Survival at 60°C, 30 minutes

NH3 from peptone

Pyogenic Streptococcus pyogenes Streptococcus thermophilus Lactococcus lactis subsp. lactis

Enterococcus Enterococcus



Adapted from Sherman, 1937 2Names reflect current nomenclature

These early papers also reflect the changing nature of bacterial taxonomy and nomenclature. In the Orla-Jensen treatise, for example, a description is given of the dairy lactic acid bacterium that was originally called Bacterium lactis (by Lister in 1878), then changed, successively to Streptococcus acidi lactici, Bacterium lactis acidi, Bacterium Guntheri, Streptococcus lacti-cus, and then Streptococcus lactis. In 1985, Streptococcus lactis became Lactococcus lactis subsp. lactis (Schleifer et al., 1985). Interestingly, Orla-Jensen noted that "it would be tempting to employ the name Lactococcus" for these bacteria (which was the genus name first used by the Dutch turn-of-the-century microbiologist, Beijerinck).

Although the initial classification schemes for lactic acid bacteria were based on phenotypic properties, serological reactions have also been used to differentiate streptococci.The Lance-field reactions (named after Rebecca Lancefield) were based on the antigenic properties of cell wall-associated components, resulting in more than thirteen distinct groups.The dairy streptococci (now L. lactis subsp. lactis and L. lactis subsp. cremoris) were found to possess the group

Box 3—2. Who's Who in Microbiology: Identifying the Organisms in Starter Cultures (Continued)

N antigen and became referred to as the Group N streptococci.Although the other major dairy streptococci, Streptococcus thermophilus, was not antigenic and could not be serologically grouped, the Lancefield groupings were and still are useful for classifying pathogenic streptococci, including Streptococcus pyogenes and Streptococcus pneumonia (Group A) and Streptococcus agalactiae (Group B), as well as enteric streptococci (Group D; now referred to as En-terococcus). However, one taxonomist suggested that "serology is best forgotten when working with dairy streptococci" (Garvie, 1984).

As noted above, phenotypic traits are still used successfully to classify lactic acid bacteria.The most informative distinguishing characteristics include: (1) temperature and pH ranges of growth; (2) tolerance to sodium chloride, methylene blue, and bile salts; (3) production of ammonia from arginine; and (4) carbohydrate fermentation patterns. Some of these are physiologically relevant in fermented foods (e.g., salt tolerance and growth at low or high temperature), whereas others are simply diagnostic (e.g., inhibition by methylene blue).

Based on these criteria (as well as microscopic morphology), it is, experimentally, rather easy to perform the relevant tests and to obtain a presumptive identification of a given lactic acid bacterium. Kits based on sugar fermentation profiles are also available that can be used to identify species of lactic acid bacteria. Specific classification schemes for lactic acid bacteria, based primarily on phenotypes, are described in Bergey's Manual (Holt et al., 1994). Other identification schemes that rely on membrane fatty acid composition, enzyme structure similarities, and other properties also exist, although they are now less often used.

Despite the value of traditional identification schemes, there is no doubt that the methods described above lack the power and precision of genome-based techniques. In fact, advances in nucleic acid-based bacterial fingerprinting methods have led, not only to new identification tools, but also to renewed interest in bacterial taxonomy. The ability to distinguish between strains of the same species is important not only for identification purposes, but also because it provides a basis for understanding the phylogenetic and evolutionary relationships between starter culture organisms. Although morphological, biochemical, and other phenotypic characteristics remain useful for genus and species identification, molecular approaches that rely on nucleotide sequences have proven to be more reliable, more reproducible, and more robust. Several techniques, in particular, are widely used, including pulsed field gel electrophoresis

Despite the numerous changes that have occurred in microbial classification and nomenclature (especially since the early 1980s), it is still essential that the microbial contents of a starter culture be accurately described and that species identification be based on the best available taxonomical information. First, culture suppliers need to include the correct species names on their products, since Generally Regarded As Safe (GRAS) status is affirmed only for specific organisms. A cheese culture claiming to contain species of Streptococcus thermophilus, but actually containing closely related strains of Enterococcus, would be mis-labeled. Second, accurate identification is necessary for the simple reason that culture propagation and production processes require knowledge of the organism's nutritional and maintenance needs, which are species-dependent. The growth requirements of Lactococcus lactis subsp. lactis are different, albeit only slightly, from the closely related Lactococcus lactis subsp. cremoris. Finally, should the product manufacturer prefer to include species information on the label of a fermented product, the species name should be correct. (This would be voluntary, since it is not required.) For example, yogurt and other cultured dairy products that contain probiotic bacteria often include species information, since some consumers may actually look for and recognize the names of particular species.

Box 3—2. Who's Who in Microbiology: Identifying the Organisms in Starter Cultures (Continued)

(PFGE), restriction fragment length polymorphism (RFLP), ribotyping, 16s ribosomal RNA sequence analysis, and various PCR-based protocols.These precise methods of strain identification have also become important for commercial reasons because proprietary organisms, with unique characteristics and production capabilities, are developed and used in fermentation processes.

Finally, the reader should be aware, as authors of Bergey's Manual have warned, that bacteria do not really care into what group they are classified or what they are called—rather, classification is done for the sole benefit of microbiologists (Staley and Krieg, 1986).Therefore, there is a certain degree of arbitrary decision-making involved in the classification process, despite efforts by taxonomists to be totally objective. In fact, as these authors noted, in bold type, "There is no 'official' classification of bacteria." Names of bacteria can certainly be 'valid' (as per the List of Bacterial Names with Standing in Nomenclature, available online at, but the value of a given classification system depends entirely on its acceptance by the microbiology community.


Garvie, E.I. 1984.Taxonomy and identification of bacteria important in cheese and fermented dairy products, p. 35-66. In F.L. Davies and B.D. Law (ed.). Advances in the Microbiology and Biochemistry of Cheese and Fermented Milk. Elsevier Science Publishers. London. Holt, J.G., N.R. Krieg, PH.A. Sneath, J.T. Staley, and S.T.Williams. 1994. Bergey's Manual of Determinative

Bacteriology. Ninth Edition.Williams and Wilkins. Baltimore, Maryland. Krieg, N.R., 1988. Bacterial classification: an overview. Can. J. Microbiol. 34:536-540. Orla-Jensen, S. 1919The lactic acid bacteria. Copenhagen.

Schleifer, K.H., J. Kraus, C. Dvorak, R. Kilpper-Balz, M.D. Collins, and W. Fischer. 1985.Transfer of Streptococcus lactis and related streptococci to the genus Lactococcus gen. nov. System.Appl. Microbiol. 6:183195.

Sherman,J.M. 1937.The streptococci. Bacteriol. Rev. 1:3-97.

Staley,J.T., and N.R. Krieg. 1986. Classification of procaryotic organisms: an overview, p. 965-968.In P.H.A. Sneath, N.S. Mair, M.E. Sharpe, and J.G. Holt (ed.). Bergey's Manual of Systematic Bacteriology, Volume 2. Williams and Wilkins. Baltimore, Maryland.

For such products, the name of the organism (and perhaps even the strain) is especially important because probiotic activity depends on the actual species or strain. Unfortunately, species declarations on consumer products are often incorrect or are outdated.

Starter Culture Math

Using starter cultures not only ensures a consistent and predictable fermentation, it also addresses a more fundamental problem, namely how to produce enough cells to accommodate the inocula demands of large-scale fermentations. For most fermented foods, the first requirement of a starter cul ture is that it initiate a fermentation promptly and rapidly. Although exceptions exists (e.g., in the case of the secondary, carbon dioxide-evolving fermentation that occurs late in the Swiss cheese process), it is usually necessary that the fermentation commence shortly after the culture is added. And while a short lag phase may certainly be acceptable or even expected, a long lag phase is generally a sign that the culture has suffered a loss of cell via-bility.Thus, for a starter culture to function effectively, it must contain a large number of viable microorganisms.

As the mass or volume of the food (or liquid) starting material increases, either larger starter culture volumes or greater starter culture cell concentrations are required. For example, if a 1% starter culture inoculum is ordinarily used for a given product, then 1 kg (or 1 L) of starter culture would be necessary to inoculate 100 kg (or 100 L) of the food substrate material.This modest-sized inoculum could easily be produced from a colony or stock culture, simply by one or two successive transfers through an intermediate culture (e.g., 0.1 ml or one colony into 10 ml, followed by 10 ml into 1 L).

When pure cultures, rather than backslop-ping techniques, were first introduced a century ago in the dairy industry, this approach of making intermediate and mother cultures was generally how cultures were prepared and used. As the size of the industry increased, however, such that larger and larger starter culture volumes were required, it was no longer feasible for cheese manufacturers to prepare cultures in this manner. In other words, a cheese plant receiving 1 million L of milk per day would need 10,000 L of culture, plus all of the intermediate cultures (and incubations) necessary to reach this volume.

To address this situation, two general types of cultures are manufactured and sold to the fermented food and beverage industries. The first type, often referred to in the dairy industry as bulk cultures, are used to inoculate a bulk tank. The bulk starter culture essentially is the equivalent of several intermediate cultures that traditionally have been required to build up the culture. After a suitable incubation period in the appropriate culture medium, the fully-grown bulk culture (which is akin to a mother culture) is then used to inoculate the raw material. The starter culture organisms comprising the bulk culture will remain viable for many hours, provided they are protected against acid damage, oxygen, hydrogen peroxide, or other inhibitory end-products. In the cheese and fermented dairy products industry, bulk cultures are routinely used to inoculate production vats throughout a manufacturing day. Maintaining culture viability is still an important issue, however, as will be discussed below.

A bulk culture is not warranted for many fermented foods simply because raw material volumes are more modest. That is, the amount of culture necessary to inoculate the fermentation substrate can easily be met using the culture as supplied directly from the manufacturer. For example, bakers' yeasts may be supplied as yeast cakes, which are added directly to the dough ingredients just prior to mixing. Similarly, meat starter cultures, whether in frozen can form or lyophilized packets, are added directly to the meat mixture. Even dairy cultures that are designed to be inoculated directly into the food material are now available, thus eliminating the need to prepare bulk cultures.

In the dairy industry, these cultures are referred to as direct-to-vat set cultures.They have several advantages that have made them very popular. They eliminate the labor, hardware, and capital costs associated with the construction, preparation, and maintenance of bulk starter culture systems.They also eliminate leftover or wasted bulk culture. Although they were initially produced as frozen concentrates, packaged in cans, they are now available as pourable pellets or lyophilized powders, making it easy to dispense the exact amount necessary for inoculation.

As described below, eliminating the bulk culture fermentation also means that bacterio-phages have one less opportunity to infect the culture and cause trouble.The use of these cultures can also reduce mixed strain compositional variability. However, when direct-to-vat set cultures are used to inoculate large volumes, the culture must be highly concentrated to deliver a sufficient inoculum into the raw material. Freezing, centrifugation, filtration, lyophiliza-tion, and other concentration steps may indirectly reduce culture viability, ultimately leading to slow-starting fermentations.Improvements in concentration technologies have minimized some of these problems. Still, direct-to-vat cultures are generally more expensive than bulk cultures, and, despite their convenience, may not be economical for all operations.

Culture Composition

Mixed or undefined cultures

Mixed or undefined cultures were once the main type of culture used throughout the fermented foods industry. They typically contain blends of different organisms, representing several genera, species, or even strains.The actual identities of the organisms in a mixed culture are often not known, and the individual species may or may not have been characterized microbiologically or biochemically. Even the proportion of different organisms in a mixed culture is not necessarily constant from one product lot to another.

Nevertheless, undefined cultures are still used as starter cultures for many applications because they have a proven history of successful use. In Italy, the famous Parmigiano Reggiano cheese is made using a mixed culture obtained by overnight incubation of the whey collected from that day's cheese production. Moreover, depending on the product, mixed cultures may have a particular advantage. For example, mixed starter cultures are commonly used in the Netherlands for the manufacture of Gouda, Edam, and related cheese varieties.The fermentation part of the cheese-making process can be quite long, and it is not unusual for a given strain to be infected and subsequently killed by bacterio-phages that inhabit the cheese plant environment (or that are present within the culture itself). However, given the diversity of lactic acid bacteria present in these mixed cultures, there are likely strains that are resistant to that particular bacteriophage and that can then complete the fermentation. In fact, it is well established that frequent exposure to different bacteriophages provides a natural and effective mechanism for ensuring that phage-resistant strains will be present in repeatedly propagated mixed cultures.

Despite their proven track record, undefined mixed cultures are not without problems. In particular, product quality may be inconsistent and fermentation rates may vary widely, affect ing production schedules. For small-scale operations, where time is somewhat flexible and where quality variations may be more tolerable, these issues are not so serious. For large production facilities, however, where precise schedules are critical and consistent product quality is expected, mixed cultures have become less common. Starter cultures comprised of defined strains, with more precise biological and biochemical properties, are now prevalent. These cultures, referred to as defined cultures, simply refer to cultures that contain microbiologically characterized strains.

Defined cultures

Defined starter cultures can be comprised of a single individual strain or as a blend of two or more strains.The origin of the defined strains that are used commercially varies. Some were simply present in traditional mixed cultures and others have been isolated from natural sources. For example, in the case of dairy starters, milk production habitats and cheese serve as good sources; in the case of wine starters, grapes and wine-making equipment are good locations to find suitable yeasts. Defined strains must be identified and characterized for relevant metabolic and physiological properties, phage-resistance (in the case of dairy strains), and other desirable traits.For fungal starters, safety is an important issue and the inability of a putative culture strain to produce mycotoxins is an essential criteria. In addition, when multiple strains are assembled into a culture blend, all strains must be compatible. That is, one strain must not produce inhibitory agents (e.g., bacte-riocins, hydrogen peroxide, or killer factors) that would affect growth of other organisms or that would cause one strain to dominate over others.

Ever since the defined strain concept for dairy starters was introduced in the 1970s, there has been debate about whether single, paired, or multiple strain blends are preferred (Box 3-3). Multiple strains are used for some applications, such as yogurt cultures, which

Box 3-3. Defined Strain Cultures: How Many Strains Are Enough?

Most large cheese factories in the United States rely on defined multiple strain starter cul-tures.These cultures, which typically contain up to five different strains, perform in a timely and predictable manner and generally yield products with consistent quality attributes. Perhaps most importantly, they provide the manufacturer with a reasonable level of assurance that phages will not be a problem, because if one of the strains was suddenly attacked and lysed, other strains in the culture would act as back-ups and would see the fermentation through to completion.

However popular the multiple strain approach has become, it is limited for several reasons. First, in the case of Cheddar cheese manufacture, there are not that many phage-unrelated strains of Lactococcus lactis subsp. lactis that have the appropriate cheese-making properties. There are even fewer distinct strains of Lactococcus lactis subsp. cremoris, which is generally considered to be the "better" cheese-making strain.Although it is relatively easy to isolate spontaneous phage-insensitive mutants (derived from good cheese-making strains), these derivative strains often revert to a phage-sensitive phenotype or else lose plasmids or acquire other mutations that make them unsuitable for cheese-making.

Second, it is not so easy to produce and then maintain multiple strain blends that contain the correct strains and in the correct ratio. In theory, strains present in the multiple strain culture are selected based on good cheese-making properties. However, in reality, there may still be variation in product quality using multiple culture blends, at least more so than would occur if only one or two strains were used. Also, whenever multiple strain starters are used, phage diversity within the production environment would likely increase, whereas when a single or paired strain starter is used, the types of phages that one would expect to see in the cheese plant would be limited.

Given the advantages of single or paired strain starters, why didn't such cultures quickly become the norm in the cheese industry? Indeed, the very idea of defined strain cultures (first conceived in the 1930s by researchers in New South Wales, and later employed by New Zealand researchers) was based on the use of phage-resistant, single-strain cultures (recounted by O'Toole, 2004).A simple test was also devised to provide a reliable basis for predicting whether a strain will remain resistant to bacteriophages during repeated cheese making trials (Figure 1; Heap and Lawrence, 1976).

Obviously, reliance on a single strain carries a perceived risk (despite the advantages noted above), because the appearance of an infective phage could quickly decimate the culture and ruin the fermentation. Even if the single strain was known to be highly phage resistant, it would be hard to imagine that the cheese plant manager would get much sleep. Indeed, in the United States and Europe, cheese manufacturers were initially quite reluctant to adopt single strain starters (Sandine in 1977 described this as a possibly "frightening" practice). However, research in the 1980s from several groups supported the earlier results from New Zealand and demonstrated that defined single or paired strains could be used successfully without phage infections (Hynd, 1976; Richardson et al., 1980;Thunell et al., 1984;Timmons et al., 1988).

Another way to assuage the worries of the sleepless cheese manager would be to use a strain genetically configured to resist bacteriophage infection (reviewed in Walker and Klaenhammer, 2003). Specifically, genes encoding for phage resistance could be introduced into the genome of the selected organism, conferring a bacteriophage-resistant phenotype.As described later in this chapter (and in Chapter 5), there are multiple mechanisms by which lactic acid bacteria can become phage-resistant.These strains can be constructed such that a single strain is stacked with several different phage-resistant systems and is therefore insensitive to a variety of phage. Alternatively, isogenic derivatives, each carrying a different phage resistance system, can be constructed and used in a rotation program.

Box 3-3. Defined Strain Cultures: How Many Strains Are Enough? (Continued)

Brew Your Own Beer

Brew Your Own Beer

Discover How To Become Your Own Brew Master, With Brew Your Own Beer. It takes more than a recipe to make a great beer. Just using the right ingredients doesn't mean your beer will taste like it was meant to. Most of the time it’s the way a beer is made and served that makes it either an exceptional beer or one that gets dumped into the nearest flower pot.

Get My Free Ebook

Post a comment