Putri_whole_thesis.pdf (6.11 MB)
Understanding thermophilic spore-forming bacteria in milk powders
thesisposted on 2023-05-27, 10:51 authored by Putri, TP
Thermophilic spore-forming bacteria such as Geobacillus spp. are common contaminants of milk powder processing plants. Their spores can remain viable throughout the entire dairy powder industrial process, including pasteurisation (72 ¬¨‚àûC for 15s) and the even hotter evaporator sections. Geobacillus spp. can form biofilms in dairy processing equipment that, over extended run times (e.g., >16 h), can deposit unacceptably high spore loads in end-product. While not dangerous to human health, this potentially leads to quality defects and price reductions, requiring that the process be stopped and the plant fully cleaned. The objective of this study was to: a. investigate the kinetics of Geobacillus growth, biofilm development and spore formation as a function of temperature and water activity, b. to develop an understanding of how these factors influence thermophile attachment and biofilm formation on stainless steel and affect the time before spore release into milk being processed into milk powder, and, based on this knowledge, c. explore options to extend run times of dairy powder plants Growth studies were undertaken, and a stainless-steel, laboratory bench-scale flow-through reactor was built and used, to investigate: 1. The effect of temperature (45 to 75 ¬¨‚àûC), media composition and water activity (0.959 to 0.992) on growth rates of 16 Geobacillus spp. originally isolated from milk powder processing plants; 2. the kinetics of attachment, biofilm formation and eventual release of new spores from spores inoculated into the flow-through reactor system via milk and with regard to spore inoculum levels, milk flow rates and temperature; 3. the effect on Geobacillus spp. growth of temperature step changes in the flow-through-system, and 4. the potential to disperse biofilms on stainless steel surfaces in the flow-through reactor using exogenously provided nitric oxide (NO). Growth rates of the 16 strains of Geobacillus spp. were modelled as function of temperature using a four-parameter square-root ('Ratkowsky') model. The model was developed to predict the growth of Geobacillus spp. under time-varying temperature conditions and to identify temperatures optimal for growth and biofilm formation. Over 300 growth curves were generated at temperatures in the range 45 to 75 ¬¨‚àûC, using different incubation methods, enumeration methods and growth media, although not all data sets were used because many were deemed to be unreliable due to insufficient or erratic growth of the spore-forming thermophiles under apparently well-controlled growth conditions, a phenomenon reported anecdotally by others. The studies showed that growth occurred in the temperature range 45 to ~70 ¬¨‚àûC with fastest growth occurring at ~60 ¬¨‚àûC. Consistent with published reports the generation time at 60 ¬¨‚àûC was estimated to be ~ 22 ‚Äö- 25 min. Studies at different water activities (a\\(_w\\)) suggested that the minimum water activity for growth was ~0.975. The results also showed that growth rate variability of all 16 strains is large compared to growth rate variability reported for non-spore forming cells. Nonetheless, no strain had growth rates that were systematically different to that of the pooled data. Growth rates observed for G. stearothermophilus W14 were representative of the average response of all strains and this strain was selected for use in subsequent studies. Growth rate data of G. stearothermophilus strain W14 under either anaerobic (100% N\\(_2\\)) or aerobic (20% O\\(_2\\)) conditions were also generated using a bioreactor (fermenter apparatus) at 55 ¬¨‚àûC, 60 ¬¨‚àûC, 65 ¬¨‚àûC, and 70 ¬¨‚àûC. The bioreactor was employed mainly to enable a comparison of aerobic and anaerobic growth rates by reliably producing anaerobic conditions. G. stearothermophilus strain W14 showed fastest growth at ~60 ¬¨‚àûC in anaerobic conditions with a doubling time of 26 min for vegetative cells, which was similar to growth rates under analogous aerobic conditions. The kinetics of cell and spore attachment to stainless steel were studied by inoculating the flow-through system with spores of G. stearothermophilus W14 in the milk flowing through the system and monitoring the change over time in vegetative cells and spores in the milk leaving the system (i.e., the 'effluent'). Studies were conducted at temperatures from 45 to 70 ¬¨‚àûC. At near optimal temperatures (i.e., 60 ‚Äö- 65 ¬¨‚àûC) viable cell and spores counts initially decreased in the milk effluent but began to increase consistently after 3-6 h indicating attachment, germination and proliferation of cells and production of spores. Different milk flow rates applied to the system (5, 10, 20, 40 mL/min) showed no significant differences in the time for the generation of vegetative cells or spores. Spore inocula, fed into the system as pulses (1 h) or continuously added (usually for ~24 h but up to 40 h in some experiments), showed significant differences in the time to attachment and detectable proliferation, and as a function of temperature. In the pulsed system, the spore counts increased above the inoculum level after ~8 h of milk flow at near optimal temperatures, whereas in the continuous system levels in the effluent milk increased after ~4 h. In the pulsed system, a lower spore inoculum fed to the system (<10\\(^2\\) CFU/mL), resulted in a longer time before spore counts increased in the effluent (>8 h). Swabbing internal surfaces of the flow-through equipment at the end of runs‚ÄövÑvp at 45 ¬¨‚àûC showed early stage biofilm growth (>10\\(^3\\) CFU/cm\\(^2\\)) whereas runs at 65 ¬¨‚àûC showed high counts (>10\\(^8\\) log CFU/cm\\(^2\\)). The flow-through system was a useful way to study, under 'commercially-relevant' conditions, the attachment, growth and sporulation of thermophilic spore formers on warm stainless steel surfaces in milk powder plants and provided a system to study potential interventions against biofilm formation by thermophilic spore-forming bacteria. Based on Knight et al. (2004), who reported a method to minimise biofilm build up using temperature cycling to interrupt the growth cycle of non-spore forming bacteria in whole milk processing, temperature cycling studies were undertaken using the flow-through system to evaluate whether powder plant run times could be extended using the same approach. A number of experiments were conducted with temperature of the system altered systematically during the run‚ÄövÑvp. In some runs, there was reduced attachment and outgrowth of thermophilic spore-forming bacteria, suggesting that temperature cycling could extend run times if applied at the sites of most rapid attachment, growth and biofilm formation. The result showed that the temperature cycling should include temperatures near the limits of the temperature growth range of G. stearothermophilus to be able to significantly retard growth. Together with the temperature model, these data can be used as a foundation to estimate the expected benefits of manipulation of temperatures of milk powder processes. Nitric oxide (NO) has been reported to disrupt bacterial biofilms. Its potential for use to disrupt G. stearothermophilus biofilms in milk processing equipment was also studied using the flow-through system. At realistic and commercially relevant levels, NO did not significantly delay the time for unacceptable spore levels to occur in the milk effluent, although there was evidence of a reduction in final spore loads. However, recent studies have suggested that NO, while effective against Gram-negative biofilms, will not be ineffective against spore-forming Gram-positive bacterial biofilms. In short, the application of NO to extend run times is not supported, however, by the results of this study. Considering the ecology and physiology of G. stearothermophilus, and related species, the results of this study have reinforced that thermophile contamination during dairy powder processing will likely continue to be a difficult problem to address. This is because of the ability of spores to survive processing and cleaning, the rapid growth rate of the microorganism, its ability to form biofilms, and the inevitable production of spores in biofilms. Cell differentiation, including spore-formation, occurs through quorum sensing (QS) mechanisms and appears to be a 'bet-hedging' mechanism. It is concluded that thorough sanitation procedures will still be required at the end of each processing session to minimise residual fouling on stainless steel because the time to unacceptable spore loads also depends on initial contamination levels. While temperature cycling produced some effects, under some circumstances, further research is required to determine whether this approach can be manipulated and optimised to achieve commercially significant extension of powder plant run times.
Rights statementCopyright 2017 the author