Limiting Growth: Microbial Shelf-Life Testing
February 01, 1998 - Article
Print
Comments
Limiting Growth: Microbial Shelf-Life Testing

February 1998 -- QA/QC

By: Michael S. Curiale, Ph.D.
Contributing Editor

  Just about every consumer has encountered spoiled food, whether it was moldy bread, soured milk, bad fish or the things formerly called "leftovers" hiding in the nether regions of the refrigerator. If any of these were eaten, a negative physiological or psychological reaction might have followed. If pathogenic organisms were present, an unforgettable bout with some uncomfortable gastrointestinal malady may have resulted. And in rare cases, foodborne pathogens can prove deadly. However, the presence of pathogens is not always signaled by food spoilage. Still, the risk of an encounter with either spoiled or dangerous foods can be significantly reduced by knowing and understanding shelf life and stability.

  Shelf life represents the useful storage life of food. At the end of shelf life, food is developing characteristics -- changes in taste, aroma, texture or appearance -- that are deemed unacceptable or undesirable. The underlying cause for the change may be microbiological, chemical or physical. Microbiological spoilage is exemplified by the above attributes. Examples of chemical and physical deterioration include rancidity and freezer burn.

  Establishing the microbiological shelf life for many foods becomes important at some point in their history. The determination may be required during product development, after they have been established on the market, or at some time in between. The reasons behind determining shelf life range from product design goal, to formulation change, to packaging or storage change, to changes in microbiological criteria, or a simple desire to know. Regardless of the when or why, numerous variables must be considered in the experimental design of the shelf-life study to approach a useful result.

Spoilage susceptibility

  Microorganisms possess specific growth requirements for temperature, moisture, acidity, nutrients and time. For microorganisms to grow, cultural conditions must be within a certain range. If minimum conditions are not satisfied, growth will not occur.

  In general, organisms grow at temperatures between 0° and 55°C, at pH values between 2 and 10, and at water-activity levels greater than 0.6. These limit ranges are not absolute, and the boundaries around them are not usually sharply defined. Optimal growth generally occurs in the middle region of the various ranges, and slows as the boundaries are approached. Nutrients are not limiting in most foods, but inhibitory substances may block proper utilization of the food. Oxygen is required for the growth of some organisms. For others, it is optional, and for still others, it is a poison. Oxygen and other gases are available from the atmosphere or from air trapped in the product. By manipulating the food's composition, pH, water, etc., different groups of organisms are either activated or inactivated, and their growth rates are either accelerated or slowed.

  Assessing a food item in terms of its microbial growth requirements makes it possible to determine its potential for spoilage. However, studies are generally required to confirm expectation.

Qualifying criteria

  The taste, odor and appearance of a food are the ultimate criteria used by consumers to judge a food's acceptability. In the laboratory as well, organoleptic evaluation of a food is a direct method for determining shelf life. The food is prepared and periodically examined for changes in appearance, aroma, texture and taste until it becomes unacceptable. The organoleptic determination is easily accomplished by those familiar with the product's desired characteristics. Shelf life based on organoleptic analysis, however, may vary significantly from consumer to consumer, since tastes, expectations and ability to detect changes differ greatly.

  The organoleptic quality of food changes as its micro-flora -- bacteria, yeast and mold -- grow and metabolize available nutrients. The sensory changes at first might be subtle, but they eventually make the food unacceptable. Generally, sensory changes are not detectable until the microbial population is high. The number of organisms required to cause spoilage varies with the food item and the type(s) of microorganisms growing in it. Shelf life may be estimated on the basis of microbial density. As a rule of thumb, 10 million bacteria per gram, 100,000 yeast per gram, or visible mold, signal the end of microbiological shelf life. Noticeable degradation of the product is likely at these levels. Whether the changes are acceptable is determined by the organoleptic evaluation.

  High numbers of microorganisms are normal in certain foods, but indicate deterioration in other foods. Therefore, it is desirable to know, even in the absence of objectionable organoleptic changes, the microbiological state of food as it nears the end of shelf life. For the delivery of a product with maximum quality, the shelf life of a product should be determined by organoleptic and microbiological examination.

Designing a study

  Many factors must be considered when designing a microbiological shelf-life study. Among these are: temperature, water content, time, types of microorganisms, suitability of analyses, sampling and replication. Shelf-life studies for each product should be designed specifically for that product because of the number of variables that must be considered.

  Temperature. Of the factors influencing microbial growth -- water, acidity, temperature, nutrients, preservatives and atmosphere -- all but temperature become essentially fixed at the time of product formulation, processing and packaging. Normally, these factors are not intentionally altered in a shelf-life study. Storage temperature usually determines the length of microbiological shelf life of perishable foods.

  In general, as the temperature increases, the microbial growth rate increases. At temperatures near freezing, organisms either grow very slowly or not at all. As the temperature increases toward the optimum, metabolic activity and growth rate increase. At this temperature, growth is fastest. As the temperature increases beyond the optimum, the growth rate begins to slow. At some maximum temperature, growth stops and higher temperatures begin to kill the cells. Each species of organism has a different minimum, optimum and maximum growth temperature range. Moreover, differences may be observed among isolates of the same species. The important point about temperature and growth is that when the storage temperature of a product changes, not only does the shelf life change, but the types of spoilage flora also will likely change.

  Because of the important relationship between growth rate and storage temperature, the most useful shelf-life information is obtained for product kept at its intended storage temperature. Refrigerated products are stored in the refrigerator, and room temperature products are stored at ambient conditions. Small changes in storage temperature may have a significant effect on shelf life. A few degrees may determine the difference between good shelf life and premature spoilage.

  Unfortunately, in the real world, refrigerator and room temperatures are not standardized. Refrigeration can mean anything from 28° to 55°F, and room temperature might fall anywhere between 60° to 95°F. Shelf life at 30°F may be very different from that at 50°F, although both temperatures may represent refrigeration conditions. To produce a meaningful study and to compare different studies, temperatures used must be known. This is most easily accomplished if the study temperatures are fixed and not varied. Temperatures of 40°F (4°C) and 75°F (24°C) are commonly used for refrigerator and room temperature storage, respectively.

  In the real-world conditions of refrigeration during distribution and retail presentation storage of perishable foods, temperatures often cycle between low and high temperatures. Temperature cycling in laboratory studies of shelf life introduces conditions that make data interpretation difficult, and temperature cycling usually is not recommended. A significantly better understanding of shelf life can be obtained when several storage temperatures are used. For refrigerated foods, studies may be conducted at fixed temperatures in each of three ranges (38° to 40°F, 45° to 50°F, and 50° to 55°F). Useful room temperatures are 75°, 85° and 95°F. From the information gained, inferences can often be made about other temperatures.

  Water. Water level determines the characteristics of many foods. Some foods are expected to be dry, some appear moist, and others obviously contain water. Water is essential for microbial growth, and if the amount of free water changes, a food's susceptibility to spoilage may change. For example, if a dry product that is resistant to spoilage becomes damp, it will likely spoil. In contrast, a moist food will not spoil if it dries. Food packaging plays an essential role in the control of moisture, and has a significant effect on shelf life. There is exchange of moisture between the atmosphere and the food. This exchange continues until the food reaches equilibrium with the atmosphere.

  Hermetically sealed packages contain a limited amount of air, and the smaller the head space, the quicker equilibrium is attained between food and air. For hermetically sealed samples, humidity control need not be considered as a study variable as long as the package remains intact.

  Most foods are packaged to limit the rate of water exchange so that little moisture exchange occurs during the life of the product. Thus, humidity control and/or monitoring is required mainly for foods that are: subjected to temperature extremes; exposed to the atmosphere (such as cakes, pies and pastries); or packaged in air-permeable containers. Relative humidities of 40%, 60% and 80% represent a practical range for experimentation.

  Duration. A study's duration should at least match the target shelf life for the food being considered. If a 60-day shelf life is desired for a refrigerated item stored at 40°F, the study should be designed for a minimum of 60 days. Similarly, if six months is expected at room temperature, then the study should be at least that long. A study may be designed to exceed the shelf-life goal if expectations are met and the point of spoilage needs to be determined.

  If a product fails halfway through a designed shelf-life study, there is little point in continuing the analysis. On the other hand, if the product is stable during one segment of a study (for example, no microbial activity is observed), the study should be continued to the next segment. Sterile products do not require repeated testing beyond the time expected for outgrowth of any contaminating microorganisms. It is not unusual for microbial levels to stay constant, and even decrease, over a period of hours, days or weeks before beginning to increase.

  The microbiological shelf life of a food designed to be stored at one temperature cannot be confidently determined more quickly by storing the food at a higher temperature because microbial growth is influenced by temperature. While it is true that organisms grow faster when warmer, it is not yet possible to predict the result for another incubation temperature. For rough estimation, a two- to four-fold increase in growth rate is estimated for an 18° to 20°F temperature increase.

  If the shelf life was found to be 10 days at 60°F, at 40°F, it is estimated to be 20 to 40 days. This broad range for a prediction is not very useful. Moreover, it is entirely possible that the organisms that grew at the elevated temperature (60°F) do not grow, or are not the main spoilage organism, at the desired storage temperature (40°F). For these reasons, accelerated microbiological shelf-life predictions are not useful.

Methodical approach

  Microbial growth in foods for estimation of shelf life is most commonly monitored using agar plating procedures. The procedures are quantitative for the number of viable organisms present at the time of analysis. Because of differences in growth requirements among the different types of microorganisms that may be found in food, no single procedure is available to enumerate all microorganisms. However, a simple, useful procedure is the aerobic plate count, which detects organisms that form colonies on plate count agar usually incubated at 35°C for 48 hours.

  Many organisms are not detected using mesophilic incubation of the aerobic plate count. These include: organisms that grow only at low or high temperatures; most lactic acid bacteria; strict anaerobes; and yeast and mold. Thus, the plating procedures are usually selected on the basis of the type or types of organisms anticipated or known to be present in the food. If the "right" procedures are not selected, it is very possible to have obvious microbiological spoilage, but no experimental data to support the organoleptic observations.

Sensible sampling

  How often a food is analyzed for microorganisms during the shelf-life study must be decided with care in order to detect significant microbiological events. To better understand why this is important, the typical growth cycle of a population of microorganisms should be understood. The growth cycle consists of four phases. The beginning of the cycle is the "lag phase." Increases in cell numbers are not observed during this time. In the second phase, the cell number increases exponentially: one cell becomes two, then four, and so on. In the stationary phase, neither the rate of growth nor the number of cells continues to increase. Growth stops at a density usually not exceeding 1,010 bacterial cells or 106 yeast cells per gram. The final phase of the cycle is aptly called the "death phase," since cell viability decreases. Another cycle will not begin until the cells are diluted into a fresh growth medium, such as food.

  During the time the microorganisms in food are in the lag phase, the food appears to be microbiologically stable. Once the cells enter into the growth phase and begin to multiply, the product begins to change and is considered unstable. At some point along the microbial growth curve, the food usually will spoil. Therefore, for shelf life, the significant points concerning the microbial growth cycle are: the duration of the lag phase; the growth rate; and the microbial count at the end of the growth phase. The end of shelf life -- 10 million bacteria or 100,000 yeast per gram -- usually occurs near the end of the growth phase.

  To identify the different transition points along the growth path, the food is sampled periodically to quantify the number of organisms present. An excessive period between samplings increases the risk of under- or over-estimating shelf life. The more analyses that are completed, the more accurate will be the shelf-life determination. For most foods, the anticipated shelf-life time is divided into five to 12 intervals for sample collection and analyses. The number of intervals chosen is generally an estimation based on experience with similar foods.

  The distribution of microorganisms in a sample of food, or even between samples of food from the same production lot, is not necessarily uniform. For example, if one in five bottles of food solution contains a spoilage organism, then only one of five may display evidence of spoilage. In another example, a spoilage organism present in a solid or viscous food may exhibit localized spoilage while another area without the organism is free of spoilage. Sampling plans must take into account the possible distribution of microorganisms within the lot. This is especially critical when the initial levels are lower than 10 cells per gram, which is the normal sensitivity of the agar plating procedures used for analysis. Single packages represent the most easily distinguishable analytical unit.

  Conducting a meaningful study requires fewer samples if an even distribution of organisms exists in the product. For the homogeneous product, each analytical unit is more likely to be identical to the next. For example, liquids are homogeneous if carefully mixed each time before sampling. Each unit of liquid will represent the content of the whole. However, in an unmixed sample, the distribution from top to bottom or side to side might be uneven. In a viscous sample, thorough mixing may not be possible, resulting in a non-uniform distribution of organisms. Any one sample may not be typical of the whole product lot. For these products, multiple subsamples need to be analyzed to represent the whole. Generally, at least three carefully selected samples of a heterogeneous product are needed to obtain an acceptable representation of microbiological activity in the product.

  A shelf-life study conducted on a single batch of food is valid for that food and any other production lot that is identical. If the microorganism type or number differs significantly among batches, shelf-life duration may differ. Replication of the study always will enhance the accuracy of the prediction. Periodic determinations of shelf life help to provide assurance that the product remains consistent over time with respect to spoilage rate. Changes in formulation, processing and packaging conditions call for reevaluating a product's shelf life.

Making a challenge

  A food may exhibit an exceptionally long shelf life even though the temperature, pH, water and nutrient levels permit microbial growth. The long shelf life may result from the absence of microorganisms in the samples tested, or it may occur because the contaminating organisms will not grow in the particular product formulation. Understanding the stability of these foods in the event of a chance contamination requires a microbiological shelf-life study in which a product is challenged by inoculating it with appropriate spoilage organisms.

  In a challenge study, the product is inoculated with known spoilage microorganisms. The inoculated samples are then treated and stored in accordance with the shelf-life study guidelines. Adding organisms to the foods adds several more variables to the study. The types of organisms and the number of strains of each type to be used need to be decided. In addition, an inoculation level must be selected. The spoilage organism used in the challenge study is usually one that has been isolated previously from similar foods that have spoiled. For example, lactobacilli and yeast are the most common spoilage organisms of salad dressings and sauces. The more isolates included in the challenge study, the greater will be the confidence in the accuracy of the shelf-life assessment. In practice, five isolates of lactobacilli, five of yeast, and five of mold represent a reasonable selection for a salad-dressing challenge study.

  The number of organisms added to the food is generally significantly higher than what would normally be found as a result of contamination during processing. The inoculation levels used are generally greater than 10 per gram, offering easy observation of the presence of the challenge organisms; 10 organisms per gram represents the limit of sensitivity of the agar plate count procedures normally used for enumeration. Lower levels can be detected, but usually at significantly greater expense and with lower accuracy.

  When the level of the challenge organisms does not increase during shelf-life storage, the product formulation is resistant to microbial growth. It is stable in the sense that the number of microorganisms does not increase. However, if the organisms are present in sufficient number, it is still possible that the metabolic activity of the nongrowing cells will cause undesirable changes in the product.

  In most foods susceptible to spoilage, the organisms do not begin to multiply immediately. Instead, the count remains relatively constant for a period of time before growth is observed. The period of no growth is analogous to the lag phase of the microbial growth cycle. A fraction of the challenge organisms may die soon after being added to the test sample.

  If the sample has a low initial inoculation level and die-off occurs, one might incorrectly conclude that the product is stable. Using high inoculation levels will prevent this error. A level of about 10,000 cells per gram is useful for observing either decreases or increases in levels, even if an initial 100-fold die-off is observed.

  Die-off after inoculation most likely results from shock caused by an abrupt change in environment for which the cells are not preconditioned. The die-off can sometimes be avoided or reduced by first adapting the organisms to the product's nutrients, acidity or water activity. In the real world, contamination of product by both unadapted and adapted organisms occurs. The use of organisms that are not specifically adapted for growth in the product simulates organisms originating in the environment and entering the food through contact. Adaptation simulates product-to-product contamination.

  Challenge studies using pathogens are conducted to measure the behavior of those microorganisms in foods and formats similar to studies in which spoilage microorganisms are used. The purpose for using pathogens is to measure their growth, inhibition or die-off in a food. Commonly used pathogens are Salmonella, Listeria monocytogenes, Staphylococcus aureus, Bacillus cereus, Clostridium perfringens, Yersinia enterocolitica, Clostridium botulinum, and Escherichia coli. If the pathogens do not grow, the food is considered stable with respect to the ability of the food system to inhibit their growth.

  Inhibiting changes in the organism's environment influences the duration of the lag phase of the pathogens. The greater the inhibition, the longer the lag phase. Even shifts in incubation temperature between that used to propagate the organism for the study and the storage temperature of the product may change the length of the lag period. Therefore, the organisms' preparation conditions must be carefully chosen to account for a study's specific needs.

Mathematical modeling

  The influences of atmosphere, temperature, water, pH, preservatives and nutrients on the microbial growth are easily measured. Collecting sufficient data makes it possible to derive a mathematical equation of growth. This could allow a quick estimation of shelf life by plugging the required variables for the food into the equation. Mathematical models that include some of the growth-controlling variables for bacterial pathogens are available. The models are interesting, but not necessarily accurate when applied to foods. They are useful for playing "what if" games, such as: "How much longer is the growth of Salmonella delayed by decreasing the pH of the food from 5.6 to 5.4?" Future refinements will most certainly make models more useful. However, a long time will pass before they can reliably replace experimental shelf-life studies.

  Accurate prediction of shelf life necessitates a carefully planned and executed series of experimental studies. Shelf life should be reevaluated in the event ingredient, formulation, processing, packaging or storage changes are anticipated. The knowledge gained from these studies reliably promotes confidence that the product delivered to the customer is safe and of high quality.


  Michael S. Curiale is director of microbiological research of Silliker Laboratories Group, Inc., Homewood, IL. He received his B.S. degrees in biological sciences and chemistry from the University of Illinois, and his doctorate in genetics from Oregon State University. Curiale has authored or co-authored more than 25 research articles on topics ranging from shelf-life and challenge studies, to rapid-detection methods and is a featured lecturer at Silliker's "Food Safety and Stability for Product Developers" short course. Curiale may be contacted at 708/225-1435 or by e-mail at silk-sh@ix.netcom.com.


Print
Comments
comments powered by Disqus
Related News
News
Kancor North America launched Oxikan—a full range of natural antioxidants derived from rosemary to
News
Arla Foods Ingredients will debut a “toolbox" of natural improvement solutions designed
News
Certain compounds, when added to wine, may curb the chemical reaction that causes wine to look,
News
SunOpta Grains and Foods Group, of SunOpta, Inc., added organic and non-GMO stabilized wholegrain