CA1218894A - Propionates and metabolites of propionibacteria affecting microbial growth - Google Patents

Abstract

PROPIONATES AND METABOLITES OF
PROPIONIBACTERIA AFFECTING MICROBIAL GROWTH
ABSTRACT OF THE DISCLOSURE
Metabolites of propionibacteria, when added to a food product, inhibit the growth of spoilage microorganisms. Propionates added to food products inhibit the growth of yeasts responsible for spoilage of certain food products. And, metabolites of propionibacteria can be separated into fractions that, respectively, inhibit and stimulate the growth of microorganisms.

Description

12~8899~

P~OPIONATES AND MUTILATES OF
PROPIONIBACTERIA AFFECTING MICROBIAL GROWTH
BRIEF SUMMARY OF THE INVENT ION
The present invention relates to chemical substances which affect microbial growth. More specifically, it relates to substances which inhibit spoilage microorganisms in food products and to substances which stimulate the growth and reproduction of desirable microorganisms such as microorganisms useful in the manufacture of cultured or fermented foods.
The need for improved methods of food and feed preservation is great; activities of bacteria molds and yeasts render millions of pounds of food inedible annually and the problem is especially acute in countries with inadequate refrigeration In the United Kingdom, it is estimated that 50 million loaves of bread are lost each year due to mold contamination (Soiler, "Factors affecting the use of mold inhibitors in bread and cake," Microbial Inhibitors in Food, Stockholm 1964, p. 211). Feed also may spoil, be rendered unpalatable or even become toxic as a result of microbial activity (Rod ricks and Levitt, "Toxigenic Fungi", in Compendium of Methods for Microbiological Fermentation of Foods, Am. Pub. Health Assign. Ed MEL. Speck, 1976).
Among these spoilage microorganisms, the yeasts and molds are economically the most significant because they are so versatile from nutritional and growth-temperature standpoints and because their spores are so ubiquitous. Minimizing their effect in food presently depends on sepsis and sanitation, exclusion of oxygen and use of chemical additives (Ayes, Mandate and Sardine, Microbiology of Foods, W. H. Freeman and Co., 1980, p. 140).
These procedures are only partially successful. For example, Suriyarachchi and Fleet (Apply.
and Environ. Microbial. 42:574, 1981) found that 45% of A

128 retail yogurt samples had yeast counts of greater than 1000 per gram; nine different genera were represented. Thus, almost half the product on the market has a limited shelf-life as compared to products made under good manufacturing practices where no greater than 1 yeast cell per gram should be present initially (Davis, Dairy In. 35:139, 1970). Such a product should have a shelf life of at least 30 days at 5C (Kruger, J.
Dairy Sat. 59:344, 1975). Temperature abuse is common in retail channels (Body felt and Davidson, J. Milk and Food Tuitional. 38:734, 1975) and even longer shelf-life times are being sought as product manufacturing sites are being localized in fewer but larger plants which distribute products over immense distances. For example, one of the well-known brands of yogurt available throughout the United States is manufactured in only two localities. Distribution time therefore can represent a significant proportion of the shelf life of the product, depending on conditions of manufacture and temperature history.
For some foods, bacteria are more important spoilage agents than yeasts or molds. Cottage cheese and market milk products such as pasteurized whole milk, skim milk, half and half, and whipping cream are 25 examples of such products. Psychotropic bacteria, able to grow at refrigeration temperatures (35 to 50F), rapidly spoil these products in the commercial marketplace. Such spoilage is a continuing and serious problem for the dairy industry. The only control practice-s now available to the industry for psychotropic bacterial spoilage are sanitation and the prevention of temperature abuse. The nature of the industry and methods of product manufacturing and handling, however, make these practices inadequate to 35 prevent the spoilage problem.
Prop ionic acid (CH3CH2COOH - The Merck Index, Thea Ed ! Merck and Co., Inc., 1976) is known as a i2~8~394 mold inhibitor, especially for use in spillage ~Draughon et at, J. Food Sat. 47:1018, 1982), bread and in certain food wrappers such as those for cheese (Ayes et at., Microbiology of Foods, W. H. Freeman and Co. 1980, p.
140; Moon, Pro. Am. Sock Microbial., 1981, p. 29).
Wilfred and Anderson ("Preappoints Control Microbial Growth in Fruits, Vegetables", Food Industries, 17, pp. 622-624, 726, 728, 730, 732, 734, June 1945) reported that sodium preappoint incorporated in concentrations of 0.2~ into nutrient broth at pi 4.5 inhibited Pseudomonas fluoresces along with some other bacteria. Concentrations of 0.5~ at pi 5.0 inhibited two different molds and at pi 4.5 concentrations of 2.0%
were needed to inhibit SaccharomYces ellipsoids yeast 5 and concentrations of 3.0% were needed to inhibit yeast. Thus, low concentrations were not effective against yeast. For figs, a dip into 15% calcium preappoint or incorporation of 0.5% calcium preappoint in a fig puree inhibited growth of mold on the figs. Young berries were treated by dipping into 5% sodium preappoint or 10 percent sodium preappoint or sprayed with a 10 percent solution. These treatments inhibited mold growth.
Apple slices were treated with a 0.5 percent calcium preappoint solution and these "proved to be less susceptible to damage by molds than were similar slices not treated with preappoint. However, The fruit carried a prop ionic odor and tended to become more noticeably gray than did untreated slices".
Peas were treated with a 5 percent preappoint solution at pi 6.3. In this study, controls were reported to be faded in color, to have a very disagreeable odor, and to be very slimy. Preappoint treated lots showed slight color change, no sliminess 35 and little abnormal odor. However, such differences were not found by Wilfred and Anderson in a separate study wherein peas were held for 4.5 hours after 12~8894 treatment with calcium preappoint (presumably a 5 percent solution) and then scalded, frozen, and stored for one month. At the end of the storage period, prop ionic flavor was not detected in the treated peas;
but the differences between controls and treated peas for flavor and skin texture were insignificant.
Further, Wilfred and Anderson state that experiments involving dipping of lima beans in 5 percent sodium preappoint solution definitely inhibited bacterial growth for a number of hours, as in peas.
Such inhibition of bacteria for a few hours is meaningless for food preservation. The authors also state "preappoints should be used with due regard to its limitations, such as the pi of the product, its microbial flora, and the concentration of preappoint likely to impact foreign flavor or odor to the food."
An authoritative text (Handbook of Food Additives, end Ed., CRC Press, 1972, pp. 137-141) references the work of Wilfred and Anderson done in 1945 and then those who wrote this section in the CRC
Handbook reviewed other appropriate literature on prop ionic acid, sodium preappoint and calcium preappoint and concluded that "preappoints are more active against molds than sodium bonniest, but have essentially no activity against yeasts. They have little action against bacteria with notable exception of their ability to inhibit the organisms which cause rope." It is also reported that preappoints "are suitable for yeast raised as well as other baked goods" and "because preappoints inhibit molds and spares yeast" they are used in breads (the Handbook of Food Additives, end Ed., CRC Press, 1972, pp. 137-141).
Microbial metabolizes, especially so-called antibiotics, which inhibit the growth of microorganisms 35 are well-known. Indeed, a large segment of the pharmaceutical industry is based on the sale of purified antimicrobial which find uses in medicine and-to some ., ~Z:18894 extent also in the food industry.
A considerable body ox literature exists on the propionibacteria which produce prop ionic acid. Their growth and metabolism have been reviewed (Hutting and Reinhold, J. Milk Food Tuitional., 35:295, 358 and 463, 1972) as well as their contribution to the flavor and microbiology ox Swiss cheese (Langsrund and Reinhold, J.
Milk Food Tuitional. 36:487, 531 and 593, 1973; 37:26, 1974). Early literature (Skew and Sherman, J. Dairy Sat. 6:303, 1923) reported that propionibacteria produced acetic and prop ionic acids and the production of other volatile, namely acetaldehyde, propionaldehyde, ethanol, propanol and dim ethyl sulfide, by these bacteria was noted by Canaan and Bills (J.
Dairy Sat. 51:797, 1968). Dustily production by propionibacteria was reported by Lee et at. (Can. J.
Microbial. 16:1231, 1970). The Handbook of Food Additives, end Ed., (CRC Press, 1972), pp. 137-141, provides background information on prop ionic acid and its salts, including uses physical and chemical properties, antimicrobial activity, safety, regulatory status, applications, handling, storage and assay. The same type of information on acetic acid and acetates is presented in this reference on pages 147-150.
Propionibacteria also produce acetate along with preappoint and COY as end products of lactic acid metabolism (Allen et at., J. Bacterial. 87:171, 1962).
Propionibacteria are known to also produce substantial amounts of succinic acid as well as acetic acid (Wood and Workman, "Mechanism of Glucose Dissimilation by the Prop ionic Acid Bacteria," Become. J. 30:618-623, 1936;
Wood and Workman, "The Relationship of Bacterial Utilization of COY to Succinic Acid Formation", Become. J. 34:129-137 (1940); Leaver, Wood and 35 Stjerholm, "The Fermentation of Three Carbon Substrates by C. Propionicum and Propionibacterium", J. Bacterial.
70:521-530, 1955.
f . I, lZ~8~394 Acetic acid is the main constituent of vinegar and has a definite characteristic odor and flavor. The Handbook of Food Additives, end Ed., (CRC Press, 1972), refers to work which shows that homology of prop ionic acid (such as succinic) have tastes and odors which would be noticeable in foods such as baked goods.
Propionibacteria are used in the production of Swiss cheese and here only a small amount (only about 10~ of the total inoculum which is, in turn, only about 1-2~ of lo the milk used) produces distinctive flavor characteristics. Therefore, it is not anticipated that a nutrient growth medium containing propionibacteria could be used as a liquid suspension, after condensing or after drying as an additive to foods or feeds without producing an undesirable change in flavor or odor.
Propionibacteria are reported to produce an anti viral component (Ramanathan, Read and Cutting, "Purification of Propionin, An Anti viral Agent from Propionibacteria", Pro. Sock Exp. Blot. Med.
123:271-273, 1966; Ramanathan, Walynec and Cutting, "Anti viral Principles of Propionibacteria", Isolation and Activity of Propionics 8 and C", Pro. Sock Exp.
Blot. Med. 129:73-77, 1968~
Despite the many reported techniques in the art of food preservation and a great deal of ongoing research concerning preappoints and propionibacteria, there was until now no simple, natural additive substance that could effectively inhibit such difficult spoilage microorganisms as yeasts, molds and, particularly, slime-producing psychotropic bacteria.
It is now discovered, quite surprisingly, that prop ionic acid, in concentrations so low that flavor and aroma are not adversely affected, inhibits yeast growth in certain foods. It is also discovered that a mature propionibacterium growth medium can provide prolonged inhibition of yeasts, some bacteria and mold without providing an undesirable flavor of succinic acid or ~8894 prop ionic acid or vinegar. The unexpected findings disclosed are especially dramatic in light of some of the low concentrations which provide microbial inhibition as described in following examples.
An antimicrobial food additive can be obtained by growing propionibacteria, e.g. Propionibacterium Sherman, P. freudenreichii, P. pentosaceum, P.
thin, P. arabinosum, P. rub rum, P. lensenii, P.
peterssonii, and related species (as identified in Mali et at., Jan. J. Microbial. 14:1185, 1968) in a milk cheese whey or broth medium or other suitable nutrient mixtures. The resulting growth liquid is then added to food and feed products to inhibit yeasts, molds and spoilage bacteria. To facilitate storage and shipment, the growth liquid may be dried to form a powder, or frozen before use as an antimicrobial food additive. Such powdered or liquid natural metabolizes of propionibacteria can be incorporated into various foods and feeds to render them less susceptible to spoilage by growth and/or enzymatic activity of yeasts, molds and bacteria.
The growth medium for such Propionibacterium species may be formulated with milk or whey containing yeast extractives or broth media containing appropriate growth nutrients The growth liquid, after development of the propionibacteria up to 106 to 101 cells per ml, may be heat treated (pasteurized) to kill the inoculated and adventitious bacteria prior to use in liquid, condensed, dried, or frozen form. It is added in various concentrations preferred between 0.01 and 10% of total weight) to food or feed where it functions to inhibit yeasts, molds or certain bacteria. This inhibition enables the shelf life and storage times of the food or feed to ox increased.
Furthermore, it is discovered that the growth liquid, after development of the propionibacteria, can be mechanically separated into fractions, one of which I

- B -inhibits yeasts, molds, and bacteria, and another of which is stipulatory to the growth of useful bacteria such as Streptococcus lactic, Streptococcus creamers, Lactobacillus bul~aricus, Streptococcus thermophilus, S Lactobacillus acidoPhilus, Leuconostoc species, Lactobacillus Planetarium, and Pedro coccus cerevisiae.
It is therefore a general object of the present invention to extend the shelf life of food products subject to microbial spoilage Another object is to provide a substance which can be added to a food product to inhibit the growth of mold, yeast and some bacteria without harming the flavor, aroma, or other characteristics of the food product.
A specific object is to extend the shelf life of dairy foods, cultured foods, high acidity foods, and specifically cottage cheese, yogurt, kisses, fruit juice, sausages, and other ground meat products.
An additional object is to provide a method which uses naturally produced substances, such as metabolizes of Propionibacterium species elaborated in a suitable growth medium, in the preservation of food and feed.
It is also an object to provide antimicrobial metabolizes of propionibacteria in a dried or frozen form for simplicity of storage and shipment.
Further an object of the present invention is to provide such a method to inhibit microorganisms which are potentially injurious to human health.
A still further object is to provide an antimicrobial food additive substance comprising a growth mixture containing propionibacteria and metabolizes thereof, with the bacteria being viable or made not viable depending upon whether it is desired to 35 produce additional amounts of metabolizes, including COY, after the antimicrobial substance is added to the food product.

.
.~.. ;,. , 12~3 !39~
g Still another object is to provide a growth medium in which propionibacteria can grow and produce metabolizes having properties that effect the growth of other microbes.
A related object is to separate and separately utilize fractions of the mixture of metabolizes of a propionibacterium culture, e.g. to inhibit and to stimulate microbial growth, respectively A related object is to provide a growth stimulant useful in a starter culture of commercial fermentation bacteria.
These and other objects will become increasingly apparent by reference to the following description.
BRIEF DESCRIPTION OF THE DRAWINGS
In the drawings:
Figs. 1-4 are graphs showing the effect of a supernatent fraction of a growth mixture of P. Sherman on the growth of gram negative psychotropic cottage cheese spoilage bacteria in lactose broth versus the amount of the supernatent fraction applied.

DETAILED DESCRIPTION
There are several aspects to the present invention as set forth below. It has been found possible to inhibit spoilage microorganisms and thereby extend the shelf life of many food products without adversely affecting flavor or aroma by adding preappoints, a growth mixture containing a propionibacterium culture with its metabolizes or a fraction of such a growth mixture. Such substances, which should be widely accepted as safe for human consumption, are surprisingly excellent inhibitors of spoilage microorganisms.
Examples of the present invention are set forth hereinafter. It is intended that they be only 1;~18~3~4 illustrative. Propionibacterium strains are available from the American Type Culture Collection (ATTICS). The other cultures are widely available or can be obtained from Oregon State University, Corvallis, Oregon, without cost.

A. ACIDS AND SALTS
As mentioned above, certain acids, in particular prop ionic acid, were thought to have an inhibitory effect on molds, certain bacteria, and fungus, but no meaningful effect on yeasts. Prop ionic acid is felt to be so compatible with yeast that it was recommended as an additive for bread dough wherein yeast vitality is essential It has now been discovered that prop ionic acid inhibits yeast in certain food products and is an effective preservative in foods such as yogurt, wherein yeast contamination is a major difficulty. Furthermore, the amount of prop ionic acid required for success is so unexpectedly low, as compared to other acids and past reports, that prop ionic acid can be present in such a small amount that it has no adverse effect on flavor or aroma of the food product.

1. Succinic Acid a. Mold and Yeast The success of prop ionic acid in inhibiting yeast is particularly surprising in view of the fact that other closely related acids and salts are ineffective. The following examples show that a closely related compound, succinic acid, could not be used to inhibit the growth of yeast.

Potato dextrose ajar (3.9 gym per 100 ml of water) was prepared, autoclave, and the pi was lowered by adding 1.4 gym of tartaric acid. Then, 8 different ,:

lZ18~394 concentrations of succinic acid (0, 0.5%, 0.75%, I 2%
, 3%, and 4%) were prepared in the potato dextrose ajar and the pi of each was adjusted to 3.5 with hydrochloric acid; two plates of each were poured and dried for 24 hours and then one plate of each concentration was inoculated with 0.2 ml of yeast cells and the other plate of each concentration of succinic acid in potato dextrose ajar was inoculated with 0.2 ml of mold cells.
The yeast and mold inkwell were prepared by transferring a loop of either yeast or mold found growing on commercially available yogurt products into sterile water and then vortexing before use.
The succinic acid did not show inhibition of mold or yeast in the above experiments as all of the plates, including both the control plates with no succinic acid and those plates with succinic acid, supported growth of the mold and the yeast.

b. Bacteria Despite the lack of success with yeast and mold, it was found that succinate ions, from an addition of succinic acid or soluble succinate salts, are effective to inhibit Pseudomonas bacteria in certain substances.

A second set of similar plates of varying succinic acid concentrations in potato dextrose ajar were prepared, as in Example 1. Differences were that the second set of plates were adjusted to a pi between 4.5 and 5.5 USinCJ hydrochloric acid or sodium hydroxide and the plates were inoculated with 0.2 ml of Pseudomonas fluoresces.
The Pseudomonas control plates with no succinic 35 acid supported substantial growth after 30 hours and growth was also observed on the 0.25~, 0.75%, and 1%
succinic acid plates. There was no growth observed on lo 394 the 0.5~, 2%, 3%, or 4% succinic acid plates after 48 hours. Since the pi was adjusted to between 4.5 and 5.5, it is possible that inhibition was due to a low pi of the plates, and not to the succinic acid, since Pseudomonas does not grow well below pi 5. Therefore, the experiment described as follows was conducted to determine whether succinic acid can be inhibitory to Pseudomonas.
In a separate experiment, plate count ajar 10 (2.35 gym for 100 ml of water) was prepared, autoclave, and placed in separate containers and sufficient succinic acid was added to produce plates which contained either 0, .25, .5, .75, 1, 2, 3, or I of succinic acid. The pi was adjusted to 4.7 for one set of plates; and the pi was adjusted to 7.0 for another set of plates. Then, each set of plates was inoculated with either Pseudomonas fluoresces or Pseudomonas alcaligenes, and these plates were incubated at room temperature for 24 to 48 hours.
The plates inoculated with Pseudomonas fluoresces at pi 4.7 all became contaminated with mold and were not evaluated for Pseudomonas growth. Those plates inoculated with Pseudomonas fluoresces at pi 7.0 showed growth of Pseudomonas fluoresces within 24 hours when the concentration of succinic acid was I or less.
The 3% and 4% succinic acid plates did not exhibit growth of Pseudomonas fluoresces in 24 hours but did become positive for growth at 48 hours.
Of the Pseudomonas alcaliqenes plates, the control plate and those which contained succinic acid at pi 7.0, all supported growth of the organism at 24 hours. In addition, the control plate at pi 4.7 was positive for growth of Pseudomonas alcaliqenes after 48 hours of incubation. However, all plates containing 35 1/2% succinic acid or more and adjusted to pi 4.7 did not support Pseudomonas alcaliqenes growth after 48 hours of incubation.

Jo.
..~

~21~3~39~

It is clear from the above data that succinic acid can inhibit the growth of Pseudomonas alcaliqenes at pi 4.7 in fairly low concentrations but is not very active at pi 7Ø The data also demonstrate that Pseudomonas alcaliqenes is not as susceptible to the effects of succinic acid at pi 7 as is Pseudomonas fluoresces.

2. Prop ionic Acid Although prop ionic acid is known to have certain valuable abilities as an inhibitor of microbial growth, it was discovered by experimentation that there are additional, previously unexpected uses for preappoints.
a. Bacteria It was found that preappoints are quite effective to inhibit Pseudomonas bacteria.

In order to further investigate the effects of pure succinic acid and pure prop ionic acid against Pseudomonas fluoresces and Pseudomonas alcaliqenes, the following experiment, wherein the pi of the growth media were adjusted to 5.2, was conducted. The pi of 5.2 was chosen because this is about the highest pi which one is likely to encounter in good cottage cheese. Plate count ajar was prepared as previously described and adjusted to contain 0.5, 0.75, 1, 2, or 3% of succinic acid, and separately prepared was plate count ajar containing 0.5%, 0.75%, I 2%, 3%, and 4%, of prop ionic acid. The final pi was adjusted to 5.2 for all of the media; and the plates were allowed to dry and inoculated with 0.2 ml of either Pseudomonas fluoresces or Pseudomonas alcaliqenes cultures. The plates were then incubated at room temperature for 48 hours. The Pseudomonas alcaliqenes control plates showed growth in 36 hours;
I' : , ~Z~8~394 and the Pseudomonas fluoresces control plates supported growth within 48 hours.
The only concentration of succinic acid which inhibited growth of any pseudomonas was the 3%
concentration of succinic acid which did inhibit the growth of Pseudomonas fluoresces. However, all other concentrations of succinic acid allowed growth of Pseudomonas fluoresces and Pseudomonas alcaliqenes.
The Pseudomonas alcaliqenes and Pseudomonas fluoresces were both inhibited by all concentrations of prop ionic acid, including the lowest concentration tested which was 0.5%.

b. Yeast Perhaps the most surprising discovery, so contrary to conventional wisdom, is that preappoints are effective in inhibiting yeast in food products over a substantial period of time and that the concentration of preappoint ions can be sufficiently low that there is no adverse effect on flavor or aroma of even a delicate food. This discovery is of particularly great value in the preservation of yogurt wherein yeast contamination is a substantial problem. However preappoint ions from added prop ionic acid or preappoint salts, should be effective when mixed into any bendable food product.
Although the effect of pi on yeast inhibition by preappoints has not been fully explored, it appears that preappoints are effective at oh's below 5.2 and clearly so at pi 3.35 and below.
Prop ionic acid in concentrations of 0.2% has been reported to inhibit mold (Handbook of Food Additives, Second Edition, CRC Press, 1972, pages 137-141). However, prop ionic acid has been reported not to inhibit yeast and, in fact, concentrations of 0.3%
35 are used in the manufacture of bread or rolls to act as an inhibitor of mold, but the yeast which is used for production of bread or rolls is not inhibited.

12~8~394 In order to preserve yogurt or other similar food products, a nontoxic source of preappoint ions is provided and the pi adjusted, if necessary, so that the desired degree of yeast inhibition is achieved. The optimum preappoint concentration and pi will vary somewhat depending on the composition of the food product, but can be determined by experimentation.
With plain yogurt, the preappoint source is best added before fermentation to avoid any blending of the yogurt thereafter. The preappoint source may be added at any time to mixed, eye. fruit-flavored, yogurts and many other food products.

15 Yogurt spoilage occurs both through contamination of mold and yeast. Therefore, the following experiment was conducted in order to determine if prop ionic acid would inhibit mold and/or yeast which develops in yogurt. Commercially available yogurts (Yoplait*strawberry, apple, and plain) were purchased from a grocery store and maintained at room temperature for several days in order to allow the mold and yeast present to develop. The mold and yeasts which developed in these yogurt samples were isolated on plates and allowed to develop. They were then maintained separately as "pure" cultures of either mold or yeast.
Potato dextrose ajar (3.5 g of potato dextrose ajar powder and 0.85 9 of tartaric acid in 100 ml of water) was prepared, autoclave, and mixed with prop ionic acid to produce either 0.25%, 0.5% or 0.75~
prop ionic acid. The control plates with no prop ionic acid had a pi of 3.6 and the plates containing prop ionic acid had a pi of between 3.25 and 3.35. Therefore, control plates with no prop ionic acid were Allah prepared 35 with addition of hydrochloric acid so they were available at pi 2.1 and 2.8. After all plates were prepared, the mold and yeast which had developed in the * Trade Mark 12~ 39~

commercially purchased yogurt was transferred to each plate and the plates were incubated at 33C for 48 hours and then maintained at room temperature.
Those control plates with 0% prop ionic acid at pi 3.6 had yeast colonies present after 36 hours and mold developed after 2 days at room temperature. Those control plates containing added hydrochloric acid (pi 2.8) also supported the growth of yeast and mold after 48 hours at room temperature; and a slide of the organisms confirmed that yeast was growing. The control plate at pi 2.1 supported the growth of mold only after 7 days and did not support the growth of yeast at this low phi After 7 days, the plates containing all concentrations of prop ionic acid did not support any growth of either mold or yeast. It is clear from the above data that prop ionic acid in a concentration as low as 0.25 percent inhibits the growth of both mold and yeast strains which contaminate commercially available dairy products. The results also indicate that preappoints are effective mold and yeast inhibitors at concentrations sufficiently low that an effective amount of such substances would not adversely affect the flavor or aroma of even a delicate food product. Such results are not taught by the prior literature, since usual concentrations of prop ionic acid which inhibit mold are 0.2%. Furthermore, the baking industry routinely uses calcium preappoint which is not inhibitory to the growth of yeast at 0.31% (Handbook of Food Additives, Second Edition, CRC Press, 1972, pages 137-141).

B. MIXED METABOLIZES OF PROPIONIBACTERIA
A further discovery is that Propionibacterium cultures can be used to produce a growth mixture, including metabolizes, that inhibits mold, yeast, and certain bacteria in any of a wide variety of food products. Some inhibition of spoilage microbes is A

~Z~889~

probably due to the presence of prop ionic acid as a metabolize of propionibacteria. But, the degree of inhibition achieved is much greater than would be expected for the amount of prop ionic acid in the metabolizes. This indicates that an unidentified substance or substances in the growth mixture is responsible for the excellent ability of the growth mixture to extend the shelf life of food products.
As mentioned above, viable propionibacteria are used in the manufacture of Swiss cheese to form eyes by the production of COY and to impart the characteristic Swiss cheese flavor. And, it is now discovered that there are advantages to adding viable propionibacteria directly into a batch for other cultured dairy products. In most food products, however, the presence of viable propionibacteria would be unacceptable because eyes would not be desired and a COY release would bloat packaging materials. Thus, as described in certain of the following examples, propane- bacteria can also be grown in a liquid growth medium which is subsequently heated or otherwise treated to render the bacteria not viable. The result is a stable mixture which is an effective additive for the inhibition of spoilage bacteria in food products.
To facilitate storage and shipping, a propionibacteria growth mixture may be frozen or concentrated, e.g. by spray-drying, or freeze-drying, to form a powder.
A growth mixture according to the present invention is most readily used by mixing with a bendable food product, but should also be effective to treat the surface of solid food products, or the interior of such products, e.g. by injection. The optimum amount to be used will depend on the composition 35 of the particular food product to be treated, but can be determined by simple experimentation.
In most instances, substantial improvements in I

1~:18~394 shelf life can be obtained by adding the growth mixture in an amount sufficiently small that it will have no deleterious effect on the flavor or aroma of the food product. The growth mixture includes the numerous metabolizes of propionibacteria, more than one of which metabolizes appear to be active in microbial inhibition. But, as an indication of the quantity of total metabolizes in a growth mixture, it is convenient to refer to the amount of preappoint radical present.
Typically, a growth mixture will be effective when mixed with a bendable food product if the mixture is present in an amount sufficient that the preappoint radical contributed by the growth mixture is from 5xlO 7 to 0.1 weight percent of the food product.
Example 5 illustrates, generally, the effectiveness of propionibacteria growth mixtures.

Six different strains of ProPionibacterium were grown in a commercially available (PHASE 4, ~alloway-West Co., Fond Du lag, Wisconsin) bulk starter medium wherein 39 g of the medium was added to 500 ml of water, pasteurized at 85C for 45 minutes, cooled to 30C, and then inoculated with 10 ml of 6 different cultures grown in tomato juice media for 24 hours. The ProPionibacterium cultures tested were: p-31-c 13673;
8262; 9615; 9616; 9617. These were all obtained from the American Type Culture Collection in Rockville, Maryland.
After times 0, 16, 18, 21, and 24 hours, 25 ml of the growth media were removed by pipette, mixed with 25 ml of double strength potato dextrose ajar (39 g for 500 ml of water) and the mixture was autoclave. After the potato dextrose ajar and sample were mixed, the 35 sample was brought down to pi 3.0 to 3.7 with hydrochloride acid and poured into plates. The plates were dried overnight and then mold or yeast was streaked , .~. ...

~Z~8~39~

onto each plate the next day and the plates were incubated at 33C for 24 hours. Plates containing the zero hour propionibacteria growth medium showed a large amount of growth of mold and yeast. The samples collected at 16, 18, 21, and 24 hours all allowed the growth of yeast (most after 24 hours) but mold was inhibited on these plates, even though the samples collected and used for zero hour plates did have extensive mold growth.
For p-31-c, yeast was apparent after 20 hours and there was no mold for 60 hours. For 13673, yeast was present after 24 hours and no mold was apparent after 72 hours. For 8262, there was both yeast and mold after 24 hours. For strains 9615, 9616, and 9617, yeast was present after 24 hours; but no mold appeared during 72 hours of observation. Samples of growth media containing 9617 were somewhat inhibitory to yeast in this experiment as indicated by less growth of yeast on plates containing samples of 9617.
Duplicates of the above plates were prepared after continuing to grow the propionibacteria from 24 up to 80 hours and utilizing samples collected at times 0, 16, 18, 21, 24, and 80 hours after beginning of growth.
Each plate was divided into 3 parts and 1 part was streaked with mold and the other two parts were each streaked with 2 different kinds of yeast. The results for these duplicate samples showed that for strain p-31-c, none of the samples prevented complete growth of yeast or mold, except the 18-hour sample which prevented mold growth. For strain 13673, the 80-hour sample was inhibitory to mold, but none of these samples prevented yeast or mold growth. For strain 8262, none of these samples were inhibitory to mold or yeast. For strain 9615, the 24-hour sample and the 80-hour sample were inhibitory to mold and the 80-hour sample was inhibitory to yeast. For strain 9616, the 80-hour sample was inhibitory to mold and slightly inhibitory to yeast.

*, ., lZ~8~4 For strain 9617, the 24-hour and 80-hour samples were inhibitory to yeast and the 18, 21, 24 and 80-hour samples were all inhibitory to mold.
These data show that strains 9615, 9616, 9617, and 13673 which have been allowed to grow for 80 hours under the conditions described herein can inhibit growth of yeast or both yeast and mold, and that strains 9616 and 9617 were most active in inhibiting yeast and mold.
An additional experiment was conducted wherein each of the cultures were sampled 18 hours after growth and 100 hours after growth and the samples which were collected were 1 ml and 5 ml and 10 ml of the growth medium at 18 hours and 1 ml and 5 ml at 100 hours.
These samples were then mixed with 20 ml of potato dextrose ajar (58.5 g for 1000 ml, autoclave and poured into plates after adjusting the pi to between 3.0 and

3.6 using concentrated hydrochloric acid). The plates were dried overnight and inoculated as spread plates with 0.2 ml of each of 2 different kinds of yeast or with mold. Plates were then incubated at 30C for 30 hours. Results showed that 10 ml of strain 9615 was inhibitory to mold after 18 hours of growth and 10 ml of strain 9617 was inhibitory to mold after 18 hours of growth. The other samples were not inhibitory after only 18 hours of growth. However, after 100 hours of growth, strain 13673 was inhibitory to mold and yeast for the S-ml sample and to mold with the l-ml sample.
Strain p-31-c was not inhibitory. Strain 8262 was inhibitory to mold with the 5-ml sample only. Strain 9615 was inhibitory to mold with the l-ml sample and inhibitory to both mold and yeast with the 5-ml samples. Strain 9616 was inhibitory to mold with the l-ml sample and inhibitory to yeast and mold with the 5-ml sample. Strain 9617 was inhibitory to mold with 35 the l-ml sample and inhibitory to both yeast and mold with the 5-ml sample. It should be noted that the most inhibitory to mold of all these strains was 9616 where 1 A

ml of the growth medium inhibited the growth of molds on the plates for over 75 hours, which is longer than any of the other growth media inhibited mold at such a low concentration. Control plates showed the presence of mold and yeast after only 30 hours of incubation.
As mentioned above, growth mixtures according to the present invention will perform somewhat differently in different food products. The following examples illustrate ways to determine how best to use lo ProPionibacterium growth mixtures according to the present invention with specific food products.

1. Yogurt For the purpose of this disclosure, yogurt is defined as a fermented milk product made by allowing Lactobacillus bulgaricus and Streptococcus thermophilus to grow in pasteurized milk until a firm coagulum is formed.
Commercial yogurt products typically have a pi of 3.5 to 4.5 and a titratable acidity in the range of 0.9 to 2.0 percent expressed as lactic acid. Most commercial yogurts have a fat content of two to four percent. Yogurt may contain coloring and flavoring ingredients, including artificial and natural fruit flavorings, whole fruit and fruit syrups. Yogurt, particularly the unflavored variety, has a characteristic flavor and aroma that would be expected to be compromised if large amounts of preappoints were added.
It is the expectation of many consumers and a requirement in many countries that yogurt contain viable fermentation microorganisms. Such bacteria are consumed to aid digestion. To retain viable bacteria, yogurt cannot be given a final pasteurization to preserve the 35 product for shipment and storage. Thus, yogurt is particularly susceptible to spoilage microorganisms present at the manufacturing and packaging site. Due to ~Z~8~94 its composition, mold and yeast growth are the most common spoilage problems.
As demonstrated by the following examples, the shelf life of yogurt can be extended greatly, without pasteurization that would harm the fermentation bacteria, by adding ProPionib-acteri-um to a yogurt batch before fermentation or by adding a Propionibacterium growth mixture according to the present invention, either before or after yogurt fermentation. Quite surprisingly, it was found that addition of the propionihacteria or the growth mixture to a yogurt-making batch had no noticeable detrimental effect on the activity of the fermentation bacteria.
A number of examples report the addition to yogurt of a growth mixture wherein Propionibacterium cells have been made not viable. This is expected to be the best commercial procedure. But, the use of a growth mixture with viable Propane bacterium might be advantageous due to continued production of metabolizes 2Q in the yogurt. In a "Swiss-style" mixed fruit yogurt, eyes formed by COY released from Propionibacteria might prove to be a desirable sensory characteristic.

A commercially available home style yogurt maker (Salton*yogurt maker) was used to produce cups of yogurt . The starter cultures employed were commercially available for the production of yogurt cry. Hansen's Laboratories, Milwaukee, Wisconsin). The culture used 30 was R6. The commercially available frozen culture was thawed and added to milk and allowed to mature prior to use. Then 1 ml of this culture was added to 200 ml of 2% milk and the yogurt making instructions followed.
Excellent yogurt was produced which had a good body, and 35 was a uniform product with a final pi of 4.2. In a separate yogurt preparation, the inoculum consisted of ,~_ 1/2 ml of commercially available Hansen's culture * Trade Mark :1218~394 combined with 3 ml of Propionibacterium culture. An excellent yogurt was also produced with good body and a final pi of 3.9. In addition, yogurt was prepared by mixing 1/2 ml of Hansen's culture with 3 ml of Propionibacterium in 200 ml of I milk which had been fortified with 9 q of nonfat dry milk. The final pi was 3.98 and an excellent, uniform body yogurt was produced. Thus, the presence of the ProPionibacterium did not inhibit the production of yogurt in any way.
This is quite important since Pro~ionibacterium are known to produce carbon dioxide which might disrupt yogurt formation, but this was not observed. A
combination of lactobacillus and propionibacteria bus been reported to increase carbon dioxide production ("Stimulating Effect of Lactobacilli on the Growth of Propionibacteria in Cheese," F.F.J. Nieuwenhof, J.
Standhouders and G. Hut, Net. Milk Dairy J. 23, p.
287-239, I969).

In this example, yogurt was prepared by inoculating pasteurized 6.25~ nonfat dry milk in water with ProPionibacterium~ waiting 12 hours and then inoculating with commercially available Hansen's*yogurt culture. The initial inoculum of the Propionibacterium was either 1, 2, 4, or 10 I of the mature cultures. At the end of 12 hours, the pi values were, respectively, 5.5, 5.9, 5.2, and 4.2. A small amount of carbon dioxide was produced during this time period as determined by visual observation. Each of these media was then inoculated with 1 ml of Hansen's yogurt culture and the milk was allowed to incubate for an additional 8 hours. At the end of this time, the pi values of the samples were, respectively, 4.5, 4.4, 4.5~ and 4.6 and the viscosity, texture and flavor of the yogurt were very good.
In addition, an experiment was conducted ,.
* Trade Mark ~Z~8~94 wherein 10 ml of ProPionibacterium was inoculated into the nonfat dry milk, allowed to incubate for 12 hours, and then heated to 85C for 40 minutes to kill the propionibacteria and then cooled to inoculation temperature and inoculated with Hansen's yogurt culture. This also produced an excellent yogurt with very high viscosity and smooth, fine texture. Thus, it has been demonstrated here that in the production of yogurt, the milk could be preincubated with propionibacteria and then inoculated with yogurt producing cultures either with or without pasteurization in order to kill the propionibacteria, and an excellent quality yogurt can be produced.
Since propionibacteria are known to grow slowly at refrigeration temperature and to produce prop ionic acid under these conditions (Hutting and Reinhold, The Propionic-Acid Bacteria-A Review, J. Milk Food Tuitional.
Vol. 35, No. 5, pp. 295-30; 1972) this example teaches that the propionibacteria can be grown to substantial numbers in milk and then yogurt cultures can be added to produce yogurt, and then the product could be stored in the refrigerator and, with time, viable propionibacteria could produce prop ionic acid or some metabolizes which would be inhibitory to the development of mold and yeast or some bacteria as taught elsewhere herein.
A similar effect could be obtained by inoculating yogurt milk with large numbers of Propionibacterium cells immediately prior to inoculation with yogurt bacteria. The large number of I propionibacteria might be obtained either by concentration or growing to large numbers with internal or external naturalization of the culture media. There is no teaching in the prior art that the propionibacteria can be used either by preincubation or by simultaneous inoculation with yogurt producing bacteria in order to produce a good quality yogurt which would be inhibitory to mold and yeast growth.

. .

SLY

Yogurt was prepared by pasteurizing 2% milk at 90C for 5 minutes, cooling to 30C and pouring 200 ml each into different autoclave clear, plastic, 8 ox wide mouth jars with screw top caps. Nonfat dry milk powder (9 gym) and sucrose (11.5 gym) were added to each cup and mixed well by shaking.
Each cup was placed in a constant temperature water bath (42-44C) and sterile, disposable pipes were used to inoculate with cultures as shown in the following tables. The screw-top caps were tightened, container contents mixed well by shaking, and the water bath temperature adjusted to 35C i 0.5C, and the inoculated samples were incubated overnight.
Samples were collected the next morning and allowed to stand undisturbed, without opening the caps, at room temperature or in the refrigerator as indicated in the following tables. Each sample was visually inspected (through the clear plastic cup) periodically for the development of mold, sample viscosity and gas formation as evidenced by bubble or "eye" formation.
The results are given in Tables I and II.

~Z18894 TV ¦ _ I S
_ f Jo ., Jo us Jo ,_ t Jo ,. _ z . S .

n I I. J
I I- 1_ I

'.~

lZ~3894 a c o _ u a -- Ox o o ., 3 8 o U
I, ox _ C U
o _ o o a _ _, u u 8 o I
., O O
_ O
I
O o C ox.
a o _ O
o. o a u "
g --Jo "U O
U C OX _ O
_ I. a c 9 E I-I, o o " n O
Lo U

0, o o C o O _ .. O O D _ a o I

a (' I I a u o n a 8 Jo E 5 I. u 0 u a I o c _ _ _, -_ Jo , .

~2~8~39~

The data in Table I show that the inclusion of propionibacteria along with usual yogurt-producing organisms allows production of a good consistency yogurt which can retain a good consistency if there is an appropriate balance in the amount of each culture used.
Those samples which developed gas bubbles or "eyes" may be desirable products if the gas production is not excessive since the appearance is not unpleasant and might be useful in marketing 'iSwiss-style" yogurt.
The room temperature data also show that mold developed in 7 days in the control sample (A-7) but was delayed by incorporation of propionibacteria until the Thea day or longer. This "doubling" of the "shelf-life"
is a substantial improvement.

.

C : ' Lo _ L I
C, A

~Z~889~

O I. 1 2 , , O

_ . . S . , r `'~

Jo I _ if l , V A
l .

I mm m c: o m n I' = S

', I

~21t3894 The data in Table II confirm the data in Table I in that the presence of propionibacteria increase substantially the length of time prior to appearance of mold; also the growth of mold, once it does appear, is much more rapid or "heavier" in the controls without propionibacteria.
Gas chromotographic data indicate that no much prop ionic acid was produced in these products. The data are shown below Sample Identifi-Parts per Million ligation Number of Prop ionic Acid Aye Controls B-7 7.9 A-2 6.9 A-4 9.5 B-2 16.7 This example teaches that propionibacteria can be incorporated in the production of yogurt and will extend the shelf-life of the yogurt.

Propionibacterium Sherman (ATTICS Strain 9617) was grown in 800 ml of water containing either 62.4 g of a commercially available bulk starter medium (Formula l-Phase 4, Galloway West Company, Fun Du lag, Wisconsin) or in a formula (formula 2) which contained in 800 ml of water, whey (28.8 g), yeast (4.0 g), diammonium phosphate (10 g), and citric acid MindWrite (1 g).
Each formula was pasteurized at 85C for 45 minutes, and then cooled to 31C and inoculated with 10 ml of Propionibacterium Sherman 9617 which had been grown for 20 hours at 31C in milk. The organisms were allowed to incubate at 30C for 90 hours and at that time 1 ml and 5 ml samples of the growth media were transferred from each of the flasks and separately mixed with 20 ml of potato dextrose ajar (29.25 g of potato dextrose ajar in 500 ml of water) and the pi was reduced ,:

~L8~39~

to 3.1 to 3.7 before pouring the plates. Three plates were made for each dilution and they were dried overnight at room temperature, inoculated 24 hours later with 0.2 ml of mold and 0.2 ml each of two different types of yeast. Yeasts and mold had been isolated from commercially available yogurt. These plates were then incubated at 31C for 48 hours. The results were the same for the two formulae tested in that 1 ml of the growth media did not inhibit yeast or mold and 5 ml of the growth media did not inhibit yeast but did inhibit mold growth. After 5 days, there was no growth of mold on any of the plates which had been inoculated with 5 ml of the growth media. It should be noted that control plates supported growth of mold and yeast after only 30 hours.
The above growth media which had been allowed to grow for 90 hours were allowed to continue to grow for an additional 10 hours and then the pi was raised to between 7.5 and 8.0 with a slurry of calcium hydroxide and the 800 ml of growth media for each formula was divided into two 400 ml portions and dried two different ways. The first 400 ml portion of each growth medium was evaporated to semi dryness using a rotary evaporator and a water vacuum at a temperature of about 70C and then the powder was placed in a vacuum oven at 80C and dried; finally the powder was pulverized using a hammer mill. The initially light cream colored growth medium produced quite a dark brown powder.
A quantity (300 ml) of each formula (1 and 2 above) was lyophilized by freeze-drying after 100 hours of growth. This produced a creamy white powder.
Powder (5 g of each of the above dried powders) was added to 50 ml of water and 3.9 g of potato dextrose ajar was added to 50 ml of water and the liquids were autoclave separately, mixed together, and the pi was lowered to 3.6 using hydrochloric acid before the plates were poured. Plates were then allowed to dry overnight I.. .:~

12~B894 at room temperature. Three plates were made for each powder and 0.2 ml of yeast (two separate types) and 0.2 ml of mold were each transferred to each plate as a spread plate. The mold and yeast had been isolated from a commercially available contaminated yogurt sample as one loopful of culture which was transferred to 5 ml of sterile water which was vortexes well and then transferred as a 0.2 ml sample to each plate. Plates were then incubated at 31C for 40 hours. The results lo were quite spectacular in that there was no growth of either type of yeast or of mold on any of the plates which contained either the freeze-dried or the heat dried powder. However, there was extensive mold and yeast growth on the control plates.
Yogurt was prepared by using 200 ml of I milk and following the usual procedures for preparation of yogurt with inoculation of 0.5 ml of commercially-available Hansen's yogurt producing bacteria. The incubation temperature was 35C for 17 hours. The control sample was fortified with 9 g of whey and 11.5 g of sucrose; 20% of a commercially available strawberry fruit flavor was added after the yogurt was formed. A "plain" control was also prepared which did not have the added fruit flavor. In addition, a control fortified with 9 g of nonfat dry milk and 11.5 g of sucrose with added fruit Flavor was also prepared.
The initial pi of each of these preparations was 6.45 to 6.5 and the final pi was 4.15 to 4.4. Each produced a very smooth, uniform yogurt product with an excellent consistency and each of these exhibited mold growth on the surface beginning between the Thea and Thea day after production of the yogurt.
A yogurt formula was also prepared containing the same ingredients as the control above except that 9 g of whey was replaced with 7 g of whey plus 2 g of the heat dried powder produced by growing Propionibacterium in Formula l. Flavor was also added after formation of ~21~3894 the yogurt. The initial pi of the milk before formation of yogurt was 6.5 and the final pi was 4.6. The final product was discolored because the heat dried powder was discolored and the product was somewhat separated and was not uniform in viscosity. Mold growth was inhibited and did not appear on the surface of this formula until the Thea day.
An additional yogurt sample was prepared which contained only 1 g of Formula 1 and 8 g of whey and this product was very similar to the preceding product in appearance and the mold did not appear on the surface of this product until the Thea day. It should also be noted that there appeared to be some yeast present since carbon dioxide was slowly evolved from this product.
lo Yogurt was also prepared which contained 6.3 g of the heat dried powder from Formula 2 identified above and this product was prepared in 200 ml of 2% milk which had been fortified with 11.5 g of sucrose. After preparation, the fruit flavor was added. This product had an acceptable consistency and did not show any evidence of mold growth by the Thea day.
The freeze dried powders identified above were also used to produce yogurt. The control yogurt was made from 200 ml of 2% milk fortifies with 9 g of whey and flavor was added after production. In addition, a control was prepared which was not flavored and a control was prepared which was fortified with 9 g of nonfat dry milk and was flavored. All controls produced excellent products and developed mold growth after the Thea day of storage at room temperature.
The samples containing the freeze dried powder were all flavored after production and were all made from 200 ml of 2% milk. The first sample was fortified with 4-1/2 g of whey and 4-1/2 g of powder which had been freeze dried and was produced by growing the Propionibacterium in a commercially-available medium (Formula 1). The second sample contained 2 g of the ~Z~8894 same freeze dried powder and 7 g of whey; the third sample contained 4-1/2 g of the freeze dried powder from Formula 2 above and 4-1/2 g of whey; the fourth sample was fortified with 2 g of the freeze dried powder from Formula 2 above, and 7 g of whey. In all of these cases, a product with excellent consistency was formed and there was no apparent yeast or mold growth after 6 days in any of these products treated with the freeze dried powders. The color was much better than when the powder had been heat dried although there was some odor which appeared to be from the whey base growth media in these final products.
This present example shows that a product can be produced by growing propionibacterium on whey base media which, after drying, will be effective inhibiting yeast and mold growth in dairy products.
It is apparent from other examples herein that the Propionibacteria can be grown in nonfat dry milk and a product with a very acceptable odor and color can be produced after drying the growth media.

In 800 ml of water was prepared whey (100 g), yeast extract to g), and diammonium phosphate (8 g) and this is referred to as Formula 1 for this example.
Formula 2 for this example consisted of whey (100 g), yeast (5 g), and diammonium phosphate (12 g). Each formula was pasteurized at 85C for 45 minutes, the temperature was allowed to cool to 35C, and the growth media were then inoculated with 10 ml of Propionibacterium Sherman (10 ml) strain 9617 which had been grown in sodium lactate broth. These culture media were allowed to grow for 90 hours and then 5 ml of each culture was mixed separately with 20 ml of potato dextrose ajar (19.5 g for 250 ml) and the pi was lowered to 3.4 using hydrochloric acid before the plates were poured. Three plates were made for each formula.

j I..

~Z1~3894 Plates were dried overnight and then inoculated with 0.2 ml of either mold or each of two different kinds of yeast and then incubated for 24 hours at 31C. Yeast and mold growth appeared on control plates within 36 hours but yeast did not appear on any of the plates containing the growth media until after 6 days and mold did not appear on any of the plates containing growth media for up to lo days.
The culture was allowed to continue to grow for lo an additional 6 hours or a total of 96 hours and then the samples were divided and a portion of each growth medium was freeze dried and a portion was heat dried.
The powders so prepared were mixed (5 g of powder plus 50 ml of water) with potato dextrose ajar (3.9 g of ajar plus 50 ml of water). It should be noted tot the pi of the growth cultures was not raised before drying, which is different from a previous example. Each of the solutions prepared were autoclave and then the potato dextrose ajar and the freeze dried or heat dried powder solutions were combined, the pi of each plate was adjusted to 3.0 to 3.6 with hydrochloric acid, and the plates were poured. After drying, each plate was inoculated with 0.2 ml of yeast or 0.2 ml of mold, and the plates were incubated for 72 hours at 31C. It should be noted that the pi of some of the dried powder plus water solutions was raised to about 7.5 prior to autoclaving and the pi was not raised up to 7.5 before autoclaving for some other preparations.
When the pi of the heat-dried powder plus water mixture was increased prior to autoclaving, there was complete inhibition of yeast and mold for both Formulas l and 2 prior to 72 hours and then some mold did appear on the plates after 72 hours of incubation.
For the freeze-dried powder when the pi was raised prior to autoclaving, the plates were negative for yeast and mold for at least 48 hours.
For the freeze dried powder (Formula 2) when . .

~Z18894 the pi was not raised prior to autoclaving, the plates were negative for both yeast and mold for 48 hours.
The growth medium which had not been dried was centrifuged and the supernatant was filtered through a 0.22 micron diameter pore size filter and S ml of the filtrate for each formula was mixed with potato dextrose ajar and the plates were inoculated as indicated above.
The filtrates were inhibitory to both yeast and mold as indicated by a more rapid growth of yeast and mold on control plates than with those plates which contained filtrate. After 72 hours, the supernatant from Formula 1 above did allow the growth of yeast and was marginal with respect to slight growth of mold. However, the supernatant from Formula 2 completely prevented yeast and mold grown for the entire 72 hours.
This experiment shows clearly that propionibacteria growth medium can be converted to a dry powder either by heat drying or freeze drying and the powder produced will inhibit yeast and mold. In addition, this example demonstrates that the growth medium supernatant liquid contains inhibitory substance(s) which can be used to inhibit yeast and/or mold.

Yogurt was prepared by inoculating 200 ml of 2%
milk fortified with nonfat dry milk (4.5 g) and sugar (11.5 g) as a basal formula. Commercially available yogurt cultures (Hansen's) were used.
Propionibacterium (strain 9617) was grown for 96 hours in a mixture of whey (200 g), diammonium phosphate (12 g), and yeast extract (5 g), in 800 ml of water (identification symbol p-a) or in a formula of nonfat dry milk (200 g), diammonium phosphate (12 g), yeast (5 g), in 800 ml of water (identification p-b).
Each of these growth media were then either freeze dried or heat dried as described elsewhere herein. The i. .

~LZ~8~394 following combinations of ingredients were then used to prepare yogurt. The basal media also initially contained sodium caseinate (1.0 g) unless specified otherwise below and in some cases a fruit flavor (10.0 g) was added after the yogurt had been formed. The following 12 formulae were tested.
1) Basal medium plus p-a (4.5 g) which had been freeze dried, 2) Basal medium plus p-a (4.5 g) which had been heat dried, 3) Basal medium plus p-a (4.5 g) which had been freeze dried and no flavor,

4) Basal medium plus p-a (4.5 g) which had been heat dried and no flavor,

5) Basal medium plus p-a (4.5 g) which had been heat dried and no sodium caseinate,

6) Basal medium plus p-a which had been heat dried, 4.5 g and no sodium caseinate and no flavor,

7) Basal medium plus p-b which had been freeze dried, 4.5 g,

8) Basal medium plus p-b which had been heat dried, 4.5 g,

9) Basal medium plus p-b which had been freeze dried, 4.5 9 and no flavor,

10) Basal medium plus p-b which had been heat dried, 4.5 g and no flavor,

11) Basal medium plus p-b which had been heat dried, 4.5 g and no sodium caseinate,

12) Basal medium plus p-b which had been heat dried, 4.5 g and no sodium caseinate and no flavor.

~Z~889~

Each of the above products produced an excellent yogurt with excellent viscosity and texture.
Those yogurts which had been prepared with p-b often had a solid or precipitate on the bottom which was readily dispersed and mixable with the products which contained flavor since the stirring of the flavor into the yogurt distributed any solid which had settled to the bottom.
It is possible that the source of the solid was coagulated proteins which developed during the growth of ProPionibacterium prior either to freeze drying or heat drying of the growth medium when nonfat dry milk was used as part of the nutrient medium, since a layer of solid or precipitated material did not develop in those formulae prepared with p-a which employed whey rather than nonfat dry milk.
The above products were prepared and stirred and three additional cups of yogurt which were purchased commercially were placed in identical containers and stirred and set out with these 12 products for a taste test by 3 individuals experienced in dairy microbiology and yogurt production. Combined, these individuals had 50 years experience in working with dairy products and have been involved in taste panels for judging yogurt.
In general, the products prepared in this example were judged to be equal to the commercially-purchased products, and the preferred formulae were numbers 3, 4, 5, and 12 although the three "taste experts" did not agree on the order of superiority.
These products were kept in the refrigerator and examined daily. A commercially available flavored product exhibited large amounts of gas production and a yeasty odor and flavor 6 days after purchasing. A plain yogurt which had been purchased showed growth of mold on the surface after 20 days in the refrigerator. An additional flavored yogurt which had been purchased also showed physical separation and had a very bad yeasty odor. There was no appearance of yeast in the formulas Lo`' aye -- Jo --1-12 above after 26 days and the first appearance of mold appeared on the surface of a few samples after 26 days. Most of the samples did not exhibit mold or yeast growth for a prolonged period although the exact appearance date of contamination was not recorded in this experiment.
In a separate experiment with similar formulations stored at room temperature, mold often appeared on the surface of control formulations in four or five days for products which had been flavored but those which contained 2.25~ of the powdered growth medium of propionibacteria did not demonstrate mold growth until almost three times as long (i.e. 12 to 14 days) and in one exceptional case mold appearance was delayed for over 40 days, even though the container was opened and checked daily.

2. Cozily "Cozily*" it a commercially available "non cultured yogurt" product containing stabilizers, carbide rates and protein Chocolate, apricot, and blueberry flavored Kisses we're obtained commercially;
and each was divided into controls and test samples.
The controls developed gas bubbles at room temperature after three days and mold developed on the top of the chocolate and apricot control icily aster five days, although mold did not appear in the blueberry product at that time. In addition, the chocolate flavored control developed mold after 16 days in the refrigerator and the apricot flavored product developed mold after 21 days in the refrigerator.
The test samples were incorporated with 2-1/4-of the powder produced by drying propionibacterium growth medium as described in Example 9 herein, and the products were then stored at either room or refrigeration (2-5C) temperature in a study design I- * Trade Mar ~218894 which paralleled the controls described in the above paragraph. After 30 days, there was no mold or yeast in any of the Kisses products which had the dried, powdered propionibacteria growth medium added at either room temperature or refrigeration temperature. This example further indicates that the materials produced by growing propionibacterium can be effective at inhibiting both yeast and mold in dairy products. The appearance, odor and flavor of these products with the extended shelf-life were all excellent.

3. Fruit Juice he shelf life variety of fruit juices and related products was extended using propionibacteria according to the present invention Apple cider which had been filtered prior to packaging and another sample which had not been filtered, grape juice frozen concentrate, concentrated raspberry yogurt flavor and concentrated cherry yogurt flavor for use in yogurt were commercially obtained.
Small paper cups were prepared containing about 100 g of each type of apple cider and of the concentrated grape juice and of the grape juice after it had been diluted according to manufacturer's instructions, and of the cherry yogurt flavor and of the raspberry yogurt flavor.
Five cups of each was prepared with one cup of each serving as a control. Other cups were inoculated with yeast and/or mold obtained from commercial yogurt.
A second cup of each food product was inoculated with 1 ml of viable yeast cells only; and a third cup was inoculated with 1 ml of viable mold only. The fourth cup was inoculated with 1 ml of viable yeast and I of a heat-dried powder produced from growing propionibacteria medium according to Example 9, and the fifth cup was inoculated with 1 ml of viable mold and 3% of the ~21138~4 powdered propionibacteria growth medium. These were then stored at room temperature (uncovered). The following key was used to identify the effects:
1. Apple cider, filtered a. Mold 2. Unfiltered apple cider b. Yeast 3. Diluted grape juice c. Mold plus ProPionibacterium growth medium 4. Concentrated grape d. Yeast plus juice Propionibacterium growth medium 5. Raspberry yogurt e. Control flavor On the second day, cup l-b (apple cider filtered and inoculated with yeast) developed a yeasty odor and cup 3-b (diluted grape juice inoculated with yeast) also developed a yeasty odor. On the third day, cup l-a developed mold and cup 3-a also developed mold, and cup 4-b (concentrated grape juice inoculated with yeast) developed yeast On the Thea day cup l-e had mold, and cup 4-a had mold. However, cups c and d which were the products inoculated with mold and yeast and the powdered Propionibacterium growth medium did not have any mold or yeast even after four days at room temperature as contrasted to those samples which did not contain the Propionibacterium growth medium as described above.
This demonstrates a dramatic usefulness of the propionibacterium growth medium in inhibiting yeast and mold in a variety of food products.

P. Sherman ATTICS 9616 was grown in frozen-reconstituted orange juice fortified with 0.1%
yeast extract for two days at 30C~ A commercial orange juice concentrate was diluted 50%, 10~, 5% and 1%. The pi was adjusted to 5.0 with 1.0 molar Noah. Three sets . .

lZ~8894 of the above dilutions were made. One set was autoclave, the second set pasteurized, and the third set received no heat treatment. The three sets were inoculated with I P. Sherman culture and incubated at 30C for five days. Centrifugation was carried out on all the three sets and the supernatant was filter sterilized and used (5%) against four different molds.
Controls were made from the orange juice receiving the above treatment without P. Sherman inoculation but inoculated with the different molds.
Results appear in Table III:

, , .

a + + + + + + + +
++++ .++++
two ++++ ++++
v Jo ++++ I++
o lo ++++

N l l + l l l +
1++1 11+1 O 1++1 11+1 .' .' I fill fill 11+1 +1+1 I ~.1+1 11+1 H O Jo D I I + I
O 11+1 +1+1 .
O
' e 61~161 Eye I
o o ::
Jo mu l mu :: I
I O C O X e I: U Lo O 1:: U L. O
o C I C I I lo D

I :

lZ1~

These results show that propionibacteria grown in orange juice produce substances inhibitory to molds and yeast.

4. Cottage Cheese For the purpose of this disclosure, cottage cheese is defined as a soft, uncured cheese made by coagulating pasteurized skim milk and separating the curd from the non coagulated liquid.
Typically, cottage cheese has at least 20~
solids by weight. There are two common types of cottage cheese plain and creamed. The latter is raised in fat content by stirring in cream or a creaming mixture.
US. Government standards require 4% fat content in creamed cottage cheese, but partially-creamed cottage cheeses, having a fat content of 0.5 to 2% are also sold in the United States.
Coagulation may be accomplished by the addition of lactic acid-producing bacteria, a suitable milk-clotting enzyme and/or a food grade acid.
Propionibacteria can be incorporated into cottage cheese with an effect similar to that described above for yogurt and other food products. The bacteria can be produced separately from the cottage cheese and concentrated after growth and prior to utilization or just grown to high numbers and then added to the cottage cheese "cream". Alternatively, the cottage cheese "cream" can be cultured with the propionibacteria prior to creaming. Further, the propionibacteria can be used along with usual cultures to "set" the cottage cheese from the beginning of the cottage cheese manufacture as in the case of yogurt.
To produce one antimicrobial growth mixture particularly suitable for use in cottage cheese and other dairy products, a Propionibacterium culture is started in any suitable lactate broth to condition the culture to metabolize lactic acid. The resulting ~Z~88g4 culture is then used to inoculate a much larger volume of a growth medium that may include as nutritive ingredients a source of milk protein, or milk protein derivative, or a milk protein substitute and lactic acid, for example skim milk acidified to pi 5.0 to 5.5 with lactic acid. A higher initial pi could be used, but the medium would then be suitable to sustain the growth of numerous possible contaminating microorganisms. The inoculated medium is allowed to grow at a favorable temperature for as long as necessary to form a growth in the mixture that is active against spoilage microorganisms.
The immediately following examples and Example 21 below illustrate the usefulness of propionibacteria growth mixtures in the preservation of cottage cheese and, particularly, in the inhibition of psychrotrophic gram negative, slime-producing cottage cheese spoilage organisms.

Flasks (100-ml capacity containing magnetic stirrer bars) of nonfat milk containing 0.1% yeast extract were pasteurized at 85C for 45 minutes. They were cooled to 30C and acidified to pi 5.3 with 10%
lactic acid. ProPionibacterium Sherman (ATTICS Strain 9616) was added at the rate of I from a 48-hour old culture grown up in sodium lactate broth (Tripticase, 10.0 g; yeast extract, 10.0 g; 60% sodium lactate solution, 16.7 ml; monopotassium phosphate, 0.25 g;
30 manganese sulfate, 0.005 g or 0.5 ml of a 0.1 M
solution; and water, 1000 ml; pi =7.0 before autoclaving at 121C for 15 minutes) at 30C. Flasks were placed on a six station magnetic stirrer and slowly agitated during the entire incubation period.
A control flask uninoculated with the propionibacteria was similarly treated. Samples were taken at 24, 48, 72 and 96 hours and assayed for acetic i I

~2~8~394 and prop ionic acids using a Model AYE Hewlett Packard gas chromatography equipped with a Model AYE
integrator. Samples were also tested for inhibitory activity against a psychrotrophic gram negative, slime-producing cottage cheese spoilage organism supplied by H. P. Hood, Inc., 56 Roland St., Boston, Mass. This organism was found by electron microscopy to contain both coliform (pericricnous flagella) and pseudomonas (monotrichous flagella) cells. These organisms are important cottage cheese spoilage bacteria (Dr. Paul Swenson, H. P. Hood, Inc., personal communication) and are now maintained in and available from the Department of Microbiology at Oregon State University.
The test for inhibition against this spoilage bacterium was carried out by adding 1%, 2%, I 4%, 5 and 10% (v/v) amounts of autoclave (121C for 15 minutes) P. Sherman 9616 milk culture taken at the various time intervals of growth (and autoclaving as obtained or neutralizing to pi 7.0 with calcium hydroxide prior to autoclaving) to 100 ml of crystal violet tetrazolium ajar (CUT ajar - Tryptone 5.0 g;
yeast extract, 2.5 g; glucose, 1.0 g; distilled water, 1000 ml; ajar, 15.0 g; adjust to pi 7.1; after autoclaving at 121C for 15 minutes, add filter-sterilized crystal violet at 0.001 g per liter and 2, 3, 5, triphenyltetrazolium chloride at 0.05 g per liter). Each lot of medium containing the various concentrations of the autoclave milk culture was then acidified to pi 5.3 with sterile 10% tartaric acid.
Several dilutions of an overnight lactose broth (Beef extract, 3.0 g; petunia, 5.0 g; lactose, 5.0 g;
distilled water, 1000 ml; pi 6.8-7.0) culture of the psychrotrophic spoilage organism were then made and 1.0 ml allocates added to sterile putter plates. About 10 to 15 ml of CUT ajar then was added and the plates incubated at room temperature (25C) for 48 hours. On i . -.;

~Z18~39~
- I -this medium the psychrotrophic spoilage organism grew as large, (l to 5 mm diameter) deep red, glistening colonies. The amount of inhibition in comparison to control plates containing no autoclave P. Sherman culture could then be calculated as follows for each concentration of culture taken at the different time intervals:
I Collins on Control - Colonies on Test Plate loo Colonies on Control lo Results obtained in analyzing the neutralized (pi 7.0) samples for acetic acid and prop ionic acids were as follows:
TABLE IV
Mel (Pam) ours Acetic Acid Prop ionic Acid Essentially the same data were obtained for unneutraliæed samples.
In analyzing the samples for inhibition of the psychrotrophic, slime-producing cottage cheese spoilage organism, the following results were obtained:
TABLE V
Percent inhibition of cottage cheese spoilage organism by various amounts of P. Sherman 9616 milk culture grown in pasteurized* milk.
Culture Added Time Sampled (hours) (%1 24 48 72 96 O O O O O
l 62 50 lo 96 2 56 lo lo 90 3 42 lo 44 97 4 88 lo 46 98 96 lo 52 97 3510 50 lo 36 97 *Milk pasteurized at 85C for 45 minutes and then acidified with lactic acid to pi 5.3.
A

lZ~L8~394 In another experiment, the Propionibacterium she manic 9616 culture was grown under the same conditions in nonfat milk except the milk was autoclave and not acidified to pi 5.3 with lactic acid. This was done because microscopic examination of the culture grown in the pasteurized (85Cm 45 minutes) nonfat milk revealed sporeformers emerging by 48-72 hours. In fact, the counts of heat tolerant forms (survive 60C for 30 minutes) at zero time and at 72 hours were 100 per ml and 107 per ml, respectively. It was felt this might confound interpretation of the data in the pasteurized samples. The following data were obtained with the autoclave milk grown cultures.

TABLE VI
__ Mel _ m) Hoarsest c Acid Prop ionic Acid 72 5~2 713 . _ _ _ __ In this case, we see the maximum acetate and preappoint production delayed until at least 96 hours in contrast to the cultures which were acidified with lactic acid to pi 5.3. This difference apparently is due to the fact that acetate and preappoint are produced from lactate which the propionibacteria did not have to produce in the acidified samples but which they produced gradually in the non acidified samples.
Data on inhibition of the cottage cheese spoilage organism are in the following table.

~Z1~3~394 TABLE VII
Percent inhibition of cottage cheese spoilage organism by various amounts of P. Sherman culture grown in auto craved* milk.
Culture Added Time Sampled (hours) (%) _ 48 96_ O O O O

_ _ . _ . _ . . , _ _ *Milk autoclave 15 minutes at 121C.

If one considers thaw prop ionic and acetic acids may be responsible for the inhibition seen, the time at which these acids are maximally produced should agree with the times at which samples show maximum inhibition of the spoilage bacteria. This, however, is not the case. In the case of the pasteurized, acidified growth culture, maximum acetate and preappoint occurred by 24 hours but maximum inhibition not until at least 48 hours. In case of the autoclave, non acidified milk, maximum acetate and preappoint occurred at 96 hours but maximum inhibition of spoilage bacteria at 24 to 48 2 hours. Therefore, these data suggest that some natural metabolizes other than acetate and preappoint may also be involved in the inhibition of the cottage cheese spoilage bacteria.

Propionibacterium Sherman (ATTICS Strain 9716) were grown in a sodium lactate solution for 48 hours as described in Example 15. five hundred gallons of skim milk were then pasteurized at 190F for 45 minutes, and subsequently cooled to 86F. The cooled milk was acidified using 85~ reagent grade lactic acid to a pi ox 5.3 and then inoculated with 0.5% of the .

12ï8894 ProPionibacterium Sherman culture. The inoculated milk was slowly agitated during incubation for 98 hours, and thereafter neutralized with sodium hydroxide to pi 7Ø The neutralized liquid was pasteurized at 1~5 for 20 minutes, cooled to ambient temperature (about 75F), pumped through sterile lines into six-gallon sterile plastic bags and then frozen.
When thawed, the liquid medium was very active in inhibiting the growth of the psychorotropic gram negative, slime-producing cottage cheese spoilage organism mentioned in Example 15.

220 gallons of milk fortified with 0.1% yeast extract were heat-treated at 85C for 45 minutes and then cooled rapidly to 30C. 86~ lactic acid (Sigma Grade) was then added with agitation until the pi was lowered to 5.3. The milk was then inoculated with a mature culture of 1.25 Propionibacterium Sherman (ATTICS 9716); and the culture allowed to grow for 48 hours. The final mature growth mixture (bacterial soup) was pasteurized, a portion retained as a liquid, and the remainder spray-dried using a commercial box-type spray dryer. The liquid and spray dried powders were incorporated into cottage cheese dressing to evaluate their effect on the keeping quality of the final creamed cottage cheese. The liquid and powder were substituted in the control dressing formulation for whole milk (for liquid) and nonfat dry milk (for spray dried powder) on a comparable solids substitution basis. The dressing was formulated at the upper level of inhibitor addition to be tested and lower levels were obtained by back blending at specified ratios with a control dressing. The final inhibitor addition level was calculated to be the percent of inhibitor on a solids basis per pound of finished cottage cheese.
A control cottage cheese was made along with all , ~8894 testing variables. Initially, in order to establish an effective but not excessive addition level of the inhibitor and also a reasonable inoculation level of the surface slime organism, various samples were evaluated with different inhibitor addition and inoculation levels. The slime inoculant was mixed in the cottage cheese at specified counts per pound of cheese. Control and inhibitor samples, both inoculated and non inoculated, were packed and sealed with heat sealed foil lids. Samples were set in storage for evaluation at 45 and 50 F.
Evaluations indicated a positive inhibitory effect of the spray-dried powder in test sample versus control in that the storage time at 45 and 50F was increased 7 to 10 days before surface slime growth appeared. Results with liquid which had 8.8% solids were similar. The percent inhibitor in the finished product was 0.00% (control) 0.12~, 0.24%, 0.36~ and 0.48%. All samples were inoculated with spoilage bacteria at 2000/lb. or 8000/lb. along with a non inoculated set of each inhibitor level as a control.
Results showing the number of days before surface slime growth appeared in samples from a 4000 lb. batch of cottage cheese are summarized in Table VIII:

TABLE VIII
Days at 50F
When Spoilage Inoculation level of Appeared Sample Spoilage Bacteria 16 Control 8000/lb.
23 Control 2000/lb.
23 0.12% Inhibitor 8000/lb.
44 Control Non-inoculated 56 Owls% Inhibitor 2000/lb.

Flavor evaluations indicate that all 3 non-inoculated inhibitor-containing samples were acceptable after 44 days at 50F.

121~389~

Thus, the inhibitor extended significantly the number of days at 50F before surface slime growth appeared.
Further testing of these products was carried out in cottage cheese as above with the following conditions:
(a) Control - No inhibitor added.
(b) Spray dried inhibitor addition - 0.30%
lovely.
(c) Spray dried inhibitor addition -reconstituted and pasteurized at 255F - 36 sec. - 0.30~ level.
(d) Skim based liquid - 14.50~ solids - 0.15%
level.

Condition (c) was set up to safeguard against possible microbial contamination by the inhibitor powder. Condition (d) was included to compare the inhibitory activity of the liquid product versus the spray dried sample.
The following table shows results of these tests where the number of days before the appearance of spoilage slime is given:
TABLE IX
Days When Conditions Spoilage Appeared 6 days Control - Inoculated 14 days Condition (b) - Inoculated - 1 of 9 samples 17 days Condition (b) - 100%
Condition (c) - 100 From these observations it is clear that the spray dried material also displays inhibitory activity I

sly but it appears somewhat less active than the liquid samples, especially since samples under condition (d) showed no surface growth.

5. Other Food Products The foregoing examples describe some specific applications of the present invention. It is anticipated that propionibacteria growth mixtures will prove to be safe and effective for use in the preservation of other food products. Results similar to those obtained for yogurt and cottage cheese are expected for the use of propionibacteria metabolizes to preserve such dairy products as fluid milk, half and half, whipping cream, sour cream, and the like wherein psychotropic bacteria are a common spoilage organism.
The metabolizes should also be particularly suitable for use in the preservation of bendable food products such as ground meat or in sausages.
It is also anticipated that such growth mixtures can be applied in liquid form to solid food products such as fresh fruits and vegetables. Application could be by spraying, immersion, or injection.

C. SEPARATED METABOLIZES OF PROPIONIBACTERIA
As discussed above in Example 10, the supernatant liquid of a propionibacteria liquid growth mixture was effective in inhibiting spoilage microorganisms. It has been further discovered that the supernatant is the most effective fraction in inhibiting mold and yeast. And, surprisingly, it is now found that solids separated from the supernatant actually stimulate microbial growth. It is anticipated that the solids fraction can be added to growth media and the media then used for an accelerated growth of such commercial useful microorganisms as Streptococcus lactic, Streptococcus creamers, Lactobacillus bul~aricus, Streptococcus thermophilus, Lactobacillus acidoPhilus~ Leuconostoc Jo ~X18~94 species, Lactobacillus planters, and Pedro coccus cerevisiae.
The unique properties of propionibacteria growth mixture fractions are illustrated by the following examples.

Forty liters of milk fortified with 0.1% yeast extract were heat-treated at 85C for 45 minutes in a Fermacell fermenter model FG-50. The milk was cooled rapidly to 30C and then acid-treated with 86~ lactic acid (Sigma Grade) to pi 5.3. The milk was then inoculated with 2% of a 48 hour-old culture of Propionibacterium Sherman (Strain ATTICS 9616). After four days, samples were taken aseptically and placed in a sterile flask. Two 500-ml amounts of culture were centrifuged at 10,000 rum for 10 minutes. The supernatant was placed into sterile bottles while the sediment was centrifuged and washed three times with distilled water. One 250-ml portion of supernatant was neutralized with 1.0 molar Noah to pi 7.0 and the other 250-ml portion was used unneutralized. The clean sediment and the two portions of supernatant were left at room temperature. In two days, the sediment developed a sharp putrid odor and was obviously spoiled. The supernatant fractions, on the other hand, were clear and fresh and smelled like Swiss cheese.
They have remained fresh and clear for over three months. The two supernatants showed no sign of contamination. This reveals the wide range of inhibitory activity residing in the supernatant portion in contrast to the sediment which rapidly spoiled.
Fractionation of such inhibitory activity in this manner was unexpected.

Ten percent of the two portions of the , 1~113894 supernatant material from Example 18 was added to an acidified malt extract ajar pi 3.5 (bacto-malt Extract B-186) and poured onto plates. The plates were allowed to solidify and dry overnight and 0.2 ml of seven different molds (Asperqillus nicer, Fusarium oxysporum, Respace species, Mocker species, yogurt mold, cottage cheese mold, and Penicillium rcqueforti) were spread over the ajar surface. The controls were made with only acidified malt ajar inoculated with the molds. Ten percent of the sediment was subjected to the same treatment. In two days the controls showed nominal mold growth while the plates containing the sediment were very heavily overgrown with the molds. In the supernatant-containing plates, six of the molds were totally inhibited while P. roqueforti was only partially inhibited. The sediment clearly stimulated mold growth so these plates had more growth than the controls.
The same experiment was done using two yeasts tKluYveromyces fragile and a yeast isolated from commercially available orange juice). A pronounced inhibition by the supernatant material was seen again in this case. This inhibition by a culture fraction of Propane bacterium, especially of yeasts, and the stimulant effect of the sediment was unexpected.

In experiments to test for inhibition of a gram negative cottage cheese spoilage organism (slime producer) supplied by H. P. Hood, Inc., 56 Roland St., Boston, Massachusetts, lactose broth (Disco) was acidified with 10% tartaric acid to pi 5.0 and dispensed into test tubes for sterilization at 121C for 15 minutes. The supernatant material from Example 18 was added to two sets of test tubes containing the lactose broth to make final concentrations of 40%, 20%, 10%, 5%, 2.5%, and 1.2%. One set was inoculated wit 0.2 ml of the slime producer and the other was not inoculated. A
A

~Z18894 control was 10 ml of lactose broth (pi 5.0) inoculated with slime producer. All the test tubes were incubated at 30C for 24 hours. Turbimeteric readings were taken on all the test tubes at 600 no with a Perkin-Elmer spectrophotometer model 35.
Results were as follows:

do ~2~88~4 n Jo ..
US
TV o o Al or us . C
. Q VOW O O I N (I') Lo a owe o Ye v Jo OWE
a I v a o w . U
or + a 0 o N I Us 0 V
W 000000¢ 1~5 C U
X Ø
It O 0 v S so ill S

TV
v vow o I o or v v z duo o o o o o E
z a Jo v O o . E
111 N O ED 0 'r I ."
I O O O O O O V U
. I IT V
a u c + I 1`'1 o ox v c I 111 Lo US O O O O O I O O S

0 Us V
S ., O O O 11' Us N .,~ ... .0 U 1-1 or N Jo Z C

f ,,~" Jo ~Z~889~
59 .
This example shows, therefore, that the natural metabolizes of P. Sherman will inhibit gram negative psychrotrophic bacterium which spoil refrigerated food such as cottage cheese.
Graphs showing these results appear as Figs. 1 and 2. Fig. 1 shows inhibition of gram negative psychrotrophic cottage cheese spoilage bacterium by 4 day-old neutralized ON P. Sherman supernatant (inhibitor). Fig. 2 shows inhibition of gram negative psychrotrophic cottage cheese spoilage bacterium by 4 day-old unneutralized (4/U) P. Sherman supernatant (inhibitor).

Milk (220 gallons) fortified with 0.1% yeast extract in a closed vat at a commercial dairy center in Port land, Oregon, was inoculated with 1.25~
Propane bacterium Sherman (ATTICS 9616) culture after being treated the same way as in the Example 18.
Samples were taken at 48 hours and treated with rennet (1/1000) to precipitate the proteins. Centrifugation was carried out at 5,000 rum for 5 minutes. The supernatant was filtered on a Whitman No. 1 filter paper. The filtrate was then neutralized to pi 7.0 with 1.0 molar Noah.
Two gram negative organisms, the cottage cheese slime producer of Example 15, and a gram negative isolate from raw goat's milk, were tested as before for the inhibitory effect of the filtered supernatant in lactose broth preadjusted to pi 5Ø Final concentrations of the inhibitor were 20%, 15~, 10%, 5%, 2.5% and 1.2%. Results are summarized in Tables XI and XII.

~LZ1~894 Twill I
tnhlbltloo of the sly producer (SUP).

Z lnhlbltorSP I I Difference Z lnhlbl~lon 20'. 0.~440.878 0.066 92%
15% owe 0.069 92Z
10% 0.8630.667 0.196 80%
5% 0.9500.547 0.403 51%
2.5~ 0.9350.439 0.496 40~
1.2;: 0.8610.371 0.490 40;:
Control a. 0.830 T~BLB XII
Inhibition of rho goat r~llk solace (GO).

Z lnl-lbl~or GO + I I Difference X lnhlbltlon _ _ 20Z 0.917 OBOE 0.038 89,:
15,. 0.812 0.774 0.038 80%
lo 0.718 0.665 0.053 70%
5Z 0.574 0.512 0.062 67Z
2.5,. 0.526 0.457 0.069 63Z
1.2% 0.490 0.372 0.118 37%
Control a. 0.186 ~Z~8894 - 6] -These results appear in graphic form in Figs. 3 and 4. Fig 3 is a graph showing inhibition of gram negative psychrotrophic cottage cheese spoilage bacterium by supernatant of 2-day culture of P.
Sherman (inhibitor). Fig. 4 is a graph showing inhibition of gram negative bacterium isolated from raw goat's milk by 2 day-old supernatant of P. Sherman.
While we have described and given examples of preferred embodiments of our inventions, it will be apparent to those skilled in the art that changes and modifications may be made without departing from our inventions in their broader aspects. We therefore intend the appended claims to cover all such changes and modifications as fall within the true spirit and scope of our inventions.

A

Claims (34)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A process for preserving food products comprising providing in a food product, that is normally devoid of propionibacteria and is subject to spoilage by growth of an undesired microorganism, a material containing at least one metabolite formed by growing a culture of a propionibacterium, the material being provided in an amount sufficient to inhibit the undesired microorganism and comprising one or more metabolites that, in combination, have a greater inhibitory effect than a weight of propionic acid equal to the propionic acid content of the one or more metabolites.

2. The process of claim 1 wherein the food product is a fermented product and wherein the material is provided by including a Propionibacterium culture in a batch before fermentation.

3. The process of claim 1 wherein the food product comprises yogurt.

4. The process of claim 1 wherein the food product comprises cottage cheese.

5. The process of claim 1 wherein the food product comprises creamed cottage cheese and the material is formed by providing viable propionibacteria in the cottage cheese cream.

6. The process of claim 1 wherein the food product comprises a fruit or vegetable juice.

7. The process of claim 1 wherein the food product comprises ground meat.

8. The process of claim 1 wherein the food product comprises milk, half and half, whipping cream, or sour cream.

9. The process of claim 1 wherein the food product comprises bread.

10. The process of claim 1 wherein the food product comprises bread and wherein the material, provided in bread dough, is present in an amount sufficient to inhibit mold and in an amount insufficient to prevent yeast from performing its leavening function.

11. The process of claim 1 wherein the material is added in an amount sufficient to inhibit mold.

12. The process of claim 1 wherein the material is added in an amount sufficient to inhibit yeast.

13. The process of claim 1 wherein the material is added in an amount sufficient to inhibit psychotropic bacteria of the type which produce slime, off flavor, off odor, or off appearance in the food product.

14. The process of claim 1 wherein the material consists essentially of all the metabolites formed by growth of the culture.

15. The process of claim 1 for preserving a blendable food product comprising intimately mixing the material and the food product.

16. The process of claim 1 for preserving a solid food product comprising applying the material to the surface of the food product.

17. The process of claim 1 for preserving a solid food product comprising injecting the material into the food product.

18. The process of claim 1 wherein the material includes propionibacteria cells of the culture, at least some of the cells being viable.

19. The process of claim 1 wherein the material includes propionibacteria cells of the culture, the cells having been rendered not viable.

20. The process of claim 1 wherein the material is provided by growing the propionibacterium culture in a liquid growth medium to provide a liquid mixture containing bacterial metabolites.

21. The process of claim 20 further comprising condensing, by spray-drying, the liquid mixture so that the mixture, which includes the metabolites, can be added as a powder to the food product.

22. The process of claim 1 wherein the material is provided by growing the propionibacterium culture in a liquid growth medium to provide a liquid mixture containing bacterial metabolites; freezing the liquid mixture; and thawing the mixture, which includes the metabolites, for contact with the food product as a liquid.

23. The process of claim 1 wherein the material is provided by growing a propionibacterium culture in a liquid growth medium so that a mixture containing metabolites is formed; and after growing the culture, separating the mixture into fractions, one of which is the material.

24. The process of claim 1 wherein one of the metabolites contained in the material is propionic acid.

25. An extended shelf life food product comprising a food product and an added material containing at least one metabolite produced by an independently grown culture of propionibacterium, the material being present in an amount sufficient to inhibit an undesired microorganism and comprising one or more metabolites that, in combination, have greater inhibitory effect on the undesired microorganism than a weight of propionic acid equal to the propionic acid content of the one or more metabolites.

26. A process for preserving a food product comprising growing a propionibacterium culture in a medium compatible with a food product to produce a material, the material comprising one or more metabolites that, in combination, have greater inhibitory effect on spoilage microorganisms than does a weight of propionic acid equal to the propionic acid content of the one or more metabolites; and providing the material in a food product that is normally devoid of propionibacteria and is subject to spoilage by growth of an undesired organism.

27. The process of claim 26 wherein the medium consists essentially of skim milk and lactic acid in amounts sufficient to allow production of an inhibitory metabolite.

28. The process of claim 26 wherein the medium consists essentially of skim milk, yeast extract and lactic acid in an amount sufficient to allow the production of an inhibitory metabolite.

29. The process of claim 26 wherein the medium consists essentially of whey.

30. The process of claim 26 wherein the medium consists essentially of whey and yeast extract.

31. The process of claim 26 wherein, after growing the culture, the mixture is separated into fractions, one of which is the material.

32. A process for the production of an antimicrobial food preservation additive comprising growing a propionibacterium culture in a liquid growth medium so that a mixture containing metabolites is formed; and after growing the culture, separating the mixture into fractions, one of which is an active fraction containing one or more metabolites that, in combination, have a greater antimicrobial effect than a weight of propionic acid equal to the propionic acid content of the one or more metabolites.

33. The process of claim 32 wherein the separating comprises separating suspended solids from the liquid growth medium, the remaining liquid fraction being the active fraction.

34. A food preservation additive adapted to inhibit an undesired microorganism, the additive comprising one or more propionibacterium metabolites that, in combination, have greater inhibitory effect on the undesired microorganism than a weight of propionic acid equal to the propionic acid content of the one or more metabolites.