The glutamate decarboxylase acid resistance mechanism affects survival of Listeria monocytogenes LO28 in modified atmosphere-packaged foods
Gillian A. Francis, Food Science Research Centre, Department of Life Sciences, University of Limerick, Limerick, Ireland.
Aims: The contribution of the glutamate decarboxylase (GAD) acid resistance system to survival and growth of Listeria monocytogenes LO28 in modified atmosphere-packaged foods was examined.
Methods and Results: The survival and growth of the wild-type LO28 and four GAD deletion mutants (ΔgadA, ΔgadB, ΔgadC, ΔgadAB) in packaged foods (minced beef, lettuce, dry coleslaw mix) during storage at 4, 8 and 15°C were studied. Survival and growth patterns varied with strain, product type, gas atmosphere and storage temperature. In minced beef, the wild-type LO28 survived better (P < 0·05) than the GAD mutant strains at 8 and 15°C. In both packaged vegetables at all storage temperatures, the wild-type strain survived better (P < 0·05) than the double mutant ΔgadAB. The requirement for the individual gad genes varied depending on the packaged food. In the case of lettuce, gadA played the most important role, while the gadB and gadC genes played the greatest role in packaged coleslaw (at 15°C).
Conclusions: This work demonstrates that elements of the GAD system play significant roles in survival of L. monocytogenes LO28 during storage in modified atmosphere-packaged foods.
Significance and Impact of the Study: A better understanding of how L. monocytogenes behaves in modified atmosphere-packaged foods, and how it responds to elevated carbon dioxide atmospheres.
Modified atmosphere packaging (MAP) involves packaging foods in a modified gaseous mixture chosen to control the product’s biochemical and enzymatic reactions, reduce microbial spoilage and extend the shelf-life. The elevated carbon dioxide (CO2) concentrations within modified atmosphere (MA) packages of food may inhibit the growth of a number of micro-organisms (Daniels et al. 1985). The extent of inhibition by CO2 varies with the micro-organism, CO2 concentration, temperature of incubation and type of food (Enfors et al. 1979; Dixon and Kell 1989; Hudson et al. 1994).
A number of studies have indicated that MAP can enhance the survival and growth of facultatively anaerobic pathogens, such as Listeria monocytogenes (Hintlian and Hotchkiss 1986; Brackett 1994; Francis and O’Beirne 1997). In addition, L. monocytogenes has been shown to survive and grow on a range of packaged meats (Wimpfheimer et al. 1990; Devlieghere et al. 2001; Mataragas et al. 2003) and minimally processed vegetables (Beuchat and Brackett 1990; Carlin et al. 1995; Francis and O’Beirne 1997, 2001a; Jacxsens et al. 1999) during storage at refrigeration temperatures.
In addition to its ability to grow at low temperatures (Bell and Kyriakides 1998), L. monocytogenes is able to survive a variety of other environmental stresses, including low pH. One well-characterized adaptive response is to acid stress, the so-called acid tolerance response (ATR), in which L. monocytogenes cells exposed to pH 5·5 for a short period can withstand a normally lethal pH environment (Hill et al. 1995; Hill and Gahan 2000). The ability of L. monocytogenes to induce an ATR has been shown to play an important role in its survival in acidic foods and in its survival in vivo (Gahan et al. 1996; Gahan and Hill 1999; Hill et al. 2002). In addition, acid-adapted L. monocytogenes cells demonstrated enhanced survival in MA packages of vegetables which had elevated in-pack CO2 levels (i.e. 30%) (Francis and O’Beirne 2001b).
Listeria monocytogenes is one of a relatively small number of bacteria known to possess a glutamate decarboxylase (GAD)-dependent stress response system that plays a major role in its acid tolerance. When the cell is exposed to low pH, GAD catalyses the conversion of glutamate to γ-amino butyrate and CO2, thus removing a proton from the cytoplasm and resulting in an increase in cytoplasmic pH. Previous investigations have shown that L. monocytogenes LO28 possesses a GAD system that contains at least two independently regulated GAD, encoded by the gadA and gadB genes, and a putative glutamate/γ-amino butyrate antiporter, encoded by gadC (Cotter et al. 2001a). More recently, Cotter et al. (2005) demonstrated that strain LO28 possesses two additional gad genes, and contains a total of three decarboxylase enzymes and two antiporters, as does the sequenced L. monocytogenes EGDe (Glaser et al. 2001).
Analysis of a set of four GAD deletion mutants demonstrated a dramatic decrease in the level of GAD activity in the order of wild-type LO28 > ΔgadA > ΔgadB = ΔgadC > ΔgadAB, with a significantly enhanced acid sensitivity of the mutants, indicating that the GAD system may play an important role in gastric transit, and is necessary for prolonged survival of L. monocytogenes in acid foods (Cotter et al. 2001a,b). In a recent study, Jydegaard-Axelsen et al. (2004) observed an increase in expression of the gad genes (gadA, gadB, gadC) when strain LO28 was grown in CO2, suggesting that strain LO28 responds to CO2 as it would to acid using the GAD system.
In this paper, the contribution of the GAD system to the survival and growth of L. monocytogenes LO28 during storage in MA-packaged food products was examined. The effects of product type (minced beef, lettuce, coleslaw mix), gas atmospheres (varying CO2 levels) and storage temperatures (4, 8 and 15°C) were also investigated.
Materials and methods
Preparation of the food products
Minced beef The meat sample (lean beef from the shin and shank) was minced in the on-site abattoir. It was then transferred back to the laboratory in a sterile sealed bag to prevent cross-contamination.
Lettuce and dry coleslaw mix Heads of Irish Iceberg lettuce were purchased from a local supplier. Outer and damaged leaves and the core of the lettuce heads were removed and discarded. Inner leaves were sliced manually using a sharp knife to approximately 10-mm strips. Dry coleslaw mix, consisting of 80% shredded cabbage and 20% shredded carrot, was obtained from a local supplier.
The wild-type strain was L. monocytogenes LO28. Nonpolar mutants of gadA, gadB and gadC with 336, 405 and 444 bp deletions respectively, were created individually (Cotter et al. 2001a). LO28 ΔgadAB was a double mutant in which both gadA and gadB genes had internal deletions, resulting in negligible GAD activity. The culture medium for bacterial growth was tryptone soya broth (CM 129), supplemented with 0·6% yeast extract (Oxoid L21, TSB-YE).
Preparation of inocula
Listeria monocytogenes strains were maintained at −20°C in TSB-YE supplemented with 15% (v/v) glycerol. Resuscitation was achieved by thawing cultures at room temperature (17–22°C) followed by a loop transfer in TSB-YE (10 ml) and overnight incubation at 37°C. Cultures were centrifuged (5000 g, 15 min), the cells were washed once with sterile phosphate-buffered saline (PBS; Oxoid BR014), resuspended and diluted in PBS to the desired concentrations to allow for contamination of food products at initial levels of approximately 105 CFU g−1 of the product.
Inoculation of meat and vegetable products and storage
Minced beef Meat portions (750 g) and 30 ml of the cell suspensions were transferred aseptically into a sterile blender and blended to ensure an even distribution of the inoculum. Inoculated meat portions (25 g) were transferred aseptically into bags (12 cm × 12 cm), composed of oriented polypropylene packaging film (Cannings Packaging Ltd., Dublin, Ireland), which were later heat-sealed after flushing bags with 20% CO2 and 80% oxygen (O2).
Lettuce and dry coleslaw mix Vegetable portions (25 g) were transferred aseptically into bags (10 cm × 10 cm), composed of 35 μm oriented polypropylene packaging film (Cannings Packaging Ltd.). After appropriate dilution, ten 10-μl aliquots of the cell suspensions were distributed over the vegetables contained within the packages. After inoculation, packages were heat-sealed and were gently shaken to assist inoculum distribution. Immediately after sealing, all food samples were transferred to storage temperatures of 4, 8 and 15°C and stored for a period of 12 days.
Microbiological analyses were carried out on day 0 (day of inoculation) and at regular intervals throughout the storage period. At each sampling, duplicate packs from the same experiment were analysed for populations of L. monocytogenes and mesophilic bacteria. The 25-g sample from each pack was aseptically transferred into a stomacher bag. Samples were homogenized for 2 min at high speed with 225 ml of sterile PBS using a Seward laboratory stomacher (Model 400; AGB Scientific, Ireland). Serial dilutions of each homogenized sample were made in PBS and were surface spread (100 μl per plate) in duplicate onto appropriate media. Numbers of L. monocytogenes were determined on Listeria-selective agar (LSA; Oxoid CM 856 and SR140) after incubation for 48 h at 35°C. Total counts of mesophilic bacteria were enumerated on plate count agar (PCA; Oxoid, CM 325) after incubation at 35°C for 48 h.
Gaseous atmospheres inside the packages
On each sampling date, gases within three of each package type were analysed using an O2 and CO2 gas analyser (Model TIA-III LV, PBI-Dansensor; PBI Development, Ringsted, Denmark).
pH of food samples
On each of the sampling dates, the pH of the food samples was analysed. A 10-g sample from each package was homogenized for 2 min with 90 ml of distilled water using a Waring blender. The pH of the homogenized sample was measured during storage using a digital pH meter (pH meter 3310; Jenway Ltd, Dunmow, UK).
Measurement of levels of free glutamate in foods
The levels of free l-glutamic acid in the food samples were measured using an l-glutamic acid food analysis kit (Boehringer Mannheim, Indianapolis, USA) according to the manufacturer’s instructions.
All experiments were carried out in duplicate and replicated thrice. On each analysis date, duplicate samples were serially diluted and plated in duplicate. Reported populations therefore represent the means of six values. Figures show the means of six values ± standard deviations (SD). Results were analysed by analysis of variance and least significant difference testing at the P < 0·05 level.
In order to determine whether the GAD system plays a role in the survival of L. monocytogenes in MAP foods, the wild-type strain LO28 and a set of four mutant strains were inoculated onto MAP meats and vegetables, and differences in their survival rates were followed during storage. Figure 1 shows the survival of the wild-type strain, and the GAD deletion mutants on packaged minced beef during storage at 4, 8 and 15°C. At 4°C, the GAD mutants behaved similar to the wild-type strain, with all populations decreasing during storage (Fig. 1a). At 8 and 15°C, however, the wild-type strain survived significantly better (P < 0·05) than the mutant strains, but there was no major difference between mutants.
Figure 2 shows survival and growth of the wild-type and GAD mutant strains on packaged lettuce during storage at 4 and 8°C. At 4°C, the ΔgadB and ΔgadC mutants behaved similar to the wild-type strain, with populations not changing significantly (P > 0·05) during storage on lettuce (Fig. 2a). However, growth of the ΔgadA and ΔgadAB mutants was significantly (P < 0·05) reduced during storage, with ΔgadAB demonstrating a 1-log reduction in populations during storage. At 8°C, there was no significant difference (P > 0·05) in the survival and growth of the wild-type or ΔgadB and ΔgadC strains; populations of all strains increased by approximately 1-log cycle during storage (Fig. 2b). However, ΔgadA and ΔgadAB did not grow as well as the other mutants. Populations of ΔgadA increased between days 0 and 2, however, numbers declined in the storage thereafter to initial population densities. Populations of the ΔgadAB mutant, which had the two listerial GAD homologues deleted, remained at initial levels throughout the storage period.
The survival of the wild-type and mutant strains on MAP coleslaw mix was also examined during storage at 4, 8 and 15°C (Fig. 3). For all storage temperatures, populations of the wild-type and mutant strains decreased (P < 0·05) during storage. At 4°C, the ΔgadA, ΔgadB and ΔgadC mutants behaved similar to the wild-type strain, with all populations declining by approximately 1-log cycle during storage (Fig. 3a). However, populations of the ΔgadAB mutant were significantly lower (P < 0·05) than the wild-type or other mutant strains, with numbers decreasing by 2-log cycles during the storage period. At 8°C, ΔgadA behaved similar to the wild-type strain with numbers declining by 1·5-log cycles during storage. The ΔgadB and ΔgadC mutants behaved similar to the wild-type strain up to day 9; however, by day 12, numbers were 1-log cycle lower. Populations of ΔgadAB were significantly lower (P < 0·05) than the wild-type or other mutant strains, with numbers declining by 3-log cycles during the storage period. During storage at 15°C, GAD deletion mutations resulted in a dramatic decrease in survival on packaged coleslaw, where the order of survival was LO28 > ΔgadA > ΔgadC > ΔgadB > ΔgadAB. Viable populations of the wild-type LO28 remained at the end of the 12-day storage period. However, viable populations of ΔgadC were undetectable on day 9, while populations of ΔgadB and ΔgadAB could not be detected on days 7 and 5, respectively. The results indicate that the wild-type strain survived and/or grew significantly better (P < 0·05) than the double mutant ΔgadAB on both packaged vegetables at all storage temperatures. Thus, the GAD system plays an important role in the survival of L. monocytogenes during storage in MAP vegetables.
The fresh-cut vegetables were sealed in packages initially enclosing air. During storage, the gas atmospheres within the packages were modified, mainly as a result of the respiration of the vegetables. The concentrations of O2 and CO2 achieved within packs varied with the packaged product and storage temperature. With packages of lettuce stored at 4 and 8°C, O2 levels fell to 2–5% and 1–3%, and CO2 levels increased to 9–10% and 12–13%, respectively. Dry coleslaw mix had a higher respiration rate and higher levels of CO2 were reached in packs of coleslaw than those attained in packs of lettuce. With coleslaw, CO2 levels rose to 12–13% during storage at 4°C, and to 25–27% and 34–35% during storage at 8 and 15°C, respectively. In the case of minced beef, packages were flushed with 20% CO2 and 80% O2 and these gas levels persisted during storage.
The total aerobic mesophilic microflora proliferated on all products during storage (Fig. 4). Generally, increasing the storage temperature from 4°C to 15°C increased the rate and extent of microbial growth. Initial microbial counts on all products were approximately 6 log CFU g−1. Populations on minced beef increased to 7·5, 8·5 and 9·0 log CFU g−1 during storage at 4, 8 and 15°C, respectively. On lettuce, microbial counts reached final population densities of c. 7·3 log CFU g−1 at 4 and 8°C, and 8·0 log CFU g−1 at 15°C. On coleslaw mix, counts of mesophilic microflora also increased more rapidly at the higher temperatures; however, no difference in final population densities was observed.
Changes in pH of food samples were measured during storage and they varied depending on the packaged product (Fig. 5). The initial pH of minced beef was c. 6·0 and it did not vary over the duration of storage, regardless of storage temperature. The pH of the lettuce samples increased during storage from 5·7 to a maximum of 6·6, and there was no effect of storage temperature on pH changes. In the case of coleslaw mix, however, storage temperature had a major effect on changes in pH. At 4°C, the initial pH was c. 6·2 and it remained at this approximate value throughout the storage period. At 8°C, the pH gradually dropped from c. 6·2 on day 0 to pH 5·8 on day 5, and to pH 4·1 on day 12. At 15°C, the pH decreased rapidly from c. 6·2 to pH 4 on day 5 and it stayed at pH 4 for the remaining storage period.
The glutamate levels were also measured; free glutamate levels per 100 g were 11, 10 and 7 mg, respectively for coleslaw, minced beef and lettuce.
This work demonstrates reduced survival and/or growth of the GAD mutant strains in a number of MAP foods, especially in those containing elevated CO2 levels. Fernández et al. (1997) reported that elevated CO2 concentrations were inhibitory to L. monocytogenes at refrigeration temperatures. CO2 inhibits the growth of bacteria by: (i) affecting cellular enzymes and decreasing the rate of metabolic reactions (Ranson et al. 1960); (ii) CO2 product repression of carboxylases and decarboxylases (King and Nagel 1975); (iii) disrupting cell membrane structural integrity and/or specific functions (Sears and Eisenberg 1961); (iv) decreasing the substrate and intra-cellular pH (Wolfe 1980), or by a combination of these mechanisms (Dixon and Kell 1989).
The survival/growth patterns of the wild-type LO28 and the mutant strains varied depending on the type of packaged food. The reasons for the differences in survival in the various foods may be related to a number of factors including, different CO2 levels, different product pH values, different substrate compositions, nutrient (glutamate) concentrations and availability and/or the indigenous microflora and associated competitive interactions on foods. Differences in survival between the wild-type LO28 and the mutants were especially evident in packaged coleslaw during storage at 15°C, where the order of survival was LO28 > ΔgadA > ΔgadC > ΔgadB > ΔgadAB. The reasons for the reduced survival of the mutants were presumably because of direct inhibitory effects of CO2, as well as indirect effects of CO2 resulting in acidification of the coleslaw. From the gas analysis and pH studies, the high CO2 levels (c. 35%) within the coleslaw packages at 15°C resulted in rapid decrease of coleslaw pH. An important factor contributing to bacterial survival in acidic and other stressful environments is the ability to regulate their intracellular pH (pHi) and membrane potential, a process primarily driven by the controlled movement of cations across the cell membrane (Hill et al. 1995). If CO2 growth inhibition is caused by acidification of the food, the acid-tolerant strain LO28 most likely used the GAD system to re-establish pHi, and differences in survival of the various GAD mutants were presumably because of differences in acid resistance and ability to control pHi. Jydegaard-Axelsen et al. (2004) observed an increase in expression of GAD genes (gadA, gadB, gadC) in L. monocytogenes LO28 grown in atmospheres of 100% CO2 or 100% nitrogen (N2). The similarity in response towards CO2 and N2 indicated that L. monocytogenes LO28 responded to CO2 as it would to acid using the GAD system, and indicates overlap between the CO2 response and the response to acidity and anaerobiosis, respectively.
The requirement for the individual gad genes varied depending on the food and/or storage conditions. The packaged coleslaw contained anti-listerial shredded carrots (Francis and O’Beirne 2001a), had low pH values and high in-pack CO2 levels (at 15°C) and under these stressful conditions, the gadB and gadC genes played an important role. In the case of packaged lettuce, gadA played the most obvious role. The packaged lettuce may be considered a less stressful environment for Listeria, as it had a higher pH and the packages contained lower CO2 levels and in this environment, gadA appeared to play the greatest role. Cotter et al. (2005) reported similar results; gadA played a role in survival/growth at relatively mild acidic conditions, whereas the gadB and gadC genes were required for survival in low pH conditions. While the CO2 levels within packs of lettuce were not very high, they appeared to be high enough to reduce the growth of the ΔgadA and gadAB mutants. Other factors such as the presence of competitive microflora on lettuce, dominated mainly by enterobacteriaceae and pseudomonads (Francis and O’Beirne 1998a; Jacxsens et al. 2003), may have caused additional stress and reduced survival/growth of these mutants (Francis and O’Beirne 1998a,b).
The GAD mutants behaved similar to the wild-type strain on minced beef stored at 4°C, but their survival was impaired at 8 and 15°C. The reason for reduced survival of the mutants at 8 and 15°C was presumably because of direct inhibitory effects of elevated CO2 levels, although it is unclear why no inhibitory effects were observed at 4°C. The additional impact of the rapidly growing competitive mesophilic microflora on the meat at 8 and 15°C possibly contributed to the generation of more stressful conditions for the mutants because of nutrient depletion, changes in the chemical composition of the product and/or production of bactericidal agents. The high levels of CO2 within packages of meat combined with the higher temperatures may have selected for growth of lactic acid bacteria (LAB) (Francis and O’Beirne 1998b), and allowed the more acid-resistant wild-type to survive better than the acid-sensitive mutants (Vescovo et al. 1996; Francis and O’Beirne 1998a,b).
This study demonstrates that the GAD acid resistance system plays an important role in the survival and/or growth of L. monocytogenes LO28 in MAP foods, especially those containing elevated CO2 atmospheres. The reasons for reduced survival/growth of the mutants in the different packaged products appear because of a number of reasons including, direct inhibitory effects from high CO2 levels, indirect effects of CO2 resulting in acidification of the product and possible promotion of growth of competitive microflora such as LAB, and/or general inhibitory effects from the competitive indigenous microflora of foods. Each gad gene played a different role and the requirement for individual gad genes varied depending on the food or environment. In more stressful environments (i.e. higher CO2 conditions), the gadB and gadC genes played the greatest role, whereas in less stressful environments (i.e. lower CO2 levels) gadA appears to be more important.
This work was funded by the Irish Government under the National Development Plan 2000–2006.