Introduction

Composting is a biological decomposition process as the result of microbial activities. Composting has been used as a practical and effective way for reducing human pathogen populations in various types of animal wastes. The finished compost can then be applied to agricultural land, home gardens, ornamental nurseries, and greenhouses as the value-added fertilizer and soil amendment [23, 24]. A potential risk of pathogenic bacteria including Salmonella spp., Escherichia coli O157:H7, and Listeria monocytogenes in farm environment is generally low [27, 29, 36]. However, as shown in recent fresh produce-related outbreaks due to E. coli O157:H7 and Salmonella spp. contaminations, concerns about contamination of vegetables with fecal bacteria in the agricultural environment have increased [6, 10, 34]. Raw or inadequately composted manure has been identified as a potential preharvest source of contamination [5].

Bacterial regrowth in compost indicates that either a few survived cells during composting have multiplied under favorable growth conditions or that recontamination has occurred after active composting. The majority of research on pathogen regrowth was conducted in biosolids. These studies investigated the effect of different environmental factors such as moisture content [4, 12, 40], temperature [15], available nutrients [15, 35], and the population and diversity of indigenous microorganisms [33, 35] on Salmonella regrowth in sterile and/or nonsterile composted biosolids. However, inoculum levels were relatively high (ca. 103–105 CFU/g) in these studies which may not be encountered in composted materials. Additionally, there was the lack of information on the regrowth of E. coli O157:H7 or L. monocytogenes in compost.

Previous studies of manure or manure-based composts were focused on the survival of pathogens such as Salmonella and E. coli O157:H7 in stored composts and/or compost-amended soils. In these studies, high populations of seeded pathogens were monitored under laboratory and field settings. A few studies reported the regrowth of pathogens during composting of animal waste under field condition [8, 32]. For example, Cekmecelioglu et al. [8] enumerated Salmonella populations during windrow composting of food waste and cow manure in both summer and winter. Under winter conditions, inconsistent inactivation of pathogens and regrowth to high levels on several sampling days was observed. It was not clear whether the regrowth was related to rainfall as this information was not reported. Also, there was no data indicating the possibility of fecal contamination by wild animals. Recently, Shepherd et al. [32] reported the increase of avirulent E. coli O157:H7 counts in surface samples during on-farm dairy manure composting for a short period of time immediately following rainfall. The increase of E. coli O157:H7 counts in surface samples of compost could result from the bacterial proliferation when the pathogen encountered appropriate moisture levels in compost after rainfall. Although these studies suggest the possibility of pathogen regrowth in compost, there is a lack of information on the effect of environmental factors on regrowth of pathogen from a few cells in manure-based composts.

Physiological stage of pathogenic microorganisms is an important factor affecting the fate of pathogens in compost. Most composting studies investigating the fate of pathogens were conducted with nonstressed cultures [19, 20, 24], a condition not representative of the microorganisms in complex composting systems. Benito et al. [3] evaluated the response of natural isolates of E. coli O157:H7 to various stresses and concluded that it is important to include stress-resistant strains of target pathogens for preservation treatment. Growth conditions for pathogen typically encountered during the composting such as limited nutrients should be addressed.

The objective of this study was to investigate the regrowth potential of Salmonella spp., E. coli O157:H7, and L. monocytogenes in compost under the selected environmental conditions, using water extract of dairy compost as a model system.

Materials and Methods

Preparation of Compost Extracts

Commercial Black Kow dairy compost (Black Gold Compost Company, Oxford, FL, USA) was purchased at a local store. As listed on its label, the compost contains 0.5% total nitrogen, 0.5% available phosphate (P2O2), 0.5% soluble potash (K2O), and no more than 1% chlorine. Compost was extracted with sterile tap water at different ratios (1:2, 1:5, and 1:10, w/v) in a shaking incubator (100 rpm, 22°C) for 24 h and centrifuged at 5,000×g for 20 min. The supernatant was transferred to a sterile 50 ml centrifuge tube and stored at freezer (−18°C) until used. For sterile water extract preparation, compost extracts of both 1:10 and 1:2 ratios were filtered through 0.2 μm syringe filter (Corning, NY, USA) and stored in freezer until used. Tryptic soy agar (TSA; Becton Dickinson, Sparks, MD, USA) was used to enumerate microorganisms in compost extract.

Dairy compost at different ages (days 0, 7, 14, 30, and 60 after the onset of composting) was taken from the compost heaps of a concurrent field study performed by our lab. The compost mixture consisted of a cow manure/sawdust–calf feces mixture, wasted feed, old hay, and vegetable wastes (squash and plant residues) at a ratio of 10:2:2:2, respectively. All materials used were obtained from a single dairy farm, and none of the collected was subjected to any treatment before composting. The materials were thoroughly mixed with the aid of a front-end loader, and heaps were conical shape of ca. 1.2 × 2 m. During composting, the heaps were on the concrete slab (25 × 16 m) surrounded by a gated fence and were not covered or protected from environmental conditions. The heaps were turned on days 3, 7, 14, 21, and 30 of composting with the use of a front-end loader. Compost samples from days 7, 14, and 30 were taken after the compost heaps were turned, whereas composts from days 0 and 60 were taken from the surface. Water extracts of the compost were prepared as described above.

Bacterial Cultures

A five-strain mixture of each Salmonella spp., E. coli O157:H7, and L. monocytogenes was used as the inoculum. Salmonella enteritidis H2292, Salmonella Newport H9113, Salmonella Poona H9301 (kindly provided by Dr. Doyle, University of Georgia), Salmonella Typhimurium DT 104 ATCC 700408 ISSA with green fluorescent protein (GFP; kindly provided by Dr. Sofos, Colorado State University), and avirulent Salmonella Typhimurium 8243 (kindly provided by Dr. Curtis, Washington University) for Salmonella spp., strains F06M-0923-21 (spinach outbreak strain from California Dept. of Health), F07M-020-1 (Taco John's outbreak strain from California Dept. of Health), H1730 (human isolates associated with lettuce consumption, kindly provided by Dr. Harrison, University of Georgia), ATCC 43895 ISEH GFP (kindly provided by Dr. Sofos, Colorado State University), and avirulent B6914 (kindly provided by Dr. Fratamico, Eastern Regional Research Center, USDA-ARS) for E. coli O157:H7, strains 201, LCDC, Scott A, 101M, and 109 from our stock cultures for L. monocytogenes were used. These strains were induced to be resistant to 100 μg/ml of rifampin (Fisher Scientific, Fair Lawn, NJ, USA) through gradient plate method [28] and stored at −80°C in tryptic soy broth (TSB; Becton Dickinson) with 20% glycerol. For each species, the lag phase and growth rate of rifampin resistant strains were compared to the wild-type strain. The stability of the rifampin-resistant strains was tested up to 21 generations.

Inoculum Preparation and Inoculation Procedure

Each strain was streaked onto TSA supplemented with 100 μg/ml of rifampin and grown at 37°C for 24 h. Single colonies of each strain were transferred into TSB, 1/2 strength of TSB, or 1/10 strength of TSB supplemented with rifampin and grown in a shaking incubator (100 rpm, 37°C) for 24 h. Each culture was centrifuged at 5,000×g for 20 min, washed twice, and resuspended in 0.85% saline. After being adjusted to OD600 as 0.5 (ca. 108 CFU/ml), five equal volumes of rifampin-resistant strains were combined and serially diluted to the desired concentration (ca. 103 CFU/ml). The five-strain mixture of each pathogen was inoculated into compost extract with 1:100 ratio (v/v) to achieve a final concentration of ca. 10 CFU/ml of compost extract and then stored at 35°C or 22°C for 7 days. For nonsterile compost extract, samples were taken and analyzed at days 1, 3, 5, and 7 during storage, whereas 4-, 8-, 16-, and 36-h sampling intervals were added in addition to the same sampling intervals for sterile compost extract.

Microbiological Analysis

The population of each pathogen in compost extract was enumerated by plating tenfold serial dilutions, in duplicate, onto Xylose lysine tergitol 4 (XLT-4, Becton Dickinson), TSA, and Oxford medium (Becton Dickinson) supplemented with rifampin (100 μg/ml) for Salmonella spp., E. coli O157:H7, and L. monocytogenes, respectively. The plates were incubated at 37°C for 24~48 h for the enumeration of each pathogen. When the direct plating method failed to detect any pathogens (detection limit, 10 CFU/ml), selective enrichment method was used. One milliliter of compost extract was transferred into 9 ml of universal preenrichment broth (UPB, Acumedia Manufacturers, Lansing, MI, USA) and incubated at 37°C for 24 h, then transferred into Rappaport–Vassiliadis (RV) broth, TSB, and Fraser broth (Becton Dickinson) supplemented with rifampin for detection of Salmonella spp., E. coli O157:H7, and L. monocytogenes, respectively. After incubation at 37°C for 24 h, a loopful of each secondary enrichment culture was streaked onto XLT-4, sorbitol MacConkey agar (Becton Dickinson), and Oxford medium supplemented with rifampin for Salmonella spp., E. coli O157:H7, and L. monocytogenes, respectively. Randomly selected colonies grown on each selective media were confirmed by a latex agglutination test (Oxoid, Basingstoke, Hampshire, UK) for E. coli O157:H7 and real-time polymerase chain reaction assay for Salmonella spp. and L. monocytogenes [37].

Pulsed-Field Gel Electrophoresis

A five-strain mixture of each pathogen was used to address strain variation in terms of growth and survival in the compost ecosystem. The strains exhibiting extended survival or growth in compost extracts were identified by pulsed-field gel electrophoresis (PFGE). The restriction enzyme XbaI for Salmonella and E. coli O157:H7 and AscI for L. monocytogenes were used to digest the DNAs extracted from the randomly selected isolates. PFGE for Salmonella and E. coli O157:H7 was performed using the protocol described by CDC/PulseNet [7], whereas L. monocytogenes strains were analyzed using a protocol by Graves and Swaminathan [13]. Following restrictive enzyme digestion, PFGE for E. coli O157:H7 and Salmonella was performed with the CHEF MAPPER system (Bio-Rad Laboratories, Hercules, CA, USA) in a 1% agarose gel and 0.5× Tris–borate–EDTA buffer at 6 V/cm for 18 h at 14°C, with initial and final switch times of 2.2 and 54.2 s, respectively. For L. monocytogenes, running time was 22 h with initial and final switch times of 4.0 and 40.0 s, respectively. Lambda DNA ladder (Bio-Rad) was used as molecular size standard. The gel was stained with ethidium bromide, the bands were visualized with UV transillumination, and the gel image was captured with the GelDoc 1000 system (Bio-Rad).

Growth Rate and Generation Time

The growth curve of individual strain of E. coli O157:H7 was determined in sterile compost extract (1:2 ratio). The OD600 of each culture in 96-microwell plate (Corning, Corning, NY, USA) was measured at every 2 h up to 10 h and then every 4 h through 48 h from the time of culture inoculation. The viable cell counts at the selected time periods were determined using an Autoplate 4000 spiral plater (Spiral Biotech, Bethesda, MD, USA). The growth rate and generation time of individual strain were determined from two points in the exponential phase of the growth curves using following equations.

$$ {\text{Growth rate constant }}\left( \mu \right)~ = ~{{\left( {\left( {\log_{10} N_t --~\log_{10} N_0 } \right) \times 2.303} \right)} \mathord{\left/ {\vphantom {{\left( {\left( {\log_{10} N_t --~\log_{10} N_0 } \right) \times 2.303} \right)} {\left( {t - t_0 } \right)}}} \right. } {\left( {t - t_0 } \right)}} $$
$$ {\text{Generation time }}(g)~ = ~{{\left( {\log_{10} N_t --\log_{10} N_0 } \right)} \mathord{\left/ {\vphantom {{\left( {\log_{10} N_t --\log_{10} N_0 } \right)} {0.301;}}} \right. } {0.301;}} $$

N t and N 0 are the number of cells at time t and time t 0.

pH Determination

The pH was determined with an Orion pH meter Model 310 (Orion Research, Boston, MA, USA).

Statistical Analysis

For each trial, two compost extracts of each condition were prepared, and duplicate samples were taken from each extract at selected sampling intervals. At least two trials were conducted for each experiment. Bacterial counts were converted to log CFU/ml and subjected to analysis of variance using SAS (SAS Institute, ver. 9.1, Cary, NC, USA). P value below 0.05 was considered significantly different. Duncan's multiple range test was used to determine the pathogen populations differed at various nutritional concentration, growth medium, extract type (sterile or nonsterile), and compost at different ages. The same statistical tests were applied to determine if pathogen populations were significantly influenced by storage time at each condition.

Results

Determination of Pathogen Regrowth in Compost Extracts

Water extract of commercially available dairy compost was used as a model system to evaluate the effect of temperature, nutrient availability, and indigenous microbial population on pathogen regrowth in compost. Rifampin resistant cultures of each species were not significantly different from the wild-type strain in duration of lag phase and growth rate. Up to 21 generations, no reversion of rifampin resistance to susceptibility was observed (data not shown).

Tables 1 and 2 presented the plate count data of Salmonella spp., E. coli O157:H7, and L. monocytogenes in compost extract at 35°C and 22°C, respectively. At 35°C, both Salmonella spp. and E. coli O157:H7 regrew in compost extract with ratios of 1:2 and 1:5 but not 1:10. The population increases ranged from 0.37 to 2.07 log CFU/ml, with increased regrowth of pathogens in more concentrated compost extract (1:2; Table 1). However, no regrowth was detected for L. monocytogenes in all compost extracts. When tested at room temperature (22°C), all three pathogens regrew in the compost extracts with population increases ranging from 0.27 to 1.77 log CFU/ml (Table 2). Both E. coli O157:H7 and L. monocytogenes multiplied in all compost extracts at 22°C, whereas Salmonella spp. regrew in both 1:2 and 1:5, but not 1:10 compost extracts. The population increases were transient at 35°C compared to 22°C. For example, both Salmonella and E. coli O157:H7 at 35°C were only detectable after enrichment at 5 days, while at 22°C, both pathogens were still culturable after 7 days.

Table 1 The potential regrowth and persistence of Salmonella, E. coli O157:H7, and L. monocytogenes previously grown in TSB in nonsterile water extract of compost with different nutrient concentrations at 35°C
Full size table
Table 2 The potential regrowth and persistence of Salmonella, E. coli O157:H7, and L. monocytogenes previously grown in TSB in nonsterile water extract of compost with different nutrient concentrations at 22°C
Full size table

The population of mesophilic background microorganisms at 35°C increased rapidly by ca. 2.57 and 1.86 log CFU/ml on day 1, from 4.78 and 4.15 log CFU/ml of initial levels in 1:2 and 1:10 ratio of compost extracts, respectively, whereas at 22°C, mesophiles on day 1 regrew much slower than at 35°C with population increases of 1.86 and 1.06 log CFU/ml in 1:2 and 1:10 ratio of compost extracts, respectively. Maximum population was ca. 7.4 and 7.6 log CFU/ml on day 3 at 22°C and 35°C, respectively, regardless of compost extract ratios.

Effect of Physiological Stage of Pathogenic Inocula on the Regrowth Potential in Compost Extracts

The plate count data indicated that both Salmonella and E. coli O157:H7 adapted to the reduced nutrient media regrew to higher (P < 0.05) populations than the control cultures at 22°C (Table 3). Both Salmonella and E. coli O157:H7 previously grown in 1/10 strength of TSB increased in population by ca. 2 log CFU/ml by day 3 and remained at that level during sampling period, whereas L. monocytogenes grown in the same medium grew by only 0.5 log CFU/ml at day 1, and there was no significant difference in regrowth potential among previous growth media.

Table 3 The potential regrowth and persistence of Salmonella spp., E. coli O157:H7, and L. monocytogenes grown previously at 35°C with different growth media in nonsterile water extract of compost (1:2) at 22°C
Full size table

Effect of Indigenous Microorganisms on Pathogen Regrowth in Compost Extracts

To verify the effect of competitive background microflora on pathogen regrowth in compost extract, we compared pathogen regrowth in both sterile and nonsterile compost extracts. All pathogens in sterile water extract with 1:10 ratio of compost extract started to regrow within a day at 22°C with population increases as ca. 0.99, 1.30, and 0.33 log CFU/ml for Salmonella spp., E. coli O157:H7, and L. monocytogenes, respectively (Fig. 1). E. coli O157:H7 is the only pathogen which did not continue regrowth after 2 days of incubation. In the more concentrated compost extract (1:2), the averages of maximum population increase were 5.46, 5.79, and 4.09 CFU/ml for Salmonella spp., E. coli O157:H7, and L. monocytogenes, respectively, up to 7 days after incubation. In nonsterile compost extract, both Salmonella spp. and E. coli O157:H7 regrew only in the most concentrated compost extract (1:2) with much less population increases, i.e., ca. 1.4 log CFU/ml, whereas L. monocytogenes cell populations increased only ca. 0.7 log CFU/ml in compost extract of 1:2 ratio.

Figure 1
figure 1

The potential regrowth and persistence of Salmonella spp., E. coli O157:H7, and L. monocytogenes previously grown in 1/10 strength of TSB in sterile (solid line) and nonsterile (dashed line) compost extract at 22°C (n = 8, error bars = SD)

Full size image

The growth rate of each bacterium in the concentrated sterile compost extract (1:2) was approximately two times higher (P < 0.05) than in the diluted extract (1:10) regardless of species (Table 4). Apparently, the regrowth rate for E. coli O157:H7 was the highest among the three species.

Table 4 Growth rate constant and generation time of Salmonella spp., E. coli O157:H7, and L. monocytogenes cocktail previously grown in 1/10 strength of TSB in sterile compost extract
Full size table

Strain Identification of Extended Survival in Compost Extracts

The PFGE patterns of Salmonella for all isolates matched either serotypes Poona H9301 or Newport H9113. For L. monocytogenes, strains LCDC and 109 were predominant. In regard to E. coli O157:H7, spinach outbreak strain, FO6M-0923-21 which was originally isolated from cow feces on the implicated farm of spinach outbreak or avirulent (stx 1− and stx 2−) B6914 strain, was the predominant strain that regrew in the compost extract.

Strain Variation of E. coli O157:H7 Regrowth in Sterile Compost Extract

The growth curve of individual strain of E. coli O157:H7 was determined in sterile compost extract with concentrated nutrients (1:2). Each strain had ca. 8 h of lag phase and reached maximum populations of ca. 7 log CFU/ml on day 2 (Table 5). However, there was a significant difference in growth rate and generation time among strains, showing the shortest generation time of 2.1 h for F06M-0923-21 (spinach outbreak strain; Table 5).

Table 5 Growth rate constant and generation time of individual strain of E. coli O157:H7 previously grown in 1/10 strength of TSB in sterile compost extract (1:2)
Full size table

Evaluation of Pathogen Regrowth in Compost Extract of Different Ages

During composting, the levels of available nutrients, diversity, and populations of indigenous microflora in compost can vary significantly. To determine the regrowth of pathogens during composting, dairy compost samples at different ages were taken from the compost heaps of our field study. Average temperatures in the compost heaps were above 50°C for 11 days of the first 14 days of composting, with a maximal average temperature of 60.5°C on day 11 (data not shown). Compost samples (0, 7, 14, 30, and 60 days after the onset of composting) were evaluated. There was no noticeable difference in pHs and total mesophillic bacterial counts among composts of different ages (Table 6). Populations of total coliforms and E. coli were high initially and declined rapidly during composting. Compost extracts of all ages supported the regrowth of Salmonella spp. and E. coli O157:H7 with least increase of plate counts in 7-day-old compost extract, whereas L. monocytogenes grew only in the compost extract of 0 and 60 days of composting (Fig. 2). The regrowth potential of all pathogens in different ages of compost extracts ranged from 0.22 to 2.32 log CFU/ml with most population increases in initial compost extracts (Table 6). L. monocytogenes had the least regrowth potential among the three pathogens.

Table 6 The physical and biological characteristics of compost with different ages
Full size table
Figure 2
figure 2

The potential regrowth and persistence of Salmonella spp., E. coli O157:H7, and L. monocytogenes previously grown in 1/10 strength of TSB in nonsterile water extract of compost with different ages at 22°C (n = 12, error bars = SD)

Full size image

Discussion

Compost is an excellent nutrient source for agricultural crops, but it is also known as a potential source of contamination with food-borne pathogens such as E. coli O157:H7, Salmonella, and L. monocytogenes. Generally, the risk of pathogens in compost is very low; however, a few cells which may survive during composting or be introduced to the finished compost may multiply and increase the risk of contamination of crops when compost is used as an organic fertilizer. Previous regrowth studies have been conducted in biosolids with initial inoculum more than 103 CFU/g, focusing on Salmonella regrowth. There are no studies, to date, providing any information that the pathogens can regrow from a few cells in manure-based compost. In this study, we investigated the effect of various environmental factors on pathogen regrowth in compost extracts.

Compost tea is a water extract of compost which is produced by mixing compost and water for a defined period, either actively aerating or non-aerating and with or without nutrient additives [26]. Compost tea is used as a spray for biocontrol of foliar diseases or soil drench to promote plant growth and/or control root diseases [2, 14, 39]. A few studies have examined the microbiological safety of compost tea. Duffy et al. [11] reported that the numbers of E. coli O157:H7 and Salmonella Thompson were increased in compost tea made from various types of compost with over 0.5% molasses, showing different regrowth potentials with pathogen type and compost. More recently, Ingram and Millner [17] revealed that compost tea supplemented with commercially available nutrient solution such as blend of molasses, bat guano, sea bird guano, soluble kelp, citric acid, Epsom salts, ancient scabbed minerals, and calcium carbonate and the mixture of powdered soluble kelp, liquid humic acids, and glacial rock dust caused population increase of E. coli O157:H7 or Salmonella by 10–1,000 folds, whereas compost tea itself did not support the growth of both pathogens during aerated or non-aerated brewing cycle. These studies suggest that compost tea could serve as a potential vehicle to transmit E. coli O157:H7 and Salmonella to fresh produce. Although the compost extract was used as a model system in this study for pathogen regrowth in compost, it has some relevance to compost tea in terms of composition (compost mixed with water) and passively aerated system (sterile flask, loosely closed in a shaking incubator for 24 h) [26]. However, our study revealed various regrowth potential among bacterial species in compost extract without additional nutrients. For example, both Salmonella spp. and E. coli O157:H7 grew at both incubation temperatures in more concentrated water extracts (1:2 and 1:5 ratios), whereas L. monocytogenes grew only at 22°C, not 35°C (Tables 1 and 2).

Among the three pathogens tested, L. monocytogenes had the least regrowth potential. It might be explained by the fact that gram-positive bacteria generally have more requirements for nutrients than gram-negative bacteria [18]. The psychrotrophic nature of L. monocytogenes might allow regrowth of this microorganism only at 22°C. Both Salmonella and E. coli O157:H7 at 35°C were only detectable after enrichment at 5 days, while at 22°C, both pathogens were still detectable after 7 days. Available nutrients in the concentrated compost extract (1:2) could allow initial rapid regrowth of both pathogenic and indigenous microorganisms, and then pathogens may be outcompeted when nutrients became limited at 35°C. Lower incubation temperature (22°C) results in slower growth of mesophilic microorganisms and may balance the growth of pathogenic microorganisms and the background microflora.

Previous studies reported similar findings that pathogen regrowth was possible in nonsterile biosolids in the presence of indigenous microorganisms [22, 30]. Russ and Yanko [30] reported that Salmonella did not grow extensively but remained transient. According to Hussong et al. [15], Salmonella died off within 7 days in non-irradiated composted biosolids at 36°C, while in irradiated and moist compost, the pathogen regrew to 8 log CFU/g. Soares [35] revealed when sufficient moisture levels (e.g., 30%) were allowed in composted biosolids, E. coli and total coliforms multiplied in nonsterile compost with the addition of external carbon source. In the present study, each pathogenic bacterium grew from a few cells to maximum 2 log CFU/ml within 1 day and thereafter declined. Although direct comparison of above studies with our present study was not possible due to differences in compost materials (biosolids vs. manure-based compost) and compost types (solid vs. water extract), these results clearly indicate that pathogen regrowth is a complex response as affected by multiple environmental factors. Furthermore, as shown in present study, potential regrowth of pathogens may become higher when sterilization is used as an alternative final treatment for compost, due to the lack of indigenous microorganisms in compost.

In preharvest environments, microorganisms are usually exposed to some stresses including, but not limited to, poor nutrient availability, suboptimal temperatures, and desiccation. Sublethal exposure to various stresses may enhance the survival of bacteria under the subsequent stress conditions and result in cross-protection against other stresses [1]. For example, acid adaptation at pH 4.8 for 18 h at 37°C for E. coli O157:H7 enhanced subsequent acid tolerance, heat tolerance, and freeze–thaw tolerance [21, 31]. In order to simulate the nutrient conditions in the compost environment, the bacterial cultures were grown in the reduced nutrient media (1/2 and 1/10 strength of TSB) and then inoculated into water extract of compost (1:2 ratio) to observe the regrowth at 22°C. Cultures of both Salmonella and E. coli O157:H7 grown previously in reduced nutrient media (1/10 strength of TSB) regrew to higher (P < 0.05) populations than in the control (full strength of TSB), whereas there was no difference (P > 0.05) in population of L. monocytogenes among growth medium. Bacterial adaptation to limited nutrients requires changes in gene expression that allows the microorganisms to utilize nutrients more efficiently and extends survival [9]. The sigma factor RpoS, which is widely present in many gram-negative bacteria including E. coli O157:H7 and Salmonella, is important for adaptation under nutrient-limited conditions [16, 38]. Previous exposure of Salmonella and E. coli O157:H7 to reduced nutrient media might result in a substantial increase in RpoS expression which then contributes to the regrowth of these pathogens in water extract with limited nutrient availability.

Available nutrients and microbial communities are expected to change with the composting process. Sidhu et al. [33] observed the decline in Salmonella growth in composted biosolids with maturity along with the decline in available organic carbon and nitrogen with maturity. Millner et al. [25] reported that the selected fractions of compost microorganisms had different suppression of Salmonella growth in compost at different temperatures. Our results demonstrate different regrowth potentials for each pathogen in composts at different ages. These observations indicate that the differences of nutrient availability and microbial communities among composts of different ages might have a combined effect on suppressing the pathogens in compost. Further studies of competitive inhibition of indigenous background microflora against pathogen regrowth in compost need to be explored.

In conclusion, dairy compost extract, even for finished ones, contains sufficient nutrients for pathogen regrowth. However, pathogen regrowth in compost was suppressed by indigenous microflora to a high extent. Lower temperature (22°C) may balance the growth of pathogenic microorganisms and the background microflora. Pathogens adapted to low nutrients can regrow well in compost extract. Regrowth potentials of all pathogens were ca. 0.7–1.4 and ca. 4–6 log CFU/ml for nonsterile and sterile compost extracts (1:2 ratio), respectively. Both Salmonella and E. coli O157:H7 regrew easier than L. monocytogenes under the same experimental conditions. These results demonstrate that nutrient availability, species and physiological stage of pathogens, competitive microbial flora in compost, and incubation temperature were important factors affecting the regrowth of food-borne pathogens in the compost ecosystem. Meeting the time and temperature criteria for thermophilic composting does not guarantee the complete destruction of all pathogens. Due to the limitation of sensitivity for pathogen detection, false negative results for compost analysis may occur. Our results clearly indicate the pathogen regrowth could occur from a few survived cells to hazardous levels in compost under appropriate conditions. Even with both the time–temperature exposure data and before/after composting test results, appropriate storage conditions for finished compost are necessary for compost producers to document that their process meets the standards for pathogen limits established for the use of composted fecal materials as a soil amendment. Furthermore, the use of compost tea needs to be cautious since current study shows pathogen could regrow in nonsterile compost extract without additional nutrients. Further studies on determining the combination of these environmental factors and specific competitive inhibitory microorganisms in compost that would suppress or inhibit the growth of pathogens are currently undertaken in our lab.