Effect of Temperature on Salmonella enterica subsp. enterica ser. Javiana, Listeria monocytogenes, and Listeria innocua Persistence in Hydroponic Nutrient Solution

1. Introduction

According to the Food and Agriculture Organization of the United Nations (FAO), by 2050 arable land per person is expected to diminish to one-third of the land that was available in 1970 [1]. Factors that may lead to depletion of arable lands include climatic changes, increase in urbanization, population growth, reduction in freshwater resources, soil fertility loss, and over-farming practices [1]. To meet the consumer’s increasing demand for fresh produce year-round, growers are adopting controlled environment agriculture (CEA) production systems.

Controlled environment agriculture includes various structures (e.g., greenhouses and warehouses) and growing systems (e.g., hydroponic systems and soilless substrates) as an alternative cultivation method to traditional, in-field production. In CEA systems, environmental conditions such as temperature, light, humidity, and pH are typically maintained by automated control systems making it favorable for the growth of fresh produce.

Hydroponic production is a system where plants are grown by dipping their roots partially or fully into nutrient solution containing macro- and micronutrients. These nutrients are the same as those found naturally in the soil which promote the development of plants in the field. There are different types of soil free systems such as wicking, deep water culture (DWC), drip, ebb and flow, nutrient film technique (NFT), and aeroponics. Some advantages and disadvantages of the listed hydroponic systems have been discussed by [2]. Although hydroponic cultivation takes place in closed and controlled environments, these systems still face microbial contamination risks including contaminated water used for irrigation, workers’ personal hygiene, internal and external hygienic conditions of the growing space, and improper, or lack of, sanitization of hydroponic systems between the production cycles. Although foodborne disease outbreaks related to hydroponically grown fresh produce are limited [3], it is essential to develop and instill food safety strategies to mitigate microbial contamination risks for such systems [4].

A review by Riggio et al. (2019) indicated that distinct types of hydroponic systems have been used to study the internalization and dissemination of human enteric (bacterial and viral) pathogens during hydroponic cultivation. Since water is the essential component of hydroponic cultivation and is used in the preparation of nutrient solution, contamination of leafy greens via irrigation water containing human pathogens such as Salmonella enterica subsp. enterica, Listeria monocytogenes, and human norovirus (HuNoV) may lead to food safety risks during hydroponic production [5,6,7,8]. Studies have shown that the route of entry for human pathogens into fresh produce can be via plant roots that are immersed in hydroponic nutrient solution [9,10,11,12,13]. However, there are little to no studies [14,15,16] focusing on bacterial survival in hydroponic nutrient solution in the absence of plants within the system. Of the studies that have considered bacterial survival in nutrient solution, none have evaluated the impact of seasonal temperature variations. Therefore, the present study aimed to investigate the persistence of human bacterial pathogens in hydroponic nutrient solution subjected to different temperatures. During this study, experiments were performed over a 21-day period as a preliminary step for investigating the persistence and internalization of human pathogens in a recirculating DWC hydroponic systems used for the production of lettuce. This 21-day period corresponds to the timeframe for the development of lettuce from seedling to mature plant stage thus mimicking a more realistic scenario of hydroponic cultivation.

2. Materials and Methods

2.1. Preparation of bacterial cultures

Bacterial cultures used in this study include Listeria monocytogenes (FSL R2–574, strain F2365, Cornell University, Institute for the Advancement of Food and Nutrition Sciences (IAFNS) Collection), Listeria innocua (FSL C2-0008, Cornell University; IAFNS Collection), and Salmonella enterica subsp. enterica serotype Javiana (ATCC BAA1593, American Type Culture Collection, Manassas, VA). Salmonella Javiana from frozen 50% glycerol stock was streaked onto tryptic soy agar (TSA; Difco, Sparks, MD) plates using sterile inoculation loops followed by incubation for 18 to 24 h at 37°C. Listeria monocytogenes and L. innocua were also taken from frozen 50% glycerol stocks, streaked onto separate brain heart infusion (BHI; BD, Franklin Lakes, NJ) agar plates, and incubated for 24 h at 35°C. Pure isolated colonies from the streaked plates were cultured separately in 10 ml of BHI broth (Listeria spp.), and S. Javiana into tryptic soy broth (TSB; Difco) for 24 h at 35°C with shaking at 125 rpm. Bacterial cells for each bacteria species were harvested by centrifugation at 4000 x g for 10 min at 4°C and washed twice in 10 ml of 1X Phosphate Buffer Saline (PBS, pH 7.4) solution. The final cell pellet was re-suspended in 10 ml 1X PBS solution at approximately 9 log colony forming units (CFU)/ml. L. innocua was utilized in this study to validate whether it would serve as an appropriate surrogate for L. monocytogenes during our present and the future studies. This strain of L. innocua is also determined to be an appropriate surrogate for use in produce safety research based on the curated list of produce relevant strains published by Harrand et al. [17]

2.2. Preparation of nutrient solution

Five hundred milliliters of nutrient solution were prepared by mixing 7.5 ml of stock solution containing macronutrients and micronutrients (Table 1), 0.1 ml sodium thiosulphate (Sigma Chemical Co., St. Louis, MO) stock solution (12.5 ppm), and 492.4 ml of tap water. The average pH and electrical conductivity of nutrient solution during this study was 6.01±0.07 and 2.62±0.09 mS/cm.

2.3. Inoculation of nutrient solution

Each of the three separate 250 ml glass bottles received 100 ml nutrient solution. One hundred microliters of each bacterium (9 log CFU/ml) were inoculated in separate bottles containing the prepared nutrient solution. The final bacterial concentration was approximately 6 log CFU/ml. The three glass bottles with inoculated nutrient solution and a glass bottle containing 100 ml nutrient solution only (negative control) were placed in a shaking incubator. The incubator was maintained at 15, 25, 30, or 37°C with shaking at 50 rpm throughout 21 days of study. This 21-day study period represents the growth stage of lettuce from seedling to mature plant.

2.4. Sample collection and bacterial quantification

Post inoculation, 1 ml of sample was collected from all four glass bottles on day 0, 1, 3, 5, 7, 14, and 21 and serially diluted. One hundred microliters of each dilution were plated on Oxford Listeria Agar Base with modified Oxford Listeria supplement (MOX; Difco) for detection of L. monocytogenes and L. innocua, and Xylose Lysine Tergitol™ 4 agar (XLT4) plates (Hardy Diagnostics, Santa Maria, CA) were used for detection of S. Javiana. Colony forming units were counted and reported as log CFU/ml.

2.5. Data analysis

All data were analyzed using JMP® Pro 16 (SAS Institute, Inc., Cary, NC, USA) software. Three trials (n=3) were performed for each temperature (15, 25, 30, or 37°C). The persistence data of each bacterium in nutrient solution was log transformed prior to analysis. This study was a completely randomized block design with a split-split plot, and data were analyzed using a mixed effect model to compare mean log CFU/ml obtained from each sampling point for all three bacteria at four different temperatures. Least mean squares were calculated to compare mean log CFU/ml. Tukey-Kramer honest significant difference test was used to compare mean values. A value of p < 0.05 was considered as statistically significant. Each data point (log CFU/ml) of individual bacterium from each temperature was used for the scatter plot, and decimal reduction values (D-values) were calculated using the slope from the linear regression equation (Figure 1). Here, D-values indicate one log (90%) reduction of bacteria (in days) in hydroponic nutrient solution for each temperature.

3. Results

3.1. Persistence of bacteria in hydroponic nutient solution at each temperature

Statistical analysis indicated that a 3-way interaction effect between temperature, sampling day, and bacteria had a significant effect on bacterial persistence in nutrient solution (p<0.0001), so interpretations cannot be made about main effects or 2-way interactions. The estimated mean log reduction of each bacterium at different temperatures in nutrient solution is illustrated in Figure 2.

At 15°C, there was no significant difference in the concentration of L. innocua and L. monocytogenes during day 0 to 7 of sampling However, significant differences were observed on day 14 and 21 compared to other sampling days for both L. innocua and L. monocytogenes (p≤0.0001). There was little to no effect of temperature on the concentrations of S. Javiana recovered throughout 21 days of sampling. Moreover, L. innocua, L. monocytogenes and S. Javiana were detected throughout the 21 days of study at 15°C with an overall reduction of 2.1, 5.1, and 1.1 log CFU/ml, respectively (Table 2).

At 25°C, after sampling on day 3, significant differences were observed in the concentrations of L. innocua and L. monocytogenes on day 5, 7, 14, and 21. (p<0.0001). Whereas no significant differences were observed in bacterial recovery of S. Javiana, indicating that lower temperatures such as 15 and 25°C may have minimal effect on persistence of S. Javiana in hydroponic nutrient solution. Also, L. innocua and S. Javiana were detected throughout the 21 days of study with a ~5.5 and 1.2 log CFU/ml reduction, respectively (Table 3). However, L. monocytogenes was only detectable up to 7 days at 25°C with an estimated 6 log CFU/ml reduction.

At 30°C, after sampling on day 1 significant differences were noticed in concentrations of L. innocua and L. monocytogenes (p<0.0001). Whereas significant difference in concentrations of S. Javiana were observed after sampling day 5 (p<0.0001). Also S. Javiana survived throughout 21 days of the study with a ~4 log reduction, conversely L. innocua persisted only for 7 days, and L. monocytogenes persisted for one day in nutrient solution with a ~6 log reduction. This indicated that temperature had some influence on the survivability of S. Javiana in hydroponic nutrient solution over the period of 21 days.

At 37°C, a significant difference in L. innocua and L. monocytogenes were observed after sampling on day 1 (p<0.0001), contrarily significant differences were observed in S. Javiana concentrations after sampling day 7 (p<0.0001). Furthermore, S. Javiana persisted throughout the study period of 21 days with a reduction of 4.7 log CFU/ml, whereas L. monocytogenes and L. innocua survived in nutrient solution for 3 and 14 days, respectively, with a reduction of ~6 and 5.4 log CFU/ml indicating that higher temperatures may affect persistence of all three bacteria in the nutrient solution.

3.2. Estimated decimal reduction values of bacteria in hydroponic nutrient solution by temperature (in days)

The D-values, or time to achieve a 1 log reduction, of L. innocua, L. monocytogenes, and S. Javiana in hydroponic nutrient solution were determined at 15, 25, 30, and 37°C (Table 6). As evidenced by pathogen log reduction at each temperature over time (Table 2, Table 3, Table 4 and Table 5), the D-values of S. Javiana indicate greater stability in nutrient solution than L. innocua and L. monocytogenes at all temperatures. In addition, D-values differed across temperature for each bacteria evaluated. For instance, at 15°C and 25°C D-values for S. Javiana were estimated to be 82 and 26 days, respectively, compared to D-values of ~4 days at 30°C and 37°C. Similar trends were observed for L. monocytogenes and L. innocua across temperatures (Table 6).

4. Discussion

Water is the primary and essential component of hydroponic production. Fertilizer salts that are vital for the growth of leafy crops are generally mixed with water and used for hydroponic cultivation. Previous studies utilize various modified formulations of nutrient solution to evaluate the internalization and dissemination of enteric pathogens such as Salmonella, L. monocytogenes, Shiga toxin-producing E. coli, and human norovirus, within hydroponic systems for leafy greens production [8,9,13,18,19]. Unlike studies that have focused on investigating the internalization and propagation of human pathogens within leafy greens, the present study focuses on S. Javiana, L. monocytogenes, and L. innocua persistence in hydroponic nutrient solution without plants at different temperatures (15, 25, 30, and 37°C).

Hydroponic nutrient solutions are prepared using macro and micronutrients that are essential for leafy crop growth. Nutrient solution components include fertilizer salts such as calcium, copper, magnesium, sulfur, zinc, chlorine, and sodium [20]. Stokes et al. [21] discussed various bacteria and their dependency on specific nutrients including hydrogen, oxygen, phosphorus, carbon, sulfur, and potassium for their growth. The commonality of nutrients needed for hydroponic production of leafy greens and bacterial growth provides justification for the investigation of bacterial pathogen survival in nutrient solution and characterization of potential food safety risk.

The present study illustrated that bacterial persistence in nutrient solution can be temperature dependent. In a study by Xylia et al. [13], S. Enteriditis (6 log CFU/ml) was inoculated in hydroponic nutrient solution with different pH (5, 6, 7, and 8) and maintained at 21°C or 37°C for 21 h to determine bacterial growth kinetics. Xylia and coauthors observed about a 2 log CFU/ml (from 6 to 8 log CFU/ml) increase in S. Enteriditis in nutrient solution after 21 h at pH 5 and 6 regardless of temperature. This is contrary to the present study where an increase in S. Javiana was not observed after 24 h in nutrient solution (pH ~6) across all temperatures. Conversely, a minimum log reduction between 0.04 and 0.14 in S. Javiana was recorded across all temperatures after 24 h. The discrepancy in Salmonella growth kinetics between the present study and Xylia et al. [13] could be due to differences between Salmonella serovars, inoculum preparation (2x wash in PBS in the present study versus suspension in BHI), or the nutrient solution composition. Unfortunately, the composition of the nutrient solution was not provided by Xylia et al. [13].

Similarly, Shaw et al. [22] evaluated the persistence of bacterial pathogens in nutrient solution; however, the authors only considered microbial persistence at a single temperature over a 24 h period. Specifically, Shaw and co-authors inoculated spent nutrient solution (i.e., nutrient solution utilized over 4 weeks for growing basil in a DWC system), fresh nutrient solution, and distilled water separately with E. coli O157:H7, non-O157 Shiga toxin-producing E. coli (STEC; cocktail including O26:H11, O45:H2, O103:H2, O111:H2, O121:H19, and O145NM), and Salmonella (cocktail of S. Enteritidis and S. Typhimurium). The inoculated nutrient solutions and water were held at 21°C, and samples were collected over the 24 h period. Shaw et al. (2016) reported an estimated 3 log CFU/ml increase for all pathogen types across all three solutions. E. coli O157:H7 bacterial load was higher in spent nutrient solution compared to water and fresh nutrient solution. Conversely, the concentrations of Salmonella and non-O157 STEC were higher in fresh nutrient solution and water compared to spent nutrient solution. This may be due to the presence of root associated microorganisms released into nutrient solution possibly leading to the suppression of Salmonella and non-O157 STEC in the spent nutrient solution used in the Shaw et al. (2016). Indeed, Dong et al. [22] reported the presence of a rich microbiome during hydroponic production of various types of microgreens and lettuce cultivars. This rich microbial community may impact pathogens due to nutrient competition or the presence of secondary metabolites secreted by the rhizosphere microbial community during hydroponic lettuce production [24,25]. However, reasons for the inconsistent impact of the microbial community across pathogen types are unclear and warrant further investigation.

In another study [14], mineral nutrient solution (MNS) was used for the hydroponic cultivation of radish. One of their objectives was to inoculate MNS (pH 6) with eight pathogenic and/or spoilage bacteria (Citrobacter freundii, Enterobacter spp., E. coli, Klebsiella oxytoca, Serratia grimesii, Pseudomonas putida, Stenotrophomonas maltophilia, and L. monocytogenes) and evaluate bacterial survival in MNS over a 28-day period. A single bacteria type was added to each system at 7 log CFU/ml. Due to diurnal variation, the average temperature of the MNS throughout production was 38.6°C (day) and 8.8°C (night). The study authors reported a microbial reduction of 3.1 to 5.1 log CFU/ml in MNS for all eight bacteria by the end of 28 days. Specifically, Settanni et al. [14] reported a 4-log reduction for L. monocytogenes after 28 days which is similar to the present study where L. monocytogenes decreased by up to 6 log CFU/ml across all temperatures within the 21-day period (Table 2 to 5).

In a study by Lopez-Galvez et al. [15], the microbiological quality of different irrigation water sources used during hydroponic production of tomatoes in greenhouses was analyzed. The irrigation water types included reclaimed water with or without fertilizer solution and surface water with or without fertilizer solution. Five separate fertilizer solutions were used during hydroponic production: monopotassium phosphate (F1), potassium nitrate (F2), calcium nitrate (F3), microelements solution (F4), and nitric acid (F5). The authors also collected drainage water samples (irrigated with reclaimed water or surface water) from growth substrates used during production. Water samples were collected over 13 weeks and analyzed for indicator bacteria (generic E. coli and Listeria spp.) and pathogenic bacteria (STEC, L. monocytogenes, and Salmonella spp.). The authors reported higher levels of Listeria spp. in both surface and reclaimed water with fertilizer solutions compared to water sources without fertilizers along with drainage water from the growth substrates. As a result, individual fertilizer solutions were then analyzed for the presence of Listeria spp., and elevated counts (4.5 log CFU/100ml) within the potassium nitrate solution were observed; however, none were positive for L. monocytogenes by real time PCR. Meanwhile, 8 out of 104 water samples were presumptively positive for Salmonella including reclaimed water with and without fertilizer, surface water with and without fertilizer, and drainage water from the rockwool growth substrate. During the study by Lopez-Galvez et al. [15], the ambient temperature of the greenhouse was between 15.0°C and 29.1°C, and electrical conductivity and pH were maintained at ≤ 2 dS/m and 5.5, respectively, within the hydroponic systems. However, it is difficult to determine the impact of temperature on bacterial persistence since the internal temperature of the nutrient solution within hydroponic systems were not recorded over the 13-week period. Overall, Lopez-Galvez and co-authors [15] demonstrate the potential for fertilizer solutions to introduce microbial contamination within a hydroponic system via recirculating nutrient solutions.

The recent outbreak and recall associated with hydroponic production of leafy greens has created greater awareness regarding the associated food safety challenges within CEA [3,26]. The present study indicates that temperature may be an important factor in the persistence of human pathogens in hydroponic nutrient solution during production. Specifically, the data demonstrate that temperature has a greater impact on the persistence of L. monocytogenes and L. innocua over a 21-day period compared to S. Javiana, specifically at lower temperatures such as 15 and 20°C. Based on these data, it could be anticipated that Salmonella might pose a greater food safety risk during hydroponic production of leafy greens when compared to Listeria spp. As a result, there may be a higher probability of internalization of Salmonella in leafy greens grown in hydroponic systems.