13–19, but did not affect the slow decline in numbers of E. coli or enterococci.

Conclusions:Escherichia coli and enterococci may survive at least 19 weeks at 9–21°C in bovine manure/soil, with E. coli surviving better.

Significance and Impact of the Study: Quantification of E. coli or enterococci in late spring/early summer soil may be useful in indicating recent application of bovine manure.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Land-spreading of bovine manure is a common fertilization practice for Wisconsin soil, much of which is used for vegetable production. As bovine manure is known sporadically to contain human pathogenic bacteria, such as Escherichia coli O157:H7 and Salmonella spp. (Wells et al. 1991; Zhao et al. 1995; Pell 1997), vegetable growers may choose to test the soil after manure incorporation to determine whether planting or harvesting of vegetables is safe. The faecal indicator bacteria, E. coli and Enterococcus spp., may be appropriate test organisms because they are common in bovine manure (Jay 1992; Calicioglu et al. 1999) and may persist for several weeks in soil (Van Donsel et al. 1967; Chandler and Craven 1978, 1980; Recorbet et al. 1992; Stoddard et al. 1998). The present study was undertaken to: evaluate the survival of E. coli and enterococci in bovine manure incorporated in soil; determine the effects of soil type and watering interval on survival of these indicator bacteria in manure/soil mixtures; determine whether analyses for either indicator group could be recommended for determining whether bovine manure has been recently applied to soil.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Incorporation of manure and soil; storage temperatures

Fresh loamy sand (LS) and silty clay loam (SCL) soils were collected from the University of Wisconsin-Madison, Agricultural Research Stations at Hancock and West Madison, WI, respectively. The LS was 84% sand, 4% silt, 12% clay and pH 7·5; the SCL was 2% sand, 70% silt, 28% clay and pH 7·2, as determined at the University of Wisconsin-Madison Soil and Plant Analysis Laboratory (SPAL). Fresh manure was obtained from six lactating Holstein and Jersey cows fed a standard corn- and soybean-based ration at the UW-Madison Dairy Research Center. Analysis of a representative manure sample at SPAL showed that the manure contained 83·4% (w/w) moisture, 0·56% N, 0·24% P2O5 and 0·17% K2O. In the laboratory, 400 g LS or SCL were mixed with 17 g of cow manure in a sterile plastic tub (17·8 × 17·8 × 18·4 cm; Servin Saver, Nasco, Fort Atkinson, WI, USA), with three tubs prepared for each soil type. These proportions of soil and manure represent application of 67·2 t ha–1 wet manure distributed in a 15 cm depth. Initially, 50 ml deionized water were added to each tub containing manure and LS, while 100 ml deionized water were added to each tub containing manure and SCL. These watering treatments were previously determined to exceed the saturation capacity of the respective soils, at typical field moisture levels. The two soil types differ markedly in water-holding capacity, with 100 g oven-dried LS and SCL absorbing an average of 35 and 64 ml water, respectively. Each of the six tubs of watered soil/manure was loosely covered with aluminium foil and the weight of each covered tub was measured. Each day, these tubs were placed in a laminar flow hood, with ambient fluorescent lighting (in the room outside the hood) and temperature at 22 ± 1°C, for 10 h, followed by storing in an unlit low temperature incubator at 9 ± 2°C for 14 h. This cycle was repeated for 19 weeks.

Watering the soil samples

Part 1: (Weeks 0–12).

The soil/manure mixture in each tub was watered weekly with deionized water to reach the day 0 wetted weight, minus the weight of collected samples. Thus, the water content of each soil fluctuated between less than saturation and greater than saturation. After water addition, the tubs were shaken uniformly in a circular motion to mix the soil and water thoroughly.

Part 2: (Weeks 13–19).

Each of three tubs of SCL was randomly assigned to be watered twice a week (denoted as a 3 day watering interval), once a week, or bi-weekly. The same assignments were made for the three tubs of LS. At the appropriate time interval, each tub was watered to the day 0 weight, minus the weight of previously collected samples.

Microbiological analyses

Part 1 (Weeks 0–12).

On day 0, manure and both soils were analysed separately, and after mixing. Every week thereafter, a sample from each of the six tubs of soil/manure was quantitatively analysed for presumptive enterococci and E. coli, and qualitatively analysed for E. coli. Without mixing the soil, the sample was collected by inserting a sterile spatula straight downwards into the soil and removing a 2·0–2·5 g core, which was transferred to a sterile blender jar. The weight of each sample was recorded; 99 ml Butterfield’s phosphate diluent (NutraMax) were added, the contents mixed for 30 s, allowed to rest for 30 s, and mixed for an additional 30 s. Further decimal dilutions were made using the same diluent, and 1 ml of the blended mixture was pipetted to a single PetrifilmTME. coli/Coliform count plate (3M Microbiology Products, St Paul, MN, USA) for each dilution. For each appropriate dilution, another 1 ml of mixture was used for pour-plating with kanamycin esculin azide (KEA) agar (Oxoid). To test qualitatively for presumptive E. coli, 0·1 ml of a 10–2 dilution was inoculated in 9 ml lauryl sulphate tryptone (LST) (Difco) broth, containing 3·6 mg 5-bromo-4-chloro-3-indolyl β-D-glucuronide (X-GlcA; Sigma), and an inverted Durham tube. All PetrifilmTME. coli/Coliform count plates, KEA agar plates and LST broth + X-GlcA tubes were incubated at 35°C for 24 h (presumptive E. coli) or 48 h (presumptive enterococci).

The number of blue colonies with associated gas on the Petrifilm plates, and colonies surrounded by brown colour on KEA agar, were counted as presumptive E. coli and enterococci, respectively. The presence of presumptive E. coli in LST broth + X-GlcA was indicated by blue coloration and gas formation in the Durham tube.

To confirm presumptive E. coli in each soil type, a loopful of broth from one positive LST broth + X-GlcA tube and two presumptive colonies from PetrifilmTME. coli/Coliform count plates were streaked on Levine’s eosin methylene blue agar (L-EMB; Difco) and incubated at 35°C for 24 h. After 24 h, dark colonies with a shiny metallic sheen on L-EMB agar were selected and streaked on nutrient agar (NA; Difco), and incubated at 35°C for 24 h. The presumptive E. coli isolates were further analysed by Gram stain, cell morphology, oxidase reaction (Becton Dickinson Microbiology System, Sparks, MD, USA) and API 20E biochemical characterization (BioMerieux, Hazelwood, MO, USA).

Two presumptive enterococci colonies were randomly selected and streaked on NA with added yeast extract (4·0 g; Difco), proteose peptone no. 3 (5·0 g; Difco) and glucose (20·0 g; Sigma). This medium, denoted NA + YPG, gave superior growth of presumptive enterococci isolates compared with NA. The streaked NA + YPG plates were incubated anaerobically using a BBL GasPak System with GasPak Plus gas generation kit (Becton Dickinson) at 35°C for 48 h. At the end of the incubation period, confirmed enterococci were defined as Gram-positive, catalase-negative cocci having pyrrolidonyl arylamidase activity (BBL DrySlide PYR kit; Becton Dickinson).

Part 2: (Weeks 13–19).

The method of sampling was changed in Part 2 because of poor durability of the blender jars. A stomacher bag containing a filter (Nasco Whirl-Pak, Atkinson, WI, USA; Seward Stomacher 400 Filter bag, Fisher Scientific, Itasca, IL, USA; or Filtra-bag, Tekmar-Dohrmann, Cincinnati, OH, USA) was used to mix each sample, instead of a blender jar. The sample and diluent were manually shaken for 30 s, allowed to rest for 30 s and then shaken again for 30 s. A comparison of the sampling methods was made using paired samples from three additional tubs of manure/soil per soil type after initial watering and 3 day storage. Results showed that although the stomacher bag method recovered more cells in all but one sample, there was no statistically significant (P < 0·05) difference between the two homogenization methods, as determined by a paired t-test (Minitab for Windows, Release 12; Minitab, Inc., State College, PA, USA). Interestingly, sample-to-sample variation was much greater for the blender jar method than for the stomacher bag method (standard deviations of 1·1–2·1 and 0·3–0·4 log cfu g–1, respectively).

Enumerations of presumptive E. coli and enterococci for each of three samples per tub were determined as in Part 1. For each tub of soil/manure, one randomly-chosen sample was qualitatively tested for the presence of presumptive E. coli using the LST broth + X-GlcA method. One colony each of presumptive E. coli and enterococci was picked for confirmation testing as in Part 1.

Statistical analyses

Values of log cfu g–1 were calculated for each sample. Using simple linear regression for data from a given sample container, the rate of decline in log cfu g–1, after reaching maximum log cfu g–1, was estimated. For Part 1 data, a two-sample t-test was used to compare the two indicator groups in terms of the rate of decline in log cfu g–1 for a given soil. Similarly, soil types were compared in terms of rate of log cfu g–1 decline for a given indicator group. To perform analysis of variance, the General Linear Model (GLM; Minitab version 12·21, State College, PA, USA) was used to determine whether indicator group, soil type or an indicator–soil interaction affected the rate of log cfu g–1 decline. It was assumed that the two groups of indicator bacteria were statistically and biologically independent. For Part 2 data, the difference between the two indicator groups, in terms of rate of log cfu g–1 decline, was calculated for each tub. These difference values were then analysed using a two-way analysis of variance (SAS version 8, SAS Institute, Cary, NC, USA) to determine whether difference values were affected by soil types or watering interval. Analysis of variance was carried out as for Part 1 data to test for effects of soil types or watering interval on rate of log cfu g–1 decline for a given indicator group.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

All presumptive E. coli isolates from Petrifilm plates were confirmed as E. coli. Confirmed E. coli isolates were obtained from each of the positive LST + X-GlcA tests. Of presumptive enterococci isolates, 79% were confirmed as Enterococcus spp. Numbers of presumptive E. coli increased by about 1·5–2 log cfu g–1 during the first 2 weeks, regardless of soil type. Numbers of presumptive enterococci increased by about 1 log cfu g–1 in LS but did not show an initial increase in SCL. Thereafter, numbers of presumptive enterococci fell to levels about 1·5 log cfu g–1 lower than on day 0. Presumptive E. coli numbers did not decrease to day 0 levels during Part 1. Numbers of presumptive enterococci declined at a significantly (P < 0·05) greater rate than numbers of presumptive E. coli in SCL, but not in LS (Figs 1 and 2, Table 1a and b). The soil type and group of indicator bacteria both significantly affected rate of log cfu g–1 decline, but no significant soil type × indicator group interaction occurred (Table 1e). For both indicator bacterial groups, the rate of log cfu g–1 decline was significantly greater in LS than in SCL (Tables 1c and d).

image

Figure 1.  Survival of (▮) Escherichia coli and (●) enterococci in bovine manure incorporated into loamy sand during 12 weeks of storage under a temperature regime of 10 h at 22 ± 1°C and 14 h at 9 ± 2°C. Points indicate mean (n=3) log cfu g–1 and error bars indicate ± 1 S.D. Sample analysis was not done in week 10

Download figure to PowerPoint

image

Figure 2.  Survival of (▮) Escherichia coli and (●) enterococci in bovine manure incorporated into silty clay loam during 12 weeks of storage under a temperature regime of 10 h at 22 ± 1°C and 14 h at 9 ± 2°C. Points indicate mean (n=3) log cfu g–1 and error bars indicate ± 1 S.D. Sample analysis was not done in week 10

Download figure to PowerPoint

Table 1.   Two-sample t-tests (a–d) and analysis of variance (e) for survival of Escherichia coli and enterococci in bovine manure incorporated into loamy sand (LS) or silty clay loam (SCL) stored for 12 weeks under a temperature regime of 10 h at 22 ± 1 °C and 14 h at 9 ± 2 °C. A P-value <0·05 indicates a statistically significant difference or effect Thumbnail image of

All isolates from PetrifilmTM plates and LST + X-GlcA tubes were confirmed as E. coli, and 81% of the presumptive enterococci isolates were confirmed. The two E. coli analysis methods were clearly more specific. Generally, small decreases in numbers of E. coli and presumptive enterococci occurred in Part 2 (≤ 1 log cfu g–1). Overall, neither of the soil types nor two of the three watering intervals significantly affected the difference in rates of log cfu g–1 decline between E. coli and entercocci (Table 2). However, enterococci decreased in numbers significantly more rapidly than E. coli when a 7 day watering interval was used. When data for each indicator group were analysed separately, neither soil type nor watering interval had a significant effect on rate of log cfu g–1 decline.

Table 2.   Analysis of variance to determine whether differences in rate of log cfu g−1 decline between Escherichia coli and entercocci are affected by soil type or watering interval during weeks 13–19 of storage Thumbnail image of

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

The temperature regime used in this study is comparable with average late spring/early summer Wisconsin temperatures. Under this regime, E. coli survived better than the enterococci. These results are consistent with the finding of Van Donsel et al. (1967), that E. coli survived better during June–August than Enterococcus faecalis when applied to two Ohio soils. Our results differ from the findings of Stoddard et al. (1998) that faecal coliforms did not survive as well as faecal streptococci in manure applied to soil. We found soil type to significantly affect indicator bacteria survival. Zibilske and Weaver (1978) also reported that soil type affected faecal bacterial survival, although numerous other factors may also have been significant. The initial increase in E. coli numbers in Part 1 is similar to findings in other studies (Gudding 1975; Chandler and Craven 1980), but the rate of E. coli die-off was much slower than that reported elsewhere (Tannock and Smith 1972; Gudding 1975; Chandler and Craven 1978; Turpin et al. 1993).

Part 2 results did not support the hypothesis that watering interval would significantly affect the survival of E. coli and enterococci. Other studies have shown soil moisture to affect survival of faecal bacteria (Zibilske and Weaver 1978; Chandler and Craven 1980), but soil type, inoculum cell conditions, exposure of inoculated soil to sunlight (Tannock and Smith 1972), protozoan predation (Recorbet et al. 1992), frost (Tannock and Smith 1972) and temperature (Guy and Small 1977; Zibilske and Weaver 1978) may also be important.

Both the PetrifilmTM plate and LST + X-GlcA broth methods detected E. coli. In situations where qualitative analysis for this indicator organism is sufficient, the latter method is simple and easily interpreted. Obtaining a composite sample from several locations in the field would be advised, based on the sample-to-sample variation observed in this study.

Other studies have shown that survival of Salmonella typhimurium in soil parallels that of E. coli (Chandler and Craven 1978,1980), and that E. coli O157:H7 and Salm. typhimurium die at similar rates in bovine manure and manure slurry (Himathongkham et al. 1999). The results of this study suggest that quantitative analyses of Wisconsin soils in late spring/early summer for either E. coli or enterococci may be useful in determining the likelihood of recent bovine manure application and, thus, the potential risk of enteric pathogens being present.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

The authors gratefully acknowledge the assistance of Landon Sego in statistical analyses. This research was financially supported by a grant from the United States Department of Agriculture Special Food Safety Program.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References
  • 1
    Calicioglu, M., Buege, D.R., Ingham, S.C., Luchansky, J.B. (1999) Recovery of Escherichia coli Biotype I and Enterococcus spp. during refrigerated storage of beef carcasses inoculated with a fecal slurry. Journal of Food Protection 62, 944947.
  • 2
    Chandler, D.S. & Craven, J.A. (1978) Environmental factors affecting Escherichia coli and Salmonella typhimurium numbers on land used for effluent disposal. Australian Journal of Agricultural Research 29, 577585.
  • 3
    Chandler, D.S. & Craven, J.A. (1980) Relationship of soil moisture to survival of Escherichia coli and Salmonella typhimurium in soils. Australian Journal of Agricultural Research 31, 547555.
  • 4
    Gudding, R. (1975) The persistence of Salmonella typhimurium in various types of manure with and without admixture of silage effluent. Acta Veterinaria Scandinavica 16, 115125.
  • 5
    Guy, E.M. & Small, J.A. (1977) Survival of streptococci and coliforms of bovine faecal origin in drainage water and soil stored at different temperatures. New Zealand Journal of Agricultural Research 20, 1318.
  • 6
    Himathongkham, S., Bahari, S., Riemann, H., Cliver, D. (1999) Survival of Escherichia coli O157:H7 and Salmonella typhimurium in cow manure and cow manure slurry. FEMS Microbiology Letters 178, 251257.DOI: 10.1016/s0378-1097(99)00364-x
  • 7
    Jay, M. (1992) Modern Food Microbiology, 4th edn. New York: Chapman & Hall.
  • 8
    Pell, A.N. (1997) Manure and microbes: public and animal health problem? Journal of Dairy Science 80, 26732681.
  • 9
    Recorbet, G., Steinberg, C., Faurie, G. (1992) Survival in soil of genetically engineered Escherichia coli as related to inoculum density, predation and competition. FEMS Microbiology Ecology 101, 251260.
  • 10
    Stoddard, C.S., Coyne, M.S., Grove, J.H. (1998) Fecal bacteria survival and infiltration through a shallow agricultural soil: timing and tillage effects. Journal of Environmental Quality 27, 15161523.
  • 11
    Tannock, G.W. & Smith, J.M.B. (1972) Studies on the survival of Salmonella typhimurium and Salmonella bovis-morbificans on soil and sheep faeces. Research in Veterinary Science 13, 150153.
  • 12
    Turpin, P.E., Maycroft, K.A., Rowlands, C.L., Wellington, E.M.H. (1993) Viable but non-culturable salmonellas in soil. Journal of Applied Bacteriology 74, 421427.
  • 13
    Van Donsel, D.J., Geldreich, E.E., Clarke, N.A. (1967) Seasonal variations in survival of indicator bacteria in soil and their contribution to storm-water pollution. Applied Microbiology 15, 13621370.
  • 14
    Wells, J.G., Shipman, L.D., Greene, K.D. et al. 1991) Isolation of Escherichia coli serotype O157:H7 and other Shiga-like toxin producing E. coli from dairy cattle. Journal of Clinical Microbiology 29, 985989.
  • 15
    Zhao, T., Doyle, M.P., Shere, J., Garber, L. (1995) Prevalence of enterohemmorhagic Escherichia coli O157:H7 in a survey of dairy herds. Applied and Environmental Microbiology 61, 12901293.
  • 16
    Zibilske, L.M. & Weaver, R.W. (1978) Effect of environmental factors on survival of Salmonella typhimurium in soil. Journal of Environmental Quality 7, 593597.