Survival of Listeria monocytogenes on Blueberries and Raspberries Stored at 4 °C and −18 °C

TL;DR

This study shows that Listeria monocytogenes can survive on blueberries and raspberries stored in the fridge or freezer, highlighting the need for better food safety practices.

Contribution

The study provides new insights into the survival patterns of L. monocytogenes on berries under refrigerated and frozen storage conditions.

Findings

L. monocytogenes populations remained stable at 4 °C but decreased significantly at −18 °C over time.

High-inoculated berries showed greater die-off at frozen temperatures compared to low-inoculated berries.

Die-off modeling revealed different decay patterns depending on inoculation level and storage temperature.

Abstract

Raw and minimally processed berries are not subjected to any processing kill steps and are stored under cold conditions to extend shelf-life. This study evaluated the growth and survival of high and low populations of L. monocytogenes on blueberries and raspberries stored under refrigerated and frozen temperatures. Fresh berries (10 g) were inoculated with a five-strain cocktail of L. monocytogenes to target 5.5 or 2 log CFU/g, stored at 4 and −18 ± 2 °C, and enumerated for up to 14 days post inoculation (dpi) at 4 °C and 168 dpi at −18 °C. Significant differences were evaluated (p ≤ 0.05), and die-off was modeled, using Davies test to determine breakpoints. No significant changes (p > 0.05) in L. monocytogenes populations were observed over time on berries stored at 4 °C, regardless of inoculation level. At −18 °C, significant reductions were observed over 168 dpi, with greater declines for high-inoculated blueberries and raspberries (2.63 and 2.13 log CFU/g, respectively), compared to those that were low-inoculated (0.67 and 0.46 log CFU/g, respectively). Die-off modeling indicated linear decay for both low-inoculated berry types and biphasic patterns for both high-inoculated frozen blueberries and high-inoculated raspberries stored at 4 °C. Results suggest that while L. monocytogenes does not grow, blueberries and raspberries support survival during refrigerated and frozen storage, emphasizing the need for stringent pre- and postharvest practices to limit contamination.

Figures

Full text

1. Introduction

In 2023, Washington State ranked first among states for the production of cultivated blueberries and second for raspberries [1]. Approximately 75% of Washington’s blueberry production is processed (which includes freezing), while the remaining 25% is sold in the fresh market and over 90% of the WA red raspberries production is frozen [2]. Due to the importance of berries to the farmgate of the state, any failure in food safety practices during berry production can have severe consequences for Washington growers. Since blueberries and raspberries grow in open-air field environments, they are susceptible to preharvest foodborne contamination due to their exposure to various external factors [3].

Foodborne pathogens, including Listeria monocytogenes, have been repeatedly isolated from preharvest environments, including soil, water, manure, and decaying vegetation, indicating that these reservoirs contribute to contamination risks for fresh fruits and vegetables [3,4,5,6]. Foodborne pathogens can also be introduced to, and persist within, produce packinghouses and processing environments [7,8,9,10], further increasing the risk of contamination throughout the production chain.

L. monocytogenes poses a significant food safety concern for raw and minimally processed (e.g., frozen) berries, due to its ability to persist and potentially grow at low temperatures like 4–5 °C [11]. L. monocytogenes has been repeatedly linked to outbreaks associated with fresh and minimally processed produce, including the 2011 cantaloupe outbreak and the 2014 caramel apple outbreak, as well as multiple recalls involving fresh, fresh-cut, and frozen fruit and vegetables [12,13,14,15,16]. While fresh berries have been infrequently linked to foodborne outbreaks and recalls, a 2023 incident involving frozen mixes of strawberries, blueberries, blackberries, and raspberries were recalled for suspicion of contamination with L. monocytogenes during postharvest handling and freezing [16]. In most cases, listeriosis causes mild illness; however, for at-risk populations, it can also lead to severe symptoms, including fever, headache, confusion, and miscarriage in pregnant women [17].

Previous studies have also demonstrated that the starting inoculation concentration may play a role in L. monocytogenes growth and/or survival patterns over time on produce surfaces [18,19,20,21,22]. Although high-inoculation concentrations in laboratory studies are required to observe substantial die-offs, such starting concentrations are less likely to mimic natural contamination scenarios. Thus, including both low and high-inoculation scenarios in L. monocytogenes research is essential for understanding bacterial survival, growth trends, and the effectiveness of risk assessment and pre- and postharvest intervention strategies. Therefore, the purpose of this research was to determine the growth and survival pattern of varying levels of L. monocytogenes on blueberries and raspberries stored at refrigeration and freezer temperatures, providing quantitative information relevant to contamination persistence, predictive modeling, and risk-based decision-making in the berry supply chain.

2. Materials and Methods

2.1. Experimental Design

Blueberries (Vaccinium sect. Cyanococcus) and raspberries (Rubus idaeus) were purchased from local markets on the same day of each experiment. To remove background microflora, berries were washed in 25 ppm of free chlorine (Sodium Hypochlorite; Clorox Company, Oakland, CA, USA) for 2 min, rinsed with distilled water, and air dried for 30 min before inoculation and storage. For each experimental combination, three independent biological replicate experiments were initiated on different days. For each of the three replications, triplicate samples were analyzed at each time point for the following combinations: berry (blueberry and raspberry), temperature (4 °C and −18 °C), and inoculation level [low (2 log CFU/g) and high (5.5 log CFU/g)].

2.2. Cold Storage

Cold storage temperatures of 4 °C and −18 °C were selected to represent standard refrigerated and frozen storage conditions commonly used in retail, household, and commercial cold-chain environments, as recommended by the Food and Drug Administration for proper food storage [23]. Refrigeration at approximately 4 °C has been previously employed in food safety research to evaluate the survival behavior of L. monocytogenes under proper cold storage conditions, as this temperature is known to significantly slow bacterial growth while allowing assessment of long-term persistence on fresh produce [11,20,21]. Similarly, frozen storage at temperatures near −18 °C (often approximated as −20 °C in microbial survival studies) is used to characterize pathogen persistence under extended frozen storage rather than complete inactivation.

2.3. L. monocytogenes Inoculum Preparation

A cocktail was prepared with five strains of L. monocytogenes that included LIS0234 (2012 recall of prepackaged raw, diced yellow onions food isolate [24]); LIS0133 (2010 outbreak of fresh-cut celery; environmental isolate [25]); LIS0110 (outbreak of whole cantaloupes; [13]); FSL107 (outbreak of coleslaw [26], and MDD262 (environmental isolate from cantaloupe packinghouse [27]). Strains were previously adapted to grow in the presence of 80 μg/mL of rifampicin (Thermo Fisher Scientific, Ward Hill, MA, USA). Each strain was streaked from frozen (−80 °C) onto tryptic soy agar (TSA; Difco Becton, Dickinson and Company, Sparks, MD, USA) supplemented with 80 μg/mL of rifampin (TSA-R) and was then incubated at 35 °C for 24 h. One isolated colony of each strain were each transferred into 10 mL of tryptic soy broth (TSB; Difco Becton, Dickinson and Company, Sparks, MD, USA) supplemented with 80 μg/mL of rifampin (TSB-R) and incubated at 35 °C for 24 h. Following incubation, 10 µL of liquid culture was later transferred into 10 mL of fresh TSB-R and incubated 35 °C for 24 h. Each broth culture was centrifuged at 3000× g for 5 min, after which the supernatant was discarded. The resulting pellet was washed twice with 5 mL of peptone (Difco Becton, Dickinson and Company, Sparks, MD, USA) and resuspended in 5 mL of peptone. The inoculum cocktail was prepared by combining equal volumes of each previously centrifuged strain to produce an inoculum of approximately 10^9^ CFU/mL. Depending on the target starting concentration, the cocktail was further diluted 100 to 100,000-fold in peptone water.

2.4. Inoculation of Samples

Blueberries or raspberries (10 g) were spot inoculated with 20 μL of the prepared L. monocytogenes cocktail on the surface. The target concentration for high-inoculation was ~5.5 log CFU/g and for the low-inoculation was 2 log CFU/g. Both blueberries and raspberries were then held at room temperature for 1 h under a biological hood to facilitate drying. Berries were either immediately processed (0 d) or transferred to sterile stomacher bags (Thermo Fisher Scientific, Ward Hill, MA, USA) and stored at 4 ± 2 °C and −18 ± 2 °C.

2.5. Enumeration of L. monocytogenes

Populations of L. monocytogenes on blueberries and raspberries were enumerated at 0, 0.167 (4 h), 1, 2, 3, 5, 7, 10, and 14 d for 4 ± 2 °C and 0 d, 1, 3, 7, 14, 28, 56, 84, 112, 140, and 168 d for −18 ± 2 °C. Blueberries and raspberries were hand massaged in 20 mL of 0.1% peptone water for 90 s, alternating between 30 s of rubbing, 30 s of shaking, and 30 s of rubbing. Subsequent dilutions were made using 0.1% peptone water. Samples were plated in duplicate onto TSA-R and incubated at 30 ± 5 °C for 48 h. An additional 0.25 mL from the bag was also plated, in duplicate, onto four TSA-R agar plates each (1 mL total) to lower the limit of detection (0.48 log CFU/g).

2.6. Statistical Analysis

All statistical analysis were conducted using R (version 2024.04.2). Descriptive statistics including the mean and standard deviation of log CFU/g were calculated for each berry type, temperature, timepoint and inoculum combination. Analyses were made comparing populations (CFU/g) at day 0 for each level of inoculation and berry type (5.69 ± 0.50 high-inoculated blueberries; 2.12 ± 1.20 low inoculated blueberries; 4.78 ± 0.56 high inoculated raspberries; and 1.77 ± 0.73 for low inoculated raspberries). To assess significant differences in L. monocytogenes counts, Tukey’s Honest Significant Difference post hoc test was applied (p ≤ 0.05). Additionally, to model L. monocytogenes daily die-off rate, a Davies test was used to determine if linear models for each combination were significant for a nonconstant regression parameter (breakpoint). If the Davies tests were significant for at least one breakpoint, a segmented linear regression model was built.

3. Results

3.1. Survival of L. monocytogenes on High-Inoculated Blueberries and Raspberries

At 0 d, initial L. monocytogenes levels on high-inoculated blueberries and raspberries were 5.69 ± 0.50 and 4.78 ± 0.56 log CFU/g, respectively (Figure 1, Table S1). Due to decay, high-inoculated raspberries held at 4 °C could not be enumerated on 14 d. At 4 °C, no significant changes in high-inoculated L. monocytogenes concentrations were observed on either blueberries or raspberries over the entire study duration. In contrast, storage at −18 °C led to statistically significant decreases in L. monocytogenes concentrations on blueberries and raspberries over time (Figure 1, Table S1).

On high-inoculated blueberries stored at −18 °C, L. monocytogenes populations were significantly different on 1 d (4.39 ± 1.72 log CFU/g) compared to day 0. However, populations on 2 d (4.71 ± 1.22 log CFU/g) and 7 d (4.86 ± 0.90 log CFU/g) were not significantly different from the initial levels. From 14 d onward, L. monocytogenes counts showed statistically significant reductions relative to 0 d. By 168 d, the population declined to 3.06 ± 0.73 log CFU/g, representing a total reduction of 2.63 log CFU/g or about 46.2% of the population density. For high-inoculated raspberries stored at −18 °C, a statistically significant reduction in L. monocytogenes compared to 0 d was observed at 56 d, when levels declined to 3.38 ± 0.87 log CFU/g. Although further decreases were observed over time, with a loss of 44.56% of the population, reaching 2.65 ± 0.36 log CFU/g by 168 d, the concentrations from day 56 onward were not significantly different.

3.2. Survival of L. monocytogenes on Low-Inoculated Blueberries and Raspberries

For the low-inoculum, the initial concentration of L. monocytogenes at 0 d on blueberries and raspberries were 2.12 ± 1.20 and 1.77 ± 0.73 log CFU/g, respectively (Figure 2, Table S2). Due to decay, low-inoculated raspberries held at 4 °C could not be enumerated on 10 d and 14 d. Similarly to high-inoculated berries, no significant differences in low-inoculated L. monocytogenes concentrations were observed on either blueberries or raspberries held at 4 °C over the study duration. L. monocytogenes populations on blueberries at −18 °C fluctuated over time, with the greatest occurring within the first seven days post-inoculation. A mild numerical increase was observed at day 14 that did not differ significantly from day 0 (p > 0.05). However, by 168 d of storage, L. populations significantly decreased by 0.67 log CFU/g, reaching 1.45 ± 0.89 log CFU/g. For low-inoculated raspberries stored at −18 °C, a statistically significant reduction in L. monocytogenes compared to 0 d was observed by 10 d (1.64 ± 0.73 log CFU/g). By 168 d, the population significantly declined to 1.31 ± 0.78 log CFU/g, representing a reduction of 0.46 log CFU/g.

3.3. L. monocytogenes Die-Off Models

For berries held at refrigeration temperatures, only the high-inoculated raspberries were significant for a breakpoint in the L. monocytogenes die-off pattern by the Davies test, with one significant breakpoint at 1.31 d (95% confidence interval [CI]: 0.07, 2.55 d; Table 1). Based on the biphasic die-off pattern, L. monocytogenes demonstrated an initial daily die-off rate of −0.38 log CFU/g/day (CI: −0.84, −0.08) from 0 to 1.31 d. Following 1.31 d, the estimated rate of change for L. monocytogenes was 0.05 log CFU/g/day (CI: −0.02 to 0.11), indicating a non-significant increase in population over time. The high- and low-inoculated blueberries, as well as the low-inoculated raspberries stored at 4 °C, did not exhibit significant breakpoints, indicating a better fit by a linear pattern. The daily log-linear rate of change for L. monocytogenes on blueberries for high starting inoculation levels was significant (p = 0.04) with a die-off of −0.04 log CFU/g/day (CI: −0.04, −0.03). However, the daily rate of change for the low-inoculated blueberries and raspberries were non-significant (p = 0.06 and 0.20, respectively).

For berries stored at frozen temperatures, high-inoculated blueberries and raspberries each exhibited one significant breakpoint in L. monocytogenes based on the Davies test, whereas no significant breakpoints were observed for low-inoculated berries (Table 1). The daily log-linear die-off rates of low-inoculation of L. monocytogenes on blueberries and raspberries at −18 °C were −0.00 (CI: −0.00, 0.00) and −0.01 (CI: −0.01, 0.00) log CFU/g/day. However, while the rate was statistically significant for raspberries (p ≤ 0.01), it was not for blueberries (p = 0.18), demonstrating a stable L. monocytogenes population on frozen blueberries under low-inoculation conditions. High-inoculated frozen blueberries showed a biphasic die-off pattern with a daily die-off of −0.10 log CFU/g/day (CI: −0.11, −0.01) until 15.54 d (CI: 8.08, 23.01), followed by a slower decline of −0.01 log CFU/g/day (CI: −0.14, −0.01) for the remainder of the study duration (168 d). For high-inoculated raspberries, the die-off was −0.02 log CFU/g/day (CI: −0.03, −0.01) from day 0 to day 84.00 (CI: 55.77, 112.22), followed by a non-significant change for the reminder of the 168 d (−0.00 log CFU/g/day; CI: −0.01, 0.00).

4. Discussion

The results of the present study showed that fridge and freezer conditions did not support the growth of L. monocytogenes populations on the surface of blueberries and raspberries. Previous research has suggested that L. monocytogenes growth, survival, and die-off dynamics on intact produce surfaces, including berries, differs substantially by commodity type, storage temperature, and surface characteristics [28,29,30,31]. For example, after spot inoculating the surface of whole strawberries at 6 or 8 log CFU/g and storing at 4 °C for 7 d or 24 °C for 48 h, L. monocytogenes did not grow under any of the conditions tested but remained detectable, indicating survival [30]. A study looking at fresh highbush blueberries stored at 4 °C or 12 °C found that L. monocytogenes, initially inoculated at ~6 log CFU/g, gradually decreased by a total of 0.5–1 log CFU/g over a 10 d period [31]. On the other hand, L. monocytogenes (spot inoculated at ~3.5–4.5 log CFU/g) grew and survived on the intact surfaces of 9 out of 10 whole fruits and vegetables tested (i.e., blackberry, raspberry, blueberry, lemon, mandarin orange, sweet cherry, tomato, cauliflower, and broccoli; but, not carrot) at 2 °C, 12 °C, and 22 °C showing an initial growth followed by subsequent declines [32]. Specifically, at 2 °C, L. monocytogenes on blueberries and raspberries showed small but significant initial increases of 0.64 and 0.42 log CFU/g before declining by 1.24 and 1.62 log CFU/g, respectively over 28 d [32].

It is important to note that while significant growth was not observed under refrigerated conditions, significant reductions were also not observed, demonstrating that populations remained stagnant over up to 14 d. This is also echoed by the segmented model results, demonstrating that, other than the initial phase of the biphasic pattern for high-inoculated raspberries, rates of change were either not statically significant (p ≥ 0.05) or not biological significant, as the magnitude of −0.04 log CFU/g/d does not have a meaningful impact over 7–14 d. Nonetheless, good agricultural practices and good manufacturing practices remain critical for preventing initial contamination.

Variations in population dynamics between the present study and previous berry studies may reflect differences in the initial L. monocytogenes populations, as the current study observed distinct differences in patterns between the two inoculation levels tested. The present study used two initial inoculation levels: ~5.5 log CFU/g and ~2 log CFU/g. Previous research examining inoculation type (e.g., dip vs. spot; [33], wet vs. dry [32]) and inoculation levels [18,19,21,34] for fresh produce have shown that these parameters substantially influence L. monocytogenes die-off, survival, and/or growth patterns. Findings from the present study indicate that L. monocytogenes on high-inoculated berries generally followed a biphasic die-off pattern characterized by an initial period of faster decline followed by a slower, prolonged decline or stable phase. In contrast, berries inoculated at lower levels were better described by a consistent linear pattern, showing stability or decreases that were not biologically meaningful over time.

When data from 29 published journal articles representing 130 data sets for L. monocytogenes on fresh produce was modeled to understand factors influencing growth rate, storage temperature as well as initial and final cell concentrations were the remaining factors in the final, best fitting model [34]. Model outputs indicated that initial cell concentration versus growth rate had a linear rate of −0.72, demonstrating that as initial inoculation concentration increases the maximum growth rate tends to decrease. Additionally, when L. monocytogenes was inoculated onto parsley at 3, 7, or 8 log CFU/leaf and stored at 20 °C with 100% relative humidity (RH), populations converged to ~5 log CFU/leaf regardless of the initial level, whereas under 45% RH, populations declined in a biphasic manner with an initial inoculum of 7 and 9 log CFU/leaf and linearly for 3 log CFU/leaf [18]. High-inoculation increases competition for limited nutrients and resistant bacteria use the available resources rapidly, causing faster initial die-offs [35,36,37]. Previous studies have indicated that contamination of fresh produce with L. monocytogenes occurs at relatively low concentrations, with the pathogen often detected at levels below 1–2 log CFU/g when present [38,39]. Therefore, the use of lower initial inoculum concentrations in laboratory studies has been recommended to better reflect realistic contamination scenarios and to more accurately characterize the growth and/or survival dynamics of L. monocytogenes on produce surfaces, in order to make the resulting estimates of pathogen behavior, risk, and intervention efficacy more relevant to real-world produce handling and food safety decision-making [38,39,40,41].

The recommended storage temperature for fresh berries is 0–4 °C, while for frozen it is −18 °C or lower, as these conditions slow microbial growth by decreasing the rate of the enzymatic and metabolic reactions necessary for its microbial replication, thus minimizing spoilage and pathogen proliferation [29,42]. The results of the present study show that no significant growth was supported by berries under recommended storage conditions and significant reduction was observed on both blueberries and raspberries at −18 °C, with viable counts by the end of the experiments. This persistence is likely associated with the presence of sublethal injured or stressed cells as a survival strategy that allows cells to remain alive with a slow-downed metabolism rather than full inactivation. Temperature stress responses and adaptive mechanisms further contribute to the ability of L. monocytogenes to survive adverse conditions encountered during the storage and processing of foods [37,43]. Research has shown that as the storage temperature of produce increases, so does the growth rate. Igo et al. [34] reported a linear relationship between produce storage temperature and maximum L. monocytogenes growth rate, with the growth rate increasing by 0.081 log CFU/d for every 1 °C rise in temperature. This was also observed on broccoli and cauliflower, where L. monocytogenes growth rates were 1.69, 0.25, 0.05, and −0.01 log and 2.16, 0.43, 0.10, and −0.01 log CFU/g/d, respectively, at 23 + 2, 12 ± 2, 4 ± 2, and −18 ± 28 °C [20]. Although temperatures above the recommended storage conditions were not examined in the present study, it was reported previously that L. monocytogenes populations on blueberries decreased over 14 d at all temperatures tested; however, final populations were significantly higher on blueberries stored at 15 °C compared with those stored at 4 °C and 10 °C [42]. Thus, the use of temperature as a control measure as part of a hurdle approach to manage L. monocytogenes risk for blueberries and raspberries remains essential to effectively suppress microbial growth and reduce the likelihood of pathogen proliferation.

5. Conclusions

Overall, results of the present study demonstrate that refrigeration and frozen storage conditions do not support the growth of L. monocytogenes on the surfaces of blueberries and raspberries, regardless of initial contamination level. Under refrigeration, L. monocytogenes populations remained largely stable over time, while frozen storage resulted in gradual but statistically significant reductions, particularly for high-inoculated berries. Differences in survival patterns between high and low-inoculum levels underscore the influence of initial contamination on die-off dynamics, with high-inoculation generally producing biphasic declines and low-inoculation resulting in more stable or minimally declining populations. The present study reinforces the critical role of temperature control in limiting L. monocytogenes growth. L. monocytogenes populations remained viable throughout storage, even under frozen conditions, highlighting the critical need for robust good agricultural practices and stringent manufacturing and handling controls that limit the initial contamination of berries. However, this study has several limitations. Laboratory conditions did not fully replicate commercial handling and storage environments, which can include fluctuating temperatures, freeze–thaw cycles, and interactions with naturally occurring fungi and other microorganisms. Future research should address these limitations by investigating the effects of variable temperature conditions, freeze–thaw events, and pathogen-fungal interactions to better simulate commercial fruit decay and storage scenarios.