If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Assessment of the response of indigenous microflora and inoculated Bacillus licheniformis endospores in reconstituted skim milk to microwave and conventional heating systems by flow cytometry
Heat treatment is one of the most widely used processing technologies in the dairy industry. Its primary purpose is to destroy microorganisms, both pathogenic and spoilage, to ensure the product is safe and has a reasonable shelf life. In this study microwave volumetric heating (MVH) was compared with a conventional tubular heat exchanger (THE), in terms of the effects of each at a range of temperatures (75°C, 85°C, 95°C, 105°C, 115°C, and 125°C) on indigenous microflora viability and the germination of inoculated Bacillus licheniformis endospores in reconstituted skim milk. To assess the heat treatment–related effects on microbial viability, classical agar-based tests were applied to obtain the counts of 4 various microbiological groups including total bacterial, thermophilic bacterial, mesophilic aerobic bacterial endospore, and thermophilic aerobic bacterial endospore counts, and additional novel insights into cell permeability and spore germination profiles post-heat treatment were obtained using real-time flow cytometry (FC) methods. No significant differences in the plate counts of the indigenous microorganisms tested, the plate counts of the inoculated B. licheniformis, or the relative percentage of germinating endospores were observed between MVH- and THE-treated samples, at equal temperatures in the range specified above, indicating that both methods inactivated inoculated endospores to a similar degree (up to 70% as measured by FC and 5 log reduction as measured by plate counting for some treatments of inoculated endospores). Furthermore, increased cell permeability of indigenous microflora was observed by FC after MVH compared with THE treatment of uninoculated skim milk, which was reflected in lower total bacterial count at a treatment temperature of 105°C. This work demonstrates the utility of FC as a rapid method for assessing cell viability and spore inactivation for postthermal processing in dairy products and overall provides evidence that MVH is at least as effective at eliminating native microflora and inoculated B. licheniformis endospores as THE.
Conventional pasteurization methods using HTST combinations are a well-established dairy industry standard for ensuring adequate food safety in short shelf life refrigerated products. On the other hand, the advent of high heat temperature (HHT, <135°C) technologies and the ability to sterilize fluid milk has allowed for liquid dairy products to be stored for longer periods at room temperature, albeit the effects on nutritional and organoleptic quality have led to challenges regarding consumer acceptability in some markets (
). These challenges have driven consumer and processor interest in innovative minimal process technologies that can enhance consumer experience without compromising food safety. Microwave volumetric heating (MVH) is an emerging technology that can serve as an alternative to conventional heating, and for many applications, microwave heating has gained worldwide popularity for the processing of food on an industrial scale (
). Microwave volumetric heating processing has the potential to reduce the time of energy exposure, to guarantee a specified thermal load, and to improve product quality in comparison to conventional, indirect heating technologies (
). Because heat is normally transferred from a hot surface to the interior of the product by convection and conduction during indirect heating, this results in a temperature gradient between the outside and inside of the product. In contrast, during MVH, heat is generated inside the food during the short time when microwaves penetrate through it. Microwaves have a greater penetration depth, and this property coupled with a volumetric flow can lead to rapid heating rates, minimizing temperature differentials within the product matrix (
In the dairy industry, bacterial endospores constitute a persistent microbiological challenge, especially in milk powders. With a higher heat resistance compared with vegetative cells, bacterial endospores are able to survive heat treatment at 80°C for 10 min, which is more intensive than standard commercial HTST (72–75°C for 15–20 s) pasteurization (
). Dormant endospores can be activated by pasteurization or other heat treatments during dairy processing, germinate, and outgrow to form cells that multiply in finished products (
). In contrast to the prevalence of mesophilic endospores in liquid milk products, thermophilic bacterial endospores, which grow optimally at 55°C, are more common in dairy powders (
Recently, the occurrence of Bacillus licheniformis, a thermophilic sporeformer that can also recover at psychrotrophic temperature (below 7°C), has been identified as a concern in dairy processes and finished products (
). Thermophilic sporeformers, in their endospore form, are extremely resistant to heat treatment, and can germinate and outgrow under various downstream process and storage conditions, ranging from 5°C to 60°C, which may explain their predominance in the dairy manufacturing environment. Certain thermophilic sporeformers may also give rise to spoilage or food poisoning (
). It is essential that new analytical techniques to monitor, in real time, the prevalence and behavior of microbes having spoilage or food safety concerns are developed to safeguard consumers and processors. Flow cytometry (FC) is a method that uniquely allows a series of rapid, multiparametric measurements to be obtained from large numbers of individual cells based on cell structure, physiology, and viability (
). At the current level of knowledge, it is unclear how MVH compares to conventional heat treatment in terms of the relative efficiency of microbial inactivation of a broad range of species. While the goal of HHT treatment is to reduce the microbial load in milk to levels that significantly prolong shelf life, certain species survive these treatments, with the endospores of spore-forming bacteria making up a significant proportion of surviving cells. It needs to be demonstrated that MVH is at least as effective as HHT in inducing lethal or sublethal damage to endospores before it can gain widespread application. The objective of this study was to assess the bacterial and spore response in reconstituted skim milk (RSM) to MVH and tubular heat exchanger (THE) treatments by both classic plate counting and novel flow cytometry methods to gain a more in-depth physiological profiling of the effects of process treatments on resultant viability and permeabilization responses at the population and subpopulation level. In this study, various heat treatments (at 75, 85, 95, 105, 115, and 125°C, respectively, for 15 s) were applied to RSM using either MVH or THE, and the effects of both modes of heating on the product's indigenous microflora including total bacterial count (TBC), thermophilic bacteria, mesophilic endospores, and thermophilic endospores and inoculated B. licheniformis endospores were measured and compared using conventional microbiological techniques and FC.
MATERIALS AND METHODS
Preparation of RSM
Skim milk was purchased from a local dairy processor and heat-treated at 73°C for 30 s followed by evaporation (to 48% DM) and spray drying (inlet temperature at 180°C and outlet at 85°C) to obtain skim milk powder (Moorepark Technology Ltd.). The skim milk powder was reconstituted (referred to henceforth as RSM) in advance of each trial using a high shear mixer, whereby 6 kg of powder was rehydrated in 54 kg of water at 50°C.
Compositional Analyses of the RSM
The composition of the RSM was determined using following international standard methods: total protein content was tested by the Kjeldahl method (ISO 8968–2/IDF Standard 2001;
Two experimental runs were carried out examining the effect of MVH and THE on the indigenous microflora contained in RSM (without endospore inoculation), and 3 runs involving RSM inoculated with B. licheniformis endospores. For compositional, microbiological, and FC analysis, 2 samples of approximately 100-mL aliquots of unheated RSM, spore-inoculated RSM before heating (if applicable), and heat-treated RSM after each MVH or THE heating temperature of 75, 85, 95, 105, 115, and 125°C for 15 s were collected using 50-mL sterile tubes for individual analysis. All microbiological samples were kept on ice and tested within the same day of processing and tested within 4 h of sampling.
Preparation and Inoculation of Spore Suspensions
Pure endospores of B. licheniformis ATCC 14580 were prepared as per
Immediately before heat treatment, aliquots of B. licheniformis ATCC 14580 endospore suspensions were inoculated into RSM to final levels of 6 log cfu·mL−1, entailing the addition of approximately 6 × 5 mL volumes of such endospore suspensions into 60 L of RSM before heat treatment.
Continuous-Flow Microwave and Conventional Heat Treatment
An indirect THE UHT pilot plant (MicroThermics) was used as the conventional heat treatment, consisting of preheating, final heating, holding, and cooling sections. The MVH treatments were performed using a continuous-flow microwave AMT 3-kW heating unit (Advanced Microwave Technologies). The system was incorporated into the MicroThermics flow scheme as a final heating section, whereby all other heating, holding, and cooling sections were provided by the THE system, ensuring that the true effect of heating technology could be ascertained without additional artifacts of plant design influencing the observations. Six final heating temperatures of 75, 85, 95, 105, 115, and 125°C were used to cover the temperature range of medium to high heat treatment temperatures applied in dairy industry. The actual readings of preheating and final temperatures applied by the MVH system and the MicroThermics unit are shown in Figure 1, with a defined holding time of 15 s applied to all treatments.
Figure 1Schematic flowchart of the heating processes during microwave treatment. The dotted outlines show the preheated temperatures by the tubular heat exchanger.
Samples of unheated RSM, endospore-inoculated (if applicable) RSM, MVH, and THE heat-treated RSM were tested for TBC, thermophilic bacteria, mesophilic aerobic bacterial endospores, and thermophilic aerobic bacterial endospores. The details of endospore and bacterial testing methods used are described in Table 1. For the TBC and thermophilic bacteria counts, 0.1 mL of sample was spread-plated in duplicate and incubated at 30°C for 3 d and 55°C for 2 d as required by the testing methods, respectively. For mesophilic and thermophilic endospore counts, samples were first heat-treated at 80°C for 10 min (
). From each sample, 1 mL was pour-plated in duplicate on the appropriate agar. Tryptic soy agar (TSA; Becton Dickinson) was used for all bacterial analyses and plate count skim milk agar (Merck) was used for endospore tests. Five milliliters of overlay of the same medium was applied on the plate surface for endospore tests post agar solidification to prevent the spread of colonies. Agar plates were incubated at the required time, temperature, and atmospheric conditions as outlined in Table 1. After incubation, colonies were counted, and results expressed as cfu·mL−1 of skim milk (Figure 2). Spreading colonies were counted as single colonies if less than one-quarter of the agar surface was covered; if more than one-quarter of the agar surface was covered, the result was discarded and recorded as a plate with spreading colonies.
Table 1Details of the methods used for microbial and spore enumeration
Figure 2Boxplots of microbial counts (cells and spores) in microwave and conventional heat-treated skim milk depending on spore inoculation and heat treatment temperature. Microwave volumetric heating (MVH; blue rectangles) and tubular heat exchanger (THE; red rectangles). Each rectangular box consists of the mean (as a horizontal line), and the minimum and maximum values as the lower and upper lines of the rectangle, respectively. TBC = total bacterial count.
Sample Preparation for FC Analysis of Indigenous Microflora
Before staining and FC analysis, duplicate samples from each treatment underwent cleanup. For the SYBR Gold/PI staining regimen (see below), 100 μL of sample was added to 900 μL of 2% trisodium citrate solution (Merck), the suspension mixed, incubated at 30°C for 30 min with shaking at 350 rpm, and then centrifuged at 5,000 × g for 5 min at 20°C. Subsequently, fatty material was removed using a sterile cotton bud, the supernatant removed, and the pellet resuspended in a solution of 5 mg·mL−1 papain (Merck). This was incubated at room temperature for 30 min with shaking at 350 rpm. Following centrifugation at 5,000 × g for 5 min at 20°C, the pellet was resuspended in 97 μL of staining buffer [PBS, pH 7.4 (Merck) containing 0.1% BSA (Merck)] to which was added 2 μL of a 10 mg·L−1 solution of phloxine B (Merck), and this was incubated in the dark at 37°C for 30 min to eliminate autofluorescence (
Photobleaching with phloxine B sensitizer to reduce food matrix interference for detection of Escherichia coli serotype O157:H7 in fresh spinach by flow cytometry.
). For sample staining, 1 µL of nucleic acid dye SYBR Gold (Thermo Fisher Scientific) working stock (a 1:100 dilution of the reagent supplied as a 10,000× concentrate in dimethyl sulfoxide, aliquoted, and stored at −20°C until needed) was added to each sample, which were incubated for a further 15 min. Finally, samples were centrifuged at 5,000 × g for 5 min, the supernatant removed, and the pellet resuspended in 300 µL of staining buffer containing propidium iodide (PI), a dye that only enters cells with permeabilized membranes (Merck), at a concentration of 5 µg·mL−1. This final suspension was then analyzed using FC.
Measuring Endospore Germination Using FC
The capacity of conventional- and microwave-treated B. licheniformis endospores inoculated into RSM to germinate was measured using 2 FC methods, one based on nucleic acid staining/membrane permeability (SYBR Gold/PI) and another based on esterase activity and membrane permeability [carboxyfluorescein diacetate (cFDA)/PI)]. Duplicate samples from each of the heat treatment conditions were cleaned as above. For the SYBR Gold/PI staining regimen, following the incubation step with papain, the pellet resulting from centrifugation at 5,000 × g for 5 min was resuspended in 1 mL of PBS containing 10 mM of the germinant L-alanine (Merck), which stimulates endospore germination, and was incubated at 30°C for 1 h. Tubes were then centrifuged at 5,000 × g for 5 min at 20°C and the pellet stained and analyzed using FC. For cFDA (Thermo Fisher Scientific)/PI staining, the inoculated RSM samples were diluted 1:10 in 2% trisodium citrate solution and incubated at 30°C for 20 min. Cells were collected by centrifugation at 5,000 × g for 5 min and washed once in PBS. The cells were resuspended in PBS supplemented with 10 mM L-alanine and incubated at 30°C for 1 h. Samples were centrifuged at 5,000 × g for 5 min, resuspended in PBS and incubated with cFDA (final concentration 50 μM) for 15 min at 30°C with shaking at 350 rpm. The PI was then added to a final concentration of 30 μM to the samples and incubated at room temperature for 15 min followed by FC analysis. For both staining regimens, positive controls for germination consisted of approximately 6 log cfu·mL−1 pure B. licheniformis endospores suspended in PBS containing 10 mM L-alanine and negative controls featured the same amount of endospores suspended in PBS only. These controls were stained as per cleaned RSM samples and used to identify regions on cytographs corresponding to germinating endospores.
FC Analysis of Samples
Flow cytometry was carried out using an Accuri C6 (Becton Dickinson) flow cytometer according to
Protein A-mediated binding of Staphylococcus spp. to antibodies in flow cytometric assays and reduction of this binding by using Fc receptor blocking reagent.
. The cFDA [excitation maximum (λex) 494 nm, emission maximum (λem) 521 nm] fluorescence was captured using instrument's FL1 detector.
For the analysis of SYBR Gold/PI-stained samples of uninoculated RSM, the nucleic acid fluorescence signal from SYBR Gold was used to construct a “cell” gate in a cytograph of green fluorescence versus forward scatter (FSC; signal area in both cases). A second gate in a cytograph of FSC versus side scatter area (SSC) was used to remove further debris and aggregates. Following this, in the case of the analysis of the RSM's native microflora, the proportion of damaged cells was quantified in a plot of SYBR Gold versus PI fluorescence (Figure 3; area in both cases). Through the use of pure stimulated cultures of B. licheniformis endospores, a region corresponding to outgrowing endospores was identified in plots of SYBR Gold versus PI fluorescence and this was subsequently used to quantify the percentage of germinated endospores in treated samples (Figure 4). Another region, which was observed to become populated as the intensity of heat applied to samples increased and was thought to correspond to damaged endospores, was also drawn in this plot and used to quantify damaged endospores (Figure 4). For cFDA/PI analysis, a threshold of 2,000 on the FSC channel was set. Twenty-microliter volumes of each sample were acquired. Positive controls described in the Measuring Endospore Germination Using FC section were used to gate the endospore-specific region on an FSC versus SSC dot plot. Events within this region were then gated on sequential FSC versus cFDA, SSC versus cFDA, and cFDA versus PI dotplots using ungerminated and germinated control B. licheniformis endospores to identify cFDA-cleaving active endospores not permeable to PI (Figure 5; Figure 6A). This template was then used to gate active endospores in the inoculated milk samples. The events/μL in the final gate of the sequence (cFDA/PI plot) for each heat treatment, relating to active, germinating B. licheniformis endospores, was expressed as a percentage of the number of events in this gate for the unheated, inoculated, milk sample controls analyzed on the same day (relative percentage of active endospores).
Figure 3Analysis by flow cytometry of the percentage of permeabilized cells among reconstituted skim milk's native microflora. The proportion of damaged cells was quantified in a plot of SYBR Gold pulse area (A; Thermo Fisher Scientific) versus propidium iodide (PI)-A fluorescence (right panel, A) after gating on SYBR Gold-A versus forward scatter A (FSC-A) to remove debris (left panel, A), followed by further debris removal using a plot of SYBR Gold-A versus side scatter A (SSC-A; middle panel, B). (A) Untreated control sample showing gating sequence, (B) microwave volumetric heating (MVH) at 75°C, (C) MVH at 95°C, (D) MVH at 125°C, (E) tubular heat exchanger (THE) at 75°C, (F) THE at 95°C, and (G) THE at 125°C.
Figure 4SYBR Gold pulse area (A; Thermo Fisher Scientific) versus propidium iodide (PI) fluorescence pulse height (H) in combination with flow cytometry to quantify the percentage of germinated endospores in treated samples at different temperatures. (A) Spores without germinant, (B) spores with germinant, (C) spores inoculated into reconstituted skim milk (RSM), no heat, germinant. (D) spores inoculated into RSM, no heat, no germinant, (E) spores inoculated into RSM, tubular heat exchanger (THE) at 75°C, and germinant, and (F) spores inoculated into RSM, microwave volumetric heating (MVH) at 125°C, and germinant. FSC-A = forward scatter pulse area; SSC-A = side scatter pulse area.
Figure 4SYBR Gold pulse area (A; Thermo Fisher Scientific) versus propidium iodide (PI) fluorescence pulse height (H) in combination with flow cytometry to quantify the percentage of germinated endospores in treated samples at different temperatures. (A) Spores without germinant, (B) spores with germinant, (C) spores inoculated into reconstituted skim milk (RSM), no heat, germinant. (D) spores inoculated into RSM, no heat, no germinant, (E) spores inoculated into RSM, tubular heat exchanger (THE) at 75°C, and germinant, and (F) spores inoculated into RSM, microwave volumetric heating (MVH) at 125°C, and germinant. FSC-A = forward scatter pulse area; SSC-A = side scatter pulse area.
Figure 5(A) The number of permeabilized indigenous cells in reconstituted skim milk (RSM) exposed to varying levels of microwave volumetric heating (MVH; gray bars) or tubular heat exchanger (THE; white bars) as measured using flow cytometry. The percentage damaged (B) or germinating (C) endospores of Bacillus licheniformis inoculated into RSM. Error bars show SEM.
Figure 6Flow cytographs of carboxyfluorescein diacetate (cFDA)/propidium iodide (PI)-stained uninoculated and Bacillus licheniformis-inoculated reconstituted skim milk (RSM) heated using microwave volumetric heating (MVH) and tubular heat exchanger (THE) methods. (A) Samples acquired by flow cytometry were gated by light scatter and fluorescence parameters through successive plots of forward scatter pulse area (FSC-A) versus side scatter area (SSC-A), cFDA pulse area (A) versus FSC-A, cFDA-A versus SSC pulse area (A), and cFDA-A versus PI-A, as demonstrated for unheated spore controls in PBS exposed to germinant. Events in the gated region of the final cFDA-A versus PI-A plot represent active, cFDA-cleaving spores. The percentage value displayed in the cFDA-A versus PI-A cytograph represents the percentage of “active spores” in the gated “spores” region of the FSC-A versus SSC-A plot. The final analysis plots (cFDA-A vs. PI-A) are shown for (B) unheated spore controls in PBS not exposed to germinant; (C) unheated RSM; (D) inoculated unheated RSM; inoculated RSM MVH treated to (E) 75°C, (F) 95°C, (G) 105°C, and (H) 125°C; and inoculated RSM THE treated to (I) 75°C, (J) 95°C, (K) 105°C, and (L) 125°C. All samples C to L were exposed to germinant before analysis.
The bacterial and endospore counts from each sample were converted to log10 cfu·mL−1. The average and standard deviation of the values at each sampling point were calculated and graphed using GraphPad Prism 7.02 (GraphPad Software Inc.). Where the numbers were below the detection limit, an arbitrary value of 0 log cfu·mL−1 was applied. The results of the bacterial and endospore counts were analyzed using GraphPad Prism 7.02 to generate a boxplot for each test method. Two-way ANOVA was carried out on the data wherever possible to determine the significant sources of variation. In Figure 2, each rectangular box consists of the mean (as a horizontal line), and the minimum and maximum values as the lower and upper lines of the rectangle, respectively.
The FCS 3.0 files were analyzed using FCSExpress Lite Standalone Research Version 6.06.0014 32-bit (DeNovo Software). Cytograms and FC histogram overlays for publication were also prepared using this package. For all other data analysis and graph preparation, Microsoft Excel for Office 365 (16.0.10730.20344; Microsoft Corporation) 64 bit was used.
RESULTS AND DISCUSSION
RSM Composition
The composition of the RSM was determined as follows: total protein content was 3.92%, fat content was 0.07%, moisture was 90.08%, and ash was 0.78%.
Microbial Enumeration
The results of testing 4 groups of microbes in endospore-inoculated and uninoculated RSM processed through MVH and THE are presented in Figure 2.
TBC
In uninoculated RSM, the median TBC values for unheated RSM were 3.44 and 3.50 log cfu·mL−1 before the treatment of MVH and THE. After MVH treatment at 75°C, 85°C, and 95°C, the TBC reduced to between 2.38 and 1.70 log cfu·mL−1, whereas for THE-treated milk the TBC was ~2.10 log cfu·mL−1 for the aforementioned treatment temperatures. A significant difference was observed at 105°C, where MVH-treated TBC was about 1 log cfu·mL−1 lower than THE-treated TBC, at 0.89 log cfu·mL−1. After heat treatment at 115°C and 125°C, there was no significant difference between the heating technologies, as the median values of TBC were <1 log cfu·mL−1. These observations could point toward an effect of heating technology/mechanism of heat transfer leading to higher lethality at 105°C for MVH compared with THE. In contrast, at higher heating temperatures (115/125°C) the holding step likely masks differences between the technologies (i.e., it is less important how the final temperature was achieved relative to the time the product spent at that temperature).
In endospore-inoculated RSM, TBC increased by about 3 to 6.1 log cfu·mL−1 after inoculation. There was a limited reduction in TBC of around 0.5 log cfu·mL−1 after application of either technology at 75°C, 85°C, and 95°C, but treatment at 115°C and 125°C gave rise to a reduction in counts to <1 cfu·mL−1. In whole milk for B. anthracis a 6 log cfu·mL−1 reduction in endospore viability occurred through heating at 120°C for 16 s (
), which is in broad agreement with the levels of inactivation observed in this study at temperatures >115°C. A static microwave treatment for 5 min at a temperature of 110°C killed all biofilm-contained B. cereus endospores (
), comparable to the inactivation levels of both technologies tested in this study.
Two-way ANOVA indicated that heating temperature (P < 0.0001), heating technology (P < 0.05), and interaction between them (P < 0.05) were all significant sources of variation in uninoculated RSM, but in endospore-inoculated RSM, only heating temperature was significantly related to reduction in count (P < 0.001).
Both heating technologies applied here have comparable ability to reduce or eliminate bacterial counts in RSM.
applied a prototype microwave pasteurizer to human milk, using holding temperatures of 62.5°C or 66°C for 3 or 5 min, and found that this gave rise to a complete inactivation of the indigenous vegetative bacteria, whereas in this study, great reduction of TBC was observed after the heat treatment at 115°C and onward for 15 s. The great differences of temperature and time scheme used to achieve a satisfactory TBC reduction in various dairy ingredients highlight the difficulty when drawing comparisons regarding the efficacy of a given microwave treatment as reported in the literature owing to the multitude of factors which affect the lethality of the treatment. These include temperature, holding times, microwave wattage, and equipment design and implementation thereof (
For mesophilic endospores in uninoculated RSM, there was no significant difference (P > 0.05) between unheated and heated RSM at 2 ± 0.3 log cfu·mL−1 within the temperature range 75 to 105°C for either technologies. After heat treatment at 115°C and 125°C, mesophilic endospores were reduced to <1 log cfu·mL−1 for both heating technologies.
Mesophilic endospores increased from ~1.7 to 6.0 log cfu·mL−1 after endospore inoculation. Heat treatment at 75 to 95°C had no significant effect on mesophilic endospore numbers, but treatment at 105°C reduced the endospore counts to 2 to 2.5 log cfu·mL−1. After heat treatment at 115°C and 125°C, endospore numbers were further decreased to 0.7 to 1.1 log cfu·mL−1. Compared with uninoculated RSM, where a significant reduction in mesophilic endospores occurred at 115°C, the decrease at 105°C observed for inoculated RSM indicated that those endospores were less heat resistant than the native population.
reported a significant difference in spore-forming bacteria at a species level when comparing raw milk and dairy powders thereof. Endospores with a higher heat resistance normally found in powder products were likely linked to processing artifacts.
Thermophilic Bacteria and Endospores
The number of thermophilic bacteria and endospores were both low in uninoculated RSM, with mean values <1.2 log cfu·mL−1 in all samples analyzed. For both technologies, no significant difference (P > 0.05) was observed under any heating temperature. However, an increase in counts was observed for both thermophilic bacteria and endospores with increasing heating temperature, which was counterintuitive. The bacterial count increased after MVH treatment at 95°C and 105°C, whereas spore counts increased after 75°C under both treatments in uninoculated RSM. Counts gradually decreased to nearly 0 log cfu·mL−1 as heating temperature rose from 95°C to 125°C. The THE-treated RSM had slightly higher thermophilic endospore counts than MVH-treated samples. Data of thermophilic endospores for MVH-treated RSM at 85°C are missing due to uncountable plates caused by colony spread.
After endospore inoculation, thermophilic bacteria and endospores increased from 0.5 and 1.0 log cfu·mL−1, respectively, to ~6 log cfu·mL−1. There was no significant reduction in thermophilic bacteria and endospores after heat treatment from 75°C to 95°C. The mean values for thermophilic bacteria and endospores were reduced to around 1.5 and 1.0 log cfu·mL−1 at 105°C and decreased to <0.5 log cfu·mL−1 at 125°C.
It should be noted that the thermophilic bacteria detection method was applied to samples after MVH or THE treatment, the results of which might reflect the presence of thermophilic endospores in RSM before the heat treatment. It is possible that the thermal processes triggered endospore germination and subsequent outgrowth during enumeration of thermophilic bacteria, hence the observed increases at 95°C and 105°C. The phenomenon of heat- or process-induced endospore germination was also reported by
in a previous study whereby an increase in the thermophilic bacteria count in heat-treated skim milk was observed. The effect of multiple heating steps, both from the process and the enumeration methodology, may have affected the results observed. This may explain why the mean values of thermophilic endospores were lower than 6.0 log cfu·mL−1 compared with other tests after spore inoculation: greater damage to cells occurred after these multiple heating steps. Additionally, the counts of thermophilic bacteria and endospores might be underestimated when dealing with low numbers that were close to the detection limits of the classic microbiological plating methods used
The large variability in box plots shown in Figure 2 for heating temperature 105°C is of note: this temperature has been observed as a key temperature for bacterial and spore inactivation. The variability should be considered as an overall function of the various heat resistances of the individual cell of bacteria and endospores as well as due to the occurrence of plates where colonies spread, rendering them uncountable even with the application of overlay. As described in the Materials and Methods section, these plates were considered as a single colony depending on the size or discarded, thus explaining the large variability inherent in the plate counting technique.
Flow Cytometric Examination of the Indigenous Microflora of Skim Milk
For the SYBR Gold/PI staining regimen, results from the 2 experiments examining the effect of the 2 heating technologies on the indigenous microflora of RSM were combined and analyzed together (n = 51). For each run, the number of PI-permeabilized cells per mL of the mean of the untreated control samples was subtracted from that of each test sample and expressed as the cfu·mL−1 of permeable events compared with the control. These data are presented in Figure 3A. A single factor ANOVA found no significant difference between any of the treatments (df = 50; P = 0.11). However, 4 of the 6 MVH treatments gave rise to higher numbers of permeable cells compared with the control (in the range of 5.5 log cfu·mL−1 in the case of treatment at 85°C, to 5.7 log cfu·mL−1 in the case of 125°C). The extent of permeabilization was lesser in the case of THE treatment, where the range of increase was 4.7 log cfu·mL−1 in the case of treatment at 85°C, to 5.2 log cfu·mL−1 in the case of 125°C. For MVH treatment, a linear increase in the number of permeabilized cells was seen from treatment at 105°C to 125°C.
Unlike plate counting, FC allows the quantification of cells, which are not capable of growth on a medium such as that used for TBC. For control RSM the background cfu·mL−1 of permeabilized cells was 5.6 log cfu·mL−1, representing about 4% of events identified as cells, and giving an FC TBC of ~6.6 log cfu·mL−1. Although this figure is 3 log greater than that produced by plate counting, it is not uncommon for FC counts to greatly exceed plate counts (
). While there is a discrepancy between FC and plate counting with regard to absolute numbers, the general patterns in the data are in broad agreement. For the 3 lower temperature treatments, slightly larger reductions in TBC for MVH treatment reflected a higher permeability at 85°C as revealed by FC. For the 3 higher temperatures, the linear increase in permeabilities for MVH-treated RSM, not seen in the case of THE-treated RSM, is mirrored in the significant difference in TBC observed at 105°C, where MVH treatment resulted in a 1 log lower count than THE-treated TBC. Combining both methods of examining the damage done to the indigenous microflora of RSM allows the tentative assumption that MVH causes increases in cell permeability above that caused by heating alone, especially at 105°C, 110°C, and 115°C, and that this permeability is reflected in lower TBC for 4 of the 6 treatment temperatures.
It is important to keep in mind that the indigenous microflora of the RSM consisted of a heterogeneous group of microorganisms. Genera present would be expected to be Bacillus, Microbacterium, Micrococcus, Enterococcus, Lactobacillus, and Corynebacterium (
). Of the 3 log cfu·mL−1 TBC, 2 of these were mesophilic endospores, and one was thermophilic bacteria/spores. In the case of mesophilic endospores, reductions in counts were only seen at 115°C and 125°C. Mesophilic bacteria saw numbers begin to fall from 85°C upward, with total elimination occurring at the 2 highest temperatures for both technologies. For thermophilic endospores, reductions were seen from 95°C upward. The spike in permeability seen in the case of MVH treatment at 85°C could be explained by the permeabilizing of mesophilic bacteria, whereas the linear increase in permeability recorded from 105°C onward for MVH could be explained by the permeabilizing of mesophilic and thermophilic endospores.
In terms of mechanisms of action of the MVH and THE technologies investigated in this study, FC has shown that at least a proportion of the indigenous vegetative cells and endospores in the RSM were killed through membrane damage. Heat treatment has already been demonstrated to cause damage to the membranes of B. cereus vegetative cells (
speculated that, beyond the effect of the delivered heat, microwaves may destabilize cell membranes through electroporation as well as causing vibrational damage to other cellular components such as structural proteins and enzymes. The cell wall of Escherichia coli and fecal coliforms displayed structural damage following microwave irradiation, whereas heating through boiling did not cause the same damage (
). Some authors, however, disagree with the idea that microwave energy disrupts the structure of molecules with which it interacts independent of those effects attributable to heat (
), even though there is growing evidence that microwave and ohmic heating and other novel forms of physical food preservation exert nonthermal effects on bacteria (
Flow Cytometric Examination of the B. Licheniformis Endospores Inoculated into Skim Milk
Quantification of Damaged and Germinating Endospores Using SYBR Gold/PI Staining
For the SYBR Gold/PI staining regimen, results from the 3 experiments examining the effect of MVH and THE treatment on B. licheniformis inoculated into RSM were combined and analyzed together (n = 72). For the quantification of damaged (highly PI-permeable) endospores, the percentage of events falling into the gate “damaged” for each treatment were expressed as a percentage of the mean percentage of “damaged” events for the controls from that run. These data are presented in Figure 5. A single factor ANOVA found no significant differences between the treatments (df = 71; P = 0.11). Most treatments, however, showed an increase in the percentage of damaged endospores relative to the control, with all MVH treatments showing an increase in damage except the treatment at 105°C. The greatest increase in damage (234% relative to the control) was seen for MVH treatment at 125°C.
Plate counting revealed minimal decreases in the log cfu·mL−1 of inoculated B. licheniformis endospores following treatment at 75°C, 85°C, and 95°C: the decrease in TBC was 0.5 log and there was essentially no decrease in mesophilic endospores, mesophilic bacteria, or thermophilic endospores. However, with the application of 105°C, large decreases in endospore survival are seen. At this temperature for both heating technologies a decrease in endospore log cfu·mL−1 of circa 2 to 3 units was recorded. At 115°C and 125°C, reductions in counts to <1 cfu·mL−1 from a starting level of ~6 log cfu·mL−1 were seen. The FC-mediated quantification of the percentage of damaged endospores mirrors this pattern. The highest levels of endospore damage cluster at 115°C and 125°C, indicating that the mechanism whereby both heating technologies inactivate endospores is through damage to permeability barriers. At 125°C for MVH ~230% of endospores relative to the control show permeability to PI, the highest level seen for this set of treatments. For the 3 lowest temperature treatments, MVH also displays higher levels of permeability than THE. This could point toward MVH causing higher levels of damage to permeability barriers above that caused by THE, even though reduction in viability as measured using plate counting was not significantly different between the technologies.
While it has been known for some time that lethal and sublethal heat treatment of endospores produce an increase in permeability to dyes such as PI (
Monitoring changes in germination and permeability of Bacillus cereus endospores following chemical, heat and enzymatic treatments using flow cytometry.
J. Rapid Methods Automation Microbiol.2008; 16: 164-182
), understanding of the specific actions of microwave radiation over and above thermal effects has been lacking. Knowledge of endospore structure, along with previous studies into the mechanisms of thermal destruction of endospores, posit several endospore structures as targets of heating, the destruction of which opens up endospores to the entry of dyes such as PI: inner membrane/core (
). Core proteins and enzymes contained in the endospore coat are also damaged by heat. Whether these are targets for the action of microwaves remains to be resolved. A study by
into the effect of microwave radiation on B. subtilis endospores shed some light on the mechanisms of action of both the microwaves themselves and the thermal heating produced by them. While endospore cortex structure was disrupted by high temperatures, microwave radiation without thermal heating left the cortex intact. Both high temperatures and microwaves alone caused the inner membrane to become permeable, although the quantity of dipicolinic acid released by this opening up of the core varied depending on whether MVH or THE were applied. In the case of B. megaterium, B. stearothermophilus, Clostridium sporogenes, and Thermoanaerobacterium thermosaccharolyticum,
also concluded that microwave radiation causes damage to endospores over and above that caused by thermal effects, and attributed the dramatic fragmentation of endospores, observed using electron microscopy, to an “explosion of internal pressure generated within the core.” Electron microscopy also confirmed endospore cortex hydrolysis and swelling, and the rupturing of the endospore coat and inner membrane in the case of B. licheniformis endospores exposed to 2.0 kW of microwave radiation (
using confocal microscopy together with SYTO 9 and PI provided evidence for the lethal effect of microwaves through membrane disruption. The SYTO 9-stained Geobacillus stearothermophilus endospores after heat treatment were composed of 2 populations: a refractive subpopulation impermeable to SYTO 9 and a permeabilized subpopulation positive for SYTO 9 (
). These authors concluded that viability loss and permeabilization during heat treatment, although closely related, seem to be 2 different mechanisms that occur independently.
For the quantification of germinating endospores, the percentage of events falling into the gate “germinating endospores” for each treatment were expressed as a percentage of the mean percentage “germinating endospores” for the controls from that run. These data are presented in Figure 5. A single factor ANOVA found significant differences between the treatments (df = 71; P < 0.01). A Tukey-Kramer honestly significant difference post-hoc test found significant differences between the following: THE treatment at 75°C and MVH at 105°C, 115°C, and 125°C; MVH at 75°C and MVH at 125°C; and MVH treatment at 85°C and MVH treatment at 125°C. At lower intensities of MVH treatment (75°C and 85°C), germination is stimulated (through heat activation; see
) relative to unheated controls. With the application of further MVH, levels of germination fall below that of controls, with reduction of >70% seen following treatment at 125°C. The THE treatment does not exert the same stimulatory effect at lower intensities, with levels reduced by 20% following treatment at 75°C. A “cliff effect” was seen for both heat treatments from 105°C onward: levels of germination drop to <40% of controls from this point. This “cliff” was not seen in the case of the percentage of damaged endospores relative to untreated endospores (Figure 5). However, it is from treatment at 105°C onward that the number of permeabilized cells in MVH-treated RSM begins to increase linearly (Figure 4). From treatment at 105°C onward, levels of reduction of germination are slightly lower for THE treatment compared with microwave treatment, although not significantly different. The “cliff” seen in the reduction of the percentage of germinating cells is also reflected in plate counts, where numbers of surviving thermophilic endospores are reduced from ~6 log cfu·mL−1 to <1 log cfu·mL−1 from treatment at 105°C onward. This effect is also measured using the cFDA/PI FC method for quantifying germination (Figure 6). It would therefore seem as if heating at 105°C for 15 s constitutes a critical threshold for the inactivation of B. licheniformis endospores in RSM, regardless of the manner of achieving this temperature.
It was previously shown that as endospores germinate, they become permeable to SYTO 9, a nucleic acid dye, and through the measurement of the corresponding increase in fluorescence of individual endospores the extent of germination of a community can be measured (
). Although the FC method of measuring the germination of endospores may have produced data that successfully mirror traditional plate counting data, the problem exists of how to reconcile one method, which reports that, for example, ~20% of endospores relative to the control (~1% of the endospore population) heated to 125°C using THE are capable of germination (FC), with another that states that less than 1 log out of 6 log cfu·mL−1 are viable (plate counting). In many cases the ability to germinate does not correspond with the further ability to outgrow and produce a colony on solid medium (
). Germination is a degradative biophysical process that can be triggered by lethal treatments such as pressure or heat, and can proceed without a vegetative cell arising from the process. If the proteins or enzymes required to reactivate the metabolism of the germinating spore have been damaged by the treatment applied, the germinating endospore is inactivated (
in utilizing a very similar staining regimen (SYTO 16 and PI) to that described here along with FC to measure the high pressure- and heat-induced germination and inactivation of B. subtilis endospores, recognized that correspondence between plate counting and FC would be minimal as 2 separate processes were being measured: pressure-induced germination leading to inactivation and survival/outgrowth.
, also using SYTO 16 and PI, measured the percentage of “germinated inactivated” B. subtilis endospores produced in response to thermosonication. These authors ascribed the failure of germinated endospores to proceed to outgrowth by the fact that the permeabilized plasma membrane of germinated thermosonicated endospores originated from a dysfunctional, leaky inner membrane of the treated endospore.
Quantification of Germinating Endospores Using CFDA/PI Staining
Flow cytometry with the fluorescent cell viability dye combination cFDA/PI was used to detect active endospores in the RSM matrix following heat treatments by both MVH and THE methods. Flow cytometry results were obtained within 3 h of completing the heat treatments. Endospore preparations, including control endospore preparations in buffer and test endospores inoculated into RSM, displayed low detectable cellular activity, as determined by low intracellular esterase-driven cleavage of cFDA to a fluorescent product, and low population of the carboxyfluorescein-positive, PI-negative “active endospore” gate (Figures 6B and 6C). However, treatment of the cells with the germinant l-alanine at a 10 μM concentration resulted in migration of 55% of the control endospore population to the “active endospore” gate (Figure 6A), indicating that live endospores are present in the endospore preparation and could be detected using cFDA after treatment with germinant, as has been demonstrated with Bacillus cereus endospores previously (
). Similarly, active endospores were detected in B. licheniformis-inoculated, unheated RSM samples incubated with l-alanine (Figure 6D), albeit in a higher background of non-spore RSM matrix-derived events compared with endospores in PBS (Figure 6A), as indicated by the difference in percentage of active endospores within the FSC and SSC “spores” region for both samples. Of the events in the “spores” gate, 55% and 11% represent active endospores for control endospores in PBS, and test endospores in RSM, respectively. In RSM samples, endospores were differentiated from RSM matrix events through the light scatter and fluorescence-based gating strategy employed.
Based on this ability to detect active endospores in an RSM background following incubation with the germinant l-alanine, MVH, and THE-treated inoculated RSM samples were assessed for temperature-induced changes in the number of active endospores. For MVH and THE methods, the population of active endospores in the RSM decreased moderately between 75°C and 95°C, and the percentage of active endospores within the “spores” region decreased from 10.8% to 7.4%, and 10.5% to 7%, respectively. A pronounced decrease in active endospores occurred at 105°C with MVH and THE treatment, with only 1% and 0.4%, respectively, of events in the “spores” gate representing active endospores (Figure 6E–6L). This correlates with the data obtained from plate counting, and from the FC assessment with SYBR Gold and PI of permeabilized and intact endospores.
The concentration (spores/μL) of gated “active endospores” for heat-treated inoculated RSM samples was expressed as a percentage of the active endospores in the inoculated, unheated RSM sample (Figure 7). This relative percentage of active endospores decreased with increasing temperature. Between 95°C and 105°C, for both heating methods, there was a pronounced and significant decrease in active endospores from 42% to 9% with MVH, and from 54% to 9% with THE heating (Students t-test, P < 0.05). Thereafter, the relative percentage of active endospores decreased minimally to 6 to 7% at 115°C and 125°C. This indicates that the majority of B. licheniformis endospores are inactivated at 105°C for both heat treatments. For MVH, there was also a significant difference between the relative percentage of active endospores at 75°C (90%) and 95°C (42%, P < 0.05). However, there was no significant difference in the relative percentage of active endospores between MVH and THE heating methods at any of the temperatures tested, indicating that both heating methods inactivate the inoculated endospores to a similar degree.
Figure 7The relative percentage of active Bacillus licheniformis spores in inoculated reconstituted skim milk (RSM) after each heat treatment compared with active spores in unheated inoculated RSM. The concentration of events (events/μL) gated by flow cytometry as active spores by carboxyfluorescein diacetate (cFDA)/propidium iodide (PI) staining was expressed as a percentage of the active spore concentration in unheated inoculated RSM. The bar chart shows the average relative percentages of 3 independent trials. MVH = microwave volumetric heating; THE = tubular heat exchanger. Bars with letters A, B, C differ at P < 0.05; bars with letters a, b differ at P < 0.05; and bars with letters y, z differ at P < 0.05, as determined by Student's t-test. Error bars represent 1 SD.
In this study, we found that heat treatment of RSM by MVH produced a comparable reduction in TBC, mesophilic endospores, and thermophilic bacteria and endospores compared with THE, while also demonstrating similar levels of inactivation, permeability damage, and reduction of germination in inoculated B. licheniformis endospores. The FC data for damage and germination were consistent with plate count data and demonstrated the potential of this technology for rapid at-line monitoring of the effectiveness of heat treatments in a dairy production context. The FC provided a range of physiological data beyond classical plate counting, which could be integrated into an overall dairy quality control and risk prevention strategy in the future. While continuous microwave heating was found to be comparable to classical heating technologies under the conditions tested, further studies could assess microbial inactivation in products with higher DM contents, whereby the mechanism of heat transfer may improve thermal inactivation kinetics.
ACKNOWLEDGMENTS
The research was funded by the Department of Agriculture, Food and the Marine (DAFM) under the Food Institutional Research Measure, project number 14/F/883 and by Enterprise Ireland under the Dairy Processing Technology Centre, project number TC 2014 0016. The authors have not stated any conflicts of interest.
REFERENCES
Buehler A.J.
Martin N.H.
Boor K.J.
Wiedmann M.
Psychrotolerant spore-former growth characterization for the development of a dairy spoilage predictive model.
Photobleaching with phloxine B sensitizer to reduce food matrix interference for detection of Escherichia coli serotype O157:H7 in fresh spinach by flow cytometry.
Protein A-mediated binding of Staphylococcus spp. to antibodies in flow cytometric assays and reduction of this binding by using Fc receptor blocking reagent.
Monitoring changes in germination and permeability of Bacillus cereus endospores following chemical, heat and enzymatic treatments using flow cytometry.
J. Rapid Methods Automation Microbiol.2008; 16: 164-182
00665-2&id=gr1.jpg){kind=link}
00665-2&id=gr2.jpg){kind=link}
00665-2&id=gr3.jpg){kind=link}
00665-2&id=gr4a.jpg){kind=link}
00665-2&id=gr4b.jpg){kind=link}
00665-2&id=gr5.jpg){kind=link}
00665-2&id=gr6.jpg){kind=link}
00665-2&id=gr7.jpg){kind=link}