SEARCH

SEARCH BY CITATION

Keywords:

  • Enterohaemorrhagic;
  • food-borne pathogens;
  • manure treatment;
  • pre-harvest control

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

Aims: The objective of this study was to investigate alkaline treatments of cattle manure to kill coliforms, Escherichia coli O157:H7 and Salmonella Typhimurium DT104 based on their inhibition by carbonate ion and ammonia.

Methods and Results: Pure cultures of S. Typhimurium DT104 and E. coli O157:H7 strains were treated with sodium carbonate and ammonia to determine threshold inhibitory concentrations. Fresh cattle manure samples were inoculated with the same strains and their survival was determined after addition of sodium hydroxide, ammonium sulphate, sodium carbonate and/or urea. Control of COinline image and NH3 concentrations in manure by pH adjustment to 9·5 with sodium hydroxide to more than 5 and 30 mmol l−1, respectively, killed more than 106 cells g−1 in 7 days. Addition of sodium carbonate enhanced the killing effect of NaOH by increasing the COinline image and NH3 concentrations. Addition of 100 mmol l−1 urea, produced high levels of COinline image and NH3 and decreased all bacterial counts by at least 106 cells g−1 after 7 days.

Conclusions: Reduction of food-borne pathogens in manure can be achieved by a combination of high concentrations of COinline image and NH3 which are pH-dependent parameters.

Significance and Impact of Study: Addition of urea could provide a simple manure treatment by combining both antimicrobial factors.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

Gastrointestinal infections by Salmonella are the second leading bacterial food-borne disease, and it has been estimated that this pathogen infects approx. 1·4 million people in the USA (Mead et al. 1999). Escherichia coli O157:H7 is also considered a major public health hazard as it causes more than 73 000 illnesses every year in the USA. Escherichia coli O157:H7 causes enterohaemorrhagic colitis that can lead to serious complications such as haemolytic uraemic syndrome (Nataro and Kaper 1998). In recent years, the number of food-borne outbreaks in North America caused by E. coli O157:H7 and the multi-antibiotic resistant S. Typhimurium DT104 have increased (Akkina et al. 1999; Olsen et al. 2000). Many outbreaks caused by these pathogens have been linked to the consumption of contaminated ground beef, dairy products and fresh produce fertilized with cattle manure.

Investigations from some of those outbreaks have indicated that animal manure was the potential source of food contamination (USDA:APHIS 1997). It has long been recognized that animal manure can be a reservoir of food-borne pathogens because these organisms are part of livestock's natural intestinal flora. Salmonella frequently colonizes the gastrointestinal tract of a variety of animals (USDA:APHIS 1995; CEAH/USDA/APHIS 1997) and E. coli O157:H7 is typically shed in the faeces of cattle (Elder et al. 2000). Both organisms are released into the environment via faecal shedding and a number of studies have demonstrated that they can survive in manure for long periods of time (Kudva et al. 1998; Himathongkham et al. 1999; McGee et al. 2001). These pathogens can be further disseminated into the environment by the practice of spreading manure onto fields to serve as crop fertilizer.

Treatments such as anaerobic digestion and composting can reduce faecal pathogens in animal manure, but their effectiveness is highly dependent upon strict process controls. A study investigating composting of dairy waste reported that coliforms were not eliminated after 30 days at temperatures greater than 50°C (Mote et al. 1988). Because of the increased operational cost, few farms can afford to use anaerobic digestion and composting to treat manure. Typical cattle farms employ manure management that includes storing manure in pits or lagoons for several months and spreading it at the end of each season. Because of the potential dissemination of E. coli O157:H7 and S. Typhimurium DT104 from cattle manure, a number of methods are currently being investigated to reduce pathogen prevalence.

The role of ammonia concentration in animal manure has been investigated to control the populations of Salmonella (Turnbull and Snoeyenbos 1973; Himathongkham et al. 2000). Turnbull and Snoeyenbos (1973) were able to reduce the population of S. Typhimurium in dried poultry litter by 9 log cycles after treatment with ammonia gas for 11 days and they concluded that ammonia killed this bacterium because of the high pH. A recent report indicated that ammonia could also be effective in inactivating E. coli O157:H7 in bovine manure and suggested that the accumulation of free ammonia was one of the most important factors to kill pathogenic bacteria (Himathongkham et al. 2000). This latter report provided evidence for the killing of Salmonella and E. coli O157:H7 in pure cultures, but the influence of the concentration of other manure components was not considered.

One of the recently proposed technologies that are being studied for manure treatment is the use of carbonate to kill food-borne pathogens. Recent work by Diez-Gonzalez et al. (2000) have shown that carbonate anion was very effective in killing coliforms and E. coli in cattle manure, but its effect on Salmonella and E. coli O157:H7 was only demonstrated in pure culture experiments. In the same paper, it was reported that addition of sodium carbonate killed E. coli and it was speculated that cattle urine supplied a significant fraction of carbonate anion. In a recent paper, Arthurs et al. (2001) compared the effect of different carbonate sources on the reduction of E. coli in manure slurries, but little consideration was given to the carbonate naturally produced by the hydrolysis of urea. This latter paper reported that at pH 9·5 or higher in the presence of 16 g kg−1 sodium carbonate, E. coli was completely eliminated after 5 days, however, the potential effect of naturally present ammonia concentration was not discussed.

This project was undertaken to further investigate the potential alkali treatment of cattle manure with carbonate and ammonia, and it was specifically aimed to determine the changes in carbonate and ammonia concentration in stored manure and the impact of those changes in reducing the coliforms, S. Typhimurium DT104 and E. coli O157:H7. The utilization of urea as a viable and simple alternative source of carbonate and ammonia was also investigated.

Strains and culture conditions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

Escherichia coli O157:H7 (ATCC 43895, CDC EDL 933) was originally isolated from raw hamburger implicated in an haemorrhagic colitis outbreak (Riley et al. 1983). Escherichia coli O157:H7 strain C-984, 380-94 and SEA 13B88 carrying plasmids encoding green-fluorescent protein (gfp) and ampicillin resistance were kindly provided by Dr Pina Fratamico USDA-ARS (Fratamico et al. 1997). Salmonella Typhimurium phage type DT104 (ATCC 700408) is a strain resistant to ampicillin, chloramphenicol, tetracycline, streptomycin and sulphonamide. Bacterial strains were cultivated in Luria-Bertani (LB) broth at 37°C.

Antimicrobial effect of carbonate anion and ammonia in pure cultures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

The LB broth was supplemented with sodium carbonate at three different concentrations (75, 150 and 250 mmol l−1) and the pH was adjusted to 8·5, 9·0 and 9·5 with HCl (3 mol l−1) to obtain different concentrations of carbonate anion (COinline image). Ammonium chloride was also added to LB broth at four different concentrations (75, 150, 300 and 500 mmol l−1) and adjusted to the same pH values as carbonate-supplemented medium to obtain different concentrations of ammonia (NH3). Carbonate- or ammonia-containing LB media (5 ml) were filter-sterilized and dispensed into sterile 10-ml tubes. These tubes were inoculated with 10 μl of overnight E. coli O157:H7 strain ATCC43895 and S. Typhimurium DT104 cultures, and incubated at 37°C. The inoculated cultures were sampled at 6 and 24 h and the cell suspensions were 10-fold serially diluted in sterile 96-well microtitre plates containing LB broth. After incubation of plates at 37°C for 24 h the bacterial count was determined according to the most probable number (MPN) technique.

Manure treatments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

Fresh faecal and urine samples were collected directly from dairy cows (n = 3 and n = 3, respectively) from the Dairy Barn at the University of Minnesota. Manure slurries were prepared by mixing 1 l urine or water and 2·2 kg faeces. Manure slurry samples (300 g) were dispensed into 500-ml sterile containers with a loose lid and were incubated aerobically at room temperature. Manure samples were subjected to addition of sodium hydroxide, sodium carbonate, ammonium sulphate and urea, supplemented alone or in combination. From 5 to 10 ml of a sodium hydroxide (10 mol l−1) solution was thoroughly mixed with manure samples to adjust the pH to 9·5 by stirring manually. Powder sodium carbonate (Na2CO3), ammonium sulphate, ammonium chloride and urea at different concentrations were added directly to manure samples and manually mixed. After manure samples were mixed with the different chemical treatments, they were inoculated with 107–108 cells g−1 of overnight LB-grown cultures of E. coli O157:H7 (mixture of strains C-984, 380-94, SEA 13B88) or S. Typhimurium DT104. The containers with the manure samples were kept at room temperature and small samples were collected at times for chemical and microbiological analyses described below.

Bacteria enumeration

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

The count of total coliforms was determined using the Food and Drug Administration's (FDA) method (Hitchins et al. 1998) that uses the MPN technique in lauryl tryptose broth (LST) and confirmation with brilliant green bile broth (BGLB) media. Salmonella Typhimurium DT104 was determined by 10-fold serially diluting manure samples in triplicate into tubes containing 4·5 ml LB broth that had been supplemented with tetracycline (6 mg l−1), streptomycin (30 mg l−1) and chloramphenicol (6 mg l−1) as selective agents, and sodium thiosulphate (6·8 g l−1) and ferric ammonium citrate (0·8 g l−1) as differential reagents. Antibiotic-containing LB broth tubes were incubated at 37°C for 24 h, and those cultures that became black were struck onto xylose lysine dextrose (XLD; Difco Inc., Sparks, MD, USA) petri plates containing the same concentration of antibiotics for Salmonella identification. XLD plates were incubated at 37°C for 24 h, the plates were examined for the presence of black colonies, and the MPN was calculated based on the original positive dilution tubes. Ampicillin-resistant and gfp-producing E. coli O157:H7 was determined using the same method as described for S. Typhimurium DT104 except that the LB tubes contained only ampicillin (100 mg l−1), and confirmation of positive tubes was made by plating and incubating in ampicillin-containing LB-agar medium. Positive colonies were identified by a distinctive green colour when the plates were exposed to ultraviolet light (365 nm) in a Gel-Doc 8000 imaging equipment (UVP Inc., Upland, CA, USA).

Analytical assays

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

The concentration of total carbonate in manure was determined by mixing 10-g samples with 10-ml 3 mmol l−1 HCl in sealed tubes, and measuring the volume of water displaced by CO2. The liberated CO2 volume was converted to total carbonate concentration using a standard curve of solutions of known carbonate concentration and their liberated volume. The COinline image concentration was calculated based on the total carbonate concentration, the manure pH, the pKa values of carbonic acid (6·35 and 10·33) and the Henderson–Hasselbalch equation (log[base]/[acid] = pH − pKa). Total ammonia concentration in manure was assayed using the colorimetric method of Chaney and Marbach (1962). Faecal samples (0·5 g, n = 3) were put into a saline solution (4·5 ml) and 50 μl (n = 3) of 10- and 100-fold diluted faecal suspension were used for total ammonia assay. The concentrations of NH3 were calculated based on the total ammonia concentration, the manure pH, the Henderson–Hasselbalch equation and a pKa value of 9·25 and total ammonia amount measured in above procedure. Manure pH was determined by direct measurement with a pH electrode after 10-min equilibration.

These assays were performed at 2, 4 and 7 days in the same time with sampling for enumeration of bacteria.

Statistical analysis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

Data was the mean value of at least two independent experiments and the coefficients of variation were less than 10%. Significant differences between treatments were determined using Student's t-test (Moore and McCabe 1999).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

The viability of cultures of E. coli O157:H7 and S. Typhimurium DT104 incubated in LB media at pH values of 8·5, 9·0 and 9·5 was not decreased after 24 h, but their numbers were reduced when sodium carbonate or ammonium chloride was supplemented in combination at the respective pH values. After sets of cultures were incubated at each of those pH values containing 75, 150 and 250 mmol l−1 sodium carbonate, the bacterial count was determined after 6 h, and a relationship between the calculated carbonate anion (COinline image) and reduction of viability was noted (Fig. 1a). If the concentration of carbonate anion was less than 1 mmol l−1, the count of both bacteria was not reduced and Salmonella numbers actually increased as much as 10-fold. However, when the concentration of COinline image was increased to values greater than 2 mmol l−1, reductions in bacterial count were observed with concentration increments. At 10 mmol l−1 COinline image, the count of S. Typhimurium DT104 decreased by 1·5 logs and the viability of E. coli O157:H7 declined approx. 3·5 logs. At very high concentrations (>30 mm), the count of both bacteria declined more than 5 log cells ml−1 after 6 h.

image

Figure 1. Effect of calculated COinline image (a) and NH3 (b) concentrations on the survival of Escherichia coli O157:H7 (•) and Salmonella Typhimurium DT104 (bsl00001) in cultures incubated for 6 h at 37°C

Download figure to PowerPoint

In similar pure culture experiments carbonate was replaced by 75, 150, 300 and 500 mmol l−1 ammonium chloride at the same pH values and a relationship between the ammonia concentration and the reduction of bacterial count after 6 h was determined (Fig. 1b). At ammonia concentrations smaller than 5 mmol l−1S. Typhimurium and E. coli O157:H7 were able to grow, but as the concentration increased their viability was decreased. The Salmonella count decreased at an approximate rate of 0·14 log cell reduction per mmol l−1 ammonia, but the inactivation of E. coli O157:H7 was only 0·03 logs per mmol l−1 ammonia. More than 40 mmol l−1 ammonia caused a complete reduction in the Salmonella population, but the E. coli count was not reduced more than 5 log cells ml−1 until the ammonia concentration was 180 mmol l−1.

Untreated fresh cattle manure samples had concentrations of naturally present carbonate (measured as CO2 released) and total ammonia (NH3 + NHinline image) of approx. 50 and 100 mmol l−1, respectively, and these values could increase up to 115 and 300 mmol l−1 after 3 days of storage (Fig. 2a). The carbonate concentration gradually declined after the first week of storage to approx. 75 mmol l−1 after 25 days. The pH of untreated manure remained relatively constant throughout the experiments and it was never higher than 8·3 (Fig. 2b). The initial counts of coliform bacteria in manure samples were approx. 8 log cells g−1 manure, and this value was reduced to approx. 6 log cells g−1 manure after 25 days (Fig. 2c). In manure samples that had been inoculated with 108 cells g−1 manure of S. Typhimurium DT104, the final count at the end of the experiment was almost 106 cells g−1 manure (Fig. 2c). The concentrations of calculated COinline image and NH3 were always less than 1·5 and 5 mmol l−1, respectively (data not shown).

image

Figure 2. Total concentrations of carbonate (•) and ammonia (○) (a), pH (b) and count of coliforms (bsl00066) and artificially inoculated Salmonella Typhimurium DT104 (▵) (c) in untreated cattle manure stored at room temperature

Download figure to PowerPoint

When the pH of manure was adjusted to 9·5, the concentration of total carbonate increased by approx. 80 mmol l−1 within 2 days and reached a total of 190 mmol l−1 after 7 days (data not shown). The concentration of carbonate in those samples that had been supplemented with sodium carbonate after pH-9·5 adjustment also had an increase in carbonate concentration from 60 to 80 mmol l−1 within 2 days of storage. After 7 days, they had a carbonate concentration that was equal to the amount added initially plus approx. 190 mmol l−1. The total ammonia concentration of pH-9·5 adjusted and carbonate-added samples increased from approx. 130 to 300 mmol l−1 after 2 days and remained at that level after 7 days.

The pH of all pH-9·5 adjusted and carbonate-treated samples declined throughout the experiment, but in samples that were supplemented with 80 mmol l−1 sodium carbonate, the final pH was approx. 8·7 (Fig. 3a). The initial COinline image concentration of manure samples increased from 0·07 mmol l−1 to approx. 10 mmol l−1 by adjusting the pH to 9·5 with sodium hydroxide (Fig. 3b). Addition of sodium carbonate in combination with pH 9·5 increased the initial COinline image concentration even further, but after 7 days of incubation the COinline image level declined to less than 7 mmol l−1 (Fig. 3b). Adjustment of pH also caused an increase in the NH3 concentration at time 0 from 1·4 to 90 mmol l−1 and this concentration continued to rise during the first 2 days to a maximum value of 150 mm (Fig. 3c). The NH3 concentration of pH-adjusted samples (with and without carbonate addition) declined during the rest of the experiment, but at least 35 mmol l−1 NH3 remained after 7 days.

image

Figure 3. The effect of pH adjustment and sodium carbonate addition on the pH (a), COinline image concentration (b) and NH3 concentration of cattle manure (mixture of faeces and urine). No addition (•), 0 mmol l−1 Na2CO3 plus pH-9·5 adjustment (bsl00001), 40 mmol l−1 Na2CO3 plus pH-9·5 adjustment (bsl00066), and 80 mmol l−1 Na2CO3 plus pH 9·5 adjustment (bsl00046)

Download figure to PowerPoint

The population of coliforms in manure samples was gradually reduced from 107 to 101 cells g−1 after 7 days, by initially adjusting the pH to 9·5 with sodium hydroxide (Fig. 4a). Addition of 40 and 80 mmol l−1 sodium carbonate to pH-adjusted samples caused a marked coliform reduction in their viable count and almost no survivor was recovered after 4 days. The viable count of E. coli O157:H7 and S. Typhimurium DT104 strains that were added to untreated manure samples remained at approx. 7 log cells g−1 during 7 days (Fig. 4b,c). By initially adjusting the manure pH to 9·5, the population of E. coli O157:H7 was reduced to 5 log cells g−1 after 2 days and it was less than 1 log cells g−1 by day 7. Addition of 80 mmol l−1 sodium carbonate caused an almost complete inactivation of serotype O157:H7, but could still be recovered after 7 days.

image

Figure 4. Survival of coliform (a), Escherichia coli O157:H7 (b) and Salmonella Typhimurium DT104 in cattle manure (mixture of faeces and urine) as affected by pH adjustment and sodium carbonate addition. No addition (•), 0 mmol l−1 Na2CO3 plus pH-9·5 adjustment (bsl00001), 40 mmol l−1 Na2CO3 plus pH-9·5 adjustment (bsl00066), and 80 mmol l−1 Na2CO3 plus pH-9·5 adjustment (bsl00046)

Download figure to PowerPoint

Salmonella Typhimurium DT104 was not affected by pH adjustment during the first 2 days of treatment, but it was decreased by as much as 6 log cells g−1 after 7 days (Fig. 4c). Mixing with either 40 or 80 mmol l−1 sodium carbonate reduced the strain DT104 count to less than 100 cells after 2 days, and caused complete inactivation after 4 days. No detectable levels of coliforms, E. coli O157:H7 and S. Typhimurium DT104 were detected in carbonate-treated samples after 25 days (data not shown).

Samples of manure that were prepared by substituting water with cattle urine and adjusted to pH 9·5 with sodium hydroxide, had a constant concentration of total carbonate when 150 mmol l−1 sodium carbonate was initially added, but the COinline image concentration rapidly decreased from 20 to 3·5 mmol l−1 in 2 days (Fig. 5a). Similar pH-9·5 adjusted samples that were supplemented with 300 mm ammonium chloride had less than 10 and 2 mmol l−1 total carbonate and COinline image, respectively. The NH3 concentrations of NH4Cl-treated samples markedly declined more than 10-fold during the experiment (Fig. 5b), but the total ammonia concentration decreased less than twofold. The NH3 concentrations of carbonate treated samples were never higher than 25 mmol l−1 (Fig. 5b). The pH of manure samples that did not contain urine declined throughout the experiment to final values of 7·2 and 8·0 for Na2CO3- and NH4Cl-treated samples, respectively (Fig. 5c). Treatment with 150 mmol l−1 Na2CO3 did not decrease coliforms, E. coli O157:H7 and Salmonella, but addition of 300 mmol l−1 NH4Cl reduced their counts by 3, 2·5 and 4 log cells g−1, respectively (Fig. 6).

image

Figure 5. The effect of addition of 150 mmol l−1 sodium carbonate (closed symbols) and 300 mmol l−1 ammonium chloride (open symbols) on the pH (a), total carbonate (•, ○) and COinline image (bsl00001, □) concentrations (b), and total ammonia (bsl00001, □) and NH3 (•, ○) concentrations (c) of cattle manure

Download figure to PowerPoint

image

Figure 6. Survival of coliforms (a), Escherichia coli O157:H7 (b) and Salmonella Typhimurium DT104 in cattle manure prepared by mixing faeces with water and adjusting pH to 9·5 with sodium hydroxide, as influenced by the addition of 150 mmol l−1 sodium carbonate (•) and 300 mmol l−1 ammonium chloride (○)

Download figure to PowerPoint

In regular manure samples (combination of faeces and urine) that were treated with 100–300 mmol l−1 ammonium sulphate [(NH4)2SO4], the carbonate concentration was never greater than 100 mmol l−1, and the pH values were lower than 8 with little difference from that of untreated samples (data not shown). The COinline image concentration remained below 0·5 mmol l−1 and it decreased fivefold after 7 days. The NH3 concentration in untreated samples increased from 120 to 300 mmol l−1 after 7 days. In (NH4)2SO4 -supplemented samples, the concentration of total ammonia increased up to 400 mmol l−1 after 7 days, but the concentration of NH3 decreased after 2 days. Addition of 200 and 300 mmol l−1 (NH4)2SO4 caused an approximate reduction of 3·5- and 2-log in the coliform and Salmonella count, respectively, after 7 days. The population of E. coli O157:H7 was largely unaffected by (NH4)2SO4 addition (data not shown).

The pH of manure samples that were mixed with 100, 200 and 300 mmol l−1 urea increased to 8·5 or higher (Fig 7a), and the total carbonate concentration increased from 75 mmol l−1 to approx. 210, 300 and 370 mmol l−1, respectively, after 2 days. After this time the pH of all urea-treated samples decreased slightly, but remained greater than 8·0. The COinline image concentration increased in the first 2 days after addition of urea and declined slowly throughout the rest of the experiments (Fig. 7b). The total ammonia concentration of urea-treated samples increased during the first 2 days to concentrations from 300 to 700 mmol l−1 depending on the initial urea concentration. When the concentration of urea was 100 mmol l−1, the total ammonia concentration continued to accumulate to 400 mmol l−1, but at concentrations of urea greater than 200 mmol l−1 it decreased gradually throughout the rest of the experiments (data not shown). The final concentrations of NH3 were approx. 40, 125 and 245 mmol l−1 for the 100, 200 and 300 mmol l−1 urea treatments (Fig. 7c). The populations of coliforms, E. coli O157:H7 and S. Typhimurium DT104 were reduced more than 7 logs by urea treatments (Fig. 8) after 7 days. Salmonella Typhimurium DT104 was not even detected after 4 days in most of the treatments.

image

Figure 7. The effect of urea addition on the pH (a), COinline image concentration (b) and NH3 concentration of cattle manure (mixture of faeces and urine). Urea was added to obtain concentrations of 0 (•), 100 (bsl00001), 200 (bsl00066), and 300 mmol l−1 (bsl00046)

Download figure to PowerPoint

image

Figure 8. Survival of coliforms (a), Escherichia coli O157:H7 (b) and Salmonella Typhimurium DT104 in cattle manure (mixture of faeces and urine) influenced by addition of urea alone. Urea was added to obtain concentrations of 0 (•), 100 (bsl00001), 200 (bsl00066), and 300 mmol l−1 (bsl00046)

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

The antibacterial effect of bicarbonate was reported by Corral et al. (1988), but recent work indicated that the carbonate anion (COinline image) was the fraction responsible for killing microorganisms (Diez-Gonzalez et al. 2000). Diez-Gonzalez et al. (2000) reported that carbonate could kill a variety of bacterial strains after exposure to 150 mmol l−1 sodium carbonate after 24 h, and because this effect was not observed with equimolar solutions of ammonium chloride, an osmotic effect was unlikely to be causing bacterial reduction. In this research a killing effect of pH as high as 9·5 was not observed in the absence of carbonate, but more than 5-log reductions in the viable count of S. Typhimurium DT104 and E. coli O157:H7 were observed after only 6 h when the concentration of COinline image was greater than 30 mm. The extent of killing after 6 h was highly correlated with the COinline image concentration and this observation supported the idea that the carbonate anion was the killing form of carbonate.

A similar relation of bacterial reduction to COinline image concentration was previously established for faecal E. coli in cattle manure after 5 and 10 days of treatment with carbonate (Arthurs et al. 2001. In that report, the authors concluded that the concentration of COinline image had to be greater than 1 mmol l−1 in order to observe significant reductions, and this value was similar to the one estimated in the present paper. The minimal killing concentration of COinline image determined with pure cultures of S. Typhimurium DT104 and E. coli O157:H7 was approx. 2 mmol l−1, and bacterial growth was observed at lower concentrations. The antibacterial threshold concentration of COinline image was similar for both strains, but it appeared that S. Typhimurium DT104 was more resistant to carbonate than E. coli O157:H7 (Fig. 1a).

It has long been recognized that ammonia (NH3) is capable of inhibiting and killing microorganisms, and its importance on the survival of food-borne pathogens in animal manure has been investigated (Turnbull and Snoeyenbos 1973). Himathongkham et al. (2000) recently observed that the effect of NH3 on the viability of pure cultures of S. Typhimurium and E. coli O157:H7 was dependent on the ‘free ammonia’ concentration, but the authors did not assess minimum inhibitory levels and sensitivity differences among strains. In the present research, it appeared that the level of NH3 at which reductions started to be observed for both microorganisms was approx. 5 mm, but E. coli O157:H7 strain 43895 was markedly more resistant than S. Typhimurium DT104 (Fig. 1b). This relationship between the concentration of NH3 and the loss of viability could be used to assess the survival of these pathogens in manure containing NH3.

Previous reports have indicated that Salmonella strains can survive in cattle manure for more than 100 days at room temperature (Forshell and Ekesbo 1993) and in the present study the population of S. Typhimurium DT104 was only reduced to 2 logs after 25 days (Fig. 2b). During this time, the concentration of COinline image was always less than 2 mmol l−1 despite the fact that the total concentration of carbonate initially increased. Arthurs et al. (2001) reported that the concentration of total carbonate of untreated manure slurries remained at approx. 50 mmol l−1 for 10 days. However, the concentration of carbonate in undiluted manure increased from 50 to 125 mmol l−1 after 3 days and it decreased to approx. 75 mmol l−1 at day 25 (Fig. 2a). These changes in the carbonate concentration were probably because of urea hydrolysis and subsequent CO2 losses at low pH.

Untreated manure samples had consistently pH values less than 8·5 and under these conditions the concentrations of COinline image and NH3 were less than 0·5 and 4 mmol l−1, respectively, which were not enough to reduce the bacterial count (Fig. 2). When the pH in natural manure was adjusted to 9·5, the amount of COinline image and NH3 were increased to 10 and 100 mmol l−1, respectively (Fig. 3). Although the gradual decrease in pH caused a decline in the concentrations of COinline image and NH3, these values remained above the threshold levels determined from Fig. 1. As a result, pH-adjustment alone caused reductions in coliform, E. coli O157:H7 and S. Typhimurium DT104 of more than 6 logs after 7 days (Fig. 4). The addition of 40 and 80 mmol l−1 sodium carbonate to pH-adjusted samples enhanced the bacterial reduction not only because of the higher concentration of COinline image but also because of the increased concentration of NH3 as a result of the buffering effect of carbonate.

The effect of carbonate and ammonia was then investigated separately by eliminating urine from the manure mixture (Fig. 5). In manure samples that were treated with sodium carbonate alone, the total carbonate concentration remained at approx. 150 mmol l−1 during 7 days, but the pH decreased more rapidly than in carbonate-treated, urine-containing samples. This rapid decline caused a logarithmic reduction in the concentration of COinline image, which was not sufficient to cause bacterial inhibition. Similarly, the NH3 concentrations of ammonium chloride-treated samples were steadily decreased by a rapid pH decline, despite the observation that the total ammonia concentration was more than 150 mmol l−1. The ammonia levels in the first 4 days of the experiment were higher than 20 mmol l−1 that caused approx. 3 log bacterial reduction during this period (Fig. 6). After the concentration decreased to less than 20 mmol l−1 no more reduction was observed. These results indicated that the presence of urea naturally present in urine might be an important factor to ensure the antimicrobial activity of carbonate and ammonia by serving as their natural source and as buffer.

Addition of more than 100 mmol l−1 ammonium sulphate to manure samples appeared to reduce the total carbonate concentration (data not shown), but the total ammonia concentration increased more than 150 mmol l−1 after 7 days. Because the pH was always less than 8, the NH3 concentrations were never higher than 30 mmol l−1 and the carbonate concentration was less than 1 mmol l−1. Under these conditions the extent of coliform and Salmonella reductions were very limited and the count of E. coli O157:H7 was almost not affected. This experiment supported the idea that Salmonella was more sensitive to ammonia than E. coli O157:H7, and suggested that the minimal NH3 concentration at which significant bacterial reduction in manure should be at least 30 mmol l−1.

From these experiments it is clear that the effect of carbonate and ammonia treatments are highly dependent on alkaline pH, and this observation supports the idea that the killing species are COinline image and NH3. However, their antimicrobial mechanisms appear to be markedly different. Jarvis et al. (2001) recently compared the killing effect of carbonate to that of EDTA in E. coli and reported that in cells treated with both compounds, the level of six enzymes was markedly repressed. Because those enzymes had divalent metal cofactors in their structure, the authors concluded that the antimicrobial effect of COinline image was probably the result of chelating of ions such as Ca2+ and Mg2+. Ammonia is considered a very toxic substance for any type of organism, but the actual killing mechanism has not been fully elucidated. It is believed that NH3 can cause a rapid alkalinization of the cytoplasm as it can easily cross the cell membrane by simple diffusion and reduce the proton concentration as NHinline image is formed at lower intracellular pH. Because of their different killing mechanisms it is likely that COinline image and NH3 could be used in combination for manure treatment.

Despite the fact that ammonia can have antimicrobial activity, it is not normally utilized as a sanitizer or disinfectant because of its high volatility and toxicity, and these characteristics would preclude any direct application to treat animal manure. In addition, the concentration of ammonia might be markedly affected by its volatilization caused at the time of mixing and spreading of manure, which can have serious implications for its value as fertilizer, the health of farmers and the environment. Urea, a compound naturally present in urine, is the major source of ammonia in manure and addition of synthetic urea could be used to increase not only the ammonia concentration, but also the carbonate levels. Commercial urea is produced by chemical synthesis from natural gas and it is widely used as crop fertilizer (Briggs 1967). It has been estimated that the US utilizes approx. 3·9 million tonnes of urea as nitrogen fertilizer and urea accounts for approx. 20% of the total amount of agricultural fertilizers (EPA 1999). Urea is also used as a feed supplement for livestock to increase their nitrogen intake (Briggs 1967). Because of its availability and safety, urea could be a suitable ingredient for manure treatment.

Supplementation of manure samples with urea caused an increase in pH because of the enhanced production of ammonia and carbonate from urea hydrolysis (Fig. 7). Samples treated with at least 100 mmol l−1 urea had concentrations of COinline image and NH3 higher than 2 mm and 40 mmol l−1, respectively, which were greater than the threshold values estimated above. At these concentrations, marked reductions in all bacterial counts were observed, however, the specific contribution of any of those chemical species could not be determined. The rate of bacterial reduction when samples were treated with urea appeared to be slower than in pH-adjusted, carbonate treated samples, but this effect could be explained by the additional time that urea hydrolysis took to increase the pH and the concentrations of COinline image and NH3.

Untreated manure samples prepared in a ratio of 2·2 : 1 of faeces to urine had approx. 150 mmol l−1 urea, but this amount was not sufficient to cause any antimicrobial effect (Fig. 2). The bacterial reduction observed with at least 100 mmol l−1 urea addition suggested that the total concentration of urea for the elimination of pathogens in manure should be at least 250 mmol l−1 under similar conditions. The manure composition in the farm environment is often more complex and diluted than the manure mixtures used in this report. It should be noted that the concentration of urea, pH and storage conditions can be quite variable and this variability could play an important role in the effectiveness of any manure treatment.

The successful application of a manure treatment that would entail simple mixing of animal waste with chemical substances is highly dependent upon a variety of factors such as safety, low cost, easy handling and effectiveness. Alkalinization of manure with sodium hydroxide appeared to be the simplest and most cost-effective procedure as significant bacterial reductions were observed. However, because of the variability of manure composition and changing concentrations of carbonate and ammonia, its effectiveness might be limited. Supplementation with carbonate and urea would ensure that the bacteriocidal effect would be achieved. The results of this study provide valuable information that could lead to novel manure treatments to control pathogens in animal manure using the principles of ammonia and carbonate inhibition.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References

Funding for this project was provided by the Minnesota Agricultural Experiment Station. The authors would like to thank Dr Pina Fratamico for providing the E. coli O157:H7 strains and Dr Sita Tatini for critically reviewing this manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Strains and culture conditions
  6. Antimicrobial effect of carbonate anion and ammonia in pure cultures
  7. Manure treatments
  8. Bacteria enumeration
  9. Analytical assays
  10. Statistical analysis
  11. Results
  12. Discussion
  13. Acknowledgements
  14. References
  • Akkina, J.E., Hogue, A.T., Angulo, F.J., Johnson, R., Petersen, K.E., Saini, P.K., Fedorka-Cray, P.J. and Schlosser, W.D. (1999) Epidemiologic aspects, control, and importance of multiple-drug resistant Salmonella Typhimurium DT104 in the United States. Journal of the American Veterinary Medicine Association 214, 790798.
  • Arthurs, C.E., Jarvis, G.N. and Russell, J.B. (2001) The effect of various carbonate sources on the survival of Escherichia coli in dairy cattle manure. Current Microbiology 43, 220224.
  • Briggs, M.H. (1967) Urea as a Protein Supplement. London: Pergamon Press.
  • CEAH/USDA/APHIS (1997) Shedding of Salmonella by finisher hogs in the U.S. Animal and Plant Health Inspection Service. Veterinary Services, Fort Collins, CO, USA.
  • Chaney, A.L. and Marbach, E.P. (1962) Modified reagents for determination of urea and ammonia. Clinical Chemistry 8, 130132.
  • Corral, L.G., Post, L.S. and Montville, T.J. (1988) Antimicrobial activity of sodium bicarbonate. Journal of Food Science 53, 981982.
  • Diez-Gonzalez, F., Jarvis, G.N., Adamovich, D.A. and Russell, J.B. (2000) Use of carbonate and alkali to eliminate Escherichia coli from dairy cattle manure. Environmental Science and Technology 34, 12751279.
  • Elder, R.O., Keen, U.E., Siragusa, G.R., Barkocy-Gallagher, G.A., Koohmaraie, M. and Laegreid, W.W. (2000) Correlation of enterohemorrhagic Escherichia coli O157 prevalence in feces, hides, and carcasess of beef cattle during processing. Proceedings of the National Academy of Sciences 97, 29993003.
  • EPA (1999) Background Report on Fertilizer Use, Contaminants and Regulation, pp. 1012. Washington, DC: National Program Chemicals Division.
  • Forshell, L.P. and Ekesbo, I. (1993) Survival of salmonellas in composted and not composted solid animal manures. Journal of Veterinary Medicine 40, 653658.
  • Fratamico, P.M., Deng, M.Y., Strobaugh, T.P. and Palumbo, S.A. (1997) Construction and characterization of Escherichia coli O157:H7 strains expressing firefly luciferase and green fluorescent protein and their use in survival studies. Journal of Food Protection 60, 11671173.
  • Himathongkham, S., Bahari, S., Riemann, H. and Cliver, D. (1999) Survival of Escherichia coli O157:H7 and Salmonella Typhimurium in cow manure and cow manure slurry. FEMS Microbiology Letters 178, 251257.
  • Himathongkham, S., Riemann, H., Bahari, S., Nuanualsuwan, S., Kass, P. and Cliver, D.O. (2000) Survival of Salmonella Typhimurium and Escherichia coli O157:H7 in poultry manure and manure slurry at sublethal temperatures. Avian Diseases 44, 853860.
  • Hitchins, W.D., Feng, P., Watkins, W.D., Rippey, S.R. and Chandler, L.A. (1998) Escherichia coli and the coliform bacteria. FDA Bacteriological Analytical Manual. Washington, DC: AOAC International.
  • Jarvis, G.N., Fields, M.W., Adamovich, D.A., Arthurs, C.E. and Russell, J.B. (2001) The mechanism of carbonate killing of Escherichia coli. Letters in Applied Microbiology 33, 196200.
  • Kudva, I.T., Blanch, K. and Hovde, C.J. (1998) Analysis of Escherichia coli O157:H7 survival in ovine or bovine manure and manure slurry. Applied Environmental Microbiology 64, 31663174.
  • McGee, P., Bolton, D.J., Sheridan, J.J., Earley, B. and Leonard, N. (2001) The survival of Escherichia coli O157:H7 in slurry form cattle fed different diets. Letters in Applied Microbiology 32, 152155.
  • Mead, P.S., Slutsker, L., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C., Griffin, P.M. and Tauxe, R.V. (1999) Food-related illness and death in the United States. Emerging Infectious Diseases 5, 607625.
  • Moore, D.S. and McCabe, G.P. (1999) Introduction to the Practice of Statistics. New York: W. H. Freeman and Co.
  • Mote, C.R., Emerton, B.L., Allison, J.S., Dowlen, H.H. and Oliver, S.P. (1988) Survival of coliform bacteria in static compost piles of dairy waste solids intended for freestall bedding. Journal of Dairy Science 71, 16761681.
  • Nataro, J.P. and Kaper, J.B. (1998) Diarrheagenic Escherichia coli. Clinical Microbiology Reviews 11, 142201.
  • Olsen, S.J., MacKinon, L.C., Goulding, J.S. and Slutsker, L. (2000) Surveillance for Foodborne Disease Outbreaks – United States, 1993–1997. Morbidity and Mortality Weekly Report 49, 151.
  • Riley, L.W., Remis, R.S., Helgerson, S.D., McGee, H.B., Wells, J.G., Davis, B.R., Herbert, R.J., Olcott, E.S., Johnson, L.M., Hargrett, N.T., Blake, P.A. and Cohen, M.L. (1983) Hemorrhagic colitis associated with a rare Escherichia coli serotype. New England Journal of Medicine 308, 681685.
  • Turnbull, P.C.B. and Snoeyenbos, G.H. (1973) The roles of ammonia, water activity, and pH in the salmonellacidal effect of long-used poultry litter. Avian Diseases 17, 7286.
  • USDA:APHIS (1995) Salmonella Shedding by Feedlot Cattle. Forth Collins, CO: Centers for Epidemiology and Animal Health.
  • USDA:APHIS (1997) An Update: Escherichia coli O157:H7 in Humans and Cattle. Forth Collins, CO: CEAH.